BRITISH MEDICAL JOURNAL VOLUME 291 2 NOVEMBER 1985 1231 Br Med J (Clin Res Ed): first published as 10.1136/bmj.291.6504.1231 on 2 November 1985. Downloaded from

PAPERS AND SHORT REPORTS

Acne: double blind clinical and laboratory trial of tetracycline, oestrogen-cyproterone acetate, and combined treatment

R GREENWOOD, L BRUMMITT, B BURKE, W J CUNLIFFE

Abstract Introduction Since the recent introduction of a drug regimen containing 2 mg Interest in the hormonal treatment of acne in women has been of the antiandrogen cyproterone acetate and 50 [ig ethinyl- present since the sebaceous gland was found to be sensitive to oestradiol (Diane; oestrogen-cyproterone acetate) several un- androgens.' Oestrogens have been used effectively to treat acne but controlied reports have extolled the benefits ofthis drug. Double the dose required causes unacceptable side effects.2 Plainly, the blind studies, however, are lacking. Sixty two patients with ideal hormonal treatment would be one that affected only the skin, moderate or moderately severe acne were therefore included in a but no successful topical hormonal treatment is yet available. Oral double blind trial of treatment for six months comparing tetra- and topical antiandrogenic treatments have been tried since cycline alone, oestrogen-cyproterone acetate alone, and a Neumann and Elger first reported the effect of cyproterone acetate combination ofthese agents. Sebum excretion rates and bacterial on mouse sebaceous glands.' In addition to being antiandrogenic, counts were measured before, during, and after treatment, at the cyproterone acetate is progestogenic, such that in combination with same time as a clinical assessment was made. oestrogen it provides contraceptive cover on a par with conventional acne assessed overali had combined oral and has used extensively in At six months the (as by grade) contraceptives been http://www.bmj.com/ improved by 68% in the antibiotic treated group and by 74% in Europe. In the treatment ofacne and related conditions it was given the oestrogen-cyproterone treated group. The group given a initially as the so called reverse sequential regimetx of Hammerstein combination of both agents improved by 82%, which was and Cupceancu,4 when 50-200 mg cyproterone acetate daily for 10 significantly better (p<0025) than the improvement in the tetra- days of the menstrual cycle was combined with 40-60 [tg ethinyl- cycline treated patients. No significant difference was found oestradiol given daily for 21 days of the cycle. More recently a between the groups given oestrogen-cyproterone alone and the preparation containing 2 mg cyproterone acetate and 50 [ig combined treatment. The sebum excretion rate was suppressed ethinyloestradiol (Diane) has been introduced as useful for the by 25% in the patients in both groups receiving oestrogen- treatment ofacne in women. In particular it is not known how this cyproterone but not in the group given antibiotics alone. treatment compares with antibiotic treatment of acne. We set up a on 2 October 2021 by guest. Protected copyright. Oestrogen-cyproterone acetate is as effective as antibiotics in study to examine this point. Furthermore, as a combination treating acne in women, and adding antibiotics offers no ad- treatment of the oestrogen-cyproterone preparation and antibiotics vantage over using oestrogen-cyproterone on its own, although in may be seen as possibly superior to a single entity this was this study the combination was more effective than tetracycline compared as well. Women who had already tried antibiotics for alone at six months. their acne were chosen for a double blind randomised study. Oestrogen-cyproterone and antibiotic treatment influence different aetiological factors of acne and so the effect on sebum excretion and surface bacteria was assessed.

