Renal Iron Load in Sickle Cell Disease Correlates with Hemolysis and Transfusion History, but Not with Hepatic Iron

Sickle cell disease (SCD) is characterised by chronic hemolysis and frequent, painful vaso-occlusive episodes, eventually leading to organ damage. Overall, renal failure is found in approximately 11% of SCD patients, and incidence increases with age. Both the vaso-occlusive and hemolytic aspects of SCD have been implicated in renal failure, as assessed by glomerular filtration rate (GFR) and microalbuminurea. Additionally, the increasing use of blood transfusion therapy in SCD has resulted in a sub-population with secondary iron overload, also leading to organ damage. Differential organ iron loading has been noted in the two major hemoglobin disorders; cardiac iron toxicity is common in β-thalassemia patients with high liver iron concentration (LIC), whereas this is uncommon in SCD patients for an equivalent LIC. Relatively limited research has been carried out regarding renal iron loading in SCD. This study investigates the clinical significance and etiology of renal iron accumulation in SCD patients.

41 patients from the specialist hematology clinic at King's College Hospital, London had already had an assessment of hepatic iron loading by spin density projection assisted R2-MRI (FerriScan^®) as part of their clinical care programme and/or as part of another study approved by the NHS Research Ethics Committee (REC 05/Q0703/21). One patient was excluded due to prior chelation therapy; the remaining 40 were chelation naïve. Owing to the retrospective nature of the study, the REC confirmed that informed consent was not required of patients for this study. Renal R2 (R-R2) values were derived from monoexponential fits to kidney image data obtained during LIC assessment in order to assess kidney iron concentration. R-R2 data were also collected from 17 healthy control participants to provide a reference range. Control data were acquired with approval from the Fremantle Hospital Human Research Ethics Committee (08/404) and the University of Western Australia Human Research Ethics Committee.

Clinical data were collected retrospectively for the 2 year period prior to the MRI scan. Thresholds for renal hyperfiltration and microalbuminurea were MDRD eGFR ≥140 ml/min/1.73m² and albumin:creatinine ratio ≥4.5 mg/mmol, respectively. Lifetime blood transfusion history (top-up units) was recorded up to the date of the R2-MRI scan.

Mean R-R2 (expressed in units of s⁻¹) was significantly higher (p<0.0001) in the SCD patients (26.87 ± SD 8.89 s⁻¹) compared with healthy controls (17.77 ± SD 2.94 s⁻¹). R-R2 values correlated with transfusion history (r²=0.23, p=0.004) but not with LIC. Highly significant correlations were seen between R-R2 and bilirubin (r²=0.39, p<0.0001) and lactate dehydrogenase levels (r²=0.37, p<0.0001), known markers of hemolysis. SCD patients with co-inherited α thalassemia had a lower mean R2 (23.45 ± SD 6.1 s⁻¹) than those without (28.94 ± 9.6 SD s⁻¹), which approached significance (p=0.070).

Patients were stratified into those with evidence of renal hyperfiltration and those without. Mean R-R2 was significantly higher (p=0.019) in patients with renal hyperfiltration (according to the described MDRD eGFR criteria) than those without (30.48 ± SD 6.6 s⁻¹vs. 23.3 ± SD 10.3 s⁻¹, respectively). A weak but significant negative correlation was observed between renal R2 and age (r²=0.12, p=0.03). The renal hyperfiltration group were also significantly younger (p=0.003) than those with normal filtration rates (29.8 ± SD 11.26 vs. 42.3 ± SD 11.24 years). No significant relationship was observed between R-R2 and microalbuminurea.

We showed that R-R2 is not related to liver iron concentration, but does correlate with transfusion history and markers of hemolysis. Given the lack of correlation with LIC, we believe that the correlation with transfusion history is not causative, but rather reflects the increased clinical severity; those patients with more severe disease and anemia requiring more frequent transfusions. Though not significant, mean renal R2 was noticeably lower in patients who have co-inherited α-thalassemia, in keeping with the reduced hemolysis and protective effect of α-thalassemia on renal impairment in SCD.

St Pierre:Resonance Health: Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Consultancy, Research Funding, Speakers Bureau.