Contact Dermatitis

Diabetes mellitus is a morbid and costly condition that carries a high burden of disease for patients (both with and without a diagnosis) and for society as a whole. The economic burden of diabetes in the United States recently was estimated at nearly $250 billion annually,¹ and this number continues to rise. The cutaneous manifestations of diabetes are diverse and far-reaching, ranging from benign cosmetic concerns to severe dermatologic conditions. Given the wide range of etiology for diabetes mellitus and its existence on a spectrum of severity, it is perhaps not surprising that some of these entities are the subject of debate (vis-à-vis the strength of association between these skin conditions and diabetes) and can manifest in different forms. However, it is clear that the cutaneous manifestations of diabetes are equally as important to consider and manage as the systemic complications of the disease. In analyzing associations with diabetes, it is important to note that given such a high incidence of diabetes among the general population and its close association with other disease states, such as the metabolic syndrome, studies aimed at determining direct relationships to this entity must be well controlled for confounding factors, which may not even always be possible. Regardless, it is important for dermatologists and dermatology residents to recognize and understand the protean cutaneous manifestations of diabetes mellitus, and this column will explore skin findings that are characteristic of diabetes (Table 1) as well as other dermatoses with a reported but less clear association with diabetes (Table 2).

Skin Findings Characteristic of Diabetes

Diabetic Thick Skin

The association between diabetes and thick skin is well described as either a mobility-limiting affliction of the joints of the hands (cheiroarthropathy) or as an asymptomatic thickening of the skin. It has been estimated that 8% to 36% of patients with insulin-dependent diabetes develop some form of skin thickening²; one series also found this association to be true for patients with non–insulin-dependent diabetes mellitus (NIDDM).³ Skin thickening is readily observable on clinical presentation or ultrasonography, with increasing thickness in many cases associated with long-term disease progression. This increasing thickness was shown histopathologically to be a direct result of activated fibroblasts and increased collagen polymerization, with some similar features to progressive systemic sclerosis.⁴ Interestingly, even clinically normal skin showed some degree of fibroblast activation in diabetic patients, but collagen fibers in each case were smaller in diameter than those found in progressive systemic sclerosis. This finding clearly has implications on quality of life, as a lack of hand mobility due to the cheiroarthropathy can be severely disabling. Underlining the need for strict glycemic control, it has been suggested that tight control of blood sugar levels can lead to improvement in diabetic thick skin; however, reports of improvement are based on a small sample population.⁵ Huntley papules are localized to areas on the dorsum of the hands overlying the joints, demonstrating hyperkeratosis and enlarged dermal papillae.⁶ They also can be found in a minority of patients without diabetes. Interestingly, diabetic thick skin also has been associated with neurologic disorders in diabetes.⁷ Diabetic thick skin was found to be significantly (P<.05) correlated with diabetic neuropathy, independent of duration of diabetes, age, or glycosylated hemoglobin levels, though no causal or etiologic link between these entities has been proven.

Yellow Nails

Nail changes are well described in diabetes, ranging from periungual telangiectases to complications from infections, such as paronychia; however, a well-recognized finding, especially in elderly diabetic patients, is a characteristic yellowing of the nails, reported to affect up to 40% of patients with diabetes.⁸ The mechanism behind it likely includes accumulation of glycation end products, which also has been thought to lead to yellowing of the skin, and vascular impairment.⁹ These nails tend to exhibit slow growth, likely resulting from a nail matrix that is poorly supplied with blood, and also can be more curved than normal with longitudinal ridges (onychorrhexis).¹⁰ It is important, however, not to attribute yellow nails to diabetes without considering other causes of yellow nails, such as onychomycosis, yellow nail syndrome, and yellow nails associated with lymphedema or respiratory tract disease (eg, pleural effusion, bronchiectasis).¹¹

Diabetic Dermopathy

Colloquially known as shin spots, diabetic dermopathy is perhaps the most common skin finding in this patient population, though it also can occur in up to 1 in 5 individuals without diabetes.¹² Although it is very common, it is not a condition that should be overlooked, as numerous studies have shown an increase in microangiopathic complications such as retinopathy in patients with diabetic dermopathy.^13,14 Although follow-up studies may be necessary to fully characterize the relationship between shin spots and diabetes, it certainly is reasonable to be more wary of diabetic patients presenting with many shin spots, as the general consensus is that these areas represent postinflammatory hyperpigmentation and cutaneous atrophy in the setting of poor vascular supply, which should prompt analysis of other areas that might be affected by poor vasculature, such as an ophthalmologic examination. Antecedent and perhaps unnoticed trauma has been implicated given a possible underlying neuropathy, but this theory has not been supported by studies.

Bullosis Diabeticorum

Bullosis diabeticorum is a rare but well-described occurrence of self-resolving, nonscarring blisters that arise on the extremities of diabetic patients. This entity should be distinguished from other primary autoimmune blistering disorders and from simple mechanobullous lesions. Several types of bullosis diabeticorum have been described, with the most classic form showing an intraepidermal cleavage plane.¹⁵ These lesions tend to resolve in weeks but can be recurrent. The location of the pathology underlines its nonscarring nature, though similar lesions have been reported showing a cleavage plane in the lamina lucida of the dermoepidermal junction, which underlines the confusion in the literature surrounding diabetic bullae.¹⁶ Some may even use this term interchangeably with trauma or friction-induced blisters, which diabetics may be prone to develop due to peripheral neuropathy. Confounding reports have stated there is a correlation between bullosis diabeticorum and neuropathy as well as the acral location of these blisters. Although many authors cite the incidence of bullosis diabeticorum being 0.5%,¹⁷ no population-based studies have confirmed this figure and some have speculated that the actual incidence is higher.¹⁸ In the end, the term bullosis diabeticorum is probably best reserved for a rapidly appearing blister on the extremities of diabetic patients with at most minimal trauma, with a lesion containing sterile fluid and negative immunofluorescence. The mechanism for these blisters is thought to be microangiopathy, with scant blood supply to the skin causing it to be more prone to acantholysis and blister formation.¹⁹ This theory was reinforced in a study showing a reduced threshold for suction blister formation in diabetic patients.²⁰ Care should be taken to prevent secondary infections at these sites.

Acanthosis Nigricans

Acanthosis nigricans, which consists of dark brown plaques in the flexural areas, especially the posterior neck and axillae, is a common finding in diabetic patients and is no doubt familiar to clinicians. The pathophysiology of these lesions has been well studied and is a prototype for the effects of insulin resistance in the skin. In this model, high concentrations of insulin binding to insulinlike growth factor receptor in the skin stimulate keratinocyte proliferation,²¹ leading to the clinical appearance and the histologic finding of hyperkeratosis and papillomatosis, which in turn is responsible for the observed hyperpigmentation. It is an important finding, especially in those without a known history of diabetes, as it can also signal an underlying endocrinopathy (eg, Cushing syndrome, acromegaly, polycystic ovary syndrome) or malignancy (ie, adenocarcinoma of the gastrointestinal tract). Several distinct mechanisms of insulin resistance have been described, including insulin resistance due to receptor defects, such as those seen with insulin resistance in NIDDM; autoimmune processes; and postreceptor defects in insulin action.²² Keratolytics and topical retinoids have been used to ameliorate the appearance of these lesions.

Necrobiosis Lipoidica (Diabeticorum)

Necrobiosis lipoidica diabeticorum was first described by Urback²³ in 1932, but reports of similar lesions were described in nondiabetic patients soon after. The dermatologic community has since come to realize that perhaps a more accurate nomenclature is necrobiosis lipoidica to fully encompass this entity. Clinically, lesions appear as erythematous papules and plaques that expand into a larger well-circumscribed plaque with a waxy yellowish atrophic center, often with telangiectases, and usually presenting in the pretibial area. Lesions can become ulcerated in up to one-third of cases. Necrobiosis lipoidica also is defined by characteristic histologic findings, including important features such as palisaded granulomas arranged in a tierlike fashion, necrotizing vasculitis, collagen degradation, and panniculitis. Necrobiosis lipoidica is still relatively rare, developing in approximately 0.3% of patients with diabetes,²⁴ though its relationship with insulin resistance and diabetes is strong. Approximately two-thirds of patients with necrobiosis lipoidica have diabetes and an even higher number go on to develop diabetes or have a positive family history of diabetes.

Although these figures are interesting, the data are nearly a half-century-old, and it is unclear if these findings still hold true today. The etiology of necrobiosis lipoidica also remains elusive, with theories focusing on the role of microangiopathy, immunoglobulin deposition leading to vasculitis, structural abnormalities in collagen or fibroblasts, and trauma; however, the true nature of this condition is likely some combination of these factors.²⁵ These lesions are difficult to treat, especially at an advanced stage. Management with topical steroids to limit the inflammatory progression of the lesions is the mainstay of therapy.

Scleredema

Scleredema adultorum (Buschke disease) refers to indurated plaques over the posterior neck and upper back. It is usually thought of as 3 distinct forms. The form that is known to occur in diabetic patients is sometimes referred to as scleredema diabeticorum; the other 2 occur as postinfectious, usually Streptococcus, or malignancy-related forms. The prevalence of scleredema diabeticorum among diabetic individuals most frequently is reported as 2.5%²⁶; however, it is worth noting that other estimates have been as high as 14%.²⁷ Although there has been some correlation between poorly controlled NIDDM, treatment and tight glucose control does not seem to readily resolve these lesions with only few conflicting case studies serving as evidence for and against this premise.^28-30 The lesions often are recalcitrant toward a wide variety of treatment approaches. Histopathologic analysis generally reveals a thickened dermis with large collagen bundles, with clear spaces between the collagen representing mucin and increased numbers of mast cells. Proposed mechanisms include stimulation of collagen synthesis by fibroblasts and retarded collagen degradation, likely due to excess glucose.³¹

Dermatoses Demonstrating an Association With Diabetes

Granuloma Annulare

Granuloma annulare (GA) is a dermatologic condition existing in numerous forms. The generalized form has been suggested to have some association with diabetes. The lesions of GA are classically round, flesh-colored to erythematous papules arising in the dermis that may start on the dorsal extremities where the localized form typically presents, though larger annular plaques or patches may exist in the generalized form. Histologically, GA has a characteristic granulomatous infiltrate and palisaded granulomatous dermatitis, depending on the stage of the evolution. Many studies dating back to the mid-20th century have attempted to elucidate a link between GA and diabetes, with numerous reports showing conflicting results across their study populations.^32-36 This issue is further muddled by links between generalized GA and a host of other diseases, such as malignancy, thyroid disease, hepatitis, and human immunodeficiency virus infection. The usual course of GA is spontaneous resolution, including a peculiar phenomenon noted in the literature whereby biopsy of one of the lesions led to clearance of other lesions on the body.³⁷ However, the generalized form may be more difficult to treat, with therapeutic approaches including topical steroids, light therapy, and systemic immunomodulators.

Lichen Planus

A recent small population study in Turkey demonstrated a strong relationship between lichen planus and abnormal glucose tolerance. In this study of 30 patients with lichen planus, approximately half (14/30) had abnormal glucose metabolism and a quarter (8/30) had known diabetes, but larger studies are needed to clarify this relationship.³⁸ Prior to this report, a link between oral lichen planus and diabetes had been shown in larger case series.^39,40 Clinically, one may see white plaques with a characteristic lacy reticular pattern in the mouth. At other cutaneous sites, lichen planus generally appears as pruritic, purple, flat-topped polygonal papules. The clinical finding of lichen planus also is linked with many other disease states, most notably hepatitis C virus, but also thymoma, liver disease, and inflammatory bowel disease, among other associations.⁴¹

Vitiligo

As an autoimmune entity, it stands to reason that vitiligo may be seen more commonly associated with insulin-dependent diabetes, which has been shown to hold true in one study, while no association was found between later-onset NIDDM and vitiligo.⁴² Given the nonspecific nature of this association and the relatively common presentation of vitiligo, no special consideration is likely needed when examining a patient with vitiligo, but its presence should remind the clinician that these autoimmune entities tend to travel together.

Acquired Perforating Dermatosis

Although the classic presentation of acquired perforating dermatosis (Kyrle disease) is linked to renal failure, diabetes also has been connected to its presentation. Extremely rare outside of the setting of chronic renal failure, acquired perforating dermatosis occurs in up to 10% of dialysis patients.^43,44 It is characterized by papules with a central keratin plug, representing transepidermal elimination of keratin, collagen, and other cellular material; its etiology has not been elucidated. The connection between acquired perforating dermatosis and diabetes is not completely clear; it would seem that renal failure is a prerequisite for itspresentation. A large proportion of renal failure necessitating hemodialysis occurs in patients with diabetic nephropathy, which may explain the coincidence of diabetes, renal failure, and acquired perforating dermatosis.⁴⁵ The presentation of this cutaneous finding should not, however, affect treatment of the underlying conditions. Symptom relief in the form of topical steroids can be used as a first-line treatment of these often pruritic lesions.

Eruptive Xanthomas

The link between diabetes and eruptive xanthomas seems to be a rather tenuous one, hinging on the fact that many diabetic patients have abnormalities in carbohydrate and lipid metabolism. A central feature of eruptive xanthomas is an elevation in triglycerides, which can occur in diabetes. Indeed, it has been estimated that only 0.1% of diabetics will develop eruptive xanthomas,⁴⁶ and its main importance may be to prompt the physician to treat the hypertriglyceridemia and consider other concerning possibilities such as acute pancreatitis.

Psoriasis

Psoriasis is a common dermatologic condition that has been shown to have a far-reaching impact both on patients’ quality of life and cardiovascular risk profiles. Data have emerged linking psoriasis with diabetes as an independent risk factor⁴⁷; although this retrospective study had its limitations, it certainly is interesting to note that patients with psoriasis may have an increased risk for developing diabetes. Perhaps more importantly, though, this study also implied that patients with severe psoriasis may present with diabetes that is more difficult to control, evidenced by increased treatment with systemic therapies as opposed to milder forms of intervention such as diet control.⁴⁷ There almost certainly are other confounding factors and further studies would serve to reveal the strength of this association, but it is certainly an intriguing concept. Echoing these findings, a more recent nationwide study from Denmark demonstrated that psoriasis is associated with increased incidence of new-onset diabetes, adjusting for numerous confounding factors.⁴⁸ The relationship between psoriasis and diabetes is worth noting as evidence continues to emerge.

Conclusion

Given the diverse cutaneous manifestations of diabetes, it is important to distinguish those that are directly related to diabetes from those that suggest there may be another underlying process. For example, a new patient presenting to a primary care physician with acanthosis nigricans and yellow nails should immediately trigger a test for a hemoglobin A_1c (glycated hemoglobin) level to investigate for diabetes; however, clinicians also should be wary of patients with acanthosis nigricans who report early satiety, as this asso-ciation may be a sign of underlying malignancy. Conversely, the presence of yellow nails in a patient with chronic diabetes should not be ignored. The physician should consider onychomycosis and query the patient about possible respiratory symptoms. In the case of a multisystem disease such as diabetes, it may be challenging to reconcile seemingly disparate skin findings, but having a framework to approach the cutaneous manifestations of diabetes can help to properly identify and treat an individual patient’s afflictions.

