Postpartum hemorrhage in Sweden : temporal trends and novel approaches for risk assessment

<h4>Background</h4><p dir="ltr">Postpartum hemorrhage (PPH)—in Sweden defined as blood loss >1000 mL after delivery—is the largest contributor to maternal morbidity in high-income countries. Severe cases can require interventions such as hysterectomy or, more commonly, blood transfusions. Additional complications include maternal infection, iron deficiency anemia, fatigue and impaired breast-feeding, all of which can contribute towards a negative childbirth experience. The incidence of PPH has increased in high-income countries over recent decades, with the underlying causes remaining unclear. Up to 40% of postpartum hemorrhages occur in women without known risk factors. Identifying new risk factors and improving the ability to predict women at risk is therefore central for improving the care and management of birthing women.</p><h4>Materials and methods</h4><p dir="ltr">Study I used the Swedish Pregnancy Register to examine the accuracy of diagnostic codes (according to the International Statistical Classification of Diseases and Related Health Problems, 10th edition) for PPH in Sweden. We also investigated whether this accuracy varied by obstetric characteristics. Study II used the Swedish Medical Birth Register to investigate temporal trends in PPH in Sweden, stratified by Robson groups, and whether these changes could be attributed to changes in risk factors. Study III used the Stockholm-Gotland Perinatal Cohort to investigate the relationship between first stage labor duration and PPH. Study IV used the Stockholm-Gotland Perinatal Cohort to examine the accuracy of the Californian Maternal Quality Care Collaborative’s (CMQCC) risk assessment tool for identifying women at risk of PPH in Sweden. Study V used the Swedish Medical Birth Register to examine whether crowding of deliveries was more pronounced in births complicated by PPH, compared to non- PPH births. We further investigated whether this association varied by hospital size, or whether the delivery occurred during summer or on a weekend.</p><h4>Main findings</h4><p dir="ltr">I. The sensitivity for the diagnostic codes was 88.5% (95% confidence interval; CI: 88.2– 88.7) and the specificity was 99.4% (95% CI: 99.4–99.4). The sensitivity for PPH was higher in vaginal (92.9%; 95% CI: 92.6–93.2) than cesarean deliveries (79.0%; 95% CI: 78.4–79.7).</p><p dir="ltr">II. The rate of PPH increased with 34% between 2000 and 2016 (5.4%–7.3%). Adjusting for changes in risk factors did not attenuate this increase in PPH in the overall population or when stratified by Robson group. In the cohort of singleton, term deliveries, with the fetus in vertex presentation, induction of labor: Robson groups 2a (nulliparous) and 4a (parous without prior CD) contributed largely to the overall increase in PPH.</p><p dir="ltr">III. The risk of PPH increased with longer first stage labor duration (adjusted risk ratio for duration ≥12.1 h vs <7.7 h: 1.5, 95% CI: 1.41‒1.66). This risk was partly mediated by a prolonged second stage of labor or cesarean delivery in the second stage.</p><p dir="ltr">IV. The accuracy of the PPH risk stratification tool was poor, with a sensitivity of 53.6%, indicating that near half of all the PPH cases occurred in deliveries that were classified as low-risk.</p><p dir="ltr">V. The risk of PPH was not increased if the delivery ward was crowded around the time of birth (odds ratio; OR: 0.97, 95% CI: 0.95‒0.99). These results were not modified by hospital size or having a weekend or summer delivery.</p><h4>Conclusions</h4><p dir="ltr">Given the demonstrated high accuracy of PPH diagnostic codes, these can be used for population-based studies when blood loss volume is unavailable. The increasing trend in PPH is of concern, particularly given that this trend could not be explained by changes in population characteristics or obstetric management. Whether this increase, in part, could be due to changes in quantification practices or labor management that are not captured in the registers, was not possible to evaluate. The validity of the CMQCC tool for classifying women in Sweden at risk for PPH was low and this tool should not be implemented in a Swedish setting. Our findings suggest that first stage of labor duration should be incorporated into in risk assessments for PPH during labor while there was no indication that delivery ward crowding increased the risk of PPH. These insights have the potential to advance future research on PPH, inform clinical decision-making and improve maternal health outcomes.</p><h3>List of scientific papers</h3><p dir="ltr">I. <b>Ladfors LV</b>, Muraca GM, Butwick A, Edgren G, Stephansson O. Accuracy of postpartum hemorrhage coding in the Swedish Pregnancy Register. Acta Obstet Gynecol Scand. 2021 Feb; 100(2):322-330. <a href="https://doi.org/10.1111/aogs.13994" rel="noreferrer" target="_blank">https://doi.org/10.1111/aogs.13994</a></p><p dir="ltr">II. <b>Ladfors LV</b>, Muraca GM, Zetterqvist J, Butwick AJ, Stephansson O. Postpartum haemorrhage trends in Sweden using the Robson ten group classification system: a population-based cohort study. BJOG. 2022 Mar;129(4):562-571. <a href="https://doi.org/10.1111/1471-0528.16931" rel="noreferrer" target="_blank">https://doi.org/10.1111/1471-0528.16931</a></p><p dir="ltr">III. <b>Ladfors LV</b>, Liu X, Sandström A, Lundborg L, Butwick AJ, Muraca GM, Snowden JM, Ahlberg M, Stephansson O. Risk of postpartum hemorrhage with increasing first stage labor duration. Scientific reports. 2024 Sep; vol. 14,1 22152. <a href="https://doi.org/10.1038/s41598-024-72963-2" rel="noreferrer" target="_blank">https://doi.org/10.1038/s41598-024-72963-2</a></p><p dir="ltr">IV. <b>Ladfors LV</b>, Butwick A, Stephansson O. A validation of The California Maternal Quality Care Collaborative obstetric hemorrhage risk assessment tool in a Swedish population. Am J Obstet Gynecol MFM. 2024 Jan;6(1):101240. <a href="https://doi.org/10.1016/j.ajogmf.2023.101240" rel="noreferrer" target="_blank">https://doi.org/10.1016/j.ajogmf.2023.101240</a></p><p dir="ltr">V. <b>Ladfors LV,</b> Holowko N, Liu C, Lundborg L, Ahlberg M, Granath F, Stephansson O. The relationship between crowding in the delivery ward and the risk of postpartum hemorrhage. Acta Obstet Gynecol Scand. 2025 Jul;104(7):1295-1303. <a href="https://doi.org/10.1111/aogs.15137" rel="noreferrer" target="_blank">https://doi.org/10.1111/aogs.15137</a></p>