Abstract Background and Aims Elevated alkaline phosphatase (ALP) levels are widely recognized as a marker of poor outcomes in chronic kidney disease (CKD). This association remains poorly understood, with arguments pointing to the skeletal and non-skeletal isozyme fractions as being decisive factors. The liver and intestinal isoenzymes of ALP are able to dephosphorylate and thus inactivate a number of substrates which have been implicated in the promotion of inflammation, including the bacterial endotoxin lipopolysaccharide (LPS). Present study aimed to test the hypothesis whether association of total ALP with all-cause mortality is mainly driven by non-skeletal ALP and its link with metabolic endotoxemia and inflammation. Method Laboratory parameters of mineral and bone disease, inflammation (C-reactive protein, CRP), and metabolic endotoxemia (Lipopolysaccharide binding protein, LBP) were assessed in 431 CKD patients enrolled in the Leuven Mild-to-Moderate CKD Study between Nov 2005 and Sept 2006. Non-skeletal ALP levels were estimated as residuals of the linear regression of log transformed total and bone-specific ALP. Spearman correlation, multivariate regression, and Cox proportional hazard analyses were performed to investigate the associations between skeletal and non-skeletal ALP, LBP, CRP and mortality. Data were censored at start of renal replacement therapy. Results Median age was 63 years, 54% were men, 19% had diabetes, 31% had established cardiovascular disease. At baseline, non-skeletal ALP was associated with inflammatory parameters (CRP: rho = 0.33, P<0.001) and with metabolic endotoxemia (LBP: rho = 0.32, P<0.001). Non-skeletal ALP was a significant predictor of elevated CRP (>3 mg/dL) in multivariate models (odds ratio (OR) per standard deviation (SD) increase 1.97; 95% CI 1.21-3.21; P = 0.006). This association was attenuated by further adjustment for LBP (OR 1.40; 95% CI 0.69-2.80; P = 0.3). During a median follow-up of 8.0 (interquartile range, 2.6-14.7) years 131 patients died. High non-skeletal ALP associated with increased risk of all-cause mortality, even after adjusting for demographics, Framingham risk factors and comorbidities (Hazard ratio (HR) per SD increase 1.28; 95% CI 1.05 – 1.56; P = 0.016). This association was attenuated by further adjustment for LBP (HR 1.22; 95% CI 1.00-1.50; P = 0.051). High bone-specific ALP was not associated with increased risk of all-cause mortality. Conclusion In patients with CKD not yet on dialysis, increased non-skeletal ALP (as opposed to skeletal ALP) is associated with inflammation and increased risk of all-cause mortality. These associations are at least partly driven by metabolic endotoxemia.
Read full abstract