Objective: To evaluate the prognostic significance of preoperative DNA flow cytometry compared with other clinical and histo­logic variables in cervical carcinoma. Study design: Sixty-four patients with FIGO Stage lb-III cervical cancer treated with radical abdominal hysterectomy and syste­matic pelvic lymphadenectomy were analyzed. The mean follow-up was 3.4 (range 0.3-9.8) years. DNA flow cytometry was per­formed with fresh tumor tissue. Four biopsies were recut from the surgical specimen within 30 minutes of the operation. The ecto­cervix was divided into four quadrants and a specimen obtained from each. DNA-low-grade tumors (diploid, near-diploid, tetraploid and near-tetraploid) were distinguished from DNA-high-grade tumors (aneuploid and hypoploid). Carcinomas with more than one non-diploid stem line were considered heterogeneous. An S phase fraction >7% was classified as low, 7% -<14% as moderate, and 2:14 as high. DNA ploidy, DNA heterogeneity, S phase fraction and various clinical and histological variables were related to disease-free survival. Results: In the univariate analysis patients with DNA-low-grade carcinomas had significantly better disease-free survival than patients with DNA-high-grade tumors (82% vs 45%, p = 0.021). Carcinomas with an S-phase fraction< 7% were associated with better disease-free survival (0.8) than those with an S-phase fraction 7% - >14% (0.62) and those with ≥14% (0.64), but this was not statistically significant. Cox stepwise regression analysis showed DNA-heterogeneity, age, grade, parametrial involvement and extrapelvic metastasis to be independent prognostic factors. Conclusion: DNA ploidy and DNA heterogeneity are of prognostic importance in cervical cancer. DNA flow cytometry may be used preoperatively to identify low-risk and high-risk patients within a given stage.