Variability of antimicrobial prescribing in patients with acute cellulitis

Clinical guidelines concerning treatment of infection are incorporated into prescribing formularies and antimicrobial stewardship policies. The extent to which these influence prescribing is uncertain. In this study, we sought to examine antimicrobial prescribing patterns in patients with cellulitis.

Consecutive adults admitted to hospital due to acute cellulitis between 2008 and 2010 were studied. Data collected were clinical and laboratory markers of sepsis, antimicrobial agent, route of administration, number of i.v. dosages, duration of antimicrobial treatment, and hospital length of stay. Three groups were defined by prescribing that was (i) identical to formulary, (ii) modified appropriately due to microbiological data or prior drug allergy, and (iii) nonformulary prescribing. Comparisons were made between groups using Mann-Whitney tests.

There were 306 patients: 167 men (54.6%), median age 66 (range 18-100) years. Prescribing was consistent with formulary recommendations in 253 (82.7%), modified appropriately in 24 (7.8%), and nonformulary in 29 (9.5%). Median [interquartile range (IQR)] duration of hospital stay was 5 (3-8), 7 (5-9, P = 0.026), and 7 (5-14, P = 0.0006) days, and overall duration of antimicrobial therapy was 12 (9-16), 13 (8-15), and 15 (12-19, P = 0.0479) days, respectively. No differences were observed in clinical or laboratory markers of sepsis.

Prescribing patterns accorded with prevailing guidelines in the majority of patients. Nonetheless, there was nonformulary prescribing in 10% of patients, and this could not be explained by clinical or laboratory measures of disease severity. Further work is needed to explore the factors that contribute to nonformulary prescribing in this group of patients.