Between December 1990 and January 1994, bone marrow (BM) samples from 151 patients with multiple myeloma (MM), including 117 patients evaluated at diagnosis, were collected for cytogenetic analysis. A total of 129 patients had assessable metaphases (100 patients at diagnosis). Cytogenetic studies were performed on BM cells after longterm cultures (6 days) with stimulation of cultures by granulocyte-macrophage colony-stimulating factor (GM-CSF), GM-CSF plus interleukin (IL)-6, IL-3 plus IL-6, or GM-CSF plus IL-3 plus IL-6 to improve myeloma cell growth, and 91 patients had an additional unstimulated culture. Sixty-six patients (51%) had cytogenetic abnormalities, including 47 of 100 patients at diagnosis (47%) and 17 of 24 patients at relapse (71%; P = .04). The aberration rate increased with stage (P = .007), BM plasmacytosis (P = .003), beta 2 microglobulin level (P = .001), C-reactive protein (CRP) level (P = .001), and Ki-67 (P = .007). The abnormality detection rate was higher in stimulated than unstimulated cultures, and the difference was statistically significant (P < .01). Hyperdiploidy was observed in 39 patients (30% of patients with an assessable karyotype) and hypodiploidy in 19 patients (15%). Among numeric changes, gains predominantly involved chromosomes 3, 5, 7, 9, 11, 15, 19 and losses, chromosomes 8, 13, 14, and X. The most frequent loss was loss of chromosome 13, observed in 22 patients (15%), including 18 patients at diagnosis (12%). We observed frequent structural changes of chromosomes 1 (15%) and 14 (10%) but also a 5% incidence of 19q13 abnormality and two patients with translocation t(1;16)(p11;p11). By using the proportional hazard univariate model, patients with abnormal karyotypes were demonstrated to have 2.5-fold greater chance of death than patients with normal karyotypes (P < .014). Despite a multivariate approach with the same model, the respective roles of karyotype abnormality, age, stage, and beta 2 microglobulin level could not be clearly ascertained. From these results we conclude that cytogenetic analysis using stimulation of cultures by cytokine(s) may be a promising method to identify about 50% of cytogenetic abnormalities in patients with newly diagnosed MM. Cytogenetic analysis may help to define a high-risk population that would benefit from intensive therapeutic approaches.