Image by Betty Pace

The US Food and Drug Administration (FDA) has granted approval for L-glutamine oral powder (Endari), the first treatment approved to treat sickle cell disease (SCD) in the US in nearly 20 years.

L-glutamine oral powder, a product developed by Emmaus Medical Inc., is intended to reduce severe complications of SCD in patients age 5 and older.

The FDA’s approval of L-glutamine was supported by efficacy data from a phase 3 trial.

The trial enrolled 230 adults and children with SCD, and they were randomized to receive L-glutamine or placebo.

Patients who received L-glutamine had fewer sickle cell crises, hospitalizations, cumulative hospital days, and cases of acute chest syndrome than patients who received placebo.

Results from this trial were presented at the 2014 ASH Annual Meeting.

The FDA approval of L-glutamine was also supported by safety data from 298 patients treated with L-glutamine and 111 patients treated with placebo in the phase 2 and phase 3 studies.

Based on these data, L-glutamine was considered well-tolerated in pediatric and adult patients. The most common adverse events (occurring in more than 10% of patients receiving L-glutamine) were constipation, nausea, headache, abdominal pain, cough, pain in extremity, back pain, and chest pain (non-cardiac).

Image by Betty Pace

The US Food and Drug Administration (FDA) has granted approval for L-glutamine oral powder (Endari), the first treatment approved to treat sickle cell disease (SCD) in the US in nearly 20 years.

L-glutamine oral powder, a product developed by Emmaus Medical Inc., is intended to reduce severe complications of SCD in patients age 5 and older.

The FDA’s approval of L-glutamine was supported by efficacy data from a phase 3 trial.

The trial enrolled 230 adults and children with SCD, and they were randomized to receive L-glutamine or placebo.

Patients who received L-glutamine had fewer sickle cell crises, hospitalizations, cumulative hospital days, and cases of acute chest syndrome than patients who received placebo.

Results from this trial were presented at the 2014 ASH Annual Meeting.

The FDA approval of L-glutamine was also supported by safety data from 298 patients treated with L-glutamine and 111 patients treated with placebo in the phase 2 and phase 3 studies.

Based on these data, L-glutamine was considered well-tolerated in pediatric and adult patients. The most common adverse events (occurring in more than 10% of patients receiving L-glutamine) were constipation, nausea, headache, abdominal pain, cough, pain in extremity, back pain, and chest pain (non-cardiac).

Image by Betty Pace

The US Food and Drug Administration (FDA) has granted approval for L-glutamine oral powder (Endari), the first treatment approved to treat sickle cell disease (SCD) in the US in nearly 20 years.

L-glutamine oral powder, a product developed by Emmaus Medical Inc., is intended to reduce severe complications of SCD in patients age 5 and older.

The FDA’s approval of L-glutamine was supported by efficacy data from a phase 3 trial.

The trial enrolled 230 adults and children with SCD, and they were randomized to receive L-glutamine or placebo.

Patients who received L-glutamine had fewer sickle cell crises, hospitalizations, cumulative hospital days, and cases of acute chest syndrome than patients who received placebo.

Results from this trial were presented at the 2014 ASH Annual Meeting.

The FDA approval of L-glutamine was also supported by safety data from 298 patients treated with L-glutamine and 111 patients treated with placebo in the phase 2 and phase 3 studies.

Based on these data, L-glutamine was considered well-tolerated in pediatric and adult patients. The most common adverse events (occurring in more than 10% of patients receiving L-glutamine) were constipation, nausea, headache, abdominal pain, cough, pain in extremity, back pain, and chest pain (non-cardiac).

Study Overview

Objective. To examine the screening practices of family practitioners (FPs) and pediatricians for type 2 diabetes (T2D) in adolescents.

Design. Cross-sectional study.

Setting and participants. The researchers randomly sampled 700 pediatricians and 700 FPs who participated in direct patient care using the American Medical Association Physician Masterfile using a mail survey. Exclusion criteria included providers who were residents, hospital staff, retirees, or employed by federally owned medical facilities, certified with a subspecialty, or over age 70.

Main outcome measures. Providers were given a hypothetical case of an obese, female, teenaged patient with concurrent associated risk factors for T2D (family history of T2D, minority race, signs of insulin resistance) and asked what initial screening tests they would order. Respondents were then informed of the updated American Diabetes Association (ADA) guidelines that added hemoglobin A1c as a screening test to diagnose diabetes. The survey then asked if knowing this change in recommendation has changed or will change their screening practices in adolescents.

Main results. 1400 surveys were mailed. After 2 were excluded due to mailing issues, 52% of providers provided responses. Of these, 129 providers reported that they did not care for adolescents (age 10–17), resulting in 604 providers in the final sample, 398 pediatricians and 335 FPs.

The vast majority (92%) said they would screen the hypothetical case for diabetes, with most initially ordering a fasting test (fasting plasma glucose or 2-hour glucose tolerance test) (63%) or A1c test (58%). Of the 58% who planned to order HbA1c, only 35% ordered it in combination with a fasting test. HbA1c was significantly more likely to be ordered by pediatricians than by FPs (P = 0.001). After being presented with the new guidelines, 84% said then would now order HbA1c, a 27% increase.

Conclusion. In response to information about the new guidelines, providers were more likely to order A1c as part of initial testing. Due to the lower test performance in children and increased cost of the test, the use of HbA1c without fasting tests may result in missed diagnosis of T2D in adolescents as well as increased health care costs.

Commentary

Rates of childhood obesity continue to rise throughout the United States. Obese children are at risk for numerous comorbidities such as hypertension, hyperlipidemia, and T2D [1,2]. It is important for providers to use effective screening tools for risk assessment of prediabetes/T2DM in children.

The standard tests for diagnosing diabetes are the fasting plasma glucose test and the 2-hour plasma glucose test. While accurate, these tests are not convenient. In 2010, the ADA added an easier method of testing for T2D: an HbA1c, with results greater than or equal to 6.5% indicating diabetes [3]. However, this recommendation is controversial, given studies suggesting that HbA1c is not as reliable in children as it is in adults [4–6]. The ADA itself acknowledges that there are limited data in the pediatric population.

In this study, most providers were unaware of the 4-year-old revised guidelines offering the A1c option but are planning to apply the guidelines going forward. According to the study, this would result in a 27% increase in providers utilizing HbA1c.

Should increased uptake of A1c as an initial screening test be a concern? Using it in combination with other tests may be useful for assessing which adolescents will need further testing [3–6]. Additionally, by starting with a test that can be performed in the office with no regard to fasting time, it is possible that more cases of T2D will be found by primary care providers treating adolescents.

A weakness of the study is the potential for response bias related to mailed surveys. An additional weakness is that the researchers utilized only 1 hypothetical situation. Providing additional hypothetical situations may have allowed for further understanding of screening practices. The investigators also did not include nurse practitioners or physician assistants in their sample, a growing percentage of whom may care for adolescent populations at risk for T2D or be primary referral sources.

Applications for Clinical Practice

Providers can use HbA1c to screen for diabetes in nonfasting adolescents at risk for diabetes. While the test may not be as accurate in pediatric patients, utilizing HbA1C as directed by the ADA may aid in diagnosing patients that may otherwise miss follow-up appointments to complete a fasting test.

                                                                                                                                                                                           —Jennifer L. Nahum, MSN, CPNP-AC, PPCNP-BC, and Allison Squires, PhD, RN

Study Overview

Objective. To examine the screening practices of family practitioners (FPs) and pediatricians for type 2 diabetes (T2D) in adolescents.

Design. Cross-sectional study.

Setting and participants. The researchers randomly sampled 700 pediatricians and 700 FPs who participated in direct patient care using the American Medical Association Physician Masterfile using a mail survey. Exclusion criteria included providers who were residents, hospital staff, retirees, or employed by federally owned medical facilities, certified with a subspecialty, or over age 70.

Main outcome measures. Providers were given a hypothetical case of an obese, female, teenaged patient with concurrent associated risk factors for T2D (family history of T2D, minority race, signs of insulin resistance) and asked what initial screening tests they would order. Respondents were then informed of the updated American Diabetes Association (ADA) guidelines that added hemoglobin A1c as a screening test to diagnose diabetes. The survey then asked if knowing this change in recommendation has changed or will change their screening practices in adolescents.

Main results. 1400 surveys were mailed. After 2 were excluded due to mailing issues, 52% of providers provided responses. Of these, 129 providers reported that they did not care for adolescents (age 10–17), resulting in 604 providers in the final sample, 398 pediatricians and 335 FPs.

The vast majority (92%) said they would screen the hypothetical case for diabetes, with most initially ordering a fasting test (fasting plasma glucose or 2-hour glucose tolerance test) (63%) or A1c test (58%). Of the 58% who planned to order HbA1c, only 35% ordered it in combination with a fasting test. HbA1c was significantly more likely to be ordered by pediatricians than by FPs (P = 0.001). After being presented with the new guidelines, 84% said then would now order HbA1c, a 27% increase.

Conclusion. In response to information about the new guidelines, providers were more likely to order A1c as part of initial testing. Due to the lower test performance in children and increased cost of the test, the use of HbA1c without fasting tests may result in missed diagnosis of T2D in adolescents as well as increased health care costs.

Commentary

Rates of childhood obesity continue to rise throughout the United States. Obese children are at risk for numerous comorbidities such as hypertension, hyperlipidemia, and T2D [1,2]. It is important for providers to use effective screening tools for risk assessment of prediabetes/T2DM in children.

The standard tests for diagnosing diabetes are the fasting plasma glucose test and the 2-hour plasma glucose test. While accurate, these tests are not convenient. In 2010, the ADA added an easier method of testing for T2D: an HbA1c, with results greater than or equal to 6.5% indicating diabetes [3]. However, this recommendation is controversial, given studies suggesting that HbA1c is not as reliable in children as it is in adults [4–6]. The ADA itself acknowledges that there are limited data in the pediatric population.

In this study, most providers were unaware of the 4-year-old revised guidelines offering the A1c option but are planning to apply the guidelines going forward. According to the study, this would result in a 27% increase in providers utilizing HbA1c.

Should increased uptake of A1c as an initial screening test be a concern? Using it in combination with other tests may be useful for assessing which adolescents will need further testing [3–6]. Additionally, by starting with a test that can be performed in the office with no regard to fasting time, it is possible that more cases of T2D will be found by primary care providers treating adolescents.

A weakness of the study is the potential for response bias related to mailed surveys. An additional weakness is that the researchers utilized only 1 hypothetical situation. Providing additional hypothetical situations may have allowed for further understanding of screening practices. The investigators also did not include nurse practitioners or physician assistants in their sample, a growing percentage of whom may care for adolescent populations at risk for T2D or be primary referral sources.

Applications for Clinical Practice

Providers can use HbA1c to screen for diabetes in nonfasting adolescents at risk for diabetes. While the test may not be as accurate in pediatric patients, utilizing HbA1C as directed by the ADA may aid in diagnosing patients that may otherwise miss follow-up appointments to complete a fasting test.

                                                                                                                                                                                           —Jennifer L. Nahum, MSN, CPNP-AC, PPCNP-BC, and Allison Squires, PhD, RN

Study Overview

Objective. To examine the screening practices of family practitioners (FPs) and pediatricians for type 2 diabetes (T2D) in adolescents.

Design. Cross-sectional study.

Setting and participants. The researchers randomly sampled 700 pediatricians and 700 FPs who participated in direct patient care using the American Medical Association Physician Masterfile using a mail survey. Exclusion criteria included providers who were residents, hospital staff, retirees, or employed by federally owned medical facilities, certified with a subspecialty, or over age 70.

Main outcome measures. Providers were given a hypothetical case of an obese, female, teenaged patient with concurrent associated risk factors for T2D (family history of T2D, minority race, signs of insulin resistance) and asked what initial screening tests they would order. Respondents were then informed of the updated American Diabetes Association (ADA) guidelines that added hemoglobin A1c as a screening test to diagnose diabetes. The survey then asked if knowing this change in recommendation has changed or will change their screening practices in adolescents.

Main results. 1400 surveys were mailed. After 2 were excluded due to mailing issues, 52% of providers provided responses. Of these, 129 providers reported that they did not care for adolescents (age 10–17), resulting in 604 providers in the final sample, 398 pediatricians and 335 FPs.

The vast majority (92%) said they would screen the hypothetical case for diabetes, with most initially ordering a fasting test (fasting plasma glucose or 2-hour glucose tolerance test) (63%) or A1c test (58%). Of the 58% who planned to order HbA1c, only 35% ordered it in combination with a fasting test. HbA1c was significantly more likely to be ordered by pediatricians than by FPs (P = 0.001). After being presented with the new guidelines, 84% said then would now order HbA1c, a 27% increase.

Conclusion. In response to information about the new guidelines, providers were more likely to order A1c as part of initial testing. Due to the lower test performance in children and increased cost of the test, the use of HbA1c without fasting tests may result in missed diagnosis of T2D in adolescents as well as increased health care costs.

Commentary

Rates of childhood obesity continue to rise throughout the United States. Obese children are at risk for numerous comorbidities such as hypertension, hyperlipidemia, and T2D [1,2]. It is important for providers to use effective screening tools for risk assessment of prediabetes/T2DM in children.

The standard tests for diagnosing diabetes are the fasting plasma glucose test and the 2-hour plasma glucose test. While accurate, these tests are not convenient. In 2010, the ADA added an easier method of testing for T2D: an HbA1c, with results greater than or equal to 6.5% indicating diabetes [3]. However, this recommendation is controversial, given studies suggesting that HbA1c is not as reliable in children as it is in adults [4–6]. The ADA itself acknowledges that there are limited data in the pediatric population.

In this study, most providers were unaware of the 4-year-old revised guidelines offering the A1c option but are planning to apply the guidelines going forward. According to the study, this would result in a 27% increase in providers utilizing HbA1c.

Should increased uptake of A1c as an initial screening test be a concern? Using it in combination with other tests may be useful for assessing which adolescents will need further testing [3–6]. Additionally, by starting with a test that can be performed in the office with no regard to fasting time, it is possible that more cases of T2D will be found by primary care providers treating adolescents.

A weakness of the study is the potential for response bias related to mailed surveys. An additional weakness is that the researchers utilized only 1 hypothetical situation. Providing additional hypothetical situations may have allowed for further understanding of screening practices. The investigators also did not include nurse practitioners or physician assistants in their sample, a growing percentage of whom may care for adolescent populations at risk for T2D or be primary referral sources.

Applications for Clinical Practice

Providers can use HbA1c to screen for diabetes in nonfasting adolescents at risk for diabetes. While the test may not be as accurate in pediatric patients, utilizing HbA1C as directed by the ADA may aid in diagnosing patients that may otherwise miss follow-up appointments to complete a fasting test.

                                                                                                                                                                                           —Jennifer L. Nahum, MSN, CPNP-AC, PPCNP-BC, and Allison Squires, PhD, RN

PARIS – Percutaneous left atrial appendage closure with an Amplatzer device in patients with nonvalvular atrial fibrillation was associated with significantly lower rates of all-cause and cardiovascular mortality, compared with oral anticoagulation, in a large propensity score–matched observational registry study.

Left atrial appendage closure (LAAC) also bested oral anticoagulation (OAC) with warfarin or a novel oral anticoagulant (NOAC) in terms of net clinical benefit on the basis of the device therapy’s greater protection against stroke and systemic embolism coupled with a trend, albeit not statistically significant, for fewer bleeding events, Steffen Gloekler, MD, reported at the annual congress of the European Association of Percutaneous Cardiovascular Interventions.

The Watchman LAAC device, commercially available both in Europe and the United States, has previously been shown to be superior to OAC in terms of efficacy and noninferior regarding safety. But there have been no randomized trials of an Amplatzer device versus OAC. This lack of data was the impetus for Dr. Gloekler and his coinvestigators to create a meticulously propensity-matched observational registry.

Five hundred consecutive patients with AF who received an Amplatzer Cardiac Plug or its second-generation version, the Amplatzer Amulet, during 2009-2014 were tightly matched to an equal number of AF patients on OAC based on age, sex, body mass index, left ventricular ejection fraction, renal function, coronary artery disease status, hemoglobin level, CHA₂DS₂-VASc score, and HAS-BLED score. During a mean 2.7 years, or 2,645 patient-years, of follow-up, the composite primary efficacy endpoint, composed of stroke, systemic embolism, and cardiovascular or unexplained death occurred in 5.6% of the LAAC group, compared with 7.8% of controls in the OAC arm, for a statistically significant 30% relative risk reduction. Disabling stroke occurred in 0.7% of Amplatzer patients versus 1.5% of controls. The ischemic stroke rate was 1.5% in the device therapy group and 2% in the OAC arm.

All-cause mortality occurred in 8.3% of Amplatzer patients and 11.6% of the OAC group, for a 28% relative risk reduction. The cardiovascular death rate was 4% in the Amplatzer group, compared with 6.5% of controls, for a 36% risk reduction.

