Continuous haemodynamic effects of left tilting and supine positions during Caesarean section under spinal anaesthesia with a noninvasive cardiac output monitor system

This article is accompanied by the following Invited Commentary:

Lyons G, Kranke P. Uterine tilt for caesarean section. Eur J Anaesthesiol 2019; 36:6–7.

Editor,

In full-term parturients, aortocaval compression may affect maternal haemodynamics, for example supine hypotension, and foetal outcomes. Although the incident of significant supine hypotension at term is unknown, current clinical practices and guidelines suggest a tilt range of 12 to 15° for full-term parturients undergoing elective Caesarean section under neuraxial blockade. ^1,2 In clinical practice, however, physicians tend to apply a lesser degree of left tilt because of subjective feelings of ‘insecurity’ from parturients and objective difficulty of this surgical position. ³ In truth, the degree of tilt is usually unmeasured and arbitrary. In addition, residual inferior vena cava compression may also be observed at lesser degrees of tilt, thus compromising the effectiveness of the tilt position on maternal haemodynamics. ⁴

In this study, we introduced the noninvasive cardiac output (CO) monitoring (NICOM^™; Cheetah Medical Inc., Newton, Massachusetts, USA) system to compare the maternal haemodynamic effects of the supine and 15° left tilt positions in full-term parturients under spinal anaesthesia for elective Caesarean delivery.

Following approval by the Institutional Review Board of Changhua Christian Hospital, Chairperson Ying-Ming Chiu, protocol number 150605, on the 11 August 2015, every patient provided written informed consent to participate in the study. Eighty full-term singleton parturients scheduled for elective Caesarean delivery were recruited into the study. Cases involving current labour, multiple pregnancy, foetal distress and a history of hypertension, pre-eclampsia, diabetes, heart disease, placenta praevia or obesity (BMI > 35 kg m⁻²) were excluded from our study.

All parturients were allocated to either the control group (supine position after spinal insertion) or study group (15° left lateral table tilt after spinal insertion) by a computerised random number table and the sealed envelope technique.

In the operating room, 750 ml of 0.9% saline was transfused into all parturients via a 20-gauge cannula. Monitoring comprised pulse oximetry, electrocardiography and noninvasive blood pressure (BP) measurement via a cuff on the left arm. A bioreactance-based NICOM system was used in this study. This system was automatically calibrated, while the patient was in the supine position, via four electrodes placed on the thorax. Cardiovascular variables, including heart rate, CO, cardiac index (CI), stroke volume (SV), SV variation, total peripheral resistance (TPR) and TPR index (TPRI) were recorded every minute. Automated BP measurements were taken every 2.5 min throughout the study. Baseline variables were recorded with the patient in the supine position before spinal insertion.

A standard spinal technique was performed with the patient in the right lateral position; in brief, 2.4 ml of hyperbaric bupivacaine with 0.2 mg of morphine and 10 μg of fentanyl was injected via a 25-gauge spinal needle inserted at the L3 to L4 interspace. All participants were placed in the supine position immediately after spinal injection. Patients allocated to the tilt position group were then placed on an operating table tilted to a table angle of 15° measured by protractor. The level of sensory block was determined using a pinprick and alcohol pad (i.e. cold sensation).

Patients received 5 to 10 mg of intravenous ephedrine if the systolic pressure was less than 80 mmHg or less than 80% of the baseline. The total administered amount of ephedrine and all episodes of nausea and vomiting were recorded. Apgar scores at 1 and 5 min were recorded as neonatal outcome.

Student's t test and χ² test were used to analyse demographic characteristics between the two groups. As data are collected on the same units across successive points in time, multiple linear regression model with generalised estimating equation (GEE) was used to estimate the effect of posture and other associated predictors on cardiovascular variables. SPSS statistical software version 22 (SPSS Inc., Chicago, Illinois, USA) was used for the analysis, and a P value less than 0.05 was considered statistically significant.

