Performance of a transesophageal echocardiography probe at temperature monitoring during simulated hypothermia and rewarming

Madeline Larochelle; Margarita Popova; David Mackenzie; Andrew Fried; Peter Croft; Joshua Rehberg; Christina Wilson

doi:10.15441/ceem.24.321

Abstract

Objective

To determine whether a transesophageal echocardiography (TEE) probe can accurately measure temperature and be used to monitor temperature changes over time without overheating in an experimental model of hypothermia and rewarming.

Methods

A 6-L water bath was heated with a sous vide immersion circulator to 24, 28, 32, and 36 °C to simulate severe hypothermia, moderate hypothermia, mild hypothermia, and normothermia, respectively. A TEE probe, esophageal temperature probe, and bladder temperature probe were used to measure temperature every 60 seconds for 15 minutes.

Results

The TEE probe reported temperatures with a mean difference of 0.60 °C (95% confidence interval [CI], 0.51 to 0.69 °C) compared to the sous vide immersion circulator. The esophageal probe and bladder probe reported temperatures with a mean difference of –0.19 °C (95% CI, –0.23 to –0.14 °C) and –0.20 °C (95% CI, –0.26 to –0.14 °C), respectively.

Conclusion

During this simulation, the TEE tip temperature did not increase beyond the expected changes produced by water temperature. The probe temperature was less accurate than the esophageal and bladder temperature probes but demonstrated precision in monitoring temperature changes and stable hypothermia. Based on this study, TEE probes should not be relied upon for an accurate initial temperature but can likely be used to monitor changes in temperature over time.

Keywords: Transesophageal echocardiography; Resuscitation; Hypothermia; Cardiac arrest

INTRODUCTION

Temperature monitoring during resuscitation from cardiac arrest during accidental hypothermia is essential to monitor the response to rewarming efforts and to guide further resuscitation [1]. Rectal, bladder, and esophageal temperature probes can provide continuous temperature measurement but have potential limitations. Active rewarming with peritoneal lavage or warm water instillation into the bladder can artificially elevate rectal and bladder temperature readings [2]. Esophageal temperature measurement is often considered superior to rectal and bladder temperature measurements given its proximity to the heart and major blood vessels and has been shown to respond more quickly to dynamic changes in core temperature [3]. Transesophageal echocardiography (TEE) is increasingly being used to guide resuscitation during cardiac arrest [4]. TEE probes routinely measure and display the probe tip temperature during use and could potentially be used to monitor temperature during hypothermic cardiac arrest as an adjunct or alternative to other temperature sensors. Simultaneous cardiac ultrasound imaging and temperature monitoring with the TEE probe could streamline resuscitation efforts by reducing the need for additional interventions by eliminating the need for an additional temperature monitoring probe.

Ultrasound probe manufacturers warn that operators should expect heating of the TEE probe with routine use. Continued operation in two-dimensional (2D) mode results in the lowest transducer surface temperatures, and using a still image can reduce temperatures further. In manufacturer studies, there was a rise in TEE tip temperature of 4.9 °C with simulated use and of 10.1 °C at the controls and settings anticipated to produce the maximum temperature [5]. A study in patients undergoing cardiac surgery with cardiopulmonary bypass showed that temperature monitoring with a TEE probe during periods of cooling and rewarming was superior to monitoring with a nasopharyngeal probe without any report of probe overheating [6].

This study uses an experimental model of hypothermia and rewarming to determine if a TEE probe can accurately measure temperature and be used to monitor temperature changes over time without overheating. In a prior study, temperatures were measured during cardiopulmonary bypass and only reached mild hypothermia; we aim to evaluate TEE temperature performance in severe, moderate, and mild hypothermia ranges [6].

METHODS

Study design

A 6-L water bath was heated to a starting temperature of 24 °C using a sous vide immersion circulator (Anova Precision Cooker 3.0, Anova Applied Electronics Inc), a heating device that circulates water to maintain a uniform water temperature. The starting temperature of 24 °C was selected to simulate severe hypothermia [1]. Once the water bath was at temperature, as determined by the sous vide immersion circulator, a TEE probe (SonoSite TEEx 8-3 MHz Transducer, Fujifilm SonoSite Inc), an esophageal temperature probe (9F General Purpose Temperature 400 Series, DeRoyal Industries Inc), and a bladder temperature probe (14F Surestep Temperature Sensing Foley System, BD) were added to the water bath (Fig. 1). The tip of each device was placed in approximately the same location and depth in the water bath and was monitored throughout the experiment, making adjustments as needed. The water bath was maintained at 24 °C for 15 minutes, and the temperature reading of each device was recorded every 60 seconds. After 15 minutes, the water temperature was raised to 28 °C to simulate moderate hypothermia [1]. The temperatures were recorded every 30 seconds until the water bath reached a steady temperature of 28 °C as indicated by the immersion circulator. The temperature was maintained at 28 °C for 15 minutes, and temperatures were recorded every 60 seconds. This process was repeated for a temperature of 32 °C to represent mild hypothermia and 36 °C to represent normothermia [1]. Each time a target temperature was reached, a brief resuscitative TEE protocol was simulated [7], in which an operator performed an omniplane to 130°, saved an image, performed an omniplane back to 0°, and saved another image. This was all conducted without removing the probe from the water bath. Apart from this brief protocol, continuous 2D video monitoring without Doppler was maintained throughout the experiment.

