Hair & Nails

Hair loss has various etiologies. Correct diagnosis of hair disorders is complex and requires the evaluation of clinical presentation, history, physical examination, and laboratory test results. In the patient with a sexually transmitted disease (STD), alopecia may be an important associated finding and can provide clues to diagnosis. This review focuses on the relationship between hair loss and STDs. Specifically, we review alopecia in association with syphilis and human immunodeficiency virus (HIV) infection and the medications used to treat these infections. In addition, we review the literature regarding the putative association between alopecia areata and cytomegalovirus (CMV). There are multiple mechanisms involved in hair loss in these diseases, including the diseases themselves, systemic sequelae of these infections, autoimmune phenomena, and side effects of medications. 

Syphilis

When considering the STDs associated with hair loss, syphilis is usually the first STD described because of the large incidence of the disease and its many reported cases of associated hair loss. This is especially important due to the increasing number of current cases of syphilis. Hair loss does not occur in primary syphilis except when associated with a primary chancre of scalp. Hair loss in secondary syphilis, also known as latent syphilis, occurs infrequently; various series report an incidence of 2.9% to 7%.^1,2 There are 2 types of secondary syphilitic alopecia. The first is an uncommon symptomatic type found in association with an actual secondary lesion (usually papulosquamous) on the scalp. The second is termed essential syphilitic alopecia, which designates hair loss in the absence of visible syphilitic scalp lesions. Essential syphilitic alopecia has been divided into 3 types: the classic patchy "moth-eaten" alopecia (Figure), a generalized thinning of the hair, and the moth-eaten type in combination with general thinning of the hair. Of these, patchy moth-eaten alopecia occurs most frequently. The diffuse hair loss of essential syphilitic alopecia as the only manifestation of syphilis is uncommon. Cuozzo et al³ described 2 patients in whom the first sign of disease was alopecia.

PLEASE REFER TO THE PDF TO VIEW THE FIGURE

Moth-eaten alopecia of syphilis is a characteristic manifestation of secondary syphilis that usually affects the scalp and occasionally other areas such as the eyebrows, beard, and pubic area.⁴ This form of alopecia may be confused with trichotillomania, traction alopecia, and alopecia areata.⁵ Pareek⁴ described a case of an unusual location of patchy moth-eaten alopecia that presented on the anterior side of the lower legs of a 30-year-old man in conjunction with patchy alopecia on the scalp and thinning of the eyebrows. With penicillin administration, hair of the legs, scalp, and eyebrows started to grow; the hair was fully regrown within 6 months, which suggests good prognosis with treatment instigation for syphilitic alopecia of all areas.

Jordaan and Louw⁵ systematically documented the histopathologic features of 12 patients with moth-eaten alopecia. Characteristic features included follicular plugging; a sparse, perivascular and perifollicular lymphocytic infiltrate; telogenization; and follicle-oriented melanin clumping.⁵ van der Willigen et al⁶ conducted a study of hair roots in 11 and 8 patients with primary and secondary syphilis, respectively. A decreased number of anagen hair roots; an increased number of catagen hair roots, dysplastic/dystrophic hair roots, and anagen hair roots with sheaths; and more than 20% angulation were observed in both groups.⁶ In addition, Lee and Hsu⁷ noted the histopathologic similarity between alopecia syphilitica and alopecia areata. They reported the histopathologic findings of alopecia syphilitica from 9 patients with secondary syphilis and acute hair loss. The alopecia was moth-eaten in 4 patients and was diffuse but slightly moth-eaten in 5. Microscopically, the dermoepidermal interface was not involved. The number of hair follicles was diminished, with increased numbers of catagens and telogens. Lymphocytic infiltration was present around the hair bulbs and fibrous tracts in 8 patients, and plasma cells were present in 4 biopsy specimens. Except for the follicular changes, the findings resembled those of macular/maculopapular syphilides outside the scalp. With the follicular changes, the overall patterns closely resembled alopecia areata. Results of the modified Steiner stain did not reveal spirochetes in any of the patients and failed to differentiate between alopecia syphilitica and alopecia areata. Comparing the alopecia syphilitica patients with 13 patients with alopecia areata, the authors found only a few differentiating features. They concluded that the presence of peribulbar eosinophils strongly suggests alopecia areata.⁷ Without peribulbar eosinophils, the presence of plasma cells, abundant lymphocytes in the isthmus, or peribulbar lymphoid aggregates suggests alopecia syphilitica. Elston et al⁸ observed several cases of syphilis with numerous eosinophils in the peribulbar infiltrate and noted that it can be indistinguishable from alopecia areata.

When an associated skin rash or lymphadenopathy is present, the diagnosis of syphilis may be suggested and confirmed by positive serology test results. If such findings are not present, a biopsy specimen to differentiate from other forms of alopecia should be obtained. Because moth-eaten alopecia and alopecia areata have similar resemblance microscopically, syphilis serologic tests are needed.

The treatment of syphilis also has been shown to be a cause of alopecia. Pareek⁹ described the association of syphilitic alopecia and Herxheimer reaction. A 25-year-old man presented with syphilis with widespread thinning of the scalp hair, eyebrows, and pubic area; the scalp showed patchy moth-eaten alopecia. He was treated with 1 to 2 megaunits of procaine penicillin daily for 10 days. Six hours after the first injection, the patient's temperature rose to 103°F; in addition to malaise, headache, flush, and sore throat, he had a transient skin rash and marked loss of hair. All the symptoms disappeared by the next day. Two to 3 weeks later, the lymphadenopathy had disappeared, and the patient's eyebrows and pubic hair started to regrow. The scalp hair was fully regrown 10 weeks from the onset of treatment. The author concluded that diffuse and extensive hair loss after the first injection of penicillin was part of the Herxheimer reaction.⁹

HIV

Hair loss is common in patients with HIV; in black patients, this loss may be associated with hair straightening.¹⁰ Possible causes of hair loss frequently are present in patients with HIV, including chronic HIV infection itself, acute and chronic systemic infections, local infections, nutritional deficits, immune and endocrine dysregulation, and exposure to multiple drugs.¹⁰ Alopecia areata and alopecia universalis also have been reported in patients with HIV.^11-14

Smith et al¹⁰ studied and reviewed the clinical and histopathologic features of hair loss in 10 patients with HIV. They noted that the most characteristic change in the hair of patients with HIV was hair loss with straightening, sometimes associated with fine hair texture and an increased tendency for broken hairs. These changes are seen in late-stage disease, most commonly in black patients. Each patient had telogen effluvium, and it was observed that any chronic or acute infection (including HIV) can lead to this condition. Nutritional deficits, often prominent in HIV patients, may lead to or potentiate telogen effluvium. Secondary infections and changes in bowel mucosa may lead to specific nutritional deficiencies even before evidence of clinical wasting is seen. In addition to caloric and protein malnutrition that may affect hair growth, minerals such as copper, zinc, and selenium are decreased in patients with HIV. Elevated levels of interleukin 6 and tumor necrosis factor α, which increase epidermal proliferation, may predispose patients to abnormal keratinization by increasing the proliferative rate and nutritional requirements.¹⁰

Endocrine regulation is another important factor in hair growth. In late-stage HIV disease, androgen levels decrease while estradiol levels increase. Although thyroid hormone levels are normal in advanced HIV, thyroid functions are elevated to more than expected for the amount of wasting and may contribute to the change of hair texture,10 autoimmune mechanism, associated diseases, and HIV medication side effects.

In the Smith et al¹⁰ study, scanning electron microscopy was performed on plucked and pulled hairs of 10 patients with late-stage HIV-1 infection. In addition, scalp biopsy specimens were examined in both vertical and transverse sections. All patients had telogen effluvium. Numerous apoptotic or necrotic keratinocytes were seen in the upper external root sheath follicular epithelium; a mild to moderate perifollicular mononuclear cell infiltrate, often containing eosinophils, also was seen. Additionally, the mononuclear infiltrate was seen surrounding and within the basaloid cells of the follicles in telogen phase; the midfollicular area had the most marked inflammatory infiltrate. Variable dystrophy of the hair shafts also was a consistent feature. Although telogen effluvium is a common response to a wide spectrum of biologic stresses, the presence of apoptotic or necrotic keratinocytes within the upper end of the external root sheath epithelium, as well as dystrophy of hairs, may be markers of hair loss in patients with HIV-1 infection.¹⁰

Autoimmune alopecia, including alopecia areata and alopecia universalis, can be seen in association with HIV.^11-15 Ostlere et al¹¹ first reported a case of alopecia universalis that developed in a patient 2 years after HIV antibody was detected. The patient showed loss of all scalp hair, eyelashes, eyebrows, and body hair. Two possible mechanisms for the development of alopecia were suggested. The first was that HIV induced nonspecific polyclonal B-cell activation with production of autoantibody either directly or via activated T cells; this supports a humoral theory of alopecia areata pathogenesis. Alternatively, the authors postulated that HIV induced a change in the balance between helper and suppressor cells, which resulted in aberrant cell-mediated immune effect at the hair follicles.¹¹ Werninghaus and Kaminer¹² described a similar patient with alopecia universalis; a biopsy specimen revealed perifollicular fibrosis without inflammation.

Stewart and Smoller¹³ described an HIV-positive patient with altered T-lymphocyte subsets in whom alopecia universalis developed. Results of a skin biopsy of the patient's scalp demonstrated a classic perifollicular lymphocytic infiltrate; results of immunophenotyping of the same specimen revealed that most cells were CD4⁺ lymphocytes. During the active loss of hair, the patient's ratio of CD4/CD8 cells was decreased; however, the ratio normalized during the period of hair regrowth. Their data suggested that systemic immune dysfunction, as seen in HIV infection, may be more important in mediating alopecia areata than localized immune responses. Because of the proposed mechanism of alopecia areata developing in this patient (ie, influx of CD4⁺ lymphocytes to the perifollicular regions of skin when the CD4/CD8 cells ratio is low), the authors were surprised that alopecia areata is not more common in patients with HIV infection.¹³

Cho et al¹⁴ described the association of vitiligo and alopecia areata in patients with HIV. They noted that the development of autoimmune diseases, though not life threatening, is an interesting phenomenon that may result from immune dysfunction or from B-cell infection by HIV, Epstein-Barr virus, or other unknown viruses. They described a 47-year-old man who had vitiligo and alopecia areata approximately 2 years after testing positive for HIV antibodies.¹⁴ Grossman et al¹⁵ described an HIV-seropositive man with acquired eyelash trichomegaly and alopecia areata. They noted that this combination of clinical manifestations is intriguing because the new onset of elongated eyelashes in patients with acquired immunodeficiency syndrome usually has been associated with severe immunosuppression, and alopecia areata has a presumed autoimmune etiology that requires T-cell activation. They concluded that the occurrence of these dichotomous conditions illustrates the potential selective pathogenesis of progressive HIV infection.¹⁵

Medications used in the treatment of HIV can play a role in hair loss. Geletko et al¹⁶ reported a 33-year-old HIV-infected man who developed alopecia areata after beginning therapy with zidovudine, a nucleoside analogue reverse transcriptase inhibitor. The alopecia reversed after the drug was discontinued. The authors proposed that patients with lower CD4⁺ counts may be more predisposed to zidovudine-induced alopecia than those in the earlier stages of HIV with higher CD4⁺ counts.¹⁶

Indinavir-related alopecia was described by d'Arminio Monforte et al.¹⁷ Of 337 patients given indinavir in combination with nucleoside analogues, 5 patients with HIV developed severe alopecia, which was evident clinically after a mean of 50 days of treatment. All patients were receiving triple therapy that included indinavir. Three patients had diffuse shedding of hair involving the entire scalp, and 2 had circumscribed circular areas of alopecia resulting in complete severe hair loss.¹⁷ Bouscarat et al¹⁸ reported 10 more cases of hair loss associated with indinavir therapy in patients receiving triple antiviral treatment that included indinavir. Hair loss developed during the first 6 months of indinavir therapy and initially involved the lower limbs. Progressive hair regrowth occurred within 4 months after indinavir was replaced by other treatments.¹⁸

Ginarte et al¹⁹ described significant alopecia induced by indinavir plus ritonavir therapy in 3 patients a few weeks after beginning treatment. The authors noted that patients receiving indinavir often experience retinoidlike effects such as alopecia, xerosis, and cheilitis. Nonscarring alopecia can develop in patients receiving indinavir, with or without retinoid effects.¹⁹ Hair loss also has been noted with the use of crixivan.²⁰

CMV

CMV is a prevalent viral pathogen.²¹ Most people with acute CMV experience an inapparent infection. The virus usually is spread through close personal contact, including sexual transmission. There has been debate over the link of alopecia areata with CMV. In 1995, Skinner et al²² described using polymerase chain reaction (PCR) techniques to find evidence of CMV DNA in paraffin block sections of lesions of alopecia areata. Of 21 patient biopsy specimens, 10 had alopecia areata and 11 had other hair loss conditions. Of the 10 alopecia areata samples, 9 were positive for CMV; no other hair loss samples were positive for CMV.²² Skinner et al²³ theorized that CMV may achieve latency in the hair root. Reactivation of CMV was thought to be one of the pathogenic mechanisms in alopecia areata; the authors argued that a lymphocytic surveillance of not-quite-latent CMV would explain much of the behavior of alopecia areata, which has a tendency for intermittent relapses and remissions.²³

The association between alopecia areata and CMV was refuted by Garcia-Hernandez et al,²⁴ who used 3 different PCR assays to detect CMV DNA in skin punch biopsy specimens of 3 patient groups: 40 patients with alopecia areata, 3 patients with HIV and alopecia areata, and 12 patients with other types of alopecia. PCR assays are known to be the most sensitive assay for CMV detection; this study used different PCR assays to achieve maximum sensitivity for CMV. No CMV DNA amplification was found in any of the specimens.²⁴

Offidani et al²⁵ further contradicted this association. The purpose of their study was to clarify the role of CMV infection and to demonstrate the absence of replication of other autoimmune disease–related herpesviruses (eg, Epstein-Barr virus) in the pathogenesis of alopecia areata. After extraction of mRNA from tissue samples of 4 patients with active patchy alopecia areata, reverse transcriptase PCR was carried out using primers specific for some viral members of the β Herpesviridae subfamily of the Herpesviridae family (eg, CMV, Epstein-Barr virus, herpes simplex virus). The authors could not detect any replication of the CMV or other β Herpesviridae in the samples collected, which supports the hypothesis that CMV is not the triggering factor in alopecia areata, neither as a reactivator of the immune response nor as a trigger of the autoimmunity.²⁵

Conclusion

Although many etiologies exist for hair loss, STDs should not be overlooked in a sexually active patient presenting with an otherwise unexplainable cause of this condition. A full workup, including clinical history, physical examination, and laboratory tests, should include STDs in the differential diagnosis (Table).

PLEASE REFER TO THE PDF TO VIEW THE TABLE

Hair loss has various etiologies. Correct diagnosis of hair disorders is complex and requires the evaluation of clinical presentation, history, physical examination, and laboratory test results. In the patient with a sexually transmitted disease (STD), alopecia may be an important associated finding and can provide clues to diagnosis. This review focuses on the relationship between hair loss and STDs. Specifically, we review alopecia in association with syphilis and human immunodeficiency virus (HIV) infection and the medications used to treat these infections. In addition, we review the literature regarding the putative association between alopecia areata and cytomegalovirus (CMV). There are multiple mechanisms involved in hair loss in these diseases, including the diseases themselves, systemic sequelae of these infections, autoimmune phenomena, and side effects of medications. 

Syphilis

When considering the STDs associated with hair loss, syphilis is usually the first STD described because of the large incidence of the disease and its many reported cases of associated hair loss. This is especially important due to the increasing number of current cases of syphilis. Hair loss does not occur in primary syphilis except when associated with a primary chancre of scalp. Hair loss in secondary syphilis, also known as latent syphilis, occurs infrequently; various series report an incidence of 2.9% to 7%.^1,2 There are 2 types of secondary syphilitic alopecia. The first is an uncommon symptomatic type found in association with an actual secondary lesion (usually papulosquamous) on the scalp. The second is termed essential syphilitic alopecia, which designates hair loss in the absence of visible syphilitic scalp lesions. Essential syphilitic alopecia has been divided into 3 types: the classic patchy "moth-eaten" alopecia (Figure), a generalized thinning of the hair, and the moth-eaten type in combination with general thinning of the hair. Of these, patchy moth-eaten alopecia occurs most frequently. The diffuse hair loss of essential syphilitic alopecia as the only manifestation of syphilis is uncommon. Cuozzo et al³ described 2 patients in whom the first sign of disease was alopecia.

PLEASE REFER TO THE PDF TO VIEW THE FIGURE

Moth-eaten alopecia of syphilis is a characteristic manifestation of secondary syphilis that usually affects the scalp and occasionally other areas such as the eyebrows, beard, and pubic area.⁴ This form of alopecia may be confused with trichotillomania, traction alopecia, and alopecia areata.⁵ Pareek⁴ described a case of an unusual location of patchy moth-eaten alopecia that presented on the anterior side of the lower legs of a 30-year-old man in conjunction with patchy alopecia on the scalp and thinning of the eyebrows. With penicillin administration, hair of the legs, scalp, and eyebrows started to grow; the hair was fully regrown within 6 months, which suggests good prognosis with treatment instigation for syphilitic alopecia of all areas.

Jordaan and Louw⁵ systematically documented the histopathologic features of 12 patients with moth-eaten alopecia. Characteristic features included follicular plugging; a sparse, perivascular and perifollicular lymphocytic infiltrate; telogenization; and follicle-oriented melanin clumping.⁵ van der Willigen et al⁶ conducted a study of hair roots in 11 and 8 patients with primary and secondary syphilis, respectively. A decreased number of anagen hair roots; an increased number of catagen hair roots, dysplastic/dystrophic hair roots, and anagen hair roots with sheaths; and more than 20% angulation were observed in both groups.⁶ In addition, Lee and Hsu⁷ noted the histopathologic similarity between alopecia syphilitica and alopecia areata. They reported the histopathologic findings of alopecia syphilitica from 9 patients with secondary syphilis and acute hair loss. The alopecia was moth-eaten in 4 patients and was diffuse but slightly moth-eaten in 5. Microscopically, the dermoepidermal interface was not involved. The number of hair follicles was diminished, with increased numbers of catagens and telogens. Lymphocytic infiltration was present around the hair bulbs and fibrous tracts in 8 patients, and plasma cells were present in 4 biopsy specimens. Except for the follicular changes, the findings resembled those of macular/maculopapular syphilides outside the scalp. With the follicular changes, the overall patterns closely resembled alopecia areata. Results of the modified Steiner stain did not reveal spirochetes in any of the patients and failed to differentiate between alopecia syphilitica and alopecia areata. Comparing the alopecia syphilitica patients with 13 patients with alopecia areata, the authors found only a few differentiating features. They concluded that the presence of peribulbar eosinophils strongly suggests alopecia areata.⁷ Without peribulbar eosinophils, the presence of plasma cells, abundant lymphocytes in the isthmus, or peribulbar lymphoid aggregates suggests alopecia syphilitica. Elston et al⁸ observed several cases of syphilis with numerous eosinophils in the peribulbar infiltrate and noted that it can be indistinguishable from alopecia areata.

When an associated skin rash or lymphadenopathy is present, the diagnosis of syphilis may be suggested and confirmed by positive serology test results. If such findings are not present, a biopsy specimen to differentiate from other forms of alopecia should be obtained. Because moth-eaten alopecia and alopecia areata have similar resemblance microscopically, syphilis serologic tests are needed.

The treatment of syphilis also has been shown to be a cause of alopecia. Pareek⁹ described the association of syphilitic alopecia and Herxheimer reaction. A 25-year-old man presented with syphilis with widespread thinning of the scalp hair, eyebrows, and pubic area; the scalp showed patchy moth-eaten alopecia. He was treated with 1 to 2 megaunits of procaine penicillin daily for 10 days. Six hours after the first injection, the patient's temperature rose to 103°F; in addition to malaise, headache, flush, and sore throat, he had a transient skin rash and marked loss of hair. All the symptoms disappeared by the next day. Two to 3 weeks later, the lymphadenopathy had disappeared, and the patient's eyebrows and pubic hair started to regrow. The scalp hair was fully regrown 10 weeks from the onset of treatment. The author concluded that diffuse and extensive hair loss after the first injection of penicillin was part of the Herxheimer reaction.⁹

HIV

Hair loss is common in patients with HIV; in black patients, this loss may be associated with hair straightening.¹⁰ Possible causes of hair loss frequently are present in patients with HIV, including chronic HIV infection itself, acute and chronic systemic infections, local infections, nutritional deficits, immune and endocrine dysregulation, and exposure to multiple drugs.¹⁰ Alopecia areata and alopecia universalis also have been reported in patients with HIV.^11-14

Smith et al¹⁰ studied and reviewed the clinical and histopathologic features of hair loss in 10 patients with HIV. They noted that the most characteristic change in the hair of patients with HIV was hair loss with straightening, sometimes associated with fine hair texture and an increased tendency for broken hairs. These changes are seen in late-stage disease, most commonly in black patients. Each patient had telogen effluvium, and it was observed that any chronic or acute infection (including HIV) can lead to this condition. Nutritional deficits, often prominent in HIV patients, may lead to or potentiate telogen effluvium. Secondary infections and changes in bowel mucosa may lead to specific nutritional deficiencies even before evidence of clinical wasting is seen. In addition to caloric and protein malnutrition that may affect hair growth, minerals such as copper, zinc, and selenium are decreased in patients with HIV. Elevated levels of interleukin 6 and tumor necrosis factor α, which increase epidermal proliferation, may predispose patients to abnormal keratinization by increasing the proliferative rate and nutritional requirements.¹⁰

Endocrine regulation is another important factor in hair growth. In late-stage HIV disease, androgen levels decrease while estradiol levels increase. Although thyroid hormone levels are normal in advanced HIV, thyroid functions are elevated to more than expected for the amount of wasting and may contribute to the change of hair texture,10 autoimmune mechanism, associated diseases, and HIV medication side effects.

In the Smith et al¹⁰ study, scanning electron microscopy was performed on plucked and pulled hairs of 10 patients with late-stage HIV-1 infection. In addition, scalp biopsy specimens were examined in both vertical and transverse sections. All patients had telogen effluvium. Numerous apoptotic or necrotic keratinocytes were seen in the upper external root sheath follicular epithelium; a mild to moderate perifollicular mononuclear cell infiltrate, often containing eosinophils, also was seen. Additionally, the mononuclear infiltrate was seen surrounding and within the basaloid cells of the follicles in telogen phase; the midfollicular area had the most marked inflammatory infiltrate. Variable dystrophy of the hair shafts also was a consistent feature. Although telogen effluvium is a common response to a wide spectrum of biologic stresses, the presence of apoptotic or necrotic keratinocytes within the upper end of the external root sheath epithelium, as well as dystrophy of hairs, may be markers of hair loss in patients with HIV-1 infection.¹⁰

Autoimmune alopecia, including alopecia areata and alopecia universalis, can be seen in association with HIV.^11-15 Ostlere et al¹¹ first reported a case of alopecia universalis that developed in a patient 2 years after HIV antibody was detected. The patient showed loss of all scalp hair, eyelashes, eyebrows, and body hair. Two possible mechanisms for the development of alopecia were suggested. The first was that HIV induced nonspecific polyclonal B-cell activation with production of autoantibody either directly or via activated T cells; this supports a humoral theory of alopecia areata pathogenesis. Alternatively, the authors postulated that HIV induced a change in the balance between helper and suppressor cells, which resulted in aberrant cell-mediated immune effect at the hair follicles.¹¹ Werninghaus and Kaminer¹² described a similar patient with alopecia universalis; a biopsy specimen revealed perifollicular fibrosis without inflammation.

Stewart and Smoller¹³ described an HIV-positive patient with altered T-lymphocyte subsets in whom alopecia universalis developed. Results of a skin biopsy of the patient's scalp demonstrated a classic perifollicular lymphocytic infiltrate; results of immunophenotyping of the same specimen revealed that most cells were CD4⁺ lymphocytes. During the active loss of hair, the patient's ratio of CD4/CD8 cells was decreased; however, the ratio normalized during the period of hair regrowth. Their data suggested that systemic immune dysfunction, as seen in HIV infection, may be more important in mediating alopecia areata than localized immune responses. Because of the proposed mechanism of alopecia areata developing in this patient (ie, influx of CD4⁺ lymphocytes to the perifollicular regions of skin when the CD4/CD8 cells ratio is low), the authors were surprised that alopecia areata is not more common in patients with HIV infection.¹³

Cho et al¹⁴ described the association of vitiligo and alopecia areata in patients with HIV. They noted that the development of autoimmune diseases, though not life threatening, is an interesting phenomenon that may result from immune dysfunction or from B-cell infection by HIV, Epstein-Barr virus, or other unknown viruses. They described a 47-year-old man who had vitiligo and alopecia areata approximately 2 years after testing positive for HIV antibodies.¹⁴ Grossman et al¹⁵ described an HIV-seropositive man with acquired eyelash trichomegaly and alopecia areata. They noted that this combination of clinical manifestations is intriguing because the new onset of elongated eyelashes in patients with acquired immunodeficiency syndrome usually has been associated with severe immunosuppression, and alopecia areata has a presumed autoimmune etiology that requires T-cell activation. They concluded that the occurrence of these dichotomous conditions illustrates the potential selective pathogenesis of progressive HIV infection.¹⁵

Medications used in the treatment of HIV can play a role in hair loss. Geletko et al¹⁶ reported a 33-year-old HIV-infected man who developed alopecia areata after beginning therapy with zidovudine, a nucleoside analogue reverse transcriptase inhibitor. The alopecia reversed after the drug was discontinued. The authors proposed that patients with lower CD4⁺ counts may be more predisposed to zidovudine-induced alopecia than those in the earlier stages of HIV with higher CD4⁺ counts.¹⁶

