Sensitivity and specificity of an algorithm based on medico-administrative data to identify hospitalized patients with major bleeding presenting to an emergency department

Burden of status epilepticus: prognosis and cost driving factors, insight from a nationwide retrospective cohort study of the French health insurance database

BACKGROUND

Status epilepticus (SE) imposes a significant burden in terms of in-hospital mortality and costs, but the relationship between SE causes, patient comorbidities, mortality, and cost remains insufficiently understood. We determined the in-hospital mortality and cost-driving factors of SE using a large and comprehensive database.

METHODS

We conducted a retrospective cohort study involving patients experiencing their first hospitalization with an ICD-10 code diagnosis of SE, spanning from January 1, 2015, to December 31, 2019, using the French health insurance database which covers 99% of population. Patient characteristics, SE causes, Intensive Care Unit (ICU) admissions, mechanical ventilation, discharge status, and health insurance costs were extracted for each hospitalization.

RESULTS

We identified 52,487 patients hospitalized for a first SE. In-hospital mortality occurred in 11,464 patients (21.8%), with associated factors including age (Odds Ratio [OR], 10.3, 95% Confidence Interval [CI] 7.87-13.8 for ages over 80 compared to 10-19), acute causes (OR, 15.3, 95% CI 13.9-16.8 for hypoxic cause), tumors (OR, 1.75, 95% CI 1.63-1.8), comorbidities (OR, 3.00, 95% CI 2.79-3.24 for 3 or more comorbidities compared to 0), and prolonged mechanical ventilation (OR, 2.61, 95% CI 2.42-2.82). The median reimbursed cost for each SE hospitalization was 6517€ (3364-13,354), with cost factors mirroring those of in-hospital mortality.

CONCLUSION

Causes and co-morbidities are major determinants of mortality and hospital costs in status epilepticus, and factors associated with higher mortality are also often associated with higher costs. Further studies are needed to identify their long-term effects.

Validation of a Major and Clinically Relevant Nonmajor Bleeding Phenotyping Algorithm on Electronic Health Records

PURPOSE

Bleeding is an important health outcome of interest in epidemiological studies. We aimed to develop and validate rule-based algorithms to identify (1) major bleeding and (2) all clinically relevant bleeding (CRB) (composite of major and all clinically relevant nonmajor bleeding) within real-world electronic healthcare data.

METHODS

We took a random sample (n = 1630) of inpatient admissions to Singapore public healthcare institutions in 2019 and 2020, stratifying by hospital and year. We included patients of all age groups, sex, and ethnicities. Presence of major bleeding and CRB were ascertained by two annotators through chart review. A total of 630 and 1000 records were used for algorithm development and validation, respectively. We formulated two algorithms: sensitivity- and positive predictive value (PPV)-optimized algorithms. A combination of hemoglobin test patterns and diagnosis codes were used in the final algorithms.

RESULTS

During validation, diagnosis codes alone yielded low sensitivities for major bleeding (0.16) and CRB (0.24), although specificities and PPV were high (>0.97). For major bleeding, the sensitivity-optimized algorithm had much higher sensitivity and negative predictive values (NPVs) (sensitivity = 0.94, NPV = 1.00), however false positive rates were also relatively high (specificity = 0.90, PPV = 0.34). PPV-optimized algorithm had improved specificity and PPV (specificity = 0.96, PPV = 0.52), with little reduction in sensitivity and NPV (sensitivity = 0.88, NPV = 0.99). For CRB events, our algorithms had lower sensitivities (0.50-0.56).

CONCLUSIONS

The use of diagnosis codes alone misses many genuine major bleeding events. We have developed major bleeding algorithms with high sensitivities, which can ascertain events within populations of interest.

COVID-19 pandemic and trends in clinical outcomes and medication use for patients with established atrial fibrillation: A nationwide analysis of claims data

Study objective

The COVID-19 pandemic disrupted multiple aspects of the health care system, including the diagnosis and control of chronic conditions. This study aimed to quantify pandemic-related changes in the rates of clinical events among patients with atrial fibrillation (AF).

