Blood Donor Age, Sex Do Not Affect Outcomes After All

Diana Swift

April 27, 2017

Blood donor age and sex are not associated with recipient survival, and are therefore not relevant to donated blood allocation, according to a Scandinavian study of almost 1 million transfusions.

The new findings, published online April 24 in JAMA Internal Medicine, contrast with those from a 2016 Canadian study. That study, which analyzed outcomes for more than 30,000 transfusion recipients, suggested blood from young or female donors was significantly associated with an increased risk for death among transfusion recipients compared with blood from older or male donors.

The Scandinavian researchers sought to replicate these findings, but after meticulously adjusting for the total number of transfusions, they found the association was eliminated.

"Even among the patients who received multiple units of blood from very young or very old donors, absolute mortality differences compared with patients who received no such units of blood were consistently below 0.5%," write Gustaf Edgren, MD, PhD, from the Department of Medical Epidemiology and Biostatistics at the Karolinska Institutet in Stockholm, Sweden, and colleagues. They attribute the earlier findings of increased risk by donor age and sex not to biological effects but, rather, to the result of residual confounding from incomplete adjustment for the total number of transfusions.

Using the same analytic methods as the Canadian group, Dr Edgren and colleagues studied Swedish (58.8%) and Danish recipients included in the SCANDAT2 database who were transfused at least once with red blood cells between 2003 and 2012. The median age of patients at first transfusion was 73 years, and 58.6% were women.

Whereas crude analysis replicated some of the Canadian findings of age- and sex-linked risk, after careful adjustment for the total number of transfusions, neither donor age nor sex showed an association with patient mortality.

Specifically, the initial analysis showed a U-shaped relationship between donor age and recipient mortality, with the lowest occurrence in recipients of blood from the most common donor age group (40 - 49 years) and significant and increasing hazard ratios (HR) in recipients of blood from donors of the youngest and oldest (the least common) age groups. For example, with blood from donors younger than 20 years, the crude HR was 1.12 (95% confidence interval [CI], 1.10 - 1.14), and with that from donors aged 70 years and older, it was 1.25 (95% CI, 1.08 - 1.44).

Higher mortality risk also initially emerged in those receiving blood from female donors (HR, 1.07, 95% CI, 1.07 - 1.07).

Adjusting for the number of transfusions with a linear term, the researchers found the associations were attenuated but still present. Closer examination of the association between number of transfusions and survival, however, revealed a nonlinear pattern, and after additional statistical adjustment to account for this nonlinearity, donor age and sex were no longer associated with recipient mortality.

The fully adjusted HR per transfusion from a donor younger than age 20 years fell to 0.98 (95% CI, 0.96 - 1.00), whereas the per unit HR for blood from a female donor fell to 0.99 (95% CI, 0.99 - 1.00).

"Assuming our approach is correct, these findings indicate that, with regard to recipient outcomes, neither of these donor characteristics need be considered when allocating red blood cell units for transfusion," Dr Edgren and coauthors write. " In addition, we believe these data reinforce the importance of extreme caution in assessing epidemiologic analyses in this field, given the tremendous clinical and logistical implications of false-positive findings."

Responding to the divergent findings, Michaël Chassé, MD, from Laval University in Quebec City, Quebec, who was lead author of the 2016 Canadian study, conceded that there may be residual confounding in his analysis. "Which statistical model is the best and represents the truth remains unclear and supports the need for prospective randomized confirmation," he told Medscape Medical News.

"It is also possible that the included population and red blood cell components used are different in the different cohorts, so one sees an effect in one cohort but not in another." He points out that the Canadian cohort had a significantly higher proportion of donors younger than 20 years (5.9% vs 1.9%). "This provides more power to detect a difference in the youngest strata," he said.

After two reanalyses, including one using the Scandinavian strategy, the Canadian group's conclusions remained largely unchanged, he added.

In an invited editorial, Nareg Roubinian, MD, MPHTM, from the Blood Systems Research Institute in San Francisco, California, and colleagues found the study results reassuring for current practice and the authors' explanation for their negative findings plausible. "They provide a convincing argument that differences in the statistical approach for controlling confounding by the total number of transfused components likely explained the divergent results of the 2 studies," Dr Roubinian and coauthors write.

They suggest that the researchers could have controlled for confounding with the simpler approach of analyzing only patients who had received a single category of blood product (eg, only from female donors), but this method would exclude patients with mixed exposures and would discount the dose effect.

According to the commentators, the disparate findings between the two studies underline the need for rigorous statistical methods in observational research on transfusion. "This conclusion also highlights the challenges and limitations of interpreting novel findings discovered by mining observational data rather than testing a priori hypotheses derived from biological mechanisms," they write.

They also note that Dr Chassé and colleagues provided the Scandinavian team with methodological information that aided their analysis, "exemplifying the true spirit of scientific collaboration."

Yet even with collaborations and improvements in statistical methods, Dr Roubinian and colleagues conclude that "rigorous attention is needed to avoid taking the wrong pathway. In the end, we are optimistic that future efforts to understand health outcomes associated with blood transfusion will benefit from these types of studies."

This study was supported by the Swedish Research Council, the Swedish Heart-Lung Foundation, the Swedish Society for Medical Research, the Karolinska Institutet, and the Danish Council for Independent Research. The authors and the commentators have disclosed no relevant financial relationships.

JAMA Intern Med. Published online April 24, 2017. Article abstract, Editorial extract

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: