Hidradenitis suppurativa (HS) is a common and debilitating inflammatory disorder of the pilosebaceous unit that presents with recurrent scarring inflammatory nodules and sinus tracts in the intertriginous folds of the body. It is a complex condition that requires multimodal management to address the medical, surgical, and psychosocial needs of affected patients. However, it can be difficult to coordinate all that goes into HS management beyond the standard therapeutic ladder of topical and oral antimicrobials, intralesional corticosteroids, biologics, and surgery. In this article, I will outline 5 important aspects of HS treatment that often are overlooked.

Talk About Pathophysiology

Patients with HS often have limited understanding of their condition. One common misperception is that HS is an infectious disease and that disease activity is associated with poor hygiene.¹ Dispelling this myth may help patients avoid unnecessary hygiene practices, decrease perceived stigma, and enhance your therapeutic alliance.

The current model of HS pathophysiology implicates an aberrant inflammatory response to the cutaneous bacterial microbiome, which leads to follicular occlusion and then rupture of debris and bacteria into the surrounding dermis. Immune cells and inflammatory mediators such as nuclear factor κB and tumor necrosis factor α respond to the disruption. Chronic lesions develop due to tissue repair with scarring and re-epithelialization.^2,3 Although most patients probably are not interested in the esoteric details, I typically make a point of explaining to patients that HS is a chronic inflammatory disease and provide reassurance that it is not a sign of poor hygiene. 

Counsel on Smoking Cessation

Most HS patients use tobacco. As many as 75% of HS patients are active smokers and another 10% to 15% are former smokers. Although there is mixed evidence that disease activity correlates with smoking status, the Hidradenitis Suppurativa Foundation in the United States and Canada concluded in the 2019 North American Clinical Management Guidelines for Hidradenitis Suppurativa that due to the overall health risks of smoking, we should recommend cessation to our patients.⁴

Laser Hair Removal Works

Don’t forget about laser hair removal! Evidence from randomized controlled trials supports the use of the Nd:YAG laser in the treatment of HS. Treat the entire affected anatomic area and use stacked double pulses on active nodules (typical settings: 10-mm spot size; 10-millisecond pulse duration and 35–50 J/cm² in Fitzpatrick skin types I–III; 20-millisecond pulse duration and 25–40 J/cm² in Fitzpatrick skin types IV–VI).⁴ Especially if it is covered by your patient’s insurance, Nd:YAG is a great adjunctive treatment to consider. The guidelines also recommend long-pulsed alexandrite and diode lasers as well as intense pulsed light, all of which result in follicular destruction, though these treatments have less supporting evidence.⁴

Have a Plan for Flares

Intralesional injection of triamcinolone is a mainstay of HS treatment and provides patients with rapid relief of symptoms during a flare.⁵ One case series found that there was a notable decrease in pain, size, and drainage after just 1 day of treatment with intralesional triamcinolone 10 mg/mL (0.2–2.0 mL).⁶

Intralesional steroid injection is a great tool for quieting an active disease flare while simultaneously instating ongoing treatment for preventive management. However, even when disease control is optimized, patients may still experience intermittent flares of disease. For some patients, it may be appropriate to have a plan in place for a return to clinic during the beginning of a flare to obtain intralesional steroids. The ability to come in on short notice may help avoid visits to the emergency department and urgent care where your patients may receive treatments such as short courses of antibiotics or incision and drainage that may deviate from your overall treatment plan.

Consider Childbearing Status

Don’t forget to consider childbearing plans and childbearing potential when treating female patients with HS. Pregnancy is a frequent consideration in HS patients, as HS affects 3 to 4 times more women than men and typically presents after puberty (second or third decades of life). Many of the medications in the HS armamentarium are contraindicated in pregnancy including tetracyclines, retinoids, and hormonal agents. Surgery should be avoided in pregnant patients whenever possible, particularly in the first trimester. Relatively safe options include topical antibiotics such as clindamycin and metronidazole, as well as tumor necrosis factor α inhibitors, which are classified as category B in pregnancy.⁵

Before making treatment decisions in pregnant and breastfeeding patients, consult the US Food and Drug Administration recommendations. Perng et al⁷ reviewed current management strategies for HS in pregnant and breastfeeding women, and their review article in the Journal of the American Academy of Dermatology is an excellent resource.

Final Thoughts

Comprehensive management of HS may include a combination of medication and procedures, lifestyle modification, management of comorbidities, and social support. Formulating a good treatment plan may be a challenge but can drastically improve your patient’s quality of life.

Hidradenitis suppurativa (HS) is a common and debilitating inflammatory disorder of the pilosebaceous unit that presents with recurrent scarring inflammatory nodules and sinus tracts in the intertriginous folds of the body. It is a complex condition that requires multimodal management to address the medical, surgical, and psychosocial needs of affected patients. However, it can be difficult to coordinate all that goes into HS management beyond the standard therapeutic ladder of topical and oral antimicrobials, intralesional corticosteroids, biologics, and surgery. In this article, I will outline 5 important aspects of HS treatment that often are overlooked.

Talk About Pathophysiology

Patients with HS often have limited understanding of their condition. One common misperception is that HS is an infectious disease and that disease activity is associated with poor hygiene.¹ Dispelling this myth may help patients avoid unnecessary hygiene practices, decrease perceived stigma, and enhance your therapeutic alliance.

The current model of HS pathophysiology implicates an aberrant inflammatory response to the cutaneous bacterial microbiome, which leads to follicular occlusion and then rupture of debris and bacteria into the surrounding dermis. Immune cells and inflammatory mediators such as nuclear factor κB and tumor necrosis factor α respond to the disruption. Chronic lesions develop due to tissue repair with scarring and re-epithelialization.^2,3 Although most patients probably are not interested in the esoteric details, I typically make a point of explaining to patients that HS is a chronic inflammatory disease and provide reassurance that it is not a sign of poor hygiene. 

Counsel on Smoking Cessation

Most HS patients use tobacco. As many as 75% of HS patients are active smokers and another 10% to 15% are former smokers. Although there is mixed evidence that disease activity correlates with smoking status, the Hidradenitis Suppurativa Foundation in the United States and Canada concluded in the 2019 North American Clinical Management Guidelines for Hidradenitis Suppurativa that due to the overall health risks of smoking, we should recommend cessation to our patients.⁴

Laser Hair Removal Works

Don’t forget about laser hair removal! Evidence from randomized controlled trials supports the use of the Nd:YAG laser in the treatment of HS. Treat the entire affected anatomic area and use stacked double pulses on active nodules (typical settings: 10-mm spot size; 10-millisecond pulse duration and 35–50 J/cm² in Fitzpatrick skin types I–III; 20-millisecond pulse duration and 25–40 J/cm² in Fitzpatrick skin types IV–VI).⁴ Especially if it is covered by your patient’s insurance, Nd:YAG is a great adjunctive treatment to consider. The guidelines also recommend long-pulsed alexandrite and diode lasers as well as intense pulsed light, all of which result in follicular destruction, though these treatments have less supporting evidence.⁴

Have a Plan for Flares

Intralesional injection of triamcinolone is a mainstay of HS treatment and provides patients with rapid relief of symptoms during a flare.⁵ One case series found that there was a notable decrease in pain, size, and drainage after just 1 day of treatment with intralesional triamcinolone 10 mg/mL (0.2–2.0 mL).⁶

Intralesional steroid injection is a great tool for quieting an active disease flare while simultaneously instating ongoing treatment for preventive management. However, even when disease control is optimized, patients may still experience intermittent flares of disease. For some patients, it may be appropriate to have a plan in place for a return to clinic during the beginning of a flare to obtain intralesional steroids. The ability to come in on short notice may help avoid visits to the emergency department and urgent care where your patients may receive treatments such as short courses of antibiotics or incision and drainage that may deviate from your overall treatment plan.

Consider Childbearing Status

Don’t forget to consider childbearing plans and childbearing potential when treating female patients with HS. Pregnancy is a frequent consideration in HS patients, as HS affects 3 to 4 times more women than men and typically presents after puberty (second or third decades of life). Many of the medications in the HS armamentarium are contraindicated in pregnancy including tetracyclines, retinoids, and hormonal agents. Surgery should be avoided in pregnant patients whenever possible, particularly in the first trimester. Relatively safe options include topical antibiotics such as clindamycin and metronidazole, as well as tumor necrosis factor α inhibitors, which are classified as category B in pregnancy.⁵

Before making treatment decisions in pregnant and breastfeeding patients, consult the US Food and Drug Administration recommendations. Perng et al⁷ reviewed current management strategies for HS in pregnant and breastfeeding women, and their review article in the Journal of the American Academy of Dermatology is an excellent resource.

Final Thoughts

Comprehensive management of HS may include a combination of medication and procedures, lifestyle modification, management of comorbidities, and social support. Formulating a good treatment plan may be a challenge but can drastically improve your patient’s quality of life.

Hidradenitis suppurativa (HS) is a common and debilitating inflammatory disorder of the pilosebaceous unit that presents with recurrent scarring inflammatory nodules and sinus tracts in the intertriginous folds of the body. It is a complex condition that requires multimodal management to address the medical, surgical, and psychosocial needs of affected patients. However, it can be difficult to coordinate all that goes into HS management beyond the standard therapeutic ladder of topical and oral antimicrobials, intralesional corticosteroids, biologics, and surgery. In this article, I will outline 5 important aspects of HS treatment that often are overlooked.

Talk About Pathophysiology

Patients with HS often have limited understanding of their condition. One common misperception is that HS is an infectious disease and that disease activity is associated with poor hygiene.¹ Dispelling this myth may help patients avoid unnecessary hygiene practices, decrease perceived stigma, and enhance your therapeutic alliance.

The current model of HS pathophysiology implicates an aberrant inflammatory response to the cutaneous bacterial microbiome, which leads to follicular occlusion and then rupture of debris and bacteria into the surrounding dermis. Immune cells and inflammatory mediators such as nuclear factor κB and tumor necrosis factor α respond to the disruption. Chronic lesions develop due to tissue repair with scarring and re-epithelialization.^2,3 Although most patients probably are not interested in the esoteric details, I typically make a point of explaining to patients that HS is a chronic inflammatory disease and provide reassurance that it is not a sign of poor hygiene. 

Counsel on Smoking Cessation

Most HS patients use tobacco. As many as 75% of HS patients are active smokers and another 10% to 15% are former smokers. Although there is mixed evidence that disease activity correlates with smoking status, the Hidradenitis Suppurativa Foundation in the United States and Canada concluded in the 2019 North American Clinical Management Guidelines for Hidradenitis Suppurativa that due to the overall health risks of smoking, we should recommend cessation to our patients.⁴

Laser Hair Removal Works

Don’t forget about laser hair removal! Evidence from randomized controlled trials supports the use of the Nd:YAG laser in the treatment of HS. Treat the entire affected anatomic area and use stacked double pulses on active nodules (typical settings: 10-mm spot size; 10-millisecond pulse duration and 35–50 J/cm² in Fitzpatrick skin types I–III; 20-millisecond pulse duration and 25–40 J/cm² in Fitzpatrick skin types IV–VI).⁴ Especially if it is covered by your patient’s insurance, Nd:YAG is a great adjunctive treatment to consider. The guidelines also recommend long-pulsed alexandrite and diode lasers as well as intense pulsed light, all of which result in follicular destruction, though these treatments have less supporting evidence.⁴

Have a Plan for Flares

Intralesional injection of triamcinolone is a mainstay of HS treatment and provides patients with rapid relief of symptoms during a flare.⁵ One case series found that there was a notable decrease in pain, size, and drainage after just 1 day of treatment with intralesional triamcinolone 10 mg/mL (0.2–2.0 mL).⁶

Intralesional steroid injection is a great tool for quieting an active disease flare while simultaneously instating ongoing treatment for preventive management. However, even when disease control is optimized, patients may still experience intermittent flares of disease. For some patients, it may be appropriate to have a plan in place for a return to clinic during the beginning of a flare to obtain intralesional steroids. The ability to come in on short notice may help avoid visits to the emergency department and urgent care where your patients may receive treatments such as short courses of antibiotics or incision and drainage that may deviate from your overall treatment plan.

Consider Childbearing Status

Don’t forget to consider childbearing plans and childbearing potential when treating female patients with HS. Pregnancy is a frequent consideration in HS patients, as HS affects 3 to 4 times more women than men and typically presents after puberty (second or third decades of life). Many of the medications in the HS armamentarium are contraindicated in pregnancy including tetracyclines, retinoids, and hormonal agents. Surgery should be avoided in pregnant patients whenever possible, particularly in the first trimester. Relatively safe options include topical antibiotics such as clindamycin and metronidazole, as well as tumor necrosis factor α inhibitors, which are classified as category B in pregnancy.⁵

Before making treatment decisions in pregnant and breastfeeding patients, consult the US Food and Drug Administration recommendations. Perng et al⁷ reviewed current management strategies for HS in pregnant and breastfeeding women, and their review article in the Journal of the American Academy of Dermatology is an excellent resource.

Final Thoughts

Comprehensive management of HS may include a combination of medication and procedures, lifestyle modification, management of comorbidities, and social support. Formulating a good treatment plan may be a challenge but can drastically improve your patient’s quality of life.

In the United States, endoscopists who participate in advanced endoscopy fellowship programs use rectal NSAIDs more often than pancreatic duct stent placement to prevent post-ERCP pancreatitis (PEP), according to research published online in Gastrointestinal Endoscopy. In addition, methods for PEP prophylaxis in clinical practice are not implemented to the extent that the current evidence warrants.

“Future studies should not only further clarify the optimal PEP prophylaxis strategy, but should also focus on strategies to improve the implementation of evidence-based PEP prophylaxis techniques,” wrote Patrick Avila, MD, MPH, gastroenterologist at the University of California, San Francisco, and colleagues.
 

A survey of American endoscopists

Approximately 4% of patients who undergo biliopancreatic endoscopy develop PEP, which has a mortality rate of 3%. The American Society for Gastrointestinal Endoscopy (ASGE) recommends prophylactic pancreatic duct stent placement and rectal NSAIDs to reduce the incidence and severity of PEP in patients at high risk. The ASGE further suggests that rectal indomethacin may decrease the risk and severity of PEP in patients at average risk.

The European Society for Gastrointestinal Endoscopy (ESGE) recommends rectal indomethacin for all patients undergoing ERCP (endoscopic retrograde cholangiopancreatography). Several surveys of European endoscopists, however, indicate that relatively few respondents have adopted the recommended prophylactic techniques. The literature does not contain information about practice patterns among American endoscopists, and Dr. Avila and colleagues decided to investigate this question.

The researchers developed a 16-question online survey to assess current practice patterns with regard to PEP prophylaxis. They defined ERCPs that entailed a high risk for PEP as any that involved pancreatic or precut sphincterotomy, traumatic biliary sphincterotomy, balloon dilation of the biliary sphincter, injection of the pancreatic duct, extensive pancreatic duct instrumentation, difficult cannulation, suspected sphincter of Oddi dysfunction, previous PEP, or a female patient. Dr. Avila and colleagues distributed the survey to 233 advanced endoscopists involved in advanced endoscopy fellowship training programs.
 

Respondents had years of experience

Sixty-two endoscopists (26.7%) completed the survey. Respondents’ mean age was 47 years, and most respondents (74.6%) had been performing ERCP for more than 5 years. Almost all respondents (95%) worked at a tertiary referral center, and all worked with fellows.

All respondents reported having used pancreatic duct stent placement to prevent PEP. Most responders (72%) used pancreatic duct stent placement only in patients at high risk of PEP, and 64% reported using pancreatic duct stent placement in 25% or less of the ERCPs that they had performed. Four respondents (6.8%) used pancreatic duct stent placement for PEP in more than half of ERCPs. Among endoscopists who rarely use pancreatic duct stent placement for PEP, the major reasons cited included concern about increased risk of PEP with failed pancreatic duct insertion, the belief that stents do not provide additional benefit beyond pharmacologic prophylaxis, and difficulties in following up patients to ensure stent passage.

About 98% of respondents reported using rectal NSAIDs for PEP. Thirty-four respondents (59.7%) used this treatment only for patients at high risk, and 23 respondents (40.1%) used it for patients at average risk. Among respondents who used rectal NSAIDs to prevent PEP, 67.8% used the treatment in half or more of ERCPs. The NSAID of choice was indomethacin for all respondents. One respondent reported never using rectal NSAIDs for PEP because of doubts about its efficacy, in addition to cost and availability.

In addition, 49 respondents (83.0%) reported using rapid intravenous fluids to prevent PEP. None reported using octreotide or antibiotics to prevent PEP.

Results may reflect recall bias

The survey reveals “a significant divergence from the scientific evidence in how PEP techniques are used in routine clinical practice,” wrote Dr. Avila and colleagues. Several studies, including a randomized controlled trial, support the use of rectal NSAIDs as prophylaxis as patients at average risk of PEP, but less than half of respondents reported using it. The ASGE guidelines state that this treatment is “reasonable,” but do not advocate for it. “If appropriate, adopting a stronger stance in our practice guidelines may lead to further widespread use of rectal NSAIDs in this group of patients,” wrote Dr. Avila and colleagues.

Pancreatic duct stent placement is a difficult procedure to perform. The success rate in one British study was 51%, and a study of expert pancreaticobiliary endoscopists found a failure rate of 7%. It therefore may not be surprising that pancreatic duct stent placement was used less often than rectal NSAIDs among respondents, according to the authors.

Dr. Avila and colleagues acknowledged that the survey’s low response rate could have introduced nonresponse bias into the findings. They also stated that the study may have been affected by selection and recall biases. The results thus may not be generalizable to other practice settings, they concluded.

The authors did not report any study funding or disclosures.

* This story was updated on 12/2/2019.

[email protected]

SOURCE: Avila P et al. Gastrointest Endosc. 2019 Nov 16. doi: 10.1016/j.gie.2019.11.013.

In the United States, endoscopists who participate in advanced endoscopy fellowship programs use rectal NSAIDs more often than pancreatic duct stent placement to prevent post-ERCP pancreatitis (PEP), according to research published online in Gastrointestinal Endoscopy. In addition, methods for PEP prophylaxis in clinical practice are not implemented to the extent that the current evidence warrants.

“Future studies should not only further clarify the optimal PEP prophylaxis strategy, but should also focus on strategies to improve the implementation of evidence-based PEP prophylaxis techniques,” wrote Patrick Avila, MD, MPH, gastroenterologist at the University of California, San Francisco, and colleagues.
 

A survey of American endoscopists

Approximately 4% of patients who undergo biliopancreatic endoscopy develop PEP, which has a mortality rate of 3%. The American Society for Gastrointestinal Endoscopy (ASGE) recommends prophylactic pancreatic duct stent placement and rectal NSAIDs to reduce the incidence and severity of PEP in patients at high risk. The ASGE further suggests that rectal indomethacin may decrease the risk and severity of PEP in patients at average risk.

The European Society for Gastrointestinal Endoscopy (ESGE) recommends rectal indomethacin for all patients undergoing ERCP (endoscopic retrograde cholangiopancreatography). Several surveys of European endoscopists, however, indicate that relatively few respondents have adopted the recommended prophylactic techniques. The literature does not contain information about practice patterns among American endoscopists, and Dr. Avila and colleagues decided to investigate this question.

The researchers developed a 16-question online survey to assess current practice patterns with regard to PEP prophylaxis. They defined ERCPs that entailed a high risk for PEP as any that involved pancreatic or precut sphincterotomy, traumatic biliary sphincterotomy, balloon dilation of the biliary sphincter, injection of the pancreatic duct, extensive pancreatic duct instrumentation, difficult cannulation, suspected sphincter of Oddi dysfunction, previous PEP, or a female patient. Dr. Avila and colleagues distributed the survey to 233 advanced endoscopists involved in advanced endoscopy fellowship training programs.
 

Respondents had years of experience

Sixty-two endoscopists (26.7%) completed the survey. Respondents’ mean age was 47 years, and most respondents (74.6%) had been performing ERCP for more than 5 years. Almost all respondents (95%) worked at a tertiary referral center, and all worked with fellows.

All respondents reported having used pancreatic duct stent placement to prevent PEP. Most responders (72%) used pancreatic duct stent placement only in patients at high risk of PEP, and 64% reported using pancreatic duct stent placement in 25% or less of the ERCPs that they had performed. Four respondents (6.8%) used pancreatic duct stent placement for PEP in more than half of ERCPs. Among endoscopists who rarely use pancreatic duct stent placement for PEP, the major reasons cited included concern about increased risk of PEP with failed pancreatic duct insertion, the belief that stents do not provide additional benefit beyond pharmacologic prophylaxis, and difficulties in following up patients to ensure stent passage.

About 98% of respondents reported using rectal NSAIDs for PEP. Thirty-four respondents (59.7%) used this treatment only for patients at high risk, and 23 respondents (40.1%) used it for patients at average risk. Among respondents who used rectal NSAIDs to prevent PEP, 67.8% used the treatment in half or more of ERCPs. The NSAID of choice was indomethacin for all respondents. One respondent reported never using rectal NSAIDs for PEP because of doubts about its efficacy, in addition to cost and availability.

In addition, 49 respondents (83.0%) reported using rapid intravenous fluids to prevent PEP. None reported using octreotide or antibiotics to prevent PEP.

Results may reflect recall bias

The survey reveals “a significant divergence from the scientific evidence in how PEP techniques are used in routine clinical practice,” wrote Dr. Avila and colleagues. Several studies, including a randomized controlled trial, support the use of rectal NSAIDs as prophylaxis as patients at average risk of PEP, but less than half of respondents reported using it. The ASGE guidelines state that this treatment is “reasonable,” but do not advocate for it. “If appropriate, adopting a stronger stance in our practice guidelines may lead to further widespread use of rectal NSAIDs in this group of patients,” wrote Dr. Avila and colleagues.

Pancreatic duct stent placement is a difficult procedure to perform. The success rate in one British study was 51%, and a study of expert pancreaticobiliary endoscopists found a failure rate of 7%. It therefore may not be surprising that pancreatic duct stent placement was used less often than rectal NSAIDs among respondents, according to the authors.

Dr. Avila and colleagues acknowledged that the survey’s low response rate could have introduced nonresponse bias into the findings. They also stated that the study may have been affected by selection and recall biases. The results thus may not be generalizable to other practice settings, they concluded.

The authors did not report any study funding or disclosures.

* This story was updated on 12/2/2019.

[email protected]

SOURCE: Avila P et al. Gastrointest Endosc. 2019 Nov 16. doi: 10.1016/j.gie.2019.11.013.

In the United States, endoscopists who participate in advanced endoscopy fellowship programs use rectal NSAIDs more often than pancreatic duct stent placement to prevent post-ERCP pancreatitis (PEP), according to research published online in Gastrointestinal Endoscopy. In addition, methods for PEP prophylaxis in clinical practice are not implemented to the extent that the current evidence warrants.

“Future studies should not only further clarify the optimal PEP prophylaxis strategy, but should also focus on strategies to improve the implementation of evidence-based PEP prophylaxis techniques,” wrote Patrick Avila, MD, MPH, gastroenterologist at the University of California, San Francisco, and colleagues.
 

A survey of American endoscopists

Approximately 4% of patients who undergo biliopancreatic endoscopy develop PEP, which has a mortality rate of 3%. The American Society for Gastrointestinal Endoscopy (ASGE) recommends prophylactic pancreatic duct stent placement and rectal NSAIDs to reduce the incidence and severity of PEP in patients at high risk. The ASGE further suggests that rectal indomethacin may decrease the risk and severity of PEP in patients at average risk.

The European Society for Gastrointestinal Endoscopy (ESGE) recommends rectal indomethacin for all patients undergoing ERCP (endoscopic retrograde cholangiopancreatography). Several surveys of European endoscopists, however, indicate that relatively few respondents have adopted the recommended prophylactic techniques. The literature does not contain information about practice patterns among American endoscopists, and Dr. Avila and colleagues decided to investigate this question.

The researchers developed a 16-question online survey to assess current practice patterns with regard to PEP prophylaxis. They defined ERCPs that entailed a high risk for PEP as any that involved pancreatic or precut sphincterotomy, traumatic biliary sphincterotomy, balloon dilation of the biliary sphincter, injection of the pancreatic duct, extensive pancreatic duct instrumentation, difficult cannulation, suspected sphincter of Oddi dysfunction, previous PEP, or a female patient. Dr. Avila and colleagues distributed the survey to 233 advanced endoscopists involved in advanced endoscopy fellowship training programs.
 

Respondents had years of experience

Sixty-two endoscopists (26.7%) completed the survey. Respondents’ mean age was 47 years, and most respondents (74.6%) had been performing ERCP for more than 5 years. Almost all respondents (95%) worked at a tertiary referral center, and all worked with fellows.

All respondents reported having used pancreatic duct stent placement to prevent PEP. Most responders (72%) used pancreatic duct stent placement only in patients at high risk of PEP, and 64% reported using pancreatic duct stent placement in 25% or less of the ERCPs that they had performed. Four respondents (6.8%) used pancreatic duct stent placement for PEP in more than half of ERCPs. Among endoscopists who rarely use pancreatic duct stent placement for PEP, the major reasons cited included concern about increased risk of PEP with failed pancreatic duct insertion, the belief that stents do not provide additional benefit beyond pharmacologic prophylaxis, and difficulties in following up patients to ensure stent passage.

About 98% of respondents reported using rectal NSAIDs for PEP. Thirty-four respondents (59.7%) used this treatment only for patients at high risk, and 23 respondents (40.1%) used it for patients at average risk. Among respondents who used rectal NSAIDs to prevent PEP, 67.8% used the treatment in half or more of ERCPs. The NSAID of choice was indomethacin for all respondents. One respondent reported never using rectal NSAIDs for PEP because of doubts about its efficacy, in addition to cost and availability.

In addition, 49 respondents (83.0%) reported using rapid intravenous fluids to prevent PEP. None reported using octreotide or antibiotics to prevent PEP.

Results may reflect recall bias

The survey reveals “a significant divergence from the scientific evidence in how PEP techniques are used in routine clinical practice,” wrote Dr. Avila and colleagues. Several studies, including a randomized controlled trial, support the use of rectal NSAIDs as prophylaxis as patients at average risk of PEP, but less than half of respondents reported using it. The ASGE guidelines state that this treatment is “reasonable,” but do not advocate for it. “If appropriate, adopting a stronger stance in our practice guidelines may lead to further widespread use of rectal NSAIDs in this group of patients,” wrote Dr. Avila and colleagues.

Pancreatic duct stent placement is a difficult procedure to perform. The success rate in one British study was 51%, and a study of expert pancreaticobiliary endoscopists found a failure rate of 7%. It therefore may not be surprising that pancreatic duct stent placement was used less often than rectal NSAIDs among respondents, according to the authors.

Dr. Avila and colleagues acknowledged that the survey’s low response rate could have introduced nonresponse bias into the findings. They also stated that the study may have been affected by selection and recall biases. The results thus may not be generalizable to other practice settings, they concluded.

The authors did not report any study funding or disclosures.

* This story was updated on 12/2/2019.

[email protected]

SOURCE: Avila P et al. Gastrointest Endosc. 2019 Nov 16. doi: 10.1016/j.gie.2019.11.013.

Electronic cigarettes and other “vaping” devices have been increasing in popularity among youth and adults since their introduction in the US market in 2007.¹ This increase is partially driven by a public perception that vaping is harmless, or at least less harmful than cigarette smoking.² Vaping fans also argue that current smokers can use vaping as nicotine replacement therapy to help them quit smoking.³

We disagree. Research on the health effects of vaping, though still limited, is accumulating rapidly and making it increasingly clear that this habit is far from harmless. For youth, it is a gateway to addiction to nicotine and other substances. Whether it can help people quit smoking remains to be seen. And recent months have seen reports of serious respiratory illnesses and even deaths linked to vaping.⁴

In December 2016, the US Surgeon General warned that e-cigarette use among youth and young adults in the United States represents a “major public health concern,”⁵ and that more adolescents and young adults are now vaping than smoking conventional tobacco products.

This article reviews the issue of vaping in the United States, as well as available evidence regarding its safety.

YOUTH AT RISK

Retail sales of e-cigarettes and vaping devices approach an annual $7 billion.⁶ A 2014–2015 survey found that 2.4% of the general US population were current users of e-cigarettes, and 8.5% had tried them at least once.³

In 2014, for the first time, e-cigarette use became more common among US youth than traditional cigarettes.⁵

The odds of taking up vaping are higher among minority youth in the United States, particularly Hispanics.⁹ This trend is particularly worrisome because several longitudinal studies have shown that adolescents who use e-cigarettes are 3 times as likely to eventually become smokers of traditional cigarettes compared with adolescents who do not use e-cigarettes.^10–12

If US youth continue smoking at the current rate, 5.6 million of the current population under age 18, or 1 of every 13, will die early of a smoking-related illness.¹³ 

RECENT OUTBREAK OF VAPING-ASSOCIATED LUNG INJURY

As of November 5, 2019, there had been 2,051 cases of vaping-associated lung injury in 49 states (all except Alaska), the District of Columbia, and 1 US territory reported to the US Centers for Disease Control and Prevention (CDC), with 39 confirmed deaths.⁴ The reported cases include respiratory injury including acute eosinophilic pneumonia, organizing pneumonia, acute respiratory distress syndrome, and hypersensitivity pneumonitis.¹⁴

Most of these patients had been vaping tetrahydrocannabinol (THC), though many used both nicotine- and THC-containing products, and others used products containing nicotine exclusively.⁴ Thus, it is difficult to identify the exact substance or substances that may be contributing to this sudden outbreak among vape users, and many different product sources are currently under investigation.

One substance that may be linked to the epidemic is vitamin E acetate, which the New York State Department of Health has detected in high levels in cannabis vaping cartridges used by patients who developed lung injury.¹⁵ The US Food and Drug Administration (FDA) is continuing to analyze vape cartridge samples submitted by affected patients to look for other chemicals that can contribute to the development of serious pulmonary illness.

Source: The US Food and Drug Administration

Vape pens consist of similar elements but are not necessarily similar in appearance to a conventional cigarette, and may look more like a pen or a USB flash drive. In fact, the Juul device is recharged by plugging it into a USB port.

Vaping devices have many street names, including e-cigs, e-hookahs, vape pens, mods, vapes, and tank systems.

The first US patent application for a device resembling a modern e-cigarette was filed in 1963, but the product never made it to the market.¹⁶ Instead, the first commercially successful e-cigarette was created in Beijing in 2003 and introduced to US markets in 2007.

Newer-generation devices have larger batteries and can heat the liquid to higher temperatures, releasing more nicotine and forming additional toxicants such as formaldehyde. Devices lack standardization in terms of design, capacity for safely holding e-liquid, packaging of the e-liquid, and features designed to minimize hazards of use.

Not just nicotine

Many devices are designed for use with other drugs, including THC.¹⁷ In a 2018 study, 10.9% of college students reported vaping marijuana in the past 30 days, up from 5.2% in 2017.¹⁸

Other substances are being vaped as well.¹⁹ In theory, any heat-stable psychoactive recreational drug could be aerosolized and vaped. There are increasing reports of e-liquids containing recreational drugs such as synthetic cannabinoid receptor agonists, crack cocaine, LSD, and methamphetamine.¹⁷

Freedom, rebellion, glamour

Sales have risen rapidly since 2007 with widespread advertising on television and in print publications for popular brands, often featuring celebrities.²⁰ Spending on advertising for e-cigarettes and vape devices rose from $6.4 million in 2011 to $115 million in 2014—and that was before the advent of Juul (see below).²¹

Marketing campaigns for vaping devices mimic the themes previously used successfully by the tobacco industry, eg, freedom, rebellion, and glamour. They also make unsubstantiated claims about health benefits and smoking cessation, though initial websites contained endorsements from physicians, similar to the strategies of tobacco companies in old cigarette ads. Cigarette ads have been prohibited since 1971—but not e-cigarette ads. Moreover, vaping products appear as product placements in television shows and movies, with advocacy groups on social media.²²

By law, buyers have to be 18 or 21

Vaping devices can be purchased at vape shops, convenience stores, gas stations, and over the Internet; up to 50% of sales are conducted online.²⁴

Fruit flavors are popular

Zhu et al²⁵ estimated that 7,700 unique vaping flavors exist, with fruit and candy flavors predominating. The most popular flavors are tobacco and mint, followed by fruit, dessert and candy flavors, alcoholic flavors (strawberry daiquiri, margarita), and food flavors.²⁵ These flavors have been associated with higher usage in youth, leading to increased risk of nicotine addiction.²⁶

WHAT IS JUUL?

