Dissecting cellulitis of the scalp: a retrospective study of 51 patients and review of literature

Dear Editor, Dissecting cellulitis of the scalp (DCS), also known as perifolliculitis capitis abscedens et suffodiens, or Hoffmann disease, is a rare primary neutrophilic cicatricial alopecia of unknown aetiology. DCS has been considered to be part of the ‘follicular occlusion tetrad’ along with hidradenitis suppurativa, acne conglobata and pilonidal cyst, suggesting a common pathogenic mechanism: deep follicular occlusion, follicular rupture, then follicular infection.1 2 3 Since the first description of DCS, the details of only 72 patients have been published. Several treatments have been tried; among them, only systemic retinoids achieved complete remission. We conducted a retrospective study of patients with DCS from two dermatology departments in Paris, France, from 1996 to 2013. DCS was clinically defined as cicatricial alopecia associated with nodules or abscesses of the scalp. Clinical findings, clinical examination, triggers and association with another disease from the ‘follicular occlusion tetrad’ were collected. Bacteriological and mycological samples were systematically examined to rule out an infectious aetiology. Between 1996 and 2013, 51 patients (50 men, one woman) were seen at the two centres. Nineteen patients of the 29 whose phototype was mentioned (65%) had a dark phototype (phototype 4, 5 or 6). None had a noteworthy family history, including family history of DCS. None had notable personal medical history. Mean age at DCS onset was 26·6 years (range 15·0–62·0). Mean duration of symptoms before diagnosis was 34·3 months (range 4·0–120·0). Subcutaneous nodules and abscesses were principally located on the vertex (25 patients; 49%), and were diffuse over the entire scalp in five patients (Fig. 1). The severity of DCS was scored by the clinician and based on the following criteria: pain, number of nodules, surface of the affected scalp, inflammation and suppuration. Two per cent, 61% and 25% of patients were considered as having mild, moderate and severe DCS, respectively. A traumatic trigger was found in only five patients: hair shaving; neurosurgery for epilepsy or after wearing a helmet. Nodules were painful in 13 patients (25%) and itchy in four (8%). The pattern of disease progression appeared to be more chronic with progressive onset of lesions (44 patients; 86%) than acute (seven patients; 14%). No squamous cell carcinoma was reported. Severe and diffuse dissecting cellulitis of the scalp associated with abscesses of the axillary and pubic areas, suggestive of hidradenitis suppurativa. Among the 51 patients, six also had hidradenitis suppurativa (12%), eight had acne conglobata (16%) and two had both acne conglobata and hidradenitis suppurativa (4%); none had pilonidal cyst, psoriasis, inflammatory bowel disease or inflammatory arthritis. Among the treatments used, 40 patients (78%) received antibiotics (doxycycline, pristinamycin, rifampicin or a combination of several antibiotics), with moderate improvement seen in all patients; however, the patients relapsed after discontinuation of treatment. Eight patients (16%) received zinc supplementation with no efficacy. Eight patients (16%) had surgical treatment (excision or abscess drainage) to reduce pain, but there was no effect on DCS progression. Thirty‐six patients (71%) received systemic retinoids: 35 patients received isotretinoin (0·5–0·8 mg kg−1 daily) with complete remission in 33 patients at 3 months but frequent relapse after discontinuation; one patient received acitretin at 30 mg daily but was lost to follow‐up. Only one patient received a tumour necrosis factor‐α inhibitor (infliximab) without any improvement. Ten patients (20%) received other treatments, including dapsone, topical or systemic corticosteroids, topical or systemic antifungals, and topical erythromycin, without efficacy. The mean follow‐up was 11·2 months for all patients, and 6·7 months for patients treated by isotretinoin. Compared with data in the literature, age of onset was the same but our patients were more frequently male (98% in this study vs. 84% in the literature),1 2 3 4 5 6 7 8 9 10 and 65% of patients in this study had a dark phototype vs. 48% in literature. Unlike our series, two patients in the literature had a family history of DCS.11 Clinical findings were consistent with data in the literature: nodules and abscesses predominated on the vertex area (49% in our series vs. 56% in the literature), followed by the occipital area (25% in our series vs. 50% in the literature). The temporal and parietal areas were less affected by DCS (6% and 4% in our series vs. 1% and 3% in the literature, respectively). Six patients initially presented only one unique nodule of the scalp, with progressive worsening and an increase in the number of nodules. This progression suggests that the recently described entity of ‘pseudocyts of the scalp’ could be an early stage of DCS.12 In our series, few patients had a disease associated with the follicular tetrad. These results suggest that CDS and hidradenitis suppurativa are two associated diseases but do not belong to the same clinical entity; also, DCS affects more young afro‐Caribbean men but hidradenitis suppurativa affects more young afro‐Caribbean women. The pathophysiology of DCS remains unclear. DCS is a neutrophilic alopecia, suggesting an important role for neutrophils, especially via interleukin‐1. Also, the association with hidradenitis suppurativa reinforces the role of neutrophils in the onset of both diseases. The male dominance, the young age of onset, the occurrence in patients with a dark phototype (especially afro‐Caribbean patients) and the cases of familiar DCS reported in the literature suggest a genetic risk factor.11 The male dominance and the predominance of lesions on vertex could also suggest a hormonal risk factor. The predominance of patients with a dark phototype, especially afro‐Caribbean patients, raises the question of the role of hair type, which is often coarse and frizzy, as well as an initial traumatic factor, such as hair shaving. Finally, microbiota may play an important role as alloantigens in the pathophysiology of DCS.13 The loss of immune tolerance to alloantigens in the hair follicle may lead to an inflammatory reaction; the improvement of lesions after antibiotics favours this hypothesis. The initial traumatic factor could also lead to a break in the skin barrier. Treatment of DCS is difficult. The most effective treatment seen in our study was isotretinoin 0·5–0·8 mg kg−1 daily, with complete remission in 92% of patients. However, this remission did not seem to last, with frequent relapses after discontinuation of isotretinoin. Accepted as a communication at the Annual Meeting of the French Society of Dermatology (10–13 December 2014). Funding sources: none. Conflicts of interest: none declared.