Underbody carbon-fiber contact warming versus underbody forced-air warming to prevent hypothermia during laparoscopic gynecologic surgery: A randomized trial

Abstract

Background

Studies comparing intraoperative warming devices report discordant or out-of-date results. This trial compared two underbody warming devices.

Methods

Thirty patients undergoing elective prolonged laparoscopic gynecologic surgery were randomized to underbody warming by forced air (n = 15) or contact with a carbon fibre blanket (n = 15). The main outcome was esophageal temperature at the end of surgery. We also compared temperature throughout surgery and need for rescue warming, blood loss, fluids infused, urine output, and adverse events. Outcomes were compared with χ² or Fisher exact tests, t-tests, and mixed effects models as appropriate.

Results

No median (interquartile range) differences between forced-air and contact warming were found in initial or final temperatures (36.2 °C [36.0, 36.2] vs 36.3 °C [35.9, 36.6] and 36.6 °C [36.2, 36.8] vs 36.3 °C [35.6, 36.5]). Temperature slightly increased over time in the forced-air group and slightly decreased in contact group (0.11 °C/h (0.02, 0.19) vs −0.05 °C/h (−0.13, 0.03), p = 0.008). A single patient required rescue warming (in contact group after 4.75 h). Surgery took longer in the contact group (3.2 h [2.5, 3.8] vs 4.0 h [2.9, 5.6] h, p = 0.042). Two surgeons complained of dizziness related to ambient heat in the forced-air group. No differences were found in the remaining variables.

Conclusions

During use of the underbody forced-air and carbon-fibre warming devices tested, we recorded only very slight differences in temperature changes over the course of surgery. The variations can be considered clinically unimportant as no significant difference was evident at the end of surgery.

Resumen

Antecedentes

Los estudios comparativos de los dispositivos de calentamiento intraoperatorios reportan resultados discordantes u obsoletos. Este ensayo comparó dos dispositivos de calentamiento corporal bajo el cuerpo.

Métodos

Aleatorizamos a treinta pacientes sometidas a cirugía laparoscópica electiva prolongada en los grupos de calentamiento por aire forzado (n = 15) o de contacto con manta de fibra de carbono (n = 15). El resultado primario fue la temperatura esofágica al final de la cirugía. También comparamos la temperatura a lo largo de la cirugía y la necesidad de calentamiento de rescate, hemorragia, líquidos infundidos, gasto urinario y episodios adversos. Los resultados se compararon mediante las pruebas χ² o exacta de Fisher, prueba t y modelos de efectos mixtos, según el caso.

Resultados

No se encontraron diferencias medianas (rango intercuartílico) entre el calentamiento por aire forzado y por contacto en las temperaturas iniciales o finales (36,2 °C [36, 36,2] vs 36,3 °C [35,9, 36,6] y 36,6 °C [36,2, 36,8] vs 36,3 °C [35,6, 36,5]). La temperatura se incrementó ligeramente con el tiempo en el grupo de aire forzado, y descendió ligeramente en el grupo de contacto (0,11 °C/h (0,02, 0,19) vs -0,05 °C/h (−0,13, 0,03), p = 0,008). Una única paciente requirió calentamiento de rescate (en el grupo de contacto, transcurridas 4,75 h). La cirugía fue más prolongada en el grupo de contacto (3,2 h [2,5, 3,8] vs 4 h [2,9, 5,6] h, p = 0,042). Dos cirujanos se quejaron de mareo relacionado con la temperatura ambiente en el grupo de aire forzado. No se encontraron diferencias en las variables restantes.

Conclusiones

Durante el uso de dispositivos de calentamiento por aire forzado y fibra de carbono registramos únicamente diferencias ligeras en cuanto a cambios de temperatura a lo largo de la cirugía. Las variaciones pueden considerarse sin importancia, desde el punto de vista clínico, ya que no se evidenció diferencia significativa al final de la cirugía.

Introduction

Intraoperative hypothermia is a common adverse event during surgery. In order to avoid well-known complications of hypothermia, guidelines recommend using active warming strategies when there is risk of hypothermia, such as in surgical procedures lasting longer than 30 min under general anesthesia.^1,2

Making an evidence-based choice of warming system is complicated due to differences in type (with or without forced air) and placement (full or partial body, over- or underbody). Forced-air warming and resistive heating systems are widely used in routine practice, and some studies have reported similar efficacy for both systems^3–8 while others found better results with forced air.^9–12

Disadvantages of forced-air systems are the ambient noise and convection currents they produce close to the surgical site.¹³ Air currents could theoretically increase the risk of surgical site infection,¹⁴ although no robust evidence has emerged to support this hypothesis.¹⁵ Carbon-fibre and other resistive systems that warm by direct contact, on the other hand, are more rigid and less able to adapt to the patient’s position.¹⁵ Other constraints on choice of warming are the practical limitations imposed by surgical equipment. During laparoscopic gynaecologic surgery, for example, the abdomen must be exposed, so over-body blankets can only partially cover the patient.

