Handheld infrared thermometer to evaluate cellulitis: the HI-TEC study. (December 2021)

• Record Type: Journal Article
• Authors: Demir, Koray K.; McDonald, Emily G.; de L'Étoile-Morel, Samuel; Schweitzer, Lorne; Butler-Laporte, Guillaume; Cheng, Matthew P.; Lee, Todd C.
• Is Part Of: Clinical microbiology and infection. Volume 27:Number 12(2021)
• Page Start: 1814
• Page End: 1819
• Publication Date: 2021-12
• Subjects: Cellulitis -- Infrared thermometer -- Pseudocellulitis -- Diagnosis -- Erysipelas; Medical microbiology -- Periodicals; Diagnostic microbiology -- Periodicals; Communicable diseases -- Periodicals; Infection -- Periodicals; 616.01
• DOI: 10.1016/j.cmi.2021.03.025
• Languages: English
• ISSNs: 1198-743X

Abstract

Abstract: Objectives: Differentiating cellulitis from pseudocellulitis is challenging, and misdiagnosis leads to unnecessary antimicrobial use and increased healthcare expenditure. Clinical diagnosis remains the criterion standard and may involve expert consultation. Our objective was to evaluate the usefulness of a handheld infrared thermometer to improve diagnostic certainty in cases of suspected cellulitis. Methods: We conducted a cross-sectional study from August 2018 to January 2020 at a tertiary-care hospital in Montreal, Canada. We enrolled adult patients with suspected limb cellulitis. Using the infrared thermometer, we compared the average temperature of the affected area with that of the contralateral limb, and we used Youden's method to determine the optimal temperature difference which best differentiated cellulitis from pseudocellulitis as determined by an independent and blinded infectious diseases specialist. We used bootstrapping to estimate 95% confidence intervals for the sensitivity, specificity, and area under the receiver operating curve. Results: Of 65 patients screened for enrolment, 52 patients were recruited (median age: 64 years, IQR 52–76); 39 of these were diagnosed with cellulitis and 13 were not. The mean temperature difference between affected and unaffected limbs was 2.6°C (95%CI 2.1–3.1°C) for patients with cellulitis and 0.4°C (95%CI –1.2°C to 2.1°C) for patients without (p < 0.001). An average temperature difference between limbs of 0.8°C or more was 95% sensitive (95%CI 74–100%) and 69% specific (95%CI 44–95%) for the diagnosis of cellulitis (c-statistic 0.82). Conclusions: In this proof-of-concept single-centre study, a handheld infrared thermometer was a useful aid to differentiate cellulitis from pseudocellulitis.