Background: 40-80 % of patients with myelodysplastic syndrome (MDS) become transfusion dependent during their disease course and are at risk for the development of alloimmunization. Red blood cell (RBC) alloantibodies can make finding compatible blood for transfusion more difficult, expensive and time consuming. Allommunization rates of approximately 30-47% have been reported in patients with sickle cell disease and the transfusion of RBCs prophylactically matched for Rh antigens E and C, and K antigens reduced the rate of alloimmunization from 3% to 0.5% per unit (Vichinsky et al, 2001). In 2007, our hospital instituted a policy of transfusing prophylactic Rh and K matched blood to MDS patients. The objectives of this study were to compare the rates of alloimmunization in MDS patients who received prophylactic Rh and K matched blood compared to those that did not and identify potential risk factors for alloimmunization. Methods: 193 Transfusion dependent MDS patients were identified out of 387 patients registered and prospectively followed in a local MDS registry. Transfusion dependence was defined as the receipt of at least 1 unit of PRBC every 8 weeks for a minimum of 16 weeks. Records of transfusions received up to May 1, 2014 were collected from blood bank databases of the hospitals at which patients were transfused. Patients were classified according to whether phenotyping had been performed, the location of transfusions (transfused only at our institution, transfused only at an outside institution or transfused at both sites) and whether prophylactic Rh (E, C antigens) and K matched blood was transfused. Data were descriptively analyzed and we conducted univariate and multivariate logistic regression using p< 0.05 as statistically significant to identify risk factors for alloimmunization.

Results: 176 MDS patients with complete transfusion records are included, 73 transfused at Sunnybrook, 92 transfused in community hospitals and 11 at both. The median age was 72 yrs (range 22-89), 60% were male, and 8%, 43% and 27% had very low, low and intermediate risk R-IPSS scores respectively. Median follow up was 2.9 years (IQR 1.6-5) 3.49 SD). Blood groups O, A, B, AB and O were 45%, 38%, 15% and 2% respectively, while 85% were RhD+. The median time from diagnosis until first transfusion was 4 months (IQR: 0.2-14), with 51 patients having received at least 1 transfusion prior to diagnosis at a median time of 0.9 months. 4.5% had a pre-existing allo-antibody at time of MDS diagnosis. With a median follow up from diagnosis of 3 years (IQR:1.6-5)), the median number of RBC units transfused was 38 (IQR: 15-98)) and 36 (20%) patients developed new alloantibodies (median 2 (IQR (1-2.5) alloantibodies). The median number of RBC units until first allo antibody was 13.5 units (range 0-121) and 1.25 years from diagnosis (95% CI:0.4-2.1). The majority of the alloantibodies were in the Rh (n=28) and K (n=14) groups (80%) and co-existed 27% of the time. More patients transfused at our hospital received prophylactic Rh K matched blood sometimes or always (60% versus 26%) and rates of allo-immunization were decreased by 65% (absolute rate of alloimmunization 10% versus 29%). By multivariate analysis analysis, number of rbc transfused (p<.0001), receiving prophylactic phenotype matched blood (p=.0008) and location of transfusions (Sunnybrook versus elsewhere (p=.03)) were independent risk factors for alloimmunization.

Conclusions: 20-30% of RBC transfusion dependent MDS patients will become allo-immunized to clinically significant blood group antigens, the majority being Rh and K antigens. The practice of phenotyping at baseline and prophylactically transfusing Rh and Kell matched blood decreases rates of alloimmunization up to 65% and should be strongly considered for routine transfusion practice in centres that treat MDS.

Table 1.
All Patients (n=176)Sunnybrook
(n=73)
Community
(n=92)
Ever phenotyped 45% 64% 28% 
Phenotyped before 1st transfusion 20% 38% 8% 
Developed allo-antibodies 20% 10% 29% 
Received prophylactic Rh K matched blood
(developed alloantibodies)
Never
Sometimes
Always 



58% (16%)
24% (42%)
18% (6%) 



40% (3%)
23% (23%)
37% (7%) 



74% (22%)
22% (60%)
4% (0%) 
All Patients (n=176)Sunnybrook
(n=73)
Community
(n=92)
Ever phenotyped 45% 64% 28% 
Phenotyped before 1st transfusion 20% 38% 8% 
Developed allo-antibodies 20% 10% 29% 
Received prophylactic Rh K matched blood
(developed alloantibodies)
Never
Sometimes
Always 



58% (16%)
24% (42%)
18% (6%) 



40% (3%)
23% (23%)
37% (7%) 



74% (22%)
22% (60%)
4% (0%) 

Figure 1.

Allo Antibody free survival according to number of red cell units transfused in patients that developed an allo-antibody

Figure 1.

Allo Antibody free survival according to number of red cell units transfused in patients that developed an allo-antibody

Close modal
Disclosures

Wells:Celgene: Honoraria, Other, Research Funding; Novartis: Honoraria, Research Funding; Alexion: Honoraria, Research Funding. Buckstein:Celgene: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution