Mineral Metabolism Disorders in Chronic Kidney Disease

Abstract

CHRONIC KIDNEY DISEASE IS ONE OF THE MOST RAPidly increasing chronic diseases in the United States. More than 20 million US adults have an estimated glomerular filtration rate of less than 60 mL/min/1.73m, which represents loss of more than half of normal kidney function. Disturbances in phosphorus, calcium, and vitamin D metabolic pathways develop early during chronic kidney disease and eventually affect most patients during the course of their disease. Data from cell culture and animal models suggest that mineral metabolism disorders might play a causal role in the development of vascular calcification, arterial stiffness, and ventricular hypertrophy. Biomarker studies of serum concentrations of phosphorus, calcium, and parathyroid hormone provide complementary data for evaluating whether these metabolic pathways might contribute to the development of clinical disease outcomes in humans. In this issue of issue of JAMA, Palmer and colleagues compiled and summarized findings from numerous and diverse clinical outcome studies of serum concentrations of calcium, phosphorus, and parathyroid hormone among individuals who have chronic kidney disease. Their task was complicated by marked differences across studies in terms of patient populations, data collection methods, and analytic strategies. Using a meta-analytic approach, the authors found evidence for a summary association of higher serum concentrations of phosphorus with all-cause mortality, but did not find associations for serum concentrations of either calcium or parathyroid hormone. In a subgroup analysis, higher serum concentrations of parathyroid hormone were associated with mortality among patients with chronic kidney disease who were not yet receiving dialysis; however, the validity of this subgroup finding is unclear because serum concentrations of parathyroid hormone serve as a general marker of the extent and duration of kidney dysfunction beyond that of estimated glomerular filtration rate among patients not receiving dialysis. Inconsistent associations of mineral metabolism markers with clinical outcomes may indicate no true relationship with disease, but also may reflect inherent deficiencies of these factors as biomarkers. For example, serum concentrations of parathyroid hormone fluctuate according to a diurnal pattern within an individual and are strongly influenced by differences in laboratory assays. Serum concentrations of phosphorus and calcium represent minute fractions of total body mineral content, are regulated within a narrow range by counterbalancing hormones, and may be influenced by genetic differences in physiological set points. For these reasons, current etiologic research of phosphorus and calcium metabolism has generally moved beyond the evaluation of serum mineral concentrations. For example, fibroblast growth factor 23, a bone-derived hormone that regulates phosphorus homeostasis, is more strongly associated with mortality and subclinical cardiovascular disease than serum concentration of phosphorus among patients who have chronic kidney disease. Many of the studies cited in the systematic review by Palmer et al are limited by the use of registry data or electronic medical record system data, which imprecisely characterize comorbid conditions and may lead to bias. For example, the use of questionnaires or diagnosis codes to determine whether a patient has heart failure or peripheral vascular disease is likely to misclassify these conditions, limiting the ability to properly adjust for them in the analyses. Obtaining serum measurements for phosphorus, calcium, and parathyroid hormone from electronic medical record systems will preferentially focus on patients who undergo more frequent clinical laboratory testing, possibly due to illness, obscuring the validity of findings. Associations of a biomarker with disease in etiologic research are generally insufficiently strong enough to provide a basis for clinical decision making in individual patients. Even if serum concentrations of phosphorus, calcium, and parathyroid hormone were consistently associated with clinical outcomes across published studies, the distributions of these markers are likely to overlap substantially among patients who develop disease compared with those who do not, precluding their use as clinical prediction tools. Despite this fact, national and international committees have constructed guidelines that recommend target ranges of serum concentrations of phosphorus, calcium, and parathy-