Topical Clindamycin Versus Systemic Tetracycline in the Treatment of Hidradenitis Suppurativa

Topical clindamycin versus systemic tetracycline in the treatment of hidradenitis suppurativa

Gregor B. E. Jemec, MD, DMedSci,a and Peter Wendelboe, MDb Copenhagen and Århus, Denmark

Background: Antibiotics are often used to treat hidradenitis, but only topical clindamycin has been shown to be effective in a randomized controlled trial. The paucity of these trials may be the result of difficulties in disease assessment. Objective: We compare topical clindamycin with systemic tetracycline in the treatment of hidradenitis suppurativa, and study clinical disease assessment. Methods: A total of 46 patients with stage 1 or 2 hidradenitis suppurativa were treated in a double-blind, double dummy controlled trial. Results: No significant difference was found between the two types of treatment. Patients’ global assessment of disease was significantly worse than physician’s assessment in 3 of 5 evaluations (P = .0096 to .015), but the correlation between patients’ and physicians’ assessments was satisfactory after only one visit (rs = .761 to .895). Soreness was the key factor in patients’ overall assessment of the disease. Conclusion: Systemic therapy with tetracyclines did not show better results than topical therapy with clindamycin. Subjective factors, particularly soreness, appear to be a key factor in patients’ assessment of the disease and should, therefore, be included as an outcome variable in future therapy studies. (J Am Acad Dermatol 1998;39:971-4.)

Different types of antibiotic therapy are com- ously published trial, and compared the clinical monly used in the treatment of hidradenitis suppu- efficacy of topical clindamycin against that of sys- rativa although viable bacteria are only found in temic tetracycline focusing on the disease assess- approximately 50% of all lesions.1,2 Only topical ment in an attempt to provide a better criteria for clindamycin has, however, been tested in a ran- future trials in hidradenitis. domized controlled trial (RCT) in a study by Clemmensen.3 To our knowledge, the efficacy of MATERIALS AND METHODS other antibiotics has never been shown in an RCT. The protocol was approved by the local Ethics Clinically relevant and objective disease assess- Committee for Copenhagen. Patients were recruited ment is a prerequisite for rational studies of thera- from those referred for treatment or patients in whom py, and no standard disease assessment in hidradenitis suppurativa was an incidental finding on hidradenitis exists. Hurley4 has proposed a staging physical examination. After informed consent was of the disease that is of great general value, but obtained, a total of 46 patients (39 women and 7 men) focuses on elements such as scarring and sinus with hidradenitis suppurativa were included in the tract formation. These are long-term changes and study. A group of 12 patients were not available for may, therefore, not be sufficiently dynamic for out- evaluation (7 did not show up for scheduled visits, 2 discontinued treatment because of gastrointestinal come analysis in shorter clinical trials. upset, and 1 each because of lack of effect, concomitant We have conducted an RCT based on the previ- treatment with another antibiotic, and suspected allergic reaction to topical medication). From the Department of Dermatology, Bispebjerg Hospital, The characteristics of the patients who completed University of Copenhagen,a and the Department of Dermatology, the study are given in Table I. Treatments were com- Marselisborg Hospital, Århus University.b pared in a double-blind double dummy study by means Reprint requests: Gregor B. E. Jemec, Strandvejen 97, 3.tv., DK-2900 Hellerup, Denmark. of parallel groups and computerized blinded random- Copyright © 1998 by the American Academy of Dermatology, Inc. ization. Uniform containers, placebo tablets, and place- 0190-9622/98/$5.00 + 0 16/1/93688 bo lotion were provided by the Upjohn Company. After

971 Journal of the American Academy of Dermatology 972 Jemec and Wendelboe December 1998

Table I. Baseline data about the patients enrolled, given as median and 95% confidence intervals

