To our knowledge, this is the longest study that shows a comprehensive analysis of the real-world scenarios when DOAC concentrations are measured 24/7, including changes with time in the number of tests requested over 10 years till 2023 (together with the COVID-19 pandemic), indications in out- and inpatients (including emergency settings), and time trends in TAT for dabigatran, rivaroxaban, and apixaban. Previous studies did not address such a broad spectrum of clinical and laboratory aspects in the years 2013–2023. Regarding the number of tests requested in a post-pandemic time (which to our knowledge, has not been specifically assessed in available papers) our observation on a stable number of DOAC measurements in every day practice is of interest. Despite limited indications for the DOAC measurements, on average every week at least one test was performed in in- or outpatients mostly during the day shift. We observed that the total number of requested tests increased until 2017 and then remained constant until the pandemic, when it dropped by 50% and returned to the previous counts in 2022. The lack of an increase in DOAC testing may be related to increasing physicians’ experience with the DOAC use and recommendations for DOAC testing solely in a few specific clinical situations; most physicians do not need the exact DOAC levels in everyday practice in non-emergency settings. Moreover, DOAC measurements are relatively expensive tests, which might also affect clinical decision making at least in some countries taking into account that no strong clinical guidelines support such testing. However, our study shows also the use of DOAC measurements in patients in whom DOACs are currently not recommended in particular, in patients with ESUS. It was hypothesized that oral anticoagulants may be more effective than antiplatelet therapy in preventing recurrence stroke in patients with ESUS [26, 27]. Observational studies indicated that the use of DOACs in ESUS patients may bring benefits as compared to aspirin [28, 29]. However, in the years 2018–2024, randomized trials with dabigatran, rivaroxaban, and apixaban showed that the DOACs were not superior to aspirin in reducing the risk of recurrent ESUS and were associated with a higher risk of bleeding [30,31,32]. Therefore, in the 2024 ESC guidelines for the management of AF, DOACs are not recommended instead of aspirin for the prevention of secondary stroke in patients with ESUS. Since our analysis covered 10 years of DOAC use, our study showed that physicians prescribed DOACs to a subset of ESUS patients, as reflected by the present results. In recent years the use of DOACs, in this indication has been reduced. To minimize the number of patients with ESUS on DOACs in our center, we conducted additional training for physicians. In our facility, a real-life TAT shorter than 60 min is hard to achieve, which in emergency situations could be suboptimal and thus encourages both the use of reversal agents in the case of bleeding and another treatment in acute ischemic stroke considered for thrombolysis. It should be noted however, that in acute stroke patients with normal renal function and who did not take the DOAC for at least 48 h, the management should not differ from that in non-anticoagulated individuals. In patients who are still on DOACs, mechanical thrombectomy is encouraged preferentially. Experts stated that if DOACs concentrations are < 50 ng/ml thrombolysis is allowed [33].

Our findings indicate that DOAC measurements on a 24/7 basis should be available in tertiary centres to help guide the patient management and more intensive efforts are needed to ensure obtaining the results within 30 min, if necessary, in the case of routine coagulation parameters.

The indications for anticoagulation observed in our patients were mostly VTE (34%) and AF (33%), however, we showed a higher percentage of “other” indications (7%) compared to previous reports, e.g., 1%15 or 4% [20]. The differences may be due at least in part to potential disease coding errors especially in complex cases. On the other hand, in everyday practice DOACs are increasingly used off-label in prevention of thromboembolism, which is also reflected in particular by a relatively large proportion of patients on DOACs for ESUS as it has been discussed above. Most likely, the off-label use of DOACs stimulates attending physicians to determine the levels of DOACs to optimize such a therapy despite no conclusive evidence that it is effective at all. Moreover, our data showed that DOAC measurements were requested in patients with left ventricular thrombus following acute myocardial infarction, which is another common off-label indication for these anticoagulants. However, given compelling data on similar therapeutic efficacy and safety of DOACs and VKA in such patients, now DOACs represent a reasonable alternative to VKA in this indication at least in patients in whom a therapeutic INR range is difficult to achieve consistently or in whom frequent INR checks are impossible [34].

