There were no significant differences in mortality or readmission due to VTE recurrence among patients who received an abbreviated versus a non-abbreviated DOAC lead-in dose regimen for the treatment of acute VTE. Furthermore, there were no significant differences in bleeding events between the two cohorts.

Our observed prescribing pattern for abbreviated lead-in therapy (17%) was consistent with the findings of Williams et al. (22%)[14] and Alshaya et al. (20%) [15]. Interestingly, phase III studies of apixaban and rivaroxaban primarily included patients who received less than 48 h of a parenteral agent before transitioning to an oral agent [5, 6, 21]. However, in clinical practice, clinicians may encounter patients who require a longer duration of parenteral therapy beyond 48 h to ensure clinical stability. In such cases, clinicians may perceive the extended parenteral duration as sufficient anticoagulation, thus not necessitating additional intense anticoagulation with an oral DOAC lead-in phase. Our data supported this and showed a longer duration of parenteral agent in the abbreviated group compared to the non-abbreviated group, which was similar to the findings of Ageno et al. [22] Moreover, the greater proportion with a previous VTE history in the abbreviated lead-in group suggested a more complex clinical presentation, which may explain the longer parenteral therapy duration, leading to an abbreviated lead-in phase.

Additionally, recent studies have provided conflicting evidence regarding VTE recurrence and bleeding risks in patients receiving an abbreviated lead-in with apixaban and rivaroxaban. Contrary to our findings, Jackson et al. [17] reported an increased incidence of bleeding with abbreviated apixaban lead-in therapy, which was attributed to more bleeding risk factors, including the use of P2Y12 inhibitors, older age, and poor renal function. However, this study was limited to apixaban and did not include data on rivaroxaban [17]. Additionally, their patient population consisted of predominantly male (96%) veterans, when compared to our study, which included patients receiving care at an academic medical center, and 53% were male. In another study examining clinician adherence to standard lead-in therapy for apixaban and rivaroxaban, Korayem et al. [11] found an elevated risk of bleeding and recurrent VTE in patients transitioned directly from parenteral agent to maintenance dosing of DOAC. Their findings highlighted worse outcomes with no DOAC lead-in therapy. The design differences between our study and the previously mentioned studies may explain the inconsistencies in the findings. These studies had a longer 90-day window for assessing recurrent VTE or bleeding events, whereas our study was limited to a 30-day follow-up window. Generally, patients receiving the shortened lead-in tended to be older, with more cardiac comorbidities and higher use of antiplatelet agents [14]. Therefore, the increased bleeding risk observed in Korayem et al. [11] and Jackson et al. [17] may be confounded by bleeding risk factors, rather than the duration of the DOAC lead-in therapy.

Our study rates of bleeding and recurrent VTE in the abbreviated DOAC lead-in group were similar to those observed in the landmark trials, AMPLIFY and EINSTEIN. These trials included patients undergoing the standard lead-in therapy (7 days for apixaban and 21 days for rivaroxaban) [5, 6]. Additionally, the study by Alshaya et al. [15] evaluated the relative safety and efficacy of a shorter lead-in dose for apixaban and rivaroxaban by also considering the days of parenteral therapy. Notably, our cohort had a higher rate of previous VTE (21% vs. 11%). This may be attributable to an older patient population (69 vs. 53 years of age), and more comorbid conditions, including 30% of patients having active malignancy compared to 3% [15]. They also documented outcomes within a 90-day window, whereas our study investigated outcomes within 30 days. The longer follow-up could introduce additional influencing factors that negate the clinical significance of variations in DOAC lead-in therapy. In addition, patients were excluded if they received greater than 48 h of a parenteral agent before transitioning to a DOAC, whereas these patients were included in the present study. Interestingly, despite the differences, our findings were consistent in showing no significant increase in recurrent VTE or bleeding risk with the abbreviated lead-in compared to the standard regimen. Additionally, a recent retrospective study comparing the abbreviated versus non-abbreviated lead-in dosing of DOAC included patients who received either ≥ 48 h of parenteral or no parenteral agent at all before transition to DOAC [23]. Our study had similar primary outcomes, which included recurrent VTE or bleeding events, per ISTH definitions. However, our study design time frame to capture these events was shorter at 30 days compared to 6 months, and our primary outcomes included all-cause mortality. Despite these differences, this study had similar findings to our study, with similar safety and effectiveness between groups.

Several limitations should be considered when interpreting the findings. Given the retrospective nature of the study, despite stringent manual review and adjustment for covariates, the risk of misclassification and residual confounding remains. Moreover, the single-center design may limit the generalizability of findings, along with the inability to capture readmissions and adverse outcomes that may have occurred outside the health system. Sample size limitations and low event rates, especially across the abbreviated lead-in dose cohort, could have impacted our ability to detect potentially clinically significant associations. Additionally, while the abbreviated versus non-abbreviated lead-in dose categories were determined based on a combination of inpatient and outpatient records, patient adherence to these regimens was not assessed. Such categorization can also be subject to immortal time bias and warrants further prospective study. Finally, the difference in lead-in therapy durations between apixaban and rivaroxaban (i.e. 7 days vs. 21 days, respectively) may have impacted both adherence to therapy, as well as the categorization of patients, in addition to pharmacokinetic differences.