The present meta-analysis was undertaken to (1) assess erythroid response rates in myelodysplastic syndromes (MDS) patients treated with epoetin alfa as a monotherapy, (2) gain further insights into predictors of response rates, and (3) compare the erythroid response rates observed with epoetin alfa and darbepoetin alfa. A systematic review of studies from 1990 to 2006 in MDS patients treated with epoetin alfa or darbepoetin alfa was performed and yielded 30 studies evaluating a total of 1,314 patients (epoetin alfa: 22 studies, 925 patients; darbepoetin alfa: eight studies, 389 patients). Pooled estimates of erythroid response rates, stratified by the International Working Group criteria (IWGc) and treatment group, were calculated using random-effects meta-analysis methods. Univariate meta-regression analyses were further conducted to identify study characteristics associated with erythroid response rate. The pooled estimate of erythroid response rate was significantly higher for epoetin alfa IWGc studies (57.6%) as compared to non-IWGc studies (31.6%; p < 0.001). Study factors predictive of higher response rate in the epoetin alfa IWGc studies included higher proportion of patients with RA/RARS (p < 0.001), lower mean baseline serum erythropoietin level (p = 0.007), and fixed dosing regimen (p < 0.001). There was no significant difference in the pooled erythroid response rates between the two agents (epoetin alfa: 57.6% vs. darbepoetin alfa: 59.4%; p = 0.828). The current study reported significantly higher erythroid response rates predominantly in the more recent studies that primarily utilized IWGc to define response. With the use of standardized patient selection and response evaluation methods, epoetin alfa and darbepoetin alfa yielded comparable erythroid response rates in MDS patients.