Main Findings
Our study found that APH in the setting of PP has significant implications for mothers intraoperatively and postpartum, including a tendency for preterm emergency caesarean sections under general anaesthesia, with greater blood loss requiring transfusion, and longer postpartum stay.
The incidence of PP in our population was 0.8%, compared to the incidence reported in the current literature at 0.2–0.4%.8,15 This is likely due to the large number of high-risk obstetric referrals received at the study hospital known to be a major tertiary centre.
In our study, the majority (62%) of women with placenta praevia experienced antenatal haemorrhage, which is consistent with a previous study14 and slightly greater than reported (51.6%) within the systematic review and meta-analysis of twenty-nine articles with 4687 individuals by Fan et al.6 Despite excluding all placenta accreta cases, our study examines the largest number of women diagnosed solely with placenta praevia compared to any previous individual article noted by Fan et al.6
We found that maternal characteristics, though may predispose a diagnosis of placenta praevia, made no difference in risk of antepartum bleeding in women with placenta praevia, alike a prior study by Mastrolia et al.16 However, the diagnosis of major PP significantly increases the risk of APH (OR 2.88p <0.001) compared to minor PP. This result is supported by another large retrospective cohort of 306 women by Bahar et al.17 (OR 3.18, 95% CI 1.58–6.4, p =0.001) and similarly reported in a population of 121 women by Bhide et al.18 comparing APH in major versus minor PP (57.1% versus 47.5% p <0.05). Besides the degree of PP, developing research suggests that the likelihood of APH might also be related to the placental edge thickness19 and echo-free space in the lower edge of the placenta.20
The increased risk of preterm delivery in women with PP is well established.21,22 In our study, women with PP who experienced bleeding were significantly more likely to deliver earlier than those without bleeding (median 35.4 versus 38.0 weeks,p <0.001). This finding is consistent with results by Lam et al.,14 who also found that newborns delivered from mothers with APH were smaller and more frequently required nursery admission. Therefore the earlier timing of delivery in placenta praevia seems to correlate with the incident of at least one bleeding episode and degree of PP, which in itself increases the tendency to bleed.
Pivano et al.23 published a scoring system that predicted the risk of emergency caesarean for women with placenta praevia based on the type of PP, frequency and intensity of antenatal bleeding and gestational age at sentinel bleed. In our population, women who had experienced APH were more likely to undergo an emergency caesarean sections than their asymptomatic counterparts for lower uterine segment (61% vs 25%) and classical incisions (8.5% vs 1.4%), which correlates well with significant findings by Love et al.24 (63% vs 25% p <0.001) and Fishman et al.25 (OR 17.7 95% CI 6.1–51.7).
Interestingly, in our study there was a novel finding of three-fold increased risk of undergoing general anaesthesia in women with bleeding that has not been reported before in the literature. This might be explained by the urgency of delivery as suggested by earlier gestations and greater proportion of emergency caesareans in the antepartum bleeding cohort. The use of general anaesthesia in caesarean sections is associated with a greater volume of blood loss than neuroaxial anaesthesia.26
We found that both syntocinon bolus and infusions were used more frequently intra-operatively in situations where bleeding had occurred. Otherwise, there was no difference in the median number of uterotonics used between the compared cohorts. Likewise, besides use of surgicell (OR 3.56, 95% CI 1.45–8.73) and Bakri balloon (OR 10.3, 95% CI 1.35–78.2), there were no differences in additional surgical techniques required for managing bleeding in the APH group versus those without APH. A plausible explanation for the lack of statistical significance in both medical and additional surgical techniques, including hysterectomy, might be the heterogeneity of surgical and anaesthetic experience and individual preferences within the study hospital, which may range from junior obstetric trainees to skilled specialists.
Women with PP who bled antenatally were likely to have significantly greater volume of blood loss postoperatively than their asymptomatic counterparts (IRR 1.20, 95% CI 1.05–1.37), requiring over three-fold the number of blood transfusions. Mastrolia et al.16similarly found a tendency for blood transfusions in the presence of APH, unlike previous studies.14,24
Besides a marginally longer postpartum hospital stay for those who experienced APH, there was no difference in postoperative complications. The current literature has scarcely focused on postpartum maternal outcomes in PP,27,28 and for those that have evaluated this aspect have also found no significant difference between the presence and absence of APH.14