Diabetes mellitus is a morbid and costly condition that carries a high burden of disease for patients (both with and without a diagnosis) and for society as a whole. The economic burden of diabetes in the United States recently was estimated at nearly $250 billion annually,¹ and this number continues to rise. The cutaneous manifestations of diabetes are diverse and far-reaching, ranging from benign cosmetic concerns to severe dermatologic conditions. Given the wide range of etiology for diabetes mellitus and its existence on a spectrum of severity, it is perhaps not surprising that some of these entities are the subject of debate (vis-à-vis the strength of association between these skin conditions and diabetes) and can manifest in different forms. However, it is clear that the cutaneous manifestations of diabetes are equally as important to consider and manage as the systemic complications of the disease. In analyzing associations with diabetes, it is important to note that given such a high incidence of diabetes among the general population and its close association with other disease states, such as the metabolic syndrome, studies aimed at determining direct relationships to this entity must be well controlled for confounding factors, which may not even always be possible. Regardless, it is important for dermatologists and dermatology residents to recognize and understand the protean cutaneous manifestations of diabetes mellitus, and this column will explore skin findings that are characteristic of diabetes (Table 1) as well as other dermatoses with a reported but less clear association with diabetes (Table 2).

Skin Findings Characteristic of Diabetes

Diabetic Thick Skin

The association between diabetes and thick skin is well described as either a mobility-limiting affliction of the joints of the hands (cheiroarthropathy) or as an asymptomatic thickening of the skin. It has been estimated that 8% to 36% of patients with insulin-dependent diabetes develop some form of skin thickening²; one series also found this association to be true for patients with non–insulin-dependent diabetes mellitus (NIDDM).³ Skin thickening is readily observable on clinical presentation or ultrasonography, with increasing thickness in many cases associated with long-term disease progression. This increasing thickness was shown histopathologically to be a direct result of activated fibroblasts and increased collagen polymerization, with some similar features to progressive systemic sclerosis.⁴ Interestingly, even clinically normal skin showed some degree of fibroblast activation in diabetic patients, but collagen fibers in each case were smaller in diameter than those found in progressive systemic sclerosis. This finding clearly has implications on quality of life, as a lack of hand mobility due to the cheiroarthropathy can be severely disabling. Underlining the need for strict glycemic control, it has been suggested that tight control of blood sugar levels can lead to improvement in diabetic thick skin; however, reports of improvement are based on a small sample population.⁵ Huntley papules are localized to areas on the dorsum of the hands overlying the joints, demonstrating hyperkeratosis and enlarged dermal papillae.⁶ They also can be found in a minority of patients without diabetes. Interestingly, diabetic thick skin also has been associated with neurologic disorders in diabetes.⁷ Diabetic thick skin was found to be significantly (P<.05) correlated with diabetic neuropathy, independent of duration of diabetes, age, or glycosylated hemoglobin levels, though no causal or etiologic link between these entities has been proven.

Yellow Nails

Nail changes are well described in diabetes, ranging from periungual telangiectases to complications from infections, such as paronychia; however, a well-recognized finding, especially in elderly diabetic patients, is a characteristic yellowing of the nails, reported to affect up to 40% of patients with diabetes.⁸ The mechanism behind it likely includes accumulation of glycation end products, which also has been thought to lead to yellowing of the skin, and vascular impairment.⁹ These nails tend to exhibit slow growth, likely resulting from a nail matrix that is poorly supplied with blood, and also can be more curved than normal with longitudinal ridges (onychorrhexis).¹⁰ It is important, however, not to attribute yellow nails to diabetes without considering other causes of yellow nails, such as onychomycosis, yellow nail syndrome, and yellow nails associated with lymphedema or respiratory tract disease (eg, pleural effusion, bronchiectasis).¹¹

Diabetic Dermopathy

Colloquially known as shin spots, diabetic dermopathy is perhaps the most common skin finding in this patient population, though it also can occur in up to 1 in 5 individuals without diabetes.¹² Although it is very common, it is not a condition that should be overlooked, as numerous studies have shown an increase in microangiopathic complications such as retinopathy in patients with diabetic dermopathy.^13,14 Although follow-up studies may be necessary to fully characterize the relationship between shin spots and diabetes, it certainly is reasonable to be more wary of diabetic patients presenting with many shin spots, as the general consensus is that these areas represent postinflammatory hyperpigmentation and cutaneous atrophy in the setting of poor vascular supply, which should prompt analysis of other areas that might be affected by poor vasculature, such as an ophthalmologic examination. Antecedent and perhaps unnoticed trauma has been implicated given a possible underlying neuropathy, but this theory has not been supported by studies.

Bullosis Diabeticorum

Bullosis diabeticorum is a rare but well-described occurrence of self-resolving, nonscarring blisters that arise on the extremities of diabetic patients. This entity should be distinguished from other primary autoimmune blistering disorders and from simple mechanobullous lesions. Several types of bullosis diabeticorum have been described, with the most classic form showing an intraepidermal cleavage plane.¹⁵ These lesions tend to resolve in weeks but can be recurrent. The location of the pathology underlines its nonscarring nature, though similar lesions have been reported showing a cleavage plane in the lamina lucida of the dermoepidermal junction, which underlines the confusion in the literature surrounding diabetic bullae.¹⁶ Some may even use this term interchangeably with trauma or friction-induced blisters, which diabetics may be prone to develop due to peripheral neuropathy. Confounding reports have stated there is a correlation between bullosis diabeticorum and neuropathy as well as the acral location of these blisters. Although many authors cite the incidence of bullosis diabeticorum being 0.5%,¹⁷ no population-based studies have confirmed this figure and some have speculated that the actual incidence is higher.¹⁸ In the end, the term bullosis diabeticorum is probably best reserved for a rapidly appearing blister on the extremities of diabetic patients with at most minimal trauma, with a lesion containing sterile fluid and negative immunofluorescence. The mechanism for these blisters is thought to be microangiopathy, with scant blood supply to the skin causing it to be more prone to acantholysis and blister formation.¹⁹ This theory was reinforced in a study showing a reduced threshold for suction blister formation in diabetic patients.²⁰ Care should be taken to prevent secondary infections at these sites.

Acanthosis Nigricans

Acanthosis nigricans, which consists of dark brown plaques in the flexural areas, especially the posterior neck and axillae, is a common finding in diabetic patients and is no doubt familiar to clinicians. The pathophysiology of these lesions has been well studied and is a prototype for the effects of insulin resistance in the skin. In this model, high concentrations of insulin binding to insulinlike growth factor receptor in the skin stimulate keratinocyte proliferation,²¹ leading to the clinical appearance and the histologic finding of hyperkeratosis and papillomatosis, which in turn is responsible for the observed hyperpigmentation. It is an important finding, especially in those without a known history of diabetes, as it can also signal an underlying endocrinopathy (eg, Cushing syndrome, acromegaly, polycystic ovary syndrome) or malignancy (ie, adenocarcinoma of the gastrointestinal tract). Several distinct mechanisms of insulin resistance have been described, including insulin resistance due to receptor defects, such as those seen with insulin resistance in NIDDM; autoimmune processes; and postreceptor defects in insulin action.²² Keratolytics and topical retinoids have been used to ameliorate the appearance of these lesions.

Necrobiosis Lipoidica (Diabeticorum)

Necrobiosis lipoidica diabeticorum was first described by Urback²³ in 1932, but reports of similar lesions were described in nondiabetic patients soon after. The dermatologic community has since come to realize that perhaps a more accurate nomenclature is necrobiosis lipoidica to fully encompass this entity. Clinically, lesions appear as erythematous papules and plaques that expand into a larger well-circumscribed plaque with a waxy yellowish atrophic center, often with telangiectases, and usually presenting in the pretibial area. Lesions can become ulcerated in up to one-third of cases. Necrobiosis lipoidica also is defined by characteristic histologic findings, including important features such as palisaded granulomas arranged in a tierlike fashion, necrotizing vasculitis, collagen degradation, and panniculitis. Necrobiosis lipoidica is still relatively rare, developing in approximately 0.3% of patients with diabetes,²⁴ though its relationship with insulin resistance and diabetes is strong. Approximately two-thirds of patients with necrobiosis lipoidica have diabetes and an even higher number go on to develop diabetes or have a positive family history of diabetes.

Although these figures are interesting, the data are nearly a half-century-old, and it is unclear if these findings still hold true today. The etiology of necrobiosis lipoidica also remains elusive, with theories focusing on the role of microangiopathy, immunoglobulin deposition leading to vasculitis, structural abnormalities in collagen or fibroblasts, and trauma; however, the true nature of this condition is likely some combination of these factors.²⁵ These lesions are difficult to treat, especially at an advanced stage. Management with topical steroids to limit the inflammatory progression of the lesions is the mainstay of therapy.

Scleredema

Scleredema adultorum (Buschke disease) refers to indurated plaques over the posterior neck and upper back. It is usually thought of as 3 distinct forms. The form that is known to occur in diabetic patients is sometimes referred to as scleredema diabeticorum; the other 2 occur as postinfectious, usually Streptococcus, or malignancy-related forms. The prevalence of scleredema diabeticorum among diabetic individuals most frequently is reported as 2.5%²⁶; however, it is worth noting that other estimates have been as high as 14%.²⁷ Although there has been some correlation between poorly controlled NIDDM, treatment and tight glucose control does not seem to readily resolve these lesions with only few conflicting case studies serving as evidence for and against this premise.^28-30 The lesions often are recalcitrant toward a wide variety of treatment approaches. Histopathologic analysis generally reveals a thickened dermis with large collagen bundles, with clear spaces between the collagen representing mucin and increased numbers of mast cells. Proposed mechanisms include stimulation of collagen synthesis by fibroblasts and retarded collagen degradation, likely due to excess glucose.³¹

Dermatoses Demonstrating an Association With Diabetes

Granuloma Annulare

Granuloma annulare (GA) is a dermatologic condition existing in numerous forms. The generalized form has been suggested to have some association with diabetes. The lesions of GA are classically round, flesh-colored to erythematous papules arising in the dermis that may start on the dorsal extremities where the localized form typically presents, though larger annular plaques or patches may exist in the generalized form. Histologically, GA has a characteristic granulomatous infiltrate and palisaded granulomatous dermatitis, depending on the stage of the evolution. Many studies dating back to the mid-20th century have attempted to elucidate a link between GA and diabetes, with numerous reports showing conflicting results across their study populations.^32-36 This issue is further muddled by links between generalized GA and a host of other diseases, such as malignancy, thyroid disease, hepatitis, and human immunodeficiency virus infection. The usual course of GA is spontaneous resolution, including a peculiar phenomenon noted in the literature whereby biopsy of one of the lesions led to clearance of other lesions on the body.³⁷ However, the generalized form may be more difficult to treat, with therapeutic approaches including topical steroids, light therapy, and systemic immunomodulators.

Lichen Planus

A recent small population study in Turkey demonstrated a strong relationship between lichen planus and abnormal glucose tolerance. In this study of 30 patients with lichen planus, approximately half (14/30) had abnormal glucose metabolism and a quarter (8/30) had known diabetes, but larger studies are needed to clarify this relationship.³⁸ Prior to this report, a link between oral lichen planus and diabetes had been shown in larger case series.^39,40 Clinically, one may see white plaques with a characteristic lacy reticular pattern in the mouth. At other cutaneous sites, lichen planus generally appears as pruritic, purple, flat-topped polygonal papules. The clinical finding of lichen planus also is linked with many other disease states, most notably hepatitis C virus, but also thymoma, liver disease, and inflammatory bowel disease, among other associations.⁴¹

Vitiligo

As an autoimmune entity, it stands to reason that vitiligo may be seen more commonly associated with insulin-dependent diabetes, which has been shown to hold true in one study, while no association was found between later-onset NIDDM and vitiligo.⁴² Given the nonspecific nature of this association and the relatively common presentation of vitiligo, no special consideration is likely needed when examining a patient with vitiligo, but its presence should remind the clinician that these autoimmune entities tend to travel together.

Acquired Perforating Dermatosis

Although the classic presentation of acquired perforating dermatosis (Kyrle disease) is linked to renal failure, diabetes also has been connected to its presentation. Extremely rare outside of the setting of chronic renal failure, acquired perforating dermatosis occurs in up to 10% of dialysis patients.^43,44 It is characterized by papules with a central keratin plug, representing transepidermal elimination of keratin, collagen, and other cellular material; its etiology has not been elucidated. The connection between acquired perforating dermatosis and diabetes is not completely clear; it would seem that renal failure is a prerequisite for itspresentation. A large proportion of renal failure necessitating hemodialysis occurs in patients with diabetic nephropathy, which may explain the coincidence of diabetes, renal failure, and acquired perforating dermatosis.⁴⁵ The presentation of this cutaneous finding should not, however, affect treatment of the underlying conditions. Symptom relief in the form of topical steroids can be used as a first-line treatment of these often pruritic lesions.

Eruptive Xanthomas

The link between diabetes and eruptive xanthomas seems to be a rather tenuous one, hinging on the fact that many diabetic patients have abnormalities in carbohydrate and lipid metabolism. A central feature of eruptive xanthomas is an elevation in triglycerides, which can occur in diabetes. Indeed, it has been estimated that only 0.1% of diabetics will develop eruptive xanthomas,⁴⁶ and its main importance may be to prompt the physician to treat the hypertriglyceridemia and consider other concerning possibilities such as acute pancreatitis.

Psoriasis

Psoriasis is a common dermatologic condition that has been shown to have a far-reaching impact both on patients’ quality of life and cardiovascular risk profiles. Data have emerged linking psoriasis with diabetes as an independent risk factor⁴⁷; although this retrospective study had its limitations, it certainly is interesting to note that patients with psoriasis may have an increased risk for developing diabetes. Perhaps more importantly, though, this study also implied that patients with severe psoriasis may present with diabetes that is more difficult to control, evidenced by increased treatment with systemic therapies as opposed to milder forms of intervention such as diet control.⁴⁷ There almost certainly are other confounding factors and further studies would serve to reveal the strength of this association, but it is certainly an intriguing concept. Echoing these findings, a more recent nationwide study from Denmark demonstrated that psoriasis is associated with increased incidence of new-onset diabetes, adjusting for numerous confounding factors.⁴⁸ The relationship between psoriasis and diabetes is worth noting as evidence continues to emerge.

Conclusion

Given the diverse cutaneous manifestations of diabetes, it is important to distinguish those that are directly related to diabetes from those that suggest there may be another underlying process. For example, a new patient presenting to a primary care physician with acanthosis nigricans and yellow nails should immediately trigger a test for a hemoglobin A_1c (glycated hemoglobin) level to investigate for diabetes; however, clinicians also should be wary of patients with acanthosis nigricans who report early satiety, as this asso-ciation may be a sign of underlying malignancy. Conversely, the presence of yellow nails in a patient with chronic diabetes should not be ignored. The physician should consider onychomycosis and query the patient about possible respiratory symptoms. In the case of a multisystem disease such as diabetes, it may be challenging to reconcile seemingly disparate skin findings, but having a framework to approach the cutaneous manifestations of diabetes can help to properly identify and treat an individual patient’s afflictions.

Diabetes mellitus is a morbid and costly condition that carries a high burden of disease for patients (both with and without a diagnosis) and for society as a whole. The economic burden of diabetes in the United States recently was estimated at nearly $250 billion annually,¹ and this number continues to rise. The cutaneous manifestations of diabetes are diverse and far-reaching, ranging from benign cosmetic concerns to severe dermatologic conditions. Given the wide range of etiology for diabetes mellitus and its existence on a spectrum of severity, it is perhaps not surprising that some of these entities are the subject of debate (vis-à-vis the strength of association between these skin conditions and diabetes) and can manifest in different forms. However, it is clear that the cutaneous manifestations of diabetes are equally as important to consider and manage as the systemic complications of the disease. In analyzing associations with diabetes, it is important to note that given such a high incidence of diabetes among the general population and its close association with other disease states, such as the metabolic syndrome, studies aimed at determining direct relationships to this entity must be well controlled for confounding factors, which may not even always be possible. Regardless, it is important for dermatologists and dermatology residents to recognize and understand the protean cutaneous manifestations of diabetes mellitus, and this column will explore skin findings that are characteristic of diabetes (Table 1) as well as other dermatoses with a reported but less clear association with diabetes (Table 2).