Underbody warming—the least studied method overall—is therefore a logical choice.

The aim of this randomized trial was to compare 2 currently available underbody warming approaches, one using a forced-air mattress and the other a carbon-fibre electric blanket.

Methods

Trial design and setting

This single-centre, randomized, parallel group, open clinical trial was approved by our institutional review board (Hospital del Mar Clinical Research Ethics Committee, # 2019/8860/I Chair Marta Guix MD) and prospectively registered at Clinical Trials.gov (NCT04410068). The study was conducted in accordance with the amended Declaration of Helsinki, and all patients gave their written informed consent. The study was conducted in the surgical area of Hospital del Mar. This manuscript adheres to the CONSORT guidelines for reporting clinical trials.¹⁶

Patients

We recruited consecutive patients undergoing elective major laparoscopic surgery for gynaecologic cancer with an expected surgical time of at least 2 h. Exclusion criteria were active infection, initial temperature higher than 37 °C or lower than 35 °C, diabetes, or autonomic dysfunction.

Interventions

Thirty patients were randomized to warming with an underbody forced-air warming system (Bair Hugger™ 54500, 3 M, Saint Paul, MN, USA) (forced air group, n = 15) or with an underbody carbon fibre electric blanket (Warmtac™, Tacklen Medical Technology, Terrassa, Spain) (contact group, n = 15). Both devices were placed under the torso and the head.

Before surgery, axillary temperature was determined in order to verify inclusion criteria, the temperature in the operating theatre was set at 20 °C and was regularly checked during the procedure, and the warming devices were set to their maximum temperature. During surgery, patient temperature was monitored with an oesophageal thermometer (ER 400-12 Level 1, Smiths Medical, Minneapolis, MN, USA) placed immediately after intubation. Correct positioning of the temperature probe was checked by laryngoscopy and temperature was recorded every 15 min during the procedure. Warming devices were disconnected if the patient’s temperature reached 37.5 °C.

Anaesthesia technique

General anaesthesia was the same for all patients. Anaesthesia was induced with propofol (1.5–2.5 mg/kg), fentanyl (2−3 µg/kg), and rocuronium (0.6 mg/kg) and maintained with desflurane titrated to achieve the bispectral index (Medtronic, Minneapolis, MN) between 40 and 60. Rocuronium was titrated to achieve adequate neuromuscular blockade for surgery (0–1 responses to train-of-four stimulation), and boluses of fentanyl were titrated to maintain heart rate and blood pressure within 20% above or below preoperative values. Patients were mechanically ventilated (Avance CS2, General Electric Healthcare, Fairfield, CT, USA) in volume-control mode. The fresh gas flow rate was set at 1 L/min, inspired oxygen concentration at 60%, and tidal volume at 6 mL/kg of ideal body weight. Positive end-expiratory pressure was started at 5–10 cmH2O and adjusted to achieve best respiratory compliance. Respiratory rate was set to achieve an end-tidal carbon dioxide (ETCO2) concentration of between 34 and 39 mmHg. Temperature, blood pressure, 5-lead electrocardiography, respiratory rate, ETCO2, and pulse oximetry were continuously monitored throughout surgery (Carescape B650, General Electric Healthcare, Chicago, IL, USA), and readings were automatically registered in the anaesthesia chart through the hospital’s intranet (IMASIS software).

A balanced crystalloid solution (PlasmaLyte, Baxter International Inc, Deerfield, IL, USA) was infused through a fluid warming system at 40 °C (Ranger™, 3 M, Saint Paul, MN, USA) starting at 1 mL/kg/h and modified according to the anaesthesiologist’s criteria.

Outcomes

The primary outcome was the difference in oesophageal temperature between the 2 warming system groups (forced air vs contact) at the end of surgery. We also compared temperature changes throughout surgery. To correct for the influence of duration of surgery, we compared the rate of change in temperature measured with the formula (initial temperature – final temperature)/surgical time. If oesophageal temperature decreased to 35 °C, rescue warming was provided with an additional upper body blanket (WarmTouch™, Medtronic, Minneapolis, MN) and the incident was registered. Other secondary outcomes were surgical blood loss reflected by haemoglobin concentration at the end of surgery minus the preoperative concentration. Urine output and adverse events related to the heating systems, such as skin lesions or other complications eventually linked to the warming system, were also recorded.

