Occurrence of hemolytic reactions on the same day as immune globulin product administrations during 2008 to 2014

• Published in: Transfusion (Philadelphia, Pa.) Vol. 58; no. 1; pp. 70 - 80
• Main Authors: Sridhar, Gayathri, Ekezue, Bola F., Izurieta, Hector S., Forshee, Richard A., Selvam, Nandini, Mintz, Paul D., Anderson, Steven A., Menis, Mikhail D.
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.01.2018
• Subjects: Anemia; Confidence intervals; Globulins; Hemolysis; Hemolytic anemia; Immunoglobulins; Intravenous administration; Males; Manufacturing industry; Regression analysis; Risk analysis; Risk factors; Sex; Statistical analysis
• ISSN: 0041-1132; 1537-2995
• DOI: 10.1111/trf.14384

BACKGROUND Hemolytic reactions (HRs) are rare serious adverse events after immune globulin (IG) use. Our large claims‐based study evaluated occurrence of same‐day hemolysis after administration of different IG products and potential risk factors, during the 2008 to 2014 study period. STUDY DESIGN AND METHODS We conducted a retrospective cohort study using a large commercial administrative database. The study included individuals exposed to IG products as identified by procedure codes. HRs were ascertained using ICD‐9‐CM diagnosis codes. Unadjusted same‐day hemolysis rates (per 1000 persons) were estimated overall, by age, sex, and IG products. Multivariable regression analyses were used to evaluate potential risk factors. RESULTS Of 20,440 persons exposed, 211 (10.3 per 1000) had same‐day HRs. The median numbers of doses for IG users with versus without same‐day hemolysis were one and six, respectively. The unadjusted product‐specific HR rates ranged from 1.92 for subcutaneous product Hizentra to 17.99 for intravenous Octagam. The multivariable regression analyses showed significantly increased same‐day HR risk in males and in IG users with histories of hemolysis, pneumonia, and hereditary hemolytic anemias. Compared to Gammagard Liquid, significantly elevated overall hemolysis risk was identified with Octagam (odds ratio, 2.36; 95% confidence interval, 1.04‐5.35), using Firth's method to account for small sample size bias. CONCLUSION The study showed variation in the same‐day IG‐related hemolysis by age, sex, and IG products administered. The results suggest importance of underlying health conditions, especially prior hemolysis, and first IG product dose. Differences in HR occurrence may also be explained by product manufacturing processes, indications, routes, and rates of administration, which warrant further investigation.