Abstract BACKGROUND AND AIMS Recently, a growing body of evidence point to the relationships between inflammation and fibroblast growth factor 23 (FGF23) in adults with and without chronic kidney disease (CKD). On the other hand, some haematological indices derived from the complete blood count appear as simple and inexpensive biomarkers of systemic inflammation. Therefore, this study aimed to evaluate the association between the haematological markers of inflammation and FGF23 in non-dialysis CKD. METHOD This single centre, cross-sectional study prospectively enrolled 90 subjects with moderate to severe stable CKD [median estimated glomerular filtration rate (eGFR) 25 (95% confidence interval 95% CI 24.9–30.5) mL/min, 46% in stage G3, urinary albumin-to-creatinine ratio (ACR) 221 (95% CI 498–954) mg/g], mostly males (61%), aged 62 (95% CI 57–64) years. Patients on renal replacement therapy (RRT) and those with active malignancies, infectious and inflammatory diseases were excluded. Demographic, past medical history (CKD vintage and etiology, comorbidities, chronic medications) and laboratory data were collected. Haematological markers of systemic inflammation [red blood cell distribution width (RDW), platelet distribution width (PDW), neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR)] were obtained from the complete blood count. Other measured lab parameters were: c-terminal FGF23 (cFGF23), intact parathyroid hormone (iPTH), serum calcidiol (25OHD), total calcium (tCa), phosphate (PO4), alkaline phosphatase, C-reactive protein, albumin, transferrin, ferritin and transferrin saturation, as markers of mineral metabolism, inflammatory, nutritional and iron status, respectively. Associations among studied parameters were assessed by Spearman rank test, multivariate linear regression and logistic regression models. Non-parametric variables were log-transformed. RESULTS The median cFGF23 was 4.65 (95% CI 7.57–14.4) pg/mL. Subjects in the highest cFGF23 quartile [n = 22, cFGF23 20.8 (95% CI 20.8–40.8) pg/mL] had higher RDW (P = 0.02), iPTH (P < 0.001), PO4 (P = 0.005) and ACR (P = 0.004), but lower tCa (P = 0.04), haemoglobin (P = 0.01) and eGFR (P < 0.001) as compared to those in the lowest quartile [n = 23, cFGF23 0.84 (95% CI 0.71–1.10) pg/mL]. Moreover, they had higher proportions of arterial hypertension (P = 0.01) and previous treatment with intravenous iron (P = 0.03). A trend to higher PLR (P = 0.052), NLR (P = 0.064), and ferritin (P = 0.076) was also observed. However, beside the expected bivariate correlations with eGFR, ACR, iPTH, tCa and PO4, cFGF23 was further correlated solely with haemoglobin (rs = –0.33, P = 0.002). In a multivariate linear regression model which explained 34% of the log(cFGF23) variance, log(PLR) (B 1.59, 95% CI 0.14–3.03, P = 0.03) and log(eGFR) (B –1.65, 95% CI –2.51 to –0.80, P < 0.001) were independent predictors. After adjustment for history of arterial hypertension and previous intravenous iron treatment, in a model of logistic regression (Nagelkerke R2 0.57, Chi2 16.7, P = 0.03), log(eGFR) was found as independent predictor of the log(FGF23) [odds ratio (OR) 0.07, 95% CI 0.01–0.36, P = 0.002]. However, RDW was retained in the last step of this model with a borderline significance (OR 4.3, 95% CI 0.98–19.02, P = 0.052). CONCLUSION Taken together, our findings suggest that, in addition to the kidney function, RDW and PLR seem to predict cFGF23 in non-inflammed, stable, non-dialysis CKD adults. Thus, one might speculate in favour of the association between chronic inflammatory state and FGF23 metabolism and in support of the revelatory role of some haematological markers of inflammation, a hypothesis that needs further research.