Blood in, blood out : aspects of blood transfusions and donations in the realms of obstetric care and beyond

<p dir="ltr">Blood transfusion therapy is a cornerstone of modern obstetric care, commonly administered during childbirth to manage acute hemorrhage. Meanwhile, blood donation represents a vital public health act sustained by healthy individuals, many of whom are women of reproductive age. This thesis examines red-cell transfusion and whole blood donation from multiple perspectives, including clinical patterns, recurrence, long-term outcomes, and the effects of pre- pregnancy donation on maternal and offspring health. It combines four register- based cohort studies using Swedish health data and blood donation data integrated within the SCANDAT3-S database to evaluate how blood, both received and given, intersects with maternal and neonatal health.</p><p dir="ltr">Aims . To describe the patterns and temporal trends of red-cell transfusion use in obstetric care in Sweden.</p><p dir="ltr">· To assess the risk of recurrent transfusion in delivery.</p><p dir="ltr">· To investigate whether red-cell transfusion during childbirth is associated with long-term risks of autoimmune disease and non-Hodgkin lymphoma.</p><p dir="ltr">· To evaluate whether blood donation before pregnancy is associated with adverse maternal or neonatal outcomes.</p><p dir="ltr">Study I. Transfusion practices in obstetric care in Sweden have evolved, but national-level data on usage patterns remain limited. This study analyzed a nationwide cohort of 1.6 million deliveries in Sweden between 2003 and 2017. In addition to describing overall transfusion patterns, we examined temporal trends and volume using logistic regression. For analyses of inter-hospital variations in transfusion rates, we used direct standardization to the year 2003, adjusting for parity and maternal age. Overall, 3 percent of women were transfused annually, and there were no considerable changes over the study period. We saw an increasing proportion of low-volume and a decline in high-volume transfusions. The most common number of red-cell units transfused per delivery was two, reflecting a broader pattern where even-numbered transfusions-particularly two or four units-dominated clinical practice. We saw substantial variation in the proportion of transfused deliveries across hospitals.</p><p dir="ltr">Summary: While overall transfusion rates remained stable, hospital-level variation and the use of paired units may signal inappropriate use and warrant further investigation.</p><p dir="ltr">Study II. Although transfusion during childbirth is typically viewed as an event determined by the specific circumstances of each delivery, some women experience repeated transfusions across deliveries. We analyzed a nationwide cohort of 825,000 women with deliveries between 2000 and 2017 to investigate the recurrence of transfusion. We used logistic regression to estimate the odds ratios of requiring a repeated transfusion in relation to specific risk factors. Women transfused in their first delivery were overall approximately five times more likely to be transfused in their second delivery (adjusted odds ratio [aOR] 5.4, 95% confidence intervals [CI]; 5.0-5.8) than non-transfused, non- hemorrhaged women. The overall odds of repeat transfusion were similar regardless of hemorrhage diagnosis in the first delivery. When considering specific subtypes, we saw the most pronounced recurrence risk associated with prior diagnoses of atony (aOR 6.7; 95% CI 6.1-7.4), placental retention (aOR 4.4; 95% CI, 3.8-5.1), and caesarian section with hemorrhage (aOR 6.8; 95% CI, 5.8- 8.0). Certain conditions in the second pregnancy (e.g., preeclampsia, placenta previa, iron deficiency anemia) increased the odds of recurrence. Having a sister transfused in childbirth was associated with higher odds of transfusion in childbirth (aOR 1.8; 95% CI, 1.5-2.1).</p><p dir="ltr">Summary: A prior transfusion is a strong risk factor for future transfusion, underscoring the importance of individualized risk management.