Indinavir-related alopecia was described by d'Arminio Monforte et al.¹⁷ Of 337 patients given indinavir in combination with nucleoside analogues, 5 patients with HIV developed severe alopecia, which was evident clinically after a mean of 50 days of treatment. All patients were receiving triple therapy that included indinavir. Three patients had diffuse shedding of hair involving the entire scalp, and 2 had circumscribed circular areas of alopecia resulting in complete severe hair loss.¹⁷ Bouscarat et al¹⁸ reported 10 more cases of hair loss associated with indinavir therapy in patients receiving triple antiviral treatment that included indinavir. Hair loss developed during the first 6 months of indinavir therapy and initially involved the lower limbs. Progressive hair regrowth occurred within 4 months after indinavir was replaced by other treatments.¹⁸

Ginarte et al¹⁹ described significant alopecia induced by indinavir plus ritonavir therapy in 3 patients a few weeks after beginning treatment. The authors noted that patients receiving indinavir often experience retinoidlike effects such as alopecia, xerosis, and cheilitis. Nonscarring alopecia can develop in patients receiving indinavir, with or without retinoid effects.¹⁹ Hair loss also has been noted with the use of crixivan.²⁰

CMV

CMV is a prevalent viral pathogen.²¹ Most people with acute CMV experience an inapparent infection. The virus usually is spread through close personal contact, including sexual transmission. There has been debate over the link of alopecia areata with CMV. In 1995, Skinner et al²² described using polymerase chain reaction (PCR) techniques to find evidence of CMV DNA in paraffin block sections of lesions of alopecia areata. Of 21 patient biopsy specimens, 10 had alopecia areata and 11 had other hair loss conditions. Of the 10 alopecia areata samples, 9 were positive for CMV; no other hair loss samples were positive for CMV.²² Skinner et al²³ theorized that CMV may achieve latency in the hair root. Reactivation of CMV was thought to be one of the pathogenic mechanisms in alopecia areata; the authors argued that a lymphocytic surveillance of not-quite-latent CMV would explain much of the behavior of alopecia areata, which has a tendency for intermittent relapses and remissions.²³

The association between alopecia areata and CMV was refuted by Garcia-Hernandez et al,²⁴ who used 3 different PCR assays to detect CMV DNA in skin punch biopsy specimens of 3 patient groups: 40 patients with alopecia areata, 3 patients with HIV and alopecia areata, and 12 patients with other types of alopecia. PCR assays are known to be the most sensitive assay for CMV detection; this study used different PCR assays to achieve maximum sensitivity for CMV. No CMV DNA amplification was found in any of the specimens.²⁴

Offidani et al²⁵ further contradicted this association. The purpose of their study was to clarify the role of CMV infection and to demonstrate the absence of replication of other autoimmune disease–related herpesviruses (eg, Epstein-Barr virus) in the pathogenesis of alopecia areata. After extraction of mRNA from tissue samples of 4 patients with active patchy alopecia areata, reverse transcriptase PCR was carried out using primers specific for some viral members of the β Herpesviridae subfamily of the Herpesviridae family (eg, CMV, Epstein-Barr virus, herpes simplex virus). The authors could not detect any replication of the CMV or other β Herpesviridae in the samples collected, which supports the hypothesis that CMV is not the triggering factor in alopecia areata, neither as a reactivator of the immune response nor as a trigger of the autoimmunity.²⁵

Conclusion

Although many etiologies exist for hair loss, STDs should not be overlooked in a sexually active patient presenting with an otherwise unexplainable cause of this condition. A full workup, including clinical history, physical examination, and laboratory tests, should include STDs in the differential diagnosis (Table).

PLEASE REFER TO THE PDF TO VIEW THE TABLE

Hair loss has various etiologies. Correct diagnosis of hair disorders is complex and requires the evaluation of clinical presentation, history, physical examination, and laboratory test results. In the patient with a sexually transmitted disease (STD), alopecia may be an important associated finding and can provide clues to diagnosis. This review focuses on the relationship between hair loss and STDs. Specifically, we review alopecia in association with syphilis and human immunodeficiency virus (HIV) infection and the medications used to treat these infections. In addition, we review the literature regarding the putative association between alopecia areata and cytomegalovirus (CMV). There are multiple mechanisms involved in hair loss in these diseases, including the diseases themselves, systemic sequelae of these infections, autoimmune phenomena, and side effects of medications. 

Syphilis

When considering the STDs associated with hair loss, syphilis is usually the first STD described because of the large incidence of the disease and its many reported cases of associated hair loss. This is especially important due to the increasing number of current cases of syphilis. Hair loss does not occur in primary syphilis except when associated with a primary chancre of scalp. Hair loss in secondary syphilis, also known as latent syphilis, occurs infrequently; various series report an incidence of 2.9% to 7%.^1,2 There are 2 types of secondary syphilitic alopecia. The first is an uncommon symptomatic type found in association with an actual secondary lesion (usually papulosquamous) on the scalp. The second is termed essential syphilitic alopecia, which designates hair loss in the absence of visible syphilitic scalp lesions. Essential syphilitic alopecia has been divided into 3 types: the classic patchy "moth-eaten" alopecia (Figure), a generalized thinning of the hair, and the moth-eaten type in combination with general thinning of the hair. Of these, patchy moth-eaten alopecia occurs most frequently. The diffuse hair loss of essential syphilitic alopecia as the only manifestation of syphilis is uncommon. Cuozzo et al³ described 2 patients in whom the first sign of disease was alopecia.

PLEASE REFER TO THE PDF TO VIEW THE FIGURE

Moth-eaten alopecia of syphilis is a characteristic manifestation of secondary syphilis that usually affects the scalp and occasionally other areas such as the eyebrows, beard, and pubic area.⁴ This form of alopecia may be confused with trichotillomania, traction alopecia, and alopecia areata.⁵ Pareek⁴ described a case of an unusual location of patchy moth-eaten alopecia that presented on the anterior side of the lower legs of a 30-year-old man in conjunction with patchy alopecia on the scalp and thinning of the eyebrows. With penicillin administration, hair of the legs, scalp, and eyebrows started to grow; the hair was fully regrown within 6 months, which suggests good prognosis with treatment instigation for syphilitic alopecia of all areas.

Jordaan and Louw⁵ systematically documented the histopathologic features of 12 patients with moth-eaten alopecia. Characteristic features included follicular plugging; a sparse, perivascular and perifollicular lymphocytic infiltrate; telogenization; and follicle-oriented melanin clumping.⁵ van der Willigen et al⁶ conducted a study of hair roots in 11 and 8 patients with primary and secondary syphilis, respectively. A decreased number of anagen hair roots; an increased number of catagen hair roots, dysplastic/dystrophic hair roots, and anagen hair roots with sheaths; and more than 20% angulation were observed in both groups.⁶ In addition, Lee and Hsu⁷ noted the histopathologic similarity between alopecia syphilitica and alopecia areata. They reported the histopathologic findings of alopecia syphilitica from 9 patients with secondary syphilis and acute hair loss. The alopecia was moth-eaten in 4 patients and was diffuse but slightly moth-eaten in 5. Microscopically, the dermoepidermal interface was not involved. The number of hair follicles was diminished, with increased numbers of catagens and telogens. Lymphocytic infiltration was present around the hair bulbs and fibrous tracts in 8 patients, and plasma cells were present in 4 biopsy specimens. Except for the follicular changes, the findings resembled those of macular/maculopapular syphilides outside the scalp. With the follicular changes, the overall patterns closely resembled alopecia areata. Results of the modified Steiner stain did not reveal spirochetes in any of the patients and failed to differentiate between alopecia syphilitica and alopecia areata. Comparing the alopecia syphilitica patients with 13 patients with alopecia areata, the authors found only a few differentiating features. They concluded that the presence of peribulbar eosinophils strongly suggests alopecia areata.⁷ Without peribulbar eosinophils, the presence of plasma cells, abundant lymphocytes in the isthmus, or peribulbar lymphoid aggregates suggests alopecia syphilitica. Elston et al⁸ observed several cases of syphilis with numerous eosinophils in the peribulbar infiltrate and noted that it can be indistinguishable from alopecia areata.

When an associated skin rash or lymphadenopathy is present, the diagnosis of syphilis may be suggested and confirmed by positive serology test results. If such findings are not present, a biopsy specimen to differentiate from other forms of alopecia should be obtained. Because moth-eaten alopecia and alopecia areata have similar resemblance microscopically, syphilis serologic tests are needed.

The treatment of syphilis also has been shown to be a cause of alopecia. Pareek⁹ described the association of syphilitic alopecia and Herxheimer reaction. A 25-year-old man presented with syphilis with widespread thinning of the scalp hair, eyebrows, and pubic area; the scalp showed patchy moth-eaten alopecia. He was treated with 1 to 2 megaunits of procaine penicillin daily for 10 days. Six hours after the first injection, the patient's temperature rose to 103°F; in addition to malaise, headache, flush, and sore throat, he had a transient skin rash and marked loss of hair. All the symptoms disappeared by the next day. Two to 3 weeks later, the lymphadenopathy had disappeared, and the patient's eyebrows and pubic hair started to regrow. The scalp hair was fully regrown 10 weeks from the onset of treatment. The author concluded that diffuse and extensive hair loss after the first injection of penicillin was part of the Herxheimer reaction.⁹

HIV

Hair loss is common in patients with HIV; in black patients, this loss may be associated with hair straightening.¹⁰ Possible causes of hair loss frequently are present in patients with HIV, including chronic HIV infection itself, acute and chronic systemic infections, local infections, nutritional deficits, immune and endocrine dysregulation, and exposure to multiple drugs.¹⁰ Alopecia areata and alopecia universalis also have been reported in patients with HIV.^11-14

Smith et al¹⁰ studied and reviewed the clinical and histopathologic features of hair loss in 10 patients with HIV. They noted that the most characteristic change in the hair of patients with HIV was hair loss with straightening, sometimes associated with fine hair texture and an increased tendency for broken hairs. These changes are seen in late-stage disease, most commonly in black patients. Each patient had telogen effluvium, and it was observed that any chronic or acute infection (including HIV) can lead to this condition. Nutritional deficits, often prominent in HIV patients, may lead to or potentiate telogen effluvium. Secondary infections and changes in bowel mucosa may lead to specific nutritional deficiencies even before evidence of clinical wasting is seen. In addition to caloric and protein malnutrition that may affect hair growth, minerals such as copper, zinc, and selenium are decreased in patients with HIV. Elevated levels of interleukin 6 and tumor necrosis factor α, which increase epidermal proliferation, may predispose patients to abnormal keratinization by increasing the proliferative rate and nutritional requirements.¹⁰

Endocrine regulation is another important factor in hair growth. In late-stage HIV disease, androgen levels decrease while estradiol levels increase. Although thyroid hormone levels are normal in advanced HIV, thyroid functions are elevated to more than expected for the amount of wasting and may contribute to the change of hair texture,10 autoimmune mechanism, associated diseases, and HIV medication side effects.

In the Smith et al¹⁰ study, scanning electron microscopy was performed on plucked and pulled hairs of 10 patients with late-stage HIV-1 infection. In addition, scalp biopsy specimens were examined in both vertical and transverse sections. All patients had telogen effluvium. Numerous apoptotic or necrotic keratinocytes were seen in the upper external root sheath follicular epithelium; a mild to moderate perifollicular mononuclear cell infiltrate, often containing eosinophils, also was seen. Additionally, the mononuclear infiltrate was seen surrounding and within the basaloid cells of the follicles in telogen phase; the midfollicular area had the most marked inflammatory infiltrate. Variable dystrophy of the hair shafts also was a consistent feature. Although telogen effluvium is a common response to a wide spectrum of biologic stresses, the presence of apoptotic or necrotic keratinocytes within the upper end of the external root sheath epithelium, as well as dystrophy of hairs, may be markers of hair loss in patients with HIV-1 infection.¹⁰

Autoimmune alopecia, including alopecia areata and alopecia universalis, can be seen in association with HIV.^11-15 Ostlere et al¹¹ first reported a case of alopecia universalis that developed in a patient 2 years after HIV antibody was detected. The patient showed loss of all scalp hair, eyelashes, eyebrows, and body hair. Two possible mechanisms for the development of alopecia were suggested. The first was that HIV induced nonspecific polyclonal B-cell activation with production of autoantibody either directly or via activated T cells; this supports a humoral theory of alopecia areata pathogenesis. Alternatively, the authors postulated that HIV induced a change in the balance between helper and suppressor cells, which resulted in aberrant cell-mediated immune effect at the hair follicles.¹¹ Werninghaus and Kaminer¹² described a similar patient with alopecia universalis; a biopsy specimen revealed perifollicular fibrosis without inflammation.

