FGF23 and Cardiovascular Structure and Function in Advanced Chronic Kidney Disease : Kidney360

Journal Logo

Original Investigation: Chronic Kidney Disease

FGF23 and Cardiovascular Structure and Function in Advanced Chronic Kidney Disease

Halim, Arvin1; Burney, Heather N.2; Li, Xiaochun2; Li, Yang2; Tomkins, Claudia3; Siedlecki, Andrew M.4; Lu, Tzong-shi4; Kalim, Sahir5; Thadhani, Ravi6; Moe, Sharon1; Ting, Stephen M.S.7; Zehnder, Daniel8; Hiemstra, Thomas F.9; Lim, Kenneth1

Author Information
Kidney360 3(9):p 1529-1541, September 29, 2022. | DOI: 10.34067/KID.0002192022
  • Open
  • Infographic
  • SDC

Introduction

Fibroblast growth factor 23 (FGF23) is a potent circulating phosphatonin produced by bone (1,2). There is accumulating evidence that FGF23 is intricately related to cardiovascular health in patients with CKD. FGF23 levels rise as CKD progresses, increasing more than 1000-fold by ESKD (3). After kidney transplantation, FGF23 levels decline, and this is paralleled with improved cardiovascular disease (CVD) burden (4,5). However, both basic and clinical studies have demonstrated conflicting evidence as to whether FGF23 imparts a protective (1,2,6–8) or harmful role in regulating CVD development (3,9–13). Moreover, published clinical studies on FGF23 to date have focused mainly on mortality outcomes and resting cardiac end points such as left ventricular (LV) geometric indexes. However, the effects of these latter single measures of structural or morphologic alterations and their coupling to cardiovascular performance and functional intolerance are largely unknown (14). This has made it difficult to determine accurately the role and potential utility of FGF23 as a biomarker for assessing cardiovascular functional performance in patients with CKD.

CVD development involves complex interactions between target organ systems, including the heart (LV hypertrophy and fibrosis), vascular (arteriosclerosis and calcification), musculoskeletal (sarcopenia and bone-mineral disorders), and respiratory (impaired lung function) systems, together with widespread ultrastructural and molecular changes as CKD progresses (14). These alterations collectively contribute to impairment of the cardiovascular system to fulfill its primary function to be an effective oxygen transport system (the Fick Principle) (15). The assessment of oxygen uptake, notably at maximal exercise (VO2Max) has been widely accepted as a robust and reproducible index of cardiovascular functional capacity. Advances in cardiopulmonary exercise testing (CPET) have enabled the assessment of ventilatory gas exchange measurements during graded exercise, including VO2Max with relative ease (16–18). Significantly, VO2Max has been shown to be powerfully predictive of survival in heart failure, chronic lung disease (19–22), and perioperative demands of major surgery (23–25). Additionally, submaximal indexes such as anaerobic threshold (VO2AT) have recently been shown to be robust predictors of cardiovascular morbidity, intensive care unit admission, and survival after kidney transplantation (24,26,27).

Our preliminary work in the Cardiopulmonary Exercise Testing in Renal Failure and After Kidney Transplantation (CAPER) Study demonstrated that CPET can accurately detect changes in cardiovascular functional capacity (as assessed by VO2Max) in advanced CKD patients and those who underwent kidney transplantation prospectively at 1 year, notably in the absence of significant changes in LV mass index (LVMI) (28). A comprehensive or aggregate end point such as VO2Max, which considers the complexity of the entire oxygen transport system and its component alterations during kidney failure, is therefore emerging as a particularly appropriate index for assessing cardiovascular improvement or decline in patients with impaired kidney function. To date, no studies have examined the relationship between FGF23 and cardiovascular functional end points in advanced CKD patients and kidney transplant recipients. Given that conventional cardiovascular assessment tools such as the Framingham risk score fail to predict or reflect cardiovascular outcomes in the CKD population (29) accurately, a biomarker that adds relevant prognostic information above and beyond these conventional assessment tools has yet to be identified (30). Assessment of the role of FGF23 with end points of functional and prognostic relevance such as VO2Max compared with conventional resting imaging of morphologic alterations is therefore imperative. The present study sought to determine whether FGF23 is associated with cardiovascular functional capacity as assessed by CPET in parallel with conventional LV geometric measures by echocardiography in the CAPER cohort.

Materials and Methods

Study Design

We examined data collected in the CAPER cohort (28). The CAPER study was a three-arm prospective nonrandomized controlled cohort study. Recruitment and participant retention in the study have been previously described (28). Briefly, in the original CAPER cohort, a total of 253 patients were recruited to the study, but 53 patients were lost to follow-up throughout the 12-month period. A total of 235 participants were included in the present study from three groups: patients with advanced stage 5 CKD who underwent kidney transplantation (n=65; KTR), waitlisted nontransplanted patients with stage 5 CKD (n=87; NTWC), and hypertensive non-CKD controls (n=83; HtC). All 235 patients contributed to baseline data. However, only 186 patients contributed to 1-year follow-up data. For the KTR group, participants were enrolled in the original CAPER cohort within 4 weeks preceding kidney transplant. The 2-month and 1-year timepoints for the KTR group are post transplantation. Inclusion criteria included patients who were aged ≥18 years and who were either waitlisted or scheduled for kidney transplantation. In parallel, individuals with treated essential hypertension but without evidence of CVD (heart failure, ischemic heart disease, or cerebrovascular disease), diabetes, or secondary causes of hypertension were recruited at random from the community through a primary care database. All patients with preexisting chronic lung disease were excluded.

All patients recruited underwent CPET and transthoracic echocardiography in parallel. For patients who were hemodialysis dependent, these assessments were carried out on the first nondialysis day that was at least 12 hours after the last dialysis session in order to avoid the effects of hemodialysis-induced myocardial stunning (31) and to minimize the effect of volume load variability on the indexes of cardiovascular structure and function (32) . All blood samplings and clinical assessments were performed before exercise testing.

Study End Points

The primary end point of this study was cardiovascular functional capacity as assessed by VO2Max. Our secondary end points included ventilatory gas exchange measures (VO2AT), cardiac structural echocardiographic indexes (LVMI), and mechanical and hemodynamic measures.

CPET and Echocardiography

Briefly, CPET was conducted using an electronically braked upright cycle ergometer to maximal tolerance incorporating an individualized work rate and executed by an experienced blinded investigator as previously described (28). Each patient rested for 3 minutes, followed by 3 minutes of unloaded pedaling before workload increments, and a continuous 12-lead electrocardiogram was recorded. The test was terminated at maximal exhaustion, accompanied by attainment of a respiratory exchange ratio (RER; ratio of CO2 production to O2 consumption) of ≥1.10. Individualized predicted VO2Max was calculated according to the formula proposed by Wasserman et al. (33) The VO2 at the point of AT, expressed as percentage of predicted VO2Max, was determined by the V slope method in conjunction with analyses of the ventilatory equivalents (VE/VO2 and VE/VCO2) and end-tidal gas tension (PETO2 and PETCO2) plots (33).

Two-dimensional Doppler and tissue Doppler transthoracic echocardiography were performed using the Vivid 7 (GE Healthcare) ultrasound system according to a standardized study protocol. Further details are provided in the Supplemental Methods (19,34–36).

Laboratory Investigation

Plasma intact FGF23 concentrations were determined using an FGF23 ELISA Kit (cat. no. CY-4000; Kainos Laboratories, Inc.). All other biochemical laboratory tests were performed at the University Hospitals Coventry and Warwickshire NHS Trust clinical laboratory using standard protocols as previously described (28).

Statistical Analyses

Demographic data, comorbidities, laboratory, echocardiographic, and CPET measures at baseline were summarized by descriptive statistics and stratified by baseline quartiles of FGF23 on the logarithmic scale. Relative frequencies and percentages were calculated for categorical variables. For continuous variables, mean±SD were presented if normally distributed, or median [interquartile range] otherwise. Comparisons between baseline FGF23 groups were made by ANOVA or Kruskal–Wallis test for continuous variables and chi-squared tests for categorical variables. Unadjusted differences in each outcome variable over time were assessed for each group using paired t tests. Regression analyses using linear mixed models were conducted to assess associations between FGF23 and each outcome variable, while accounting for repeated measure correlations within each patient, where baseline FGF23 values on the logarithmic scale (i.e., logFGF23) and the change of logFGF23 since baseline (i.e., ΔlogFGF23) were included to evaluate independent associations with outcome variables, adjusted by other investigator-specified variables observed from baseline. P values <0.05 were regarded as statistically significant. Statistical analyses were conducted using SAS v9.4 (SAS Institute) and RStudio v1.2.5033 (RStudio).

Study Oversight

The study protocol was approved by the Black Country Research Ethics Committee (REC reference number: 09/H1202/113) and adhered to the Declaration of Helsinki. Written informed consent was obtained from all eligible participants.

