Federal Practitioner

Plantar hyperpigmentation (also known as plantar melanosis [increased melanin], volar pigmented macules, benign racial melanosis, acral pigmentation, acral ethnic melanosis, or mottled hyperpigmentation of the plantar surface) is a benign finding in many individuals and is especially prevalent in those with darker skin tones. Acral refers to manifestation on the hands and feet, volar on the palms and soles, and plantar on the soles only. Here, we focus on plantar hyper-pigmentation. We use the terms ethnic and racial interchangeably.

It is critically important to differentiate benign hyperpigmentation, which is common in patients with skin of color, from melanoma. Although rare, Black patients in the United States experience high morbidity and mortality from acral melanoma, which often is diagnosed late in the disease course.¹

There are many causes of hyperpigmentation on the plantar surfaces, including benign ethnic melanosis, nevi, melanoma, infections such as syphilis and tinea nigra, conditions such as Peutz-Jeghers syndrome and Laugier-Hunziker syndrome, and postinflammatory hyperpigmentation secondary to atopic dermatitis and psoriasis. We focus on the most common causes, ethnic melanosis and nevi, as well as melanoma, which is the deadliest cause.

Epidemiology

In a 1980 study (N=251), Black Americans had a high incidence of plantar hyperpigmentation, with 52% of affected patients having dark brown skin and 31% having light brown skin.²

The epidemiology of melanoma varies by race/ethnicity. Melanoma in Black individuals is relatively rare, with an annual incidence of approximately 1 in 100,000 individuals.³ However, when individuals with skin of color develop melanoma, they are more likely than their White counterparts to have acral melanoma (acral lentiginous melanoma), one of the deadliest types.¹ In a case series of Black patients with melanoma (N=48) from 2 tertiary care centers in Texas, 30 of 40 primary cutaneous melanomas (75%) were located on acral skin.⁴ Overall, 13 patients developed stage IV disease and 12 died due to disease progression. All patients who developed distant metastases or died of melanoma had acral melanoma.⁴ Individuals of Asian descent also have a high incidence of acral melanoma, as shown in research from Japan.^5-9

Key Clinical Features in Individuals With Darker Skin Tones

Dermoscopy is an evidence-based clinical examination method for earlier diagnosis of cutaneous melanoma, including on acral skin.^10,11 Benign nevi on the volar skin as well as the palms and soles tend to have one of these 3 dermoscopic patterns: parallel furrow, lattice, or irregular fibrillar. The pattern that is most predictive of volar melanoma is the parallel ridge pattern (PRP) (Figures A and B [insets]), which showed a high specificity (99.0%) and very high negative predictive value (97.7%) for malignant melanoma in a Japanese population.⁷ The PRP data from this study cannot be applied reliably to Black individuals, especially because benign ethnic melanosis and other benign conditions can demonstrate PRP.¹² Reliance on the PRP as a diagnostic clue could result in unneccessary biopsies in as many as 50% of Black patients with benign plantar hyperpigmentation.² Furthermore, biopsies of the plantar surface can be painful and cause pain while walking.

It has been suggested that PRP seen on dermoscopy in benign hyperpigmentation such as ethnic melanosis and nevi may preserve the acrosyringia (eccrine gland openings on the ridge), whereas PRP in melanoma may obliterate the acrosyringia.¹³ This observation is based on case reports only and needs further study. However, if validated, it could be a useful diagnostic clue.

In a retrospective cohort study of skin cancer in Black individuals (n=165) at a New York City–based cancer center from 2000 to 2020, 68% of patients were diagnosed with melanomas—80% were the acral subtype and 75% displayed a PRP. However, the surrounding uninvolved background skin, which was visible in most cases, also demonstrated a PRP.¹⁴ Because of the high morbidity and mortality rates of acral melanoma, clinicians should biopsy or immediately refer patients with concerning plantar hyperpigmentation to a dermatologist.

Health disparity highlight

The mortality rate for acral melanoma in Black patients is disproportionately high for the following reasons^15,16:

• Patients and health care providers do not expect to see melanoma in Black patients (it truly is rare!), so screening and education on sun protection are limited.

• Benign ethnic melanosis makes it more difficult to distinguish between early acral melanoma and benign skin changes.

• Black patients and other US patient populations with skin of color may be less likely to have health insurance, which contributes to inequities in access to health care. As of 2022, the uninsured rates for nonelderly American Indian and Alaska Native, Hispanic, Native Hawaiian and Other Pacific Islander, Black, and White individuals were 19.1%, 18.0%, 12.7%, 10.0%, and 6.6%, respectively.¹⁷

Multi-institutional registries could improve understanding of acral melanoma in Black patients.⁴ More studies are needed to help differentiate between the dermoscopic finding of PRP in benign ethnic melanosis vs malignant melanoma.

Plantar hyperpigmentation (also known as plantar melanosis [increased melanin], volar pigmented macules, benign racial melanosis, acral pigmentation, acral ethnic melanosis, or mottled hyperpigmentation of the plantar surface) is a benign finding in many individuals and is especially prevalent in those with darker skin tones. Acral refers to manifestation on the hands and feet, volar on the palms and soles, and plantar on the soles only. Here, we focus on plantar hyper-pigmentation. We use the terms ethnic and racial interchangeably.

It is critically important to differentiate benign hyperpigmentation, which is common in patients with skin of color, from melanoma. Although rare, Black patients in the United States experience high morbidity and mortality from acral melanoma, which often is diagnosed late in the disease course.¹

There are many causes of hyperpigmentation on the plantar surfaces, including benign ethnic melanosis, nevi, melanoma, infections such as syphilis and tinea nigra, conditions such as Peutz-Jeghers syndrome and Laugier-Hunziker syndrome, and postinflammatory hyperpigmentation secondary to atopic dermatitis and psoriasis. We focus on the most common causes, ethnic melanosis and nevi, as well as melanoma, which is the deadliest cause.

Epidemiology

In a 1980 study (N=251), Black Americans had a high incidence of plantar hyperpigmentation, with 52% of affected patients having dark brown skin and 31% having light brown skin.²

The epidemiology of melanoma varies by race/ethnicity. Melanoma in Black individuals is relatively rare, with an annual incidence of approximately 1 in 100,000 individuals.³ However, when individuals with skin of color develop melanoma, they are more likely than their White counterparts to have acral melanoma (acral lentiginous melanoma), one of the deadliest types.¹ In a case series of Black patients with melanoma (N=48) from 2 tertiary care centers in Texas, 30 of 40 primary cutaneous melanomas (75%) were located on acral skin.⁴ Overall, 13 patients developed stage IV disease and 12 died due to disease progression. All patients who developed distant metastases or died of melanoma had acral melanoma.⁴ Individuals of Asian descent also have a high incidence of acral melanoma, as shown in research from Japan.^5-9

Key Clinical Features in Individuals With Darker Skin Tones

Dermoscopy is an evidence-based clinical examination method for earlier diagnosis of cutaneous melanoma, including on acral skin.^10,11 Benign nevi on the volar skin as well as the palms and soles tend to have one of these 3 dermoscopic patterns: parallel furrow, lattice, or irregular fibrillar. The pattern that is most predictive of volar melanoma is the parallel ridge pattern (PRP) (Figures A and B [insets]), which showed a high specificity (99.0%) and very high negative predictive value (97.7%) for malignant melanoma in a Japanese population.⁷ The PRP data from this study cannot be applied reliably to Black individuals, especially because benign ethnic melanosis and other benign conditions can demonstrate PRP.¹² Reliance on the PRP as a diagnostic clue could result in unneccessary biopsies in as many as 50% of Black patients with benign plantar hyperpigmentation.² Furthermore, biopsies of the plantar surface can be painful and cause pain while walking.

It has been suggested that PRP seen on dermoscopy in benign hyperpigmentation such as ethnic melanosis and nevi may preserve the acrosyringia (eccrine gland openings on the ridge), whereas PRP in melanoma may obliterate the acrosyringia.¹³ This observation is based on case reports only and needs further study. However, if validated, it could be a useful diagnostic clue.

In a retrospective cohort study of skin cancer in Black individuals (n=165) at a New York City–based cancer center from 2000 to 2020, 68% of patients were diagnosed with melanomas—80% were the acral subtype and 75% displayed a PRP. However, the surrounding uninvolved background skin, which was visible in most cases, also demonstrated a PRP.¹⁴ Because of the high morbidity and mortality rates of acral melanoma, clinicians should biopsy or immediately refer patients with concerning plantar hyperpigmentation to a dermatologist.

Health disparity highlight

The mortality rate for acral melanoma in Black patients is disproportionately high for the following reasons^15,16:

• Patients and health care providers do not expect to see melanoma in Black patients (it truly is rare!), so screening and education on sun protection are limited.

• Benign ethnic melanosis makes it more difficult to distinguish between early acral melanoma and benign skin changes.

• Black patients and other US patient populations with skin of color may be less likely to have health insurance, which contributes to inequities in access to health care. As of 2022, the uninsured rates for nonelderly American Indian and Alaska Native, Hispanic, Native Hawaiian and Other Pacific Islander, Black, and White individuals were 19.1%, 18.0%, 12.7%, 10.0%, and 6.6%, respectively.¹⁷

Multi-institutional registries could improve understanding of acral melanoma in Black patients.⁴ More studies are needed to help differentiate between the dermoscopic finding of PRP in benign ethnic melanosis vs malignant melanoma.

Plantar hyperpigmentation (also known as plantar melanosis [increased melanin], volar pigmented macules, benign racial melanosis, acral pigmentation, acral ethnic melanosis, or mottled hyperpigmentation of the plantar surface) is a benign finding in many individuals and is especially prevalent in those with darker skin tones. Acral refers to manifestation on the hands and feet, volar on the palms and soles, and plantar on the soles only. Here, we focus on plantar hyper-pigmentation. We use the terms ethnic and racial interchangeably.

It is critically important to differentiate benign hyperpigmentation, which is common in patients with skin of color, from melanoma. Although rare, Black patients in the United States experience high morbidity and mortality from acral melanoma, which often is diagnosed late in the disease course.¹

There are many causes of hyperpigmentation on the plantar surfaces, including benign ethnic melanosis, nevi, melanoma, infections such as syphilis and tinea nigra, conditions such as Peutz-Jeghers syndrome and Laugier-Hunziker syndrome, and postinflammatory hyperpigmentation secondary to atopic dermatitis and psoriasis. We focus on the most common causes, ethnic melanosis and nevi, as well as melanoma, which is the deadliest cause.

Epidemiology

In a 1980 study (N=251), Black Americans had a high incidence of plantar hyperpigmentation, with 52% of affected patients having dark brown skin and 31% having light brown skin.²

The epidemiology of melanoma varies by race/ethnicity. Melanoma in Black individuals is relatively rare, with an annual incidence of approximately 1 in 100,000 individuals.³ However, when individuals with skin of color develop melanoma, they are more likely than their White counterparts to have acral melanoma (acral lentiginous melanoma), one of the deadliest types.¹ In a case series of Black patients with melanoma (N=48) from 2 tertiary care centers in Texas, 30 of 40 primary cutaneous melanomas (75%) were located on acral skin.⁴ Overall, 13 patients developed stage IV disease and 12 died due to disease progression. All patients who developed distant metastases or died of melanoma had acral melanoma.⁴ Individuals of Asian descent also have a high incidence of acral melanoma, as shown in research from Japan.^5-9

Key Clinical Features in Individuals With Darker Skin Tones

Dermoscopy is an evidence-based clinical examination method for earlier diagnosis of cutaneous melanoma, including on acral skin.^10,11 Benign nevi on the volar skin as well as the palms and soles tend to have one of these 3 dermoscopic patterns: parallel furrow, lattice, or irregular fibrillar. The pattern that is most predictive of volar melanoma is the parallel ridge pattern (PRP) (Figures A and B [insets]), which showed a high specificity (99.0%) and very high negative predictive value (97.7%) for malignant melanoma in a Japanese population.⁷ The PRP data from this study cannot be applied reliably to Black individuals, especially because benign ethnic melanosis and other benign conditions can demonstrate PRP.¹² Reliance on the PRP as a diagnostic clue could result in unneccessary biopsies in as many as 50% of Black patients with benign plantar hyperpigmentation.² Furthermore, biopsies of the plantar surface can be painful and cause pain while walking.

It has been suggested that PRP seen on dermoscopy in benign hyperpigmentation such as ethnic melanosis and nevi may preserve the acrosyringia (eccrine gland openings on the ridge), whereas PRP in melanoma may obliterate the acrosyringia.¹³ This observation is based on case reports only and needs further study. However, if validated, it could be a useful diagnostic clue.

In a retrospective cohort study of skin cancer in Black individuals (n=165) at a New York City–based cancer center from 2000 to 2020, 68% of patients were diagnosed with melanomas—80% were the acral subtype and 75% displayed a PRP. However, the surrounding uninvolved background skin, which was visible in most cases, also demonstrated a PRP.¹⁴ Because of the high morbidity and mortality rates of acral melanoma, clinicians should biopsy or immediately refer patients with concerning plantar hyperpigmentation to a dermatologist.

Health disparity highlight

The mortality rate for acral melanoma in Black patients is disproportionately high for the following reasons^15,16:

• Patients and health care providers do not expect to see melanoma in Black patients (it truly is rare!), so screening and education on sun protection are limited.

• Benign ethnic melanosis makes it more difficult to distinguish between early acral melanoma and benign skin changes.

• Black patients and other US patient populations with skin of color may be less likely to have health insurance, which contributes to inequities in access to health care. As of 2022, the uninsured rates for nonelderly American Indian and Alaska Native, Hispanic, Native Hawaiian and Other Pacific Islander, Black, and White individuals were 19.1%, 18.0%, 12.7%, 10.0%, and 6.6%, respectively.¹⁷

Multi-institutional registries could improve understanding of acral melanoma in Black patients.⁴ More studies are needed to help differentiate between the dermoscopic finding of PRP in benign ethnic melanosis vs malignant melanoma.

Allergic rhinoconjunctivitis causes onerous symptoms of sneezing, rhinorrhea, postnasal drip, nasal congestion, and itchy, watery eyes. It is a common condition that affects 10% to 25% of the US population and up to 23% of military members with increased symptoms during deployments.^1-3 Allergen immunotherapy (AIT), commonly known as allergy shots, is an effective treatment for allergic rhinoconjunctivitis, especially for patients whose symptoms are not controlled by allergy medications.⁴ Many military personnel who would like to receive AIT cannot continue with their immunotherapy because of frequent moves, deployments, and temporary duty assignments. This case report highlights the difficulty of managing AIT in the Military Health System.

