Study Design
The Ethical Committee of Ningbo NO.7 Hospital approved this study, which follows the tenets of the Declaration of Helsinki, and we pre-registered it at http://www.chictr.org.cn/index.aspx (ChiCTR1800019117). This study adheres the applicable CONSORT guidelines. We enrolled healthy pregnant women undergoing elective cesarean deliveries under spinal anesthesia after obtaining their informed consents. American Society of Anesthesiologists physical status I-II parturients, aged 18 to 40 years, with more than 37-week gestations, singleton pregnancies, and scheduled for cesarean delivery under spinal anesthesia were eligible for enrollment. We excluded women with coagulation abnormalities, thyroid disease, cesarean delivery using epidural or general anesthesia, and baseline temperatures ≥37.5˚C.
Study Protocol
After obtaining the signed informed consents, we randomly allocated eligible participants to either the control or the intervention groups. Randomization was computer-generated using Microsoft Excel’s random number generator, and we concealed allocations using sequentially numbered opaque sealed envelopes.
All parturients fasted for eight hours before the cesarean section. Once in the preoperative waiting area, the parturients in the intervention group received 30 minutes of upper body preoperative warming using a forced-air warming device (EQ-5000 230V, Smiths Medical ASD, Rockland, USA) set to 43˚C and nurses established intravenous accesses. The women in the intervention group received Ringer’s lactate solution pre-warmed to 37˚C through a 3MRangerTMFluid Warmer until the end of the procedure. We monitored the patients during the interventions. We discontinued the intervention in cases in which the parturients experienced adverse side effects related to warming such as diaphoresis or nausea and vomiting, or if the core thermometer was >37.5˚C.
After prewarming, we immediately transferred the term parturients to the operating room (OR).Participants in the intervention group received 30 minutes of upper body preoperative warming in the preoperative waiting area, and received IV fluid warming during the observation period(preoperative waiting area, OR and PACU ).The women in the control group received usual care consisting of no active warming and they received the intravenous fluid at room temperature throughout the procedure(preoperative waiting area, OR and PACU ). We recorded data on vital signs including heart rate, blood pressure, hemoglobin peripheral saturation, and baseline core temperature in the preoperative area. The same operator measured patients’ core temperatures using an infrared tympanic thermometer (PRO6000, Braun, Marlborough, MA USA 01752) with disposable covers, and recorded the average value of three measurements. The hospital maintained central control of the temperatures of the preoperative area, OR, and post-anesthesia care unit (PACU), and we obtained the temperature readings from the thermostat.
An anesthesiologist not involved in the study applied all spinal anesthesias at the L3-4 interspace, with 2 mL of 0.5% plain bupivacaine, using a 25-gauge Quincke needle. The surgeon commenced the operations once a sensory blockade above the T4 level was achieved according to the results of pinprick tests. After the operation, all patients were transferred to the PACU covered with a cotton sheet and a blanket.
We obtained values for core temperature, maternal thermal comfort scores, and the incidences of shivering and hypothermia at the following timepoints: T0 = baseline, T1 = pre-spinal, T2 = post-spinal, T3 = after 15 minutes in the OR, T4 = after 30 minutes in the OR, T5 = surgery end, T6 = PACU arrival, T7 = after 15 minutes in the PACU, T8 = after 30 minutes in the PACU. According to Guidelines[21], we defined maternal hypothermia as a core temperature <36˚C. We assessed thermal comfort scores using a verbal numerical scale on which we defined 0 as completely unsatisfied with the “thermal comfort” and 100 as completely satisfied. We graded shivering during and after the cesarean section according to the Bedside Shivering Assessment Scale (0, no shivering; 1, shivering localized to the core and neck; 2, shivering including the upper extremities; 3, total body shivering)[22]. The anesthesiologist provided meperidine according to their own criteria. A midwife recorded neonatal axillary temperature, and Apgar scores at 1 and 5 minutes after birth. Based on our institutional guidelines, if the core temperature was lower than 35.5°C, rescue warming would performed for the parturients by using a forced-air warming device.
We defined bradycardia as a heart rate<50 beats/min, and treated it with 0.5 mg of intravenous atropine. When the systemic pressure decreased more than 30% of the baseline pressure or dropped below 90 mmHg, we administered ephedrine (5 mg).Mean arterial pressure and heart rate was measured at baseline, prespinal, postspinal and at the end of the procedure.
We recorded demographic data (age, height, weight, parity, and gravidity) and surgical and anesthetic variables (Preoperative and total volume of intravenous fluids, estimated blood loss, duration of surgery, and the ambient temperatures in the preoperative area, OR, and PACU).
Statistical Analyses
The primary outcome measure was the core temperature change between two groups from baseline to the end of the surgical procedure.Secondary outcomes included thermal comfort scores during the operation, the incidence of shivering and hypothermia (<36˚C), the core temperature on the arrival at the PACU, neonatal axillary temperature at birth, and Apgar scores at 1 and 5 minutes).
Analysis of covariance for repeated measures was under taken to calculate the sample size. A bonferroni correction for multiple pairwise comparisons was used, giving an adjusted P value level of significance (P<0.01).A clinically significant difference in the core temperature between study groups was set at 0.4°C according to our pilot trial with a standard deviation of 0.5°C,which was also consistent with Chung et al' s study[23].A sample size of120 patients, including 20% dropouts, was estimated to provide 90% power for detecting a statistically significant difference between groups at an α level of 0.01.
We expressed normally distributed continuous data as means ± SDs, and compared variables between study groups using the Student t test. Nonparametric data are presented as medians (interquartile ranges), and compared between study groups using the Mann–Whitney U test. We investigated associations among discrete variables using the c2 or Fisher exact tests. Two-way repeated measures ANOVA was applied with change from baseline as the dependent variable, and the intervention, time, and the treatment multiplied by time interaction as independent variables. We also used two-way repeated measures ANOVA to assess the core temperature change and the thermal comfort between groups at each timepoint. We performed all statistical analyses using the SPSS software (version 22.0, SPSS, Chicago, IL, USA). We considered P-values <0.05 as statistically significant.