Statistical analysis

Statistical analysis was performed in Microsoft Excel (Microsoft Corp). Bland-Altman plots were created to analyze the agreement between the sous vide immersion circulator temperature/water bath temperature and the TEE probe, bladder temperature probe, and esophageal temperature probe.

RESULTS

Continuous temperature monitoring was performed for a total of 70 minutes, with 78 temperature recordings for each device (Fig. 2). There were three episodes of water bath heating, each between 2 and 5 minutes. Fig. 3 displays the Bland-Altman plots of all three devices compared to the reported temperature of the sous vide immersion circulator. The difference between the immersion circulator and each device is plotted as a function of the mean of the two measurements. The TEE probe reported temperatures with a mean difference of 0.60 °C (95% confidence interval [CI], 0.51 to 0.69 °C). The upper level of agreement was 1.38 °C (95% CI, 1.23 to 1.54 °C) and the lower level of agreement was –0.18 °C (95% CI, –0.02 to –0.34°C) (Fig. 3A). Compared to the immersion circulator, the esophageal probe showed a mean difference of –0.19 °C (95% CI, –0.23 to –0.14 °C). The upper level of agreement was 0.22 °C (95% CI, 0.14 to 0.30 °C) and the lower level of agreement was –0.59 °C (95% CI, –0.67 to –0.51 °C) (Fig. 3B). For the bladder temperature probe, the mean temperature difference was –0.20 °C (95% CI, –0.26 to –0.14 °C). The upper level of agreement was 0.30 °C (95% CI, 0.20 to 0.41 °C) and the lower level of agreement was –0.70 °C (95% CI, –0.80 to –0.60 °C) (Fig. 3C). With the circulator as the reference standard, the TEE probe and both monitoring devices effectively tracked temperature changes. The mean TEE temperature was higher than the reference standard, while the mean esophageal and bladder probes were lower. The mean difference was lowest with the esophageal probe and highest with the TEE probe, but all modalities tracked temperature changes accurately.

DISCUSSION

Based on manufacturer testing, this TEE probe has the potential to heat accurately under certain circumstances [5]. During this simulation, using only 2D mode for an extended period of time and occasional omniplane, the TEE tip temperature did not increase unexpectedly other than changes from the water bath temperature. The probe temperature was an average of 0.6 °C warmer than the reported sous vide immersion circulator temperature and was less accurate than both the esophageal and bladder temperature probes. This temperature difference is unique to the TEE probe that was used in this experiment and may vary between TEE probe types. The thermometer on the TEE probe used in this experiment cannot be calibrated; if calibration was possible, temperature reporting might have been more accurate.

The mean difference in temperature between the TEE probe and the sous vide immersion circulator during rewarming episodes fell below the lower level of agreement of –0.18 °C, whereas the mean temperature difference during steady state water bath temperature was 0.6 °C, suggesting a TEE probe lag during rewarming phases. The TEE probe quickly reached a steady state temperature upon water bath equilibrium. Rewarming during hypothermic cardiac arrest typically is performed very slowly, so the lag in rapid temperature change is unlikely to be clinically relevant. The esophageal and bladder temperature probes also recorded temperatures lower than the reported water bath temperature during episodes of warming. This suggests that the water bath temperature was inaccurate or that these temperature probes also lagged.

This study has several limitations. The major limitation is that this is an experimental model using a water bath and might not be applicable to humans. This simulation tested a single TEE probe and was less reliable than simultaneous testing of multiple probes of the same design. Our findings cannot be generalized to other TEE probes on the market. There was no gold standard for the water bath temperature. While we assigned the sous vide immersion circulator temperature as the standard for comparison, this temperature may have been inaccurate, limiting the conclusions. Use of the Bland-Altman plot, even in the absence of a gold standard, offers insight into the relative accuracy and consistency of the different temperature probes. The water bath temperature also may have varied in different parts of the container, although efforts were made to position the devices in the same area of the water bath.

In this simulated model of hypothermia and rewarming, the TEE probe demonstrated precision in monitoring stable hypothermia and rewarming episodes but was less accurate than esophageal and bladder temperature probes. During an extended period of continuous 2D ultrasound without Doppler, the TEE probe did not heat up outside of the expected water bath temperature changes. Based on this study, TEE probes may not be ideal to determine an accurate initial core temperature but can likely be used to monitor changes in temperature over time. Future in vivo data may further inform the use of TEE to monitor rewarming.