The composite safety endpoint, comprising all major procedural adverse events and major or life-threatening bleeding during follow-up, occurred in 3.6% of the Amplatzer group and 4.6% of the OAC group, for a 20% relative risk reduction that is not significant at this point because of the low number of events. Major, life-threatening, or fatal bleeding occurred in 2% of Amplatzer recipients versus 5.5% of controls, added Dr. Gloekler of University Hospital in Bern, Switzerland.

The net clinical benefit, a composite of death, bleeding, or stroke, occurred in 8.1% of the Amplatzer group, compared with 10.9% of controls, for a significant 24% reduction in relative risk in favor of device therapy.

Of note, at 2.7 years of follow-up only 55% of the OAC group were still taking an anticoagulant: 38% of the original 500 patients were on warfarin, and 17% were taking a NOAC. At that point, 8% of the Amplatzer group were on any anticoagulation therapy.

Discussion of the study focused on that low rate of medication adherence in the OAC arm. Dr. Gloekler’s response was that, after looking at the literature, he was no longer surprised by the finding that only 55% of the control group were on OAC at follow-up.

“If you look in the literature, that’s exactly the real-world adherence for OACs. Even in all four certification trials for the NOACs, the rate of discontinuation was 30% after 2 years – and these were controlled studies. Ours was observational, and it depicts a good deal of the problem with any OAC in my eyes,” Dr. Gloekler said.

Patients on warfarin in the real-world Amplatzer registry study spent on average a mere 30% of time in the therapeutic international normalized ratio range of 2-3.

“That means 70% of the time patients are higher and have an increased bleeding risk or they are lower and don’t have adequate stroke protection,” he noted.

This prompted one observer to comment, “We either have to do a better job in our clinics with OAC or we have to occlude more appendages.”

A large pivotal U.S. trial aimed at winning FDA approval for the Amplatzer Amulet for LAAC is underway. Patients with AF are being randomized to the approved Watchman or investigational Amulet at roughly 100 U.S. and 50 foreign sites.

Dr. Gloekler reported receiving research funds for the registry from the Swiss Heart Foundation and Abbott.

[email protected]

PARIS – Percutaneous left atrial appendage closure with an Amplatzer device in patients with nonvalvular atrial fibrillation was associated with significantly lower rates of all-cause and cardiovascular mortality, compared with oral anticoagulation, in a large propensity score–matched observational registry study.

Left atrial appendage closure (LAAC) also bested oral anticoagulation (OAC) with warfarin or a novel oral anticoagulant (NOAC) in terms of net clinical benefit on the basis of the device therapy’s greater protection against stroke and systemic embolism coupled with a trend, albeit not statistically significant, for fewer bleeding events, Steffen Gloekler, MD, reported at the annual congress of the European Association of Percutaneous Cardiovascular Interventions.

The Watchman LAAC device, commercially available both in Europe and the United States, has previously been shown to be superior to OAC in terms of efficacy and noninferior regarding safety. But there have been no randomized trials of an Amplatzer device versus OAC. This lack of data was the impetus for Dr. Gloekler and his coinvestigators to create a meticulously propensity-matched observational registry.

Five hundred consecutive patients with AF who received an Amplatzer Cardiac Plug or its second-generation version, the Amplatzer Amulet, during 2009-2014 were tightly matched to an equal number of AF patients on OAC based on age, sex, body mass index, left ventricular ejection fraction, renal function, coronary artery disease status, hemoglobin level, CHA₂DS₂-VASc score, and HAS-BLED score. During a mean 2.7 years, or 2,645 patient-years, of follow-up, the composite primary efficacy endpoint, composed of stroke, systemic embolism, and cardiovascular or unexplained death occurred in 5.6% of the LAAC group, compared with 7.8% of controls in the OAC arm, for a statistically significant 30% relative risk reduction. Disabling stroke occurred in 0.7% of Amplatzer patients versus 1.5% of controls. The ischemic stroke rate was 1.5% in the device therapy group and 2% in the OAC arm.

All-cause mortality occurred in 8.3% of Amplatzer patients and 11.6% of the OAC group, for a 28% relative risk reduction. The cardiovascular death rate was 4% in the Amplatzer group, compared with 6.5% of controls, for a 36% risk reduction.

The composite safety endpoint, comprising all major procedural adverse events and major or life-threatening bleeding during follow-up, occurred in 3.6% of the Amplatzer group and 4.6% of the OAC group, for a 20% relative risk reduction that is not significant at this point because of the low number of events. Major, life-threatening, or fatal bleeding occurred in 2% of Amplatzer recipients versus 5.5% of controls, added Dr. Gloekler of University Hospital in Bern, Switzerland.

The net clinical benefit, a composite of death, bleeding, or stroke, occurred in 8.1% of the Amplatzer group, compared with 10.9% of controls, for a significant 24% reduction in relative risk in favor of device therapy.

Of note, at 2.7 years of follow-up only 55% of the OAC group were still taking an anticoagulant: 38% of the original 500 patients were on warfarin, and 17% were taking a NOAC. At that point, 8% of the Amplatzer group were on any anticoagulation therapy.

Discussion of the study focused on that low rate of medication adherence in the OAC arm. Dr. Gloekler’s response was that, after looking at the literature, he was no longer surprised by the finding that only 55% of the control group were on OAC at follow-up.

“If you look in the literature, that’s exactly the real-world adherence for OACs. Even in all four certification trials for the NOACs, the rate of discontinuation was 30% after 2 years – and these were controlled studies. Ours was observational, and it depicts a good deal of the problem with any OAC in my eyes,” Dr. Gloekler said.

Patients on warfarin in the real-world Amplatzer registry study spent on average a mere 30% of time in the therapeutic international normalized ratio range of 2-3.

“That means 70% of the time patients are higher and have an increased bleeding risk or they are lower and don’t have adequate stroke protection,” he noted.

This prompted one observer to comment, “We either have to do a better job in our clinics with OAC or we have to occlude more appendages.”

A large pivotal U.S. trial aimed at winning FDA approval for the Amplatzer Amulet for LAAC is underway. Patients with AF are being randomized to the approved Watchman or investigational Amulet at roughly 100 U.S. and 50 foreign sites.

Dr. Gloekler reported receiving research funds for the registry from the Swiss Heart Foundation and Abbott.

[email protected]

PARIS – Percutaneous left atrial appendage closure with an Amplatzer device in patients with nonvalvular atrial fibrillation was associated with significantly lower rates of all-cause and cardiovascular mortality, compared with oral anticoagulation, in a large propensity score–matched observational registry study.

Left atrial appendage closure (LAAC) also bested oral anticoagulation (OAC) with warfarin or a novel oral anticoagulant (NOAC) in terms of net clinical benefit on the basis of the device therapy’s greater protection against stroke and systemic embolism coupled with a trend, albeit not statistically significant, for fewer bleeding events, Steffen Gloekler, MD, reported at the annual congress of the European Association of Percutaneous Cardiovascular Interventions.

The Watchman LAAC device, commercially available both in Europe and the United States, has previously been shown to be superior to OAC in terms of efficacy and noninferior regarding safety. But there have been no randomized trials of an Amplatzer device versus OAC. This lack of data was the impetus for Dr. Gloekler and his coinvestigators to create a meticulously propensity-matched observational registry.

Five hundred consecutive patients with AF who received an Amplatzer Cardiac Plug or its second-generation version, the Amplatzer Amulet, during 2009-2014 were tightly matched to an equal number of AF patients on OAC based on age, sex, body mass index, left ventricular ejection fraction, renal function, coronary artery disease status, hemoglobin level, CHA₂DS₂-VASc score, and HAS-BLED score. During a mean 2.7 years, or 2,645 patient-years, of follow-up, the composite primary efficacy endpoint, composed of stroke, systemic embolism, and cardiovascular or unexplained death occurred in 5.6% of the LAAC group, compared with 7.8% of controls in the OAC arm, for a statistically significant 30% relative risk reduction. Disabling stroke occurred in 0.7% of Amplatzer patients versus 1.5% of controls. The ischemic stroke rate was 1.5% in the device therapy group and 2% in the OAC arm.

All-cause mortality occurred in 8.3% of Amplatzer patients and 11.6% of the OAC group, for a 28% relative risk reduction. The cardiovascular death rate was 4% in the Amplatzer group, compared with 6.5% of controls, for a 36% risk reduction.

The composite safety endpoint, comprising all major procedural adverse events and major or life-threatening bleeding during follow-up, occurred in 3.6% of the Amplatzer group and 4.6% of the OAC group, for a 20% relative risk reduction that is not significant at this point because of the low number of events. Major, life-threatening, or fatal bleeding occurred in 2% of Amplatzer recipients versus 5.5% of controls, added Dr. Gloekler of University Hospital in Bern, Switzerland.

The net clinical benefit, a composite of death, bleeding, or stroke, occurred in 8.1% of the Amplatzer group, compared with 10.9% of controls, for a significant 24% reduction in relative risk in favor of device therapy.

Of note, at 2.7 years of follow-up only 55% of the OAC group were still taking an anticoagulant: 38% of the original 500 patients were on warfarin, and 17% were taking a NOAC. At that point, 8% of the Amplatzer group were on any anticoagulation therapy.

Discussion of the study focused on that low rate of medication adherence in the OAC arm. Dr. Gloekler’s response was that, after looking at the literature, he was no longer surprised by the finding that only 55% of the control group were on OAC at follow-up.

“If you look in the literature, that’s exactly the real-world adherence for OACs. Even in all four certification trials for the NOACs, the rate of discontinuation was 30% after 2 years – and these were controlled studies. Ours was observational, and it depicts a good deal of the problem with any OAC in my eyes,” Dr. Gloekler said.

Patients on warfarin in the real-world Amplatzer registry study spent on average a mere 30% of time in the therapeutic international normalized ratio range of 2-3.

“That means 70% of the time patients are higher and have an increased bleeding risk or they are lower and don’t have adequate stroke protection,” he noted.

This prompted one observer to comment, “We either have to do a better job in our clinics with OAC or we have to occlude more appendages.”

A large pivotal U.S. trial aimed at winning FDA approval for the Amplatzer Amulet for LAAC is underway. Patients with AF are being randomized to the approved Watchman or investigational Amulet at roughly 100 U.S. and 50 foreign sites.

Dr. Gloekler reported receiving research funds for the registry from the Swiss Heart Foundation and Abbott.

[email protected]

Outpatient care represents the largest share of mental health treatment expenditures, and it continues to get larger, while components such as retail drug prescriptions and inpatient care have declined, according to the National Center of Health Statistics.

In 2014, outpatient care took a $65.5-billion slice (about 35%) out of the $186-billion mental health care spending pie, compared with the $51.1 billion (27%) spent on retail drug prescriptions, which was the next-largest portion. Inpatient care was third with $30.3 billion in spending (16% of the total), followed by residential care at $23.2 billion (12%), and insurance administration at $15.9 billion (9%), the NCHS reported in “Health, United States, 2016.”

[email protected]

Outpatient care represents the largest share of mental health treatment expenditures, and it continues to get larger, while components such as retail drug prescriptions and inpatient care have declined, according to the National Center of Health Statistics.

In 2014, outpatient care took a $65.5-billion slice (about 35%) out of the $186-billion mental health care spending pie, compared with the $51.1 billion (27%) spent on retail drug prescriptions, which was the next-largest portion. Inpatient care was third with $30.3 billion in spending (16% of the total), followed by residential care at $23.2 billion (12%), and insurance administration at $15.9 billion (9%), the NCHS reported in “Health, United States, 2016.”

[email protected]

Outpatient care represents the largest share of mental health treatment expenditures, and it continues to get larger, while components such as retail drug prescriptions and inpatient care have declined, according to the National Center of Health Statistics.

In 2014, outpatient care took a $65.5-billion slice (about 35%) out of the $186-billion mental health care spending pie, compared with the $51.1 billion (27%) spent on retail drug prescriptions, which was the next-largest portion. Inpatient care was third with $30.3 billion in spending (16% of the total), followed by residential care at $23.2 billion (12%), and insurance administration at $15.9 billion (9%), the NCHS reported in “Health, United States, 2016.”

[email protected]

The Food and Drug Administration approved L-glutamine oral powder for reducing severe complications of sickle cell disease in patients aged 5 years and older.

The approval was based on placebo-controlled phase II and phase III trials suggesting that L-glutamate offered moderate benefit to patients with this rare, serious, and potentially fatal blood disorder.

The Food and Drug Administration approved L-glutamine oral powder for reducing severe complications of sickle cell disease in patients aged 5 years and older.

The approval was based on placebo-controlled phase II and phase III trials suggesting that L-glutamate offered moderate benefit to patients with this rare, serious, and potentially fatal blood disorder.

The Food and Drug Administration approved L-glutamine oral powder for reducing severe complications of sickle cell disease in patients aged 5 years and older.

The approval was based on placebo-controlled phase II and phase III trials suggesting that L-glutamate offered moderate benefit to patients with this rare, serious, and potentially fatal blood disorder.

LUGANO, SWITZERLAND – Tazemetostat, a first-in-class experimental agent that inhibits an oncogenic protein, shows efficacy in patients with heavily pretreated, relapsed/refractory follicular lymphoma (FL) and diffuse large B cell lymphoma (DLBCL), interim results from a phase II study suggest.

Among patients with relapsed/refractory FL who had mutations in EZH2 (enhancer of zeste homolog 2), a member of a family of proteins that are involved in epigenetic gene silencing, the overall response rate (ORR) was 92%, reported Franck Morschhauser, MD, PhD, of the University of Lille, France.

Neil Osterweil/Frontline Medical News

Large multicenter study

Dr. Morschhauser reported interim results from a global, multi-center open-label study of tazemetostat in six cohorts of patients with relapsed/refractory FL (two monotherapy cohorts of 45 patients each) or DLBCL (three monotherapy cohorts of 60 patients each). A sixth cohort consisting of 70 patients with DLBCL treated with tazemetostat and prednisolone was added in 2017.

In the ongoing trial, patients receive oral tazemetostat 800 mg twice daily until disease progression or withdrawal from study, and are being followed for ORR, progression-free survival, overall survival, duration-of response, safety, and pharmacokinetics.

The longest follow-up at the time of data cutoff was approximately 18 months. Among 13 evaluable patients with FL with EZH2 mutations, the ORR was 92%, including one complete response (CR) and 11 partial responses (PR). In contrast, the ORR for 54 patients with FL and wild-type EZH2 was 28%, consisting of three CRs and 11 PRs. One patient with mutated EZH2 and 23 with wild-type EZH2 had stable disease.

Among 17 patients with DLBCL and EZH2 mutations, the ORR was 29%, consisting of 5 PR. For 119 patients with wild-type EZH2, the ORR was 15%, consisting of 10 CR and 8 PR. Six patients with mutations and 22 with wild-type EZH2 had stable disease.

Among the patients with FL, 75% had significant reduction in tumor burden.

The time to response ranged from 2 months to 1 year, with a median of approximately 4 months.

The variability in time to response “makes it a little bit tricky to calculate response duration,” Dr. Morschhauser said.

The drug had a “favorable” safety profile, with treatment-related adverse events of grade 3 or greater in more than 5% of patients including thrombocytopenias in 6% of patients, anemias in 4%, and neutropenias in 6%. Treatment-emergent adverse events leading to dose reductions occurred in 4% of patients, and those leading to drug discontinuation or study withdrawal occurred in 12% of patients.

In a retrospective analysis, the investigators performed molecular profiling studies using next-generation sequencing to look for predictors of response to tazemetostat. They found that patients most likely to respond to tazemetostat were those with activating mutations in EZH2 and MYD88. In contrast, patients with mutations HIST1H1E or MYC were not likely to respond.

Thomas E. Witzig, MD, of the Mayo Clinic in Rochester, Minn., the invited discussant, said that the study is important because “it provides proof of principle that attacking the methylation issue, attacking one of these enzymes, is very important and can produce single-agent responses.

“It also demonstrates the value of mutation status, and this trial knowledge of that mutation status has actually changed the trial design, so that now they are only putting patients on with mutations,” he said.

The trial also raises the possibility of targeting other parts of the methylation pathway to treat cancer, he added.

The study was sponsored by Epizyme, the maker of tazemetostat. Dr. Morschhauser disclosed receiving honoraria from and serving on advisory boards for both companies. Dr. Witzig has disclosed grants for clinical trials from Novartis and Wyeth, and he has served on advisory boards for Cephalon, Novartis, and Wyeth.

LUGANO, SWITZERLAND – Tazemetostat, a first-in-class experimental agent that inhibits an oncogenic protein, shows efficacy in patients with heavily pretreated, relapsed/refractory follicular lymphoma (FL) and diffuse large B cell lymphoma (DLBCL), interim results from a phase II study suggest.

Among patients with relapsed/refractory FL who had mutations in EZH2 (enhancer of zeste homolog 2), a member of a family of proteins that are involved in epigenetic gene silencing, the overall response rate (ORR) was 92%, reported Franck Morschhauser, MD, PhD, of the University of Lille, France.

Neil Osterweil/Frontline Medical News

Large multicenter study

Dr. Morschhauser reported interim results from a global, multi-center open-label study of tazemetostat in six cohorts of patients with relapsed/refractory FL (two monotherapy cohorts of 45 patients each) or DLBCL (three monotherapy cohorts of 60 patients each). A sixth cohort consisting of 70 patients with DLBCL treated with tazemetostat and prednisolone was added in 2017.

