Clinical Usefulness of Imaging and Blood Cultures in Cellulitis Evaluation

Cellulitis is a commonly occurring acute bacterial skin infection of the dermis and subcutaneous skin. In the United States, there are approximately 14.5 million cases of cellulitis annually, resulting in 650 000 hospital admissions.¹ The Infectious Disease Society of America (IDSA) released evidence-based guidelines that advise against imaging except in patients with cellulitis and febrile neutropenia, and against blood cultures except in patients who were highly immunocompromised, exhibiting systemic toxic effects, or who had sustained animal bites.² This study evaluates the clinical usefulness and cost of blood cultures and imaging in patients with presumed cellulitis.

This retrospective cohort study was deemed exempt by the Partners Human Research Committee institutional review board following review. Between October 2014 and February 2017, we enrolled 183 patients seen in the Massachusetts General Hospital Emergency Department with a presumed diagnosis of uncomplicated cellulitis who were subsequently admitted to inpatient medicine or an emergency department observation unit. We ascertained demographics, past medical and surgical history, and cellulitis risk factors by reviewing patient medical records or interviewing patients in person. After discharge, patient laboratory, microbiological, and imaging study results were recorded. Only imaging relevant to the suspected skin infections was recorded. The cost of diagnostic imaging and blood cultures was then estimated based on annual cellulitis hospitalization rates and with the assumption that other academic medical centers order testing at a similar rate as this study site. The costs of diagnostic studies were determined using HealthCare BlueBook,³ a free online tool calculating prices based on actual amounts paid on claims by health plans.

Of the 183 patients with uncomplicated cellulitis, 60 (32.8%) received blood cultures, and growth was detected in 1 patient (1.7%). According to IDSA guidelines, only 18 patients (10%) tested were appropriate candidates to undergo blood cultures based on their history and vital signs. Over two-thirds of the patients (n = 124 [67.8%]) received imaging. Nearly half the patients (n = 83 [45.4%]) underwent 1 form of imaging, while 30 (16.4%) underwent 2 forms, and the remaining patients underwent 3 (n = 9 [4.9%]) or 4 (n = 2 [1.1%]) imaging modalities (Table 1); 84 patients (46.0%) underwent ultrasonography, while 53 (29.0%) received a radiograph, 29 (16.0%) received computed tomography, and 11 (6.0%) received magnetic resonance imaging (Table 2). None of the patients imaged were febrile or neutropenic, indicating that all imaging performed was contraindicated by IDSA guidelines. Imaging results changed diagnosis and management in 8 patients (6.5%), who had alternative diagnoses of hematoma (n = 1), abscess (n = 5), and osteomyelitis (n = 2). While 4 instances (36.4%) of magnetic resonance imaging led to changes in patient management, only a small fraction of computed tomography (n = 2 patients [6.9%]), radiographs (n = 2 patients [1.5%]), and ultrasonography (n = 1 patient [1.0%]) changed diagnosis and treatment.

Comparing patients who did and did not receive imaging, we found that those who underwent imaging had a significantly higher percentage of chronic lymphedema (n = 36 [29.0%] vs n = 9 [15.3%]; P = .04) and higher mean levels of serum glucose (133.1 vs 116.3; P = .03). There was no significant difference in other cellulitis risk factors or laboratory values between the 2 groups.

Based on estimates of imaging and blood culture costs, the yearly cellulitis hospitalization rate, and the rate at which imaging and blood cultures were ordered within this patient cohort, the cost of these largely clinically useless diagnostic studies is approximately $226.9 million dollars annually.

The majority of patients with cellulitis received 1 or more imaging modalities, despite IDSA guidelines that recommend against imaging except in patients who are also experiencing febrile neutropenia.² Results of blood cultures and imaging seldom altered diagnosis or treatment. As previous studies have suggested,^4-6 radiologic imaging and blood cultures have low clinical usefulness for evaluation and treatment of cellulitis. In addition, they portend significant cost to the health care system. Imaging and blood cultures should be pursued only in patients who are severely immunocompromised or experiencing systemic toxic effects.

Corresponding Author: Daniela Kroshinsky, MD, MPH, Department of Dermatology, Massachusetts General Hospital, 50 Staniford St, Ste 200, Boston, MA 02114 (dkroshinsky@mgh.harvard.edu).

Accepted for Publication: January 26, 2018.

Published Online: April 2, 2018. doi:10.1001/jamainternmed.2018.0625

Author Contributions: Dr Kroshinsky and Ms Ko had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Ko, St John, Strazzula, Kroshinsky

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Ko.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Ko.

Administrative, technical, or material support: Garza-Mayers, St. John, Vedak, Kroshinsky.

Study supervision: Kroshinsky.

Conflict of Interest Disclosures: None reported.

Funding/Support: This study was supported in part by The Dermatology Foundation.

Role of the Funder/Sponsor: The funder/sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contributions: We are indebted to the Dermatology Foundation; Sowmya R. Rao, PhD, for her paid assistance with statistical analysis; Blair Alden Parry, BA, CRCC, and the Massachusetts General Hospital Emergency Department Research Team for their assistance in patient screening and enrollment; as well as Evan Stein, MD, for his assistance in manuscript editing. Unless otherwise specified, no one was compensated for their contributions.