Case Presentation

A 34-year-old active-duty US Air Force male surgeon with a medical history of allergic rhinoconjunctivitis was referred to the allergy clinic for evaluation and consideration of AIT. His symptoms included rhinorrhea, sneezing, nasal congestion, and itchy, watery eyes. The symptoms had been present for several years, occurring predominantly in the spring and fall, but also perennially when exposed to animals such as cats, dogs, and horses. The patient was raised on a ranch where he was exposed to these animals.

The patient had prior skin testing at the University of Nebraska Medical Center (UNMC) for aeroallergens and was positive for trees, grasses, weeds, molds, dust mites, cats, dogs, and horses. He received AIT at UNMC with great success for18 months. Regrettably, the patient discontinued AIT following a military move to Keesler Air Force Base in Mississippi. The patient’s examination was notable for injected conjunctiva, nasal mucosa edema, and a cobblestone throat. His symptoms were not alleviated with oral cetirizine and nasal fluticasone.

His skin testing was positive for trees, weeds, mold, cats, dogs, dust mites, and horsehair (Table). The risks and benefits of AIT were discussed with the patient, who elected to proceed with restarting AIT and received counseling on aeroallergen avoidance. The patient was unable to continue AIT at Keesler Medical Center because of a military deployment.

Discussion

There are several barriers to receiving AIT for active-duty patients with allergies. Due to previous skin test extracts, our patient had become desensitized to them. Though he had received aeroallergen immunotherapy with success for 18 months, the patients had to restart the build up phase of AIT due to a military-related move.

For patients to benefit from AIT, they must build up and maintain their immunotherapy injections for at least 3 to 5 years.⁴ The build-up period of immunotherapy lasts about 3 to 4 months. Patients typically receive weekly injections until they reach a maintenance immunotherapy dose of 0.5 mL of a 1:1 concentration ratio.⁴

Frequent deployments or temporary duty assignments are other barriers to AIT for active-duty patients. AIT is not usually given on deployments or temporary duty assignments unless the patient is located near a major military medical center. The US Air Force and Army operate allergy extender clinics at smaller bases and overseas locations to facilitate the maintenance of immunotherapy for military patients. Primary care physicians act as allergy extenders. These smaller allergy clinics are supervised by regional allergists at major military medical centers via telehealth and electronic/telephonic communication. These allergy clinics are not more widely available because there are not enough allergists and allergy medical technicians.

Allergen immunotherapy is not standardized, meaning civilian allergists use different aeroallergen immunotherapy formulations. While AIT is standardized in the US military through the Extract Laboratory Management System (ELMS), many active-duty patients are evaluated by civilian allergists in the TRICARE system who do not use ELMS, and when they move, AIT is not maintained.

Because up to 25% of active-duty personnel suffer from allergic rhinoconjunctivitis and AIT is not administered in many deployed settings, this issue could affect mission readiness and capabilities.^3-6 These personnel may suffer from frequent and severe nasal and ocular allergy symptoms without being able to continue AIT. There is the potential for adverse effects on the military missions because of these impaired military personnel.^5,6

Potential steps to improve the availability of allergen immunotherapy in the deployed setting include training deployed physicians, medical technicians, and other health care practitioners in administering and treating AIT so deployed personnel can receive therapy. Additionally, AIT should be standardized and ordered via the ELMS. Civilian allergists should be highly encouraged to use ELMS. This would create standardization of AIT for all active-duty allergy patients. The allergy extender system could be expanded to all military treatment facilities to provide easy access to allergen immunotherapy. The US Navy has the fewest allergists and allergy extenders, and would need to expand its network of allergy extenders to provide AIT at its health care facilities.

Conclusions

We present an active-duty servicemember with allergic rhinoconjunctivitis to trees, grasses, weeds, cats, dogs, dust mites, mold, and horses who had intermittent therapy that was interrupted by deployments. Our case highlights the difficulty of managing AIT in the military health system due to frequent moves, deployments, and temporary duty assignments. We also suggest steps that could help expand AIT for military personnel, including those deployed internationally.

Allergic rhinoconjunctivitis causes onerous symptoms of sneezing, rhinorrhea, postnasal drip, nasal congestion, and itchy, watery eyes. It is a common condition that affects 10% to 25% of the US population and up to 23% of military members with increased symptoms during deployments.^1-3 Allergen immunotherapy (AIT), commonly known as allergy shots, is an effective treatment for allergic rhinoconjunctivitis, especially for patients whose symptoms are not controlled by allergy medications.⁴ Many military personnel who would like to receive AIT cannot continue with their immunotherapy because of frequent moves, deployments, and temporary duty assignments. This case report highlights the difficulty of managing AIT in the Military Health System.

Case Presentation

A 34-year-old active-duty US Air Force male surgeon with a medical history of allergic rhinoconjunctivitis was referred to the allergy clinic for evaluation and consideration of AIT. His symptoms included rhinorrhea, sneezing, nasal congestion, and itchy, watery eyes. The symptoms had been present for several years, occurring predominantly in the spring and fall, but also perennially when exposed to animals such as cats, dogs, and horses. The patient was raised on a ranch where he was exposed to these animals.

The patient had prior skin testing at the University of Nebraska Medical Center (UNMC) for aeroallergens and was positive for trees, grasses, weeds, molds, dust mites, cats, dogs, and horses. He received AIT at UNMC with great success for18 months. Regrettably, the patient discontinued AIT following a military move to Keesler Air Force Base in Mississippi. The patient’s examination was notable for injected conjunctiva, nasal mucosa edema, and a cobblestone throat. His symptoms were not alleviated with oral cetirizine and nasal fluticasone.

His skin testing was positive for trees, weeds, mold, cats, dogs, dust mites, and horsehair (Table). The risks and benefits of AIT were discussed with the patient, who elected to proceed with restarting AIT and received counseling on aeroallergen avoidance. The patient was unable to continue AIT at Keesler Medical Center because of a military deployment.

Discussion

There are several barriers to receiving AIT for active-duty patients with allergies. Due to previous skin test extracts, our patient had become desensitized to them. Though he had received aeroallergen immunotherapy with success for 18 months, the patients had to restart the build up phase of AIT due to a military-related move.

For patients to benefit from AIT, they must build up and maintain their immunotherapy injections for at least 3 to 5 years.⁴ The build-up period of immunotherapy lasts about 3 to 4 months. Patients typically receive weekly injections until they reach a maintenance immunotherapy dose of 0.5 mL of a 1:1 concentration ratio.⁴

Frequent deployments or temporary duty assignments are other barriers to AIT for active-duty patients. AIT is not usually given on deployments or temporary duty assignments unless the patient is located near a major military medical center. The US Air Force and Army operate allergy extender clinics at smaller bases and overseas locations to facilitate the maintenance of immunotherapy for military patients. Primary care physicians act as allergy extenders. These smaller allergy clinics are supervised by regional allergists at major military medical centers via telehealth and electronic/telephonic communication. These allergy clinics are not more widely available because there are not enough allergists and allergy medical technicians.

Allergen immunotherapy is not standardized, meaning civilian allergists use different aeroallergen immunotherapy formulations. While AIT is standardized in the US military through the Extract Laboratory Management System (ELMS), many active-duty patients are evaluated by civilian allergists in the TRICARE system who do not use ELMS, and when they move, AIT is not maintained.

Because up to 25% of active-duty personnel suffer from allergic rhinoconjunctivitis and AIT is not administered in many deployed settings, this issue could affect mission readiness and capabilities.^3-6 These personnel may suffer from frequent and severe nasal and ocular allergy symptoms without being able to continue AIT. There is the potential for adverse effects on the military missions because of these impaired military personnel.^5,6

Potential steps to improve the availability of allergen immunotherapy in the deployed setting include training deployed physicians, medical technicians, and other health care practitioners in administering and treating AIT so deployed personnel can receive therapy. Additionally, AIT should be standardized and ordered via the ELMS. Civilian allergists should be highly encouraged to use ELMS. This would create standardization of AIT for all active-duty allergy patients. The allergy extender system could be expanded to all military treatment facilities to provide easy access to allergen immunotherapy. The US Navy has the fewest allergists and allergy extenders, and would need to expand its network of allergy extenders to provide AIT at its health care facilities.

Conclusions

We present an active-duty servicemember with allergic rhinoconjunctivitis to trees, grasses, weeds, cats, dogs, dust mites, mold, and horses who had intermittent therapy that was interrupted by deployments. Our case highlights the difficulty of managing AIT in the military health system due to frequent moves, deployments, and temporary duty assignments. We also suggest steps that could help expand AIT for military personnel, including those deployed internationally.

Allergic rhinoconjunctivitis causes onerous symptoms of sneezing, rhinorrhea, postnasal drip, nasal congestion, and itchy, watery eyes. It is a common condition that affects 10% to 25% of the US population and up to 23% of military members with increased symptoms during deployments.^1-3 Allergen immunotherapy (AIT), commonly known as allergy shots, is an effective treatment for allergic rhinoconjunctivitis, especially for patients whose symptoms are not controlled by allergy medications.⁴ Many military personnel who would like to receive AIT cannot continue with their immunotherapy because of frequent moves, deployments, and temporary duty assignments. This case report highlights the difficulty of managing AIT in the Military Health System.

Case Presentation

A 34-year-old active-duty US Air Force male surgeon with a medical history of allergic rhinoconjunctivitis was referred to the allergy clinic for evaluation and consideration of AIT. His symptoms included rhinorrhea, sneezing, nasal congestion, and itchy, watery eyes. The symptoms had been present for several years, occurring predominantly in the spring and fall, but also perennially when exposed to animals such as cats, dogs, and horses. The patient was raised on a ranch where he was exposed to these animals.

The patient had prior skin testing at the University of Nebraska Medical Center (UNMC) for aeroallergens and was positive for trees, grasses, weeds, molds, dust mites, cats, dogs, and horses. He received AIT at UNMC with great success for18 months. Regrettably, the patient discontinued AIT following a military move to Keesler Air Force Base in Mississippi. The patient’s examination was notable for injected conjunctiva, nasal mucosa edema, and a cobblestone throat. His symptoms were not alleviated with oral cetirizine and nasal fluticasone.

His skin testing was positive for trees, weeds, mold, cats, dogs, dust mites, and horsehair (Table). The risks and benefits of AIT were discussed with the patient, who elected to proceed with restarting AIT and received counseling on aeroallergen avoidance. The patient was unable to continue AIT at Keesler Medical Center because of a military deployment.

Discussion

There are several barriers to receiving AIT for active-duty patients with allergies. Due to previous skin test extracts, our patient had become desensitized to them. Though he had received aeroallergen immunotherapy with success for 18 months, the patients had to restart the build up phase of AIT due to a military-related move.

For patients to benefit from AIT, they must build up and maintain their immunotherapy injections for at least 3 to 5 years.⁴ The build-up period of immunotherapy lasts about 3 to 4 months. Patients typically receive weekly injections until they reach a maintenance immunotherapy dose of 0.5 mL of a 1:1 concentration ratio.⁴

Frequent deployments or temporary duty assignments are other barriers to AIT for active-duty patients. AIT is not usually given on deployments or temporary duty assignments unless the patient is located near a major military medical center. The US Air Force and Army operate allergy extender clinics at smaller bases and overseas locations to facilitate the maintenance of immunotherapy for military patients. Primary care physicians act as allergy extenders. These smaller allergy clinics are supervised by regional allergists at major military medical centers via telehealth and electronic/telephonic communication. These allergy clinics are not more widely available because there are not enough allergists and allergy medical technicians.

Allergen immunotherapy is not standardized, meaning civilian allergists use different aeroallergen immunotherapy formulations. While AIT is standardized in the US military through the Extract Laboratory Management System (ELMS), many active-duty patients are evaluated by civilian allergists in the TRICARE system who do not use ELMS, and when they move, AIT is not maintained.

Because up to 25% of active-duty personnel suffer from allergic rhinoconjunctivitis and AIT is not administered in many deployed settings, this issue could affect mission readiness and capabilities.^3-6 These personnel may suffer from frequent and severe nasal and ocular allergy symptoms without being able to continue AIT. There is the potential for adverse effects on the military missions because of these impaired military personnel.^5,6

Potential steps to improve the availability of allergen immunotherapy in the deployed setting include training deployed physicians, medical technicians, and other health care practitioners in administering and treating AIT so deployed personnel can receive therapy. Additionally, AIT should be standardized and ordered via the ELMS. Civilian allergists should be highly encouraged to use ELMS. This would create standardization of AIT for all active-duty allergy patients. The allergy extender system could be expanded to all military treatment facilities to provide easy access to allergen immunotherapy. The US Navy has the fewest allergists and allergy extenders, and would need to expand its network of allergy extenders to provide AIT at its health care facilities.

Conclusions

We present an active-duty servicemember with allergic rhinoconjunctivitis to trees, grasses, weeds, cats, dogs, dust mites, mold, and horses who had intermittent therapy that was interrupted by deployments. Our case highlights the difficulty of managing AIT in the military health system due to frequent moves, deployments, and temporary duty assignments. We also suggest steps that could help expand AIT for military personnel, including those deployed internationally.

Obstructive sleep apnea (OSA), the repetitive collapse of posterior oropharynx during sleep resulting in hypoxia and/or arousals from sleep, is the most common form of sleep disordered breathing and a common chronic respiratory disorders among middle-aged adults. OSA can lead to significant health problems, such as worsened cardiometabolic disease and cognitive impairment, which can increase morbidity and mortality.¹

The gold standard for OSA diagnosis is polysomnography (PSG), although home sleep studies can be performed for select patients. OSA diagnoses are based on the number of times per hour of sleep a patient’s airway narrows or collapses, reducing or stopping airflow, scored as hypopnea or apnea events, respectively. An Apnea-Hypopnea Index (AHI) score of 5 to 14 events/hour is considered mild OSA, 15 to 30 events/hour moderate OSA, and ≥ 30 events/hour severe OSA.²

Treatment commonly includes positive airway pressure (PAP) but more than one-half of patients are not adherent to continuous PAP (CPAP) treatment after about 90 days.³ Efficacy of treatments vary as a function of disease severity and etiology, which—in addition to the classic presentation of obesity with large neck/narrowupper airway—includes craniofacial abnormalities, altered muscle function in the upper airway, pharyngeal neuropathy, and fluid shifts to the neck.