</p><p dir="ltr">Study III. The long-term effects of obstetric transfusion on immune-mediated conditions remain poorly understood. In this study, we followed a large register- based cohort of over one million women, starting in 1987, for up to 25 years to assess the risk of autoimmune disease and non-Hodgkin lymphoma. Cox proportional hazard models were used, adjusting for both fixed and time- dependent maternal factors. To reduce the risk of reverse causation, a six-month lag period was applied before the start of follow-up after each delivery. Transfusion in childbirth was modestly associated with increased risks for later diagnosis of systemic lupus erythematosus (SLE) (adjusted hazard ratio [aHR] 1.38; 95% CI, 1.01-1.87) and systemic sclerosis (aHR 1.89; 95% CI, 1.21-3.21), but not associated with risk of non-Hodgkin lymphoma or rheumatoid arthritis. However, the absolute risk difference at 25 years was slight. Sensitivity analyses, applying various stratifications and restrictions, generally supported the main findings.</p><p dir="ltr">Summary: Although the absolute risk difference was small, transfusion in childbirth was associated with an increased risk of developing SLE and systemic sclerosis.</p><p dir="ltr">Study IV. While blood donors are typically healthier than the general population, frequent donation may deplete iron stores and influence pregnancy outcomes. We analyzed a register-based cohort of 2.5 million deliveries between 1985 and 2017 to investigate any potential adverse effects of blood donations on pregnancy and childbirth. We used mixed-effects linear models to estimate birth weight by donation intensity, and logistic regression was applied to assess maternal and neonatal outcomes. When comparing births to non-donors to those to women who donated at least once within five years of delivery, we found that donors were less likely to experience adverse pregnancy outcomes and that the birth weight of their offspring was slightly higher. However, when considering only donors, high-frequency donors (29 donations within five years of delivery) had lower birth weight babies and higher risks of preeclampsia (aOR 1.18; 95% CI, 1.06-1.31), preterm birth (aOR 1.33; 95% CI, 1.22-1.45), and SGA infants (aOR 1.17; 95% CI, 1.03-1.34) than low-frequency donors (1-2 donations). Specific temporal analyses using interaction terms supported the notion of a biological effect. Sensitivity analyses confirmed the main findings.</p><p dir="ltr">Summary: High-frequency whole blood donation before pregnancy may have adverse effects on maternal health and neonatal outcomes.</p><p dir="ltr">In conclusion, the thesis reveals that transfusion and donation intersect with maternal health. It underscores the need for consistent transfusion practices, monitoring for recurrence risk, awareness of potential long-term immune effects, and cautious iron management in frequent female donors.</p><h3>List of scientific papers</h3><p dir="ltr">I. <b>Brynolf A.</b>, Zhao J., Wikman A., Öberg S., Sandström A., Edgren G. (2021), Patterns of red-cell transfusion use in obstetric practice in Sweden 2003-2017: A nationwide study. Vox Sang, 116: 821-830. Epub 2021 Feb 2. PMID: 33528029<br><a href="https://doi.org/10.1111/vox.13074">https://doi.org/10.1111/vox.13074</a><br><br></p><p dir="ltr">Il. <b>Brynolf A,</b> Sandstrom A, Edgren G. Risk of recurrent red-cell transfusion in delivery: A nationwide longitudinal study. BJOG. 2024 Mar;131(4):455-462. doi: 10.1111/1471-0528.17672. Epub 2023 Sep 25. PMID: 37749750.<br><a href="https://doi.org/10.1111/1471-0528.17672">https://doi.org/10.1111/1471-0528.17672</a><br><br></p><p dir="ltr">III. <b>Brynolf A,</b> Sandström A, Hjalgrim H, Edgren G. Association Between Red-Cell Transfusion in Childbirth and Long-Term Risk of Lymphoma and Autoimmune Disease: A Swedish Nationwide Cohort Study. Am J Hematol. 2025 Apr;100(4):735-739. doi: 10.1002/ajh.27610. Epub 2025 Jan 27. PMID: 39868856.<br><a href="https://doi.org/10.1002/ajh.27610">https://doi.org/10.1002/ajh.27610</a><br><br></p><p dir="ltr">IV. <b>Brynolf A,</b> Sandström A, Öberg S, Toss F, Edgren G. Whole blood donations before delivery and the risk of adverse pregnancy, delivery, and neonatal outcomes: A Swedish longitudinal cohort study. [Manuscript]</p>