Case Reports

Case Reports

Patient 1
A 65-year-old woman presented to the dermatology clinic in July with a pruritic rash of 2 days’ duration that started on the back and spread diffusely. The patient gardened regularly. Physical examination showed inflammatory papules and pustules on the back (Figure 1), as well as the groin, breasts, and ears. There was a punctate black dot in the center of some papules, and dermoscopy revealed ticks (Figure 2). Removal and microscopic examination confirmed larval-stage lone star ticks (Figure 3). The patient was prescribed topical steroids for pruritus as well as oral doxycycline for prophylaxis against tick-borne illnesses.

Patient 2
A 54-year-old man presented to the same clinic in July with pruritic lesions on the back, legs, ankles, and scrotum of 3 days’ duration that first appeared 24 hours after performing yardwork. Physical examination revealed diffusely distributed papules, pustules, and vesicles on the back (Figure 4). Some papules featured a punctate black dot in the center (similar to patient 1), and dermoscopy again revealed ticks. Removal and microscopic examination confirmed larval-stage ticks. The patient was treated with topical steroids and oral antihistamines for pruritus as well as prophylactic oral doxycycline.

Comment

Ticks are well-known human parasites, representing the second most common vector of human infectious disease.¹ Ticks have 3 motile stages: larva (or “seed”), nymph, and adult. They can bite humans during all stages. Larval ticks, distinguished by having 6 legs rather than 8 legs in nymphs and adults, can attack in droves and cause an infestation that presents as diffuse, pruritic, erythematous papules and pustules.^2-4 The first report of larval tick infestation in humans may have been in 1728 by William Byrd who described finding ticks on the skin that were too small to see without a microscope.⁵

Identification
The ticks in both of our cases were lone star ticks (Amblyomma americanum). The larval stage of A americanum is a proven cause of cutaneous reaction.^6,7 A PubMed search of articles indexed for MEDLINE as well as a Google Scholar search using the terms tick, seed tick, or tick bite in combination with rash, eruption, infestation, papule, pustule, or pruritic revealed 6 reported cases of larval tick infestation in the literature (including our case); 5 were caused by A americanum and 1 by Ixodes dammini (now known as Ixodes scapularis); all occurred in July or August.^3,7-10 This time frame is consistent with the general tick life cycle across species: Adults feed from April to June, then lay eggs that hatch into larval ticks within 4 to 6 weeks. After hatching, larval ticks climb grass and weeds awaiting a passing host.⁴

Diagnosis
Larval tick infestation remains a frequently misdiagnosed etiology of diffuse pruritic papules and pustules, especially in urban settings where physicians are less likely to be familiar with this type of manifestation.^3,9-11 Larval ticks are submillimeter in size and difficult to appreciate with the naked eye, contributing to misdiagnosis. A punctate black dot may sometimes be seen in papules; however, dermoscopy is critical for accurate diagnosis, as hemorrhagic crust is a frequent misdiagnosis.

Management
In addition to symptomatic therapy, both of our patients received doxycycline as antibiotic prophylaxis for tick-borne illnesses given that a high number of ticks had been attached for more than 2 days.^12,13 Antibiotic prophylaxis for tick-borne illness is controversial. The exception is Lyme disease transmitted by nymphal or adult I scapularis when specific conditions are met: the bite must have occurred in an endemic area, doxycycline cannot be contraindicated, estimated duration of attachment is at least 36 hours, and prophylaxis must be started within 72 hours of tick removal.¹³ There are no official recommendations for the A americanum species or for larval-stage ticks of any species. Larval-stage ticks acting as vectors for disease transmission is not well documented in recent literature, and there currently is limited evidence supporting prophylactic antibiotics for larval tick bites. The presence of spotted fever rickettsioses has been reported (with the exception of Rickettsia rickettsii and Ehrlichia chaffeensis) in larval A americanum ticks, suggesting a theoretical possibility that they could act as disease vectors.^3,8,11,14-17 At a minimum, both prompt tick removal and close patient follow-up is warranted.

Conclusion

Human infestation with larval ticks is a common occurrence but can present a diagnostic challenge to an unfamiliar physician. We encourage consideration of larval tick infestation as the etiology of multiple or diffuse pruritic papules with a history of outdoor exposure.

Case Reports

Patient 1
A 65-year-old woman presented to the dermatology clinic in July with a pruritic rash of 2 days’ duration that started on the back and spread diffusely. The patient gardened regularly. Physical examination showed inflammatory papules and pustules on the back (Figure 1), as well as the groin, breasts, and ears. There was a punctate black dot in the center of some papules, and dermoscopy revealed ticks (Figure 2). Removal and microscopic examination confirmed larval-stage lone star ticks (Figure 3). The patient was prescribed topical steroids for pruritus as well as oral doxycycline for prophylaxis against tick-borne illnesses.

Patient 2
A 54-year-old man presented to the same clinic in July with pruritic lesions on the back, legs, ankles, and scrotum of 3 days’ duration that first appeared 24 hours after performing yardwork. Physical examination revealed diffusely distributed papules, pustules, and vesicles on the back (Figure 4). Some papules featured a punctate black dot in the center (similar to patient 1), and dermoscopy again revealed ticks. Removal and microscopic examination confirmed larval-stage ticks. The patient was treated with topical steroids and oral antihistamines for pruritus as well as prophylactic oral doxycycline.

Comment

Ticks are well-known human parasites, representing the second most common vector of human infectious disease.¹ Ticks have 3 motile stages: larva (or “seed”), nymph, and adult. They can bite humans during all stages. Larval ticks, distinguished by having 6 legs rather than 8 legs in nymphs and adults, can attack in droves and cause an infestation that presents as diffuse, pruritic, erythematous papules and pustules.^2-4 The first report of larval tick infestation in humans may have been in 1728 by William Byrd who described finding ticks on the skin that were too small to see without a microscope.⁵

Identification
The ticks in both of our cases were lone star ticks (Amblyomma americanum). The larval stage of A americanum is a proven cause of cutaneous reaction.^6,7 A PubMed search of articles indexed for MEDLINE as well as a Google Scholar search using the terms tick, seed tick, or tick bite in combination with rash, eruption, infestation, papule, pustule, or pruritic revealed 6 reported cases of larval tick infestation in the literature (including our case); 5 were caused by A americanum and 1 by Ixodes dammini (now known as Ixodes scapularis); all occurred in July or August.^3,7-10 This time frame is consistent with the general tick life cycle across species: Adults feed from April to June, then lay eggs that hatch into larval ticks within 4 to 6 weeks. After hatching, larval ticks climb grass and weeds awaiting a passing host.⁴

Diagnosis
Larval tick infestation remains a frequently misdiagnosed etiology of diffuse pruritic papules and pustules, especially in urban settings where physicians are less likely to be familiar with this type of manifestation.^3,9-11 Larval ticks are submillimeter in size and difficult to appreciate with the naked eye, contributing to misdiagnosis. A punctate black dot may sometimes be seen in papules; however, dermoscopy is critical for accurate diagnosis, as hemorrhagic crust is a frequent misdiagnosis.

Management
In addition to symptomatic therapy, both of our patients received doxycycline as antibiotic prophylaxis for tick-borne illnesses given that a high number of ticks had been attached for more than 2 days.^12,13 Antibiotic prophylaxis for tick-borne illness is controversial. The exception is Lyme disease transmitted by nymphal or adult I scapularis when specific conditions are met: the bite must have occurred in an endemic area, doxycycline cannot be contraindicated, estimated duration of attachment is at least 36 hours, and prophylaxis must be started within 72 hours of tick removal.¹³ There are no official recommendations for the A americanum species or for larval-stage ticks of any species. Larval-stage ticks acting as vectors for disease transmission is not well documented in recent literature, and there currently is limited evidence supporting prophylactic antibiotics for larval tick bites. The presence of spotted fever rickettsioses has been reported (with the exception of Rickettsia rickettsii and Ehrlichia chaffeensis) in larval A americanum ticks, suggesting a theoretical possibility that they could act as disease vectors.^3,8,11,14-17 At a minimum, both prompt tick removal and close patient follow-up is warranted.

Conclusion

Human infestation with larval ticks is a common occurrence but can present a diagnostic challenge to an unfamiliar physician. We encourage consideration of larval tick infestation as the etiology of multiple or diffuse pruritic papules with a history of outdoor exposure.

Case Reports

Patient 1
A 65-year-old woman presented to the dermatology clinic in July with a pruritic rash of 2 days’ duration that started on the back and spread diffusely. The patient gardened regularly. Physical examination showed inflammatory papules and pustules on the back (Figure 1), as well as the groin, breasts, and ears. There was a punctate black dot in the center of some papules, and dermoscopy revealed ticks (Figure 2). Removal and microscopic examination confirmed larval-stage lone star ticks (Figure 3). The patient was prescribed topical steroids for pruritus as well as oral doxycycline for prophylaxis against tick-borne illnesses.

Patient 2
A 54-year-old man presented to the same clinic in July with pruritic lesions on the back, legs, ankles, and scrotum of 3 days’ duration that first appeared 24 hours after performing yardwork. Physical examination revealed diffusely distributed papules, pustules, and vesicles on the back (Figure 4). Some papules featured a punctate black dot in the center (similar to patient 1), and dermoscopy again revealed ticks. Removal and microscopic examination confirmed larval-stage ticks. The patient was treated with topical steroids and oral antihistamines for pruritus as well as prophylactic oral doxycycline.

Comment

Ticks are well-known human parasites, representing the second most common vector of human infectious disease.¹ Ticks have 3 motile stages: larva (or “seed”), nymph, and adult. They can bite humans during all stages. Larval ticks, distinguished by having 6 legs rather than 8 legs in nymphs and adults, can attack in droves and cause an infestation that presents as diffuse, pruritic, erythematous papules and pustules.^2-4 The first report of larval tick infestation in humans may have been in 1728 by William Byrd who described finding ticks on the skin that were too small to see without a microscope.⁵

Identification
The ticks in both of our cases were lone star ticks (Amblyomma americanum). The larval stage of A americanum is a proven cause of cutaneous reaction.^6,7 A PubMed search of articles indexed for MEDLINE as well as a Google Scholar search using the terms tick, seed tick, or tick bite in combination with rash, eruption, infestation, papule, pustule, or pruritic revealed 6 reported cases of larval tick infestation in the literature (including our case); 5 were caused by A americanum and 1 by Ixodes dammini (now known as Ixodes scapularis); all occurred in July or August.^3,7-10 This time frame is consistent with the general tick life cycle across species: Adults feed from April to June, then lay eggs that hatch into larval ticks within 4 to 6 weeks. After hatching, larval ticks climb grass and weeds awaiting a passing host.⁴

Diagnosis
Larval tick infestation remains a frequently misdiagnosed etiology of diffuse pruritic papules and pustules, especially in urban settings where physicians are less likely to be familiar with this type of manifestation.^3,9-11 Larval ticks are submillimeter in size and difficult to appreciate with the naked eye, contributing to misdiagnosis. A punctate black dot may sometimes be seen in papules; however, dermoscopy is critical for accurate diagnosis, as hemorrhagic crust is a frequent misdiagnosis.

Management
In addition to symptomatic therapy, both of our patients received doxycycline as antibiotic prophylaxis for tick-borne illnesses given that a high number of ticks had been attached for more than 2 days.^12,13 Antibiotic prophylaxis for tick-borne illness is controversial. The exception is Lyme disease transmitted by nymphal or adult I scapularis when specific conditions are met: the bite must have occurred in an endemic area, doxycycline cannot be contraindicated, estimated duration of attachment is at least 36 hours, and prophylaxis must be started within 72 hours of tick removal.¹³ There are no official recommendations for the A americanum species or for larval-stage ticks of any species. Larval-stage ticks acting as vectors for disease transmission is not well documented in recent literature, and there currently is limited evidence supporting prophylactic antibiotics for larval tick bites. The presence of spotted fever rickettsioses has been reported (with the exception of Rickettsia rickettsii and Ehrlichia chaffeensis) in larval A americanum ticks, suggesting a theoretical possibility that they could act as disease vectors.^3,8,11,14-17 At a minimum, both prompt tick removal and close patient follow-up is warranted.

Conclusion

Human infestation with larval ticks is a common occurrence but can present a diagnostic challenge to an unfamiliar physician. We encourage consideration of larval tick infestation as the etiology of multiple or diffuse pruritic papules with a history of outdoor exposure.

THE CASE

A 20-year-old man presented to our family medicine clinic with right wrist pain 4 days after falling on his wrist and hand while playing basketball. He denied any other previous injury or trauma. The pain was unchanged since the injury occurred.

Examination demonstrated mild edema over the palmar and ulnar aspect of the patient’s right wrist with no apparent ecchymosis. He had normal range of motion of his right wrist and hand. However, he experienced pain with active and passive wrist extension and ulnar deviation. There was significant tenderness in the palmar and ulnar aspects of his right wrist just distal to the ulnar styloid process.

THE DIAGNOSIS

Standard plain x-rays of the right wrist revealed an isolated fracture of the body of the triquetrum (FIGURE 1). Since the patient refused to have a cast placed, his wrist was immobilized with a wrist brace. By Day 16 post injury, the pain and edema had improved significantly. After talking with the patient about the potential risks and benefits of continuing to play basketball—and despite our recommendation that he not play—he decided to continue playing since he was a college basketball prospect.

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

At 4 weeks post injury, x-rays demonstrated mild interval healing (FIGURE 2). At the 8-week visit, the patient had only very mild pain and tenderness, and x-ray images showed improvement (FIGURE 3). Within a few months, his symptoms resolved completely. No further imaging was performed.

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

DISCUSSION

In general, carpal fractures are uncommon.¹ The triquetrum is the second most commonly injured carpal bone, involved in up to 18% of all carpal fractures.^2,3 Triquetrum fractures most commonly occur as isolated injuries and are typically classified in 2 general categories: avulsion fractures (dorsal cortex or volar cortex) and fractures of the triquetrum body.^4-8 Isolated avulsion fractures of the triquetral dorsal cortex are relatively common, occurring in about 95% of triquetrum injuries.^4-9 Isolated fractures of the triquetrum body are less common, occurring in about 4% of triquetrum injuries, and can go unnoticed on conventional x-rays.^4-9

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

Basketball presents a unique risk for hand or wrist fracture due to its high-impact nature, hard playing surfaces, and frequent use of the hands for dribbling, shooting, rebounding, and passing the ball.

In a retrospective study of sports-related fractures conducted at the Royal Infirmary of Edinburgh, basketball had the highest incidence of carpal injuries compared with other sports, including football, rugby, skiing, snowboarding, and ice-skating.⁴ Similarly, a retrospective study conducted at the University of California, Los Angeles, found that of all Division 1 collegiate athletes at the school, basketball players had the highest incidence of primary fractures, and the most common fracture location was the hand.¹⁰

Continue to: An injury that's easy to miss

Because the incidence of hand and wrist injuries is high among basketball players, it is imperative that triquetrum body fractures are not missed or misdiagnosed as more common hand and wrist injuries, such as triquetral dorsal avulsion fractures.

Our patient, who had an isolated triquetrum body fracture, presented with focal tenderness on the palmar and ulnar aspects of his wrist and pain with ulnar deviation. Since triquetral body fractures often have a clinical presentation quite similar to that of triquetral dorsal avulsion fractures, patients presenting with symptoms of wrist tenderness and pain should be treated with a high degree of clinical suspicion.

Triquetral fractures can be missed in up to 20% of x-rays.

With our patient, anteroposterior and lateral x-rays were sufficient to demonstrate an isolated triquetrum body fracture; however, triquetral fractures can be missed in up to 20% of x-rays.⁴ Both magnetic resonance imaging and computerized tomography are useful in diagnosing occult triquetrum fractures and should be used to confirm clinical suspicion when traditional x-rays are inconclusive.^11,12

Management varies

Management of isolated triquetrum body fractures varies depending on the fracture pattern and the status of bone consolidation. Triquetral body fractures typically heal well; it’s very rare that there is a nonunion. As our patient’s fracture was nondisplaced and stable, brace immobilization for 4 weeks was sufficient to facilitate healing and restore long-term hand and wrist functionality. This course of treatment is consistent with other cases of nondisplaced triquetrum body fractures reported in the literature.¹³

Long-term outcomes. The literature is sparse regarding the long-term functional outcome of nonsurgical treatment for nondisplaced triquetrum body fractures. Multiple carpal fractures, displaced triquetrum body fractures, and persistent pain for multiple months after nonsurgical management all indicate the need for referral to orthopedic surgery. In instances of fracture displacement or nonunion, management tends to be surgical, with open reduction and internal fixation (ORIF) used in multiple cases of nonunion for isolated triquetrum body fractures.^3,14 Any diagnostic imaging that reveals displacement, malunion, or nonunion of the fracture is an indication for referral to an orthopedic surgeon.

Continue to: Return to play

Return to play. There is no evidence-based return-to-play recommendation for patients with a triquetrum fracture. However, our patient continued to play basketball through the early stages of injury management because he was a collegiate prospect. While medical, social, and economic factors should be considered when discussing treatment options with athletes, injuries should be managed so that there is no long-term loss of function or risk of injury exacerbation. When discussing early return from injury with athletes who have outside pressure to return to play, it’s important to make them aware of the associated long- and short-term risks.¹⁵

THE TAKEAWAY

Management of an isolated triquetrum body fracture is typically straightforward; however, if the fracture is displaced, refer the patient to an orthopedic surgeon as ORIF may be required. For this reason, it’s important to be able to promptly identify isolated triquetrum body fractures and to avoid confusing them with triquetrum dorsal avulsion fractures.

Depending on the sport played and the severity of the injury, athletes with conservatively managed nondisplaced triquetral body fractures may be candidates for early return to play. Nonetheless, athletes should understand both the short- and the long-term risks of playing with an injury, and they should never be advised to continue playing with an injury if it jeopardizes their well-being or the long-term functionality of the affected body part.

CORRESPONDENCE
Morteza Khodaee, MD, MPH, University of Colorado School of Medicine, AFW Clinic, 3055 Roslyn Street, Denver, CO 80238; [email protected]

THE CASE

A 20-year-old man presented to our family medicine clinic with right wrist pain 4 days after falling on his wrist and hand while playing basketball. He denied any other previous injury or trauma. The pain was unchanged since the injury occurred.

Examination demonstrated mild edema over the palmar and ulnar aspect of the patient’s right wrist with no apparent ecchymosis. He had normal range of motion of his right wrist and hand. However, he experienced pain with active and passive wrist extension and ulnar deviation. There was significant tenderness in the palmar and ulnar aspects of his right wrist just distal to the ulnar styloid process.

THE DIAGNOSIS

Standard plain x-rays of the right wrist revealed an isolated fracture of the body of the triquetrum (FIGURE 1). Since the patient refused to have a cast placed, his wrist was immobilized with a wrist brace. By Day 16 post injury, the pain and edema had improved significantly. After talking with the patient about the potential risks and benefits of continuing to play basketball—and despite our recommendation that he not play—he decided to continue playing since he was a college basketball prospect.

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

At 4 weeks post injury, x-rays demonstrated mild interval healing (FIGURE 2). At the 8-week visit, the patient had only very mild pain and tenderness, and x-ray images showed improvement (FIGURE 3). Within a few months, his symptoms resolved completely. No further imaging was performed.

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

DISCUSSION

In general, carpal fractures are uncommon.¹ The triquetrum is the second most commonly injured carpal bone, involved in up to 18% of all carpal fractures.^2,3 Triquetrum fractures most commonly occur as isolated injuries and are typically classified in 2 general categories: avulsion fractures (dorsal cortex or volar cortex) and fractures of the triquetrum body.^4-8 Isolated avulsion fractures of the triquetral dorsal cortex are relatively common, occurring in about 95% of triquetrum injuries.^4-9 Isolated fractures of the triquetrum body are less common, occurring in about 4% of triquetrum injuries, and can go unnoticed on conventional x-rays.^4-9

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

Basketball presents a unique risk for hand or wrist fracture due to its high-impact nature, hard playing surfaces, and frequent use of the hands for dribbling, shooting, rebounding, and passing the ball.

In a retrospective study of sports-related fractures conducted at the Royal Infirmary of Edinburgh, basketball had the highest incidence of carpal injuries compared with other sports, including football, rugby, skiing, snowboarding, and ice-skating.⁴ Similarly, a retrospective study conducted at the University of California, Los Angeles, found that of all Division 1 collegiate athletes at the school, basketball players had the highest incidence of primary fractures, and the most common fracture location was the hand.¹⁰

Continue to: An injury that's easy to miss

Because the incidence of hand and wrist injuries is high among basketball players, it is imperative that triquetrum body fractures are not missed or misdiagnosed as more common hand and wrist injuries, such as triquetral dorsal avulsion fractures.

Our patient, who had an isolated triquetrum body fracture, presented with focal tenderness on the palmar and ulnar aspects of his wrist and pain with ulnar deviation. Since triquetral body fractures often have a clinical presentation quite similar to that of triquetral dorsal avulsion fractures, patients presenting with symptoms of wrist tenderness and pain should be treated with a high degree of clinical suspicion.

Triquetral fractures can be missed in up to 20% of x-rays.

With our patient, anteroposterior and lateral x-rays were sufficient to demonstrate an isolated triquetrum body fracture; however, triquetral fractures can be missed in up to 20% of x-rays.⁴ Both magnetic resonance imaging and computerized tomography are useful in diagnosing occult triquetrum fractures and should be used to confirm clinical suspicion when traditional x-rays are inconclusive.^11,12

Management varies

Management of isolated triquetrum body fractures varies depending on the fracture pattern and the status of bone consolidation. Triquetral body fractures typically heal well; it’s very rare that there is a nonunion. As our patient’s fracture was nondisplaced and stable, brace immobilization for 4 weeks was sufficient to facilitate healing and restore long-term hand and wrist functionality. This course of treatment is consistent with other cases of nondisplaced triquetrum body fractures reported in the literature.¹³

Long-term outcomes. The literature is sparse regarding the long-term functional outcome of nonsurgical treatment for nondisplaced triquetrum body fractures. Multiple carpal fractures, displaced triquetrum body fractures, and persistent pain for multiple months after nonsurgical management all indicate the need for referral to orthopedic surgery. In instances of fracture displacement or nonunion, management tends to be surgical, with open reduction and internal fixation (ORIF) used in multiple cases of nonunion for isolated triquetrum body fractures.^3,14 Any diagnostic imaging that reveals displacement, malunion, or nonunion of the fracture is an indication for referral to an orthopedic surgeon.