The Juul device (Juul Labs, www.juul.com) was developed in 2015 by 2 Stanford University graduates. Their goal was to produce a more satisfying and cigarette-like vaping experience, specifically by increasing the amount of nicotine delivered while maintaining smooth and pleasant inhalation. They created an e-liquid that could be vaporized effectively at lower temperatures.²⁷

While more than 400 brands of vaping devices are currently available in the United States,³ Juul has held the largest market share since 2017,²⁸ an estimated 72.1% as of August 2018.²⁹ The surge in popularity of this particular brand is attributed to its trendy design that is similar in size and appearance to a USB flash drive,²⁹ and its offering of sweet flavors such as “crème brûlée” and “cool mint.”

On April 24, 2018, in view of growing concern about the popularity of Juul products among youth, the FDA requested that the company submit documents regarding its marketing tactics, as well as research on the effects of this marketing on product design and public health impact, and information about adverse experiences and complaints.³⁰ The company was forced to change its marketing to appeal less to youth. Now it offers only 3 flavors: “Virginia tobacco,” “classic tobacco,” and “menthol,” although off-brand pods containing a variety of flavors are still available. And some pods are refillable, so users can essentially vape any substance they want.

Although the Juul device delivers a strong dose of nicotine, it is small and therefore easy to hide from parents and teachers, and widespread use has been reported among youth in middle and high schools. Hoodies, hemp jewelry, and backpacks have been designed to hide the devices and allow for easy, hands-free use. YouTube searches for terms such as “Juul,” “hiding Juul at school,” and “Juul in class,” yield thousands of results.³¹ A 2017 survey reported that 8% of Americans age 15 to 24 had used Juul in the month prior to the survey.³² “To juul” has become a verb.

Each Juul starter kit contains the rechargeable inhalation device plus 4 flavored pods. In the United States, each Juul pod contains nearly as much nicotine as 1 pack of 20 cigarettes in a concentration of 3% or 5%. (Israel and Europe have forced the company to replace the 5% nicotine pods with 1.7% nicotine pods.³³) A starter kit costs $49.99, and additional packs of 4 flavored liquid cartridges or pods cost $15.99.³⁴ Other brands of vape pens cost between $15 and $35, and 10-mL bottles of e-liquid cost approximately $⁷.

What is ‘dripping’?

Hard-core vapers seeking a more intense experience are taking their vaping devices apart and modifying them for “dripping,” ie, directly dripping vape liquids onto the heated coils for inhalation. In a survey, 1 in 4 high school students using vape devices also used them for dripping, citing desires for a thicker cloud of vapor, more intense flavor, “a stronger throat hit,” curiosity, and other reasons.³⁵ Dripping involves higher temperatures, which leads to higher amounts of nicotine delivered, along with more formaldehyde, acetaldehyde, and acetone (see below).³⁶

Studies of vape liquids consistently confirm the presence of toxic substances in the resulting vape aerosol.^37–40 Depending on the combination of flavorings and solvents in a given e-liquid, a variety of chemicals can be detected in the aerosol from various vaping devices. Chemicals that may be detected include known irritants of respiratory mucosa, as well as various carcinogens. The list includes:

• Organic volatile compounds such as propylene glycol, glycerin, and toluene
• Aldehydes such as formaldehyde (released when propylene glycol is heated to high temperatures), acetaldehyde, and benzaldehyde
• Acetone and acrolein
• Carcinogenic nitrosamines
• Polycyclic aromatic hydrocarbons
• Particulate matter
• Metals including chromium, cadmium, nickel, and lead; and particles of copper, nickel, and silver have been found in electronic nicotine delivery system aerosol in higher levels than in conventional cigarette smoke.⁴¹

The specific chemicals detected can vary greatly between brands, even when the flavoring and nicotine content are equivalent, which frequently results in inconsistent and conflicting study findings. The chemicals detected also vary with the voltage or power used to generate the aerosol. Different flavors may carry varying levels of risk; for example, mint- and menthol-flavored e-cigarettes were shown to expose users to dangerous levels of pulegone, a carcinogenic compound banned as a food additive in 2018.⁴² The concentrations of some of these chemicals are sufficiently high to be of toxicologic concern; for example, one study reported the presence of benzaldehyde in e-cigarette aerosol at twice the workplace exposure limit.⁴³

Biologic effects

In an in vitro study,⁴⁴ 57% of e-liquids studied were found to be cytotoxic to human pulmonary fibroblasts, lung epithelial cells, and human embryonic stem cells. Fruit-flavored e-liquids in particular caused a significant increase in DNA fragmentation. Cell cultures treated with e-cigarette liquids showed increased oxidative stress, reduced cell proliferation, and increased DNA damage,⁴⁴ which may have implications for carcinogenic risk.

In another study,⁴⁵ exposure to e-cigarette aerosol as well as conventional cigarette smoke resulted in suppression of genes related to immune and inflammatory response in respiratory epithelial cells. All genes with decreased expression after exposure to conventional cigarette smoke also showed decreased expression with exposure to e-cigarette smoke, which the study authors suggested could lead to immune suppression at the level of the nasal mucosa. Diacetyl and acetoin, chemicals found in certain flavorings, have been linked to bronchiolitis obliterans, or “popcorn lung.”⁴⁶

Nicotine is not benign

The nicotine itself in many vaping liquids should also not be underestimated. Nicotine has harmful neurocognitive effects and addictive properties, particularly in the developing brains of adolescents and young adults.⁴⁷ Nicotine exposure during adolescence negatively affects memory, attention, and emotional regulation,⁴⁸ as well as executive functioning, reward processing, and learning.⁴⁹

The brain undergoes major structural remodeling in adolescence, and nicotine acetylcholine receptors regulate neural maturation. Early exposure to nicotine disrupts this process, leading to poor executive functioning, difficulty learning, decreased memory, and issues with reward processing.

Fetal exposure, if nicotine products are used during pregnancy, has also been linked to adverse consequences such as deficits in attention and cognition, behavioral effects, and sudden infant death syndrome.⁵

Much to learn about toxicity

Partly because vaping devices have been available to US consumers only since 2007, limited evidence is available regarding the long-term effects of exposure to the aerosol from these devices in humans.¹ Many of the studies mentioned above were in vitro studies or conducted in mouse models. Differences in device design and the composition of the e-liquid among device brands pose a challenge for developing well-designed studies of the long-term health effects of e-cigarette and vape use. Additionally, devices may have different health impacts when used to vape cannabis or other drugs besides nicotine, which requires further investigation.

E-CIGARETTES AND SMOKING CESSATION

Conventional cigarette smoking is a major public health threat, as tobacco use is responsible for 480,000 deaths annually in the United States.⁵⁰

And smoking is extremely difficult to quit: as many as 80% of smokers who attempt to quit resume smoking within the first month.⁵¹ The chance of successfully quitting improves by over 50% if the individual undergoes nicotine replacement therapy, and it improves even more with counseling.⁵⁰

There are currently 5 types of FDA-approved nicotine replacement therapy products (gum, patch, lozenge, inhaler, nasal spray) to help with smoking cessation. In addition, 2 non-nicotine prescription drugs (varenicline and bupropion) have been approved for treating tobacco dependence.

Can vaping devices be added to the list of nicotine replacement therapy products? Although some manufacturers try to brand their devices as smoking cessation aids, in one study,⁵² one-third of e-cigarette users said they had either never used conventional cigarettes or had formerly smoked them.

Bullen et al⁵³ randomized smokers interested in quitting to receive either e-cigarettes, nicotine patches, or placebo (nicotine-free) e-cigarettes and followed them for 6 months. Rates of tobacco cessation were less than predicted for the entire study population, resulting in insufficient power to determine the superiority of any single method, but the study authors concluded that nicotine e-cigarettes were “modestly effective” at helping smokers quit, and that abstinence rates may be similar to those with nicotine patches.⁵³

Hajek et al⁵⁴ randomized 886 smokers to e-cigarette or nicotine replacement products of their choice. After 1 year, 18% of e-cigarette users had stopped smoking, compared with  9.9% of nicotine replacement product users. However, 80% of the e-cigarette users were still using e-cigarettes after 1 year, while only 9% of nicotine replacement product users were still using nicotine replacement therapy products after 1 year.

While quitting conventional cigarette smoking altogether has widely established health benefits, little is known about the health benefits of transitioning from conventional cigarette smoking to reduced conventional cigarette smoking with concomitant use of e-cigarettes.

Campagna et al⁵⁵ found no beneficial health effects in smokers who partially substituted conventional cigarettes for e-cigarettes.

Many studies found that smokers use e-cigarettes to maintain their habit instead of quitting entirely.⁵⁶ It has been suggested that any slight increase in effectiveness in smoking cessation by using e-cigarettes compared with other nicotine replacement products could be linked to satisfying of the habitual smoking actions, such as inhaling and bringing the hand to the mouth,²⁴ which are absent when using other nicotine replacement methods such as a nicotine patch.

As with safety information, long-term outcomes regarding the use of vape devices for smoking cessation have not been yet established, as this option is still relatively new.

Another worrisome trend involving electronic nicotine delivery systems is their marketing and branding, which appear to be aimed directly at adolescents and young adults. Juul and other similar products cannot be sold to anyone under the age of 18 (or 21 in 18 states, including California, Massachusetts, New York, and now Ohio). Despite this, Juul and similar products continue to increase in popularity among middle school and high school students.⁵⁷

While smoking cessation and health improvement are cited as reasons for vaping among middle-aged and older adults, adolescents and young adults more often cite flavor, enjoyment, peer use, and curiosity as reasons for use.

Adolescents are more likely to report interest in trying a vape product flavored with menthol or fruit than tobacco, and commonly hold the belief that fruit-flavored e-cigarettes are less harmful than tobacco-flavored e-cigarettes.⁵⁸ Harrell et al⁵⁹ polled youth and young adults who used flavored e-cigarettes, and 78% said they would no longer use the product if their preferred flavor were not available. In September 2019, Michigan became the first state to ban the sale of flavored e-cigarettes in stores and online. Similar bills have been introduced in California, Massachusetts, and New York.⁶⁰

Myths and misperceptions abound among youth regarding smoking vs vaping. Young people view regular cigarette smoking negatively, as causing cancer, bad breath, and asthma exacerbations. Meanwhile, they believe marijuana is safer and less addictive than traditional cigarette smoking.⁶¹ Youth exposed to e-cigarette advertisements viewed e-cigarettes as healthier, more enjoyable, “cool,” safe, and fun.⁶¹ The overall public health impact of increasing initiation of smoking, particularly among youth and young adults, should not be underestimated.

SECONDHAND VAPE AND OTHER EXPOSURE RISKS

Cigarette smoking has been banned in many public places, in view of a large body of scientific evidence about the harmful effects of secondhand smoke. Advocates for allowing vaping in public places say that vaping emissions do not harm bystanders, but evidence is insufficient to support this claim.⁶² One study showed that passive exposure to e-cigarette aerosol generated increases in serum levels of cotinine (a nicotine metabolite) similar to those with passive exposure to conventional cigarette smoke.⁵

Accidental nicotine poisoning in children as a result of ingesting e-cigarette liquid is also a major concern,⁶³ particularly with sweet flavors such as bubblegum or cheesecake that may be attractive to children.

Calls to US poison control centers with respect to e-cigarettes and vaping increased from 1 per month in September 2010 to 215 in February 2014, with 51% involving children under age 5.⁶⁴ This trend resulted in the Child Nicotine Poisoning Prevention Act, which passed in 2015 and went into effect in 2016, requiring packaging that is difficult to open for children under age 5.⁵

Device malfunctions or battery failures have led to explosions that have resulted in substantial injuries to users, as well as house and car fires.⁴⁹

HOW DO WE DISCOURAGE ADOLESCENT USE?

There are currently no established treatment approaches for adolescents who have become addicted to vaping. A review of the literature regarding treatment modalities used to address adolescent use of tobacco and marijuana provides insight that options such as nicotine replacement therapy and counseling modalities such as cognitive behavioral therapy may be helpful in treating teen vaping addiction. However, more research is needed to determine the effectiveness of these treatments in youth addicted to vaping.

Given that youth who vape even once are more likely to try other types of tobacco, we recommend that parents and healthcare providers start conversations by asking what the young person has seen or heard about vaping. Young people can also be asked what they think the school’s response should be: Do they think vaping should be banned in public places, as cigarettes have been banned? What about the carbon footprint? What are their thoughts on the plastic waste, batteries, and other toxins generated by the e-cigarette industry?

New US laws ban the sale of e-cigarettes and vaping devices to minors in stores and online. These policies are modeled in many cases on environmental control policies that have been previously employed to reduce tobacco use, particularly by youth. For example, changing laws to mandate sales only to individuals age 21 and older in all states can help to decrease access to these products among middle school and high school students.

As with tobacco cessation, education will not be enough. Support of legislation that bans vaping in public places, increases pricing to discourage adolescent use, and other measures used successfully to decrease conventional cigarette smoking can be deployed to decrease the public health impact of e-cigarettes. We recommend further regulation of specific harmful chemicals and clear, detailed ingredient labeling to increase consumer understanding of the risks associated with these products. Additionally, we recommend eliminating flavored e-cigarettes, which are the most appealing type for young users, and raising prices of e-cigarettes and similar products to discourage use by youth.

If current cigarette smokers want to use e-cigarettes to quit, we recommend that clinicians counsel them to eventually completely stop use of traditional cigarettes and switch to using e-cigarettes, instead of becoming a dual user of both types of products or using e-cigarettes indefinitely. After making that switch, they should then work to gradually taper usage and nicotine addiction by reducing the amount of nicotine in the e-liquid. Clinicians should ask patients about use of e-cigarettes and vaping devices specifically, and should counsel nonsmokers to avoid initiation of use.

EVIDENCE OF HARM CONTINUES TO EMERGE

Data about respiratory effects, secondhand exposure, and long-term smoking cessation efficacy are still limited, and it remains as yet unknown what combinations of solvents, flavorings, and nicotine in a given e-liquid will result in the most harmful or least harmful effects. In addition, while much of the information about the safety of these components has been obtained using in vitro or mouse models, increasing reports of serious respiratory illness and rising numbers of deaths linked to vaping make it clear that these findings likely translate to harmful effects in humans.

E-cigarettes may ultimately prove to be less harmful than traditional cigarettes, but it seems likely that with further time and research, serious health risks of e-cigarette use will continue to emerge.

Electronic cigarettes and other “vaping” devices have been increasing in popularity among youth and adults since their introduction in the US market in 2007.¹ This increase is partially driven by a public perception that vaping is harmless, or at least less harmful than cigarette smoking.² Vaping fans also argue that current smokers can use vaping as nicotine replacement therapy to help them quit smoking.³

We disagree. Research on the health effects of vaping, though still limited, is accumulating rapidly and making it increasingly clear that this habit is far from harmless. For youth, it is a gateway to addiction to nicotine and other substances. Whether it can help people quit smoking remains to be seen. And recent months have seen reports of serious respiratory illnesses and even deaths linked to vaping.⁴

In December 2016, the US Surgeon General warned that e-cigarette use among youth and young adults in the United States represents a “major public health concern,”⁵ and that more adolescents and young adults are now vaping than smoking conventional tobacco products.

This article reviews the issue of vaping in the United States, as well as available evidence regarding its safety.

YOUTH AT RISK

Retail sales of e-cigarettes and vaping devices approach an annual $7 billion.⁶ A 2014–2015 survey found that 2.4% of the general US population were current users of e-cigarettes, and 8.5% had tried them at least once.³

In 2014, for the first time, e-cigarette use became more common among US youth than traditional cigarettes.⁵

The odds of taking up vaping are higher among minority youth in the United States, particularly Hispanics.⁹ This trend is particularly worrisome because several longitudinal studies have shown that adolescents who use e-cigarettes are 3 times as likely to eventually become smokers of traditional cigarettes compared with adolescents who do not use e-cigarettes.^10–12

If US youth continue smoking at the current rate, 5.6 million of the current population under age 18, or 1 of every 13, will die early of a smoking-related illness.¹³ 

RECENT OUTBREAK OF VAPING-ASSOCIATED LUNG INJURY

As of November 5, 2019, there had been 2,051 cases of vaping-associated lung injury in 49 states (all except Alaska), the District of Columbia, and 1 US territory reported to the US Centers for Disease Control and Prevention (CDC), with 39 confirmed deaths.⁴ The reported cases include respiratory injury including acute eosinophilic pneumonia, organizing pneumonia, acute respiratory distress syndrome, and hypersensitivity pneumonitis.¹⁴

Most of these patients had been vaping tetrahydrocannabinol (THC), though many used both nicotine- and THC-containing products, and others used products containing nicotine exclusively.⁴ Thus, it is difficult to identify the exact substance or substances that may be contributing to this sudden outbreak among vape users, and many different product sources are currently under investigation.

One substance that may be linked to the epidemic is vitamin E acetate, which the New York State Department of Health has detected in high levels in cannabis vaping cartridges used by patients who developed lung injury.¹⁵ The US Food and Drug Administration (FDA) is continuing to analyze vape cartridge samples submitted by affected patients to look for other chemicals that can contribute to the development of serious pulmonary illness.

Source: The US Food and Drug Administration

Vape pens consist of similar elements but are not necessarily similar in appearance to a conventional cigarette, and may look more like a pen or a USB flash drive. In fact, the Juul device is recharged by plugging it into a USB port.

Vaping devices have many street names, including e-cigs, e-hookahs, vape pens, mods, vapes, and tank systems.

The first US patent application for a device resembling a modern e-cigarette was filed in 1963, but the product never made it to the market.¹⁶ Instead, the first commercially successful e-cigarette was created in Beijing in 2003 and introduced to US markets in 2007.

Newer-generation devices have larger batteries and can heat the liquid to higher temperatures, releasing more nicotine and forming additional toxicants such as formaldehyde. Devices lack standardization in terms of design, capacity for safely holding e-liquid, packaging of the e-liquid, and features designed to minimize hazards of use.

Not just nicotine

Many devices are designed for use with other drugs, including THC.¹⁷ In a 2018 study, 10.9% of college students reported vaping marijuana in the past 30 days, up from 5.2% in 2017.¹⁸

Other substances are being vaped as well.¹⁹ In theory, any heat-stable psychoactive recreational drug could be aerosolized and vaped. There are increasing reports of e-liquids containing recreational drugs such as synthetic cannabinoid receptor agonists, crack cocaine, LSD, and methamphetamine.¹⁷

Freedom, rebellion, glamour

Sales have risen rapidly since 2007 with widespread advertising on television and in print publications for popular brands, often featuring celebrities.²⁰ Spending on advertising for e-cigarettes and vape devices rose from $6.4 million in 2011 to $115 million in 2014—and that was before the advent of Juul (see below).²¹

Marketing campaigns for vaping devices mimic the themes previously used successfully by the tobacco industry, eg, freedom, rebellion, and glamour. They also make unsubstantiated claims about health benefits and smoking cessation, though initial websites contained endorsements from physicians, similar to the strategies of tobacco companies in old cigarette ads. Cigarette ads have been prohibited since 1971—but not e-cigarette ads. Moreover, vaping products appear as product placements in television shows and movies, with advocacy groups on social media.²²

By law, buyers have to be 18 or 21

Vaping devices can be purchased at vape shops, convenience stores, gas stations, and over the Internet; up to 50% of sales are conducted online.²⁴

Fruit flavors are popular

Zhu et al²⁵ estimated that 7,700 unique vaping flavors exist, with fruit and candy flavors predominating. The most popular flavors are tobacco and mint, followed by fruit, dessert and candy flavors, alcoholic flavors (strawberry daiquiri, margarita), and food flavors.²⁵ These flavors have been associated with higher usage in youth, leading to increased risk of nicotine addiction.²⁶

WHAT IS JUUL?

The Juul device (Juul Labs, www.juul.com) was developed in 2015 by 2 Stanford University graduates. Their goal was to produce a more satisfying and cigarette-like vaping experience, specifically by increasing the amount of nicotine delivered while maintaining smooth and pleasant inhalation. They created an e-liquid that could be vaporized effectively at lower temperatures.²⁷

While more than 400 brands of vaping devices are currently available in the United States,³ Juul has held the largest market share since 2017,²⁸ an estimated 72.1% as of August 2018.²⁹ The surge in popularity of this particular brand is attributed to its trendy design that is similar in size and appearance to a USB flash drive,²⁹ and its offering of sweet flavors such as “crème brûlée” and “cool mint.”

On April 24, 2018, in view of growing concern about the popularity of Juul products among youth, the FDA requested that the company submit documents regarding its marketing tactics, as well as research on the effects of this marketing on product design and public health impact, and information about adverse experiences and complaints.³⁰ The company was forced to change its marketing to appeal less to youth. Now it offers only 3 flavors: “Virginia tobacco,” “classic tobacco,” and “menthol,” although off-brand pods containing a variety of flavors are still available. And some pods are refillable, so users can essentially vape any substance they want.

Although the Juul device delivers a strong dose of nicotine, it is small and therefore easy to hide from parents and teachers, and widespread use has been reported among youth in middle and high schools. Hoodies, hemp jewelry, and backpacks have been designed to hide the devices and allow for easy, hands-free use. YouTube searches for terms such as “Juul,” “hiding Juul at school,” and “Juul in class,” yield thousands of results.³¹ A 2017 survey reported that 8% of Americans age 15 to 24 had used Juul in the month prior to the survey.³² “To juul” has become a verb.

Each Juul starter kit contains the rechargeable inhalation device plus 4 flavored pods. In the United States, each Juul pod contains nearly as much nicotine as 1 pack of 20 cigarettes in a concentration of 3% or 5%. (Israel and Europe have forced the company to replace the 5% nicotine pods with 1.7% nicotine pods.³³) A starter kit costs $49.99, and additional packs of 4 flavored liquid cartridges or pods cost $15.99.³⁴ Other brands of vape pens cost between $15 and $35, and 10-mL bottles of e-liquid cost approximately $⁷.

What is ‘dripping’?

Hard-core vapers seeking a more intense experience are taking their vaping devices apart and modifying them for “dripping,” ie, directly dripping vape liquids onto the heated coils for inhalation. In a survey, 1 in 4 high school students using vape devices also used them for dripping, citing desires for a thicker cloud of vapor, more intense flavor, “a stronger throat hit,” curiosity, and other reasons.³⁵ Dripping involves higher temperatures, which leads to higher amounts of nicotine delivered, along with more formaldehyde, acetaldehyde, and acetone (see below).³⁶

Studies of vape liquids consistently confirm the presence of toxic substances in the resulting vape aerosol.^37–40 Depending on the combination of flavorings and solvents in a given e-liquid, a variety of chemicals can be detected in the aerosol from various vaping devices. Chemicals that may be detected include known irritants of respiratory mucosa, as well as various carcinogens. The list includes:

• Organic volatile compounds such as propylene glycol, glycerin, and toluene
• Aldehydes such as formaldehyde (released when propylene glycol is heated to high temperatures), acetaldehyde, and benzaldehyde
• Acetone and acrolein
• Carcinogenic nitrosamines
• Polycyclic aromatic hydrocarbons
• Particulate matter
• Metals including chromium, cadmium, nickel, and lead; and particles of copper, nickel, and silver have been found in electronic nicotine delivery system aerosol in higher levels than in conventional cigarette smoke.⁴¹

The specific chemicals detected can vary greatly between brands, even when the flavoring and nicotine content are equivalent, which frequently results in inconsistent and conflicting study findings. The chemicals detected also vary with the voltage or power used to generate the aerosol. Different flavors may carry varying levels of risk; for example, mint- and menthol-flavored e-cigarettes were shown to expose users to dangerous levels of pulegone, a carcinogenic compound banned as a food additive in 2018.⁴² The concentrations of some of these chemicals are sufficiently high to be of toxicologic concern; for example, one study reported the presence of benzaldehyde in e-cigarette aerosol at twice the workplace exposure limit.⁴³

Biologic effects

In an in vitro study,⁴⁴ 57% of e-liquids studied were found to be cytotoxic to human pulmonary fibroblasts, lung epithelial cells, and human embryonic stem cells. Fruit-flavored e-liquids in particular caused a significant increase in DNA fragmentation. Cell cultures treated with e-cigarette liquids showed increased oxidative stress, reduced cell proliferation, and increased DNA damage,⁴⁴ which may have implications for carcinogenic risk.

In another study,⁴⁵ exposure to e-cigarette aerosol as well as conventional cigarette smoke resulted in suppression of genes related to immune and inflammatory response in respiratory epithelial cells. All genes with decreased expression after exposure to conventional cigarette smoke also showed decreased expression with exposure to e-cigarette smoke, which the study authors suggested could lead to immune suppression at the level of the nasal mucosa. Diacetyl and acetoin, chemicals found in certain flavorings, have been linked to bronchiolitis obliterans, or “popcorn lung.”⁴⁶

Nicotine is not benign

The nicotine itself in many vaping liquids should also not be underestimated. Nicotine has harmful neurocognitive effects and addictive properties, particularly in the developing brains of adolescents and young adults.⁴⁷ Nicotine exposure during adolescence negatively affects memory, attention, and emotional regulation,⁴⁸ as well as executive functioning, reward processing, and learning.⁴⁹

The brain undergoes major structural remodeling in adolescence, and nicotine acetylcholine receptors regulate neural maturation. Early exposure to nicotine disrupts this process, leading to poor executive functioning, difficulty learning, decreased memory, and issues with reward processing.

Fetal exposure, if nicotine products are used during pregnancy, has also been linked to adverse consequences such as deficits in attention and cognition, behavioral effects, and sudden infant death syndrome.⁵

Much to learn about toxicity

Partly because vaping devices have been available to US consumers only since 2007, limited evidence is available regarding the long-term effects of exposure to the aerosol from these devices in humans.¹ Many of the studies mentioned above were in vitro studies or conducted in mouse models. Differences in device design and the composition of the e-liquid among device brands pose a challenge for developing well-designed studies of the long-term health effects of e-cigarette and vape use. Additionally, devices may have different health impacts when used to vape cannabis or other drugs besides nicotine, which requires further investigation.

E-CIGARETTES AND SMOKING CESSATION

Conventional cigarette smoking is a major public health threat, as tobacco use is responsible for 480,000 deaths annually in the United States.⁵⁰

And smoking is extremely difficult to quit: as many as 80% of smokers who attempt to quit resume smoking within the first month.⁵¹ The chance of successfully quitting improves by over 50% if the individual undergoes nicotine replacement therapy, and it improves even more with counseling.⁵⁰

There are currently 5 types of FDA-approved nicotine replacement therapy products (gum, patch, lozenge, inhaler, nasal spray) to help with smoking cessation. In addition, 2 non-nicotine prescription drugs (varenicline and bupropion) have been approved for treating tobacco dependence.

Can vaping devices be added to the list of nicotine replacement therapy products? Although some manufacturers try to brand their devices as smoking cessation aids, in one study,⁵² one-third of e-cigarette users said they had either never used conventional cigarettes or had formerly smoked them.

Bullen et al⁵³ randomized smokers interested in quitting to receive either e-cigarettes, nicotine patches, or placebo (nicotine-free) e-cigarettes and followed them for 6 months. Rates of tobacco cessation were less than predicted for the entire study population, resulting in insufficient power to determine the superiority of any single method, but the study authors concluded that nicotine e-cigarettes were “modestly effective” at helping smokers quit, and that abstinence rates may be similar to those with nicotine patches.⁵³

Hajek et al⁵⁴ randomized 886 smokers to e-cigarette or nicotine replacement products of their choice. After 1 year, 18% of e-cigarette users had stopped smoking, compared with  9.9% of nicotine replacement product users. However, 80% of the e-cigarette users were still using e-cigarettes after 1 year, while only 9% of nicotine replacement product users were still using nicotine replacement therapy products after 1 year.

While quitting conventional cigarette smoking altogether has widely established health benefits, little is known about the health benefits of transitioning from conventional cigarette smoking to reduced conventional cigarette smoking with concomitant use of e-cigarettes.

Campagna et al⁵⁵ found no beneficial health effects in smokers who partially substituted conventional cigarettes for e-cigarettes.

Many studies found that smokers use e-cigarettes to maintain their habit instead of quitting entirely.⁵⁶ It has been suggested that any slight increase in effectiveness in smoking cessation by using e-cigarettes compared with other nicotine replacement products could be linked to satisfying of the habitual smoking actions, such as inhaling and bringing the hand to the mouth,²⁴ which are absent when using other nicotine replacement methods such as a nicotine patch.

As with safety information, long-term outcomes regarding the use of vape devices for smoking cessation have not been yet established, as this option is still relatively new.

Another worrisome trend involving electronic nicotine delivery systems is their marketing and branding, which appear to be aimed directly at adolescents and young adults. Juul and other similar products cannot be sold to anyone under the age of 18 (or 21 in 18 states, including California, Massachusetts, New York, and now Ohio). Despite this, Juul and similar products continue to increase in popularity among middle school and high school students.⁵⁷

While smoking cessation and health improvement are cited as reasons for vaping among middle-aged and older adults, adolescents and young adults more often cite flavor, enjoyment, peer use, and curiosity as reasons for use.

Adolescents are more likely to report interest in trying a vape product flavored with menthol or fruit than tobacco, and commonly hold the belief that fruit-flavored e-cigarettes are less harmful than tobacco-flavored e-cigarettes.⁵⁸ Harrell et al⁵⁹ polled youth and young adults who used flavored e-cigarettes, and 78% said they would no longer use the product if their preferred flavor were not available. In September 2019, Michigan became the first state to ban the sale of flavored e-cigarettes in stores and online. Similar bills have been introduced in California, Massachusetts, and New York.⁶⁰

Myths and misperceptions abound among youth regarding smoking vs vaping. Young people view regular cigarette smoking negatively, as causing cancer, bad breath, and asthma exacerbations. Meanwhile, they believe marijuana is safer and less addictive than traditional cigarette smoking.⁶¹ Youth exposed to e-cigarette advertisements viewed e-cigarettes as healthier, more enjoyable, “cool,” safe, and fun.⁶¹ The overall public health impact of increasing initiation of smoking, particularly among youth and young adults, should not be underestimated.

SECONDHAND VAPE AND OTHER EXPOSURE RISKS

Cigarette smoking has been banned in many public places, in view of a large body of scientific evidence about the harmful effects of secondhand smoke. Advocates for allowing vaping in public places say that vaping emissions do not harm bystanders, but evidence is insufficient to support this claim.⁶² One study showed that passive exposure to e-cigarette aerosol generated increases in serum levels of cotinine (a nicotine metabolite) similar to those with passive exposure to conventional cigarette smoke.⁵

Accidental nicotine poisoning in children as a result of ingesting e-cigarette liquid is also a major concern,⁶³ particularly with sweet flavors such as bubblegum or cheesecake that may be attractive to children.

Calls to US poison control centers with respect to e-cigarettes and vaping increased from 1 per month in September 2010 to 215 in February 2014, with 51% involving children under age 5.⁶⁴ This trend resulted in the Child Nicotine Poisoning Prevention Act, which passed in 2015 and went into effect in 2016, requiring packaging that is difficult to open for children under age 5.⁵

Device malfunctions or battery failures have led to explosions that have resulted in substantial injuries to users, as well as house and car fires.⁴⁹

HOW DO WE DISCOURAGE ADOLESCENT USE?

There are currently no established treatment approaches for adolescents who have become addicted to vaping. A review of the literature regarding treatment modalities used to address adolescent use of tobacco and marijuana provides insight that options such as nicotine replacement therapy and counseling modalities such as cognitive behavioral therapy may be helpful in treating teen vaping addiction. However, more research is needed to determine the effectiveness of these treatments in youth addicted to vaping.

Given that youth who vape even once are more likely to try other types of tobacco, we recommend that parents and healthcare providers start conversations by asking what the young person has seen or heard about vaping. Young people can also be asked what they think the school’s response should be: Do they think vaping should be banned in public places, as cigarettes have been banned? What about the carbon footprint? What are their thoughts on the plastic waste, batteries, and other toxins generated by the e-cigarette industry?

New US laws ban the sale of e-cigarettes and vaping devices to minors in stores and online. These policies are modeled in many cases on environmental control policies that have been previously employed to reduce tobacco use, particularly by youth. For example, changing laws to mandate sales only to individuals age 21 and older in all states can help to decrease access to these products among middle school and high school students.