In the ongoing trial, patients receive oral tazemetostat 800 mg twice daily until disease progression or withdrawal from study, and are being followed for ORR, progression-free survival, overall survival, duration-of response, safety, and pharmacokinetics.

The longest follow-up at the time of data cutoff was approximately 18 months. Among 13 evaluable patients with FL with EZH2 mutations, the ORR was 92%, including one complete response (CR) and 11 partial responses (PR). In contrast, the ORR for 54 patients with FL and wild-type EZH2 was 28%, consisting of three CRs and 11 PRs. One patient with mutated EZH2 and 23 with wild-type EZH2 had stable disease.

Among 17 patients with DLBCL and EZH2 mutations, the ORR was 29%, consisting of 5 PR. For 119 patients with wild-type EZH2, the ORR was 15%, consisting of 10 CR and 8 PR. Six patients with mutations and 22 with wild-type EZH2 had stable disease.

Among the patients with FL, 75% had significant reduction in tumor burden.

The time to response ranged from 2 months to 1 year, with a median of approximately 4 months.

The variability in time to response “makes it a little bit tricky to calculate response duration,” Dr. Morschhauser said.

The drug had a “favorable” safety profile, with treatment-related adverse events of grade 3 or greater in more than 5% of patients including thrombocytopenias in 6% of patients, anemias in 4%, and neutropenias in 6%. Treatment-emergent adverse events leading to dose reductions occurred in 4% of patients, and those leading to drug discontinuation or study withdrawal occurred in 12% of patients.

In a retrospective analysis, the investigators performed molecular profiling studies using next-generation sequencing to look for predictors of response to tazemetostat. They found that patients most likely to respond to tazemetostat were those with activating mutations in EZH2 and MYD88. In contrast, patients with mutations HIST1H1E or MYC were not likely to respond.

Thomas E. Witzig, MD, of the Mayo Clinic in Rochester, Minn., the invited discussant, said that the study is important because “it provides proof of principle that attacking the methylation issue, attacking one of these enzymes, is very important and can produce single-agent responses.

“It also demonstrates the value of mutation status, and this trial knowledge of that mutation status has actually changed the trial design, so that now they are only putting patients on with mutations,” he said.

The trial also raises the possibility of targeting other parts of the methylation pathway to treat cancer, he added.

The study was sponsored by Epizyme, the maker of tazemetostat. Dr. Morschhauser disclosed receiving honoraria from and serving on advisory boards for both companies. Dr. Witzig has disclosed grants for clinical trials from Novartis and Wyeth, and he has served on advisory boards for Cephalon, Novartis, and Wyeth.

LUGANO, SWITZERLAND – Tazemetostat, a first-in-class experimental agent that inhibits an oncogenic protein, shows efficacy in patients with heavily pretreated, relapsed/refractory follicular lymphoma (FL) and diffuse large B cell lymphoma (DLBCL), interim results from a phase II study suggest.

Among patients with relapsed/refractory FL who had mutations in EZH2 (enhancer of zeste homolog 2), a member of a family of proteins that are involved in epigenetic gene silencing, the overall response rate (ORR) was 92%, reported Franck Morschhauser, MD, PhD, of the University of Lille, France.

Neil Osterweil/Frontline Medical News

Large multicenter study

Dr. Morschhauser reported interim results from a global, multi-center open-label study of tazemetostat in six cohorts of patients with relapsed/refractory FL (two monotherapy cohorts of 45 patients each) or DLBCL (three monotherapy cohorts of 60 patients each). A sixth cohort consisting of 70 patients with DLBCL treated with tazemetostat and prednisolone was added in 2017.

In the ongoing trial, patients receive oral tazemetostat 800 mg twice daily until disease progression or withdrawal from study, and are being followed for ORR, progression-free survival, overall survival, duration-of response, safety, and pharmacokinetics.

The longest follow-up at the time of data cutoff was approximately 18 months. Among 13 evaluable patients with FL with EZH2 mutations, the ORR was 92%, including one complete response (CR) and 11 partial responses (PR). In contrast, the ORR for 54 patients with FL and wild-type EZH2 was 28%, consisting of three CRs and 11 PRs. One patient with mutated EZH2 and 23 with wild-type EZH2 had stable disease.

Among 17 patients with DLBCL and EZH2 mutations, the ORR was 29%, consisting of 5 PR. For 119 patients with wild-type EZH2, the ORR was 15%, consisting of 10 CR and 8 PR. Six patients with mutations and 22 with wild-type EZH2 had stable disease.

Among the patients with FL, 75% had significant reduction in tumor burden.

The time to response ranged from 2 months to 1 year, with a median of approximately 4 months.

The variability in time to response “makes it a little bit tricky to calculate response duration,” Dr. Morschhauser said.

The drug had a “favorable” safety profile, with treatment-related adverse events of grade 3 or greater in more than 5% of patients including thrombocytopenias in 6% of patients, anemias in 4%, and neutropenias in 6%. Treatment-emergent adverse events leading to dose reductions occurred in 4% of patients, and those leading to drug discontinuation or study withdrawal occurred in 12% of patients.

In a retrospective analysis, the investigators performed molecular profiling studies using next-generation sequencing to look for predictors of response to tazemetostat. They found that patients most likely to respond to tazemetostat were those with activating mutations in EZH2 and MYD88. In contrast, patients with mutations HIST1H1E or MYC were not likely to respond.

Thomas E. Witzig, MD, of the Mayo Clinic in Rochester, Minn., the invited discussant, said that the study is important because “it provides proof of principle that attacking the methylation issue, attacking one of these enzymes, is very important and can produce single-agent responses.

“It also demonstrates the value of mutation status, and this trial knowledge of that mutation status has actually changed the trial design, so that now they are only putting patients on with mutations,” he said.

The trial also raises the possibility of targeting other parts of the methylation pathway to treat cancer, he added.

The study was sponsored by Epizyme, the maker of tazemetostat. Dr. Morschhauser disclosed receiving honoraria from and serving on advisory boards for both companies. Dr. Witzig has disclosed grants for clinical trials from Novartis and Wyeth, and he has served on advisory boards for Cephalon, Novartis, and Wyeth.

Opioid prescribing in the United States declined overall between 2010 and 2015, but remained stable or increased in some counties, according to a report from the Centers for Disease Control and Prevention. The findings were published online in the CDC’s Morbidity and Mortality Weekly Report.

“The bottom line remains: We have too many people getting too many prescriptions at too high a dose,” Anne Schuchat, MD, acting director of the CDC, said in a July 6 teleconference.

Opioid prescribing in the United States declined overall between 2010 and 2015, but remained stable or increased in some counties, according to a report from the Centers for Disease Control and Prevention. The findings were published online in the CDC’s Morbidity and Mortality Weekly Report.

“The bottom line remains: We have too many people getting too many prescriptions at too high a dose,” Anne Schuchat, MD, acting director of the CDC, said in a July 6 teleconference.

Opioid prescribing in the United States declined overall between 2010 and 2015, but remained stable or increased in some counties, according to a report from the Centers for Disease Control and Prevention. The findings were published online in the CDC’s Morbidity and Mortality Weekly Report.

“The bottom line remains: We have too many people getting too many prescriptions at too high a dose,” Anne Schuchat, MD, acting director of the CDC, said in a July 6 teleconference.

The Food and Drug Administration has approved abatacept, a selective T-cell costimulation modulator, for treating adults with active psoriatic arthritis (PsA), the manufacturer, Bristol-Myers Squibb, has announced.

Approval of abatacept (Orencia) was based on two randomized, double-blind, placebo-controlled studies (PsA-I and PsA-II) in 594 adults with PsA for more than 7 years, according to the July 6 announcement. Patients had active PsA (at least three swollen joints and at least three tender joints), despite previous disease-modifying antirheumatic drug (DMARD) therapy and had one qualifying psoriatic skin lesion measuring at least 2 cm in diameter. The studies included patients treated with TNF inhibitors (TNFi) previously.

[email protected]

The Food and Drug Administration has approved abatacept, a selective T-cell costimulation modulator, for treating adults with active psoriatic arthritis (PsA), the manufacturer, Bristol-Myers Squibb, has announced.

Approval of abatacept (Orencia) was based on two randomized, double-blind, placebo-controlled studies (PsA-I and PsA-II) in 594 adults with PsA for more than 7 years, according to the July 6 announcement. Patients had active PsA (at least three swollen joints and at least three tender joints), despite previous disease-modifying antirheumatic drug (DMARD) therapy and had one qualifying psoriatic skin lesion measuring at least 2 cm in diameter. The studies included patients treated with TNF inhibitors (TNFi) previously.

[email protected]

The Food and Drug Administration has approved abatacept, a selective T-cell costimulation modulator, for treating adults with active psoriatic arthritis (PsA), the manufacturer, Bristol-Myers Squibb, has announced.

Approval of abatacept (Orencia) was based on two randomized, double-blind, placebo-controlled studies (PsA-I and PsA-II) in 594 adults with PsA for more than 7 years, according to the July 6 announcement. Patients had active PsA (at least three swollen joints and at least three tender joints), despite previous disease-modifying antirheumatic drug (DMARD) therapy and had one qualifying psoriatic skin lesion measuring at least 2 cm in diameter. The studies included patients treated with TNF inhibitors (TNFi) previously.

[email protected]

Interstitial lung disease (ILD) encompasses a diverse group of disorders that cause inflammation and fibrosis of the lung parenchyma. The clinical manifestations, disease course, management and prognosis of ILD vary depending on the underlying subtype, making accurate classification and diagnosis an important initial step. While a comprehensive list of ILD contains dozens of disorders, the majority of patients will fall into 1 of 3 categories: exposure-related ILD, connective tissue disease-related ILD (CT-ILD), and the idiopathic interstitial pneumonias (Table).

An essential first step in the evaluation of every hospitalized patient with ILD is establishing a diagnosis. A common mistake among clinicians who diagnose patients with ILD is not realizing that ILD is a collection of diseases with different etiologies, natural histories, and treatments. A careful evaluation should be performed in every hospitalized patient with ILD to ensure an accurate diagnosis, ideally in the context of a multidisciplinary conference with pulmonary, radiology, pathology, and other specialties, as appropriate. A multidisciplinary panel of the American Thoracic Society/European Respiratory Society recently published a revised classification of ILD based on a combination of clinical, radiologic, and histopathologic findings, which may aid in refining the diagnosis.¹

There are 3 main scenarios in which the hospital physician will encounter patients with ILD.

Acute presentation of new-onset disease. While many ILDs present insidiously, some cases present acutely and require hospitalization. The most common of these are acute hypersensitivity pneumonitis (HP), CT-ILD (in particular, myositis-related and systemic lupus erythematosus-related), drug-induced ILD (eg, amiodarone, nitrofurantoin), cryptogenic organizing pneumonia (COP), acute eosinophilic pneumonia (AEP), and acute interstitial pneumonia (AIP).

Acute presentation of established (chronic) disease. Patients with chronic forms of ILD can present to the hospital with an acute exacerbation of disease. This can be caused by extra-parenchymal complications, including pulmonary embolism, pneumothorax, and pleural effusion; parenchymal complications such as infectious pneumonia, aspiration pneumonitis, and congestive heart failure; or without an identifiable cause. This latter presentation is most commonly seen in idiopathic pulmonary fibrosis (IPF).^2,3

Elective hospitalization for diagnostic surgical lung biopsy. Patients with ILD may be hospitalized electively for a laparoscopic surgical lung biopsy as part of their diagnostic evaluations.

Physicians caring for a hospitalized ILD patient must be familiar with the clinical presentations, diagnostic approach, medical management, and outpatient follow-up recommended in these 3 settings. We will summarize these areas and provide answers to commonly encountered clinical questions in the hospitalized patient with ILD.

Acute onset (or worsening) of dyspnea is the primary presenting symptom in most patients hospitalized for ILD. This symptom should be further characterized by assessing the degree of dyspnea and the extent of exercise limitation, as both impact overall disease severity and prognosis.⁴ Cough is the second most common symptom, and can be nonproductive, as is common in IPF, or be associated with secretions if parenchymal infection or acute bronchitis is present.⁵ Pleuritic chest pain, pleural effusion, and/or the presence of extrapulmonary features, including dysphagia, joint pain and swelling, or cutaneous thickening may suggest the presence of a CT-ILD. Because most forms of ILD present with only nonspecific symptoms, a careful history and physical examination are essential.

History

A comprehensive patient history is the backbone of diagnosing any ILD. History-taking should focus on severity and temporal progression of symptoms, presence of pre-existing systemic conditions associated with ILD, symptoms of extrapulmonary disease, and exposures to substances that can cause pulmonary injury, including a detailed history of occupations and hobbies, medications, smoking, and familial lung disease.^6-9 Physicians must try to exclude other diagnoses that could result in a similar acute presentation, including congestive heart failure and infection. Considering the complex and extensive recommended history-taking, physicians may find it helpful to use a standardized questionnaire, as provided by the American College of Chest Physicians.¹⁰

Laboratory Testing

All patients presenting to the hospital with a suspected ILD should undergo careful assessment for the presence of connective tissue disease, including patients without clear symptoms because ILD can be the presenting manifestation. We routinely test for antinuclear antibody titer and pattern, rheumatoid factor, anticyclic citrullinated peptide, creatinine kinase, and aldolase as the initial screening panel in most patients, with further testing directed by the findings on history and physical examination. Pulmonary function tests are used routinely to monitor disease progression in the outpatient setting; however, in the hospitalized ILD patient, they are often difficult to perform and have no real diagnostic value. Similarly, arterial blood gas is not routinely used as part of the initial inpatient evaluation.

Imaging

All hospitalized patients with a known or suspected ILD should undergo chest imaging, assuming they are stable enough to do so. While the chest radiograph can provide a low-cost initial assessment of the degree of lung involvement and presence of accompanying abnormalities, computed tomography (CT) scanning is the diagnostic test of choice.¹¹ The pattern and distribution of abnormalities on CT scan can greatly assist with the differential diagnosis in patients presenting with a new ILD, while the presence and pattern of new opacities superimposed on chronic changes can inform the differential and the prognosis of an ILD exacerbation.¹² High-resolution CT provides the most sensitive imaging modality for diffuse ILD. The addition of prone and expiratory images are helpful in differentiating mild lung disease from atelectasis and detecting air trapping, respectively.¹³ However, since pulmonary embolism is a common extraparenchymal finding routinely considered in the differential of a patient presenting with a known or suspected ILD, physicians should consider ordering a CT pulmonary angiogram with additional high-resolution images. Most important, radiographic evaluation should include a review of all available prior chest imaging to assess both the tempo and the nature of radiographic findings.

Bronchoscopy

Bronchoscopy (with bronchoalveolar lavage [BAL], transbronchial lung biopsy [TBLB] and/or transbronchial needle aspiration [TBNA]) is not a routinely used diagnostic tool in the hospitalized ILD patient. However, it should be considered in certain circumstances.⁷ Cell count and differential can be helpful in diagnosing AEP (greater than 40% eosinophilia) or acute HP (greater than 50% lymphocytosis), while the addition of microbiologic and cytologic analysis can assist with the diagnosis of infectious etiologies (including pneumocystis pneumonia) or malignancy.^14,15 Bronchoscopy with BAL has limited sensitivity for many infections and the procedure is associated with a small risk of worsened hypoxemia. Transbronchial lung biopsy, and to a lesser extent TBNA, carry the added risk of pneumothorax and bleeding. In the majority of cases of ILD, TBLB and TBNA have limited diagnostic utility given the small amount of lung tissue sampled. In cases of suspected IPF, where the identification of the histologic pattern is needed for definitive diagnosis, tissue from TBLB cannot be used to make a conclusive diagnosis.^16,17 However, both TBNA and TBLB are useful in the diagnosis of granulomatous disorders, such as sarcoidosis, where the diagnostic yield ranges from 80% to 90% and 50% to 75%, respectively.^18,19

A newer bronchoscopic approach to sampling the lung using a bronchoscopically-placed cryoprobe (termed transbronchial cryobiopsy) has uncertain diagnostic utility and safety in the acute setting. This procedure involves intubation, sedation, and bronchoscopy allowing for the passage of an endobronchial cryoprobe through the bronchoscope and into the periphery of the lung. Several cryobiopies are generally taken from the same pulmonary subsegment. Despite a large number of recent publications on this topic, none of them have provided a clear sense of the diagnostic yield and safety.^20,21 Transbronchial cryobiopsy remains a highly controversial procedure in the clinical setting, and we would not recommend its use until further data are available.²²

Surgical Lung Biopsy

In the outpatient setting, a surgical lung biopsy is often useful when the ILD diagnosis cannot be made from the clinical context and imaging. However, patients presenting with acute respiratory failure from ILD are at greatly increased risk of complications from nonelective biopsy including pneumothorax, hemothorax, acute exacerbation of ILD, ICU admission, mechanical ventilation, and in-hospital mortality.^23,24 Acute histological findings can also make it difficult to appreciate the underlying pattern of fibrosis, reducing the diagnostic utility.^25-27 In our experience, surgical lung biopsy rarely alters the treatment of ILD patients presenting in acute respiratory failure. We believe that surgical lung biopsy should be reserved for the rare hospitalized patients in whom the clinician believes the results would clearly change management and that the substantial risk is worth taking.^5,28

The inpatient management of ILD is a large topic and difficult to comprehensively cover in a single review. Therefore, in this section, we will review 6 key management questions that address both general and specific treatment decisions that frequently arise in the care of hospitalized ILD patients (Figure).

