POST-MORTEM HEPATIC AND BONE MARROW IRON CONTENT IN HEMODIALYSIS PATIENTS: A PROSPECTIVE COHORT STUDY

Abstract Background and Aims Studies using T2 MRI liver scans among Hemodialysis (HD) patients raised concern about the presence of iron overload in this population, regularly treated with intravenous (IV) iron. Histological evidence of tissue iron overload is scarce, since the majority of studies were performed in pre-erythropoiesis- stimulating agents (ESA’s) era, when blood transfusions were common. Primary objective: to quantify iron in the liver and bone marrow by biochemical and histological analysis, in adult CKD stage 5-HD. Secondary objectives: To explore association of clinical, laboratorial parameters, IV iron therapy and iron stores. Method After approval of local Ethical committee and informed consent from families, liver biopsy and bone marrow aspirate were obtained in the first 24h post-mortem from 21 chronic HD patients with anemia or under anemia treatment who died in Hospital Fernando Fonseca. Exclusion criteria: blood transfusion in the previous 2 weeks, acute or chronic liver disease, HIV infection, known hematologic or oncologic disease. Clinical, laboratorial and anemia therapy data were retrieved from hospital registry and outpatient HD centers. Biochemical liver iron content (LIC) was quantified by atomic absorption spectrophotometry. Histological semi-quantitative grading of iron storage was made in the liver and bone marrow using Scheuer’s and Gale’s criteria of grading Perls’ stain, respectively. Results Of 21 patients included, 10 (47,6%) were male, median (IQR) age 76.0 (67.5-85.5) years old, 18 (85.7%) white, 3 (14.3%) black, dialysis vintage was 47.0 (12.5-104.0) months. Charlson Comorbidity index was 10.0 (7.5-11.0), 7 (33%) patients had diabetes, and 11 (52.4%) used an arteriovenous fistula as vascular access. The cause of death was infection (n=9, 42.9%), cardiovascular (N=6, 28.6%), HD withdrawal (n=2, 9.5%) and unknown =3 (14.3%). Median (IQR) hemoglobin was 9.8 (8.5-11.4) g/dl and 11 (52.3%) patients had hemoglobin <10 g/dl. Ferritin was 494.0 (136.0-850.5) ng/ml and TSAT 19.9 % (13.3-26.0). 19 (90.5%) patients were receiving IV iron therapy. Median (IQR) IV iron administered in the previous 6 and 12 months before death was 800 (300-1250) mg and 1500 (650-2175) mg, respectively. All patients were on ESA therapy, median (IQR) dose 5000 (3000-9000) UI/week and erythropoietin resistance index was 9.6 (4.2-16.6). Median (IQR) liver iron content determined by atomic absorption was 42.5 (22.9-69.7) µmol/g. 9 patients (42.9%) had normal LIC (<36 μmol/g), while the remainder had mild to moderate overload. Median (IQR) Scheuer grade was 2 (1-3) and 13 (62%) of liver biopsies had increased (Scheuer grade > 1) iron deposition at histology. Median (IQR) grade of Perls staining in the bone marrow was 3 (3-4) and 9 (45%) had increased (Gale’s grade >3) iron content in the bone marrow. Iron semi-quantitative scores in liver and bone marrow had strong positive correlation (r=0.71, p<0.001). There was a strong positive correlation between LIC and ferritin (r=0.86, p < 0.001) and also TSAT (r=0.56, n=16, p=0.02). Hemoglobin was negatively associated with LIC (r= -0.46, p=0.04), and with iron content in the bone marrow (p=0.04). LIC did not associate with ESA dose, C-reactive protein, dialysis vintage or other clinical parameters. There was no statistically significant association between the dose of IV iron administered in the previous 6 and 12 months with LIC, ferritin,TSAT or iron scores in bone marrow and liver. Conclusion In these HD patients, there was biochemical and histological evidence of iron accumulation in liver and bone marrow. Ferritin and TSAT showed strong correlation with iron deposits, but none was found with the dose of IV iron administered. In this study, anemia severity was associated with higher degree of iron storage both in the liver and bone marrow, suggesting a multilevel blocking mechanism of iron’s utilization.