As with tobacco cessation, education will not be enough. Support of legislation that bans vaping in public places, increases pricing to discourage adolescent use, and other measures used successfully to decrease conventional cigarette smoking can be deployed to decrease the public health impact of e-cigarettes. We recommend further regulation of specific harmful chemicals and clear, detailed ingredient labeling to increase consumer understanding of the risks associated with these products. Additionally, we recommend eliminating flavored e-cigarettes, which are the most appealing type for young users, and raising prices of e-cigarettes and similar products to discourage use by youth.

If current cigarette smokers want to use e-cigarettes to quit, we recommend that clinicians counsel them to eventually completely stop use of traditional cigarettes and switch to using e-cigarettes, instead of becoming a dual user of both types of products or using e-cigarettes indefinitely. After making that switch, they should then work to gradually taper usage and nicotine addiction by reducing the amount of nicotine in the e-liquid. Clinicians should ask patients about use of e-cigarettes and vaping devices specifically, and should counsel nonsmokers to avoid initiation of use.

EVIDENCE OF HARM CONTINUES TO EMERGE

Data about respiratory effects, secondhand exposure, and long-term smoking cessation efficacy are still limited, and it remains as yet unknown what combinations of solvents, flavorings, and nicotine in a given e-liquid will result in the most harmful or least harmful effects. In addition, while much of the information about the safety of these components has been obtained using in vitro or mouse models, increasing reports of serious respiratory illness and rising numbers of deaths linked to vaping make it clear that these findings likely translate to harmful effects in humans.

E-cigarettes may ultimately prove to be less harmful than traditional cigarettes, but it seems likely that with further time and research, serious health risks of e-cigarette use will continue to emerge.

Electronic cigarettes and other “vaping” devices have been increasing in popularity among youth and adults since their introduction in the US market in 2007.¹ This increase is partially driven by a public perception that vaping is harmless, or at least less harmful than cigarette smoking.² Vaping fans also argue that current smokers can use vaping as nicotine replacement therapy to help them quit smoking.³

We disagree. Research on the health effects of vaping, though still limited, is accumulating rapidly and making it increasingly clear that this habit is far from harmless. For youth, it is a gateway to addiction to nicotine and other substances. Whether it can help people quit smoking remains to be seen. And recent months have seen reports of serious respiratory illnesses and even deaths linked to vaping.⁴

In December 2016, the US Surgeon General warned that e-cigarette use among youth and young adults in the United States represents a “major public health concern,”⁵ and that more adolescents and young adults are now vaping than smoking conventional tobacco products.

This article reviews the issue of vaping in the United States, as well as available evidence regarding its safety.

YOUTH AT RISK

Retail sales of e-cigarettes and vaping devices approach an annual $7 billion.⁶ A 2014–2015 survey found that 2.4% of the general US population were current users of e-cigarettes, and 8.5% had tried them at least once.³

In 2014, for the first time, e-cigarette use became more common among US youth than traditional cigarettes.⁵

The odds of taking up vaping are higher among minority youth in the United States, particularly Hispanics.⁹ This trend is particularly worrisome because several longitudinal studies have shown that adolescents who use e-cigarettes are 3 times as likely to eventually become smokers of traditional cigarettes compared with adolescents who do not use e-cigarettes.^10–12

If US youth continue smoking at the current rate, 5.6 million of the current population under age 18, or 1 of every 13, will die early of a smoking-related illness.¹³ 

RECENT OUTBREAK OF VAPING-ASSOCIATED LUNG INJURY

As of November 5, 2019, there had been 2,051 cases of vaping-associated lung injury in 49 states (all except Alaska), the District of Columbia, and 1 US territory reported to the US Centers for Disease Control and Prevention (CDC), with 39 confirmed deaths.⁴ The reported cases include respiratory injury including acute eosinophilic pneumonia, organizing pneumonia, acute respiratory distress syndrome, and hypersensitivity pneumonitis.¹⁴

Most of these patients had been vaping tetrahydrocannabinol (THC), though many used both nicotine- and THC-containing products, and others used products containing nicotine exclusively.⁴ Thus, it is difficult to identify the exact substance or substances that may be contributing to this sudden outbreak among vape users, and many different product sources are currently under investigation.

One substance that may be linked to the epidemic is vitamin E acetate, which the New York State Department of Health has detected in high levels in cannabis vaping cartridges used by patients who developed lung injury.¹⁵ The US Food and Drug Administration (FDA) is continuing to analyze vape cartridge samples submitted by affected patients to look for other chemicals that can contribute to the development of serious pulmonary illness.

Source: The US Food and Drug Administration

Vape pens consist of similar elements but are not necessarily similar in appearance to a conventional cigarette, and may look more like a pen or a USB flash drive. In fact, the Juul device is recharged by plugging it into a USB port.

Vaping devices have many street names, including e-cigs, e-hookahs, vape pens, mods, vapes, and tank systems.

The first US patent application for a device resembling a modern e-cigarette was filed in 1963, but the product never made it to the market.¹⁶ Instead, the first commercially successful e-cigarette was created in Beijing in 2003 and introduced to US markets in 2007.

Newer-generation devices have larger batteries and can heat the liquid to higher temperatures, releasing more nicotine and forming additional toxicants such as formaldehyde. Devices lack standardization in terms of design, capacity for safely holding e-liquid, packaging of the e-liquid, and features designed to minimize hazards of use.

Not just nicotine

Many devices are designed for use with other drugs, including THC.¹⁷ In a 2018 study, 10.9% of college students reported vaping marijuana in the past 30 days, up from 5.2% in 2017.¹⁸

Other substances are being vaped as well.¹⁹ In theory, any heat-stable psychoactive recreational drug could be aerosolized and vaped. There are increasing reports of e-liquids containing recreational drugs such as synthetic cannabinoid receptor agonists, crack cocaine, LSD, and methamphetamine.¹⁷

Freedom, rebellion, glamour

Sales have risen rapidly since 2007 with widespread advertising on television and in print publications for popular brands, often featuring celebrities.²⁰ Spending on advertising for e-cigarettes and vape devices rose from $6.4 million in 2011 to $115 million in 2014—and that was before the advent of Juul (see below).²¹

Marketing campaigns for vaping devices mimic the themes previously used successfully by the tobacco industry, eg, freedom, rebellion, and glamour. They also make unsubstantiated claims about health benefits and smoking cessation, though initial websites contained endorsements from physicians, similar to the strategies of tobacco companies in old cigarette ads. Cigarette ads have been prohibited since 1971—but not e-cigarette ads. Moreover, vaping products appear as product placements in television shows and movies, with advocacy groups on social media.²²

By law, buyers have to be 18 or 21

Vaping devices can be purchased at vape shops, convenience stores, gas stations, and over the Internet; up to 50% of sales are conducted online.²⁴

Fruit flavors are popular

Zhu et al²⁵ estimated that 7,700 unique vaping flavors exist, with fruit and candy flavors predominating. The most popular flavors are tobacco and mint, followed by fruit, dessert and candy flavors, alcoholic flavors (strawberry daiquiri, margarita), and food flavors.²⁵ These flavors have been associated with higher usage in youth, leading to increased risk of nicotine addiction.²⁶

WHAT IS JUUL?

The Juul device (Juul Labs, www.juul.com) was developed in 2015 by 2 Stanford University graduates. Their goal was to produce a more satisfying and cigarette-like vaping experience, specifically by increasing the amount of nicotine delivered while maintaining smooth and pleasant inhalation. They created an e-liquid that could be vaporized effectively at lower temperatures.²⁷

While more than 400 brands of vaping devices are currently available in the United States,³ Juul has held the largest market share since 2017,²⁸ an estimated 72.1% as of August 2018.²⁹ The surge in popularity of this particular brand is attributed to its trendy design that is similar in size and appearance to a USB flash drive,²⁹ and its offering of sweet flavors such as “crème brûlée” and “cool mint.”

On April 24, 2018, in view of growing concern about the popularity of Juul products among youth, the FDA requested that the company submit documents regarding its marketing tactics, as well as research on the effects of this marketing on product design and public health impact, and information about adverse experiences and complaints.³⁰ The company was forced to change its marketing to appeal less to youth. Now it offers only 3 flavors: “Virginia tobacco,” “classic tobacco,” and “menthol,” although off-brand pods containing a variety of flavors are still available. And some pods are refillable, so users can essentially vape any substance they want.

Although the Juul device delivers a strong dose of nicotine, it is small and therefore easy to hide from parents and teachers, and widespread use has been reported among youth in middle and high schools. Hoodies, hemp jewelry, and backpacks have been designed to hide the devices and allow for easy, hands-free use. YouTube searches for terms such as “Juul,” “hiding Juul at school,” and “Juul in class,” yield thousands of results.³¹ A 2017 survey reported that 8% of Americans age 15 to 24 had used Juul in the month prior to the survey.³² “To juul” has become a verb.

Each Juul starter kit contains the rechargeable inhalation device plus 4 flavored pods. In the United States, each Juul pod contains nearly as much nicotine as 1 pack of 20 cigarettes in a concentration of 3% or 5%. (Israel and Europe have forced the company to replace the 5% nicotine pods with 1.7% nicotine pods.³³) A starter kit costs $49.99, and additional packs of 4 flavored liquid cartridges or pods cost $15.99.³⁴ Other brands of vape pens cost between $15 and $35, and 10-mL bottles of e-liquid cost approximately $⁷.

What is ‘dripping’?

Hard-core vapers seeking a more intense experience are taking their vaping devices apart and modifying them for “dripping,” ie, directly dripping vape liquids onto the heated coils for inhalation. In a survey, 1 in 4 high school students using vape devices also used them for dripping, citing desires for a thicker cloud of vapor, more intense flavor, “a stronger throat hit,” curiosity, and other reasons.³⁵ Dripping involves higher temperatures, which leads to higher amounts of nicotine delivered, along with more formaldehyde, acetaldehyde, and acetone (see below).³⁶

Studies of vape liquids consistently confirm the presence of toxic substances in the resulting vape aerosol.^37–40 Depending on the combination of flavorings and solvents in a given e-liquid, a variety of chemicals can be detected in the aerosol from various vaping devices. Chemicals that may be detected include known irritants of respiratory mucosa, as well as various carcinogens. The list includes:

• Organic volatile compounds such as propylene glycol, glycerin, and toluene
• Aldehydes such as formaldehyde (released when propylene glycol is heated to high temperatures), acetaldehyde, and benzaldehyde
• Acetone and acrolein
• Carcinogenic nitrosamines
• Polycyclic aromatic hydrocarbons
• Particulate matter
• Metals including chromium, cadmium, nickel, and lead; and particles of copper, nickel, and silver have been found in electronic nicotine delivery system aerosol in higher levels than in conventional cigarette smoke.⁴¹

The specific chemicals detected can vary greatly between brands, even when the flavoring and nicotine content are equivalent, which frequently results in inconsistent and conflicting study findings. The chemicals detected also vary with the voltage or power used to generate the aerosol. Different flavors may carry varying levels of risk; for example, mint- and menthol-flavored e-cigarettes were shown to expose users to dangerous levels of pulegone, a carcinogenic compound banned as a food additive in 2018.⁴² The concentrations of some of these chemicals are sufficiently high to be of toxicologic concern; for example, one study reported the presence of benzaldehyde in e-cigarette aerosol at twice the workplace exposure limit.⁴³

Biologic effects

In an in vitro study,⁴⁴ 57% of e-liquids studied were found to be cytotoxic to human pulmonary fibroblasts, lung epithelial cells, and human embryonic stem cells. Fruit-flavored e-liquids in particular caused a significant increase in DNA fragmentation. Cell cultures treated with e-cigarette liquids showed increased oxidative stress, reduced cell proliferation, and increased DNA damage,⁴⁴ which may have implications for carcinogenic risk.

In another study,⁴⁵ exposure to e-cigarette aerosol as well as conventional cigarette smoke resulted in suppression of genes related to immune and inflammatory response in respiratory epithelial cells. All genes with decreased expression after exposure to conventional cigarette smoke also showed decreased expression with exposure to e-cigarette smoke, which the study authors suggested could lead to immune suppression at the level of the nasal mucosa. Diacetyl and acetoin, chemicals found in certain flavorings, have been linked to bronchiolitis obliterans, or “popcorn lung.”⁴⁶

Nicotine is not benign

The nicotine itself in many vaping liquids should also not be underestimated. Nicotine has harmful neurocognitive effects and addictive properties, particularly in the developing brains of adolescents and young adults.⁴⁷ Nicotine exposure during adolescence negatively affects memory, attention, and emotional regulation,⁴⁸ as well as executive functioning, reward processing, and learning.⁴⁹

The brain undergoes major structural remodeling in adolescence, and nicotine acetylcholine receptors regulate neural maturation. Early exposure to nicotine disrupts this process, leading to poor executive functioning, difficulty learning, decreased memory, and issues with reward processing.

Fetal exposure, if nicotine products are used during pregnancy, has also been linked to adverse consequences such as deficits in attention and cognition, behavioral effects, and sudden infant death syndrome.⁵

Much to learn about toxicity

Partly because vaping devices have been available to US consumers only since 2007, limited evidence is available regarding the long-term effects of exposure to the aerosol from these devices in humans.¹ Many of the studies mentioned above were in vitro studies or conducted in mouse models. Differences in device design and the composition of the e-liquid among device brands pose a challenge for developing well-designed studies of the long-term health effects of e-cigarette and vape use. Additionally, devices may have different health impacts when used to vape cannabis or other drugs besides nicotine, which requires further investigation.

E-CIGARETTES AND SMOKING CESSATION

Conventional cigarette smoking is a major public health threat, as tobacco use is responsible for 480,000 deaths annually in the United States.⁵⁰

And smoking is extremely difficult to quit: as many as 80% of smokers who attempt to quit resume smoking within the first month.⁵¹ The chance of successfully quitting improves by over 50% if the individual undergoes nicotine replacement therapy, and it improves even more with counseling.⁵⁰

There are currently 5 types of FDA-approved nicotine replacement therapy products (gum, patch, lozenge, inhaler, nasal spray) to help with smoking cessation. In addition, 2 non-nicotine prescription drugs (varenicline and bupropion) have been approved for treating tobacco dependence.

Can vaping devices be added to the list of nicotine replacement therapy products? Although some manufacturers try to brand their devices as smoking cessation aids, in one study,⁵² one-third of e-cigarette users said they had either never used conventional cigarettes or had formerly smoked them.

Bullen et al⁵³ randomized smokers interested in quitting to receive either e-cigarettes, nicotine patches, or placebo (nicotine-free) e-cigarettes and followed them for 6 months. Rates of tobacco cessation were less than predicted for the entire study population, resulting in insufficient power to determine the superiority of any single method, but the study authors concluded that nicotine e-cigarettes were “modestly effective” at helping smokers quit, and that abstinence rates may be similar to those with nicotine patches.⁵³

Hajek et al⁵⁴ randomized 886 smokers to e-cigarette or nicotine replacement products of their choice. After 1 year, 18% of e-cigarette users had stopped smoking, compared with  9.9% of nicotine replacement product users. However, 80% of the e-cigarette users were still using e-cigarettes after 1 year, while only 9% of nicotine replacement product users were still using nicotine replacement therapy products after 1 year.

While quitting conventional cigarette smoking altogether has widely established health benefits, little is known about the health benefits of transitioning from conventional cigarette smoking to reduced conventional cigarette smoking with concomitant use of e-cigarettes.

Campagna et al⁵⁵ found no beneficial health effects in smokers who partially substituted conventional cigarettes for e-cigarettes.

Many studies found that smokers use e-cigarettes to maintain their habit instead of quitting entirely.⁵⁶ It has been suggested that any slight increase in effectiveness in smoking cessation by using e-cigarettes compared with other nicotine replacement products could be linked to satisfying of the habitual smoking actions, such as inhaling and bringing the hand to the mouth,²⁴ which are absent when using other nicotine replacement methods such as a nicotine patch.

As with safety information, long-term outcomes regarding the use of vape devices for smoking cessation have not been yet established, as this option is still relatively new.

Another worrisome trend involving electronic nicotine delivery systems is their marketing and branding, which appear to be aimed directly at adolescents and young adults. Juul and other similar products cannot be sold to anyone under the age of 18 (or 21 in 18 states, including California, Massachusetts, New York, and now Ohio). Despite this, Juul and similar products continue to increase in popularity among middle school and high school students.⁵⁷

While smoking cessation and health improvement are cited as reasons for vaping among middle-aged and older adults, adolescents and young adults more often cite flavor, enjoyment, peer use, and curiosity as reasons for use.

Adolescents are more likely to report interest in trying a vape product flavored with menthol or fruit than tobacco, and commonly hold the belief that fruit-flavored e-cigarettes are less harmful than tobacco-flavored e-cigarettes.⁵⁸ Harrell et al⁵⁹ polled youth and young adults who used flavored e-cigarettes, and 78% said they would no longer use the product if their preferred flavor were not available. In September 2019, Michigan became the first state to ban the sale of flavored e-cigarettes in stores and online. Similar bills have been introduced in California, Massachusetts, and New York.⁶⁰

Myths and misperceptions abound among youth regarding smoking vs vaping. Young people view regular cigarette smoking negatively, as causing cancer, bad breath, and asthma exacerbations. Meanwhile, they believe marijuana is safer and less addictive than traditional cigarette smoking.⁶¹ Youth exposed to e-cigarette advertisements viewed e-cigarettes as healthier, more enjoyable, “cool,” safe, and fun.⁶¹ The overall public health impact of increasing initiation of smoking, particularly among youth and young adults, should not be underestimated.

SECONDHAND VAPE AND OTHER EXPOSURE RISKS

Cigarette smoking has been banned in many public places, in view of a large body of scientific evidence about the harmful effects of secondhand smoke. Advocates for allowing vaping in public places say that vaping emissions do not harm bystanders, but evidence is insufficient to support this claim.⁶² One study showed that passive exposure to e-cigarette aerosol generated increases in serum levels of cotinine (a nicotine metabolite) similar to those with passive exposure to conventional cigarette smoke.⁵

Accidental nicotine poisoning in children as a result of ingesting e-cigarette liquid is also a major concern,⁶³ particularly with sweet flavors such as bubblegum or cheesecake that may be attractive to children.

Calls to US poison control centers with respect to e-cigarettes and vaping increased from 1 per month in September 2010 to 215 in February 2014, with 51% involving children under age 5.⁶⁴ This trend resulted in the Child Nicotine Poisoning Prevention Act, which passed in 2015 and went into effect in 2016, requiring packaging that is difficult to open for children under age 5.⁵

Device malfunctions or battery failures have led to explosions that have resulted in substantial injuries to users, as well as house and car fires.⁴⁹

HOW DO WE DISCOURAGE ADOLESCENT USE?

There are currently no established treatment approaches for adolescents who have become addicted to vaping. A review of the literature regarding treatment modalities used to address adolescent use of tobacco and marijuana provides insight that options such as nicotine replacement therapy and counseling modalities such as cognitive behavioral therapy may be helpful in treating teen vaping addiction. However, more research is needed to determine the effectiveness of these treatments in youth addicted to vaping.

Given that youth who vape even once are more likely to try other types of tobacco, we recommend that parents and healthcare providers start conversations by asking what the young person has seen or heard about vaping. Young people can also be asked what they think the school’s response should be: Do they think vaping should be banned in public places, as cigarettes have been banned? What about the carbon footprint? What are their thoughts on the plastic waste, batteries, and other toxins generated by the e-cigarette industry?

New US laws ban the sale of e-cigarettes and vaping devices to minors in stores and online. These policies are modeled in many cases on environmental control policies that have been previously employed to reduce tobacco use, particularly by youth. For example, changing laws to mandate sales only to individuals age 21 and older in all states can help to decrease access to these products among middle school and high school students.

As with tobacco cessation, education will not be enough. Support of legislation that bans vaping in public places, increases pricing to discourage adolescent use, and other measures used successfully to decrease conventional cigarette smoking can be deployed to decrease the public health impact of e-cigarettes. We recommend further regulation of specific harmful chemicals and clear, detailed ingredient labeling to increase consumer understanding of the risks associated with these products. Additionally, we recommend eliminating flavored e-cigarettes, which are the most appealing type for young users, and raising prices of e-cigarettes and similar products to discourage use by youth.

If current cigarette smokers want to use e-cigarettes to quit, we recommend that clinicians counsel them to eventually completely stop use of traditional cigarettes and switch to using e-cigarettes, instead of becoming a dual user of both types of products or using e-cigarettes indefinitely. After making that switch, they should then work to gradually taper usage and nicotine addiction by reducing the amount of nicotine in the e-liquid. Clinicians should ask patients about use of e-cigarettes and vaping devices specifically, and should counsel nonsmokers to avoid initiation of use.

EVIDENCE OF HARM CONTINUES TO EMERGE

Data about respiratory effects, secondhand exposure, and long-term smoking cessation efficacy are still limited, and it remains as yet unknown what combinations of solvents, flavorings, and nicotine in a given e-liquid will result in the most harmful or least harmful effects. In addition, while much of the information about the safety of these components has been obtained using in vitro or mouse models, increasing reports of serious respiratory illness and rising numbers of deaths linked to vaping make it clear that these findings likely translate to harmful effects in humans.

E-cigarettes may ultimately prove to be less harmful than traditional cigarettes, but it seems likely that with further time and research, serious health risks of e-cigarette use will continue to emerge.

The American Society of Hematology (ASH) plans to honor 10 researchers with awards and lectures at this year’s annual meeting, scheduled to take place Dec. 7-10 in Orlando.

Richard Aster, MD, will receive the 2019 Wallace H. Coulter Award for Lifetime Achievement in Hematology. Dr. Aster, of the Medical College of Wisconsin and Versiti Blood Center of Wisconsin in Milwaukee, “will be honored for his significant contributions to hematology through research, mentorship, and education throughout his 62-year career,” according to ASH.

Dr. Aster is known for his research on immune diseases that affect blood cells, particularly platelets. His work has led to improvements in platelet preparation, storage, and matching. In addition, he and his team developed the standard techniques for diagnosing immune thrombocytopenia and heparin-induced thrombocytopenia-thrombosis.

Other ASH awardees include William Eaton, MD, PhD, and Richard A. Larson, MD, who will receive the 2019 Henry M. Stratton Medal “for their seminal contributions to basic and clinical/translational hematology research, respectively.”

Dr. Eaton, of the National Institute of Diabetes and Digestive and Kidney Diseases conducts research on sickle cell disease, and his work contributed to the development of hydroxyurea. Dr. Larson, of the University of Chicago is the namesake of the Larson regimen (CALGB 8811) for acute lymphoblastic leukemia, and he played a key role in research that led to the U.S. approval of midostaurin.

Philip Greenberg, MD, will deliver the 2019 E. Donnall Thomas Lecture in recognition of “his outstanding contributions to the field of immunotherapy.” Dr. Greenberg, of theFred Hutchinson Cancer Research Center and the University of Washington in Seattle, is known for contributing to the development of T-cell adoptive immune therapy. His lecture will focus on the engineering of T cells to target acute myeloid leukemia and other malignancies.

Sriram Krishnaswamy, PhD, and Jeffrey I. Weitz, MD, will deliver the 2019 Ernest Beutler Lecture in recognition of “their significant research contributions to the understanding and treatment of blood clots.” The researchers will each deliver one part of the lecture at the meeting, and both will discuss research related to novel anticoagulants.

Dr. Krishnaswamy, of the University of Pennsylvania and Children’s Hospital of Philadelphia, is considered an authority on the function of surface-dependent coagulation complexes. Dr. Weitz, of McMaster University in Hamilton, Ont., has conducted research that led to the development of novel anticoagulants.

Emmanuelle Passegué, PhD, will receive the 2019 William Dameshek Prize “for her outstanding contributions to the understanding of hematopoietic stem cells.” Dr. Passegué, of Columbia University Irving Medical Center in New York, conducts research focused on changes to hematopoietic stem cells in the contexts of myeloid malignancies and physiological aging.

Griffin Rodgers, MD, will receive the ASH Award for Leadership in Promoting Diversity “for his extraordinary commitment to diversity and inclusion in hematology.” As director of the National Institute of Diabetes and Digestive and Kidney Diseases, Dr. Rodgers has worked to promote diversity in the scientific workforce and in clinical trials.

Leonard Zon, MD, and Michael R. DeBaun, MD, will receive the 2019 ASH Mentor Award “for their sustained, outstanding commitment to the training and career development of early career hematologists.”

Dr. DeBaun, of Vanderbilt University School of Medicine in Nashville, Tenn., has mentees ranging from high school students with sickle cell disease to tenured faculty members at medical schools. Dr. Zon, of Harvard University and Boston Children’s Hospital, organizes mentoring events for postdocs, graduate students, and technicians.

[email protected]

The American Society of Hematology (ASH) plans to honor 10 researchers with awards and lectures at this year’s annual meeting, scheduled to take place Dec. 7-10 in Orlando.

Richard Aster, MD, will receive the 2019 Wallace H. Coulter Award for Lifetime Achievement in Hematology. Dr. Aster, of the Medical College of Wisconsin and Versiti Blood Center of Wisconsin in Milwaukee, “will be honored for his significant contributions to hematology through research, mentorship, and education throughout his 62-year career,” according to ASH.

Dr. Aster is known for his research on immune diseases that affect blood cells, particularly platelets. His work has led to improvements in platelet preparation, storage, and matching. In addition, he and his team developed the standard techniques for diagnosing immune thrombocytopenia and heparin-induced thrombocytopenia-thrombosis.

Other ASH awardees include William Eaton, MD, PhD, and Richard A. Larson, MD, who will receive the 2019 Henry M. Stratton Medal “for their seminal contributions to basic and clinical/translational hematology research, respectively.”

Dr. Eaton, of the National Institute of Diabetes and Digestive and Kidney Diseases conducts research on sickle cell disease, and his work contributed to the development of hydroxyurea. Dr. Larson, of the University of Chicago is the namesake of the Larson regimen (CALGB 8811) for acute lymphoblastic leukemia, and he played a key role in research that led to the U.S. approval of midostaurin.

Philip Greenberg, MD, will deliver the 2019 E. Donnall Thomas Lecture in recognition of “his outstanding contributions to the field of immunotherapy.” Dr. Greenberg, of theFred Hutchinson Cancer Research Center and the University of Washington in Seattle, is known for contributing to the development of T-cell adoptive immune therapy. His lecture will focus on the engineering of T cells to target acute myeloid leukemia and other malignancies.

Sriram Krishnaswamy, PhD, and Jeffrey I. Weitz, MD, will deliver the 2019 Ernest Beutler Lecture in recognition of “their significant research contributions to the understanding and treatment of blood clots.” The researchers will each deliver one part of the lecture at the meeting, and both will discuss research related to novel anticoagulants.

Dr. Krishnaswamy, of the University of Pennsylvania and Children’s Hospital of Philadelphia, is considered an authority on the function of surface-dependent coagulation complexes. Dr. Weitz, of McMaster University in Hamilton, Ont., has conducted research that led to the development of novel anticoagulants.

Emmanuelle Passegué, PhD, will receive the 2019 William Dameshek Prize “for her outstanding contributions to the understanding of hematopoietic stem cells.” Dr. Passegué, of Columbia University Irving Medical Center in New York, conducts research focused on changes to hematopoietic stem cells in the contexts of myeloid malignancies and physiological aging.

Griffin Rodgers, MD, will receive the ASH Award for Leadership in Promoting Diversity “for his extraordinary commitment to diversity and inclusion in hematology.” As director of the National Institute of Diabetes and Digestive and Kidney Diseases, Dr. Rodgers has worked to promote diversity in the scientific workforce and in clinical trials.

Leonard Zon, MD, and Michael R. DeBaun, MD, will receive the 2019 ASH Mentor Award “for their sustained, outstanding commitment to the training and career development of early career hematologists.”

Dr. DeBaun, of Vanderbilt University School of Medicine in Nashville, Tenn., has mentees ranging from high school students with sickle cell disease to tenured faculty members at medical schools. Dr. Zon, of Harvard University and Boston Children’s Hospital, organizes mentoring events for postdocs, graduate students, and technicians.

[email protected]

The American Society of Hematology (ASH) plans to honor 10 researchers with awards and lectures at this year’s annual meeting, scheduled to take place Dec. 7-10 in Orlando.

Richard Aster, MD, will receive the 2019 Wallace H. Coulter Award for Lifetime Achievement in Hematology. Dr. Aster, of the Medical College of Wisconsin and Versiti Blood Center of Wisconsin in Milwaukee, “will be honored for his significant contributions to hematology through research, mentorship, and education throughout his 62-year career,” according to ASH.

Dr. Aster is known for his research on immune diseases that affect blood cells, particularly platelets. His work has led to improvements in platelet preparation, storage, and matching. In addition, he and his team developed the standard techniques for diagnosing immune thrombocytopenia and heparin-induced thrombocytopenia-thrombosis.

Other ASH awardees include William Eaton, MD, PhD, and Richard A. Larson, MD, who will receive the 2019 Henry M. Stratton Medal “for their seminal contributions to basic and clinical/translational hematology research, respectively.”

Dr. Eaton, of the National Institute of Diabetes and Digestive and Kidney Diseases conducts research on sickle cell disease, and his work contributed to the development of hydroxyurea. Dr. Larson, of the University of Chicago is the namesake of the Larson regimen (CALGB 8811) for acute lymphoblastic leukemia, and he played a key role in research that led to the U.S. approval of midostaurin.

Philip Greenberg, MD, will deliver the 2019 E. Donnall Thomas Lecture in recognition of “his outstanding contributions to the field of immunotherapy.” Dr. Greenberg, of theFred Hutchinson Cancer Research Center and the University of Washington in Seattle, is known for contributing to the development of T-cell adoptive immune therapy. His lecture will focus on the engineering of T cells to target acute myeloid leukemia and other malignancies.

Sriram Krishnaswamy, PhD, and Jeffrey I. Weitz, MD, will deliver the 2019 Ernest Beutler Lecture in recognition of “their significant research contributions to the understanding and treatment of blood clots.” The researchers will each deliver one part of the lecture at the meeting, and both will discuss research related to novel anticoagulants.

Dr. Krishnaswamy, of the University of Pennsylvania and Children’s Hospital of Philadelphia, is considered an authority on the function of surface-dependent coagulation complexes. Dr. Weitz, of McMaster University in Hamilton, Ont., has conducted research that led to the development of novel anticoagulants.

Emmanuelle Passegué, PhD, will receive the 2019 William Dameshek Prize “for her outstanding contributions to the understanding of hematopoietic stem cells.” Dr. Passegué, of Columbia University Irving Medical Center in New York, conducts research focused on changes to hematopoietic stem cells in the contexts of myeloid malignancies and physiological aging.

Griffin Rodgers, MD, will receive the ASH Award for Leadership in Promoting Diversity “for his extraordinary commitment to diversity and inclusion in hematology.” As director of the National Institute of Diabetes and Digestive and Kidney Diseases, Dr. Rodgers has worked to promote diversity in the scientific workforce and in clinical trials.

Leonard Zon, MD, and Michael R. DeBaun, MD, will receive the 2019 ASH Mentor Award “for their sustained, outstanding commitment to the training and career development of early career hematologists.”

Dr. DeBaun, of Vanderbilt University School of Medicine in Nashville, Tenn., has mentees ranging from high school students with sickle cell disease to tenured faculty members at medical schools. Dr. Zon, of Harvard University and Boston Children’s Hospital, organizes mentoring events for postdocs, graduate students, and technicians.

[email protected]

The plan was in motion before I got on the plane.

When your leukemia came back suddenly 3 years after your stem cell transplant, it was devastating. But we had a plan. Your cancer developed a new mutation we could target with a chemotherapy drug. If we got you into a second remission, we could consolidate it by infusing more of your donor’s stem cells.

We met in the hospital, but I was adamant to “keep” you as my patient when you got to the clinic. I made swaps to see you, to get the continuity I value so much but often lose as a fellow, rotating from clinic to hospital to clinic.

I was grateful to see how well you dealt with the chemotherapy. Typically it’s a tough regimen, but you hardly had side effects. Between visits, I would check your blood counts on my phone, watching your blast count fall and your normal blood cells rise. Watching the cancer disappear.

Your lumbar punctures were negative, negative, negative – my favorite word, I told you. I was involved in long email threads coordinating the timing of your stem cell infusion with the remission we were achieving.

We were on our way.

One day, your lumbar puncture came back with a few “atypical” cells. I called the pathologist, and upon further review they were convinced the cells were reactive, not cancer. The next lumbar puncture was normal, but it was hard to ignore.

“Are you worried?” I asked my attending in clinic.

“I’m always worried,” she said. Neither of us truly believed the leukemia was back, but with the odds against us, we pored over every detail, always on the alert for a clue to an outcome we feared.