Stewart and Smoller¹³ described an HIV-positive patient with altered T-lymphocyte subsets in whom alopecia universalis developed. Results of a skin biopsy of the patient's scalp demonstrated a classic perifollicular lymphocytic infiltrate; results of immunophenotyping of the same specimen revealed that most cells were CD4⁺ lymphocytes. During the active loss of hair, the patient's ratio of CD4/CD8 cells was decreased; however, the ratio normalized during the period of hair regrowth. Their data suggested that systemic immune dysfunction, as seen in HIV infection, may be more important in mediating alopecia areata than localized immune responses. Because of the proposed mechanism of alopecia areata developing in this patient (ie, influx of CD4⁺ lymphocytes to the perifollicular regions of skin when the CD4/CD8 cells ratio is low), the authors were surprised that alopecia areata is not more common in patients with HIV infection.¹³

Cho et al¹⁴ described the association of vitiligo and alopecia areata in patients with HIV. They noted that the development of autoimmune diseases, though not life threatening, is an interesting phenomenon that may result from immune dysfunction or from B-cell infection by HIV, Epstein-Barr virus, or other unknown viruses. They described a 47-year-old man who had vitiligo and alopecia areata approximately 2 years after testing positive for HIV antibodies.¹⁴ Grossman et al¹⁵ described an HIV-seropositive man with acquired eyelash trichomegaly and alopecia areata. They noted that this combination of clinical manifestations is intriguing because the new onset of elongated eyelashes in patients with acquired immunodeficiency syndrome usually has been associated with severe immunosuppression, and alopecia areata has a presumed autoimmune etiology that requires T-cell activation. They concluded that the occurrence of these dichotomous conditions illustrates the potential selective pathogenesis of progressive HIV infection.¹⁵

Medications used in the treatment of HIV can play a role in hair loss. Geletko et al¹⁶ reported a 33-year-old HIV-infected man who developed alopecia areata after beginning therapy with zidovudine, a nucleoside analogue reverse transcriptase inhibitor. The alopecia reversed after the drug was discontinued. The authors proposed that patients with lower CD4⁺ counts may be more predisposed to zidovudine-induced alopecia than those in the earlier stages of HIV with higher CD4⁺ counts.¹⁶

Indinavir-related alopecia was described by d'Arminio Monforte et al.¹⁷ Of 337 patients given indinavir in combination with nucleoside analogues, 5 patients with HIV developed severe alopecia, which was evident clinically after a mean of 50 days of treatment. All patients were receiving triple therapy that included indinavir. Three patients had diffuse shedding of hair involving the entire scalp, and 2 had circumscribed circular areas of alopecia resulting in complete severe hair loss.¹⁷ Bouscarat et al¹⁸ reported 10 more cases of hair loss associated with indinavir therapy in patients receiving triple antiviral treatment that included indinavir. Hair loss developed during the first 6 months of indinavir therapy and initially involved the lower limbs. Progressive hair regrowth occurred within 4 months after indinavir was replaced by other treatments.¹⁸

Ginarte et al¹⁹ described significant alopecia induced by indinavir plus ritonavir therapy in 3 patients a few weeks after beginning treatment. The authors noted that patients receiving indinavir often experience retinoidlike effects such as alopecia, xerosis, and cheilitis. Nonscarring alopecia can develop in patients receiving indinavir, with or without retinoid effects.¹⁹ Hair loss also has been noted with the use of crixivan.²⁰

CMV

CMV is a prevalent viral pathogen.²¹ Most people with acute CMV experience an inapparent infection. The virus usually is spread through close personal contact, including sexual transmission. There has been debate over the link of alopecia areata with CMV. In 1995, Skinner et al²² described using polymerase chain reaction (PCR) techniques to find evidence of CMV DNA in paraffin block sections of lesions of alopecia areata. Of 21 patient biopsy specimens, 10 had alopecia areata and 11 had other hair loss conditions. Of the 10 alopecia areata samples, 9 were positive for CMV; no other hair loss samples were positive for CMV.²² Skinner et al²³ theorized that CMV may achieve latency in the hair root. Reactivation of CMV was thought to be one of the pathogenic mechanisms in alopecia areata; the authors argued that a lymphocytic surveillance of not-quite-latent CMV would explain much of the behavior of alopecia areata, which has a tendency for intermittent relapses and remissions.²³

The association between alopecia areata and CMV was refuted by Garcia-Hernandez et al,²⁴ who used 3 different PCR assays to detect CMV DNA in skin punch biopsy specimens of 3 patient groups: 40 patients with alopecia areata, 3 patients with HIV and alopecia areata, and 12 patients with other types of alopecia. PCR assays are known to be the most sensitive assay for CMV detection; this study used different PCR assays to achieve maximum sensitivity for CMV. No CMV DNA amplification was found in any of the specimens.²⁴

Offidani et al²⁵ further contradicted this association. The purpose of their study was to clarify the role of CMV infection and to demonstrate the absence of replication of other autoimmune disease–related herpesviruses (eg, Epstein-Barr virus) in the pathogenesis of alopecia areata. After extraction of mRNA from tissue samples of 4 patients with active patchy alopecia areata, reverse transcriptase PCR was carried out using primers specific for some viral members of the β Herpesviridae subfamily of the Herpesviridae family (eg, CMV, Epstein-Barr virus, herpes simplex virus). The authors could not detect any replication of the CMV or other β Herpesviridae in the samples collected, which supports the hypothesis that CMV is not the triggering factor in alopecia areata, neither as a reactivator of the immune response nor as a trigger of the autoimmunity.²⁵

Conclusion

Although many etiologies exist for hair loss, STDs should not be overlooked in a sexually active patient presenting with an otherwise unexplainable cause of this condition. A full workup, including clinical history, physical examination, and laboratory tests, should include STDs in the differential diagnosis (Table).

PLEASE REFER TO THE PDF TO VIEW THE TABLE

Folliculitis is a common complaint seen in dermatology practice. The differential diagnosis of folliculitis is broad and includes Demodex folliculitis. In humans, the Demodex mite species Demodex folliculorum and Demodex brevis have been found to inhabit the pilosebaceous unit. D folliculorum typically is found in the follicular infundibulum; D brevis exists in the sebaceous and meibomian glands.¹ Although the prevalence of Demodex approaches 100% in middle-aged and elderly adults,¹ mite density normally is low in healthy skin.²Demodex mites are considered pathogenic only when they are found in large numbers or in an intradermal location³; therefore, it has been suggested that D folliculorum may play a role in various papular and pustular eruptions of the head and neck, such as demodicosis and rosacea.³ We examine a case of scalp folliculitis secondary to Demodex infection and the role that this organism plays in the pathogenesis of folliculitis, as well as the available treatment options. 

Case Report

A 57-year-old white man presented to our department in June 2004 with an "infected scalp" and scalp irritation for 2 months. The patient was diagnosed with bacterial folliculitis and treated with clindamycin 1% gel twice daily for 1 month. He presented for follow-up in July 2004 with continued complaint of scalp pruritus and rash (Figure 1). Results of an examination showed a deep pink 10X7-cm plaque on the scalp with hyperkeratosis and pustules. An ectoparasite wet mount prepared from one of the pustules revealed the presence of several Demodex mites (Figure 2). The patient was treated with sulfacetamide 10% plus sulfur 5% cream twice daily, in addition to a 2-week course of selenium sulfide 2.5% shampoo once daily. When the patient was seen for follow-up in September 2004, his entire scalp had cleared (Figure 3). He was instructed to continue the selenium sulfide 2.5% shampoo twice weekly for 6 months to prevent recurrence.

Please refer to the PDF to view the figures

Comment

The Demodex mite is a ubiquitous arthropod measuring approximately 0.1 to 0.4 mm in length. Typically, it infests areas around the eyelids, nose, and ear canals in human hosts.⁴ The life cycle of the mite is 18 to 24 days. The female mite lays 20 to 24 eggs in a hair follicle where the eggs are nourished by the surrounding pilosebaceous unit. The eggs hatch and the nymphs continue to live in the follicle where their main source of food is human glandular secretions.⁵ The mite primarily is an asymptomatic inhabitant of human pilosebaceous follicles and poses no harm to the host.¹

The role of D folliculorum in cutaneous disease in humans remains controversial. The pathogenicity is difficult to establish secondary to the localization of the disease, the widespread prevalence of infection with the D folliculorum mite, and the obligate nature of the parasite; therefore, the detection of the presence of the mite is not, in and of itself, enough evidence to establish pathogenicity.⁶ Results of immunohistochemical staining have shown that helper T lymphocytes predominate in the dermal infiltrate of demodicosis suggesting a possible role of cell-mediated immune response and delayed hypersensitivity.⁷ There also is evidence for a humoral immune response component with increased macrophages and Langerhans cells in the presence of infestation with Demodex.⁷

Demodex mites have been implicated as a causative agent in rosacea and pustular folliculitis.⁶ It is important to consider the possibility that the vascular changes of rosacea create an environment that is favorable to the multiplication of Demodex mites and their penetration into the dermis.⁸ Forton and Seys⁶ reported that Demodex mites are associated with the inflammatory symptoms of rosacea and that the mites are present in greater numbers and higher frequencies in patients with rosacea. Additionally, a study by Georgala et al⁷ evaluated the importance of D folliculorum in the etiology and course of rosacea and showed that D folliculorum was found in 83 (90.2%) of 92 rosacea subjects studied but in only 11 (11.9%) of the 92 controls, thereby concluding that although Demodex mites may not be the cause of rosacea, they may represent an important cofactor. Finally, Vollmer⁹ examined 388 follicles in 24 resections of skin for the presence of histologic folliculitis and Demodex mites. Results showed that Demodex mites were found in 87 (42%) of 208 follicles with inflammation but in just 18 (10%) of 180 follicles without inflammation. Furthermore, 87 (83%) of 105 follicles with Demodex showed inflammation, which demonstrated a nonrandom association between these 2 entities.⁹ A study by Meinking et al¹⁰ supported the rapid clearing of papulopustular dermatosis of the scalp and granulomatous rosacea when treated with scabicidal preparations such as permethrin or ivermectin, thereby supporting the pathogenic role of Demodex in papulopustular eruptions.

On the other hand, a review by Aylesworth and Vance¹¹ found that 117 (10%) of 1124 skin biopsies and 198 (12%) of 1692 follicles incidentally revealed that follicular mites were found in patients with various unrelated skin disorders, thereby suggesting that Demodex is a normal inhabitant of the hair follicle and is not pathogenic.¹¹ Other histologic evidence that failed to show a correlation between Demodex presence and skin disease was an examination of the results of 108 biopsy specimens of rosacea, of which only 20 (19%) contained Demodex.¹² There was no correlation between Demodex mites and skin disease in a study of 29 biopsy samples of the head and neck by Nutting and Green.¹³ We must note that the reported prevalence of Demodex presence is partially determined by the preciseness of the detection method used.

There are several treatment options available for demodicosis. In our case, the patient cleared with a combination of sulfacetamide 10% plus sulfur 5% cream, in addition to selenium sulfide 2.5% shampoo. Other commonly used treatment options include ivermectin,¹⁰ topical antibiotics, and topical retinoids.

The persistence of the patient's folliculitis despite treatment with clindamycin 1% gel; rapid clearance after therapy with sulfacetamide 10% plus sulfur 5% cream twice daily and selenium sulfide 2.5% shampoo once daily is initiated; and positive results of the ectoparasite wet mount suggest a pathogenic role of Demodex in causing the patient's symptoms. Although the link between folliculitis and Demodex infection remains controversial, this case demonstrates the importance of considering the possible role of Demodex in the differential diagnosis of rosacea and papulopustular eruptions of the head and neck. 

Folliculitis is a common complaint seen in dermatology practice. The differential diagnosis of folliculitis is broad and includes Demodex folliculitis. In humans, the Demodex mite species Demodex folliculorum and Demodex brevis have been found to inhabit the pilosebaceous unit. D folliculorum typically is found in the follicular infundibulum; D brevis exists in the sebaceous and meibomian glands.¹ Although the prevalence of Demodex approaches 100% in middle-aged and elderly adults,¹ mite density normally is low in healthy skin.²Demodex mites are considered pathogenic only when they are found in large numbers or in an intradermal location³; therefore, it has been suggested that D folliculorum may play a role in various papular and pustular eruptions of the head and neck, such as demodicosis and rosacea.³ We examine a case of scalp folliculitis secondary to Demodex infection and the role that this organism plays in the pathogenesis of folliculitis, as well as the available treatment options. 

Case Report

A 57-year-old white man presented to our department in June 2004 with an "infected scalp" and scalp irritation for 2 months. The patient was diagnosed with bacterial folliculitis and treated with clindamycin 1% gel twice daily for 1 month. He presented for follow-up in July 2004 with continued complaint of scalp pruritus and rash (Figure 1). Results of an examination showed a deep pink 10X7-cm plaque on the scalp with hyperkeratosis and pustules. An ectoparasite wet mount prepared from one of the pustules revealed the presence of several Demodex mites (Figure 2). The patient was treated with sulfacetamide 10% plus sulfur 5% cream twice daily, in addition to a 2-week course of selenium sulfide 2.5% shampoo once daily. When the patient was seen for follow-up in September 2004, his entire scalp had cleared (Figure 3). He was instructed to continue the selenium sulfide 2.5% shampoo twice weekly for 6 months to prevent recurrence.

