Accuracy of Infrared Ear Thermometry in Children: A Meta-Analysis and Systematic Review

Background. Accurate determination of temperature is important, especially in the diagnosis and treatment of febrile illnesses in the pediatric population. False negative measurement can lead to miss febrile and false positive measurement can cause excessive medical care. Temperatures can be measured at various sites, but we have not found the ideal thermometer yet. As a relatively new and popular alternative over traditional methods, infrared ear thermometry has many advantages, but its accuracy remains a major concern. Design. Systematic review and meta-analysis. Data Sources. Medline, Ovid, Elsevier, Google Scholar, Cochrane Library. Study Selection. Cross-sectional, prospective design. Data Extraction. Two investigators independently assessed selected studies and extracted data. Disagreements were resolved by discussion with other reviewers. Results. Mean tympanic temperature was always lower than rectal temperature. The overall pooled (random effects) mean difference between tympanic and rectal temperature was 0.22 degrees C (95% limits of agreement -0.44 degrees C to 1.30 degrees C), which is similar to the within rectal device groups (mercury, 0.21 degrees C, -0.44 degrees C to 1.27 degrees C; electronic, 0.24 degrees C, -0.46 degrees C to 1.34 degrees C). In febrile children group, the pooled mean temperature difference between tympanic and rectal temperature was 0.15 degrees C (95% limits of agreement -0.32 degrees C to 1.10 degrees C). Conclusion. The mean difference was large and the 95% limits of agreement was wide. The accuracy of infrared ear thermometry in children is poor, and it cannot replace rectal thermometry in clinical practice of children.