Deoxyribonucleic acid (DNA) ploidy image analysis was used postoperatively to predict recurrence of 112 clinically localized adenocarcinomas of the prostate. All men underwent radical retropubic prostatectomy between 1978 and 1991. Patients with positive lymph nodes or positive seminal vesicles were excluded because progression is nearly inevitable in these men. The minimum followup for men without progression was 5 years (range 5 to 15). Patients were considered to have clinically evident disease progression based on local recurrence (8%), distant metastases (4%) and/or an isolated elevation of serum prostate specific antigen (87%). Of the tumors 43% were diploid and 57% were nondiploid. In a multivariate analysis comparing grade, ploidy, capsular penetration and surgical margins, Gleason sum was the best predictor of progression (p < 0.0001). Nevertheless, a subset of patients remained with well to moderately differentiated Gleason grade tumors (Gleason sum 6 or less) who failed. DNA ploidy was able to predict recurrence in this particular group (p = 0.034). In addition, we compared different methods of tissue preparation to determine which best predicted progression. We found that ploidy analysis on tissue sections was more predictive than ploidy performed on disaggregated tissue. In summary, our study revealed that DNA ploidy analysis can offer additional prognostic information following radical prostatectomy for men with low grade prostatic adenocarcinoma.