By now, the stem cell infusion was all set up. The donor was ready; so was the medical team; so were you. It was exciting. I thought of how a different attending described his interest in leukemia: There’s a subset you get to cure. Yes, you were going to be one of them.

Your big day coincided with a vacation I had scheduled months before. I was sorry I would be missing the actual moment, but happy I would come back to good news.

I left my coat and badge at the hospital, packed my bags, and got on the plane. I refrained from immediately checking your blood counts on my phone as soon as we landed. That night, jet lagged, I let myself look before I go to sleep. Relief. Your numbers still looked good.

da-kuk/Getty Images

Every day, I explored. My Internet was spotty during my travels, and when I would I finally get service I would peek at your latest blood tests.

Day 1. Cooled lava canyons. Black sand beaches. Circulating blast count: 0%.

Day 2: Glacier tour. A national park. Geysers. Blast count: 2%.

Day 3: We drive along the shore to see a famous waterfall, where you can climb a set of winding stairs to the top.

I check my phone before we start the climb. No service.

And so we begin. The wind cuts as I count steps. 403, 404, 405 … and 406. We are there. The air is thin, the world quiet. My nose is running from the cold.

We hike a bit, and I glance down again. Still no signal. It’s probably for the best. The scenery is spectacular.

Two miles later, I get service. I open the blood work first. Circulating blast count: 5%. But the other counts are okay. It could still be reactive, I say to myself, though on a deeper level I think of my attending’s words: I’m always worried. The stem cell infusion is scheduled for tomorrow.

I hear the rush of the water hitting the rocks below. Icicles form to our left. Sheep graze on our right. I appreciate the feeling of my muscles aching as we climb, higher and higher, a reminder of where I am and my place in it.

At the very top, we pause to take photos. And I get a signal again. I open the bone marrow biopsy report and skim the pathologist’s words. My eyes glue on the summary: 80% blasts, compatible with relapsed leukemia.

I let out an audible gasp.

Do you know? How will they tell you? I am painfully aware of the distance between us, in so many ways.

I want to be present. And soon I will be back, and I will be visiting in the hospital, and we will be having hard conversations and thinking about hard decisions.

But I’m not there right now. Someone else is. Here, now, I realize what I cannot do. The best way I can be present for you later is to be present where I am now.

I stuff my phone in my backpack and zip it closed. I step carefully forward on the rocks, slippery from the rain. My nose is running again, but not from the cold.

“What do you think?” my partner asks.

“The views are incredible,” I say.

Minor details of this story were changed to protect privacy.

Dr. Yurkiewicz is a fellow in hematology and oncology at Stanford (Calif.) University. Follow her on Twitter @ilanayurkiewicz and listen to her each week on the Blood & Cancer podcast.

The plan was in motion before I got on the plane.

When your leukemia came back suddenly 3 years after your stem cell transplant, it was devastating. But we had a plan. Your cancer developed a new mutation we could target with a chemotherapy drug. If we got you into a second remission, we could consolidate it by infusing more of your donor’s stem cells.

We met in the hospital, but I was adamant to “keep” you as my patient when you got to the clinic. I made swaps to see you, to get the continuity I value so much but often lose as a fellow, rotating from clinic to hospital to clinic.

I was grateful to see how well you dealt with the chemotherapy. Typically it’s a tough regimen, but you hardly had side effects. Between visits, I would check your blood counts on my phone, watching your blast count fall and your normal blood cells rise. Watching the cancer disappear.

Your lumbar punctures were negative, negative, negative – my favorite word, I told you. I was involved in long email threads coordinating the timing of your stem cell infusion with the remission we were achieving.

We were on our way.

One day, your lumbar puncture came back with a few “atypical” cells. I called the pathologist, and upon further review they were convinced the cells were reactive, not cancer. The next lumbar puncture was normal, but it was hard to ignore.

“Are you worried?” I asked my attending in clinic.

“I’m always worried,” she said. Neither of us truly believed the leukemia was back, but with the odds against us, we pored over every detail, always on the alert for a clue to an outcome we feared.

By now, the stem cell infusion was all set up. The donor was ready; so was the medical team; so were you. It was exciting. I thought of how a different attending described his interest in leukemia: There’s a subset you get to cure. Yes, you were going to be one of them.

Your big day coincided with a vacation I had scheduled months before. I was sorry I would be missing the actual moment, but happy I would come back to good news.

I left my coat and badge at the hospital, packed my bags, and got on the plane. I refrained from immediately checking your blood counts on my phone as soon as we landed. That night, jet lagged, I let myself look before I go to sleep. Relief. Your numbers still looked good.

da-kuk/Getty Images

Every day, I explored. My Internet was spotty during my travels, and when I would I finally get service I would peek at your latest blood tests.

Day 1. Cooled lava canyons. Black sand beaches. Circulating blast count: 0%.

Day 2: Glacier tour. A national park. Geysers. Blast count: 2%.

Day 3: We drive along the shore to see a famous waterfall, where you can climb a set of winding stairs to the top.

I check my phone before we start the climb. No service.

And so we begin. The wind cuts as I count steps. 403, 404, 405 … and 406. We are there. The air is thin, the world quiet. My nose is running from the cold.

We hike a bit, and I glance down again. Still no signal. It’s probably for the best. The scenery is spectacular.

Two miles later, I get service. I open the blood work first. Circulating blast count: 5%. But the other counts are okay. It could still be reactive, I say to myself, though on a deeper level I think of my attending’s words: I’m always worried. The stem cell infusion is scheduled for tomorrow.

I hear the rush of the water hitting the rocks below. Icicles form to our left. Sheep graze on our right. I appreciate the feeling of my muscles aching as we climb, higher and higher, a reminder of where I am and my place in it.

At the very top, we pause to take photos. And I get a signal again. I open the bone marrow biopsy report and skim the pathologist’s words. My eyes glue on the summary: 80% blasts, compatible with relapsed leukemia.

I let out an audible gasp.

Do you know? How will they tell you? I am painfully aware of the distance between us, in so many ways.

I want to be present. And soon I will be back, and I will be visiting in the hospital, and we will be having hard conversations and thinking about hard decisions.

But I’m not there right now. Someone else is. Here, now, I realize what I cannot do. The best way I can be present for you later is to be present where I am now.

I stuff my phone in my backpack and zip it closed. I step carefully forward on the rocks, slippery from the rain. My nose is running again, but not from the cold.

“What do you think?” my partner asks.

“The views are incredible,” I say.

Minor details of this story were changed to protect privacy.

Dr. Yurkiewicz is a fellow in hematology and oncology at Stanford (Calif.) University. Follow her on Twitter @ilanayurkiewicz and listen to her each week on the Blood & Cancer podcast.

The plan was in motion before I got on the plane.

When your leukemia came back suddenly 3 years after your stem cell transplant, it was devastating. But we had a plan. Your cancer developed a new mutation we could target with a chemotherapy drug. If we got you into a second remission, we could consolidate it by infusing more of your donor’s stem cells.

We met in the hospital, but I was adamant to “keep” you as my patient when you got to the clinic. I made swaps to see you, to get the continuity I value so much but often lose as a fellow, rotating from clinic to hospital to clinic.

I was grateful to see how well you dealt with the chemotherapy. Typically it’s a tough regimen, but you hardly had side effects. Between visits, I would check your blood counts on my phone, watching your blast count fall and your normal blood cells rise. Watching the cancer disappear.

Your lumbar punctures were negative, negative, negative – my favorite word, I told you. I was involved in long email threads coordinating the timing of your stem cell infusion with the remission we were achieving.

We were on our way.

One day, your lumbar puncture came back with a few “atypical” cells. I called the pathologist, and upon further review they were convinced the cells were reactive, not cancer. The next lumbar puncture was normal, but it was hard to ignore.

“Are you worried?” I asked my attending in clinic.

“I’m always worried,” she said. Neither of us truly believed the leukemia was back, but with the odds against us, we pored over every detail, always on the alert for a clue to an outcome we feared.

By now, the stem cell infusion was all set up. The donor was ready; so was the medical team; so were you. It was exciting. I thought of how a different attending described his interest in leukemia: There’s a subset you get to cure. Yes, you were going to be one of them.

Your big day coincided with a vacation I had scheduled months before. I was sorry I would be missing the actual moment, but happy I would come back to good news.

I left my coat and badge at the hospital, packed my bags, and got on the plane. I refrained from immediately checking your blood counts on my phone as soon as we landed. That night, jet lagged, I let myself look before I go to sleep. Relief. Your numbers still looked good.

da-kuk/Getty Images

Every day, I explored. My Internet was spotty during my travels, and when I would I finally get service I would peek at your latest blood tests.

Day 1. Cooled lava canyons. Black sand beaches. Circulating blast count: 0%.

Day 2: Glacier tour. A national park. Geysers. Blast count: 2%.

Day 3: We drive along the shore to see a famous waterfall, where you can climb a set of winding stairs to the top.

I check my phone before we start the climb. No service.

And so we begin. The wind cuts as I count steps. 403, 404, 405 … and 406. We are there. The air is thin, the world quiet. My nose is running from the cold.

We hike a bit, and I glance down again. Still no signal. It’s probably for the best. The scenery is spectacular.

Two miles later, I get service. I open the blood work first. Circulating blast count: 5%. But the other counts are okay. It could still be reactive, I say to myself, though on a deeper level I think of my attending’s words: I’m always worried. The stem cell infusion is scheduled for tomorrow.

I hear the rush of the water hitting the rocks below. Icicles form to our left. Sheep graze on our right. I appreciate the feeling of my muscles aching as we climb, higher and higher, a reminder of where I am and my place in it.

At the very top, we pause to take photos. And I get a signal again. I open the bone marrow biopsy report and skim the pathologist’s words. My eyes glue on the summary: 80% blasts, compatible with relapsed leukemia.

I let out an audible gasp.

Do you know? How will they tell you? I am painfully aware of the distance between us, in so many ways.

I want to be present. And soon I will be back, and I will be visiting in the hospital, and we will be having hard conversations and thinking about hard decisions.

But I’m not there right now. Someone else is. Here, now, I realize what I cannot do. The best way I can be present for you later is to be present where I am now.

I stuff my phone in my backpack and zip it closed. I step carefully forward on the rocks, slippery from the rain. My nose is running again, but not from the cold.

“What do you think?” my partner asks.

“The views are incredible,” I say.

Minor details of this story were changed to protect privacy.

Dr. Yurkiewicz is a fellow in hematology and oncology at Stanford (Calif.) University. Follow her on Twitter @ilanayurkiewicz and listen to her each week on the Blood & Cancer podcast.

A 44-year-old woman presents with an 8-year history of intermittent heartburn, and in the past year she has been experiencing her symptoms daily. She says the heartburn is constant and is worse immediately after eating spicy or acidic foods. She says she has had no dysphagia, weight loss, or vomiting. Her symptoms have persisted despite taking a histamine (H)₂-receptor antagonist twice daily plus a proton pump inhibitor (PPI) before breakfast and dinner for more than 3 months.

She has undergone upper endoscopy 3 times in the past 8 years. Each time, the esophagus was normal with a regular Z-line and normal biopsy results from the proximal and distal esophagus.

The patient believes she has severe gastroesophageal reflux disease (GERD) and asks if she is a candidate for fundoplication surgery.

HEARTBURN IS A SYMPTOM; GERD IS A CONDITION

A distinction should be made between heartburn—the symptom of persistent retrosternal burning and discomfort—and gastroesophageal reflux disease—the condition in which reflux of stomach contents causes troublesome symptoms or complications.¹ While many clinicians initially diagnose patients who have heartburn as having GERD, there are many other potential causes of their symptoms.

For patients with persistent heartburn, an empiric trial of a once-daily PPI is usually effective, but one-third of patients continue to have heartburn.^2,3 The most common cause of this PPI-refractory heartburn is functional heartburn, a functional or hypersensitivity disorder of the esophagus.⁴

PATHOPHYSIOLOGY IS POORLY UNDERSTOOD

DIAGNOSTIC EVALUATION

Clinicians have several tests available for diagnosing these conditions.

Upper endoscopy

Upper endoscopy is recommended for patients with heartburn that does not respond to a 3-month trial of a PPI.⁹ Endoscopy is also indicated in any patient who has any of the following “alarm symptoms” that could be due to malignancy or peptic ulcer:

• Dysphagia
• Odynophagia
• Vomiting
• Unexplained weight loss or anemia
• Signs of gastrointestinal bleeding
• Anorexia
• New onset of dyspepsia in a patient over age 60.

During upper endoscopy, the esophagus is evaluated for reflux esophagitis, Barrett esophagus, and other inflammatory disorders such as infectious esophagitis. But even if the esophageal mucosa appears normal, the proximal and distal esophagus should be biopsied to rule out an inflammatory disorder such as eosinophilic or lymphocytic esophagitis.

If endoscopic and esophageal biopsy results are inconclusive, a workup for an esophageal motility disorder is the next step. Dysphagia is the most common symptom of these disorders, although the initial presenting symptom may be heartburn or regurgitation that persists despite PPI therapy.

Manometry is used to test for motility disorders such as achalasia and esophageal spasm.¹⁰ After applying a local anesthetic inside the nares, the clinician inserts a flexible catheter (about 4 mm in diameter) with 36 pressure sensors spaced at 1-cm intervals into the nares and passes it through the esophagus and lower esophageal sphincter. The patient then swallows liquid, and the sensors relay the esophageal response, creating a topographic plot that shows esophageal peristalsis and lower esophageal sphincter relaxation.

Achalasia is identified by incomplete lower esophageal sphincter relaxation combined with 100% failed peristalsis in the body of the esophagus. Esophageal spasms are identified by a shortened distal latency, which corresponds to premature contraction of the esophagus during peristalsis.¹¹

Esophageal pH testing

Measuring esophageal pH levels is an important step to quantify gastroesophageal reflux and determine if symptoms occur during reflux events. According to the updated Porto GERD consensus group recommendations,¹² a pH test is positive if the acid exposure time is greater than 6% of the testing period. Testing the pH differentiates between GERD (abnormal acid exposure), reflux hypersensitivity (normal acid exposure, strong correlation between symptoms and reflux events), and functional heartburn (normal acid exposure, negative correlation between reflux events and symptoms).⁵ For this test, a pH probe is placed in the esophagus transnasally or endoscopically. The probe records esophageal pH levels for 24 to 96 hours in an outpatient setting. Antisecretory therapy needs to be withheld for 7 to 10 days before the test.

Transnasal pH probe. For this approach, a thin catheter is inserted through the nares and advanced until the tip is 5 cm proximal to the lower esophageal sphincter. (The placement is guided by the results of esophageal manometry, which is done immediately before pH catheter placement.) The tube is secured with clear tape on the side of the patient’s face, and the end is connected to a portable recorder that compiles the data. The patient pushes a button on the recorder when experiencing heartburn symptoms. (A nurse instructs the patient on proper procedure.) After 24 hours, the patient either removes the catheter or has the clinic remove it. The pH and symptom data are downloaded and analyzed.

Transnasal pH testing can be combined with impedance measurement, which can detect nonacid reflux or weakly acid reflux. However, the clinical significance of this measurement is unclear, as multiple studies have found total acid exposure time to be a better predictor of response to therapy than weakly acid or nonacid reflux.¹²

Wireless pH probe. This method uses a disposable, catheter-free, capsule device to measure esophageal pH. The capsule, about the size of a gel capsule or pencil eraser, is attached to the patient’s esophageal lining, usually during upper endoscopy. The capsule records pH levels in the lower esophagus for 48 to 96 hours and transmits the data wirelessly to a receiver the patient wears. The patient pushes buttons on the receiver to record symptom-specific data when experiencing heartburn, chest pain, regurgitation, or cough. The capsule detaches from the esophagus spontaneously, generally within 7 days, and is passed out of the body through a bowel movement.

Diagnosing functional heartburn

CASE CONTINUED: NORMAL RESULTS ON TESTING

Based on these results, her condition is diagnosed as functional heartburn, consistent with the Rome IV criteria.⁵

Patient education is key

Patient education about the pathogenesis, natural history, and treatment options is the most important aspect of treating any functional gastrointestinal disorder. This includes the “brain-gut connection” and potential mechanisms of dysregulation. Patient education along with assessment of symptoms should be part of every visit, especially before discussing treatment options.

Patients whose condition is diagnosed as functional heartburn need reassurance that the condition is benign and, in particular, that the risk of progression to esophageal adenocarcinoma is minimal in the absence of Barrett esophagus.¹³ Also important to point out is that the disorder may spontaneously resolve: resolution rates of up to 40% have been reported for other functional gastrointestinal disorders.¹⁴

Antisecretory medications may work for some

A PPI or H₂-receptor antagonist is the most common first-line treatment for heartburn symptoms. Although most patients with functional heartburn experience no improvement in symptoms with an antisecretory agent, a small number report some relief, which suggests that acid-suppression therapy may have an indirect impact on pain modulation in the esophagus.¹⁵ In patients who report symptom relief with an antisecretory agent, we suggest continuing the medication tapered to the lowest effective dose, with repeated reassurance that the medication can be discontinued safely at any time.

Antireflux surgery should be avoided

Antireflux surgery should be avoided in patients with normal pH testing and no objective finding of reflux, as this is associated with worse subjective outcomes than in patients with abnormal pH test results.¹⁶

Neuromodulators

It is important to discuss with patients the concept of neuromodulation, including the fact that antidepressants are often used because of their effects on serotonin and norepinephrine, which decrease visceral hypersensitivity.

The selective serotonin reuptake inhibitor citalopram has been shown to reduce esophageal hypersensitivity,¹⁷ and a tricyclic antidepressant has been shown to improve quality of life.¹⁸ These results have led experts to recommend a trial of a low dose of either type of medication.¹⁹ The dose of tricyclic antidepressant often needs to be increased sequentially every 2 to 4 weeks.

Interestingly, melatonin 6 mg at bedtime has also shown efficacy for functional heartburn, potentially due to its antinociceptive properties.²⁰

Alternative and complementary therapies

Many esophageal centers use cognitive behavioral therapy and hypnotherapy as first-line treatment for functional esophageal disorders. Here again, it is important for the patient to understand the rationale of therapy for functional gastrointestinal disorders, given the stigma in the general population regarding psychotherapy.

Cognitive behavioral therapy has been used for functional gastrointestinal disorders for many years, as it has been shown to modulate visceral perception.²¹ Although published studies are limited, research regarding other functional esophageal disorders suggests that patients who commit to long-term behavioral therapy have had a significant improvement in symptoms.²²

The goal of esophageal-directed behavioral therapy is to promote focused relaxation using deep breathing techniques, which can help patients manage esophageal hypervigilance, especially if symptoms continue despite neuromodulator therapy. Specifically, hypnotherapy has been shown to modulate functional chest pain through the visceral sensory pathway and also to suppress gastric acid secretion.^21,23 A study of a 7-week hypnotherapy program reported significant benefits in heartburn relief and improved quality of life in patients with functional heartburn.²⁴ The data support the use of behavioral therapies as first-line therapy or as adjunctive therapy for patients already taking a neuromodulator.

CASE FOLLOW-UP: IMPROVEMENT WITH TREATMENT

During a follow-up visit, the patient is given several printed resources, including the Rome Foundation article on functional heartburn.⁵ We again emphasize the benign nature of functional heartburn, noting the minimal risk of progression to esophageal adenocarcinoma, as she had no evidence of Barrett esophagus on endoscopy. And we discuss the natural course of functional heartburn, including the spontaneous resolution rate of about 40%.

For treatment, we present her the rationale for using neuromodulators and reassure her that these medications are for treatment of visceral hypersensitivity, not for anxiety or depression. After the discussion, the patient opts to start amitriptyline therapy at 10 mg every night at bedtime, increasing the dose by 10 mg every 2 weeks until symptoms improve, up to 75–100 mg every day.

After 3 months, the patient reports a 90% improvement in symptoms while on amitriptyline 30 mg every night. She is also able to taper her antisecretory medications once symptoms are controlled. We plan to continue amitriptyline at the current dose for 6 to 12 months, then discuss a slow taper to see if her symptoms spontaneously resolve.

A 44-year-old woman presents with an 8-year history of intermittent heartburn, and in the past year she has been experiencing her symptoms daily. She says the heartburn is constant and is worse immediately after eating spicy or acidic foods. She says she has had no dysphagia, weight loss, or vomiting. Her symptoms have persisted despite taking a histamine (H)₂-receptor antagonist twice daily plus a proton pump inhibitor (PPI) before breakfast and dinner for more than 3 months.

She has undergone upper endoscopy 3 times in the past 8 years. Each time, the esophagus was normal with a regular Z-line and normal biopsy results from the proximal and distal esophagus.

The patient believes she has severe gastroesophageal reflux disease (GERD) and asks if she is a candidate for fundoplication surgery.

HEARTBURN IS A SYMPTOM; GERD IS A CONDITION

A distinction should be made between heartburn—the symptom of persistent retrosternal burning and discomfort—and gastroesophageal reflux disease—the condition in which reflux of stomach contents causes troublesome symptoms or complications.¹ While many clinicians initially diagnose patients who have heartburn as having GERD, there are many other potential causes of their symptoms.

For patients with persistent heartburn, an empiric trial of a once-daily PPI is usually effective, but one-third of patients continue to have heartburn.^2,3 The most common cause of this PPI-refractory heartburn is functional heartburn, a functional or hypersensitivity disorder of the esophagus.⁴

PATHOPHYSIOLOGY IS POORLY UNDERSTOOD

DIAGNOSTIC EVALUATION

Clinicians have several tests available for diagnosing these conditions.

Upper endoscopy

Upper endoscopy is recommended for patients with heartburn that does not respond to a 3-month trial of a PPI.⁹ Endoscopy is also indicated in any patient who has any of the following “alarm symptoms” that could be due to malignancy or peptic ulcer:

• Dysphagia
• Odynophagia
• Vomiting
• Unexplained weight loss or anemia
• Signs of gastrointestinal bleeding
• Anorexia
• New onset of dyspepsia in a patient over age 60.

During upper endoscopy, the esophagus is evaluated for reflux esophagitis, Barrett esophagus, and other inflammatory disorders such as infectious esophagitis. But even if the esophageal mucosa appears normal, the proximal and distal esophagus should be biopsied to rule out an inflammatory disorder such as eosinophilic or lymphocytic esophagitis.

If endoscopic and esophageal biopsy results are inconclusive, a workup for an esophageal motility disorder is the next step. Dysphagia is the most common symptom of these disorders, although the initial presenting symptom may be heartburn or regurgitation that persists despite PPI therapy.

Manometry is used to test for motility disorders such as achalasia and esophageal spasm.¹⁰ After applying a local anesthetic inside the nares, the clinician inserts a flexible catheter (about 4 mm in diameter) with 36 pressure sensors spaced at 1-cm intervals into the nares and passes it through the esophagus and lower esophageal sphincter. The patient then swallows liquid, and the sensors relay the esophageal response, creating a topographic plot that shows esophageal peristalsis and lower esophageal sphincter relaxation.

Achalasia is identified by incomplete lower esophageal sphincter relaxation combined with 100% failed peristalsis in the body of the esophagus. Esophageal spasms are identified by a shortened distal latency, which corresponds to premature contraction of the esophagus during peristalsis.¹¹

Esophageal pH testing

Measuring esophageal pH levels is an important step to quantify gastroesophageal reflux and determine if symptoms occur during reflux events. According to the updated Porto GERD consensus group recommendations,¹² a pH test is positive if the acid exposure time is greater than 6% of the testing period. Testing the pH differentiates between GERD (abnormal acid exposure), reflux hypersensitivity (normal acid exposure, strong correlation between symptoms and reflux events), and functional heartburn (normal acid exposure, negative correlation between reflux events and symptoms).⁵ For this test, a pH probe is placed in the esophagus transnasally or endoscopically. The probe records esophageal pH levels for 24 to 96 hours in an outpatient setting. Antisecretory therapy needs to be withheld for 7 to 10 days before the test.

Transnasal pH probe. For this approach, a thin catheter is inserted through the nares and advanced until the tip is 5 cm proximal to the lower esophageal sphincter. (The placement is guided by the results of esophageal manometry, which is done immediately before pH catheter placement.) The tube is secured with clear tape on the side of the patient’s face, and the end is connected to a portable recorder that compiles the data. The patient pushes a button on the recorder when experiencing heartburn symptoms. (A nurse instructs the patient on proper procedure.) After 24 hours, the patient either removes the catheter or has the clinic remove it. The pH and symptom data are downloaded and analyzed.

Transnasal pH testing can be combined with impedance measurement, which can detect nonacid reflux or weakly acid reflux. However, the clinical significance of this measurement is unclear, as multiple studies have found total acid exposure time to be a better predictor of response to therapy than weakly acid or nonacid reflux.¹²

Wireless pH probe. This method uses a disposable, catheter-free, capsule device to measure esophageal pH. The capsule, about the size of a gel capsule or pencil eraser, is attached to the patient’s esophageal lining, usually during upper endoscopy. The capsule records pH levels in the lower esophagus for 48 to 96 hours and transmits the data wirelessly to a receiver the patient wears. The patient pushes buttons on the receiver to record symptom-specific data when experiencing heartburn, chest pain, regurgitation, or cough. The capsule detaches from the esophagus spontaneously, generally within 7 days, and is passed out of the body through a bowel movement.

Diagnosing functional heartburn

CASE CONTINUED: NORMAL RESULTS ON TESTING

Based on these results, her condition is diagnosed as functional heartburn, consistent with the Rome IV criteria.⁵

Patient education is key

Patient education about the pathogenesis, natural history, and treatment options is the most important aspect of treating any functional gastrointestinal disorder. This includes the “brain-gut connection” and potential mechanisms of dysregulation. Patient education along with assessment of symptoms should be part of every visit, especially before discussing treatment options.

Patients whose condition is diagnosed as functional heartburn need reassurance that the condition is benign and, in particular, that the risk of progression to esophageal adenocarcinoma is minimal in the absence of Barrett esophagus.¹³ Also important to point out is that the disorder may spontaneously resolve: resolution rates of up to 40% have been reported for other functional gastrointestinal disorders.¹⁴

Antisecretory medications may work for some

A PPI or H₂-receptor antagonist is the most common first-line treatment for heartburn symptoms. Although most patients with functional heartburn experience no improvement in symptoms with an antisecretory agent, a small number report some relief, which suggests that acid-suppression therapy may have an indirect impact on pain modulation in the esophagus.¹⁵ In patients who report symptom relief with an antisecretory agent, we suggest continuing the medication tapered to the lowest effective dose, with repeated reassurance that the medication can be discontinued safely at any time.

Antireflux surgery should be avoided

Antireflux surgery should be avoided in patients with normal pH testing and no objective finding of reflux, as this is associated with worse subjective outcomes than in patients with abnormal pH test results.¹⁶

Neuromodulators

It is important to discuss with patients the concept of neuromodulation, including the fact that antidepressants are often used because of their effects on serotonin and norepinephrine, which decrease visceral hypersensitivity.

The selective serotonin reuptake inhibitor citalopram has been shown to reduce esophageal hypersensitivity,¹⁷ and a tricyclic antidepressant has been shown to improve quality of life.¹⁸ These results have led experts to recommend a trial of a low dose of either type of medication.¹⁹ The dose of tricyclic antidepressant often needs to be increased sequentially every 2 to 4 weeks.

Interestingly, melatonin 6 mg at bedtime has also shown efficacy for functional heartburn, potentially due to its antinociceptive properties.²⁰

Alternative and complementary therapies

Many esophageal centers use cognitive behavioral therapy and hypnotherapy as first-line treatment for functional esophageal disorders. Here again, it is important for the patient to understand the rationale of therapy for functional gastrointestinal disorders, given the stigma in the general population regarding psychotherapy.

Cognitive behavioral therapy has been used for functional gastrointestinal disorders for many years, as it has been shown to modulate visceral perception.²¹ Although published studies are limited, research regarding other functional esophageal disorders suggests that patients who commit to long-term behavioral therapy have had a significant improvement in symptoms.²²

The goal of esophageal-directed behavioral therapy is to promote focused relaxation using deep breathing techniques, which can help patients manage esophageal hypervigilance, especially if symptoms continue despite neuromodulator therapy. Specifically, hypnotherapy has been shown to modulate functional chest pain through the visceral sensory pathway and also to suppress gastric acid secretion.^21,23 A study of a 7-week hypnotherapy program reported significant benefits in heartburn relief and improved quality of life in patients with functional heartburn.²⁴ The data support the use of behavioral therapies as first-line therapy or as adjunctive therapy for patients already taking a neuromodulator.

CASE FOLLOW-UP: IMPROVEMENT WITH TREATMENT

During a follow-up visit, the patient is given several printed resources, including the Rome Foundation article on functional heartburn.⁵ We again emphasize the benign nature of functional heartburn, noting the minimal risk of progression to esophageal adenocarcinoma, as she had no evidence of Barrett esophagus on endoscopy. And we discuss the natural course of functional heartburn, including the spontaneous resolution rate of about 40%.

For treatment, we present her the rationale for using neuromodulators and reassure her that these medications are for treatment of visceral hypersensitivity, not for anxiety or depression. After the discussion, the patient opts to start amitriptyline therapy at 10 mg every night at bedtime, increasing the dose by 10 mg every 2 weeks until symptoms improve, up to 75–100 mg every day.

After 3 months, the patient reports a 90% improvement in symptoms while on amitriptyline 30 mg every night. She is also able to taper her antisecretory medications once symptoms are controlled. We plan to continue amitriptyline at the current dose for 6 to 12 months, then discuss a slow taper to see if her symptoms spontaneously resolve.

A 44-year-old woman presents with an 8-year history of intermittent heartburn, and in the past year she has been experiencing her symptoms daily. She says the heartburn is constant and is worse immediately after eating spicy or acidic foods. She says she has had no dysphagia, weight loss, or vomiting. Her symptoms have persisted despite taking a histamine (H)₂-receptor antagonist twice daily plus a proton pump inhibitor (PPI) before breakfast and dinner for more than 3 months.

She has undergone upper endoscopy 3 times in the past 8 years. Each time, the esophagus was normal with a regular Z-line and normal biopsy results from the proximal and distal esophagus.

The patient believes she has severe gastroesophageal reflux disease (GERD) and asks if she is a candidate for fundoplication surgery.

HEARTBURN IS A SYMPTOM; GERD IS A CONDITION

A distinction should be made between heartburn—the symptom of persistent retrosternal burning and discomfort—and gastroesophageal reflux disease—the condition in which reflux of stomach contents causes troublesome symptoms or complications.¹ While many clinicians initially diagnose patients who have heartburn as having GERD, there are many other potential causes of their symptoms.

For patients with persistent heartburn, an empiric trial of a once-daily PPI is usually effective, but one-third of patients continue to have heartburn.^2,3 The most common cause of this PPI-refractory heartburn is functional heartburn, a functional or hypersensitivity disorder of the esophagus.⁴

PATHOPHYSIOLOGY IS POORLY UNDERSTOOD

DIAGNOSTIC EVALUATION

Clinicians have several tests available for diagnosing these conditions.

Upper endoscopy

Upper endoscopy is recommended for patients with heartburn that does not respond to a 3-month trial of a PPI.⁹ Endoscopy is also indicated in any patient who has any of the following “alarm symptoms” that could be due to malignancy or peptic ulcer:

• Dysphagia
• Odynophagia
• Vomiting
• Unexplained weight loss or anemia
• Signs of gastrointestinal bleeding
• Anorexia
• New onset of dyspepsia in a patient over age 60.

During upper endoscopy, the esophagus is evaluated for reflux esophagitis, Barrett esophagus, and other inflammatory disorders such as infectious esophagitis. But even if the esophageal mucosa appears normal, the proximal and distal esophagus should be biopsied to rule out an inflammatory disorder such as eosinophilic or lymphocytic esophagitis.

If endoscopic and esophageal biopsy results are inconclusive, a workup for an esophageal motility disorder is the next step. Dysphagia is the most common symptom of these disorders, although the initial presenting symptom may be heartburn or regurgitation that persists despite PPI therapy.

Manometry is used to test for motility disorders such as achalasia and esophageal spasm.¹⁰ After applying a local anesthetic inside the nares, the clinician inserts a flexible catheter (about 4 mm in diameter) with 36 pressure sensors spaced at 1-cm intervals into the nares and passes it through the esophagus and lower esophageal sphincter. The patient then swallows liquid, and the sensors relay the esophageal response, creating a topographic plot that shows esophageal peristalsis and lower esophageal sphincter relaxation.