Please refer to the PDF to view the figures

Comment

The Demodex mite is a ubiquitous arthropod measuring approximately 0.1 to 0.4 mm in length. Typically, it infests areas around the eyelids, nose, and ear canals in human hosts.⁴ The life cycle of the mite is 18 to 24 days. The female mite lays 20 to 24 eggs in a hair follicle where the eggs are nourished by the surrounding pilosebaceous unit. The eggs hatch and the nymphs continue to live in the follicle where their main source of food is human glandular secretions.⁵ The mite primarily is an asymptomatic inhabitant of human pilosebaceous follicles and poses no harm to the host.¹

The role of D folliculorum in cutaneous disease in humans remains controversial. The pathogenicity is difficult to establish secondary to the localization of the disease, the widespread prevalence of infection with the D folliculorum mite, and the obligate nature of the parasite; therefore, the detection of the presence of the mite is not, in and of itself, enough evidence to establish pathogenicity.⁶ Results of immunohistochemical staining have shown that helper T lymphocytes predominate in the dermal infiltrate of demodicosis suggesting a possible role of cell-mediated immune response and delayed hypersensitivity.⁷ There also is evidence for a humoral immune response component with increased macrophages and Langerhans cells in the presence of infestation with Demodex.⁷

Demodex mites have been implicated as a causative agent in rosacea and pustular folliculitis.⁶ It is important to consider the possibility that the vascular changes of rosacea create an environment that is favorable to the multiplication of Demodex mites and their penetration into the dermis.⁸ Forton and Seys⁶ reported that Demodex mites are associated with the inflammatory symptoms of rosacea and that the mites are present in greater numbers and higher frequencies in patients with rosacea. Additionally, a study by Georgala et al⁷ evaluated the importance of D folliculorum in the etiology and course of rosacea and showed that D folliculorum was found in 83 (90.2%) of 92 rosacea subjects studied but in only 11 (11.9%) of the 92 controls, thereby concluding that although Demodex mites may not be the cause of rosacea, they may represent an important cofactor. Finally, Vollmer⁹ examined 388 follicles in 24 resections of skin for the presence of histologic folliculitis and Demodex mites. Results showed that Demodex mites were found in 87 (42%) of 208 follicles with inflammation but in just 18 (10%) of 180 follicles without inflammation. Furthermore, 87 (83%) of 105 follicles with Demodex showed inflammation, which demonstrated a nonrandom association between these 2 entities.⁹ A study by Meinking et al¹⁰ supported the rapid clearing of papulopustular dermatosis of the scalp and granulomatous rosacea when treated with scabicidal preparations such as permethrin or ivermectin, thereby supporting the pathogenic role of Demodex in papulopustular eruptions.

On the other hand, a review by Aylesworth and Vance¹¹ found that 117 (10%) of 1124 skin biopsies and 198 (12%) of 1692 follicles incidentally revealed that follicular mites were found in patients with various unrelated skin disorders, thereby suggesting that Demodex is a normal inhabitant of the hair follicle and is not pathogenic.¹¹ Other histologic evidence that failed to show a correlation between Demodex presence and skin disease was an examination of the results of 108 biopsy specimens of rosacea, of which only 20 (19%) contained Demodex.¹² There was no correlation between Demodex mites and skin disease in a study of 29 biopsy samples of the head and neck by Nutting and Green.¹³ We must note that the reported prevalence of Demodex presence is partially determined by the preciseness of the detection method used.

There are several treatment options available for demodicosis. In our case, the patient cleared with a combination of sulfacetamide 10% plus sulfur 5% cream, in addition to selenium sulfide 2.5% shampoo. Other commonly used treatment options include ivermectin,¹⁰ topical antibiotics, and topical retinoids.

The persistence of the patient's folliculitis despite treatment with clindamycin 1% gel; rapid clearance after therapy with sulfacetamide 10% plus sulfur 5% cream twice daily and selenium sulfide 2.5% shampoo once daily is initiated; and positive results of the ectoparasite wet mount suggest a pathogenic role of Demodex in causing the patient's symptoms. Although the link between folliculitis and Demodex infection remains controversial, this case demonstrates the importance of considering the possible role of Demodex in the differential diagnosis of rosacea and papulopustular eruptions of the head and neck. 

Folliculitis is a common complaint seen in dermatology practice. The differential diagnosis of folliculitis is broad and includes Demodex folliculitis. In humans, the Demodex mite species Demodex folliculorum and Demodex brevis have been found to inhabit the pilosebaceous unit. D folliculorum typically is found in the follicular infundibulum; D brevis exists in the sebaceous and meibomian glands.¹ Although the prevalence of Demodex approaches 100% in middle-aged and elderly adults,¹ mite density normally is low in healthy skin.²Demodex mites are considered pathogenic only when they are found in large numbers or in an intradermal location³; therefore, it has been suggested that D folliculorum may play a role in various papular and pustular eruptions of the head and neck, such as demodicosis and rosacea.³ We examine a case of scalp folliculitis secondary to Demodex infection and the role that this organism plays in the pathogenesis of folliculitis, as well as the available treatment options. 

Case Report

A 57-year-old white man presented to our department in June 2004 with an "infected scalp" and scalp irritation for 2 months. The patient was diagnosed with bacterial folliculitis and treated with clindamycin 1% gel twice daily for 1 month. He presented for follow-up in July 2004 with continued complaint of scalp pruritus and rash (Figure 1). Results of an examination showed a deep pink 10X7-cm plaque on the scalp with hyperkeratosis and pustules. An ectoparasite wet mount prepared from one of the pustules revealed the presence of several Demodex mites (Figure 2). The patient was treated with sulfacetamide 10% plus sulfur 5% cream twice daily, in addition to a 2-week course of selenium sulfide 2.5% shampoo once daily. When the patient was seen for follow-up in September 2004, his entire scalp had cleared (Figure 3). He was instructed to continue the selenium sulfide 2.5% shampoo twice weekly for 6 months to prevent recurrence.

Please refer to the PDF to view the figures

Comment

The Demodex mite is a ubiquitous arthropod measuring approximately 0.1 to 0.4 mm in length. Typically, it infests areas around the eyelids, nose, and ear canals in human hosts.⁴ The life cycle of the mite is 18 to 24 days. The female mite lays 20 to 24 eggs in a hair follicle where the eggs are nourished by the surrounding pilosebaceous unit. The eggs hatch and the nymphs continue to live in the follicle where their main source of food is human glandular secretions.⁵ The mite primarily is an asymptomatic inhabitant of human pilosebaceous follicles and poses no harm to the host.¹

The role of D folliculorum in cutaneous disease in humans remains controversial. The pathogenicity is difficult to establish secondary to the localization of the disease, the widespread prevalence of infection with the D folliculorum mite, and the obligate nature of the parasite; therefore, the detection of the presence of the mite is not, in and of itself, enough evidence to establish pathogenicity.⁶ Results of immunohistochemical staining have shown that helper T lymphocytes predominate in the dermal infiltrate of demodicosis suggesting a possible role of cell-mediated immune response and delayed hypersensitivity.⁷ There also is evidence for a humoral immune response component with increased macrophages and Langerhans cells in the presence of infestation with Demodex.⁷

Demodex mites have been implicated as a causative agent in rosacea and pustular folliculitis.⁶ It is important to consider the possibility that the vascular changes of rosacea create an environment that is favorable to the multiplication of Demodex mites and their penetration into the dermis.⁸ Forton and Seys⁶ reported that Demodex mites are associated with the inflammatory symptoms of rosacea and that the mites are present in greater numbers and higher frequencies in patients with rosacea. Additionally, a study by Georgala et al⁷ evaluated the importance of D folliculorum in the etiology and course of rosacea and showed that D folliculorum was found in 83 (90.2%) of 92 rosacea subjects studied but in only 11 (11.9%) of the 92 controls, thereby concluding that although Demodex mites may not be the cause of rosacea, they may represent an important cofactor. Finally, Vollmer⁹ examined 388 follicles in 24 resections of skin for the presence of histologic folliculitis and Demodex mites. Results showed that Demodex mites were found in 87 (42%) of 208 follicles with inflammation but in just 18 (10%) of 180 follicles without inflammation. Furthermore, 87 (83%) of 105 follicles with Demodex showed inflammation, which demonstrated a nonrandom association between these 2 entities.⁹ A study by Meinking et al¹⁰ supported the rapid clearing of papulopustular dermatosis of the scalp and granulomatous rosacea when treated with scabicidal preparations such as permethrin or ivermectin, thereby supporting the pathogenic role of Demodex in papulopustular eruptions.

On the other hand, a review by Aylesworth and Vance¹¹ found that 117 (10%) of 1124 skin biopsies and 198 (12%) of 1692 follicles incidentally revealed that follicular mites were found in patients with various unrelated skin disorders, thereby suggesting that Demodex is a normal inhabitant of the hair follicle and is not pathogenic.¹¹ Other histologic evidence that failed to show a correlation between Demodex presence and skin disease was an examination of the results of 108 biopsy specimens of rosacea, of which only 20 (19%) contained Demodex.¹² There was no correlation between Demodex mites and skin disease in a study of 29 biopsy samples of the head and neck by Nutting and Green.¹³ We must note that the reported prevalence of Demodex presence is partially determined by the preciseness of the detection method used.

There are several treatment options available for demodicosis. In our case, the patient cleared with a combination of sulfacetamide 10% plus sulfur 5% cream, in addition to selenium sulfide 2.5% shampoo. Other commonly used treatment options include ivermectin,¹⁰ topical antibiotics, and topical retinoids.

The persistence of the patient's folliculitis despite treatment with clindamycin 1% gel; rapid clearance after therapy with sulfacetamide 10% plus sulfur 5% cream twice daily and selenium sulfide 2.5% shampoo once daily is initiated; and positive results of the ectoparasite wet mount suggest a pathogenic role of Demodex in causing the patient's symptoms. Although the link between folliculitis and Demodex infection remains controversial, this case demonstrates the importance of considering the possible role of Demodex in the differential diagnosis of rosacea and papulopustular eruptions of the head and neck. 

Median canaliform nail dystrophy is a nail abnormality that typically involves one or both thumbnails. The first case of this disorder was recorded by Heller¹ in 1928. Median canaliform nail dystrophy presents as a central longitudinal groove of the nail plate, extending proximally from the end of the nail.² This condition is usually not inherited. However, it may be acquired following trauma to the nail. We describe a man with familial median canaliform nail dystrophy and discuss the differential diagnosis.

Case Report

A 68-year-old man presented with bilateral dystrophy of his thumbnails. The nail abnormality initially appeared at 34 years of age with no preceding trauma to the digits. His older brother and his mother also had developed the same nail changes as young adults. Neither the patient nor his brother or mother rubbed their proximal thumbnail fold with the tip of their second finger; the absence of this behavioral activity was repeatedly confirmed by both the patient and his wife during several subsequent office visits.

Examination of both thumbnails showed an asymptomatic distal fissure with a fir tree–like pattern (Figure). Proximally, the nail plates showed transverse grooves. In addition, the lunula was red and enlarged.

Please refer to the PDF to view the figure

Comment

Median canaliform nail dystrophy appears as a long longitudinal groove extending from either the proximal nail fold or a more distal portion of the nail plate to the end of the nail. Lateral extensions of this fissure create a conspicuous inverted fir tree–like pattern. In severe cases, the nail can split along the groove.³ Thickening of the proximal nail fold, enlargement of the lunula, and redness of the lunula also may occur.^4-9

The diagnosis of this condition is usually established based on clinical features because pathologic correlation is rarely available. However, specimens for microscopic evaluation have occasionally been provided. Parakeratosis, as well as an accumulation of melanin within and between the nail bed keratinocytes, was demonstrated in the evaluation of an affected nail by Heller¹⁰ in 1927. Subsequently, parakeratosis and intranuclear pigmentation were found within the longitudinal canal of the affected nail plate of a 12-year-old girl with median canaliform nail dystrophy who was described by Robinson and Weidman¹¹ in 1948.

Median canaliform nail dystrophy may present following trauma to the nail plate or nail matrix.^3-7,12-15 In addition, coexisting conditions such as either soft tissue in the nail defect or dental caries have been observed in some patients with median canaliform nail dystrophy. In one case, a 19-year-old woman presented with a flabby filament of fleshy tissue that was observed within the dystrophic nail canal.¹⁴ The tissue was extracted, and the nail abnormality resolved. Subsequently, the nail dystrophy, including the associated tissue, reappeared.¹⁴ Tooth decay associated with median canaliform nail dystrophy was reported in a 23-year-old woman with a deformity that involved many of the nails on both of her hands. Her nail condition spontaneously cleared after 3 carious teeth were extracted.¹⁶

Medication was postulated as the causative factor for the development of median canaliform nail dystrophy in 3 patients who were receiving isotretinoin. The first, reported by Bottomley and Cunliffe¹⁷ in 1992, was a 38-year-old woman who developed median canaliform nail dystrophy 6 weeks after beginning treatment with isotretinoin. Her thumbnail returned to normal 4 weeks after she discontinued the drug.¹⁷ The second patient, described by Griego et al⁴ in 1995, was an 18-year-old man who developed median canaliform nail dystrophy of both thumbnails after starting therapy with isotretinoin for his acne. The nail disfigurement became distinct after 4 months of treatment; his new thumbnail dystrophy resolved 5 months after he discontinued the medication.⁴ A third patient was reported by Dharmagunawardena and Charles-Holmes¹² in 1997. They described a 19-year-old man who developed median canaliform nail dystrophy in both thumbnails within 4 weeks after starting treatment with isotretinoin for his acne. His nails returned to normal 3 months after completing a 5-month course of isotretinoin therapy.¹²

Familial median canaliform nail dystrophy has not been associated with any systemic syndromes. In our patient and his family, the nail dystrophy was not congenital but rather appeared as an acquired abnormality of the nails in adulthood.

