Epidemiology and risk factors for recurrent severe lower limb cellulitis: a longitudinal cohort study

Abstract

Objective: To describe the epidemiology and risk factors for recurrence of severe lower leg cellulitis (LLC).

Methods: A longitudinal cohort study using state-wide data of adults presenting with recurrent LLC following a primary episode occurring between January 2002 and December 2013. The incidence of recurrent LLC was estimated using the cumulative incidence function, accounting for mortality. Independent risk factors for recurrence were identified using Fine-Gray sub-distribution and Cox proportional hazards models.

Results: Of 36 276 patients presenting with their first episode of LLC, 4598 had at least one recurrence during the follow-up period. The cumulative incidence of first, second, and third recurrences at 12 months since previous infection was 6.3% (95% CI 6.0-6.5), 17.2% (95% CI 16.1-18.4), and 29.4% (95% CI 26.8-31.9), respectively, and at 5 years was 13.9% (95% CI 13.5-14.3), 35.9% (95% CI 34.2-37.5), and 52.9% (95% CI 49.5-56.2), respectively. The length of hospitalization increased from 3 days for the primary episode to 4 and 5 days for first and all subsequent recurrences, respectively. Independent risk factors for recurrent LLC included age, ethnicity (Indigenous Australian), local factors relating to lower leg pathology, conditions that commonly result in peripheral oedema, and systemic conditions that may be associated with increased leg size.

Conclusions: LLC recurrences are frequent, and each episode increases the likelihood of subsequent recurrence and length of hospitalization. These data provide context and scope to develop workable and effective strategies to prevent secondary episodes for all cases of primary LLC.

Keywords: Case-control; Cellulitis; Cellulitis recurrence; Epidemiology; Infectious diseases.