Results

Baseline Characteristics of the Study Population

Characteristics of the study population according to quartiles (Q1–4) of baseline logFGF23 levels are included in Table 1. The mean (SD) age of the study population was 49.1 (12.98) years. There was no significant difference in age (P=0.08) and sex (P=0.5) across the groups. A slightly higher mean (SD) body mass index (BMI) was found in patients in Q1 (27.9±4 kg/m2) and Q4 (27.9±6 kg/m2) compared with those in Q2 (26±3.9 kg/m2) and Q3 (25.5±3.8 kg/m2; P=0.004). Mean arterial pressure was higher in Q1 (104.5±10 mm Hg) and incrementally declined in Q2 (101.4±12.3 mm Hg), Q3 (100.2±11.7 mm Hg), and Q4 (94.9±14.5 mm Hg; P<0.001).

Table 1. - Baseline characteristics of the study population
Characteristics FGF23 Quartile 1 FGF23 Quartile 2 FGF23 Quartile 3 FGF23 Quartile 4 P Value
<48.98 pg/ml 48.98–524.81 pg/ml 537.03–3801.89 pg/ml >3890.45 pg/ml
N=57 N=59 N=60 N=59
Men, n (%) 28 (49) 30 (51) 36 (60) 36 (61) 0.5
Age, yr, mean±SD 52.9±7.5 48.1±14.5 47±15.3 48.7±12.6 0.08
BMI (kg/m2), mean±SD 27.9±4 26±3.9 25.5±3.8 27.9±6 0.004 a
Systolic BP (mm Hg), mean±SD 141.9±14.6 139.3±17.9 136.9±18.3 130.3±21.7 0.006 a
Diastolic BP (mm Hg), mean±SD 85.8±10.3 82.5±11.8 81.8±10.9 77.2±12.7 0.001 a
MAP (mm Hg), mean±SD 104.5±10 101.4±12.3 100.2±11.7 94.9±14.5 <0.001 a
Comorbidities, n (%)
 Hypertension 56 (98) 55 (93) 57 (95) 53 (90) 0.3
 Smoking (ever) 32 (56) 29 (49) 33 (56) 30 (52) 0.9
 Diabetes 0 (0) 5 (9) 6 (10) 8 (14) 0.02 a
 Dyslipidemia 12 (21) 21 (36) 26 (43) 22 (37) 0.08
Dialysis status, n (%)
 Hemodialysis 0 (0) 10 (17) 33 (55) 50 (85) <0.001 a
 Peritoneal dialysis 0 (0) 4 (7) 6 (10) 9 (15)
 Predialysis 1 (2) 18 (31) 21 (35) 0 (0)
 Control 56 (98) 27 (46) 0 (0) 0 (0)
Dialysis vintage (months), median [IQR] 9.0 [2, 36] 32 [12, 64] 36 [16, 72] 0.03 a
Laboratory values
 eGFR (ml/min per 1.73 m2), median [IQR] 94 [84, 105] 19 [13, 83] 8.5 [7, 13.5] 6 [5, 7] <0.001 a
 Calcium (mmol/L), mean±SD 2.2±0.1 2.2±0.1 2.2±0.1 2.3±0.2 <0.001 a
 Phosphate (mmol/L), mean±SD 1.1±0.3 1.2±0.3 1.5±0.3 1.8±0.4 <0.001 a
 Albumin (mg/L), mean±SD 45.5±5.9 44.7±3.6 43.7±3.8 42.9±4 0.008 a
 25(OH)D (ng/ml), median [IQR] 43 [31, 111] 15 [11, 22] 15 [10, 18] 16 [12, 23] 0.03 a
 PTH (pmol/L), median [IQR] 3.3 [2.9, 4.4] 5.7 [3.5, 19.8] 20.3 [7.7, 36.2] 34.5 [14.8, 73.3] <0.001 a
 Hemoglobin (g/dl), mean±SD 14.2±1.3 12.7±1.9 11.7±1.3 11.9±1.5 <0.001 a
 HbA1c (%), mean±SD 5.8±0.4 5.7±0.8 5.5±0.6 5.6±1.2 0.1
 CRP (mg/L), median [IQR] 3 [3, 4] 3 [0, 4] 0 [0, 8] 4.5 [0, 9] 0.2
Characteristics, comorbidities, and laboratory values for the entire study population stratified by quartiles of baseline logFGF23. Raw baseline FGF23 values (pg/ml) are presented, and correlate to the following logarithmic FGF23 values: Q1 <1.69 log10(pg/ml), Q2 1.69–2.72 log10(pg/ml), Q3 2.73–3.58 log10(pg/ml), Q4 >3.59 log10(pg/ml). Data are presented as means±SD, median [interquartile range], or frequencies (%). Comparisons were made by ANOVA or Kruskal–Wallis test for continuous variables and chi-squared test for categorical variables. Conversion factors for units: calcium in mmol/L to mg/dl, ÷0.2495; phosphate in mmol to mg/dl, ÷0.3229; 25-hydroxyvitamin D in ng/ml to nmol/L, ×2.496. BMI, body mass index; MAP, mean arterial pressure; 25(OH)D, 25-hydroxyvitamin D, inactive vitamin D; PTH, parathyroid hormone; HbA1c, hemoglobin A1c; CRP, C-reactive protein; FGF23, fibroblast growth factor 23.
aStatistically significant P values of <0.05.

The numbers of diabetic patients were none (0%) in Q1, five (9%) in Q2, six (10%) in Q3, and eight (14%) in Q4 (P=0.02). Substantially more patients were on dialysis in the higher logFGF23 quartiles (P<0.001), whereas all but one patient were hypertensive controls in the lowest logFGF23 quartile. Median dialysis vintage was significantly higher in Q4 (36 [16, 72] months) and Q3 (32 [12, 64] months) compared with Q2 (9 [2, 36] months; P=0.03). Patients in higher quartiles of logFGF23 had reduced serum albumin (P=0.008), inactive 25-hydroxyvitamin D (25(OH)D; P=0.03), and lower hemoglobin (P<0.001) concentrations. Laboratory values that increased with increasing logFGF23 quartiles include phosphate (P<0.001) and parathyroid hormone (P<0.001). Patients in the highest logFGF23 quartile had slightly higher calcium concentrations (P<0.001).

Cardiovascular Functional and Structural Alterations

Baseline echocardiographic and cardiovascular functional capacity findings are presented in Table 2, and major end points of interest are illustrated in Figure 1. Patients in the highest logFGF23 quartiles had significantly reduced maximal oxygen consumption (VO2Max: Q4 18.6±5.2 ml/min per kilogram, Q3 19.6±4.9 ml/min per kilogram, Q2 22.4±6.4 ml/min per kilogram, Q1 24.2±4.8 ml/min per kilogram; P<0.001). Percent predicted VO2Max followed a similar pattern (P<0.001). Patients in higher quartiles of logFGF23 had reduced submaximal oxygen consumption at anaerobic threshold (VO2AT: Q4 11.4±2.5 ml/min per kilogram, Q3 11.5±2 ml/min per kilogram, Q2 13.2±3 ml/min per kilogram, Q1 14.1±2.6 ml/min per kilogram; P<0.001), and similarly had reduced percent predicted VO2AT (P<0.001). Among patients in the higher logFGF23 quartiles, we found reduced maximal workload (Q4 99.5±36.4 W, Q3 109.9±42.5 W, Q2 133.7±56.4 W, Q1 151.1±45.4 W; P<0.001), reduced endurance time (Q4 10.3±2.1 minutes, Q3 10.6±1.9 minutes, Q2 10.9±1.9 minutes, Q1 11.7±1.9 minutes; P<0.001), and lower heart rate at peak exercise (Q4 132±23.8 beats/min, Q3 133.1±27 beats/min, Q2 147.7±24.6 beats/min, Q1 153.6±18.1 beats/min; P<0.001). All patients achieved a respiratory exchange ratio (RER) of more than 1.