The etiology of median canaliform nail dystrophy is unknown.^5,7,13,16-18 It usually is an acquired condition. Nail matrix trauma may precede the onset; however, an associated nail injury has often not occurred.^{3-7,12-16,19-21} This nail dystrophy is not considered to be inherited. The familial occurrence of median canaliform nail dystrophy has rarely been described. Indeed, to the best of our knowledge, in addition to our patient, only 3 families with median canaliform nail dystrophy have been described.^20,22,23 In the first such family, a 16-year-old girl with bilateral median canaliform nail dystrophy of her thumbnails since the age of 13 years had a mother with similar-appearing thumbnails.²⁰ A second such family also included a mother and daughter.²² Long longitudinal grooves were present in the daughter's left thumbnail since the age of 11 years; her mother had a similar dystrophy involving her right thumbnail that began when she was 12 years old. Her mother, currently 34 years old, still has recurrent episodes of spontaneously resolving median canaliform nail dystrophy. The family had no history of other hereditary diseases.²² The third family in which median canaliform nail dystrophy occurred was reported by Bossi²³ in the Italian literature in 1965. Our patient and his brother and mother represent the fourth such family.

The differential diagnosis of median canaliform nail dystrophy includes habit tic deformity (Table). It also includes other causes of longitudinal splits in the nail plate such as direct trauma to the nail unit. In addition, digital mucous cyst (synovial cyst), lichen striatus, nail-patella syndrome, pterygium, Raynaud disease, and trachyonychia are other conditions in which a longitudinal nail defect has been described.^5,7,30,31

Please refer to the PDF to view the table

Habit tic deformity is usually present in one or both thumbnails and results in alteration of the normal nail growth. It is caused by the constant or habitual rubbing of the thumb's proximal nail fold by the tip of the second digit. The subsequent damage to the nail matrix causes clinical changes in the nail plate that appear different than those of median canaliform nail dystrophy. The habit tic deformity produces transverse ridges along the central nail plate depression instead of a longitudinal groove with lateral projections. The depth of the central nail plate canal depends on the intensity of the inflicted trauma by the index finger to the matrix of the thumbnail. In addition, the lunula may appear red and enlarged.^9,29 Also, the proximal nail fold may be swollen.^5,13

Median canaliform nail dystrophy has occasionally been described to periodically disappear; often, the nail defect reappears in these individuals.^{4-7,13,15,17,24} In some patients, the central nail defect is replaced by a longitudinal ridge^5,6; however, in most patients, such as ours, the condition does not resolve spontaneously. Keeping the nail length short and buffing the surface of the nail can prevent the edge of the nail plate from catching on clothing and other objects.⁵ Covering the nail plate with tape or a nail wrap also can aid in ensuring that jagged edges are not present.^4,7 

Conclusion

Familial median canaliform nail dystrophy has rarely been described. Our patient had adult onset of his condition involving both thumbnails with associated red macrolunula. His brother and his mother also experienced the same nail dystrophy. Including our patient and his family, median canaliform nail dystrophy has only been reported in 4 families. The mode of inheritance for median canaliform nail dystrophy in these families remains to be determined. 

Median canaliform nail dystrophy is a nail abnormality that typically involves one or both thumbnails. The first case of this disorder was recorded by Heller¹ in 1928. Median canaliform nail dystrophy presents as a central longitudinal groove of the nail plate, extending proximally from the end of the nail.² This condition is usually not inherited. However, it may be acquired following trauma to the nail. We describe a man with familial median canaliform nail dystrophy and discuss the differential diagnosis.

Case Report

A 68-year-old man presented with bilateral dystrophy of his thumbnails. The nail abnormality initially appeared at 34 years of age with no preceding trauma to the digits. His older brother and his mother also had developed the same nail changes as young adults. Neither the patient nor his brother or mother rubbed their proximal thumbnail fold with the tip of their second finger; the absence of this behavioral activity was repeatedly confirmed by both the patient and his wife during several subsequent office visits.

Examination of both thumbnails showed an asymptomatic distal fissure with a fir tree–like pattern (Figure). Proximally, the nail plates showed transverse grooves. In addition, the lunula was red and enlarged.

Please refer to the PDF to view the figure

Comment

Median canaliform nail dystrophy appears as a long longitudinal groove extending from either the proximal nail fold or a more distal portion of the nail plate to the end of the nail. Lateral extensions of this fissure create a conspicuous inverted fir tree–like pattern. In severe cases, the nail can split along the groove.³ Thickening of the proximal nail fold, enlargement of the lunula, and redness of the lunula also may occur.^4-9

The diagnosis of this condition is usually established based on clinical features because pathologic correlation is rarely available. However, specimens for microscopic evaluation have occasionally been provided. Parakeratosis, as well as an accumulation of melanin within and between the nail bed keratinocytes, was demonstrated in the evaluation of an affected nail by Heller¹⁰ in 1927. Subsequently, parakeratosis and intranuclear pigmentation were found within the longitudinal canal of the affected nail plate of a 12-year-old girl with median canaliform nail dystrophy who was described by Robinson and Weidman¹¹ in 1948.

Median canaliform nail dystrophy may present following trauma to the nail plate or nail matrix.^3-7,12-15 In addition, coexisting conditions such as either soft tissue in the nail defect or dental caries have been observed in some patients with median canaliform nail dystrophy. In one case, a 19-year-old woman presented with a flabby filament of fleshy tissue that was observed within the dystrophic nail canal.¹⁴ The tissue was extracted, and the nail abnormality resolved. Subsequently, the nail dystrophy, including the associated tissue, reappeared.¹⁴ Tooth decay associated with median canaliform nail dystrophy was reported in a 23-year-old woman with a deformity that involved many of the nails on both of her hands. Her nail condition spontaneously cleared after 3 carious teeth were extracted.¹⁶

Medication was postulated as the causative factor for the development of median canaliform nail dystrophy in 3 patients who were receiving isotretinoin. The first, reported by Bottomley and Cunliffe¹⁷ in 1992, was a 38-year-old woman who developed median canaliform nail dystrophy 6 weeks after beginning treatment with isotretinoin. Her thumbnail returned to normal 4 weeks after she discontinued the drug.¹⁷ The second patient, described by Griego et al⁴ in 1995, was an 18-year-old man who developed median canaliform nail dystrophy of both thumbnails after starting therapy with isotretinoin for his acne. The nail disfigurement became distinct after 4 months of treatment; his new thumbnail dystrophy resolved 5 months after he discontinued the medication.⁴ A third patient was reported by Dharmagunawardena and Charles-Holmes¹² in 1997. They described a 19-year-old man who developed median canaliform nail dystrophy in both thumbnails within 4 weeks after starting treatment with isotretinoin for his acne. His nails returned to normal 3 months after completing a 5-month course of isotretinoin therapy.¹²

Familial median canaliform nail dystrophy has not been associated with any systemic syndromes. In our patient and his family, the nail dystrophy was not congenital but rather appeared as an acquired abnormality of the nails in adulthood.

The etiology of median canaliform nail dystrophy is unknown.^5,7,13,16-18 It usually is an acquired condition. Nail matrix trauma may precede the onset; however, an associated nail injury has often not occurred.^{3-7,12-16,19-21} This nail dystrophy is not considered to be inherited. The familial occurrence of median canaliform nail dystrophy has rarely been described. Indeed, to the best of our knowledge, in addition to our patient, only 3 families with median canaliform nail dystrophy have been described.^20,22,23 In the first such family, a 16-year-old girl with bilateral median canaliform nail dystrophy of her thumbnails since the age of 13 years had a mother with similar-appearing thumbnails.²⁰ A second such family also included a mother and daughter.²² Long longitudinal grooves were present in the daughter's left thumbnail since the age of 11 years; her mother had a similar dystrophy involving her right thumbnail that began when she was 12 years old. Her mother, currently 34 years old, still has recurrent episodes of spontaneously resolving median canaliform nail dystrophy. The family had no history of other hereditary diseases.²² The third family in which median canaliform nail dystrophy occurred was reported by Bossi²³ in the Italian literature in 1965. Our patient and his brother and mother represent the fourth such family.

The differential diagnosis of median canaliform nail dystrophy includes habit tic deformity (Table). It also includes other causes of longitudinal splits in the nail plate such as direct trauma to the nail unit. In addition, digital mucous cyst (synovial cyst), lichen striatus, nail-patella syndrome, pterygium, Raynaud disease, and trachyonychia are other conditions in which a longitudinal nail defect has been described.^5,7,30,31

Please refer to the PDF to view the table

Habit tic deformity is usually present in one or both thumbnails and results in alteration of the normal nail growth. It is caused by the constant or habitual rubbing of the thumb's proximal nail fold by the tip of the second digit. The subsequent damage to the nail matrix causes clinical changes in the nail plate that appear different than those of median canaliform nail dystrophy. The habit tic deformity produces transverse ridges along the central nail plate depression instead of a longitudinal groove with lateral projections. The depth of the central nail plate canal depends on the intensity of the inflicted trauma by the index finger to the matrix of the thumbnail. In addition, the lunula may appear red and enlarged.^9,29 Also, the proximal nail fold may be swollen.^5,13

Median canaliform nail dystrophy has occasionally been described to periodically disappear; often, the nail defect reappears in these individuals.^{4-7,13,15,17,24} In some patients, the central nail defect is replaced by a longitudinal ridge^5,6; however, in most patients, such as ours, the condition does not resolve spontaneously. Keeping the nail length short and buffing the surface of the nail can prevent the edge of the nail plate from catching on clothing and other objects.⁵ Covering the nail plate with tape or a nail wrap also can aid in ensuring that jagged edges are not present.^4,7 

Conclusion

Familial median canaliform nail dystrophy has rarely been described. Our patient had adult onset of his condition involving both thumbnails with associated red macrolunula. His brother and his mother also experienced the same nail dystrophy. Including our patient and his family, median canaliform nail dystrophy has only been reported in 4 families. The mode of inheritance for median canaliform nail dystrophy in these families remains to be determined. 

Median canaliform nail dystrophy is a nail abnormality that typically involves one or both thumbnails. The first case of this disorder was recorded by Heller¹ in 1928. Median canaliform nail dystrophy presents as a central longitudinal groove of the nail plate, extending proximally from the end of the nail.² This condition is usually not inherited. However, it may be acquired following trauma to the nail. We describe a man with familial median canaliform nail dystrophy and discuss the differential diagnosis.

Case Report

A 68-year-old man presented with bilateral dystrophy of his thumbnails. The nail abnormality initially appeared at 34 years of age with no preceding trauma to the digits. His older brother and his mother also had developed the same nail changes as young adults. Neither the patient nor his brother or mother rubbed their proximal thumbnail fold with the tip of their second finger; the absence of this behavioral activity was repeatedly confirmed by both the patient and his wife during several subsequent office visits.

Examination of both thumbnails showed an asymptomatic distal fissure with a fir tree–like pattern (Figure). Proximally, the nail plates showed transverse grooves. In addition, the lunula was red and enlarged.

Please refer to the PDF to view the figure

Comment

Median canaliform nail dystrophy appears as a long longitudinal groove extending from either the proximal nail fold or a more distal portion of the nail plate to the end of the nail. Lateral extensions of this fissure create a conspicuous inverted fir tree–like pattern. In severe cases, the nail can split along the groove.³ Thickening of the proximal nail fold, enlargement of the lunula, and redness of the lunula also may occur.^4-9

The diagnosis of this condition is usually established based on clinical features because pathologic correlation is rarely available. However, specimens for microscopic evaluation have occasionally been provided. Parakeratosis, as well as an accumulation of melanin within and between the nail bed keratinocytes, was demonstrated in the evaluation of an affected nail by Heller¹⁰ in 1927. Subsequently, parakeratosis and intranuclear pigmentation were found within the longitudinal canal of the affected nail plate of a 12-year-old girl with median canaliform nail dystrophy who was described by Robinson and Weidman¹¹ in 1948.

Median canaliform nail dystrophy may present following trauma to the nail plate or nail matrix.^3-7,12-15 In addition, coexisting conditions such as either soft tissue in the nail defect or dental caries have been observed in some patients with median canaliform nail dystrophy. In one case, a 19-year-old woman presented with a flabby filament of fleshy tissue that was observed within the dystrophic nail canal.¹⁴ The tissue was extracted, and the nail abnormality resolved. Subsequently, the nail dystrophy, including the associated tissue, reappeared.¹⁴ Tooth decay associated with median canaliform nail dystrophy was reported in a 23-year-old woman with a deformity that involved many of the nails on both of her hands. Her nail condition spontaneously cleared after 3 carious teeth were extracted.¹⁶

Medication was postulated as the causative factor for the development of median canaliform nail dystrophy in 3 patients who were receiving isotretinoin. The first, reported by Bottomley and Cunliffe¹⁷ in 1992, was a 38-year-old woman who developed median canaliform nail dystrophy 6 weeks after beginning treatment with isotretinoin. Her thumbnail returned to normal 4 weeks after she discontinued the drug.¹⁷ The second patient, described by Griego et al⁴ in 1995, was an 18-year-old man who developed median canaliform nail dystrophy of both thumbnails after starting therapy with isotretinoin for his acne. The nail disfigurement became distinct after 4 months of treatment; his new thumbnail dystrophy resolved 5 months after he discontinued the medication.⁴ A third patient was reported by Dharmagunawardena and Charles-Holmes¹² in 1997. They described a 19-year-old man who developed median canaliform nail dystrophy in both thumbnails within 4 weeks after starting treatment with isotretinoin for his acne. His nails returned to normal 3 months after completing a 5-month course of isotretinoin therapy.¹²

Familial median canaliform nail dystrophy has not been associated with any systemic syndromes. In our patient and his family, the nail dystrophy was not congenital but rather appeared as an acquired abnormality of the nails in adulthood.