Design/setting/participants

In this retrospective cohort study, we identified individuals with established AF at any time before 2019 using de-identified Optum's Clinformatics® Data Mart, and followed them from 3/18/2019 to death, or disenrollment, or the end of the study (09/30/2021).

Main outcome

Rates of clinical event, including all-cause hospitalization, ischemic stroke, and bleeding. We constructed interrupted time series to test changes in outcomes after the onset of the COVID-19 pandemic (3/11/2020, date of pandemic declaration). We then identified the first month after the start of the pandemic in which outcomes returned to pre-pandemic levels.

Results

A total of 561,758 patients, with a mean age of 77 ± 9.9 years, were included in the study. The monthly incidence rate of all-cause hospitalization decreased from 2.8 % in the period immediately before the pandemic declaration to 1.7 % in the period immediately after, with p-value for level change<0.001. The rate of new ischemic stroke diagnoses decreased from 0.28 % in the period immediately before pandemic declaration to 0.20 % in the period immediately after, and the rate of major bleeding diagnoses from 0.81 % to 0.59 %, both p-values for level change<0.01. The incidence rate of ischemic stroke and bleeding events returned to pre-pandemic levels in October and November 2020, respectively.

Conclusions

The COVID-19 pandemic was associated with a decrease in health care visits for ischemic stroke and bleeding in a nationwide cohort of patients with established AF.

Risk of Major Bleeding Associated with Concomitant Direct-Acting Oral Anticoagulant and Clopidogrel Use: A Retrospective Cohort Study

BACKGROUND AND AIM

Combined anticoagulant-antiplatelet therapy is often indicated in adults with cardiovascular disease and atrial fibrillation or venous thromboembolism. The study aim was to assess the comparative risk of bleeding between rivaroxaban and apixaban when combined with clopidogrel.

METHODS

We conducted a retrospective cohort study of commercially insured US adults newly treated with a combination of rivaroxaban+clopidogrel or apixaban+clopidogrel (2015-2018) using Merative™ Marketscan Research Databases. We used propensity score-based inverse probability of treatment weighting (IPTW) to balance the treatment groups. Weighted Cox proportional hazards regression was used to estimate the risk of major bleeding.

RESULTS

The study cohort included 2895 rivaroxaban+clopidogrel users and 3628 apixaban+clopidogrel users. The median (range) duration of follow up was 61 (73) days. Rivaroxaban+clopidogrel users had a similar risk of major bleeding compared with apixaban+clopidogrel users (IPTW incidence rate per 100 person-years 7.96 vs 7.38; IPTW hazard ratio [HR] 1.13 [95% CI 0.78-1.63]). In the subcohort of adults who were treated with DOAC or clopidogrel monotherapy prior to the combined therapy, the risk of major bleeding did not differ by the drug of monotherapy (IPTW HR for rivaroxaban+clopidogrel group: 0.66 [95% CI 0.33-1.32]; IPTW HR for apixaban+clopidogrel group: 1.10 [95% CI 0.55-2.23]) CONCLUSIONS: In our study of commercially insured US adults, the concomitant use of rivaroxaban+clopidogrel and apixaban+clopidogrel conferred a similar risk of major bleeding. DOAC versus clopidogrel monotherapy prior to the concomitant therapy did not influence the risk of major bleeding.

Atrial fibrillation as a novel risk factor for retinal stroke: A protocol for a population-based retrospective cohort study

Central retinal artery occlusion (CRAO; retinal stroke or eye stroke) is an under-recognized, disabling form of acute ischemic stroke which causes severe visual loss in one eye. The classical risk factor for CRAO is ipsilateral carotid stenosis; however, nearly half of patients with CRAO do not have high-grade carotid stenosis, suggesting that other cardiovascular risk factors may exist for CRAO. Specifically, prior studies have suggested that cardioembolism, driven by underlying atrial fibrillation, may predispose patients to CRAO. We describe the design of an observational, population-based study in this protocol. We evaluate two specific objectives: 1) To determine if atrial fibrillation is an independent risk factor for CRAO after adjusting for medical and cardiovascular risk; 2) To determine if use of oral anticoagulation can modify the risk of CRAO for patients with atrial fibrillation. This protocol lays out our strategy for cohort definition, case and control definition, comorbidity ascertainment, and statistical methods.