Sample size

Sample size calculation was based on considering that a between-group difference of 0.5 °C in oesophageal temperature could be defined as clinically relevant, accepting an alpha risk of 0.05 and a beta risk of 0.2 in a 2-sided test. Using the sample size calculator GRANMO (https://www.datarus.eu/en/applications/granmo/), we determined that 11 patients would be required for each group in order to rule out the null hypothesis, assuming a standard deviation (SD) of 0.4 ⁰C.

Randomization and allocation concealment

Randomization was performed with a computer using an unstratified list with a 1:1 ratio. No restrictions were applied. Randomized allocations were concealed by authors M.C. and L.G., who prepared sequentially numbered, opaque, sealed envelopes. Authors M.C., C.R-C. and L.G. enrolled patients, and all authors performed clinical work. The anaesthesiologist in charge of patient care opened the randomized assignment envelope before starting each procedure.

Statistical analysis

Quantitative variables were described as mean (SD) or median (interquartile range [IQR]) and qualitative variables as number and percentage. Bivariate analyses of categorical variables were performed using the χ² or Fisher exact tests, as appropriate; quantitative variables were compared using the Student t-test.

Temperature changes throughout surgery were compared using a mixed effects model for repeated measures, taking into account main effects of group and time (as a continuous variable) and their interactions.

STATA version 15.1 (StataCorp, College Station, TX, USA) was used for statistical analysis; p values less than 0.05 were considered statistically significant.

Results

Thirty-one patients were enrolled from December 2019 to February 2022 and randomly assigned to forced air or contact warming. One patient allocated to the contact group was excluded because of a baseline temperature lower than 35 °C. Thus, data for 15 patients per group were included in the final analysis (Fig. 1). Baseline patient characteristics and surgeries are shown in Table 1.

Figure 1.

Flowchart of patients enrolled, lost, and analyzed.

Table 1.

Patient characteristics at baseline.

	Forced-air group (n = 15)	Contact group (n = 15)	p Value
Age (years)	46 (39−52)	53 (43−72)	0.081
Body mass index (kg/m^2)	25.1 (19.8−28.3)	26.6 (25.1−29.5)	0.191
ASA physical status 1/2/3	1/10/4	4/7/4
Type of surgery (N.)
Endometrial cancer	1	4
Cervical cancer	5	3
Endometriosis	3	4
Ovarian cancer	1	2
Uterine myomatosis	5	2

ASA: American Society of Anesthesiologists. Data are median (IQR).

Comparisons of intraoperative characteristics (Table 2) revealed significant differences only in surgical duration (p = 0.04). Oesophageal temperature measured at the beginning and end of surgery was similar in both groups. Between-group comparison of temperature changes during surgery showed significant differences (interaction term p value < 0.001). However, mean temperature values over 36 °C were recorded in both groups (Fig. 2), and rescue warming was required for only one patient (in the contact blanket group) after 4.75 h of surgery.

Table 2.

Intraoperative parameters and outcomes.

	Forced-air group (n = 15)	Contact group (n = 15)	p Value
Pre-induction axillary temperature (ºC)	36.0 (35.7−36.2)	36.1 (35.6−36.4)	0.694
Initial oesophageal temperature (ºC)	36.2 (36.0−36.2)	36.3 (35.9−36.6)	0.184
Final temperature (ºC)	36.6 (36.2−36.8)	36.3 (35.6−36.5)	0.213
Rate of change in temperature (ºC/h)	0.11 (0.02−0.19)	−0.05 (-0.13−0.03)	0.008
Operating room temperature (ºC)	20.5 (20.1−20.8)	20.5 (20.2−20.9)	0.28
Duration of surgery (h)	3.2 (2.5−3.8)	4.0 (2.9−5.6)	0.042
Urine output (mL/h)	54.3 (26.7−111.4)	61.8 (47.3−91.8)	0.419
Intraoperative IV fluids (mL/h)	195.7 (92.6−240)	175 (145.9−255)	0.576
Blood loss^a (g/dL)	0.8 (0.2−1.7)	0.5 (0.2−2.1)	0.947
Rescue warming (n)	0	1	1

IV: intravenous.

Data are median (IQR).

a

Measured as decrease in haemoglobin concentration (final concentration – baseline).

Figure 2.

Plot of mean (SD) temperatures at each time point. Significant differences between these trajectories were observed (p-value for interaction term on mixed model < 0.001).

Analysis of the rate of change in median (IQR) temperature in each group showed that temperature increased slightly over time with forced-air warming (0.11 °C/h [0.02, 0.19 °C/h]) and decreased slightly over time with contact blanket warming (–0.05 °C/h [–0.13, 0.03 °C/h]) (p = 0.008) (Table 2, Fig. 3). No patient needed disconnection of the warming device due to a temperature of 37.5 °C or higher.