The etiology of median canaliform nail dystrophy is unknown.^5,7,13,16-18 It usually is an acquired condition. Nail matrix trauma may precede the onset; however, an associated nail injury has often not occurred.^{3-7,12-16,19-21} This nail dystrophy is not considered to be inherited. The familial occurrence of median canaliform nail dystrophy has rarely been described. Indeed, to the best of our knowledge, in addition to our patient, only 3 families with median canaliform nail dystrophy have been described.^20,22,23 In the first such family, a 16-year-old girl with bilateral median canaliform nail dystrophy of her thumbnails since the age of 13 years had a mother with similar-appearing thumbnails.²⁰ A second such family also included a mother and daughter.²² Long longitudinal grooves were present in the daughter's left thumbnail since the age of 11 years; her mother had a similar dystrophy involving her right thumbnail that began when she was 12 years old. Her mother, currently 34 years old, still has recurrent episodes of spontaneously resolving median canaliform nail dystrophy. The family had no history of other hereditary diseases.²² The third family in which median canaliform nail dystrophy occurred was reported by Bossi²³ in the Italian literature in 1965. Our patient and his brother and mother represent the fourth such family.

The differential diagnosis of median canaliform nail dystrophy includes habit tic deformity (Table). It also includes other causes of longitudinal splits in the nail plate such as direct trauma to the nail unit. In addition, digital mucous cyst (synovial cyst), lichen striatus, nail-patella syndrome, pterygium, Raynaud disease, and trachyonychia are other conditions in which a longitudinal nail defect has been described.^5,7,30,31

Please refer to the PDF to view the table

Habit tic deformity is usually present in one or both thumbnails and results in alteration of the normal nail growth. It is caused by the constant or habitual rubbing of the thumb's proximal nail fold by the tip of the second digit. The subsequent damage to the nail matrix causes clinical changes in the nail plate that appear different than those of median canaliform nail dystrophy. The habit tic deformity produces transverse ridges along the central nail plate depression instead of a longitudinal groove with lateral projections. The depth of the central nail plate canal depends on the intensity of the inflicted trauma by the index finger to the matrix of the thumbnail. In addition, the lunula may appear red and enlarged.^9,29 Also, the proximal nail fold may be swollen.^5,13

Median canaliform nail dystrophy has occasionally been described to periodically disappear; often, the nail defect reappears in these individuals.^{4-7,13,15,17,24} In some patients, the central nail defect is replaced by a longitudinal ridge^5,6; however, in most patients, such as ours, the condition does not resolve spontaneously. Keeping the nail length short and buffing the surface of the nail can prevent the edge of the nail plate from catching on clothing and other objects.⁵ Covering the nail plate with tape or a nail wrap also can aid in ensuring that jagged edges are not present.^4,7 

Conclusion

Familial median canaliform nail dystrophy has rarely been described. Our patient had adult onset of his condition involving both thumbnails with associated red macrolunula. His brother and his mother also experienced the same nail dystrophy. Including our patient and his family, median canaliform nail dystrophy has only been reported in 4 families. The mode of inheritance for median canaliform nail dystrophy in these families remains to be determined. 

Onychotillomania is an uncommon condition characterized by self-destruction of the fingernails and/or toenails by compulsive manipulation. We report 2 cases of onychotillomania with differing presentations in a young man and in an older man. Onychotillomania may be a form of obsessive-compulsive disorder (OCD), and we discuss the psychologic factors and current treatments for this condition.

Emotional and psychologic factors have the ability to influence the underlying disease process in at least 33% of patients with a dermatologic condition.1 In some cases, such as onychotillomania, a psychiatric condition can be the underlining cause. Onychotillomania, a condition whose true incidence is unknown, is characterized by the compulsive or irresistible urge in patients to pick at, pull off, or harmfully bite or chew their nails. This urge may be conscious or unconscious. The word onychotillomania is derived from the Greek onycho, nail; tillo, to pull; and mania, madness or frenzy. In psychiatry, onychotillomania has been classified as an impulse control disorder that includes conditions such as compulsive gambling, kleptomania, pyromania, habit-tic deformity, and obsessive-compulsive disorder (OCD). The more well-documented trichotillomania, or pulling of the hair, is estimated to occur in up to 1 in 200 individuals.2 The incidence of onychotillomania is thought to be much lower and widely underreported. However, the incidence may surpass that of trichotillomania when nail biting, nail chewing, or habit-tic deformity is included, though this thought is controversial. In this report, we document 2 patients with slightly different presentations of onychotillomania and the approaches to their therapy.

Case Reports

Patient 1—A 72-year-old white man was referred to the dermatology clinic with an 8-month history of fingernail loss and pain. On physical examination, he was missing 2 nails on the left hand. Prominent longitudinal ridging was observed on the remaining nails, which were thick and yellow and showed some loss of the distal nail plate. All nails on the right hand were normal. Results of a biopsy showed epithelial necrosis with no evidence of lichen planus or inflammation. Fungal culture results were negative.

The patient was confrontational during the visit, protesting the nail examination. He continually drew back in protest, not wanting his nails examined. His wife reported the same, stating that he would slap her hands away anytime she tried to look at his nails. He also reported that there was a "clear goo" under his nails that he thought was necessary to remove by picking. Past medical history was significant for essential tremors, chronic obstructive pulmonary disease, and congestive heart failure. Medications included primidone and gabapentin for essential tremor and alprazolam at night for insomnia.

Based on examination findings and the patient's own admission, the diagnosis of onychotillomania was made. Results of plain film radiographs that were obtained to rule out osteomyelitis were negative. Attempts to use occlusive dressings were made, but the patient refused to keep the nails covered because he was unable to manipulate the remaining nails or nail beds. Also, referral for psychiatric evaluation was vehemently refused. The patient did not return for follow-up.

Patient 2—A 22-year-old white man presented to the dermatology clinic with a several-month history of pain in his toenails. On physical examination, he was missing all nails on his right foot. He had blood on and under all the remaining toenails, blood on all nail beds, and blood under most of his fingernails and on his fingertips. On questioning, the patient adamantly denied pulling his nails, even after confronted with the blood evidence on his fingers and fingernails. His mother reported that he constantly picked at his toenails.

Due to our suspicion of onychotillomania with secondary infection, the patient was treated initially with oral cephalexin followed by placement of an Unna boot on the involved foot with modifications to cover the entire foot and digits. This was changed once weekly. After one month, new healthy nail was noted, but the patient refused our recommendation for psychiatric evaluation and did not follow-up with us as recommended.

Comment

In both of our cases, the diagnosis of onychotillomania was made by the obvious physical signs on examination, as well as by self-admission in patient 1. Although we believed a psychiatric evaluation was necessary for proper treatment, it was refused by both patients.

Psychodermatologic problems can be grouped into 3 categories. Psychophysiologic disorders are exacerbated by emotional stress and include atopic dermatitis and psoriasis.3 Primary psychiatric disorders (anxiety, depression, delusion, and OCD) may present as delusions of parasitosis, neurotic excoriations, trichotillomania, and onychotillomania. In secondary psychiatric disorders, patients endure psychologic or emotional distress as a result of physical or visual deformity caused by primary skin disorders such as acne, leprosy, psoriasis, and vitiligo.4

The term onychotillomania previously has been used to include nail biting in addition to physical deformities caused by self-induced damage to the nails or periungual tissues by picking or pulling. However, it is generally reserved for the manual removal of the nail plate. Examination of individuals with onychotillomania may show periungual erosions and crusts associated with nail-plate surface abnormalities.5 The damage simply may be diminished or missing nails. The matrix melanocytes can be stimulated by chronic trauma, which may result in longitudinal melanonychia.6

Onychotillomania has been classified as a habit-tic deformity that may occur after psychological and emotional stress or as a form of OCD.7 However, the habit-tic deformity may not fit the true definition of onychotillomania, though pharmacologic treatment is similar. Paranoid delusions and psychoses also have been associated with onychotillomania,8 as has Smith-Magenis syndrome. This congenital anomaly associated with mental retardation is estimated to occur in 1 in 25,000 individuals. Self-mutilatory behavior is seen in 70% of patients and includes onychotillomania.9 The differential diagnosis also should include Lesch-Nyhan syndrome.

Treatment of the underlying psychologic disorders should be considered in those with onychotillomania. In addition to onychotillomania, the more common manifestations of OCD in dermatology include trichotillomania, onychophagia, acne excoriee, and neurotic excoriations.10 OCD most frequently is manifested in childhood, though behavior such as obsessive hand washing, AIDS phobia, and other psychosomatic dermatoses can be observed in all age groups.

However, not every patient with onychotillomania has OCD as the underlying psychopathology. Before coming to a conclusion that a patient with onychotillomania has OCD, one must rule out other psychiatric diagnostic possibilities, mainly delusion and simple habit disorder (habit-tic deformity). The key distinction between obsession and delusion is the presence or absence of insight on the part of the patient. Obsessive patients have more insight than delusional patients do. Often, patients with OCD are apologetic for their behavior.10 Patients with habit-tic deformity may be differentiated from true onychotillomania in that they may only rub their nails unconsciously and not actually pick off their nails.

It is important for the underlying psychiatric disorder to be defined by psychiatric evaluation and subsequent treatment with psychoactive medications.11 Common treatments for OCD include individual psychotherapy and behavioral therapy. In addition, there are 3 oral medications commonly used for their anti–obsessive-compulsive effect, namely, clomipramine, fluoxetine, and fluvoxamine.9 Paroxetine, sertraline, the mixed uptake inhibitor venlafaxine, and citalopram are the latest additions for the treatment of OCD.12 Fluoxetine hydrochloride also has been found helpful, specifically in the treatment of onychotillomania.12 In addition, pimozide has been used specifically to treat onychotillomania.8 Topical treatments also have been approached using distasteful preparations applied to the nails to discourage nail biting and chewing.5 The physical barrier method appeared to work quite well in our younger patient, though it was not effective in the treatment of our older patient.

Onychotillomania has been cited in the literature to include both nail pulling and nail biting, but our 2 patients exhibited the most classic form of onychotillomania of picking and pulling of the nails, as the term was originally coined. Neither of our patients had what would be classified as habit-tic deformity. We speculate that onychotillomania can be divided into a nail-pulling/picking group, a nail-biting group, and a combination of the 2. In any case, it is advisable to explore etiologies other than self-inflicted trauma, such as mechanical or friction trauma, fungal infection, or another form of nail dystrophy. However, when the diagnosis of onychotillomania is reached, in addition to occlusion to prevent the self-induced damage, referral to and treatment by a psychiatrist is warranted because of the strong association with underlying psychiatric conditions, namely, OCD. 

Onychotillomania is an uncommon condition characterized by self-destruction of the fingernails and/or toenails by compulsive manipulation. We report 2 cases of onychotillomania with differing presentations in a young man and in an older man. Onychotillomania may be a form of obsessive-compulsive disorder (OCD), and we discuss the psychologic factors and current treatments for this condition.

Emotional and psychologic factors have the ability to influence the underlying disease process in at least 33% of patients with a dermatologic condition.1 In some cases, such as onychotillomania, a psychiatric condition can be the underlining cause. Onychotillomania, a condition whose true incidence is unknown, is characterized by the compulsive or irresistible urge in patients to pick at, pull off, or harmfully bite or chew their nails. This urge may be conscious or unconscious. The word onychotillomania is derived from the Greek onycho, nail; tillo, to pull; and mania, madness or frenzy. In psychiatry, onychotillomania has been classified as an impulse control disorder that includes conditions such as compulsive gambling, kleptomania, pyromania, habit-tic deformity, and obsessive-compulsive disorder (OCD). The more well-documented trichotillomania, or pulling of the hair, is estimated to occur in up to 1 in 200 individuals.2 The incidence of onychotillomania is thought to be much lower and widely underreported. However, the incidence may surpass that of trichotillomania when nail biting, nail chewing, or habit-tic deformity is included, though this thought is controversial. In this report, we document 2 patients with slightly different presentations of onychotillomania and the approaches to their therapy.

Case Reports

Patient 1—A 72-year-old white man was referred to the dermatology clinic with an 8-month history of fingernail loss and pain. On physical examination, he was missing 2 nails on the left hand. Prominent longitudinal ridging was observed on the remaining nails, which were thick and yellow and showed some loss of the distal nail plate. All nails on the right hand were normal. Results of a biopsy showed epithelial necrosis with no evidence of lichen planus or inflammation. Fungal culture results were negative.

