AUTHORS
Elizabeth Herrity, DNP, Yulia Lin, MD2, Jeannie Callum, MD, FRCPC3, Lani Lieberman, MD, FRCPC4
AFFILIATIONS
1 University of Toronto, Toronto, Ontario, Canada
2 Sunnybrook Health Science Centre, Toronto, Ontario, Canada, University of Toronto, Toronto, Ontario, Canada
3 Queen's University, Kingston Health Sciences Centre, Kingston, Ontario, Canada
4 Toronto General Hospital: University Health Network, Toronto, Ontario, Canada, University of Toronto, Toronto, Ontario, Canada
ABSTRACT
Introduction
Fetomaternal hemorrhage (FMH) can be quantified using an acid elution test (Kleihauer Betke test (KB)) or hemoglobin F quantitation by flow cytometry (FC). FC is more precise and reproducible, but it requires dedicated lab instrumentation and trained personnel, limiting its 24/7 availability. KB is widely available and commonly used in routine practice, but is labor-intensive, relies on manual counting, is prone to interobserver variability, and may overestimate FMH in the presence of elevated maternal F cells. The objective of this study was to compare the diagnostic performance of KB with FC and assess the impact of discordant results on Rh immune globulin (RhIG) dosing and clinical management.
Patients and Methods
This retrospective multicenter study included pregnant patients (pts) who underwent paired KB and FC testing within 96 hours between June 2001 and June 2009, conducted at 2 Canadian tertiary-care centers. KB positivity was defined as FMH ≥1.0 mL and FC positivity as ≥0.7 mL. Using FC as the reference standard as KB was followed by FC in all cases, KB sensitivity and specificity were calculated at FC thresholds of ≥0.7, ≥2, ≥5, ≥15 and ≥30 mL and stratified by maternal RhD status. Discordant pairs (e.g., defined as a case in which KB and FC classifications disagreed) were categorized by discrepancy volume. For RhD-negative pts, the RhIG dose administered and the doses that would have been recommended based on the KB and FC estimates were compared. For RhD-positive pts with important discordant results, explanations and outcomes were reviewed.
Results
1,690 unique FMH events from 1,589 pregnant pts were included. Most tests were for RhD-positive pts for diagnostic indications (1,388/1,690; 82%). FC was positive in 414 (24.5%), compared with 109 events (6.4%) positive by KB. Most FMH were <2 mL by KB (1,617/1,690; 96%) and FC (1,637/1,690; 97%). FC-confirmed FMH ≥5 mL was rare (11 events (0.7%)).
KB had low sensitivity (18%), but high specificity (97%) compared with FC. Sensitivity increased with FC-defined FMH volumes to 73% at ≥5 mL (n=3), 80% at ≥15 mL (n=1), and 67% at ≥30 mL (n=1). Specificity remained consistently high across thresholds (94% to 97%).
KB and FC were discordant in 390/1,690 events (23%). False-negative KB results accounted for 341/390 discordant pairs (87%); however, 338/341 involved discrepancies <5 mL and would have been covered by a standard 300ug postpartum RhIG dose. Three KB false-negatives occurred with FC-confirmed FMH ≥5 mL; all involved RhD-positive pts where RhIG is not indicated. In the 3 cases of false-negative KB, two neonates required red blood cell transfusion (FC volume = 248 mL and 7.2 mL, respectively). The first of these was premature and died following complications; no causal relationship between the false-negative result and the death was established.
KB false positives accounted for 49/390 (13%) discordant pairs, 40 had discrepancies <5 mL and 9 had discrepancies ≥ 5 mL. Elevated maternal HbF or hemoglobinopathy were investigated in 7 cases (≥ 10 mL); hemoglobin electrophoresis was negative in 5 and was not performed in 2. Based on the KB estimates, these 7 cases would require a mean of 3.7 additional RhIG vials compared with FC, with a range of 1–14 additional vials. However, only 2 involved RhD-negative pts, and KB would have recommended an additional 300ug in each case.
Quantitative disagreement was identified in 17 concordant-positive pairs, in which both tests detected FMH, but the estimated volumes differed by ≥5 mL. Five of the 17 pairs would have overestimated RhIG vial recommendations based on KB vs FC estimates (2 vs 1, n=2; 3 vs 1, n=1; 9 vs 1, n=1; 15 vs 1, n=1). Of note, only 2 involved Rh-D negative pts, of which only one would have received an additional RhIG vial.
Conclusion
Compared with FC, KB demonstrated high specificity but low overall sensitivity, as it failed to detect small-volume FMH. Nearly all missed bleeds were <5 mL and would have been covered by standard 300ug RhIG prophylaxis. No RhIG underdosing or patient harm attributable to test discordance was identified, supporting the use of KB as a safe and practical initial quantitative test for FMH assessment. KB occasionally overestimated FMH volume, which could lead to unnecessary additional RhIG administration. FC should be used selectively when the KB-estimated volume is unexpectedly large, inconsistent with the clinical presentation, or would substantially alter RhIG dosing.
DISCLOSURES
Elizabeth Herrity, DNP: No relevant conflicts of interest to declare.