Achalasia is identified by incomplete lower esophageal sphincter relaxation combined with 100% failed peristalsis in the body of the esophagus. Esophageal spasms are identified by a shortened distal latency, which corresponds to premature contraction of the esophagus during peristalsis.¹¹

Esophageal pH testing

Measuring esophageal pH levels is an important step to quantify gastroesophageal reflux and determine if symptoms occur during reflux events. According to the updated Porto GERD consensus group recommendations,¹² a pH test is positive if the acid exposure time is greater than 6% of the testing period. Testing the pH differentiates between GERD (abnormal acid exposure), reflux hypersensitivity (normal acid exposure, strong correlation between symptoms and reflux events), and functional heartburn (normal acid exposure, negative correlation between reflux events and symptoms).⁵ For this test, a pH probe is placed in the esophagus transnasally or endoscopically. The probe records esophageal pH levels for 24 to 96 hours in an outpatient setting. Antisecretory therapy needs to be withheld for 7 to 10 days before the test.

Transnasal pH probe. For this approach, a thin catheter is inserted through the nares and advanced until the tip is 5 cm proximal to the lower esophageal sphincter. (The placement is guided by the results of esophageal manometry, which is done immediately before pH catheter placement.) The tube is secured with clear tape on the side of the patient’s face, and the end is connected to a portable recorder that compiles the data. The patient pushes a button on the recorder when experiencing heartburn symptoms. (A nurse instructs the patient on proper procedure.) After 24 hours, the patient either removes the catheter or has the clinic remove it. The pH and symptom data are downloaded and analyzed.

Transnasal pH testing can be combined with impedance measurement, which can detect nonacid reflux or weakly acid reflux. However, the clinical significance of this measurement is unclear, as multiple studies have found total acid exposure time to be a better predictor of response to therapy than weakly acid or nonacid reflux.¹²

Wireless pH probe. This method uses a disposable, catheter-free, capsule device to measure esophageal pH. The capsule, about the size of a gel capsule or pencil eraser, is attached to the patient’s esophageal lining, usually during upper endoscopy. The capsule records pH levels in the lower esophagus for 48 to 96 hours and transmits the data wirelessly to a receiver the patient wears. The patient pushes buttons on the receiver to record symptom-specific data when experiencing heartburn, chest pain, regurgitation, or cough. The capsule detaches from the esophagus spontaneously, generally within 7 days, and is passed out of the body through a bowel movement.

Diagnosing functional heartburn

CASE CONTINUED: NORMAL RESULTS ON TESTING

Based on these results, her condition is diagnosed as functional heartburn, consistent with the Rome IV criteria.⁵

Patient education is key

Patient education about the pathogenesis, natural history, and treatment options is the most important aspect of treating any functional gastrointestinal disorder. This includes the “brain-gut connection” and potential mechanisms of dysregulation. Patient education along with assessment of symptoms should be part of every visit, especially before discussing treatment options.

Patients whose condition is diagnosed as functional heartburn need reassurance that the condition is benign and, in particular, that the risk of progression to esophageal adenocarcinoma is minimal in the absence of Barrett esophagus.¹³ Also important to point out is that the disorder may spontaneously resolve: resolution rates of up to 40% have been reported for other functional gastrointestinal disorders.¹⁴

Antisecretory medications may work for some

A PPI or H₂-receptor antagonist is the most common first-line treatment for heartburn symptoms. Although most patients with functional heartburn experience no improvement in symptoms with an antisecretory agent, a small number report some relief, which suggests that acid-suppression therapy may have an indirect impact on pain modulation in the esophagus.¹⁵ In patients who report symptom relief with an antisecretory agent, we suggest continuing the medication tapered to the lowest effective dose, with repeated reassurance that the medication can be discontinued safely at any time.

Antireflux surgery should be avoided

Antireflux surgery should be avoided in patients with normal pH testing and no objective finding of reflux, as this is associated with worse subjective outcomes than in patients with abnormal pH test results.¹⁶

Neuromodulators

It is important to discuss with patients the concept of neuromodulation, including the fact that antidepressants are often used because of their effects on serotonin and norepinephrine, which decrease visceral hypersensitivity.

The selective serotonin reuptake inhibitor citalopram has been shown to reduce esophageal hypersensitivity,¹⁷ and a tricyclic antidepressant has been shown to improve quality of life.¹⁸ These results have led experts to recommend a trial of a low dose of either type of medication.¹⁹ The dose of tricyclic antidepressant often needs to be increased sequentially every 2 to 4 weeks.

Interestingly, melatonin 6 mg at bedtime has also shown efficacy for functional heartburn, potentially due to its antinociceptive properties.²⁰

Alternative and complementary therapies

Many esophageal centers use cognitive behavioral therapy and hypnotherapy as first-line treatment for functional esophageal disorders. Here again, it is important for the patient to understand the rationale of therapy for functional gastrointestinal disorders, given the stigma in the general population regarding psychotherapy.

Cognitive behavioral therapy has been used for functional gastrointestinal disorders for many years, as it has been shown to modulate visceral perception.²¹ Although published studies are limited, research regarding other functional esophageal disorders suggests that patients who commit to long-term behavioral therapy have had a significant improvement in symptoms.²²

The goal of esophageal-directed behavioral therapy is to promote focused relaxation using deep breathing techniques, which can help patients manage esophageal hypervigilance, especially if symptoms continue despite neuromodulator therapy. Specifically, hypnotherapy has been shown to modulate functional chest pain through the visceral sensory pathway and also to suppress gastric acid secretion.^21,23 A study of a 7-week hypnotherapy program reported significant benefits in heartburn relief and improved quality of life in patients with functional heartburn.²⁴ The data support the use of behavioral therapies as first-line therapy or as adjunctive therapy for patients already taking a neuromodulator.

CASE FOLLOW-UP: IMPROVEMENT WITH TREATMENT

During a follow-up visit, the patient is given several printed resources, including the Rome Foundation article on functional heartburn.⁵ We again emphasize the benign nature of functional heartburn, noting the minimal risk of progression to esophageal adenocarcinoma, as she had no evidence of Barrett esophagus on endoscopy. And we discuss the natural course of functional heartburn, including the spontaneous resolution rate of about 40%.

For treatment, we present her the rationale for using neuromodulators and reassure her that these medications are for treatment of visceral hypersensitivity, not for anxiety or depression. After the discussion, the patient opts to start amitriptyline therapy at 10 mg every night at bedtime, increasing the dose by 10 mg every 2 weeks until symptoms improve, up to 75–100 mg every day.

After 3 months, the patient reports a 90% improvement in symptoms while on amitriptyline 30 mg every night. She is also able to taper her antisecretory medications once symptoms are controlled. We plan to continue amitriptyline at the current dose for 6 to 12 months, then discuss a slow taper to see if her symptoms spontaneously resolve.

Most tricyclic antidepressants (TCAs) have US Food and Drug Administration approval for treatment of depression and anxiety disorders, but they are also a viable off-label option that should be considered by clinicians in specialties beyond psychiatry, especially for treating pain syndromes. Given the ongoing epidemic of opioid use disorder, increasing attention has been drawn to alternative strategies for chronic pain management, renewing an interest in the use of TCAs.

This review summarizes the pharmacologic properties of TCAs, their potential indications in conditions other than depression, and safety considerations.

BRIEF HISTORY OF TRICYCLICS

TCAs were originally designed in the 1950s and marketed later for treating depression. Due to their adverse effects and lethality in overdose quantities, over time they have been largely replaced by selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) in depression management. However, TCAs have been applied to conditions other than depression with varying degrees of efficacy and safety.

TCA PHARMACOLOGY

TCAs are absorbed in the small intestine and undergo first-pass metabolism in the liver. They bind extensively to proteins, leading to interactions with other protein-bound drugs. They are widely distributed throughout the systemic circulation because they are highly lipophilic, resulting in systemic effects including central nervous system manifestations.

Peak plasma concentration is at about 2 to 6 hours, and elimination half-life is around 24 hours for most agents, providing a long duration of action. Clearance depends on cytochrome P450 oxidative enzymes.¹

MECHANISMS OF ACTION

TCAs inhibit reuptake of norepinephrine and serotonin, resulting in accumulation of these neurotransmitters in the presynaptic cleft. They also block postsynaptic histamine, alpha-adrenergic, and muscarinic-acetylcholine receptors, causing a variety of adverse effects, including dry mouth, confusion, cognitive impairment, hypotension, orthostasis, blurred vision, urinary retention, drowsiness, and sedation.¹

Research suggests that TCAs relieve pain centrally through a descending pathway that inhibits transmission of pain signals in the spinal cord, as well as peripherally through complex anti-neuroimmune actions.² Norepinephrine appears to play a more important role in this process than serotonin, although both are deemed necessary for the “dual action” often cited in pain management,¹ which is also the rationale for widespread use of SNRIs to control pain.

Table 1 compares neurotransmitter reuptake mechanisms, adverse effect profiles, and typical dosages for depression for commonly prescribed TCAs.

POTENTIAL USES

Headache and migraine

TCAs have been shown to be effective for managing and preventing chronic headache syndromes.^3,4 Amitriptyline has been the most studied of the TCAs for both chronic daily and episodic migraine headache, showing the most efficacy among diverse drug classes (angiotensin II receptor blockers, anticonvulsants, beta-blockers, SSRIs) compared with placebo. However, in head-to-head trials, amitriptyline was no more effective than SSRIs, venlafaxine, topiramate, or propranolol.⁴ Jackson et al⁴ suggested that prophylactic medication choices should be tailored to patient characteristics and expected adverse effects, and specifically recommended that TCAs—particularly amitriptyline—be reserved for patients who have both migraine and depression.

Neuropathic pain

Neuropathic pain is defined as pain secondary to a lesion or disease of the somatosensory nervous system⁵ and is the pathomechanistic component of a number of conditions, including postherpetic neuralgia,⁶ diabetic and nondiabetic painful polyneuropathy,⁷ posttraumatic or postsurgical neuropathic pain⁸ (including plexus avulsion and complex regional pain syndrome⁹), central poststroke pain,¹⁰ spinal cord injury pain,¹¹ and multiple sclerosis-associated pain.¹²

As a group, TCAs appear to have a role as first-line agents for managing these varied neuropathic pain syndromes. In a recent meta-analysis,¹³ 16 (89%) of 18 placebo-controlled trials of TCAs (mainly amitriptyline at 25–150 mg/day) for these pain conditions were positive, with a combined number needed to treat of 3.6, suggesting a role for TCAs in these conditions. Of note, the TCAs desipramine¹⁴ and nortriptyline¹⁵ have demonstrated little evidence of efficacy in neuropathic pain syndromes.

Chronic low back pain

Chronic low back pain is a leading cause of loss of work, excessive healthcare expenditure, and disability in the United States. It can be due to numerous spinal conditions, including degenerative disk disease, spinal stenosis, lumbar spondylosis, and spinal arthropathy.

TCAs have been used to treat chronic low back pain for decades and have been repeatedly shown to be more effective than placebo in reducing pain severity.^16,17 A double-blind controlled trial¹⁸ from 1999 compared the effects of the TCA maprotiline (up to 150 mg daily), the SSRI paroxetine (up to 30 mg daily), and placebo and found a statistically significant reduction in back pain with maprotiline compared with paroxetine and placebo. However, a 2008 meta-analysis suggested little evidence that TCAs were superior to placebo.¹⁹

Evidence of TCA efficacy for back pain was reported in 2018 with a well-designed 6-month double-blind randomized controlled trial²⁰ comparing low-dose amitriptyline (25 mg) with an active comparator (benztropine 1 mg). The authors reported that amitriptyline was effective in reducing pain and pain-related disability without incurring serious adverse effects. They suggested continued use of TCAs for chronic low back pain if complicated with pain-related disability, insomnia, depression, or other comorbidity, although they called for further large-scale studies. They also cautioned that patients started the trial with symptoms similar to the adverse effects of TCAs themselves; this has implications for monitoring of symptoms as well as TCA adverse effects while using these drugs.

Fibromyalgia is a common, frustrating, noninflammatory pain syndrome characterized by diffuse hyperalgesia and multiple comorbidities.²¹ Although sleep hygiene, exercise, cognitive-behavioral therapy, some gabapentinoids (pregabalin), and a combination of these therapies have demonstrated efficacy, TCAs also offer robust benefits.

A meta-analysis of 9 placebo-controlled TCA trials showed large effect sizes for pain reduction, fatigue reduction, improved sleep quality, and reduced stiffness and tenderness, with the most significant of these improvements being for sleep.²² A separate meta-analysis calculated that the number needed to treat with amitriptyline for a positive outcome is 4.9.²³ Recent systematic reviews have supported these findings, listing TCAs as second-line agents after pregabalin, duloxetine, and milnacipran.²⁴

Of note, TCA monotherapy rarely produces a complete response in patients with moderate to severe fibromyalgia, chronic widespread pain, or significant comorbidities (depression, anxiety). Supplementation with cognitive-behavioral therapy, physical therapy, functional restoration, and other modalities is strongly recommended.

Abdominal and gastrointestinal pain

TCAs have been applied to a number of gastrointestinal syndromes with or without pain. Patients with irritable bowel syndrome have long been known to benefit from TCAs; the number needed to treat for symptomatic benefit over placebo is 3.5.^25,26

Although there is no substantial evidence that TCAs are useful in reducing active inflammation in inflammatory bowel disease, a study involving 81 patients found that residual noninflammatory gastrointestinal symptoms (such as diarrhea and pain) responded to TCAs, including nortriptyline and amitriptyline, with greater benefit for ulcerative colitis than for Crohn disease.²⁷

TCAs have also shown prophylactic benefit in cyclic vomiting syndrome, with a clinical response in over 75% of patients in controlled cohort studies.²⁸

The efficacy of TCAs in other abdominal or gastrointestinal syndromes is unclear or modest at best.²⁹ However, few alternative treatments exist for these conditions. Amitriptyline may help symptoms of functional dyspepsia,³⁰ but nortriptyline has proven ineffective in gastroparesis.³¹ Nonetheless, some authors²⁹ suggest considering TCAs on an individualized basis, with proper monitoring, in many if not most functional gastrointestinal disorders, especially when paired with behavioral therapies.

Pelvic and urogynecologic symptoms

Chronic pelvic pain affects up to 24% of women³² and 5% to 10% of men.³³ TCAs have shown efficacy in treating chronic pelvic pain with or without comorbid depression.³⁴ Amitriptyline and to a lesser extent nortriptyline are the TCAs most often prescribed. Pain relief appears to be independent of antidepressant effects and may be achieved at low doses; initial dosing ranges from 10 to 25 mg at bedtime, which may be increased to 100 mg as tolerated.³⁴

Based on a randomized, double-blind trial,³⁵ amitriptyline was recommended as a treatment option for interstitial cystitis or bladder pain, with the greatest symptom improvement in patients tolerating a daily dose of 50 mg.

Another study³⁶ randomized 56 women with chronic pelvic pain to amitriptyline or  gabapentin, or a combination of the drugs for 24 months. Although each regimen resulted in significant reduction in pain, fewer adverse effects occurred with gabapentin than amitriptyline. Poor compliance and early discontinuation of amitriptyline were common due to anticholinergic effects.

In small uncontrolled studies,³⁷ about half of women with chronic pelvic pain became pain-free after 8 weeks of treatment with nortriptyline and imipramine.

Randomized controlled studies are needed to confirm potential benefits of TCAs in chronic urologic and pelvic pain.

Insomnia

Insomnia affects 23% to 56% of people in the United States, Europe, and Asia³⁸ and is the reason for more than 5.5 million primary care visits annually.³⁹ TCAs (especially doxepin, maprotiline, and amitriptyline⁴⁰) have been shown to be an effective treatment, with an 82% increase in somnolence compared with placebo, as well as measurably improved total sleep time, enhanced sleep efficiency, reduced latency to persistent sleep, and decreased wake times after sleep onset.³⁸

Dosing should be kept at a minimum to minimize harsh anticholinergic effects and avoid daytime sedation. Patients should be advised to take new doses or dose escalations earlier in the night to ensure less hangover sedation the next morning.

For patients with insomnia and comorbid depression, the American Academy of Sleep Medicine suggests the addition of a low dose (eg, 10–25 mg) of a TCA at nighttime to complement preexisting, full-dose, non-TCA antidepressants, while monitoring for serotonin syndrome and other potential but exceedingly rare drug-drug interactions.⁴¹

Psychiatric indications other than depression

Beyond the known benefits in major depressive disorder, TCAs have been shown to be effective for obsessive-compulsive disorder, panic disorder, posttraumatic stress disorder, bulimia nervosa, and childhood enuresis.⁴² Given the shortage of mental health clinicians and the high prevalence of these conditions, nonpsychiatrist physicians should be familiar with the therapeutic potential of TCAs for these indications.

ADVERSE EFFECTS

Adverse effects vary among TCAs. Common ones include blurred vision, dry mouth, constipation, urinary retention, hypotension, tachycardia, tremor, weight gain, and sexual dysfunction.⁴³ Tertiary amines are generally more sedating than secondary amines and cause more anticholinergic effects (Table 1).

Despite widespread perceptions that TCAs are less tolerable than newer antidepressants, studies repeatedly suggest that they have an adverse-effect burden similar to that of SSRIs and SNRIs, although SSRIs have a greater tendency to produce nausea, whereas TCAs are more likely to cause constipation.⁴⁴

Discontinuation syndrome

Abrupt discontinuation or unintentionally missed doses of TCAs have been associated with a discontinuation syndrome in about 40% of users.⁴⁵ Patients should be warned about this possibility and the syndrome’s potential effects: dizziness, insomnia, headaches, nausea, vomiting, flulike achiness, and restlessness. Rebound depression, anxiety, panic, or other psychiatric symptoms may also occur. Symptoms generally present within 2 to 5 days after dose discontinuation and last 7 to 14 days.⁴⁵

However, all TCAs have a long half-life, allowing for sufficient coverage with once-daily dosing and thus carry a lower risk of discontinuation syndrome than many other antidepressants (78% with venlafaxine; 55% with paroxetine).⁴⁵

To discontinue therapy safely, the dosage should be reduced gradually. As is pharmacologically expected, the greatest likelihood of discontinuation syndrome is associated with longer duration of continuous treatment.

CONTRAINDICATIONS

Cardiac conduction abnormalities

TCAs should not be prescribed to patients who have right bundle branch block, a severe electrolyte disturbance, or other cardiac conduction deficit or arrhythmia that can prolong the QTc interval and elevate the risk of lethal arrhythmia.^46,47 Cardiac effects from TCAs are largely dose-dependent. Nevertheless, a baseline electrocardiogram can be obtained to assess cardiac risk, and dose escalation can proceed if results are normal (eg, appropriate conduction intervals, QTc ≤ 450 ms).

Advanced age

For elderly patients, TCAs should be prescribed with caution and sometimes not at all,⁴⁸ because anticholinergic effects may worsen preexisting urinary retention (including benign prostatic hyperplasia), narrow-angle glaucoma, imbalance and gait issues, and cognitive impairment and dementia. Dehydration and orthostatic hypotension are contraindications for TCAs, as they may precipitate falls or hypotensive shock.

Epilepsy

TCAs should also be used with caution in patients with epilepsy, as they lower the seizure threshold.

Concomitant monoamine oxidase inhibitor treatment

Giving TCAs together with monoamine oxidase inhibitor antidepressants should be avoided, given the risk of hypertensive crisis.

Suicide risk

TCAs are dangerous and potentially lethal in overdose and so should not be prescribed to suicidal or otherwise impulsive patients.

Pregnancy

TCAs are in pregnancy risk category C (animal studies show adverse effects on fetus; no adequate or well-controlled studies in humans; potential benefits may warrant use despite risks). Using TCAs during pregnancy has very rarely led to neonatal withdrawal such as irritability, jitteriness, and convulsions, as well as fetal QTc interval prolongation.⁴⁹

The American College of Obstetricians and Gynecologists recommends that therapy for depression during pregnancy be individualized, incorporating the expertise of the patient’s mental health clinician, obstetrician, primary healthcare provider, and pediatrician. In general, they recommend that TCAs should be avoided if possible and that alternatives such as SSRIs or SNRIs should be considered.⁵⁰

TCAs are excreted in breast milk, but they have not been detected in the serum of nursing infants, and no adverse events have been reported.

OVERDOSE IS HIGHLY DANGEROUS

Severe morbidity and death are associated with TCA overdose, characterized by  convulsions, cardiac arrest, and coma (the “3 Cs”). These dangers occur at much higher rates with TCAs than with other antidepressants.⁴³ Signs and symptoms of toxicity develop rapidly, usually within the first hour of overdose. Manifestations of overdose include prolonged QTc, cardiac arrhythmias, tachycardia, hypertension, severe hypotension, agitation, seizures, central nervous system depression, hallucinations, seizures, and coma.

Overdose management includes activated charcoal, seizure control, cardioversion, hydration, electrolyte stabilization, and other intensive care.

OFF-LABEL TCA MANAGEMENT

Dosing recommendations for off-label use of TCAs vary based on the condition, the medication, and the suggestions of individual authors and researchers. In general, dosing ranges for pain and other nondepression indications may be lower than for severe depression (Table 2).¹

As with any pharmacologic titration, monitoring for rate-limiting adverse effects is recommended. We suggest caution, tailoring the approach to the patient, and routinely assessing for adverse effects and other safety considerations.

In addition, we strongly recommend supplementing TCA therapy with nonpharmacologic strategies such as lifestyle changes, dietary modifications, exercise, physical therapy, and mental health optimization.

Most tricyclic antidepressants (TCAs) have US Food and Drug Administration approval for treatment of depression and anxiety disorders, but they are also a viable off-label option that should be considered by clinicians in specialties beyond psychiatry, especially for treating pain syndromes. Given the ongoing epidemic of opioid use disorder, increasing attention has been drawn to alternative strategies for chronic pain management, renewing an interest in the use of TCAs.

This review summarizes the pharmacologic properties of TCAs, their potential indications in conditions other than depression, and safety considerations.

BRIEF HISTORY OF TRICYCLICS

TCAs were originally designed in the 1950s and marketed later for treating depression. Due to their adverse effects and lethality in overdose quantities, over time they have been largely replaced by selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) in depression management. However, TCAs have been applied to conditions other than depression with varying degrees of efficacy and safety.

TCA PHARMACOLOGY

TCAs are absorbed in the small intestine and undergo first-pass metabolism in the liver. They bind extensively to proteins, leading to interactions with other protein-bound drugs. They are widely distributed throughout the systemic circulation because they are highly lipophilic, resulting in systemic effects including central nervous system manifestations.

Peak plasma concentration is at about 2 to 6 hours, and elimination half-life is around 24 hours for most agents, providing a long duration of action. Clearance depends on cytochrome P450 oxidative enzymes.¹

MECHANISMS OF ACTION

TCAs inhibit reuptake of norepinephrine and serotonin, resulting in accumulation of these neurotransmitters in the presynaptic cleft. They also block postsynaptic histamine, alpha-adrenergic, and muscarinic-acetylcholine receptors, causing a variety of adverse effects, including dry mouth, confusion, cognitive impairment, hypotension, orthostasis, blurred vision, urinary retention, drowsiness, and sedation.¹

Research suggests that TCAs relieve pain centrally through a descending pathway that inhibits transmission of pain signals in the spinal cord, as well as peripherally through complex anti-neuroimmune actions.² Norepinephrine appears to play a more important role in this process than serotonin, although both are deemed necessary for the “dual action” often cited in pain management,¹ which is also the rationale for widespread use of SNRIs to control pain.

Table 1 compares neurotransmitter reuptake mechanisms, adverse effect profiles, and typical dosages for depression for commonly prescribed TCAs.

POTENTIAL USES

Headache and migraine

TCAs have been shown to be effective for managing and preventing chronic headache syndromes.^3,4 Amitriptyline has been the most studied of the TCAs for both chronic daily and episodic migraine headache, showing the most efficacy among diverse drug classes (angiotensin II receptor blockers, anticonvulsants, beta-blockers, SSRIs) compared with placebo. However, in head-to-head trials, amitriptyline was no more effective than SSRIs, venlafaxine, topiramate, or propranolol.⁴ Jackson et al⁴ suggested that prophylactic medication choices should be tailored to patient characteristics and expected adverse effects, and specifically recommended that TCAs—particularly amitriptyline—be reserved for patients who have both migraine and depression.

Neuropathic pain

Neuropathic pain is defined as pain secondary to a lesion or disease of the somatosensory nervous system⁵ and is the pathomechanistic component of a number of conditions, including postherpetic neuralgia,⁶ diabetic and nondiabetic painful polyneuropathy,⁷ posttraumatic or postsurgical neuropathic pain⁸ (including plexus avulsion and complex regional pain syndrome⁹), central poststroke pain,¹⁰ spinal cord injury pain,¹¹ and multiple sclerosis-associated pain.¹²

As a group, TCAs appear to have a role as first-line agents for managing these varied neuropathic pain syndromes. In a recent meta-analysis,¹³ 16 (89%) of 18 placebo-controlled trials of TCAs (mainly amitriptyline at 25–150 mg/day) for these pain conditions were positive, with a combined number needed to treat of 3.6, suggesting a role for TCAs in these conditions. Of note, the TCAs desipramine¹⁴ and nortriptyline¹⁵ have demonstrated little evidence of efficacy in neuropathic pain syndromes.

Chronic low back pain

Chronic low back pain is a leading cause of loss of work, excessive healthcare expenditure, and disability in the United States. It can be due to numerous spinal conditions, including degenerative disk disease, spinal stenosis, lumbar spondylosis, and spinal arthropathy.

TCAs have been used to treat chronic low back pain for decades and have been repeatedly shown to be more effective than placebo in reducing pain severity.^16,17 A double-blind controlled trial¹⁸ from 1999 compared the effects of the TCA maprotiline (up to 150 mg daily), the SSRI paroxetine (up to 30 mg daily), and placebo and found a statistically significant reduction in back pain with maprotiline compared with paroxetine and placebo. However, a 2008 meta-analysis suggested little evidence that TCAs were superior to placebo.¹⁹

Evidence of TCA efficacy for back pain was reported in 2018 with a well-designed 6-month double-blind randomized controlled trial²⁰ comparing low-dose amitriptyline (25 mg) with an active comparator (benztropine 1 mg). The authors reported that amitriptyline was effective in reducing pain and pain-related disability without incurring serious adverse effects. They suggested continued use of TCAs for chronic low back pain if complicated with pain-related disability, insomnia, depression, or other comorbidity, although they called for further large-scale studies. They also cautioned that patients started the trial with symptoms similar to the adverse effects of TCAs themselves; this has implications for monitoring of symptoms as well as TCA adverse effects while using these drugs.

Fibromyalgia is a common, frustrating, noninflammatory pain syndrome characterized by diffuse hyperalgesia and multiple comorbidities.²¹ Although sleep hygiene, exercise, cognitive-behavioral therapy, some gabapentinoids (pregabalin), and a combination of these therapies have demonstrated efficacy, TCAs also offer robust benefits.

A meta-analysis of 9 placebo-controlled TCA trials showed large effect sizes for pain reduction, fatigue reduction, improved sleep quality, and reduced stiffness and tenderness, with the most significant of these improvements being for sleep.²² A separate meta-analysis calculated that the number needed to treat with amitriptyline for a positive outcome is 4.9.²³ Recent systematic reviews have supported these findings, listing TCAs as second-line agents after pregabalin, duloxetine, and milnacipran.²⁴

Of note, TCA monotherapy rarely produces a complete response in patients with moderate to severe fibromyalgia, chronic widespread pain, or significant comorbidities (depression, anxiety). Supplementation with cognitive-behavioral therapy, physical therapy, functional restoration, and other modalities is strongly recommended.

Abdominal and gastrointestinal pain

TCAs have been applied to a number of gastrointestinal syndromes with or without pain. Patients with irritable bowel syndrome have long been known to benefit from TCAs; the number needed to treat for symptomatic benefit over placebo is 3.5.^25,26

Although there is no substantial evidence that TCAs are useful in reducing active inflammation in inflammatory bowel disease, a study involving 81 patients found that residual noninflammatory gastrointestinal symptoms (such as diarrhea and pain) responded to TCAs, including nortriptyline and amitriptyline, with greater benefit for ulcerative colitis than for Crohn disease.²⁷

TCAs have also shown prophylactic benefit in cyclic vomiting syndrome, with a clinical response in over 75% of patients in controlled cohort studies.²⁸

The efficacy of TCAs in other abdominal or gastrointestinal syndromes is unclear or modest at best.²⁹ However, few alternative treatments exist for these conditions. Amitriptyline may help symptoms of functional dyspepsia,³⁰ but nortriptyline has proven ineffective in gastroparesis.³¹ Nonetheless, some authors²⁹ suggest considering TCAs on an individualized basis, with proper monitoring, in many if not most functional gastrointestinal disorders, especially when paired with behavioral therapies.

Pelvic and urogynecologic symptoms

Chronic pelvic pain affects up to 24% of women³² and 5% to 10% of men.³³ TCAs have shown efficacy in treating chronic pelvic pain with or without comorbid depression.³⁴ Amitriptyline and to a lesser extent nortriptyline are the TCAs most often prescribed. Pain relief appears to be independent of antidepressant effects and may be achieved at low doses; initial dosing ranges from 10 to 25 mg at bedtime, which may be increased to 100 mg as tolerated.³⁴

Based on a randomized, double-blind trial,³⁵ amitriptyline was recommended as a treatment option for interstitial cystitis or bladder pain, with the greatest symptom improvement in patients tolerating a daily dose of 50 mg.

Another study³⁶ randomized 56 women with chronic pelvic pain to amitriptyline or  gabapentin, or a combination of the drugs for 24 months. Although each regimen resulted in significant reduction in pain, fewer adverse effects occurred with gabapentin than amitriptyline. Poor compliance and early discontinuation of amitriptyline were common due to anticholinergic effects.

In small uncontrolled studies,³⁷ about half of women with chronic pelvic pain became pain-free after 8 weeks of treatment with nortriptyline and imipramine.

Randomized controlled studies are needed to confirm potential benefits of TCAs in chronic urologic and pelvic pain.

Insomnia

Insomnia affects 23% to 56% of people in the United States, Europe, and Asia³⁸ and is the reason for more than 5.5 million primary care visits annually.³⁹ TCAs (especially doxepin, maprotiline, and amitriptyline⁴⁰) have been shown to be an effective treatment, with an 82% increase in somnolence compared with placebo, as well as measurably improved total sleep time, enhanced sleep efficiency, reduced latency to persistent sleep, and decreased wake times after sleep onset.³⁸

Dosing should be kept at a minimum to minimize harsh anticholinergic effects and avoid daytime sedation. Patients should be advised to take new doses or dose escalations earlier in the night to ensure less hangover sedation the next morning.

For patients with insomnia and comorbid depression, the American Academy of Sleep Medicine suggests the addition of a low dose (eg, 10–25 mg) of a TCA at nighttime to complement preexisting, full-dose, non-TCA antidepressants, while monitoring for serotonin syndrome and other potential but exceedingly rare drug-drug interactions.⁴¹

Psychiatric indications other than depression

Beyond the known benefits in major depressive disorder, TCAs have been shown to be effective for obsessive-compulsive disorder, panic disorder, posttraumatic stress disorder, bulimia nervosa, and childhood enuresis.⁴² Given the shortage of mental health clinicians and the high prevalence of these conditions, nonpsychiatrist physicians should be familiar with the therapeutic potential of TCAs for these indications.

ADVERSE EFFECTS

Adverse effects vary among TCAs. Common ones include blurred vision, dry mouth, constipation, urinary retention, hypotension, tachycardia, tremor, weight gain, and sexual dysfunction.⁴³ Tertiary amines are generally more sedating than secondary amines and cause more anticholinergic effects (Table 1).

Despite widespread perceptions that TCAs are less tolerable than newer antidepressants, studies repeatedly suggest that they have an adverse-effect burden similar to that of SSRIs and SNRIs, although SSRIs have a greater tendency to produce nausea, whereas TCAs are more likely to cause constipation.⁴⁴

Discontinuation syndrome

Abrupt discontinuation or unintentionally missed doses of TCAs have been associated with a discontinuation syndrome in about 40% of users.⁴⁵ Patients should be warned about this possibility and the syndrome’s potential effects: dizziness, insomnia, headaches, nausea, vomiting, flulike achiness, and restlessness. Rebound depression, anxiety, panic, or other psychiatric symptoms may also occur. Symptoms generally present within 2 to 5 days after dose discontinuation and last 7 to 14 days.⁴⁵

However, all TCAs have a long half-life, allowing for sufficient coverage with once-daily dosing and thus carry a lower risk of discontinuation syndrome than many other antidepressants (78% with venlafaxine; 55% with paroxetine).⁴⁵

To discontinue therapy safely, the dosage should be reduced gradually. As is pharmacologically expected, the greatest likelihood of discontinuation syndrome is associated with longer duration of continuous treatment.