The patient was confrontational during the visit, protesting the nail examination. He continually drew back in protest, not wanting his nails examined. His wife reported the same, stating that he would slap her hands away anytime she tried to look at his nails. He also reported that there was a "clear goo" under his nails that he thought was necessary to remove by picking. Past medical history was significant for essential tremors, chronic obstructive pulmonary disease, and congestive heart failure. Medications included primidone and gabapentin for essential tremor and alprazolam at night for insomnia.

Based on examination findings and the patient's own admission, the diagnosis of onychotillomania was made. Results of plain film radiographs that were obtained to rule out osteomyelitis were negative. Attempts to use occlusive dressings were made, but the patient refused to keep the nails covered because he was unable to manipulate the remaining nails or nail beds. Also, referral for psychiatric evaluation was vehemently refused. The patient did not return for follow-up.

Patient 2—A 22-year-old white man presented to the dermatology clinic with a several-month history of pain in his toenails. On physical examination, he was missing all nails on his right foot. He had blood on and under all the remaining toenails, blood on all nail beds, and blood under most of his fingernails and on his fingertips. On questioning, the patient adamantly denied pulling his nails, even after confronted with the blood evidence on his fingers and fingernails. His mother reported that he constantly picked at his toenails.

Due to our suspicion of onychotillomania with secondary infection, the patient was treated initially with oral cephalexin followed by placement of an Unna boot on the involved foot with modifications to cover the entire foot and digits. This was changed once weekly. After one month, new healthy nail was noted, but the patient refused our recommendation for psychiatric evaluation and did not follow-up with us as recommended.

Comment

In both of our cases, the diagnosis of onychotillomania was made by the obvious physical signs on examination, as well as by self-admission in patient 1. Although we believed a psychiatric evaluation was necessary for proper treatment, it was refused by both patients.

Psychodermatologic problems can be grouped into 3 categories. Psychophysiologic disorders are exacerbated by emotional stress and include atopic dermatitis and psoriasis.3 Primary psychiatric disorders (anxiety, depression, delusion, and OCD) may present as delusions of parasitosis, neurotic excoriations, trichotillomania, and onychotillomania. In secondary psychiatric disorders, patients endure psychologic or emotional distress as a result of physical or visual deformity caused by primary skin disorders such as acne, leprosy, psoriasis, and vitiligo.4

The term onychotillomania previously has been used to include nail biting in addition to physical deformities caused by self-induced damage to the nails or periungual tissues by picking or pulling. However, it is generally reserved for the manual removal of the nail plate. Examination of individuals with onychotillomania may show periungual erosions and crusts associated with nail-plate surface abnormalities.5 The damage simply may be diminished or missing nails. The matrix melanocytes can be stimulated by chronic trauma, which may result in longitudinal melanonychia.6

Onychotillomania has been classified as a habit-tic deformity that may occur after psychological and emotional stress or as a form of OCD.7 However, the habit-tic deformity may not fit the true definition of onychotillomania, though pharmacologic treatment is similar. Paranoid delusions and psychoses also have been associated with onychotillomania,8 as has Smith-Magenis syndrome. This congenital anomaly associated with mental retardation is estimated to occur in 1 in 25,000 individuals. Self-mutilatory behavior is seen in 70% of patients and includes onychotillomania.9 The differential diagnosis also should include Lesch-Nyhan syndrome.

Treatment of the underlying psychologic disorders should be considered in those with onychotillomania. In addition to onychotillomania, the more common manifestations of OCD in dermatology include trichotillomania, onychophagia, acne excoriee, and neurotic excoriations.10 OCD most frequently is manifested in childhood, though behavior such as obsessive hand washing, AIDS phobia, and other psychosomatic dermatoses can be observed in all age groups.

However, not every patient with onychotillomania has OCD as the underlying psychopathology. Before coming to a conclusion that a patient with onychotillomania has OCD, one must rule out other psychiatric diagnostic possibilities, mainly delusion and simple habit disorder (habit-tic deformity). The key distinction between obsession and delusion is the presence or absence of insight on the part of the patient. Obsessive patients have more insight than delusional patients do. Often, patients with OCD are apologetic for their behavior.10 Patients with habit-tic deformity may be differentiated from true onychotillomania in that they may only rub their nails unconsciously and not actually pick off their nails.

It is important for the underlying psychiatric disorder to be defined by psychiatric evaluation and subsequent treatment with psychoactive medications.11 Common treatments for OCD include individual psychotherapy and behavioral therapy. In addition, there are 3 oral medications commonly used for their anti–obsessive-compulsive effect, namely, clomipramine, fluoxetine, and fluvoxamine.9 Paroxetine, sertraline, the mixed uptake inhibitor venlafaxine, and citalopram are the latest additions for the treatment of OCD.12 Fluoxetine hydrochloride also has been found helpful, specifically in the treatment of onychotillomania.12 In addition, pimozide has been used specifically to treat onychotillomania.8 Topical treatments also have been approached using distasteful preparations applied to the nails to discourage nail biting and chewing.5 The physical barrier method appeared to work quite well in our younger patient, though it was not effective in the treatment of our older patient.

Onychotillomania has been cited in the literature to include both nail pulling and nail biting, but our 2 patients exhibited the most classic form of onychotillomania of picking and pulling of the nails, as the term was originally coined. Neither of our patients had what would be classified as habit-tic deformity. We speculate that onychotillomania can be divided into a nail-pulling/picking group, a nail-biting group, and a combination of the 2. In any case, it is advisable to explore etiologies other than self-inflicted trauma, such as mechanical or friction trauma, fungal infection, or another form of nail dystrophy. However, when the diagnosis of onychotillomania is reached, in addition to occlusion to prevent the self-induced damage, referral to and treatment by a psychiatrist is warranted because of the strong association with underlying psychiatric conditions, namely, OCD. 

Onychotillomania is an uncommon condition characterized by self-destruction of the fingernails and/or toenails by compulsive manipulation. We report 2 cases of onychotillomania with differing presentations in a young man and in an older man. Onychotillomania may be a form of obsessive-compulsive disorder (OCD), and we discuss the psychologic factors and current treatments for this condition.

Emotional and psychologic factors have the ability to influence the underlying disease process in at least 33% of patients with a dermatologic condition.1 In some cases, such as onychotillomania, a psychiatric condition can be the underlining cause. Onychotillomania, a condition whose true incidence is unknown, is characterized by the compulsive or irresistible urge in patients to pick at, pull off, or harmfully bite or chew their nails. This urge may be conscious or unconscious. The word onychotillomania is derived from the Greek onycho, nail; tillo, to pull; and mania, madness or frenzy. In psychiatry, onychotillomania has been classified as an impulse control disorder that includes conditions such as compulsive gambling, kleptomania, pyromania, habit-tic deformity, and obsessive-compulsive disorder (OCD). The more well-documented trichotillomania, or pulling of the hair, is estimated to occur in up to 1 in 200 individuals.2 The incidence of onychotillomania is thought to be much lower and widely underreported. However, the incidence may surpass that of trichotillomania when nail biting, nail chewing, or habit-tic deformity is included, though this thought is controversial. In this report, we document 2 patients with slightly different presentations of onychotillomania and the approaches to their therapy.

Case Reports

Patient 1—A 72-year-old white man was referred to the dermatology clinic with an 8-month history of fingernail loss and pain. On physical examination, he was missing 2 nails on the left hand. Prominent longitudinal ridging was observed on the remaining nails, which were thick and yellow and showed some loss of the distal nail plate. All nails on the right hand were normal. Results of a biopsy showed epithelial necrosis with no evidence of lichen planus or inflammation. Fungal culture results were negative.

The patient was confrontational during the visit, protesting the nail examination. He continually drew back in protest, not wanting his nails examined. His wife reported the same, stating that he would slap her hands away anytime she tried to look at his nails. He also reported that there was a "clear goo" under his nails that he thought was necessary to remove by picking. Past medical history was significant for essential tremors, chronic obstructive pulmonary disease, and congestive heart failure. Medications included primidone and gabapentin for essential tremor and alprazolam at night for insomnia.

Based on examination findings and the patient's own admission, the diagnosis of onychotillomania was made. Results of plain film radiographs that were obtained to rule out osteomyelitis were negative. Attempts to use occlusive dressings were made, but the patient refused to keep the nails covered because he was unable to manipulate the remaining nails or nail beds. Also, referral for psychiatric evaluation was vehemently refused. The patient did not return for follow-up.

Patient 2—A 22-year-old white man presented to the dermatology clinic with a several-month history of pain in his toenails. On physical examination, he was missing all nails on his right foot. He had blood on and under all the remaining toenails, blood on all nail beds, and blood under most of his fingernails and on his fingertips. On questioning, the patient adamantly denied pulling his nails, even after confronted with the blood evidence on his fingers and fingernails. His mother reported that he constantly picked at his toenails.

Due to our suspicion of onychotillomania with secondary infection, the patient was treated initially with oral cephalexin followed by placement of an Unna boot on the involved foot with modifications to cover the entire foot and digits. This was changed once weekly. After one month, new healthy nail was noted, but the patient refused our recommendation for psychiatric evaluation and did not follow-up with us as recommended.

Comment

In both of our cases, the diagnosis of onychotillomania was made by the obvious physical signs on examination, as well as by self-admission in patient 1. Although we believed a psychiatric evaluation was necessary for proper treatment, it was refused by both patients.

Psychodermatologic problems can be grouped into 3 categories. Psychophysiologic disorders are exacerbated by emotional stress and include atopic dermatitis and psoriasis.3 Primary psychiatric disorders (anxiety, depression, delusion, and OCD) may present as delusions of parasitosis, neurotic excoriations, trichotillomania, and onychotillomania. In secondary psychiatric disorders, patients endure psychologic or emotional distress as a result of physical or visual deformity caused by primary skin disorders such as acne, leprosy, psoriasis, and vitiligo.4

The term onychotillomania previously has been used to include nail biting in addition to physical deformities caused by self-induced damage to the nails or periungual tissues by picking or pulling. However, it is generally reserved for the manual removal of the nail plate. Examination of individuals with onychotillomania may show periungual erosions and crusts associated with nail-plate surface abnormalities.5 The damage simply may be diminished or missing nails. The matrix melanocytes can be stimulated by chronic trauma, which may result in longitudinal melanonychia.6

Onychotillomania has been classified as a habit-tic deformity that may occur after psychological and emotional stress or as a form of OCD.7 However, the habit-tic deformity may not fit the true definition of onychotillomania, though pharmacologic treatment is similar. Paranoid delusions and psychoses also have been associated with onychotillomania,8 as has Smith-Magenis syndrome. This congenital anomaly associated with mental retardation is estimated to occur in 1 in 25,000 individuals. Self-mutilatory behavior is seen in 70% of patients and includes onychotillomania.9 The differential diagnosis also should include Lesch-Nyhan syndrome.

Treatment of the underlying psychologic disorders should be considered in those with onychotillomania. In addition to onychotillomania, the more common manifestations of OCD in dermatology include trichotillomania, onychophagia, acne excoriee, and neurotic excoriations.10 OCD most frequently is manifested in childhood, though behavior such as obsessive hand washing, AIDS phobia, and other psychosomatic dermatoses can be observed in all age groups.

However, not every patient with onychotillomania has OCD as the underlying psychopathology. Before coming to a conclusion that a patient with onychotillomania has OCD, one must rule out other psychiatric diagnostic possibilities, mainly delusion and simple habit disorder (habit-tic deformity). The key distinction between obsession and delusion is the presence or absence of insight on the part of the patient. Obsessive patients have more insight than delusional patients do. Often, patients with OCD are apologetic for their behavior.10 Patients with habit-tic deformity may be differentiated from true onychotillomania in that they may only rub their nails unconsciously and not actually pick off their nails.

It is important for the underlying psychiatric disorder to be defined by psychiatric evaluation and subsequent treatment with psychoactive medications.11 Common treatments for OCD include individual psychotherapy and behavioral therapy. In addition, there are 3 oral medications commonly used for their anti–obsessive-compulsive effect, namely, clomipramine, fluoxetine, and fluvoxamine.9 Paroxetine, sertraline, the mixed uptake inhibitor venlafaxine, and citalopram are the latest additions for the treatment of OCD.12 Fluoxetine hydrochloride also has been found helpful, specifically in the treatment of onychotillomania.12 In addition, pimozide has been used specifically to treat onychotillomania.8 Topical treatments also have been approached using distasteful preparations applied to the nails to discourage nail biting and chewing.5 The physical barrier method appeared to work quite well in our younger patient, though it was not effective in the treatment of our older patient.

Onychotillomania has been cited in the literature to include both nail pulling and nail biting, but our 2 patients exhibited the most classic form of onychotillomania of picking and pulling of the nails, as the term was originally coined. Neither of our patients had what would be classified as habit-tic deformity. We speculate that onychotillomania can be divided into a nail-pulling/picking group, a nail-biting group, and a combination of the 2. In any case, it is advisable to explore etiologies other than self-inflicted trauma, such as mechanical or friction trauma, fungal infection, or another form of nail dystrophy. However, when the diagnosis of onychotillomania is reached, in addition to occlusion to prevent the self-induced damage, referral to and treatment by a psychiatrist is warranted because of the strong association with underlying psychiatric conditions, namely, OCD.