This secondary analysis of the data from the Predict-PPH study determined the incidence of severe postpartum haemorrhage (PPH) and its associated antepartum risk factors in anaemic pregnant women in health institutions in the metropolitan city of Lagos, Nigeria, a country with one of the highest burdens of maternal mortality worldwide (13). We recorded that about one in 13 anaemic pregnant women enrolled in the study developed Severe PPH during childbirth. In addition, the identified independent risk factors of sPPH were obesity, antepartum bleeding in the current pregnancy, co-existing uterine fibroids in pregnancy, delivery gestational age beyond 38 weeks, and caesarean birth in the current pregnancy.
The incidence of 7.4% for severe PPH in anaemic pregnant women recorded in our study is much higher than the previously reported modest rates of 1.2% in Uganda and United States (16,17), 1.4% in Australia (18), 1.6% in China (19), and 2.1% in Japan (20) but only slightly higher than the range of 3.9% and 4.5% reported in nationwide studies conducted in two European countries of Netherlands (21) and United Kingdom (22) respectively. The observed disparities in the findings between our study and these previous studies can be attributed to several potential factors. First, unlike these previous studies that were conducted among the general pregnant women population, our study exclusively focused on women affected by anaemia in pregnancy which has been linked extensively with an elevated risk of PPH within and outside sub-Saharan Africa (4,9,23). In contrast, the previous studies did not specifically target this subset of pregnant women. Secondly, our study employed a more precise and objective measurement of postpartum blood loss using the calibrated blood collection V-drape receptacle rather than the subjective and less accurate blood loss estimation by visual inspection (24,25). Thirdly, the variation in the incidence of severe PPH may also stem from differences in diagnostic criteria, medical interventions, and healthcare systems (26). For instance, the lack of sufficiently available essential treatments such as heat-stable carbetocin, tranexamic acid, fresh frozen plasma, and platelet concentrates in our setting (27) could contribute to an elevated incidence of severe PPH. Additionally, our study was conducted in referral hospital settings, mostly patronized by women with more complex and high-risk pregnancies. This is further corroborated by a recent nationwide retrospective study in Japan, which demonstrated an increasing trend in PPH cases managed in large referral hospitals over time (28).
The five antepartum risk factors of severe PPH, starting in order of importance in our study, were delivery by caesarean birth in the current pregnancy, co-existing uterine fibroids in pregnancy, maternal obesity, antepartum haemorrhage (APH) in the index pregnancy, and advanced delivery gestational age beyond 38 weeks. Compared to vaginal birth, the risk of severe PPH is 16.8 times higher with caesarean delivery. This corroborated the findings by Ekin et al. in Turkey (29), Liu et al. in China (19), and Davey et al. in Australia (18), who reported a similar link between caesarean births and the risk of severe PPH. This is mostly because severe PPH is represented as a large volume of blood loss at caesarean section (30). Secondly, at caesarean delivery, there is an increased tendency for more accurate measurement of blood loss (30), usually due to intensive monitoring by a combination of experts, including anaesthetists and obstetricians. Lastly, women undergoing caesarean delivery are likely to experience more blood loss as the standard physiological mechanism involved in uterine contraction and retraction is usually disturbed during lower-segment caesarean section (31,32).
Uterine fibroids are the most prevalent benign growths in women of reproductive age, with a higher frequency among individuals of African descent (33). More than one in 10 pregnant women in the current study had co-existing uterine fibroids which is similar to the estimated 11% prevalence previously reported by Laughlin et al. in a predominantly African-American pregnant women population in four metropolitan areas in three states (North Carolina, Tennessee, and Texas) of the United States (34). In our current study, the risk of severe PPH in pregnant women with uterine fibroids is six-fold higher than in women without fibroids and this is even greater than the 20% higher risk among women with uterine fibroids in a 2011 large hospital-based retrospective cross-sectional study in mainland China (35). These findings are, however, at variance to that of other studies in the United States by Roberts et al. in 1999 (36) and Coronado et al. in 2000 (37) that found no increase in the rate of PPH in women with fibroids. The link between fibroids and PPH risk is attributed to the significantly increased risk of adverse pregnancy events associated with PPH, such as abnormal placentation (placental praevia and abruption) and caesarean section (38,39), and the risk of uterine atony due to overdistension of uterine smooth muscle and impaired contractility in during the postpartum period (40).
There is a four-fold higher risk of severe PPH in obese mothers as reported in our study. Although our finding is not consistent with the inverse association seen between maternal obesity and severe PPH by Butwick and other authors in two different studies conducted in 2014 (41) and 2017 (42), it corroborated the findings from two population-wide studies conducted among mothers in both Sweden (43) and Japan (44). The potential mechanisms for this positive link include the impaired uterine response to control bleeding that occurs due to the disruption of hormonal balance including insulin resistance seen in obese women (45). Second, the larger uterine size and volume and more adipose tissue that is associated with obesity, make it difficult for the uterus to contract effectively after childbirth. Also, the increased likelihood of requiring operative interventions such as caesarean sections during childbirth in obese mothers is often associated with a higher risk of PPH compared to vaginal deliveries (18,29).
Women who have APH have a three times higher risk of developing severe PPH. This is consistent with the findings in other previous studies (17,19,30,42) where APH is depicted as an important predictor of severe PPH. This may be due to the link between certain causes of APH, such as placenta previa, abruption and accreta spectrum, which may also be associated with an increased risk of PPH (19). For example, a compromised placenta might not function optimally during pregnancy, potentially leading to bleeding during pregnancy, childbirth and the postpartum period. Similarly, women with APH are likely to be considered for caesarean delivery which has a strong link to severe PPH as we have reported in this study and other prior studies. Gestational age may be an important and underappreciated risk factor for PPH (46) as corroborated in this current study where we recorded a slight 23% adjusted higher risk of PPH in women at or beyond 39 weeks gestational age. Nonetheless, in a large international retrospective cohort study involving women who underwent hospital-based delivery in California (United States) and Sweden, the highest risks of PPH were documented during the 41 to 42 weeks of gestation (46). While earlier research has observed reduced contractility in the myometrium of women experiencing post-term pregnancies (47), further investigations are necessary to delve into the interconnection between gestational age at delivery, myometrial contractility, altered placentation, and other factors contributing to severe PPH.
This study has the following strengths. First, it is the first prospective study, to our knowledge, that addressed the risk factors of severe PPH specifically among women with prenatal anaemia in Nigeria. Second, we determined prenatal haemoglobin concentration in the study using capillary haemoglobinometry (HemoCue® B-Hemoglobin system), which allows for accurate and almost similar measurements as the reference laboratory standard (48). Third, we used the calibrated blood collection V-drape receptacle that measured postpartum blood loss more precisely and objectively compared to visual blood loss estimation (24). Our study has a few limitations. First, the study findings are only generalizable to clinical settings, as the participating pregnant women were enrolled in the hospitals, thus excluding a major proportion of women in low-income settings who deliver mostly at home. Second, although we analyzed a large sample size of anaemic pregnant women enrolled in the Predict-PPH study, our study may not have adequate power to detect all the possible risk factors of severe PPH.