Predictors of 1- and 2-year mortality in patients with rectal cancer

Abstract

Aim: Tools are needed to aid in the assessment of the prognosis of patients with rectal cancer regarding the risk of medium-term mortality. The aim of this study was to develop and validate clinical prediction rules for 1- and 2-year mortality in patients undergoing surgery for rectal cancer.

Method: A prospective cohort study of patients diagnosed with rectal cancer who underwent surgery was carried out. The main outcomes were mortality at 1 and 2 years after surgery. Background, clinical parameters and diagnostic test findings were evaluated as possible predictors. Multivariable survival models were used in the statistical analyses.

Results: Predictors of 1-year mortality were being a current smoker [hazard ratio (HR) 4.98], having a Charlson index adjusted by age > 5 (HR 2.61), the presence of vascular, perineural or lymphatic invasion (HR 3.30), the presence of residual tumour at the operation (R-stage) (HR 8.64) and TNM stage (HR for TNM IV 5.10) [concordance index (C-index) 0.799 (95% CI: 0.71-0.89)]. Age greater than 80 years (HR 2.19), being a current smoker (HR 2.20), the pre-intervention haemoglobin level (HR 2.02), need for blood transfusion (HR 2.12), vascular, perineural or lymphatic invasion (HR 2.59), R-stage of the operation (HR 6.13) and TNM stage (HR for TNM IV 4.43) were predictors of 2-year mortality [C-index 0.779 (0.718-0.840)]. Adjuvant chemotherapy was an additional predictor at both outcome durations.

Conclusion: These clinical parameters show good predictive values and are easy and quick-to-use tools to help in clinical decision making.

Keywords: Rectal cancer; clinical prediction rules; mortality; prospective study.