The wide prognostic variability of myelodysplastic syndromes (MDS) complicates decision-making regarding the choice and evaluation of alternative treatments to transfusional and antiinfectious supportive measures. Due to its simplicity the Bournemouth scoring system seems to have achieved wide acceptance for establishing the prognosis in MDS patients. The aims of this study were to examine the Bournemouth system in a series of 370 patients with MDS and to evaluate the capability of the prognostic index recently proposed by our group to better define the outcome predicted by the former. The Bournemouth scoring system identified 3 risk groups, A (0-1 points), B (2-3 points) and C (4 points), in the whole series (p less than 0.0001) and it allowed us to stratify refractory anemia (RA), RA with excess of blasts (RAEB) and RAEB in transformation (RAEB-t) patients into two distinct prognostic groups (p = 0.03 and p = 0.01, respectively). This scoring system did not show a significant value in RA with ringed sideroblasts (RARS) and chronic myelomonocytic leukemia (CMML) patients (p greater than 0.05). Our prognostic index clearly segregated patients in the whole series into low- (0-1 points), intermediate- (2-3 points) and high-risk (4-5 points) groups (p less than 0.00001) as well as stratifying ARSA (p = 0.0005), CMML (p less than 0.0001) and RAEB and RAEB-t patients (p less than 0.00001) into different prognostic subset, although it failed to demonstrate a significant predictive value in RA patients.(ABSTRACT TRUNCATED AT 250 WORDS)