Accuracy of anemia screening by point-of-care hemoglobin testing in patients seeking abortion

Accurate and timely diagnosis of anemia in pregnancy is necessary for safe management of abortion care. Screening for anemia in abortion care is often accomplished using commercially-available point-of-care tests from capillary blood, but the validity of this test has not been investigated in the context of abortion care. We sought to determine the accuracy of a capillary hemoglobin (Hb) among pregnant patients seeking induced abortion. We conducted a retrospective study of patients seeking abortion care at the University of Illinois at Chicago. We identified 108 subjects with paired capillary Hb and venous complete blood count (CBC) hemoglobin measurements within 7 days of each other and within 14 days before abortion. Agreement analysis was performed using Passing-Bablok regression and Bland-Altman plots. More patients were deemed anemic by capillary than by venous Hb measurement (32% vs 19%, p = 0.030). Capillary Hb correlated with venous Hb (r = 0.85, p < 0.001). The average bias for capillary Hb was -1.1 ± 1.0 g/dL. Step-wise, multivariable linear regression identified venous Hb as the only determinant of capillary Hb, and failed to identify any other predictors of bias. The agreement analysis between capillary and venous Hb by Passing-Bablok regression demonstrated systematic and proportional differences. Results from capillary Hb may be biased toward diagnosing anemia and should be interpreted with caution. Misdiagnosing anemia in abortion care can have several consequences and may prevent timely medical abortion or cause delayed procedural abortion. Clinical correlation, and possibly confirmation by venous complete blood count measurement, should be considered before clinical decision-making based solely upon the capillary point-of-care assay.