Background

The American Academy of Sleep Medicine (AASM) estimates that 10% to 17% of adults in the United States have OSA.⁴ Compared with civilians, the military population generally is younger and healthier. Service members have access to regular health care with yearly physical examinations, exercise scheduled into the workday, and mandatory height/weight and fitness standards. Because obesity is a major risk factor for OSA, and the incidence of obesity is relatively low in the military population (estimated at 18.8% in 2021 vs 39.8% among all US adults aged 20 to 39 years), it might be expected that incidence of OSA would be correspondingly low.^5,6 However, there is evidence of a rapidly increasing incidence of OSA in military populations. A 2021 study revealed that OSA incidence rates increased from 11 to 333 per 10,000 between 2005 and 2019 across all military branches and demographics, with the highest rate among Army personnel.⁷ An earlier study revealed a 600% increase in OSA incidence among Army personnel between 2003 and 2011.⁸

Several factors likely contributed to this increase, including expanding obesity and greater physician awareness and availability of sleep study centers. Rogers and colleagues found that 40% to 50% of incident OSA diagnoses among military personnel occur within 12 months of separation, suggesting that the secondary gains associated with military disability benefits might motivate OSA evaluation.⁹ It is possible that secondary gain is a factor because an OSA diagnosis can range from a 0% to 100% disability rating, depending on the severity.¹⁰ This disability claim is based on evidence that untreated OSA can negatively affect long-term health and mission readiness.⁸ For example, untreated OSA can lead to hypertension, which contributes to a long list of adverse health and wellness consequences. Most importantly for the military, OSA has been shown to increase daytime sleepiness and reduce cognitive performance.¹⁰

The current first-line treatment for OSA is CPAP, which improves symptoms of daytime sleepiness, hypertension management, and daytime alertness.¹¹ Despite its efficacy, nonadherence rates range from 29% to 83%.^12-15 Nonadherence factors include lifestyle changes, adverse effects (eg, nasal congestion), and lack of education on proper use.¹¹ Lifestyle changes needed to increase the likelihood of successful therapy, such as regular sleep schedules and proper CPAP cleaning and maintenance, are difficult for military personnel because of the nature of continuous or sustained operations that might require shift work and/or around-the-clock (ie, 24-hour, 7 days a week) task performance. Traveling with CPAP is an added burden for service members deployed to combat operations (ie, added luggage, weight, maintenance). Although alternate treatments such as oral appliances (ie, custom dental devices) are available, they generally are less effective than CPAP.² Oral appliances could be a reasonable alternative treatment for some patients who cannot manage their OSA with behavioral modifications and are intolerant or unable to effectively use CPAP. This could include patients in the military who are deployed to austere environments.

Surgically implanted hypoglossal nerve stimulator (HGNS) treatment may provide long-term health benefits to service members. After the device is implanted near the hypoglossal nerve, electrical stimulation causes the tongue to move forward, which opens the airway in the anteroposterior dimension. The most important consideration is the mechanism of airway collapse. HGNS is not effective for patients whose OSA events are caused by circumferential collapse of other airway muscles. The cause of airway collapse is ascertained before surgery with drug-induced sleep endoscopy, a procedure that allows visualization of conformational changes in the upper airway during OSA events.

METHODS

To identify eligible studies, we employed PICOS: Population (patients aged ≥ 18 years with a history of OSA), Intervention (HGNS), Comparator (standard of care PAP therapy), Outcome (AHI or Epworth Sleepiness Scale [ESS], and Study (randomized control trial [RCT] or clinical trial). Studies were excluded if they were not written in English or included pediatric populations. The ESS is a subjective rating scale used to determine and quantify a patient’s level of daytime sleepiness, using a 4-point scale for the likelihood of falling asleep totaled across 8 different situations.¹⁸ Daytime sleepiness is considered lower normal(0-5 points), higher normal (6-10 points), mild or moderate excessive (11-15 points), and severe excessive (16-24 points).

Literature Search

We conducted a review of PubMed and Scopus for RCTs and controlled trials published from 2013 to 2023 that included the keywords and phrases: obstructive sleep apnea and either hypoglossal nerve stimulation or upper airway stimulation. The final literature search was performed December 8, 2023.

Two authors independently assessed the titles and abstracts of studies identified in the literature search based on the predefined inclusion criteria. If it was not clear whether an article met inclusion criteria based on its title and/or abstract, the 2 review authors assessed the full text of study and resolved any disagreement through consensus. If consensus was not obtained, a third author was consulted. No duplicates were identified. The PRISMA study selection process is presented in the Figure.

Data extraction was performed by 1 independent reviewer. A second author reviewed the extracted data. Any identified discrepancies were resolved through discussion and consensus. If consensus was not obtained, a third author was consulted. Study data included methods (study design and study objective), participants mean age, inclusion criteria, exclusion criteria, interventions and comparators, and primary study outcomes.

The quality of evidence was assessed using a rating of 1 to 5 based on a modified version of the Oxford Centre for Evidence-based Medicine Levels of Evidence and Grades of Recommendation.¹⁹ A rating of 1 indicated a properly powered and conducted RCT, 2 demonstrated a well-designed controlled trial without randomization or prospective comparative cohort trial, 3 designated a case-control study or retrospective cohort study, 4 signified a case series with or without intervention or a cross-sectional study, and 5 denoted an opinion of respected authorities or case reports. Two reviewers independently evaluated the quality of evidence. Any identified discrepancies were resolved through discussion and consensus. If consensus was not obtained, a third review author was consulted.

We identified 30 studies; 19 articles did not meet inclusion criteria. The remaining 11 articles were divided into 4 cohorts. Five articles were based on data from the STAR trial, a multicenter study that included adults with moderate-to-severe OSA and inadequate adherence to CPAP.^20-24 Four articles used the same patient selection criteria as the STAR trial for a long-term German postmarket study of upper airway stimulation efficacy with OSA.^25-28 The third and fourth cohorts each consist of 31 patients with moderate-to-severe OSA with CPAP nonadherence or failure.^29,30 The STAR trial included follow-up at 5 years, and the German-postmarket had a follow-up at3 years. The remaining 2 cohorts have 1-year follow-ups.

The Scopus review identified 304 studies; 299 did not meet inclusion criteria and 1 was part of the STAR trial.³¹ The remaining 4 articles were classified as distinct cohorts. Huntley and colleagues included patients from Thomas Jefferson University (TJU) and University of Pittsburgh (UP) academic medical centers.³² The Pordzik and colleagues cohort received implantation at a tertiary medical center, an RCCT, and a 1:1 comparator trial (Table 1).^33-35

STAR Trial

This multicenter, prospective, single-group cohort study was conducted in the US, Germany, Belgium, Netherlands, and France. The STAR trial included 126 patients who were not CPAP therapy adherent. Patients were excluded if they had AHI < 20 or > 50, central sleep apnea > 25% of total AHI, anatomical abnormalities that prevent effective assessment of upper-airway stimulation, complete concentric collapse of the retropalatal airway during drug-induced sleep, neuromuscular disease, hypoglossal-nerve palsy, severe restrictive or obstructive pulmonary disease, moderate-to-severe pulmonary arterial hypertension, severe valvular heart disease, New York Heart Association class III or IV heart failure, recent myocardial infarction or severe cardiac arrhythmias (within the past 6 months), persistent uncontrolled hypertension despite medication use, active psychiatric illness, or coexisting nonrespiratory sleep disorders that would confound functional sleep assessment. Primary outcome measures included the AHI and oxygen desaturation index (ODI) with secondary outcomes using the ESS, the Functional Outcomes of Sleep Questionnaire (FOSQ), and the percentage of sleep time with oxygen saturation < 90%. Of 126 patients who received implantation, 71 underwent an overnight PSG evaluation at 5-year follow-up. Mean (SD) AHI at baseline was reduced with HGNS treatment to from 32.0 (11.8) to 12.4 (16.3). Mean (SD) ESS for 92 participants with 2 measurements declined from 11.6 (5.0) at baseline to 6.9 (4.7) at 5-year follow-up.

The STAR trial included a randomized controlled withdrawal study for 46 patients who had a positive response to therapy to evaluate efficacy and durability of upper airway stimulation. Patients were randomly assigned to therapy maintenance or therapy withdrawal groups for ≥ 1 week. The short-term withdrawal effect was assessed using the original trial outcome measures and indicated that both the withdrawal and maintenance groups showed improvements at 12 months compared with the baseline. However, after the randomized withdrawal, the withdrawal group’s outcome measures deteriorated to baseline levels while the maintenance group showed no change. At 18 months of therapy, outcome measures for both groups were similar to those observed with therapy at 12 months.²⁴ The STAR trial included self-reported outcomes at baseline, 12 months, and 24 months that used ESS to measure daytime sleepiness. These results included subsequent STAR trial reports.^20-24,31

The German Postmarket Cohort

This multicenter, prospective, single-arm study used selection criteria that were based on those used in the STAR trial and included patients with moderate-to-severe OSA and nonadherence to CPAP. Patients were excluded if they had a BMI > 35, AHI < 15 or > 65; central apnea index > 25% of total AHI; or complete concentric collapse at the velopharynx during drug-induced sleep. Measured outcomes included AHI, ODI, FOSQ, and ESS. Among the 60 participants, 38 received implantation and a 3-year follow-up. Mean (SD) AHI decreased from 31.2 (13.2) at baseline to 13.1 (14.1) at follow-up, while mean (SD) ESS decreased from 12.8 (5.3) at baseline to 6.0 (3.2) at follow-up.^25-28

Munich Cohort

This single-center, prospective clinical trial included patients with AHI > 15 and < 65, central apnea index < 25% of total AHI, and nonadherence to CPAP. Patients were excluded if they had a BMI > 35, anatomical abnormalities that would prevent effective assessment of upper-airway stimulation; all other exclusion criteria matched those used in the STAR trial. Among 31 patients who received implants and completed a 1-year follow-up, mean (SD) AHI decreased from 32.9 (11.2) at baseline to 7.1 (5.9) at follow-up and mean (SD) ESS decreased from 12.6 (5.6) at baseline to 5.9 (5.2) at follow-up.²⁹

Kezirian and Colleagues Cohort

This prospective, single-arm, open-label study was conducted at 4 Australian and 4 US sites. Selection criteria included moderate-to-severe OSA with failure of CPAP, AHI of 20 to 100 with ≥ 15 events/hour occurring in sleep that was non-REM (rapid eye movement) sleep, BMI ≤ 40 (Australia) or ≤ 37 (US), and a predominance of hypopneas (≥ 80% of disordered breathing events during sleep). Patients were excluded if they had earlier upper airway surgery, markedly enlarged tonsils, uncontrolled nasal obstruction, severe retrognathia, > 5% central or mixed apneic events, incompletely treated sleep disorders other than OSA, or a major disorder of the pulmonary, cardiac, renal, or nervous systems. Data were reported for 31 patients whose mean (SD) AHI declined from 45.4 (17.5) at baseline to 25.3 (20.6) at 1-year follow-up and mean (SD) ESS score declined from 12.1 (4.6) at baseline to 7.9 (3.8) 1 year later.³⁰

The TJU and UP cohorts are composed of patients who underwent implantation between May 2014 and August 2016 at 2 academic centers.^31,32 Selection criteria was consistent with that used in the STAR trial, and patients completed postoperative titration PSG and outpatient follow-up (48 patients at TJU and 49 at UP). Primary outcomes included AHI, ESS, and O₂ nadir. Secondary outcomes consisted of surgical success and percentage of patients tolerating optimal titration setting at follow-up. Postoperative outcomes were assessed during the titration PSG. Time from initial ESS to postoperative PSG at TJU was 1.7 years and at UP was 1.9 years. Time from initial AHI to postoperative PSG at TJU was 90.4 days and 85.2 days at UP. At TJU, mean (SD) AHI and ESS dropped from 35.9 (20.8) and 11.1 (3.8), respectively at baseline to 6.3 (11.5) and 5.8 (3.4), respectively at follow-up. At UP, mean (SD) AHI and ESS fell from 35.3 (15.3) and 10.9 (4.9), respectively at baseline to 6.3 (6.1) and 6.6 (4.5), respectively at follow-up. There were no site-related differences in rates of AHI, ESS, or surgical success.³¹

Pordzik and Colleagues Cohort

This cohort of 29 patients underwent implantation between February 2020 and June 2022 at a tertiary university medical center with both pre- and postoperative PSG. Selection criteria was consistent with that of the German postmarket cohort. Postoperative PSG was completed a mean (SD) 96.3 (27.0) days after device activation. Mean (SD) AHI dropped from 38.6 (12.7) preoperatively to 24.4 (13.3) postoperatively. Notably, this cohort showed a much lower decrease of postoperative AHI than reported by the STAR trial and UP/TJU cohort.³³

Stimulation vs Sham Trial

This multicenter, double-blinded, randomized, crossover trial assessed the effect of HGNS (stim) vs sham stimulation (sham) in 86 patients that completed both phases of the trial. Primary outcomes included AHI and ESS. Secondary outcomes included FOSQ. No carryover effect was found during the crossover phase. The difference between the phases was−15.5 (95% CI, −18.3 to −12.8) for AHI and −3.3 (95% CI, −4.4 to −2.2) for ESS.³⁴

Comparator

The comparator study used propensity score matching to compare outcomes of HGNS and PAP therapy. Primary outcomes included sleepiness, AHI, and effectiveness with outcome measures of AHI and ESS collected at baseline and 12 months postimplantation. The article reported that 126 of 227 patients were matched 1:1. Both groups showed improvement in AHI and ESS. Mean (SD) AHI for the HGNS group at baseline started at 33.9 (15.1) and decreased to 8.1 (6.3). Mean (SD) ESS for the HGNS group at baseline was 15.4 (3.5) and decreased to 7.5 (4.7). In the PAP comparator group, mean (SD) baseline AHI was 36.8 (21.6) and at follow-up was 6.6 (8.0) and mean (SD) ESS was 14.6 (3.9) at baseline and 10.8 (5.6) at follow-up.³⁵

DISCUSSION

The current clinical data on HGNS suggest that this treatment is effective in adults with moderate-to-severe OSA and effects are sustained at long-term follow-up, as measured by AHI reduction and improvements in sleep related symptoms and quality of life (Table 2). These results have been consistent across several sites.

The STAR trial included a randomized control withdrawal group, for whom HGNS treatment was withdrawn after the 12-month follow-up, and then restored at 18 months.²¹ This revealed that withdrawal of HGNS treatment resulted in deterioration of both objective and subjective measures of OSA and sleepiness. The beneficial effects of HGNS were restored when treatment was resumed.²⁴ Additionally, the RCCT revealed that therapeutic stimulation via HGNS significantly reduced subjective and objective measures of OSA.³⁴ These studies provide definitive evidence of HGNS efficacy.

Currently, a diagnosis of OSA on PAP is classified as a 50% military disability rating. This rating is based primarily on epidemiologic evidence that untreated OSA is a costly disease that leads to other chronic illnesses that increases health care utilization.⁹ HGNS requires an initially invasive procedure and higher upfront costs, but it could result in reduced health care use and long-term costs because of improved adherence to treatment—compared with CPAP—that results in better outcomes.