Patients and methods

Leeds Foundation for Dermatological Research, Department ofDermatology, The We recruited 92 women aged 16-30 years with acne. All were referred General Infirmary, Leeds LS1 3EX from their general practitioners and they thereby represented a normal R GREENWOOD, MA, MRCP, consultant dermatologist population with acne. Most had had low dose antibiotic treatment. A L BRUMMITT, ssc, research associate separate observer allocated the patients at random to one of the following B BURKE, MB, CHB, clinical assistant three treatment groups. Treatment was for six months. W J CUNLIFFE, MD, Frcp, consultant dermatologist Group 1 received the oestrogen-cyproterone preperation for 21 days out of 28 with tetracycline 500 mg twice daily continuously (combination treat- Correspondence to: Dr Cunliffe. ment). 1232 BRITISH MEDICAL JOURNAL VOLUME 291 2 NOVEMBER 1985 Br Med J (Clin Res Ed): first published as 10.1136/bmj.291.6504.1231 on 2 November 1985. Downloaded from Group 2 received oestrogen-cyproterone for 21 days out of28 plus placebo treatment (Diane plus tetracycline) produced significantly greater improve- tetracycline taken twice daily continuously. ment than did tetracycline alone, both for overall grade (fig 2; p<0 025) and Group 3 received placebo oestrogen-cyproterone for 21 days out of28 and for inflamed lesions (fig 4; p<0-001). Oestrogen-cyproterone alone was not tetracycline 500 mg twice daily continuously. significantly different from either of the two treatments. Whether receiving placebo or active tetracycline the patients were told to Sebum excretion rate (fig 5>-Surprisingly, the antibiotic treated group take the antibiotic half an hour before food with a sip of water to maximise showed an increase of the order of 7-13% in the sebum excretion rate, this absorption.7 No topical treatment was given. being significant only at the end of the second month. Both oestrogen- Patients were assessed at an initial baseline visit, then every two months cyproterone treated groups showed a significant reduction in sebum for six months. A final assessment was made two months after stopping excretion at two, four, and six months. Maximum suppression of 26-9% in treatment. Each patient was observed clinically by one observer. Clinical the group given the oestrogen-cyproterone preparation alone was seen by evaluation comprised grading and counting oflesions.8 The sebum excretion four months, and the combined treatment produced suppression of23-7% at rate was measured at each visit by the gravimetric technique.9 10 Micro- six months. The effects of both hormonal regimens were significantly organisms on the surface ofthe right cheek of57 patients (22 in group 1, 18 in different from that of tetracycline alone (p<0-001), and at two months group 2, 17 in group 3) were sampled using the scrub technique of patients given oestrogen-cyproterone alone showed greater suppression of Williamson and Kligman."I Diluted samples were spread on either rein- their sebum excretion than that achieved by the combined treatment forced clostridial medium (Oxoid) containing 6,ug furazolidone (Eaton (p<0 05). Laboratories) per ml for propionibacteria or blood agar base (Oxoid) for Bacteriology-None ofthe treatments caused a significant reduction in the micrococcaceae. Aerobic plates were incubated for 48 hours at 37°C and density of micrococcaceae on cheek skin. Both antibiotic regimens achieved anaerobic cultures for seven days in a cold catalyst anaerobic jar (atmosphere a significant reduction in counts of propionibacteria of the order of 1 log 10% carbon dioxide, 90% hydrogen) at 37°C. Counts were expressed as cycle, but there was no significant difference between these treatments in the log,0 colony forming units per cm2 skin. reductions caused (fig 6). The numbers of patients dropping out and the reasons for doing so were Relapse data-Although not part of the original protocol, we followed up analysed by x2 and Fisher's exact tests. Paired t tests were used to compare where possible those patients who could attend two months after stopping variation in the weights of the rather small number ofpatients (48 out of92) treatment. We were able to investigate 20 (80%) of the 25 patients given the who had their weight recorded at both the first and fourth visits. combined treatments, 14 (67%) of the 21 given the oestrogen-cyproterone Sebum excretion rates and the clinical assessments of lesions were preparation alone, and 9 (56%) of the 16 given the antibiotic. Counts of compared within groups between visits by paired t tests and the results inflamed and non-inflamed lesions showed no change; the improved overall converted to percentage for display. For comparison between groups the grade was maintained in the hormonal treatment groups but the tetracycline Wilcoxon test and the Kruskal-Wallis one way analysis of variance were group showed a relapse of their acne (fig 1). Intergroup analyses showed that used. the antibiotic treated group fared significantly less well than patients given Rises and falls were also considered, since an apparent lack of change in a the combined regimen (p<0005); on the face improvement was maintained mean or median may be the result of large variations in counterbalancing best by both the combined treatment group (p<0-001) and the group given directions, and this was done using the sign test. the oestrogen-cyproterone preparation alone (p<0025) (fig 1). The sebum excretion rate returned towards pretreatment values after both regimens of oestrogen-cyproterone (fig 5). Results The numbers recruited for the three groups starting the trial were uneven because of an error of randomisation. The difference, however, was not statistically significant and had no biasing effect on the results. Discussion Attendance failures and side effects-Thirty patients dropped out of the This study confirmed several predicted results. It showed the trial, about equal numbers dropping out of each treatment group (Table). benefit of tetracycline alone and of the oestrogen-cyproterone preparation alone. There are few studies showing long term benefit of oral tetracycline taken over six months.'2 Most are of shorter Reasonsforpatients dropping out oftrial duration, usually three or four months or less. This study showed that with tetracycline continuing improvement was seen at two, Oestrogen- four, and six months as assessed by the overall grades. Spot counts Combination cyproterone Tetracycline suggested that a steady state ofimprovement occurred between four

