Original article

Annales de dermatologie et de v?n?r?ologie 2001; 128: 224-228
? Masson, Paris, 2001

Acne flares and deteriorations during oral isotretinoin therapy

M.?Chivot?<sup>(1)

(1)</sup>Service du Pr Morel, H?pital Saint-Louis, 1, avenue Claude Vellefaux, 75475 Paris Cedex 10.

Tir?s ? part : M.?CHIVOT , at the above address.

SUMMARY

Background

Isotretinoin treatment for acne can lead to inflammatory flare-ups or an aggravation, occasionally leading to acne fulminans. The purpose of this work was to examine our cases and to propose a classification system for management.

Patients and methods

Over a 3-year period, we selected patients referred to our dermatology unit for paradoxical aggravation of acne under isotretinoin treatment. We recorded clinical data, drug prescriptions and the course of the flare-up.

Results

Over 3 years (1995-1998) we observed 32 cases of acne flare-up in patients taking isotretinoin, 6 women and 26 men.

Discussion

Four types of aggravation could be identified depending on their date of onset, the skin signs, and the presence or not of general signs. Systemic corticosteroids are generally required, together with a lower daily dose of isotretinoin and local care (excision of open and closed comedons). Factors predictive of aggravation are young age, male sex and sebaceous retention.

Conclusion

Acne fuminans is exceptional in patients taking isotretinoin compared with the number of patients treated. Clinicians should nevertheless be aware of the risk in order to make the diagnosis and provide appropriate care.

R?SUM?

Pouss?es inflammatoires et aggravations d'acn? sous isotr?tino?ne orale.

M.?Chivot

Introduction

Lors du traitement de l'acn? par l'isotr?tino?ne (Roaccutane?), il a ?t? observ? des pouss?es inflammatoires, des aggravations, voire des acn?s fulminans. Le but de ce travail ?tait d'exposer nos observations, et de proposer une classification et une conduite ? tenir.

Malades et m?thodes

Pendant une p?riode de 3 ans, nous avons s?lectionn? des malades adress?s par des confr?res pour une «?aggravation paradoxale?» de l'acn? sous isotr?tino?ne. Nous avons recueilli les param?tres concernant l'?tat des malades, le traitement de la pouss?e et leur devenir.

R?sultats

En 3 ans (1995-1998), nous avons observ? 32 cas de pouss?es inflammatoires sous isotr?tino?ne, 6 femmes et 26 hommes.

Commentaires

Nous avons pu isoler 4 types d'aggravations selon leur date d'apparition, leur symptomatologie cutan?e et l'existence ou non de signes g?n?raux. La prise en charge repose essentiellement sur la corticoth?rapie par voie g?n?rale, sur la diminution de la dose quotidienne d'isotr?tino?ne et sur les soins locaux (ex?r?se des com?dons ouverts et ferm?s). Les facteurs pr?dictifs d'une aggravation sont les hommes jeunes avec une r?tention s?bac?e.

Conclusion

Si les acn?s fulminans sous isotr?tino?ne sont exceptionnelles par rapport au nombre de malades trait?s, les aggravations sont suffisamment nombreuses pour qu'elles soient reconnues et trait?es.

---

Isotretinoin (Roaccutane?) therapy for acne has been associated with inflammation, exacerbation, and acne fulminans [1, 2, 3, 4, 5, 6, 7]. No regrouping or classification of these entities exists to our knowledge. We report our observations and propose a classification scheme and therapeutic measures.

Patients and methods

Over a three-year period (1995-1998) and within the framework of consultations for acne, we received 218 patients referred by dermatologists confronted with therapeutic challenges. Twenty-two patients developed inflammation with isotretinoin and were referred for paradoxical aggravation of acne. Two male patients were lost from follow-up, and analysis included 30 patients. For each patient data were recorded for age, gender, history of sebaceous gland obstruction before treatment, initial localization before the flare, initial dose of isotretinoin, delay before aggravation, type of inflammatory lesion, results of bacteriological analysis, treatment (suspension, renewal, or modification of the dose of isotretinoin, systemic steroids), and evolution.

Results

The study included twenty-four 15 to 23 year-old men (mean age 18) and six 16 to 45 year-old women (mean age 23.5). Nineteen patients had papules, pustules, and painful subcutaneous nodules. In the other 11 these lesions were associated with ulcerated necrotic elements or purulent discharge. Arthralgia, myalgia, fever of 38?C or higher, or alteration in general health developed in three patients, suggesting a diagnosis of acne fulminans. Initially acne was limited to the face in 12 patients and to the chest and back in 18 others. The mean initial dose of isotretinoin was 0.6 mg/kg/day, and in 12 cases the dose was lower than 0.5?mg/kg/day.

In thirteen patients who were not receiving antibiotic therapy, material from papules or pustules was obtained for bacterial culture. Common saprophytic flora grew from the cultures of 11 patients, as well as Staphylococcus lugdunensis in one case and Klebsiella oxytoca in one. Seventeen patients did not undergo bacteriological exploration. Isotretinoin was stopped at the onset of the flare or shortly after in 26 patients. Four patients improved when the dose was reduced and later increased to the initial dose in conjunction with systemic steroids. Twenty-two of 30 patients were treated with oral prednisone at 0.5 mg/kg/day until complete disappearance of the inflammation, with durations of 10 days to 8 months. The prednisone was discontinued after an appropriate taper. Improvement was remarkable in all cases. In 2 patients the dose required a temporary increase owing to a rapid taper. Prior to their referral to our clinic, 19 patients had received antibiotic therapy with cyclines, pristinamycin, or roxithromycin. The patient with Klebsiella oxytoca had been treated with ofloxacin, and the patient with Staphylococcus lugdunensis received topical clindamycin (Dalacine T topic?).

Three of 6 women reported regular menstrual cycles since puberty and had not had hormonal therapies. Three others had irregular menses and two had hormone levels checked six months after the end of treatment (testosterone, DHEAS, and 17-OHP, assayed in the first five days of the cycle), which were normal. One patient declined discontinuation of her oral contraceptive and did not undergo hormone level analysis.

The outcome was good in 20 patients. Among these, complete long-term resolution was achieved with continuation (1 case) or renewal (13 cases) of isotretinoin, with cyclines (2 cases), spontaneously after the flare ceased (3 cases), or with Androcur? (1 case). Open and closed comedones were extracted in 8 patients. The other 10 patients continued to suffer from moderate inflammatory acne. Isotretinoin was not re-introduced in these patients because of moderate relapse of inflammation after re-introduction of the product (6 cases) or because the patients refused (4 cases). Among the three patients with acne fulminans one refused and two accepted renewal of isotretinoin; of these two, one tolerated the second course well and the other stopped because of moderate relapse of inflammation.

Classification

We classified these 30 patients into 3 groups depending on the types of lesion present: inflammatory nodules, ulcerated or necrotic elements, and fluctuant or pyogenic lesions. We also noted whether the onset of the flare was abrupt or progressive and whether it was accompanied by systemic symptoms.

Acute inflammatory flares

Acute inflammatory flares after the first month (19 patients, among which were 6 women) were marked by the absence of improvement and by progressive, occasionally painful, eruption of papules, pustules, and nodules at sites of pre-existing acne. Inflammation was aggravated by continuation of treatment or increased dosage. Symptomatology was not consistently intense fig.?1 and 2.

Eruptive inflammatory flares

Eruptive inflammatory flares (8 patients) occurred abruptly during the second or third month of treatment with a clinical form similar to acne fulminans fig.?3and 4 but without systemic signs, articular symptoms, or biological signs of inflammation. Painful, deep-seated, and inflammatory purulent nodules were associated with ulcerated, necrotic elements and fluctuant collections of pus. The eruption occurred in areas previously unaffected by acne.

Acne fulminans

Acne fulminans developed in three patients, who presented with abrupt onset of skin lesions of the type described for the preceding group, accompanied by arthralgia, myalgia, malaise, and a fever of 38?C or higher. Leukocytosis, elevated erythrocyte sedimentation rate, or increased levels of C-reactive protein were noted. Antibiotic therapy was ineffective fig.?5 and 6.

Discussion

Acne fulminans provoked by isotretinoin has been the subject of several publications [1, 2, 3, 4, 5, 6, 7]. The literature does not detail the different entities covered by the term "aggravation" of acne by isotretinoin [2, 8]. It is of debatable value to describe these forms as a single entity. The only factor common to the eruptions is the paradoxical effect of isotretinoin. The most complete series is that of Clark and Cunliffe [2]. Over a period of 5 years these authors studied 980 patients treated with isotretinoin. They reported a 6?p.?100 rate of exacerbation, among which 3?p.?100 were severe, but they did not specify whether acne fulminans was diagnosed. These severe flares appear rare, but the majority of cases may not be reported.

The responsibility of isotretinoin in the extent of these flares has not been formally established, but indirect arguments strongly favor the hypothesis. These are based on the observation that improvement occurs universally upon suspension of isotretinoin or reduction of the dose, and conversely that increases in dosage can aggravate inflammatory symptoms. These aggravations are more common in young males.

Pre-existing sebaceous gland occlusion has been emphasized as a predisposing factor [2, 8, 9]. This prompts re-evaluation of open and closed comedones. We noted that in 7 of our patients these were not a consistent factor in exacerbations but rather required distinct attention. In addition, the role of the initial dose is controversial. Some authors note aggravation at doses of 1 mg/kg/day [2, 10], but cases of acne fulminans have been reported with doses lower than 0.5 mg/kg/day [1, 11]. Twelve of our patients had initially received doses lower than 0.5 mg/kg/day. However, an increase in dose during an inflammatory flare led to further deterioration in most cases. The initial site of acne does not appear to be a predictor of aggravation, as in our patients acne flared on both the trunk and face.

Classification

Based on the literature, we feel it is possible to outline four types of exacerbation by adding to the three types observed in our series the inflammatory flare of the first month. This classification should permit a pragmatic approach in the diagnosis and treatment of these flares.

Flares of the first month

The flares of the first month (type 1) are manifested by inflammatory aggravation of pre-existing lesions and generally occur during the first two weeks after initiation of treatment. They were rapidly identified when the product was launched in 1986 and appear in the summary of product characteristics (SPC) under the heading "adverse events". The patient must be told of the transient character of this aggravation, which generally resolves spontaneously before the patient consults during the first month. Sometimes it is particularly severe and may distress the patient.

Subintrant inflammatory flares

Subintrant inflammatory flares after the first month (type 2) possibly correspond to an absence of response. The clinical picture is an eruption of papules and nodules continuously recurring in pre-existing acne sites. The patient does not improve and the inflammation increases during treatment or with an increase in dose. The lesions can be painful, but symptomatology is not always severe fig.?1 and 2. Inflammation may develop progressively. These flares can be distinguished from those of the first month, which resolve with continued treatment.

Fulminant inflammatory flares

The fulminant inflammatory flares (type 3) rarely occur at the start of treatment. During the second or third month of treatment or later the patient suddenly develops an eruption similar to acne fulminans but without systemic signs fig.?3 and 4. The findings and symptoms are cutaneous with no articular pain or biological signs of inflammation. The lesions flare with deep and painful purulent nodules. Ulcerated and necrotic elements co-exist with fluctuant purulent pockets. The eruption may appear in areas otherwise untouched by acne.

Acne fulminans

Acne fulminans (type 4) is a severe form of ulcerating acne that develops abruptly. The cutaneous lesions resemble those of type 3 but are associated with arthralgia, myalgia, fever of 38?C or higher, and occasional malaise. Leukocytosis develops and is accompanied by elevation of the erythrocyte sedimentation rate or C-reactive protein as well as absence of response to antibiotic therapy fig.?5 and 6.

The presumed pathogenesis of acne would be a type III immune reaction according to the classification of Gell and Coombs, with involvement of an Arthus phenomenon. Free Propionibacterium acnes antigens, present in excess of circulating immune complexes, can lead to a generalized inflammatory reaction with fever and articular involvement. This hypothesis is supported by the clear response of this type of inflammation to steroids. Isotretinoin might have a paradoxical effect, perhaps due to its effect on pilosebaceous units with loss of epithelial integrity through desmosome release. This would allow direct contact between large amounts of P. acnes antigens and the immune system, with consequent activation of an inflammatory reaction [4]. Certain authors suggest that a genetic predisposition exists, since cases of acne fulminans have been described in patients' siblings and in monozygotic twins [12]. The existence of a nosological relationship between the different types of flares is debatable, since the inflammatory process in acne fulminans does not seem to occur in other types of flares. However, a point common to all these forms is an excessive response to treatment in the pilosebaceous epithelium.

Treatment of flares

Treatment of flares is controversial. The following have been proposed: continuation of isotretinoin alone [1, 12] or associated with systemic steroid therapy [7, 13]; reduction of the dose of isotretinoin in association with systemic steroid therapy [14, 15]; and suspension of isotretinoin [16, 17] and institution of systemic steroid therapy [4, 6, 10, 14, 17, 18] at doses of 0.5 to 1 mg/kg/day.

In 26 of our 30 patients we discontinued isotretinoin at the onset of the flare or shortly thereafter and initiated oral steroid therapy (0.5 mg/kg/day). The steroid course was maintained until the flare resolved, with a range of 10 days to several months. In our experience more severe inflammation requires longer therapy. Suspension of steroids must be progressive and slow, as rapid reduction may lead to increased lesions. By tapering the dose slowly, we were able to stop corticosteroid therapy in all our patients. The currently favored corticosteroid is prednisone (Cortancyl?), which has been established to be the anti-inflammatory best tolerated by the gastric mucosa [19]. We have no experience with the non-steroidal anti-inflammatory drugs used in recent reports [8, 13].

In acne fulminans (type 4), it may be useful to prescribe level II analgesics (codeine, dextropropoxyphene) for painful skin lesions. Treatment of Type 3 flares is identical to that of acne fulminans, but the course of steroid therapy is generally shorter. In type 3 flares, we strongly recommend that the dose of isotretinoin not be increased since it will provoke exacerbation of the lesions. Rather, we recommend that the dose be lowered and that a short steroid course (8 to 10 days) be initiated at a dose of 0.5 mg/kg/day. This halts the inflammatory process, and treatment with isotretinoin can be continued at the reduced dose or at the initial dose, depending on the response. Finally, in type 1 flares no change in treatment is necessary.

In our experience it is helpful to excise microcysts and comedones and to incise and drain sinuses. In addition, the presence of unusual bacteria is due to secondary colonization of pre-existing inflammation, as their eradication does not necessarily coincide with clinical improvement. Treatment of these organisms with appropriate antibiotics, as revealed by culture and sensitivity testing, is justified. However, prescription of an antibiotic without microbiological testing is expensive and ineffective; 19 of our patients had received antibiotics without relief before being referred to us.

Outcome of the patients

The inflammatory flares, even fulminant types, eventually resolve, but this does not reflect the "cure" of acne. Generally, isotretinoin is prescribed to patients for whom classical treatments have failed. At times a patient's acne improves after a sufficient total dose of isotretinoin but is not resolved, and one is confronted with the difficult problem of treating these cases later on. It is therefore preferable to administer very low doses (10 to 20 mg/day) of isotretinoin in conjunction with corticosteroids, rather than to discontinue isotretinoin therapy. Once the flare is under control, the corticosteroids can be tapered and the isotretinoin dose progressively increased. In this way one avoids the patients' refusal to renew the product, believed to be responsible for the flare. In our series of patients in whom the acne had not been "cured" and who refused to renew treatment with isotretinoin, acne remained inflammatory and no therapeutic solution was possible. The physical and emotional implications can be substantial.

To cite the present paper, use exclusively following reference: Chivot M. Pouss?es inflammatoires et aggravations d'acn? sous isotr?tino?ne orale (full text in english on e2med.com/ad). Ann Dermatol Venereol 2001; 128: 224-8.

Acknowledgements: We gratefully acknowledge the assistance of Doctor Antoine Petit with literature research and Doctor Michel Canesi for the development of the computer program.

Figure 1.
Type 2: A 22 year-old woman, after 6 months of 0.5 mg/kg/day of Roaccutane?.

Figure 2.
Type 2: Revelation of micro cysts, probably responsible for the lack of response to treatment.

Figure 3.
Type 3: A 19 year-old male after 8 months of treatment with 0.7 mg/kg/day of Roaccutane?.

Figure 4.
Type 3: A 19 year-old male after 8 months of treatment with 0.7 mg/kg/day of Roaccutane?.

Figure 5.
Type 4: Acne fulminans in a young 17 year-old male after 3 months of treatment with 1 mg/kg/day of Roaccutane?.

Figure 6.
Type 4: Acne fulminans in a young 17 year-old male after 3 months of treatment with 1 mg/kg/day of Roaccutane?.

REFERENCES

[1] Choi EH and Bang D. Acne fulminans and 13-cis-retinoic acid. J Dermatol 1992;19:378-83.

[2] Clark SM, Cunliffe WJ. Acne flare and isotretinoin. Incidence and treatment. Br J Dermatol 1995;133(Suppl?45): 26.

[3] D'incan M, Reverte M, Loriette Y, Ferrier-Lebouedec MC, Souteyrand P. Acn? fulminans avec ost?olyse aseptique survenue sous isotr?tino?ne. Rev Int Pediatr 1994;252:25-7.

[4] Faverge B, Attou D, Boncin M, Gratecos L. Acn? fulminans de l'adolescent et isotr?tino?ne. Arch Pediatr 1996;3:188-9.

[5] Jenkinson HA. Isotretinoin-induced acne fulminans. Br J Dermatol 1992;127(Suppl 40): 62-3.

[6] Joly P, Prost C, Gaudemar M, Revuz J. Acn? fulminans d?clench?e par la prise d'isotr?tino?ne. Ann Dermatol Venereol 1991;118:369-72.

[7] Mantoux F, Benet L, Gorlier C, Ortonne JP, Lacour JP. Acne fulminans triggered by isotretinoin. Ann Dermatol Venereol 1998;125(Suppl 1): 1554-5.

[8] Lehucher-Ceyrac D, Chaspoux C, Sulimovic L, Morel P, Lefrancq H. Aggravation de l'acn? sous isotr?tino?ne. Ann Dermatol Venereol 1998;125:496-9.

[9] Bottomley WW, Cunliffe WJ. Severe flares of acne following isotretinoin. Acta Derm Venereol 1993;73:74.

[10] Ogylvie ALJ, Ghane Y, Von Dendriesch P. Akne fulminans und generalisierts pustul?ses exanthem w?hrend der Behandlung f?r einer akne vulgaris mit isotretinoin. Z Hautkrank 1995;70:637-40.

[11] Katz RA, Jorgensen H, Nigra TP. Flare of cystic acne from oral isotretinoin. J Am Acad Dermatol 1983;8:132-3.

[12] Blanc D, Zultak M, Wendling D, Lonchampt F. Eruptive pyogenic granuloma and acne fulminans in two siblings treated with isotretinoin. Dermatologica 1988;177:16-8.

[13] Kellet JK, Beck MH, Chalmers RC. Erythema nodosum and circulating immune complex in acne fulminans after treatment with isotretinoin. Br Med J 1985;290:820.

[14] Konrad E, Rahmel B, Rassner G. Acne fulminans bei isotretinoin-behandlung. Akt Dermatol 1991;17:182-3.

[15] Traupe H, Von Muhlendahl E, Br?mswig J, Happle R. Acne of the fulminans type following testosterone therapy in three excessively tall boys. Arch Dermatol 1988;124:414-7.

[16] Lasserre J, Aquilina C. Aggravation paradoxale de l'acn? par l'isotr?tino?ne. Nouv Dermatol 1995;3:171-2.

[17] Sigal-Nahum M. Diagnostic. Abstract Dermato 1992;17-8.

[18] Karvonen SL Acne fulminans: report of clinical findings and treatment of twenty-four patients. J Am Acad Dermatol 1993;28:272-9.

[19] Brion N, Guillevin L, Le Parc JM. La corticoth?rapie en pratique. Ed Masson. Paris, 1998; p?92.

Rechercher dans

? Masson, Paris, 2001

?Résumé
?Texte intégral PDF
?Retour au Sommaire

Texte intégral précédent ? ? ? Texte intégral suivant