Iron overload has increasingly been recognized as a cause of morbidity and mortality among allogenic hematopoietic stem cell transplant (HSCT) recipients. Estimation of liver iron is considered the gold standard method that reflects the overall body iron burden.
We attempted to correlate the severity of liver iron overload to various clinical parameters in allogenic HSCT recipients.
We studied 49 adult allogenic HSCT recipients who had a minimum of one liver biopsy performed after transplant. All patients had been diagnosed, transplanted and followed up in our institution since 1998. Biopsies were performed due to abnormal liver function tests under the initial clinical suspicion of GVHD with a mean of 185 days (range: 23-844 days) post-transplant. Sections from each biopsy were stained with hematoxylin and eosin, Masson trichome and Perl's iron stains. We used the histological semiquantitative grading of liver iron (grade 0-4), evidence of GVHD (score 0-3) and fibrosis (stage 0-4). Iron overload was defined as a score of 2 and above (correlating with hepatic iron concentration >1.4 mg/g dry weight). Various clinical parameters were also assessed including age, sex, underlying malignancy, number of transfusions, source and type of stem cell transplant, liver function tests, clinical grading of GVHD, significant bacteremia, other infections, survival and cause of death.
In this cohort, 28 of 49 patients had iron overload. The overall prevalence of iron overload was 57% (95% confidence interval, 42%-71%). Iron overload in 7 of the 49 patients (14%) was the only histological explanation of liver dysfunction. The degree of iron overload was directly correlated to the number of transfusions (r=0.69, p<0.001), serum ferritin (r=0.55, p=0.004) and the development of significant bacteremia (r=0.33, p=0.019). The correlation to the histologic liver-GVHD grade was close to statistical significance (p=0.062). 37 of the 49 patients (76%) died after HSCT at a mean of 677.8 days (range: 52-3821 days) post-transplant. The overall mortality did not correlate statistically to the degree of iron overload. The mortality was higher in patients with high serum bilirubin (direct or total) and increasing clinical degree of acute GVHD. Sixteen patients died secondary to infection (bacterial=15, fungal=1). However, there was no significant evidence of increased risk of infection-related mortality with iron overload (Hazard Ratio = 1.16, 95% CI, 0.76-1.77, p = 0.496).
Iron overload is a common adverse effect of allogenic HSCT (57% prevalence). Iron overload occurred in some of our patients after as few as 10 red blood cell units. In this cohort, iron overload statistically correlated to the number of transfusions, development of significant bacteremia and clinical acute GVHD. We did not find a statistically significant association with mortality, likely due to our small sample size and complex clinical picture. Iron overload was the only histological explanation of abnormal liver functions in 14% of our patients. More studies are needed to determine the impact of iron overload in long-term morbidity and mortality in allogenic transplant patients.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.