Initial serum ferritin predicts number of therapeutic phlebotomies to iron depletion in secondary iron overload.

BACKGROUND

Therapeutic phlebotomy is increasingly used in patients with transfusional siderosis to mitigate organ injury associated with iron overload (IO). Laboratory response variables and therapy duration are not well characterized in such patients.

METHODS

We retrospectively evaluated 99 consecutive patients undergoing therapeutic phlebotomy for either transfusional IO (TIO, n = 88; 76% had undergone hematopoietic transplantation) or nontransfusional indications (hyperferritinemia or erythrocytosis; n = 11). Complete blood cell count, serum ferritin (SF), transferrin saturation, and transaminases were measured serially. Phlebotomy goal was an SF level of less than 300 μg/L.

RESULTS

Mean SF levels before phlebotomy among TIO and nontransfusional subjects were 3093 and 396 μg/L, respectively. Transfusion burden in the TIO group was 94 ± 108 (mean ± SD) RBC units; approximately half completed therapy with 24 ± 23 phlebotomies (range, 1-103). One-third were lost to follow-up. Overall, 15% had mild adverse effects, including headache, nausea, and dizziness, mainly during first phlebotomy. Prior transfusion burden correlated poorly with initial ferritin and total number of phlebotomies to target in the TIO group. However, number of phlebotomies to target was strongly correlated with initial SF (R(2) = 0.8; p < 0.0001) in both TIO and nontransfusional groups. ALT decreased significantly with serial phlebotomy in all groups (mean initial and final values, 61 and 39 U/L; p = 0.03).

CONCLUSIONS

Initial SF but not transfusion burden predicted number of phlebotomies to target in patients with TIO. Despite good treatment tolerance, significant losses to follow-up were noted. Providing patients with an estimated phlebotomy number and follow-up duration, and thus a finite endpoint, may improve compliance. Hepatic function improved with iron offloading.