When should hospitalized ILD patients be treated with antibiotics?

Infection and acute presentations of ILD have many similar clinical and radiographic features, making it difficult to distinguish between the two, or exclude infection as the causative role in an acute exacerbation.² In many ILD patients, the risk of infection is higher than in the general population, due to the acute and chronic use of immunosuppression. Until firm guidelines on the use of antibiotics in hospitalized patients with acute respiratory symptoms are available, we recommend considering the empiric use of antibiotics in ILD patients in respiratory failure, in addition to a thorough infectious workup.

When should hospitalized ILD patients be treated with corticosteroids?

Clinical experience supports the use of corticosteroids in the acute management of most rapidly progressive ILDs presenting with respiratory failure, including AEP, COP, acute HP, drug-induced ILD, and some cases of CT-ILD. Patients with AEP tend to respond rapidly to corticosteroids. In a series of 137 patients with AEP, 127 (92%) received corticosteroids, with defervescence and improved dyspnea within 48 to 72 hours and resolution of all symptoms after a median of 7 (4 to 10) days.²⁹ Cryptogenic organizing pneumonia is similarly corticosteroid-responsive, with patients typically started on doses of 1mg/kg of prednisone followed by a slow taper due to the risk of relapse.³⁰ For the majority of acute CT-ILD, oral prednisone is the initial treatment, often in combination with a second immunosuppressive agent such as mycophenolate.

No proven therapies are available for acute exacerbations of IPF (AE-IPF), including the use of corticosteroids. The most recent international guidelines on the management of AE-IPF conditionally recommends the use of corticosteroids, although this recommendation is largely based on anecdotal reports and clearly states that randomized studies are needed.³ When corticosteroids are used, we recommend high doses (eg, 1 to 2 mg/kg of prednisone) with close clinical monitoring. Consider stopping corticosteroids after 3 to 5 days if there is no evidence of clinical improvement. Prolonged courses of corticosteroids should be avoided.

What additional pharmacologic therapies should be considered in the treatment of hospitalized ILD patients?

Immunomodulators. Patients presenting acutely with a new-onset ILD or with an acute exacerbation of a chronic ILD often receive corticosteroids, sometimes in concert with an immunomodulator. This is most commonly seen in the acute management of CTD- ILD and in chronic HP, where mycophenolate mofetil, and to a lesser extent, cyclophosphamide and azathioprine for CT-ILD are used in combination with corticosteroids. The rationale for this is both therapeutic synergy and a desire to limit the long-term exposure to corticosteroids. Similarly, multiple observational cohort studies have investigated the role of combination or tandem immunosuppression in the treatment AE-IPF. Although cyclosporine, cyclophosphamide, azathioprine, rituximab and tacrolimus have all been studied, their efficacy remains uncertain.³ Until these therapies are better studied, they have no routine role in the management of AE-IPF.

Antifibrotics. Nintedanib and pirfenidone are 2 antifibrotic agents approved for the treatment of IPF. Clinical trials suggest that, in addition to slowing disease progression, these therapies may help prevent AE-IPF. The data are most robust in studies of nintedanib. A phase 2 trial with 432 subjects demonstrated a delay in time to the first investigator-reported acute exacerbation.³¹ Two follow-up phase 3 trials showed a reduction in centrally adjudicated AE-IPF in the pooled nintedanib groups compared to placebo.³² An initial phase 2 trial of pirfenidone showed a reduction in acute exacerbations in patients on pirfenidone, but this finding was not replicated in follow-up studies.^33-35 Because of their potential role in preventing acute exacerbations and emerging evidence to suggest that continuation of antifibrotics may lead to better outcomes during an acute exacerbation, these drugs should not generally be stopped during a hospitalization for ILD. However, no evidence supports their initiation during acute exacerbations, and we do not recommend starting antifibrotics in the hospitalized setting for newly diagnosed patients. Starting and stopping antifibrotics should be reserved for outpatient management.

When should noninvasive and mechanical ventilation be considered?

We recommend carefully considering the use of noninvasive ventilation (NIV) and intubation in every ILD patient in respiratory distress, as an acutely reversible process may be present. In patients requiring mechanical ventilation, every effort should be made to minimize potential damage by reducing the fraction of inspired oxygen (to prevent potential hyperoxic injury) and reducing tidal volumes (to minimize barotrauma). Patients with a chronic ILD, particularly IPF, who require NIV or mechanical ventilation will generally have poor outcomes.

Studies suggest that NIV prevents mechanical ventilation in only the minority of patients presenting with an AE-IPF and is associated with high in-hospital mortality and a median survival following hospital discharge of only 60 days.^36-38 The majority of patients with IPF requiring mechanical ventilation will not survive the intensive care unit. In a series of 23 patients presenting with acute respiratory failure and IPF, 22 of the 23 patients died while receiving mechanical ventilation, with a median survival of 3 days. In a more recent study of 34 patients with acute respiratory failure and IPF, 15 subjects underwent mechanical ventilation with an in-hospital mortality rate of 100%.³⁹ Given the overall poor survival associated with AE-IPF, mechanical ventilation should be carefully considered with the patient and family as part of an overall goals-of-care conversation prior to initiation.

When should hospitalized ILD patients be referred for inpatient lung transplant evaluation?

A subset of hospitalized patients with ILD will not respond to supportive and pharmacologic care, particularly those with advanced lung fibrosis. In these cases, lung transplantation may be the only remaining treatment option. This is particularly true for patients presenting with IPF, and it is 1 of the most common indications for lung transplantation. Patients with respiratory failure and ILD should be evaluated early in the hospital course for transplantation or considered for transfer to a transplant center. General contraindications to transplant are age older than 70 years, underweight or elevated BMI (generally higher than 30), malignancy within the last 2 years (with the exception of cutaneous squamous and basal cell tumors), untreatable major organ dysfunction other than the lung, noncurable chronic extrapulmonary infection (chronic active viral hepatitis B, hepatitis C, human immunodeficiency virus), significant chest wall deformity, untreatable psychiatric or psychologic disease, substance addiction within the last 6 months, or lack of dependable social support.⁴⁰ In select patients with ILD and gas exchange abnormalities, mechanical ventilation or extracorporeal membrane oxygenation may be used to bridge a patient to lung transplantation.⁴¹

What should you tell your ILD patient to expect at discharge?

Accurate diagnosis is important not only for acute inpatient management, but for informing long-term prognosis. Acute-onset ILD tends to be more reversible, to be responsive to medical therapy, and to have a more favorable overall outcome. On the other hand, acute exacerbations of established ILD, particularly IPF, can have a more unfavorable and treatment-refractory course. Once a diagnosis is established, it is important both to provide patients with information and ensure appropriate outpatient follow-up. The Pulmonary Fibrosis Foundation (the largest U.S. advocacy and support organization for patients with ILD) provides information on ILD to patients and families and can serve as an important educational source.⁴² Prior to discharge, it is important to evaluate the oxygen needs of patients at rest and with exertion. Referral to an ILD center at discharge is important whenever possible, to monitor clinical symptoms and lung function, initiate or assess response to treatment, and provide supportive care, including oxygen therapy, pulmonary rehabilitation, and outpatient lung transplant referral.

ILD is a group of heterogeneous disorders characterized by lung inflammation and fibrosis. Although the onset of disease is typically insidious, patients can present acutely requiring hospitalization. Inpatient management varies significantly depending on ILD subtype, and, therefore, accurate diagnosis is key in determining treatment and prognosis. As we develop an improved understanding of the mechanisms of acute presentations of ILD, and our approaches to detection and treatment improve as a result of clinical trials, we anticipate continued modifications to this shared framework.

Disclosure

Dr. Collard reports personal fees from Alkermes, aTyr Pharmaceuticals, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Global Blood Therapeutics, Genoa, ImmuneWorks, Moerae Matrix, Navitor, Parexel, Patara, Pharma Capital Partners, Prometic, Takeda, Toray, and Xfibra, outside the submitted work. Drs. Farrand and Shah report no financial conflicts of interest.

Interstitial lung disease (ILD) encompasses a diverse group of disorders that cause inflammation and fibrosis of the lung parenchyma. The clinical manifestations, disease course, management and prognosis of ILD vary depending on the underlying subtype, making accurate classification and diagnosis an important initial step. While a comprehensive list of ILD contains dozens of disorders, the majority of patients will fall into 1 of 3 categories: exposure-related ILD, connective tissue disease-related ILD (CT-ILD), and the idiopathic interstitial pneumonias (Table).

An essential first step in the evaluation of every hospitalized patient with ILD is establishing a diagnosis. A common mistake among clinicians who diagnose patients with ILD is not realizing that ILD is a collection of diseases with different etiologies, natural histories, and treatments. A careful evaluation should be performed in every hospitalized patient with ILD to ensure an accurate diagnosis, ideally in the context of a multidisciplinary conference with pulmonary, radiology, pathology, and other specialties, as appropriate. A multidisciplinary panel of the American Thoracic Society/European Respiratory Society recently published a revised classification of ILD based on a combination of clinical, radiologic, and histopathologic findings, which may aid in refining the diagnosis.¹

There are 3 main scenarios in which the hospital physician will encounter patients with ILD.

Acute presentation of new-onset disease. While many ILDs present insidiously, some cases present acutely and require hospitalization. The most common of these are acute hypersensitivity pneumonitis (HP), CT-ILD (in particular, myositis-related and systemic lupus erythematosus-related), drug-induced ILD (eg, amiodarone, nitrofurantoin), cryptogenic organizing pneumonia (COP), acute eosinophilic pneumonia (AEP), and acute interstitial pneumonia (AIP).

Acute presentation of established (chronic) disease. Patients with chronic forms of ILD can present to the hospital with an acute exacerbation of disease. This can be caused by extra-parenchymal complications, including pulmonary embolism, pneumothorax, and pleural effusion; parenchymal complications such as infectious pneumonia, aspiration pneumonitis, and congestive heart failure; or without an identifiable cause. This latter presentation is most commonly seen in idiopathic pulmonary fibrosis (IPF).^2,3

Elective hospitalization for diagnostic surgical lung biopsy. Patients with ILD may be hospitalized electively for a laparoscopic surgical lung biopsy as part of their diagnostic evaluations.

Physicians caring for a hospitalized ILD patient must be familiar with the clinical presentations, diagnostic approach, medical management, and outpatient follow-up recommended in these 3 settings. We will summarize these areas and provide answers to commonly encountered clinical questions in the hospitalized patient with ILD.

Acute onset (or worsening) of dyspnea is the primary presenting symptom in most patients hospitalized for ILD. This symptom should be further characterized by assessing the degree of dyspnea and the extent of exercise limitation, as both impact overall disease severity and prognosis.⁴ Cough is the second most common symptom, and can be nonproductive, as is common in IPF, or be associated with secretions if parenchymal infection or acute bronchitis is present.⁵ Pleuritic chest pain, pleural effusion, and/or the presence of extrapulmonary features, including dysphagia, joint pain and swelling, or cutaneous thickening may suggest the presence of a CT-ILD. Because most forms of ILD present with only nonspecific symptoms, a careful history and physical examination are essential.

History

A comprehensive patient history is the backbone of diagnosing any ILD. History-taking should focus on severity and temporal progression of symptoms, presence of pre-existing systemic conditions associated with ILD, symptoms of extrapulmonary disease, and exposures to substances that can cause pulmonary injury, including a detailed history of occupations and hobbies, medications, smoking, and familial lung disease.^6-9 Physicians must try to exclude other diagnoses that could result in a similar acute presentation, including congestive heart failure and infection. Considering the complex and extensive recommended history-taking, physicians may find it helpful to use a standardized questionnaire, as provided by the American College of Chest Physicians.¹⁰

Laboratory Testing

All patients presenting to the hospital with a suspected ILD should undergo careful assessment for the presence of connective tissue disease, including patients without clear symptoms because ILD can be the presenting manifestation. We routinely test for antinuclear antibody titer and pattern, rheumatoid factor, anticyclic citrullinated peptide, creatinine kinase, and aldolase as the initial screening panel in most patients, with further testing directed by the findings on history and physical examination. Pulmonary function tests are used routinely to monitor disease progression in the outpatient setting; however, in the hospitalized ILD patient, they are often difficult to perform and have no real diagnostic value. Similarly, arterial blood gas is not routinely used as part of the initial inpatient evaluation.

Imaging

All hospitalized patients with a known or suspected ILD should undergo chest imaging, assuming they are stable enough to do so. While the chest radiograph can provide a low-cost initial assessment of the degree of lung involvement and presence of accompanying abnormalities, computed tomography (CT) scanning is the diagnostic test of choice.¹¹ The pattern and distribution of abnormalities on CT scan can greatly assist with the differential diagnosis in patients presenting with a new ILD, while the presence and pattern of new opacities superimposed on chronic changes can inform the differential and the prognosis of an ILD exacerbation.¹² High-resolution CT provides the most sensitive imaging modality for diffuse ILD. The addition of prone and expiratory images are helpful in differentiating mild lung disease from atelectasis and detecting air trapping, respectively.¹³ However, since pulmonary embolism is a common extraparenchymal finding routinely considered in the differential of a patient presenting with a known or suspected ILD, physicians should consider ordering a CT pulmonary angiogram with additional high-resolution images. Most important, radiographic evaluation should include a review of all available prior chest imaging to assess both the tempo and the nature of radiographic findings.

Bronchoscopy

Bronchoscopy (with bronchoalveolar lavage [BAL], transbronchial lung biopsy [TBLB] and/or transbronchial needle aspiration [TBNA]) is not a routinely used diagnostic tool in the hospitalized ILD patient. However, it should be considered in certain circumstances.⁷ Cell count and differential can be helpful in diagnosing AEP (greater than 40% eosinophilia) or acute HP (greater than 50% lymphocytosis), while the addition of microbiologic and cytologic analysis can assist with the diagnosis of infectious etiologies (including pneumocystis pneumonia) or malignancy.^14,15 Bronchoscopy with BAL has limited sensitivity for many infections and the procedure is associated with a small risk of worsened hypoxemia. Transbronchial lung biopsy, and to a lesser extent TBNA, carry the added risk of pneumothorax and bleeding. In the majority of cases of ILD, TBLB and TBNA have limited diagnostic utility given the small amount of lung tissue sampled. In cases of suspected IPF, where the identification of the histologic pattern is needed for definitive diagnosis, tissue from TBLB cannot be used to make a conclusive diagnosis.^16,17 However, both TBNA and TBLB are useful in the diagnosis of granulomatous disorders, such as sarcoidosis, where the diagnostic yield ranges from 80% to 90% and 50% to 75%, respectively.^18,19

A newer bronchoscopic approach to sampling the lung using a bronchoscopically-placed cryoprobe (termed transbronchial cryobiopsy) has uncertain diagnostic utility and safety in the acute setting. This procedure involves intubation, sedation, and bronchoscopy allowing for the passage of an endobronchial cryoprobe through the bronchoscope and into the periphery of the lung. Several cryobiopies are generally taken from the same pulmonary subsegment. Despite a large number of recent publications on this topic, none of them have provided a clear sense of the diagnostic yield and safety.^20,21 Transbronchial cryobiopsy remains a highly controversial procedure in the clinical setting, and we would not recommend its use until further data are available.²²

Surgical Lung Biopsy

In the outpatient setting, a surgical lung biopsy is often useful when the ILD diagnosis cannot be made from the clinical context and imaging. However, patients presenting with acute respiratory failure from ILD are at greatly increased risk of complications from nonelective biopsy including pneumothorax, hemothorax, acute exacerbation of ILD, ICU admission, mechanical ventilation, and in-hospital mortality.^23,24 Acute histological findings can also make it difficult to appreciate the underlying pattern of fibrosis, reducing the diagnostic utility.^25-27 In our experience, surgical lung biopsy rarely alters the treatment of ILD patients presenting in acute respiratory failure. We believe that surgical lung biopsy should be reserved for the rare hospitalized patients in whom the clinician believes the results would clearly change management and that the substantial risk is worth taking.^5,28

The inpatient management of ILD is a large topic and difficult to comprehensively cover in a single review. Therefore, in this section, we will review 6 key management questions that address both general and specific treatment decisions that frequently arise in the care of hospitalized ILD patients (Figure).

When should hospitalized ILD patients be treated with antibiotics?

Infection and acute presentations of ILD have many similar clinical and radiographic features, making it difficult to distinguish between the two, or exclude infection as the causative role in an acute exacerbation.² In many ILD patients, the risk of infection is higher than in the general population, due to the acute and chronic use of immunosuppression. Until firm guidelines on the use of antibiotics in hospitalized patients with acute respiratory symptoms are available, we recommend considering the empiric use of antibiotics in ILD patients in respiratory failure, in addition to a thorough infectious workup.