Prediction of clinical outcomes after percutaneous coronary intervention: Machine-learning analysis of the National Inpatient Sample

BACKGROUND

This study aimed to develop a multiclass machine-learning (ML) model to predict all-cause mortality, ischemic and hemorrhagic events in unselected hospitalized patients undergoing percutaneous coronary intervention (PCI).

METHODS

This retrospective study included 1,815,595 unselected weighted hospitalizations undergoing PCI from the National Inpatient Sample (2016-2019). Five most common ML algorithms (logistic regression, support vector machine (SVM), naive Bayes, random forest (RF), and extreme gradient boosting (XGBoost)) were trained and tested with 101 input features. The study endpoints were different combinations of all-cause mortality, ischemic cerebrovascular events (CVE) and major bleeding. An area under the curve (AUC) with 95% confidence interval (95% CI) was selected as a performance metric.

RESULTS

The study population was split to a training cohort of 1,186,880 PCI discharges, validation cohort (for calibration) of 296,725 hospitalizations and a test cohort of 331,990 PCI discharges. A total of 98,180 (5.4%) hospital entries included study outcomes. Logistic regression, SVM, naive Bayes, and RF model demonstrated AUCs of 0.83 (95% CI 0.82-0.84), 0.84 (95% CI 0.83-0.86), 0.81 (95% CI 0.80-0.82), and 0.83 (95% CI 0.81-0.84), retrospectively. The XGBoost classifier performed the best with an AUC of 0.86 (95% CI 0.85-0.87) with excellent calibration. We then built a web-based application that provides predictions based on the XGBoost model.

CONCLUSION

We derived the multi-task XGBoost classifier based on 101 features to predict different combinations of all-cause death, ischemic CVE and major bleeding. Such models may be useful in benchmarking and risk prediction using routinely collected administrative data.

In hospitalized patients undergoing therapeutic plasma exchange, major bleeding prevalence depends on the bleeding definition: An analysis of The Recipient Epidemiology and Donor Evaluation Study-III

BACKGROUND

Major bleeding in patients undergoing therapeutic plasma exchange (TPE) has been studied in large databases; but without standardizing bleeding definitions. Therefore, we used standardized definitions to evaluate major bleeding in hospitalized patients undergoing TPE using public use data files from the Recipient Epidemiology and Donor Evaluation Study-III (REDS-III).

STUDY DESIGN AND METHODS

In a retrospective cross-sectional analysis, we identified TPE-treated adults in a first inpatient encounter. We evaluated major bleeding prevalence using (1) International Classification of Diseases (ICD) or Current Procedural Terminology (CPT) codes, (2) packed red blood cell (PRBC) transfusion, or (3) hemoglobin (Hgb) decline. Patients with major bleeding prior to their first TPE were excluded from the analysis.

RESULTS

Among 779 patients undergoing TPE, major bleeding by at least one of the three bleeding definitions occurred in 135 patients (17.3%). For each of the ICD/CPT, PRBC, and Hgb definitions, the prevalence of major bleeding was 2.8% (n = 31), 7.4% (n = 81), and 5.4% (n = 59), respectively. Only 3.7% of bleeds (5/135) were captured by all three definitions and 19.3% (26/135) exclusively by any two pairwise definitions. The addition of PRBC transfusion and Hgb decline to ICD/CPT code definitions increased bleeding prevalence threefold.

CONCLUSION

Among hospitalized adults undergoing TPE in the REDS-III study, the prevalence of major bleeding was 17.3%. The addition of PRBC and Hgb decline to ICD codes increased bleeding prevalence threefold. Future studies are needed to develop validated models that identify patients at risk for major bleeding during TPE.

Clinical use of low-dose parenteral anticoagulation, incidence of major bleeding and mortality: a multi-centre cohort study using the French national health data system

PURPOSE

Low-dose parenteral anticoagulation has demonstrated its efficacy for venous thromboembolism prophylaxis in randomized trials. However, current practice is not widely documented. In ambulatory settings, we aimed to provide an overview of the clinical use of low-dose parenteral anticoagulation in France and to assess the incidence of major bleeding and death rates.