Table 2. - Baseline cardiovascular measures of the study population
Characteristics FGF23 Quartile 1 FGF23 Quartile 2 FGF23 Quartile 3 FGF23 Quartile 4 P Value
<48.98 pg/ml 48.98–524.81 pg/ml 537.03–3801.89 pg/ml >3890.45 pg/ml
N=57 N=59 N=60 N=59
CPET indexes
 VO2Max, ml/min per kilogram 24.2±4.8 22.4±6.4 19.6±4.9 18.6±5.2 <0.001 a
 VO2Max, % of predicted VO2Max 95.4±2.01 78.4±17.8 68.4±16.6 69.5±18.3 <0.001 a
 VO2AT, ml/min per kilogram 14.1±2.6 13.2±3 11.5±2 11.4±2.5 <0.001 a
 VO2AT, % of predicted VO2AT 55.8±11.8 46.9±12 40.6±9.5 42.8±9.6 <0.001 a
 O2 pulse, ml/beat 12.4±3.7 11.5±3.7 11.3±3.9 11.1±3.4 0.2
 RER at peak exercise 1.2±0.1 1.2±0.1 1.3±0.1 1.2±0.1 0.2
 HR at peak exercise, beats/min 153.6±18.1 147.7±24.6 133.1±27 132±23.8 <0.001 a
 HR at peak exercise, % of predicted 91.9±10.1 85.9±13 77.1±15.1 77.3±14.7 <0.001 a
 Maximal workload, watt 151.1±45.4 133.7±56.4 109.9±42.5 99.5±36.4 <0.001 a
 Endurance time, min 11.7±1.9 10.9±1.9 10.6±1.9 10.3±2.1 <0.001 a
Echocardiography
 LVMI, g/m2 87.6±16.2 95.6±31.8 108.6±38.5 113±36 <0.001 a
 LVEF, % 66.9±6 62.5±6.8 60.7±9.2 60.2±9.2 <0.001 a
 LVIDd, cm 4.5±0.5 4.5±0.6 4.6±0.8 4.7±0.7 0.3
 LVEDVI, ml/m2 44.7±10.2 46.3±14.1 51.3±19.7 49.6±15.3 0.08
 LVESVI, ml/m2 14.7±4.7 17.6±7.7 21±11.4 20.6±9.5 <0.001 a
 E/mean e′ 8.2±1.7 8±2.3 8.4±3.1 10.3±5.5 <0.001 a
 LA diameter, cm 3.7±0.5 3.5±0.6 3.6±0.9 3.8±0.7 0.08
 LA volume index, ml/m2 26.3±8.3 23.8±10.1 28.5±16.3 29±14 0.1
CPET and echocardiographic measures for each quartile of baseline logFGF23. Data are presented as means±SD. Raw baseline FGF23 values (pg/ml) are presented and correlate to the following logarithmic FGF23 values: Q1 <1.69 log10(pg/ml), Q2 1.69–2.72 log10(pg/ml), Q3 2.73–3.58 log10(pg/ml), Q4 >3.59 log10(pg/ml). Comparisons were made by ANOVA for multigroup comparison. Cardiovascular functional capacity variables: VO2Max, oxygen consumption at maximal exercise; VO2AT, oxygen consumption at the point of anaerobic threshold; O2 pulse, ratio of oxygen consumption and heart rate; RER, respiratory exchange ratio; HR, heart rate. Echocardiographic variables: LVIDd, left ventricular internal dimension diastole; LVEDVI, LV end-diastolic volume index; LVESVI, LV end-systolic volume index; LA, left atrial; Transmitral E/mean e′, ratio of early transmitral flow velocity to annular mitral velocity (averaged of septal and lateral); FGF23, fibroblast growth factor 23.
aStatistically significant P values of <0.05.

F1
Figure 1.:
Cardiovascular functional capacity measures and LVMI by FGF23 levels. Mean, median, and interquartile ranges of unadjusted baseline values of (A) VO2Max, (B) VO2AT, (C) maximal workload, and (D) LVMI between quartiles of logFGF23. Increase in logFGF23 is coupled with decreasing trends in cardiovascular functional capacity measures VO2Max, VO2AT, and maximal workload, and an increasing trend in LVMI. LVMI, left ventricular mass index; FGF23, fibroblast growth factor 23; VO2Max, oxygen consumption at maximal exercise; VO2AT, oxygen consumption at the point of anaerobic threshold.

LVMI was lowest in Q1 (87.6±16.2 g/m2) and incrementally increased in Q2 (95.6±31.8 g/m2), Q3 (108.6±38.5 g/m2), and Q4 (113±36 g/m2; P<0.001) as logFGF23 levels increased. Patients in the highest logFGF23 quartile had the lowest LV ejection fraction (Q4 60%±9%, Q3 61%±9%, Q2 63%±7%, Q1 67%±6%; P<0.001). These patients also exhibited higher LV end-systolic volume index (Q4 20.6±9.5 ml/min2, Q3 21±11.4 ml/min2, Q2 17.6±7.7 ml/min2, Q1 14.7±4.7 ml/min2; P<0.001) and higher estimated LV filling pressure (E/mean e′: Q4 10.3±5.5, Q3 8.4±3.1, Q2 8±2.3, Q1 8.2±1.7; P<0.001).

Changes in FGF23, Cardiovascular Functional Capacity, and Cardiac Structure before and after Kidney Transplantation

We next examined logFGF23 levels and cardiovascular functional-structural measures prospectively in each of the three groups: KTR, NTWC, and HtC. Results are illustrated in Figure 2, and the mean difference between each timepoint are illustrated in Figure 3.

F2
Figure 2.:
Changes in cardiovascular functional measures, FGF23, and LVMI before and after kidney transplant. Unadjusted mean values of primary CPET and echocardiographic indexes based on patient group. (A) KTR patients had increased VO2Max at 1 year after transplantation, approaching HtC levels. NTWC had a slight decline in VO2Max, whereas VO2Max in HtC was relatively stable. (B) Maximal workload in KTR progressively increased and approached HtC levels by 1 year. Over the course of a year, NTWC had a slight decrease in maximal workload, and HtC had relatively stable maximal workload levels. (C) LogFGF23 levels declined by 2 months and maintained lower levels in the KTR group, nearing logFGF23 levels of HtC. LogFGF23 levels stayed relatively stable, if not slightly increased, in NTWC and HtC groups. (D) LVMI stayed relatively stable for all three groups in comparison between baseline and 1 year. CPET, cardiopulmonary exercise testing. KTR group, patients with advanced stage 5 CKD who underwent kidney transplantation; NTWC group, waitlisted nontransplanted patients with stage 5 CKD; HtC group, hypertensive non-CKD controls.
F3
Figure 3.:
Mean differences in cardiovascular measures and FGF23 between baseline, 2 months, and 1 year. Data are presented as means±SD for values at each time point and means for differences between time points. (A) Values of VO2Max are in ml/min per kilogram. KTR patients exhibited a significant increase in VO2Max between 2 months and 1 year and between baseline and 1 year. NTWC had a significant decline in VO2Max between 2 months and 1 year and between baseline and 1 year. HtC had a slight but nonsignificant increase in VO2Max between baseline and 1 year. Values of maximal workload are in watts. KTR had a significant increase in maximal workload between 2 months and 1 year and between baseline and 1 year. NTWC had significant decrease in maximal workload between 2 months and 1 year. HtC had no significant change in maximal workload between baseline and 1 year. (B) KTR had a significant major decline in logFGF23 between baseline and 2 months, a smaller yet significant decline between 2 months and 1 year, and an overall significant decline between baseline and 1 year. NTWC had no significant changes in logFGF23. HtC had a small significant increase in logFGF23 between baseline and 1 year. KTR had a significant increase in LVMI between baseline and 2 months but had a significant decline between 2 months and 1 year. Overall, KTR had a no significant change in LVMI between baseline and 1 year. NTWC and HtC both had no significant changes in LVMI.

Our group previously reported significant improvement in VO2Max, VO2AT, maximal workload, endurance time, and LV ejection fraction in the KTR group, whereas VO2Max and VO2AT significantly declined in the NTWC group (28). VO2Max (P=0.008) and maximal workload (P<0.001) significantly improved at 1 year in the KTR group in parallel with reductions in logFGF23 (P<0.001) to near normal levels. VO2Max remained impaired (P=0.005) in the NTWC group at 1 year in parallel with persistently high logFGF23 levels (P=0.3). By 2 months post KTR, logFGF23 levels were significantly reduced compared with baseline (P<0.001) and remained low at 1 year (P=0.004). Patients in the HtC group exhibited a small but significant increase in logFGF23 from baseline to 1 year (P<0.001). LogFGF23 levels in patients in the NTWC group did not significantly change between timepoints (P=0.1–0.4). LVMI did not significantly change at 1 year in either the KTR (P=0.6) or NTWC (P>0.9) groups.

Association between FGF23 and Cardiovascular Functional Capacity

First we examined the relationship between logFGF23 and longitudinal VO2Max using stepwise multivariable regression analysis of the entire cohort (Table 3). Baseline logFGF23 was significantly associated with longitudinal VO2Max after adjusting for age, sex, systolic BP, smoking status, dyslipidemia, albumin, hemoglobin, and changes in logFGF23 from baseline (ΔlogFGF23 β=–2.33 [SEM 0.32]; P<0.001) in Model 1. Similarly, we found that ΔlogFGF23 was also significantly associated with longitudinal VO2Max after adjusting for the same covariates and baseline logFGF23 (β=–0.9 [SEM 0.25]; P<0.001). In Model 2, baseline logFGF23 (β=–2.3 [SEM 0.32]; P<0.001) and ΔlogFGF23 (β=–0.89 [SEM 0.25]; P<0.001) both remained significantly associated with longitudinal VO2Max after the addition of diabetes to the model. In our final Model 3, baseline logFGF23 (β=–1.18 [SEM 0.53]; P=0.03) and Δlog(FGF23) (β=–0.85 [SEM 0.25]; P<0.001) remained significantly associated with longitudinal VO2Max after the addition of dialysis status to the model.

