Evaluating Risk Factors for Short Versus Long Hospital Stay for Cellulitis to Guide the Development of a Long-acting Lipoglycopeptide Use Decision Algorithm

Abstract

Background:

Cellulitis is a common cause of hospitalization imposing significant burden on healthcare systems. Treatment disposition is dependent on extent and severity of infection with a range of possibilities from ambulatory to inpatient management. Limited evidence is available, aside from expert opinion, to guide the use of long-acting lipoglycopeptides (LaLGPs) to reduce hospitalizations.

Objective:

Identify patient-specific risk factors associated with short-stay hospitalizations (<72 hours) for cellulitis to guide development of an emergency department (ED) LaLGP decision algorithm.

Methods:

Adult patients with cellulitis treated with antibiotic monotherapy were screened. Three cohorts were identified based on length of hospitalization (ED discharge, <72 hours, ≥72 hours). The primary outcome was to identify risk factors associated with short-stay hospitalization. Patients were excluded if they required surgical intervention in the operating room, received polymicrobial therapy, required intensive care unit admission, received cellulitis treatment in the past 30 days, or were treated for multiple infections.

Results:

A total of 161 patients were analyzed (ED discharge, n = 75; Short-stay, n = 46; and Long-stay, n = 40). Need for incision and drainage (odds ratio [OR] 1.8; 95% confidence interval [CI]: 0.76-2.36), advancing age (OR 1.02; 95% CI 1.00-1.05), or presenting with fever (OR 32.7; 95% CI 5.89-616.5) were associated with short-stay hospitalization compared to ED discharge. Diabetes (OR 3.02, 95% CI 1.31-7.11), presence of purulence (OR 3.86, 95% 1.09-13.79), and history of methicillin-resistant Staphylococcus aureus (MRSA) (OR 4.31, 95% CI 1.08-17.96) were independently associated with long-stay hospitalizations.

Conclusion and Relevance:

Fever was the only factor independently associated with short-stay hospitalization, compared to patients discharged directly from the ED. Diabetes, purulence, and history of MRSA were associated with longer hospitalizations. Inclusion of these factors in a decision algorithm may help guide use of LaLGPs for cellulitis in the ED, potentially reducing cellulitis hospitalization rates.