The reviewed studies have several limitations that warrant caution when determining the possible benefits of HGNS treatment. The primary limitation is the lack of active control groups, therefore precluding a direct comparison of the short- and long-term effectiveness of HGNS vs other treatments (eg, CPAP). This is especially problematic because in the reviewed studies HGNS treatment efficacy is reported as a function of the mean—and SD—percent reduction in the AHI, whereas the efficacy of CPAP treatment usually is defined in terms of “adequacy of titration” as suggested by the AASM.³⁶ It has been reported that with CPAP treatment, 50% to 60% of OSA patients achieve AASM-defined optimal improvement of respiratory disturbance index of < 5/hour during a polysomnographic sleep recording of ≥ 15 minutes duration that includes REM sleep in the supine position.³⁷ In most of the reviewed studies, treatment success was more liberally defined as a decrease of AHI by ≥ 50%, regardless of the resulting AHI. It is notable that among the reviewed HGNS studies, the TJU and UP cohorts achieved the best outcome in short-term follow-up of 2 months with a mean (SD) AHI of 6.3 (11.5) and 6.4 (6.1), respectively. Among those cohortsassessed at a 12-month follow-up, the Munich cohort achieved the best outcome with a mean (SD) AHI of 7.1 (5.9).

Although the metrics reported in the reviewed studies are not directly comparable, the reported findings strongly suggest that HGNS generally is less effective than CPAP. How important are these differences? With findings that HGNS “reliably produces clinically meaningful (positive) effects on daytime sleepiness, daytime functioning, and sleep quality,” does it really matter if the outcome metrics for HGNS are a little less positive than those produced by CPAP?³⁸ For individual military OSA patients the answer is yes. This is because in military operational environments—especially during deployment—sleep restriction is nearly ubiquitous, therefore any mild residual deficits in sleep quality and daytime alertness resulting from nominally adequate, but suboptimal OSA treatment, could be exacerbated by sleep restriction, therefore placing the service member and the mission at increased risk.³⁹

Another limitation is the narrow inclusion criteria these studies employed, which limits the generalizability of the findings. Participants in the reviewed clinical trials were selected from a patient population that was mostly middle-aged, White, and obese or overweight. In a Medical Surveillance Monthly Report study, OSA was found to be highest among service members aged > 40 years, male, obese, and Black/non-Hispanic (although it should be noted that more than one-half of enlisted service members aged ≤ 25 years).^40,41 Obesity has been noted as a growing concern for the military as the military population is beginning to mirror the civilian population in terms of being overweight or obese despite height and weight standards. HGNS might not be as successful in military populations with different demographics. Moreover, HGNS has been shown to have greater AHI reduction among those with higher BMI.³⁰ Although obese service members have a 6-fold higher 12-year incidence rate of OSA than service members without obesity, this nevertheless suggests that general level of HGNS efficacy might be lower among the military patient population, because obesity is less prevalent in the military than the general population.⁹

Ethnicity has been found to be a relevant factor, with the highest incidence rate of OSA among non-Hispanic Black males, a demographic that was underrepresented in cohorts included in this review. Further studies will be needed to determine the extent to which findings from HGNS treatment studies are generalizable to the broader OSA patient population.

HGNS Implementation Challenges

Current impediments to widespread use of HGNS as an OSA treatment include no standardized guidance for titration and follow-on care, which varies based on the resources available. Titrating a new device for HGNS requires experienced sleep technicians who have close relationships with device representatives and can troubleshoot problems. Technical expertise, which currently is rare, is required if there are complications after placement or if adjustments to voltage settings are needed over time. In addition, patients may require multiple specialists making it easy to get lost to follow-up after implantation. This is particularly challenging in a transient community, such as the military, because there is no guarantee that a service member will have access to the same specialty care at the next duty station.

Although some evidence suggests that HGNS is a viable alternative treatment for some patients with OSA, the generalizability of these findings to the military patient population is unclear. Specialized facilities and expertise are needed for the surgical procedure and follow-up requirements, which currently constitute significant logistical constraints. As with any implantable device, there is a risk of complications including infection that could result in medical evacuation from a theater of operations. If the device malfunctions or loses effectiveness in a deployed environment, the service member might not have immediate access to medical support, potentially leading to undertreatment of OSA. In future battlefield scenarios in multidomain operations, prolonged, far-forward field care will become the new normal because the military is not expected to have air superiority or the ability to quickly evacuate service members to a higher level of medical care.⁴²

In deployed environments, the potential limitations of HGNS become increasingly risky for the service member and the overall mission. Considering these factors, it will be important to evaluate the practicality of HGNS as a treatment option in military populations. Military-specific challenges associated with HGNS that require further study, include guidance for patient selection outside academic centers, guidance on long-term postsurgical care and device maintenance, duty limitation and military retention considerations, and limitations in training and combat environments. The military medical community needs to conduct its own studies in appropriately selected service members to guide clinical practice.

CONCLUSIONS

HGNS treatment results in improvement of both AHI and ESS scores and could be a deployable treatment option for military patients with OSA. However, HGNS has not been found to be as effective as CPAP, although the current literature is limited by small sample sizes, homogeneous populations that do not reflect the demographics of the military, and mostly short follow-up periods. Future studies should be focused on collecting data on HGNS from demographic groups that are more representative of the military OSA patient population and identifying the subpopulation of patients who derive the greatest benefit from HGNS, so that this treatment can be better individually targeted. Until data on existing military patients is published, it is not possible to fully weigh risks and benefits in this population and generalize civilian guidance to the military.

Obstructive sleep apnea (OSA), the repetitive collapse of posterior oropharynx during sleep resulting in hypoxia and/or arousals from sleep, is the most common form of sleep disordered breathing and a common chronic respiratory disorders among middle-aged adults. OSA can lead to significant health problems, such as worsened cardiometabolic disease and cognitive impairment, which can increase morbidity and mortality.¹

The gold standard for OSA diagnosis is polysomnography (PSG), although home sleep studies can be performed for select patients. OSA diagnoses are based on the number of times per hour of sleep a patient’s airway narrows or collapses, reducing or stopping airflow, scored as hypopnea or apnea events, respectively. An Apnea-Hypopnea Index (AHI) score of 5 to 14 events/hour is considered mild OSA, 15 to 30 events/hour moderate OSA, and ≥ 30 events/hour severe OSA.²

Treatment commonly includes positive airway pressure (PAP) but more than one-half of patients are not adherent to continuous PAP (CPAP) treatment after about 90 days.³ Efficacy of treatments vary as a function of disease severity and etiology, which—in addition to the classic presentation of obesity with large neck/narrowupper airway—includes craniofacial abnormalities, altered muscle function in the upper airway, pharyngeal neuropathy, and fluid shifts to the neck.

Background

The American Academy of Sleep Medicine (AASM) estimates that 10% to 17% of adults in the United States have OSA.⁴ Compared with civilians, the military population generally is younger and healthier. Service members have access to regular health care with yearly physical examinations, exercise scheduled into the workday, and mandatory height/weight and fitness standards. Because obesity is a major risk factor for OSA, and the incidence of obesity is relatively low in the military population (estimated at 18.8% in 2021 vs 39.8% among all US adults aged 20 to 39 years), it might be expected that incidence of OSA would be correspondingly low.^5,6 However, there is evidence of a rapidly increasing incidence of OSA in military populations. A 2021 study revealed that OSA incidence rates increased from 11 to 333 per 10,000 between 2005 and 2019 across all military branches and demographics, with the highest rate among Army personnel.⁷ An earlier study revealed a 600% increase in OSA incidence among Army personnel between 2003 and 2011.⁸

Several factors likely contributed to this increase, including expanding obesity and greater physician awareness and availability of sleep study centers. Rogers and colleagues found that 40% to 50% of incident OSA diagnoses among military personnel occur within 12 months of separation, suggesting that the secondary gains associated with military disability benefits might motivate OSA evaluation.⁹ It is possible that secondary gain is a factor because an OSA diagnosis can range from a 0% to 100% disability rating, depending on the severity.¹⁰ This disability claim is based on evidence that untreated OSA can negatively affect long-term health and mission readiness.⁸ For example, untreated OSA can lead to hypertension, which contributes to a long list of adverse health and wellness consequences. Most importantly for the military, OSA has been shown to increase daytime sleepiness and reduce cognitive performance.¹⁰

The current first-line treatment for OSA is CPAP, which improves symptoms of daytime sleepiness, hypertension management, and daytime alertness.¹¹ Despite its efficacy, nonadherence rates range from 29% to 83%.^12-15 Nonadherence factors include lifestyle changes, adverse effects (eg, nasal congestion), and lack of education on proper use.¹¹ Lifestyle changes needed to increase the likelihood of successful therapy, such as regular sleep schedules and proper CPAP cleaning and maintenance, are difficult for military personnel because of the nature of continuous or sustained operations that might require shift work and/or around-the-clock (ie, 24-hour, 7 days a week) task performance. Traveling with CPAP is an added burden for service members deployed to combat operations (ie, added luggage, weight, maintenance). Although alternate treatments such as oral appliances (ie, custom dental devices) are available, they generally are less effective than CPAP.² Oral appliances could be a reasonable alternative treatment for some patients who cannot manage their OSA with behavioral modifications and are intolerant or unable to effectively use CPAP. This could include patients in the military who are deployed to austere environments.

Surgically implanted hypoglossal nerve stimulator (HGNS) treatment may provide long-term health benefits to service members. After the device is implanted near the hypoglossal nerve, electrical stimulation causes the tongue to move forward, which opens the airway in the anteroposterior dimension. The most important consideration is the mechanism of airway collapse. HGNS is not effective for patients whose OSA events are caused by circumferential collapse of other airway muscles. The cause of airway collapse is ascertained before surgery with drug-induced sleep endoscopy, a procedure that allows visualization of conformational changes in the upper airway during OSA events.

METHODS

To identify eligible studies, we employed PICOS: Population (patients aged ≥ 18 years with a history of OSA), Intervention (HGNS), Comparator (standard of care PAP therapy), Outcome (AHI or Epworth Sleepiness Scale [ESS], and Study (randomized control trial [RCT] or clinical trial). Studies were excluded if they were not written in English or included pediatric populations. The ESS is a subjective rating scale used to determine and quantify a patient’s level of daytime sleepiness, using a 4-point scale for the likelihood of falling asleep totaled across 8 different situations.¹⁸ Daytime sleepiness is considered lower normal(0-5 points), higher normal (6-10 points), mild or moderate excessive (11-15 points), and severe excessive (16-24 points).

Literature Search

We conducted a review of PubMed and Scopus for RCTs and controlled trials published from 2013 to 2023 that included the keywords and phrases: obstructive sleep apnea and either hypoglossal nerve stimulation or upper airway stimulation. The final literature search was performed December 8, 2023.

Two authors independently assessed the titles and abstracts of studies identified in the literature search based on the predefined inclusion criteria. If it was not clear whether an article met inclusion criteria based on its title and/or abstract, the 2 review authors assessed the full text of study and resolved any disagreement through consensus. If consensus was not obtained, a third author was consulted. No duplicates were identified. The PRISMA study selection process is presented in the Figure.

Data extraction was performed by 1 independent reviewer. A second author reviewed the extracted data. Any identified discrepancies were resolved through discussion and consensus. If consensus was not obtained, a third author was consulted. Study data included methods (study design and study objective), participants mean age, inclusion criteria, exclusion criteria, interventions and comparators, and primary study outcomes.

The quality of evidence was assessed using a rating of 1 to 5 based on a modified version of the Oxford Centre for Evidence-based Medicine Levels of Evidence and Grades of Recommendation.¹⁹ A rating of 1 indicated a properly powered and conducted RCT, 2 demonstrated a well-designed controlled trial without randomization or prospective comparative cohort trial, 3 designated a case-control study or retrospective cohort study, 4 signified a case series with or without intervention or a cross-sectional study, and 5 denoted an opinion of respected authorities or case reports. Two reviewers independently evaluated the quality of evidence. Any identified discrepancies were resolved through discussion and consensus. If consensus was not obtained, a third review author was consulted.

We identified 30 studies; 19 articles did not meet inclusion criteria. The remaining 11 articles were divided into 4 cohorts. Five articles were based on data from the STAR trial, a multicenter study that included adults with moderate-to-severe OSA and inadequate adherence to CPAP.^20-24 Four articles used the same patient selection criteria as the STAR trial for a long-term German postmarket study of upper airway stimulation efficacy with OSA.^25-28 The third and fourth cohorts each consist of 31 patients with moderate-to-severe OSA with CPAP nonadherence or failure.^29,30 The STAR trial included follow-up at 5 years, and the German-postmarket had a follow-up at3 years. The remaining 2 cohorts have 1-year follow-ups.

The Scopus review identified 304 studies; 299 did not meet inclusion criteria and 1 was part of the STAR trial.³¹ The remaining 4 articles were classified as distinct cohorts. Huntley and colleagues included patients from Thomas Jefferson University (TJU) and University of Pittsburgh (UP) academic medical centers.³² The Pordzik and colleagues cohort received implantation at a tertiary medical center, an RCCT, and a 1:1 comparator trial (Table 1).^33-35

STAR Trial

This multicenter, prospective, single-group cohort study was conducted in the US, Germany, Belgium, Netherlands, and France. The STAR trial included 126 patients who were not CPAP therapy adherent. Patients were excluded if they had AHI < 20 or > 50, central sleep apnea > 25% of total AHI, anatomical abnormalities that prevent effective assessment of upper-airway stimulation, complete concentric collapse of the retropalatal airway during drug-induced sleep, neuromuscular disease, hypoglossal-nerve palsy, severe restrictive or obstructive pulmonary disease, moderate-to-severe pulmonary arterial hypertension, severe valvular heart disease, New York Heart Association class III or IV heart failure, recent myocardial infarction or severe cardiac arrhythmias (within the past 6 months), persistent uncontrolled hypertension despite medication use, active psychiatric illness, or coexisting nonrespiratory sleep disorders that would confound functional sleep assessment. Primary outcome measures included the AHI and oxygen desaturation index (ODI) with secondary outcomes using the ESS, the Functional Outcomes of Sleep Questionnaire (FOSQ), and the percentage of sleep time with oxygen saturation < 90%. Of 126 patients who received implantation, 71 underwent an overnight PSG evaluation at 5-year follow-up. Mean (SD) AHI at baseline was reduced with HGNS treatment to from 32.0 (11.8) to 12.4 (16.3). Mean (SD) ESS for 92 participants with 2 measurements declined from 11.6 (5.0) at baseline to 6.9 (4.7) at 5-year follow-up.