When should hospitalized ILD patients be treated with corticosteroids?

Clinical experience supports the use of corticosteroids in the acute management of most rapidly progressive ILDs presenting with respiratory failure, including AEP, COP, acute HP, drug-induced ILD, and some cases of CT-ILD. Patients with AEP tend to respond rapidly to corticosteroids. In a series of 137 patients with AEP, 127 (92%) received corticosteroids, with defervescence and improved dyspnea within 48 to 72 hours and resolution of all symptoms after a median of 7 (4 to 10) days.²⁹ Cryptogenic organizing pneumonia is similarly corticosteroid-responsive, with patients typically started on doses of 1mg/kg of prednisone followed by a slow taper due to the risk of relapse.³⁰ For the majority of acute CT-ILD, oral prednisone is the initial treatment, often in combination with a second immunosuppressive agent such as mycophenolate.

No proven therapies are available for acute exacerbations of IPF (AE-IPF), including the use of corticosteroids. The most recent international guidelines on the management of AE-IPF conditionally recommends the use of corticosteroids, although this recommendation is largely based on anecdotal reports and clearly states that randomized studies are needed.³ When corticosteroids are used, we recommend high doses (eg, 1 to 2 mg/kg of prednisone) with close clinical monitoring. Consider stopping corticosteroids after 3 to 5 days if there is no evidence of clinical improvement. Prolonged courses of corticosteroids should be avoided.

What additional pharmacologic therapies should be considered in the treatment of hospitalized ILD patients?

Immunomodulators. Patients presenting acutely with a new-onset ILD or with an acute exacerbation of a chronic ILD often receive corticosteroids, sometimes in concert with an immunomodulator. This is most commonly seen in the acute management of CTD- ILD and in chronic HP, where mycophenolate mofetil, and to a lesser extent, cyclophosphamide and azathioprine for CT-ILD are used in combination with corticosteroids. The rationale for this is both therapeutic synergy and a desire to limit the long-term exposure to corticosteroids. Similarly, multiple observational cohort studies have investigated the role of combination or tandem immunosuppression in the treatment AE-IPF. Although cyclosporine, cyclophosphamide, azathioprine, rituximab and tacrolimus have all been studied, their efficacy remains uncertain.³ Until these therapies are better studied, they have no routine role in the management of AE-IPF.

Antifibrotics. Nintedanib and pirfenidone are 2 antifibrotic agents approved for the treatment of IPF. Clinical trials suggest that, in addition to slowing disease progression, these therapies may help prevent AE-IPF. The data are most robust in studies of nintedanib. A phase 2 trial with 432 subjects demonstrated a delay in time to the first investigator-reported acute exacerbation.³¹ Two follow-up phase 3 trials showed a reduction in centrally adjudicated AE-IPF in the pooled nintedanib groups compared to placebo.³² An initial phase 2 trial of pirfenidone showed a reduction in acute exacerbations in patients on pirfenidone, but this finding was not replicated in follow-up studies.^33-35 Because of their potential role in preventing acute exacerbations and emerging evidence to suggest that continuation of antifibrotics may lead to better outcomes during an acute exacerbation, these drugs should not generally be stopped during a hospitalization for ILD. However, no evidence supports their initiation during acute exacerbations, and we do not recommend starting antifibrotics in the hospitalized setting for newly diagnosed patients. Starting and stopping antifibrotics should be reserved for outpatient management.

When should noninvasive and mechanical ventilation be considered?

We recommend carefully considering the use of noninvasive ventilation (NIV) and intubation in every ILD patient in respiratory distress, as an acutely reversible process may be present. In patients requiring mechanical ventilation, every effort should be made to minimize potential damage by reducing the fraction of inspired oxygen (to prevent potential hyperoxic injury) and reducing tidal volumes (to minimize barotrauma). Patients with a chronic ILD, particularly IPF, who require NIV or mechanical ventilation will generally have poor outcomes.

Studies suggest that NIV prevents mechanical ventilation in only the minority of patients presenting with an AE-IPF and is associated with high in-hospital mortality and a median survival following hospital discharge of only 60 days.^36-38 The majority of patients with IPF requiring mechanical ventilation will not survive the intensive care unit. In a series of 23 patients presenting with acute respiratory failure and IPF, 22 of the 23 patients died while receiving mechanical ventilation, with a median survival of 3 days. In a more recent study of 34 patients with acute respiratory failure and IPF, 15 subjects underwent mechanical ventilation with an in-hospital mortality rate of 100%.³⁹ Given the overall poor survival associated with AE-IPF, mechanical ventilation should be carefully considered with the patient and family as part of an overall goals-of-care conversation prior to initiation.

When should hospitalized ILD patients be referred for inpatient lung transplant evaluation?

A subset of hospitalized patients with ILD will not respond to supportive and pharmacologic care, particularly those with advanced lung fibrosis. In these cases, lung transplantation may be the only remaining treatment option. This is particularly true for patients presenting with IPF, and it is 1 of the most common indications for lung transplantation. Patients with respiratory failure and ILD should be evaluated early in the hospital course for transplantation or considered for transfer to a transplant center. General contraindications to transplant are age older than 70 years, underweight or elevated BMI (generally higher than 30), malignancy within the last 2 years (with the exception of cutaneous squamous and basal cell tumors), untreatable major organ dysfunction other than the lung, noncurable chronic extrapulmonary infection (chronic active viral hepatitis B, hepatitis C, human immunodeficiency virus), significant chest wall deformity, untreatable psychiatric or psychologic disease, substance addiction within the last 6 months, or lack of dependable social support.⁴⁰ In select patients with ILD and gas exchange abnormalities, mechanical ventilation or extracorporeal membrane oxygenation may be used to bridge a patient to lung transplantation.⁴¹

What should you tell your ILD patient to expect at discharge?

Accurate diagnosis is important not only for acute inpatient management, but for informing long-term prognosis. Acute-onset ILD tends to be more reversible, to be responsive to medical therapy, and to have a more favorable overall outcome. On the other hand, acute exacerbations of established ILD, particularly IPF, can have a more unfavorable and treatment-refractory course. Once a diagnosis is established, it is important both to provide patients with information and ensure appropriate outpatient follow-up. The Pulmonary Fibrosis Foundation (the largest U.S. advocacy and support organization for patients with ILD) provides information on ILD to patients and families and can serve as an important educational source.⁴² Prior to discharge, it is important to evaluate the oxygen needs of patients at rest and with exertion. Referral to an ILD center at discharge is important whenever possible, to monitor clinical symptoms and lung function, initiate or assess response to treatment, and provide supportive care, including oxygen therapy, pulmonary rehabilitation, and outpatient lung transplant referral.

ILD is a group of heterogeneous disorders characterized by lung inflammation and fibrosis. Although the onset of disease is typically insidious, patients can present acutely requiring hospitalization. Inpatient management varies significantly depending on ILD subtype, and, therefore, accurate diagnosis is key in determining treatment and prognosis. As we develop an improved understanding of the mechanisms of acute presentations of ILD, and our approaches to detection and treatment improve as a result of clinical trials, we anticipate continued modifications to this shared framework.

Disclosure

Dr. Collard reports personal fees from Alkermes, aTyr Pharmaceuticals, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Global Blood Therapeutics, Genoa, ImmuneWorks, Moerae Matrix, Navitor, Parexel, Patara, Pharma Capital Partners, Prometic, Takeda, Toray, and Xfibra, outside the submitted work. Drs. Farrand and Shah report no financial conflicts of interest.

Interstitial lung disease (ILD) encompasses a diverse group of disorders that cause inflammation and fibrosis of the lung parenchyma. The clinical manifestations, disease course, management and prognosis of ILD vary depending on the underlying subtype, making accurate classification and diagnosis an important initial step. While a comprehensive list of ILD contains dozens of disorders, the majority of patients will fall into 1 of 3 categories: exposure-related ILD, connective tissue disease-related ILD (CT-ILD), and the idiopathic interstitial pneumonias (Table).

An essential first step in the evaluation of every hospitalized patient with ILD is establishing a diagnosis. A common mistake among clinicians who diagnose patients with ILD is not realizing that ILD is a collection of diseases with different etiologies, natural histories, and treatments. A careful evaluation should be performed in every hospitalized patient with ILD to ensure an accurate diagnosis, ideally in the context of a multidisciplinary conference with pulmonary, radiology, pathology, and other specialties, as appropriate. A multidisciplinary panel of the American Thoracic Society/European Respiratory Society recently published a revised classification of ILD based on a combination of clinical, radiologic, and histopathologic findings, which may aid in refining the diagnosis.¹

There are 3 main scenarios in which the hospital physician will encounter patients with ILD.

Acute presentation of new-onset disease. While many ILDs present insidiously, some cases present acutely and require hospitalization. The most common of these are acute hypersensitivity pneumonitis (HP), CT-ILD (in particular, myositis-related and systemic lupus erythematosus-related), drug-induced ILD (eg, amiodarone, nitrofurantoin), cryptogenic organizing pneumonia (COP), acute eosinophilic pneumonia (AEP), and acute interstitial pneumonia (AIP).

Acute presentation of established (chronic) disease. Patients with chronic forms of ILD can present to the hospital with an acute exacerbation of disease. This can be caused by extra-parenchymal complications, including pulmonary embolism, pneumothorax, and pleural effusion; parenchymal complications such as infectious pneumonia, aspiration pneumonitis, and congestive heart failure; or without an identifiable cause. This latter presentation is most commonly seen in idiopathic pulmonary fibrosis (IPF).^2,3

Elective hospitalization for diagnostic surgical lung biopsy. Patients with ILD may be hospitalized electively for a laparoscopic surgical lung biopsy as part of their diagnostic evaluations.

Physicians caring for a hospitalized ILD patient must be familiar with the clinical presentations, diagnostic approach, medical management, and outpatient follow-up recommended in these 3 settings. We will summarize these areas and provide answers to commonly encountered clinical questions in the hospitalized patient with ILD.

Acute onset (or worsening) of dyspnea is the primary presenting symptom in most patients hospitalized for ILD. This symptom should be further characterized by assessing the degree of dyspnea and the extent of exercise limitation, as both impact overall disease severity and prognosis.⁴ Cough is the second most common symptom, and can be nonproductive, as is common in IPF, or be associated with secretions if parenchymal infection or acute bronchitis is present.⁵ Pleuritic chest pain, pleural effusion, and/or the presence of extrapulmonary features, including dysphagia, joint pain and swelling, or cutaneous thickening may suggest the presence of a CT-ILD. Because most forms of ILD present with only nonspecific symptoms, a careful history and physical examination are essential.

History

A comprehensive patient history is the backbone of diagnosing any ILD. History-taking should focus on severity and temporal progression of symptoms, presence of pre-existing systemic conditions associated with ILD, symptoms of extrapulmonary disease, and exposures to substances that can cause pulmonary injury, including a detailed history of occupations and hobbies, medications, smoking, and familial lung disease.^6-9 Physicians must try to exclude other diagnoses that could result in a similar acute presentation, including congestive heart failure and infection. Considering the complex and extensive recommended history-taking, physicians may find it helpful to use a standardized questionnaire, as provided by the American College of Chest Physicians.¹⁰

Laboratory Testing

All patients presenting to the hospital with a suspected ILD should undergo careful assessment for the presence of connective tissue disease, including patients without clear symptoms because ILD can be the presenting manifestation. We routinely test for antinuclear antibody titer and pattern, rheumatoid factor, anticyclic citrullinated peptide, creatinine kinase, and aldolase as the initial screening panel in most patients, with further testing directed by the findings on history and physical examination. Pulmonary function tests are used routinely to monitor disease progression in the outpatient setting; however, in the hospitalized ILD patient, they are often difficult to perform and have no real diagnostic value. Similarly, arterial blood gas is not routinely used as part of the initial inpatient evaluation.

Imaging

All hospitalized patients with a known or suspected ILD should undergo chest imaging, assuming they are stable enough to do so. While the chest radiograph can provide a low-cost initial assessment of the degree of lung involvement and presence of accompanying abnormalities, computed tomography (CT) scanning is the diagnostic test of choice.¹¹ The pattern and distribution of abnormalities on CT scan can greatly assist with the differential diagnosis in patients presenting with a new ILD, while the presence and pattern of new opacities superimposed on chronic changes can inform the differential and the prognosis of an ILD exacerbation.¹² High-resolution CT provides the most sensitive imaging modality for diffuse ILD. The addition of prone and expiratory images are helpful in differentiating mild lung disease from atelectasis and detecting air trapping, respectively.¹³ However, since pulmonary embolism is a common extraparenchymal finding routinely considered in the differential of a patient presenting with a known or suspected ILD, physicians should consider ordering a CT pulmonary angiogram with additional high-resolution images. Most important, radiographic evaluation should include a review of all available prior chest imaging to assess both the tempo and the nature of radiographic findings.

Bronchoscopy

Bronchoscopy (with bronchoalveolar lavage [BAL], transbronchial lung biopsy [TBLB] and/or transbronchial needle aspiration [TBNA]) is not a routinely used diagnostic tool in the hospitalized ILD patient. However, it should be considered in certain circumstances.⁷ Cell count and differential can be helpful in diagnosing AEP (greater than 40% eosinophilia) or acute HP (greater than 50% lymphocytosis), while the addition of microbiologic and cytologic analysis can assist with the diagnosis of infectious etiologies (including pneumocystis pneumonia) or malignancy.^14,15 Bronchoscopy with BAL has limited sensitivity for many infections and the procedure is associated with a small risk of worsened hypoxemia. Transbronchial lung biopsy, and to a lesser extent TBNA, carry the added risk of pneumothorax and bleeding. In the majority of cases of ILD, TBLB and TBNA have limited diagnostic utility given the small amount of lung tissue sampled. In cases of suspected IPF, where the identification of the histologic pattern is needed for definitive diagnosis, tissue from TBLB cannot be used to make a conclusive diagnosis.^16,17 However, both TBNA and TBLB are useful in the diagnosis of granulomatous disorders, such as sarcoidosis, where the diagnostic yield ranges from 80% to 90% and 50% to 75%, respectively.^18,19

A newer bronchoscopic approach to sampling the lung using a bronchoscopically-placed cryoprobe (termed transbronchial cryobiopsy) has uncertain diagnostic utility and safety in the acute setting. This procedure involves intubation, sedation, and bronchoscopy allowing for the passage of an endobronchial cryoprobe through the bronchoscope and into the periphery of the lung. Several cryobiopies are generally taken from the same pulmonary subsegment. Despite a large number of recent publications on this topic, none of them have provided a clear sense of the diagnostic yield and safety.^20,21 Transbronchial cryobiopsy remains a highly controversial procedure in the clinical setting, and we would not recommend its use until further data are available.²²

Surgical Lung Biopsy

In the outpatient setting, a surgical lung biopsy is often useful when the ILD diagnosis cannot be made from the clinical context and imaging. However, patients presenting with acute respiratory failure from ILD are at greatly increased risk of complications from nonelective biopsy including pneumothorax, hemothorax, acute exacerbation of ILD, ICU admission, mechanical ventilation, and in-hospital mortality.^23,24 Acute histological findings can also make it difficult to appreciate the underlying pattern of fibrosis, reducing the diagnostic utility.^25-27 In our experience, surgical lung biopsy rarely alters the treatment of ILD patients presenting in acute respiratory failure. We believe that surgical lung biopsy should be reserved for the rare hospitalized patients in whom the clinician believes the results would clearly change management and that the substantial risk is worth taking.^5,28

The inpatient management of ILD is a large topic and difficult to comprehensively cover in a single review. Therefore, in this section, we will review 6 key management questions that address both general and specific treatment decisions that frequently arise in the care of hospitalized ILD patients (Figure).

When should hospitalized ILD patients be treated with antibiotics?

Infection and acute presentations of ILD have many similar clinical and radiographic features, making it difficult to distinguish between the two, or exclude infection as the causative role in an acute exacerbation.² In many ILD patients, the risk of infection is higher than in the general population, due to the acute and chronic use of immunosuppression. Until firm guidelines on the use of antibiotics in hospitalized patients with acute respiratory symptoms are available, we recommend considering the empiric use of antibiotics in ILD patients in respiratory failure, in addition to a thorough infectious workup.

When should hospitalized ILD patients be treated with corticosteroids?

Clinical experience supports the use of corticosteroids in the acute management of most rapidly progressive ILDs presenting with respiratory failure, including AEP, COP, acute HP, drug-induced ILD, and some cases of CT-ILD. Patients with AEP tend to respond rapidly to corticosteroids. In a series of 137 patients with AEP, 127 (92%) received corticosteroids, with defervescence and improved dyspnea within 48 to 72 hours and resolution of all symptoms after a median of 7 (4 to 10) days.²⁹ Cryptogenic organizing pneumonia is similarly corticosteroid-responsive, with patients typically started on doses of 1mg/kg of prednisone followed by a slow taper due to the risk of relapse.³⁰ For the majority of acute CT-ILD, oral prednisone is the initial treatment, often in combination with a second immunosuppressive agent such as mycophenolate.

