Purpose: Surgery is the most effective treatment for renal cell carcinoma with tumor thrombus but predictors of outcome and patient survival are variable. Co-morbidity may affect therapeutic decision making and survival, although to our knowledge this factor has not been studied in patients with tumor thrombus. We analyzed the Charlson co-morbidity index as a predictor of outcome after surgery. Material and Methods: From 1970 to 1998, 303 patients underwent surgical resection. The Charlson index, surgical era, completeness of resection, patient age, sex, tumor level, TNM stage, grade and perinephric fat invasion were studied retrospectively as univariate and multivariate predictors of outcome. Results: The level of tumor thrombus was 0 (renal vein only) in 127 patients, and I to IV in 66, 58, 36 and 16, respectively. At 5 years overall, cause specific and metastasis-free survival were 32%, 42% and 41%, while at 10 years they were 21%, 32% and 30%, respectively. For the whole cohort significant multivariate predictors of cause specific survival were metastasis (p = 0.0001), grade (p = 0.0001), perinephric fat involvement (p = 0.02) and tumor levels 0 versus I to IV (p = 0.048). The Charlson index did not predict outcome (univariate model p = 0.65). Conclusions: Characteristics of the primary tumor remained the most important predictors of cause specific survival in this cohort. The Charlson index did not predict cause specific survival in this cohort of surgically treated patients. Prospective assessment of co-morbidity in patients treated with surgery versus conservative therapy is warranted.
