Accuracy of Infrared Thermography for Detecting Febrile Critically Ill Patients

Overview

Accurate determination of critically ill patient of being febrile or not is an essential part of management critically ill patients as it prompt investigating the underlying cause and initiating therapeutic action. Pulmonary artery catheter thermistor is considered the gold standard for temperature measurement . Central non-vascular thermometer such as esophageal, bladder and rectal thermometer showed excellent correlation and agreement with pulmonary artery catheter thermistor and has been accepted as alternative methods for core body temperature assessment. However, those methods are invasive and cannot be tolerated in conscious patients. Peripheral thermometer such as oral, axillary and tympanic membrane thermometer are either impractical in unconscious patient and/or carry the risk of trauma and infection. Infrared thermography (IRT) is a non-contact and non-invasive imaging approach that enable real-time estimation of body temperature by detecting infrared emission. IRT had been used for screening for mass detection of febrile patients at airport at times of infectious disease outbreak. In adult population, IRT showed good accuracy in detecting febrile patients in emergency department; however, those studies used 37.7⁰ C as the fever threshold and the reference standard was either oral 6 or tympanic membrane thermometer. No studies to the best of our knowledge had evaluated the IRT accuracy in detecting fever in critically ill patients using esophageal thermometer as a reference method.

The enrolled patient will be sedated according to the ICU protocol. A calibrated esophageal thermometer will be inserted by the attending intensivist through the nostril or oral cavity into the lower third of the esophagus to approximately 10-20 cm. IRT will be performed using FLIR C2 compact thermal camera. The focal length of the thermal camera will be adjusted at 0.5 meter from the patient head. scanning will conducted over a period of 10-seconds. The researcher will record the highest reading obtained from medial canthus. No cold fomentations will be applied to the head nor the axilla of the patient for 30 minutes at least before measuring IRT. FLIR C2 thermal camera will be calibrated before use to eliminate the effect of the ambient room temperature. The researcher of will obtain the IRT temperature will be blinded to the esophageal probe reading.