Prognostic parameters in myelodysplastic syndromes: a multiple regression analysis

Eur J Haematol. 1988 Feb;40(2):158-62. doi: 10.1111/j.1600-0609.1988.tb00814.x.

Abstract

The influence on survival of 21 basic clinical and hematologic parameters was evaluated in 72 patients with previously untreated myelodysplastic syndromes (MDS). Only five parameters were significant by both survival curves and multiple regression analyses: hemoglobin level, bone marrow (BM) cellularity (estimated from trephine BM biopsies), BM blast percentage, age and BM erythro/myeloid (E/M) ratio. Using these parameters, multiple regression analysis enabled us to predict 34% of the survival of all MDS patients (p less than 0.002), 38% of that of patients who had stable disease (p less than 0.04) and over 80% of that of patients who developed acute leukemia (p less than 0.02). High BM cellularity was the most predictive factor for the development of leukemia. No factor was predictive for patients who died of cytopenic or other complications.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Humans
  • Myelodysplastic Syndromes / blood
  • Myelodysplastic Syndromes / mortality
  • Myelodysplastic Syndromes / physiopathology*
  • Prognosis
  • Regression Analysis
  • Time Factors