Complete blood count monitoring for patients on anticoagulation with direct oral anticoagulants for atrial fibrillation and/or venous thromboembolism.

Abstract Introduction: Serial complete blood count (CBC) tests can identify potentially significant lab abnormalities for patients using the direct oral anticoagulants (DOACs). This includes low hemoglobin values, hematocrit abnormalities, and abnormal platelet values. While various CBC monitoring strategies have been proposed, the optimal CBC monitoring strategy has not been defined. It is unclear whether monitoring should differ based on factors like age, comorbidity, and indication for anticoagulation. CBC testing can be important to identify anemia. Patients found to have anemia may need further diagnostic evaluation (e.g., endoscopy to assess for bleeding) and/or treatment (e.g., iron replacement for iron deficiency anemia). We sought to evaluate CBC utilization for patients on DOACs. For patients who had CBCs done, we sought to determine the prevalence and incidence of low hemoglobin values. In addition, we sought to explore the incidence of abnormal platelet and hematocrit values. Methods: We used the Michigan Anticoagulation Quality Improvement Initiative (MAQI2) DOAC registry data, which includes chart abstracted data from six health systems in Michigan. We included patients from 1/2010 to 12/2024 on long term anticoagulation for atrial fibrillation (AF), and/or venous thromboembolism (VTE). Patients with less than 6 months of follow-up, on dialysis, or receiving chemotherapy were excluded. Records were reviewed up to 6 months before anticoagulation initiation to determine the baseline hemoglobin. Anemia was defined as a hemoglobin <12.0 g/dL for females and <13.5 g/dL for males, using the most recent CBC before anticoagulant initiation. Patients were classified as having baseline anemia, a normal baseline hemoglobin, or missing data based on this review. After starting anticoagulation, CBC results were reviewed at 6-month intervals. In addition to hemoglobin abnormalities (defined above), platelets were abnormal if <150,000 or >400,000 platelets per microliter, and hematocrit was abnormal when above or below 40-54% for men and 36-48% for women. Descriptive statistics and prevalences are reported. Results: A total of 3,677 patients met the inclusion criteria. Before the initiation of anticoagulation, 1,247 (33.9%) patients had anemia, while no record of a baseline CBC was available for 380 patients (10.3%). Of the initial study cohort, 2,579/3,677 (70.1%) patients had at least one CBC within the first 6 months of follow-up. Among patients with normal baseline CBC testing, with CBCs tested in the first 6 months, 25.9% had one or more abnormality. Labs showed abnormal platelet values, hematocrit, and low hemoglobin values for 9.2%, 8.2%, and 9.9% of CBCs respectively. A total of 1,775/3,295 (53.9%) patients had at least one CBC between 6-12 months after DOAC initiation. Among patients with normal baseline CBC testing, with CBCs tested between 6-12 months, 29.5% of CBCs had one or more abnormality. Labs showed abnormal platelet values, hematocrit, and low hemoglobin values for 9.3%, 10.2%, and 12.0% of CBCs respectively. A total of 1,706/2,466 (69.2%) patients had at least one CBC between 12-24 months after DOAC initiation. Among patients with normal baseline CBC testing, with CBCs tested between 12-24 months, 31.4% of CBCs had one or more abnormality. Labs showed abnormal platelet values, hematocrit, and low hemoglobin values for 11.4%, 10.7%, and 13.4% of CBCs respectively. The median (interquartile range) of time to initial CBC after enrollment was 4 months (3.9 months). A total of 256/3,295 patients (7.8%) did not have a CBC checked during the initial 12 months after starting anticoagulation. Within the first 12 months of anticoagulation, 5,780 normal CBCs were resulted, 72% for patients with AF. The anticoagulant prescriber for patients with normal CBCs were most commonly cardiologists (37.0%), internal medicine (24.4%), and other providers (15.0%) while hematologists were less often prescribers for patients with normal CBCs (0.6%). Conclusions: Most patients have a baseline CBC before DOAC initiation for AF and/or VTE with nearly one-third being anemic. In follow-up, the majority of patients had CBC monitoring, with this testing often revealing abnormalities. The rate of CBC abnormality at baseline among patients without testing and the response to new CBC abnormalities needs further investigation.