We conducted a cross-sectional study to determine whether occupational exposure to low levels of ethylene oxide can cause hematological abnormalities. Blood samples were collected from a group of 47 hospital workers who were exposed to ethylene oxide during a mean period of 6.6 years (standard error, 1.1). Ethylene oxide range levels measured were <0.01 to 0.06 ppm. The control group, individually matched by age, sex, and smoking habits, consisted of 88 workers from the administrative sector. We found significant differences between the exposed and the control group in the frequency of workers with white blood cells lower than the normal range. Although there was no significant difference in the absolute mean number of the total white blood cells, we found an elevation in the absolute mean number of monocytes and eosinophils (P< 0.01) and a decrease (P< 0.01) in the absolute mean number of lymphocytes in the exposed group compared with the control group. We also found an elevation (P< 0.01) in the percentage of hematocrit and the mean absolute number of the red blood cells, and a decrease (P< 0.01) in the mean absolute number of platelets, in the exposed group compared with the control group. The mean absolute number of eosinophils, red blood cells, and percentage of hematocrit was significantly higher, and the mean absolute number of lymphocytes and platelets was significantly lower, in the subgroups with a higher cumulative dose of exposure. A positive dose-response was found between cumulative dose exposure and the absolute mean number of eosinophils. In view of our findings, we suggest that the use of complete blood cells with differential in routine medical surveillance and for early detection of hygiene problems should be reexamined with special attention to the eosinophils count.