METHODS

A population-based prospective cohort study using the French national health data system (SNIIRAM) identified 142,815 adults living in five well-defined geographical areas who had a course of low-dose parenteral anticoagulants (a total of 150,389 courses) in the period 2013-2015. The main outcome measures were the types of low-dose parenteral anticoagulant, the duration and the clinical context. Adjusted incidence rate ratios (IRR) were derived from Poisson models.

RESULTS

Enoxaparin was the most frequently prescribed anticoagulant (58.9%) followed by tinzaparin (27.3%) and fondaparinux (10.9%). Patients receiving unfractionated heparin (N = 766, 0.53%) were older, more frequently had renal disease (48.75%) and had a higher modified HAS-B(L)ED score (≥ 3 in 61.6%) than patients receiving low-molecular weight heparin (LMWH). Surgical thrombo-prophylaxis was the most frequent indication (47.6%), followed by medical prophylaxis (29.9%). Course durations were in line with regulatory agency specifications. Only 43 (0.028%) major bleeding events and 478 (0.32%) deaths were observed. Adjusted IRRs for major bleeding or death were not significantly different for dalteparin/nadroparin, tinzaparin or fondaparinux compared to enoxaparin.

CONCLUSION

Very low incidence rates of major bleeding and all-cause mortality were observed. Our study confirms the safety of LMWHs and fondaparinux in thrombo-prophylaxis in ambulatory settings.

TRIAL REGISTRATION

ClinicalTrials.gov identifier: NCT02886533.

Long-Term Outcomes for Patients With Acute Coronary Syndrome and Nonvalvular Atrial Fibrillation

Short-term outcomes are worse for patients with acute coronary syndrome (ACS) with a history of nonvalvular atrial fibrillation (NVAF). However, long-term prognosis remains unclear. We linked administrative health databases to identify patients hospitalized with ACS (ST-elevation myocardial infarction [STEMI], non-STEMI [NSTEMI], and unstable angina) between 2008 and 2019 in Alberta, Canada. Patients were stratified according to history of NVAF before hospitalization. The primary outcome was a composite of all-cause mortality, hospitalization for myocardial infarction, or stroke at 3 years. Cox models were constructed to estimate the association between ACS, NVAF, and outcomes. Of 54,309 ACS hospitalizations, 6,351 patients (11.7%) had a history of NVAF. Compared with patients without NVAF, patients with previous NVAF were older (75.6 ± 11.6 vs 64.9 ± 13.4 years), women (35.1% vs 30.0%), had higher comorbid burden (Charlson co-morbidity index 3.0 vs 1.0), and more often presented with NSTEMI (57.5% vs 49.0%). The primary outcome occurred in 37.0% of patients with previous NVAF and 17.4% without (p <0.001). In the multivariable analysis, there was a 1.14-fold (95% confidence interval [CI] 1.09 to 1.20) higher risk of the primary outcome in patients with previous NVAF. There was a significant association with STEMI (adjusted harazard ratio [aHR] 1.24, 95% CI 1.12 to 1.36) and NSTEMI (aHR 1.12, 95% CI 1.06 to 1.19) but not with unstable angina (aHR 1.04, 95% CI 0.90 to 1.22). In conclusion, in this population-based study, we identified that a history of NVAF at ACS presentation is associated with worse long-term prognosis, particularly for STEMI and NSTEMI.

Red Blood Cell Transfusion at a Hemoglobin Threshold of Seven g/dL in Critically Ill Patients: A Regression Discontinuity Study

RATIONALE

In critically ill patients, a hemoglobin transfusion threshold of <7.0 G/dl compared to <10.0 G/dl improves organ dysfunction. However, it is unclear if transfusion at a hemoglobin of <7.0 g/dL is superior to no transfusion.

OBJECTIVES

To compare levels of organ dysfunction between transfusion and no transfusion at a hemoglobin threshold of <7.0 G/dl among critically ill patients using quasi-experimental regression discontinuity methods.

