Association of Diagnostic Blood Loss from Phlebotomy and Hospital-Acquired Anemia During Admission with Acute Myocardial Infarction

Abstract

Introduction: Acute, hospital-acquired anemia (HAA) during admission with AMI is associated with higher mortality and worse health status, and often occurs in the absence of recognized bleeding. Diagnostic blood loss from phlebotomy is readily modifiable, but the relationship between diagnostic blood loss and HAA is unclear. Methods: We studied 17,676 AMI patients in the Health Facts database who were not anemic at admission and did not undergo coronary bypass surgery, focusing on the development of moderate-severe HAA (Hgb decline to < 11 g/dl during hospitalization), since this degree of HAA has been shown to be prognostically important. Health Facts included the lab tests, as well as date and time of every blood draw. Patients¡¯ total diagnostic blood loss was calculated by multiplying the number and types of blood tubes drawn by the standard blood volume for each tube type. Hierarchical modified Poisson regression was used to test the association between phlebotomy volume and the development of moderate-severe HAA, accouting for clustering by hospital site and adjusting for demographics, comorbidities, disease severity and treatment variables. Sensitivity analyses were conducted after excluding patients with documented bleeding and after stratifying by length of stay (LOS). Results: Moderate-severe HAA developed in 3,549 patients (20%). Total diagnostic blood loss ranged from 12ml to 1864ml, and mean phlebotomy volume was significantly higher in patients with (182¡À149 ml) vs. without HAA (86.2¡À55.9 ml, p<0.001). The risk of developing HAA was greater with increasing phlebotomy volume (per 50 ml: RR 1.17 (95% CI1.13-1.21)). After multivariable adjustment, each 50 ml of blood drawn for laboratory tests was associated with a 14% increase in risk of HAA (RR 1.14 (95% CI 1.11-1.17)). Results were similar in patients without documented bleeding (RR 1.14 (95% CI1.11-1.16) per 50 ml) and stratified by LOS (LOS ¡Ü 4 days: RR 1.30 (95% CI 1.15-1.46) per 50 ml; LOS ¡Ý 4 days RR 1.11 (95% CI 1.09-1.14) per 50 ml). Conclusion: Blood loss from more frequent phlebotomy is independently associated with the development of moderate-severe HAA. These findings suggest that HAA may be preventable by implementing strategies to limit both the number of blood draws and the volume of blood removed for diagnostic testing.