Continue to: Return to play

Return to play. There is no evidence-based return-to-play recommendation for patients with a triquetrum fracture. However, our patient continued to play basketball through the early stages of injury management because he was a collegiate prospect. While medical, social, and economic factors should be considered when discussing treatment options with athletes, injuries should be managed so that there is no long-term loss of function or risk of injury exacerbation. When discussing early return from injury with athletes who have outside pressure to return to play, it’s important to make them aware of the associated long- and short-term risks.¹⁵

THE TAKEAWAY

Management of an isolated triquetrum body fracture is typically straightforward; however, if the fracture is displaced, refer the patient to an orthopedic surgeon as ORIF may be required. For this reason, it’s important to be able to promptly identify isolated triquetrum body fractures and to avoid confusing them with triquetrum dorsal avulsion fractures.

Depending on the sport played and the severity of the injury, athletes with conservatively managed nondisplaced triquetral body fractures may be candidates for early return to play. Nonetheless, athletes should understand both the short- and the long-term risks of playing with an injury, and they should never be advised to continue playing with an injury if it jeopardizes their well-being or the long-term functionality of the affected body part.

CORRESPONDENCE
Morteza Khodaee, MD, MPH, University of Colorado School of Medicine, AFW Clinic, 3055 Roslyn Street, Denver, CO 80238; [email protected]

THE CASE

A 20-year-old man presented to our family medicine clinic with right wrist pain 4 days after falling on his wrist and hand while playing basketball. He denied any other previous injury or trauma. The pain was unchanged since the injury occurred.

Examination demonstrated mild edema over the palmar and ulnar aspect of the patient’s right wrist with no apparent ecchymosis. He had normal range of motion of his right wrist and hand. However, he experienced pain with active and passive wrist extension and ulnar deviation. There was significant tenderness in the palmar and ulnar aspects of his right wrist just distal to the ulnar styloid process.

THE DIAGNOSIS

Standard plain x-rays of the right wrist revealed an isolated fracture of the body of the triquetrum (FIGURE 1). Since the patient refused to have a cast placed, his wrist was immobilized with a wrist brace. By Day 16 post injury, the pain and edema had improved significantly. After talking with the patient about the potential risks and benefits of continuing to play basketball—and despite our recommendation that he not play—he decided to continue playing since he was a college basketball prospect.

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

At 4 weeks post injury, x-rays demonstrated mild interval healing (FIGURE 2). At the 8-week visit, the patient had only very mild pain and tenderness, and x-ray images showed improvement (FIGURE 3). Within a few months, his symptoms resolved completely. No further imaging was performed.

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

DISCUSSION

In general, carpal fractures are uncommon.¹ The triquetrum is the second most commonly injured carpal bone, involved in up to 18% of all carpal fractures.^2,3 Triquetrum fractures most commonly occur as isolated injuries and are typically classified in 2 general categories: avulsion fractures (dorsal cortex or volar cortex) and fractures of the triquetrum body.^4-8 Isolated avulsion fractures of the triquetral dorsal cortex are relatively common, occurring in about 95% of triquetrum injuries.^4-9 Isolated fractures of the triquetrum body are less common, occurring in about 4% of triquetrum injuries, and can go unnoticed on conventional x-rays.^4-9

IMAGES COURTESY OF UNIVERSITY OF COLORADO, DENVER

Basketball presents a unique risk for hand or wrist fracture due to its high-impact nature, hard playing surfaces, and frequent use of the hands for dribbling, shooting, rebounding, and passing the ball.

In a retrospective study of sports-related fractures conducted at the Royal Infirmary of Edinburgh, basketball had the highest incidence of carpal injuries compared with other sports, including football, rugby, skiing, snowboarding, and ice-skating.⁴ Similarly, a retrospective study conducted at the University of California, Los Angeles, found that of all Division 1 collegiate athletes at the school, basketball players had the highest incidence of primary fractures, and the most common fracture location was the hand.¹⁰

Continue to: An injury that's easy to miss

Because the incidence of hand and wrist injuries is high among basketball players, it is imperative that triquetrum body fractures are not missed or misdiagnosed as more common hand and wrist injuries, such as triquetral dorsal avulsion fractures.

Our patient, who had an isolated triquetrum body fracture, presented with focal tenderness on the palmar and ulnar aspects of his wrist and pain with ulnar deviation. Since triquetral body fractures often have a clinical presentation quite similar to that of triquetral dorsal avulsion fractures, patients presenting with symptoms of wrist tenderness and pain should be treated with a high degree of clinical suspicion.

Triquetral fractures can be missed in up to 20% of x-rays.

With our patient, anteroposterior and lateral x-rays were sufficient to demonstrate an isolated triquetrum body fracture; however, triquetral fractures can be missed in up to 20% of x-rays.⁴ Both magnetic resonance imaging and computerized tomography are useful in diagnosing occult triquetrum fractures and should be used to confirm clinical suspicion when traditional x-rays are inconclusive.^11,12

Management varies

Management of isolated triquetrum body fractures varies depending on the fracture pattern and the status of bone consolidation. Triquetral body fractures typically heal well; it’s very rare that there is a nonunion. As our patient’s fracture was nondisplaced and stable, brace immobilization for 4 weeks was sufficient to facilitate healing and restore long-term hand and wrist functionality. This course of treatment is consistent with other cases of nondisplaced triquetrum body fractures reported in the literature.¹³

Long-term outcomes. The literature is sparse regarding the long-term functional outcome of nonsurgical treatment for nondisplaced triquetrum body fractures. Multiple carpal fractures, displaced triquetrum body fractures, and persistent pain for multiple months after nonsurgical management all indicate the need for referral to orthopedic surgery. In instances of fracture displacement or nonunion, management tends to be surgical, with open reduction and internal fixation (ORIF) used in multiple cases of nonunion for isolated triquetrum body fractures.^3,14 Any diagnostic imaging that reveals displacement, malunion, or nonunion of the fracture is an indication for referral to an orthopedic surgeon.

Continue to: Return to play

Return to play. There is no evidence-based return-to-play recommendation for patients with a triquetrum fracture. However, our patient continued to play basketball through the early stages of injury management because he was a collegiate prospect. While medical, social, and economic factors should be considered when discussing treatment options with athletes, injuries should be managed so that there is no long-term loss of function or risk of injury exacerbation. When discussing early return from injury with athletes who have outside pressure to return to play, it’s important to make them aware of the associated long- and short-term risks.¹⁵

THE TAKEAWAY

Management of an isolated triquetrum body fracture is typically straightforward; however, if the fracture is displaced, refer the patient to an orthopedic surgeon as ORIF may be required. For this reason, it’s important to be able to promptly identify isolated triquetrum body fractures and to avoid confusing them with triquetrum dorsal avulsion fractures.

Depending on the sport played and the severity of the injury, athletes with conservatively managed nondisplaced triquetral body fractures may be candidates for early return to play. Nonetheless, athletes should understand both the short- and the long-term risks of playing with an injury, and they should never be advised to continue playing with an injury if it jeopardizes their well-being or the long-term functionality of the affected body part.

CORRESPONDENCE
Morteza Khodaee, MD, MPH, University of Colorado School of Medicine, AFW Clinic, 3055 Roslyn Street, Denver, CO 80238; [email protected]

There are 4 variants of psoriasis: plaque, guttate, pustular, and erythroderma (in order of prevalence).² Guttate psoriasis is characterized by small, 2- to 10-mm, raindroplike lesions on the skin.¹ It accounts for approximately 2% of total psoriasis cases and is commonly triggered by group A streptococcal pharyngitis or tonsillitis.^3,4

Hand-foot-and-mouth disease (HFMD) is an illness most commonly caused by a coxsackievirus A infection but also can be caused by other enteroviruses.^5,6 Coxsackievirus is a serotype of the Enterovirus species within the Picornaviridae family.⁷ Hand-foot-and-mouth disease is characterized by a brief fever and vesicular rashes on the palms, soles, or buttocks, as well as oropharyngeal ulcers.⁸ Typically, the rash is benign and short-lived.⁹ In rare cases, neurologic complications develop. There have been no reported cases of guttate psoriasis following a coxsackievirus A infection.

The involvement of coxsackievirus B in the etiopathogenesis of psoriasis has been previously reported.¹⁰ We report the case of guttate psoriasis following presumed coxsackievirus A HFMD.

Case Report

A 56-year-old woman presented with a vesicular rash on the hands, feet, and lips. The patient reported having a sore throat that started around the same time that the rash developed. The severity of the sore throat was rated as moderate. No fever was reported. One day prior, the patient’s primary care physician prescribed a tapered course of prednisone for the rash. The patient reported a medical history of herpes zoster virus, sunburn, and genital herpes. She was taking clonazepam and had a known allergy to penicillin.

Physical examination revealed erythematous vesicular and papular lesions on the extensor surfaces of the hands and feet. Vesicles also were noted on the vermilion border of the lip. Examination of the patient’s mouth showed blisters and shallow ulcerations in the oral cavity. A clinical diagnosis of coxsackievirus A HFMD was made, and the treatment plan included triamcinolone acetonide ointment 0.025% applied twice daily for 2 weeks and oral valacyclovir hydrochloride 1 g taken 3 times daily for 7 days. A topical emollient also was recommended for the lips when necessary. The lesions all resolved within a 2-week period with no sequela.

The patient returned 1 month later, citing newer red abdominal skin lesions. Fever was denied. She reported that both prescribed treatments had not been helping for the newer lesions. She noticed similar lesions on the groin and brought them to the attention of her gynecologist. Physical examination revealed salmon pink papules and plaques with silvery scaling involving the abdomen, bilateral upper extremities and ears, and scalp. The patient was then clinically diagnosed with guttate psoriasis. A shave biopsy of a representative lesion on the abdomen was performed. The treatment plan included betamethasone dipropionate cream 0.05% applied twice daily for 2 weeks, clobetasol propionate solution 0.05% applied twice daily for 14 days (for the scalp), and hydrocortisone valerate cream 0.2% applied twice daily for 14 days (for the groin).

The skin biopsy shown in the Figure was received in 10% buffered formalin, measuring 5×4×1 mm of skin. Sections showed an acanthotic epidermis with foci of spongiosis and hypergranulosis covered by mounds of parakeratosis infiltrated by neutrophils. Superficial perivascular and interstitial lymphocytic inflammation was present. Tortuous blood vessels within the papillary dermis also were present. Results showed psoriasiform dermatitis with mild spongiosis. Periodic acid–Schiff stain did not reveal any fungal organisms. These findings were consistent with a diagnosis of guttate psoriasis.

Comment

Enterovirus-derived HFMD likely is caused by coxsackie-virus A. Current evidence supports the theory that guttate psoriasis can be environmentally triggered in genetically susceptible individuals, often but not exclusively by a streptococcal infection. The causative agent elicits a T-cell–mediated reaction leading to increased type 1 helper T cells, IFN-γ, and IL-2 cytokine levels. HLA-Cw^∗0602–positive patients are considered genetically susceptible and more likely to develop guttate psoriasis following an environmental trigger. Based on the coincidence in timing of both diagnoses, this reported case of guttate psoriasis may have been triggered by a coxsackievirus A infection.

There are 4 variants of psoriasis: plaque, guttate, pustular, and erythroderma (in order of prevalence).² Guttate psoriasis is characterized by small, 2- to 10-mm, raindroplike lesions on the skin.¹ It accounts for approximately 2% of total psoriasis cases and is commonly triggered by group A streptococcal pharyngitis or tonsillitis.^3,4

Hand-foot-and-mouth disease (HFMD) is an illness most commonly caused by a coxsackievirus A infection but also can be caused by other enteroviruses.^5,6 Coxsackievirus is a serotype of the Enterovirus species within the Picornaviridae family.⁷ Hand-foot-and-mouth disease is characterized by a brief fever and vesicular rashes on the palms, soles, or buttocks, as well as oropharyngeal ulcers.⁸ Typically, the rash is benign and short-lived.⁹ In rare cases, neurologic complications develop. There have been no reported cases of guttate psoriasis following a coxsackievirus A infection.

The involvement of coxsackievirus B in the etiopathogenesis of psoriasis has been previously reported.¹⁰ We report the case of guttate psoriasis following presumed coxsackievirus A HFMD.

Case Report

A 56-year-old woman presented with a vesicular rash on the hands, feet, and lips. The patient reported having a sore throat that started around the same time that the rash developed. The severity of the sore throat was rated as moderate. No fever was reported. One day prior, the patient’s primary care physician prescribed a tapered course of prednisone for the rash. The patient reported a medical history of herpes zoster virus, sunburn, and genital herpes. She was taking clonazepam and had a known allergy to penicillin.

Physical examination revealed erythematous vesicular and papular lesions on the extensor surfaces of the hands and feet. Vesicles also were noted on the vermilion border of the lip. Examination of the patient’s mouth showed blisters and shallow ulcerations in the oral cavity. A clinical diagnosis of coxsackievirus A HFMD was made, and the treatment plan included triamcinolone acetonide ointment 0.025% applied twice daily for 2 weeks and oral valacyclovir hydrochloride 1 g taken 3 times daily for 7 days. A topical emollient also was recommended for the lips when necessary. The lesions all resolved within a 2-week period with no sequela.

The patient returned 1 month later, citing newer red abdominal skin lesions. Fever was denied. She reported that both prescribed treatments had not been helping for the newer lesions. She noticed similar lesions on the groin and brought them to the attention of her gynecologist. Physical examination revealed salmon pink papules and plaques with silvery scaling involving the abdomen, bilateral upper extremities and ears, and scalp. The patient was then clinically diagnosed with guttate psoriasis. A shave biopsy of a representative lesion on the abdomen was performed. The treatment plan included betamethasone dipropionate cream 0.05% applied twice daily for 2 weeks, clobetasol propionate solution 0.05% applied twice daily for 14 days (for the scalp), and hydrocortisone valerate cream 0.2% applied twice daily for 14 days (for the groin).

The skin biopsy shown in the Figure was received in 10% buffered formalin, measuring 5×4×1 mm of skin. Sections showed an acanthotic epidermis with foci of spongiosis and hypergranulosis covered by mounds of parakeratosis infiltrated by neutrophils. Superficial perivascular and interstitial lymphocytic inflammation was present. Tortuous blood vessels within the papillary dermis also were present. Results showed psoriasiform dermatitis with mild spongiosis. Periodic acid–Schiff stain did not reveal any fungal organisms. These findings were consistent with a diagnosis of guttate psoriasis.

Comment

Enterovirus-derived HFMD likely is caused by coxsackie-virus A. Current evidence supports the theory that guttate psoriasis can be environmentally triggered in genetically susceptible individuals, often but not exclusively by a streptococcal infection. The causative agent elicits a T-cell–mediated reaction leading to increased type 1 helper T cells, IFN-γ, and IL-2 cytokine levels. HLA-Cw^∗0602–positive patients are considered genetically susceptible and more likely to develop guttate psoriasis following an environmental trigger. Based on the coincidence in timing of both diagnoses, this reported case of guttate psoriasis may have been triggered by a coxsackievirus A infection.

There are 4 variants of psoriasis: plaque, guttate, pustular, and erythroderma (in order of prevalence).² Guttate psoriasis is characterized by small, 2- to 10-mm, raindroplike lesions on the skin.¹ It accounts for approximately 2% of total psoriasis cases and is commonly triggered by group A streptococcal pharyngitis or tonsillitis.^3,4

Hand-foot-and-mouth disease (HFMD) is an illness most commonly caused by a coxsackievirus A infection but also can be caused by other enteroviruses.^5,6 Coxsackievirus is a serotype of the Enterovirus species within the Picornaviridae family.⁷ Hand-foot-and-mouth disease is characterized by a brief fever and vesicular rashes on the palms, soles, or buttocks, as well as oropharyngeal ulcers.⁸ Typically, the rash is benign and short-lived.⁹ In rare cases, neurologic complications develop. There have been no reported cases of guttate psoriasis following a coxsackievirus A infection.

The involvement of coxsackievirus B in the etiopathogenesis of psoriasis has been previously reported.¹⁰ We report the case of guttate psoriasis following presumed coxsackievirus A HFMD.

Case Report

A 56-year-old woman presented with a vesicular rash on the hands, feet, and lips. The patient reported having a sore throat that started around the same time that the rash developed. The severity of the sore throat was rated as moderate. No fever was reported. One day prior, the patient’s primary care physician prescribed a tapered course of prednisone for the rash. The patient reported a medical history of herpes zoster virus, sunburn, and genital herpes. She was taking clonazepam and had a known allergy to penicillin.

Physical examination revealed erythematous vesicular and papular lesions on the extensor surfaces of the hands and feet. Vesicles also were noted on the vermilion border of the lip. Examination of the patient’s mouth showed blisters and shallow ulcerations in the oral cavity. A clinical diagnosis of coxsackievirus A HFMD was made, and the treatment plan included triamcinolone acetonide ointment 0.025% applied twice daily for 2 weeks and oral valacyclovir hydrochloride 1 g taken 3 times daily for 7 days. A topical emollient also was recommended for the lips when necessary. The lesions all resolved within a 2-week period with no sequela.

The patient returned 1 month later, citing newer red abdominal skin lesions. Fever was denied. She reported that both prescribed treatments had not been helping for the newer lesions. She noticed similar lesions on the groin and brought them to the attention of her gynecologist. Physical examination revealed salmon pink papules and plaques with silvery scaling involving the abdomen, bilateral upper extremities and ears, and scalp. The patient was then clinically diagnosed with guttate psoriasis. A shave biopsy of a representative lesion on the abdomen was performed. The treatment plan included betamethasone dipropionate cream 0.05% applied twice daily for 2 weeks, clobetasol propionate solution 0.05% applied twice daily for 14 days (for the scalp), and hydrocortisone valerate cream 0.2% applied twice daily for 14 days (for the groin).

The skin biopsy shown in the Figure was received in 10% buffered formalin, measuring 5×4×1 mm of skin. Sections showed an acanthotic epidermis with foci of spongiosis and hypergranulosis covered by mounds of parakeratosis infiltrated by neutrophils. Superficial perivascular and interstitial lymphocytic inflammation was present. Tortuous blood vessels within the papillary dermis also were present. Results showed psoriasiform dermatitis with mild spongiosis. Periodic acid–Schiff stain did not reveal any fungal organisms. These findings were consistent with a diagnosis of guttate psoriasis.

Comment

Enterovirus-derived HFMD likely is caused by coxsackie-virus A. Current evidence supports the theory that guttate psoriasis can be environmentally triggered in genetically susceptible individuals, often but not exclusively by a streptococcal infection. The causative agent elicits a T-cell–mediated reaction leading to increased type 1 helper T cells, IFN-γ, and IL-2 cytokine levels. HLA-Cw^∗0602–positive patients are considered genetically susceptible and more likely to develop guttate psoriasis following an environmental trigger. Based on the coincidence in timing of both diagnoses, this reported case of guttate psoriasis may have been triggered by a coxsackievirus A infection.

Photolichenoid dermatitis is an uncommon eruptive dermatitis of variable clinical presentation. It has a histopathologic pattern of lichenoid inflammation and is best characterized as a photoallergic reaction.¹ Photolichenoid dermatitis was first described in 1954 in association with the use of quinidine in the treatment of malaria.² Subsequently, it has been associated with various medications, including trimethoprim-sulfamethoxazole, azithromycin, and nonsteroidal anti-inflammatory drugs.^1,2 Photolichenoid dermatitis has been documented in patients with human immunodeficiency virus (HIV) with variable clinical presentations. Photolichenoid dermatitis in patients with HIV has been described both with and without an associated photosensitizing systemic agent, suggesting that HIV infection is an independent risk factor for the development of this eruption in patients with HIV.^3-6

Case Report

A 62-year-old African man presented for evaluation of asymptomatic hypopigmented and depigmented patches in a photodistributed pattern. The eruption began the preceding summer when he noted a pink patch on the right side of the forehead. It progressed over 2 months to involve the face, ears, neck, and arms. His medical history was negative. The only medication he was taking was hydroxychloroquine, which was prescribed by another dermatologist when the patient first developed the eruption. The patient was unsure of the indication for the medication and admitted to poor compliance. A review of systems was negative. There was no personal or family history of autoimmune disease. A detailed sexual history and illicit drug history were not obtained. Physical examination revealed hypopigmented and depigmented patches, some with overlying erythema and collarettes of fine scale. The patches were photodistributed on the face, conchal bowls, neck, dorsal aspect of the hands, and extensor forearms (Figures 1 and 2). Macules of repigmentation were noted within some of the patches. There also were large hyperpigmented patches with peripheral hypopigmentation on the legs.

A punch biopsy taken from the left posterior neck revealed a patchy bandlike lymphocytic infiltrate in the superficial dermis with lymphocytes present at the dermoepidermal junction and scattered dyskeratotic keratinocytes extending into the mid spinous layer (Figure 3). Histopathologic findings were consistent with photolichenoid dermatitis.

Comment

Prevalence of Photosensitive Eruptions
Photodermatitis is an uncommon clinical manifestation of HIV occurring in approximately 5% of patients who are HIV positive.³ Photosensitive eruptions previously described in association with HIV include porphyria cutanea tarda, pseudoporphyria, chronic actinic dermatitis, granuloma annulare, photodistributed dyspigmentation, and lichenoid photodermatitis.⁷ These HIV-associated photosensitive eruptions have been found to disproportionally affect patients of African and Native American descent.^5,7,8 Therefore, a new photodistributed eruption in a patient of African or Native American descent should prompt evaluation of possible underlying HIV infection.

