To be or Not to be? Exploring the Monitoring of Direct Oral Anticoagulants (DOACs) in Routine Clinical Practice

1. Introduction Monitoring the levels of direct-acting oral anticoagulants (DOACs) can be useful in some circumstances. However, the current literature does not provide clear indications for requesting these levels, nor are there guidelines for subsequent clinical management. Our main objective was to explore the indications for requesting DOAC levels in a tertiary hospital. Secondly, if the result leads to a change in anticoagulation management and if it implies a higher risk of events. 2. Methods The DOAC levels requested for a total of 300 patients from January 2018 to December 2022 at the University Clinical Hospital of Valencia were retrospectively collected. Demographic, analytical, and clinical data of these patients were obtained. The range of levels was established according to the 2021 EHRA guidelines for each DOAC. 3. Results The median age was 76 years, and the median weight was 76.8 kg. 54% were men. The main reason for anticoagulation was atrial fibrillation (87%). The most commonly used drug was apixaban (44%). 81% of the patients had levels within the established normal range. 17% of the patients experienced a clinical event during the anticoagulation period. The main reason for requesting levels was extreme weight (40%). Secondly, drug interactions (20%), and thirdly, advanced chronic kidney disease (CKD) (16%). Other reasons listed by frequency were: urgent surgery or bleeding, thrombosis despite correct dosing, hemorrhagic predisposition, regulated preoperative regimen, high thrombotic risk, CHILD-PUG C cirrhosis, and malabsorption. In 5% of cases, the reason was unknown. Regarding the presence of events, 15% of those with levels within range experienced a clinical event. For patients with levels out of range, 23% experienced a clinical event, with no statistically significant differences. Disaggregating into hemorrhagic and thrombotic events, 25% of patients with high levels experienced a hemorrhagic event, while only 12% of those within or below range did, although this difference was not statistically significant. For thrombotic events, the occurrence between patients with low levels or those within or above range was similar. The studied population was disaggregated according to glomerular filtration rate (GFR) by the CKD-EPI formula: 68% of patients had a GFR >50, 28% had a GFR between 25-50, and 4% had a GFR <25. The distribution of levels and the presence of events were analyzed in these three groups. In the group with a GFR >50, 18% of patients had levels out of range, and 16% had experienced an event. The distribution was similar in the group with a GFR between 25-50. In those with a GFR <25, 28% had levels out of range, and 37% had events. 32% of patients had cancer. Of those with cancer, 23% had levels out of range and 22% had an event, compared to 18% with levels out of range and 14% with events in those without cancer. The presence of a higher percentage of overall events in anticoagulated patients with cancer compared to those without cancer was statistically significant. Therapeutic changes were made in 7% of patients. In 86% of this patients, the reason for the change was levels out of range, resulting in a statistically significant higher percentage of therapeutic changes in patients with levels out of range compared to those with levels within range. The patients who had a therapeutic change for another reason were: due to hemorrhagic/ischemic complications and one by patient choice. 4. Conclusion The main reasons for requesting DOAC levels in our center are extreme weight, drug interactions, and advanced CKD. Regarding clinical impact, we observed a higher percentage of overall and hemorrhagic events in patients out of range and those out of range above respectively, although this was not statistically significant, possibly due to the low event rate relative to the total. Cancer patients had a higher percentage of levels out of range and clinical events compared to anticoagulated patients without cancer. Patients with a GFR <25 had a higher percentage of clinical events and levels out of range, although this was not statistically significant, likely due to the small number of patients in this group (11 patients). Although in our case, the request for levels did result in a therapeutic change for a significant number of patients, prospective studies would likely be necessary to establish the utility and indications for level monitoring.