We pooled data from three biomonitoring studies using the cytokinesis-block micronucleus assay in peripheral blood lymphocytes in combination with fluorescence in situ hybridization. Centromere-positive micronuclei (C+MN) were classified in two groups: those containing one centromere (C1+MN) and those with two or more (Cx+MN). The three studies evaluated untreated cancer patients, welders, and pathologists/anatomists exposed to formaldehyde. The total number of subjects included in the pooled re-analysis was 113. A higher frequency of C+MN was observed in cancer patients and exposed workers, who showed significant differences from controls in all studies. C1+MN were particularly increased in the group of pathologists/anatomists, who showed a 3.29 times higher frequency than controls (95% CI: 2.04-5.30). A borderline increase in Cx+MN was observed in welders when compared to the corresponding control group (FR: 1.31; 95% CI: 0.99-1.74). An evident effect of gender was found, with significantly increased frequencies of all endpoints measuring aneuploidy in females (C+MN, C1+MN, and Cx+MN). Alcohol consumption had a significant effect on total MN frequency and particularly on C+MN and C1+MN. In conclusion, scoring the number of centromeric signals in the micronucleus assay provides additional information about the mechanism of action of various genotoxic agents, and the role of confounding factors may be more specifically accounted for. Indeed, C+MN could be efficiently used in biomonitoring studies as an independent biomarker of exposure and early biological effect. The use of centromeric signals allows the identification of two further endpoints, representing two alternative pathways of chromosome loss, i.e., impaired chromosome migration, leading to increased C1+MN frequency, and centrosome amplification, possibly leading to Cx+MN with two or more centromeric signals.