Using preoperative albumin levels as a surrogate marker for outcomes after radical cystectomy for bladder cancer

Abstract

Objective: To evaluate preoperative albumin levels as a marker for comparing survival outcomes after cystectomy in patients with bladder cancer.

Materials and methods: We performed a retrospective record review using our bladder cancer database of 238 patients from 2004 to 2011. Of these, we included 187 patients with sufficient data for analysis, aged 35 years or older, who survived to undergo cystectomy. Serum albumin levels were routinely checked the day before cystectomy. Overall survival and cancer-specific survival by albumin levels were compared using Kaplan-Meier and Cox proportional hazards regression models. Complication rates between albumin groups were compared by a 2-sample test of proportions.

Results: Thirty-one patients (16.5%) were in the low-albumin cohort (defined as albumin <3.5 g/dL), and 156 patients had albumin levels within normal reference ranges. Multivariable analysis showed overall survival at 3 years was 41% and 56% (adjusted hazard ratio, 1.76; P = .04) and cancer-specific survival was 57% and 72% (hazard ratio, 1.57; P = .22) in the low- and normal-albumin groups, respectively. Overall complication rates were significantly higher in the cohort with low albumin than in those with normal albumin (87% vs 65%; P = .014).

Conclusion: Our single-institution retrospective study demonstrates that patients with low preoperative albumin levels had an increased overall mortality and cancer-specific mortality risk than those with normal albumin levels. Albumin may therefore be a reflection of disease state as well as nutritional status.