According to our pilot study, the minimum sample size in each group was estimated to be 37 using the statistical software G*POWER statistical software (version 3.1.9.2; Franz Faul, Universität Kiel, Kiel, Germany) with the formula for calculation of samples of repeated measures to detect an effect size of 0.25 on each domain score, with 80% power and 5% type 1 errors.

A total of 102 parturients were screened for eligibility for the study from 9/9/2015 to 22/3/2016, and 80 parturients met the inclusion criteria. Of the initial 80 parturients, one was excluded because of foetal distress after admission and data of two patients were excluded from the analysis because of failed spinal anaesthesia or a failed data record. The maternal demographics of the two groups did not differ significantly and a comparison of baseline cardiovascular variables similarly revealed no intergroup differences.

Mean cardiovascular variable values from the period from spinal insertion to foetal delivery are presented graphically in Fig. 1. The SBP, mean arterial pressure (MAP), CI and TPRI did not differ significantly among different time points between groups. Table 1 presents a multiple linear regression model analysis of various parameters using the GEE method. After controlling for other predictors, the haemodynamic variables did not differ significantly between the lateral tilt and supine groups.

Fig. 1: Time courses of SBP, mean arterial pressure, cardiac index and total peripheral resistance index values in the supine () and left-tilt (○) groups. Patients received spinal insertion in the lateral position; SA, spinal anaesthesia. All haemodynamic variables are not significantly different between two groups.

Table 1: The effect of posture and other associated predictors on cardiovascular variables analysed using the generalised estimating equation method

The incidence of nausea was 39.5% in the supine group and 35.9% in the tilted group (P = 0.36), and the corresponding incidences of vomiting were 2.6 and 3.9%, respectively (P > 0.05). The mean doses of ephedrine administered to the supine and tilted groups were 12.1 (SD = 11.2) and 13.6 (SD = 12) (P > 0.05). Similarly, the Apgar scores at 1 and 5 min did not differ between the groups.

Previous studies reported the advantages and disadvantages of the obstetric lateral-tilt position in combination with a neuraxial block, using BP as an indirect measure of clinical outcome. ^5,6 In contrast, our study used other cardiovascular parameters, including CO and TPR, to assess aortocaval compression in parturients under spinal anaesthesia.

In our study, there were no significant differences between the left-tilt and supine groups regarding haemodynamic parameters such as SBP, MAP, CI and TPRI. Both groups also had similar incidence rates of nausea and vomiting. In addition, we observed that CO and TPR, when analysed via multiple linear regression with the GEE method, correlated positively with BMI. Therefore, we selected CI and TPRI, rather than CO and TPR, as our study measurements.

Although both British (The National Institute for Health and Care Excellence) and American guidelines suggest a 12 to 15° tilt during Caesarean section under regional anaesthesia, ^1,2 we did not observe a significant effect of the tilted position in our study. The explanation of these results might be that the sample size is too small to ensure the incidence of supine hypotension.

A recent study reported that maternal position between supine and 15° tilt in elective Caesarean section did not alter neonatal acid-base status, ⁷ which we did not access. However, Apgar score in our study showed no difference.

Nevertheless, the results may not be applicable to parturients with moderate risk (i.e. pre-eclampsia or history of cardiac disease). Further studies using the NICOM system to monitor parturients with moderate risk during spinal anaesthesia may provide more information for diagnosis and treatment.

In this study, the NICOM system was introduced to compare left-tilt and supine position in full-term parturients under spinal anaesthesia on haemodynamics. Present guidelines suggest a tilt position for full-term parturients undergoing elective Caesarean section under neuraxial blockade; however, we observed that the haemodynamic parameters were not statistically different between the two groups. Neither were the incidence of nausea and vomiting. Nevertheless, further investigation could focus on the monitoring of parturients with moderate risk, who do not need invasive monitoring.

Acknowledgements relating to this article

Assistance with the letter: none.

Financial support and sponsorship: none.

Conflicts of interest: none.