CONTRAINDICATIONS

Cardiac conduction abnormalities

TCAs should not be prescribed to patients who have right bundle branch block, a severe electrolyte disturbance, or other cardiac conduction deficit or arrhythmia that can prolong the QTc interval and elevate the risk of lethal arrhythmia.^46,47 Cardiac effects from TCAs are largely dose-dependent. Nevertheless, a baseline electrocardiogram can be obtained to assess cardiac risk, and dose escalation can proceed if results are normal (eg, appropriate conduction intervals, QTc ≤ 450 ms).

Advanced age

For elderly patients, TCAs should be prescribed with caution and sometimes not at all,⁴⁸ because anticholinergic effects may worsen preexisting urinary retention (including benign prostatic hyperplasia), narrow-angle glaucoma, imbalance and gait issues, and cognitive impairment and dementia. Dehydration and orthostatic hypotension are contraindications for TCAs, as they may precipitate falls or hypotensive shock.

Epilepsy

TCAs should also be used with caution in patients with epilepsy, as they lower the seizure threshold.

Concomitant monoamine oxidase inhibitor treatment

Giving TCAs together with monoamine oxidase inhibitor antidepressants should be avoided, given the risk of hypertensive crisis.

Suicide risk

TCAs are dangerous and potentially lethal in overdose and so should not be prescribed to suicidal or otherwise impulsive patients.

Pregnancy

TCAs are in pregnancy risk category C (animal studies show adverse effects on fetus; no adequate or well-controlled studies in humans; potential benefits may warrant use despite risks). Using TCAs during pregnancy has very rarely led to neonatal withdrawal such as irritability, jitteriness, and convulsions, as well as fetal QTc interval prolongation.⁴⁹

The American College of Obstetricians and Gynecologists recommends that therapy for depression during pregnancy be individualized, incorporating the expertise of the patient’s mental health clinician, obstetrician, primary healthcare provider, and pediatrician. In general, they recommend that TCAs should be avoided if possible and that alternatives such as SSRIs or SNRIs should be considered.⁵⁰

TCAs are excreted in breast milk, but they have not been detected in the serum of nursing infants, and no adverse events have been reported.

OVERDOSE IS HIGHLY DANGEROUS

Severe morbidity and death are associated with TCA overdose, characterized by  convulsions, cardiac arrest, and coma (the “3 Cs”). These dangers occur at much higher rates with TCAs than with other antidepressants.⁴³ Signs and symptoms of toxicity develop rapidly, usually within the first hour of overdose. Manifestations of overdose include prolonged QTc, cardiac arrhythmias, tachycardia, hypertension, severe hypotension, agitation, seizures, central nervous system depression, hallucinations, seizures, and coma.

Overdose management includes activated charcoal, seizure control, cardioversion, hydration, electrolyte stabilization, and other intensive care.

OFF-LABEL TCA MANAGEMENT

Dosing recommendations for off-label use of TCAs vary based on the condition, the medication, and the suggestions of individual authors and researchers. In general, dosing ranges for pain and other nondepression indications may be lower than for severe depression (Table 2).¹

As with any pharmacologic titration, monitoring for rate-limiting adverse effects is recommended. We suggest caution, tailoring the approach to the patient, and routinely assessing for adverse effects and other safety considerations.

In addition, we strongly recommend supplementing TCA therapy with nonpharmacologic strategies such as lifestyle changes, dietary modifications, exercise, physical therapy, and mental health optimization.

Most tricyclic antidepressants (TCAs) have US Food and Drug Administration approval for treatment of depression and anxiety disorders, but they are also a viable off-label option that should be considered by clinicians in specialties beyond psychiatry, especially for treating pain syndromes. Given the ongoing epidemic of opioid use disorder, increasing attention has been drawn to alternative strategies for chronic pain management, renewing an interest in the use of TCAs.

This review summarizes the pharmacologic properties of TCAs, their potential indications in conditions other than depression, and safety considerations.

BRIEF HISTORY OF TRICYCLICS

TCAs were originally designed in the 1950s and marketed later for treating depression. Due to their adverse effects and lethality in overdose quantities, over time they have been largely replaced by selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) in depression management. However, TCAs have been applied to conditions other than depression with varying degrees of efficacy and safety.

TCA PHARMACOLOGY

TCAs are absorbed in the small intestine and undergo first-pass metabolism in the liver. They bind extensively to proteins, leading to interactions with other protein-bound drugs. They are widely distributed throughout the systemic circulation because they are highly lipophilic, resulting in systemic effects including central nervous system manifestations.

Peak plasma concentration is at about 2 to 6 hours, and elimination half-life is around 24 hours for most agents, providing a long duration of action. Clearance depends on cytochrome P450 oxidative enzymes.¹

MECHANISMS OF ACTION

TCAs inhibit reuptake of norepinephrine and serotonin, resulting in accumulation of these neurotransmitters in the presynaptic cleft. They also block postsynaptic histamine, alpha-adrenergic, and muscarinic-acetylcholine receptors, causing a variety of adverse effects, including dry mouth, confusion, cognitive impairment, hypotension, orthostasis, blurred vision, urinary retention, drowsiness, and sedation.¹

Research suggests that TCAs relieve pain centrally through a descending pathway that inhibits transmission of pain signals in the spinal cord, as well as peripherally through complex anti-neuroimmune actions.² Norepinephrine appears to play a more important role in this process than serotonin, although both are deemed necessary for the “dual action” often cited in pain management,¹ which is also the rationale for widespread use of SNRIs to control pain.

Table 1 compares neurotransmitter reuptake mechanisms, adverse effect profiles, and typical dosages for depression for commonly prescribed TCAs.

POTENTIAL USES

Headache and migraine

TCAs have been shown to be effective for managing and preventing chronic headache syndromes.^3,4 Amitriptyline has been the most studied of the TCAs for both chronic daily and episodic migraine headache, showing the most efficacy among diverse drug classes (angiotensin II receptor blockers, anticonvulsants, beta-blockers, SSRIs) compared with placebo. However, in head-to-head trials, amitriptyline was no more effective than SSRIs, venlafaxine, topiramate, or propranolol.⁴ Jackson et al⁴ suggested that prophylactic medication choices should be tailored to patient characteristics and expected adverse effects, and specifically recommended that TCAs—particularly amitriptyline—be reserved for patients who have both migraine and depression.

Neuropathic pain

Neuropathic pain is defined as pain secondary to a lesion or disease of the somatosensory nervous system⁵ and is the pathomechanistic component of a number of conditions, including postherpetic neuralgia,⁶ diabetic and nondiabetic painful polyneuropathy,⁷ posttraumatic or postsurgical neuropathic pain⁸ (including plexus avulsion and complex regional pain syndrome⁹), central poststroke pain,¹⁰ spinal cord injury pain,¹¹ and multiple sclerosis-associated pain.¹²

As a group, TCAs appear to have a role as first-line agents for managing these varied neuropathic pain syndromes. In a recent meta-analysis,¹³ 16 (89%) of 18 placebo-controlled trials of TCAs (mainly amitriptyline at 25–150 mg/day) for these pain conditions were positive, with a combined number needed to treat of 3.6, suggesting a role for TCAs in these conditions. Of note, the TCAs desipramine¹⁴ and nortriptyline¹⁵ have demonstrated little evidence of efficacy in neuropathic pain syndromes.

Chronic low back pain

Chronic low back pain is a leading cause of loss of work, excessive healthcare expenditure, and disability in the United States. It can be due to numerous spinal conditions, including degenerative disk disease, spinal stenosis, lumbar spondylosis, and spinal arthropathy.

TCAs have been used to treat chronic low back pain for decades and have been repeatedly shown to be more effective than placebo in reducing pain severity.^16,17 A double-blind controlled trial¹⁸ from 1999 compared the effects of the TCA maprotiline (up to 150 mg daily), the SSRI paroxetine (up to 30 mg daily), and placebo and found a statistically significant reduction in back pain with maprotiline compared with paroxetine and placebo. However, a 2008 meta-analysis suggested little evidence that TCAs were superior to placebo.¹⁹

Evidence of TCA efficacy for back pain was reported in 2018 with a well-designed 6-month double-blind randomized controlled trial²⁰ comparing low-dose amitriptyline (25 mg) with an active comparator (benztropine 1 mg). The authors reported that amitriptyline was effective in reducing pain and pain-related disability without incurring serious adverse effects. They suggested continued use of TCAs for chronic low back pain if complicated with pain-related disability, insomnia, depression, or other comorbidity, although they called for further large-scale studies. They also cautioned that patients started the trial with symptoms similar to the adverse effects of TCAs themselves; this has implications for monitoring of symptoms as well as TCA adverse effects while using these drugs.

Fibromyalgia is a common, frustrating, noninflammatory pain syndrome characterized by diffuse hyperalgesia and multiple comorbidities.²¹ Although sleep hygiene, exercise, cognitive-behavioral therapy, some gabapentinoids (pregabalin), and a combination of these therapies have demonstrated efficacy, TCAs also offer robust benefits.

A meta-analysis of 9 placebo-controlled TCA trials showed large effect sizes for pain reduction, fatigue reduction, improved sleep quality, and reduced stiffness and tenderness, with the most significant of these improvements being for sleep.²² A separate meta-analysis calculated that the number needed to treat with amitriptyline for a positive outcome is 4.9.²³ Recent systematic reviews have supported these findings, listing TCAs as second-line agents after pregabalin, duloxetine, and milnacipran.²⁴

Of note, TCA monotherapy rarely produces a complete response in patients with moderate to severe fibromyalgia, chronic widespread pain, or significant comorbidities (depression, anxiety). Supplementation with cognitive-behavioral therapy, physical therapy, functional restoration, and other modalities is strongly recommended.

Abdominal and gastrointestinal pain

TCAs have been applied to a number of gastrointestinal syndromes with or without pain. Patients with irritable bowel syndrome have long been known to benefit from TCAs; the number needed to treat for symptomatic benefit over placebo is 3.5.^25,26

Although there is no substantial evidence that TCAs are useful in reducing active inflammation in inflammatory bowel disease, a study involving 81 patients found that residual noninflammatory gastrointestinal symptoms (such as diarrhea and pain) responded to TCAs, including nortriptyline and amitriptyline, with greater benefit for ulcerative colitis than for Crohn disease.²⁷

TCAs have also shown prophylactic benefit in cyclic vomiting syndrome, with a clinical response in over 75% of patients in controlled cohort studies.²⁸

The efficacy of TCAs in other abdominal or gastrointestinal syndromes is unclear or modest at best.²⁹ However, few alternative treatments exist for these conditions. Amitriptyline may help symptoms of functional dyspepsia,³⁰ but nortriptyline has proven ineffective in gastroparesis.³¹ Nonetheless, some authors²⁹ suggest considering TCAs on an individualized basis, with proper monitoring, in many if not most functional gastrointestinal disorders, especially when paired with behavioral therapies.

Pelvic and urogynecologic symptoms

Chronic pelvic pain affects up to 24% of women³² and 5% to 10% of men.³³ TCAs have shown efficacy in treating chronic pelvic pain with or without comorbid depression.³⁴ Amitriptyline and to a lesser extent nortriptyline are the TCAs most often prescribed. Pain relief appears to be independent of antidepressant effects and may be achieved at low doses; initial dosing ranges from 10 to 25 mg at bedtime, which may be increased to 100 mg as tolerated.³⁴

Based on a randomized, double-blind trial,³⁵ amitriptyline was recommended as a treatment option for interstitial cystitis or bladder pain, with the greatest symptom improvement in patients tolerating a daily dose of 50 mg.

Another study³⁶ randomized 56 women with chronic pelvic pain to amitriptyline or  gabapentin, or a combination of the drugs for 24 months. Although each regimen resulted in significant reduction in pain, fewer adverse effects occurred with gabapentin than amitriptyline. Poor compliance and early discontinuation of amitriptyline were common due to anticholinergic effects.

In small uncontrolled studies,³⁷ about half of women with chronic pelvic pain became pain-free after 8 weeks of treatment with nortriptyline and imipramine.

Randomized controlled studies are needed to confirm potential benefits of TCAs in chronic urologic and pelvic pain.

Insomnia

Insomnia affects 23% to 56% of people in the United States, Europe, and Asia³⁸ and is the reason for more than 5.5 million primary care visits annually.³⁹ TCAs (especially doxepin, maprotiline, and amitriptyline⁴⁰) have been shown to be an effective treatment, with an 82% increase in somnolence compared with placebo, as well as measurably improved total sleep time, enhanced sleep efficiency, reduced latency to persistent sleep, and decreased wake times after sleep onset.³⁸

Dosing should be kept at a minimum to minimize harsh anticholinergic effects and avoid daytime sedation. Patients should be advised to take new doses or dose escalations earlier in the night to ensure less hangover sedation the next morning.

For patients with insomnia and comorbid depression, the American Academy of Sleep Medicine suggests the addition of a low dose (eg, 10–25 mg) of a TCA at nighttime to complement preexisting, full-dose, non-TCA antidepressants, while monitoring for serotonin syndrome and other potential but exceedingly rare drug-drug interactions.⁴¹

Psychiatric indications other than depression

Beyond the known benefits in major depressive disorder, TCAs have been shown to be effective for obsessive-compulsive disorder, panic disorder, posttraumatic stress disorder, bulimia nervosa, and childhood enuresis.⁴² Given the shortage of mental health clinicians and the high prevalence of these conditions, nonpsychiatrist physicians should be familiar with the therapeutic potential of TCAs for these indications.

ADVERSE EFFECTS

Adverse effects vary among TCAs. Common ones include blurred vision, dry mouth, constipation, urinary retention, hypotension, tachycardia, tremor, weight gain, and sexual dysfunction.⁴³ Tertiary amines are generally more sedating than secondary amines and cause more anticholinergic effects (Table 1).

Despite widespread perceptions that TCAs are less tolerable than newer antidepressants, studies repeatedly suggest that they have an adverse-effect burden similar to that of SSRIs and SNRIs, although SSRIs have a greater tendency to produce nausea, whereas TCAs are more likely to cause constipation.⁴⁴

Discontinuation syndrome

Abrupt discontinuation or unintentionally missed doses of TCAs have been associated with a discontinuation syndrome in about 40% of users.⁴⁵ Patients should be warned about this possibility and the syndrome’s potential effects: dizziness, insomnia, headaches, nausea, vomiting, flulike achiness, and restlessness. Rebound depression, anxiety, panic, or other psychiatric symptoms may also occur. Symptoms generally present within 2 to 5 days after dose discontinuation and last 7 to 14 days.⁴⁵

However, all TCAs have a long half-life, allowing for sufficient coverage with once-daily dosing and thus carry a lower risk of discontinuation syndrome than many other antidepressants (78% with venlafaxine; 55% with paroxetine).⁴⁵

To discontinue therapy safely, the dosage should be reduced gradually. As is pharmacologically expected, the greatest likelihood of discontinuation syndrome is associated with longer duration of continuous treatment.

CONTRAINDICATIONS

Cardiac conduction abnormalities

TCAs should not be prescribed to patients who have right bundle branch block, a severe electrolyte disturbance, or other cardiac conduction deficit or arrhythmia that can prolong the QTc interval and elevate the risk of lethal arrhythmia.^46,47 Cardiac effects from TCAs are largely dose-dependent. Nevertheless, a baseline electrocardiogram can be obtained to assess cardiac risk, and dose escalation can proceed if results are normal (eg, appropriate conduction intervals, QTc ≤ 450 ms).

Advanced age

For elderly patients, TCAs should be prescribed with caution and sometimes not at all,⁴⁸ because anticholinergic effects may worsen preexisting urinary retention (including benign prostatic hyperplasia), narrow-angle glaucoma, imbalance and gait issues, and cognitive impairment and dementia. Dehydration and orthostatic hypotension are contraindications for TCAs, as they may precipitate falls or hypotensive shock.

Epilepsy

TCAs should also be used with caution in patients with epilepsy, as they lower the seizure threshold.

Concomitant monoamine oxidase inhibitor treatment

Giving TCAs together with monoamine oxidase inhibitor antidepressants should be avoided, given the risk of hypertensive crisis.

Suicide risk

TCAs are dangerous and potentially lethal in overdose and so should not be prescribed to suicidal or otherwise impulsive patients.

Pregnancy

TCAs are in pregnancy risk category C (animal studies show adverse effects on fetus; no adequate or well-controlled studies in humans; potential benefits may warrant use despite risks). Using TCAs during pregnancy has very rarely led to neonatal withdrawal such as irritability, jitteriness, and convulsions, as well as fetal QTc interval prolongation.⁴⁹

The American College of Obstetricians and Gynecologists recommends that therapy for depression during pregnancy be individualized, incorporating the expertise of the patient’s mental health clinician, obstetrician, primary healthcare provider, and pediatrician. In general, they recommend that TCAs should be avoided if possible and that alternatives such as SSRIs or SNRIs should be considered.⁵⁰

TCAs are excreted in breast milk, but they have not been detected in the serum of nursing infants, and no adverse events have been reported.

OVERDOSE IS HIGHLY DANGEROUS

Severe morbidity and death are associated with TCA overdose, characterized by  convulsions, cardiac arrest, and coma (the “3 Cs”). These dangers occur at much higher rates with TCAs than with other antidepressants.⁴³ Signs and symptoms of toxicity develop rapidly, usually within the first hour of overdose. Manifestations of overdose include prolonged QTc, cardiac arrhythmias, tachycardia, hypertension, severe hypotension, agitation, seizures, central nervous system depression, hallucinations, seizures, and coma.

Overdose management includes activated charcoal, seizure control, cardioversion, hydration, electrolyte stabilization, and other intensive care.

OFF-LABEL TCA MANAGEMENT

Dosing recommendations for off-label use of TCAs vary based on the condition, the medication, and the suggestions of individual authors and researchers. In general, dosing ranges for pain and other nondepression indications may be lower than for severe depression (Table 2).¹

As with any pharmacologic titration, monitoring for rate-limiting adverse effects is recommended. We suggest caution, tailoring the approach to the patient, and routinely assessing for adverse effects and other safety considerations.

In addition, we strongly recommend supplementing TCA therapy with nonpharmacologic strategies such as lifestyle changes, dietary modifications, exercise, physical therapy, and mental health optimization.

Perceptions regarding the use of gabapentin for alcohol use disorder (AUD) have shifted over time.^1–4 Early on, the drug was deemed to be benign and effective.^4–6 But more and more, concerns are being raised about its recreational use to achieve euphoria,⁷ and the drug is often misused by vulnerable populations, particularly those with opioid use disorder.^7–9

Given the large number of gabapentin prescriptions written off-label for AUD, it is incumbent on providers to understand how to prescribe it responsibly.^7–9 To that end, this article focuses on the benefits—and concerns—of this treatment option. We describe the effects of gabapentin on the central nervous system and how it may mitigate alcohol withdrawal and increase the likelihood of abstinence. In addition, we review clinical trials that evaluated potential roles of gabapentin in AUD, discuss the drug’s misuse potential, and suggest a framework for its appropriate use in AUD management.

ALCOHOL USE DISORDER IS COMMON AND SERIOUS

AUD affects about 14% of US adults and represents a significant health burden,¹ often with severe clinical and social implications. It manifests as compulsive drinking and loss of control despite adverse consequences on various life domains.¹⁰ It is generally associated with cravings, tolerance, and withdrawal symptoms upon cessation. Alcohol withdrawal is characterized by tremors, anxiety, sweating, nausea, and tachycardia, and in severe cases, may involve hallucinations, seizures, and delirium tremens. Untreated, alcohol withdrawal can be fatal.¹⁰

Even though psychosocial treatments for AUD by themselves are associated with high relapse rates, pharmacotherapy is underutilized. Three drugs approved by the US Food and Drug Administration (FDA) are available to treat it, but they are often poorly accepted and have limited efficacy. For these reasons, there is considerable interest in finding alternatives. Gabapentin is one of several agents that have been studied (Table 1). The topic has been reviewed in depth by Soyka and Müller.¹¹

GABAPENTIN REDUCES EXCITATION

The anticonvulsant gabapentin is FDA-approved for treating epilepsy, postherpetic neuralgia, and restless leg syndrome.^8,12–14 It binds and selectively impedes voltage-sensitive calcium channels, the pores in cell membrane that permit calcium to enter a neuron in response to changes in electrical currents.¹⁵

Gabapentin is believed to decrease excitation of the central nervous system in multiple ways:

• It reduces the release of glutamate, a key component of the excitatory system¹⁶
• It increases the concentration of gamma-aminobutyric acid (GABA), the main inhibitory neurotransmitter in the brain7
• By binding the alpha-2-delta type 1 subunit of voltage-sensitive calcium channels,^8,15–17 it inhibits excitatory synapse formation independent of calcium channel activity¹⁶
• By blocking excitatory neurotransmission, it also may indirectly increase the concentration of GABA in the central nervous system1^6,17
• It modulates action of glutamic acid decarboxylase (involved in the synthesis of GABA) and glutamate synthesizing enzyme to increase GABA and decrease glutamate.¹⁷

The actions of alcohol on the brain are also complex.¹⁸ Alpha-2-delta type 1 subunits of calcium channels are upregulated in the reward centers of the brain by addictive substances, including alcohol.¹⁶ Alcohol interacts with corticotropin-releasing factor and several neurotransmitters,¹⁸ and specifically affects neuropathways involving norepinephrine, GABA, and glutamate.¹⁹ Alcohol has reinforcing effects mediated by the release of dopamine in the nucleus accumbens.²⁰

Acutely, alcohol promotes GABA release and may also reduce GABA degradation, producing sedative and anxiolytic effects.²¹ Chronic alcohol use leads to a decrease in the number of GABAA receptors. Clinically, this downregulation manifests as tolerance to alcohol’s sedating effects.²¹

Alcohol affects the signaling of glutamatergic interaction with the N-methyl-d-aspartate (NMDA) receptor.²² Glutamate activates this receptor as well as the voltage-gated ion channels, modifying calcium influx and increasing neuronal excitability.^22,23 Acutely, alcohol has an antagonistic effect on the NMDA receptor, while chronic drinking upregulates (increases) the number of NMDA receptors and voltage-gated calcium channels.^22,23

Alcohol withdrawal increases excitatory effects

Patients experiencing alcohol withdrawal have decreased GABA-ergic functioning and increased glutamatergic action throughout the central nervous system.^19,24

Withdrawal can be subdivided into an acute phase (lasting up to about 5 days) and a protracted phase (of undetermined duration). During withdrawal, the brain activates its “stress system,” leading to overexpression of corticotropin-releasing factor in the amygdala. Protracted withdrawal dysregulates the prefrontal cortex, increasing cravings and worsening negative emotional states and sleep.¹⁶

GABAPENTIN FOR ALCOHOL WITHDRAWAL

Benzodiazepines are the standard treatment for alcohol withdrawal.^3,24 They relieve symptoms and can prevent seizures and delirium tremens,²⁴ but they are sedating and cause psychomotor impairments.³ Because of the potential for addiction, benzodiazepine use is limited to acute alcohol withdrawal.³

Gabapentin shows promise as an agent that can be used in withdrawal and continued through early abstinence without the highly addictive potential of benzodiazepines.¹⁶ It is thought to affect drinking behaviors during early abstinence by normalizing GABA and glutamate activity.^2,16

Early preclinical studies in mouse models found that gabapentin decreases anxiogenic and epileptic effects of alcohol withdrawal. Compared with other antidrinking medications, gabapentin has the benefits of lacking elimination via hepatic metabolism, few pharmacokinetic interactions, and good reported tolerability in this population.

Inpatient trials show no benefit over standard treatments

Bonnet et al²⁵ conducted a double-blind placebo-controlled trial in Germany in inpatients experiencing acute alcohol withdrawal to determine whether gabapentin might be an effective adjunct to clomethiazole, a GABAA modulator commonly used in Europe for alcohol withdrawal. Participants (N = 61) were randomized to receive placebo or gabapentin (400 mg every 6 hours) for 72 hours, with tapering over the next 3 days. All patients could receive rescue doses of clomethiazole, using a symptom-triggered protocol.

The study revealed no differences in the amount of clomethiazole administered between the 2 groups, suggesting that gabapentin had no adjunctive effect. Side effects (vertigo, nausea, dizziness, and ataxia) were mild and comparable between groups.

Nichols et al²⁶ conducted a retrospective cohort study in a South Carolina academic psychiatric hospital to assess the adjunctive effect of gabapentin on the as-needed use of benzodiazepines for alcohol withdrawal. The active group (n = 40) received gabapentin as well as a symptom-triggered alcohol withdrawal protocol of benzodiazepine. The control group (n = 43) received only the symptom-triggered alcohol withdrawal protocol without gabapentin.

No effect was found of gabapentin use for benzodiazepine treatment of alcohol withdrawal. It is notable that Bonnet et al and Nichols et al had similar findings despite their studies being conducted in different countries using distinct comparators and methods.

Bonnet et al,²⁷ in another study, tried a different design to investigate a possible role for gabapentin in inpatient alcohol withdrawal. The study included 37 patients with severe alcohol withdrawal (Clinical Institute Withdrawal Assessment of Alcohol Scale, Revised [CIWA-Ar] > 15).

All participants received gabapentin 800 mg. Those whose CIWA-Ar score improved within 2 hours were considered “early responders” (n = 27) and next received 2 days of gabapentin 600 mg 4 times a day before starting a taper. The nonresponders whose CIWA-Ar score worsened (associated with greater anxiety and depressive symptoms; n = 10) were switched to standard treatment with clomethiazole (n = 4) or clonazepam (n = 6). Scores of 3 early responders subsequently worsened; 2 of these participants developed seizures and were switched to standard treatment.

The authors concluded that gabapentin in a dose of 3,200 mg in the first 24 hours is useful only for milder forms of alcohol withdrawal. Hence, subsequent efforts on the use of gabapentin for alcohol withdrawal have focused on outpatients.

Outpatient trials reveal benefits over benzodiazepines

Myrick et al³ compared gabapentin vs lorazepam in 100 outpatients seeking treatment for alcohol withdrawal. Participants were randomized to 1 of 4 groups: gabapentin 600 mg, 900 mg, or 1,200 mg, or lorazepam 6 mg, each tapering over 4 days. Alcohol withdrawal was measured by the CIWA-Ar score. Only 68 patients completed all follow-up appointments to day 12.

Gabapentin 600 mg was discontinued because of seizures in 2 patients, but it was generally well tolerated and was associated with diminished symptoms of alcohol withdrawal, especially at the 1,200 mg dose. The gabapentin groups experienced less anxiety and sedation and fewer cravings than the lorazepam group. Those treated with lorazepam fared worse for achieving early abstinence and were more likely to return to drinking when the intervention was discontinued. However, significant relapse by day 12 occurred in both groups.

The authors concluded that gabapentin was at least as effective as lorazepam in the outpatient treatment of alcohol withdrawal, with the 1,200-mg gabapentin dosage being more effective than 900 mg. At 1,200 mg, gabapentin was associated with better sleep, less anxiety, and better self-reported ability to work than lorazepam, and at the 900-mg dose it was associated with less depression than lorazepam.

Stock et al²⁸ conducted a randomized, double-blind study of gabapentin in acute alcohol withdrawal in 26 military veterans in an outpatient setting. Patients were ran­domized to one of the following:

• Gabapentin 1,200 mg orally for 3 days, followed by 900 mg, 600 mg, and 300 mg for 1 day each (n = 17)
• Chlordiazepoxide 100 mg orally for 3 days, followed by 75 mg, 50 mg, and 25 mg for 1 day each (n = 9).

Withdrawal scores improved similarly in both groups. Early on (days 1–4), neither cravings nor sleep differed significantly between groups; but later (days 5–7), the gabapentin group had superior scores for these measures. Gabapentin was also associated with significantly less sedation than chlordiazepoxide and trended to less alcohol craving.

Data suggest that gabapentin offers benefits for managing mild alcohol withdrawal. Improved residual craving and sleep measures are clinically important because they are risk factors for relapse. Mood and anxiety also improve with gabapentin, further indicating a therapeutic effect.

Gabapentin’s benefits for moderate and severe alcohol withdrawal have not been established. Seizures occurred during withdrawal despite gabapentin treatment, but whether from an insufficient dose, patient susceptibility, or lack of gabapentin efficacy is not clear. Best results occurred at the 1,200-mg daily dose, but benefits may not apply to patients with severe withdrawal. In addition, many studies were small, limiting the strength of conclusions.

Across most studies of gabapentin for alcohol withdrawal, advantages included a smoother transition into early abstinence due to improved sleep, mood, and anxiety, alleviating common triggers for a return to drinking. Gabapentin also carries less reinforcing potential than benzodiazepines. These qualities fueled interest in trying gabapentin to improve long-term abstinence.

GABAPENTIN FOR RELAPSE PREVENTION

Although naltrexone and acamprosate are the first-line treatments for relapse prevention, they do not help all patients and are more effective when combined with cognitive behavioral therapy.^1,29,30 For patients in whom standard treatments are not effective or tolerated, gabapentin may provide a reasonable alternative, and several randomized controlled trials have examined its use for this role.

Gabapentin alone is better than placebo

Furieri and Nakamura-Palacios⁴ assessed the use of gabapentin for relapse prevention in Brazilian outpatients (N = 60) who had averaged 27 years of drinking and consumed 17 drinks daily for the 90 days before baseline. After detoxification with diazepam and vitamins, patients were randomized to either gabapentin 300 mg twice daily or placebo for 4 weeks.

Compared with placebo, gabapentin significantly reduced cravings and lowered the percentage of heavy drinking days and the number of drinks per day, with a significant increase in the percentage of abstinent days. These self-reported measures correlated with decreases in gamma-glutamyl transferase, a biological marker for heavy drinking.

Brower et al³¹ investigated the use of gabapentin in 21 outpatients with AUD and insomnia who desired to remain abstinent. They were randomized to gabapentin (up to 1,500 mg at night) or placebo for 6 weeks. Just 14 participants completed the study; all but 2 were followed without treatment until week 12.

Gabapentin was associated with significantly lower relapse rates at 6 weeks (3 of 10 in the gabapentin group vs 9 of 11 in the placebo group) and at 12 weeks (6 of 10 in the gabapentin group vs 11 of 11 in the placebo group, assuming the 2 patients lost to follow-up relapsed). No difference between groups was detected for sleep measures in this small study. However, other studies have found that gabapentin for AUD improves measures of insomnia and daytime drowsiness—predictors of relapse—compared with other medications.¹⁶

High-dose gabapentin is better

Mason et al² randomized 150 outpatients with alcohol dependence to 12 weeks of daily treatment with either gabapentin (900 mg or 1,800 mg) or placebo after at least 3 days of abstinence. All participants received counseling. Drinking quantity and frequency were assessed by gamma-glutamyl transferase testing.

Patients taking gabapentin had better rates of abstinence and cessation of heavy drinking than those taking placebo. During the 12-week study, the 1,800-mg daily dose showed a substantially higher abstinence rate (17%) than either 900 mg  (11%) or placebo (4%). Significant dose-related improvements were also found for heavy drinking days, total drinking quantity, and frequency of alcohol withdrawal symptoms that predispose to early relapse, such as poor sleep, cravings, and poor mood. There were also significant linear dose effects on rates of abstinence and nondrinking days at the 24-week posttreatment follow-up.

Gabapentin plus naltrexone is better than naltrexone alone

Anton et al⁵ examined the efficacy of gabapentin combined with naltrexone during early abstinence. The study randomly assigned 150 people with AUD to one of the following groups:

• 16 weeks of naltrexone (50 mg/day) alone
• 6 weeks of naltrexone (50 mg/day) plus gabapentin (up to 1,200 mg/day), followed by 10 weeks of naltrexone alone
• Placebo.

All participants received medical management.

Over the first 6 weeks, those receiving naltrexone plus gabapentin had a longer interval to heavy drinking than those taking only naltrexone. By week 6, about half of those taking placebo or naltrexone alone had a heavy drinking day, compared with about 35% of those taking naltrexone plus gabapentin. Those receiving the combination also had fewer days of heavy drinking, fewer drinks per drinking day, and better sleep than the other groups. Participants in the naltrexone-alone group were more likely to drink heavily during periods in which they reported poor sleep. No significant group differences were found in measures of mood.