Table 3. - Association of FGF23 with VO2Max in all patients before and after kidney transplantation
Variable (Baseline) VO2Max
Model 1 Model 2 Model 3
β (SEM) P Value β (SEM) P Value β (SEM) P Value
Age, yr −0.15 (0.02) <0.001 a −0.15 (0.02) <0.001 a −0.16 (0.02) <0.001 a
Sex, men versus women 3.77 (0.54) <0.001 a 3.8 (0.54) <0.001 a 3.87 (0.53) <0.001 a
Systolic BP, mm Hg 0.02 (0.01) 0.2 0.02 (0.01) 0.2 0.02 (0.01) 0.1
Smoking status, former versus current 1.53 (0.75) 0.04 a 1.66 (0.75) 0.03 a 1.68 (0.75) 0.03 a
Smoking status, never versus current 0.3 (0.7) 0.7 0.41 (0.7) 0.6 0.4 (0.7) 0.6
Dyslipidemia, yes versus no −2.36 (0.6) <0.001 a −2.16 (0.6) <0.001 a −2.08 (0.59) <0.001 a
Albumin, mg/L 0.12 (0.06) 0.05 0.12 (0.06) 0.06 0.1 (0.06) 0.1
Hemoglobin, g/dl 0.45 (0.18) 0.01 a 0.41 (0.18) 0.03 a 0.19 (0.2) 0.3
logFGF23 −2.33 (0.32) <0.001 a −2.3 (0.32) <0.001 a −1.18 (0.53) 0.03 a
ΔlogFGF23 (from baseline) −0.9 (0.25) <0.001 a −0.89 (0.25) <0.001 a −0.85 (0.25) <0.001 a
Diabetes, yes versus no −1.82 (1.03) 0.08 −1.11 (1.06) 0.3
Dialysis status, dialysis versus control −3.67 (1.38) 0.008 a
Dialysis status, predialysis versus control −2.61 (1.12) 0.02 a
Table of multivariable, stepwise regression modeling results comparing longitudinal VO2Max (ml/min per kilogram) and clinically relevant covariates. In Model 1, baseline logFGF23 was significantly associated with longitudinal VO2Max after adjusting for age, sex, systolic BP, smoking, dyslipidemia, albumin, hemoglobin, and changes in logFGF23 from baseline to 1 year (ΔlogFGF23). ΔlogFGF23 was also significantly associated with longitudinal VO2Max after adjusting for baseline logFGF23 and the same variables. In Model 2, both baseline logFGF23 and ΔlogFGF23 remained significantly associated with longitudinal VO2Max after adjusting for diabetes and the same respective variables in Model 1. In Model 3, both baseline logFGF23 and ΔlogFGF23 were significantly associated with longitudinal VO2Max after adjusting for dialysis status and predialysis status and respective variables in Model 2. FGF23, fibroblast growth factor 23.
aStatistically significant P values of <0.05.

Next, we restricted multivariable regression analysis to the KTR group only to assess if associations remain significant in patients who had undergone transplantation (Table 4). Baseline logFGF23 was significantly associated with longitudinal VO2Max after adjusting for age, sex, systolic BP, smoking status, dyslipidemia, albumin, hemoglobin, and ΔlogFGF23 (β=–2.09 [SEM 0.71]; P=0.005) in Model 1. ΔlogFGF23 was also significantly associated with longitudinal VO2Max after adjusting for the same covariates and baseline logFGF23 (β=–0.85 [SEM 0.3]; P=0.006). In Model 2, baseline logFGF23 (β=–2.12 [SEM 0.71]; P=0.004) and Δlog(FGF23) (β=–0.85 [SEM 0.3]; P=0.007) both remained significantly associated with longitudinal VO2Max after the addition of diabetes to the model. In Model 3, baseline logFGF23 (β=–2.98 [SEM 0.92]; P=0.002) and ΔlogFGF23 (β=–0.87 [SEM 0.3]; P=0.006) remained significantly associated with longitudinal VO2Max after the addition of dialysis status to the model.

Table 4. - Association of FGF23 with VO2Max in KTR group before and after kidney transplantation
Variable (Baseline) VO2Max
Model 1 Model 2 Model 3
β (SEM) P Value β (SEM) P Value β (SEM) P Value
Age, yr −0.18 (0.04) <0.001 a −0.17 (0.04) <0.001 a −0.17 (0.04) <0.001 a
Sex, men versus women 4.95 (1.05) <0.001 a 4.83 (1.08) <0.001 a 5.04 (1.08) <0.001 a
Systolic BP, mm Hg 0.05 (0.02) <0.05 a 0.05 (0.03) 0.04 a 0.06 (0.03) 0.03 a
Smoking status, former versus current 2.58 (1.55) 0.1 2.65 (1.55) 0.09 3.15 (1.56) <0.05 a
Smoking status, never versus current 1.59 (1.48) 0.3 1.67 (1.49) 0.3 1.98 (1.48) 0.2
Dyslipidemia, yes versus no −3.05 (1.13) 0.009 a −2.93 (1.16) 0.01 a −3.19 (1.15) 0.008 a
Albumin, mg/L 0.03 (0.16) 0.9 0.04 (0.16) 0.8 0.08 (0.16) 0.6
Hemoglobin, g/dl 0.06 (0.35) 0.9 0.05 (0.35) 0.9 0.02 (0.35) 1
logFGF23 −2.09 (0.71) 0.005 a −2.12 (0.71) 0.004 a −2.98 (0.92) 0.002 a
ΔlogFGF23 (from baseline) −0.85 (0.3) 0.006 a −0.85 (0.3) 0.007 a −0.87 (0.3) 0.006 a
Diabetes, yes versus no −0.86 (2.02) 0.7 −1.37 (2.02) 0.5
Dialysis status, dialysis versus predialysis 1.99 (1.4) 0.2
Table of multivariable, stepwise regression modeling results comparing longitudinal VO2Max (ml/min per kilogram) and clinically relevant covariates. In Model 1, baseline logFGF23 was significantly associated with longitudinal VO2Max after adjusting for age, sex, systolic BP, smoking, dyslipidemia, albumin, hemoglobin, and ΔlogFGF23. ΔlogFGF23 was also significantly associated with longitudinal VO2Max after adjusting for baseline logFGF23 and the same variables. In Model 2, both baseline logFGF23 and ΔlogFGF23 remained significantly associated with longitudinal VO2Max after adjusting for diabetes and the same respective variables in Model 1. In Model 3, both baseline logFGF23 and ΔlogFGF23 were both significantly associated with longitudinal VO2Max after adjusting for dialysis versus predialysis status and respective variables in Model 2. The second status in either/or variables is the reference status. KTR group, patients with advanced stage 5 CKD who underwent kidney transplantation; FGF23, fibroblast growth factor 23.
aStatistically significant P values of <0.05.

Multivariable regression analysis for longitudinal LVMI of the entire cohort (Table 5) demonstrated that baseline logFGF23 levels were associated with longitudinal LVMI after adjustment for age, sex, BMI, systolic BP, smoking, dyslipidemia, albumin, hemoglobin, and ΔlogFGF23 (β=5.66 [SEM 1.83]; P=0.002) in Model 1. However, ΔlogFGF23 was not significantly associated with longitudinal LVMI (β=1.92 [SEM 1.69]; P=0.3) after adjusting for the same covariates and baseline logFGF23. In Model 2, baseline logFGF23 levels remained significantly associated with longitudinal LVMI (β=5.58 [SEM 1.83]; P=0.003) after the addition of diabetes to the model, whereas ΔlogFGF23 was not significantly associated with longitudinal LVMI (β=1.89 [SEM 1.69]; P=0.3). In Model 3, neither baseline logFGF23 (β=2.56 [SEM 3.15]; P=0.4) nor ΔlogFGF23 (β=1.68 [SEM 1.7]; P=0.3) was significantly associated with longitudinal LVMI after addition of dialysis status to the model.

