Comparison of safety and efficacy of liberal versus restrictive red blood cell transfusion thresholds on the quality of life in patients with myelodysplastic syndromes: a systematic review and meta-analysis

TL;DR

A review finds that liberal red blood cell transfusion strategies may improve quality of life for MDS patients but increase transfusion use and iron overload risks.

Contribution

This study provides the first systematic review and meta-analysis comparing liberal versus restrictive transfusion thresholds in MDS patients.

Findings

Liberal transfusion strategies improved quality of life compared to restrictive approaches (pooled effect size 0.54).

Liberal transfusion required 4 additional RBC units per patient and increased ferritin levels by 868 µg/L.

No significant differences were found in mortality or transfusion reactions between the two strategies.

Abstract

Myelodysplastic syndromes (MDS) are clonal stem cell disorders managed by risk stratification: lower-risk disease receives erythropoiesis-stimulating agents; higher-risk disease receives azacitidine. Red blood cell (RBC) transfusions manage symptomatic anaemia and improve quality of life (QoL) but carry risks of iron overload and alloimmunisation. No standardised transfusion strategy exists, requiring systematic evidence synthesis comparing liberal versus restrictive haemoglobin (Hb) thresholds for their effects on quality of life and transfusion-related complications. We performed a systematic review evaluating liberal versus restrictive RBC transfusion thresholds in adults with MDS not undergoing curative treatment such as stem cell transplantation. Primary outcome was health-related QoL measured by validated instruments. Secondary outcomes included mortality, transfusion reaction, iron overload, RBC utilisation and rise in ferritin level. Of 4,295 records screened, 212 articles underwent full-text review, with three RCTs meeting inclusion criteria. Liberal transfusion strategies were associated with improved QoL compared to restrictive approaches, with pooled standardised mean difference (Hedges g) of 0.54 (95% CI 0.06–1.02; p = 0.33; I²=9.7%). Subgroup analysis of EQ‑5D outcomes across studies revealed minimal but statistically significant difference between strategies (pooled mean difference 0.084; 95% CI: 0.033 to 0.134). Mortality (hazard ratio 0.913; 95% CI 0.167–4.98) and transfusion reactions (risk difference − 0.01; 95% CI-0.10 to 0.09) did not differ. Notably, liberal thresholds required average of 4 additional RBC units per patient (95% CI 1.43–6.79), raising concerns about iron overload (rise in ferritin; mean difference 868 µg/L; 95% CI 482–1255). Limited evidence suggests that liberal transfusion strategies in MDS may improve short-term QoL but at the cost of increased transfusion burden, with no clear impact on survival or safety. A patient-centred, multidisciplinary approach remains essential when tailoring transfusion thresholds.

The study protocol has been registered in PROSPERO (CRD420251085221). https://www.crd.york.ac.uk/PROSPERO/view/CRD420251085221.

The online version contains supplementary material available at 10.1007/s00277-026-06789-5.

Figures