METHODS

We performed regression discontinuity analysis using hemoglobin measurements from patients admitted to ICUs in three cohorts (MIMIC-IV, eICU, and Premier Inc.), estimating the change in organ dysfunction (modified sequential organ failure assessment score) in the 24-72-hour window following each hemoglobin measurement. We compared hemoglobin levels just above and below 7.0 g/dL using a 'fuzzy' discontinuity approach, based on the concept that measurement noise pseudorandomizes similar hemoglobin levels on either side of the transfusion threshold.

RESULTS

A total of 11,181, 13,664, and 167,142 patients were included in the MIMIC-IV, eICU, and Premier cohorts, respectively. Patient characteristics below the threshold did not differ from those above the threshold, except that crossing below the threshold resulted in a >20% absolute increase in transfusion rates in all three cohorts. Transfusion was associated with increases in hemoglobin level in the subsequent 24-72 hours (MIMIC-IV 2.4 [95% CI 1.1, 3.6] g/dL; eICU 0.7 [95% CI 0.3, 1.2] g/dL; Premier 1.9 [95% CI 1.5, 2.2] g/dL), but not with improvement in organ dysfunction (MIMIC-IV 4.6 [95% CI -1.2, 10] points; eICU 4.4 [95% CI 0.9, 7.8] points; Premier 1.1 [95% CI -0.2, 2.3] points), compared to no transfusion.

CONCLUSIONS

Transfusion was not associated with improved organ dysfunction compared to no transfusion at a hemoglobin threshold of 7.0 g/dL, suggesting that evaluation of transfusion targets other than a hemoglobin threshold of 7.0 G/dl may be warranted.

Review Article: Gastrointestinal Bleeding Risk with Direct Oral Anticoagulants

PURPOSE

Although direct oral anticoagulants (DOACs) are associated with an overall favourable safety profile, the risk of gastrointestinal bleeding with DOACs compared with vitamin K antagonists (VKAs) remains controversial. Accordingly, we aimed to provide a focused overview of the risk of gastrointestinal bleeding associated with dabigatran, rivaroxaban, apixaban and edoxaban and its management.

METHODS

We reviewed published studies reporting on DOACs with gastrointestinal bleeding as an outcome, including randomised controlled trials (RCTs), retrospective database studies and large-scale prospective cohort studies.

RESULTS

Cumulative evidence confirms no notable difference in major gastrointestinal bleeding risk between DOACs and VKAs. Moreover, gastrointestinal bleeding in DOAC-treated patients seems less severe and requires less intensive management. The main cause of upper gastrointestinal bleeding in DOAC-treated patients appears to be gastroduodenal ulcers, whereas lower gastrointestinal bleedings are mainly due to diverticula followed by angiodysplasia and haemorrhoids. The lack of head-to-head RCTs with DOACs precludes drawing conclusions on the DOAC with the lowest gastrointestinal bleeding risk. Prescribing physicians should be aware of risk factors for DOAC-related gastrointestinal bleeding (e.g. age > 65, heavy alcohol use, uncontrolled hypertension, hepatic or renal dysfunction, active cancer, anaemia) and adopt preventive measures accordingly. Management of DOAC-associated major gastrointestinal bleeding involves temporary discontinuation of the DOAC, investigation of the bleeding source and treatment of bleeding with fluid resuscitation combined with transfusion and endoscopic haemostasis.

CONCLUSION

DOACs as a class do not increase the risk of major gastrointestinal bleeding compared to VKAs, which supports their continued use for different anticoagulant indications.

Temporal Trends in in-Hospital Bleeding and Transfusion in a Contemporary Canadian ST-Elevation Myocardial Infarction Patient Population

Background

Although ST-elevation myocardial infarction (STEMI) management has evolved substantially over the past decade, its effect on bleeding and transfusion rates are largely unknown in a contemporary population.

Methods

Our study cohort included patients 20 years of age or older who were hospitalized for STEMI between 2007 and 2016 across all Canadian provinces, except Quebec. Unadjusted rates of bleeding and of transfusion during STEMI episodes were calculated overall and for each province according to fiscal year. Patients were stratified into 4 groups according to their bleeding/transfusion. Characteristics, treatment, and outcomes were compared between groups. Multivariate logistic regression modelling was used to assess the association between bleeding and transfusion on in-hospital mortality.