Skin Findings Characteristic of Diabetes

Diabetic Thick Skin

The association between diabetes and thick skin is well described as either a mobility-limiting affliction of the joints of the hands (cheiroarthropathy) or as an asymptomatic thickening of the skin. It has been estimated that 8% to 36% of patients with insulin-dependent diabetes develop some form of skin thickening²; one series also found this association to be true for patients with non–insulin-dependent diabetes mellitus (NIDDM).³ Skin thickening is readily observable on clinical presentation or ultrasonography, with increasing thickness in many cases associated with long-term disease progression. This increasing thickness was shown histopathologically to be a direct result of activated fibroblasts and increased collagen polymerization, with some similar features to progressive systemic sclerosis.⁴ Interestingly, even clinically normal skin showed some degree of fibroblast activation in diabetic patients, but collagen fibers in each case were smaller in diameter than those found in progressive systemic sclerosis. This finding clearly has implications on quality of life, as a lack of hand mobility due to the cheiroarthropathy can be severely disabling. Underlining the need for strict glycemic control, it has been suggested that tight control of blood sugar levels can lead to improvement in diabetic thick skin; however, reports of improvement are based on a small sample population.⁵ Huntley papules are localized to areas on the dorsum of the hands overlying the joints, demonstrating hyperkeratosis and enlarged dermal papillae.⁶ They also can be found in a minority of patients without diabetes. Interestingly, diabetic thick skin also has been associated with neurologic disorders in diabetes.⁷ Diabetic thick skin was found to be significantly (P<.05) correlated with diabetic neuropathy, independent of duration of diabetes, age, or glycosylated hemoglobin levels, though no causal or etiologic link between these entities has been proven.

Yellow Nails

Nail changes are well described in diabetes, ranging from periungual telangiectases to complications from infections, such as paronychia; however, a well-recognized finding, especially in elderly diabetic patients, is a characteristic yellowing of the nails, reported to affect up to 40% of patients with diabetes.⁸ The mechanism behind it likely includes accumulation of glycation end products, which also has been thought to lead to yellowing of the skin, and vascular impairment.⁹ These nails tend to exhibit slow growth, likely resulting from a nail matrix that is poorly supplied with blood, and also can be more curved than normal with longitudinal ridges (onychorrhexis).¹⁰ It is important, however, not to attribute yellow nails to diabetes without considering other causes of yellow nails, such as onychomycosis, yellow nail syndrome, and yellow nails associated with lymphedema or respiratory tract disease (eg, pleural effusion, bronchiectasis).¹¹

Diabetic Dermopathy

Colloquially known as shin spots, diabetic dermopathy is perhaps the most common skin finding in this patient population, though it also can occur in up to 1 in 5 individuals without diabetes.¹² Although it is very common, it is not a condition that should be overlooked, as numerous studies have shown an increase in microangiopathic complications such as retinopathy in patients with diabetic dermopathy.^13,14 Although follow-up studies may be necessary to fully characterize the relationship between shin spots and diabetes, it certainly is reasonable to be more wary of diabetic patients presenting with many shin spots, as the general consensus is that these areas represent postinflammatory hyperpigmentation and cutaneous atrophy in the setting of poor vascular supply, which should prompt analysis of other areas that might be affected by poor vasculature, such as an ophthalmologic examination. Antecedent and perhaps unnoticed trauma has been implicated given a possible underlying neuropathy, but this theory has not been supported by studies.

Bullosis Diabeticorum

Bullosis diabeticorum is a rare but well-described occurrence of self-resolving, nonscarring blisters that arise on the extremities of diabetic patients. This entity should be distinguished from other primary autoimmune blistering disorders and from simple mechanobullous lesions. Several types of bullosis diabeticorum have been described, with the most classic form showing an intraepidermal cleavage plane.¹⁵ These lesions tend to resolve in weeks but can be recurrent. The location of the pathology underlines its nonscarring nature, though similar lesions have been reported showing a cleavage plane in the lamina lucida of the dermoepidermal junction, which underlines the confusion in the literature surrounding diabetic bullae.¹⁶ Some may even use this term interchangeably with trauma or friction-induced blisters, which diabetics may be prone to develop due to peripheral neuropathy. Confounding reports have stated there is a correlation between bullosis diabeticorum and neuropathy as well as the acral location of these blisters. Although many authors cite the incidence of bullosis diabeticorum being 0.5%,¹⁷ no population-based studies have confirmed this figure and some have speculated that the actual incidence is higher.¹⁸ In the end, the term bullosis diabeticorum is probably best reserved for a rapidly appearing blister on the extremities of diabetic patients with at most minimal trauma, with a lesion containing sterile fluid and negative immunofluorescence. The mechanism for these blisters is thought to be microangiopathy, with scant blood supply to the skin causing it to be more prone to acantholysis and blister formation.¹⁹ This theory was reinforced in a study showing a reduced threshold for suction blister formation in diabetic patients.²⁰ Care should be taken to prevent secondary infections at these sites.

Acanthosis Nigricans

Acanthosis nigricans, which consists of dark brown plaques in the flexural areas, especially the posterior neck and axillae, is a common finding in diabetic patients and is no doubt familiar to clinicians. The pathophysiology of these lesions has been well studied and is a prototype for the effects of insulin resistance in the skin. In this model, high concentrations of insulin binding to insulinlike growth factor receptor in the skin stimulate keratinocyte proliferation,²¹ leading to the clinical appearance and the histologic finding of hyperkeratosis and papillomatosis, which in turn is responsible for the observed hyperpigmentation. It is an important finding, especially in those without a known history of diabetes, as it can also signal an underlying endocrinopathy (eg, Cushing syndrome, acromegaly, polycystic ovary syndrome) or malignancy (ie, adenocarcinoma of the gastrointestinal tract). Several distinct mechanisms of insulin resistance have been described, including insulin resistance due to receptor defects, such as those seen with insulin resistance in NIDDM; autoimmune processes; and postreceptor defects in insulin action.²² Keratolytics and topical retinoids have been used to ameliorate the appearance of these lesions.

Necrobiosis Lipoidica (Diabeticorum)

Necrobiosis lipoidica diabeticorum was first described by Urback²³ in 1932, but reports of similar lesions were described in nondiabetic patients soon after. The dermatologic community has since come to realize that perhaps a more accurate nomenclature is necrobiosis lipoidica to fully encompass this entity. Clinically, lesions appear as erythematous papules and plaques that expand into a larger well-circumscribed plaque with a waxy yellowish atrophic center, often with telangiectases, and usually presenting in the pretibial area. Lesions can become ulcerated in up to one-third of cases. Necrobiosis lipoidica also is defined by characteristic histologic findings, including important features such as palisaded granulomas arranged in a tierlike fashion, necrotizing vasculitis, collagen degradation, and panniculitis. Necrobiosis lipoidica is still relatively rare, developing in approximately 0.3% of patients with diabetes,²⁴ though its relationship with insulin resistance and diabetes is strong. Approximately two-thirds of patients with necrobiosis lipoidica have diabetes and an even higher number go on to develop diabetes or have a positive family history of diabetes.

Although these figures are interesting, the data are nearly a half-century-old, and it is unclear if these findings still hold true today. The etiology of necrobiosis lipoidica also remains elusive, with theories focusing on the role of microangiopathy, immunoglobulin deposition leading to vasculitis, structural abnormalities in collagen or fibroblasts, and trauma; however, the true nature of this condition is likely some combination of these factors.²⁵ These lesions are difficult to treat, especially at an advanced stage. Management with topical steroids to limit the inflammatory progression of the lesions is the mainstay of therapy.

Scleredema

Scleredema adultorum (Buschke disease) refers to indurated plaques over the posterior neck and upper back. It is usually thought of as 3 distinct forms. The form that is known to occur in diabetic patients is sometimes referred to as scleredema diabeticorum; the other 2 occur as postinfectious, usually Streptococcus, or malignancy-related forms. The prevalence of scleredema diabeticorum among diabetic individuals most frequently is reported as 2.5%²⁶; however, it is worth noting that other estimates have been as high as 14%.²⁷ Although there has been some correlation between poorly controlled NIDDM, treatment and tight glucose control does not seem to readily resolve these lesions with only few conflicting case studies serving as evidence for and against this premise.^28-30 The lesions often are recalcitrant toward a wide variety of treatment approaches. Histopathologic analysis generally reveals a thickened dermis with large collagen bundles, with clear spaces between the collagen representing mucin and increased numbers of mast cells. Proposed mechanisms include stimulation of collagen synthesis by fibroblasts and retarded collagen degradation, likely due to excess glucose.³¹

Dermatoses Demonstrating an Association With Diabetes

Granuloma Annulare

Granuloma annulare (GA) is a dermatologic condition existing in numerous forms. The generalized form has been suggested to have some association with diabetes. The lesions of GA are classically round, flesh-colored to erythematous papules arising in the dermis that may start on the dorsal extremities where the localized form typically presents, though larger annular plaques or patches may exist in the generalized form. Histologically, GA has a characteristic granulomatous infiltrate and palisaded granulomatous dermatitis, depending on the stage of the evolution. Many studies dating back to the mid-20th century have attempted to elucidate a link between GA and diabetes, with numerous reports showing conflicting results across their study populations.^32-36 This issue is further muddled by links between generalized GA and a host of other diseases, such as malignancy, thyroid disease, hepatitis, and human immunodeficiency virus infection. The usual course of GA is spontaneous resolution, including a peculiar phenomenon noted in the literature whereby biopsy of one of the lesions led to clearance of other lesions on the body.³⁷ However, the generalized form may be more difficult to treat, with therapeutic approaches including topical steroids, light therapy, and systemic immunomodulators.

Lichen Planus

A recent small population study in Turkey demonstrated a strong relationship between lichen planus and abnormal glucose tolerance. In this study of 30 patients with lichen planus, approximately half (14/30) had abnormal glucose metabolism and a quarter (8/30) had known diabetes, but larger studies are needed to clarify this relationship.³⁸ Prior to this report, a link between oral lichen planus and diabetes had been shown in larger case series.^39,40 Clinically, one may see white plaques with a characteristic lacy reticular pattern in the mouth. At other cutaneous sites, lichen planus generally appears as pruritic, purple, flat-topped polygonal papules. The clinical finding of lichen planus also is linked with many other disease states, most notably hepatitis C virus, but also thymoma, liver disease, and inflammatory bowel disease, among other associations.⁴¹

Vitiligo

As an autoimmune entity, it stands to reason that vitiligo may be seen more commonly associated with insulin-dependent diabetes, which has been shown to hold true in one study, while no association was found between later-onset NIDDM and vitiligo.⁴² Given the nonspecific nature of this association and the relatively common presentation of vitiligo, no special consideration is likely needed when examining a patient with vitiligo, but its presence should remind the clinician that these autoimmune entities tend to travel together.

Acquired Perforating Dermatosis

Although the classic presentation of acquired perforating dermatosis (Kyrle disease) is linked to renal failure, diabetes also has been connected to its presentation. Extremely rare outside of the setting of chronic renal failure, acquired perforating dermatosis occurs in up to 10% of dialysis patients.^43,44 It is characterized by papules with a central keratin plug, representing transepidermal elimination of keratin, collagen, and other cellular material; its etiology has not been elucidated. The connection between acquired perforating dermatosis and diabetes is not completely clear; it would seem that renal failure is a prerequisite for itspresentation. A large proportion of renal failure necessitating hemodialysis occurs in patients with diabetic nephropathy, which may explain the coincidence of diabetes, renal failure, and acquired perforating dermatosis.⁴⁵ The presentation of this cutaneous finding should not, however, affect treatment of the underlying conditions. Symptom relief in the form of topical steroids can be used as a first-line treatment of these often pruritic lesions.

Eruptive Xanthomas

The link between diabetes and eruptive xanthomas seems to be a rather tenuous one, hinging on the fact that many diabetic patients have abnormalities in carbohydrate and lipid metabolism. A central feature of eruptive xanthomas is an elevation in triglycerides, which can occur in diabetes. Indeed, it has been estimated that only 0.1% of diabetics will develop eruptive xanthomas,⁴⁶ and its main importance may be to prompt the physician to treat the hypertriglyceridemia and consider other concerning possibilities such as acute pancreatitis.

Psoriasis

Psoriasis is a common dermatologic condition that has been shown to have a far-reaching impact both on patients’ quality of life and cardiovascular risk profiles. Data have emerged linking psoriasis with diabetes as an independent risk factor⁴⁷; although this retrospective study had its limitations, it certainly is interesting to note that patients with psoriasis may have an increased risk for developing diabetes. Perhaps more importantly, though, this study also implied that patients with severe psoriasis may present with diabetes that is more difficult to control, evidenced by increased treatment with systemic therapies as opposed to milder forms of intervention such as diet control.⁴⁷ There almost certainly are other confounding factors and further studies would serve to reveal the strength of this association, but it is certainly an intriguing concept. Echoing these findings, a more recent nationwide study from Denmark demonstrated that psoriasis is associated with increased incidence of new-onset diabetes, adjusting for numerous confounding factors.⁴⁸ The relationship between psoriasis and diabetes is worth noting as evidence continues to emerge.

Conclusion

Given the diverse cutaneous manifestations of diabetes, it is important to distinguish those that are directly related to diabetes from those that suggest there may be another underlying process. For example, a new patient presenting to a primary care physician with acanthosis nigricans and yellow nails should immediately trigger a test for a hemoglobin A_1c (glycated hemoglobin) level to investigate for diabetes; however, clinicians also should be wary of patients with acanthosis nigricans who report early satiety, as this asso-ciation may be a sign of underlying malignancy. Conversely, the presence of yellow nails in a patient with chronic diabetes should not be ignored. The physician should consider onychomycosis and query the patient about possible respiratory symptoms. In the case of a multisystem disease such as diabetes, it may be challenging to reconcile seemingly disparate skin findings, but having a framework to approach the cutaneous manifestations of diabetes can help to properly identify and treat an individual patient’s afflictions.

The Diagnosis: Elephantiasis Nostras Verrucosa

Elephantiasis nostras verrucosa (ENV) is a chronic deforming disorder characterized by hyperkeratosis and papillomatosis of the epidermis with underlying woody fibrosis of the dermis and the subcutaneous tissue in the setting of chronic nonfilarial lymphedema. Mossy papules, plaques, and cobblestonelike nodules are classic clinical features of ENV (Figure). The cobblestone lesion often is foul smelling and may be colonized with multiple bacteria and fungi.¹

The differential diagnosis includes filariasis, chromoblastomycosis, and venous stasis dermatitis. Filariasis is infection with the filarial worms Wuchereria bancrofti, Brugia malayi, or Brugia timori. These parasites are transmitted to humans through the bite of an infected mosquito and develop into adult worms in the lymphatic vessels, causing lymphedema. Elephantiasis, the classic sign of late-stage disease, also causes swelling of the legs and genital organs that can be disfiguring. Thus travel history to Southeast Asia and Africa regions is particularly important.

Chromoblastomycosis is a chronic fungal infection of the skin and subcutaneous tissue caused by traumatic inoculation of a specific group of dematiaceous fungi, usually Fonsecaea pedrosoi, Phialophora verrucosa, Cladosporium carrionii, or Fonsecaea compacta. Chromoblastomycosis-induced lymphedema may mimic elephantiasis. Thus it is important to perform fungal culture and fungal scraping, which will show typical thick-walled, cigar-colored, sclerotic cells (known as Medlar bodies).

