Tissue Iron Overload Is Common Post Transplantation (Allo BMT) and Is Associated with Red Cell Transfusion Load and HFE Genotype.

In order to analyse the incidence of iron overload after allo BMT and assess the role of venesection in preventing complications, we retrospectively analysed 168 consecutive patients undergoing allo BMT at our institution from 1998–2003 surviving at least one year. Iron studies were performed routinely pre-BMT, at D100, one & two years post BMT. Iron overload was defined by at least one of the following criteria i)liver biopsy (n=24), one of : a) dry weight iron concentration >80μmol/g; b) iron index >1.9; c) Perl’s stain grade 3 or 4, ii) CT liver iron >1.0mg/ml (n=13) iii) raised ferritin >1000 μg/L and transferrin saturation >55% on 2 occasions, persisting >6/12 post BMT (n=11), iv) venesection >5g iron (n=1). Using these criteria, iron overload occurred in 49/168 (29%) pts. 12/119 in the non-overloaded group had further investigation but did not meet the criteria; liver biopsy (n=10) or CT (n=2). Elevated ferritin, particularly early post-transplant, did not reliably predict for iron overload, with 55/91 evaluable patients having values >1000μg/L at D100 not fulfilling the criteria for iron overload. There was no difference between overloaded and non-overloaded patients with respect to age or sex. Acute (15/49 vs. 26/113) or chronic liver GVHD (25/46 vs. 47/105) was not different between the two groups (both p>0.05). Only 3 patients were hepatitis B sAg+ or hepatitis C Ab+. The iron overloaded group was more likely to i) have been transplanted for acute leukaemia (29/49 vs. 33/119; p 0.0002) ii) be C282Y heterozygotes (11/46 vs. 10/110, p 0.02) (iii) been transfused more units of red cells (mean 42 vs. 19; p<0.0001) and iv) have persistently abnormal liver function post-transplant, ALT (IU/ml; normal <55) at 1 year 77 vs. 52 and at 2 years 67 vs. 37 (all p<0.05). There was no effect of hetero- or homozygosity for H63D. 63 patients were analysed for the S65C, V59M and Q283P mutations. One patient was heterozygous for the S65C mutation (non-overloaded group). A mean of 12.3 units were venesected in 22 patients (range 2–46), all of whom had received >25 units of red cells. ALT fell significantly (mean pre venesection 189 IU/ml, post 36, p<0.05), as did transferrin saturation (mean pre venesection 68%, post 29%, p<0.05). We conclude that tissue iron overload is common after BMT, that biochemical measures of iron stores (ferritin and transferrin saturation) may be unreliable in this context, particularly in the early post BMT period and that radiological or histological assessment to distinguish hyperferritinaemia due to inflammation from true tissue iron overload may be required. Patients at risk of iron overload (transfusions >25 units, C282Y heterozygotes) should be closely monitored and early venesection therapy instituted to minimise organ damage.