Results

Using 108,832 STEMI episodes, rates of in-hospital bleeding and transfusion declined between 2007 and 2016 from 3.9% to 2.8% (P < 0.0001) and 4.7% to 3.8% (P < 0.0001), respectively. However, variation in bleeding and transfusion rates were observed across Canadian provinces. Patients with bleeding or transfusion, were older, female, and had more comorbidities. Compared with patients who did not bleed or receive a transfusion, individuals who bled, were transfused, or bled and were transfused, had higher in-hospital mortality (18.6%, 30.3%, and 30.4%, respectively [P < 0.0001]). The association remained after adjustment: bleeding (odds ratio [OR], 2.0; 95% confidence interval [CI], 1.8-2.4), transfusion (OR, 4.4; 95% CI, 3.9-4.9), and bleeding and transfusion (OR, 3.8; 95% CI, 3.2-4.6).

Conclusions

The proportion of Canadian STEMI patients who experienced in-hospital bleeding and transfusion has decreased over the past 9 years. However, patients with bleed or transfusion remain at higher risk of adverse outcomes.

Major bleeding risk and mortality associated with antiplatelet drugs in real-world clinical practice. A prospective cohort study

Background

Major bleedings other than gastrointestinal (GI) and intracranial (ICH) and mortality rates associated with antiplatelet drugs in real-world clinical practice are unknown. The objective was to estimate major bleeding risk and mortality among new users of antiplatelet drugs in real-world clinical practice.

Methods and findings

A population-based prospective cohort using the French national health data system (SNIIRAM), identified 69,911 adults living within five well-defined geographical areas, who were new users of antiplatelet drugs in 2013–2015 and who had not received any antithrombotics in 2012. Among them, 63,600 started a monotherapy and 6,311 a dual regimen. Clinical data for all adults referred for bleeding was collected from all emergency departments within these areas, and medically validated. Databases were linked using common key variables. The main outcome measure was time to major bleeding (GI, ICH and other bleedings). Secondary outcomes were death, and event-free survival (EFS). Hazard ratios (HR) were derived from adjusted Cox proportional hazard models. We used Inverse Propensity of Treatment Weighting as a stratified sensitivity analysis according to the antiplatelet monotherapy indication: primary prevention without cardiovascular (CV) risk factors, with CV risk factors, and secondary prevention. We observed 250 (0.36%) major haemorrhages, 81 ICH, 106 GI and 63 other types of bleeding. Incidences were twice as high in dual therapy as in monotherapy. Compared to low-dose aspirin (≤ 100 mg daily), high-dose (> 100 up to 325 mg daily) was associated with an increased risk of ICH (HR = 1.80, 95%CI 1.10 to 2.95). EFS was improved by high-dose compared to low-dose aspirin (1.41, 1.04 to 1.90 and 1.32, 1.03 to 1.68) and clopidogrel (1.30, 0.73 to 2.3 and 1.7, 1.24 to 2.34) respectively in primary prevention with and without CV risk factors.

Conclusion

The incidence of major bleeding and mortality was low. In monotherapy, low-dose aspirin was the safest therapeutic option whatever the indication.

Trial registration

NCT02886533.

Hospital databases for the identification of adverse drug reactions: A 2-year multicentre study in 9 French general hospitals

AIMS

To estimate the actual number of adverse drug reactions (ADRs), we used the French medical administrative database (PMSI) in addition to ADRs spontaneously reported in the French Pharmacovigilance Database (FPVDB).

METHODS

Capture-recapture method was applied to these 2 sources (PMSI and FPVDB), checking their independence via a third data source. The study ran from 1 July 2014 to 30 June 2016 in 9 French general hospitals. From PMSI, all discharge summaries including a selection of 10th International Classification of Diseases codes related to ADRs were analysed. This selection was based on the results of a previous study. All ADRs corresponding to these codes, spontaneously reported in the FPVDB, were included.