Presenting Sign of HIV Infection
We report a case of photolichenoid dermatitis presenting with loss of pigmentation as a presenting sign of HIV. The patient had no known history of HIV or prior opportunistic infections and was not taking any medications at the time of onset or presentation to clinic. Similar cases of photodistributed depigmentation with lichenoid inflammation on histopathology occurring in patients with HIV have been previously described.^4-6,9 In these cases, most patients were of African descent with previously diagnosed advanced HIV and CD4 counts of less than 50 cells/mL³. The additional clinical findings of lichenoid papules and plaques were noted in several of these cases.^5,6

Exposure to Photosensitizing Drugs
Photodermatitis in patients with HIV often is attributed to exposure to a photosensitizing drug. Many reported cases are retrospective and identify a temporal association between the onset of photodermatitis following the initiation of a photosensitizing drug. The most commonly implicated drugs have included nonsteroidal anti-inflammatory drugs, trimethoprim-sulfamethoxazole, and azithromycin. Other potential offenders may include saquinavir, dapsone, ketoconazole, and efavirenz.^3,5 In cases in which temporal association with a new medication could not be identified, the photodermatitis often has been presumed to be due to polypharmacy and the potential synergistic effect of multiple photosensitizing drugs.^3,5-8

Advanced HIV
There are several reported cases of photodermatitis occurring in patients who were not exposed to systemic photosensitizers. These patients had advanced HIV, meeting criteria for AIDS with a CD4 count of less than 200 cells/mL³. The majority of patients had an even lower CD4 count of less than 50 cells/mL³. Clinical presentations have included photodistributed lichenoid papules and plaques as well as depigmented patches.^4,5,8,10

Evaluating HIV as a Risk Factor for Photodermatitis
Discerning the validity of the correlation between photodermatitis and HIV is difficult, as all previously reported cases are case reports and small retrospective case series. One study of 34 patients with HIV and photodermatitis showed that there was no significant increase in incidence of photodermatitis in patients who were exposed to a photosensitizing drug vs those who were not,³ which further validates that HIV infection may be an independent risk factor in the development of photodermatitis.

Conclusion

This case represents an uncommon presentation of photolichenoid dermatitis as the presenting sign of HIV infection.¹⁰ Although most reported cases of photodermatitis in HIV are attributed to photosensitizing drugs, we propose that HIV may be an independent risk factor for the development of photodermatitis. We recommend consideration of HIV testing in patients who present with photodistributed depigmenting eruptions, even in the absence of a photosensitizing drug, particularly in patients of African and Native American descent.

Photolichenoid dermatitis is an uncommon eruptive dermatitis of variable clinical presentation. It has a histopathologic pattern of lichenoid inflammation and is best characterized as a photoallergic reaction.¹ Photolichenoid dermatitis was first described in 1954 in association with the use of quinidine in the treatment of malaria.² Subsequently, it has been associated with various medications, including trimethoprim-sulfamethoxazole, azithromycin, and nonsteroidal anti-inflammatory drugs.^1,2 Photolichenoid dermatitis has been documented in patients with human immunodeficiency virus (HIV) with variable clinical presentations. Photolichenoid dermatitis in patients with HIV has been described both with and without an associated photosensitizing systemic agent, suggesting that HIV infection is an independent risk factor for the development of this eruption in patients with HIV.^3-6

Case Report

A 62-year-old African man presented for evaluation of asymptomatic hypopigmented and depigmented patches in a photodistributed pattern. The eruption began the preceding summer when he noted a pink patch on the right side of the forehead. It progressed over 2 months to involve the face, ears, neck, and arms. His medical history was negative. The only medication he was taking was hydroxychloroquine, which was prescribed by another dermatologist when the patient first developed the eruption. The patient was unsure of the indication for the medication and admitted to poor compliance. A review of systems was negative. There was no personal or family history of autoimmune disease. A detailed sexual history and illicit drug history were not obtained. Physical examination revealed hypopigmented and depigmented patches, some with overlying erythema and collarettes of fine scale. The patches were photodistributed on the face, conchal bowls, neck, dorsal aspect of the hands, and extensor forearms (Figures 1 and 2). Macules of repigmentation were noted within some of the patches. There also were large hyperpigmented patches with peripheral hypopigmentation on the legs.

A punch biopsy taken from the left posterior neck revealed a patchy bandlike lymphocytic infiltrate in the superficial dermis with lymphocytes present at the dermoepidermal junction and scattered dyskeratotic keratinocytes extending into the mid spinous layer (Figure 3). Histopathologic findings were consistent with photolichenoid dermatitis.

Comment

Prevalence of Photosensitive Eruptions
Photodermatitis is an uncommon clinical manifestation of HIV occurring in approximately 5% of patients who are HIV positive.³ Photosensitive eruptions previously described in association with HIV include porphyria cutanea tarda, pseudoporphyria, chronic actinic dermatitis, granuloma annulare, photodistributed dyspigmentation, and lichenoid photodermatitis.⁷ These HIV-associated photosensitive eruptions have been found to disproportionally affect patients of African and Native American descent.^5,7,8 Therefore, a new photodistributed eruption in a patient of African or Native American descent should prompt evaluation of possible underlying HIV infection.

Presenting Sign of HIV Infection
We report a case of photolichenoid dermatitis presenting with loss of pigmentation as a presenting sign of HIV. The patient had no known history of HIV or prior opportunistic infections and was not taking any medications at the time of onset or presentation to clinic. Similar cases of photodistributed depigmentation with lichenoid inflammation on histopathology occurring in patients with HIV have been previously described.^4-6,9 In these cases, most patients were of African descent with previously diagnosed advanced HIV and CD4 counts of less than 50 cells/mL³. The additional clinical findings of lichenoid papules and plaques were noted in several of these cases.^5,6

Exposure to Photosensitizing Drugs
Photodermatitis in patients with HIV often is attributed to exposure to a photosensitizing drug. Many reported cases are retrospective and identify a temporal association between the onset of photodermatitis following the initiation of a photosensitizing drug. The most commonly implicated drugs have included nonsteroidal anti-inflammatory drugs, trimethoprim-sulfamethoxazole, and azithromycin. Other potential offenders may include saquinavir, dapsone, ketoconazole, and efavirenz.^3,5 In cases in which temporal association with a new medication could not be identified, the photodermatitis often has been presumed to be due to polypharmacy and the potential synergistic effect of multiple photosensitizing drugs.^3,5-8

Advanced HIV
There are several reported cases of photodermatitis occurring in patients who were not exposed to systemic photosensitizers. These patients had advanced HIV, meeting criteria for AIDS with a CD4 count of less than 200 cells/mL³. The majority of patients had an even lower CD4 count of less than 50 cells/mL³. Clinical presentations have included photodistributed lichenoid papules and plaques as well as depigmented patches.^4,5,8,10

Evaluating HIV as a Risk Factor for Photodermatitis
Discerning the validity of the correlation between photodermatitis and HIV is difficult, as all previously reported cases are case reports and small retrospective case series. One study of 34 patients with HIV and photodermatitis showed that there was no significant increase in incidence of photodermatitis in patients who were exposed to a photosensitizing drug vs those who were not,³ which further validates that HIV infection may be an independent risk factor in the development of photodermatitis.

Conclusion

This case represents an uncommon presentation of photolichenoid dermatitis as the presenting sign of HIV infection.¹⁰ Although most reported cases of photodermatitis in HIV are attributed to photosensitizing drugs, we propose that HIV may be an independent risk factor for the development of photodermatitis. We recommend consideration of HIV testing in patients who present with photodistributed depigmenting eruptions, even in the absence of a photosensitizing drug, particularly in patients of African and Native American descent.

Photolichenoid dermatitis is an uncommon eruptive dermatitis of variable clinical presentation. It has a histopathologic pattern of lichenoid inflammation and is best characterized as a photoallergic reaction.¹ Photolichenoid dermatitis was first described in 1954 in association with the use of quinidine in the treatment of malaria.² Subsequently, it has been associated with various medications, including trimethoprim-sulfamethoxazole, azithromycin, and nonsteroidal anti-inflammatory drugs.^1,2 Photolichenoid dermatitis has been documented in patients with human immunodeficiency virus (HIV) with variable clinical presentations. Photolichenoid dermatitis in patients with HIV has been described both with and without an associated photosensitizing systemic agent, suggesting that HIV infection is an independent risk factor for the development of this eruption in patients with HIV.^3-6

Case Report

A 62-year-old African man presented for evaluation of asymptomatic hypopigmented and depigmented patches in a photodistributed pattern. The eruption began the preceding summer when he noted a pink patch on the right side of the forehead. It progressed over 2 months to involve the face, ears, neck, and arms. His medical history was negative. The only medication he was taking was hydroxychloroquine, which was prescribed by another dermatologist when the patient first developed the eruption. The patient was unsure of the indication for the medication and admitted to poor compliance. A review of systems was negative. There was no personal or family history of autoimmune disease. A detailed sexual history and illicit drug history were not obtained. Physical examination revealed hypopigmented and depigmented patches, some with overlying erythema and collarettes of fine scale. The patches were photodistributed on the face, conchal bowls, neck, dorsal aspect of the hands, and extensor forearms (Figures 1 and 2). Macules of repigmentation were noted within some of the patches. There also were large hyperpigmented patches with peripheral hypopigmentation on the legs.

A punch biopsy taken from the left posterior neck revealed a patchy bandlike lymphocytic infiltrate in the superficial dermis with lymphocytes present at the dermoepidermal junction and scattered dyskeratotic keratinocytes extending into the mid spinous layer (Figure 3). Histopathologic findings were consistent with photolichenoid dermatitis.

Comment

Prevalence of Photosensitive Eruptions
Photodermatitis is an uncommon clinical manifestation of HIV occurring in approximately 5% of patients who are HIV positive.³ Photosensitive eruptions previously described in association with HIV include porphyria cutanea tarda, pseudoporphyria, chronic actinic dermatitis, granuloma annulare, photodistributed dyspigmentation, and lichenoid photodermatitis.⁷ These HIV-associated photosensitive eruptions have been found to disproportionally affect patients of African and Native American descent.^5,7,8 Therefore, a new photodistributed eruption in a patient of African or Native American descent should prompt evaluation of possible underlying HIV infection.

Presenting Sign of HIV Infection
We report a case of photolichenoid dermatitis presenting with loss of pigmentation as a presenting sign of HIV. The patient had no known history of HIV or prior opportunistic infections and was not taking any medications at the time of onset or presentation to clinic. Similar cases of photodistributed depigmentation with lichenoid inflammation on histopathology occurring in patients with HIV have been previously described.^4-6,9 In these cases, most patients were of African descent with previously diagnosed advanced HIV and CD4 counts of less than 50 cells/mL³. The additional clinical findings of lichenoid papules and plaques were noted in several of these cases.^5,6

Exposure to Photosensitizing Drugs
Photodermatitis in patients with HIV often is attributed to exposure to a photosensitizing drug. Many reported cases are retrospective and identify a temporal association between the onset of photodermatitis following the initiation of a photosensitizing drug. The most commonly implicated drugs have included nonsteroidal anti-inflammatory drugs, trimethoprim-sulfamethoxazole, and azithromycin. Other potential offenders may include saquinavir, dapsone, ketoconazole, and efavirenz.^3,5 In cases in which temporal association with a new medication could not be identified, the photodermatitis often has been presumed to be due to polypharmacy and the potential synergistic effect of multiple photosensitizing drugs.^3,5-8

Advanced HIV
There are several reported cases of photodermatitis occurring in patients who were not exposed to systemic photosensitizers. These patients had advanced HIV, meeting criteria for AIDS with a CD4 count of less than 200 cells/mL³. The majority of patients had an even lower CD4 count of less than 50 cells/mL³. Clinical presentations have included photodistributed lichenoid papules and plaques as well as depigmented patches.^4,5,8,10

Evaluating HIV as a Risk Factor for Photodermatitis
Discerning the validity of the correlation between photodermatitis and HIV is difficult, as all previously reported cases are case reports and small retrospective case series. One study of 34 patients with HIV and photodermatitis showed that there was no significant increase in incidence of photodermatitis in patients who were exposed to a photosensitizing drug vs those who were not,³ which further validates that HIV infection may be an independent risk factor in the development of photodermatitis.

Conclusion

This case represents an uncommon presentation of photolichenoid dermatitis as the presenting sign of HIV infection.¹⁰ Although most reported cases of photodermatitis in HIV are attributed to photosensitizing drugs, we propose that HIV may be an independent risk factor for the development of photodermatitis. We recommend consideration of HIV testing in patients who present with photodistributed depigmenting eruptions, even in the absence of a photosensitizing drug, particularly in patients of African and Native American descent.

Infection with Mycobacterium haemophilum, a rare, slow-growing organism, most commonly presents as ulcerating cutaneous lesions and subcutaneous nodules in immunocompromised adults.¹ The most common clinical presentation in adults includes cutaneous lesions, nodules, cysts, and papules, with signs and symptoms of erythema, pain, pruritus, and drainage.² Disseminated disease states of septic arthritis, pulmonary infiltration, and osteomyelitis, though life-threatening, are less common manifestations reported in highly immunocompromised persons.³

Infection with M haemophilum presents a challenge to the dermatology community because it is infrequently suspected and misidentified, resulting in delayed diagnosis. Additionally, M haemophilum is an extremely fastidious organism that requires heme-supplemented culture media and a carefully regulated low temperature for many consecutive weeks to yield valid culture results.¹ These features contribute to complications and delays in diagnosis of an already overlooked source of infection.

We discuss the clinical presentation, diagnosis, and treatment of 3 unusual cases of cutaneous M haemophilum infection involving the upper arms. The findings in these cases highlight the challenges inherent in diagnosis as well as the obstacles that arise in providing effective, long-term treatment of this infection.

Case Reports

Patient 1
A 69-year-old woman with a medical history of a single functioning kidney and moderate psoriasis managed with low-dosage methotrexate presented with an erythematous nonhealing wound on the left forearm that developed after she was scratched by a dog. The pustules, appearing as bright red, tender, warm abscesses, had been present for 3 months and were distributed on the left proximal and distal dorsal forearm (Figure 1A). The patient reported no recent travel, sick contacts, allergies, or new medications.

A shave biopsy was initially obtained. Swab specimens were sent for bacterial, fungal, and mycobacterial culture following discontinuation of methotrexate. Initial histopathologic analysis revealed aggregates of histiocytes and multinucleated giant cells within the dermis, surrounded by infiltrates of lymphocytes and neutrophils (Figure 2), consistent with a dermal noncaseating granulomatosis. Acid-fast bacilli (AFB), periodic acid–Schiff, Gram, and Grocott-Gomori methenamine-silver stains were negative for pathogenic microorganisms. There was no evidence of vasculitis.

Despite negative special stains, an infectious cause was still suspected. Oral doxycycline monohydrate 100 mg twice daily, oral fluconazole 200 mg daily, and econazole cream 1% were prescribed because of concern for mycobacterial infection and initial growth of Candida parapsilosis in the swab culture.

A punch biopsy also was performed at this time for both repeat histopathologic analysis and tissue culture. Follow-up appointments were scheduled every 2 weeks. Staining by AFB of the repeat histopathologic specimen was negative.

The patient demonstrated symptomatic and aesthetic improvement (Figure 1B) during consecutive regular follow-up appointments while culture results were pending. No lesions appeared above the left elbow and she had no lymphadenopathy. Results of blood chemistry analyses and complete blood cell count throughout follow-up were normal.

The final tissue culture report obtained 7 weeks after initial presentation showed growth of M haemophilum despite a negative smear. The swab culture that initially was taken did not grow pathogenic organisms.

The patient was referred to an infectious disease specialist who confirmed that the atypical mycobacterial infection likely was the main source of the cutaneous lesions. She was instructed to continue econazole cream 1% and was given prescriptions for clarithromycin 500 mg twice daily, ciprofloxacin 500 mg twice daily, and rifampin 300 mg twice daily for a total duration of 12 to 18 months. The patient has remained on this triple-drug regimen and demonstrated improvement in the lesions. She has been off methotrexate while on antibiotic therapy.

Patient 2
A 79-year-old man with a medical history of chronic lymphocytic leukemia, basal cell carcinoma, and squamous cell carcinoma presented with a nonhealing, painful, red lesion on the left forearm of 1 week’s duration. Physical examination revealed a violaceous nontender plaque with erosions and desquamation that was initially diagnosed as a carbuncle. The patient reported a similar eruption on the right foot that was successfully treated with silver sulfadiazine by another physician.

Biopsy was performed by the shave method for histologic analysis and tissue culture. Doxycycline 100 mg twice daily was prescribed because of high suspicion of infection. Histologic findings revealed granulomatous inflammation with pseudoepitheliomatous hyperplasia, reported as squamous cell carcinoma. A second opinion confirmed suspicion of an infectious process; the patient remained on doxycycline. During follow-up, the lesion progressed to a 5-cm plaque studded with pustules and satellite papules. Multiple additional tissue cultures were performed over 2 months until “light growth” of M haemophilum was reported.

The patient showed minimal improvement on tetracycline antibiotics. His condition was complicated by a photosensitivity reaction to doxycycline on the left and right forearms, hands, and nose. Consequently, triamcinolone was prescribed, doxycycline was discontinued, and minocycline 100 mg twice daily and ciprofloxacin 500 mg twice daily were prescribed.

Nine months after initial presentation, the lesions were still present but remarkably improved. The antibiotic regimen was discontinued after 11 months.

Patient 3
A 77-year-old woman with a history of rheumatoid arthritis treated with methotrexate and abatacept as well as cutaneous T-cell lymphoma treated with narrowband UVB radiation presented to the emergency department with fever and an inflamed right forearm (Figure 3A). Initial bacterial cultures of the wound and blood were negative.

Arm edema and pustules slowly resolved, but the eschar and verrucous plaques continued to slowly progress while the patient was off immunosuppression. She was kept off antibiotics until mycobacterial culture was positive at 4 weeks, at which time she was placed on doxycycline and clarithromycin. Final identification of M haemophilum was made at 6 weeks; consequently, doxycycline was discontinued and she was referred to infectious disease for multidrug therapy. She remained afebrile during the entire 6 weeks until cultures were final.

While immunosuppressants were discontinued and clarithromycin was administered, the plaque changed from an edematous pustular dermatitis to a verrucous crusted plaque. Neither epitrochlear nor axillary lymphadenopathy was noted during the treatment period. The infectious disease specialist prescribed azithromycin, ethambutol, and rifampin, which produced marked improvement (Figure 3B). The patient has remained off immunosuppressive therapy while on antibiotics.

Comment

Clinical Presentation and Diagnosis
Mycobacterium haemophilum is a rare infectious organism that affects primarily immunocompromised adults but also has been identified in immunocompetent adults and pediatric patients.² Commonly affected immunosuppressed groups include solid organ transplant recipients, bone marrow transplant recipients, human immunodeficiency virus–positive patients, and patients with rheumatoid arthritis.

The infection typically presents as small violaceous papules and pustules that become painful and erythematous, with progression and draining ulceration in later stages.² In our cases, all lesions tended to evolve into a verrucous plaque that slowly resolved with antibiotic therapy.

Due to the rarity of this infection, the initial differential diagnosis can include infection with other mycobacteria, Sporothrix, Staphylococcus aureus, and other fungal pathogens. Misdiagnosis is a common obstacle in the treatment of M haemophilum due to its rarity, often negative AFB stains, and slow growth on culture media; therefore, tissue culture is essential to successful diagnosis and management. The natural reservoir of M haemophilum is unknown, but infection has been associated with contaminated water sources.¹ In one case (patient 1), symptoms developed after a dog scratch; the other 2 patients were unaware of injury to the skin.Laboratory diagnosis of M haemophilum is inherently difficult and protracted. The species is a highly fastidious and slow-growing Mycobacterium that requires cooler (30°C) incubation for many weeks on agar medium enriched with hemin or ferric ammonium citrate to obtain valid growth.¹ To secure timely diagnosis, the organism’s slow agar growth warrants immediate tissue culture and biopsy when an immunocompromised patient presents with clinical features of atypical infection of an extremity. Mycobacterium haemophilum infection likely is underreported because of these difficulties in diagnosis.

Management
Although there are no standard guidelines for antibiotic treatment of M haemophilum, the current literature recommends triple-drug therapy with clarithromycin, ciprofloxacin, and rifamycin for at least 12 to 24 months.²

Upon clinical suspicion of an atypical Mycobacterium, we recommend a macrolide antibiotic over doxycycline, however, because this class of agents maintains broad coverage while being more specific for atypical mycobacteria. Although an atypical Mycobacterium was suspected early in the presentation in our cases, we discourage immediate use of triple-agent antibiotic therapy until laboratory evidence is procured to minimize antibiotic overuse in patients who do not have a final diagnosis. Single-agent therapy for prolonged treatment is discouraged for atypical mycobacterial infections because of the high risk of antibiotic resistance. Therapy should be tailored to the needs of the individual based on the extent of dissemination of disease and the severity of immunosuppression.^1,2

Additionally, underlying disease that results in immunosuppression might necessitate treatment reevaluation (as occurred in our cases) requiring cessation of immunosuppressive drugs, extended careful monitoring, and pharmacotherapeutic readjustment through the course of treatment. The degree to which antibiotics contribute to eradication of M haemophilum is unknown; therefore, it is recommended that long-term antibiotic use and treatment aimed at recovering the immunocompromised state (eg, highly active antiretroviral therapy in a patient with AIDS) be implemented.²

Conclusion

Our 3 cases of M haemophilum infection involved the upper extremities of immunosuppressed patients older than 65 years. This propensity to affect the upper extremities could possibly be due to the lower temperature required for growth of M haemophilum. Initial histopathologic study showed granulomatous and neutrophilic infiltrates, yet histopathologic specimens from all 3 patients failed to display positive AFB staining, which delayed the initial antibiotic choice. In all cases, diagnosis was made by tissue culture after swab culture failed to grow the pathogen. Furthermore, the 3 cases took approximately 6 weeks to achieve final identification of the organism. Neither clinical lymphadenopathy nor systemic spread was noted in our patients; immunosuppression was discontinued when possible.

Mycobacterium haemophilum is an uncommon but potentially life-threatening infection that should be suspected in immunocompromised adults who present with atypical cellulitis of the extremities. The ultimate diagnosis often is delayed because the organism grows slowly (as long as 8 weeks) in tissue culture. For that reason, empiric antibiotic treatment, including a macrolide, should be considered in patients with disseminated or severe infection or critical immunosuppression and in those who do not demonstrate improvement in symptoms once immunosuppressants are withheld. A prolonged course of multiple-drug antibiotic therapy has proved to be effective for treating cutaneous infection with M haemophilum.

