A High Calcium-Phosphate Product Is Associated with High C-Reactive Protein Concentrations in Hemodialysis Patients

Background: An elevated Ca×PO 4 product and C-reactive protein (CRP) have been associated with coronary artery calcification and increased cardiovascular mortality in hemodialysis (HD) patients. However, it has not been defined, so far, whether and how both parameters are related to each other. For this reason we have evaluated in a cross-sectional and in an interventional study the possible correlation between Ca×PO 4 and CRP and the effect of the correction of a high Ca×PO 4 on CRP levels. Methods: 47 uremic patients (age 65 ± 16 years) on regular chronic HD were selected from a total population of 125 prevalent patients treated at our Institution. Patients had no clinical evidence of either acute infectious or inflammatory diseases for at least 4 weeks before the study. They were on regular bicarbonate HD for 6–329 months (median 42). CRP, hemoglobin (Hb), serum albumin (sAlb), protein catabolic rate (PCRn), serum calcium (Ca), serum phosphorus (PO 4 ), Ca×PO 4 , intact PTH, Kt/V, presence of ischemic heart disease (IHD) and/or peripheral vascular disease (PVD) were recorded. CRP was Ln-transformed in all statistical analyses because of positive skewness. Results: The main findings were: LnCRP 2.17 ± 0.77 mg/l, Ca 10.1 ± 0.4 mg/dl, PO 4 5.8 ± 0.6 mg/dl, Ca×PO 4 59 ± 6 mg 2 /dl 2 , andPTHint 218 ± 195 ng/ml. 18/47 had IHD, 18/47 PVD. A significant hyperbolic correlation between Ca×PO 4 and CRP was observed. A piecewise linear regression model analysis identified a break-point for Ca×PO 4 at 55 mg 2 /dl 2 . Comparison of CRP levels after the division of the patients into two groups according to Ca×PO 4 break-point (group A, Ca×PO 4 ≤55 mg 2 /dl 2 , n = 16 patients; group B, Ca×PO 4 >55 mg 2 /dl 2 , n = 31 patients) showed that CRP levels were significantly lower in patients in group A (LnCRP 1.43 ± 0.22 mg/l) than in group B (LnCRP 2.55 ± 0.67 mg/l, p < 0.0001). Multiple regression analysis bearing LnCRP as dependent variable confirmed Ca×PO 4 as the most significant variable among the other variables examined. In 22 patients with Ca×PO 4 ≧60 mg 2 /dl 2 , we performed intensive lowering of the Ca×PO 4 product in order to reach and maintain a Ca×PO 4 ≤55 mg 2 /dl 2 for 3 months. At the end of observation, a significant reduction in Ca×PO 4 and LnCRP was observed (Ca×PO 4 pre 62.8 ± 1.9 vs. post 46.3 ± 6.2 mg 2 /dl 2 : p < 0.0001; LnCRP pre 2.32 ± 0.36 vs. post 1.83 ± 0.14 mg/l: p < 0.0001). No significant variation in the other biochemical parameters was o