Five-Year Trends in Obstetric Hemorrhage, Hemostatic Management, and Blood Product Utilization at a Tertiary Maternity Center in Kazakhstan: A Retrospective Time-Series Study.
Ayazbekov Ardak A, Kurmanova Almagul A, Khaidarov Saken S, Terlikbayeva Aigul A et al.
Background/Objectives: Obstetric hemorrhage remains a leading cause of maternal morbidity and mortality. We examined five-year institutional trends in the recorded hemorrhage rate, cesarean delivery, uterus-preserving hemostatic procedures, uterotonic and antifibrinolytic use, and blood product consumption at a high-volume tertiary maternity center in Kazakhstan. Methods: This was a retrospective time-series analysis of aggregated annual and monthly records for 2018-2022 (46,554 deliveries). The outcome was hemorrhage as entered on the hemorrhage line of the institutional register against the national threshold of 500 mL after vaginal birth and 1000 mL after cesarean; measured blood loss and severity grade were not recorded, so the study describes incidence rather than severity, and blood product figures are hospital-wide adult transfusion volumes. We calculated rates per 1000 deliveries with Wilson confidence intervals, fitted Poisson regression with delivery volume as an offset to estimate incidence rate ratios (IRRs), compared recorded hemorrhage by delivery route with Fisher's exact test, screened for seasonality and out-of-limit months with process control charts, and correlated the monthly hemorrhage rate with service-use variables before and after detrending, with false-discovery-rate correction. Results: The recorded hemorrhage rate fell from 11.39 to 3.62 per 1000 deliveries (rate ratio 0.318, 95% CI 0.217-0.466; 68.2% relative reduction), an average decline of 27.6% per year (IRR 0.724, 95% CI 0.665-0.788, p < 0.001). No reproducible seasonal pattern was found. The cesarean rate declined from 24.90% to 19.36%. Carbetocin and tranexamic acid use rose several-fold, while red cell volume per delivery fell by about 48%. Raw inverse correlations between medication use and hemorrhage were explained by the shared time trend and did not survive detrending and correction. Conclusions: The center recorded a substantial, statistically robust decline in recorded obstetric hemorrhage alongside expanded pharmacologic prophylaxis and reduced transfusion. Because the data are aggregate and neither placental pathology nor other patient case mix could be measured, the parallel trends cannot show that any intervention caused the fall, and a shift in documentation practice cannot be excluded. The findings support prospective monitoring and patient-level study. Because severe hemorrhage and the surgery used to control it may compromise later fertility, the reproductive outcomes of these women are a priority for further research.