Tese de Doutoramento Medicina 2024 Faculdade de Ciências Médicas, Universidade NOVA de Lisboa

SUMMARY The studies included in this thesis contributed to clarify the role of some non-traditional cardiovascular risk factors, such as native vitamin D and serum magnesium, in hemodialysis patients. These biomarkers / mediators consolidate the existence of a strong interconnection between bone and cardiovascular disease in chronic kidney disease. Magnesium also appears to have an important role, until now underestimated, in the synthesis and metabolism of vitamin D. In some clinical situations, magnesium supplementation can reverse the resistance to therapy with native vitamin D, ensuring adequate levels of vitamin D, with a particularly positive impact on cardiovascular mortality. It is therefore essential to ensure adequate levels of magnesium in order to optimize the benefits of vitamin D supplementation in chronic kidney disease patients. We also proposed, for the first time, to evaluate the incidence of bone fractures in a large population of prevalent hemodialysis patients in Portugal and study their possible association with markers of bone disease and cardiovascular risk factors. Main results of the studies included in this thesis: - Deficiency of native vitamin D (calcidiol) and chronic kidney disease This study included 223 patients on hemodialysis who underwent two determinations, separated by 6 months of 25-hydroxyvitamin D 25(OH)D and 1,25-dihydroxyvitamin D 1,25(OH)2D. We found deficient levels (< 15 ng/mL) of 25(OH)D in 33.6% of the patients and insufficient levels (15 - 30 ng/mL) in 45.8% of the patients. We observed a positive association between 25(OH)D and 1,25(OH)2D levels and a negative association between 25(OH)D levels and age, diabetes mellitus, and markers of inflammation such as C-reactive protein. We also identified, for the first time, cardiovascular risk factors associated with insufficiency / deficiency in native vitamin D in hemodialysis patients. In the univariable analysis, serum 25(OH)D levels were negatively associated with brain natriuretic peptide (BNP) levels, with an elevated pulse pressure (≥ 65 mmHg) and with the presence of more vascular calcifications. In the multivariable analysis, low serum 25(OH)D levels were also independently associated with high BNP values, a pulse pressure ≥ 65 mmHg and a vascular calcification score ≥ 3. We demonstrated that baseline serum levels of 25(OH)D appear to be an excellent marker of global and cardiovascular morbidity and mortality in hemodialysis patients. In a 48-month prospective evaluation, we found that baseline serum levels of 25(OH)D were significantly lower in patients who died of all causes and in those who died of cardiovascular causes. Baseline serum 25(OH)D levels were also lower in patients who required hospitalization during the study. Considering these apparent protective effects of 25(OH)D on cardiovascular system dysfunction, we carried out a prospective study with a 5-year follow-up, on 97 hemodialysis patients, in which they were supplemented with oral cholecalciferol during hemodialysis sessions. Given the lack of specific recommendations regarding supplementation doses to be used in hemodialysis patients, we decided to transpose the proposals of the KDOQI guidelines for the pre-dialysis stage of chronic kidney disease to our population. We demonstrated a significant increase in serum 25(OH)D levels after 6 months (23.1 ± 13.2 vs. 39.9 ± 11.1 ng/mL; p < 0.001) and after 60 months (23.1 ± 1 3.2 vs. 44.1 ± 11.9 ng/mL; p < 0.001) of supplementation compared to baseline values. In comparison to the values obtained after 6 and 60 months of cholecalciferol supplementation, there was a slight, but not significant, increase at the end of the study. In 94% of the patients, 25(OH)D levels became normal (> 30 ng/mL) at the end of the study, compared to only 18% at the beginning of the study. Patients whose 25(OH)D levels did not increase were all diabetic and had hypomagnesemia (serum magnesium < 0.7 mmol/L). In clinical practice, we emphasize the importance of excluding magnesium deficiency in patients who, despite high doses, do not respond to native vitamin D supplementation. Serum levels of calcium, phosphorus and intact parathormone (iPTH) were significantly reduced with supplementation. On the contrary, there was an increase in serum magnesium. We also demonstrated a reduction in the number of patients treated with active vitamin D, a decrease in doses of erythropoiesis-stimulating agent, a reduction in C-reactive protein, and an increase in albuminemia. Finally, we found that supplementation with cholecalciferol was associated with a significant reduction in plasma BNP levels, pulse pressure, and left ventricular mass index. The supplementation was well tolerated, with no documented adverse effects (namely hypercalcemia greater than 10.5 mg/dL), and none of the patients achieved serum 25(OH)D levels considered toxic (> 100 ng/mL). These results