Burden-of-Illness Associated with Bleeding-Related Hospitalizations in Atrial Fibrillation Patients: Findings from the Nationwide Readmission Database

Abstract

Introduction A paucity of contemporary data examining bleeding-related hospitalization outcomes in atrial fibrillation (AF) patients exists. Methods Adults in the Nationwide Readmissions Database (January 2016–November 2016) with AF and hospitalized for intracranial hemorrhage (ICH), gastrointestinal, genitourinary, or other bleeding were identified. Association between bleed types and outcomes were assessed using multivariable regression (gastrointestinal defined as referent) and reported as crude incidences and adjusted odds ratios (ORs) or mean differences with 95% confidence intervals (CIs). Results In total, 196,878 index bleeding-related hospitalizations were identified in this AF cohort (CHA2DS2VASc score ≥2 in 95.1%), with 70.8% classified as gastrointestinal. The overall incidences of in-hospital mortality, need for post-discharge out-of-home care, and 30-day readmission were 4.9, 50.8, and 18.2%, respectively. Multivariable regression suggested traumatic and nontraumatic ICHs were associated with higher odds of in-hospital mortality (OR = 3.99, 95% CI = 3.79, 4.19; OR = 13.09, 95% CI = 12.24, 13.99) and need for post-discharge out-of-home care (OR = 2.92, 95% CI = 2.83, 3.01; OR = 2.74, 95% CI = 2.59, 2.90), and increases in mean index hospitalization length-of-stay (8.31 days, 95% CI = 8.03, 8.60, 6.27 days, 95% CI = 5.97, 6.57) versus gastrointestinal bleeding. Genitourinary and other bleeds were associated with lower mortality (OR = 0.37, 95% CI = 0.25, 0.55; OR = 0.59, 95% CI = 0.53, 0.64) and reduced length-of-stays (−2.84 days, 95% CI = − 2.91, −2.76; −2.06 days, 95% CI = − 2.11, −2.01) versus gastrointestinal bleeding. Genitourinary bleeds were also associated with a reduced need for post-discharge out-of-home care (OR = 0.86, 95% CI = 0.77, 0.97). Conclusion The burden of bleeding-related hospitalizations was notably driven by relatively rare but severe and life-threatening ICH, and less morbid but more frequent gastrointestinal bleeding. There is need for continued research on bleeding risk factors and mitigation techniques to avoid bleeding-related patient hospitalizations.

Clinical and health care resource use burden of hospitalizations for oral factor Xa inhibitor‐associated major bleeding: A real‐world analysis of Medicare beneficiaries

“…This may imply that these risk factors relate to the need for more institutionalized services after discharge from the index hospitalization and potentially extending the health care resource use burden. The results of this study are consistent with another real‐world study of bleeding‐related hospitalizations in patients with atrial fibrillation, in which ICH was associated with higher odds of in‐hospital mortality, longer hospitalization length of stay, and need for postdischarge out‐of‐home care versus GI bleeding 18 . Previous studies have shown that discharge destination can provide highly predictive values and likelihood ratios for death and disability in patients with ICH 19,20 .…”

Andexanet alfa versus four-factor prothrombin complex concentrate for the reversal of apixaban- or rivaroxaban-associated intracranial hemorrhage: a propensity score-overlap weighted analysis

“…Secondary outcomes included thrombotic event occurrence during the first five days after reversal agent administration. The five-day time frame for thrombotic events was selected as it was a time point specifically reported in ANNEXA-4 and it reduced potential surveillance bias (which can occur when outcomes are sought with differential intensity across populations or over time, or according to care setting and/or patient characteristics) associated with post-discharge thrombotic events [6,11].…”

“…Moreover, investigators in the prospective ANNEXA-4 study actively surveilled for thrombotic events, whereas clinicians in routine practice were less likely to do so, which could have biased the detection of thrombotic events against andexanet alfa. Of note, the five-day time point for thrombotic events utilized in our study was based on the time stratifications reported within the ANNEXA-4 study (< 6, 6-14, and 15-30 days after andexanet alfa administration) [11] and a recent study by Miao and colleagues [6], which demonstrated that five to six days is the median hospital length of stay for atrial fibrillation patients experiencing an intracranial hemorrhage. Third, based upon our study's inclusion criteria, our findings are most applicable to an apixaban-or rivaroxaban-associated intracranial hemorrhage population with baseline GCS scores ≥ 8 and hematoma volumes ≤ 60 mL managed in the USA.…”

“…While DOACs are associated with a decreased risk of intracranial hemorrhage compared to vitamin K antagonists (VKAs), these life-threatening bleeds were observed in DOAC-treated patients at a rate of approximately 0.7%/ year in randomized trials [5]. The previous literature has shown that antithrombotic-related intracranial hemorrhages can result in a high clinical burden, with in-hospital mortality rates of 12.4% for traumatic and 29% for non-traumatic intracranial hemorrhage [6]. Research in the American Heart Association's Get with the Guidelines Stroke registry also found a high burden of inhospital mortality of 27% for patients with spontaneous intracerebral hemorrhage in the presence of FXa inhibitors, with odds of mortality being significantly higher for patients taking FXa inhibitors compared to those who do not take anticoagulants, but significantly lower compared to those taking warfarin [7].…”

Management of risk factors for gastrointestinal bleeding in patients receiving anticoagulant therapy

“…У пациентов, принимающих ПОАК, ниже риск смертельных и жизнеугрожающих, в т.ч. внутричерепных и внутримозговых, кровотечений, оказывающих серьезное влияние на прогноз, и чаще приводящих к инвалидности и летальному исходу по сравнению с варфарином [24].…”