Previous research has suggested that receipt of a red blood cell (RBC) transfusion from a female donor who has ever been pregnant increases mortality risk among transfusion recipients, but a large study published in JAMA provides conflicting results, finding no association between donor’s sex or pregnancy history and risk of death among recipients.
“While our findings will not particularly affect clinical care directly, as donor sex and parity is not considered when allocating blood units, our results should comfort clinicians that the current practice is safe,” lead author Gustaf Edgren, MD, PhD, of the Karolinska University Hospital in Stockholm, Sweden, told ASH Clinical News.
In this study, the researchers reviewed data from three large cohorts of U.S. and Scandinavian transfusion recipients:
- Kaiser Permanente Northern California (KPNC) from 2013 to 2016 (n=34,662)
- Recipient Epidemiology and Donor Evaluation Study-III (REDS-III) from 2013 to 2016 (n=93,724)
- Scandinavian Donations and Transfusions (SCANDAT) from 2003 to 2012 (n=918,996)
Patient data were analyzed for the following “exposures:” the number of RBC transfusions from female donors, the pregnancy history of female donors (previously pregnant or parous), and whether donations were from sex-discordant donors (i.e., male donor to female recipient or female donor to male recipient).
The investigators then categorized patients according to how many RBC transfusion units they received from donors meeting any of these exposure variables (0, 1-2, 3-4, 5-6, ≥7).
Reference groups, comprising transfusion recipients who received an equal number of RBC units as recipients in other exposure groups, were identified for each exposure group.
The primary outcome of the analysis was in-hospital mortality, which was defined as any death that occurred during the hospitalization in which transfusion took place. A secondary outcome – long-term mortality – was assessed in only the KPNC and SCANDAT cohorts and included any deaths following hospital discharge.
In the KPNC, REDS-III, and SCANDAT cohorts, patients had a median of three (range = 2-5), two (range = 1-5), and three (range = 2-5) RBC transfusions, respectively. The percentage of transfusions for each variable were as follows:
- transfusions from female donors: KPNC (39%), REDS-III (43%), and SCANDAT (41%)
- transfusions from previously pregnant/parous donors: KPNC (9%), REDS-III (18%), and SCANDAT (25%)
- transfusions from sex-discordant donors: KPNC (44%), REDS-III (49%), and SCANDAT (50%)
In-hospital mortality occurred in 3,217 patients in the KPNC cohort, 8,519 patients in the REDS-III cohort, and 198,537 patients in the SCANDAT cohort. Overall, the authors found no associations between in-hospital mortality and the three exposure variables, regardless of the number of RBC units patients received.
However, researchers observed a higher mortality rate in the KPNC cohort in people who had received three to four or five to six RBC units from sex-discordant donors, compared with patients who received no sex-discordant donations (hazard ratios [HRs] = 1.22 [95% CI 1.05-1.41] and 1.31 [95% CI 1.03-1.66]), although p values were not reported. “These findings are not likely related to true biologic effects,” the authors noted, explaining that “similar associations were not observed in either of the two larger cohorts [and] the findings from the KPNC data did not follow a clear dose-response pattern.”
Results were similar in an analyses of long-term mortality risk in the KPNC and SCANDAT cohorts: There were no differences in mortality among patients who received RBC units from a previously pregnant or parous donor (HRs = 1.01 and 1.00, respectively; p values not reported), even among patients who received at least seven RBC units.
According to Dr. Edgren, the findings from this study were informative but not surprising. “For example, red cell units from previously pregnant female donors seemed to increase mortality only among young male patients,” he said. “We found this puzzling, and in a way, this may have meant we didn’t really expect to find a difference in our larger investigation.”
Limitations of the study include its observational design, which may have led to confounding or bias due to unobserved factors. The study also did not control for the disease that led to hospitalization. In addition, the researchers only examined mortality in their analysis, making it “possible that there might have been other negative effects of blood units from donors of different sex that didn’t result in an increased mortality,” Dr. Edgren explained.
The authors reported no conflicts of interest.
Edgren G, Murphy EL, Brambilla DJ, et al. Association of blood donor sex and prior pregnancy with mortality among red blood cell transfusion recipients. JAMA. 2019;321:2183-92.