treatment alone alone http://www.bmj.com/ and six months, but the discrepancy between grade and counts may Lack of compliance 4 2 5 be explained by the observation that inflamed acne lesions show Lack of efficacy 2 - I Wished to'become pregnant 1 - 2 various intensities of inflammation.8 In this study we did not Sebum excretion rate measurements distinguish between the more active inflamed and less active too time consuming 1 5 1 inflamed lesions. Clinical studies with oestrogen-cyproterone Side effects: preparation using less refined assessment techniques have shown Chest pain 1 - Headaches 1 that improvement is maintained over a six month treatment period,6 Menstrual irregularity 1 and our findings support this. As with tetracycline alone, oestrogen- Breast tendemess 1 on 2 October 2021 by guest. Protected copyright. Increased blood pressure - 1 cyproterone acetate produced a maintained improvement over the Dysmcnorrhoca - 1 - second, fourth, and sixth months. There have been no other studies to show whether adding antibiotics brings greater benefit. Both Total 12 9 9 drugs work in acne through different mechanisms. Oestrogen- cyproterone reduces the sebum excretion rate'"; antibiotics can affect the function of Propionibacterium acnes'4 and probably affect the host inflammatory mechanisms. It seemed likely that there may The commonest reason for exclusion from the final results was lack of compliance (n= 11). Seven failed because of the time consuming nature of be synergism between the oestrogen-cyproterone preparation and the measurements of sebum excretion rate. Side effects were common but tetracycline. This study, however, showed limited clinical benefit of usually minor. One patient taking the oestrogen-cyproterone preparation the drug combination. developed mild but definite hypertension necessitating withdrawal; the Our study produced a few unexpected results. The tetracycline blood pressure had returned to normal two months after stopping the drug. treated group showed an overall increase ofthe sebum excretion rate There was a small but significant weight gain in both oestrogen-cyproterone of the order of 7-13%, just reaching significance at two months. treated groups. Other studies have failed to show an effect of antibiotics on sebum Grades of acne and lesion counts-All three treatment groups showed a excretion.5 Our observation was unlikely to be a technical error gradual improvement in facial grade and total acne grade (figs 1 and 2). This since the patients started the treatment at different points in time overall improvement was reflected in a significant reduction in non-inflamed and other carried out did not unusual and inflamed lesions. The reduction was significant at all points compared studies concurrently give any with baseline. At the end of six months the improvement in lesion counts results. The effect of another antimicrobial agent, benzoyl perox- varied between 51% and 800/o (figs 3 and 4). There were only two significant ide, on the sebum excretion rate is controversial'"'7 but in our differences among the treatment groups after six months. The combined laboratory we find that it increases the rate by 22%.8 We do not BRITISH MEDICAL JOURNAL VOLUME 291 2 NOVEMBER 1985 1233 Br Med J (Clin Res Ed): first published as 10.1136/bmj.291.6504.1231 on 2 November 1985. Downloaded from Treatment phase Fol low up Treatment Follow up -I - 0 0- 10 10. 20 20-