RESULTS

In PMSI, 56.9% of hospital stays were related to an ADR (628 out of 1104). In the FPVDB, we retained 115 cases. A total of 43 ADRs were common to the 2 databases. In both sources, the most frequently reported ADRs were cutaneous (33.1 and 19.1%) and renal (25.2% and 11.6%). The most frequently suspected drugs were anti-infectives in PMSI (31.1%) and antineoplastic drugs in the FPVDB (30.4%). Using the capture-recapture method, the estimated number of ADRs was 1657 [95% CI: 1273 to 2040].

CONCLUSION

The use of the PMSI could constitute an additional tool for the estimation of the actual number of ADRs in French hospitals. A model involving a third data source enabled the independence of the 2 sources (PMSI and FPVDB) to be checked before applying the capture-recapture method.

Rivaroxaban was found to be noninferior to warfarin in routine clinical care: A retrospective noninferiority cohort replication study

Purpose

To compare the effectiveness and safety of a drug in daily practice with the outcomes of a target non‐inferiority trial by rigorously mimickingin an observational study the trial's design features.

Methods

This cohort study was conducted using the British Clinical Practice Research Datalink (CPRD) to emulate the ROCKET AF (Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation) trial. Patients with atrial fibrillation who were newly prescribed (>=12 months of no use) either rivaroxaban or warfarinfrom October 2008 to December 2017 were included. Non‐inferiority of rivaroxaban to warfarin in the prevention of stroke or systemic embolism was assessed in different analysis populations (intention‐to‐treat [ITT], per‐protocol [PP], and as‐treated populations) using a hazardratio (HR) of 1.46 as the non‐inferiority margin. Major bleeding (safety outcome) was also assessed and compared to that of the target trial. All outcomes were analyzed using Cox‐proportional hazard analyses.

Results

We included 25,473 incident users of rivaroxaban (n=4,008) or warfarin(n=21,465). Similar to the trial, non‐inferiority in the primary out come was demonstrated in all three analysis populations: HR=1.04 (95%CI 0.84 to 1.30) (ITT), HR=0.98 (95%CI 0.70 to 1.38) (PP), and HR=1.11 (95%CI 0.86 to 1.42) (as‐treated). Risk of major bleeding was also similar to the target trial.

Conclusion

The results of this study provide supportive evidence to the effectiveness of rivaroxaban and adds knowledge on the usefulness of emulating a non‐inferiority trial to assess drug effectiveness.

Major bleeding risk associated with oral anticoagulant in real clinical practice. A multicentre 3-year period population-based prospective cohort study

AIMS

The objective was to compare major bleeding risk of direct oral anticoagulants (DOACs; per type and dose) with vitamin K antagonists (VKAs), irrespective of indication, using real-world data.

METHODS

A population-based prospective cohort study, using the French national health data system (SNIIRAM), identified 47 469 adults living within 5 well-defined geographical areas, who were new users of oral anticoagulants in the period 2013-2015: 20 205 VKA users, 19 579 rivaroxaban users, 4225 dabigatran users and 3460 apixaban users. From all emergency departments within these areas, clinical data for all adults referred for bleeding was collected and medically validated. The databases were linked for common key variables. The main outcome measure was major bleeding: intracranial haemorrhage, major gastrointestinal bleeding and other major bleeding events. Hazard ratios were derived from adjusted Cox proportional hazard models. We used propensity score weighting as a sensitivity analysis, with separate analyses according to indications (atrial fibrillation or venous thromboembolism).

RESULTS

Compared to VKAs, high and low-dose DOACs were associated with a reduced risk of intracranial haemorrhage (adjusted hazard ratio 0.55, 95% confidence interval 0.37-0.82 and 0.54, 0.26-1.12 respectively), and a reduced risk of other major bleeding events (0.41, 0.29-0.58 and 0.41, 0.22-0.79 respectively), irrespective of duration and indication. Neither DOAC dose evidenced any significant difference from VKAs in terms of risk of major gastrointestinal bleeding.

CONCLUSION

There is a clear benefit of using DOACs with regard to intracranial haemorrhage. The study provides new insight into major gastrointestinal and other major bleeding events.