Assessing interacting effects of albumin and phosphate in a time-dependent Cox model: Results of a prospective cohort study including incident dialysis patients

Abstract

Background: Hypoalbuminemia and hyperphosphatemia have been shown to be strong predictors of mortality in chronic dialysis patients. Albumin and phosphate levels are closely related. However, evaluations on their combined relationship over time with survival outcomes over a long-term observation period in incident dialysis patients are sparse. Methods: We followed 235 incident dialysis patients in a prospective single-centre cohort study (INVOR-Study: Study of INcident Dialysis Patients in VORarlberg) applying a time-dependent Cox regression analysis using all measured laboratory values for up to more than seven years. The interaction effects of albumin and phosphate values were evaluated for varying albumin values at fixed phosphate values and vice versa, determining the turnover-points for both parameters: For which values of phosphate does albumin have an effect on mortality and vice versa? Results: During a median follow-up of 35.1 months 82 patients died (35%). Albumin was found to be inversely associated with the risk of all-cause mortality (HR (95% CI) 0.24 (0.15,0.37); p<0.00001), whereas increasing phosphate levels were associated with an increased mortality risk (HR (95% CI) 1.58 (0.99,2.54); p=0.057). Including an interaction term (p=0.047), high albumin levels were still significantly associated with a reduction in mortality risk, whereas the phosphate effect alone was attenuated, emphasizing its interacting effect: For albumin values higher than 3.6, increasing phosphate levels were significantly associated with an increased mortality risk, while this association was not found at lower albumin. Analyzing varying albumin values at simultaneously fixed phosphate levels, yielded a linear inverse association between albumin and mortality with attenuation of this effect at very high phosphate levels (>2.6). Conclusions: The lowest risk is found with simultaneously low phosphate and high albumin values. However, time-dependent albumin and phosphate values interact in their association with all-cause mortality in incident dialysis patients. Epidemiological studies and therapeutic guidelines aiming for target values should consider this significant interplay.