The STAR trial included a randomized controlled withdrawal study for 46 patients who had a positive response to therapy to evaluate efficacy and durability of upper airway stimulation. Patients were randomly assigned to therapy maintenance or therapy withdrawal groups for ≥ 1 week. The short-term withdrawal effect was assessed using the original trial outcome measures and indicated that both the withdrawal and maintenance groups showed improvements at 12 months compared with the baseline. However, after the randomized withdrawal, the withdrawal group’s outcome measures deteriorated to baseline levels while the maintenance group showed no change. At 18 months of therapy, outcome measures for both groups were similar to those observed with therapy at 12 months.²⁴ The STAR trial included self-reported outcomes at baseline, 12 months, and 24 months that used ESS to measure daytime sleepiness. These results included subsequent STAR trial reports.^20-24,31

The German Postmarket Cohort

This multicenter, prospective, single-arm study used selection criteria that were based on those used in the STAR trial and included patients with moderate-to-severe OSA and nonadherence to CPAP. Patients were excluded if they had a BMI > 35, AHI < 15 or > 65; central apnea index > 25% of total AHI; or complete concentric collapse at the velopharynx during drug-induced sleep. Measured outcomes included AHI, ODI, FOSQ, and ESS. Among the 60 participants, 38 received implantation and a 3-year follow-up. Mean (SD) AHI decreased from 31.2 (13.2) at baseline to 13.1 (14.1) at follow-up, while mean (SD) ESS decreased from 12.8 (5.3) at baseline to 6.0 (3.2) at follow-up.^25-28

Munich Cohort

This single-center, prospective clinical trial included patients with AHI > 15 and < 65, central apnea index < 25% of total AHI, and nonadherence to CPAP. Patients were excluded if they had a BMI > 35, anatomical abnormalities that would prevent effective assessment of upper-airway stimulation; all other exclusion criteria matched those used in the STAR trial. Among 31 patients who received implants and completed a 1-year follow-up, mean (SD) AHI decreased from 32.9 (11.2) at baseline to 7.1 (5.9) at follow-up and mean (SD) ESS decreased from 12.6 (5.6) at baseline to 5.9 (5.2) at follow-up.²⁹

Kezirian and Colleagues Cohort

This prospective, single-arm, open-label study was conducted at 4 Australian and 4 US sites. Selection criteria included moderate-to-severe OSA with failure of CPAP, AHI of 20 to 100 with ≥ 15 events/hour occurring in sleep that was non-REM (rapid eye movement) sleep, BMI ≤ 40 (Australia) or ≤ 37 (US), and a predominance of hypopneas (≥ 80% of disordered breathing events during sleep). Patients were excluded if they had earlier upper airway surgery, markedly enlarged tonsils, uncontrolled nasal obstruction, severe retrognathia, > 5% central or mixed apneic events, incompletely treated sleep disorders other than OSA, or a major disorder of the pulmonary, cardiac, renal, or nervous systems. Data were reported for 31 patients whose mean (SD) AHI declined from 45.4 (17.5) at baseline to 25.3 (20.6) at 1-year follow-up and mean (SD) ESS score declined from 12.1 (4.6) at baseline to 7.9 (3.8) 1 year later.³⁰

The TJU and UP cohorts are composed of patients who underwent implantation between May 2014 and August 2016 at 2 academic centers.^31,32 Selection criteria was consistent with that used in the STAR trial, and patients completed postoperative titration PSG and outpatient follow-up (48 patients at TJU and 49 at UP). Primary outcomes included AHI, ESS, and O₂ nadir. Secondary outcomes consisted of surgical success and percentage of patients tolerating optimal titration setting at follow-up. Postoperative outcomes were assessed during the titration PSG. Time from initial ESS to postoperative PSG at TJU was 1.7 years and at UP was 1.9 years. Time from initial AHI to postoperative PSG at TJU was 90.4 days and 85.2 days at UP. At TJU, mean (SD) AHI and ESS dropped from 35.9 (20.8) and 11.1 (3.8), respectively at baseline to 6.3 (11.5) and 5.8 (3.4), respectively at follow-up. At UP, mean (SD) AHI and ESS fell from 35.3 (15.3) and 10.9 (4.9), respectively at baseline to 6.3 (6.1) and 6.6 (4.5), respectively at follow-up. There were no site-related differences in rates of AHI, ESS, or surgical success.³¹

Pordzik and Colleagues Cohort

This cohort of 29 patients underwent implantation between February 2020 and June 2022 at a tertiary university medical center with both pre- and postoperative PSG. Selection criteria was consistent with that of the German postmarket cohort. Postoperative PSG was completed a mean (SD) 96.3 (27.0) days after device activation. Mean (SD) AHI dropped from 38.6 (12.7) preoperatively to 24.4 (13.3) postoperatively. Notably, this cohort showed a much lower decrease of postoperative AHI than reported by the STAR trial and UP/TJU cohort.³³

Stimulation vs Sham Trial

This multicenter, double-blinded, randomized, crossover trial assessed the effect of HGNS (stim) vs sham stimulation (sham) in 86 patients that completed both phases of the trial. Primary outcomes included AHI and ESS. Secondary outcomes included FOSQ. No carryover effect was found during the crossover phase. The difference between the phases was−15.5 (95% CI, −18.3 to −12.8) for AHI and −3.3 (95% CI, −4.4 to −2.2) for ESS.³⁴

Comparator

The comparator study used propensity score matching to compare outcomes of HGNS and PAP therapy. Primary outcomes included sleepiness, AHI, and effectiveness with outcome measures of AHI and ESS collected at baseline and 12 months postimplantation. The article reported that 126 of 227 patients were matched 1:1. Both groups showed improvement in AHI and ESS. Mean (SD) AHI for the HGNS group at baseline started at 33.9 (15.1) and decreased to 8.1 (6.3). Mean (SD) ESS for the HGNS group at baseline was 15.4 (3.5) and decreased to 7.5 (4.7). In the PAP comparator group, mean (SD) baseline AHI was 36.8 (21.6) and at follow-up was 6.6 (8.0) and mean (SD) ESS was 14.6 (3.9) at baseline and 10.8 (5.6) at follow-up.³⁵

DISCUSSION

The current clinical data on HGNS suggest that this treatment is effective in adults with moderate-to-severe OSA and effects are sustained at long-term follow-up, as measured by AHI reduction and improvements in sleep related symptoms and quality of life (Table 2). These results have been consistent across several sites.

The STAR trial included a randomized control withdrawal group, for whom HGNS treatment was withdrawn after the 12-month follow-up, and then restored at 18 months.²¹ This revealed that withdrawal of HGNS treatment resulted in deterioration of both objective and subjective measures of OSA and sleepiness. The beneficial effects of HGNS were restored when treatment was resumed.²⁴ Additionally, the RCCT revealed that therapeutic stimulation via HGNS significantly reduced subjective and objective measures of OSA.³⁴ These studies provide definitive evidence of HGNS efficacy.

Currently, a diagnosis of OSA on PAP is classified as a 50% military disability rating. This rating is based primarily on epidemiologic evidence that untreated OSA is a costly disease that leads to other chronic illnesses that increases health care utilization.⁹ HGNS requires an initially invasive procedure and higher upfront costs, but it could result in reduced health care use and long-term costs because of improved adherence to treatment—compared with CPAP—that results in better outcomes.

The reviewed studies have several limitations that warrant caution when determining the possible benefits of HGNS treatment. The primary limitation is the lack of active control groups, therefore precluding a direct comparison of the short- and long-term effectiveness of HGNS vs other treatments (eg, CPAP). This is especially problematic because in the reviewed studies HGNS treatment efficacy is reported as a function of the mean—and SD—percent reduction in the AHI, whereas the efficacy of CPAP treatment usually is defined in terms of “adequacy of titration” as suggested by the AASM.³⁶ It has been reported that with CPAP treatment, 50% to 60% of OSA patients achieve AASM-defined optimal improvement of respiratory disturbance index of < 5/hour during a polysomnographic sleep recording of ≥ 15 minutes duration that includes REM sleep in the supine position.³⁷ In most of the reviewed studies, treatment success was more liberally defined as a decrease of AHI by ≥ 50%, regardless of the resulting AHI. It is notable that among the reviewed HGNS studies, the TJU and UP cohorts achieved the best outcome in short-term follow-up of 2 months with a mean (SD) AHI of 6.3 (11.5) and 6.4 (6.1), respectively. Among those cohortsassessed at a 12-month follow-up, the Munich cohort achieved the best outcome with a mean (SD) AHI of 7.1 (5.9).

Although the metrics reported in the reviewed studies are not directly comparable, the reported findings strongly suggest that HGNS generally is less effective than CPAP. How important are these differences? With findings that HGNS “reliably produces clinically meaningful (positive) effects on daytime sleepiness, daytime functioning, and sleep quality,” does it really matter if the outcome metrics for HGNS are a little less positive than those produced by CPAP?³⁸ For individual military OSA patients the answer is yes. This is because in military operational environments—especially during deployment—sleep restriction is nearly ubiquitous, therefore any mild residual deficits in sleep quality and daytime alertness resulting from nominally adequate, but suboptimal OSA treatment, could be exacerbated by sleep restriction, therefore placing the service member and the mission at increased risk.³⁹

Another limitation is the narrow inclusion criteria these studies employed, which limits the generalizability of the findings. Participants in the reviewed clinical trials were selected from a patient population that was mostly middle-aged, White, and obese or overweight. In a Medical Surveillance Monthly Report study, OSA was found to be highest among service members aged > 40 years, male, obese, and Black/non-Hispanic (although it should be noted that more than one-half of enlisted service members aged ≤ 25 years).^40,41 Obesity has been noted as a growing concern for the military as the military population is beginning to mirror the civilian population in terms of being overweight or obese despite height and weight standards. HGNS might not be as successful in military populations with different demographics. Moreover, HGNS has been shown to have greater AHI reduction among those with higher BMI.³⁰ Although obese service members have a 6-fold higher 12-year incidence rate of OSA than service members without obesity, this nevertheless suggests that general level of HGNS efficacy might be lower among the military patient population, because obesity is less prevalent in the military than the general population.⁹

Ethnicity has been found to be a relevant factor, with the highest incidence rate of OSA among non-Hispanic Black males, a demographic that was underrepresented in cohorts included in this review. Further studies will be needed to determine the extent to which findings from HGNS treatment studies are generalizable to the broader OSA patient population.

HGNS Implementation Challenges

Current impediments to widespread use of HGNS as an OSA treatment include no standardized guidance for titration and follow-on care, which varies based on the resources available. Titrating a new device for HGNS requires experienced sleep technicians who have close relationships with device representatives and can troubleshoot problems. Technical expertise, which currently is rare, is required if there are complications after placement or if adjustments to voltage settings are needed over time. In addition, patients may require multiple specialists making it easy to get lost to follow-up after implantation. This is particularly challenging in a transient community, such as the military, because there is no guarantee that a service member will have access to the same specialty care at the next duty station.

Although some evidence suggests that HGNS is a viable alternative treatment for some patients with OSA, the generalizability of these findings to the military patient population is unclear. Specialized facilities and expertise are needed for the surgical procedure and follow-up requirements, which currently constitute significant logistical constraints. As with any implantable device, there is a risk of complications including infection that could result in medical evacuation from a theater of operations. If the device malfunctions or loses effectiveness in a deployed environment, the service member might not have immediate access to medical support, potentially leading to undertreatment of OSA. In future battlefield scenarios in multidomain operations, prolonged, far-forward field care will become the new normal because the military is not expected to have air superiority or the ability to quickly evacuate service members to a higher level of medical care.⁴²

In deployed environments, the potential limitations of HGNS become increasingly risky for the service member and the overall mission. Considering these factors, it will be important to evaluate the practicality of HGNS as a treatment option in military populations. Military-specific challenges associated with HGNS that require further study, include guidance for patient selection outside academic centers, guidance on long-term postsurgical care and device maintenance, duty limitation and military retention considerations, and limitations in training and combat environments. The military medical community needs to conduct its own studies in appropriately selected service members to guide clinical practice.

CONCLUSIONS

HGNS treatment results in improvement of both AHI and ESS scores and could be a deployable treatment option for military patients with OSA. However, HGNS has not been found to be as effective as CPAP, although the current literature is limited by small sample sizes, homogeneous populations that do not reflect the demographics of the military, and mostly short follow-up periods. Future studies should be focused on collecting data on HGNS from demographic groups that are more representative of the military OSA patient population and identifying the subpopulation of patients who derive the greatest benefit from HGNS, so that this treatment can be better individually targeted. Until data on existing military patients is published, it is not possible to fully weigh risks and benefits in this population and generalize civilian guidance to the military.

Obstructive sleep apnea (OSA), the repetitive collapse of posterior oropharynx during sleep resulting in hypoxia and/or arousals from sleep, is the most common form of sleep disordered breathing and a common chronic respiratory disorders among middle-aged adults. OSA can lead to significant health problems, such as worsened cardiometabolic disease and cognitive impairment, which can increase morbidity and mortality.¹

The gold standard for OSA diagnosis is polysomnography (PSG), although home sleep studies can be performed for select patients. OSA diagnoses are based on the number of times per hour of sleep a patient’s airway narrows or collapses, reducing or stopping airflow, scored as hypopnea or apnea events, respectively. An Apnea-Hypopnea Index (AHI) score of 5 to 14 events/hour is considered mild OSA, 15 to 30 events/hour moderate OSA, and ≥ 30 events/hour severe OSA.²

Treatment commonly includes positive airway pressure (PAP) but more than one-half of patients are not adherent to continuous PAP (CPAP) treatment after about 90 days.³ Efficacy of treatments vary as a function of disease severity and etiology, which—in addition to the classic presentation of obesity with large neck/narrowupper airway—includes craniofacial abnormalities, altered muscle function in the upper airway, pharyngeal neuropathy, and fluid shifts to the neck.

