Diagnostic accuracy of three different methods of temperature measurement in acutely ill geriatric patients

Objective: we examined the diagnostic accuracy of different methods of body temperature measurement to diagnose infection in geriatric patients presenting to the emergency department (ED).

Methods: this observational study was done in consecutive patients ≥75 years old presenting to the ED. Body temperature was determined by tympanal thermometry, temporal artery thermometry and rectal temperature measurement. Adjudicated final diagnosis of infection was done by two experts including patient history, clinical and laboratory findings as well as radiographic studies.

Results: a total of 427 patients were included in the data analysis (age: 82.7 ± 5.1 years). Infection was present in 105 patients (24.6%). Respiratory rate, heart rate and body temperature were significantly higher in patients with infection, blood pressure was lower (P < 0.01). Body temperature measured by tympanal and temporal artery thermometry was correlated with rectal thermometry. Body temperature was significantly higher in patients with infection compared with those without infection independent of the method of body temperature measurement (P < 0.001). The diagnostic accuracy for infection quantified by the area under curve (AUC) was comparable among rectal [AUC: 0.72 (95% CI: 0.65-0.80)] and tympanal thermometry [AUC: 0.73 (95% CI: 0.66-0.81)], but significantly lower in temporal artery thermometry [AUC: 0.65 (95% CI: 0.57-0.73; P < 0.001)]. Compared with rectal measurement tympanal thermometry showed a higher bias than temporal artery thermometry (0.54 versus 0.03°C), while its limits of agreement were more narrow (-0.14 to 1.21°C versus -0.94-1.01°C).

Conclusion: diagnostic accuracy for the identification of infection was comparable among tympanal and rectal thermometry and lower for temporal artery thermometry. Different cut-off points should be used to identify infection using tympanal (37.3°C) or rectal (37.9°C) thermometry. In general, temperature measurement is an insensitive method to identify geriatric patients with infection. Registration number clinicaltrials.com: KSMC-tempger-1.