Reliability of Three Electrothemometers

Abstract

0130 Tissue temperature is an objective means of assessing physiological function. Numerous therapeutic modalities studies have used temperature to assess their effectiveness. Manufactures of temperature measuring machines report the reliability and accuracy of their equipment [accuracy of the electrothermometers of our interest were ± 0.1°C and ± 0.5 + (.001 * temperature measurement)°C respectfully]; but we were unaware of any study which compared and verified the reliability of this equipment. PURPOSE: To determine the reliability of 3 electrothermometers from 2 different manufactures. METHODS: Three sets of six, PT6 were attached to 3 electrothermometers; 2 of the machines required PC interface. For each trial a different set of thermocouples were attached to each machine. The sensing thermocouple tips were placed in a water bath kept at a constant temperature of 18.4°C for all three trials. Water bath temperature was monitored using a mercury thermometer graded by 0.1°C. Measurements were taken every 10 seconds for each 3 minute trial. Means and standard deviates were computed for the differences and absolute differences between the electothemometers reading and that read from the mercury thermometer. An ANOVA for repeated measures followed by Tukey-Kramer post-hoc testing were used for statistical analysis. RESULTS: The 2 electrothermometers from the same manufacture (0.058 ± 0.032°C and 0.034 ± 0.021°C respectfully) did not differ statistically from each other (p<0.05, Tukey- Kramer), but both were significantly differed from the third unit (0.638 ± 0.205°C) (p<0.05, Tukey-Kramer). When looking at relative differences, the first manufactures units were consistently lower and the second manufacture was consistently greater than the temperatures read from the mercury thermometer. CONCLUSIONS: Not all equipment is equal in is stability and reliability. The electrothermometers from the same manufacture were more stable than the third as well as being within the reliability published by the manufacture (± 0.1°C) when used with PT6 thermocouples. The third machine used in our study had measurement variability greater than that indicated in the users manual (∼ ±0.5°C). Variability of ± 0.64°C can significantly influence finding when intramuscular temperature changes are usually not greater than 5°C for most modalities studied. Future reports of temperature data should include the reliability of the specific equipment used to obtain the measures.