Is surveillance of small renal masses safe in the elderly?

Objective: To determine if preoperative variables, including gender, age and tumour size, influence the decision for active surveillance of renal masses, as due to the increasing detection of incidental renal masses within the ageing population there is a need to identify reliable means of selecting patients who require therapy.

Patients and methods: We retrospectively identified all renal masses resected at our institution between 1 December 1999, and 1 October 2005. The size of tumour, patient age and gender were compared between those with and without malignancy on final pathology. The influence of these variables in predicting malignancy, high grade, and high stage were assessed by univariate and multivariate analysis using logistic regression models, with a significance level of P < 0.05. Subsets were analysed for the groups of patients with tumours of ≤ 3 or > 3 cm and those aged ≤ 75 or > 75 years.

Results: Among 466 of 501 patients with evaluable data, univariate analysis showed that both male gender and increasing size positively predicted malignancy (odds ratio 1.13 and 1.40, respectively), but age, treated as a continuous variable, did not. On multivariate analysis both remained independent predictors of malignancy (odds ratio 1.13 and 1.40, respectively). Size was the only independent predictor of high-stage and high-grade disease on both univariate and multivariate analysis. Among 156 patients with tumours of ≤ 3 cm, on multivariate analysis, male gender was only weakly associated with the risk of malignancy, whereas size remained strongly predictive (odds ratio 1.98, P = 0.076; and 2.16, P = 0.015, respectively). Neither male gender, size nor age increased the risk of high-stage or high-grade disease in this cohort. Patients who were aged > 75 years had a greater risk of high-stage disease than those aged < 75 years (odds ratio 2.64, P = 0.008). On multivariate analysis, age > 75 years remained an independent predictor of malignancy and high-stage, along with size (odds ratio 2.75, P = 0.014; and 1.35, P < 0.001).

Conclusions: Increased size of tumour increases the risk of malignancy and the likelihood of high-stage and high-grade disease. Among patients aged > 75 years there was a higher risk of malignancy and high-stage disease than in those aged ≤ 75 years. As such, the decision for observation should not be based upon age alone, and should be approached with caution in patients aged >75 years, particularly for larger lesions.