A retrospective evaluation of the risk of bias in perioperative temperature metrics

Reliability of different body temperature measurement sites during normothermic cardiac surgery

INTRODUCTION

Patients undergoing cardiac surgery can experience significant thermal changes during the perioperative period and, for that reason, it is essential to monitor temperatures with adequate accuracy and precision during cardiopulmonary bypass (CPB). The primary aim of the current study was to measure the discrepancies between temperatures at different body sites during normothermic or mild hypothermic CPB.

METHODS

48 patients undergoing cardiac surgery participated in our study. Simultaneous temperatures were measured at nasopharynx, pulmonary artery, arterial outlet, venous inlet, forehead using a heat flux sensor, and urinary bladder at 5-min intervals throughout surgery. The Bland-Altman plot for repeated measures was used to assess concordance between methods.

RESULTS

The duration of surgery was 360 min (interquartile range (IQR) 300-412), while the median cross-clamp time was 135 min (IQR 101-169). During the CPB time, the average difference between arterial outlet and nasopharyngeal temperature was -0.16°C (95% limits of agreement of ±0.93). The bias between arterial outlet and the venous inflow was 0.16°C and the 95% limits of agreement were -0.63 to 0.95°C. The Bland-Altman analysis showed an average difference between oxigenator arterial outlet and bladder probe of -0.62 (95% limits of agreement of ±1.3). The average difference between arterial outlet and Tcore™ temperatures was 0.08°C (95% limits of agreement of ±1.46). 25 patients (52.08%) presented nasopharyngeal temperatures higher than 37°C in the post-CPB period, but none of them exceeded 38°C.

CONCLUSIONS

Perfusionists should be cautious when using the nasopharyngeal site as the only surrogate of brain temperature, even in normothermic cardiac surgery because the precision of measurements is not entirely adequate.

Çocuk Hastanesinde Perioperatif Hipotermi Sıklığı ve Predispozan Faktörler

Background/Aim: Perioperative hypothermia is more common in pediatric patients than in adult patients due to increased body surface area/weight ratio and limited subcutaneous fat deposits. Therefore, active and passive warming techniques are used more frequently in the surgeries applied to pediatric patients. This study presents the prevalence of perioperative hypothermia and the predisposing conditions for perioperative hypothermia in pediatric orthopedic surgeries in which active and passive warming techniques are applied. Methods: This cross-sectional, descriptive, and observational study included 102 children admitted to the pediatric orthopedic clinic of a children’s hospital. Temperature measurements were made with a calibrated infrared tympanic thermometer in all cases while they were waiting in the service room, when they entered the operation theater, when they left the operation theater, and when they left the postoperative care unit. Their demographic data, hemogram, and thyroid hormone parameters, preoperative fasting times related to the procedure, and temperature were recorded. Also, the humidity and temperature values of the operating theater, the operation type performed, the anesthesia method applied during the operation, the complications encountered, and the time spent in the operating theater were all recorded in their follow-up forms. Data analysis was done using SPSS V21.0 and was conducted at a 95% confidence interval. Results: Hypothermia was observed in 20,58% of 102 patients included in the study. Predisposing factors were determined for perioperative hypothermia as follows; the patient's American Society of Anesthesiologists risk score is above 1, concomitant diagnosis of cerebral palsy, low hemoglobin level in the preoperative period, fasting longer than 8 hours, the low body temperature of the patient in the service room, long time stayed in the operating room, and different humidity values of the operating theatre. Conclusion: Although active and passive warming techniques are applied during the operation, perioperative hypothermia was observed in 20,58% of pediatric patients who underwent the orthopedic operation. Consideration of predisposing factors together with active and passive warming techniques may reduce the incidence of perioperative hypothermia.

Zero-Heat-Flux and Esophageal Temperature Monitoring in Orthopedic Surgery: An Observational Study

Purpose

Perioperative hypothermia prevention requires regular, accurate, and consistent temperature monitoring. Zero-heat-flux (ZHF) thermometry offers a non-invasive, measurement method that can be applied across all surgical phases. The purpose of this study was to measure agreement between the zero-heat-flux device and esophageal monitoring, sensitivity, and specificity to detect hypothermia and patient acceptability amongst patients undergoing upper and lower limb orthopedic surgery.

Patients and Methods

This prospective, observational study utilized Bland–Altman analysis and Lin’s concordance coefficient to measure agreement between devices, sensitivity and specificity to detect hypothermia and assessed patient acceptability amongst 30 patients between December 2018 and June 2019.

Results

Bias was observed between devices via Bland Altman, with bias dependent on actual temperature. The mean difference ranged from −0.16°C at 34.9°C (where the mean of ZHF was lower than the esophageal device) to 0.46°C at 37.25°C (where the mean of ZHF was higher than esophageal device), with 95% limits of agreement (max) upper LOA = 0.80 to 1.41, lower LOA = −1.12 to −0.50. Seventy-five percentage of zero-heat-flux measurements were within 0.5°C of esophageal readings. Patient acceptability was high; 96% (n=27) stated that the device was comfortable.

Conclusion

ZHF device achieved lesser measurement accuracy with core (esophageal) temperature compared to earlier findings. Nonetheless, due to continuous capability, non-invasiveness and patient reported acceptability, the device warrants further evaluation.

Title Registration

The study was registered at www.ANZCTR.org.au (reference: ACTRN12619000842167).