Context: The most prevalent side effect associated with warfarin therapy is bleeding. Routine monitoring of laboratory tests, such as complete blood counts (CBC), may provide theoretical benefit in this patient population as a means of detecting occult bleeding through hemoglobin values that decrease compared to baseline.
Objective: To describe the clinical outcomes associated with routine CBC monitoring in a large, diverse sample of anticoagulated patients.
Design: Forty-seven month retrospective, observational study.
Setting: Large non-profit, group-model health maintenance organization with a centralized clinical pharmacy anticoagulation service that routinely orders baseline CBCs on all patients enrolled in the service with follow up CBCs repeated at 3 months and yearly thereafter.
Patients: Patients continuously enrolled in the anticoagulation service for warfarin therapy monitoring for at least one year between January 1, 2000 and December 5, 2003.
Main outcome measures: Medical records were reviewed to determine clinical outcomes associated with hemoglobin decreases of > or =2.0 gm/dL. Clinical outcomes included both overt and occult bleeding events and a determination of major or minor bleeding in these events.
Results: Of the 4033 patients included in the analysis, 578 (14.3%) experienced at least one decrease in hemoglobin > or =2.0 gm/dL. Occult bleeding was confirmed in 121 patients (3.0% of all patients monitored) with a decrease of hemoglobin > or =2.0 gm/dL but only 13 of these patients (0.3% of all patients monitored) experienced major bleeding. The annual detection rate of occult bleeding through routine CBC monitoring was 0.8%.
Conclusions: Although routine CBC monitoring provides some utility in detecting occult bleeding, the yield of clinically important decreases in hemoglobin detected was low. For most anticoagulated patients, routine CBC monitoring appears to be clinically unnecessary.