Few data have been published that correlate in vitro monocyte monolayer assays (MMA) and red cell (RBC) survival in patients with alloantibodies of unknown significance. Over the past 6 years we gathered clinical correlations in 12 patients with the following antibodies: anti-Lan (three patients), -Ge (three patients), -Yta (five patients), and -Ytb (one patient). RBC survival was estimated using 51Cr studies in seven patients and follow-up of transfusion of incompatible blood in the other five. Six patients with no evidence of RBC destruction had negative MMA findings (anti-Lan [one patient], -Ge [two patients], and -Yta [three patients]). Five patients with evidence of in vivo RBC destruction had significant MMA results. The two clinically significant anti-Lans required fresh serum to give a meaningful MMA result. One patient (anti-Ytb) had an MMA result of borderline significance--normal 51Cr RBC survival at 1 hour--but a reduced T50Cr. The MMA we used appeared to predict the clinical outcome of transfusion in every patient with antibodies to high-frequency antigens whom we tested.

A new technic using polyethylene glycol (PEG) was developed to detect and enhance weak red blood cell/antibody reactions. Twenty percent PEG of 4,000 molecular weight was found to be optimal. Weakly reactive antibodies (n = 25) were tested by PEG, Polybrene, and low ionic strength saline (LISS); 64% were strongest in PEG, 28% reacted equally in PEG as in Polybrene or LISS, 8% reacted weaker in PEG than in Polybrene or LISS. Stronger antibodies (n = 11) were titrated and compared in the three technics; in 10 of 11 titrations, PEG was better or equal to Polybrene and LISS. The false positive rate with the use of PEG and anti-IgG was 1.5% with the use of random sera. Sera from patients (n = 24) with hemolytic transfusion reactions and no detectable antibody by routine technics were tested; two sera had specific antibodies by the PEG technic. This new technic should be a valuable aid in the detection and identification of weak antibodies.

A flow cytometry method was used to compare the amount of red cell (RBC)-bound IgG in 73 patients with and without immune hemolytic anemia (IHA). The positive results in 10 of the direct antiglobulin tests (DATs) were idiopathic, and those in 25 were due to methyldopa therapy; 38 of the 73 DAT-positive patients were babies born to women with IgG alloantibodies of potential clinical significance. Normal blood donors with (n = 30) and without (n = 121) positive DATs were also tested. RBCs that had been strongly sensitized (4+ indirect antiglobulin test) in vitro with different quantities of IgG anti-D, but that had similar antiglobulin test (AGT) titration scores, could easily be differentiated by flow cytometry. The mean percent fluorescence of RBCs, incubated with fluorescein-labeled anti-IgG, from neonatal patients with IHA was higher than that of RBCs from those without IHA, but there was no statistical difference in the other groups. There was considerable overlap in the respective ranges of percent fluorescence of RBCs from patients with and without IHA in all groups. It was not possible to define a clear quantitative threshold differentiating patients with IHA from those without. Although flow cytometry was more precise and reproducible than standard serology (e.g., AGT titration scores), correlations of the amount of RBC-bound IgG and in vivo hemolysis were similar.

A simple and objective assay was developed for the detection and quantitation of fetal-maternal hemorrhage with the use of flow cytometry. In vitro prepared control mixtures of 10%, 2%, 1%, 0.5%, 0.25%, 0.125%, and 0.06% D+ RBCs in D- RBCs were tested (8-11) different times by flow cytometry and gave mean % D+ results of 11.10%, 1.90%, 0.92%, 0.45%, 0.24%, 0.11%, and 0.05%. The coefficient of variation of preparing and testing these mixtures ranged from 11.0 to 15.9% for the 10-0.125% mixtures. Thus, flow cytometry was accurate, reproducible, and sensitive. Flow cytometry was compared with Du tests, rosette tests, and acid elution. The Du test was highly variable because it was not sensitive enough to detect a significant bleed (approximately 0.6%) in some cases and too sensitive (necessitating quantitation of an insignificant bleed) in others. The rosette test was too sensitive. Acid elution and flow cytometry results did not always agree; acid elution results were approximately twice as high as flow cytometry. The authors believe flow cytometric detection of D+ red blood cells to be more accurate than the detection of fetal hemoglobin by acid elution techniques, which is known to have poor reproducibility. Postpartum samples from 56 D- women who delivered D+ babies were tested. Fifty-two had fetal bleeds less than 0.3% by acid elution and flow cytometry; all had negative Du test results, but there were two false positive results with the use of the rosette technique. Four had significant bleeds (greater than or equal to 0.6%); in all four cases the flow cytometry results were lower than the acid elution results. The authors were able to quantitate a bleed of fetal RBCs, which were D+ only by the Du test, in a D- mother with the use of flow cytometry, and D+ RBCs in a mother whose RBCs were of the rare DVI mosaic phenotype. This would not have been possible with the use of the standard Du or rosette techniques.