Previous cytogenetic studies of patients with systemic sclerosis have obtained conflicting results regarding the presence of chromosomal anomalies. We studied 38 patients and 15 controls to determine whether these inconsistencies were due to differences in the subgroups of patients who were studied. Because many patients with systemic sclerosis produce autoantibodies to protein antigens that have been implicated in chromosome structure and function, we further hypothesized that the presence of these autoantibodies might correlate with the presence of chromosomal anomalies. Patients were classified into clinical subgroups based on the extent of their disease. Their sera were assayed for autoantibodies to topoisomerase I and centromere proteins (CENP-A, CENP-B, and CENP-C) by immunoblotting. Cytogenetic analyses for aneuploidy and chromosome breaks were performed. Anticentromere antibody positive (ACA+) patients had significantly more aneuploidy than either ACA negative (ACA-) patients or controls (P = 0.041). Although the patient group, when considered as a whole, had significantly greater aneuploidy than the control group (P < 0.005), patients who were ACA--did not have more aneuploidy than the controls had. Patients with Type I disease (sclerodactyly), the majority of whom were ACA+, also had significantly more aneuploidy than did either the controls or patients with Type III (diffuse) disease, most of whom were ACA- (P < 0.005). ACA+ patients also had more chromatid breaks than the controls had (P < 0.05). The correlation between the presence of ACAs and chromosomal aneuploidy suggests that aneuploidy may be the result of nondisjunction secondary to centromeric dysfunction. In support of this hypothesis, the ACA+ patients who had antibodies to CENP-C exhibited more chromosomal aneuploidy than did either anti-CENP-A or anti-CENP-B positive patients (P < 0.048). Unlike CENP-A and CENP-B, which are present at both functional and inactivated centromeres, CENP-C is present at the kinetochore of functional centromeres.