Infection with Mycobacterium haemophilum, a rare, slow-growing organism, most commonly presents as ulcerating cutaneous lesions and subcutaneous nodules in immunocompromised adults.¹ The most common clinical presentation in adults includes cutaneous lesions, nodules, cysts, and papules, with signs and symptoms of erythema, pain, pruritus, and drainage.² Disseminated disease states of septic arthritis, pulmonary infiltration, and osteomyelitis, though life-threatening, are less common manifestations reported in highly immunocompromised persons.³

Infection with M haemophilum presents a challenge to the dermatology community because it is infrequently suspected and misidentified, resulting in delayed diagnosis. Additionally, M haemophilum is an extremely fastidious organism that requires heme-supplemented culture media and a carefully regulated low temperature for many consecutive weeks to yield valid culture results.¹ These features contribute to complications and delays in diagnosis of an already overlooked source of infection.

We discuss the clinical presentation, diagnosis, and treatment of 3 unusual cases of cutaneous M haemophilum infection involving the upper arms. The findings in these cases highlight the challenges inherent in diagnosis as well as the obstacles that arise in providing effective, long-term treatment of this infection.

Case Reports

Patient 1
A 69-year-old woman with a medical history of a single functioning kidney and moderate psoriasis managed with low-dosage methotrexate presented with an erythematous nonhealing wound on the left forearm that developed after she was scratched by a dog. The pustules, appearing as bright red, tender, warm abscesses, had been present for 3 months and were distributed on the left proximal and distal dorsal forearm (Figure 1A). The patient reported no recent travel, sick contacts, allergies, or new medications.

A shave biopsy was initially obtained. Swab specimens were sent for bacterial, fungal, and mycobacterial culture following discontinuation of methotrexate. Initial histopathologic analysis revealed aggregates of histiocytes and multinucleated giant cells within the dermis, surrounded by infiltrates of lymphocytes and neutrophils (Figure 2), consistent with a dermal noncaseating granulomatosis. Acid-fast bacilli (AFB), periodic acid–Schiff, Gram, and Grocott-Gomori methenamine-silver stains were negative for pathogenic microorganisms. There was no evidence of vasculitis.

Despite negative special stains, an infectious cause was still suspected. Oral doxycycline monohydrate 100 mg twice daily, oral fluconazole 200 mg daily, and econazole cream 1% were prescribed because of concern for mycobacterial infection and initial growth of Candida parapsilosis in the swab culture.

A punch biopsy also was performed at this time for both repeat histopathologic analysis and tissue culture. Follow-up appointments were scheduled every 2 weeks. Staining by AFB of the repeat histopathologic specimen was negative.

The patient demonstrated symptomatic and aesthetic improvement (Figure 1B) during consecutive regular follow-up appointments while culture results were pending. No lesions appeared above the left elbow and she had no lymphadenopathy. Results of blood chemistry analyses and complete blood cell count throughout follow-up were normal.

The final tissue culture report obtained 7 weeks after initial presentation showed growth of M haemophilum despite a negative smear. The swab culture that initially was taken did not grow pathogenic organisms.

The patient was referred to an infectious disease specialist who confirmed that the atypical mycobacterial infection likely was the main source of the cutaneous lesions. She was instructed to continue econazole cream 1% and was given prescriptions for clarithromycin 500 mg twice daily, ciprofloxacin 500 mg twice daily, and rifampin 300 mg twice daily for a total duration of 12 to 18 months. The patient has remained on this triple-drug regimen and demonstrated improvement in the lesions. She has been off methotrexate while on antibiotic therapy.

Patient 2
A 79-year-old man with a medical history of chronic lymphocytic leukemia, basal cell carcinoma, and squamous cell carcinoma presented with a nonhealing, painful, red lesion on the left forearm of 1 week’s duration. Physical examination revealed a violaceous nontender plaque with erosions and desquamation that was initially diagnosed as a carbuncle. The patient reported a similar eruption on the right foot that was successfully treated with silver sulfadiazine by another physician.

Biopsy was performed by the shave method for histologic analysis and tissue culture. Doxycycline 100 mg twice daily was prescribed because of high suspicion of infection. Histologic findings revealed granulomatous inflammation with pseudoepitheliomatous hyperplasia, reported as squamous cell carcinoma. A second opinion confirmed suspicion of an infectious process; the patient remained on doxycycline. During follow-up, the lesion progressed to a 5-cm plaque studded with pustules and satellite papules. Multiple additional tissue cultures were performed over 2 months until “light growth” of M haemophilum was reported.

The patient showed minimal improvement on tetracycline antibiotics. His condition was complicated by a photosensitivity reaction to doxycycline on the left and right forearms, hands, and nose. Consequently, triamcinolone was prescribed, doxycycline was discontinued, and minocycline 100 mg twice daily and ciprofloxacin 500 mg twice daily were prescribed.

Nine months after initial presentation, the lesions were still present but remarkably improved. The antibiotic regimen was discontinued after 11 months.

Patient 3
A 77-year-old woman with a history of rheumatoid arthritis treated with methotrexate and abatacept as well as cutaneous T-cell lymphoma treated with narrowband UVB radiation presented to the emergency department with fever and an inflamed right forearm (Figure 3A). Initial bacterial cultures of the wound and blood were negative.

Arm edema and pustules slowly resolved, but the eschar and verrucous plaques continued to slowly progress while the patient was off immunosuppression. She was kept off antibiotics until mycobacterial culture was positive at 4 weeks, at which time she was placed on doxycycline and clarithromycin. Final identification of M haemophilum was made at 6 weeks; consequently, doxycycline was discontinued and she was referred to infectious disease for multidrug therapy. She remained afebrile during the entire 6 weeks until cultures were final.

While immunosuppressants were discontinued and clarithromycin was administered, the plaque changed from an edematous pustular dermatitis to a verrucous crusted plaque. Neither epitrochlear nor axillary lymphadenopathy was noted during the treatment period. The infectious disease specialist prescribed azithromycin, ethambutol, and rifampin, which produced marked improvement (Figure 3B). The patient has remained off immunosuppressive therapy while on antibiotics.

Comment

Clinical Presentation and Diagnosis
Mycobacterium haemophilum is a rare infectious organism that affects primarily immunocompromised adults but also has been identified in immunocompetent adults and pediatric patients.² Commonly affected immunosuppressed groups include solid organ transplant recipients, bone marrow transplant recipients, human immunodeficiency virus–positive patients, and patients with rheumatoid arthritis.

The infection typically presents as small violaceous papules and pustules that become painful and erythematous, with progression and draining ulceration in later stages.² In our cases, all lesions tended to evolve into a verrucous plaque that slowly resolved with antibiotic therapy.

Due to the rarity of this infection, the initial differential diagnosis can include infection with other mycobacteria, Sporothrix, Staphylococcus aureus, and other fungal pathogens. Misdiagnosis is a common obstacle in the treatment of M haemophilum due to its rarity, often negative AFB stains, and slow growth on culture media; therefore, tissue culture is essential to successful diagnosis and management. The natural reservoir of M haemophilum is unknown, but infection has been associated with contaminated water sources.¹ In one case (patient 1), symptoms developed after a dog scratch; the other 2 patients were unaware of injury to the skin.Laboratory diagnosis of M haemophilum is inherently difficult and protracted. The species is a highly fastidious and slow-growing Mycobacterium that requires cooler (30°C) incubation for many weeks on agar medium enriched with hemin or ferric ammonium citrate to obtain valid growth.¹ To secure timely diagnosis, the organism’s slow agar growth warrants immediate tissue culture and biopsy when an immunocompromised patient presents with clinical features of atypical infection of an extremity. Mycobacterium haemophilum infection likely is underreported because of these difficulties in diagnosis.

Management
Although there are no standard guidelines for antibiotic treatment of M haemophilum, the current literature recommends triple-drug therapy with clarithromycin, ciprofloxacin, and rifamycin for at least 12 to 24 months.²

Upon clinical suspicion of an atypical Mycobacterium, we recommend a macrolide antibiotic over doxycycline, however, because this class of agents maintains broad coverage while being more specific for atypical mycobacteria. Although an atypical Mycobacterium was suspected early in the presentation in our cases, we discourage immediate use of triple-agent antibiotic therapy until laboratory evidence is procured to minimize antibiotic overuse in patients who do not have a final diagnosis. Single-agent therapy for prolonged treatment is discouraged for atypical mycobacterial infections because of the high risk of antibiotic resistance. Therapy should be tailored to the needs of the individual based on the extent of dissemination of disease and the severity of immunosuppression.^1,2

Additionally, underlying disease that results in immunosuppression might necessitate treatment reevaluation (as occurred in our cases) requiring cessation of immunosuppressive drugs, extended careful monitoring, and pharmacotherapeutic readjustment through the course of treatment. The degree to which antibiotics contribute to eradication of M haemophilum is unknown; therefore, it is recommended that long-term antibiotic use and treatment aimed at recovering the immunocompromised state (eg, highly active antiretroviral therapy in a patient with AIDS) be implemented.²

Conclusion

Our 3 cases of M haemophilum infection involved the upper extremities of immunosuppressed patients older than 65 years. This propensity to affect the upper extremities could possibly be due to the lower temperature required for growth of M haemophilum. Initial histopathologic study showed granulomatous and neutrophilic infiltrates, yet histopathologic specimens from all 3 patients failed to display positive AFB staining, which delayed the initial antibiotic choice. In all cases, diagnosis was made by tissue culture after swab culture failed to grow the pathogen. Furthermore, the 3 cases took approximately 6 weeks to achieve final identification of the organism. Neither clinical lymphadenopathy nor systemic spread was noted in our patients; immunosuppression was discontinued when possible.

Mycobacterium haemophilum is an uncommon but potentially life-threatening infection that should be suspected in immunocompromised adults who present with atypical cellulitis of the extremities. The ultimate diagnosis often is delayed because the organism grows slowly (as long as 8 weeks) in tissue culture. For that reason, empiric antibiotic treatment, including a macrolide, should be considered in patients with disseminated or severe infection or critical immunosuppression and in those who do not demonstrate improvement in symptoms once immunosuppressants are withheld. A prolonged course of multiple-drug antibiotic therapy has proved to be effective for treating cutaneous infection with M haemophilum.

Infection with Mycobacterium haemophilum, a rare, slow-growing organism, most commonly presents as ulcerating cutaneous lesions and subcutaneous nodules in immunocompromised adults.¹ The most common clinical presentation in adults includes cutaneous lesions, nodules, cysts, and papules, with signs and symptoms of erythema, pain, pruritus, and drainage.² Disseminated disease states of septic arthritis, pulmonary infiltration, and osteomyelitis, though life-threatening, are less common manifestations reported in highly immunocompromised persons.³

Infection with M haemophilum presents a challenge to the dermatology community because it is infrequently suspected and misidentified, resulting in delayed diagnosis. Additionally, M haemophilum is an extremely fastidious organism that requires heme-supplemented culture media and a carefully regulated low temperature for many consecutive weeks to yield valid culture results.¹ These features contribute to complications and delays in diagnosis of an already overlooked source of infection.

We discuss the clinical presentation, diagnosis, and treatment of 3 unusual cases of cutaneous M haemophilum infection involving the upper arms. The findings in these cases highlight the challenges inherent in diagnosis as well as the obstacles that arise in providing effective, long-term treatment of this infection.

Case Reports

Patient 1
A 69-year-old woman with a medical history of a single functioning kidney and moderate psoriasis managed with low-dosage methotrexate presented with an erythematous nonhealing wound on the left forearm that developed after she was scratched by a dog. The pustules, appearing as bright red, tender, warm abscesses, had been present for 3 months and were distributed on the left proximal and distal dorsal forearm (Figure 1A). The patient reported no recent travel, sick contacts, allergies, or new medications.

A shave biopsy was initially obtained. Swab specimens were sent for bacterial, fungal, and mycobacterial culture following discontinuation of methotrexate. Initial histopathologic analysis revealed aggregates of histiocytes and multinucleated giant cells within the dermis, surrounded by infiltrates of lymphocytes and neutrophils (Figure 2), consistent with a dermal noncaseating granulomatosis. Acid-fast bacilli (AFB), periodic acid–Schiff, Gram, and Grocott-Gomori methenamine-silver stains were negative for pathogenic microorganisms. There was no evidence of vasculitis.

Despite negative special stains, an infectious cause was still suspected. Oral doxycycline monohydrate 100 mg twice daily, oral fluconazole 200 mg daily, and econazole cream 1% were prescribed because of concern for mycobacterial infection and initial growth of Candida parapsilosis in the swab culture.

A punch biopsy also was performed at this time for both repeat histopathologic analysis and tissue culture. Follow-up appointments were scheduled every 2 weeks. Staining by AFB of the repeat histopathologic specimen was negative.

The patient demonstrated symptomatic and aesthetic improvement (Figure 1B) during consecutive regular follow-up appointments while culture results were pending. No lesions appeared above the left elbow and she had no lymphadenopathy. Results of blood chemistry analyses and complete blood cell count throughout follow-up were normal.

The final tissue culture report obtained 7 weeks after initial presentation showed growth of M haemophilum despite a negative smear. The swab culture that initially was taken did not grow pathogenic organisms.

The patient was referred to an infectious disease specialist who confirmed that the atypical mycobacterial infection likely was the main source of the cutaneous lesions. She was instructed to continue econazole cream 1% and was given prescriptions for clarithromycin 500 mg twice daily, ciprofloxacin 500 mg twice daily, and rifampin 300 mg twice daily for a total duration of 12 to 18 months. The patient has remained on this triple-drug regimen and demonstrated improvement in the lesions. She has been off methotrexate while on antibiotic therapy.

Patient 2
A 79-year-old man with a medical history of chronic lymphocytic leukemia, basal cell carcinoma, and squamous cell carcinoma presented with a nonhealing, painful, red lesion on the left forearm of 1 week’s duration. Physical examination revealed a violaceous nontender plaque with erosions and desquamation that was initially diagnosed as a carbuncle. The patient reported a similar eruption on the right foot that was successfully treated with silver sulfadiazine by another physician.

Biopsy was performed by the shave method for histologic analysis and tissue culture. Doxycycline 100 mg twice daily was prescribed because of high suspicion of infection. Histologic findings revealed granulomatous inflammation with pseudoepitheliomatous hyperplasia, reported as squamous cell carcinoma. A second opinion confirmed suspicion of an infectious process; the patient remained on doxycycline. During follow-up, the lesion progressed to a 5-cm plaque studded with pustules and satellite papules. Multiple additional tissue cultures were performed over 2 months until “light growth” of M haemophilum was reported.

The patient showed minimal improvement on tetracycline antibiotics. His condition was complicated by a photosensitivity reaction to doxycycline on the left and right forearms, hands, and nose. Consequently, triamcinolone was prescribed, doxycycline was discontinued, and minocycline 100 mg twice daily and ciprofloxacin 500 mg twice daily were prescribed.

Nine months after initial presentation, the lesions were still present but remarkably improved. The antibiotic regimen was discontinued after 11 months.

Patient 3
A 77-year-old woman with a history of rheumatoid arthritis treated with methotrexate and abatacept as well as cutaneous T-cell lymphoma treated with narrowband UVB radiation presented to the emergency department with fever and an inflamed right forearm (Figure 3A). Initial bacterial cultures of the wound and blood were negative.

Arm edema and pustules slowly resolved, but the eschar and verrucous plaques continued to slowly progress while the patient was off immunosuppression. She was kept off antibiotics until mycobacterial culture was positive at 4 weeks, at which time she was placed on doxycycline and clarithromycin. Final identification of M haemophilum was made at 6 weeks; consequently, doxycycline was discontinued and she was referred to infectious disease for multidrug therapy. She remained afebrile during the entire 6 weeks until cultures were final.

While immunosuppressants were discontinued and clarithromycin was administered, the plaque changed from an edematous pustular dermatitis to a verrucous crusted plaque. Neither epitrochlear nor axillary lymphadenopathy was noted during the treatment period. The infectious disease specialist prescribed azithromycin, ethambutol, and rifampin, which produced marked improvement (Figure 3B). The patient has remained off immunosuppressive therapy while on antibiotics.

Comment

Clinical Presentation and Diagnosis
Mycobacterium haemophilum is a rare infectious organism that affects primarily immunocompromised adults but also has been identified in immunocompetent adults and pediatric patients.² Commonly affected immunosuppressed groups include solid organ transplant recipients, bone marrow transplant recipients, human immunodeficiency virus–positive patients, and patients with rheumatoid arthritis.

The infection typically presents as small violaceous papules and pustules that become painful and erythematous, with progression and draining ulceration in later stages.² In our cases, all lesions tended to evolve into a verrucous plaque that slowly resolved with antibiotic therapy.

Due to the rarity of this infection, the initial differential diagnosis can include infection with other mycobacteria, Sporothrix, Staphylococcus aureus, and other fungal pathogens. Misdiagnosis is a common obstacle in the treatment of M haemophilum due to its rarity, often negative AFB stains, and slow growth on culture media; therefore, tissue culture is essential to successful diagnosis and management. The natural reservoir of M haemophilum is unknown, but infection has been associated with contaminated water sources.¹ In one case (patient 1), symptoms developed after a dog scratch; the other 2 patients were unaware of injury to the skin.Laboratory diagnosis of M haemophilum is inherently difficult and protracted. The species is a highly fastidious and slow-growing Mycobacterium that requires cooler (30°C) incubation for many weeks on agar medium enriched with hemin or ferric ammonium citrate to obtain valid growth.¹ To secure timely diagnosis, the organism’s slow agar growth warrants immediate tissue culture and biopsy when an immunocompromised patient presents with clinical features of atypical infection of an extremity. Mycobacterium haemophilum infection likely is underreported because of these difficulties in diagnosis.

Management
Although there are no standard guidelines for antibiotic treatment of M haemophilum, the current literature recommends triple-drug therapy with clarithromycin, ciprofloxacin, and rifamycin for at least 12 to 24 months.²

Upon clinical suspicion of an atypical Mycobacterium, we recommend a macrolide antibiotic over doxycycline, however, because this class of agents maintains broad coverage while being more specific for atypical mycobacteria. Although an atypical Mycobacterium was suspected early in the presentation in our cases, we discourage immediate use of triple-agent antibiotic therapy until laboratory evidence is procured to minimize antibiotic overuse in patients who do not have a final diagnosis. Single-agent therapy for prolonged treatment is discouraged for atypical mycobacterial infections because of the high risk of antibiotic resistance. Therapy should be tailored to the needs of the individual based on the extent of dissemination of disease and the severity of immunosuppression.^1,2

Additionally, underlying disease that results in immunosuppression might necessitate treatment reevaluation (as occurred in our cases) requiring cessation of immunosuppressive drugs, extended careful monitoring, and pharmacotherapeutic readjustment through the course of treatment. The degree to which antibiotics contribute to eradication of M haemophilum is unknown; therefore, it is recommended that long-term antibiotic use and treatment aimed at recovering the immunocompromised state (eg, highly active antiretroviral therapy in a patient with AIDS) be implemented.²

Conclusion

Our 3 cases of M haemophilum infection involved the upper extremities of immunosuppressed patients older than 65 years. This propensity to affect the upper extremities could possibly be due to the lower temperature required for growth of M haemophilum. Initial histopathologic study showed granulomatous and neutrophilic infiltrates, yet histopathologic specimens from all 3 patients failed to display positive AFB staining, which delayed the initial antibiotic choice. In all cases, diagnosis was made by tissue culture after swab culture failed to grow the pathogen. Furthermore, the 3 cases took approximately 6 weeks to achieve final identification of the organism. Neither clinical lymphadenopathy nor systemic spread was noted in our patients; immunosuppression was discontinued when possible.

Mycobacterium haemophilum is an uncommon but potentially life-threatening infection that should be suspected in immunocompromised adults who present with atypical cellulitis of the extremities. The ultimate diagnosis often is delayed because the organism grows slowly (as long as 8 weeks) in tissue culture. For that reason, empiric antibiotic treatment, including a macrolide, should be considered in patients with disseminated or severe infection or critical immunosuppression and in those who do not demonstrate improvement in symptoms once immunosuppressants are withheld. A prolonged course of multiple-drug antibiotic therapy has proved to be effective for treating cutaneous infection with M haemophilum.

Case Report

A 79-year-old man with chronic lymphocytic leukemia (CLL) who was being treated with ibrutinib presented to the emergency department with a dry cough, ataxia and falls, and vision loss. Physical examination was remarkable for diffuse crackles heard throughout the right lung and bilateral lower extremity weakness. Additionally, he had 4 pink mobile nodules on the left side of the forehead, right side of the chin, left submental area, and left postauricular scalp, which arose approximately 2 weeks prior to presentation. The left postauricular lesion had been tender at times and had developed a crust. The cutaneous lesions were all smaller than 2 cm.

The patient had a history of squamous cell carcinoma of the skin and was under the care of a dermatologist as an outpatient. His dermatologist had described him as an active gardener; he was noted to have healing abrasions on the forearms due to gardening raspberry bushes.

Computed tomography of the head revealed a 14-mm, ring-enhancing lesion in the left paramedian posterior frontal lobe with surrounding white matter vasogenic edema (Figure 1). Computed tomography of the chest revealed a peripheral mass on the right upper lobe measuring 6.3 cm at its greatest dimension (Figure 2).

Clinically, the constellation of symptoms was thought to represent an infectious process or less likely metastatic malignancy. Biopsies of the nodule on the right side of the chin were performed and sent for culture and histologic examination. Sections from the anterior right chin showed compact orthokeratosis overlying a slightly spongiotic epidermis (Figure 3). Within the deep dermis, there was a dense mixed inflammatory infiltrate comprising predominantly neutrophils, with occasional eosinophils, lymphocytes, and histiocytes (Figure 4).

Gram stain revealed gram-variable, branching, bacterial organisms morphologically consistent with Nocardia. Grocott-Gomori methenamine-silver and periodic acid–Schiff stains also highlighted the bacterial organisms (Figure 5). An auramine-O stain was negative for acid-fast microorganisms. After 3 days on a blood agar plate, cultures of a specimen of the chin nodule grew branching filamentous bacterial organisms consistent with Nocardia.

