Mineral metabolism and the right ventricle: the Multi-Ethnic Study of Atherosclerosis (MESA).

To the Editor: CKD is associated with cardiovascular disease including myocardial infarction, heart failure, and cardiovascular mortality. Altered mineral metabolism, often seen in CKD, may contribute to cardiovascular disease by promoting adverse cardiac remodeling. Specifically, excess PTH and FGF-23 and insufficient 1,25-dihydroxyvitamin D have direct effects on left ventricle (LV)1-3 growth in experimental models. Right heart failure also contributes to morbidity and mortality across a broad range of cardiopulmonary diseases. Increased right ventricle (RV) mass has been associated with incident heart failure and cardiovascular mortality4. Compared to the LV, the RV has a distinct geometry, fiber orientation, embryologic origin, and is upstream of the low-pressure pulmonary circulation. As a result, lessons learned in the LV cannot be directly translated to the RV. The impact of disturbed mineral metabolism on the RV is unknown. We hypothesized that elevated serum concentrations of PTH and FGF-23 and lower serum concentrations of 25(OH)D are associated with increased RV mass and volumes and lower RV systolic function in MESA, a longitudinal cohort study of risk factors for subclinical atherosclerosis5. Ancillary studies have measured serum biomarkers of mineral metabolism (PTH, FGF-23, 25(OH)D, phosphorus, and calcium) and RV morphology (mass, end-diastolic volume (EDV), and ejection fraction (EF)) by cardiac magnetic resonance imaging at the baseline visit (details in Item S1). For this study, we examined 3777 of 6814 MESA participants with measurements of mineral metabolism biomarkers and RV dimensions. Cross-sectional associations of mineral metabolism biomarkers with RV measures were tested using multivariable linear regression. Study participants were racially and ethnically diverse with a normal eGFR (Table 1). Compared to those excluded, included participants had a lower prevalence of diabetes mellitus and hypertension and lower UACR and BMI (Table 1). Table 1 Characteristics of the participants in MESA PTH, FGF-23, calcium, and phosphorus were not associated with RV mass, RVEF, or RVEDV (Table 2, Table S1). Lower 25(OH)D was associated with a slightly greater RV mass in unadjusted, but not adjusted, analyses. Unadjusted associations are likely limited given confounding by body size. In adjusted analyses, lower 25(OH)D was associated with higher RVEF and lower RVEDV. Limited models adjusting for body size alone were similar to fully adjusted models (Table S1). Table 2 Associations of PTH, 25(OH)D, FGF-23, and RV structure and function. The lack of association between mineral metabolism biomarkers and RV mass in our large, community-based cohort was surprising and contrary to our hypothesis. These results contrast with published studies, including MESA data, suggesting that disturbed mineral metabolism may promote LV hypertrophy1-3, 6. Given our large cohort and precise RV mass measurements, insufficient power and misclassification seem unlikely to explain our null results. The ranges of PTH and FGF-23 were limited in this cohort with predominantly normal eGFR, however, and results might differ with more advanced kidney disease. Our null associations may suggest differential relationships of abnormal mineral metabolism on the LV relative to the RV. The ventricles are distinct7: the LV originates from the primary heart field whereas the RV arises from the anterior heart field. The LV is elliptical whereas the RV is triangular or crescentic. The LV is thicker and has more mass than the RV and as such, is better suited to handle pressure overload. The more compliant RV is better equipped to handle volume overload. Given their inherent differences and unique responses to physiologic insults, it is conceivable that mineral metabolism alterations affect each ventricle differently. Unexpectedly, lower 25(OH)D concentrations were associated with lower RVEDV and higher RVEF. This observation is counterintuitive because higher RVEF is associated with better outcomes in heart failure8 and pulmonary arterial hypertension9, but low 25(OH)D is typically associated with poor outcomes. One potential explanation is that lower 25(OH)D concentrations could lead to decreased RV compliance, such as that seen with mild fibrosis or diastolic dysfunction. This would be expected to result in a decreased RVEDV, 10 as was observed. In this relatively healthy population, compensation to maintain cardiac output would require increases in heart rate or RVEF. Unfortunately methods to identify fibrosis and diastolic dysfunction in the RV are not widely agreed on and these data were not available in MESA. Our study strengths include large size, a community-based multiethnic cohort free of baseline cardiovascular disease (reducing the likelihood of reverse causality), a comprehensive assessment of mineral metabolism markers, and precise RV measurements. Limitations include cross-sectional design; limited ranges of eGFR, PTH, and FGF-23; and lack of clinical outcomes. In conclusion, unlike associations established for the LV, mineral metabolism biomarkers were not associated with RV mass in a diverse, community-based population. Associations of 25(OH)D with RVEDV and RVEF were small, but hypothesis generating.