Background

The American Academy of Sleep Medicine (AASM) estimates that 10% to 17% of adults in the United States have OSA.⁴ Compared with civilians, the military population generally is younger and healthier. Service members have access to regular health care with yearly physical examinations, exercise scheduled into the workday, and mandatory height/weight and fitness standards. Because obesity is a major risk factor for OSA, and the incidence of obesity is relatively low in the military population (estimated at 18.8% in 2021 vs 39.8% among all US adults aged 20 to 39 years), it might be expected that incidence of OSA would be correspondingly low.^5,6 However, there is evidence of a rapidly increasing incidence of OSA in military populations. A 2021 study revealed that OSA incidence rates increased from 11 to 333 per 10,000 between 2005 and 2019 across all military branches and demographics, with the highest rate among Army personnel.⁷ An earlier study revealed a 600% increase in OSA incidence among Army personnel between 2003 and 2011.⁸

Several factors likely contributed to this increase, including expanding obesity and greater physician awareness and availability of sleep study centers. Rogers and colleagues found that 40% to 50% of incident OSA diagnoses among military personnel occur within 12 months of separation, suggesting that the secondary gains associated with military disability benefits might motivate OSA evaluation.⁹ It is possible that secondary gain is a factor because an OSA diagnosis can range from a 0% to 100% disability rating, depending on the severity.¹⁰ This disability claim is based on evidence that untreated OSA can negatively affect long-term health and mission readiness.⁸ For example, untreated OSA can lead to hypertension, which contributes to a long list of adverse health and wellness consequences. Most importantly for the military, OSA has been shown to increase daytime sleepiness and reduce cognitive performance.¹⁰

The current first-line treatment for OSA is CPAP, which improves symptoms of daytime sleepiness, hypertension management, and daytime alertness.¹¹ Despite its efficacy, nonadherence rates range from 29% to 83%.^12-15 Nonadherence factors include lifestyle changes, adverse effects (eg, nasal congestion), and lack of education on proper use.¹¹ Lifestyle changes needed to increase the likelihood of successful therapy, such as regular sleep schedules and proper CPAP cleaning and maintenance, are difficult for military personnel because of the nature of continuous or sustained operations that might require shift work and/or around-the-clock (ie, 24-hour, 7 days a week) task performance. Traveling with CPAP is an added burden for service members deployed to combat operations (ie, added luggage, weight, maintenance). Although alternate treatments such as oral appliances (ie, custom dental devices) are available, they generally are less effective than CPAP.² Oral appliances could be a reasonable alternative treatment for some patients who cannot manage their OSA with behavioral modifications and are intolerant or unable to effectively use CPAP. This could include patients in the military who are deployed to austere environments.

Surgically implanted hypoglossal nerve stimulator (HGNS) treatment may provide long-term health benefits to service members. After the device is implanted near the hypoglossal nerve, electrical stimulation causes the tongue to move forward, which opens the airway in the anteroposterior dimension. The most important consideration is the mechanism of airway collapse. HGNS is not effective for patients whose OSA events are caused by circumferential collapse of other airway muscles. The cause of airway collapse is ascertained before surgery with drug-induced sleep endoscopy, a procedure that allows visualization of conformational changes in the upper airway during OSA events.

METHODS

To identify eligible studies, we employed PICOS: Population (patients aged ≥ 18 years with a history of OSA), Intervention (HGNS), Comparator (standard of care PAP therapy), Outcome (AHI or Epworth Sleepiness Scale [ESS], and Study (randomized control trial [RCT] or clinical trial). Studies were excluded if they were not written in English or included pediatric populations. The ESS is a subjective rating scale used to determine and quantify a patient’s level of daytime sleepiness, using a 4-point scale for the likelihood of falling asleep totaled across 8 different situations.¹⁸ Daytime sleepiness is considered lower normal(0-5 points), higher normal (6-10 points), mild or moderate excessive (11-15 points), and severe excessive (16-24 points).

Literature Search

We conducted a review of PubMed and Scopus for RCTs and controlled trials published from 2013 to 2023 that included the keywords and phrases: obstructive sleep apnea and either hypoglossal nerve stimulation or upper airway stimulation. The final literature search was performed December 8, 2023.

Two authors independently assessed the titles and abstracts of studies identified in the literature search based on the predefined inclusion criteria. If it was not clear whether an article met inclusion criteria based on its title and/or abstract, the 2 review authors assessed the full text of study and resolved any disagreement through consensus. If consensus was not obtained, a third author was consulted. No duplicates were identified. The PRISMA study selection process is presented in the Figure.

Data extraction was performed by 1 independent reviewer. A second author reviewed the extracted data. Any identified discrepancies were resolved through discussion and consensus. If consensus was not obtained, a third author was consulted. Study data included methods (study design and study objective), participants mean age, inclusion criteria, exclusion criteria, interventions and comparators, and primary study outcomes.

The quality of evidence was assessed using a rating of 1 to 5 based on a modified version of the Oxford Centre for Evidence-based Medicine Levels of Evidence and Grades of Recommendation.¹⁹ A rating of 1 indicated a properly powered and conducted RCT, 2 demonstrated a well-designed controlled trial without randomization or prospective comparative cohort trial, 3 designated a case-control study or retrospective cohort study, 4 signified a case series with or without intervention or a cross-sectional study, and 5 denoted an opinion of respected authorities or case reports. Two reviewers independently evaluated the quality of evidence. Any identified discrepancies were resolved through discussion and consensus. If consensus was not obtained, a third review author was consulted.

We identified 30 studies; 19 articles did not meet inclusion criteria. The remaining 11 articles were divided into 4 cohorts. Five articles were based on data from the STAR trial, a multicenter study that included adults with moderate-to-severe OSA and inadequate adherence to CPAP.^20-24 Four articles used the same patient selection criteria as the STAR trial for a long-term German postmarket study of upper airway stimulation efficacy with OSA.^25-28 The third and fourth cohorts each consist of 31 patients with moderate-to-severe OSA with CPAP nonadherence or failure.^29,30 The STAR trial included follow-up at 5 years, and the German-postmarket had a follow-up at3 years. The remaining 2 cohorts have 1-year follow-ups.

The Scopus review identified 304 studies; 299 did not meet inclusion criteria and 1 was part of the STAR trial.³¹ The remaining 4 articles were classified as distinct cohorts. Huntley and colleagues included patients from Thomas Jefferson University (TJU) and University of Pittsburgh (UP) academic medical centers.³² The Pordzik and colleagues cohort received implantation at a tertiary medical center, an RCCT, and a 1:1 comparator trial (Table 1).^33-35

STAR Trial

This multicenter, prospective, single-group cohort study was conducted in the US, Germany, Belgium, Netherlands, and France. The STAR trial included 126 patients who were not CPAP therapy adherent. Patients were excluded if they had AHI < 20 or > 50, central sleep apnea > 25% of total AHI, anatomical abnormalities that prevent effective assessment of upper-airway stimulation, complete concentric collapse of the retropalatal airway during drug-induced sleep, neuromuscular disease, hypoglossal-nerve palsy, severe restrictive or obstructive pulmonary disease, moderate-to-severe pulmonary arterial hypertension, severe valvular heart disease, New York Heart Association class III or IV heart failure, recent myocardial infarction or severe cardiac arrhythmias (within the past 6 months), persistent uncontrolled hypertension despite medication use, active psychiatric illness, or coexisting nonrespiratory sleep disorders that would confound functional sleep assessment. Primary outcome measures included the AHI and oxygen desaturation index (ODI) with secondary outcomes using the ESS, the Functional Outcomes of Sleep Questionnaire (FOSQ), and the percentage of sleep time with oxygen saturation < 90%. Of 126 patients who received implantation, 71 underwent an overnight PSG evaluation at 5-year follow-up. Mean (SD) AHI at baseline was reduced with HGNS treatment to from 32.0 (11.8) to 12.4 (16.3). Mean (SD) ESS for 92 participants with 2 measurements declined from 11.6 (5.0) at baseline to 6.9 (4.7) at 5-year follow-up.

The STAR trial included a randomized controlled withdrawal study for 46 patients who had a positive response to therapy to evaluate efficacy and durability of upper airway stimulation. Patients were randomly assigned to therapy maintenance or therapy withdrawal groups for ≥ 1 week. The short-term withdrawal effect was assessed using the original trial outcome measures and indicated that both the withdrawal and maintenance groups showed improvements at 12 months compared with the baseline. However, after the randomized withdrawal, the withdrawal group’s outcome measures deteriorated to baseline levels while the maintenance group showed no change. At 18 months of therapy, outcome measures for both groups were similar to those observed with therapy at 12 months.²⁴ The STAR trial included self-reported outcomes at baseline, 12 months, and 24 months that used ESS to measure daytime sleepiness. These results included subsequent STAR trial reports.^20-24,31

The German Postmarket Cohort

This multicenter, prospective, single-arm study used selection criteria that were based on those used in the STAR trial and included patients with moderate-to-severe OSA and nonadherence to CPAP. Patients were excluded if they had a BMI > 35, AHI < 15 or > 65; central apnea index > 25% of total AHI; or complete concentric collapse at the velopharynx during drug-induced sleep. Measured outcomes included AHI, ODI, FOSQ, and ESS. Among the 60 participants, 38 received implantation and a 3-year follow-up. Mean (SD) AHI decreased from 31.2 (13.2) at baseline to 13.1 (14.1) at follow-up, while mean (SD) ESS decreased from 12.8 (5.3) at baseline to 6.0 (3.2) at follow-up.^25-28

Munich Cohort

This single-center, prospective clinical trial included patients with AHI > 15 and < 65, central apnea index < 25% of total AHI, and nonadherence to CPAP. Patients were excluded if they had a BMI > 35, anatomical abnormalities that would prevent effective assessment of upper-airway stimulation; all other exclusion criteria matched those used in the STAR trial. Among 31 patients who received implants and completed a 1-year follow-up, mean (SD) AHI decreased from 32.9 (11.2) at baseline to 7.1 (5.9) at follow-up and mean (SD) ESS decreased from 12.6 (5.6) at baseline to 5.9 (5.2) at follow-up.²⁹

Kezirian and Colleagues Cohort

This prospective, single-arm, open-label study was conducted at 4 Australian and 4 US sites. Selection criteria included moderate-to-severe OSA with failure of CPAP, AHI of 20 to 100 with ≥ 15 events/hour occurring in sleep that was non-REM (rapid eye movement) sleep, BMI ≤ 40 (Australia) or ≤ 37 (US), and a predominance of hypopneas (≥ 80% of disordered breathing events during sleep). Patients were excluded if they had earlier upper airway surgery, markedly enlarged tonsils, uncontrolled nasal obstruction, severe retrognathia, > 5% central or mixed apneic events, incompletely treated sleep disorders other than OSA, or a major disorder of the pulmonary, cardiac, renal, or nervous systems. Data were reported for 31 patients whose mean (SD) AHI declined from 45.4 (17.5) at baseline to 25.3 (20.6) at 1-year follow-up and mean (SD) ESS score declined from 12.1 (4.6) at baseline to 7.9 (3.8) 1 year later.³⁰

The TJU and UP cohorts are composed of patients who underwent implantation between May 2014 and August 2016 at 2 academic centers.^31,32 Selection criteria was consistent with that used in the STAR trial, and patients completed postoperative titration PSG and outpatient follow-up (48 patients at TJU and 49 at UP). Primary outcomes included AHI, ESS, and O₂ nadir. Secondary outcomes consisted of surgical success and percentage of patients tolerating optimal titration setting at follow-up. Postoperative outcomes were assessed during the titration PSG. Time from initial ESS to postoperative PSG at TJU was 1.7 years and at UP was 1.9 years. Time from initial AHI to postoperative PSG at TJU was 90.4 days and 85.2 days at UP. At TJU, mean (SD) AHI and ESS dropped from 35.9 (20.8) and 11.1 (3.8), respectively at baseline to 6.3 (11.5) and 5.8 (3.4), respectively at follow-up. At UP, mean (SD) AHI and ESS fell from 35.3 (15.3) and 10.9 (4.9), respectively at baseline to 6.3 (6.1) and 6.6 (4.5), respectively at follow-up. There were no site-related differences in rates of AHI, ESS, or surgical success.³¹

Pordzik and Colleagues Cohort

This cohort of 29 patients underwent implantation between February 2020 and June 2022 at a tertiary university medical center with both pre- and postoperative PSG. Selection criteria was consistent with that of the German postmarket cohort. Postoperative PSG was completed a mean (SD) 96.3 (27.0) days after device activation. Mean (SD) AHI dropped from 38.6 (12.7) preoperatively to 24.4 (13.3) postoperatively. Notably, this cohort showed a much lower decrease of postoperative AHI than reported by the STAR trial and UP/TJU cohort.³³

Stimulation vs Sham Trial

This multicenter, double-blinded, randomized, crossover trial assessed the effect of HGNS (stim) vs sham stimulation (sham) in 86 patients that completed both phases of the trial. Primary outcomes included AHI and ESS. Secondary outcomes included FOSQ. No carryover effect was found during the crossover phase. The difference between the phases was−15.5 (95% CI, −18.3 to −12.8) for AHI and −3.3 (95% CI, −4.4 to −2.2) for ESS.³⁴

Comparator

The comparator study used propensity score matching to compare outcomes of HGNS and PAP therapy. Primary outcomes included sleepiness, AHI, and effectiveness with outcome measures of AHI and ESS collected at baseline and 12 months postimplantation. The article reported that 126 of 227 patients were matched 1:1. Both groups showed improvement in AHI and ESS. Mean (SD) AHI for the HGNS group at baseline started at 33.9 (15.1) and decreased to 8.1 (6.3). Mean (SD) ESS for the HGNS group at baseline was 15.4 (3.5) and decreased to 7.5 (4.7). In the PAP comparator group, mean (SD) baseline AHI was 36.8 (21.6) and at follow-up was 6.6 (8.0) and mean (SD) ESS was 14.6 (3.9) at baseline and 10.8 (5.6) at follow-up.³⁵

DISCUSSION

The current clinical data on HGNS suggest that this treatment is effective in adults with moderate-to-severe OSA and effects are sustained at long-term follow-up, as measured by AHI reduction and improvements in sleep related symptoms and quality of life (Table 2). These results have been consistent across several sites.

The STAR trial included a randomized control withdrawal group, for whom HGNS treatment was withdrawn after the 12-month follow-up, and then restored at 18 months.²¹ This revealed that withdrawal of HGNS treatment resulted in deterioration of both objective and subjective measures of OSA and sleepiness. The beneficial effects of HGNS were restored when treatment was resumed.²⁴ Additionally, the RCCT revealed that therapeutic stimulation via HGNS significantly reduced subjective and objective measures of OSA.³⁴ These studies provide definitive evidence of HGNS efficacy.

Currently, a diagnosis of OSA on PAP is classified as a 50% military disability rating. This rating is based primarily on epidemiologic evidence that untreated OSA is a costly disease that leads to other chronic illnesses that increases health care utilization.⁹ HGNS requires an initially invasive procedure and higher upfront costs, but it could result in reduced health care use and long-term costs because of improved adherence to treatment—compared with CPAP—that results in better outcomes.

