Association Between Increased Liver Iron Concentration and Vitamin D Deficiency in Patients with Transfusion Dependent Hemoglobinopathies in British Columbia

Patients with transfusion dependent (TD) hemoglobinopathies are at risk of endocrinopathies and bone disease due to iron overload (IOL). Vitamin D is involved in regulation of calcium homeostasis, bone health and other clinical endpoints. Vitamin D deficiency is a common manifestation of patients with transfusional IOL; the mechanism remains unclear. Vitamin D hydroxylation occurs in the liver; whether liver IOL interferes with this step has not to date been addressed. This study investigates an association between liver iron concentration (LIC) and vitamin D levels in patients with TD hemoglobinopathies.

Patients with TD b thalassemia major (TM), hemoglobin Eb TM (Eβ TM), and congenital dyserythropoetic anaemia (CDA) attending the Inherited Bleeding and Red Blood Cell Disorder Program at the adult (St. Paul's Hospital) or pediatric (BC Children's Hospital) programs in Vancouver, Canada were included if they had an assessment of LIC done by magnetic resonance imaging (MRI) and 25-hydroxyvitamin D3 (25 OH D3) levels between January 2009 and the 31^st of December 2011. The analysis was restricted to patients >16 years of age to minimize confounding due to body mass. Clinical data collected included: gender, ethnicity, type of iron chelation therapy (ICT) and vitamin D and calcium supplementation. Laboratory data included: serum ferritin level, 25 OH D3, phosphorus (PO⁴), ionized calcium (Ca²⁺), parathyroid hormone (PTH) and thyroid stimulating hormone (TSH). Vitamin D insufficiency was: 25 OH D3 level <75 nmol/L and deficiency <60 nmol/L. Liver and cardiac iron assessment was done by T2* or R2* MRI. Liver iron was: LIC <2 mg/gmDW, normal; 2–5 mg/gmDW, mild; 5.1–7 mg/gmDW, moderate; >7 mg/gmDW severe. Cardiac IOL was: >20 ms, normal; 14–20 ms, mild; 8–15 ms, moderate; <8 ms, severe. Bone mineral density (BMD) was assessed by dual emission X-ray absorptiometry (DEXA) scans of the hip. BMD was: Z score <−1.5, osteopenia; and <−2.0, osteoperosis. Associations between variables were assessed using the Chi-square or Fischer's exact test and the Pearson correlation.

26 patients with TD: βTM, n=21; Eβ TM, n=3; and CDA, n=2 were included, 21 adults and 5 from the pediatric program. Patients characteristics were: median age 24 (range 16–51) years and 11 (42.3%) were male. Ethnicities were: Asian, 11 (42.3%); Indian, 6 (23%); unclear ethnicity, 5 (19.2%); Middle Eastern, 3 (11.5%); and Caucasian, 1 (3.8%). The mean amount of blood received annually was 9633.9 (4677–14,508) mls/year. All patients were receiving ICT: deferasirox (DFX), 16 (61.5%); desferrioxamine (DFO), 8 (30.8%); and combination DFX+DFO, 2 (7.7%). 15 (57.7%) patients received vitamin D and calcium supplementation in variable doses and formulations. Median 25 OH D3 was 66.5 (18–125) nmol/L and vitamin D deficiency/insufficiency occurred in 15 (57.7%) patients: <75 nmol/L, n=15; and <60 nmol/L, n=10. Median Ca²⁺ was 2.37 (2.2–2.52) mmol/L, PO4 1.21 (0.78–1.71) mmol/L, AST 24.5 (16–46) U/L and ALT 18 (range10–50) U/L; all within normal. Median serum ferritin level was 693.5 (223–3553) mcg/L, LIC 4.25 (1.1–31.8) mg/g dry weight (DW) and median cardiac iron concentration was 35 (13–53.9) ms. Median BMD at the hip was −1.25 (−0.2–−3.7). There was a significant association between LIC >5 mg/gDW, and vitamin D level <60 nmol/L (P= 0.034), with 10 (38.5%) patients having an LIC ≥5 mg/gDW, 5 of whom had a vitamin D level <60 nmol/L. There was an inverse correlation between LIC and vitamin D level, with a Pearson coefficient of −0.33. There was no significant association between: vitamin D level and age, ethnicity, degree of transfusion requirement (DTR), ferritin level or type of ICT; or LIC and age, DTR, ferritin or BMD (P=NS for all).

In this single center study of patients with transfusion dependent hemoglobinopathies, we found a significant association between LIC ≥5 mg/gDW and vitamin D level <60 nmol/L, with an inverse correlation between LIC and vitamin D level, suggesting that liver IOL may affect hepatic vitamin D metabolism and result in deficiency. These results warrant verification in larger numbers of patients and prospective trials and investigation of whether vitamin D insufficiency is linked to clinical endpoints. Patients may require intensification of vitamin D supplementation, and/or intensification of iron chelation therapy, particularly if a link to clinical consequences is demonstrated.

Ezzat:Novartis: Honoraria. Leitch:Novartis: Honoraria, Speakers Bureau, participated in advisory boards Other.