The role of volume weighted mean nuclear volume in predicting the prognosis of patients with primary transitional cell carcinoma of the upper urinary tract: a report of 102 new cases.

PURPOSE

We compare estimates of volume weighted mean nuclear volume (MNV) with histological grading to determine the prognosis of primary transitional cell carcinoma of the upper urinary tract using a Cox proportional hazards model.

MATERIALS AND METHODS

We retrospectively reviewed 102 patients who underwent nephroureterectomy for primary transitional cell carcinoma of the upper urinary tract at our hospital between April 1981 and March 1997. Traditional prognostic factors, such as patient age, sex, stage and grade, multiplicity and unbiased estimates of MNV were analyzed with respect to disease recurrence and survival.

RESULTS

Estimates of mean nuclear volume were significantly larger for patients with than without lymph node metastasis (p = 0.0031). No prognostic factor significantly correlated with recurrence of transitional cell carcinoma of the bladder. For pTxN0M0 cases univariate analysis revealed that histological grade (p = 0.0018), pathological T stage (p = 0.0030) and estimates of MNV (p = 0.0001) correlated significantly with disease specific survival, and multivariate stepwise regression analysis revealed that estimate of MNV was the only powerful predictor of prognosis (p = 0.0007).

CONCLUSIONS

Our results indicate that estimate of MNV is an important predictor of prognosis for transitional cell carcinoma of the upper urinary tract. We recommend MNV estimate as a supportive method for subjective histological grading.