This study evaluated reports of kidney injury associated with direct oral anticoagulants (DOACs), comparing direct thrombin inhibitors (DTIs) and factor Xa inhibitors (FXaIs). The authors combined an analysis of the FDA Adverse Event Reporting System (FAERS) spanning 2008–2024 with a systematic review of published case reports to identify patterns of adverse renal events linked to DOAC use. The primary goal was to assess class-level and agent-level differences in reporting disproportionality and to summarize clinical features from detailed case descriptions.
The analysis used FAERS data covering 2008 through 2024 to identify suspected DOAC-associated kidney injury reports. The DOACs evaluated included the DTI dabigatran and the FXaIs rivaroxaban, apixaban, and edoxaban. Disproportionality measures reported in the source were the reporting odds ratio (ROR) and the proportional reporting ratio (PRR). The authors also performed multivariable logistic regression to compare disproportionality signals between FXaIs and DTIs after adjustment. In addition, a systematic review of published case reports yielded 32 detailed cases analyzed for clinical patterns and outcomes.
Among 8,116 total reports of suspected DOAC-associated kidney injury identified in FAERS, 1,784 reports were linked to DTIs and 6,332 to FXaIs. Both classes exhibited statistically significant disproportionality signals for kidney injury in the reporting system. For DTIs, the ROR was reported as 2.081 (95% CI 1.979–2.189) and the PRR as 2.044 (95% CI 1.947–2.147). For FXaIs as a group, the ROR was 1.483 (95% CI 1.444–1.523) and the PRR was 1.472 (95% CI 1.434–1.510). These figures indicate that, relative to other drugs and events in FAERS, reports of kidney injury were disproportionately associated with DOACs in the dataset analyzed.
The source emphasized notable intra-class differences among the FXa inhibitors. In particular, edoxaban demonstrated a substantially higher reporting signal compared with other FXaIs: ROR = 4.453 (95% CI 4.021–4.932) and PRR = 4.212 (95% CI 3.831–4.632). The abstract indicates that rivaroxaban and apixaban had lower signals than edoxaban but does not provide their individual ROR/PRR values in the abstract text. These intra-class differences suggest that individual agent characteristics may influence renal adverse event reporting.
After adjustment in multivariable logistic regression, the study found that FXaIs were associated with significantly lower kidney injury reporting disproportionality than DTIs. The adjusted odds ratio reported was 0.687 (95% CI 0.650–0.727) with P < 0.001. This adjusted analysis supports a relative difference in reporting signal between the two drug classes when controlling for covariates included in the model (details of covariates and model specification are not provided in the abstract).
The authors reviewed 32 detailed published cases to compare clinical presentations and outcomes between the classes. Several statistically significant differences emerged in these case-level data: DTI-associated kidney injury more frequently featured pronounced proteinuria (P = 0.029). Cases linked to FXaIs were more often reported in settings of polypharmacy (P = 0.006) and were associated with worse renal recovery outcomes compared with DTI-associated cases (P = 0.023). The abstract does not supply exhaustive case-level details such as time to onset, baseline renal function, or treatment interventions beyond these summarized comparisons.
Based on the FAERS disproportionality signals and case-report comparisons, the authors conclude that kidney injury risk differs between DOAC classes and among individual agents. The findings highlight that DTIs (dabigatran) and the FXaI edoxaban showed particularly strong reporting signals in the dataset analyzed. The study underscores the need for individualized DOAC selection and enhanced renal monitoring in clinical practice, taking agent-specific and patient-specific factors into account when prescribing and following patients on DOAC therapy.
The information provided here is derived from the abstract and plain language summary. The source clarifies that the study combined spontaneous reporting data (FAERS) and published case reports; inherent limitations of FAERS—such as reporting bias, lack of denominator exposure data, and variable report quality—apply. Details that were not reported in the abstract include the full list of covariates used in multivariable models, precise ROR/PRR values for all individual FXaIs beyond edoxaban, and granular case-level data on timing, management, and long-term renal outcomes. These limitations should be considered when interpreting the reported disproportionality signals and case comparisons.