DISCUSSION
The principal finding of our study was that during labour and the first two hours postpartum, SI increased significantly in patients with NLA compared to those without NLA. The difference between the groups remained significant after adjusting for confounders. The increase in SI was statistically and clinically significant, with a difference of 0.04 to 0.10 observed between groups (NLA and without NLA). According to previous references, the expected variation in SI in the postpartum period ranged from 0.7 to 0.9.13–16. Thus, a change in SI above a specific level of 0.9 implies significant blood loss, haemodynamic instability, and an increased risk of adverse outcomes. Although small, a change of 0.05-0.1 in SI is of substantial relevance in this setting.
Using a prospectively collected high-quality dataset from two hospitals, we describe the most extensive set of SI values gathered longitudinally in patients with and without NLA. A previous study by Nathan et al., which included 316 patients, noted that the SI was higher in those under NLA (33% of their sample). The authors suggested that epidural anaesthesia increased the mean SI by approximately 0.046.31. Other authors have explored the SI in obstetric patients without reporting the NLA percentage used in their population.9,10,12,28
Our data showed higher HR values during the postpartum period, with similar SBP among patients with NLA compared to those without obstetric analgesia. Extensive literature describes haemodynamic changes during pregnancy and delivery, including an increase in HR and a decrease in SBP at rest.29 These adaptations are more pronounced during labour and the immediate postpartum period, probably due to pain, medications, and maternal anxiety.30 During labour, uterine contractions displace 300–500 mL of blood from the uterus into the central circulation.31 Immediately postpartum, there is an increase in cardiac output due to diminished lower extremity venous pressure, sustained myometrial contraction, and loss of low-resistance placental circulation.
Pregnancy increases the sympathetic system’s dependency on maintaining haemodynamics, including the venous return and systemic vascular resistance. This situation means that pregnant patients are at an increased risk of hypotension and haemodynamic instability induced by neuraxial analgesia and anaesthesia. Despite the advantages of epidural analgesia over other techniques, this sympathetic block might explain some of the increase in SI in our patients. Furthermore, a compensatory mechanism needs to be activated in patients under NLA to maintain haemodynamics after a decrease in systemic vascular resistance, which generates a response based on a rise in HR.22 Another potential explanation for the haemodynamic changes leading to a higher SI in patients under NLA might be the addition of some medications, such as uterotonics.32,33 Several studies suggest that oxytocin induces a profound and transient decrease in both SBP and DBP during caesarean delivery secondary to vasodilatation, with tachycardia and an increase in cardiac output. This effect has been proposed as a secondary to peripheral vasodilation.22,34,35
In a study by Munn et al.36, randomised women undergoing intrapartum caesarean delivery received a prophylactic infusion of oxytocin or bolus after delivery. A higher dose was associated with clinically significant tachycardia and hypotension (SBP of 100 mmHg or a decrease of ≥ 20% from baseline).
The increase in heart rate reported during the early first and second stages of labour coincides with uterine contractions and bearing-down efforts. This change in heart rate might be explained by the displacement of blood from the choriodecidual space, increased venous return to the heart, and increased catecholamine release due to pain and other stimuli (anxiety). Additionally, during labour, uterine contraction pain becomes more pronounced during the transition to the second stage. This pain induces a neuroendocrine stress response, increasing heart rate and cardiac output.37
Our results show that SI values between groups with and without NLA were not modified by maternal age, parity, haemoglobin levels, gestational age, or cervical dilation. Similarly, a previous study by our group reported that labour, but not maternal anaemia, was associated with lower SI values during the first and second stages of labour compared to SI values in patients during the third trimester, possibly secondary to maternal pain and anxiety.11
Our study has several limitations. The main weakness is the nonrandomised and direct relationship between NLA exposure and SI between centres, given that only one of the centres includes NLA during labour. This relationship may be inseparable from the association between the centre and SI. The unmatched analysis for oxytocin use limited an interpretation of the SI differences among the studied groups; consequently, we cannot explain whether NLA, oxytocin or both were the reasons for the significant differences in SI.
Additionally, the study design was restricted to pain management between the two clinical sites, implying differences in other interventions during labour, such as fluid management and other medications with potential influence on SI values. Despite this, we believe that the adjustments for clinical variables and the repeated measures analysis through labour and the postpartum period ameliorated biases associated with these differences in clinical practice. Another limitation was that almost all our patients were under the epidural technique; hence, the variation in SI observed should be interpreted cautiously in patients under NLA administered via different techniques. Other variables, such as top-up interventions for inadequate analgesia, duration of neuraxial analgesia, use of uterotonics other than oxytocin, and volume of blood loss, were not collected; therefore, adjustments for these possible confounders were not possible.
We did not consider the amount of blood loss in our analysis because our study did not focus on patients with PPH. However, we categorised patients according to the presence of PPH, without differences between the groups.
The strengths of this study are its prospective collection of data in different phases of the postpartum period for two groups (with and without NLA). Future studies should include a larger sample of patients with PPH and consider other potential confounders to establish reliable cut-off values for SI values.