Comment

Presentation and Diagnosis
Nocardiosis is an infrequently encountered opportunistic infection that typically targets skin, lungs, and the central nervous system (CNS). Nocardia species characteristically are gram-positive, thin rods that form beaded, right-angle, branching filaments.¹ More than 50 Nocardia species have been clinically isolated.²

Definitive diagnosis requires culture. Nocardia grows well on nonselective media, such as blood or Löwenstein-Jensen agar; growth can be enhanced with 10% CO₂. Growth can be slow, however, and takes from 48 hours to several weeks. Nocardia typically grows as buff or pigmented, waxy, cerebriform colonies at 3 to 5 days’ incubation.¹

Cause of Infection
Nocardia species are commonly found in the environment—soil, plant matter, water, and decomposing organic material—as well as in the gastrointestinal tract and skin of animals. Infection has been reported in cattle, dogs, horses, swine, birds, cats, foxes, and a few other animals.² A history of exposure, such as gardening or handling animals, should increase suspicion of Nocardia.³ Although infection is classically thought to affect immunocompromised patients, there are case reports of immunocompetent individuals developing disseminated infection.^4-7 However, infected immunocompetent individuals typically have localized cutaneous infection, which often includes cellulitis, abscesses, or sporotrichoid patterns.² Cutaneous infections typically are the result of direct inoculation of the skin through a penetrating injury.⁸

Disseminated nocardiosis can be caused by numerous species and generally is the result of primary pulmonary infection.⁹ In these cases, skin disease is present in approximately 10% of patients. Disseminated infection from cutaneous nocardiosis is uncommon; when it does occur, the most common site of dissemination is the CNS, resulting in abscess or cerebritis.¹⁰ Therefore, CNS involvement should always be ruled out on diagnosis in immunocompromised patients, even if neurologic symptoms are absent.⁹ Nearly 80% of patients with disseminated disease are, in fact, immunocompromised.⁸

Association With CLL
Chronic lymphocytic leukemia is associated with profound immunodeficiency caused by quantitative and qualitative aberrations in both innate and adaptive immunity. This perturbation of the immune system predisposes the patient to infection.^11,12 Early in the course of CLL, a patient develops neutropenia, which predisposes to bacterial infection; later, the patient develops a sustained B- and T-cell immunodeficiency that predisposes to opportunistic infection.¹³ Treatment-naïve patients with CLL are commonly diagnosed with respiratory and urinary tract infections.¹² Chronic lymphocytic leukemia patients treated with alemtuzumab or purine analogs have been reported to have the highest risk for major infection.¹⁴

Ibrutinib is a commonly used treatment of CLL because it induces apoptosis in B cells, which are abnormal in CLL. Ibrutinib functions by inhibiting the Bruton tyrosine kinase pathway, which is essential in B-cell production and maintenance.¹⁵ Studies have reported a high rate of infection in ibrutinib-treated CLL patients^14,16; salvage ibrutinib therapy has been associated with higher infection risk than primary ibrutinib therapy.^16,17 Long-term follow-up studies have shown a decreased rate of infection in ibrutinib-treated CLL after 2 years or longer of treatment, suggesting a reconstitution of normal B cells and humoral immunity with longer ibrutinib therapy.^16,17

Many infections have been identified in association with ibrutinib therapy, including invasive aspergillosis, disseminated fusariosis, cerebral mucormycosis, disseminated cryptococcosis, and Pneumocystis jirovecii pneumonia.^18-22 Disseminated nocardiosis has been reported in a few patients with CLL, though the treatment they received for CLL varied from case to case.^23-25

Identification and Treatment
Clinical and microscopic identification of Nocardia organisms can be exceedingly difficult. Primary cutaneous nocardiosis clinically presents as tumors or nodules that often have a sporotrichoid pattern along the lymphatics. In disease that disseminates to skin, nocardiosis presents as vesiculopustules or abscesses. The biopsy specimen most often shows a dense dermal and subcutaneous infiltrate of neutrophils with abscess formation. Long-standing lesions might show chronic inflammation and nonspecific granulomas.

The appearance of Nocardia organisms is quite subtle on hematoxylin and eosin staining and can be easily missed. Special stains, such as Gram and Grocott-Gomori methenamine-silver stains as well as stains for acid-fast organisms, can be invaluable in diagnosing this disease. Biopsy in immunocompromised patients when nocardiosis is part of the differential diagnosis requires extra attention because the organisms can be gram variable and only partially acid fast, as was the case in our patient. Organisms typically will be positive with silver stains.

Trimethoprim-sulfamethoxazole typically is the first-line treatment of nocardiosis. Although prognosis is excellent when disease is confined to skin, disseminated infection has 25% mortality.⁸ Diagnosticians should maintain a high index of suspicion for the disease, especially in immunocompromised patients, because clinical and imaging findings can be nonspecific.

Conclusion

Our patient’s primary risk factor for nocardiosis was his immunocompromised state. In addition, he was an avid gardener, which increased his risk for exposure to the microorganism. Given the timing of disease progression, our case most likely represents primary cutaneous nocardiosis with dissemination to brain, lungs, and other organs, leading to death, and serves as a reminder to dermatologists and pathologists to establish a broad differential diagnosis when dealing with an infectious process in immunocompromised patients.

Case Report

A 79-year-old man with chronic lymphocytic leukemia (CLL) who was being treated with ibrutinib presented to the emergency department with a dry cough, ataxia and falls, and vision loss. Physical examination was remarkable for diffuse crackles heard throughout the right lung and bilateral lower extremity weakness. Additionally, he had 4 pink mobile nodules on the left side of the forehead, right side of the chin, left submental area, and left postauricular scalp, which arose approximately 2 weeks prior to presentation. The left postauricular lesion had been tender at times and had developed a crust. The cutaneous lesions were all smaller than 2 cm.

The patient had a history of squamous cell carcinoma of the skin and was under the care of a dermatologist as an outpatient. His dermatologist had described him as an active gardener; he was noted to have healing abrasions on the forearms due to gardening raspberry bushes.

Computed tomography of the head revealed a 14-mm, ring-enhancing lesion in the left paramedian posterior frontal lobe with surrounding white matter vasogenic edema (Figure 1). Computed tomography of the chest revealed a peripheral mass on the right upper lobe measuring 6.3 cm at its greatest dimension (Figure 2).

Clinically, the constellation of symptoms was thought to represent an infectious process or less likely metastatic malignancy. Biopsies of the nodule on the right side of the chin were performed and sent for culture and histologic examination. Sections from the anterior right chin showed compact orthokeratosis overlying a slightly spongiotic epidermis (Figure 3). Within the deep dermis, there was a dense mixed inflammatory infiltrate comprising predominantly neutrophils, with occasional eosinophils, lymphocytes, and histiocytes (Figure 4).

Gram stain revealed gram-variable, branching, bacterial organisms morphologically consistent with Nocardia. Grocott-Gomori methenamine-silver and periodic acid–Schiff stains also highlighted the bacterial organisms (Figure 5). An auramine-O stain was negative for acid-fast microorganisms. After 3 days on a blood agar plate, cultures of a specimen of the chin nodule grew branching filamentous bacterial organisms consistent with Nocardia.

Comment

Presentation and Diagnosis
Nocardiosis is an infrequently encountered opportunistic infection that typically targets skin, lungs, and the central nervous system (CNS). Nocardia species characteristically are gram-positive, thin rods that form beaded, right-angle, branching filaments.¹ More than 50 Nocardia species have been clinically isolated.²

Definitive diagnosis requires culture. Nocardia grows well on nonselective media, such as blood or Löwenstein-Jensen agar; growth can be enhanced with 10% CO₂. Growth can be slow, however, and takes from 48 hours to several weeks. Nocardia typically grows as buff or pigmented, waxy, cerebriform colonies at 3 to 5 days’ incubation.¹

Cause of Infection
Nocardia species are commonly found in the environment—soil, plant matter, water, and decomposing organic material—as well as in the gastrointestinal tract and skin of animals. Infection has been reported in cattle, dogs, horses, swine, birds, cats, foxes, and a few other animals.² A history of exposure, such as gardening or handling animals, should increase suspicion of Nocardia.³ Although infection is classically thought to affect immunocompromised patients, there are case reports of immunocompetent individuals developing disseminated infection.^4-7 However, infected immunocompetent individuals typically have localized cutaneous infection, which often includes cellulitis, abscesses, or sporotrichoid patterns.² Cutaneous infections typically are the result of direct inoculation of the skin through a penetrating injury.⁸

Disseminated nocardiosis can be caused by numerous species and generally is the result of primary pulmonary infection.⁹ In these cases, skin disease is present in approximately 10% of patients. Disseminated infection from cutaneous nocardiosis is uncommon; when it does occur, the most common site of dissemination is the CNS, resulting in abscess or cerebritis.¹⁰ Therefore, CNS involvement should always be ruled out on diagnosis in immunocompromised patients, even if neurologic symptoms are absent.⁹ Nearly 80% of patients with disseminated disease are, in fact, immunocompromised.⁸

Association With CLL
Chronic lymphocytic leukemia is associated with profound immunodeficiency caused by quantitative and qualitative aberrations in both innate and adaptive immunity. This perturbation of the immune system predisposes the patient to infection.^11,12 Early in the course of CLL, a patient develops neutropenia, which predisposes to bacterial infection; later, the patient develops a sustained B- and T-cell immunodeficiency that predisposes to opportunistic infection.¹³ Treatment-naïve patients with CLL are commonly diagnosed with respiratory and urinary tract infections.¹² Chronic lymphocytic leukemia patients treated with alemtuzumab or purine analogs have been reported to have the highest risk for major infection.¹⁴

Ibrutinib is a commonly used treatment of CLL because it induces apoptosis in B cells, which are abnormal in CLL. Ibrutinib functions by inhibiting the Bruton tyrosine kinase pathway, which is essential in B-cell production and maintenance.¹⁵ Studies have reported a high rate of infection in ibrutinib-treated CLL patients^14,16; salvage ibrutinib therapy has been associated with higher infection risk than primary ibrutinib therapy.^16,17 Long-term follow-up studies have shown a decreased rate of infection in ibrutinib-treated CLL after 2 years or longer of treatment, suggesting a reconstitution of normal B cells and humoral immunity with longer ibrutinib therapy.^16,17

Many infections have been identified in association with ibrutinib therapy, including invasive aspergillosis, disseminated fusariosis, cerebral mucormycosis, disseminated cryptococcosis, and Pneumocystis jirovecii pneumonia.^18-22 Disseminated nocardiosis has been reported in a few patients with CLL, though the treatment they received for CLL varied from case to case.^23-25

Identification and Treatment
Clinical and microscopic identification of Nocardia organisms can be exceedingly difficult. Primary cutaneous nocardiosis clinically presents as tumors or nodules that often have a sporotrichoid pattern along the lymphatics. In disease that disseminates to skin, nocardiosis presents as vesiculopustules or abscesses. The biopsy specimen most often shows a dense dermal and subcutaneous infiltrate of neutrophils with abscess formation. Long-standing lesions might show chronic inflammation and nonspecific granulomas.

The appearance of Nocardia organisms is quite subtle on hematoxylin and eosin staining and can be easily missed. Special stains, such as Gram and Grocott-Gomori methenamine-silver stains as well as stains for acid-fast organisms, can be invaluable in diagnosing this disease. Biopsy in immunocompromised patients when nocardiosis is part of the differential diagnosis requires extra attention because the organisms can be gram variable and only partially acid fast, as was the case in our patient. Organisms typically will be positive with silver stains.

Trimethoprim-sulfamethoxazole typically is the first-line treatment of nocardiosis. Although prognosis is excellent when disease is confined to skin, disseminated infection has 25% mortality.⁸ Diagnosticians should maintain a high index of suspicion for the disease, especially in immunocompromised patients, because clinical and imaging findings can be nonspecific.

Conclusion

Our patient’s primary risk factor for nocardiosis was his immunocompromised state. In addition, he was an avid gardener, which increased his risk for exposure to the microorganism. Given the timing of disease progression, our case most likely represents primary cutaneous nocardiosis with dissemination to brain, lungs, and other organs, leading to death, and serves as a reminder to dermatologists and pathologists to establish a broad differential diagnosis when dealing with an infectious process in immunocompromised patients.

Case Report

A 79-year-old man with chronic lymphocytic leukemia (CLL) who was being treated with ibrutinib presented to the emergency department with a dry cough, ataxia and falls, and vision loss. Physical examination was remarkable for diffuse crackles heard throughout the right lung and bilateral lower extremity weakness. Additionally, he had 4 pink mobile nodules on the left side of the forehead, right side of the chin, left submental area, and left postauricular scalp, which arose approximately 2 weeks prior to presentation. The left postauricular lesion had been tender at times and had developed a crust. The cutaneous lesions were all smaller than 2 cm.

The patient had a history of squamous cell carcinoma of the skin and was under the care of a dermatologist as an outpatient. His dermatologist had described him as an active gardener; he was noted to have healing abrasions on the forearms due to gardening raspberry bushes.

Computed tomography of the head revealed a 14-mm, ring-enhancing lesion in the left paramedian posterior frontal lobe with surrounding white matter vasogenic edema (Figure 1). Computed tomography of the chest revealed a peripheral mass on the right upper lobe measuring 6.3 cm at its greatest dimension (Figure 2).

Clinically, the constellation of symptoms was thought to represent an infectious process or less likely metastatic malignancy. Biopsies of the nodule on the right side of the chin were performed and sent for culture and histologic examination. Sections from the anterior right chin showed compact orthokeratosis overlying a slightly spongiotic epidermis (Figure 3). Within the deep dermis, there was a dense mixed inflammatory infiltrate comprising predominantly neutrophils, with occasional eosinophils, lymphocytes, and histiocytes (Figure 4).

Gram stain revealed gram-variable, branching, bacterial organisms morphologically consistent with Nocardia. Grocott-Gomori methenamine-silver and periodic acid–Schiff stains also highlighted the bacterial organisms (Figure 5). An auramine-O stain was negative for acid-fast microorganisms. After 3 days on a blood agar plate, cultures of a specimen of the chin nodule grew branching filamentous bacterial organisms consistent with Nocardia.

Comment

Presentation and Diagnosis
Nocardiosis is an infrequently encountered opportunistic infection that typically targets skin, lungs, and the central nervous system (CNS). Nocardia species characteristically are gram-positive, thin rods that form beaded, right-angle, branching filaments.¹ More than 50 Nocardia species have been clinically isolated.²

Definitive diagnosis requires culture. Nocardia grows well on nonselective media, such as blood or Löwenstein-Jensen agar; growth can be enhanced with 10% CO₂. Growth can be slow, however, and takes from 48 hours to several weeks. Nocardia typically grows as buff or pigmented, waxy, cerebriform colonies at 3 to 5 days’ incubation.¹

Cause of Infection
Nocardia species are commonly found in the environment—soil, plant matter, water, and decomposing organic material—as well as in the gastrointestinal tract and skin of animals. Infection has been reported in cattle, dogs, horses, swine, birds, cats, foxes, and a few other animals.² A history of exposure, such as gardening or handling animals, should increase suspicion of Nocardia.³ Although infection is classically thought to affect immunocompromised patients, there are case reports of immunocompetent individuals developing disseminated infection.^4-7 However, infected immunocompetent individuals typically have localized cutaneous infection, which often includes cellulitis, abscesses, or sporotrichoid patterns.² Cutaneous infections typically are the result of direct inoculation of the skin through a penetrating injury.⁸

Disseminated nocardiosis can be caused by numerous species and generally is the result of primary pulmonary infection.⁹ In these cases, skin disease is present in approximately 10% of patients. Disseminated infection from cutaneous nocardiosis is uncommon; when it does occur, the most common site of dissemination is the CNS, resulting in abscess or cerebritis.¹⁰ Therefore, CNS involvement should always be ruled out on diagnosis in immunocompromised patients, even if neurologic symptoms are absent.⁹ Nearly 80% of patients with disseminated disease are, in fact, immunocompromised.⁸

Association With CLL
Chronic lymphocytic leukemia is associated with profound immunodeficiency caused by quantitative and qualitative aberrations in both innate and adaptive immunity. This perturbation of the immune system predisposes the patient to infection.^11,12 Early in the course of CLL, a patient develops neutropenia, which predisposes to bacterial infection; later, the patient develops a sustained B- and T-cell immunodeficiency that predisposes to opportunistic infection.¹³ Treatment-naïve patients with CLL are commonly diagnosed with respiratory and urinary tract infections.¹² Chronic lymphocytic leukemia patients treated with alemtuzumab or purine analogs have been reported to have the highest risk for major infection.¹⁴

Ibrutinib is a commonly used treatment of CLL because it induces apoptosis in B cells, which are abnormal in CLL. Ibrutinib functions by inhibiting the Bruton tyrosine kinase pathway, which is essential in B-cell production and maintenance.¹⁵ Studies have reported a high rate of infection in ibrutinib-treated CLL patients^14,16; salvage ibrutinib therapy has been associated with higher infection risk than primary ibrutinib therapy.^16,17 Long-term follow-up studies have shown a decreased rate of infection in ibrutinib-treated CLL after 2 years or longer of treatment, suggesting a reconstitution of normal B cells and humoral immunity with longer ibrutinib therapy.^16,17

Many infections have been identified in association with ibrutinib therapy, including invasive aspergillosis, disseminated fusariosis, cerebral mucormycosis, disseminated cryptococcosis, and Pneumocystis jirovecii pneumonia.^18-22 Disseminated nocardiosis has been reported in a few patients with CLL, though the treatment they received for CLL varied from case to case.^23-25

Identification and Treatment
Clinical and microscopic identification of Nocardia organisms can be exceedingly difficult. Primary cutaneous nocardiosis clinically presents as tumors or nodules that often have a sporotrichoid pattern along the lymphatics. In disease that disseminates to skin, nocardiosis presents as vesiculopustules or abscesses. The biopsy specimen most often shows a dense dermal and subcutaneous infiltrate of neutrophils with abscess formation. Long-standing lesions might show chronic inflammation and nonspecific granulomas.

The appearance of Nocardia organisms is quite subtle on hematoxylin and eosin staining and can be easily missed. Special stains, such as Gram and Grocott-Gomori methenamine-silver stains as well as stains for acid-fast organisms, can be invaluable in diagnosing this disease. Biopsy in immunocompromised patients when nocardiosis is part of the differential diagnosis requires extra attention because the organisms can be gram variable and only partially acid fast, as was the case in our patient. Organisms typically will be positive with silver stains.

Trimethoprim-sulfamethoxazole typically is the first-line treatment of nocardiosis. Although prognosis is excellent when disease is confined to skin, disseminated infection has 25% mortality.⁸ Diagnosticians should maintain a high index of suspicion for the disease, especially in immunocompromised patients, because clinical and imaging findings can be nonspecific.

Conclusion

Our patient’s primary risk factor for nocardiosis was his immunocompromised state. In addition, he was an avid gardener, which increased his risk for exposure to the microorganism. Given the timing of disease progression, our case most likely represents primary cutaneous nocardiosis with dissemination to brain, lungs, and other organs, leading to death, and serves as a reminder to dermatologists and pathologists to establish a broad differential diagnosis when dealing with an infectious process in immunocompromised patients.

Asthma attacks account for 1.8 million emergency department (ED) visits each year in the US and for 10 deaths daily.¹ Management of asthma attacks includes administration of inhaled ß₂ adrenergic agonists, inhaled anticholinergic agents, IV magnesium sulfate, and corticosteroids.² Status asthmaticus is an intense acute exacerbation of asthma that does not respond to repeated treatments of bronchodilators and corticosteroids.³ It is a medical emergency requiring immediate recognition and treatment. The decision to intubate a patient with status asthmaticus is a clinical decision based on work of breathing, respiratory acidosis, and failure to respond to medical interventions.

In refractory cases of status asthmaticus, intubation and mechanical ventilation are undertaken to provide oxygenation and ventilation until the bronchospasm resolves. However, mechanical ventilation is associated with significant risks, including high end-inspiratory pressures, barotrauma, and volutrauma.⁴ Rescue therapies include muscle relaxation, infusion of ketamine (central acting nonopioid analgesic with bronchodilatory properties), heliox, and general anesthesia.^2,4 We report a case of a patient with life-threatening asthma and status asthmaticus treated with sevoflurane general anesthesia.

Case Presentation

A 55-year-old woman whose medical history was notable for asthma, psoriasis, hypothyroidism, tobacco, and alcohol abuse, and posttraumatic stress disorder (PTSD) presented to the ED. The patient had rarely sought medical attention and had no prior ED visits or hospitalizations in the electronic health record. Her home regimen included an albuterol inhaler used as needed. Her family reported that they had found her in distress in bed in a tripod position, unable to speak and struggling to breath.

Emergency medical services found the patient cyanotic, apneic, and pulseless. She received cardiopulmonary resuscitation for 30 seconds and 1-mg IV epinephrine, and spontaneous circulation returned. The patient arrived in the ED with an oral airway in place receiving bag valve mask ventilation. The patient expelled the oral airway. She was unable to speak due to dyspnea, exhibited persistent cyanosis, fatigue due to work of breathing, and failed to respond to nebulized albuterol/ipratropium bromide, IV methylprednisolone, and magnesium sulfate. The patient met criteria for acute severe asthma, or status asthmaticus. Thus, the patient received rapid sequence induction with rocuronium and ketamine and was intubated.

According to her family, the patient had no previous intensive care unit (ICU) admissions or prior intubations. Her only asthma medication was an albuterol inhaler as needed. The patient worked as a supervisor at a window blind manufacturing company. She lived alone, smoked 2 packs of cigarettes a day for more than 30 years, had no pets, drank unknown quantities of beer, wine, and hard liquor daily, and had smoked marijuana for several years.

The patient’s physical examination was notable for diffuse expiratory wheezes. Laboratory analysis revealed white blood cell count of 13.7 k/mcL, sodium 140 mmol/L, potassium 4.9 mmol/L, chloride 105 mmol/L, CO₂ 17 mmol/L, creatinine 0.98 mg/dL, troponin 0.03 ng/mL, lactate 7.2 mmol/L. Her chest X-ray showed hyperinflation but no focal opacities, pneumothorax, or pulmonary edema. Her endotracheal tube was in good position (Figure 1). A computed tomography pulmonary angiogram showed no pulmonary embolus or emphysema. There were atelectatic changes in the dependent portion of the right lower lobe, central bronchial wall thickening, and no stigmata of air trapping (Figure 2). An echocardiogram revealed a left ventricular ejection fraction of 45%, normal right ventricle and right ventricular size and function with an estimated right ventricular systolic pressure of 40 mm Hg.