No proven therapies are available for acute exacerbations of IPF (AE-IPF), including the use of corticosteroids. The most recent international guidelines on the management of AE-IPF conditionally recommends the use of corticosteroids, although this recommendation is largely based on anecdotal reports and clearly states that randomized studies are needed.³ When corticosteroids are used, we recommend high doses (eg, 1 to 2 mg/kg of prednisone) with close clinical monitoring. Consider stopping corticosteroids after 3 to 5 days if there is no evidence of clinical improvement. Prolonged courses of corticosteroids should be avoided.

What additional pharmacologic therapies should be considered in the treatment of hospitalized ILD patients?

Immunomodulators. Patients presenting acutely with a new-onset ILD or with an acute exacerbation of a chronic ILD often receive corticosteroids, sometimes in concert with an immunomodulator. This is most commonly seen in the acute management of CTD- ILD and in chronic HP, where mycophenolate mofetil, and to a lesser extent, cyclophosphamide and azathioprine for CT-ILD are used in combination with corticosteroids. The rationale for this is both therapeutic synergy and a desire to limit the long-term exposure to corticosteroids. Similarly, multiple observational cohort studies have investigated the role of combination or tandem immunosuppression in the treatment AE-IPF. Although cyclosporine, cyclophosphamide, azathioprine, rituximab and tacrolimus have all been studied, their efficacy remains uncertain.³ Until these therapies are better studied, they have no routine role in the management of AE-IPF.

Antifibrotics. Nintedanib and pirfenidone are 2 antifibrotic agents approved for the treatment of IPF. Clinical trials suggest that, in addition to slowing disease progression, these therapies may help prevent AE-IPF. The data are most robust in studies of nintedanib. A phase 2 trial with 432 subjects demonstrated a delay in time to the first investigator-reported acute exacerbation.³¹ Two follow-up phase 3 trials showed a reduction in centrally adjudicated AE-IPF in the pooled nintedanib groups compared to placebo.³² An initial phase 2 trial of pirfenidone showed a reduction in acute exacerbations in patients on pirfenidone, but this finding was not replicated in follow-up studies.^33-35 Because of their potential role in preventing acute exacerbations and emerging evidence to suggest that continuation of antifibrotics may lead to better outcomes during an acute exacerbation, these drugs should not generally be stopped during a hospitalization for ILD. However, no evidence supports their initiation during acute exacerbations, and we do not recommend starting antifibrotics in the hospitalized setting for newly diagnosed patients. Starting and stopping antifibrotics should be reserved for outpatient management.

When should noninvasive and mechanical ventilation be considered?

We recommend carefully considering the use of noninvasive ventilation (NIV) and intubation in every ILD patient in respiratory distress, as an acutely reversible process may be present. In patients requiring mechanical ventilation, every effort should be made to minimize potential damage by reducing the fraction of inspired oxygen (to prevent potential hyperoxic injury) and reducing tidal volumes (to minimize barotrauma). Patients with a chronic ILD, particularly IPF, who require NIV or mechanical ventilation will generally have poor outcomes.

Studies suggest that NIV prevents mechanical ventilation in only the minority of patients presenting with an AE-IPF and is associated with high in-hospital mortality and a median survival following hospital discharge of only 60 days.^36-38 The majority of patients with IPF requiring mechanical ventilation will not survive the intensive care unit. In a series of 23 patients presenting with acute respiratory failure and IPF, 22 of the 23 patients died while receiving mechanical ventilation, with a median survival of 3 days. In a more recent study of 34 patients with acute respiratory failure and IPF, 15 subjects underwent mechanical ventilation with an in-hospital mortality rate of 100%.³⁹ Given the overall poor survival associated with AE-IPF, mechanical ventilation should be carefully considered with the patient and family as part of an overall goals-of-care conversation prior to initiation.

When should hospitalized ILD patients be referred for inpatient lung transplant evaluation?

A subset of hospitalized patients with ILD will not respond to supportive and pharmacologic care, particularly those with advanced lung fibrosis. In these cases, lung transplantation may be the only remaining treatment option. This is particularly true for patients presenting with IPF, and it is 1 of the most common indications for lung transplantation. Patients with respiratory failure and ILD should be evaluated early in the hospital course for transplantation or considered for transfer to a transplant center. General contraindications to transplant are age older than 70 years, underweight or elevated BMI (generally higher than 30), malignancy within the last 2 years (with the exception of cutaneous squamous and basal cell tumors), untreatable major organ dysfunction other than the lung, noncurable chronic extrapulmonary infection (chronic active viral hepatitis B, hepatitis C, human immunodeficiency virus), significant chest wall deformity, untreatable psychiatric or psychologic disease, substance addiction within the last 6 months, or lack of dependable social support.⁴⁰ In select patients with ILD and gas exchange abnormalities, mechanical ventilation or extracorporeal membrane oxygenation may be used to bridge a patient to lung transplantation.⁴¹

What should you tell your ILD patient to expect at discharge?

Accurate diagnosis is important not only for acute inpatient management, but for informing long-term prognosis. Acute-onset ILD tends to be more reversible, to be responsive to medical therapy, and to have a more favorable overall outcome. On the other hand, acute exacerbations of established ILD, particularly IPF, can have a more unfavorable and treatment-refractory course. Once a diagnosis is established, it is important both to provide patients with information and ensure appropriate outpatient follow-up. The Pulmonary Fibrosis Foundation (the largest U.S. advocacy and support organization for patients with ILD) provides information on ILD to patients and families and can serve as an important educational source.⁴² Prior to discharge, it is important to evaluate the oxygen needs of patients at rest and with exertion. Referral to an ILD center at discharge is important whenever possible, to monitor clinical symptoms and lung function, initiate or assess response to treatment, and provide supportive care, including oxygen therapy, pulmonary rehabilitation, and outpatient lung transplant referral.

ILD is a group of heterogeneous disorders characterized by lung inflammation and fibrosis. Although the onset of disease is typically insidious, patients can present acutely requiring hospitalization. Inpatient management varies significantly depending on ILD subtype, and, therefore, accurate diagnosis is key in determining treatment and prognosis. As we develop an improved understanding of the mechanisms of acute presentations of ILD, and our approaches to detection and treatment improve as a result of clinical trials, we anticipate continued modifications to this shared framework.

Disclosure

Dr. Collard reports personal fees from Alkermes, aTyr Pharmaceuticals, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Global Blood Therapeutics, Genoa, ImmuneWorks, Moerae Matrix, Navitor, Parexel, Patara, Pharma Capital Partners, Prometic, Takeda, Toray, and Xfibra, outside the submitted work. Drs. Farrand and Shah report no financial conflicts of interest.

The approach to clinical conundrums by an expert clinician is revealed through the presentation of an actual patient’s case in an approach typical of a morning report. Similarly to patient care, sequential pieces of information are provided to the clinician, who is unfamiliar with the case. The focus is on the thought processes of both the clinical team caring for the patient and the discussant.

After losing consciousness at a supermarket, a 70-year-old man was brought to the emergency department by paramedics. He subsequently developed chest pain.

Syncope can be difficult to evaluate, but chest pain may help narrow an otherwise broad differential diagnosis. If this patient has aortic stenosis or hypertrophic cardiomyopathy, effort syncope is the culprit. Cardiac dysrhythmia (eg, ventricular tachycardia), complete heart block, and supraventricular tachycardia each can cause syncope along with chest pain. Myocardial infarction and associated ventricular arrhythmia might also explain both chest pain and syncope. The paramedics might have noted an arrhythmia on the cardiac monitor; if possible, the rhythm strip should be reviewed. A pulmonary embolus can cause chest pain and, if large enough to cause right ventricular compromise, syncope.

According to witnesses at the supermarket, the patient dropped to the ground, lost consciousness, and convulsed for 30 seconds. He had no head trauma, tongue biting, urinary incontinence, or confusion afterward. Electrocardiogram (ECG) performed at the scene showed ST elevations in leads V1 to V3 with ST depressions in the inferior leads. On arrival in the emergency department, the patient described nonradiating substernal chest pressure exacerbated by deep inhalation. The pain did not improve with nitroglycerin. He recalled feeling light-headed before the syncope.

He had not received medical care for 20 years and had no known illnesses other than hypertension. He was not taking any medications. He previously worked as a welder and never smoked tobacco, drank alcohol, or used illicit drugs. The patient’s temperature was 36.4°C. Heart rate was 88 beats per minute, blood pressure 128/72 mm Hg, oxygen saturation 100% on room air, and respiratory rate 22 breaths per minute. The patient had conjunctival pallor. There was a grade 3/6 crescendo- decrescendo systolic murmur loudest at the right upper sternal border without radiation to the carotids. There was no jugular venous distention. Lungs were clear to auscultation bilaterally. There was no peripheral edema, rash, or lymphadenopathy.

Convulsive movements commonly occur during episodes of unconsciousness lasting more than 15 seconds—a phenomenon termed convulsive syncope and often is confused with seizures. These movements are usually clonic jerks of the extremities and trunk and slight twitching of the face, and occasionally tonic extension of the trunk and clenching of the jaw. Absence of tongue biting, urinary incontinence, and confusion in this patient’s case makes seizures less likely.

The distribution of ST segment changes on his ECG are concerning for myocardial infarction in the septal and inferior regions. Right-sided ECG should be performed to assess for right ventricular infarction. Although myocardial ischemia is the primary concern, some features warrant consideration of other etiologies of syncope. First, syncope is an unusual presentation of cardiac ischemia or infarct. The complaint of chest pressure exacerbated by deep inhalation is another atypical feature for myocardial ischemia. Although the patient’s oxygen saturation and heart rate are normal, pulmonary embolism remains a possibility.

The prominent crescendo-decrescendo systolic murmur at the right upper sternal border could indicate aortic stenosis; the carotids should be palpated to assess for pulsus parvus et tardus. A high-flow state associated with anemia could also lead to a midsystolic murmur. Conjunctival pallor typically is seen with hemoglobin levels of 6 g/dL or less. This finding may indicate severe anemia, which has the potential to cause myocardial ischemia and syncope.

Laboratory testing revealed a troponin of 0.04 ng/dL, hemoglobin 4.1 g/dL with MCV of 84.7 fL, white blood cell count 6,500/μL and platelet count 179,000/μL. Serum sodium was 130 mEq/L, urea nitrogen 16 mg/dL, creatinine 1.6 mg/dL, calcium 7.8 mg/dL, total protein 11.4 g/dL (reference range, 6.0-8.2), and albumin 2.2 g/dL. Erythrocyte sedimentation rate (ESR) was 20 mm/h. Serum iron was 48 μg/mL, total iron binding capacity 275 μg/dL, percent iron saturation 17% (reference range, 20-55), and ferritin 10 ng/mL (reference range, 30-400). The international normalized ratio (INR) was 1.5, prothrombin time 15.5 sec (reference range, 9.4-11.6), and partial thromboplastin time 24.7 sec (reference range, 22.9-30.6). Hepatitis C and HIV antibodies were negative as was the urine toxicology screen. Urine protein to creatinine ratio was 0.07. His hemoglobin rose to 7.9 g/dL with transfusion of 4 units packed red blood cells (RBC). His chest pain improved and inferior ST depressions resolved on follow-up ECG. Further history revealed multiple episodes of melena and hematochezia in the preceding weeks without nausea, vomiting, or abdominal pain.

The patient has a strikingly large gamma gap: 9.2. A gap larger than 4 is concerning for the presence of paraproteins. Given the possibility of a paraproteinemia (eg, multiple myeloma, plasmacytoma, Waldenström macroglobulinemia), the first step is to check serum and urine protein electrophoresis. The patient’s anemia is significant and reticulocyte index low. The low ferritin level combined with the inappropriately low reticulocyte count could result from iron deficiency anemia, another bone marrow process, or both. The patient’s syncope likely resulted from severe anemia and hypovolemia associated with hematochezia. The prolonged prothrombin time could be caused by a coagulation factor production problem, from vitamin K deficiency or underlying liver disease, or by a consumptive problem, from low-grade disseminated intravascular coagulation. It is controversial whether inhaling welding fumes causes cancer, but the patient’s age alone makes malignancy a definite possibility.

Aortic stenosis is intriguing in light of the patient’s painless melena and hematochezia. Heyde syndrome is a phenomenon in which high shear stress causes a reduction in the size of von Willebrand factor predisposing to bleeding from submucosal angiodysplasia. However, Heyde syndrome has been reported to occur in the setting of severe or critical aortic stenosis and is unlikely to be the cause here. The patient needs to be examined with both upper and lower endoscopy to rule out gastrointestinal (GI) malignancy, gastric and esophageal varices, and painless peptic ulcers. High right ventricular systolic blood pressures along with echodense material in the inferior vena cava and right atrium suggest the possibility of malignancy with vascular invasion. Tricuspid regurgitation is consistent with high right-sided pressures. As left ventricular ejection fraction is reduced, some of the high right-sided pressures could also be attributable to left heart failure. CT findings of rectal wall thickening and perirectal lymph nodes could be attributable to cancer with locally metastatic disease. Blood loss caused by this cancer would explain the severe anemia on admission as well as the low ferritin level and the iron deficiency anemia. In this 70-year-old man who has not had routine health care maintenance, the leading diagnosis is colorectal cancer. However, the markedly elevated globulin gap and elevated INR strongly suggest another process (eg, multiple myeloma, other paraproteinemia) is also present.

INR remained elevated (1.5) despite vitamin K supplementation. Peripheral blood smear showed hypochromic and normocytic RBCs with moderate rouleaux formation (Figure 1). Intravenous pantoprazole and octreotide were started. Upper and lower endoscopy revealed multiple esophageal erosions and both a polypoid mass and an ulcer within the rectum with evidence of prior bleeding. The mass was resected and the ulcer biopsied.

On hospital  day 3, the patient was transfused another unit of packed RBCs. Hemoglobin level increased from 7.4 g/dL to 8.2 g/dL, but he began to complain of headache, blurred vision, and worsened chest pain. He did not have weakness, numbness, diplopia, dysphagia, or dysarthria. External examination and extraocular movements of both eyes were normal. Visual acuity was 20/25 bilaterally. Funduscopic examination revealed mild dilation of retinal veins and retinal hemorrhages (Figure 2).

Rouleaux formation occurs as excess cathodal proteins, such as immunoglobulins or fibrinogen, adhere to RBCs and cause the cells to stack together in long chains. Classically this is associated with multiple myeloma, but can occur with Waldenström’s macroglobulinemia and other cancers or infections. It can also occur as an artifact in smear preparation but the large globulin gap in this patient supports pathologic rouleaux formation.

Venous retinopathy with hemorrhages may occur with occlusion of the arterial supply (eg, as with carotid artery obstruction) but also with hyperviscosity syndrome (HVS). Vascular disturbances throughout the body play a major role in HVS, but these changes are most easily visualized in the retina. It is interesting that the patient’s headache and blurred vision began after he received additional blood transfusions. Spuriously low hemoglobin and hematocrit levels may stem from increased plasma volume from high immunoglobulin M (IgM) concentrations in Waldenström macroglobulinemia; thus, RBC transfusions can exacerbate symptoms by elevating total RBC mass. Normocytic, normochromic anemia is characteristic of both multiple myeloma and Waldenström’s macroglobulinemia. That the patient’s chest pain recurred coincidentally with blurred vision and headache suggests the likely cause is cardiac ischemia from hyperviscosity. The serum viscosity level should be checked, and, if it is elevated, urgent serum plasmapheresis should be considered. Determining the source of excess globulin production and treating the underlying disease are crucial at this juncture.

In the general population, rectal adenocarcinoma is the most common cause of a rectal mass. In this patient, presence of a paraproteinemia may point to a different diagnosis. Extramedullary colorectal plasmacytoma can occur in the rectum but is exceedingly rare. Waldenström’s macroglobulinemia, a subtype of lymphoplasmacytic lymphoma, can be associated with a rectal lymphoma. At this point, it is not possible to confidently predict the etiology of the mass.

He was started on cyclophosphamide, bortezomib and dexamethasone for IgG κ myeloma with improvement in his headache, blurred vision, chest pain, and plasma viscosity (4 to 1.8). His hemoglobin remained stable at 10 g/dL. Neoadjuvant Capecitabine and radiation therapy were initiated for his rectal cancer.

Multiple myeloma is characterized by monoclonal proliferation of plasma cells, elevated circulating monoclonal immunoglobulin, and end-organ damage.¹ It accounts for approximately 0.8% of all new cancer diagnoses; average age at onset is 70 years. The patient described here had an unusual presentation, with GI bleeding and progression to HVS, and known risk factors for multiple myeloma (male sex, low socioeconomic status, welding career).^2,3

An early clue in the diagnosis was the patient’s large gamma gap and concurrent anemia. Gamma gap, calculated by subtracting serum albumin from serum total protein, is so named because it often reflects an elevated gamma globulin concentration. However, it actually reflects all nonalbumin serum protein. A gamma gap larger than 3.1 g/dL is an independent risk factor for death⁴ and may be associated with infection, autoimmunity, and malignancy. Although there are no screening guidelines for multiple myeloma, 73% of cases are brought to attention by anemia discovered on routine laboratory investigation.⁵ This patient’s lack of prior medical care likely contributed to his atypical presentation. Screening colonoscopy, recommended at age 50, might have identified his rectal cancer at an earlier stage.

