Association between fibroblast growth factor 23 and left ventricular hypertrophy in Egyptian patients on regular hemodialysis

Abstract Background Cardiovascular disease remains the leading cause of mortality in end-stage renal disease (ESRD) patients, with left ventricular hypertrophy (LVH) representing a prevalent and prognostically significant complication. Fibroblast growth factor 23 (FGF23), a bone-derived phosphaturic hormone markedly elevated in chronic kidney disease, has been implicated in the pathogenesis of cardiovascular abnormalities, though its clinical relevance in specific populations warrants further investigation. Aim To investigate the association between serum FGF23 levels and echocardiographically-defined LVH and systolic function in Egyptian patients undergoing maintenance hemodialysis, while concurrently evaluating traditional cardiovascular risk factors in this context. Methods A hospital-based cross-sectional study was conducted at Ain Shams University Hospital from April 2022 to December 2023. Seventy-one adult ESRD patients on regular hemodialysis were enrolled. Participants were stratified into two groups based on the median serum FGF23 level (263.99 pg/mL): low-FGF23 (≤ 263.99 pg/mL, n = 35) and high-FGF23 (> 263.99 pg/mL, n = 36). All participants underwent comprehensive evaluation including clinical assessment, laboratory investigations (with emphasis on intact FGF23 measured by ELISA), and transthoracic echocardiography. Results Among 71 patients, the overall prevalence of LVH (by qualitative echocardiographic report) was 55.7%. Patients in the high-FGF23 group had a numerically higher prevalence of LVH than the low-FGF23 group (62.9% vs. 48.6%), but this difference was not statistically significant ( p = 0.336). Receiver operating characteristic analysis showed poor discriminatory power of FGF23 for detecting LVH (AUC = 0.551, 95% CI: 0.413–0.689, p = 0.467). FGF23 as a continuous variable showed no significant univariate association with LVH (OR = 1.000, 95% CI: 0.999–1.002, p = 0.730). On univariate screening of 20 candidate variables, only male sex (OR = 3.25, 95% CI: 1.21–8.69, p = 0.019), diabetes mellitus (OR = 3.62, 95% CI: 1.15–11.43, p = 0.029), hypertension (Fisher’s exact p = 0.034), and ferritin (OR = 0.999, 95% CI: 0.998–1.0001, p = 0.077) met the p < 0.10 threshold for multivariable entry. In the model of the remaining three variables, male sex approached but did not reach significance after adjustment (OR = 2.69, 95% CI: 0.96–7.49, p = 0.059), and diabetes mellitus and ferritin were not independently significant. In a sensitivity analysis forcing FGF23 into the adjusted model, FGF23 remained non-significant (OR = 1.001, 95% CI: 0.999–1.003, p = 0.315). Conclusion A numerically higher LVH prevalence was observed in patients with higher FGF23 levels, but this was not statistically significant, and FGF23 demonstrated poor diagnostic performance for LVH in this cohort. On univariate screening, male sex, diabetes mellitus, hypertension, and ferritin were associated with LVH; after mutual adjustment, only male sex approached significance. Hypertension showed a significant association by Fisher’s exact test but could not be modeled by standard logistic regression due to the very small number of non-hypertensive patients. These findings should be interpreted as hypothesis-generating given the modest sample size and cross-sectional design.