The patient was admitted to the ICU and started on continuous infusion cisatracurium for paralysis and deep sedation to improve ventilatory synchrony and decrease auto positive end-expiratory pressure (PEEP). Mechanical ventilation was initiated with volume-cycled assist control ventilation, 6 mL/kg/ideal body weight (IBW) at 5-cm H₂O PEEP, and 1 minute ventilation of 10 liters. The patient had severe air trapping and high airway pressures. The dynamic PEEP was 22-cm H₂O (normal PEEP of 5-cm H₂O), peak airway pressure (PAP) 41-cm H₂O, and plateau pressure 31-cm H₂O. In addition, the arterial blood gas (ABG) showed severe hypercapnic respiratory acidosis without significant hypoxemia with pH 7.15, PaCO₂ 90 mm Hg, and PaO₂ 150 mm Hg.

Pressure controlled ventilation was attempted unsuccessfully due to high airway resistance. Ultimately, the patient was set on volume control with low tidal volume, 6 mL/kg/IBW, high flow 90 L/min, PEEP 0 cm of H₂O, and a low respiratory rate of 10 to achieve an inspiratory to expiratory (I:E) ratio of 1:7. Managing the ventilator to avoid dynamic hyperinflation and auto-PEEP, she remained relatively stable and improved.

By day 4 the patient’s ventilator was set to volume assist control with respiratory rate of 16, tidal volume, 6 mL/kg/IBW, PEEP 5-cm H₂O with auto PEEP of 3-cm H₂O, and fraction of inspired ABG O₂ (FiO₂) 0.35 with PAP of 46-cm H₂O and plateau pressure of 17-cm H₂O. The ABG was pH 7.32, PaCO₂ 65 mm Hg, and PaO₂ 74 mm Hg. However, on hospital day 5, she developed worsening PAP 60 to 77-cm H₂O, plateau pressures 17-cm H₂O, and a dynamic PEEP 16-cm H₂O and was unresponsive to ventilator maneuvers to lower airway pressures and improve ventilation.

The patient had been receiving continuous albuterol and ipratropium nebulizer treatments. Ketamine infusion was considered fraught with potential for a dissociative reaction due to the patient’s significant PTSD. The patient’s family requested avoidance of ketamine infusion since the patient was paralyzed and psychiatric effects could not be monitored. Heliox 80/20 mixture was considered; however, it is incompatible with the ventilator that was being used since it could not account for the density of the helium gas flow in the tidal volumes. Extracorporeal membrane oxygenation (ECMO) was not available at our facility, and the patient was not a candidate for the regional ECMO center.

On hospital day 8, the patient developed worsening respiratory acidosis. The patient’s PAP increased to > 77-cm H₂O, and her ABG revealed pH 7.22, PaCO₂ 90 mm Hg, and PaO₂ 77 mm Hg with FiO₂ 0.4. A chest X-ray demonstrated a new left lower lobe infiltrate. Fiber optic bronchoscopy was notable for scattered thick secretions throughout both lungs without obstructing mucus plug. Removal of airway secretions did not improve airway pressures or dynamic hyperinflation.

After consultation and discussion with the chief of anesthesia, the patient was placed on an anesthesia ventilator and started on sevoflurane 1.5% in the ICU. Anesthesiology was available 24 hours a day, and the anesthesiologist rounded with the intensivist frequently for this patient. The anesthesia technician worked closely with respiratory therapy regarding ventilator setting and changing the anesthesia gas scavenging charcoal canister. Within 4 hours, her gas exchange normalized (Table). The patient’s ABG was pH 7.44, PaCO₂ 52 mm Hg, and PaO₂ 69 mm Hg on FiO₂ 0.4. On volume cycled ventilation with a rate of 12, flow rate of 40 L/min, and tidal volume 6 mL/kg/IBW, the PAP decreased to 41-cm H₂O.

Within 24 hours bronchospasm improved as evidenced by decreased airway pressures, resolution of wheezing, and decreased CO₂ retention. The sevoflurane was easily weaned over the next 48 hours by decreasing the dose by 25% every 12-hour shift without rebound bronchospasm. Airway pressures and ABGs were frequently monitored during the weaning process. The patient resumed conventional mechanical ventilation, cisatracurium was discontinued, and she underwent a percutaneous tracheostomy for critical illness polymyopathy. Her respiratory muscle strength recovered more robustly than anticipated. Prior to discharge to a skilled nursing facility for continued rehabilitation, she was removed from mechanical ventilation and decannulated.

Discussion

This case illustrates the successful treatment of a patient with extreme status asthmaticus given inhalational anesthesia as supportive care while the bronchospasm and status asthmaticus abated. This is an unusual treatment in an ominous situation. Inhalational anesthetics are potent bronchodilators and have been successfully used in the management of status asthmaticus refractory to conventional therapy.⁴ Inhalational anesthetics have been shown to decrease airway resistance, dynamic hyperinflation, and intrinsic PEEP.⁵ These agents result in rapid bronchodilation by relaxing the smooth muscle and are associated with early liberation from mechanical ventilation.^5,6 Although there are no guidelines regarding which inhalational agent is best, specific dosing, duration, or titration, case reports in the literature regarding the successful use of inhalational agents in life-threatening status asthmaticus exist.^2,5,7

Caveats regarding the use of inhalational anesthetics in status asthmaticus include proarrhythmias, severe hepatic and renal toxicity. Although isoflurane is less likely to cause arrhythmia, both isoflurane and sevoflurane can cause dose-dependent hypotension by peripheral vasodilatation.⁷ Ourpatient did not manifest any adverse effects.

Additional challenges regarding the use of inhalational anesthetics for status asthmaticus include differences in ventilators and occupational hazards.⁸ Anesthesia or operating room ventilators differ from ICU ventilators in flow and pressure capabilities.⁷ The anesthesia ventilator is not capable of generating inspiratory pressures sufficient to ventilate patients with severely elevated airway resistance. Thus, the decrease inspiratory flow that occurs with increasing airway pressure limits the tidal volume delivered and consequently the minute volume. Although newer anesthesia ventilators have increased flow capabilities, they require a fully trained staff.⁸

Potential occupational exposure to these volatile anesthetic gases occurs as patients being treated may exhale considerable amounts of volatile anesthetics.⁸ An anesthesia gas scavenging device, such as a charcoal canister, must be attached to the ventilator to capture the exhaled anesthetic gases and should be changed every 12 hours.⁸ Finally, there is a potential for rebound bronchospasm as the anesthetic agent is tapered.^6,7,9-11

Conclusion

Inhalational anesthetics are an option as rescue therapy for severe life-threatening asthma when all other therapies have failed. Use of inhalational anesthetics in status asthmaticus consists of case reports of which half are in children.^2,5,7 Our patient contributes to the literature of case reports regarding using sevoflurane in refractory status asthmaticus. A decision to choose them must be a collaborative team approach with anesthesiology, pulmonary/critical care medicine, respiratory therapy, and ICU nurses, and the risks and benefits should be discussed with decision-making family members. Since there are no specific guidelines for the use of inhalational agents in status asthmaticus, close attention to inspiratory flows, gas scavenging devices, and clinical response is required. Additionally, the team must be comfortable with the plan to use an anesthesia ventilator and trained on its limitations.

Asthma attacks account for 1.8 million emergency department (ED) visits each year in the US and for 10 deaths daily.¹ Management of asthma attacks includes administration of inhaled ß₂ adrenergic agonists, inhaled anticholinergic agents, IV magnesium sulfate, and corticosteroids.² Status asthmaticus is an intense acute exacerbation of asthma that does not respond to repeated treatments of bronchodilators and corticosteroids.³ It is a medical emergency requiring immediate recognition and treatment. The decision to intubate a patient with status asthmaticus is a clinical decision based on work of breathing, respiratory acidosis, and failure to respond to medical interventions.

In refractory cases of status asthmaticus, intubation and mechanical ventilation are undertaken to provide oxygenation and ventilation until the bronchospasm resolves. However, mechanical ventilation is associated with significant risks, including high end-inspiratory pressures, barotrauma, and volutrauma.⁴ Rescue therapies include muscle relaxation, infusion of ketamine (central acting nonopioid analgesic with bronchodilatory properties), heliox, and general anesthesia.^2,4 We report a case of a patient with life-threatening asthma and status asthmaticus treated with sevoflurane general anesthesia.

Case Presentation

A 55-year-old woman whose medical history was notable for asthma, psoriasis, hypothyroidism, tobacco, and alcohol abuse, and posttraumatic stress disorder (PTSD) presented to the ED. The patient had rarely sought medical attention and had no prior ED visits or hospitalizations in the electronic health record. Her home regimen included an albuterol inhaler used as needed. Her family reported that they had found her in distress in bed in a tripod position, unable to speak and struggling to breath.

Emergency medical services found the patient cyanotic, apneic, and pulseless. She received cardiopulmonary resuscitation for 30 seconds and 1-mg IV epinephrine, and spontaneous circulation returned. The patient arrived in the ED with an oral airway in place receiving bag valve mask ventilation. The patient expelled the oral airway. She was unable to speak due to dyspnea, exhibited persistent cyanosis, fatigue due to work of breathing, and failed to respond to nebulized albuterol/ipratropium bromide, IV methylprednisolone, and magnesium sulfate. The patient met criteria for acute severe asthma, or status asthmaticus. Thus, the patient received rapid sequence induction with rocuronium and ketamine and was intubated.

According to her family, the patient had no previous intensive care unit (ICU) admissions or prior intubations. Her only asthma medication was an albuterol inhaler as needed. The patient worked as a supervisor at a window blind manufacturing company. She lived alone, smoked 2 packs of cigarettes a day for more than 30 years, had no pets, drank unknown quantities of beer, wine, and hard liquor daily, and had smoked marijuana for several years.

The patient’s physical examination was notable for diffuse expiratory wheezes. Laboratory analysis revealed white blood cell count of 13.7 k/mcL, sodium 140 mmol/L, potassium 4.9 mmol/L, chloride 105 mmol/L, CO₂ 17 mmol/L, creatinine 0.98 mg/dL, troponin 0.03 ng/mL, lactate 7.2 mmol/L. Her chest X-ray showed hyperinflation but no focal opacities, pneumothorax, or pulmonary edema. Her endotracheal tube was in good position (Figure 1). A computed tomography pulmonary angiogram showed no pulmonary embolus or emphysema. There were atelectatic changes in the dependent portion of the right lower lobe, central bronchial wall thickening, and no stigmata of air trapping (Figure 2). An echocardiogram revealed a left ventricular ejection fraction of 45%, normal right ventricle and right ventricular size and function with an estimated right ventricular systolic pressure of 40 mm Hg.

The patient was admitted to the ICU and started on continuous infusion cisatracurium for paralysis and deep sedation to improve ventilatory synchrony and decrease auto positive end-expiratory pressure (PEEP). Mechanical ventilation was initiated with volume-cycled assist control ventilation, 6 mL/kg/ideal body weight (IBW) at 5-cm H₂O PEEP, and 1 minute ventilation of 10 liters. The patient had severe air trapping and high airway pressures. The dynamic PEEP was 22-cm H₂O (normal PEEP of 5-cm H₂O), peak airway pressure (PAP) 41-cm H₂O, and plateau pressure 31-cm H₂O. In addition, the arterial blood gas (ABG) showed severe hypercapnic respiratory acidosis without significant hypoxemia with pH 7.15, PaCO₂ 90 mm Hg, and PaO₂ 150 mm Hg.

Pressure controlled ventilation was attempted unsuccessfully due to high airway resistance. Ultimately, the patient was set on volume control with low tidal volume, 6 mL/kg/IBW, high flow 90 L/min, PEEP 0 cm of H₂O, and a low respiratory rate of 10 to achieve an inspiratory to expiratory (I:E) ratio of 1:7. Managing the ventilator to avoid dynamic hyperinflation and auto-PEEP, she remained relatively stable and improved.

By day 4 the patient’s ventilator was set to volume assist control with respiratory rate of 16, tidal volume, 6 mL/kg/IBW, PEEP 5-cm H₂O with auto PEEP of 3-cm H₂O, and fraction of inspired ABG O₂ (FiO₂) 0.35 with PAP of 46-cm H₂O and plateau pressure of 17-cm H₂O. The ABG was pH 7.32, PaCO₂ 65 mm Hg, and PaO₂ 74 mm Hg. However, on hospital day 5, she developed worsening PAP 60 to 77-cm H₂O, plateau pressures 17-cm H₂O, and a dynamic PEEP 16-cm H₂O and was unresponsive to ventilator maneuvers to lower airway pressures and improve ventilation.

The patient had been receiving continuous albuterol and ipratropium nebulizer treatments. Ketamine infusion was considered fraught with potential for a dissociative reaction due to the patient’s significant PTSD. The patient’s family requested avoidance of ketamine infusion since the patient was paralyzed and psychiatric effects could not be monitored. Heliox 80/20 mixture was considered; however, it is incompatible with the ventilator that was being used since it could not account for the density of the helium gas flow in the tidal volumes. Extracorporeal membrane oxygenation (ECMO) was not available at our facility, and the patient was not a candidate for the regional ECMO center.

On hospital day 8, the patient developed worsening respiratory acidosis. The patient’s PAP increased to > 77-cm H₂O, and her ABG revealed pH 7.22, PaCO₂ 90 mm Hg, and PaO₂ 77 mm Hg with FiO₂ 0.4. A chest X-ray demonstrated a new left lower lobe infiltrate. Fiber optic bronchoscopy was notable for scattered thick secretions throughout both lungs without obstructing mucus plug. Removal of airway secretions did not improve airway pressures or dynamic hyperinflation.

After consultation and discussion with the chief of anesthesia, the patient was placed on an anesthesia ventilator and started on sevoflurane 1.5% in the ICU. Anesthesiology was available 24 hours a day, and the anesthesiologist rounded with the intensivist frequently for this patient. The anesthesia technician worked closely with respiratory therapy regarding ventilator setting and changing the anesthesia gas scavenging charcoal canister. Within 4 hours, her gas exchange normalized (Table). The patient’s ABG was pH 7.44, PaCO₂ 52 mm Hg, and PaO₂ 69 mm Hg on FiO₂ 0.4. On volume cycled ventilation with a rate of 12, flow rate of 40 L/min, and tidal volume 6 mL/kg/IBW, the PAP decreased to 41-cm H₂O.

Within 24 hours bronchospasm improved as evidenced by decreased airway pressures, resolution of wheezing, and decreased CO₂ retention. The sevoflurane was easily weaned over the next 48 hours by decreasing the dose by 25% every 12-hour shift without rebound bronchospasm. Airway pressures and ABGs were frequently monitored during the weaning process. The patient resumed conventional mechanical ventilation, cisatracurium was discontinued, and she underwent a percutaneous tracheostomy for critical illness polymyopathy. Her respiratory muscle strength recovered more robustly than anticipated. Prior to discharge to a skilled nursing facility for continued rehabilitation, she was removed from mechanical ventilation and decannulated.

Discussion

This case illustrates the successful treatment of a patient with extreme status asthmaticus given inhalational anesthesia as supportive care while the bronchospasm and status asthmaticus abated. This is an unusual treatment in an ominous situation. Inhalational anesthetics are potent bronchodilators and have been successfully used in the management of status asthmaticus refractory to conventional therapy.⁴ Inhalational anesthetics have been shown to decrease airway resistance, dynamic hyperinflation, and intrinsic PEEP.⁵ These agents result in rapid bronchodilation by relaxing the smooth muscle and are associated with early liberation from mechanical ventilation.^5,6 Although there are no guidelines regarding which inhalational agent is best, specific dosing, duration, or titration, case reports in the literature regarding the successful use of inhalational agents in life-threatening status asthmaticus exist.^2,5,7

Caveats regarding the use of inhalational anesthetics in status asthmaticus include proarrhythmias, severe hepatic and renal toxicity. Although isoflurane is less likely to cause arrhythmia, both isoflurane and sevoflurane can cause dose-dependent hypotension by peripheral vasodilatation.⁷ Ourpatient did not manifest any adverse effects.

Additional challenges regarding the use of inhalational anesthetics for status asthmaticus include differences in ventilators and occupational hazards.⁸ Anesthesia or operating room ventilators differ from ICU ventilators in flow and pressure capabilities.⁷ The anesthesia ventilator is not capable of generating inspiratory pressures sufficient to ventilate patients with severely elevated airway resistance. Thus, the decrease inspiratory flow that occurs with increasing airway pressure limits the tidal volume delivered and consequently the minute volume. Although newer anesthesia ventilators have increased flow capabilities, they require a fully trained staff.⁸

Potential occupational exposure to these volatile anesthetic gases occurs as patients being treated may exhale considerable amounts of volatile anesthetics.⁸ An anesthesia gas scavenging device, such as a charcoal canister, must be attached to the ventilator to capture the exhaled anesthetic gases and should be changed every 12 hours.⁸ Finally, there is a potential for rebound bronchospasm as the anesthetic agent is tapered.^6,7,9-11

Conclusion

Inhalational anesthetics are an option as rescue therapy for severe life-threatening asthma when all other therapies have failed. Use of inhalational anesthetics in status asthmaticus consists of case reports of which half are in children.^2,5,7 Our patient contributes to the literature of case reports regarding using sevoflurane in refractory status asthmaticus. A decision to choose them must be a collaborative team approach with anesthesiology, pulmonary/critical care medicine, respiratory therapy, and ICU nurses, and the risks and benefits should be discussed with decision-making family members. Since there are no specific guidelines for the use of inhalational agents in status asthmaticus, close attention to inspiratory flows, gas scavenging devices, and clinical response is required. Additionally, the team must be comfortable with the plan to use an anesthesia ventilator and trained on its limitations.

Asthma attacks account for 1.8 million emergency department (ED) visits each year in the US and for 10 deaths daily.¹ Management of asthma attacks includes administration of inhaled ß₂ adrenergic agonists, inhaled anticholinergic agents, IV magnesium sulfate, and corticosteroids.² Status asthmaticus is an intense acute exacerbation of asthma that does not respond to repeated treatments of bronchodilators and corticosteroids.³ It is a medical emergency requiring immediate recognition and treatment. The decision to intubate a patient with status asthmaticus is a clinical decision based on work of breathing, respiratory acidosis, and failure to respond to medical interventions.

In refractory cases of status asthmaticus, intubation and mechanical ventilation are undertaken to provide oxygenation and ventilation until the bronchospasm resolves. However, mechanical ventilation is associated with significant risks, including high end-inspiratory pressures, barotrauma, and volutrauma.⁴ Rescue therapies include muscle relaxation, infusion of ketamine (central acting nonopioid analgesic with bronchodilatory properties), heliox, and general anesthesia.^2,4 We report a case of a patient with life-threatening asthma and status asthmaticus treated with sevoflurane general anesthesia.

Case Presentation

A 55-year-old woman whose medical history was notable for asthma, psoriasis, hypothyroidism, tobacco, and alcohol abuse, and posttraumatic stress disorder (PTSD) presented to the ED. The patient had rarely sought medical attention and had no prior ED visits or hospitalizations in the electronic health record. Her home regimen included an albuterol inhaler used as needed. Her family reported that they had found her in distress in bed in a tripod position, unable to speak and struggling to breath.

Emergency medical services found the patient cyanotic, apneic, and pulseless. She received cardiopulmonary resuscitation for 30 seconds and 1-mg IV epinephrine, and spontaneous circulation returned. The patient arrived in the ED with an oral airway in place receiving bag valve mask ventilation. The patient expelled the oral airway. She was unable to speak due to dyspnea, exhibited persistent cyanosis, fatigue due to work of breathing, and failed to respond to nebulized albuterol/ipratropium bromide, IV methylprednisolone, and magnesium sulfate. The patient met criteria for acute severe asthma, or status asthmaticus. Thus, the patient received rapid sequence induction with rocuronium and ketamine and was intubated.

According to her family, the patient had no previous intensive care unit (ICU) admissions or prior intubations. Her only asthma medication was an albuterol inhaler as needed. The patient worked as a supervisor at a window blind manufacturing company. She lived alone, smoked 2 packs of cigarettes a day for more than 30 years, had no pets, drank unknown quantities of beer, wine, and hard liquor daily, and had smoked marijuana for several years.

The patient’s physical examination was notable for diffuse expiratory wheezes. Laboratory analysis revealed white blood cell count of 13.7 k/mcL, sodium 140 mmol/L, potassium 4.9 mmol/L, chloride 105 mmol/L, CO₂ 17 mmol/L, creatinine 0.98 mg/dL, troponin 0.03 ng/mL, lactate 7.2 mmol/L. Her chest X-ray showed hyperinflation but no focal opacities, pneumothorax, or pulmonary edema. Her endotracheal tube was in good position (Figure 1). A computed tomography pulmonary angiogram showed no pulmonary embolus or emphysema. There were atelectatic changes in the dependent portion of the right lower lobe, central bronchial wall thickening, and no stigmata of air trapping (Figure 2). An echocardiogram revealed a left ventricular ejection fraction of 45%, normal right ventricle and right ventricular size and function with an estimated right ventricular systolic pressure of 40 mm Hg.

The patient was admitted to the ICU and started on continuous infusion cisatracurium for paralysis and deep sedation to improve ventilatory synchrony and decrease auto positive end-expiratory pressure (PEEP). Mechanical ventilation was initiated with volume-cycled assist control ventilation, 6 mL/kg/ideal body weight (IBW) at 5-cm H₂O PEEP, and 1 minute ventilation of 10 liters. The patient had severe air trapping and high airway pressures. The dynamic PEEP was 22-cm H₂O (normal PEEP of 5-cm H₂O), peak airway pressure (PAP) 41-cm H₂O, and plateau pressure 31-cm H₂O. In addition, the arterial blood gas (ABG) showed severe hypercapnic respiratory acidosis without significant hypoxemia with pH 7.15, PaCO₂ 90 mm Hg, and PaO₂ 150 mm Hg.

Pressure controlled ventilation was attempted unsuccessfully due to high airway resistance. Ultimately, the patient was set on volume control with low tidal volume, 6 mL/kg/IBW, high flow 90 L/min, PEEP 0 cm of H₂O, and a low respiratory rate of 10 to achieve an inspiratory to expiratory (I:E) ratio of 1:7. Managing the ventilator to avoid dynamic hyperinflation and auto-PEEP, she remained relatively stable and improved.

