Clinical question: How accurate is the initial diagnosis of cellulitis in adult patients compared to subsequent specialist consultation?
Dr. Lessing
Background: Diagnosis of cellulitis leads to an estimated 650,000 hospital admissions annually. As a clinical diagnosis—with no definitive confirmatory laboratory, microbiologic, or imaging result—cellulitis is frequently misdiagnosed, leading to diagnostic delay, as well as unnecessary cost, antibiotics, and hospitalization. The authors sought to determine how often the initial diagnosis of cellulitis might be incorrect when compared to subsequent specialist reviews as a reference standard.
Study design: Systematic review and meta-analysis
Setting: Consecutive adult patients seen by generalist physicians treated for cellulitis, with subsequent secondary evaluation by a specialist in either infectious disease or dermatology. Patients were seen in inpatient, outpatient, emergency department, and observation settings.
Synopsis: Using PRISMA guidelines, 5,572 citations were screened for eight studies that met all inclusion criteria. On meta-analysis of 887 patients initially diagnosed with cellulitis, 359 (40%) were felt by the subspecialist to have a non-cellulitis diagnosis. Excluding the single outpatient study, the “pseudo cellulitis” rate was 39%. Stasis dermatitis/venous stasis was the most common noninfectious, and abscess was the most common infectious, alternative diagnosis. To assess for risk of bias, the authors used the QUADAS-2 tool and found a high risk for bias in the reference standard domain, as consultants were not blind to the initial diagnosis of cellulitis. The authors rated as unclear the risk for timing bias as the effect on diagnostic accuracy of waiting, often for a day, and of being on antibiotics, until the consultant saw the patient, is unknown.
Bottom line: Similar to prior studies, this study found cellulitis is frequently misdiagnosed. The best way to reduce cellulitis misdiagnosis is unclear, but what is clear is even common conditions are commonly misdiagnosed. Revisit the need for antibiotics or continued hospitalization if an alternative diagnosis seems likely.
Citation: Cutler TS, Jannat-Khah DP, et al. Prevalence of misdiagnosis of cellulitis: A systematic review and meta-analysis. J Hosp Med. 2023;18(3):254-61.
Dr. Lessing is an associate professor in the division of hospital medicine at the University of Colorado Anshutz Medical Campus in Aurora, Colo.