Regarding reasons for requests to measure DOAC levels, we observed that in everyday ambulatory practice, drug-drug interactions, questionable adherence, extreme overweight, and persistent bleeding tendency were the most frequent indications for DOAC measurements while the assessment of residual DOAC concentrations before invasive procedures and in acute stroke were for inpatients. Bavalia et al. [15] presented a quite different pattern of indications for DOAC testing, since among patients with known indications the suspicion of accumulation, followed by bleeding or anaemia and yearly control were reported as most common. In our practice the indication “yearly control” was not used at all, while drug-drug interactions mostly antiepileptic agents prevailed which might only in part reflect the risk of accumulation [4]. Generally routine monitoring was not reported in our center as an indication if there was no increased risk of thromboembolism or bleeding, which is in disagreement with the study by Rottenstreich et al. [14] and Denny et al. [17] in which the majority of DOAC measurements for outpatients were routine follow-up/monitoring efficacy. Regarding obese patients, available data from RCTs and real-life studies on DOACs for the treatment of VTE suggest that in patients with BMI ≥ 40 kg/m2 or weight ≥ 120 kg, apixaban and rivaroxaban appear to be effective and safe compared to VKAs. A switch to VKAs could be done in obese patients with low DOAC plasma level [35]. Of note, there are no recommendations on reducing the DOAC dose in patients with very low body weight.

For inpatients, emergency situations and perioperative evaluation were the most frequent indications in our study like in previous reports [14, 15, 17, 18, 19]. In contrast to the study by Bavalia et al. [15], in which 47% of patients had no specific indication for the requested DOAC measurements, in our study all DOAC measurements were performed with a specific indication which is undeniably an advantage despite the fact that some of them could be controversial. In our opinion, specific indications for DOAC measurements should be used to enhance the clinical impact of their results. Taken together, there is evidence that the patient profile treated in a given institution impacts the proportions of specific indications for DOACs measurements.

Emergency situations are largely considered by experts the indication of key importance, as evidenced for example by Winther-Larsen and Hvas [16], who reported that the majority of DOAC measurements (88%) were performed in an acute clinical situation predominantly acute ischemic stroke (53%). In our study, 72 patients (11.5%) with ESUS were checked for DOAC concentration before thrombolysis was administrated, which most likely resulted from a relatively lower number of acute stroke patients in our hospital in which cardiovascular diseases predominate among in- and outpatients. Nevertheless, DOAC measurements in acute stroke patients especially without access to mechanical thrombectomy is of vital therapeutic importance since prior anticoagulation with DOACs is a relative contraindication for intravenous thrombolysis [36]. It is estimated that among AF patients potentially eligible for thrombolysis, 18% had prior DOACs treatment with rapidly increasing numbers of DOAC pretreatment in recent years [37]. Even if access to thrombectomy is expected to rapidly grow worldwide, this indication for DOAC measurements is likely to remain valid in practice. Another emergency situation to determine DOAC levels is the use of reversal agents commonly associated with a few DOAC measurements [6], like in the case of idarucizumab administered in our hospital (including the first report in a patient with an acute aortic syndrome from 2016 [38] in whom dabigatran levels were determined before, during and after surgery). Such testing was not recommended by experts but our real-world data do demonstrate that in everyday practice DOACs measurements have been used more often in various clinical settings including the first months of the use of DOAC reversal agents, to help guide the therapy of patients at high bleeding risk. As suggested by van der Wall et al. [39], the decision to administer idarucizumab should be made by a multidisciplinary team and results of laboratory test may guide the decision whether or not to administer idarucizumab to avoid health care costs [39, 40]. Andexanet al.fa to reverse anticoagulant effects of FXa inhibitors has not been used in our center, but patients receiving this agent are likely to be tested in the future especially among subjects with intracranial bleeding [4, 41].