By day 4 the patient’s ventilator was set to volume assist control with respiratory rate of 16, tidal volume, 6 mL/kg/IBW, PEEP 5-cm H₂O with auto PEEP of 3-cm H₂O, and fraction of inspired ABG O₂ (FiO₂) 0.35 with PAP of 46-cm H₂O and plateau pressure of 17-cm H₂O. The ABG was pH 7.32, PaCO₂ 65 mm Hg, and PaO₂ 74 mm Hg. However, on hospital day 5, she developed worsening PAP 60 to 77-cm H₂O, plateau pressures 17-cm H₂O, and a dynamic PEEP 16-cm H₂O and was unresponsive to ventilator maneuvers to lower airway pressures and improve ventilation.

The patient had been receiving continuous albuterol and ipratropium nebulizer treatments. Ketamine infusion was considered fraught with potential for a dissociative reaction due to the patient’s significant PTSD. The patient’s family requested avoidance of ketamine infusion since the patient was paralyzed and psychiatric effects could not be monitored. Heliox 80/20 mixture was considered; however, it is incompatible with the ventilator that was being used since it could not account for the density of the helium gas flow in the tidal volumes. Extracorporeal membrane oxygenation (ECMO) was not available at our facility, and the patient was not a candidate for the regional ECMO center.

On hospital day 8, the patient developed worsening respiratory acidosis. The patient’s PAP increased to > 77-cm H₂O, and her ABG revealed pH 7.22, PaCO₂ 90 mm Hg, and PaO₂ 77 mm Hg with FiO₂ 0.4. A chest X-ray demonstrated a new left lower lobe infiltrate. Fiber optic bronchoscopy was notable for scattered thick secretions throughout both lungs without obstructing mucus plug. Removal of airway secretions did not improve airway pressures or dynamic hyperinflation.

After consultation and discussion with the chief of anesthesia, the patient was placed on an anesthesia ventilator and started on sevoflurane 1.5% in the ICU. Anesthesiology was available 24 hours a day, and the anesthesiologist rounded with the intensivist frequently for this patient. The anesthesia technician worked closely with respiratory therapy regarding ventilator setting and changing the anesthesia gas scavenging charcoal canister. Within 4 hours, her gas exchange normalized (Table). The patient’s ABG was pH 7.44, PaCO₂ 52 mm Hg, and PaO₂ 69 mm Hg on FiO₂ 0.4. On volume cycled ventilation with a rate of 12, flow rate of 40 L/min, and tidal volume 6 mL/kg/IBW, the PAP decreased to 41-cm H₂O.

Within 24 hours bronchospasm improved as evidenced by decreased airway pressures, resolution of wheezing, and decreased CO₂ retention. The sevoflurane was easily weaned over the next 48 hours by decreasing the dose by 25% every 12-hour shift without rebound bronchospasm. Airway pressures and ABGs were frequently monitored during the weaning process. The patient resumed conventional mechanical ventilation, cisatracurium was discontinued, and she underwent a percutaneous tracheostomy for critical illness polymyopathy. Her respiratory muscle strength recovered more robustly than anticipated. Prior to discharge to a skilled nursing facility for continued rehabilitation, she was removed from mechanical ventilation and decannulated.

Discussion

This case illustrates the successful treatment of a patient with extreme status asthmaticus given inhalational anesthesia as supportive care while the bronchospasm and status asthmaticus abated. This is an unusual treatment in an ominous situation. Inhalational anesthetics are potent bronchodilators and have been successfully used in the management of status asthmaticus refractory to conventional therapy.⁴ Inhalational anesthetics have been shown to decrease airway resistance, dynamic hyperinflation, and intrinsic PEEP.⁵ These agents result in rapid bronchodilation by relaxing the smooth muscle and are associated with early liberation from mechanical ventilation.^5,6 Although there are no guidelines regarding which inhalational agent is best, specific dosing, duration, or titration, case reports in the literature regarding the successful use of inhalational agents in life-threatening status asthmaticus exist.^2,5,7

Caveats regarding the use of inhalational anesthetics in status asthmaticus include proarrhythmias, severe hepatic and renal toxicity. Although isoflurane is less likely to cause arrhythmia, both isoflurane and sevoflurane can cause dose-dependent hypotension by peripheral vasodilatation.⁷ Ourpatient did not manifest any adverse effects.

Additional challenges regarding the use of inhalational anesthetics for status asthmaticus include differences in ventilators and occupational hazards.⁸ Anesthesia or operating room ventilators differ from ICU ventilators in flow and pressure capabilities.⁷ The anesthesia ventilator is not capable of generating inspiratory pressures sufficient to ventilate patients with severely elevated airway resistance. Thus, the decrease inspiratory flow that occurs with increasing airway pressure limits the tidal volume delivered and consequently the minute volume. Although newer anesthesia ventilators have increased flow capabilities, they require a fully trained staff.⁸

Potential occupational exposure to these volatile anesthetic gases occurs as patients being treated may exhale considerable amounts of volatile anesthetics.⁸ An anesthesia gas scavenging device, such as a charcoal canister, must be attached to the ventilator to capture the exhaled anesthetic gases and should be changed every 12 hours.⁸ Finally, there is a potential for rebound bronchospasm as the anesthetic agent is tapered.^6,7,9-11

Conclusion

Inhalational anesthetics are an option as rescue therapy for severe life-threatening asthma when all other therapies have failed. Use of inhalational anesthetics in status asthmaticus consists of case reports of which half are in children.^2,5,7 Our patient contributes to the literature of case reports regarding using sevoflurane in refractory status asthmaticus. A decision to choose them must be a collaborative team approach with anesthesiology, pulmonary/critical care medicine, respiratory therapy, and ICU nurses, and the risks and benefits should be discussed with decision-making family members. Since there are no specific guidelines for the use of inhalational agents in status asthmaticus, close attention to inspiratory flows, gas scavenging devices, and clinical response is required. Additionally, the team must be comfortable with the plan to use an anesthesia ventilator and trained on its limitations.

Although solar radiation is the most commonly recognized source of UV radiation (UVR), occupational exposures can contribute due to the intensity and chronicity of exposure. Arc welding is a process whereby metal is fused together by heat produced from an electric arc. The electric arc that forms between the electrode and the base metal emits radiation in the full UV spectrum including UVA (400–315 nm), UVB (315–290 nm), and UVC (290–100 nm) wavelengths. Welders, therefore, have an increased risk for broad-spectrum, intense exposure to UVR, which may play a notable role in UV-related skin disease without proper protection. We report 3 welders with skin disease attributed to occupational exposure to UVR.

Case Reports

Patient 1
A 41-year-old man presented for evaluation of treatment-resistant cutaneous lupus. During the 10-year disease course, the patient was treated by both dermatologists and rheumatologists with frequent exacerbations and poor disease control. At the time of presentation, treatment with hydroxychloroquine 200 mg twice daily, azathioprine 50 mg twice daily, intramuscular methylprednisolone acetateinjectable suspension 40 mg, and prednisone 20 mg daily was failing. Physical examination revealed polycyclic erythematous plaques typical of subacute cutaneous lupus erythematosus. A skin biopsy confirmed the diagnosis. Upon further discussion of exacerbating risk factors, the patient noted UVR exposure while working as a welder. Although he had been previously told to avoid sunlight, he did not realize that this recommendation included all forms of UV light. Once this work exposure was eliminated, he was restarted on hydroxychloroquine 200 mg twice daily and topical steroids, and he responded with complete and sustained clearance of disease. When he returned to welding, utilization of sunscreen and sun-protective clothing enabled him to maintain control of his subacute cutaneous lupus erythematosus on oral hydroxychloroquine 200 mg twice daily and topical steroids.

Patient 2
A 55-year-old man presented with numerous actinic keratoses and persistent erythema in a well-demarcated area involving the forehead, temples, and lateral cheeks but sparing the periorbital area. The patient also experienced UVR exposure from welding (up to 4 to 5 times per week during his career spanning more than 20 years). He cited frequent burns in areas where his protective equipment did not cover his skin. He also reported that he often forgoes wearing protective equipment, even though it is available, and only uses safety goggles due to the extreme heat of the working environment as well as the awkwardness of wearing full protective gear while performing certain aspects of the job.

Patient 3
A 63-year-old man presented with a growth on the left side of the upper forehead. A biopsy revealed a squamous cell carcinoma, keratoacanthoma type. He worked as a welder for 40 years until retiring 1 year prior to presentation. He welded daily and always wore a tall face shield. Although the face shield covered most of his face, the scalp and some parts of the upper face were not well protected. In addition to the keratoacanthoma, which presented just outside of the area protected by the face shield, the patient had numerous actinic keratoses on the scalp.

Comment

Welding and UVR Exposure
Arc welders endure large amounts of UVR exposure, which is substantial enough to have notable health effects. The duration of exposure, electrical current used, angle of exposure, amount of ventilation, and the distance from the welding arc play a role in overall UVR exposure.^1,2 Maximum permissible exposure (MPE) limits to UVR have been set by the International Commission on Non-Ionizing Radiation Protection and the National Institute for Occupational Safety and Health.^3,4 The quantity of radiation produced by the arc allows for an exposure time of only a few seconds to minutes before surpassing MPE to UV light.^1,5 Welders are exposed to total-body UVR doses up to 3000 times the MPE, and mean cumulative exposure calculated over an 8-hour workday can reach 9795 mJ/cm².⁶

Workers in close proximity to welders also receive large UVR doses and may not be aware of its hazardous effects. Nearby nonwelders can be exposed to 13 times the MPE of UVR.⁶ At distances up to 10 m from the arc, the irradiance is large enough to reach MPE to UVR in less than 3 hours.¹

Skin and Eye Damage From Welding
Exposure to UVR produced by the welding arc may lead to acute skin or eye reactions, chronic skin or eye disorders, or exacerbation of photosensitive diseases. Common acute problems are photokeratoconjunctivitis (welder’s flash) and skin erythema.^7,8

Actinic elastosis, actinic keratoses, ocular melanoma, and photosensitive diseases represent a spectrum of disorders that can present from chronic UV exposure in welders. In a study by Emmett et al⁷ of 152 welders and 58 controls, actinic elastosis was found to be more frequent in welders than controls. Cases of basal cell carcinoma and squamous cell carcinoma also have been reported in welders.^9,10 However, in the study by Emmett et al,⁷ a statistically significant correlation between welding and skin cancer was not documented. There were limitations in the study, such as small sample size and a young average age of welders.⁷ Future studies may be needed to further clarify the risk for skin cancer in welders.

Although there is no clear association with skin cancer, an increased risk of ocular melanoma in welders is more clearly established. A meta-analysis of 5 studies found that welding was a significant risk factor for ocular melanoma, with an odds ratio of 2.05 (95% confidence interval, 1.20-3.51).¹¹ Other reported eye damage from chronic UVR exposure includes cataracts, chronic conjunctivitis, and retinal damage.^12,13

Case reports of the following photosensitive diseases have been reported to be exacerbated or caused by UV light exposure in welders: discoid lupus erythematosus¹⁴; photodermatitis¹⁵; broadband photosensitivity with decreased minimal erythema dose to UVA, UVB, and UVC¹⁶; UVC-exacerbated atopic dermatitis¹⁷; polymorphous light eruption–like skin eruption¹⁸; and UVA-induced photoallergy to hydrochlorothiazide and ramipril.¹⁹

Prevention of Occupational Exposure to UVR
Occupational Safety and Health Administration guidelines protect workers from excessive exposure to UVR with personal protective equipment (PPE). In addition to UVR protection, PPE needs to protect welders from other risks including trauma from welding debris (slag), fires, electrical burns, and fumes. Online resources from the National Ag Safety Database,²⁰ the American Welding Society,²¹ and Occupational Safety and Health Administration^22,23 are available. These resources advise welders to work in ventilated areas with respirators specific for the metal being welded and to wear clothing and gloves that are not only fire retardant but also UV resistant.^20-23 Additional PPE should protect the head, face, and eyes.

Unfortunately, even workers well trained in prevention guidelines may not adequately protect themselves. Some welders forego PPE due to heat, thus exposing themselves to UVR damage in areas that are normally covered. Welders also may forego equipment when working on jobs requiring more detailed welds where clothing, masks, and glasses may be overly bulky and inhibit the worker’s precision. Nontraditional welders, such as artisans or handymen, may not have workplace safety education to be aware of UVR emitted from welding and may not have readily available PPE.

The Figure portrays an amateur welder working without full PPE. Although he is wearing a face shield, he is not wearing fire-retardant clothing, lacks full protective garments, and has no ventilation system.

Conclusion

It is important to recognize welding as an occupation with notable exposure to UVR. Personal protective equipment should be the mainstay of prevention. Sunscreen is a useful adjunct but does not cover UVC that is emitted in the welding arc. Screens and welding blankets can be placed around welders to contain UVR and limit nonwelder exposure. Although UVR hazards should be regulated in the workplace as part of regular safety reviews, the clinician can play a role in recognizing this source of UVR in skin disease and in encouraging the use of PPE.

Although solar radiation is the most commonly recognized source of UV radiation (UVR), occupational exposures can contribute due to the intensity and chronicity of exposure. Arc welding is a process whereby metal is fused together by heat produced from an electric arc. The electric arc that forms between the electrode and the base metal emits radiation in the full UV spectrum including UVA (400–315 nm), UVB (315–290 nm), and UVC (290–100 nm) wavelengths. Welders, therefore, have an increased risk for broad-spectrum, intense exposure to UVR, which may play a notable role in UV-related skin disease without proper protection. We report 3 welders with skin disease attributed to occupational exposure to UVR.

Case Reports

Patient 1
A 41-year-old man presented for evaluation of treatment-resistant cutaneous lupus. During the 10-year disease course, the patient was treated by both dermatologists and rheumatologists with frequent exacerbations and poor disease control. At the time of presentation, treatment with hydroxychloroquine 200 mg twice daily, azathioprine 50 mg twice daily, intramuscular methylprednisolone acetateinjectable suspension 40 mg, and prednisone 20 mg daily was failing. Physical examination revealed polycyclic erythematous plaques typical of subacute cutaneous lupus erythematosus. A skin biopsy confirmed the diagnosis. Upon further discussion of exacerbating risk factors, the patient noted UVR exposure while working as a welder. Although he had been previously told to avoid sunlight, he did not realize that this recommendation included all forms of UV light. Once this work exposure was eliminated, he was restarted on hydroxychloroquine 200 mg twice daily and topical steroids, and he responded with complete and sustained clearance of disease. When he returned to welding, utilization of sunscreen and sun-protective clothing enabled him to maintain control of his subacute cutaneous lupus erythematosus on oral hydroxychloroquine 200 mg twice daily and topical steroids.

Patient 2
A 55-year-old man presented with numerous actinic keratoses and persistent erythema in a well-demarcated area involving the forehead, temples, and lateral cheeks but sparing the periorbital area. The patient also experienced UVR exposure from welding (up to 4 to 5 times per week during his career spanning more than 20 years). He cited frequent burns in areas where his protective equipment did not cover his skin. He also reported that he often forgoes wearing protective equipment, even though it is available, and only uses safety goggles due to the extreme heat of the working environment as well as the awkwardness of wearing full protective gear while performing certain aspects of the job.

Patient 3
A 63-year-old man presented with a growth on the left side of the upper forehead. A biopsy revealed a squamous cell carcinoma, keratoacanthoma type. He worked as a welder for 40 years until retiring 1 year prior to presentation. He welded daily and always wore a tall face shield. Although the face shield covered most of his face, the scalp and some parts of the upper face were not well protected. In addition to the keratoacanthoma, which presented just outside of the area protected by the face shield, the patient had numerous actinic keratoses on the scalp.

Comment

Welding and UVR Exposure
Arc welders endure large amounts of UVR exposure, which is substantial enough to have notable health effects. The duration of exposure, electrical current used, angle of exposure, amount of ventilation, and the distance from the welding arc play a role in overall UVR exposure.^1,2 Maximum permissible exposure (MPE) limits to UVR have been set by the International Commission on Non-Ionizing Radiation Protection and the National Institute for Occupational Safety and Health.^3,4 The quantity of radiation produced by the arc allows for an exposure time of only a few seconds to minutes before surpassing MPE to UV light.^1,5 Welders are exposed to total-body UVR doses up to 3000 times the MPE, and mean cumulative exposure calculated over an 8-hour workday can reach 9795 mJ/cm².⁶

Workers in close proximity to welders also receive large UVR doses and may not be aware of its hazardous effects. Nearby nonwelders can be exposed to 13 times the MPE of UVR.⁶ At distances up to 10 m from the arc, the irradiance is large enough to reach MPE to UVR in less than 3 hours.¹

Skin and Eye Damage From Welding
Exposure to UVR produced by the welding arc may lead to acute skin or eye reactions, chronic skin or eye disorders, or exacerbation of photosensitive diseases. Common acute problems are photokeratoconjunctivitis (welder’s flash) and skin erythema.^7,8

Actinic elastosis, actinic keratoses, ocular melanoma, and photosensitive diseases represent a spectrum of disorders that can present from chronic UV exposure in welders. In a study by Emmett et al⁷ of 152 welders and 58 controls, actinic elastosis was found to be more frequent in welders than controls. Cases of basal cell carcinoma and squamous cell carcinoma also have been reported in welders.^9,10 However, in the study by Emmett et al,⁷ a statistically significant correlation between welding and skin cancer was not documented. There were limitations in the study, such as small sample size and a young average age of welders.⁷ Future studies may be needed to further clarify the risk for skin cancer in welders.

Although there is no clear association with skin cancer, an increased risk of ocular melanoma in welders is more clearly established. A meta-analysis of 5 studies found that welding was a significant risk factor for ocular melanoma, with an odds ratio of 2.05 (95% confidence interval, 1.20-3.51).¹¹ Other reported eye damage from chronic UVR exposure includes cataracts, chronic conjunctivitis, and retinal damage.^12,13

Case reports of the following photosensitive diseases have been reported to be exacerbated or caused by UV light exposure in welders: discoid lupus erythematosus¹⁴; photodermatitis¹⁵; broadband photosensitivity with decreased minimal erythema dose to UVA, UVB, and UVC¹⁶; UVC-exacerbated atopic dermatitis¹⁷; polymorphous light eruption–like skin eruption¹⁸; and UVA-induced photoallergy to hydrochlorothiazide and ramipril.¹⁹

Prevention of Occupational Exposure to UVR
Occupational Safety and Health Administration guidelines protect workers from excessive exposure to UVR with personal protective equipment (PPE). In addition to UVR protection, PPE needs to protect welders from other risks including trauma from welding debris (slag), fires, electrical burns, and fumes. Online resources from the National Ag Safety Database,²⁰ the American Welding Society,²¹ and Occupational Safety and Health Administration^22,23 are available. These resources advise welders to work in ventilated areas with respirators specific for the metal being welded and to wear clothing and gloves that are not only fire retardant but also UV resistant.^20-23 Additional PPE should protect the head, face, and eyes.

Unfortunately, even workers well trained in prevention guidelines may not adequately protect themselves. Some welders forego PPE due to heat, thus exposing themselves to UVR damage in areas that are normally covered. Welders also may forego equipment when working on jobs requiring more detailed welds where clothing, masks, and glasses may be overly bulky and inhibit the worker’s precision. Nontraditional welders, such as artisans or handymen, may not have workplace safety education to be aware of UVR emitted from welding and may not have readily available PPE.

The Figure portrays an amateur welder working without full PPE. Although he is wearing a face shield, he is not wearing fire-retardant clothing, lacks full protective garments, and has no ventilation system.

Conclusion

It is important to recognize welding as an occupation with notable exposure to UVR. Personal protective equipment should be the mainstay of prevention. Sunscreen is a useful adjunct but does not cover UVC that is emitted in the welding arc. Screens and welding blankets can be placed around welders to contain UVR and limit nonwelder exposure. Although UVR hazards should be regulated in the workplace as part of regular safety reviews, the clinician can play a role in recognizing this source of UVR in skin disease and in encouraging the use of PPE.

Although solar radiation is the most commonly recognized source of UV radiation (UVR), occupational exposures can contribute due to the intensity and chronicity of exposure. Arc welding is a process whereby metal is fused together by heat produced from an electric arc. The electric arc that forms between the electrode and the base metal emits radiation in the full UV spectrum including UVA (400–315 nm), UVB (315–290 nm), and UVC (290–100 nm) wavelengths. Welders, therefore, have an increased risk for broad-spectrum, intense exposure to UVR, which may play a notable role in UV-related skin disease without proper protection. We report 3 welders with skin disease attributed to occupational exposure to UVR.

Case Reports

Patient 1
A 41-year-old man presented for evaluation of treatment-resistant cutaneous lupus. During the 10-year disease course, the patient was treated by both dermatologists and rheumatologists with frequent exacerbations and poor disease control. At the time of presentation, treatment with hydroxychloroquine 200 mg twice daily, azathioprine 50 mg twice daily, intramuscular methylprednisolone acetateinjectable suspension 40 mg, and prednisone 20 mg daily was failing. Physical examination revealed polycyclic erythematous plaques typical of subacute cutaneous lupus erythematosus. A skin biopsy confirmed the diagnosis. Upon further discussion of exacerbating risk factors, the patient noted UVR exposure while working as a welder. Although he had been previously told to avoid sunlight, he did not realize that this recommendation included all forms of UV light. Once this work exposure was eliminated, he was restarted on hydroxychloroquine 200 mg twice daily and topical steroids, and he responded with complete and sustained clearance of disease. When he returned to welding, utilization of sunscreen and sun-protective clothing enabled him to maintain control of his subacute cutaneous lupus erythematosus on oral hydroxychloroquine 200 mg twice daily and topical steroids.

Patient 2
A 55-year-old man presented with numerous actinic keratoses and persistent erythema in a well-demarcated area involving the forehead, temples, and lateral cheeks but sparing the periorbital area. The patient also experienced UVR exposure from welding (up to 4 to 5 times per week during his career spanning more than 20 years). He cited frequent burns in areas where his protective equipment did not cover his skin. He also reported that he often forgoes wearing protective equipment, even though it is available, and only uses safety goggles due to the extreme heat of the working environment as well as the awkwardness of wearing full protective gear while performing certain aspects of the job.

