Comparison of forced-air and water-circulating warming for prevention of hypothermia during transcatheter aortic valve replacement

Abstract

Introduction: Transcatheter Aortic Valve Replacement (TAVR) procedures at our institution were complicated by perioperative hypothermia despite use of the standard of care forced-air convective warming device (the BairHugger, Augustine Medical Inc, Eden Prairie, MN, USA). To remedy this problem, we initiated a quality improvement process that investigated the use of a conductive warm water-circulating device (the Allon ThermoWrap, Menen Medical Corporation, Trevose, PA, USA), and hypothesized that it would decrease the incidence of perioperative hypothermia.

Methods: We compared two different intraoperative warming devices using a historic control. We retrospectively reviewed intraoperative records of 80 TAVRs between 6/2013 and 6/2015, 46 and 34 of which were done with the forced-air and water-circulating devices, respectively. Continuous temperature data obtained from pulmonary artery catheter, temperature upon arrival to cardiothoracic ICU (CTU), age, BSA, height, and BMI were compared.

Results: Patients warmed with both devices were similar in terms of demographic characteristics. First recorded intraoperative temperature (mean 36.26 ± SD 0.61 vs. 35.95 ± 0.46°C, p = 0.02), lowest intraoperative temperature (36.01 ± 0.58 vs. 34.89 ± 0.76°C, p<0.001), temperature at the end of the procedure (36.47 ± 0.51 vs. 35.17 ± 0.75°C, p<0.001), and temperature upon arrival to the CTU (36.35 ± 0.44 vs. 35.07 ± 0.78°C, p<0.001) were significantly higher in the water-circulating group as compared to the forced-air group.

Conclusion: A quality improvement process led to selection of a new warming device that virtually eliminated perioperative hypothermia at our institution. Patients warmed with the new device were significantly less likely to experience intraoperative hypothermia and were significantly more likely to be normothermic upon arrival to the CTU.

Fig 1. Mean temperature versus time in TAVR patients warmed with forced-air and water-circulating devices.

Temperature decreased significantly over time in the forced-air group (P < 0.001), and the trajectories for the two groups were significantly different (p = 0.004). Dashed lines represent 95% confidence interval. Mean temperature over time was estimated with a random effect mixed model using B splines for time, and included an interaction between time and device status. The number of patients present at the 1, 2, 3, and 4-hour time points in the forced-air and water-circulating groups were 46 and 34, 43 and 19, 12 and 3, and 1 and 1, respectively.

Fig 2. Temperature on arrival to the CTU.

Mean temperatures ± SD for patients warmed with the forced-air and water-circulating devices were 36.4 ± 0.4 and 35.1 ± 0.8 degrees Celsius, respectively (p < 0.001). Overall, 97% of patients warmed with the water-circulating device versus 13% of those warmed with the forced-air device were above 36.0 degrees Celsius upon arrival to the CTU (p < 0.001).

Fig 3. Degree-hours below 36 degrees Celsius.

Area between mean temperature versus time curve and below 36.0 degrees Celsius for patients warmed with forced-air and water-circulating devices. The median areas under 36.0 degrees Celsius were 0.0 and 1.6 degree-hours for the forced-air and water-circulating groups, respectively. The duration and magnitude of hypothermia was significantly greater in the forced-air warming group (p < 0.001).