The reviewed studies have several limitations that warrant caution when determining the possible benefits of HGNS treatment. The primary limitation is the lack of active control groups, therefore precluding a direct comparison of the short- and long-term effectiveness of HGNS vs other treatments (eg, CPAP). This is especially problematic because in the reviewed studies HGNS treatment efficacy is reported as a function of the mean—and SD—percent reduction in the AHI, whereas the efficacy of CPAP treatment usually is defined in terms of “adequacy of titration” as suggested by the AASM.³⁶ It has been reported that with CPAP treatment, 50% to 60% of OSA patients achieve AASM-defined optimal improvement of respiratory disturbance index of < 5/hour during a polysomnographic sleep recording of ≥ 15 minutes duration that includes REM sleep in the supine position.³⁷ In most of the reviewed studies, treatment success was more liberally defined as a decrease of AHI by ≥ 50%, regardless of the resulting AHI. It is notable that among the reviewed HGNS studies, the TJU and UP cohorts achieved the best outcome in short-term follow-up of 2 months with a mean (SD) AHI of 6.3 (11.5) and 6.4 (6.1), respectively. Among those cohortsassessed at a 12-month follow-up, the Munich cohort achieved the best outcome with a mean (SD) AHI of 7.1 (5.9).

Although the metrics reported in the reviewed studies are not directly comparable, the reported findings strongly suggest that HGNS generally is less effective than CPAP. How important are these differences? With findings that HGNS “reliably produces clinically meaningful (positive) effects on daytime sleepiness, daytime functioning, and sleep quality,” does it really matter if the outcome metrics for HGNS are a little less positive than those produced by CPAP?³⁸ For individual military OSA patients the answer is yes. This is because in military operational environments—especially during deployment—sleep restriction is nearly ubiquitous, therefore any mild residual deficits in sleep quality and daytime alertness resulting from nominally adequate, but suboptimal OSA treatment, could be exacerbated by sleep restriction, therefore placing the service member and the mission at increased risk.³⁹

Another limitation is the narrow inclusion criteria these studies employed, which limits the generalizability of the findings. Participants in the reviewed clinical trials were selected from a patient population that was mostly middle-aged, White, and obese or overweight. In a Medical Surveillance Monthly Report study, OSA was found to be highest among service members aged > 40 years, male, obese, and Black/non-Hispanic (although it should be noted that more than one-half of enlisted service members aged ≤ 25 years).^40,41 Obesity has been noted as a growing concern for the military as the military population is beginning to mirror the civilian population in terms of being overweight or obese despite height and weight standards. HGNS might not be as successful in military populations with different demographics. Moreover, HGNS has been shown to have greater AHI reduction among those with higher BMI.³⁰ Although obese service members have a 6-fold higher 12-year incidence rate of OSA than service members without obesity, this nevertheless suggests that general level of HGNS efficacy might be lower among the military patient population, because obesity is less prevalent in the military than the general population.⁹

Ethnicity has been found to be a relevant factor, with the highest incidence rate of OSA among non-Hispanic Black males, a demographic that was underrepresented in cohorts included in this review. Further studies will be needed to determine the extent to which findings from HGNS treatment studies are generalizable to the broader OSA patient population.

HGNS Implementation Challenges

Current impediments to widespread use of HGNS as an OSA treatment include no standardized guidance for titration and follow-on care, which varies based on the resources available. Titrating a new device for HGNS requires experienced sleep technicians who have close relationships with device representatives and can troubleshoot problems. Technical expertise, which currently is rare, is required if there are complications after placement or if adjustments to voltage settings are needed over time. In addition, patients may require multiple specialists making it easy to get lost to follow-up after implantation. This is particularly challenging in a transient community, such as the military, because there is no guarantee that a service member will have access to the same specialty care at the next duty station.

Although some evidence suggests that HGNS is a viable alternative treatment for some patients with OSA, the generalizability of these findings to the military patient population is unclear. Specialized facilities and expertise are needed for the surgical procedure and follow-up requirements, which currently constitute significant logistical constraints. As with any implantable device, there is a risk of complications including infection that could result in medical evacuation from a theater of operations. If the device malfunctions or loses effectiveness in a deployed environment, the service member might not have immediate access to medical support, potentially leading to undertreatment of OSA. In future battlefield scenarios in multidomain operations, prolonged, far-forward field care will become the new normal because the military is not expected to have air superiority or the ability to quickly evacuate service members to a higher level of medical care.⁴²

In deployed environments, the potential limitations of HGNS become increasingly risky for the service member and the overall mission. Considering these factors, it will be important to evaluate the practicality of HGNS as a treatment option in military populations. Military-specific challenges associated with HGNS that require further study, include guidance for patient selection outside academic centers, guidance on long-term postsurgical care and device maintenance, duty limitation and military retention considerations, and limitations in training and combat environments. The military medical community needs to conduct its own studies in appropriately selected service members to guide clinical practice.

CONCLUSIONS

HGNS treatment results in improvement of both AHI and ESS scores and could be a deployable treatment option for military patients with OSA. However, HGNS has not been found to be as effective as CPAP, although the current literature is limited by small sample sizes, homogeneous populations that do not reflect the demographics of the military, and mostly short follow-up periods. Future studies should be focused on collecting data on HGNS from demographic groups that are more representative of the military OSA patient population and identifying the subpopulation of patients who derive the greatest benefit from HGNS, so that this treatment can be better individually targeted. Until data on existing military patients is published, it is not possible to fully weigh risks and benefits in this population and generalize civilian guidance to the military.

The Olin E. Teague Veterans’ Center (OETVC) in Temple, Texas, is a teaching hospital with 189 beds that provides patients access to medical, surgical, and specialty care. In 2022, 116,359 veterans received care at OETVC and 5393 inpatient admissions were noted. The inpatient ward consists of 3 teaching teams staffed by an attending physician, a second-year internal medicine resident, and 2 to 3 interns while hospitalists staff the 3 nonteaching teams. OETVC residents receive training on both routine and complex medical problems.

Each day, teaching teams discharge patients. With the complexity of discharges, there is always a risk of patients not following up with their primary care physicians, potential issues with filling medications, confusion about new medication regiments, and even potential postdischarge questions. In 1990, Holloway and colleagues evaluated potential risk factors for readmission among veterans. This study found that discharge from a geriatrics or intermediate care bed, chronic disease diagnosis, ≥ 2 procedures performed, increasing age, and distance from a veterans affairs medical center were risk factors.¹

Several community hospital studies have evaluated readmission risk factors. One from 2000 noted that patients with more hospitalizations, lower mental health function, a diagnosis of chronic obstructive pulmonary disorder, and increased satisfaction with access to emergency care were associated with increased readmission in 90 days.² Due to the readmission risks, OETVC decided to construct a program that would help these patients successfully transition from inpatient to outpatient care while establishing means to discuss their care with a physician for reassurance and guidance.

TRANSITION OF CARE PROGRAM

Transition of care programs have been implemented and evaluated in many institutions. A 2017 systematic review of transition of care programs supported the use of tailored discharge planning and postdischarge phone calls to reduce hospital readmission, noting that 6 studies demonstrated a statistically significant reduction in 30-day readmission rate.³ Another study found that pharmacy involvement in the transition of care reduced medication-related problems following discharge.⁴

Program Goals

The foundational goal of our program was to bridge the gap between inpatient and outpatient medicine. We hoped to improve patient adherence with their discharge regimens, improve access to primary care physicians, and improve discharge follow-up. Since hospitalization can be overwhelming, we hoped to capture potential barriers to medical care postdischarge when patients return home while decreasing hospital readmissions. Our second- and third-year resident physicians spend as much time as needed going through the patient’s course of illness throughout their hospitalization and treatment plans to ensure their understanding and potential success.

This program benefits residents by providing medical education and patient communication opportunities. Residents must review the patient’s clinical trajectory before calling them. In this process, residents develop an understanding of routine and complex illness scripts, or pathways of common illnesses. They also prepare for potential questions about the hospitalization, new medications, and follow-up care. Lastly, residents can focus on communication skills. Without the time pressures of returning to a busy rotation, the residents spend as much time discussing the hospital course and ensuring patient understanding as needed.

At the beginning of each week, second- and third-year residents review the list of discharges from the 3 teaching teams. The list is generated by a medical service management analyst. The residents review patient records for inpatient services, laboratory results, medication changes, and proposed follow-up plans designed by the admission team prior to their phone call. The resident is also responsible for reviewing and reconciling discharge instructions and orders. Then, the resident calls the patient and reviews their hospitalization. If a patient does not answer, the resident leaves a voicemail that complies with the Health Insurance Portability and Accountability Act.

When patients answer the call, the resident follows a script (Appendix). Residents are encouraged to ask patients open-ended questions and address any new needs. They also discuss changes in symptoms, medications, functional status, and remind the patient about follow-up appointments. If imaging or specific orders were missed at discharge, the residents notify the chief resident, lead hospitalist, or deputy associate chief of staff for medical service. If additional laboratory tests need to be ordered, the resident devises a follow-up plan. If needed, specialty referrals can be placed. When residents feel there are multiple items that need to be addressed or if they notice any major concerns, they can recommend the patient present to the emergency department for evaluation. The chief resident, lead hospitalist, and deputy associate chair for medical service are available to assist with discussions about complex medical situations or new concerning symptoms. Residents document their encounters in the Computerized Patient Record System health record and any tests that need follow-up. This differs from the standard of care follow-up programs, which are conducted by primary care medicine nurses and do not fully discuss the hospitalization.

Implementation

This program was implemented as a 1-week elective for interested residents and part of the clinic rotation. The internal medicine medical service analyst pulls all discharges on Friday, which are then provided to the residents. The residents on rotation work through the discharges and find teaching team patients to follow up with and call.

Findings

Implementation of this program has yielded many benefits. The reminder of the importance of a primary care appointment has motivated patients to continue following up on an outpatient basis. Residents were also able to capture lapses in patient understanding. Residents could answer forgotten questions and help patients understand their admission pathology without time pressures. Residents have identified patients with hypoglycemia due to changed insulin regimens, set up specialist follow-up appointments, and provided additional education facilitating adherence. Additionally, several residents have expressed satisfaction with the ability to practice their communication skills. Others appreciated contributing to future patient successes.

While the focus on this article has been to share the program description, we have tabulated preliminary data. In January 2023, there were 239 internal medicine admissions; 158 admissions (66%) were teaching team patients, and 97 patients (61%) were called by a resident and spoken to regarding their care. There were 24 teaching team readmissions within 30 days, and 10 (42%) received a follow-up phone call. Eighty-one admitted patients were treated by nonteaching teams, 10 (12%) of whom were readmissions. Comparing 30-day readmission rates, 10 nonteaching team patients (12%), 10 teaching team patients (6.3%) who talk to a resident in the transition of care program were readmitted, and 24 teaching team patients who did not talk to a resident (10%) were readmitted.

The OETVC transition of care program was planned, formulated, and implemented without modeling after any other projects or institutions. This program aimed to utilize our residents as resources for patients.

Transition of care is defined as steps taken in a clinical encounter to assist with the coordination and continuity of patient care transferring between locations or levels of care.⁵ A 2018 study evaluating the utility of transition of care programs on adults aged ≥ 60 years found a reduction in rehospitalization rates, increased use of primary care services, and potential reduction in home health usage.⁶

In 2021, Johns Hopkins University School of Medicine implemented a program after polling residents and discovering their awareness of gaps in the transition of care.⁷ In 2002, pharmacists evaluated the impact of follow-up telephone calls to recently hospitalized patients. This group of pharmacists found that these calls were associated with increased patient satisfaction, resolution of medication-related problems and fewer emergency department returns.⁸

Our program differs from other transition of care programs in that resident physicians made the follow-up calls to patients. Residents could address all aspects of medical care, including new symptoms, new prescriptions, adverse events, and risk factors for readmission, or order new imaging and medications when appropriate. In the program, residents called all patients discharged after receiving care within their team. Calls were not based on risk assessments. The residents were able to speak with 61% of discharged patients. When readmission rates were compared between patients who received a resident follow-up phone call and those who did not, patients receiving the resident phone call were readmitted at a lower rate: 6.3% vs 10%, respectively.

While our data suggest a potential trend of decreased readmission, more follow-up over a longer period may be needed. We believe this program can benefit patients and our model can act as a template for other institutions interested in starting their own programs.

Challenges

Although our process is efficient, there have been some challenges. The discharge is created by the medical service management analyst and then sent to the chief resident, but there was concern that the list could be missed if either individual was unavailable. The chairperson for the department of medicine and their secretary are now involved in the process. To reduce unanswered telephone calls, residents use OETVC phones. Health Insurance Portability and Accountability Act-compliant voicemails providing a time for a follow-up call were implemented. As a result, veterans have answered their phones more regularly and are more aware of calls. Orders are generally placed by the chief resident, lead hospitalist, or chair of the medical service to ensure follow-up because residents are on rotation for 1 week at a time. Access to a physician also allows patients to discuss items unrelated to their hospitalization, introducing new symptoms, or situations requiring a resident to act with limited data.

CONCLUSIONS

The transition of care follow-up program described in this article may be beneficial for both internal medicine residents and patients. Second- and third-year residents are developing a better understanding of the trajectory of many illnesses and are given the opportunity to retrospectively analyze what they would do differently based on knowledge gained from their chart reviews. They are also given the opportunity to work on communication skills and explain courses of illnesses to patients in an easy-to-understand format without time constraints. Patients now have access to a physician following discharge to discuss any concerns with their hospitalization, condition, and follow-up. This program will continue to address barriers to care and adapt to improve the success of care transitions.

The Olin E. Teague Veterans’ Center (OETVC) in Temple, Texas, is a teaching hospital with 189 beds that provides patients access to medical, surgical, and specialty care. In 2022, 116,359 veterans received care at OETVC and 5393 inpatient admissions were noted. The inpatient ward consists of 3 teaching teams staffed by an attending physician, a second-year internal medicine resident, and 2 to 3 interns while hospitalists staff the 3 nonteaching teams. OETVC residents receive training on both routine and complex medical problems.

Each day, teaching teams discharge patients. With the complexity of discharges, there is always a risk of patients not following up with their primary care physicians, potential issues with filling medications, confusion about new medication regiments, and even potential postdischarge questions. In 1990, Holloway and colleagues evaluated potential risk factors for readmission among veterans. This study found that discharge from a geriatrics or intermediate care bed, chronic disease diagnosis, ≥ 2 procedures performed, increasing age, and distance from a veterans affairs medical center were risk factors.¹

Several community hospital studies have evaluated readmission risk factors. One from 2000 noted that patients with more hospitalizations, lower mental health function, a diagnosis of chronic obstructive pulmonary disorder, and increased satisfaction with access to emergency care were associated with increased readmission in 90 days.² Due to the readmission risks, OETVC decided to construct a program that would help these patients successfully transition from inpatient to outpatient care while establishing means to discuss their care with a physician for reassurance and guidance.