The patient’s anemia was likely secondary to GI hemorrhage and bone marrow suppression. His hematochezia might have been partly related to the pathophysiologic interaction of paraproteins with platelets, coagulation factors, and blood vessels.⁶ Amyloidosis of the GI tract is seen in 8% of AL amyloidosis⁷ and most frequently manifests as gastrointestinal bleeding, which is thought to be due to ischemia, vascular friability, or mucosal lesions. It less commonly presents as malabsorption or dysmotility.⁸ Although gastrointestinal amyloid is not typically associated with radiologic abnormalities, occasionally it may cause luminal wall thickening, adenopathy, and inflammatory stranding.⁹ The gold standard for diagnosis is tissue biopsy. However, presence of amyloidosis does not change the overall treatment strategy for multiple myeloma.

An interesting feature of this case is the development of HVS, which typically manifests with mucosal bleeding, blurred vision, and headache.¹⁰ HVS can be diagnosed on retinal examination with findings of venous tortuosity, dilatation, and intraretinal hemorrhage, as occurred in this case,¹¹ and is confirmed with serum viscosity measurement. The first evidence of HVS in this case might have been the spontaneous echo contrast, or “smoke,” detected on echocardiogram. Spontaneous echo contrast represents increased RBC aggregation, from interaction of RBCs and plasma proteins, at low shear rates,¹² and is associated with conditions that result in left atrial stasis, such as atrial arrhythmias and mitral stenosis. This patient did not have valvular pathology or arrhythmia, and thus the “smoke” likely reflected HVS.

Of the paraproteinemias, Waldenström’s macroglobulinemia is most often associated with HVS, likely because of the pentameric structure of IgM¹³ and the consequential large size that predisposes to vascular occlusion. Whereas HVS can occur with IgM levels as low as 3 g/dL, it typically does not occur with IgG concentrations under 15 g/dL. This patient presented with an IgG level of 8 g/dL and developed HVS symptoms only after multiple packed RBC transfusions. Elevated IgG level likely made him susceptible to HVS, which ultimately was precipitated by blood transfusion. Therefore, this patient’s initial chest pain most likely was caused by demand cardiac ischemia secondary to anemia, whereas his subsequent, posttransfusion chest pain likely resulted from hyperviscosity angina. Hyperviscosity angina—cardiac ischemia resulting from poor coronary perfusion caused by hyperviscous blood—has been described in polycythemia and connective tissue disorders.^14,15 To our knowledge, however, hyperviscosity angina has not been reported in patients with multiple myeloma. Treatment of hyperviscosity with end-organ damage typically consists of plasmapheresis, but this patient was started on urgent chemotherapy, and his symptoms improved. Untreated HVS can lead to end-organ ischemia and death.

This patient had a multitude of seemingly disparate symptoms and abnormalities that ultimately were united in a diagnosis of IgG κ multiple myeloma. Subsequently diagnosed rectal adenocarcinoma may have led to ongoing blood loss, which worsened the anemia, but had no evident relation to the primary diagnosis of multiple myeloma. This case exemplifies the fact that HVS is a rare but important iatrogenic complication of multiple myeloma treated with blood transfusion. As this patient’s hospital course progressed, the plot, and his blood, thickened.

• Multiple myeloma is occasionally associated with HVS, which manifests with mucosal bleeding, blurred vision, and headache.
• Hyperviscosity angina—cardiac ischemia resulting from poor coronary perfusion caused by hyperviscous blood—should be considered in patients with paraproteinemias and chest pain.
• Plasmapheresis reverses the clinical manifestations of HVS but not the underlying disease process (eg, Waldenström’s macroglobulinemia, multiple myeloma, leukemia, polycythemia).
• Spontaneous echo contrast represents increased RBC aggregation, from interaction of RBCs and plasma proteins, at low shear rates, and is associated with left atrial stasis, commonly from atrial fibrillation or mitral stenosis, but might be present in HVS.

Acknowledgment

The authors thank Peter Campochiaro, MD, and Whitney Green, MD, for their contributions to the images used in this article.

Disclosure

Dr. Sedighi Manesh is supported by the Jeremiah A. Barondess Fellowship in the Clinical Transaction of the New York Academy of Medicine, in collaboration with the Accreditation Council for Graduate Medical Education (ACGME). The other authors have nothing to report.

The approach to clinical conundrums by an expert clinician is revealed through the presentation of an actual patient’s case in an approach typical of a morning report. Similarly to patient care, sequential pieces of information are provided to the clinician, who is unfamiliar with the case. The focus is on the thought processes of both the clinical team caring for the patient and the discussant.

After losing consciousness at a supermarket, a 70-year-old man was brought to the emergency department by paramedics. He subsequently developed chest pain.

Syncope can be difficult to evaluate, but chest pain may help narrow an otherwise broad differential diagnosis. If this patient has aortic stenosis or hypertrophic cardiomyopathy, effort syncope is the culprit. Cardiac dysrhythmia (eg, ventricular tachycardia), complete heart block, and supraventricular tachycardia each can cause syncope along with chest pain. Myocardial infarction and associated ventricular arrhythmia might also explain both chest pain and syncope. The paramedics might have noted an arrhythmia on the cardiac monitor; if possible, the rhythm strip should be reviewed. A pulmonary embolus can cause chest pain and, if large enough to cause right ventricular compromise, syncope.

According to witnesses at the supermarket, the patient dropped to the ground, lost consciousness, and convulsed for 30 seconds. He had no head trauma, tongue biting, urinary incontinence, or confusion afterward. Electrocardiogram (ECG) performed at the scene showed ST elevations in leads V1 to V3 with ST depressions in the inferior leads. On arrival in the emergency department, the patient described nonradiating substernal chest pressure exacerbated by deep inhalation. The pain did not improve with nitroglycerin. He recalled feeling light-headed before the syncope.

He had not received medical care for 20 years and had no known illnesses other than hypertension. He was not taking any medications. He previously worked as a welder and never smoked tobacco, drank alcohol, or used illicit drugs. The patient’s temperature was 36.4°C. Heart rate was 88 beats per minute, blood pressure 128/72 mm Hg, oxygen saturation 100% on room air, and respiratory rate 22 breaths per minute. The patient had conjunctival pallor. There was a grade 3/6 crescendo- decrescendo systolic murmur loudest at the right upper sternal border without radiation to the carotids. There was no jugular venous distention. Lungs were clear to auscultation bilaterally. There was no peripheral edema, rash, or lymphadenopathy.

Convulsive movements commonly occur during episodes of unconsciousness lasting more than 15 seconds—a phenomenon termed convulsive syncope and often is confused with seizures. These movements are usually clonic jerks of the extremities and trunk and slight twitching of the face, and occasionally tonic extension of the trunk and clenching of the jaw. Absence of tongue biting, urinary incontinence, and confusion in this patient’s case makes seizures less likely.

The distribution of ST segment changes on his ECG are concerning for myocardial infarction in the septal and inferior regions. Right-sided ECG should be performed to assess for right ventricular infarction. Although myocardial ischemia is the primary concern, some features warrant consideration of other etiologies of syncope. First, syncope is an unusual presentation of cardiac ischemia or infarct. The complaint of chest pressure exacerbated by deep inhalation is another atypical feature for myocardial ischemia. Although the patient’s oxygen saturation and heart rate are normal, pulmonary embolism remains a possibility.

The prominent crescendo-decrescendo systolic murmur at the right upper sternal border could indicate aortic stenosis; the carotids should be palpated to assess for pulsus parvus et tardus. A high-flow state associated with anemia could also lead to a midsystolic murmur. Conjunctival pallor typically is seen with hemoglobin levels of 6 g/dL or less. This finding may indicate severe anemia, which has the potential to cause myocardial ischemia and syncope.

Laboratory testing revealed a troponin of 0.04 ng/dL, hemoglobin 4.1 g/dL with MCV of 84.7 fL, white blood cell count 6,500/μL and platelet count 179,000/μL. Serum sodium was 130 mEq/L, urea nitrogen 16 mg/dL, creatinine 1.6 mg/dL, calcium 7.8 mg/dL, total protein 11.4 g/dL (reference range, 6.0-8.2), and albumin 2.2 g/dL. Erythrocyte sedimentation rate (ESR) was 20 mm/h. Serum iron was 48 μg/mL, total iron binding capacity 275 μg/dL, percent iron saturation 17% (reference range, 20-55), and ferritin 10 ng/mL (reference range, 30-400). The international normalized ratio (INR) was 1.5, prothrombin time 15.5 sec (reference range, 9.4-11.6), and partial thromboplastin time 24.7 sec (reference range, 22.9-30.6). Hepatitis C and HIV antibodies were negative as was the urine toxicology screen. Urine protein to creatinine ratio was 0.07. His hemoglobin rose to 7.9 g/dL with transfusion of 4 units packed red blood cells (RBC). His chest pain improved and inferior ST depressions resolved on follow-up ECG. Further history revealed multiple episodes of melena and hematochezia in the preceding weeks without nausea, vomiting, or abdominal pain.

The patient has a strikingly large gamma gap: 9.2. A gap larger than 4 is concerning for the presence of paraproteins. Given the possibility of a paraproteinemia (eg, multiple myeloma, plasmacytoma, Waldenström macroglobulinemia), the first step is to check serum and urine protein electrophoresis. The patient’s anemia is significant and reticulocyte index low. The low ferritin level combined with the inappropriately low reticulocyte count could result from iron deficiency anemia, another bone marrow process, or both. The patient’s syncope likely resulted from severe anemia and hypovolemia associated with hematochezia. The prolonged prothrombin time could be caused by a coagulation factor production problem, from vitamin K deficiency or underlying liver disease, or by a consumptive problem, from low-grade disseminated intravascular coagulation. It is controversial whether inhaling welding fumes causes cancer, but the patient’s age alone makes malignancy a definite possibility.

Aortic stenosis is intriguing in light of the patient’s painless melena and hematochezia. Heyde syndrome is a phenomenon in which high shear stress causes a reduction in the size of von Willebrand factor predisposing to bleeding from submucosal angiodysplasia. However, Heyde syndrome has been reported to occur in the setting of severe or critical aortic stenosis and is unlikely to be the cause here. The patient needs to be examined with both upper and lower endoscopy to rule out gastrointestinal (GI) malignancy, gastric and esophageal varices, and painless peptic ulcers. High right ventricular systolic blood pressures along with echodense material in the inferior vena cava and right atrium suggest the possibility of malignancy with vascular invasion. Tricuspid regurgitation is consistent with high right-sided pressures. As left ventricular ejection fraction is reduced, some of the high right-sided pressures could also be attributable to left heart failure. CT findings of rectal wall thickening and perirectal lymph nodes could be attributable to cancer with locally metastatic disease. Blood loss caused by this cancer would explain the severe anemia on admission as well as the low ferritin level and the iron deficiency anemia. In this 70-year-old man who has not had routine health care maintenance, the leading diagnosis is colorectal cancer. However, the markedly elevated globulin gap and elevated INR strongly suggest another process (eg, multiple myeloma, other paraproteinemia) is also present.

INR remained elevated (1.5) despite vitamin K supplementation. Peripheral blood smear showed hypochromic and normocytic RBCs with moderate rouleaux formation (Figure 1). Intravenous pantoprazole and octreotide were started. Upper and lower endoscopy revealed multiple esophageal erosions and both a polypoid mass and an ulcer within the rectum with evidence of prior bleeding. The mass was resected and the ulcer biopsied.

On hospital  day 3, the patient was transfused another unit of packed RBCs. Hemoglobin level increased from 7.4 g/dL to 8.2 g/dL, but he began to complain of headache, blurred vision, and worsened chest pain. He did not have weakness, numbness, diplopia, dysphagia, or dysarthria. External examination and extraocular movements of both eyes were normal. Visual acuity was 20/25 bilaterally. Funduscopic examination revealed mild dilation of retinal veins and retinal hemorrhages (Figure 2).

Rouleaux formation occurs as excess cathodal proteins, such as immunoglobulins or fibrinogen, adhere to RBCs and cause the cells to stack together in long chains. Classically this is associated with multiple myeloma, but can occur with Waldenström’s macroglobulinemia and other cancers or infections. It can also occur as an artifact in smear preparation but the large globulin gap in this patient supports pathologic rouleaux formation.

Venous retinopathy with hemorrhages may occur with occlusion of the arterial supply (eg, as with carotid artery obstruction) but also with hyperviscosity syndrome (HVS). Vascular disturbances throughout the body play a major role in HVS, but these changes are most easily visualized in the retina. It is interesting that the patient’s headache and blurred vision began after he received additional blood transfusions. Spuriously low hemoglobin and hematocrit levels may stem from increased plasma volume from high immunoglobulin M (IgM) concentrations in Waldenström macroglobulinemia; thus, RBC transfusions can exacerbate symptoms by elevating total RBC mass. Normocytic, normochromic anemia is characteristic of both multiple myeloma and Waldenström’s macroglobulinemia. That the patient’s chest pain recurred coincidentally with blurred vision and headache suggests the likely cause is cardiac ischemia from hyperviscosity. The serum viscosity level should be checked, and, if it is elevated, urgent serum plasmapheresis should be considered. Determining the source of excess globulin production and treating the underlying disease are crucial at this juncture.

In the general population, rectal adenocarcinoma is the most common cause of a rectal mass. In this patient, presence of a paraproteinemia may point to a different diagnosis. Extramedullary colorectal plasmacytoma can occur in the rectum but is exceedingly rare. Waldenström’s macroglobulinemia, a subtype of lymphoplasmacytic lymphoma, can be associated with a rectal lymphoma. At this point, it is not possible to confidently predict the etiology of the mass.

He was started on cyclophosphamide, bortezomib and dexamethasone for IgG κ myeloma with improvement in his headache, blurred vision, chest pain, and plasma viscosity (4 to 1.8). His hemoglobin remained stable at 10 g/dL. Neoadjuvant Capecitabine and radiation therapy were initiated for his rectal cancer.

Multiple myeloma is characterized by monoclonal proliferation of plasma cells, elevated circulating monoclonal immunoglobulin, and end-organ damage.¹ It accounts for approximately 0.8% of all new cancer diagnoses; average age at onset is 70 years. The patient described here had an unusual presentation, with GI bleeding and progression to HVS, and known risk factors for multiple myeloma (male sex, low socioeconomic status, welding career).^2,3

An early clue in the diagnosis was the patient’s large gamma gap and concurrent anemia. Gamma gap, calculated by subtracting serum albumin from serum total protein, is so named because it often reflects an elevated gamma globulin concentration. However, it actually reflects all nonalbumin serum protein. A gamma gap larger than 3.1 g/dL is an independent risk factor for death⁴ and may be associated with infection, autoimmunity, and malignancy. Although there are no screening guidelines for multiple myeloma, 73% of cases are brought to attention by anemia discovered on routine laboratory investigation.⁵ This patient’s lack of prior medical care likely contributed to his atypical presentation. Screening colonoscopy, recommended at age 50, might have identified his rectal cancer at an earlier stage.

The patient’s anemia was likely secondary to GI hemorrhage and bone marrow suppression. His hematochezia might have been partly related to the pathophysiologic interaction of paraproteins with platelets, coagulation factors, and blood vessels.⁶ Amyloidosis of the GI tract is seen in 8% of AL amyloidosis⁷ and most frequently manifests as gastrointestinal bleeding, which is thought to be due to ischemia, vascular friability, or mucosal lesions. It less commonly presents as malabsorption or dysmotility.⁸ Although gastrointestinal amyloid is not typically associated with radiologic abnormalities, occasionally it may cause luminal wall thickening, adenopathy, and inflammatory stranding.⁹ The gold standard for diagnosis is tissue biopsy. However, presence of amyloidosis does not change the overall treatment strategy for multiple myeloma.

An interesting feature of this case is the development of HVS, which typically manifests with mucosal bleeding, blurred vision, and headache.¹⁰ HVS can be diagnosed on retinal examination with findings of venous tortuosity, dilatation, and intraretinal hemorrhage, as occurred in this case,¹¹ and is confirmed with serum viscosity measurement. The first evidence of HVS in this case might have been the spontaneous echo contrast, or “smoke,” detected on echocardiogram. Spontaneous echo contrast represents increased RBC aggregation, from interaction of RBCs and plasma proteins, at low shear rates,¹² and is associated with conditions that result in left atrial stasis, such as atrial arrhythmias and mitral stenosis. This patient did not have valvular pathology or arrhythmia, and thus the “smoke” likely reflected HVS.