Gabapentin enacarbil is no better than placebo

Falk et al,³² in a 2019 preliminary analysis, examined data from a trial of gabapentin enacarbil, a prodrug formulation of gabapentin. In this 6-month double-blind study, 346 people with moderate AUD at 10 sites were randomized to gabapentin enacarbil extended-release 600 mg twice a day or placebo. All subjects received a computerized behavioral intervention.

No significant differences between groups were found in drinking measures or alcohol cravings, sleep problems, depression, or anxiety symptoms. However, a dose-response analysis found significantly less drinking for higher doses of the drug.

Bottom line: Evidence of benefits mixed but risk low

The efficacy of gabapentin as a treatment for AUD has varied across studies as a function of dosing and formulation. Daily doses have ranged from 600 mg to 1,800 mg, with the highest dose showing advantages in one study for cravings, insomnia, anxiety, dysphoria, and relapse.² Thus far, gabapentin immediate-release has performed better than gabapentin enacarbil extended-release. All forms of gabapentin have been well-tolerated in AUD trials.

The 2018 American Psychiatric Association guidelines stated that gabapentin had a small positive effect on drinking outcomes, but the harm of treatment was deemed minimal, especially relative to the harms of chronic drinking.³³ The guidelines endorse the use of gabapentin in patients with moderate to severe AUD who select gabapentin from the available options, or for those who are nonresponsive to or cannot tolerate naltrexone or acamprosate, as long as no contraindications exist. It was also noted that even small effects may be clinically important, considering the significant morbidity associated with AUD.

The use of gabapentin has become controversial owing to the growing recognition that it may not be as benign as initially thought.^7–9,34 A review of US legislative actions reflects concerns about its misuse.³⁵ In July 2017, Kentucky classified it as a schedule V controlled substance with prescription drug monitoring,³⁵ as did Tennessee in 2018³⁶ and Michigan in January 2019.³⁷ Currently, 8 other states (Massachusetts, Minnesota, Nebraska, North Dakota, Ohio, Virginia, Wyoming, and West Virginia) require prescription drug monitoring of gabapentin, and other states are considering it.³⁵

Efforts to understand gabapentin misuse derive largely from people with drug use disorders. A review of postmortem toxicology reports in fatal drug overdoses found gabapentin present in 22%.³⁸ Although it was not necessarily a cause of death, its high rate of detection suggests wide misuse among drug users.

Among a cohort of 503 prescription opioid misusers in Appalachian Kentucky, 15% reported using gabapentin “to get high.” Those who reported misusing gabapentin were 6 times more likely than nonusers to be abusing opioids and benzodiazepines. The main sources of gabapentin were doctors (52%) and dealers (36%). The average cost of gabapentin on the street was less than $1.00 per pill.³⁹

Gabapentin misuse by methadone clinic patients is also reported. Baird et al⁴⁰ surveyed patients in 6 addiction clinics in the United Kingdom for gabapentin and pregabalin abuse and found that 22% disclosed misusing these medications. Of these, 38% said they did so to enhance the methadone high.

In a review article, Quintero⁴¹ also cited enhancement of methadone euphoria and treatment of opioid withdrawal as motivations for misuse. Opioid-dependent gabapentin misusers consumed doses of gabapentin 3 to 20 times higher than clinically recommended and in combination with multiple drugs.⁴ Such use can cause dissociative and psychedelic effects.

Gabapentin also potentiates the sedative effects of opioids, thus increasing the risk of falls, accidents, and other adverse events.^34,35 Risk of opioid-related deaths was increased with coprescription of gabapentin and with moderate to high gabapentin doses.³⁴

Are people with AUD at higher risk of gabapentin abuse?

Despite concerns, patients in clinical trials of gabapentin treatment for AUD were not identified as at high risk for misuse of the drug.^2,4,5,16 Further, no such trials reported serious drug-related adverse events resulting in gabapentin discontinuation or side effects that differed from placebo in frequency or severity.^2,4,5,16

Clinical laboratory studies also have found no significant interactions between alcohol and gabapentin.^42,43 In fact, they showed no influence of gabapentin on the pharmacokinetics of alcohol or on alcohol’s subjective effects. Relative to placebo, gabapentin did not affect blood alcohol levels, the degree of intoxication, sedation, craving, or alcohol self-administration.

Smith et al⁹ reported estimates that only 1% of the general population misuse gabapentin. Another review concluded that gabapentin is seldom a drug of choice.¹⁷ Most patients prescribed gabapentin do not experience cravings or loss of control, which are hallmarks of addiction. Hence, with adequate precautions, the off-label use of gabapentin for AUD is reasonable.

CLINICAL IMPLICATIONS OF GABAPENTIN PRESCRIBING

Overall, evidence for the benefit of gabapentin in AUD is mixed. Subgroups of alcoholic patients, such as those who do not respond to or tolerate standard therapies, may particularly benefit, as may those with comorbid insomnia or neuropathic pain.⁴⁴ Clinicians should prescribe gabapentin only when it is likely to be helpful and should carefully document its efficacy.^2,45

At each visit, an open and honest assessment of the benefits and risks serves to promote shared decision-making regarding initiating, continuing, or discontinuing gabapentin.

For alcohol withdrawal

Before gabapentin is prescribed for alcohol withdrawal, potential benefits (reduction of withdrawal symptoms), side effects (sedation, fatigue), and risks (falls) should be discussed with the patient.⁴⁶ Patients should also be informed that benzodiazepines are the gold standard for alcohol withdrawal and that gabapentin is not effective for severe withdrawal.⁴⁶

For relapse prevention

When initiating treatment for relapse prevention, the patient and the prescriber should agree on specific goals (eg, reduction of drinking, anxiety, and insomnia).^2,16 Ongoing monitoring is essential and includes assessing and documenting improvement with respect to these goals.

In the AUD studies, gabapentin was well tolerated.¹⁶ Frequently observed side effects including headache, insomnia, fatigue, muscle aches, and gastrointestinal distress did not occur at a statistically different rate from placebo. However, patients in studies are selected samples, and their experience may not be generalizable to clinical practice. Thus, it is necessary to exercise caution and check for comorbidities that may put patients at risk of complications.⁴⁷ Older patients and those on hemodialysis are more susceptible to gabapentin side effects such as sedation, dizziness, ataxia, and mental status changes,³⁴ and prescribers should be alert for signs of toxicity (eg, ataxia, mental status changes).^47,48

Gabapentin misuse was not observed in AUD studies,^2,4,5,16 but evidence indicates that patients with opioid use disorder, prisoners, and polydrug users are at high risk for gabapentin misuse.^39–41 In all cases, clinicians should monitor for red flags that may indicate abuse, such as missed appointments, early refill requests, demands for increased dosage, and simultaneous opiate and benzodiazepine use.⁴⁹

Acknowledgment: The authors wish to thank Nick Mulligan for his invaluable assistance with formatting and grammar.

Perceptions regarding the use of gabapentin for alcohol use disorder (AUD) have shifted over time.^1–4 Early on, the drug was deemed to be benign and effective.^4–6 But more and more, concerns are being raised about its recreational use to achieve euphoria,⁷ and the drug is often misused by vulnerable populations, particularly those with opioid use disorder.^7–9

Given the large number of gabapentin prescriptions written off-label for AUD, it is incumbent on providers to understand how to prescribe it responsibly.^7–9 To that end, this article focuses on the benefits—and concerns—of this treatment option. We describe the effects of gabapentin on the central nervous system and how it may mitigate alcohol withdrawal and increase the likelihood of abstinence. In addition, we review clinical trials that evaluated potential roles of gabapentin in AUD, discuss the drug’s misuse potential, and suggest a framework for its appropriate use in AUD management.

ALCOHOL USE DISORDER IS COMMON AND SERIOUS

AUD affects about 14% of US adults and represents a significant health burden,¹ often with severe clinical and social implications. It manifests as compulsive drinking and loss of control despite adverse consequences on various life domains.¹⁰ It is generally associated with cravings, tolerance, and withdrawal symptoms upon cessation. Alcohol withdrawal is characterized by tremors, anxiety, sweating, nausea, and tachycardia, and in severe cases, may involve hallucinations, seizures, and delirium tremens. Untreated, alcohol withdrawal can be fatal.¹⁰

Even though psychosocial treatments for AUD by themselves are associated with high relapse rates, pharmacotherapy is underutilized. Three drugs approved by the US Food and Drug Administration (FDA) are available to treat it, but they are often poorly accepted and have limited efficacy. For these reasons, there is considerable interest in finding alternatives. Gabapentin is one of several agents that have been studied (Table 1). The topic has been reviewed in depth by Soyka and Müller.¹¹

GABAPENTIN REDUCES EXCITATION

The anticonvulsant gabapentin is FDA-approved for treating epilepsy, postherpetic neuralgia, and restless leg syndrome.^8,12–14 It binds and selectively impedes voltage-sensitive calcium channels, the pores in cell membrane that permit calcium to enter a neuron in response to changes in electrical currents.¹⁵

Gabapentin is believed to decrease excitation of the central nervous system in multiple ways:

• It reduces the release of glutamate, a key component of the excitatory system¹⁶
• It increases the concentration of gamma-aminobutyric acid (GABA), the main inhibitory neurotransmitter in the brain7
• By binding the alpha-2-delta type 1 subunit of voltage-sensitive calcium channels,^8,15–17 it inhibits excitatory synapse formation independent of calcium channel activity¹⁶
• By blocking excitatory neurotransmission, it also may indirectly increase the concentration of GABA in the central nervous system1^6,17
• It modulates action of glutamic acid decarboxylase (involved in the synthesis of GABA) and glutamate synthesizing enzyme to increase GABA and decrease glutamate.¹⁷

The actions of alcohol on the brain are also complex.¹⁸ Alpha-2-delta type 1 subunits of calcium channels are upregulated in the reward centers of the brain by addictive substances, including alcohol.¹⁶ Alcohol interacts with corticotropin-releasing factor and several neurotransmitters,¹⁸ and specifically affects neuropathways involving norepinephrine, GABA, and glutamate.¹⁹ Alcohol has reinforcing effects mediated by the release of dopamine in the nucleus accumbens.²⁰

Acutely, alcohol promotes GABA release and may also reduce GABA degradation, producing sedative and anxiolytic effects.²¹ Chronic alcohol use leads to a decrease in the number of GABAA receptors. Clinically, this downregulation manifests as tolerance to alcohol’s sedating effects.²¹

Alcohol affects the signaling of glutamatergic interaction with the N-methyl-d-aspartate (NMDA) receptor.²² Glutamate activates this receptor as well as the voltage-gated ion channels, modifying calcium influx and increasing neuronal excitability.^22,23 Acutely, alcohol has an antagonistic effect on the NMDA receptor, while chronic drinking upregulates (increases) the number of NMDA receptors and voltage-gated calcium channels.^22,23

Alcohol withdrawal increases excitatory effects

Patients experiencing alcohol withdrawal have decreased GABA-ergic functioning and increased glutamatergic action throughout the central nervous system.^19,24

Withdrawal can be subdivided into an acute phase (lasting up to about 5 days) and a protracted phase (of undetermined duration). During withdrawal, the brain activates its “stress system,” leading to overexpression of corticotropin-releasing factor in the amygdala. Protracted withdrawal dysregulates the prefrontal cortex, increasing cravings and worsening negative emotional states and sleep.¹⁶

GABAPENTIN FOR ALCOHOL WITHDRAWAL

Benzodiazepines are the standard treatment for alcohol withdrawal.^3,24 They relieve symptoms and can prevent seizures and delirium tremens,²⁴ but they are sedating and cause psychomotor impairments.³ Because of the potential for addiction, benzodiazepine use is limited to acute alcohol withdrawal.³

Gabapentin shows promise as an agent that can be used in withdrawal and continued through early abstinence without the highly addictive potential of benzodiazepines.¹⁶ It is thought to affect drinking behaviors during early abstinence by normalizing GABA and glutamate activity.^2,16

Early preclinical studies in mouse models found that gabapentin decreases anxiogenic and epileptic effects of alcohol withdrawal. Compared with other antidrinking medications, gabapentin has the benefits of lacking elimination via hepatic metabolism, few pharmacokinetic interactions, and good reported tolerability in this population.

Inpatient trials show no benefit over standard treatments

Bonnet et al²⁵ conducted a double-blind placebo-controlled trial in Germany in inpatients experiencing acute alcohol withdrawal to determine whether gabapentin might be an effective adjunct to clomethiazole, a GABAA modulator commonly used in Europe for alcohol withdrawal. Participants (N = 61) were randomized to receive placebo or gabapentin (400 mg every 6 hours) for 72 hours, with tapering over the next 3 days. All patients could receive rescue doses of clomethiazole, using a symptom-triggered protocol.

The study revealed no differences in the amount of clomethiazole administered between the 2 groups, suggesting that gabapentin had no adjunctive effect. Side effects (vertigo, nausea, dizziness, and ataxia) were mild and comparable between groups.

Nichols et al²⁶ conducted a retrospective cohort study in a South Carolina academic psychiatric hospital to assess the adjunctive effect of gabapentin on the as-needed use of benzodiazepines for alcohol withdrawal. The active group (n = 40) received gabapentin as well as a symptom-triggered alcohol withdrawal protocol of benzodiazepine. The control group (n = 43) received only the symptom-triggered alcohol withdrawal protocol without gabapentin.

No effect was found of gabapentin use for benzodiazepine treatment of alcohol withdrawal. It is notable that Bonnet et al and Nichols et al had similar findings despite their studies being conducted in different countries using distinct comparators and methods.

Bonnet et al,²⁷ in another study, tried a different design to investigate a possible role for gabapentin in inpatient alcohol withdrawal. The study included 37 patients with severe alcohol withdrawal (Clinical Institute Withdrawal Assessment of Alcohol Scale, Revised [CIWA-Ar] > 15).

All participants received gabapentin 800 mg. Those whose CIWA-Ar score improved within 2 hours were considered “early responders” (n = 27) and next received 2 days of gabapentin 600 mg 4 times a day before starting a taper. The nonresponders whose CIWA-Ar score worsened (associated with greater anxiety and depressive symptoms; n = 10) were switched to standard treatment with clomethiazole (n = 4) or clonazepam (n = 6). Scores of 3 early responders subsequently worsened; 2 of these participants developed seizures and were switched to standard treatment.

The authors concluded that gabapentin in a dose of 3,200 mg in the first 24 hours is useful only for milder forms of alcohol withdrawal. Hence, subsequent efforts on the use of gabapentin for alcohol withdrawal have focused on outpatients.

Outpatient trials reveal benefits over benzodiazepines

Myrick et al³ compared gabapentin vs lorazepam in 100 outpatients seeking treatment for alcohol withdrawal. Participants were randomized to 1 of 4 groups: gabapentin 600 mg, 900 mg, or 1,200 mg, or lorazepam 6 mg, each tapering over 4 days. Alcohol withdrawal was measured by the CIWA-Ar score. Only 68 patients completed all follow-up appointments to day 12.

Gabapentin 600 mg was discontinued because of seizures in 2 patients, but it was generally well tolerated and was associated with diminished symptoms of alcohol withdrawal, especially at the 1,200 mg dose. The gabapentin groups experienced less anxiety and sedation and fewer cravings than the lorazepam group. Those treated with lorazepam fared worse for achieving early abstinence and were more likely to return to drinking when the intervention was discontinued. However, significant relapse by day 12 occurred in both groups.

The authors concluded that gabapentin was at least as effective as lorazepam in the outpatient treatment of alcohol withdrawal, with the 1,200-mg gabapentin dosage being more effective than 900 mg. At 1,200 mg, gabapentin was associated with better sleep, less anxiety, and better self-reported ability to work than lorazepam, and at the 900-mg dose it was associated with less depression than lorazepam.

Stock et al²⁸ conducted a randomized, double-blind study of gabapentin in acute alcohol withdrawal in 26 military veterans in an outpatient setting. Patients were ran­domized to one of the following:

• Gabapentin 1,200 mg orally for 3 days, followed by 900 mg, 600 mg, and 300 mg for 1 day each (n = 17)
• Chlordiazepoxide 100 mg orally for 3 days, followed by 75 mg, 50 mg, and 25 mg for 1 day each (n = 9).

Withdrawal scores improved similarly in both groups. Early on (days 1–4), neither cravings nor sleep differed significantly between groups; but later (days 5–7), the gabapentin group had superior scores for these measures. Gabapentin was also associated with significantly less sedation than chlordiazepoxide and trended to less alcohol craving.

Data suggest that gabapentin offers benefits for managing mild alcohol withdrawal. Improved residual craving and sleep measures are clinically important because they are risk factors for relapse. Mood and anxiety also improve with gabapentin, further indicating a therapeutic effect.

Gabapentin’s benefits for moderate and severe alcohol withdrawal have not been established. Seizures occurred during withdrawal despite gabapentin treatment, but whether from an insufficient dose, patient susceptibility, or lack of gabapentin efficacy is not clear. Best results occurred at the 1,200-mg daily dose, but benefits may not apply to patients with severe withdrawal. In addition, many studies were small, limiting the strength of conclusions.

Across most studies of gabapentin for alcohol withdrawal, advantages included a smoother transition into early abstinence due to improved sleep, mood, and anxiety, alleviating common triggers for a return to drinking. Gabapentin also carries less reinforcing potential than benzodiazepines. These qualities fueled interest in trying gabapentin to improve long-term abstinence.

GABAPENTIN FOR RELAPSE PREVENTION

Although naltrexone and acamprosate are the first-line treatments for relapse prevention, they do not help all patients and are more effective when combined with cognitive behavioral therapy.^1,29,30 For patients in whom standard treatments are not effective or tolerated, gabapentin may provide a reasonable alternative, and several randomized controlled trials have examined its use for this role.

Gabapentin alone is better than placebo

Furieri and Nakamura-Palacios⁴ assessed the use of gabapentin for relapse prevention in Brazilian outpatients (N = 60) who had averaged 27 years of drinking and consumed 17 drinks daily for the 90 days before baseline. After detoxification with diazepam and vitamins, patients were randomized to either gabapentin 300 mg twice daily or placebo for 4 weeks.

Compared with placebo, gabapentin significantly reduced cravings and lowered the percentage of heavy drinking days and the number of drinks per day, with a significant increase in the percentage of abstinent days. These self-reported measures correlated with decreases in gamma-glutamyl transferase, a biological marker for heavy drinking.

Brower et al³¹ investigated the use of gabapentin in 21 outpatients with AUD and insomnia who desired to remain abstinent. They were randomized to gabapentin (up to 1,500 mg at night) or placebo for 6 weeks. Just 14 participants completed the study; all but 2 were followed without treatment until week 12.

Gabapentin was associated with significantly lower relapse rates at 6 weeks (3 of 10 in the gabapentin group vs 9 of 11 in the placebo group) and at 12 weeks (6 of 10 in the gabapentin group vs 11 of 11 in the placebo group, assuming the 2 patients lost to follow-up relapsed). No difference between groups was detected for sleep measures in this small study. However, other studies have found that gabapentin for AUD improves measures of insomnia and daytime drowsiness—predictors of relapse—compared with other medications.¹⁶

High-dose gabapentin is better

Mason et al² randomized 150 outpatients with alcohol dependence to 12 weeks of daily treatment with either gabapentin (900 mg or 1,800 mg) or placebo after at least 3 days of abstinence. All participants received counseling. Drinking quantity and frequency were assessed by gamma-glutamyl transferase testing.

Patients taking gabapentin had better rates of abstinence and cessation of heavy drinking than those taking placebo. During the 12-week study, the 1,800-mg daily dose showed a substantially higher abstinence rate (17%) than either 900 mg  (11%) or placebo (4%). Significant dose-related improvements were also found for heavy drinking days, total drinking quantity, and frequency of alcohol withdrawal symptoms that predispose to early relapse, such as poor sleep, cravings, and poor mood. There were also significant linear dose effects on rates of abstinence and nondrinking days at the 24-week posttreatment follow-up.

Gabapentin plus naltrexone is better than naltrexone alone

Anton et al⁵ examined the efficacy of gabapentin combined with naltrexone during early abstinence. The study randomly assigned 150 people with AUD to one of the following groups:

• 16 weeks of naltrexone (50 mg/day) alone
• 6 weeks of naltrexone (50 mg/day) plus gabapentin (up to 1,200 mg/day), followed by 10 weeks of naltrexone alone
• Placebo.

All participants received medical management.

Over the first 6 weeks, those receiving naltrexone plus gabapentin had a longer interval to heavy drinking than those taking only naltrexone. By week 6, about half of those taking placebo or naltrexone alone had a heavy drinking day, compared with about 35% of those taking naltrexone plus gabapentin. Those receiving the combination also had fewer days of heavy drinking, fewer drinks per drinking day, and better sleep than the other groups. Participants in the naltrexone-alone group were more likely to drink heavily during periods in which they reported poor sleep. No significant group differences were found in measures of mood.

Gabapentin enacarbil is no better than placebo

Falk et al,³² in a 2019 preliminary analysis, examined data from a trial of gabapentin enacarbil, a prodrug formulation of gabapentin. In this 6-month double-blind study, 346 people with moderate AUD at 10 sites were randomized to gabapentin enacarbil extended-release 600 mg twice a day or placebo. All subjects received a computerized behavioral intervention.

No significant differences between groups were found in drinking measures or alcohol cravings, sleep problems, depression, or anxiety symptoms. However, a dose-response analysis found significantly less drinking for higher doses of the drug.

Bottom line: Evidence of benefits mixed but risk low

The efficacy of gabapentin as a treatment for AUD has varied across studies as a function of dosing and formulation. Daily doses have ranged from 600 mg to 1,800 mg, with the highest dose showing advantages in one study for cravings, insomnia, anxiety, dysphoria, and relapse.² Thus far, gabapentin immediate-release has performed better than gabapentin enacarbil extended-release. All forms of gabapentin have been well-tolerated in AUD trials.

The 2018 American Psychiatric Association guidelines stated that gabapentin had a small positive effect on drinking outcomes, but the harm of treatment was deemed minimal, especially relative to the harms of chronic drinking.³³ The guidelines endorse the use of gabapentin in patients with moderate to severe AUD who select gabapentin from the available options, or for those who are nonresponsive to or cannot tolerate naltrexone or acamprosate, as long as no contraindications exist. It was also noted that even small effects may be clinically important, considering the significant morbidity associated with AUD.

The use of gabapentin has become controversial owing to the growing recognition that it may not be as benign as initially thought.^7–9,34 A review of US legislative actions reflects concerns about its misuse.³⁵ In July 2017, Kentucky classified it as a schedule V controlled substance with prescription drug monitoring,³⁵ as did Tennessee in 2018³⁶ and Michigan in January 2019.³⁷ Currently, 8 other states (Massachusetts, Minnesota, Nebraska, North Dakota, Ohio, Virginia, Wyoming, and West Virginia) require prescription drug monitoring of gabapentin, and other states are considering it.³⁵

Efforts to understand gabapentin misuse derive largely from people with drug use disorders. A review of postmortem toxicology reports in fatal drug overdoses found gabapentin present in 22%.³⁸ Although it was not necessarily a cause of death, its high rate of detection suggests wide misuse among drug users.

Among a cohort of 503 prescription opioid misusers in Appalachian Kentucky, 15% reported using gabapentin “to get high.” Those who reported misusing gabapentin were 6 times more likely than nonusers to be abusing opioids and benzodiazepines. The main sources of gabapentin were doctors (52%) and dealers (36%). The average cost of gabapentin on the street was less than $1.00 per pill.³⁹

Gabapentin misuse by methadone clinic patients is also reported. Baird et al⁴⁰ surveyed patients in 6 addiction clinics in the United Kingdom for gabapentin and pregabalin abuse and found that 22% disclosed misusing these medications. Of these, 38% said they did so to enhance the methadone high.

In a review article, Quintero⁴¹ also cited enhancement of methadone euphoria and treatment of opioid withdrawal as motivations for misuse. Opioid-dependent gabapentin misusers consumed doses of gabapentin 3 to 20 times higher than clinically recommended and in combination with multiple drugs.⁴ Such use can cause dissociative and psychedelic effects.

Gabapentin also potentiates the sedative effects of opioids, thus increasing the risk of falls, accidents, and other adverse events.^34,35 Risk of opioid-related deaths was increased with coprescription of gabapentin and with moderate to high gabapentin doses.³⁴

Are people with AUD at higher risk of gabapentin abuse?

Despite concerns, patients in clinical trials of gabapentin treatment for AUD were not identified as at high risk for misuse of the drug.^2,4,5,16 Further, no such trials reported serious drug-related adverse events resulting in gabapentin discontinuation or side effects that differed from placebo in frequency or severity.^2,4,5,16

Clinical laboratory studies also have found no significant interactions between alcohol and gabapentin.^42,43 In fact, they showed no influence of gabapentin on the pharmacokinetics of alcohol or on alcohol’s subjective effects. Relative to placebo, gabapentin did not affect blood alcohol levels, the degree of intoxication, sedation, craving, or alcohol self-administration.

Smith et al⁹ reported estimates that only 1% of the general population misuse gabapentin. Another review concluded that gabapentin is seldom a drug of choice.¹⁷ Most patients prescribed gabapentin do not experience cravings or loss of control, which are hallmarks of addiction. Hence, with adequate precautions, the off-label use of gabapentin for AUD is reasonable.

CLINICAL IMPLICATIONS OF GABAPENTIN PRESCRIBING

Overall, evidence for the benefit of gabapentin in AUD is mixed. Subgroups of alcoholic patients, such as those who do not respond to or tolerate standard therapies, may particularly benefit, as may those with comorbid insomnia or neuropathic pain.⁴⁴ Clinicians should prescribe gabapentin only when it is likely to be helpful and should carefully document its efficacy.^2,45

At each visit, an open and honest assessment of the benefits and risks serves to promote shared decision-making regarding initiating, continuing, or discontinuing gabapentin.

For alcohol withdrawal

Before gabapentin is prescribed for alcohol withdrawal, potential benefits (reduction of withdrawal symptoms), side effects (sedation, fatigue), and risks (falls) should be discussed with the patient.⁴⁶ Patients should also be informed that benzodiazepines are the gold standard for alcohol withdrawal and that gabapentin is not effective for severe withdrawal.⁴⁶

For relapse prevention

When initiating treatment for relapse prevention, the patient and the prescriber should agree on specific goals (eg, reduction of drinking, anxiety, and insomnia).^2,16 Ongoing monitoring is essential and includes assessing and documenting improvement with respect to these goals.

In the AUD studies, gabapentin was well tolerated.¹⁶ Frequently observed side effects including headache, insomnia, fatigue, muscle aches, and gastrointestinal distress did not occur at a statistically different rate from placebo. However, patients in studies are selected samples, and their experience may not be generalizable to clinical practice. Thus, it is necessary to exercise caution and check for comorbidities that may put patients at risk of complications.⁴⁷ Older patients and those on hemodialysis are more susceptible to gabapentin side effects such as sedation, dizziness, ataxia, and mental status changes,³⁴ and prescribers should be alert for signs of toxicity (eg, ataxia, mental status changes).^47,48

Gabapentin misuse was not observed in AUD studies,^2,4,5,16 but evidence indicates that patients with opioid use disorder, prisoners, and polydrug users are at high risk for gabapentin misuse.^39–41 In all cases, clinicians should monitor for red flags that may indicate abuse, such as missed appointments, early refill requests, demands for increased dosage, and simultaneous opiate and benzodiazepine use.⁴⁹

Acknowledgment: The authors wish to thank Nick Mulligan for his invaluable assistance with formatting and grammar.

Perceptions regarding the use of gabapentin for alcohol use disorder (AUD) have shifted over time.^1–4 Early on, the drug was deemed to be benign and effective.^4–6 But more and more, concerns are being raised about its recreational use to achieve euphoria,⁷ and the drug is often misused by vulnerable populations, particularly those with opioid use disorder.^7–9

Given the large number of gabapentin prescriptions written off-label for AUD, it is incumbent on providers to understand how to prescribe it responsibly.^7–9 To that end, this article focuses on the benefits—and concerns—of this treatment option. We describe the effects of gabapentin on the central nervous system and how it may mitigate alcohol withdrawal and increase the likelihood of abstinence. In addition, we review clinical trials that evaluated potential roles of gabapentin in AUD, discuss the drug’s misuse potential, and suggest a framework for its appropriate use in AUD management.

ALCOHOL USE DISORDER IS COMMON AND SERIOUS

AUD affects about 14% of US adults and represents a significant health burden,¹ often with severe clinical and social implications. It manifests as compulsive drinking and loss of control despite adverse consequences on various life domains.¹⁰ It is generally associated with cravings, tolerance, and withdrawal symptoms upon cessation. Alcohol withdrawal is characterized by tremors, anxiety, sweating, nausea, and tachycardia, and in severe cases, may involve hallucinations, seizures, and delirium tremens. Untreated, alcohol withdrawal can be fatal.¹⁰

Even though psychosocial treatments for AUD by themselves are associated with high relapse rates, pharmacotherapy is underutilized. Three drugs approved by the US Food and Drug Administration (FDA) are available to treat it, but they are often poorly accepted and have limited efficacy. For these reasons, there is considerable interest in finding alternatives. Gabapentin is one of several agents that have been studied (Table 1). The topic has been reviewed in depth by Soyka and Müller.¹¹

GABAPENTIN REDUCES EXCITATION

The anticonvulsant gabapentin is FDA-approved for treating epilepsy, postherpetic neuralgia, and restless leg syndrome.^8,12–14 It binds and selectively impedes voltage-sensitive calcium channels, the pores in cell membrane that permit calcium to enter a neuron in response to changes in electrical currents.¹⁵

Gabapentin is believed to decrease excitation of the central nervous system in multiple ways:

• It reduces the release of glutamate, a key component of the excitatory system¹⁶
• It increases the concentration of gamma-aminobutyric acid (GABA), the main inhibitory neurotransmitter in the brain7
• By binding the alpha-2-delta type 1 subunit of voltage-sensitive calcium channels,^8,15–17 it inhibits excitatory synapse formation independent of calcium channel activity¹⁶
• By blocking excitatory neurotransmission, it also may indirectly increase the concentration of GABA in the central nervous system1^6,17
• It modulates action of glutamic acid decarboxylase (involved in the synthesis of GABA) and glutamate synthesizing enzyme to increase GABA and decrease glutamate.¹⁷

The actions of alcohol on the brain are also complex.¹⁸ Alpha-2-delta type 1 subunits of calcium channels are upregulated in the reward centers of the brain by addictive substances, including alcohol.¹⁶ Alcohol interacts with corticotropin-releasing factor and several neurotransmitters,¹⁸ and specifically affects neuropathways involving norepinephrine, GABA, and glutamate.¹⁹ Alcohol has reinforcing effects mediated by the release of dopamine in the nucleus accumbens.²⁰

Acutely, alcohol promotes GABA release and may also reduce GABA degradation, producing sedative and anxiolytic effects.²¹ Chronic alcohol use leads to a decrease in the number of GABAA receptors. Clinically, this downregulation manifests as tolerance to alcohol’s sedating effects.²¹

Alcohol affects the signaling of glutamatergic interaction with the N-methyl-d-aspartate (NMDA) receptor.²² Glutamate activates this receptor as well as the voltage-gated ion channels, modifying calcium influx and increasing neuronal excitability.^22,23 Acutely, alcohol has an antagonistic effect on the NMDA receptor, while chronic drinking upregulates (increases) the number of NMDA receptors and voltage-gated calcium channels.^22,23

Alcohol withdrawal increases excitatory effects

Patients experiencing alcohol withdrawal have decreased GABA-ergic functioning and increased glutamatergic action throughout the central nervous system.^19,24

Withdrawal can be subdivided into an acute phase (lasting up to about 5 days) and a protracted phase (of undetermined duration). During withdrawal, the brain activates its “stress system,” leading to overexpression of corticotropin-releasing factor in the amygdala. Protracted withdrawal dysregulates the prefrontal cortex, increasing cravings and worsening negative emotional states and sleep.¹⁶

GABAPENTIN FOR ALCOHOL WITHDRAWAL

Benzodiazepines are the standard treatment for alcohol withdrawal.^3,24 They relieve symptoms and can prevent seizures and delirium tremens,²⁴ but they are sedating and cause psychomotor impairments.³ Because of the potential for addiction, benzodiazepine use is limited to acute alcohol withdrawal.³

Gabapentin shows promise as an agent that can be used in withdrawal and continued through early abstinence without the highly addictive potential of benzodiazepines.¹⁶ It is thought to affect drinking behaviors during early abstinence by normalizing GABA and glutamate activity.^2,16

Early preclinical studies in mouse models found that gabapentin decreases anxiogenic and epileptic effects of alcohol withdrawal. Compared with other antidrinking medications, gabapentin has the benefits of lacking elimination via hepatic metabolism, few pharmacokinetic interactions, and good reported tolerability in this population.