TRANSITION OF CARE PROGRAM

Transition of care programs have been implemented and evaluated in many institutions. A 2017 systematic review of transition of care programs supported the use of tailored discharge planning and postdischarge phone calls to reduce hospital readmission, noting that 6 studies demonstrated a statistically significant reduction in 30-day readmission rate.³ Another study found that pharmacy involvement in the transition of care reduced medication-related problems following discharge.⁴

Program Goals

The foundational goal of our program was to bridge the gap between inpatient and outpatient medicine. We hoped to improve patient adherence with their discharge regimens, improve access to primary care physicians, and improve discharge follow-up. Since hospitalization can be overwhelming, we hoped to capture potential barriers to medical care postdischarge when patients return home while decreasing hospital readmissions. Our second- and third-year resident physicians spend as much time as needed going through the patient’s course of illness throughout their hospitalization and treatment plans to ensure their understanding and potential success.

This program benefits residents by providing medical education and patient communication opportunities. Residents must review the patient’s clinical trajectory before calling them. In this process, residents develop an understanding of routine and complex illness scripts, or pathways of common illnesses. They also prepare for potential questions about the hospitalization, new medications, and follow-up care. Lastly, residents can focus on communication skills. Without the time pressures of returning to a busy rotation, the residents spend as much time discussing the hospital course and ensuring patient understanding as needed.

At the beginning of each week, second- and third-year residents review the list of discharges from the 3 teaching teams. The list is generated by a medical service management analyst. The residents review patient records for inpatient services, laboratory results, medication changes, and proposed follow-up plans designed by the admission team prior to their phone call. The resident is also responsible for reviewing and reconciling discharge instructions and orders. Then, the resident calls the patient and reviews their hospitalization. If a patient does not answer, the resident leaves a voicemail that complies with the Health Insurance Portability and Accountability Act.

When patients answer the call, the resident follows a script (Appendix). Residents are encouraged to ask patients open-ended questions and address any new needs. They also discuss changes in symptoms, medications, functional status, and remind the patient about follow-up appointments. If imaging or specific orders were missed at discharge, the residents notify the chief resident, lead hospitalist, or deputy associate chief of staff for medical service. If additional laboratory tests need to be ordered, the resident devises a follow-up plan. If needed, specialty referrals can be placed. When residents feel there are multiple items that need to be addressed or if they notice any major concerns, they can recommend the patient present to the emergency department for evaluation. The chief resident, lead hospitalist, and deputy associate chair for medical service are available to assist with discussions about complex medical situations or new concerning symptoms. Residents document their encounters in the Computerized Patient Record System health record and any tests that need follow-up. This differs from the standard of care follow-up programs, which are conducted by primary care medicine nurses and do not fully discuss the hospitalization.

Implementation

This program was implemented as a 1-week elective for interested residents and part of the clinic rotation. The internal medicine medical service analyst pulls all discharges on Friday, which are then provided to the residents. The residents on rotation work through the discharges and find teaching team patients to follow up with and call.

Findings

Implementation of this program has yielded many benefits. The reminder of the importance of a primary care appointment has motivated patients to continue following up on an outpatient basis. Residents were also able to capture lapses in patient understanding. Residents could answer forgotten questions and help patients understand their admission pathology without time pressures. Residents have identified patients with hypoglycemia due to changed insulin regimens, set up specialist follow-up appointments, and provided additional education facilitating adherence. Additionally, several residents have expressed satisfaction with the ability to practice their communication skills. Others appreciated contributing to future patient successes.

While the focus on this article has been to share the program description, we have tabulated preliminary data. In January 2023, there were 239 internal medicine admissions; 158 admissions (66%) were teaching team patients, and 97 patients (61%) were called by a resident and spoken to regarding their care. There were 24 teaching team readmissions within 30 days, and 10 (42%) received a follow-up phone call. Eighty-one admitted patients were treated by nonteaching teams, 10 (12%) of whom were readmissions. Comparing 30-day readmission rates, 10 nonteaching team patients (12%), 10 teaching team patients (6.3%) who talk to a resident in the transition of care program were readmitted, and 24 teaching team patients who did not talk to a resident (10%) were readmitted.

The OETVC transition of care program was planned, formulated, and implemented without modeling after any other projects or institutions. This program aimed to utilize our residents as resources for patients.

Transition of care is defined as steps taken in a clinical encounter to assist with the coordination and continuity of patient care transferring between locations or levels of care.⁵ A 2018 study evaluating the utility of transition of care programs on adults aged ≥ 60 years found a reduction in rehospitalization rates, increased use of primary care services, and potential reduction in home health usage.⁶

In 2021, Johns Hopkins University School of Medicine implemented a program after polling residents and discovering their awareness of gaps in the transition of care.⁷ In 2002, pharmacists evaluated the impact of follow-up telephone calls to recently hospitalized patients. This group of pharmacists found that these calls were associated with increased patient satisfaction, resolution of medication-related problems and fewer emergency department returns.⁸

Our program differs from other transition of care programs in that resident physicians made the follow-up calls to patients. Residents could address all aspects of medical care, including new symptoms, new prescriptions, adverse events, and risk factors for readmission, or order new imaging and medications when appropriate. In the program, residents called all patients discharged after receiving care within their team. Calls were not based on risk assessments. The residents were able to speak with 61% of discharged patients. When readmission rates were compared between patients who received a resident follow-up phone call and those who did not, patients receiving the resident phone call were readmitted at a lower rate: 6.3% vs 10%, respectively.

While our data suggest a potential trend of decreased readmission, more follow-up over a longer period may be needed. We believe this program can benefit patients and our model can act as a template for other institutions interested in starting their own programs.

Challenges

Although our process is efficient, there have been some challenges. The discharge is created by the medical service management analyst and then sent to the chief resident, but there was concern that the list could be missed if either individual was unavailable. The chairperson for the department of medicine and their secretary are now involved in the process. To reduce unanswered telephone calls, residents use OETVC phones. Health Insurance Portability and Accountability Act-compliant voicemails providing a time for a follow-up call were implemented. As a result, veterans have answered their phones more regularly and are more aware of calls. Orders are generally placed by the chief resident, lead hospitalist, or chair of the medical service to ensure follow-up because residents are on rotation for 1 week at a time. Access to a physician also allows patients to discuss items unrelated to their hospitalization, introducing new symptoms, or situations requiring a resident to act with limited data.

CONCLUSIONS

The transition of care follow-up program described in this article may be beneficial for both internal medicine residents and patients. Second- and third-year residents are developing a better understanding of the trajectory of many illnesses and are given the opportunity to retrospectively analyze what they would do differently based on knowledge gained from their chart reviews. They are also given the opportunity to work on communication skills and explain courses of illnesses to patients in an easy-to-understand format without time constraints. Patients now have access to a physician following discharge to discuss any concerns with their hospitalization, condition, and follow-up. This program will continue to address barriers to care and adapt to improve the success of care transitions.

The Olin E. Teague Veterans’ Center (OETVC) in Temple, Texas, is a teaching hospital with 189 beds that provides patients access to medical, surgical, and specialty care. In 2022, 116,359 veterans received care at OETVC and 5393 inpatient admissions were noted. The inpatient ward consists of 3 teaching teams staffed by an attending physician, a second-year internal medicine resident, and 2 to 3 interns while hospitalists staff the 3 nonteaching teams. OETVC residents receive training on both routine and complex medical problems.

Each day, teaching teams discharge patients. With the complexity of discharges, there is always a risk of patients not following up with their primary care physicians, potential issues with filling medications, confusion about new medication regiments, and even potential postdischarge questions. In 1990, Holloway and colleagues evaluated potential risk factors for readmission among veterans. This study found that discharge from a geriatrics or intermediate care bed, chronic disease diagnosis, ≥ 2 procedures performed, increasing age, and distance from a veterans affairs medical center were risk factors.¹

Several community hospital studies have evaluated readmission risk factors. One from 2000 noted that patients with more hospitalizations, lower mental health function, a diagnosis of chronic obstructive pulmonary disorder, and increased satisfaction with access to emergency care were associated with increased readmission in 90 days.² Due to the readmission risks, OETVC decided to construct a program that would help these patients successfully transition from inpatient to outpatient care while establishing means to discuss their care with a physician for reassurance and guidance.

TRANSITION OF CARE PROGRAM

Transition of care programs have been implemented and evaluated in many institutions. A 2017 systematic review of transition of care programs supported the use of tailored discharge planning and postdischarge phone calls to reduce hospital readmission, noting that 6 studies demonstrated a statistically significant reduction in 30-day readmission rate.³ Another study found that pharmacy involvement in the transition of care reduced medication-related problems following discharge.⁴

Program Goals

The foundational goal of our program was to bridge the gap between inpatient and outpatient medicine. We hoped to improve patient adherence with their discharge regimens, improve access to primary care physicians, and improve discharge follow-up. Since hospitalization can be overwhelming, we hoped to capture potential barriers to medical care postdischarge when patients return home while decreasing hospital readmissions. Our second- and third-year resident physicians spend as much time as needed going through the patient’s course of illness throughout their hospitalization and treatment plans to ensure their understanding and potential success.

This program benefits residents by providing medical education and patient communication opportunities. Residents must review the patient’s clinical trajectory before calling them. In this process, residents develop an understanding of routine and complex illness scripts, or pathways of common illnesses. They also prepare for potential questions about the hospitalization, new medications, and follow-up care. Lastly, residents can focus on communication skills. Without the time pressures of returning to a busy rotation, the residents spend as much time discussing the hospital course and ensuring patient understanding as needed.

At the beginning of each week, second- and third-year residents review the list of discharges from the 3 teaching teams. The list is generated by a medical service management analyst. The residents review patient records for inpatient services, laboratory results, medication changes, and proposed follow-up plans designed by the admission team prior to their phone call. The resident is also responsible for reviewing and reconciling discharge instructions and orders. Then, the resident calls the patient and reviews their hospitalization. If a patient does not answer, the resident leaves a voicemail that complies with the Health Insurance Portability and Accountability Act.

When patients answer the call, the resident follows a script (Appendix). Residents are encouraged to ask patients open-ended questions and address any new needs. They also discuss changes in symptoms, medications, functional status, and remind the patient about follow-up appointments. If imaging or specific orders were missed at discharge, the residents notify the chief resident, lead hospitalist, or deputy associate chief of staff for medical service. If additional laboratory tests need to be ordered, the resident devises a follow-up plan. If needed, specialty referrals can be placed. When residents feel there are multiple items that need to be addressed or if they notice any major concerns, they can recommend the patient present to the emergency department for evaluation. The chief resident, lead hospitalist, and deputy associate chair for medical service are available to assist with discussions about complex medical situations or new concerning symptoms. Residents document their encounters in the Computerized Patient Record System health record and any tests that need follow-up. This differs from the standard of care follow-up programs, which are conducted by primary care medicine nurses and do not fully discuss the hospitalization.

Implementation

This program was implemented as a 1-week elective for interested residents and part of the clinic rotation. The internal medicine medical service analyst pulls all discharges on Friday, which are then provided to the residents. The residents on rotation work through the discharges and find teaching team patients to follow up with and call.

Findings

Implementation of this program has yielded many benefits. The reminder of the importance of a primary care appointment has motivated patients to continue following up on an outpatient basis. Residents were also able to capture lapses in patient understanding. Residents could answer forgotten questions and help patients understand their admission pathology without time pressures. Residents have identified patients with hypoglycemia due to changed insulin regimens, set up specialist follow-up appointments, and provided additional education facilitating adherence. Additionally, several residents have expressed satisfaction with the ability to practice their communication skills. Others appreciated contributing to future patient successes.

While the focus on this article has been to share the program description, we have tabulated preliminary data. In January 2023, there were 239 internal medicine admissions; 158 admissions (66%) were teaching team patients, and 97 patients (61%) were called by a resident and spoken to regarding their care. There were 24 teaching team readmissions within 30 days, and 10 (42%) received a follow-up phone call. Eighty-one admitted patients were treated by nonteaching teams, 10 (12%) of whom were readmissions. Comparing 30-day readmission rates, 10 nonteaching team patients (12%), 10 teaching team patients (6.3%) who talk to a resident in the transition of care program were readmitted, and 24 teaching team patients who did not talk to a resident (10%) were readmitted.

The OETVC transition of care program was planned, formulated, and implemented without modeling after any other projects or institutions. This program aimed to utilize our residents as resources for patients.

Transition of care is defined as steps taken in a clinical encounter to assist with the coordination and continuity of patient care transferring between locations or levels of care.⁵ A 2018 study evaluating the utility of transition of care programs on adults aged ≥ 60 years found a reduction in rehospitalization rates, increased use of primary care services, and potential reduction in home health usage.⁶

In 2021, Johns Hopkins University School of Medicine implemented a program after polling residents and discovering their awareness of gaps in the transition of care.⁷ In 2002, pharmacists evaluated the impact of follow-up telephone calls to recently hospitalized patients. This group of pharmacists found that these calls were associated with increased patient satisfaction, resolution of medication-related problems and fewer emergency department returns.⁸

Our program differs from other transition of care programs in that resident physicians made the follow-up calls to patients. Residents could address all aspects of medical care, including new symptoms, new prescriptions, adverse events, and risk factors for readmission, or order new imaging and medications when appropriate. In the program, residents called all patients discharged after receiving care within their team. Calls were not based on risk assessments. The residents were able to speak with 61% of discharged patients. When readmission rates were compared between patients who received a resident follow-up phone call and those who did not, patients receiving the resident phone call were readmitted at a lower rate: 6.3% vs 10%, respectively.

While our data suggest a potential trend of decreased readmission, more follow-up over a longer period may be needed. We believe this program can benefit patients and our model can act as a template for other institutions interested in starting their own programs.

Challenges

Although our process is efficient, there have been some challenges. The discharge is created by the medical service management analyst and then sent to the chief resident, but there was concern that the list could be missed if either individual was unavailable. The chairperson for the department of medicine and their secretary are now involved in the process. To reduce unanswered telephone calls, residents use OETVC phones. Health Insurance Portability and Accountability Act-compliant voicemails providing a time for a follow-up call were implemented. As a result, veterans have answered their phones more regularly and are more aware of calls. Orders are generally placed by the chief resident, lead hospitalist, or chair of the medical service to ensure follow-up because residents are on rotation for 1 week at a time. Access to a physician also allows patients to discuss items unrelated to their hospitalization, introducing new symptoms, or situations requiring a resident to act with limited data.

CONCLUSIONS

The transition of care follow-up program described in this article may be beneficial for both internal medicine residents and patients. Second- and third-year residents are developing a better understanding of the trajectory of many illnesses and are given the opportunity to retrospectively analyze what they would do differently based on knowledge gained from their chart reviews. They are also given the opportunity to work on communication skills and explain courses of illnesses to patients in an easy-to-understand format without time constraints. Patients now have access to a physician following discharge to discuss any concerns with their hospitalization, condition, and follow-up. This program will continue to address barriers to care and adapt to improve the success of care transitions.