The skin changes of venous stasis dermatitis include edema, varicosities, hyperpigmentation, atrophie blanche, diffuse red-brown discoloration representing deep dermal deposits of hemosiderin, and lipodermatosclerosis. Venous stasis dermatitis is caused by chronic venous insufficiency. It is one of the predispositions to ENV, as it can cause repeated infections and cellulitis leading to blockage of lymphatic drainage and secondary lymphedema. A study of 21 cases highlighted that chronic venous insufficiency may be an underappreciated risk factor in the pathogenesis of ENV.²

The development of verrucous cobblestonelike plaques in ENV occurs with chronic lymphedema. Chronic lymphedema can be caused by a variety of etiologies including infection, chronic venous stasis, congestive heart failure, obesity, trauma, tumor obstruction, and radiation. Our patient was obese with chronic venous stasis, congestive heart failure, and recurrent cellulitis that can lead to secondary lymphatic obstruction and edema, all contributing to the picture of ENV. Because of the unilateral presentation in our patient, we believe that recurrent infections and inflammation are the important risk factors that lead to fibrosis of the dermis and lymph channels, ultimately resulting in ENV.

The diagnosis of ENV is based on the patient’s history and characteristic skin changes. However, a biopsy should be performed if the diagnosis is not clinically apparent, if the lesion looks suspicious for malignancy, or if areas of chronic ulceration exist. Histologic findings include hyperkeratosis as well as a collarette of acanthotic epidermis extending around dilated lymph vessels in the upper dermis. Sometimes a few dilated lymph vessels present in the deeper or subcutaneous tissue. Bacterial and fungal culture should be done to exclude concomitant infection. Ultrasonography can be used to evaluate the lymphatic and venous systems.

Treatment of ENV aims at improving lymphedema and preventing infection. Physical therapy to improve lymphedema includes manual lymphatic drainage, massage, compression stockings, elastic bandages, and leg elevation. In cases associated with obesity, weight loss is recommended. In cases of recurrent cellulitis or lymphangitis, long-term antibiotic therapy with agents such as penicillins or cephalosporins is sometimes used.³ Success has been reported with the use of oral and topical retinoids, such as acitretin and tazarotene, which are thought to decrease epidermal proliferation, fibrogenesis, and inflammation.^4,5 Our patient was started on tretinoin cream. Surgery is reserved for cases of lymphedema resistant to medical therapies, including debridement,⁶ lymphatic and lymphovenous anastomosis,⁷ lymphatic transplantation, and amputation for severe cases.

Malignancies can develop in areas of chronic lymphedema such as lymphangiosarcoma, squamous cell carcinoma, Kaposi sarcoma, B-cell lymphoma, and malignant fibrous histiocytoma. The morbidity and mortality usually are from the underlying medical problems rather than ENV itself. Our patient died 1 month after the diagnosis of ENV from his poorly controlled heart failure.

The Diagnosis: Elephantiasis Nostras Verrucosa

Elephantiasis nostras verrucosa (ENV) is a chronic deforming disorder characterized by hyperkeratosis and papillomatosis of the epidermis with underlying woody fibrosis of the dermis and the subcutaneous tissue in the setting of chronic nonfilarial lymphedema. Mossy papules, plaques, and cobblestonelike nodules are classic clinical features of ENV (Figure). The cobblestone lesion often is foul smelling and may be colonized with multiple bacteria and fungi.¹

The differential diagnosis includes filariasis, chromoblastomycosis, and venous stasis dermatitis. Filariasis is infection with the filarial worms Wuchereria bancrofti, Brugia malayi, or Brugia timori. These parasites are transmitted to humans through the bite of an infected mosquito and develop into adult worms in the lymphatic vessels, causing lymphedema. Elephantiasis, the classic sign of late-stage disease, also causes swelling of the legs and genital organs that can be disfiguring. Thus travel history to Southeast Asia and Africa regions is particularly important.

Chromoblastomycosis is a chronic fungal infection of the skin and subcutaneous tissue caused by traumatic inoculation of a specific group of dematiaceous fungi, usually Fonsecaea pedrosoi, Phialophora verrucosa, Cladosporium carrionii, or Fonsecaea compacta. Chromoblastomycosis-induced lymphedema may mimic elephantiasis. Thus it is important to perform fungal culture and fungal scraping, which will show typical thick-walled, cigar-colored, sclerotic cells (known as Medlar bodies).

The skin changes of venous stasis dermatitis include edema, varicosities, hyperpigmentation, atrophie blanche, diffuse red-brown discoloration representing deep dermal deposits of hemosiderin, and lipodermatosclerosis. Venous stasis dermatitis is caused by chronic venous insufficiency. It is one of the predispositions to ENV, as it can cause repeated infections and cellulitis leading to blockage of lymphatic drainage and secondary lymphedema. A study of 21 cases highlighted that chronic venous insufficiency may be an underappreciated risk factor in the pathogenesis of ENV.²

The development of verrucous cobblestonelike plaques in ENV occurs with chronic lymphedema. Chronic lymphedema can be caused by a variety of etiologies including infection, chronic venous stasis, congestive heart failure, obesity, trauma, tumor obstruction, and radiation. Our patient was obese with chronic venous stasis, congestive heart failure, and recurrent cellulitis that can lead to secondary lymphatic obstruction and edema, all contributing to the picture of ENV. Because of the unilateral presentation in our patient, we believe that recurrent infections and inflammation are the important risk factors that lead to fibrosis of the dermis and lymph channels, ultimately resulting in ENV.

The diagnosis of ENV is based on the patient’s history and characteristic skin changes. However, a biopsy should be performed if the diagnosis is not clinically apparent, if the lesion looks suspicious for malignancy, or if areas of chronic ulceration exist. Histologic findings include hyperkeratosis as well as a collarette of acanthotic epidermis extending around dilated lymph vessels in the upper dermis. Sometimes a few dilated lymph vessels present in the deeper or subcutaneous tissue. Bacterial and fungal culture should be done to exclude concomitant infection. Ultrasonography can be used to evaluate the lymphatic and venous systems.

Treatment of ENV aims at improving lymphedema and preventing infection. Physical therapy to improve lymphedema includes manual lymphatic drainage, massage, compression stockings, elastic bandages, and leg elevation. In cases associated with obesity, weight loss is recommended. In cases of recurrent cellulitis or lymphangitis, long-term antibiotic therapy with agents such as penicillins or cephalosporins is sometimes used.³ Success has been reported with the use of oral and topical retinoids, such as acitretin and tazarotene, which are thought to decrease epidermal proliferation, fibrogenesis, and inflammation.^4,5 Our patient was started on tretinoin cream. Surgery is reserved for cases of lymphedema resistant to medical therapies, including debridement,⁶ lymphatic and lymphovenous anastomosis,⁷ lymphatic transplantation, and amputation for severe cases.

Malignancies can develop in areas of chronic lymphedema such as lymphangiosarcoma, squamous cell carcinoma, Kaposi sarcoma, B-cell lymphoma, and malignant fibrous histiocytoma. The morbidity and mortality usually are from the underlying medical problems rather than ENV itself. Our patient died 1 month after the diagnosis of ENV from his poorly controlled heart failure.

The Diagnosis: Elephantiasis Nostras Verrucosa

Elephantiasis nostras verrucosa (ENV) is a chronic deforming disorder characterized by hyperkeratosis and papillomatosis of the epidermis with underlying woody fibrosis of the dermis and the subcutaneous tissue in the setting of chronic nonfilarial lymphedema. Mossy papules, plaques, and cobblestonelike nodules are classic clinical features of ENV (Figure). The cobblestone lesion often is foul smelling and may be colonized with multiple bacteria and fungi.¹

The differential diagnosis includes filariasis, chromoblastomycosis, and venous stasis dermatitis. Filariasis is infection with the filarial worms Wuchereria bancrofti, Brugia malayi, or Brugia timori. These parasites are transmitted to humans through the bite of an infected mosquito and develop into adult worms in the lymphatic vessels, causing lymphedema. Elephantiasis, the classic sign of late-stage disease, also causes swelling of the legs and genital organs that can be disfiguring. Thus travel history to Southeast Asia and Africa regions is particularly important.

Chromoblastomycosis is a chronic fungal infection of the skin and subcutaneous tissue caused by traumatic inoculation of a specific group of dematiaceous fungi, usually Fonsecaea pedrosoi, Phialophora verrucosa, Cladosporium carrionii, or Fonsecaea compacta. Chromoblastomycosis-induced lymphedema may mimic elephantiasis. Thus it is important to perform fungal culture and fungal scraping, which will show typical thick-walled, cigar-colored, sclerotic cells (known as Medlar bodies).

The skin changes of venous stasis dermatitis include edema, varicosities, hyperpigmentation, atrophie blanche, diffuse red-brown discoloration representing deep dermal deposits of hemosiderin, and lipodermatosclerosis. Venous stasis dermatitis is caused by chronic venous insufficiency. It is one of the predispositions to ENV, as it can cause repeated infections and cellulitis leading to blockage of lymphatic drainage and secondary lymphedema. A study of 21 cases highlighted that chronic venous insufficiency may be an underappreciated risk factor in the pathogenesis of ENV.²

The development of verrucous cobblestonelike plaques in ENV occurs with chronic lymphedema. Chronic lymphedema can be caused by a variety of etiologies including infection, chronic venous stasis, congestive heart failure, obesity, trauma, tumor obstruction, and radiation. Our patient was obese with chronic venous stasis, congestive heart failure, and recurrent cellulitis that can lead to secondary lymphatic obstruction and edema, all contributing to the picture of ENV. Because of the unilateral presentation in our patient, we believe that recurrent infections and inflammation are the important risk factors that lead to fibrosis of the dermis and lymph channels, ultimately resulting in ENV.

The diagnosis of ENV is based on the patient’s history and characteristic skin changes. However, a biopsy should be performed if the diagnosis is not clinically apparent, if the lesion looks suspicious for malignancy, or if areas of chronic ulceration exist. Histologic findings include hyperkeratosis as well as a collarette of acanthotic epidermis extending around dilated lymph vessels in the upper dermis. Sometimes a few dilated lymph vessels present in the deeper or subcutaneous tissue. Bacterial and fungal culture should be done to exclude concomitant infection. Ultrasonography can be used to evaluate the lymphatic and venous systems.

Treatment of ENV aims at improving lymphedema and preventing infection. Physical therapy to improve lymphedema includes manual lymphatic drainage, massage, compression stockings, elastic bandages, and leg elevation. In cases associated with obesity, weight loss is recommended. In cases of recurrent cellulitis or lymphangitis, long-term antibiotic therapy with agents such as penicillins or cephalosporins is sometimes used.³ Success has been reported with the use of oral and topical retinoids, such as acitretin and tazarotene, which are thought to decrease epidermal proliferation, fibrogenesis, and inflammation.^4,5 Our patient was started on tretinoin cream. Surgery is reserved for cases of lymphedema resistant to medical therapies, including debridement,⁶ lymphatic and lymphovenous anastomosis,⁷ lymphatic transplantation, and amputation for severe cases.

Malignancies can develop in areas of chronic lymphedema such as lymphangiosarcoma, squamous cell carcinoma, Kaposi sarcoma, B-cell lymphoma, and malignant fibrous histiocytoma. The morbidity and mortality usually are from the underlying medical problems rather than ENV itself. Our patient died 1 month after the diagnosis of ENV from his poorly controlled heart failure.

The skin care market includes topical product formulations (eg, foams, lotions, ointments, creams) for a number of dermatologic therapeutic targets; however, there is limited information available regarding patient preference for product vehicles by specific dermatologic disease. There are few studies in the literature examining patient adherence to topical medications.¹ In the current study, 6 focus groups comprised of patients with 3 dermatologic conditions of interest—acne, atopic dermatitis (AD), or plaque psoriasis (PP)—were surveyed to gain a better understanding of treatment preferences among patients and attributes of various formulations that are most desirable and important to this patient population.

Patient preference for a vehicle is relevant to treatment adherence in patients with conditions such as acne, AD, and PP because noncompliance is a major factor in the high rates of failure that have been associated with topical dermatologic treatments.² The aesthetic attributes of a given vehicle formulation depend on the disease state being treated, the site of application, and the length of treatment.³ A limited number of studies have linked patient preferences to the attributes of topical medications. In one study, participants preferred an aqueous gel formulation compared with previously used topical treatments for AD.³ In another study, participants ranked the following properties of a hypothetical topical medication as most important: gel formulation, room temperature storage, product life of up to 18 months once opened, application with fingers, and a once-daily regimen.² A third study found that foams were preferred among other formulations (ie, creams, gels, and ointments) for a variety of disease states, including AD, PP, and seborrheic dermatitis.⁴ The current study uses a qualitative analysis to further explore patient preferences of vehicle attributes for the topical treatment of acne, AD, and PP.

Methods

Study Participants

Six focus groups were conducted with patients who were using topical prescription medications for the treatment of acne, AD, or PP from March to April 2012. Participants were recruited from Raleigh, North Carolina, and New York, New York, with 3 focus groups (1 for each condition) from each city.

Following institutional review board approval of the study processes and recruitment materials, participants meeting the following criteria were included in the study: 18 years or older; diagnosed with acne, AD, or PP by a physician (participants who currently reported more than 1 of these conditions were not eligible); diagnosed with the respective skin condition at least 6 months prior to screening; current or prior use of topical prescription medications in at least 2 different vehicle formulations (eg, cream and foam, ointment and gel); use of at least 1 topical prescription treatment 5 or more times per month; and English speaking and able to provide written consent.

Study Design

A semistructured discussion guide was developed to ensure consistency in the topics surveyed among all 6 focus groups, and the same 2 moderators conducted the discussion for all 6 groups. At the beginning of the study, after providing written informed consent, participants were given an overview of the study and were asked general questions intended to get the participants talking about their experiences with their respective conditions. To avoid or minimize bias, participants were only asked open-ended questions designed to ascertain what symptoms they experienced in relation to their respective conditions. Finally, the discussions were focused on the topical prescription treatments that participants had tried and the properties of each treatment they liked and disliked.

The focus groups were recorded (audio) and transcribed; transcriptions were then verified through an iterative process of technical and editorial review. Analysis of the results was conducted by evaluation and review of the field notes as well as the transcripts from the focus groups.

Results

A total of 54 participants were surveyed (average age, 40.9 years). Although the average age of participants in the acne and AD groups was generally the same (35.2 and 35.4 years, respectively), the average age of participants in the PP group was higher (52.2 years). The majority of participants were white females who had at least a college degree. On average, participants had been diagnosed with their respective condition approximately 15.5 years prior to screening. Participant demographics and clinical characteristics are presented in more detail in Table 1.

At the time of screening, participants reported prior or present use of topical prescription medications in various formulations for treatment of their respective conditions. The most commonly reported vehicles across all 3 conditions were creams and ointments, followed by lotions, gels, and foams.

Symptoms Across Conditions

At the beginning of the study, participants reported symptoms they experienced in association with their respective skin conditions. Itching and redness were the only symptoms reported across all 3 conditions in all 6 focus groups. Dry skin/dryness, flaking/scaling/peeling, and pain were reported by at least 1 participant in 5 of 6 focus groups. Other common symptoms reported in at least 3 focus groups included pimples/bumps/boils, sensitivity, bleeding, discoloration of skin, cracking/cuts/skin breaking, and buildup of dead or thickened skin/chunks of skin/plaque. As shown in Table 2, there was more variability in the symptoms reported across the 2 acne groups compared with those reported by the AD and PP groups. Table 2 displays all symptoms spontaneously reported across each of the 6 focus groups. Symptoms are listed according to the terms/descriptions provided by focus group participants.