Systemic treatment group Topical treatment group

Total No. of patients enrolled 24 22 Age (y) 31.8 (27.3-36.5) 33.3 (28.9-37.8) Drop-outs 8 4 Sex of evaluated patients (M/F) 3/13 3/15 Patient assessment at baseline 35.9 (20.4-51.5) 48.0 (31.5-64.5) Physician assessment at baseline 31.6 (19.1-44.0) 25.7 (13.6-37.8) Nodules at baseline 2.5 (1.2-3.8) 2.9 (1.6-4.2) Abscesses at baseline 1.9 (1.2-2.5) 0.8 (0.1-1.4) Scarring at baseline (% of area) 45.2 (12.9-77.5) 27.8 (0-58.1) Total number of adverse events in group 3 5

informed consent was obtained, consecutive patients lowing outcome variables: patient global evaluation on with early-stage (Hurley stage 1 or 2: single abscesses 100 mm Visual Analogue Scale (VAS) score, soreness without sinus tracts, or recurrent but widely separated evaluation on 100 mm VAS score, physician global lesions with sinus tracts and scarring) hidradenitis sup- evaluation on 100 mm VAS score, counting of abscess- purativa were assigned according to the prerandomized es, and counting of nodules. sequence of treatments. Dropouts were replaced, and The results were assessed by means of analysis of appropriate additional medication was delivered with- variance and a Wilcoxon 2-sample test for change from out breaking the double-blind design of the study. the baseline. The results of assessment were compared Pregnant or breast-feeding mothers were not includ- by means of Spearman correlations and stepwise ed. Additional clinical exclusion criteria were acne con- regression analysis. Results were reported with median globata (active disease on convex areas of the body); values and 95% confidence interval (CI). signs of staphylococcal infection; staphylococcosis; treatment with a systemic or topical antibiotic within RESULTS the past 7 days; known or suspected hypersensitivity to tetracycline, lincosamides, or any constituents of the No significant differences were found between medication or placebo given; signs of systemic infec- the two treatments, but significant changes tion; history of impaired renal or liver function; known occurred in the course of the study (Figs 1 and 2). severe underlying disease; treatment with steroids with- The presence of bacteria at the onset of treatment in the past 7 days; treatment with depo-steroids within did not affect the outcome. the past 6 weeks; chronic bowel diseases or diarrhea; Significant correlations were found between treatment with cyproterone acetate within the past 6 months; and more than 10 lesions (nodules/abscesses) disease features. Patients assessed the disease as from all sites together or Hurley stage 3 (diffuse significantly worse than physicians in 3 of 5 eval- involvement, multiple interconnected abscesses, and uations (P = .0096 to .015). Stepwise regression multiple sinus tracts). revealed that soreness and nodules were signifi- Before treatment an aseptic aspiration was taken cantly correlated with the patients’ overall evalua- from inflamed lesions. If aspiration was not directly tion at the first visit, whereas soreness, abscesses, possible 0.5 mL of isotonic saline was injected and and the physicians’ overall evaluation were signif- aspirated. The bacteriologic analyses have been icantly correlated at the last visit. For the physi- described in detail elsewhere.2 cians the number of abscesses was significantly All patients received a minimum of 3 months of ther- correlated with the overall disease assessment at apy with systemic as well as topical treatment, that is, the first visit, whereas the patients’ overall assess- active systemic plus topical placebo, or systemic place- ment appeared to be the only factor significantly bo plus active topical. Active systemic treatment consisted of tetracycline 1 g daily, taken as two 250 mg associated with the physicians’ overall evaluation capsules twice daily. Active topical treatment consisted at the last visit. of 1% clindamycin phosphate (Dalacin T, Upjohn Co) There was a significant correlation between in a vehicle of propylene glycol, isopropyl alcohol, and patients’ and physicians’ global assessment of the water, applied twice daily. disease at every session (all P < .001; rs = .551 Patients were seen monthly for evaluation of the fol- [baseline], rs = .762 [4 weeks], rs = .836 [8 Journal of the American Academy of Dermatology Volume 39, Number 6 Jemec and Wendelboe 973

Fig 2. Changes in objective parameters of evaluation for the 34 patients (16 tetracycline-treated and 18 clindamycin-treated) who completed the study given as median values and 95% CI. Clear bars, tetracycline group; striped bars, clindamycin group.