Table 5. - Association of FGF23 with LVMI in all patients before and after kidney transplantation
Variable (Baseline) LVMI
Model 1 Model 2 Model 3
β (SEM) P Value β (SEM) P Value β (SEM) P Value
Age, yr 0.34 (0.13) 0.01 a 0.32 (0.14) 0.02 a 0.28 (0.14) <0.05 a
Sex, men versus women 19 (3.09) <0.001 a 18.94 (3.1) <0.001 a 18.9 (3.07) <0.001 a
BMI, kg/m2 −0.28 (0.35) 0.4 −0.3 (0.36) 0.4 −0.34 (0.36) 0.3
Systolic BP, mm Hg 0.12 (0.09) 0.2 0.12 (0.09) 0.2 0.11 (0.09) 0.2
Smoking status, former versus current −5.07 (4.3) 0.2 −5.45 (4.34) 0.2 −3.95 (4.35) 0.4
Smoking status, never versus current −12.46 (4.02) 0.003 a −12.76 (4.06) 0.002 a −11.54 (4.06) 0.005 a
Dyslipidemia, yes versus no 0.89 (3.46) 0.8 0.38 (3.52) 0.9 0.38 (3.49) 0.9
Albumin, mg/L −0.11 (0.36) 0.8 −0.1 (0.36) 0.8 −0.02 (0.36) 1.0
Hemoglobin, g/dl −5.65 (1.06) <0.001 a −5.53 (1.07) <0.001 a −5.57 (1.15) <0.001 a
logFGF23 5.66 (1.83) 0.002 a 5.58 (1.83) 0.003 a 2.56 (3.15) 0.4
ΔlogFGF23 (from baseline) 1.92 (1.69) 0.3 1.89 (1.69) 0.3 1.68 (1.7) 0.3
Diabetes, yes versus no 4.79 (5.94) 0.4 2.38 (6.14) 0.7
Dialysis status, dialysis versus control 7.78 (8.07) 0.3
Dialysis status, predialysis versus control −4.03 (6.63) 0.5
Table of multivariable, stepwise regression modeling results comparing longitudinal LVMI (g/m2) and clinically relevant covariates. In Model 1, baseline logFGF23 was significantly associated with longitudinal LVMI after adjusting for age, sex, BMI, systolic BP, smoking, dyslipidemia, albumin, hemoglobin, and changes in logFGF23 from baseline to 1 year (ΔlogFGF23). ΔlogFGF23 was not associated with longitudinal LVMI in all patients after adjusting for baseline logFGF23 and the same variables. In Model 2, baseline logFGF23 remained significantly associated with longitudinal LVMI after adjusting for diabetes and the same respective variables in Model 1, whereas ΔlogFGF23 remained not significantly associated with longitudinal LVMI. In Model 3, neither baseline logFGF23 nor ΔlogFGF23 were significantly associated with LVMI after adjusting for dialysis and predialysis statuses and the same respective variables in Model 2. FGF23, fibroblast growth factor 23; LVMI, left ventricular mass index; BMI, body mass index.
aStatistically significant P values of <0.05.

Across all our stepwise regression models, including the KTR group only (Table 6), baseline logFGF23 and ΔlogFGF23 were not significantly associated with longitudinal LVMI.

Table 6. - Association of FGF23 with LVMI in KTR group before and after kidney transplantation
Variable (Baseline) LVMI
Model 1 Model 2 Model 3
β (SEM) P Value β (SEM) P Value β (SEM) P Value
Age, yr 0.21 (0.28) 0.5 0.12 (0.28) 0.7 0.12 (0.28) 0.7
Sex, men versus women 20.52 (6.6) 0.003 a 22.8 (6.78) 0.001 a 24.19 (6.76) <0.001 a
BMI, kg/m2 0.27 (0.98) 0.8 0.41 (0.97) 0.7 0.48 (0.96) 0.6
Systolic BP, mm Hg 0.21 (0.16) 0.2 0.16 (0.16) 0.3 0.18 (0.16) 0.3
Smoking status, former versus current −2.76 (9.75) 0.8 −4.05 (9.68) 0.7 −0.97 (9.8) 0.9
Smoking status, never versus current −14.69 (9.34) 0.1 −16.32 (9.32) 0.09 −14.54 (9.28) 0.1
Dyslipidemia, yes versus no −2.2 (7.34) 0.8 −4.54 (7.5) 0.5 −6.29 (7.5) 0.4
Albumin, mg/L 0.33 (1.03) 0.8 0.18 (1.03) 0.9 0.44 (1.03) 0.7
Hemoglobin, g/dl −1.68 (2.22) 0.5 −1.38 (2.21) 0.5 −1.51 (2.18) 0.5
logFGF23 6.41 (4.59) 0.2 6.91 (4.56) 0.1 1.49 (5.9) 0.8
ΔlogFGF23 (from baseline) 0.81 (2.14) 0.7 0.8 (2.14) 0.7 0.69 (2.13) 0.7
Diabetes, yes versus no 15.82 (12.72) 0.2 12.81 (12.73) 0.3
Dialysis status, dialysis versus predialysis 12.45 (8.77) 0.2
Table of multivariable, stepwise regression modeling results comparing longitudinal LVMI (g/m2) and clinically relevant covariates. In all three models, neither baseline logFGF23 nor ΔlogFGF23 were significantly associated with longitudinal LVMI after adjusting for ΔlogFGF23 and baseline logFGF23, respectively, age, sex, BMI, systolic BP, smoking, dyslipidemia, albumin, hemoglobin, diabetes (Model 2 and 3 only), and dialysis versus predialysis status (Model 3 only). The second status in either/or variables is the reference status. FGF23, fibroblast growth factor 23; LVMI, left ventricular mass index; BMI, body mass index; KTR group, patients with advanced stage 5 CKD who underwent kidney transplantation.
aStatistically significant P values of <0.05.

Discussion

In this study, we examined the relationship between FGF23 and cardiovascular functional capacity in patients with advanced CKD and after restitution of the failing kidney by transplantation. Clinical and epidemiologic studies examining the role of FGF23 in CVD to date have focused largely on the assessment of individual single-organ systems using end points, such as resting geometric cardiac changes. Although these studies are vital to help elucidate the potential role of FGF23, it is increasingly recognized that assessment of resting cardiac function alone cannot reliably predict cardiovascular functional capacity and exercise performance. In fact, the overall health of an individual is strongly linked to an integrative exercise response that takes into account the interaction of multiple target organs that form the oxygen transport system. Assessment of integrative exercise responses by CPET also more easily provides a better understanding of structural and molecular changes and its effect on function. Moreover, despite exercise tolerance being one of the most potent predictors of cardiovascular health, this is rarely assessed in patients with CKD (37).

In this study, we first showed that cardiovascular functional capacity as assessed by VO2Max was significantly impaired in patients with higher levels of FGF23. This occurred in parallel with alterations in cardiac structure, including increased LVMI, higher LV filling pressure, and reduced ejection fraction, consistent with previous reports (38–40). In a study by Mehta et al. that examined 172 patients from the Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in HFpEF trial, the authors showed that higher levels of FGF23 were associated with impaired baseline exercise capacity as assessed by VO2Peak (41). In this study, 47% of patients had CKD with an average eGFR of 63.9±22 ml/min per 1.73 m2. In our study utilizing the CAPER cohort, we found that FGF23 was significantly associated with impaired VO2Max in advanced predialytic CKD, dialysis patients, and hypertensive patients. Additionally, higher FGF23 levels were associated with impairment of submaximal indexes, including anaerobic threshold, which are independent of volitional effort. Given that reduction of cardiovascular functional capacity in CKD likely involves the impairment of multiple organ systems that form the oxygen transport chain (14), we postulate that FGF23 could potentially regulate multiple pathways in this integrated system.

At the heart, there is evidence that FGF23 may increase cardiac contractility and isometric tension of cardiomyocytes by increasing [Ca2+] influx in the acute setting (42). Long-term exposure to FGF23 could possibly disrupt Ca2+ homeostasis and activate transcriptional mechanisms that lead to cardiac hypertrophy (3,42). Moreover, in a study by Faul et al., FGF23 stimulated development of pathologic cardiac hypertrophy via FGF receptor-dependent activation of the calcineurin-NFAT signaling pathway (3). In our study, we found that FGF23 was associated with LVMI in advanced CKD, consistent with these findings. However, interestingly, this association was lost after adjustment for dialysis status and the reasons for this are unclear. Furthermore, higher FGF23 levels were associated with chronotropic incompetence. In vasculature, high FGF23 has been shown to be independently associated with vascular calcification in various stages of CKD (43). In skeletal muscle, disruption of Ca2+ by FGF23 may also contribute to CKD-associated muscle weakness and wasting (44). Elevated skeletal myoplasmic [Ca2+] leads to mitochondrial dysfunction, production of reactive oxygen species, impaired oxidative phosphorylation, and consequently muscle degeneration (45–48). We found that high levels of FGF23 were associated with reductions in maximal workload and endurance time during incremental CPET testing. Further studies are needed to investigate the effects of FGF23 that are either Klotho dependent or Klotho independent that could target the components of the Fick equation (14), which describes regulatory variables of VO2Max.