Results by Condition

Unless otherwise noted, participant responses generally were consistent across the North Carolina and New York focus groups. Results from the study, which included a total of 54 participants, suggested a high degree of similarity among preferences for topical treatment attributes across the 3 conditions. Moisturizing was the single attribute that was mentioned across all 3 conditions in all 6 focus groups as an important characteristic in a topical dermatologic treatment. Other attributes mentioned by at least 1 participant in 5 of 6 focus groups included the following: absorbs/disappears/dries quickly, available in various formulations, does not bleach or stain skin/hair/clothing, is not greasy/oily, is not sticky/tacky, is long lasting/long acting/stays on/lasts through sweating or hand washing, is fragrance or odor free, is easy to apply/simple to use, and can use all the time.

A few condition-specific attributes also were noted. An attribute was considered condition specific if it was mentioned by at least 1 participant in both groups for any condition. Preferred properties for topical medications that were specific to acne patients included the following: easy to dispense/dispenses right amount, nondrying, product goes on/spreads smoothly, container (eg, tube, pump) is not easily broken/does not leak, and creamy. Preferred properties that were specific to AD patients included the following: is not noticeable to others/conceals area, good consistency, and cooling. Preferred properties that were specific to 2 conditions included no residue among acne and AD patients and soothing among AD and PP patients. Preferred properties for topical medications across all focus groups are described in further detail in Table 3.

Regarding vehicle-type preferences, acne patients tended to prefer washes, creams, and lotions; AD patients preferred creams; and psoriasis patients preferred creams, ointments, and foams (particularly for the scalp).

Participants across all 3 conditions reported that during daytime hours (versus at night), they would be less likely to use products that are oily, shiny, thick (eg, ointments, oils); bleach or stain the skin/clothing; interfere with makeup, work, or other activities; or are visible to others. Rather, a majority of participants noted that they were more likely to use thinner, less oily products (eg, creams, lotions) during daytime hours, in social situations, or during certain activities (eg, exercise). Participants across all 3 conditions noted that they only used prescription shampoos when they had enough time to repeatedly rinse their hair to eliminate the smell. Participants across all 6 groups noted that they tended to choose creams or lotions over ointments (when available) during the summer because they considered these products to be lighter and less greasy.

Overall, participants indicated that condition-specific symptoms did not influence their preference for topical formulations; rather, they used whatever prescriptions they currently had. However, in some instances, participants noted that the location of the affected area might influence the type of product selected. For example, some participants tended not to use ointments on the scalp or in locations where their clothes might come into contact with the medicated area to avoid clothes sticking to medicated areas. Other participants used lighter, less stringent products on the face versus other locations. Participants noted that foams were preferred for more discrete localized areas.

Comment

Despite the variability in symptoms and conditions across the study population as well as participants’ varying experience with and access to topical treatments, the attributes participants valued most in topical prescription treatments were relatively consistent across the 3 conditions: moisturizing, absorbs/disappears/dries quickly, available in various formulations, does not bleach or stain skin/hair/clothing, is not greasy or oily, is not sticky or tacky, is long lasting/long acting, is fragrance or odor free, is easy to apply/simple to use, and can use all the time. Preferred vehicle attributes were generally consistent across the 3 conditions, but preferences for vehicle types tended to vary by condition. Although some vehicles were more closely associated with specific attributes (eg, the majority considered lotions to be moisturizing), no one vehicle was associated with the complete list of attributes for the ideal topical medication. Regardless of the symptoms experienced or particular situations/activities, participants noted that they tended to use the topical medications that were available to them, even if intended for different locations of the body, and nearly all reported using more than 1 type of topical treatment.

Caution should be used in interpreting these findings, as participant preferences are largely dependent on prior experience with different vehicle types. The small number of participants across the 3 conditions in this analysis also limits conclusions that can be drawn from the data sets, as each condition has different though somewhat overlapping treatment algorithms. A focused inquiry into each of these conditions through separate evaluations may provide for a more robust analysis. Future research might employ questionnaire methodology to develop items assessing the attributes of interest (eg, moisturizing ability, rate of absorption, greasiness/oiliness, stickiness/tackiness, fragrance/odor). The final set of desirable attributes may vary depending on the condition (eg, a moisturizing product may be more important to PP and AD patients than acne patients) as well as comparator products.

Acknowledgment—The authors would like to thank Jennifer Gwazdauskas, MBA, Research Triangle Park, North Carolina, of Stiefel, a GSK company, for her assistance with manuscript review.

The skin care market includes topical product formulations (eg, foams, lotions, ointments, creams) for a number of dermatologic therapeutic targets; however, there is limited information available regarding patient preference for product vehicles by specific dermatologic disease. There are few studies in the literature examining patient adherence to topical medications.¹ In the current study, 6 focus groups comprised of patients with 3 dermatologic conditions of interest—acne, atopic dermatitis (AD), or plaque psoriasis (PP)—were surveyed to gain a better understanding of treatment preferences among patients and attributes of various formulations that are most desirable and important to this patient population.

Patient preference for a vehicle is relevant to treatment adherence in patients with conditions such as acne, AD, and PP because noncompliance is a major factor in the high rates of failure that have been associated with topical dermatologic treatments.² The aesthetic attributes of a given vehicle formulation depend on the disease state being treated, the site of application, and the length of treatment.³ A limited number of studies have linked patient preferences to the attributes of topical medications. In one study, participants preferred an aqueous gel formulation compared with previously used topical treatments for AD.³ In another study, participants ranked the following properties of a hypothetical topical medication as most important: gel formulation, room temperature storage, product life of up to 18 months once opened, application with fingers, and a once-daily regimen.² A third study found that foams were preferred among other formulations (ie, creams, gels, and ointments) for a variety of disease states, including AD, PP, and seborrheic dermatitis.⁴ The current study uses a qualitative analysis to further explore patient preferences of vehicle attributes for the topical treatment of acne, AD, and PP.

Methods

Study Participants

Six focus groups were conducted with patients who were using topical prescription medications for the treatment of acne, AD, or PP from March to April 2012. Participants were recruited from Raleigh, North Carolina, and New York, New York, with 3 focus groups (1 for each condition) from each city.

Following institutional review board approval of the study processes and recruitment materials, participants meeting the following criteria were included in the study: 18 years or older; diagnosed with acne, AD, or PP by a physician (participants who currently reported more than 1 of these conditions were not eligible); diagnosed with the respective skin condition at least 6 months prior to screening; current or prior use of topical prescription medications in at least 2 different vehicle formulations (eg, cream and foam, ointment and gel); use of at least 1 topical prescription treatment 5 or more times per month; and English speaking and able to provide written consent.

Study Design

A semistructured discussion guide was developed to ensure consistency in the topics surveyed among all 6 focus groups, and the same 2 moderators conducted the discussion for all 6 groups. At the beginning of the study, after providing written informed consent, participants were given an overview of the study and were asked general questions intended to get the participants talking about their experiences with their respective conditions. To avoid or minimize bias, participants were only asked open-ended questions designed to ascertain what symptoms they experienced in relation to their respective conditions. Finally, the discussions were focused on the topical prescription treatments that participants had tried and the properties of each treatment they liked and disliked.

The focus groups were recorded (audio) and transcribed; transcriptions were then verified through an iterative process of technical and editorial review. Analysis of the results was conducted by evaluation and review of the field notes as well as the transcripts from the focus groups.

Results

A total of 54 participants were surveyed (average age, 40.9 years). Although the average age of participants in the acne and AD groups was generally the same (35.2 and 35.4 years, respectively), the average age of participants in the PP group was higher (52.2 years). The majority of participants were white females who had at least a college degree. On average, participants had been diagnosed with their respective condition approximately 15.5 years prior to screening. Participant demographics and clinical characteristics are presented in more detail in Table 1.

At the time of screening, participants reported prior or present use of topical prescription medications in various formulations for treatment of their respective conditions. The most commonly reported vehicles across all 3 conditions were creams and ointments, followed by lotions, gels, and foams.

Symptoms Across Conditions

At the beginning of the study, participants reported symptoms they experienced in association with their respective skin conditions. Itching and redness were the only symptoms reported across all 3 conditions in all 6 focus groups. Dry skin/dryness, flaking/scaling/peeling, and pain were reported by at least 1 participant in 5 of 6 focus groups. Other common symptoms reported in at least 3 focus groups included pimples/bumps/boils, sensitivity, bleeding, discoloration of skin, cracking/cuts/skin breaking, and buildup of dead or thickened skin/chunks of skin/plaque. As shown in Table 2, there was more variability in the symptoms reported across the 2 acne groups compared with those reported by the AD and PP groups. Table 2 displays all symptoms spontaneously reported across each of the 6 focus groups. Symptoms are listed according to the terms/descriptions provided by focus group participants.

Results by Condition

Unless otherwise noted, participant responses generally were consistent across the North Carolina and New York focus groups. Results from the study, which included a total of 54 participants, suggested a high degree of similarity among preferences for topical treatment attributes across the 3 conditions. Moisturizing was the single attribute that was mentioned across all 3 conditions in all 6 focus groups as an important characteristic in a topical dermatologic treatment. Other attributes mentioned by at least 1 participant in 5 of 6 focus groups included the following: absorbs/disappears/dries quickly, available in various formulations, does not bleach or stain skin/hair/clothing, is not greasy/oily, is not sticky/tacky, is long lasting/long acting/stays on/lasts through sweating or hand washing, is fragrance or odor free, is easy to apply/simple to use, and can use all the time.

A few condition-specific attributes also were noted. An attribute was considered condition specific if it was mentioned by at least 1 participant in both groups for any condition. Preferred properties for topical medications that were specific to acne patients included the following: easy to dispense/dispenses right amount, nondrying, product goes on/spreads smoothly, container (eg, tube, pump) is not easily broken/does not leak, and creamy. Preferred properties that were specific to AD patients included the following: is not noticeable to others/conceals area, good consistency, and cooling. Preferred properties that were specific to 2 conditions included no residue among acne and AD patients and soothing among AD and PP patients. Preferred properties for topical medications across all focus groups are described in further detail in Table 3.

Regarding vehicle-type preferences, acne patients tended to prefer washes, creams, and lotions; AD patients preferred creams; and psoriasis patients preferred creams, ointments, and foams (particularly for the scalp).

Participants across all 3 conditions reported that during daytime hours (versus at night), they would be less likely to use products that are oily, shiny, thick (eg, ointments, oils); bleach or stain the skin/clothing; interfere with makeup, work, or other activities; or are visible to others. Rather, a majority of participants noted that they were more likely to use thinner, less oily products (eg, creams, lotions) during daytime hours, in social situations, or during certain activities (eg, exercise). Participants across all 3 conditions noted that they only used prescription shampoos when they had enough time to repeatedly rinse their hair to eliminate the smell. Participants across all 6 groups noted that they tended to choose creams or lotions over ointments (when available) during the summer because they considered these products to be lighter and less greasy.

Overall, participants indicated that condition-specific symptoms did not influence their preference for topical formulations; rather, they used whatever prescriptions they currently had. However, in some instances, participants noted that the location of the affected area might influence the type of product selected. For example, some participants tended not to use ointments on the scalp or in locations where their clothes might come into contact with the medicated area to avoid clothes sticking to medicated areas. Other participants used lighter, less stringent products on the face versus other locations. Participants noted that foams were preferred for more discrete localized areas.

Comment

Despite the variability in symptoms and conditions across the study population as well as participants’ varying experience with and access to topical treatments, the attributes participants valued most in topical prescription treatments were relatively consistent across the 3 conditions: moisturizing, absorbs/disappears/dries quickly, available in various formulations, does not bleach or stain skin/hair/clothing, is not greasy or oily, is not sticky or tacky, is long lasting/long acting, is fragrance or odor free, is easy to apply/simple to use, and can use all the time. Preferred vehicle attributes were generally consistent across the 3 conditions, but preferences for vehicle types tended to vary by condition. Although some vehicles were more closely associated with specific attributes (eg, the majority considered lotions to be moisturizing), no one vehicle was associated with the complete list of attributes for the ideal topical medication. Regardless of the symptoms experienced or particular situations/activities, participants noted that they tended to use the topical medications that were available to them, even if intended for different locations of the body, and nearly all reported using more than 1 type of topical treatment.

Caution should be used in interpreting these findings, as participant preferences are largely dependent on prior experience with different vehicle types. The small number of participants across the 3 conditions in this analysis also limits conclusions that can be drawn from the data sets, as each condition has different though somewhat overlapping treatment algorithms. A focused inquiry into each of these conditions through separate evaluations may provide for a more robust analysis. Future research might employ questionnaire methodology to develop items assessing the attributes of interest (eg, moisturizing ability, rate of absorption, greasiness/oiliness, stickiness/tackiness, fragrance/odor). The final set of desirable attributes may vary depending on the condition (eg, a moisturizing product may be more important to PP and AD patients than acne patients) as well as comparator products.

Acknowledgment—The authors would like to thank Jennifer Gwazdauskas, MBA, Research Triangle Park, North Carolina, of Stiefel, a GSK company, for her assistance with manuscript review.

The skin care market includes topical product formulations (eg, foams, lotions, ointments, creams) for a number of dermatologic therapeutic targets; however, there is limited information available regarding patient preference for product vehicles by specific dermatologic disease. There are few studies in the literature examining patient adherence to topical medications.¹ In the current study, 6 focus groups comprised of patients with 3 dermatologic conditions of interest—acne, atopic dermatitis (AD), or plaque psoriasis (PP)—were surveyed to gain a better understanding of treatment preferences among patients and attributes of various formulations that are most desirable and important to this patient population.

Patient preference for a vehicle is relevant to treatment adherence in patients with conditions such as acne, AD, and PP because noncompliance is a major factor in the high rates of failure that have been associated with topical dermatologic treatments.² The aesthetic attributes of a given vehicle formulation depend on the disease state being treated, the site of application, and the length of treatment.³ A limited number of studies have linked patient preferences to the attributes of topical medications. In one study, participants preferred an aqueous gel formulation compared with previously used topical treatments for AD.³ In another study, participants ranked the following properties of a hypothetical topical medication as most important: gel formulation, room temperature storage, product life of up to 18 months once opened, application with fingers, and a once-daily regimen.² A third study found that foams were preferred among other formulations (ie, creams, gels, and ointments) for a variety of disease states, including AD, PP, and seborrheic dermatitis.⁴ The current study uses a qualitative analysis to further explore patient preferences of vehicle attributes for the topical treatment of acne, AD, and PP.

Methods

Study Participants

Six focus groups were conducted with patients who were using topical prescription medications for the treatment of acne, AD, or PP from March to April 2012. Participants were recruited from Raleigh, North Carolina, and New York, New York, with 3 focus groups (1 for each condition) from each city.

Following institutional review board approval of the study processes and recruitment materials, participants meeting the following criteria were included in the study: 18 years or older; diagnosed with acne, AD, or PP by a physician (participants who currently reported more than 1 of these conditions were not eligible); diagnosed with the respective skin condition at least 6 months prior to screening; current or prior use of topical prescription medications in at least 2 different vehicle formulations (eg, cream and foam, ointment and gel); use of at least 1 topical prescription treatment 5 or more times per month; and English speaking and able to provide written consent.

Study Design

A semistructured discussion guide was developed to ensure consistency in the topics surveyed among all 6 focus groups, and the same 2 moderators conducted the discussion for all 6 groups. At the beginning of the study, after providing written informed consent, participants were given an overview of the study and were asked general questions intended to get the participants talking about their experiences with their respective conditions. To avoid or minimize bias, participants were only asked open-ended questions designed to ascertain what symptoms they experienced in relation to their respective conditions. Finally, the discussions were focused on the topical prescription treatments that participants had tried and the properties of each treatment they liked and disliked.

The focus groups were recorded (audio) and transcribed; transcriptions were then verified through an iterative process of technical and editorial review. Analysis of the results was conducted by evaluation and review of the field notes as well as the transcripts from the focus groups.