.oc30- ***e 030- 0': 40-\ 0-: 40- 60 L. as cmu50-@ c) 0 u 60- 60 Q c: 70 c 70 a) M 80- Initial grade \** 10 Initial grade Oestrogen-cyproterone + tetracycline . * 2-1 ± 016 Oestrogen- cyproterone + tetracycline _-. 32 + 1 51 90- Oestrogen- cyproterone A A 17 015 90- Oestrogen- cyproterone A-A 25t 022 Tetracyctine U1-7±+0-27 Tetracycline *.... 2B8t+036 Iuu-n 100-

9 a I I 9 Fig 1 0 2 4 6 8 Fig 2 2 4 6 8 Months Months

Treatment 01 0' 10- 20 20 8 o 30 http://www.bmj.com/ .-ou U D az 40- 2w 40- A-. a 50- a) 0 cw 860- v- 60 a)cs

a) a on 2 October 2021 by guest. Protected copyright. Q *~C a c 70- 0 a) I 80- -%Z*** 80- Initial grade ***J Tritial grade Oestrogen-cyproterone + tetracycline *- 12-0 ± 2 22 Oestrogen -cyproterone + tetmcycline .-.38 79 ± 3 12 Oestrogen -cyproterone A- A14-55 ± 306 901 Oestrogen -cyproterone A -A2668 ± 262 Tetracycl ine * - a13 39 ±2 30 U - -.26 91 ± 4.01 100' 100- Tetracycline * 0 2M4n 6 6 M 4n 6 Fig 3 Months Fig 4 Months

FIG 1-Effect of three treatments on facial grade. Values indicate original acne score in arbitrary units. Compared with baseline: **p

FIG 2-Effect of three treatments on total grade. Values indicate original acne score in arbitrary units. Compared with baseline: *p<0o05; **p

FIG 3-Effect of three treatments on non-inflamed lesions. Compared with baseline: *p

