Statistics from Altmetric.com
Editor,—Distante and colleagues (
) found that C282Y homozygotes detected by testing all patients acutely admitted to hospital for non-liver associated problems had a considerably lower transferrin saturation during the acute illness (median 27%) compared with at follow up (median 71%). Indeed their strategy of performing C282Y genotyping on all acute patients with a raised transferrin saturation (>50%) would have detected only 1/14 (8%) C282Y homozygotes. This is in contrast with our study of HFE (C282Y) gene mutation and phenotypic expression in outpatients referred to a liver clinic.1 In this setting we found that transferrin saturation was a far more reliable predictor of genetic haemochromatosis than that reported by Distante et al in acutely ill inpatients. A strategy of performing C282Y genotyping on our outpatients with a raised transferrin saturation (>50%) would have detected 8/9 (89%) C282Y homozygotes in our study but only 8% of acutely ill C282Y homozygotes in the Norwegian study. In contrast, a false positive elevation of serum transferrin saturation and hepatic iron index have been reported in patients with end stage liver disease undergoing transplantation who did not have genetic haemochromatosis.2
Distante et al identified 14 homozygotes from 2027 hospitalised patients screened (0.74%), of whom 94% were Caucasian, from a high prevalence area for the HFE gene mutation. In our ethnically mixed population of 326 unselected liver outpatients from London, 79% were Caucasian. As Distante et al conclude, a serum transferrin saturation remains the screening test of choice for detecting haemochromatosis. However, we feel it needs emphasising, in the light of their findings, that screening or opportunistic testing for haemochromatosis is best restricted to relatively well subjects in the outpatient setting. Moreover, selecting target populations (such as liver2 or diabetic clinics3) is likely to identify larger numbers of homozygotes than testing all comers.
Editor,—We thank Drs Moodie and Maxwell for their comments on our paper.
Indeed, it ought to be stressed that desaturation of transferrin occurs in inflammatory states. This should be taken into account to avoid false negative results when screening for haemochromatosis both in the general population and in target groups (such as in liver or diabetic clinics).
In the London outpatient group,1-1 transferrin saturation >50% detected 89% of C282Y homozygotes. In the series of 312 liver disease patients admitted at our hospital,1-2 14/18 (78%) liver biopsy proved haemochromatosis patients had an increased transferrin saturation >50%. In line with Cotler and colleagues,1-3 we also found a false positive elevation of transferrin saturation in 27/105 (26%) patients with alcoholic liver disease and 11/132 (8.3%) with chronic non-alcoholic liver disease.
We agree with Drs Moodie and Maxwell's suggestion that transferrin saturation screening for haemochromatosis should be performed in relatively well subjects. Both their comments and our findings highlight some of the limitations of transferrin saturation as a screening parameter for genetic haemochromatosis.