Fig. 3.

Boxplots comparing the median rates of change in temperature in the forced-air and contact carbon- fibre warming groups. Horizontal bars inside the boxes show the medians, boxes show the interquartile range, whiskers show highest and lowest values within 1.5 × the interquartile range, and dots show outlier values. * p value = 0.008.

Urine output, intravenous fluids administered, and blood loss indicated by a decrease in haemoglobin concentration were similar in both groups (Table 2). No skin wounds or other complications that could be linked to the warming devices were observed. However, the surgeon complained of dizziness related to a hot environment during use of the forced-air warming device in 2 surgeries; the patients’ temperatures at that time were 36.4 °C and 36.5 °C.

Discussion

This study found that the 2 underbody warming systems — one based on forced air and the other on contact with a carbon-fibre blanket — are similarly useful for preventing inadvertent perioperative hypothermia during laparoscopic gynaecologic surgery. Patient temperature increased slightly over time in the forced-air group and decreased slightly in the contact-blanket group, but the mean temperature always remained above 36 ºC in both groups. Rescue warming was never needed in the forced-air group and needed only once in the contact-blanket group in a procedure lasting close to 5 h. Bleeding was minimal and similar in both groups and probably had no clinically relevant influence on temperatures.

Comparisons between studies are difficult because the experimental settings cover multiple scenarios (regarding type of surgery or patient age, for example, or clinical settings vs simulation trials), types of warming system, and positioning of the heat source. Our study is the first to our knowledge to compare the widely used underbody Bair-Hugger forced-air system and the relatively new carbon-fibre contact Warmtac blanket used under the body. Given the similarity of outcomes, anaesthesiologists and surgeons can consider additional advantages and disadvantages of each system.^15,17–21 The advantage of underbody forced-air warming is that heat reaches a larger body surface area. However, it is noisy and is suspected to increase the risk of infection due to air circulation,^17–19 although a systematic review found no robust evidence to support that hypothesis.¹⁵ In our study, 2 surgeons complained of dizziness due to high ambient temperature close to the patient. Thus, the possibility of staff discomfort should be taken into consideration. In contrast, a carbon-fibre blanket placed under the patient is silent, produces no air flow, and probably provides a more comfortable work area for surgeons. However, in certain surgical positions the skin surface in contact with the blanket is limited, so this system may be less effective in some procedures.

Strengths of this randomized trial were first, that the population, surgical procedure, and anaesthetic technique were homogeneous in both groups. Second, the duration of surgery was long enough to evaluate the warming systems in a challenging setting. One limitation of the study is that only women were included. Although results cannot directly be extrapolated to men, it seems logical that similar results would be obtained in both sexes in procedures of similar duration and positioning of the body. Another limitation is that the study was performed in a single hospital, although the findings would probably be similar in a multicentre study involving other similar tertiary care hospitals.

It could be argued that our sample size was small, and that this limits the strength of our findings. However, as it was larger than the 11 patients per group calculated to be required, we are confident it was adequate for the purpose of the study. When we designed the study, we considered a difference of 0.5 °C between warming devices at the end of surgery to be clinically relevant, with a standard deviation of 0.4 °C. Although a larger sample size may have revealed statistically significant differences in the final temperatures between groups, the difference would have probably remained minimal and clinically irrelevant. The median temperature of patients in each group was 36.6 °C and 36.3 °C at the end of the intervention, and we believe that any difference in temperature slightly above 36 °C, even if statistically different in larger groups, would still be clinically insignificant. As for the occasional statistically significant temperature differences during surgery, we believe they are attributable to our finding that the median temperature increased by 0.11 °C per hour in the forced-air warming group while decreasing by 0.05 °C per hour in the carbon-fibre contact group.

Conclusions

During use of the underbody forced-air warming system (Bair Hugger 54500) and the underbody carbon-fibre electric blanket (Warmtac), body temperatures varied slightly over the course of laparoscopic gynaecologic surgery, but differences were only fleetingly significant. However, the median temperatures always remained above 36 °C in both groups. The variations presumably had no clinical relevance, as no differences were observed at the end of surgery. Rescue warming was never needed in the forced-air group and needed only once in the contact- blanket group. Clinicians can therefore be confident when choosing either system based on the advantages and disadvantages each offers in specific surgical settings.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Acknowledgments

The authors thank Magda Guilera for writing assistance, Xavier Duran for his statistical assistance, and Mary Ellen Kerans for advice on English expression in a later draft.