FIG 4-Effect of three treatments on inflamed lesions. Compared with baseline: *p<005; **p<001, ***phormone treatment. There is indirect evidence that ductal cornification is under 30 hormonal control,20 and as oestrogen-cyproterone significantly * * w~~--1 reduces non-inflamed lesions it is likely that it must reduce ductal 0 2 4 6 8 cornification; whether this is a direct or indirect effect is uncertain. Months In our study, although antibiotics did significantly reduce the FIG 5-Effect of three treatments on sebum excretion rate (SER). Compared with counts of propionibacteria, the oestrogen-cyproterone preparation baseline: *p<0 05; **pbenzoyl peroxide. We do not know whether 0 2 4 the beneficial effect ofoestrogen-cyproterone would be improved by Months the concomitant application ofbenzoyl peroxide. Work is, however, FIG 6-Effect of three treatments on skin surface (log,o) counts of propionibacteria. currently underway to investigate this. Bars are SEM. We conclude that this preparation of oestrogen 50 jg and cyproterone acetate 2 mg is at least as effective as high dose (1 g daily) antibiotics. Nevertheless, the second month after the end of a on 2 October 2021 by guest. Protected copyright. think that this is a primary effect on sebaceous gland function and six month course the group of patients who had taken antibiotics suggest that benzoyl peroxide, by reducing ductal cornification," alone relapsed at a greater rate than those who had taken oestrogen- may affect sebum outflow.20 Since tetracycline also reduces non- cyproterone alone. Combining the two treatments failed to yield any inflamed lesions an effect on sebum outflow may explain the significant benefit over oestrogen-cyproterone alone but did increase in sebum excretion. These observations may be relevant in produce a greater improvement than tetracycline alone, and it is understanding the effect of the oestrogen-cyproterone preparation unlikely that such a complicated dual regimen will be ofany value in and of the combined treatment on sebum excretion rates. Both practical terms. The government recommendations for this prepara- regimens of oestrogen-cyproterone reduced the sebum excretion tion of oestrogen and cyproterone acetate are that it may be used in rate by 25% at six months, but at two months the reduction with the women with severe acne recalcitrant to long term antibiotics. Based combined treatment was significantly less than with the oestrogen- on this study, however, in which we have shown an unquestionable cyproterone preparation alone. The reduced sebosuppressive benefit of the drug similar to that oforal antibiotics, we suggest that effect of the combined treatment may be due to at least two it should be tried in the following groups ofpatients: (a) those whose mechanisms. If the antibiotic increases sebum excretion by im- oral contraception is jeopardised by antibiotic induced diarrhoea, proving ductal outflow then this will modify the sebosuppressive and (b) those whose acne has been improved by conventional treat- effect of oestrogen-cyproterone. An alternative explanation is a ment but shows no sign of further improvement after three to six possible drugs interaction. Oral contraceptive steroids undergo months. enterohepatic circulation,2' which may contribute to their normal We thank Dr K T Holland and Mrs K King for the microbiological data, function. Antibiotics may, by affecting the gut bacterial flora, Miss M Rehahn for statistical advice, and Dr P Longthorne and Dr K reduce the enterohepatic circulation and so reduce the amounts of Johnson, of Schering, for their support. BRITISH MEDICAL JOURNAL VOLUME 291 2 NOVEMBER 1985 1235 Br Med J (Clin Res Ed): first published as 10.1136/bmj.291.6504.1231 on 2 November 1985. Downloaded from References 13 Burton JL, Laschet U, Shuster S. Reduction of sebum excretion in man by the antiandrogen cyproterone acetate. BrJ Dermatol 1973;89:487-9. 14 Webster GF, McGinley KJ, Leyden JJ. Inhibition of lipase production in Propionibacterium I Hamilton JB. Male hormone stimulation is prerequisite and an incitant in common baldness. AmJ acnes by sub-minimal-inhibitory concentrations of tetracycline and erythromycin. Br J Anat 1942;71:451-80. Dermatol 1981;104:453-7. 2 Pochi PE, Strauss JS. Sebaceous gland suppression with ethinyl estradiol and diethylstilbestrol. 15 Jones DH, Cunliffe WJ, Loffler A. A comparative study of 13-cisretinoic acid and erythromycin in Arh Dermatol 1973;108:210-4. severe acne. In: Cunliffe WJ, Miller AJ, eds. Retinoid therapy. Proceedings of international 3 Neumann F, Elger W. Ihe effect of a new antiandrogenic steroid 6-chloro-17-hydroxy-1, 2- conference, London, May 1983. London: MTP Press, 1984: 293-301. methylenepregna-4, 6-diene-3, 20-dione acetate (cyprotereone acetate) on the sebaceous glands 16 Goldstein JA, Pochi PE. Failure of benzoyl peroxide to decrease sebaceous gland secretion in acne. of mice. J Invest Dermatol 1966;45:561-72. Dernatologica 1981;162:287-91. 4 Hammerstein J, Cupceancu B. Behandlung des Hirsutismus mit Cyproteroncetat. Dtsch Med 17 Fanta D, Muller MM. Effect of benzoyl peroxide on skin surface lipids. Dermatologica W'ocheschr 1969;94:829-39. 1979;158:55-9.. 5 Fanta D. Hormonal therapy of acne. Clinical and experimental principles. Vienna:Springer-Verlag, 18 Cunliffe WJ, Stainton C, Forster RA. Topical benzoyl peroxide increases the sebum excretion rate 1978. in patients with acne. BrJ Dermnatol 1978;109:577-9. 6 Mugglestone Cj, Rhodes EL. The treatment of acne with an antiandrogenloestrogen combina- 19 Cunliffe WJ, Dodman B, Ead R. Benzoyl peroxide in acne. Practitioner 1978;220:479-82. tion. Clin Exp Dermatol 1982;7:593-8. 20 Holmes RL, Williams M, Cunliffe WJ. Pilo-sebaceous duct obstruction and acne. BrJ Dernatol 7 Kirby WM, Roberts CE, Burdilk RE. Comparison of two new tetracyclines with tetracvcline and 1972;87:327-32. demethvlchlor tetracycline. In: Finland M, Savage EM, eds. Proceedings of first international 21 Orme MCE. The clinical pharmacology of oral contraceptive steroids. Br J7 Clin Pharmacol cmference on antimicrobtal agents and chemotherapy. New York: American Society for 1982;14:31-42. Microbiology, 1961:286-92. 22 Pye RJ, Meyrick G, Pye MJ, Burton JL. Effect on oral contraceptives on sebum excretion rate. Br 8 Burke BM, Cunliffe WJ. The assessment of acne vulgaris. The Leeds technique. Brj Dermatol MedJ 1977;ii:1581-2. 1984;111:83-92. 23 Cormane RH, van der Meeren HLM. Cyproterone acetate in the management of severe acne in 9 Strauss JS, Pochi PE. The quantitative gravimetric determination of sebum production. J Invest males. Arch Dermatol Res 1981;271:183-7. Dermatol 19%1;36:293-8. 24 Jones DH, King K, Miller AJ, Cunliffe WJ. A dose-response study of 13-cis-retinoic acid in acne 10 Cunliffe WJ, Shuster S. The rate of sebum excretion in man. BrJ Dermatl 1%9;81:697-704. vulgaris. BrJf Dermatol 1983;108:333-43. 11 Williamson P, Kligman AM. A new method for the quantitative investigation of cutaneous 25 Peck GL, Olsen TG, Yoder FW, et al. Prolonged remissions of cystic and conglobate acne with bacteria. J Invest Dermatol 1%5;45:498-503. 13-cis-retinoic acid. N Engl7 Med 1979;300:329-33. 12 Gould DJ, Cunliffe WJ. The long term treatment of acne vulgaris. Clin Exp Dermatol 1978;3: 249-52. (Accepted S 7ulv 1985,

