Study suggests link between blood transfusions and brain hemorrhages, but further research needed

In a recent study published in JAMA, researchers investigated whether spontaneous intracerebral hemorrhages (ICHs) among blood donors after blood donation were linked to the development of spontaneous ICH among transfusion recipients.

Background

According to recent research, cerebral amyloid angiopathy (CAA), one of the causes of repeated spontaneous ICH episodes, may be transmitted to humans by parenteral injections of pituitary hormones from contaminated cadavers. CAA has been generated in animal models by intravenous, intraperitoneal, and intracerebral injections of amyloid beta protein.

About the study

In the present retrospective cohort study, researchers determined the link between spontaneous intracerebral hemorrhage among blood donors and the risk of spontaneous intracerebral hemorrhage in those who receive blood transfusions.

The researchers hypothesized that CAA transmission by transfusion may appear as a higher risk of spontaneously occurring intracerebral hemorrhage among transfusion recipients exposed to blood from a donor with spontaneous ICH. The researchers tested the idea by examining the link between the occurrence of spontaneous ICH among blood donors and their receivers. Because CAA-associated ICH has a seven-fold greater risk of recurrence than non-CAA-related ICH, the researchers evaluated multiple episodes of ICH as a more robust proxy for CAA. They examined single and multiple occurrences of ICH independently.

The exploratory study used data provided by national blood banks and health registers of Swedish (primary group) and Danish (validation group) individuals (n=1,089,370) aged five to 80 who received an erythrocyte transfusion between 1 January 1970 (for Swedish individuals) or 1 January 1980 (for Danish individuals) and 31 December 2017. The research subjects were receiving erythrocytic transfusions from donor individuals who later had (i) one episode of spontaneous ICH, (ii) many (two or more episodes more than 30 days apart), or (iii) no episode of spontaneous ICH.

The exposure window was limited to the first six months of transfusions. The development of spontaneous ICH among transfusion recipients was the end measure, with ischemic stroke serving as a negative-type control. Individuals transfused with erythrocytes from donor individuals who had a history of ICH before blood donation were excluded from the study.

Individuals were also excluded if they received a plasma or platelet transfusion, an autologous erythrocyte unit, or an erythrocyte unit from an unidentified donor, a donor with a history of spontaneous ICH before donation, their parent, or their child during the exposure period. The International Classification of Diseases-eighth, ninth, and tenth revisions (ICD-8, 9, and 10) codes were used to identify spontaneous ICH and ischemic stroke. The hazard ratios (HRs) were calculated using Cox proportional hazard modeling after controlling for participant age, gender, blood group, index year, transfusion indication, and number of erythrocytic transfusions.

Results

In total, 329,512 Danish individuals (median age 64; 58% women) and 759,858 Swedish individuals (median age 65; 59% women) were enrolled and followed up for six years (median). In total, 8,598 Swedish and 3,695 Danish individuals were administered erythrocytic units from donors who subsequently presented with an episode of spontaneous intracerebral hemorrhage, while 862 and 448 received transfusions from donors who subsequently presented with several spontaneous ICHs.

Individuals receiving erythrocytic units from donor individuals who developed several episodes of spontaneous intracerebral hemorrhage showed a statistically significantly increased risk of developing one episode of spontaneous intracerebral hemorrhage themselves, in comparison to those who received transfusions from donor individuals who showed no signs of intracerebral hemorrhage, in the Swedish as well as Danish groups [incidence rate (unadjusted), 3.2 versus 1.1 per 1,000 individual years, adjusted HR, 2.7].

Individuals receiving transfusions from donor individuals who developed one spontaneous ICH were not significantly different in Sweden (unadjusted incidence rate, 1.4 versus 1.1 per 1,000 individual years, adjusted HR value, 1.1), nor in Denmark (unadjusted incidence rate, 1.4 versus 1.1 per 1,000 individual years; adjusted HR value, 1.1), nor in ischemic stroke.

In the sensitivity analyses, individuals receiving erythrocytic transfusions from donor individuals later developed two or more episodes of ischemic heart failure (ICH) and were significantly linked to those developing one episode of spontaneous intracerebral hemorrhage in the Swedish group (21 versus 7,686 events; unadjusted incidence rate, 2.6 versus 1.3 per 1,000 individual years; HR, 2.0).

For Danish individuals, the corresponding hazard ratio was comparable but non-significant (10 versus 3,080 events; unadjusted incidence rate, 3.0 versus 1.3 per 1,000 individual years; HR, 1.7). With a delay of five years, the hazard ratios were 1.7 and 2.0 for Swedish and Danish individuals, respectively.

Comparing individuals undergoing erythrocytic transfusions from donor individuals who subsequently developed an episode of spontaneous intracerebral hemorrhage versus individuals undergoing erythrocytic transfusions from donor individuals who showed no signs of ICH, no statistically significant differences were observed in the risk of developing ICH in Swedish transfusion recipients (114 versus 7,686 events; unadjusted incidence rate, 1.6 versus 1.3 per 1,000 individual years; HR, 1.0) and their Danish counterparts (36 versus 3,080 events; unadjusted incidence rate, 1.3 versus 1.3 per 1,000 individual years; HR, 0.8). Multiple spontaneous ICHs among individuals who underwent transfusions were rare.

Conclusion

Overall, the study findings showed that individuals who had erythrocyte transfusions from donor individuals who later got several spontaneous intracerebral hemorrhages were at considerably elevated risk of developing ICH themselves.

The findings may point to an agent transmitted by transfusions linked to spontaneous ICH but are subject to confounding and selection bias. Future studies are required to determine whether CAA transmission via transfusions explains this relationship.