Restitution of the failing kidney by transplantation has been well documented to be associated with improved cardiovascular outcomes and survival (49). In this study, we found that FGF23 significantly decreased at 2 months and remained low at 1 year after kidney transplantation. Furthermore, we found that the decline in FGF23 after transplantation was associated with improved VO2Max. These results suggest that normalization of FGF23 after transplantation could be an important mechanism involved in restoration of functional capacity. Brand et al. demonstrated that FGF23 stimulates cardiac hypertrophy by activating fibroblast growth factor receptor 4, whereas blockade of fibroblast growth factor receptor 4 ameliorates this phenotype (50). In our study, we were unable to detect significant echocardiographic changes in LVMI after transplantation. In fact, studies examining regression of LV hypertrophy after transplantation have demonstrated conflicting results (51–56). Given that diffuse myocardial fibrosis is involved in the pathogenesis of diastolic heart failure in advanced CKD, we postulate that inhibition of myocardial fibrogenesis could contribute to improved ventilatory gas exchange patterns during incremental exercise testing. Whether FGF23 could be involved in this process is unclear. Additionally, lowering FGF23 has been shown to enhance erythropoiesis and improve anemia (57,58). Reversal of other uremic factors, mineral bone disease, and sarcopenia after transplantation are also likely involved in improvement of cardiovascular health. Although our study is unable to make causal inferences, our study provides rationale for further preclinical studies evaluating the role and mechanisms of FGF23 in regulating integrative exercise responses in CKD.

A strength of this study is that we have been able to assess the link of FGF23 with cardiovascular function and structure in parallel using state-of-the-art CPET technology. CPET provides a wealth of data per single assessment, and combined with cardiac imaging, this has provided in-depth mechanistic insights into cardiovascular changes in CKD. Additionally, as previously noted in the original CAPER publication (28), because of the universal prevalence of hypertension in patients with CKD, we assessed advanced CKD patients in parallel with a control group who has essential hypertension alone. This has allowed us to differentiate cardiovascular changes in the setting of renal failure from those induced by hypertension. Our data presented here should be interpreted against the limitations of the study. First, the study is limited by its relatively small sample size. Second, the study may involve selection bias of the population because completion of CPET excludes patients who are unable to perform incremental exercise testing. CPET has been extensively used in the general heart failure population; however, data on CPET use in the CKD population have historically been scarce and only now has this become an emerging area. Another limitation of the study is the issue of patient heterogeneity, complexity of renal pathophysiology and its interactions with CVD, and competing risks, which are known challenges with all studies involving CKD patients (59). We acknowledge that it is not possible to adjust for every confounder, and further studies are needed to test for causal inference.

In summary, this is the first study to examine the relationship between FGF23 and cardiovascular functional capacity in parallel with assessment of structural adaptations before and after kidney transplantation. We found that FGF23 is associated with cardiovascular functional capacity in advanced CKD and after restitution of the failing kidney by transplantation. Because restitution of the failing kidney by transplantation resulted in reductions of FGF23 followed by improvement in VO2Max, we postulate that changes in FGF23 levels may be predictive of alterations in cardiovascular functional capacity. Our study also provides evidence that baseline FGF23 is significantly associated with the progression of maladaptive cardiac alterations, specifically LV mass in advanced CKD that was modified by dialysis status. Additionally, FGF23 was not associated with improved LV mass after kidney transplantation. The study highlights the complex relationship between cardiovascular structure and function, and the precise role of regulators of cardiovascular health such as FGF23. In addition, our study provides critical data for the potential utility and limitations of FGF23 as a biomarker of structural alterations, to help stratify cardiovascular risk in CKD, and as an assessment tool for tracking alterations in cardiovascular functional capacity in CKD.

Disclosures

A. Halim reports ownership interest in OVIBIO Corp. T.F. Hiemstra reports being an employee of GlaxoSmithKline; ownership interest in GlaxoSmithKline, Kodika Corp., and OVIBIO Corp.; research funding from UNION Therapeutics; and an advisory or leadership role for the International Clinical Trial Centre Network (SAB member) and the PHOSPHATE Trial (steering board member). S. Kalim reports participation in a speakers’ bureau for Fresenius Kabi. X. Li reports an advisory or leadership role for Health Services and Outcomes Research Methodology (member of editorial board 2014–present) and Circulation: Cardiovascular Quality and Outcomes (associated statistical editor 2016–present). K. Lim reports consultancy for Ambassadors Corp., MBX Biosciences, and OVIBIO Corp.; ownership interest in Ambassadors Corp., MBX Biosciences, and OVIBIO Corp.; and an advisory or leadership role for Ambassadors Corp. T. Lu reports consultancy for OVIBIO Corp. and ownership interest in OVIBIO Corp. S. Moe reports consultancy for Amgen, Ardelyx, and Sanifit; ownership interest in Eli Lilly (stock); research funding from Chugai (research grant), Keryx (research grant), and the National Institutes of Health (research grant); honoraria from Amgen, Ardelyx, and Sanifit; and an advisory or leadership role for the American Journal of Nephrology and the American Journal of Nutrition (editorial board). A.M. Siedlecki reports research funding from Alexion, and an advisory or leadership role for Alexion. R. Thadhani reports consultancy for Alnylum, Bayer, the US Food and Drug Administration, Fresenius Medical Care North America, Kaneka, and Thermo Fisher Scientific; ownership interest in Aggamin LLC, Comanche Pharma, Hero Health, and Tvardi; research funding from Thermo Fisher Scientific; honoraria from Alnylum, Bayer, and Thermo Fisher Scientific; patents or royalties from Gravidas Diagnostics, Thermo Fisher Scientific, and UpToDate; and an advisory or leadership role for Aggamin LLC and Vifor Pharma. D. Zehnder reports research funding for Abbott Laboratories and Amgen, and honoraria from Abbott Laboratories and Amgen. All remaining authors have nothing to disclose.

Funding

The CAPER study was funded by a British Heart Foundation grant (PG/11/66/28982). K. Lim received an Early Clinical Investigator award from the Medical Research Foundation of Oregon and a National Institutes of Health career development award (K23 DK115683) to help support the present study.

See related editorial, “Exercising the FGF23-Cardiac Axis,” on pages .

Acknowledgments

We would like to thank the Reading family and University Hospital Coventry and Warwickshire NHS Trust Charity for funding the CPET machine used in this study. We thank the National Institute of Health Research for clinical staffing support providing assistance in the administrative work and data collection.

Data Sharing Statement

Original data created for the study are or will be available upon request.

Supplemental Material

This article contains the following supplemental material online at http://kidney360.asnjournals.org/lookup/suppl/doi:10.34067/KID.0002192022/-/DCSupplemental.

Supplemental Methods.

References

1. Urakawa I, Yamazaki Y, Shimada T, Iijima K, Hasegawa H, Okawa K, Fujita T, Fukumoto S, Yamashita T: Klotho converts canonical FGF receptor into a specific receptor for FGF23. Nature 444: 770–774, 2006 https://doi.org/10.1038/nature05315
2. Kurosu H, Ogawa Y, Miyoshi M, Yamamoto M, Nandi A, Rosenblatt KP, Baum MG, Schiavi S, Hu MC, Moe OW, Kuro-O M: Regulation of fibroblast growth factor-23 signaling by klotho. J Biol Chem 281: 6120–6123, 2006 https://doi.org/10.1074/jbc.C500457200PMC2637204
3. Faul C, Amaral AP, Oskouei B, Hu MC, Sloan A, Isakova T, Gutiérrez OM, Aguillon-Prada R, Lincoln J, Hare JM, Mundel P, Morales A, Scialla J, Fischer M, Soliman EZ, Chen J, Go AS, Rosas SE, Nessel L, Townsend RR, Feldman HI, St John Sutton M, Ojo A, Gadegbeku C, Di Marco GS, Reuter S, Kentrup D, Tiemann K, Brand M, Hill JA, Moe OW, Kuro-O M, Kusek JW, Keane MG, Wolf M: FGF23 induces left ventricular hypertrophy. J Clin Invest 121: 4393–4408, 2011 https://doi.org/10.1172/JCI46122
4. Bhan I, Shah A, Holmes J, Isakova T, Gutierrez O, Burnett SM, Jüppner H, Wolf M: Post-transplant hypophosphatemia: Tertiary “hyper-phosphatoninism”? Kidney Int 70: 1486–1494, 2006 https://doi.org/10.1038/sj.ki.5001788
5. Pilmore H, Dent H, Chang S, McDonald SP, Chadban SJ: Reduction in cardiovascular death after kidney transplantation. Transplantation 89: 851–857, 2010 https://doi.org/10.1097/TP.0b013e3181caeead
6. Sitara D, Razzaque MS, St-Arnaud R, Huang W, Taguchi T, Erben RG, Lanske B: Genetic ablation of vitamin D activation pathway reverses biochemical and skeletal anomalies in Fgf-23-null animals. Am J Pathol 169: 2161–2170, 2006 https://doi.org/10.2353/ajpath.2006.060329
7. Shimada T, Urakawa I, Yamazaki Y, Hasegawa H, Hino R, Yoneya T, Takeuchi Y, Fujita T, Fukumoto S, Yamashita T: FGF-23 transgenic mice demonstrate hypophosphatemic rickets with reduced expression of sodium phosphate cotransporter type IIa. Biochem Biophys Res Commun 314: 409–414, 2004 https://doi.org/10.1016/j.bbrc.2003.12.102
8. Lim K, Lu TS, Molostvov G, Lee C, Lam FT, Zehnder D, Hsiao LL: Vascular Klotho deficiency potentiates the development of human artery calcification and mediates resistance to fibroblast growth factor 23. Circulation 125: 2243–2255, 2012 https://doi.org/10.1161/CIRCULATIONAHA.111.053405
9. Shalhoub V, Shatzen EM, Ward SC, Davis J, Stevens J, Bi V, Renshaw L, Hawkins N, Wang W, Chen C, Tsai MM, Cattley RC, Wronski TJ, Xia X, Li X, Henley C, Eschenberg M, Richards WG: FGF23 neutralization improves chronic kidney disease-associated hyperparathyroidism yet increases mortality. J Clin Invest 122: 2543–2553, 2012 https://doi.org/10.1172/JCI61405
10. Parker BD, Schurgers LJ, Brandenburg VM, Christenson RH, Vermeer C, Ketteler M, Shlipak MG, Whooley MA, Ix JH: The associations of fibroblast growth factor 23 and uncarboxylated matrix Gla protein with mortality in coronary artery disease: The Heart and Soul Study. Ann Intern Med 152: 640–648, 2010 https://doi.org/10.7326/0003-4819-152-10-201005180-00004
11. Mirza MA, Larsson A, Melhus H, Lind L, Larsson TE: Serum intact FGF23 associate with left ventricular mass, hypertrophy and geometry in an elderly population. Atherosclerosis 207: 546–551, 2009 https://doi.org/10.1016/j.atherosclerosis.2009.05.013
12. Mirza MA, Larsson A, Lind L, Larsson TE: Circulating fibroblast growth factor-23 is associated with vascular dysfunction in the community. Atherosclerosis 205: 385–390, 2009 https://doi.org/10.1016/j.atherosclerosis.2009.01.001
13. Inaba M, Okuno S, Imanishi Y, Yamada S, Shioi A, Yamakawa T, Ishimura E, Nishizawa Y: Role of fibroblast growth factor-23 in peripheral vascular calcification in non-diabetic and diabetic hemodialysis patients. Osteoporos Int 17: 1506–1513, 2006 https://doi.org/10.1007/s00198-006-0154-6
14. Lim K, McGregor G, Coggan AR, Lewis GD, Moe SM: Cardiovascular functional changes in chronic kidney disease: Integrative physiology, pathophysiology and applications of cardiopulmonary exercise testing. Front Physiol 11: 572355, 2020 https://doi.org/10.3389/fphys.2020.572355
15. Albouaini K, Egred M, Alahmar A, Wright DJ: Cardiopulmonary exercise testing and its application. Heart 93: 1285–1292, 2007 https://doi.org/10.1136/hrt.2007.121558
16. Mancini DM, Eisen H, Kussmaul W, Mull R, Edmunds LH Jr, Wilson JR: Value of peak exercise oxygen consumption for optimal timing of cardiac transplantation in ambulatory patients with heart failure. Circulation 83: 778–786, 1991 https://doi.org/10.1161/01.CIR.83.3.778
17. American Thoracic Society; American College of Chest Physicians: ATS/ACCP statement on cardiopulmonary exercise testing. Am J Respir Crit Care Med 167: 211–277, 2003 https://doi.org/10.1164/rccm.167.2.211
18. Hunt SA: ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure). J Am Coll Cardiol 46: e1–82, 2005 https://doi.org/10.1016/j.jacc.2005.08.022
19. Arena R, Sietsema KE: Cardiopulmonary exercise testing in the clinical evaluation of patients with heart and lung disease. Circulation 123: 668–680, 2011 https://doi.org/10.1161/CIRCULATIONAHA.109.914788
20. Nixon PA, Orenstein DM, Kelsey SF, Doershuk CF: The prognostic value of exercise testing in patients with cystic fibrosis. N Engl J Med 327: 1785–1788, 1992 https://doi.org/10.1056/NEJM199212173272504
21. Oga T, Nishimura K, Tsukino M, Sato S, Hajiro T: Analysis of the factors related to mortality in chronic obstructive pulmonary disease: Role of exercise capacity and health status. Am J Respir Crit Care Med 167: 544–549, 2003 https://doi.org/10.1164/rccm.200206-583OC
22. Gitt AK, Wasserman K, Kilkowski C, Kleemann T, Kilkowski A, Bangert M, Schneider S, Schwarz A, Senges J: Exercise anaerobic threshold and ventilatory efficiency identify heart failure patients for high risk of early death. Circulation 106: 3079–3084, 2002 https://doi.org/10.1161/01.CIR.0000041428.99427.06
23. Older P, Hall A, Hader R: Cardiopulmonary exercise testing as a screening test for perioperative management of major surgery in the elderly. Chest 116: 355–362, 1999 https://doi.org/10.1378/chest.116.2.355
24. Ting SM, Iqbal H, Hamborg T, Imray CH, Hewins S, Banerjee P, Bland R, Higgins R, Zehnder D: Reduced functional measure of cardiovascular reserve predicts admission to critical care unit following kidney transplantation. PLoS One 8: e64335, 2013 https://doi.org/10.1371/journal.pone.0064335
25. Wilson RJ, Davies S, Yates D, Redman J, Stone M: Impaired functional capacity is associated with all-cause mortality after major elective intra-abdominal surgery. Br J Anaesth 105: 297–303, 2010 https://doi.org/10.1093/bja/aeq128
26. Ting SM, Iqbal H, Kanji H, Hamborg T, Aldridge N, Krishnan N, Imray CH, Banerjee P, Bland R, Higgins R, Zehnder D: Functional cardiovascular reserve predicts survival pre-kidney and post-kidney transplantation. J Am Soc Nephrol 25: 187–195, 2014 https://doi.org/10.1681/ASN.2013040348
27. Guazzi M, Bandera F, Ozemek C, Systrom D, Arena R: Cardiopulmonary exercise testing: What is its value? J Am Coll Cardiol 70: 1618–1636, 2017 https://doi.org/10.1016/j.jacc.2017.08.012
28. Lim K, Ting SMS, Hamborg T, McGregor G, Oxborough D, Tomkins C, Xu D, Thadhani R, Lewis G, Bland R, Banerjee P, Fletcher S, Krishnan NS, Higgins R, Zehnder D, Hiemstra TF: Cardiovascular functional reserve before and after kidney transplant. JAMA Cardiol 5: 420–429, 2020 https://doi.org/10.1001/jamacardio.2019.5738
29. Stenvinkel P, Carrero JJ, Axelsson J, Lindholm B, Heimbürger O, Massy Z: Emerging biomarkers for evaluating cardiovascular risk in the chronic kidney disease patient: How do new pieces fit into the uremic puzzle? Clin J Am Soc Nephrol 3: 505–521, 2008 https://doi.org/10.2215/CJN.03670807
30. Park SH, Stenvinkel P, Lindholm B: Cardiovascular biomarkers in chronic kidney disease. J Ren Nutr 22: 120–127, 2012 https://doi.org/10.1053/j.jrn.2011.10.021
31. Burton JO, Jefferies HJ, Selby NM, McIntyre CW: Hemodialysis-induced repetitive myocardial injury results in global and segmental reduction in systolic cardiac function. Clin J Am Soc Nephrol 4: 1925–1931, 2009 https://doi.org/10.2215/CJN.04470709