Results

A total of 54 participants were surveyed (average age, 40.9 years). Although the average age of participants in the acne and AD groups was generally the same (35.2 and 35.4 years, respectively), the average age of participants in the PP group was higher (52.2 years). The majority of participants were white females who had at least a college degree. On average, participants had been diagnosed with their respective condition approximately 15.5 years prior to screening. Participant demographics and clinical characteristics are presented in more detail in Table 1.

At the time of screening, participants reported prior or present use of topical prescription medications in various formulations for treatment of their respective conditions. The most commonly reported vehicles across all 3 conditions were creams and ointments, followed by lotions, gels, and foams.

Symptoms Across Conditions

At the beginning of the study, participants reported symptoms they experienced in association with their respective skin conditions. Itching and redness were the only symptoms reported across all 3 conditions in all 6 focus groups. Dry skin/dryness, flaking/scaling/peeling, and pain were reported by at least 1 participant in 5 of 6 focus groups. Other common symptoms reported in at least 3 focus groups included pimples/bumps/boils, sensitivity, bleeding, discoloration of skin, cracking/cuts/skin breaking, and buildup of dead or thickened skin/chunks of skin/plaque. As shown in Table 2, there was more variability in the symptoms reported across the 2 acne groups compared with those reported by the AD and PP groups. Table 2 displays all symptoms spontaneously reported across each of the 6 focus groups. Symptoms are listed according to the terms/descriptions provided by focus group participants.

Results by Condition

Unless otherwise noted, participant responses generally were consistent across the North Carolina and New York focus groups. Results from the study, which included a total of 54 participants, suggested a high degree of similarity among preferences for topical treatment attributes across the 3 conditions. Moisturizing was the single attribute that was mentioned across all 3 conditions in all 6 focus groups as an important characteristic in a topical dermatologic treatment. Other attributes mentioned by at least 1 participant in 5 of 6 focus groups included the following: absorbs/disappears/dries quickly, available in various formulations, does not bleach or stain skin/hair/clothing, is not greasy/oily, is not sticky/tacky, is long lasting/long acting/stays on/lasts through sweating or hand washing, is fragrance or odor free, is easy to apply/simple to use, and can use all the time.

A few condition-specific attributes also were noted. An attribute was considered condition specific if it was mentioned by at least 1 participant in both groups for any condition. Preferred properties for topical medications that were specific to acne patients included the following: easy to dispense/dispenses right amount, nondrying, product goes on/spreads smoothly, container (eg, tube, pump) is not easily broken/does not leak, and creamy. Preferred properties that were specific to AD patients included the following: is not noticeable to others/conceals area, good consistency, and cooling. Preferred properties that were specific to 2 conditions included no residue among acne and AD patients and soothing among AD and PP patients. Preferred properties for topical medications across all focus groups are described in further detail in Table 3.

Regarding vehicle-type preferences, acne patients tended to prefer washes, creams, and lotions; AD patients preferred creams; and psoriasis patients preferred creams, ointments, and foams (particularly for the scalp).

Participants across all 3 conditions reported that during daytime hours (versus at night), they would be less likely to use products that are oily, shiny, thick (eg, ointments, oils); bleach or stain the skin/clothing; interfere with makeup, work, or other activities; or are visible to others. Rather, a majority of participants noted that they were more likely to use thinner, less oily products (eg, creams, lotions) during daytime hours, in social situations, or during certain activities (eg, exercise). Participants across all 3 conditions noted that they only used prescription shampoos when they had enough time to repeatedly rinse their hair to eliminate the smell. Participants across all 6 groups noted that they tended to choose creams or lotions over ointments (when available) during the summer because they considered these products to be lighter and less greasy.

Overall, participants indicated that condition-specific symptoms did not influence their preference for topical formulations; rather, they used whatever prescriptions they currently had. However, in some instances, participants noted that the location of the affected area might influence the type of product selected. For example, some participants tended not to use ointments on the scalp or in locations where their clothes might come into contact with the medicated area to avoid clothes sticking to medicated areas. Other participants used lighter, less stringent products on the face versus other locations. Participants noted that foams were preferred for more discrete localized areas.

Comment

Despite the variability in symptoms and conditions across the study population as well as participants’ varying experience with and access to topical treatments, the attributes participants valued most in topical prescription treatments were relatively consistent across the 3 conditions: moisturizing, absorbs/disappears/dries quickly, available in various formulations, does not bleach or stain skin/hair/clothing, is not greasy or oily, is not sticky or tacky, is long lasting/long acting, is fragrance or odor free, is easy to apply/simple to use, and can use all the time. Preferred vehicle attributes were generally consistent across the 3 conditions, but preferences for vehicle types tended to vary by condition. Although some vehicles were more closely associated with specific attributes (eg, the majority considered lotions to be moisturizing), no one vehicle was associated with the complete list of attributes for the ideal topical medication. Regardless of the symptoms experienced or particular situations/activities, participants noted that they tended to use the topical medications that were available to them, even if intended for different locations of the body, and nearly all reported using more than 1 type of topical treatment.

Caution should be used in interpreting these findings, as participant preferences are largely dependent on prior experience with different vehicle types. The small number of participants across the 3 conditions in this analysis also limits conclusions that can be drawn from the data sets, as each condition has different though somewhat overlapping treatment algorithms. A focused inquiry into each of these conditions through separate evaluations may provide for a more robust analysis. Future research might employ questionnaire methodology to develop items assessing the attributes of interest (eg, moisturizing ability, rate of absorption, greasiness/oiliness, stickiness/tackiness, fragrance/odor). The final set of desirable attributes may vary depending on the condition (eg, a moisturizing product may be more important to PP and AD patients than acne patients) as well as comparator products.

Acknowledgment—The authors would like to thank Jennifer Gwazdauskas, MBA, Research Triangle Park, North Carolina, of Stiefel, a GSK company, for her assistance with manuscript review.

To the Editor:

Sweet syndrome (SS) is characterized by fever; acute onset of painful erythematous papules, plaques, or nodules; peripheral neutrophilic leukocytosis; and histologic findings of a dense neutrophilic infiltrate without evidence of primary vasculitis.¹ We report a rare case of classic SS presenting with hemorrhagic bullae over photoexposed areas. Our case is notable because of the unusual nature of the clinical manifestation.

A 45-year-old woman presented with painful, fluid-filled lesions on the upper extremities of 1 week’s duration. Lesions were acute in onset and associated with fever. The patient had a history of diabetes mellitus and hypertension, which were well controlled. She also had a history of minimal itching on sun exposure as well as an upper respiratory tract infection 2 months prior to presentation. There was no history of muscle weakness, pain and/or discoloration of the fingertips, or treatment with topical or systemic agents.

On physical examination, the patient was well nourished with an average build. She was febrile (temperature, 38.5°C) with a pulse of 82 beats per minute, a blood pressure of 130/80, and a respiratory rate of 14 breaths per minute. On cutaneous examination multiple erythematous plaques with central large hemorrhagic bullae were present on the extensor aspect of the forearms and dorsum of the left hand. The smallest plaque measured 4×8 cm and the largest measured 8×15 cm (Figure 1). The lesions were tender, and Nikolsky sign was negative. Considering the clinical features, a differential diagnosis of bullous systemic lupus erythematosus, polymorphic light eruption, Jessner lymphocytic infiltrate, and SS were considered. Complete blood cell count demonstrated a hemoglobin level of 12.1 g/dL (reference range, 14.0–17.5 g/dL), total leukocyte count of 13,280/μL (reference range, 4500–11,000/μL), neutrophil count of 80% (reference range, 56%), lymphocyte count of 13% (reference range, 34%), monocyte count of 8% (reference range, 4%), and an erythrocyte sedimentation rate of 40 mm/h (reference range, 0–20 mm/h). Serum creatinine levels were 0.9 mg/dL (reference range, 0.6–1.2 mg/dL) and urea nitrogen levels were 26 mg/dL (reference range, 8–23 mg/dL). C-reactive protein was positive, antinuclear antibody was negative, and double-stranded DNA was negative. Ultrasonography of the abdomen and pelvis and a chest radiograph revealed no abnormalities.

Figure 1. Multiple erythematous plaques with central large hemorrhagic bullae were present on the extensor aspect of the forearms and dorsum of the left hand.

Histopathology from a lesion on the forearm revealed a dense, predominantly neutrophilic infiltrate located in the superficial dermis as well as prominent papillary dermal edema infiltration in the dermis with vasodilatation in some areas without leukocytoclastic vasculitis features (Figures 2 and 3). Considering the clinical and histopathologic features, a diagnosis of SS was made, and the patient was started on intravenous dexamethasone (4 mg twice daily) with a dramatic response, as the lesions almost cleared within 4 to 5 days of treatment (Figure 4).

Figure 2. Neutrophilic infiltrate in the superficial dermis and prominent papillary dermal edema infiltrate in the dermis (H&amp;E, original magnification ×40).

Sweet syndrome was first described in 1964 as acute febrile neutrophilic dermatosis.² Sweet syndrome can be subdivided into 3 groups depending on the clinical setting: classic or idiopathic, malignancy associated, and drug induced.³ Classic or idiopathic SS typically affects women in the third to fifth decades of life,⁴ as seen in our case. The proposed diagnostic criteria for SS state that patients must meet both of the 2 major criteria and 2 of 4 minor criteria for the diagnosis.³ The major criteria include acute onset of typical skin lesions and histopathologic findings consistent with SS. The minor criteria include fever (temperature >38°C) or general malaise; association with malignancy, inflammatory disease, pregnancy, or antecedent respiratory or gastrointestinal tract infection; excellent response to treatment with systemic corticosteroids or potassium iodide; and abnormal laboratory values at presentation (3 of 4 required: erythrocyte sedimentation rate >20 mm/h; leukocyte count >8000/μL; neutrophil count >70%; positive C-reactive protein). Our patient fulfilled both the major criteria and 3 of 4 minor criteria. Although the exact etiology of SS is unknown, it is widely believed that SS may be a hypersensitivity response to underlying bacterial infections such as Yersinia enterocolitica, viral infections, or tumors.⁵

Cytokine dysregulation also has been indicated in the pathogenesis of SS, an imbalance of cytokine secretion from helper T cells such as IL-2 and IFN-γ, which may stimulate the cytokine cascade leading to activation of neutrophils and release of toxic metabolites.⁶ The cutaneous manifestations of SS consist of erythematous to violaceous tender papules or nodules that often coalesce to form irregular plaques.⁷ Rare clinical manifestations include bullous lesions; oral involvement; glomerulonephritis; myositis; and ocular manifestations including conjunctivitis, episcleritis, and iridocyclitis.^5,8 Photoaggravated and photoinduced syndromes also have been reported.⁹

Figure 3. Polymorphonuclear neutrophilic infiltrate with multilobulated nucleus without leukocytoclasia (H&amp;E, original magnification ×100).

Bullous-type SS has been reported, but all the known cases were secondary to malignancy or were drug induced^8,10; they did not present in a classic or idiopathic variant. Our case is unique in that it is a report of the classic SS variant with lesions including bullae over photoexposed areas, possibly indicating a causal association of sun exposure and the development of SS.

Figure 4. Intravenous dexamethasone resulted in improvement within 4 to 5 days of treatment.

The diagnostic histopathologic features of SS include a dense, predominantly neutrophilic infiltrate located in the superficial dermis as well as prominent papillary dermal edema, which occasionally may lead to subepidermal vesiculation.^11,12 The epidermis often is normal but spongiosis may be present, and rarely neutrophils may extend into the epidermis to form subcorneal pustules.¹³ Severe edema in the papillary dermis may cause subepidermal blistering and bullous lesions.¹⁰ Systemic steroids are the therapeutic mainstay in SS. Other treatment options include methylprednisolone, potassium iodide, colchicine, indomethacin, cyclosporine, and dapsone.

To the Editor:

Sweet syndrome (SS) is characterized by fever; acute onset of painful erythematous papules, plaques, or nodules; peripheral neutrophilic leukocytosis; and histologic findings of a dense neutrophilic infiltrate without evidence of primary vasculitis.¹ We report a rare case of classic SS presenting with hemorrhagic bullae over photoexposed areas. Our case is notable because of the unusual nature of the clinical manifestation.

A 45-year-old woman presented with painful, fluid-filled lesions on the upper extremities of 1 week’s duration. Lesions were acute in onset and associated with fever. The patient had a history of diabetes mellitus and hypertension, which were well controlled. She also had a history of minimal itching on sun exposure as well as an upper respiratory tract infection 2 months prior to presentation. There was no history of muscle weakness, pain and/or discoloration of the fingertips, or treatment with topical or systemic agents.

On physical examination, the patient was well nourished with an average build. She was febrile (temperature, 38.5°C) with a pulse of 82 beats per minute, a blood pressure of 130/80, and a respiratory rate of 14 breaths per minute. On cutaneous examination multiple erythematous plaques with central large hemorrhagic bullae were present on the extensor aspect of the forearms and dorsum of the left hand. The smallest plaque measured 4×8 cm and the largest measured 8×15 cm (Figure 1). The lesions were tender, and Nikolsky sign was negative. Considering the clinical features, a differential diagnosis of bullous systemic lupus erythematosus, polymorphic light eruption, Jessner lymphocytic infiltrate, and SS were considered. Complete blood cell count demonstrated a hemoglobin level of 12.1 g/dL (reference range, 14.0–17.5 g/dL), total leukocyte count of 13,280/μL (reference range, 4500–11,000/μL), neutrophil count of 80% (reference range, 56%), lymphocyte count of 13% (reference range, 34%), monocyte count of 8% (reference range, 4%), and an erythrocyte sedimentation rate of 40 mm/h (reference range, 0–20 mm/h). Serum creatinine levels were 0.9 mg/dL (reference range, 0.6–1.2 mg/dL) and urea nitrogen levels were 26 mg/dL (reference range, 8–23 mg/dL). C-reactive protein was positive, antinuclear antibody was negative, and double-stranded DNA was negative. Ultrasonography of the abdomen and pelvis and a chest radiograph revealed no abnormalities.

Figure 1. Multiple erythematous plaques with central large hemorrhagic bullae were present on the extensor aspect of the forearms and dorsum of the left hand.

Histopathology from a lesion on the forearm revealed a dense, predominantly neutrophilic infiltrate located in the superficial dermis as well as prominent papillary dermal edema infiltration in the dermis with vasodilatation in some areas without leukocytoclastic vasculitis features (Figures 2 and 3). Considering the clinical and histopathologic features, a diagnosis of SS was made, and the patient was started on intravenous dexamethasone (4 mg twice daily) with a dramatic response, as the lesions almost cleared within 4 to 5 days of treatment (Figure 4).

Figure 2. Neutrophilic infiltrate in the superficial dermis and prominent papillary dermal edema infiltrate in the dermis (H&amp;E, original magnification ×40).

