Association of Nonoxidized Parathyroid Hormone with Cardiovascular and Kidney Disease Outcomes in Chronic Kidney Disease

Abstract

Background and objectives: In patients with CKD, elevated plasma parathyroid hormone (PTH) levels are associated with greater cardiovascular morbidity and mortality. However, the reference method for PTH measurement is disputed. It has been argued that measurement of nonoxidized PTH better reflects biologically active PTH than measurements with conventional assays.

Design, setting, participants, & measurements: PTH and nonoxidized PTH levels were measured at study baseline in 535 patients with CKD with an eGFR range between 89 and 15 ml/min per 1.73 m². Patients were followed over 5.1 years for the occurrence of acute heart failure, atherosclerotic events, CKD progression (doubling of serum creatinine or initiation of RRT), or all-cause death.

Results: Atherosclerotic events, acute heart failure, CKD progression, and deaths from any cause occurred in 116, 58, 73, and 85 patients, respectively. In Kaplan-Meier analyses, patients at the highest PTH and nonoxidized-PTH tertile (79-543 and 12-172 pg/ml, respectively) showed a higher rate of atherosclerotic events, acute heart failure, CKD progression, and death from any cause. After adjustment for eGFR and albuminuria, nonoxidized PTH was no longer associated with atherosclerotic events (hazard ratio third versus first tertile, 1.04 [95% confidence intervals, 0.62-1.75]), acute heart failure (hazard ratio third versus first tertile, 1.24 [95% confidence intervals, 0.59-2.62]), CKD progression (hazard ratio third versus first tertile, 0.93 [95% confidence intervals, 0.46-1.90]), and death from any cause (hazard ratio third versus first tertile, 1.23 [95% confidence intervals, 0.66-2.31]), and PTH lost its association with atherosclerotic events (hazard ratio third versus first tertile, 0.80 [95% confidence intervals, 0.46-1.38]) and CKD progression (hazard ratio third versus first tertile, 0.99 [95% confidence intervals, 0.46-2.10]), although it remained associated with acute heart failure (hazard ratio third versus first tertile, 2.76 [95% confidence intervals, 1.11-6.89]) and all-cause death (hazard ratio third versus first tertile, 2.35 [95% confidence intervals, 1.13-4.89]). After further adjustment for cardiovascular and kidney risk factors, PTH remained associated with all-cause death (hazard ratio third versus first tertile, 2.79 [95% confidence intervals, 1.32-5.89]), but with no other end point.

Conclusions: In a cohort of patients with CKD, PTH was associated with all-cause mortality; there was no association of nonoxidized PTH with any of the clinical outcomes examined.

Keywords: Cause of Death; Disease Progression; Humans; Renal Insufficiency, Chronic; Renal Replacement Therapy; albuminuria; cardiovascular disease; chronic kidney disease; clinical epidemiology; creatinine; heart failure; kidney; parathyroid hormone; risk factors.

Graphical abstract

Figure 1.

Impaired event-free survival for the endpoint acute heart failure in the highest tertiles of PTH and nox-PTH. Patients were stratified into tertiles by their plasma (A) PTH and (B) nox-PTH levels, then Kaplan–Meier analysis with log-rank testing was performed. nox-PTH, nonoxidized parathyroid hormone; PTH, parathyroid hormone.

Figure 2.

In Receiver Operating Characteristic (ROC) curves analyses PTH was superior to nox-PTH in the prediction of acute heart failure with the first 4 years of follow-up. Correlated ROC curves were compared by the Delong test for significance. AUC, area under the curve; 95% CI, 95% confidence interval; CKD-MBD, CKD–mineral and bone disorder; nox-PTH, nonoxidized parathyroid hormone; PTH, parathyroid hormone.