Abstract 379: Implementation of a Multi-Disciplinary Intervention to Reduce Phlebotomy Volumes in Patients Hospitalized With Acute Myocardial Infarction or Heart Failure

Abstract

Background: Hospital acquired anemia (HAA) is associated with increased mortality and worse health status in AMI patients, and frequently persists after discharge. Studies have identified diagnostic blood loss from phlebotomy as a risk factor for HAA. We studied the feasibility of implementing a phlebotomy reduction initiative, and its impact on standard vs. low volume phlebotomy tube use, phlebotomy volumes and in-hospital hemoglobin (Hgb) values among patients hospitalized with AMI or heart failure. Methods: We conducted a pre-post design, single-center study of a comprehensive diagnostic phlebotomy reduction program. On target hospital units, processes of care were implemented to ensure the use of low volume phlebotomy tubes unless standard blood samples were required for specific tests, use of stored blood for all appropriate lab tests, batching of blood draws for non-stat tests, and use of reinfusion kits to minimize blood loss from central line draws. To ensure comparability of the pre- and post-populations, patients were propensity matched using a multivariable model including demographic, clinical, comorbidity, treatment and procedural variables. After matching, conditional logistic regression was used to compare use of standard vs. low volume, total phlebotomy volume, Hgb at discharge and HAA between the two study phases. Results: There were 442 consecutive patients in the pre-intervention and 465 consecutive patients in the post-intervention cohort. At baseline, use of standard and low volume phlebotomy tubes was similar (median [IQR] 5.0 ml [3.0, 9.0] and 6.0 ml [4.0, 10.0], respectively). Use of larger volume tubes declined and use of low volume tubes increased comparing pre- and post-intervention phases (standard volume: median [IQR] 5.0 ml [3.0,9.0] vs. 3.0 ml [1.0, 6.0], p<0.001; low volume: 6.0 ml [4.0, 10.0] vs. 7.0 ml [5.0, 13.0], respectively, p=0.01). Total phlebotomy volume was numerically lower in the post-intervention phase (mean [SD] 44.8 ml [32.4-74.7] vs. 42.6 ml [27.8-75.2], p=0.09), but there were no differences in discharge Hgb (mean [SD] 12.0±2.0 vs. 12.0±2.1 ml, p=0.66) or HAA rates (42.7% vs. 41.6%, p=0.81). In propensity matched analyses, only a decline in standard volume phlebotomy tube use comparing pre- and post-phases was statistically significant. Conclusions: Implementation of a comprehensive phlebotomy reduction program is feasible in the process of routine AMI and heart failure care. Although we found a reduction in use of larger tubes after the intervention, high baseline rates of low volume blood tube use, reinfusion kits and use of stored serum specimens at our hospital likely limited the impact of the intervention on other outcomes. Further studies at hospitals with low baseline use of phlebotomy reduction strategies are needed to assess the clinical impact of interventions to limit phlebotomy with the goal of preventing HAA.