32. Parfrey PS, Foley RN, Harnett JD, Kent GM, Murray D, Barre PE: Outcome and risk factors of ischemic heart disease in chronic uremia. Kidney Int 49: 1428–1434, 1996 https://doi.org/10.1038/ki.1996.201
33. Wasserman KHJ, Sue DY, Stringer WW, Whipp BJ: Principles of Exercise Testing and Interpretation: Including Pathophysiology and Clinical Applications, 4th Ed., Philadelphia, Lippincott Williams & Wilkins, 2005
34. Guazzi M, Adams V, Conraads V, Halle M, Mezzani A, Vanhees L, Arena R, Fletcher GF, Forman DE, Kitzman DW, Lavie CJ, Myers J; European Association for Cardiovascular Prevention & RehabilitationAmerican Heart Association: EACPR/AHA Scientific Statement. Clinical recommendations for cardiopulmonary exercise testing data assessment in specific patient populations. Circulation 126: 2261–2274, 2012 https://doi.org/10.1161/CIR.0b013e31826fb946
35. Milani RV, Lavie CJ, Mehra MR: Cardiopulmonary exercise testing: How do we differentiate the cause of dyspnea? Circulation 110: e27–e31, 2004 https://doi.org/10.1161/01.CIR.0000136811.45524.2F
36. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, Picard MH, Roman MJ, Seward J, Shanewise JS, Solomon SD, Spencer KT, Sutton MS, Stewart WJ; Chamber Quantification Writing Group; American Society of Echocardiography’s Guidelines and Standards Committee; European Association of Echocardiography: Recommendations for chamber quantification: A report from the American Society of Echocardiography’s Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 18: 1440–1463, 2005 https://doi.org/10.1016/j.echo.2005.10.005
37. Ross RM: ATS/ACCP statement on cardiopulmonary exercise testing. Am J Respir Crit Care Med 167: 1451, author reply 1451, 2003 https://doi.org/10.1164/ajrccm.167.10.950
38. Ting SM, Hamborg T, McGregor G, Oxborough D, Lim K, Koganti S, Aldridge N, Imray C, Bland R, Fletcher S, Krishnan NS, Higgins RM, Townend J, Banerjee P, Zehnder D: Reduced cardiovascular reserve in chronic kidney failure: A matched cohort study. Am J Kidney Dis 66: 274–284, 2015 https://doi.org/10.1053/j.ajkd.2015.02.335
39. London GM: Left ventricular hypertrophy: Why does it happen? Nephrol Dial Transplant 18: viii2–6, 2003 https://doi.org/10.1093/ndt/gfg1083
40. London GM, Guerin AP, Marchais SJ, Pannier B, Safar ME, Day M, Metivier F: Cardiac and arterial interactions in end-stage renal disease. Kidney Int 50: 600–608, 1996 https://doi.org/10.1038/ki.1996.355
41. Ghuman J, Cai X, Patel RB, Khan SS, Hecktman J, Redfield MM, Lewis G, Shah SJ, Wolf M, Isakova T, Mehta R: Fibroblast growth factor 23 and exercise capacity in heart failure with preserved ejection fraction. J Card Fail 27: 309–317, 2021 https://doi.org/10.1016/j.cardfail.2020.09.477
42. Touchberry CD, Green TM, Tchikrizov V, Mannix JE, Mao TF, Carney BW, Girgis M, Vincent RJ, Wetmore LA, Dawn B, Bonewald LF, Stubbs JR, Wacker MJ: FGF23 is a novel regulator of intracellular calcium and cardiac contractility in addition to cardiac hypertrophy. Am J Physiol Endocrinol Metab 304: E863–E873, 2013 https://doi.org/10.1152/ajpendo.00596.2012
43. Desjardins L, Liabeuf S, Renard C, Lenglet A, Lemke HD, Choukroun G, Drueke TB, Massy ZA; European Uremic Toxin (EUTox) Work Group: FGF23 is independently associated with vascular calcification but not bone mineral density in patients at various CKD stages. Osteoporos Int 23: 2017–2025, 2012 https://doi.org/10.1007/s00198-011-1838-0
44. Richter B, Faul C: FGF23 actions on target issues—With and without klotho. Front Endocrinol (Lausanne) 9: 189, 2018 https://doi.org/10.3389/fendo.2018.00189
45. Zhou J, Dhakal K, Yi J: Mitochondrial Ca(2+) uptake in skeletal muscle health and disease. Sci China Life Sci 59: 770–776, 2016 https://doi.org/10.1007/s11427-016-5089-3
46. Takemura K, Nishi H, Inagi R: Mitochondrial dysfunction in kidney disease and uremic sarcopenia. Front Physiol 11: 565023, 2020 https://doi.org/10.3389/fphys.2020.565023
47. Glancy B, Willis WT, Chess DJ, Balaban RS: Effect of calcium on the oxidative phosphorylation cascade in skeletal muscle mitochondria. Biochemistry 52: 2793–2809, 2013 https://doi.org/10.1021/bi3015983
48. Yazdi PG, Moradi H, Yang JY, Wang PH, Vaziri ND: Skeletal muscle mitochondrial depletion and dysfunction in chronic kidney disease. Int J Clin Exp Med 6: 532–539, 2013
49. Lam NN, Kim SJ, Knoll GA, McArthur E, Lentine KL, Naylor KL, Li AH, Shariff SZ, Ribic CM, Garg AX: The risk of cardiovascular disease is not increasing over time despite aging and higher comorbidity burden of kidney transplant recipients. Transplantation 101: 588–596, 2017 https://doi.org/10.1097/TP.0000000000001155
50. Di Marco GS, Reuter S, Kentrup D, Grabner A, Amaral AP, Fobker M, Stypmann J, Pavenstädt H, Wolf M, Faul C, Brand M: Treatment of established left ventricular hypertrophy with fibroblast growth factor receptor blockade in an animal model of CKD. Nephrol Dial Transplant 29: 2028–2035, 2014 https://doi.org/10.1093/ndt/gfu190
51. Pickup LC, Law JP, Radhakrishnan A, Price AM, Loutradis C, Smith TO, Edwards NC, Steeds RP, Townend JN, Ferro CJ: Changes in left ventricular structure and function associated with renal transplantation: A systematic review and meta-analysis. ESC Heart Fail 8: 2045–2057, 2021 https://doi.org/10.1002/ehf2.13283
52. Hernández D, Ruiz-Esteban P, Gaitán D, Burgos D, Mazuecos A, Collantes R, Briceño E, Palma E, Cabello M, González-Molina M, De Mora M: Regression of cardiac growth in kidney transplant recipients using anti-m-TOR drugs plus RAS blockers: A controlled longitudinal study. BMC Nephrol 15: 65, 2014 https://doi.org/10.1186/1471-2369-15-65
53. Paoletti E, Bellino D, Signori A, Pieracci L, Marsano L, Russo R, Massarino F, Ravera M, Fontana I, Carta A, Cassottana P, Garibotto G: Regression of asymptomatic cardiomyopathy and clinical outcome of renal transplant recipients: A long-term prospective cohort study. Nephrol Dial Transplant 31: 1168–1174, 2016 https://doi.org/10.1093/ndt/gfv354
54. Becker-Cohen R, Nir A, Ben-Shalom E, Rinat C, Feinstein S, Farber B, Frishberg Y: Improved left ventricular mass index in children after renal transplantation. Pediatr Nephrol 23: 1545–1550, 2008 https://doi.org/10.1007/s00467-008-0855-5
55. Patel RK, Mark PB, Johnston N, McGregor E, Dargie HJ, Jardine AG: Renal transplantation is not associated with regression of left ventricular hypertrophy: A magnetic resonance study. Clin J Am Soc Nephrol 3: 1807–1811, 2008 https://doi.org/10.2215/CJN.01400308
56. Bienaimé F, Dechartres A, Anglicheau D, Sabbah L, Montgermont P, Friedlander G, Ravaud P, Legendre C, Prié D: The association between fibroblast growth factor 23 and renal transplantation outcome is modified by follow-up duration and glomerular filtration rate assessment method. Kidney Int Rep 2: 881–892, 2017 https://doi.org/10.1016/j.ekir.2017.05.007
57. Coe LM, Madathil SV, Casu C, Lanske B, Rivella S, Sitara D: FGF-23 is a negative regulator of prenatal and postnatal erythropoiesis. J Biol Chem 289: 9795–9810, 2014 https://doi.org/10.1074/jbc.M113.527150
58. Robach P, Calbet JA, Thomsen JJ, Boushel R, Mollard P, Rasmussen P, Lundby C: The ergogenic effect of recombinant human erythropoietin on VO2max depends on the severity of arterial hypoxemia. PLoS One 3: e2996, 2008 https://doi.org/10.1371/journal.pone.0002996
59. Rossignol P, Agarwal R, Canaud B, Charney A, Chatellier G, Craig JC, Cushman WC, Gansevoort RT, Fellström B, Garza D, Guzman N, Holtkamp FA, London GM, Massy ZA, Mebazaa A, Mol PGM, Pfeffer MA, Rosenberg Y, Ruilope LM, Seltzer J, Shah AM, Shah S, Singh B, Stefánsson BV, Stockbridge N, Stough WG, Thygesen K, Walsh M, Wanner C, Warnock DG, Wilcox CS, Wittes J, Pitt B, Thompson A, Zannad F: Cardiovascular outcome trials in patients with chronic kidney disease: Challenges associated with selection of patients and endpoints. Eur Heart J 40: 880–886, 2019 https://doi.org/10.1093/eurheartj/ehx209
Keywords:

chronic kidney disease; cardiopulmonary exercise testing (CPET); cardiovascular functional capacity; dialysis; fibroblast growth factor 23 (FGF23); heart failure; kidney transplant; LVMI; VO2Max

Copyright © 2022 by the American Society of Nephrology