Reassessment of inflammation of airways in chronic bronchitis

J BRENDAN M MULLEN, JOANNE L WRIGHT, BARRY R WIGGS, PETER D PARE, JAMES C HOGG

Abstract Introduction The term chronic bronchitis has been criticised because it is The Ciba Guest symposium of 1959 defined chronic bronchitis as associated with hypersecretion of mucus rather than bronchial "the condition ofsubjects with chronic or recurrent excessive mucus inflammation. This study was designed to establish the presence secretion in the bronchial tree."' As the diagnostic criterion is or absence of clinical chronic bronchitis and measure pulmonary clinical the relation of hypersecretion of mucus to a condition in the function in 45 patients about to undergo resection of the lung. airways in chronic bronchitis has been a source ofcontroversy. Reid The condition in the cartilaginous and small airways and the described an index of bronchial gland enlargement which she severity of the emphysema were then measured in the resected thought correlated with the amount of sputum produced.2 Sub- http://www.bmj.com/ specimen. The results from 20 patients who had clinical chronic sequent studies showed that this index was normally distributed bronchitis were compared with those in 25 patients who did not. without a clear separation between the bronchitic and non- The data show that patients with chronic bronchitis had greater bronchitic groups.36 Although chronic bronchitis implies inflam- inflammation (a) on mucosal surfaces (p<005) of all bronchi mation, the absence of documented inflammation of airways in larger than 2 mm luminal diameter and (b) around glands these cases has recently led to objection to the term bronchitis.' (p<0005) and gland ducts (p

University of British Columbia Pulmonary Research Laboratory, St Paul's Patients and methods Hospital, Vancouver, British Columbia, Canada V62 1Y6 J BRENDAN M MULLEN, MD, FRCP(C), pathologist We studied 45 patients admitted to this hospital for resection of an upper JOANNE L WRIGHT, MD, FRCP(C), pathologist lobe (n=40) or lung (n=5) for, in most cases, bronchial carcinoma. BARRY R WIGGS, Bsc, statistician Informed consent was obtained in all cases. A modified questionnaire by the PETER D PARE, MD, FRCP(C), physician British Medical Research Council was used to assess their exposure to JAMES C HOGG, MD, professor of pathology cigarettes, cough, and history of sputum. To examine the differences in condition of airways and function of lungs between patients with chronic Correspondence and requests for reprints to: Dr Hogg. bronchitis and those without the patients were divided into two groups: