Penicillin to Prevent Recurrent Cellulitis?

Abstract & Commentary

By Jennifer A. Best, MD, FACP, FHM, Assistant Professor, University of Washington School of Medicine, Seattle, WA. Dr. Best reports no financial relationships in this field of study.

Synopsis: Following an index episode of recurrent cellulitis, penicillin decreased the rate of recurrent leg cellulitis during a 12-month prophylactic period as compared with placebo, but this effect was not observed to be sustained.

Source: Thomas KS, Crook, AM, Nunn AJ, Foster KA, Mason JM, et al. Penicillin to prevent recurrent leg cellulitis. New Engl J Med 2013;368:1695-703.

Cellulitis, or infection of the skin and soft tissue, is a common admission diagnosis in hospitalized patients. Acute episodes of cellulitis are managed with courses of intravenous and/or oral antibiotics to treat the most common causative bacterial species, Staphylococcus and Streptococcus. However, as many patients have underlying conditions predisposing to recurrence, such as vascular disease, immunosuppression or skin disruption, it would be beneficial if effective prophylactic therapy were available. Though consensus guidelines recommend prophylaxis for recurrent cellulitis, few high-quality trials are available to support this recommendation. Thomas and colleagues investigated the efficacy of prophylactic low-dose penicillin for the prevention of recurrent leg cellulitis in the PATCH 1 (Prophylactic Antibiotics for Treatment of Cellulitis at Home) trial.

In this randomized, double-blind, placebo controlled trial, 274 patients with recurrent leg cellulitis were identified during hospitalization or through advertising in the United Kingdom and Ireland. All subjects were recruited within 24 weeks of their most recent recurrence. Recurrent cellulitis was defined as 2 episodes within the past 3 years. Study inclusion required substantiation of the diagnosis by dermatologist exam or record review plus patient interview for the following: warmth, tenderness, acute pain, unilateral erythema (or bilateral erythema, if symptoms temporally corresponded with the most erythematous leg) or unilateral edema. All patients with an uncertain diagnosis were excluded, as were patients using antibiotics for cellulitis prophylaxis within 6 months of recruitment, those with penicillin allergy, and those with precedent leg ulcer, surgery or trauma.

Study patients were randomized to receive penicillin 250 mg twice daily for 12 months or placebo following acute treatment for an index episode of cellulitis. Patients maintained pill counts and an adverse event and health care utilization log and received phone calls from study coordinators every 3 months during prophylaxis (0-12 months) and every 6 months during follow up (to 36 months).

The primary study outcome was time from randomization to next recurrence of leg cellulitis. Notable secondary outcomes included comparisons between the prophylaxis and follow-up phase (total recurrences and development of new edema or ulcers), adverse events (including drug reactions), healthcare utilization and cost effectiveness utilizing national reference costs and the effects of known cellulitis risk factors. Data were analyzed based on intention to treat.

The penicillin and placebo groups were similar at baseline. All patients were followed for at last 18 months. 78% of study patients reported at least 75% compliance with medications; compliance in each group was comparable. During prophylaxis, the primary outcome — median time to recurrence — was longer in the penicillin group (626 days) as compared with placebo (532 days) and the rate of recurrence was significantly lower with penicillin (22% v. 37% for placebo; HR 0.55; p < 0.01). The number needed to treat to prevent 1 case of recurrent leg cellulitis was 5. However, during the follow-up period, there were no significant differences between the two groups in the following secondary outcomes — rate of recurrence and repeat episodes, development of edema or ulceration, adverse events, healthcare utilization or cost. Predictors of prophylaxis failure included high BMI (≥33), ≥3 episodes of cellulitis and edema.

In summary, these data confirm the high rate of recurrent leg cellulitis among patients with previous episodes — among all study participants, 53% experienced at least one recurrence during the study period. Though prophylactic low-dose penicillin was effective over 1 year in increasing the time to recurrence and overall recurrence rate, this effect was not sustained to 3 years.

This study does build on a previous study of an ineffective 6-month prophylactic period, but does not clarify whether a prophylactic period longer than 12 months would extend this effect. In addition, this study does not evaluate penicillin’s effect on prophylaxis for cellulitis at other sites or whether other non-penicillin agents might be more effective.

Finally, given that some patient groups were less likely to benefit from penicillin, it is likely that one prophylactic approach may not fit all. Further research to clarify these questions will be required before prophylactic penicillin therapy is offered to eligible hospitalized patients with acute episodes of recurrent cellulitis.