Inpatient trials show no benefit over standard treatments

Bonnet et al²⁵ conducted a double-blind placebo-controlled trial in Germany in inpatients experiencing acute alcohol withdrawal to determine whether gabapentin might be an effective adjunct to clomethiazole, a GABAA modulator commonly used in Europe for alcohol withdrawal. Participants (N = 61) were randomized to receive placebo or gabapentin (400 mg every 6 hours) for 72 hours, with tapering over the next 3 days. All patients could receive rescue doses of clomethiazole, using a symptom-triggered protocol.

The study revealed no differences in the amount of clomethiazole administered between the 2 groups, suggesting that gabapentin had no adjunctive effect. Side effects (vertigo, nausea, dizziness, and ataxia) were mild and comparable between groups.

Nichols et al²⁶ conducted a retrospective cohort study in a South Carolina academic psychiatric hospital to assess the adjunctive effect of gabapentin on the as-needed use of benzodiazepines for alcohol withdrawal. The active group (n = 40) received gabapentin as well as a symptom-triggered alcohol withdrawal protocol of benzodiazepine. The control group (n = 43) received only the symptom-triggered alcohol withdrawal protocol without gabapentin.

No effect was found of gabapentin use for benzodiazepine treatment of alcohol withdrawal. It is notable that Bonnet et al and Nichols et al had similar findings despite their studies being conducted in different countries using distinct comparators and methods.

Bonnet et al,²⁷ in another study, tried a different design to investigate a possible role for gabapentin in inpatient alcohol withdrawal. The study included 37 patients with severe alcohol withdrawal (Clinical Institute Withdrawal Assessment of Alcohol Scale, Revised [CIWA-Ar] > 15).

All participants received gabapentin 800 mg. Those whose CIWA-Ar score improved within 2 hours were considered “early responders” (n = 27) and next received 2 days of gabapentin 600 mg 4 times a day before starting a taper. The nonresponders whose CIWA-Ar score worsened (associated with greater anxiety and depressive symptoms; n = 10) were switched to standard treatment with clomethiazole (n = 4) or clonazepam (n = 6). Scores of 3 early responders subsequently worsened; 2 of these participants developed seizures and were switched to standard treatment.

The authors concluded that gabapentin in a dose of 3,200 mg in the first 24 hours is useful only for milder forms of alcohol withdrawal. Hence, subsequent efforts on the use of gabapentin for alcohol withdrawal have focused on outpatients.

Outpatient trials reveal benefits over benzodiazepines

Myrick et al³ compared gabapentin vs lorazepam in 100 outpatients seeking treatment for alcohol withdrawal. Participants were randomized to 1 of 4 groups: gabapentin 600 mg, 900 mg, or 1,200 mg, or lorazepam 6 mg, each tapering over 4 days. Alcohol withdrawal was measured by the CIWA-Ar score. Only 68 patients completed all follow-up appointments to day 12.

Gabapentin 600 mg was discontinued because of seizures in 2 patients, but it was generally well tolerated and was associated with diminished symptoms of alcohol withdrawal, especially at the 1,200 mg dose. The gabapentin groups experienced less anxiety and sedation and fewer cravings than the lorazepam group. Those treated with lorazepam fared worse for achieving early abstinence and were more likely to return to drinking when the intervention was discontinued. However, significant relapse by day 12 occurred in both groups.

The authors concluded that gabapentin was at least as effective as lorazepam in the outpatient treatment of alcohol withdrawal, with the 1,200-mg gabapentin dosage being more effective than 900 mg. At 1,200 mg, gabapentin was associated with better sleep, less anxiety, and better self-reported ability to work than lorazepam, and at the 900-mg dose it was associated with less depression than lorazepam.

Stock et al²⁸ conducted a randomized, double-blind study of gabapentin in acute alcohol withdrawal in 26 military veterans in an outpatient setting. Patients were ran­domized to one of the following:

• Gabapentin 1,200 mg orally for 3 days, followed by 900 mg, 600 mg, and 300 mg for 1 day each (n = 17)
• Chlordiazepoxide 100 mg orally for 3 days, followed by 75 mg, 50 mg, and 25 mg for 1 day each (n = 9).

Withdrawal scores improved similarly in both groups. Early on (days 1–4), neither cravings nor sleep differed significantly between groups; but later (days 5–7), the gabapentin group had superior scores for these measures. Gabapentin was also associated with significantly less sedation than chlordiazepoxide and trended to less alcohol craving.

Data suggest that gabapentin offers benefits for managing mild alcohol withdrawal. Improved residual craving and sleep measures are clinically important because they are risk factors for relapse. Mood and anxiety also improve with gabapentin, further indicating a therapeutic effect.

Gabapentin’s benefits for moderate and severe alcohol withdrawal have not been established. Seizures occurred during withdrawal despite gabapentin treatment, but whether from an insufficient dose, patient susceptibility, or lack of gabapentin efficacy is not clear. Best results occurred at the 1,200-mg daily dose, but benefits may not apply to patients with severe withdrawal. In addition, many studies were small, limiting the strength of conclusions.

Across most studies of gabapentin for alcohol withdrawal, advantages included a smoother transition into early abstinence due to improved sleep, mood, and anxiety, alleviating common triggers for a return to drinking. Gabapentin also carries less reinforcing potential than benzodiazepines. These qualities fueled interest in trying gabapentin to improve long-term abstinence.

GABAPENTIN FOR RELAPSE PREVENTION

Although naltrexone and acamprosate are the first-line treatments for relapse prevention, they do not help all patients and are more effective when combined with cognitive behavioral therapy.^1,29,30 For patients in whom standard treatments are not effective or tolerated, gabapentin may provide a reasonable alternative, and several randomized controlled trials have examined its use for this role.

Gabapentin alone is better than placebo

Furieri and Nakamura-Palacios⁴ assessed the use of gabapentin for relapse prevention in Brazilian outpatients (N = 60) who had averaged 27 years of drinking and consumed 17 drinks daily for the 90 days before baseline. After detoxification with diazepam and vitamins, patients were randomized to either gabapentin 300 mg twice daily or placebo for 4 weeks.

Compared with placebo, gabapentin significantly reduced cravings and lowered the percentage of heavy drinking days and the number of drinks per day, with a significant increase in the percentage of abstinent days. These self-reported measures correlated with decreases in gamma-glutamyl transferase, a biological marker for heavy drinking.

Brower et al³¹ investigated the use of gabapentin in 21 outpatients with AUD and insomnia who desired to remain abstinent. They were randomized to gabapentin (up to 1,500 mg at night) or placebo for 6 weeks. Just 14 participants completed the study; all but 2 were followed without treatment until week 12.

Gabapentin was associated with significantly lower relapse rates at 6 weeks (3 of 10 in the gabapentin group vs 9 of 11 in the placebo group) and at 12 weeks (6 of 10 in the gabapentin group vs 11 of 11 in the placebo group, assuming the 2 patients lost to follow-up relapsed). No difference between groups was detected for sleep measures in this small study. However, other studies have found that gabapentin for AUD improves measures of insomnia and daytime drowsiness—predictors of relapse—compared with other medications.¹⁶

High-dose gabapentin is better

Mason et al² randomized 150 outpatients with alcohol dependence to 12 weeks of daily treatment with either gabapentin (900 mg or 1,800 mg) or placebo after at least 3 days of abstinence. All participants received counseling. Drinking quantity and frequency were assessed by gamma-glutamyl transferase testing.

Patients taking gabapentin had better rates of abstinence and cessation of heavy drinking than those taking placebo. During the 12-week study, the 1,800-mg daily dose showed a substantially higher abstinence rate (17%) than either 900 mg  (11%) or placebo (4%). Significant dose-related improvements were also found for heavy drinking days, total drinking quantity, and frequency of alcohol withdrawal symptoms that predispose to early relapse, such as poor sleep, cravings, and poor mood. There were also significant linear dose effects on rates of abstinence and nondrinking days at the 24-week posttreatment follow-up.

Gabapentin plus naltrexone is better than naltrexone alone

Anton et al⁵ examined the efficacy of gabapentin combined with naltrexone during early abstinence. The study randomly assigned 150 people with AUD to one of the following groups:

• 16 weeks of naltrexone (50 mg/day) alone
• 6 weeks of naltrexone (50 mg/day) plus gabapentin (up to 1,200 mg/day), followed by 10 weeks of naltrexone alone
• Placebo.

All participants received medical management.

Over the first 6 weeks, those receiving naltrexone plus gabapentin had a longer interval to heavy drinking than those taking only naltrexone. By week 6, about half of those taking placebo or naltrexone alone had a heavy drinking day, compared with about 35% of those taking naltrexone plus gabapentin. Those receiving the combination also had fewer days of heavy drinking, fewer drinks per drinking day, and better sleep than the other groups. Participants in the naltrexone-alone group were more likely to drink heavily during periods in which they reported poor sleep. No significant group differences were found in measures of mood.

Gabapentin enacarbil is no better than placebo

Falk et al,³² in a 2019 preliminary analysis, examined data from a trial of gabapentin enacarbil, a prodrug formulation of gabapentin. In this 6-month double-blind study, 346 people with moderate AUD at 10 sites were randomized to gabapentin enacarbil extended-release 600 mg twice a day or placebo. All subjects received a computerized behavioral intervention.

No significant differences between groups were found in drinking measures or alcohol cravings, sleep problems, depression, or anxiety symptoms. However, a dose-response analysis found significantly less drinking for higher doses of the drug.

Bottom line: Evidence of benefits mixed but risk low

The efficacy of gabapentin as a treatment for AUD has varied across studies as a function of dosing and formulation. Daily doses have ranged from 600 mg to 1,800 mg, with the highest dose showing advantages in one study for cravings, insomnia, anxiety, dysphoria, and relapse.² Thus far, gabapentin immediate-release has performed better than gabapentin enacarbil extended-release. All forms of gabapentin have been well-tolerated in AUD trials.

The 2018 American Psychiatric Association guidelines stated that gabapentin had a small positive effect on drinking outcomes, but the harm of treatment was deemed minimal, especially relative to the harms of chronic drinking.³³ The guidelines endorse the use of gabapentin in patients with moderate to severe AUD who select gabapentin from the available options, or for those who are nonresponsive to or cannot tolerate naltrexone or acamprosate, as long as no contraindications exist. It was also noted that even small effects may be clinically important, considering the significant morbidity associated with AUD.

The use of gabapentin has become controversial owing to the growing recognition that it may not be as benign as initially thought.^7–9,34 A review of US legislative actions reflects concerns about its misuse.³⁵ In July 2017, Kentucky classified it as a schedule V controlled substance with prescription drug monitoring,³⁵ as did Tennessee in 2018³⁶ and Michigan in January 2019.³⁷ Currently, 8 other states (Massachusetts, Minnesota, Nebraska, North Dakota, Ohio, Virginia, Wyoming, and West Virginia) require prescription drug monitoring of gabapentin, and other states are considering it.³⁵

Efforts to understand gabapentin misuse derive largely from people with drug use disorders. A review of postmortem toxicology reports in fatal drug overdoses found gabapentin present in 22%.³⁸ Although it was not necessarily a cause of death, its high rate of detection suggests wide misuse among drug users.

Among a cohort of 503 prescription opioid misusers in Appalachian Kentucky, 15% reported using gabapentin “to get high.” Those who reported misusing gabapentin were 6 times more likely than nonusers to be abusing opioids and benzodiazepines. The main sources of gabapentin were doctors (52%) and dealers (36%). The average cost of gabapentin on the street was less than $1.00 per pill.³⁹

Gabapentin misuse by methadone clinic patients is also reported. Baird et al⁴⁰ surveyed patients in 6 addiction clinics in the United Kingdom for gabapentin and pregabalin abuse and found that 22% disclosed misusing these medications. Of these, 38% said they did so to enhance the methadone high.

In a review article, Quintero⁴¹ also cited enhancement of methadone euphoria and treatment of opioid withdrawal as motivations for misuse. Opioid-dependent gabapentin misusers consumed doses of gabapentin 3 to 20 times higher than clinically recommended and in combination with multiple drugs.⁴ Such use can cause dissociative and psychedelic effects.

Gabapentin also potentiates the sedative effects of opioids, thus increasing the risk of falls, accidents, and other adverse events.^34,35 Risk of opioid-related deaths was increased with coprescription of gabapentin and with moderate to high gabapentin doses.³⁴

Are people with AUD at higher risk of gabapentin abuse?

Despite concerns, patients in clinical trials of gabapentin treatment for AUD were not identified as at high risk for misuse of the drug.^2,4,5,16 Further, no such trials reported serious drug-related adverse events resulting in gabapentin discontinuation or side effects that differed from placebo in frequency or severity.^2,4,5,16

Clinical laboratory studies also have found no significant interactions between alcohol and gabapentin.^42,43 In fact, they showed no influence of gabapentin on the pharmacokinetics of alcohol or on alcohol’s subjective effects. Relative to placebo, gabapentin did not affect blood alcohol levels, the degree of intoxication, sedation, craving, or alcohol self-administration.

Smith et al⁹ reported estimates that only 1% of the general population misuse gabapentin. Another review concluded that gabapentin is seldom a drug of choice.¹⁷ Most patients prescribed gabapentin do not experience cravings or loss of control, which are hallmarks of addiction. Hence, with adequate precautions, the off-label use of gabapentin for AUD is reasonable.

CLINICAL IMPLICATIONS OF GABAPENTIN PRESCRIBING

Overall, evidence for the benefit of gabapentin in AUD is mixed. Subgroups of alcoholic patients, such as those who do not respond to or tolerate standard therapies, may particularly benefit, as may those with comorbid insomnia or neuropathic pain.⁴⁴ Clinicians should prescribe gabapentin only when it is likely to be helpful and should carefully document its efficacy.^2,45

At each visit, an open and honest assessment of the benefits and risks serves to promote shared decision-making regarding initiating, continuing, or discontinuing gabapentin.

For alcohol withdrawal

Before gabapentin is prescribed for alcohol withdrawal, potential benefits (reduction of withdrawal symptoms), side effects (sedation, fatigue), and risks (falls) should be discussed with the patient.⁴⁶ Patients should also be informed that benzodiazepines are the gold standard for alcohol withdrawal and that gabapentin is not effective for severe withdrawal.⁴⁶

For relapse prevention

When initiating treatment for relapse prevention, the patient and the prescriber should agree on specific goals (eg, reduction of drinking, anxiety, and insomnia).^2,16 Ongoing monitoring is essential and includes assessing and documenting improvement with respect to these goals.

In the AUD studies, gabapentin was well tolerated.¹⁶ Frequently observed side effects including headache, insomnia, fatigue, muscle aches, and gastrointestinal distress did not occur at a statistically different rate from placebo. However, patients in studies are selected samples, and their experience may not be generalizable to clinical practice. Thus, it is necessary to exercise caution and check for comorbidities that may put patients at risk of complications.⁴⁷ Older patients and those on hemodialysis are more susceptible to gabapentin side effects such as sedation, dizziness, ataxia, and mental status changes,³⁴ and prescribers should be alert for signs of toxicity (eg, ataxia, mental status changes).^47,48

Gabapentin misuse was not observed in AUD studies,^2,4,5,16 but evidence indicates that patients with opioid use disorder, prisoners, and polydrug users are at high risk for gabapentin misuse.^39–41 In all cases, clinicians should monitor for red flags that may indicate abuse, such as missed appointments, early refill requests, demands for increased dosage, and simultaneous opiate and benzodiazepine use.⁴⁹

Acknowledgment: The authors wish to thank Nick Mulligan for his invaluable assistance with formatting and grammar.

In no way do I minimize the value of the imprimatur of FDA approval stating a drug, after appropriate preclinical and clinical studies, is deemed safe and effective. Whatever the agency’s shortcomings, the story of thalidomide (a drug never approved by the FDA) gives credence to the value of having a robust approval process. Arguments will likely continue forever as to whether the agency errs on the side of being too permissive or too restrictive in its approval process.

Nonetheless, I believe there are valid clinical reasons why we should continue to prescribe FDA-approved medications for nonapproved indications. In my practice, I treat some conditions that are sufficiently uncommon or heterogeneous in expression that large-scale clinical trials are logistically hard to carry out or deemed financially unviable by the corporate sponsor, even though clinical experience has informed us of a reasonable likelihood of efficacy. Sometimes drugs have “failed” in clinical trials, but experience and post hoc subset analysis of data have indicated a likely positive response in certain patients.

Although a drug that has been FDA approved has passed significant safety testing, the patients exposed to the drug when it was evaluated for treating a certain disease may be strikingly different from patients who have a different disease—the age, sex, comorbidities, and coprescribed medications may all differ significantly in the population of patients with the “off-label” disorder. Hence, appropriate caution is warranted, and if relevant, this should be explained to patients before giving them the medication.

In this issue of the Journal, 2 papers address the use of medications in an “off-label” manner. Schneider and colleagues discuss several frequent clinical uses of tricyclic antidepressants for reasons other than depression, and Modesto-Lowe and colleagues review the more controversial use of gabapentin in patients with alcohol use disorder. The hoped-for benefits in both circumstances are symptomatic, and both benefits and side effects are dose-related in ways not necessarily coinciding with those in the FDA-labeled indications.

My experience in using tricyclics as adjunctive treatment for fibromyalgia is that patients are quite sensitive to some of the side effects of the drugs (eg, oral dryness and fatigue), even in low doses. Moreover, we should expect only modest benefits, which should be explicitly described to the patient: improved quality of sleep with resultant decreased fatigue (while we watch closely for worsened fatigue from too-high dosing) and a modest reduction in pain over time as part of a multimodality treatment plan. I often find that practitioners who are less familiar with the use of these medications in this setting tend to start at lowish (but higher than often tolerated) doses, have patients take the medication too close to bedtime (resulting in some morning hangover sensation), fail to discuss the timing and degree of expected pain relief, don’t titrate the dose over time, and are not aware of the different responses that patients may experience with different medications within the same class. As with all prescribed medications, the benefits and ill effects must be frequently assessed, and particularly with these medications, one must be willing to discontinue them if appropriate outcomes are not achieved.

“Off-label” should not imply off the table as a therapeutic option. But it is incumbent on us to devote sufficient time to explain to each patient the anticipated side effects and hoped-for benefits, particularly since in most cases, we and our patients cannot refer to the results of definitive phase 3 clinical trials or patient online information sites that are totally relevant, reliable, data supported, and FDA reviewed.

In no way do I minimize the value of the imprimatur of FDA approval stating a drug, after appropriate preclinical and clinical studies, is deemed safe and effective. Whatever the agency’s shortcomings, the story of thalidomide (a drug never approved by the FDA) gives credence to the value of having a robust approval process. Arguments will likely continue forever as to whether the agency errs on the side of being too permissive or too restrictive in its approval process.

Nonetheless, I believe there are valid clinical reasons why we should continue to prescribe FDA-approved medications for nonapproved indications. In my practice, I treat some conditions that are sufficiently uncommon or heterogeneous in expression that large-scale clinical trials are logistically hard to carry out or deemed financially unviable by the corporate sponsor, even though clinical experience has informed us of a reasonable likelihood of efficacy. Sometimes drugs have “failed” in clinical trials, but experience and post hoc subset analysis of data have indicated a likely positive response in certain patients.

Although a drug that has been FDA approved has passed significant safety testing, the patients exposed to the drug when it was evaluated for treating a certain disease may be strikingly different from patients who have a different disease—the age, sex, comorbidities, and coprescribed medications may all differ significantly in the population of patients with the “off-label” disorder. Hence, appropriate caution is warranted, and if relevant, this should be explained to patients before giving them the medication.

In this issue of the Journal, 2 papers address the use of medications in an “off-label” manner. Schneider and colleagues discuss several frequent clinical uses of tricyclic antidepressants for reasons other than depression, and Modesto-Lowe and colleagues review the more controversial use of gabapentin in patients with alcohol use disorder. The hoped-for benefits in both circumstances are symptomatic, and both benefits and side effects are dose-related in ways not necessarily coinciding with those in the FDA-labeled indications.

My experience in using tricyclics as adjunctive treatment for fibromyalgia is that patients are quite sensitive to some of the side effects of the drugs (eg, oral dryness and fatigue), even in low doses. Moreover, we should expect only modest benefits, which should be explicitly described to the patient: improved quality of sleep with resultant decreased fatigue (while we watch closely for worsened fatigue from too-high dosing) and a modest reduction in pain over time as part of a multimodality treatment plan. I often find that practitioners who are less familiar with the use of these medications in this setting tend to start at lowish (but higher than often tolerated) doses, have patients take the medication too close to bedtime (resulting in some morning hangover sensation), fail to discuss the timing and degree of expected pain relief, don’t titrate the dose over time, and are not aware of the different responses that patients may experience with different medications within the same class. As with all prescribed medications, the benefits and ill effects must be frequently assessed, and particularly with these medications, one must be willing to discontinue them if appropriate outcomes are not achieved.

“Off-label” should not imply off the table as a therapeutic option. But it is incumbent on us to devote sufficient time to explain to each patient the anticipated side effects and hoped-for benefits, particularly since in most cases, we and our patients cannot refer to the results of definitive phase 3 clinical trials or patient online information sites that are totally relevant, reliable, data supported, and FDA reviewed.

In no way do I minimize the value of the imprimatur of FDA approval stating a drug, after appropriate preclinical and clinical studies, is deemed safe and effective. Whatever the agency’s shortcomings, the story of thalidomide (a drug never approved by the FDA) gives credence to the value of having a robust approval process. Arguments will likely continue forever as to whether the agency errs on the side of being too permissive or too restrictive in its approval process.

Nonetheless, I believe there are valid clinical reasons why we should continue to prescribe FDA-approved medications for nonapproved indications. In my practice, I treat some conditions that are sufficiently uncommon or heterogeneous in expression that large-scale clinical trials are logistically hard to carry out or deemed financially unviable by the corporate sponsor, even though clinical experience has informed us of a reasonable likelihood of efficacy. Sometimes drugs have “failed” in clinical trials, but experience and post hoc subset analysis of data have indicated a likely positive response in certain patients.

Although a drug that has been FDA approved has passed significant safety testing, the patients exposed to the drug when it was evaluated for treating a certain disease may be strikingly different from patients who have a different disease—the age, sex, comorbidities, and coprescribed medications may all differ significantly in the population of patients with the “off-label” disorder. Hence, appropriate caution is warranted, and if relevant, this should be explained to patients before giving them the medication.

In this issue of the Journal, 2 papers address the use of medications in an “off-label” manner. Schneider and colleagues discuss several frequent clinical uses of tricyclic antidepressants for reasons other than depression, and Modesto-Lowe and colleagues review the more controversial use of gabapentin in patients with alcohol use disorder. The hoped-for benefits in both circumstances are symptomatic, and both benefits and side effects are dose-related in ways not necessarily coinciding with those in the FDA-labeled indications.

My experience in using tricyclics as adjunctive treatment for fibromyalgia is that patients are quite sensitive to some of the side effects of the drugs (eg, oral dryness and fatigue), even in low doses. Moreover, we should expect only modest benefits, which should be explicitly described to the patient: improved quality of sleep with resultant decreased fatigue (while we watch closely for worsened fatigue from too-high dosing) and a modest reduction in pain over time as part of a multimodality treatment plan. I often find that practitioners who are less familiar with the use of these medications in this setting tend to start at lowish (but higher than often tolerated) doses, have patients take the medication too close to bedtime (resulting in some morning hangover sensation), fail to discuss the timing and degree of expected pain relief, don’t titrate the dose over time, and are not aware of the different responses that patients may experience with different medications within the same class. As with all prescribed medications, the benefits and ill effects must be frequently assessed, and particularly with these medications, one must be willing to discontinue them if appropriate outcomes are not achieved.

“Off-label” should not imply off the table as a therapeutic option. But it is incumbent on us to devote sufficient time to explain to each patient the anticipated side effects and hoped-for benefits, particularly since in most cases, we and our patients cannot refer to the results of definitive phase 3 clinical trials or patient online information sites that are totally relevant, reliable, data supported, and FDA reviewed.

A 66-year-old man presented with burning pain and erythema over the left axilla, and pustules that had ruptured and crusted over. The rash also involved the right axilla, trunk, abdomen, and face.

He said the symptoms had developed 3 days after starting to use ciprofloxacin eye drops for eye redness and purulent discharge that had been diagnosed as bacterial conjunctivitis. He was taking no other new medications. He was afebrile.

The ciprofloxacin drops were stopped. The skin lesions were treated with emollients, topical steroids, and topical mupirocin. Improvement was noted 3 days into the hospitalization, as the lesions started to crust over and dry up and no new lesions were forming. The conjunctivitis improved with topical bacitracin ointment and prednisolone drops.

ACUTE GENERALIZED EXANTHEMATOUS PUSTULOSIS

The differential diagnosis of drug-related acute generalized exanthematous pustulosis includes Stevens-Johnson syndrome, pustular psoriasis, folliculitis, and varicella infection. The characteristic features of the rash and lesions and the temporal relationship between the start of ciprofloxacin eye drops and the development of symptoms, combined with rapid resolution of symptoms within days after discontinuing the drops, accompanied by skin biopsy study showing diffuse spongiosis with scattered eosinophils and a subcorneal pustule, confirmed the diagnosis of AGEP.

Key features of AGEP include numerous small, sterile, nonfollicular pustules on an erythematous background with associated fever and sometimes neutrophilia and eosinophilia.¹ It usually begins on the face or in the intertriginous areas and then spreads to the trunk and lower limbs with rare mucosal involvement.¹ It can be associated with viral infections, but most reported cases are related to drug reactions.²

Our patient’s case was unusual because AGEP triggered by topical medications is rarely reported, especially with ophthalmic medications.³ Drugs most commonly implicated are antibiotics including penicillins, sulfonamides, and quinolones, but other drugs such as terbinafine, diltiazem, and hydroxychloroquine have also been associated.²

AGEP may present with extensive skin desquamation, as in our patient, sometimes with bullae formation and skin sloughing manifesting as AGEP with overlapping toxic epidermal necrolysis.⁴

Diagnosis entails a careful review of medications, attention to lesion morphology, compatible disease course, and a high index of suspicion. Treatment is supportive and consists of stopping the offending agent, wound care, and antipyretics. Evidence for the use of steroids is weak.⁵

TAKE-HOME POINTS

AGEP should be considered in sudden-onset pustular desquamating erythematous rash related to use of a new medication. It is important to be aware that topical and ophthalmic medications are possible triggers. A thorough medication review should be done. Antibiotics are the most commonly implicated medications. An alternative medication should be tried. Treatment is supportive, as the condition is usually self-limiting once the offending medication is discontinued. Rarely, extensive desquamation and bullae formation may occur, which may be a manifestation of overlap features with toxic epidermal necrolysis.

A 66-year-old man presented with burning pain and erythema over the left axilla, and pustules that had ruptured and crusted over. The rash also involved the right axilla, trunk, abdomen, and face.

He said the symptoms had developed 3 days after starting to use ciprofloxacin eye drops for eye redness and purulent discharge that had been diagnosed as bacterial conjunctivitis. He was taking no other new medications. He was afebrile.

The ciprofloxacin drops were stopped. The skin lesions were treated with emollients, topical steroids, and topical mupirocin. Improvement was noted 3 days into the hospitalization, as the lesions started to crust over and dry up and no new lesions were forming. The conjunctivitis improved with topical bacitracin ointment and prednisolone drops.

ACUTE GENERALIZED EXANTHEMATOUS PUSTULOSIS

The differential diagnosis of drug-related acute generalized exanthematous pustulosis includes Stevens-Johnson syndrome, pustular psoriasis, folliculitis, and varicella infection. The characteristic features of the rash and lesions and the temporal relationship between the start of ciprofloxacin eye drops and the development of symptoms, combined with rapid resolution of symptoms within days after discontinuing the drops, accompanied by skin biopsy study showing diffuse spongiosis with scattered eosinophils and a subcorneal pustule, confirmed the diagnosis of AGEP.

Key features of AGEP include numerous small, sterile, nonfollicular pustules on an erythematous background with associated fever and sometimes neutrophilia and eosinophilia.¹ It usually begins on the face or in the intertriginous areas and then spreads to the trunk and lower limbs with rare mucosal involvement.¹ It can be associated with viral infections, but most reported cases are related to drug reactions.²

Our patient’s case was unusual because AGEP triggered by topical medications is rarely reported, especially with ophthalmic medications.³ Drugs most commonly implicated are antibiotics including penicillins, sulfonamides, and quinolones, but other drugs such as terbinafine, diltiazem, and hydroxychloroquine have also been associated.²

AGEP may present with extensive skin desquamation, as in our patient, sometimes with bullae formation and skin sloughing manifesting as AGEP with overlapping toxic epidermal necrolysis.⁴

Diagnosis entails a careful review of medications, attention to lesion morphology, compatible disease course, and a high index of suspicion. Treatment is supportive and consists of stopping the offending agent, wound care, and antipyretics. Evidence for the use of steroids is weak.⁵

TAKE-HOME POINTS

AGEP should be considered in sudden-onset pustular desquamating erythematous rash related to use of a new medication. It is important to be aware that topical and ophthalmic medications are possible triggers. A thorough medication review should be done. Antibiotics are the most commonly implicated medications. An alternative medication should be tried. Treatment is supportive, as the condition is usually self-limiting once the offending medication is discontinued. Rarely, extensive desquamation and bullae formation may occur, which may be a manifestation of overlap features with toxic epidermal necrolysis.

A 66-year-old man presented with burning pain and erythema over the left axilla, and pustules that had ruptured and crusted over. The rash also involved the right axilla, trunk, abdomen, and face.

He said the symptoms had developed 3 days after starting to use ciprofloxacin eye drops for eye redness and purulent discharge that had been diagnosed as bacterial conjunctivitis. He was taking no other new medications. He was afebrile.

The ciprofloxacin drops were stopped. The skin lesions were treated with emollients, topical steroids, and topical mupirocin. Improvement was noted 3 days into the hospitalization, as the lesions started to crust over and dry up and no new lesions were forming. The conjunctivitis improved with topical bacitracin ointment and prednisolone drops.

ACUTE GENERALIZED EXANTHEMATOUS PUSTULOSIS

The differential diagnosis of drug-related acute generalized exanthematous pustulosis includes Stevens-Johnson syndrome, pustular psoriasis, folliculitis, and varicella infection. The characteristic features of the rash and lesions and the temporal relationship between the start of ciprofloxacin eye drops and the development of symptoms, combined with rapid resolution of symptoms within days after discontinuing the drops, accompanied by skin biopsy study showing diffuse spongiosis with scattered eosinophils and a subcorneal pustule, confirmed the diagnosis of AGEP.

Key features of AGEP include numerous small, sterile, nonfollicular pustules on an erythematous background with associated fever and sometimes neutrophilia and eosinophilia.¹ It usually begins on the face or in the intertriginous areas and then spreads to the trunk and lower limbs with rare mucosal involvement.¹ It can be associated with viral infections, but most reported cases are related to drug reactions.²

Our patient’s case was unusual because AGEP triggered by topical medications is rarely reported, especially with ophthalmic medications.³ Drugs most commonly implicated are antibiotics including penicillins, sulfonamides, and quinolones, but other drugs such as terbinafine, diltiazem, and hydroxychloroquine have also been associated.²

AGEP may present with extensive skin desquamation, as in our patient, sometimes with bullae formation and skin sloughing manifesting as AGEP with overlapping toxic epidermal necrolysis.⁴

Diagnosis entails a careful review of medications, attention to lesion morphology, compatible disease course, and a high index of suspicion. Treatment is supportive and consists of stopping the offending agent, wound care, and antipyretics. Evidence for the use of steroids is weak.⁵

TAKE-HOME POINTS

AGEP should be considered in sudden-onset pustular desquamating erythematous rash related to use of a new medication. It is important to be aware that topical and ophthalmic medications are possible triggers. A thorough medication review should be done. Antibiotics are the most commonly implicated medications. An alternative medication should be tried. Treatment is supportive, as the condition is usually self-limiting once the offending medication is discontinued. Rarely, extensive desquamation and bullae formation may occur, which may be a manifestation of overlap features with toxic epidermal necrolysis.