Sweet syndrome was first described in 1964 as acute febrile neutrophilic dermatosis.² Sweet syndrome can be subdivided into 3 groups depending on the clinical setting: classic or idiopathic, malignancy associated, and drug induced.³ Classic or idiopathic SS typically affects women in the third to fifth decades of life,⁴ as seen in our case. The proposed diagnostic criteria for SS state that patients must meet both of the 2 major criteria and 2 of 4 minor criteria for the diagnosis.³ The major criteria include acute onset of typical skin lesions and histopathologic findings consistent with SS. The minor criteria include fever (temperature >38°C) or general malaise; association with malignancy, inflammatory disease, pregnancy, or antecedent respiratory or gastrointestinal tract infection; excellent response to treatment with systemic corticosteroids or potassium iodide; and abnormal laboratory values at presentation (3 of 4 required: erythrocyte sedimentation rate >20 mm/h; leukocyte count >8000/μL; neutrophil count >70%; positive C-reactive protein). Our patient fulfilled both the major criteria and 3 of 4 minor criteria. Although the exact etiology of SS is unknown, it is widely believed that SS may be a hypersensitivity response to underlying bacterial infections such as Yersinia enterocolitica, viral infections, or tumors.⁵

Cytokine dysregulation also has been indicated in the pathogenesis of SS, an imbalance of cytokine secretion from helper T cells such as IL-2 and IFN-γ, which may stimulate the cytokine cascade leading to activation of neutrophils and release of toxic metabolites.⁶ The cutaneous manifestations of SS consist of erythematous to violaceous tender papules or nodules that often coalesce to form irregular plaques.⁷ Rare clinical manifestations include bullous lesions; oral involvement; glomerulonephritis; myositis; and ocular manifestations including conjunctivitis, episcleritis, and iridocyclitis.^5,8 Photoaggravated and photoinduced syndromes also have been reported.⁹

Figure 3. Polymorphonuclear neutrophilic infiltrate with multilobulated nucleus without leukocytoclasia (H&amp;E, original magnification ×100).

Bullous-type SS has been reported, but all the known cases were secondary to malignancy or were drug induced^8,10; they did not present in a classic or idiopathic variant. Our case is unique in that it is a report of the classic SS variant with lesions including bullae over photoexposed areas, possibly indicating a causal association of sun exposure and the development of SS.

Figure 4. Intravenous dexamethasone resulted in improvement within 4 to 5 days of treatment.

The diagnostic histopathologic features of SS include a dense, predominantly neutrophilic infiltrate located in the superficial dermis as well as prominent papillary dermal edema, which occasionally may lead to subepidermal vesiculation.^11,12 The epidermis often is normal but spongiosis may be present, and rarely neutrophils may extend into the epidermis to form subcorneal pustules.¹³ Severe edema in the papillary dermis may cause subepidermal blistering and bullous lesions.¹⁰ Systemic steroids are the therapeutic mainstay in SS. Other treatment options include methylprednisolone, potassium iodide, colchicine, indomethacin, cyclosporine, and dapsone.

To the Editor:

Sweet syndrome (SS) is characterized by fever; acute onset of painful erythematous papules, plaques, or nodules; peripheral neutrophilic leukocytosis; and histologic findings of a dense neutrophilic infiltrate without evidence of primary vasculitis.¹ We report a rare case of classic SS presenting with hemorrhagic bullae over photoexposed areas. Our case is notable because of the unusual nature of the clinical manifestation.

A 45-year-old woman presented with painful, fluid-filled lesions on the upper extremities of 1 week’s duration. Lesions were acute in onset and associated with fever. The patient had a history of diabetes mellitus and hypertension, which were well controlled. She also had a history of minimal itching on sun exposure as well as an upper respiratory tract infection 2 months prior to presentation. There was no history of muscle weakness, pain and/or discoloration of the fingertips, or treatment with topical or systemic agents.

On physical examination, the patient was well nourished with an average build. She was febrile (temperature, 38.5°C) with a pulse of 82 beats per minute, a blood pressure of 130/80, and a respiratory rate of 14 breaths per minute. On cutaneous examination multiple erythematous plaques with central large hemorrhagic bullae were present on the extensor aspect of the forearms and dorsum of the left hand. The smallest plaque measured 4×8 cm and the largest measured 8×15 cm (Figure 1). The lesions were tender, and Nikolsky sign was negative. Considering the clinical features, a differential diagnosis of bullous systemic lupus erythematosus, polymorphic light eruption, Jessner lymphocytic infiltrate, and SS were considered. Complete blood cell count demonstrated a hemoglobin level of 12.1 g/dL (reference range, 14.0–17.5 g/dL), total leukocyte count of 13,280/μL (reference range, 4500–11,000/μL), neutrophil count of 80% (reference range, 56%), lymphocyte count of 13% (reference range, 34%), monocyte count of 8% (reference range, 4%), and an erythrocyte sedimentation rate of 40 mm/h (reference range, 0–20 mm/h). Serum creatinine levels were 0.9 mg/dL (reference range, 0.6–1.2 mg/dL) and urea nitrogen levels were 26 mg/dL (reference range, 8–23 mg/dL). C-reactive protein was positive, antinuclear antibody was negative, and double-stranded DNA was negative. Ultrasonography of the abdomen and pelvis and a chest radiograph revealed no abnormalities.

Figure 1. Multiple erythematous plaques with central large hemorrhagic bullae were present on the extensor aspect of the forearms and dorsum of the left hand.

Histopathology from a lesion on the forearm revealed a dense, predominantly neutrophilic infiltrate located in the superficial dermis as well as prominent papillary dermal edema infiltration in the dermis with vasodilatation in some areas without leukocytoclastic vasculitis features (Figures 2 and 3). Considering the clinical and histopathologic features, a diagnosis of SS was made, and the patient was started on intravenous dexamethasone (4 mg twice daily) with a dramatic response, as the lesions almost cleared within 4 to 5 days of treatment (Figure 4).

Figure 2. Neutrophilic infiltrate in the superficial dermis and prominent papillary dermal edema infiltrate in the dermis (H&amp;E, original magnification ×40).

Sweet syndrome was first described in 1964 as acute febrile neutrophilic dermatosis.² Sweet syndrome can be subdivided into 3 groups depending on the clinical setting: classic or idiopathic, malignancy associated, and drug induced.³ Classic or idiopathic SS typically affects women in the third to fifth decades of life,⁴ as seen in our case. The proposed diagnostic criteria for SS state that patients must meet both of the 2 major criteria and 2 of 4 minor criteria for the diagnosis.³ The major criteria include acute onset of typical skin lesions and histopathologic findings consistent with SS. The minor criteria include fever (temperature >38°C) or general malaise; association with malignancy, inflammatory disease, pregnancy, or antecedent respiratory or gastrointestinal tract infection; excellent response to treatment with systemic corticosteroids or potassium iodide; and abnormal laboratory values at presentation (3 of 4 required: erythrocyte sedimentation rate >20 mm/h; leukocyte count >8000/μL; neutrophil count >70%; positive C-reactive protein). Our patient fulfilled both the major criteria and 3 of 4 minor criteria. Although the exact etiology of SS is unknown, it is widely believed that SS may be a hypersensitivity response to underlying bacterial infections such as Yersinia enterocolitica, viral infections, or tumors.⁵

Cytokine dysregulation also has been indicated in the pathogenesis of SS, an imbalance of cytokine secretion from helper T cells such as IL-2 and IFN-γ, which may stimulate the cytokine cascade leading to activation of neutrophils and release of toxic metabolites.⁶ The cutaneous manifestations of SS consist of erythematous to violaceous tender papules or nodules that often coalesce to form irregular plaques.⁷ Rare clinical manifestations include bullous lesions; oral involvement; glomerulonephritis; myositis; and ocular manifestations including conjunctivitis, episcleritis, and iridocyclitis.^5,8 Photoaggravated and photoinduced syndromes also have been reported.⁹

Figure 3. Polymorphonuclear neutrophilic infiltrate with multilobulated nucleus without leukocytoclasia (H&amp;E, original magnification ×100).

Bullous-type SS has been reported, but all the known cases were secondary to malignancy or were drug induced^8,10; they did not present in a classic or idiopathic variant. Our case is unique in that it is a report of the classic SS variant with lesions including bullae over photoexposed areas, possibly indicating a causal association of sun exposure and the development of SS.

Figure 4. Intravenous dexamethasone resulted in improvement within 4 to 5 days of treatment.

The diagnostic histopathologic features of SS include a dense, predominantly neutrophilic infiltrate located in the superficial dermis as well as prominent papillary dermal edema, which occasionally may lead to subepidermal vesiculation.^11,12 The epidermis often is normal but spongiosis may be present, and rarely neutrophils may extend into the epidermis to form subcorneal pustules.¹³ Severe edema in the papillary dermis may cause subepidermal blistering and bullous lesions.¹⁰ Systemic steroids are the therapeutic mainstay in SS. Other treatment options include methylprednisolone, potassium iodide, colchicine, indomethacin, cyclosporine, and dapsone.

To the Editor:

Acrokeratosis paraneoplastica (AP) of Bazex is a rare but distinctive acral psoriasiform dermatosis associated with internal malignancy, usually squamous cell carcinoma (SCC), of the upper aerodigestive tract.^1,2 Recognizing this paraneoplastic condition is paramount because cutaneous findings often precede the onset of symptoms associated with an occult malignancy.³

A 76-year-old woman with adenocarcinoma of the transverse colon of 3 years’ duration was referred to the dermatology department. She had a hemicolectomy and was doing well until tumor recurrence with peritoneal metastasis was detected following an exploratory laparotomy 1 year prior to presentation to us. She underwent a total hysterectomy and bilateral salpingo-oophorectomy. Palliative chemotherapy was initiated, and she completed 5 cycles of capecitabine and oxaliplatin. Long-term medications for diabetes mellitus, hypertension, and hyperlipidemia included glibenclamide, nifedipine, hydrochlorothiazide, and losartan. The patient had a progressive pruritic rash of 6 months’ duration that started on the hands and forearms and spread to involve the feet and lower limbs as well as the ears and face. She also experienced progressive thickening of the palms and soles. She had been treated with topical steroids and emollients with no improvement. Clinical examination revealed erythematous scaly patches on all of her limbs, especially on the elbows and knees, and on the ear helices and nose. There also was notable palmoplantar keratoderma with central sparing (Figure 1), onycholysis, and subungual hyperkeratosis. A skin biopsy of the forearm was performed, and histology revealed orthokeratosis, hypergranulosis and basal vacuolar alteration, and superficial perivascular lymphohistiocytic infiltrate with melanophages (Figure 2). Direct immunofluorescence studies and fungal cultures were negative. The cutaneous features of a treatment-resistant acral dermatosis supported the clinical diagnosis of AP, especially in the setting of an internal malignancy. The patient was started on palliative radiotherapy with no notable resolution of the cutaneous lesions. She was lost to follow-up.

Figure 1. Palmoplantar keratoderma with central sparing of the palms and soles, onycholysis, and subungual hyperkeratosis .

First described by Bazex et al¹ in 1965, AP is an uncommon but well-recognized paraneoplastic dermatosis associated with an underlying neoplasm. Fewer than 160 cases have been reported, and the majority of cases have been men older than 40 years. The most commonly associated malignancy was SCC (at least 50% of reported cases) involving mainly the oropharynx and larynx, with lung and esophageal SCC also described.^2-4 Only a few cases of adenocarcinoma-associated AP have been described, such as adenocarcinoma of the lung, prostate, and stomach.^3-5 In a reported case of AP associated with early colon adenocarcinoma, the patient had remarkable cutaneous resolution following successful tumor resection.⁶ Other reported rare hematologic associations included Hodgkin disease, peripheral T-cell lymphoma, and multiple myeloma.^3-5 Karabulut et al⁴ described a case associated with cholangiocarcinoma and studied the primary sites of malignancies in another 133 patients with AP (118 patients with documented cell type): oropharynx and larynx in 55 (41%); lung in 23 (17%); unknown location in 20 (15%); esophagus in 14 (11%); prostate in 3 (2%); stomach in 3 (2%); and isolated cases involving the liver, thymus, uterus, vulva, breast, urinary bladder, lymph nodes, and bone marrow.

Figure 2. Histologic examination revealed basal vacuolar alteration, hyperkeratosis, and superficial perivascular lymphohistiocytic infiltrate with melanophages (H&amp;E, original magnification ×40).

A review of 113 cases of AP showed that only 15% developed cutaneous lesions after the malignancy had been discovered; in the majority of cases (67%), cutaneous lesions were present for an average of 1 year preceding the diagnosis of malignancy.³ In our patient, the late-onset cutaneous involvement corresponded to the progression of the underlying colon adenocarcinoma. In the absence of initial cutaneous involvement or when successful treatment of a tumor results in cutaneous resolution, subsequent emergence of cutaneous lesions of AP signifies tumor progression. The evolution of cutaneous features in AP was well described by Bazex et al.¹ Stage 1 of the disease shows initial erythema and psoriasiform scaling of the fingers and toes, spreading to the nose and ear helices (hence acrokeratosis); the tumor frequently remains asymptomatic or undetected. Stage 2 shows palmoplantar keratoderma with central sparing and more extensive facial lesions; progression to stage 3 occurs if the tumor remains undetected or untreated, with further spread of psoriasiform lesions to the elbows, knees, trunk, and scalp.¹ Nail involvement occurs in nearly 75% of cases⁵; typical changes include subungual hyperkeratosis, onycholysis, longitudinal streaks, yellow pigmentation, and rarely onychomadesis. A high index of suspicion of AP is paramount when evaluating any recalcitrant acral dermatosis that fails to respond to appropriate therapy, especially in the presence of constitutional symptoms, typical bulbous enlargement of distal phalanges, or isolated involvement of helices. These findings should prompt physicians to perform an extensive search for an underlying malignancy with complete physical examination, particularly of the head and neck region, with appropriate endoscopic assessment and imaging studies.

A myriad of nonspecific histologic features of AP commonly reported include hyperkeratosis, parakeratosis, acanthosis, and dermal perivascular lymphohistiocytic infiltrate.⁷ Less common features include dyskeratotic keratinocytes, vacuolar degeneration, bandlike infiltrate, and melanin incontinence.⁷ The pathogenesis of AP remains elusive. A postulated immunologic mechanism is based on reports of immunoglobulins (IgG, IgA, IgM) and complement (C3) deposition along the basement membrane zone.⁸ Association with autoimmune disorders such as alopecia areata and vitiligo also has been reported.⁹ Another possible mechanism is cross-reactivity between antigens found in the tumor and skin, resulting in a T-cell–mediated immune response to tumorlike antigens in the epidermis, or secretion of tumor-originating growth factors responsible for the hyperkeratotic skin changes, such as epidermal growth factor, transforming growth factor a, or insulinlike growth factor 1.^3,7,10,11

Spontaneous remission of cutaneous lesions in untreated underlying malignancy is rare. Isolated reports of treatment using topical and systemic steroids, salicylic acid, topical vitamin D analogues, etretinate, and psoralen plus UVA showed minimal improvement.^5,7 The mainstay in attaining cutaneous resolution is to detect and eradicate the underlying neoplasm with surgery, chemotherapy, or radiotherapy, or combination therapy.

Our case is noteworthy because of the patient’s gender (female), underlying malignancy (adenocarcinoma of the colon), and late onset of cutaneous involvement, which are all uncommon associations related to paraneoplastic syndrome. The clinical features of AP should be recognized early to facilitate an extensive search for an occult malignancy, and late-onset cutaneous involvement also should be recognized as a marker of tumor relapse or progression.

To the Editor:

Acrokeratosis paraneoplastica (AP) of Bazex is a rare but distinctive acral psoriasiform dermatosis associated with internal malignancy, usually squamous cell carcinoma (SCC), of the upper aerodigestive tract.^1,2 Recognizing this paraneoplastic condition is paramount because cutaneous findings often precede the onset of symptoms associated with an occult malignancy.³

A 76-year-old woman with adenocarcinoma of the transverse colon of 3 years’ duration was referred to the dermatology department. She had a hemicolectomy and was doing well until tumor recurrence with peritoneal metastasis was detected following an exploratory laparotomy 1 year prior to presentation to us. She underwent a total hysterectomy and bilateral salpingo-oophorectomy. Palliative chemotherapy was initiated, and she completed 5 cycles of capecitabine and oxaliplatin. Long-term medications for diabetes mellitus, hypertension, and hyperlipidemia included glibenclamide, nifedipine, hydrochlorothiazide, and losartan. The patient had a progressive pruritic rash of 6 months’ duration that started on the hands and forearms and spread to involve the feet and lower limbs as well as the ears and face. She also experienced progressive thickening of the palms and soles. She had been treated with topical steroids and emollients with no improvement. Clinical examination revealed erythematous scaly patches on all of her limbs, especially on the elbows and knees, and on the ear helices and nose. There also was notable palmoplantar keratoderma with central sparing (Figure 1), onycholysis, and subungual hyperkeratosis. A skin biopsy of the forearm was performed, and histology revealed orthokeratosis, hypergranulosis and basal vacuolar alteration, and superficial perivascular lymphohistiocytic infiltrate with melanophages (Figure 2). Direct immunofluorescence studies and fungal cultures were negative. The cutaneous features of a treatment-resistant acral dermatosis supported the clinical diagnosis of AP, especially in the setting of an internal malignancy. The patient was started on palliative radiotherapy with no notable resolution of the cutaneous lesions. She was lost to follow-up.