Patient 3
A 63-year-old man presented with a growth on the left side of the upper forehead. A biopsy revealed a squamous cell carcinoma, keratoacanthoma type. He worked as a welder for 40 years until retiring 1 year prior to presentation. He welded daily and always wore a tall face shield. Although the face shield covered most of his face, the scalp and some parts of the upper face were not well protected. In addition to the keratoacanthoma, which presented just outside of the area protected by the face shield, the patient had numerous actinic keratoses on the scalp.

Comment

Welding and UVR Exposure
Arc welders endure large amounts of UVR exposure, which is substantial enough to have notable health effects. The duration of exposure, electrical current used, angle of exposure, amount of ventilation, and the distance from the welding arc play a role in overall UVR exposure.^1,2 Maximum permissible exposure (MPE) limits to UVR have been set by the International Commission on Non-Ionizing Radiation Protection and the National Institute for Occupational Safety and Health.^3,4 The quantity of radiation produced by the arc allows for an exposure time of only a few seconds to minutes before surpassing MPE to UV light.^1,5 Welders are exposed to total-body UVR doses up to 3000 times the MPE, and mean cumulative exposure calculated over an 8-hour workday can reach 9795 mJ/cm².⁶

Workers in close proximity to welders also receive large UVR doses and may not be aware of its hazardous effects. Nearby nonwelders can be exposed to 13 times the MPE of UVR.⁶ At distances up to 10 m from the arc, the irradiance is large enough to reach MPE to UVR in less than 3 hours.¹

Skin and Eye Damage From Welding
Exposure to UVR produced by the welding arc may lead to acute skin or eye reactions, chronic skin or eye disorders, or exacerbation of photosensitive diseases. Common acute problems are photokeratoconjunctivitis (welder’s flash) and skin erythema.^7,8

Actinic elastosis, actinic keratoses, ocular melanoma, and photosensitive diseases represent a spectrum of disorders that can present from chronic UV exposure in welders. In a study by Emmett et al⁷ of 152 welders and 58 controls, actinic elastosis was found to be more frequent in welders than controls. Cases of basal cell carcinoma and squamous cell carcinoma also have been reported in welders.^9,10 However, in the study by Emmett et al,⁷ a statistically significant correlation between welding and skin cancer was not documented. There were limitations in the study, such as small sample size and a young average age of welders.⁷ Future studies may be needed to further clarify the risk for skin cancer in welders.

Although there is no clear association with skin cancer, an increased risk of ocular melanoma in welders is more clearly established. A meta-analysis of 5 studies found that welding was a significant risk factor for ocular melanoma, with an odds ratio of 2.05 (95% confidence interval, 1.20-3.51).¹¹ Other reported eye damage from chronic UVR exposure includes cataracts, chronic conjunctivitis, and retinal damage.^12,13

Case reports of the following photosensitive diseases have been reported to be exacerbated or caused by UV light exposure in welders: discoid lupus erythematosus¹⁴; photodermatitis¹⁵; broadband photosensitivity with decreased minimal erythema dose to UVA, UVB, and UVC¹⁶; UVC-exacerbated atopic dermatitis¹⁷; polymorphous light eruption–like skin eruption¹⁸; and UVA-induced photoallergy to hydrochlorothiazide and ramipril.¹⁹

Prevention of Occupational Exposure to UVR
Occupational Safety and Health Administration guidelines protect workers from excessive exposure to UVR with personal protective equipment (PPE). In addition to UVR protection, PPE needs to protect welders from other risks including trauma from welding debris (slag), fires, electrical burns, and fumes. Online resources from the National Ag Safety Database,²⁰ the American Welding Society,²¹ and Occupational Safety and Health Administration^22,23 are available. These resources advise welders to work in ventilated areas with respirators specific for the metal being welded and to wear clothing and gloves that are not only fire retardant but also UV resistant.^20-23 Additional PPE should protect the head, face, and eyes.

Unfortunately, even workers well trained in prevention guidelines may not adequately protect themselves. Some welders forego PPE due to heat, thus exposing themselves to UVR damage in areas that are normally covered. Welders also may forego equipment when working on jobs requiring more detailed welds where clothing, masks, and glasses may be overly bulky and inhibit the worker’s precision. Nontraditional welders, such as artisans or handymen, may not have workplace safety education to be aware of UVR emitted from welding and may not have readily available PPE.

The Figure portrays an amateur welder working without full PPE. Although he is wearing a face shield, he is not wearing fire-retardant clothing, lacks full protective garments, and has no ventilation system.

Conclusion

It is important to recognize welding as an occupation with notable exposure to UVR. Personal protective equipment should be the mainstay of prevention. Sunscreen is a useful adjunct but does not cover UVC that is emitted in the welding arc. Screens and welding blankets can be placed around welders to contain UVR and limit nonwelder exposure. Although UVR hazards should be regulated in the workplace as part of regular safety reviews, the clinician can play a role in recognizing this source of UVR in skin disease and in encouraging the use of PPE.

The expanded use of targeted anticancer agents such as sorafenib has revealed a growing spectrum of adverse cutaneous eruptions. We describe 3 patients with sorafenib-induced psoriasiform dermatitis and review the literature of only 10 other similar reported cases based on a search of PubMed, Web of Science, and American Society of Clinical Oncology abstracts using the terms psoriasis or psoriasiform dermatitis and sorafenib.^1-10 We seek to increase awareness of this particular drug eruption in response to sorafenib and to describe potential effective treatment options, especially when sorafenib cannot be discontinued.

Case Reports

Patient 1
A 68-year-old man with chronic hepatitis B infection and hepatocellular carcinoma (HCC) was started on sorafenib 400 mg daily. After 2 months of treatment, he developed painful hyperkeratotic lesions on the bilateral palms and soles with formation of calluses and superficial blisters on an erythematous base that was consistent with hand-foot skin reaction (HFSR). He also had numerous erythematous thin papules and plaques with adherent white scale and yellow crust on the bilateral thighs, lower legs, forearms, dorsal hands, abdomen, back, and buttocks (Figure 1). He had no personal or family history of psoriasis, and blood tests were unremarkable. Histologic analysis of punch biopsies from the buttocks and right leg revealed focal parakeratosis with neutrophils and serous crust, acanthosis, mild spongiosis, and lymphocytes at the dermoepidermal junction and surrounding dermal vessels, consistent with psoriasiform dermatitis (Figure 2). Sorafenib was discontinued and the eruption began to resolve within a week. A lower dose of sorafenib (200 mg daily) was attempted and the psoriasiform eruption recurred.

Patient 2
An 82-year-old man with chronic hepatitis B infection and HCC with lung metastasis was treated with sorafenib 400 mg daily. One week after treatment, he developed painful, thick, erythematous lesions on acral surfaces, consistent with HFSR. The sorafenib dose was decreased to 200 mg daily and HFSR resolved. Four months later, he developed well-demarcated, erythematous, scaly plaques with peripheral pustules on the right thigh (Figure 3) and right shin. He had no personal or family history of psoriasis, and blood tests were unremarkable. Samples from the pustules were taken for bacterial culture and fungal stain, but both were negative. Histologic analysis of a punch biopsy from the right thigh revealed necrotic parakeratosis, spongiform pustules, mild acanthosis, and a perivascular lymphocytic infiltrate with many neutrophils in the dermis. These findings suggested a diagnosis of pustular psoriasis, pustular drug eruption, or acute generalized exanthematous pustulosis. Treatment was initiated with mometasone cream. The patient subsequently developed hemoptysis and ascites from sorafenib. Sorafenib was discontinued and his skin eruption gradually resolved.

Comment

Sorafenib is an oral tyrosine kinase inhibitor that blocks tumor cell proliferation and angiogenesis due to its activity against vascular endothelial growth factor (VEGF) receptor, platelet-derived growth factor receptor, stem cell growth factor receptor, and rapidly accelerated fibrosarcoma kinases.¹¹ It is primarily used for the treatment of solid tumors, such as advanced renal cell carcinoma, unresectable HCC, and thyroid carcinoma, and more recently has been expanded for treatment of AML due to potential inhibition of FMS-like tyrosine kinase 3 receptor. Although dermatologic toxicity is a common adverse event during treatment with sorafenib,¹¹ reports of psoriasiform drug eruptions are rare.

Review of Cases
Based on our literature search, there are 10 previously reported cases of psoriasiform drug eruption secondary to sorafenib. Of the 13 total cases (including the 3 patients in this report), 7 patients had a history of psoriasis; most were middle-aged men; and the treatment with sorafenib was for solid tumors, primarily HCC with the exception of patient 3 from the current report who was treated for AML (Table). In all cases, the dose of sorafenib ranged from 200 to 800 mg daily. In 5 cases, HFSR preceded (as with patient 2 in the current report) or presented concurrently (as with patient 1 in the current report) with the onset of psoriasiform rash.^1,3,5

Vascular endothelial growth factor is overexpressed on psoriatic keratinocytes, contributes to epidermal hyperplasia, and induces angiogenesis in the dermis.¹² The development of psoriasiform eruptions in patients treated with sorafenib seems paradoxical, as this drug has been considered as potential therapy for psoriasis due to its ability to block VEGF receptor signaling. Indeed, an improvement of psoriasis has been reported in 1 case of a patient treated with sorafenib¹³ and in multiple patients with psoriasis treated with other VEGF antagonists (eg, bevacizumab).¹⁴ The underlying mechanisms by which sorafenib induced or exacerbated psoriasis are not entirely clear. Palmoplantar hyperkeratosis, keratosis pilaris–like eruption, multiple cysts, eruptive keratoacanthomas, and squamous cell carcinoma have been described in patients treated with sorafenib, supporting the hypothesis that treatment with sorafenib alters keratinocyte proliferation and differentiation.¹⁵ In addition, B-Raf inhibitors such as imatinib are known to induce or exacerbate psoriasiform dermatitis.¹⁶ The activity of sorafenib resulting in psoriasis may be specific to RAF kinase inhibition, as there are no reports in the literature that describe psoriasiform dermatitis with agents that preferentially block other sorafenib targets such as VEGF receptor, stem cell growth factor receptor, or platelet-derived growth factor receptor. Future studies are needed to fully elucidate the underlying mechanisms by which sorafenib induces or exacerbates psoriasiform dermatitis and whether the severity of the drug eruption correlates with the antitumor efficacy of sorafenib.

Conclusion

Although psoriasiform drug eruptions secondary to sorafenib are not life-threatening, they impact quality of life with associated pain, pruritus, infection, and limitation of daily activities. Dose reduction or discontinuation of sorafenib resulted in resolution of the psoriasiform dermatitis; however, as demonstrated in 3 cases (including patient 3),^1,2 psoriasiform dermatitis can be managed while maintaining the patient on sorafenib so that treatment of the malignancy is not compromised.

The expanded use of targeted anticancer agents such as sorafenib has revealed a growing spectrum of adverse cutaneous eruptions. We describe 3 patients with sorafenib-induced psoriasiform dermatitis and review the literature of only 10 other similar reported cases based on a search of PubMed, Web of Science, and American Society of Clinical Oncology abstracts using the terms psoriasis or psoriasiform dermatitis and sorafenib.^1-10 We seek to increase awareness of this particular drug eruption in response to sorafenib and to describe potential effective treatment options, especially when sorafenib cannot be discontinued.

Case Reports

Patient 1
A 68-year-old man with chronic hepatitis B infection and hepatocellular carcinoma (HCC) was started on sorafenib 400 mg daily. After 2 months of treatment, he developed painful hyperkeratotic lesions on the bilateral palms and soles with formation of calluses and superficial blisters on an erythematous base that was consistent with hand-foot skin reaction (HFSR). He also had numerous erythematous thin papules and plaques with adherent white scale and yellow crust on the bilateral thighs, lower legs, forearms, dorsal hands, abdomen, back, and buttocks (Figure 1). He had no personal or family history of psoriasis, and blood tests were unremarkable. Histologic analysis of punch biopsies from the buttocks and right leg revealed focal parakeratosis with neutrophils and serous crust, acanthosis, mild spongiosis, and lymphocytes at the dermoepidermal junction and surrounding dermal vessels, consistent with psoriasiform dermatitis (Figure 2). Sorafenib was discontinued and the eruption began to resolve within a week. A lower dose of sorafenib (200 mg daily) was attempted and the psoriasiform eruption recurred.

Patient 2
An 82-year-old man with chronic hepatitis B infection and HCC with lung metastasis was treated with sorafenib 400 mg daily. One week after treatment, he developed painful, thick, erythematous lesions on acral surfaces, consistent with HFSR. The sorafenib dose was decreased to 200 mg daily and HFSR resolved. Four months later, he developed well-demarcated, erythematous, scaly plaques with peripheral pustules on the right thigh (Figure 3) and right shin. He had no personal or family history of psoriasis, and blood tests were unremarkable. Samples from the pustules were taken for bacterial culture and fungal stain, but both were negative. Histologic analysis of a punch biopsy from the right thigh revealed necrotic parakeratosis, spongiform pustules, mild acanthosis, and a perivascular lymphocytic infiltrate with many neutrophils in the dermis. These findings suggested a diagnosis of pustular psoriasis, pustular drug eruption, or acute generalized exanthematous pustulosis. Treatment was initiated with mometasone cream. The patient subsequently developed hemoptysis and ascites from sorafenib. Sorafenib was discontinued and his skin eruption gradually resolved.

Comment

Sorafenib is an oral tyrosine kinase inhibitor that blocks tumor cell proliferation and angiogenesis due to its activity against vascular endothelial growth factor (VEGF) receptor, platelet-derived growth factor receptor, stem cell growth factor receptor, and rapidly accelerated fibrosarcoma kinases.¹¹ It is primarily used for the treatment of solid tumors, such as advanced renal cell carcinoma, unresectable HCC, and thyroid carcinoma, and more recently has been expanded for treatment of AML due to potential inhibition of FMS-like tyrosine kinase 3 receptor. Although dermatologic toxicity is a common adverse event during treatment with sorafenib,¹¹ reports of psoriasiform drug eruptions are rare.

Review of Cases
Based on our literature search, there are 10 previously reported cases of psoriasiform drug eruption secondary to sorafenib. Of the 13 total cases (including the 3 patients in this report), 7 patients had a history of psoriasis; most were middle-aged men; and the treatment with sorafenib was for solid tumors, primarily HCC with the exception of patient 3 from the current report who was treated for AML (Table). In all cases, the dose of sorafenib ranged from 200 to 800 mg daily. In 5 cases, HFSR preceded (as with patient 2 in the current report) or presented concurrently (as with patient 1 in the current report) with the onset of psoriasiform rash.^1,3,5

Vascular endothelial growth factor is overexpressed on psoriatic keratinocytes, contributes to epidermal hyperplasia, and induces angiogenesis in the dermis.¹² The development of psoriasiform eruptions in patients treated with sorafenib seems paradoxical, as this drug has been considered as potential therapy for psoriasis due to its ability to block VEGF receptor signaling. Indeed, an improvement of psoriasis has been reported in 1 case of a patient treated with sorafenib¹³ and in multiple patients with psoriasis treated with other VEGF antagonists (eg, bevacizumab).¹⁴ The underlying mechanisms by which sorafenib induced or exacerbated psoriasis are not entirely clear. Palmoplantar hyperkeratosis, keratosis pilaris–like eruption, multiple cysts, eruptive keratoacanthomas, and squamous cell carcinoma have been described in patients treated with sorafenib, supporting the hypothesis that treatment with sorafenib alters keratinocyte proliferation and differentiation.¹⁵ In addition, B-Raf inhibitors such as imatinib are known to induce or exacerbate psoriasiform dermatitis.¹⁶ The activity of sorafenib resulting in psoriasis may be specific to RAF kinase inhibition, as there are no reports in the literature that describe psoriasiform dermatitis with agents that preferentially block other sorafenib targets such as VEGF receptor, stem cell growth factor receptor, or platelet-derived growth factor receptor. Future studies are needed to fully elucidate the underlying mechanisms by which sorafenib induces or exacerbates psoriasiform dermatitis and whether the severity of the drug eruption correlates with the antitumor efficacy of sorafenib.

Conclusion

Although psoriasiform drug eruptions secondary to sorafenib are not life-threatening, they impact quality of life with associated pain, pruritus, infection, and limitation of daily activities. Dose reduction or discontinuation of sorafenib resulted in resolution of the psoriasiform dermatitis; however, as demonstrated in 3 cases (including patient 3),^1,2 psoriasiform dermatitis can be managed while maintaining the patient on sorafenib so that treatment of the malignancy is not compromised.

The expanded use of targeted anticancer agents such as sorafenib has revealed a growing spectrum of adverse cutaneous eruptions. We describe 3 patients with sorafenib-induced psoriasiform dermatitis and review the literature of only 10 other similar reported cases based on a search of PubMed, Web of Science, and American Society of Clinical Oncology abstracts using the terms psoriasis or psoriasiform dermatitis and sorafenib.^1-10 We seek to increase awareness of this particular drug eruption in response to sorafenib and to describe potential effective treatment options, especially when sorafenib cannot be discontinued.

Case Reports

Patient 1
A 68-year-old man with chronic hepatitis B infection and hepatocellular carcinoma (HCC) was started on sorafenib 400 mg daily. After 2 months of treatment, he developed painful hyperkeratotic lesions on the bilateral palms and soles with formation of calluses and superficial blisters on an erythematous base that was consistent with hand-foot skin reaction (HFSR). He also had numerous erythematous thin papules and plaques with adherent white scale and yellow crust on the bilateral thighs, lower legs, forearms, dorsal hands, abdomen, back, and buttocks (Figure 1). He had no personal or family history of psoriasis, and blood tests were unremarkable. Histologic analysis of punch biopsies from the buttocks and right leg revealed focal parakeratosis with neutrophils and serous crust, acanthosis, mild spongiosis, and lymphocytes at the dermoepidermal junction and surrounding dermal vessels, consistent with psoriasiform dermatitis (Figure 2). Sorafenib was discontinued and the eruption began to resolve within a week. A lower dose of sorafenib (200 mg daily) was attempted and the psoriasiform eruption recurred.

Patient 2
An 82-year-old man with chronic hepatitis B infection and HCC with lung metastasis was treated with sorafenib 400 mg daily. One week after treatment, he developed painful, thick, erythematous lesions on acral surfaces, consistent with HFSR. The sorafenib dose was decreased to 200 mg daily and HFSR resolved. Four months later, he developed well-demarcated, erythematous, scaly plaques with peripheral pustules on the right thigh (Figure 3) and right shin. He had no personal or family history of psoriasis, and blood tests were unremarkable. Samples from the pustules were taken for bacterial culture and fungal stain, but both were negative. Histologic analysis of a punch biopsy from the right thigh revealed necrotic parakeratosis, spongiform pustules, mild acanthosis, and a perivascular lymphocytic infiltrate with many neutrophils in the dermis. These findings suggested a diagnosis of pustular psoriasis, pustular drug eruption, or acute generalized exanthematous pustulosis. Treatment was initiated with mometasone cream. The patient subsequently developed hemoptysis and ascites from sorafenib. Sorafenib was discontinued and his skin eruption gradually resolved.

Comment

Sorafenib is an oral tyrosine kinase inhibitor that blocks tumor cell proliferation and angiogenesis due to its activity against vascular endothelial growth factor (VEGF) receptor, platelet-derived growth factor receptor, stem cell growth factor receptor, and rapidly accelerated fibrosarcoma kinases.¹¹ It is primarily used for the treatment of solid tumors, such as advanced renal cell carcinoma, unresectable HCC, and thyroid carcinoma, and more recently has been expanded for treatment of AML due to potential inhibition of FMS-like tyrosine kinase 3 receptor. Although dermatologic toxicity is a common adverse event during treatment with sorafenib,¹¹ reports of psoriasiform drug eruptions are rare.

Review of Cases
Based on our literature search, there are 10 previously reported cases of psoriasiform drug eruption secondary to sorafenib. Of the 13 total cases (including the 3 patients in this report), 7 patients had a history of psoriasis; most were middle-aged men; and the treatment with sorafenib was for solid tumors, primarily HCC with the exception of patient 3 from the current report who was treated for AML (Table). In all cases, the dose of sorafenib ranged from 200 to 800 mg daily. In 5 cases, HFSR preceded (as with patient 2 in the current report) or presented concurrently (as with patient 1 in the current report) with the onset of psoriasiform rash.^1,3,5

Vascular endothelial growth factor is overexpressed on psoriatic keratinocytes, contributes to epidermal hyperplasia, and induces angiogenesis in the dermis.¹² The development of psoriasiform eruptions in patients treated with sorafenib seems paradoxical, as this drug has been considered as potential therapy for psoriasis due to its ability to block VEGF receptor signaling. Indeed, an improvement of psoriasis has been reported in 1 case of a patient treated with sorafenib¹³ and in multiple patients with psoriasis treated with other VEGF antagonists (eg, bevacizumab).¹⁴ The underlying mechanisms by which sorafenib induced or exacerbated psoriasis are not entirely clear. Palmoplantar hyperkeratosis, keratosis pilaris–like eruption, multiple cysts, eruptive keratoacanthomas, and squamous cell carcinoma have been described in patients treated with sorafenib, supporting the hypothesis that treatment with sorafenib alters keratinocyte proliferation and differentiation.¹⁵ In addition, B-Raf inhibitors such as imatinib are known to induce or exacerbate psoriasiform dermatitis.¹⁶ The activity of sorafenib resulting in psoriasis may be specific to RAF kinase inhibition, as there are no reports in the literature that describe psoriasiform dermatitis with agents that preferentially block other sorafenib targets such as VEGF receptor, stem cell growth factor receptor, or platelet-derived growth factor receptor. Future studies are needed to fully elucidate the underlying mechanisms by which sorafenib induces or exacerbates psoriasiform dermatitis and whether the severity of the drug eruption correlates with the antitumor efficacy of sorafenib.

Conclusion

Although psoriasiform drug eruptions secondary to sorafenib are not life-threatening, they impact quality of life with associated pain, pruritus, infection, and limitation of daily activities. Dose reduction or discontinuation of sorafenib resulted in resolution of the psoriasiform dermatitis; however, as demonstrated in 3 cases (including patient 3),^1,2 psoriasiform dermatitis can be managed while maintaining the patient on sorafenib so that treatment of the malignancy is not compromised.