To better define the role of exposure to myelotoxic agents in the genesis of myelodysplastic syndrome (MDS), we carried out (a) a case-control study for the determination of the relative risk (RR) of developing MDS, including 178 consecutive patients and 178 sex- and age-matched controls: (b) a study of clinicobiological features in MDS arising after occupational exposure to myelotoxic agents and in MDS in 'non-exposed' patients. The definition of the 'exposure' status was based on a predetermined questionnaire, with calculation of an 'exposure' index (hours/day x days/year x years). Cumulative exposure to pesticides or to organic solvents, for >2400 h, was recorded in 48 and 25 MDS patients, respectively, compared to 27 and four controls (P<0.00001; RR 3.74; 95% confidence interval 2.02-5.37). Older age and an excess of refractory anaemia with ringed sideroblasts and refractory anaemia with excess of blasts was noted among 'exposed' MDS-patients (group 1), compared to non-exposed MDS-patients (group 2). 68.3% patients in group 1 had clonal chromosome changes, compared with 43.2% patients in group 2. Complex karyotypes, -7/7q-, -5/5q-, +8, 7p and 17p aberrations were seen more frequently in group 1, whereas a normal karyotype, isolated 5q- or 20q- occurred more frequently in group 2. The association of exposure to myelotoxic agents with older age at presentation and with unfavourable chromosome changes accounted for the shorter survival observed in 'exposed' patients. These data show that occupational exposure to pesticides and organic solvents in our region resulted in an increased RR of developing MDS and that a distinct cytogenetic profile was associated with MDS in 'exposed' patients. These findings provide strong indirect evidence that these agents may play a role in the pathogenesis of MDS, preferentially targeting some of the chromosome regions which are frequently involved in therapy-related myeloid neoplasias.