All but one (Goldberg et al., 1990) of the studies reviewed by the committee and identified below in Table 6.46 found consistently positive odds ratios for the association between MDS and exposure to unspecified mixtures of organic solvents. Given that all the studies reviewed are case-control studies, almost all relied exclusively on self-reported exposures, which may be subject to recall bias. The difficulty in classifying MDS accurately and consistently is another limitation that the committee considered in evaluating the literature.

For exposure to specific solvents, such as benzene, the rarity of the disease and the difficulty in classifying MDS as a separate entity have limited the evaluation of associations in most studies. Additional studies are needed to elucidate and support the association between exposure to benzene and MDS. They should also attempt to separate MDS from other cancers, such as ANLL, so that the relationship between exposure to benzene and MDS can be further understood. Table 6.46 identifies the studies related to MDS and solvent exposure. Unless indicated in the table, the study populations include both men and women.