exist, but the only RCT of antibiotics describes the effect of topical clindamycin.3 Topical clindamycin was therefore compared with systemic tetracycline. No significant differences were seen between the two treatments. Two types of lesions were described: abscesses and nodules. Abscesses were reduced during the Fig 1. Changes in subjective parameters of evaluation first 3 months of treatment. Nodules, on the other for the 34 patients (16 tetracycline-treated and 18 clin- hand, only appeared to be reduced in number after damycin-treated) who completed the study given as 3 months of treatment. The time course suggests median values and 95% CI. Clear bars, tetracycline that nodules may be precursors to abscesses, as group; striped bars, clindamycin group. previously suggested by epidemiologic studies. A progressive improvement was noticed in sub- weeks], rs = .895 [12 weeks], and rs = .787 [16 jective overall assessment and physician assess- weeks]). ment, although soreness did not appear to change during the course of the study. The variation in DISCUSSION individual pain thresholds may be important in Antibiotics are commonly used in early-stage coping with hidradenitis. Coping and patients’ hidradenitis, although the evidence for their effica- general assessment of disease intensity for skin cy is limited. Several case reports and open studies diseases is also influenced by factors other than the Journal of the American Academy of Dermatology 974 Jemec and Wendelboe December 1998 diagnosis itself. One of these may be the visibility Second, an increased consensus between of the lesions in normal social intercourse. patients’ and physicians’ assessment of the disease Previous studies have suggested that acne patients was observed at the end of our study. The correla- with facial acne cope better than patients with pre- tion coefficient at baseline was only 0.551, where- dominantly truncal acne because no support or as it was 0.762 at the visit only 4 weeks later. This sympathy is elicited by hidden lesions.5 This suggests that a mutual learning process takes place mechanism may be relevant in hidradenitis, and and that this type of soft data gains validity as may explain some of the difference in the overall studies become longer. These observations suggest assessment of the disease by patients and physi- that the duration of future studies on early cian. The impact of professional knowledge on the hidradenitis should not contain less than 3 visits at physician’s assessment also is involved. A previ- 1-month intervals. ous study suggested that the impact of medically REFERENCES “benign” diseases on quality-of-life may be under- 1. Highet AS, Hay RJ, Roberts SOB. Bacterial infections. estimated by the physician, whereas medically In: Champion RH, Burton JL, Ebling FJG, editors. “malignant” diseases (eg, generalized connective Textbook of dermatology. Oxford: Blackwell Scientific tissue diseases or frank malignancy) are often Publications; 1992. p. 953-1031. underestimated by the patients.6 2. Jemec GBE, Faber M, Gutschick E, et al. Microbiology of hidradenitis suppurativa. Dermatology 1996;193:203- In the assessment of congruity between the dif- 6. ferent outcome variables studied, two key points 3. Clemmensen OJ. Topical treatment of hidradenitis sup- were noticed. First, patients and physicians con- purativa with topical clindamycin. Int J Dermatol 1983; 22:325-8. sidered different points important, and second, 4. Hurley HJ. Axillary hyperhidrosis, apocrine bromhidro- overall agreement improved in the course of the sis, hidradenitis suppurative, and familial benign pem- trial. phigus: surgical approach. In: Roenigk RK, Roenigk HHJ, editors. Dermatologic surgery: principles and prac- The most important factor in the patients’ over- tice. New York: Marcel Dekker; 1989. p. 717-43. all assessment of disease severity appears to be 5. Simpson N. Effect of isotretinoin on the quality of life in soreness. This is in good accordance with previous patients with acne. Pharmacoeconomics 1994;6:108-13. studies.7 In lieu of true objective measures of dis- 6. Jemec GBE, Wulf HC. Patient-doctor consensus on quality-of-life in dermatology. Clin Exp Dermatol 1996;21: ease activity in hidradenitis suppurativa it is, there- 177-9. fore, suggested that standard clinical outcome 7. Jemec GBE, Heidenheim M, Nielsen NH. Hidradenitis variables are supplemented with records of the suppurativa: characteristics and consequences. Clin Exp Dermatol 1996;21:419-23. patients’ subjective assessment of pain and overall disease activity.

BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the Journal of the American Academy of Dermatology are available to subscribers (only) for the 1998 issues from the Publisher at a cost of $112.00 for domestic, $139.10 for Canadian, and $130.00 for inter- national for volume 38 (January-June) and volume 39 (July-December). Shipping charges are included. Each bound volume contains a subject and author index and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby, Inc., Periodical Subscription Services, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318. USA: phone (800) 453-4351; (314) 453-4351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular journal subscription.