Figure 1. Palmoplantar keratoderma with central sparing of the palms and soles, onycholysis, and subungual hyperkeratosis .

First described by Bazex et al¹ in 1965, AP is an uncommon but well-recognized paraneoplastic dermatosis associated with an underlying neoplasm. Fewer than 160 cases have been reported, and the majority of cases have been men older than 40 years. The most commonly associated malignancy was SCC (at least 50% of reported cases) involving mainly the oropharynx and larynx, with lung and esophageal SCC also described.^2-4 Only a few cases of adenocarcinoma-associated AP have been described, such as adenocarcinoma of the lung, prostate, and stomach.^3-5 In a reported case of AP associated with early colon adenocarcinoma, the patient had remarkable cutaneous resolution following successful tumor resection.⁶ Other reported rare hematologic associations included Hodgkin disease, peripheral T-cell lymphoma, and multiple myeloma.^3-5 Karabulut et al⁴ described a case associated with cholangiocarcinoma and studied the primary sites of malignancies in another 133 patients with AP (118 patients with documented cell type): oropharynx and larynx in 55 (41%); lung in 23 (17%); unknown location in 20 (15%); esophagus in 14 (11%); prostate in 3 (2%); stomach in 3 (2%); and isolated cases involving the liver, thymus, uterus, vulva, breast, urinary bladder, lymph nodes, and bone marrow.

Figure 2. Histologic examination revealed basal vacuolar alteration, hyperkeratosis, and superficial perivascular lymphohistiocytic infiltrate with melanophages (H&amp;E, original magnification ×40).

A review of 113 cases of AP showed that only 15% developed cutaneous lesions after the malignancy had been discovered; in the majority of cases (67%), cutaneous lesions were present for an average of 1 year preceding the diagnosis of malignancy.³ In our patient, the late-onset cutaneous involvement corresponded to the progression of the underlying colon adenocarcinoma. In the absence of initial cutaneous involvement or when successful treatment of a tumor results in cutaneous resolution, subsequent emergence of cutaneous lesions of AP signifies tumor progression. The evolution of cutaneous features in AP was well described by Bazex et al.¹ Stage 1 of the disease shows initial erythema and psoriasiform scaling of the fingers and toes, spreading to the nose and ear helices (hence acrokeratosis); the tumor frequently remains asymptomatic or undetected. Stage 2 shows palmoplantar keratoderma with central sparing and more extensive facial lesions; progression to stage 3 occurs if the tumor remains undetected or untreated, with further spread of psoriasiform lesions to the elbows, knees, trunk, and scalp.¹ Nail involvement occurs in nearly 75% of cases⁵; typical changes include subungual hyperkeratosis, onycholysis, longitudinal streaks, yellow pigmentation, and rarely onychomadesis. A high index of suspicion of AP is paramount when evaluating any recalcitrant acral dermatosis that fails to respond to appropriate therapy, especially in the presence of constitutional symptoms, typical bulbous enlargement of distal phalanges, or isolated involvement of helices. These findings should prompt physicians to perform an extensive search for an underlying malignancy with complete physical examination, particularly of the head and neck region, with appropriate endoscopic assessment and imaging studies.

A myriad of nonspecific histologic features of AP commonly reported include hyperkeratosis, parakeratosis, acanthosis, and dermal perivascular lymphohistiocytic infiltrate.⁷ Less common features include dyskeratotic keratinocytes, vacuolar degeneration, bandlike infiltrate, and melanin incontinence.⁷ The pathogenesis of AP remains elusive. A postulated immunologic mechanism is based on reports of immunoglobulins (IgG, IgA, IgM) and complement (C3) deposition along the basement membrane zone.⁸ Association with autoimmune disorders such as alopecia areata and vitiligo also has been reported.⁹ Another possible mechanism is cross-reactivity between antigens found in the tumor and skin, resulting in a T-cell–mediated immune response to tumorlike antigens in the epidermis, or secretion of tumor-originating growth factors responsible for the hyperkeratotic skin changes, such as epidermal growth factor, transforming growth factor a, or insulinlike growth factor 1.^3,7,10,11

Spontaneous remission of cutaneous lesions in untreated underlying malignancy is rare. Isolated reports of treatment using topical and systemic steroids, salicylic acid, topical vitamin D analogues, etretinate, and psoralen plus UVA showed minimal improvement.^5,7 The mainstay in attaining cutaneous resolution is to detect and eradicate the underlying neoplasm with surgery, chemotherapy, or radiotherapy, or combination therapy.

Our case is noteworthy because of the patient’s gender (female), underlying malignancy (adenocarcinoma of the colon), and late onset of cutaneous involvement, which are all uncommon associations related to paraneoplastic syndrome. The clinical features of AP should be recognized early to facilitate an extensive search for an occult malignancy, and late-onset cutaneous involvement also should be recognized as a marker of tumor relapse or progression.

To the Editor:

Acrokeratosis paraneoplastica (AP) of Bazex is a rare but distinctive acral psoriasiform dermatosis associated with internal malignancy, usually squamous cell carcinoma (SCC), of the upper aerodigestive tract.^1,2 Recognizing this paraneoplastic condition is paramount because cutaneous findings often precede the onset of symptoms associated with an occult malignancy.³

A 76-year-old woman with adenocarcinoma of the transverse colon of 3 years’ duration was referred to the dermatology department. She had a hemicolectomy and was doing well until tumor recurrence with peritoneal metastasis was detected following an exploratory laparotomy 1 year prior to presentation to us. She underwent a total hysterectomy and bilateral salpingo-oophorectomy. Palliative chemotherapy was initiated, and she completed 5 cycles of capecitabine and oxaliplatin. Long-term medications for diabetes mellitus, hypertension, and hyperlipidemia included glibenclamide, nifedipine, hydrochlorothiazide, and losartan. The patient had a progressive pruritic rash of 6 months’ duration that started on the hands and forearms and spread to involve the feet and lower limbs as well as the ears and face. She also experienced progressive thickening of the palms and soles. She had been treated with topical steroids and emollients with no improvement. Clinical examination revealed erythematous scaly patches on all of her limbs, especially on the elbows and knees, and on the ear helices and nose. There also was notable palmoplantar keratoderma with central sparing (Figure 1), onycholysis, and subungual hyperkeratosis. A skin biopsy of the forearm was performed, and histology revealed orthokeratosis, hypergranulosis and basal vacuolar alteration, and superficial perivascular lymphohistiocytic infiltrate with melanophages (Figure 2). Direct immunofluorescence studies and fungal cultures were negative. The cutaneous features of a treatment-resistant acral dermatosis supported the clinical diagnosis of AP, especially in the setting of an internal malignancy. The patient was started on palliative radiotherapy with no notable resolution of the cutaneous lesions. She was lost to follow-up.

Figure 1. Palmoplantar keratoderma with central sparing of the palms and soles, onycholysis, and subungual hyperkeratosis .

First described by Bazex et al¹ in 1965, AP is an uncommon but well-recognized paraneoplastic dermatosis associated with an underlying neoplasm. Fewer than 160 cases have been reported, and the majority of cases have been men older than 40 years. The most commonly associated malignancy was SCC (at least 50% of reported cases) involving mainly the oropharynx and larynx, with lung and esophageal SCC also described.^2-4 Only a few cases of adenocarcinoma-associated AP have been described, such as adenocarcinoma of the lung, prostate, and stomach.^3-5 In a reported case of AP associated with early colon adenocarcinoma, the patient had remarkable cutaneous resolution following successful tumor resection.⁶ Other reported rare hematologic associations included Hodgkin disease, peripheral T-cell lymphoma, and multiple myeloma.^3-5 Karabulut et al⁴ described a case associated with cholangiocarcinoma and studied the primary sites of malignancies in another 133 patients with AP (118 patients with documented cell type): oropharynx and larynx in 55 (41%); lung in 23 (17%); unknown location in 20 (15%); esophagus in 14 (11%); prostate in 3 (2%); stomach in 3 (2%); and isolated cases involving the liver, thymus, uterus, vulva, breast, urinary bladder, lymph nodes, and bone marrow.

Figure 2. Histologic examination revealed basal vacuolar alteration, hyperkeratosis, and superficial perivascular lymphohistiocytic infiltrate with melanophages (H&amp;E, original magnification ×40).

A review of 113 cases of AP showed that only 15% developed cutaneous lesions after the malignancy had been discovered; in the majority of cases (67%), cutaneous lesions were present for an average of 1 year preceding the diagnosis of malignancy.³ In our patient, the late-onset cutaneous involvement corresponded to the progression of the underlying colon adenocarcinoma. In the absence of initial cutaneous involvement or when successful treatment of a tumor results in cutaneous resolution, subsequent emergence of cutaneous lesions of AP signifies tumor progression. The evolution of cutaneous features in AP was well described by Bazex et al.¹ Stage 1 of the disease shows initial erythema and psoriasiform scaling of the fingers and toes, spreading to the nose and ear helices (hence acrokeratosis); the tumor frequently remains asymptomatic or undetected. Stage 2 shows palmoplantar keratoderma with central sparing and more extensive facial lesions; progression to stage 3 occurs if the tumor remains undetected or untreated, with further spread of psoriasiform lesions to the elbows, knees, trunk, and scalp.¹ Nail involvement occurs in nearly 75% of cases⁵; typical changes include subungual hyperkeratosis, onycholysis, longitudinal streaks, yellow pigmentation, and rarely onychomadesis. A high index of suspicion of AP is paramount when evaluating any recalcitrant acral dermatosis that fails to respond to appropriate therapy, especially in the presence of constitutional symptoms, typical bulbous enlargement of distal phalanges, or isolated involvement of helices. These findings should prompt physicians to perform an extensive search for an underlying malignancy with complete physical examination, particularly of the head and neck region, with appropriate endoscopic assessment and imaging studies.

A myriad of nonspecific histologic features of AP commonly reported include hyperkeratosis, parakeratosis, acanthosis, and dermal perivascular lymphohistiocytic infiltrate.⁷ Less common features include dyskeratotic keratinocytes, vacuolar degeneration, bandlike infiltrate, and melanin incontinence.⁷ The pathogenesis of AP remains elusive. A postulated immunologic mechanism is based on reports of immunoglobulins (IgG, IgA, IgM) and complement (C3) deposition along the basement membrane zone.⁸ Association with autoimmune disorders such as alopecia areata and vitiligo also has been reported.⁹ Another possible mechanism is cross-reactivity between antigens found in the tumor and skin, resulting in a T-cell–mediated immune response to tumorlike antigens in the epidermis, or secretion of tumor-originating growth factors responsible for the hyperkeratotic skin changes, such as epidermal growth factor, transforming growth factor a, or insulinlike growth factor 1.^3,7,10,11

Spontaneous remission of cutaneous lesions in untreated underlying malignancy is rare. Isolated reports of treatment using topical and systemic steroids, salicylic acid, topical vitamin D analogues, etretinate, and psoralen plus UVA showed minimal improvement.^5,7 The mainstay in attaining cutaneous resolution is to detect and eradicate the underlying neoplasm with surgery, chemotherapy, or radiotherapy, or combination therapy.

Our case is noteworthy because of the patient’s gender (female), underlying malignancy (adenocarcinoma of the colon), and late onset of cutaneous involvement, which are all uncommon associations related to paraneoplastic syndrome. The clinical features of AP should be recognized early to facilitate an extensive search for an occult malignancy, and late-onset cutaneous involvement also should be recognized as a marker of tumor relapse or progression.

1. Patients with a documented contact allergy to caine mix should avoid all of the following except:

a. benzocaine
b. butacaine
c. lidocaine
d. procaine
e. tetracaine

2. A patient with atopic dermatitis whose condition is worsening with the use of topical steroids is referred for patch testing and found to have positivity to budesonide. Which of the following would be an appropriate topical steroid to prescribe to this patient?

a. desonide
b. desoximetasone
c. fluocinolone
d. fluocinonide
e. triamcinolone

3. A patient with a documented contact allergy to carba mix should avoid which of the following systemic medications?

a. ciprofloxacin
b. disulfiram
c. gold sodium thiomalate
d. hydroxyzine
e. piroxicam

4. Bronopol (2-bromo-2-nitropropane-1,3-diol) can cross-react with all of the following except:

a. diazolidinyl urea
b. DMDM hydantoin
c. imidazolidinyl urea
d. p-aminobenzoic acid
e. quaternium-15

5. Allergy to cocamidopropyl betaine is thought to be secondary to which of the following?

a. amidoamine
b. benzoic acid
c. bronopol
d. Myroxylon pereirae
e. N-isopropyl-N'-phenyl parapheylenediamine

1. Patients with a documented contact allergy to caine mix should avoid all of the following except:

a. benzocaine
b. butacaine
c. lidocaine
d. procaine
e. tetracaine

2. A patient with atopic dermatitis whose condition is worsening with the use of topical steroids is referred for patch testing and found to have positivity to budesonide. Which of the following would be an appropriate topical steroid to prescribe to this patient?

a. desonide
b. desoximetasone
c. fluocinolone
d. fluocinonide
e. triamcinolone

3. A patient with a documented contact allergy to carba mix should avoid which of the following systemic medications?

a. ciprofloxacin
b. disulfiram
c. gold sodium thiomalate
d. hydroxyzine
e. piroxicam

4. Bronopol (2-bromo-2-nitropropane-1,3-diol) can cross-react with all of the following except:

a. diazolidinyl urea
b. DMDM hydantoin
c. imidazolidinyl urea
d. p-aminobenzoic acid
e. quaternium-15

5. Allergy to cocamidopropyl betaine is thought to be secondary to which of the following?

a. amidoamine
b. benzoic acid
c. bronopol
d. Myroxylon pereirae
e. N-isopropyl-N'-phenyl parapheylenediamine

1. Patients with a documented contact allergy to caine mix should avoid all of the following except:

a. benzocaine
b. butacaine
c. lidocaine
d. procaine
e. tetracaine

2. A patient with atopic dermatitis whose condition is worsening with the use of topical steroids is referred for patch testing and found to have positivity to budesonide. Which of the following would be an appropriate topical steroid to prescribe to this patient?

a. desonide
b. desoximetasone
c. fluocinolone
d. fluocinonide
e. triamcinolone

3. A patient with a documented contact allergy to carba mix should avoid which of the following systemic medications?

a. ciprofloxacin
b. disulfiram
c. gold sodium thiomalate
d. hydroxyzine
e. piroxicam

4. Bronopol (2-bromo-2-nitropropane-1,3-diol) can cross-react with all of the following except:

a. diazolidinyl urea
b. DMDM hydantoin
c. imidazolidinyl urea
d. p-aminobenzoic acid
e. quaternium-15

5. Allergy to cocamidopropyl betaine is thought to be secondary to which of the following?

a. amidoamine
b. benzoic acid
c. bronopol
d. Myroxylon pereirae
e. N-isopropyl-N'-phenyl parapheylenediamine