Red blood cell transfusions from female, previously pregnant or sex-discordant donors did not appear to increase risk for mortality among transfusion recipients, according to results of a retrospective study published in JAMA.
“NHLBI is charged with conducting and supporting research that improves the blood supply and outcomes in transfused recipients,” Simone Glynn, MD, MPH, chief of the blood epidemiology and clinical therapeutics branch at NHLBI, said in a press release. “The results are reassuring in that the survival of patients who got transfused with red blood cells does not appear to be associated with whether the blood they received was donated by a man, by a woman who had been pregnant — or by one who had not. That’s important to know.”
Previous studies of whether blood donor age, sex and pregnancy history affect mortality of transfusion recipients have yielded conflicting results. Plasma transfusions from women with a history of pregnancy contain antibodies the women develop when exposed to fetal blood, and this plasma has been linked to incidence of transfusion-related acute lung injury.
In this retrospective study, Glynn and colleagues analyzed transfusion recipients from three separate cohorts: Kaiser Permanente Northern California (KPNC; n = 34,662; mean age, 69 years; 54% women), Recipient Epidemiology and Donor Evaluation Study-III (REDS-III; n = 93,724; mean age, 61 years; 52% women), and Scandinavian Donations and Transfusions database (SCANDAT; n = 918,996; mean age, 72 years; 57% women).
In the KPNC cohort, 9% of transfusions were from previously pregnant women, 39% were from female donors and 44% were from sex-discordant donors. These percentages appeared higher in the REDS-III (18% previously pregnant; 43% women; 49% sex discordant) and SCANDAT (25% parous; 41% women; 50% sex discordant) cohorts.
Median number of red blood cell transfusions per patient was three (interquartile range [IQR], 2-5) in the KPNC and SCANDAT cohorts and two (IQR, 0-2) in the REDS-III cohort.
Results showed 3,217 in-hospital deaths in the KPNC cohort, 8,519 in the REDS-III cohort, and 198,537 in the SCANDAT cohort.
Stratified Cox regression models showed no statistically significant associations between in-hospital mortality and transfusions from:
- female donors (KPNC: HR = 0.99; 95% CI, 0.96-1.03; REDS-III: HR = 1; 95% CI, 0.99-1.01; SCANDAT: HR = 1; 95% CI, 0.99-1);
- previously pregnant or parous donors (KPNC: HR = 1; 95% CI, 1-1.01; REDS-III: HR = 1.01; 95% CI, 0.98-1.03; SCANDAT: HR = 1; 95% CI, 1-1.01); and
- PAGE BREAK
- sex-discordant donors (KPNC: HR = 1.02; 95% CI, 0.99-1.05; REDS-III: HR = 0.99; 95% CI, 0.98-1; SCANDAT: HR = 1; 95% CI, 0.99-1).
The retrospective nature of the analysis, as well as a lack of data on adverse events or longer hospital stays, served as the study’s main limitations.
“We proactively address potential risks to the blood supply, and we take this seriously,” Gustaf Edgren, MD, PhD, senior researcher in the department of medicine, Solna, at Karolinska Institute in Stockholm, said in a press release. “Transfusions are very common procedures, and our findings ensure that the current practice is safe and doesn’t need to be changed.” – by John DeRosier
Disclosures: Glynn and Edgren report no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.