From a practical point of view, of key importance is the time from request for DOAC measurements to the final result, and this time was about 90 min in our study without significant differences between the anticoagulants, which was similar to a study by Denny et al. [17] but other real-world studies reported much shorter time [42, 43]. In general, it is difficult to achieve TAT shorter than 30 min mostly due to sample transport which takes an average of ~ 5–10 min even by tube, centrifugation and processing ~ 10 min, and test time 10–20 min. If reagents are refrigerated until use, additional time is required to warm, dissolve and mix them before use. In some situations, retesting of the sample is required, which further extending the TAT.

However, for emergencies (12% of tests) the TAT was shortened to 40 min in our center. The use of a pneumatic tube delivery system for transporting blood samples as well as point-of-care testing could significantly decrease TAT without a reduction in a sample quality [44, 45].

For emergency care and hospitals where specific DOAC determination is not available rapidly, the point of care testing with high negative predictive value may serve for exclusion of clinically relevant DOAC levels. As demonstrated in numerous studies, DOAC Dipstick tests for urine and plasma enable rapid DOAC measurements significantly reducing TAT which may help in immediate clinical decision-making [46,47,48].

Our analysis covered the COVID-19 pandemic, which has not been shown in other available studies on samples collected before 2020 [14,15,16], and we found that the fall by 50% occurred in 2020 compared to 2017–2019. This is mostly due to a dramatic decrease in the number of outpatients especially between March and June 2020, with a total shut down for a few weeks. Several wards at our hospital admitted solely COVID-19 patients including those previously using DOACs in secondary prevention of thromboembolism. At that time experts have recommended the use of low molecular weight heparin (LMWH) or unfractionated heparin in most COVID-19 patients in thrombosis prevention and therapy [49] as well as in patients requiring anticoagulation. Therefore, patients on DOACs were routinely switched to parenteral anticoagulants during hospitalization. Our single‑center experience indicated that in hospitalized COVID‑19 patients LMWH dosage regimens were routinely administered based in internal regulations which contributed to a reduction in DOAC measurements requested especially in 2020 [50].

In the current study we did not evaluate routinely APPT and PT in patients in whom DOAC measurements were requested. At our hospital we have developed an approach in which, since 2013, DOAC measurements may be measured on a 24/7 basis and in practice, in emergency settings this test is largely ordered along with APTT and INR/PT. The prolonged APTT can be an indicator of the presence of dabigatran and prolonged PT can be used in the case of rivaroxaban, however, they have been reported of limited practical value in emergency settings, especially as compared to plasma DOAC measurements [51]. However, in an emergency setting when a clinical decision needs to be made before laboratory results are available, these screening tests can be used as first-line tests, especially when specific DOAC-concentration assays are not available or in situation when the TAT of DOACs measurements is prolonged. The standard procedures to ensure quick decision making in an emergency setting should be implemented in every hospital.

Firstly, this study was retrospective which is associated with inherent drawbacks. However, our tertiary center has the longest experience in the measurement of plasma DOAC concentrations in Poland. Secondly, the issue of whether the DOAC measurements lead to physicians’ clinical decision-making e.g., changes in anticoagulant therapy as well as patient compliance was beyond the scope of the current study. Thirdly, our data cannot be extrapolated for hospitals with departments of general surgery, orthopaedics, or toxicology (e.g., taking DOACs for suicidal purposes) since our center is focused mainly on cardiology and neurology. Finally, information about co-morbidities or other drugs which might interfere with DOACs or data regarding solely liver disorders as well as association of clinical outcomes with specific DOAC levels were not investigated. Moreover, a detailed analysis the causes of a relatively long TAT was beyond the scope of the current study, however, certain pre- and analytical factors have been identified as responsible for prolonged TAT.

Finally, there was no data on edoxaban in our study since this agent is not available in Poland until now.