Of the paraproteinemias, Waldenström’s macroglobulinemia is most often associated with HVS, likely because of the pentameric structure of IgM¹³ and the consequential large size that predisposes to vascular occlusion. Whereas HVS can occur with IgM levels as low as 3 g/dL, it typically does not occur with IgG concentrations under 15 g/dL. This patient presented with an IgG level of 8 g/dL and developed HVS symptoms only after multiple packed RBC transfusions. Elevated IgG level likely made him susceptible to HVS, which ultimately was precipitated by blood transfusion. Therefore, this patient’s initial chest pain most likely was caused by demand cardiac ischemia secondary to anemia, whereas his subsequent, posttransfusion chest pain likely resulted from hyperviscosity angina. Hyperviscosity angina—cardiac ischemia resulting from poor coronary perfusion caused by hyperviscous blood—has been described in polycythemia and connective tissue disorders.^14,15 To our knowledge, however, hyperviscosity angina has not been reported in patients with multiple myeloma. Treatment of hyperviscosity with end-organ damage typically consists of plasmapheresis, but this patient was started on urgent chemotherapy, and his symptoms improved. Untreated HVS can lead to end-organ ischemia and death.

This patient had a multitude of seemingly disparate symptoms and abnormalities that ultimately were united in a diagnosis of IgG κ multiple myeloma. Subsequently diagnosed rectal adenocarcinoma may have led to ongoing blood loss, which worsened the anemia, but had no evident relation to the primary diagnosis of multiple myeloma. This case exemplifies the fact that HVS is a rare but important iatrogenic complication of multiple myeloma treated with blood transfusion. As this patient’s hospital course progressed, the plot, and his blood, thickened.

• Multiple myeloma is occasionally associated with HVS, which manifests with mucosal bleeding, blurred vision, and headache.
• Hyperviscosity angina—cardiac ischemia resulting from poor coronary perfusion caused by hyperviscous blood—should be considered in patients with paraproteinemias and chest pain.
• Plasmapheresis reverses the clinical manifestations of HVS but not the underlying disease process (eg, Waldenström’s macroglobulinemia, multiple myeloma, leukemia, polycythemia).
• Spontaneous echo contrast represents increased RBC aggregation, from interaction of RBCs and plasma proteins, at low shear rates, and is associated with left atrial stasis, commonly from atrial fibrillation or mitral stenosis, but might be present in HVS.

Acknowledgment

The authors thank Peter Campochiaro, MD, and Whitney Green, MD, for their contributions to the images used in this article.

Disclosure

Dr. Sedighi Manesh is supported by the Jeremiah A. Barondess Fellowship in the Clinical Transaction of the New York Academy of Medicine, in collaboration with the Accreditation Council for Graduate Medical Education (ACGME). The other authors have nothing to report.

The approach to clinical conundrums by an expert clinician is revealed through the presentation of an actual patient’s case in an approach typical of a morning report. Similarly to patient care, sequential pieces of information are provided to the clinician, who is unfamiliar with the case. The focus is on the thought processes of both the clinical team caring for the patient and the discussant.

After losing consciousness at a supermarket, a 70-year-old man was brought to the emergency department by paramedics. He subsequently developed chest pain.

Syncope can be difficult to evaluate, but chest pain may help narrow an otherwise broad differential diagnosis. If this patient has aortic stenosis or hypertrophic cardiomyopathy, effort syncope is the culprit. Cardiac dysrhythmia (eg, ventricular tachycardia), complete heart block, and supraventricular tachycardia each can cause syncope along with chest pain. Myocardial infarction and associated ventricular arrhythmia might also explain both chest pain and syncope. The paramedics might have noted an arrhythmia on the cardiac monitor; if possible, the rhythm strip should be reviewed. A pulmonary embolus can cause chest pain and, if large enough to cause right ventricular compromise, syncope.

According to witnesses at the supermarket, the patient dropped to the ground, lost consciousness, and convulsed for 30 seconds. He had no head trauma, tongue biting, urinary incontinence, or confusion afterward. Electrocardiogram (ECG) performed at the scene showed ST elevations in leads V1 to V3 with ST depressions in the inferior leads. On arrival in the emergency department, the patient described nonradiating substernal chest pressure exacerbated by deep inhalation. The pain did not improve with nitroglycerin. He recalled feeling light-headed before the syncope.

He had not received medical care for 20 years and had no known illnesses other than hypertension. He was not taking any medications. He previously worked as a welder and never smoked tobacco, drank alcohol, or used illicit drugs. The patient’s temperature was 36.4°C. Heart rate was 88 beats per minute, blood pressure 128/72 mm Hg, oxygen saturation 100% on room air, and respiratory rate 22 breaths per minute. The patient had conjunctival pallor. There was a grade 3/6 crescendo- decrescendo systolic murmur loudest at the right upper sternal border without radiation to the carotids. There was no jugular venous distention. Lungs were clear to auscultation bilaterally. There was no peripheral edema, rash, or lymphadenopathy.

Convulsive movements commonly occur during episodes of unconsciousness lasting more than 15 seconds—a phenomenon termed convulsive syncope and often is confused with seizures. These movements are usually clonic jerks of the extremities and trunk and slight twitching of the face, and occasionally tonic extension of the trunk and clenching of the jaw. Absence of tongue biting, urinary incontinence, and confusion in this patient’s case makes seizures less likely.

The distribution of ST segment changes on his ECG are concerning for myocardial infarction in the septal and inferior regions. Right-sided ECG should be performed to assess for right ventricular infarction. Although myocardial ischemia is the primary concern, some features warrant consideration of other etiologies of syncope. First, syncope is an unusual presentation of cardiac ischemia or infarct. The complaint of chest pressure exacerbated by deep inhalation is another atypical feature for myocardial ischemia. Although the patient’s oxygen saturation and heart rate are normal, pulmonary embolism remains a possibility.

The prominent crescendo-decrescendo systolic murmur at the right upper sternal border could indicate aortic stenosis; the carotids should be palpated to assess for pulsus parvus et tardus. A high-flow state associated with anemia could also lead to a midsystolic murmur. Conjunctival pallor typically is seen with hemoglobin levels of 6 g/dL or less. This finding may indicate severe anemia, which has the potential to cause myocardial ischemia and syncope.

Laboratory testing revealed a troponin of 0.04 ng/dL, hemoglobin 4.1 g/dL with MCV of 84.7 fL, white blood cell count 6,500/μL and platelet count 179,000/μL. Serum sodium was 130 mEq/L, urea nitrogen 16 mg/dL, creatinine 1.6 mg/dL, calcium 7.8 mg/dL, total protein 11.4 g/dL (reference range, 6.0-8.2), and albumin 2.2 g/dL. Erythrocyte sedimentation rate (ESR) was 20 mm/h. Serum iron was 48 μg/mL, total iron binding capacity 275 μg/dL, percent iron saturation 17% (reference range, 20-55), and ferritin 10 ng/mL (reference range, 30-400). The international normalized ratio (INR) was 1.5, prothrombin time 15.5 sec (reference range, 9.4-11.6), and partial thromboplastin time 24.7 sec (reference range, 22.9-30.6). Hepatitis C and HIV antibodies were negative as was the urine toxicology screen. Urine protein to creatinine ratio was 0.07. His hemoglobin rose to 7.9 g/dL with transfusion of 4 units packed red blood cells (RBC). His chest pain improved and inferior ST depressions resolved on follow-up ECG. Further history revealed multiple episodes of melena and hematochezia in the preceding weeks without nausea, vomiting, or abdominal pain.

The patient has a strikingly large gamma gap: 9.2. A gap larger than 4 is concerning for the presence of paraproteins. Given the possibility of a paraproteinemia (eg, multiple myeloma, plasmacytoma, Waldenström macroglobulinemia), the first step is to check serum and urine protein electrophoresis. The patient’s anemia is significant and reticulocyte index low. The low ferritin level combined with the inappropriately low reticulocyte count could result from iron deficiency anemia, another bone marrow process, or both. The patient’s syncope likely resulted from severe anemia and hypovolemia associated with hematochezia. The prolonged prothrombin time could be caused by a coagulation factor production problem, from vitamin K deficiency or underlying liver disease, or by a consumptive problem, from low-grade disseminated intravascular coagulation. It is controversial whether inhaling welding fumes causes cancer, but the patient’s age alone makes malignancy a definite possibility.

Aortic stenosis is intriguing in light of the patient’s painless melena and hematochezia. Heyde syndrome is a phenomenon in which high shear stress causes a reduction in the size of von Willebrand factor predisposing to bleeding from submucosal angiodysplasia. However, Heyde syndrome has been reported to occur in the setting of severe or critical aortic stenosis and is unlikely to be the cause here. The patient needs to be examined with both upper and lower endoscopy to rule out gastrointestinal (GI) malignancy, gastric and esophageal varices, and painless peptic ulcers. High right ventricular systolic blood pressures along with echodense material in the inferior vena cava and right atrium suggest the possibility of malignancy with vascular invasion. Tricuspid regurgitation is consistent with high right-sided pressures. As left ventricular ejection fraction is reduced, some of the high right-sided pressures could also be attributable to left heart failure. CT findings of rectal wall thickening and perirectal lymph nodes could be attributable to cancer with locally metastatic disease. Blood loss caused by this cancer would explain the severe anemia on admission as well as the low ferritin level and the iron deficiency anemia. In this 70-year-old man who has not had routine health care maintenance, the leading diagnosis is colorectal cancer. However, the markedly elevated globulin gap and elevated INR strongly suggest another process (eg, multiple myeloma, other paraproteinemia) is also present.

INR remained elevated (1.5) despite vitamin K supplementation. Peripheral blood smear showed hypochromic and normocytic RBCs with moderate rouleaux formation (Figure 1). Intravenous pantoprazole and octreotide were started. Upper and lower endoscopy revealed multiple esophageal erosions and both a polypoid mass and an ulcer within the rectum with evidence of prior bleeding. The mass was resected and the ulcer biopsied.

On hospital  day 3, the patient was transfused another unit of packed RBCs. Hemoglobin level increased from 7.4 g/dL to 8.2 g/dL, but he began to complain of headache, blurred vision, and worsened chest pain. He did not have weakness, numbness, diplopia, dysphagia, or dysarthria. External examination and extraocular movements of both eyes were normal. Visual acuity was 20/25 bilaterally. Funduscopic examination revealed mild dilation of retinal veins and retinal hemorrhages (Figure 2).

Rouleaux formation occurs as excess cathodal proteins, such as immunoglobulins or fibrinogen, adhere to RBCs and cause the cells to stack together in long chains. Classically this is associated with multiple myeloma, but can occur with Waldenström’s macroglobulinemia and other cancers or infections. It can also occur as an artifact in smear preparation but the large globulin gap in this patient supports pathologic rouleaux formation.

Venous retinopathy with hemorrhages may occur with occlusion of the arterial supply (eg, as with carotid artery obstruction) but also with hyperviscosity syndrome (HVS). Vascular disturbances throughout the body play a major role in HVS, but these changes are most easily visualized in the retina. It is interesting that the patient’s headache and blurred vision began after he received additional blood transfusions. Spuriously low hemoglobin and hematocrit levels may stem from increased plasma volume from high immunoglobulin M (IgM) concentrations in Waldenström macroglobulinemia; thus, RBC transfusions can exacerbate symptoms by elevating total RBC mass. Normocytic, normochromic anemia is characteristic of both multiple myeloma and Waldenström’s macroglobulinemia. That the patient’s chest pain recurred coincidentally with blurred vision and headache suggests the likely cause is cardiac ischemia from hyperviscosity. The serum viscosity level should be checked, and, if it is elevated, urgent serum plasmapheresis should be considered. Determining the source of excess globulin production and treating the underlying disease are crucial at this juncture.

In the general population, rectal adenocarcinoma is the most common cause of a rectal mass. In this patient, presence of a paraproteinemia may point to a different diagnosis. Extramedullary colorectal plasmacytoma can occur in the rectum but is exceedingly rare. Waldenström’s macroglobulinemia, a subtype of lymphoplasmacytic lymphoma, can be associated with a rectal lymphoma. At this point, it is not possible to confidently predict the etiology of the mass.

He was started on cyclophosphamide, bortezomib and dexamethasone for IgG κ myeloma with improvement in his headache, blurred vision, chest pain, and plasma viscosity (4 to 1.8). His hemoglobin remained stable at 10 g/dL. Neoadjuvant Capecitabine and radiation therapy were initiated for his rectal cancer.

Multiple myeloma is characterized by monoclonal proliferation of plasma cells, elevated circulating monoclonal immunoglobulin, and end-organ damage.¹ It accounts for approximately 0.8% of all new cancer diagnoses; average age at onset is 70 years. The patient described here had an unusual presentation, with GI bleeding and progression to HVS, and known risk factors for multiple myeloma (male sex, low socioeconomic status, welding career).^2,3

An early clue in the diagnosis was the patient’s large gamma gap and concurrent anemia. Gamma gap, calculated by subtracting serum albumin from serum total protein, is so named because it often reflects an elevated gamma globulin concentration. However, it actually reflects all nonalbumin serum protein. A gamma gap larger than 3.1 g/dL is an independent risk factor for death⁴ and may be associated with infection, autoimmunity, and malignancy. Although there are no screening guidelines for multiple myeloma, 73% of cases are brought to attention by anemia discovered on routine laboratory investigation.⁵ This patient’s lack of prior medical care likely contributed to his atypical presentation. Screening colonoscopy, recommended at age 50, might have identified his rectal cancer at an earlier stage.

The patient’s anemia was likely secondary to GI hemorrhage and bone marrow suppression. His hematochezia might have been partly related to the pathophysiologic interaction of paraproteins with platelets, coagulation factors, and blood vessels.⁶ Amyloidosis of the GI tract is seen in 8% of AL amyloidosis⁷ and most frequently manifests as gastrointestinal bleeding, which is thought to be due to ischemia, vascular friability, or mucosal lesions. It less commonly presents as malabsorption or dysmotility.⁸ Although gastrointestinal amyloid is not typically associated with radiologic abnormalities, occasionally it may cause luminal wall thickening, adenopathy, and inflammatory stranding.⁹ The gold standard for diagnosis is tissue biopsy. However, presence of amyloidosis does not change the overall treatment strategy for multiple myeloma.

An interesting feature of this case is the development of HVS, which typically manifests with mucosal bleeding, blurred vision, and headache.¹⁰ HVS can be diagnosed on retinal examination with findings of venous tortuosity, dilatation, and intraretinal hemorrhage, as occurred in this case,¹¹ and is confirmed with serum viscosity measurement. The first evidence of HVS in this case might have been the spontaneous echo contrast, or “smoke,” detected on echocardiogram. Spontaneous echo contrast represents increased RBC aggregation, from interaction of RBCs and plasma proteins, at low shear rates,¹² and is associated with conditions that result in left atrial stasis, such as atrial arrhythmias and mitral stenosis. This patient did not have valvular pathology or arrhythmia, and thus the “smoke” likely reflected HVS.

Of the paraproteinemias, Waldenström’s macroglobulinemia is most often associated with HVS, likely because of the pentameric structure of IgM¹³ and the consequential large size that predisposes to vascular occlusion. Whereas HVS can occur with IgM levels as low as 3 g/dL, it typically does not occur with IgG concentrations under 15 g/dL. This patient presented with an IgG level of 8 g/dL and developed HVS symptoms only after multiple packed RBC transfusions. Elevated IgG level likely made him susceptible to HVS, which ultimately was precipitated by blood transfusion. Therefore, this patient’s initial chest pain most likely was caused by demand cardiac ischemia secondary to anemia, whereas his subsequent, posttransfusion chest pain likely resulted from hyperviscosity angina. Hyperviscosity angina—cardiac ischemia resulting from poor coronary perfusion caused by hyperviscous blood—has been described in polycythemia and connective tissue disorders.^14,15 To our knowledge, however, hyperviscosity angina has not been reported in patients with multiple myeloma. Treatment of hyperviscosity with end-organ damage typically consists of plasmapheresis, but this patient was started on urgent chemotherapy, and his symptoms improved. Untreated HVS can lead to end-organ ischemia and death.

This patient had a multitude of seemingly disparate symptoms and abnormalities that ultimately were united in a diagnosis of IgG κ multiple myeloma. Subsequently diagnosed rectal adenocarcinoma may have led to ongoing blood loss, which worsened the anemia, but had no evident relation to the primary diagnosis of multiple myeloma. This case exemplifies the fact that HVS is a rare but important iatrogenic complication of multiple myeloma treated with blood transfusion. As this patient’s hospital course progressed, the plot, and his blood, thickened.

• Multiple myeloma is occasionally associated with HVS, which manifests with mucosal bleeding, blurred vision, and headache.
• Hyperviscosity angina—cardiac ischemia resulting from poor coronary perfusion caused by hyperviscous blood—should be considered in patients with paraproteinemias and chest pain.
• Plasmapheresis reverses the clinical manifestations of HVS but not the underlying disease process (eg, Waldenström’s macroglobulinemia, multiple myeloma, leukemia, polycythemia).
• Spontaneous echo contrast represents increased RBC aggregation, from interaction of RBCs and plasma proteins, at low shear rates, and is associated with left atrial stasis, commonly from atrial fibrillation or mitral stenosis, but might be present in HVS.

Acknowledgment

The authors thank Peter Campochiaro, MD, and Whitney Green, MD, for their contributions to the images used in this article.

Disclosure

Dr. Sedighi Manesh is supported by the Jeremiah A. Barondess Fellowship in the Clinical Transaction of the New York Academy of Medicine, in collaboration with the Accreditation Council for Graduate Medical Education (ACGME). The other authors have nothing to report.