Calcium, Phosphate, Renal Impairment and Coronary Artery Disease in the Cardio-renal Syndrome, The CAPRICORN-CRS Study

The interaction between cardiac and renal (dys)function has been a highly relevant, yet poorly understood phenomena to both clinicians and scientists. More than 50% of the heart failure patients suffers from renal impairment, defined as a creatinine clearance \< 60 ml/min,while renal impairment is one of the most powerful predictors of outcome in heart failure.However, the complex mechanism why renal insufficiency is associated with a worse outcome, is still not fully elucidated. It was presumed that in the setting of acute heart failure, cardiac output reduces, which is counteracted by systemic and other responses such as a decrease in renal blood flow, in order to retain circulating fluid and restore cardiac output.Yet, heart failure patients with deterioration in renal function are not necessarily those with the poorest ventricular function, lowest cardiac output or the lowest blood pressures.It is therefore hypothesized that renal impairment is not merely a marker of end-stage heart failure, but is associated with a myriad of pathophysiological processes which may influence prognosis. One of the consequences of renal insufficiency that usually remains unnoticed by cardiologists, is a disregulation in the calcium and phosphate homeostasis. Yet, especially hyperphosphatemia is present in 50% of the pre-dialysis patients and is an important independent predictor of cardiovascular morbidity and mortality in this population. The pathophysiological mechanisms behind this process are intensively studied. In vitro studies show that elevated phosphate concentrations induce differentiation of vascular smooth muscle cells (VSMC) via Cbfa1 to osteoblast-like cells. These osteoblast-like cells are capable of producing bone matrix proteins, which may subsequently regulate mineralization. Once mineralization is initiated, increased Ca x PO4 product from (ab)normal bone metabolism, secondary hyperparathyroidism, or excessive calcium intake may accelerate this process leading to vascular calcification. Serum phosphate concentrations are regulated by fibroblast growth factor 23 (FGF-23), a circulating protein which may protect the transition of VSMC into osteoblast-like cells by lowering phosphate concentrations.Studies with FGF-23 null mice revealed extensive vascular calcification of the media layer in arteries.FGF-3 was also strongly related to vascular calcification in patients with end stage renal disease (ESRD), while there was no association found with atherosclerosis in subjects with normal renal function. Fetuin-A, also a circulatory protein, inhibits the de novo formation and precipitation of mineral basic calcium phosphate, but does not dissolve it once the basic calcium phosphate is formed. Therefore, fetuin-A can prevent undesirable calcification in the circulation without inhibiting bone mineralization. Serum levels of fetuin-A in individual dialysis patients are indeed inversely correlated to coronary artery calcification on CT scan. Matrix-carboxyglutamic acid protein (MGP), a calcification inhibitor, is a vitamin-K-dependent protein synthesized by chondrocytes and VSMC.MGP knockout mice develop severe calcifications of the arterial media layer.However, only carboxylated MGP seems to inhibit the process of calcification and the inhibition of MGP carboxylation by the vitamin K antagonist warfarin resulted in extensive calcification of arteries in vitro as well as in vivo.This is explained by the fact that during warfarin treatment undercarboxylated MGP (uc-MGP) is synthesized, which is inactive. It was recently found that uc-MGP serum concentrations were negatively associated with phosphate levels and positively with fetuin-A and intact parathyroid hormone (iPTH) concentrations, suggesting that low uc-MGP concentrations may be a marker of active calcification.Besides, low uc-MGP concentrations are inversely related to both the aortic augmentation index, a marker of arterial stiffness, as well as to coronary artery calcification diagnosed on CT. Interestingly, the process of vascular calcification in renal impairment seems not to be a process of accelerated atherosclerosis, i.e plaque (de)formation, but of more severe calcification besides existing atherosclerosis. It is shown in autopsy specimens from ESRD patients and matched controls that the plaque area and volume were not different in the two groups, but the plaque was more calcified in the ESRD population. Furthermore, intimal thickness was not different between the two groups as well, whereas medial thickness was significantly greater in the ESRD patients. In addition,a 2- to 5-fold increase in coronary artery calcification is found in dialysis patients compared with age-matched non-dialysis patients with angiographically proven coronary artery disease. An elegant, new diagnostic technique which seems to be ideal for imaging coronary calcification, is the coronary artery calcification score (CAC-score) as measured by electron-beam computed tomography (EBCT).This technique enables to objectify coronary artery calcification without the application of any nephrotoxic contrast agent. Although this would be an ideal tool for diagnosing coronary artery disease in patients where classical contrast agents could potentially be harmful, there is debate about the role of CAC determination in cardiovascular medicine. Indeed, recently an AHA/ACC panel of experts stated that the role of CAC is still undetermined. Because of the variability of reported data and the large variance of calcium scores in the asymptomatic population, a clear definition of a CAC score threshold for application to the general population is problematic. For a cutoff threshold in the 100-200 range of Agatston score, most population studies have found that although the negative predictive value is very high (0.95-0.99), the positive predictive value is rather low (0.02-0.13). Although there is a positive correlation between the site and the amount of coronary artery calcium and the percent of coronary luminal narrowing at the same anatomic site, the relation is nonlinear and has large confidence limits.The relation of arterial calcification, like that of angiographic coronary artery stenosis, to the probability of plaque rupture is unknown and there is a broad variation in the correlation between CAC and angiographically proven coronary artery disease in patients without renal impairment. Interestingly however, it was found that CAC on EBCT correlated well with serum concentrations of calcium and phosphate in ESRD patients. While there is uncertainty about the exact association with plaque instability, vascular calcification does lead to arterial stiffness.Arterial stiffness can be objectified by functional parameters such as the pulse wave velocity (PWV) or the augmentation index (AIx).It has been previously demonstrated that CAC strongly correlates with these parameters of arterial stiffness.Interestingly, it was found that PWV is an independent predictor of future coronary events, while patients undergoing a percutaneous coronary intervention who have an increased AIx, are at the highest risk for a future coronary event.Furthermore, in a population of patients with renal impairment it was found that PWV significantly correlates with both vascular calcification on CT as well as serum phosphate concentrations. Via an increase of pressure throughout systole and a decrease in pressure throughout diastole, arterial stiffness results in an increase of cardiomyocyte oxygen demand. While an acute increase in load increases tension-time index and acute oxygen demand, chronic increase enhances basal oxygen requirements. This process of increased coronary blood flow requirements, associated with a decreased ability to supply, develops independently of coronary narrowing. Although small studies indeed suggest that heart failure patients have higher serum phosphate concentrations, the prevalence and prognostic role of hyperphosphatemia or any disturbances in the calcium and phosphate balance of heart failure patients with renal impairment have never been the specific subject of investigation in a larger population. The correlation between coronary artery calcification, arterial stiffness and coronary events in this population is unknown as well. We designed this study in order to investigate the prevalence of calcium and phosphate disturbances and their correlates with other markers such as fetuin-A, uc-MGP and FGF-23, vascular calcification on EBCT and functional parameters of arterial stiffness.Secondly, we want to optimize the application of the CAC score in this population, where classical contrast agents are contra-indicated, by adding serological and or functional parameters to this score. Thirdly, we aim to study the prognostic role of all these parameters.