Work

Transfusion History Is the Most Significant Predictor of Iron Overload after Allogeneic Stem Cell Transplantation

Year: 2016

Type: article

Abstract: Iron overload due to prior red blood cell transfusions is a late complication of allogeneic stem cell transplantation (SCT). Traditionally, serum ferritin has been used as a surrogate measure of iron overload. Ferritin is an acute phase reactant and levels can be elevated for a number of reasons other than iron overload including inflammation and infection. A better and simpler way to identify patients with significant iron overload who may benefit from iron removal therapy is therefore needed. Transferrin saturation and number of units of red cells previously transfused may be other simple variables that could be used to predict iron overload in recipients of allogeneic SCT. In this study we sought to compare the relative value of ferritin, transferrin saturation and number of prior transfusions as predictors of iron overload. We studied 53 long term survivors after allogeneic SCT with suspected iron overload indicated by high serum ferritin. Studies were done in post-transplant patients in remission not requiring any transfusions. We used MRI measurement of liver iron concentration (LIC) as a gold standard for a diagnosis of iron overload. Serum ferritin, transferrin saturation, and lifetime total packed red blood cell (PBRC) transfusions at the time of the MRI study were available for 48 of the 53 (91%) patients. Liver iron concentration of 7mg/g (LIC>7) of dry weight or more results in hepatic fibrosis; therefore, we used 7 mg/g as a cutoff value to define significant iron overload. We used Receiver Operating Characteristic (ROC) curve analysis to determine the predictive value of serum ferritin, transferrin saturation, and number of PBRC transfusions (units) in relation to LIC>7 mg/g. Median age was 55 (20-68), and there were 22 males and 31 females. Underlying diagnosis included AML (n=26), MDS (n=9), MPD (n=10) and other diseases (n=8). Summary statistics (median and range) for all the relevant parameters were as follows: ferritin 1950 ng/ml (range 480-8217), transferrin saturation 0.5 mg/dl (0.1-1), number of PRBC units transfused 48 (3-176). Considering ferritin, transferrin saturation, and number of transfusions as continuous variables, the ROC curve analysis showed the number of PRBC transfusions to be the most significant predictor of iron overload. The area under the curve for number of PRBC transfusions, ferritin and transferrin saturation were 0.95, 0.83, and 0.63 respectively. The prevalence of LIC>7 mg/g increased with increasing number of transfusions. It was 7%, 80%, and 100% in patients receiving less than 25 (n=14), 25-50 (n=15), and >50 (n=24) transfusions, respectively. After allogeneic transplantation, compared to serum ferritin and transferrin saturation, the number of PRBC units previously transfused is the best predictor liver iron concentration and thus iron overload.

Source: Transplantation and Cellular Therapy

DOI: 10.1016/j.bbmt.2015.11.820