A Comparison of Techniques Used for Antibody Titer Determination in Screening for Fetal Anemia Secondary to Red Blood...

The aim of the study was to compare the relationship between titer concentrations via gel microcolumn assay (GMA) and conventional tube testing (CTT) methods at a single institution that serves as a referral center for women at risk for hemolytic disease of the fetus and newborn.

Retrospective chart review on all obstetric patients treated was conducted at a single tertiary care center with RBC antibodies from November 2018 to October 2022. Patients were included in the statistical analysis if there were parallel CTT and GMA titration studies available. GMA vs. CTT titers were graphed via bubble plots, and polynomial regression analysis was used to evaluate the correlation between the two methods. Logistical regression was used to evaluate the sensitivity and specificity of GMA titers for prediction of a critical CTT titer of 16.

A total of 166 GMA titers from 87 patients had corresponding CTT titers. The most common antibody was anti-D (76) followed by anti-K (29). Polynomial regression indicated a relationship between GMA titers and CTT titers (R ² = 0.5325, p < 0.001). The relationship changed significantly with removal of anti-K antibodies (R ² = 0.6733, p < 0.001). GMA titers were predictive of a critical CTT (AUC 0.97, p < 0.001) for non-anti-K antibodies but not predictive for anti-K antibodies (p = 0.134). Overall a GMA titer of 64 was predictive of a CTT titer of 16 or higher with 90.7% (95% CI: 82.5-95.9%) sensitivity and 85.0% (95% CI: 75.3-92.0%) specificity. All GMA results of 32 or higher for anti-K antibodies had CTT of at least 16, whereas the same could not be said of the non-anti-K antibodies until GMA titers reached at least 512.

GMA titers are consistently higher than CTT titers in all antibodies tested in our study, and the correlation between the titers is statistically significant for all antibodies except for anti-K. GMA methodology is often preferred as first-line testing by the hospital transfusion service due to cost-effectiveness and ability to be automated. A predictive table was created for provider reference if their hospital's blood bank employs GMA techniques.