demonstrated that long-term cholecalciferol supplementation in hemodialysis patients was safe, allowed correction of 25(OH)D deficiency, improved mineral metabolism control, attenuated the inflammatory state, and improved cardiac dysfunction. - Serum magnesium and chronic kidney disease In a 48-month prospective study, performed on 206 prevalent hemodialysis patients, we observed a mean pre-dialysis serum magnesium concentration of 1.36 ± 0.18 mmol/L and that none of the patients presented hypomagnesemia (magnesium < 0.7 mmol/L). In univariable analysis, pre-dialysis serum magnesium levels were negatively associated with age, diabetes mellitus, pulse pressure, left ventricular mass index and vascular calcifications. There was a positive association between serum magnesium and albumin. In multivariable analysis, lower magnesium concentrations were predictors of increased pulse pressure (≥ 65 mmHg) and left ventricular mass index (≥ 140 g/m2), and of a higher vascular calcification score (≥ 3). We also found that patients with serum magnesium < 1.15 mmol/L had lower survival at the end of the 48-month follow-up. Based on these results, we studied the effects of a phosphate binder with magnesium and low calcium concentration (calcium acetate / magnesium carbonate), for the treatment of hyperphosphatemia, whose effects on the cardiovascular system remain to be determined. We evaluated, in a 48-month prospective study, the relationship between calcium acetate / magnesium carbonate therapy and markers of cardiovascular risk, already validated in this population of hemodialysis patients, such as pulse pressure, left ventricular mass index and cardiac valve calcifications. Forty patients were on calcium acetate / magnesium carbonate therapy, 52 patients were on sevelamer and 114 were not taking phosphate binders. Patients taking calcium acetate / magnesium carbonate had higher serum magnesium levels compared to the other two groups. At the end of the study, we found that patients who were on calcium acetate / magnesium carbonate showed a reduction in pulse pressure, left ventricular mass index, and aortic and mitral valve calcifications compared to patients who were not taking phosphate binders. Compared to patients on sevelamer, patients on calcium acetate / magnesium carbonate had a reduction in pulse pressure, left ventricular mass index, and aortic valve calcifications. In multivariable analysis, therapy with calcium acetate / magnesium carbonate was inversely associated with the progression of left ventricular hypertrophy and aortic valve calcifications. In this 48-month prospective study, we observed for the first time that the use of calcium acetate / magnesium carbonate in the control of hyperphosphatemia in hemodialysis patients was associated with a reduction in pulse pressure and left ventricular mass index and stabilization of cardiac valvular calcifications. - Bone fractures: incidence and risk factors in chronic kidney disease We performed a retrospective analysis on 341 prevalent patients on hemodialysis since the beginning of the dialysis technique (median of 51 months). Fifty-seven episodes of fragility fractures were identified, with a median time on hemodialysis of 47 months, which corresponded to an incidence rate of 31 per 1,000 person-years. In univariable analysis, compared to patients without fractures, those who presented fractures were more frequently female and older patients, with longer hemodialysis vintage, diabetics, had a lower body mass index and lower albumin values. Patients who had fractures were less frequently receiving active vitamin D therapy and had a higher vascular calcification score. Therapy with phosphate binders or calcimimetics was not associated with the development of fractures. A significantly higher risk of fractures was also associated with higher bone alkaline phosphatase values and iPTH levels < 300 or > 800 pg/mL, compared to levels between 300 - 800 pg/mL. Analysis of long-term fracture-free survival demonstrated that patients with iPTH < 300 (HR: 2.88) or iPTH > 800 pg/mL (HR: 3.48) had a significantly higher risk of fragility fractures. In multivariable analysis, age, female gender, longer time on dialysis, lower serum albumin levels and the presence of more vascular calcifications were independently associated with a higher risk of bone fractures. On the other hand, active vitamin D therapy was associated with a decreased risk. In this study, with a long follow-up, we observed that the incidence of fragility bone fractures in prevalent hemodialysis patients was high, and their risk appears to increase with age, female gender, presence of inflammation and more vascular calcifications. Bone disease associated with uremia also appeared to be an important factor in the development of fractures, as patients with lower or higher iPTH levels and increased bone alkaline phosphatase had a higher risk of fracture. On the other hand, active vitamin D therapy appeared to have a protective effect