Abstract
Aims
Prediction models for outcomes in atrial fibrillation (AF) are used to guide treatment. While regression models have been the analytic standard for prediction modelling, machine ...learning (ML) has been promoted as a potentially superior methodology. We compared the performance of ML and regression models in predicting outcomes in AF patients.
Methods and results
The Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF) and Global Anticoagulant Registry in the FIELD (GARFIELD-AF) are population-based registries that include 74 792 AF patients. Models were generated from potential predictors using stepwise logistic regression (STEP), random forests (RF), gradient boosting (GB), and two neural networks (NNs). Discriminatory power was highest for death STEP area under the curve (AUC) = 0.80 in ORBIT-AF, 0.75 in GARFIELD-AF and lowest for stroke in all models (STEP AUC = 0.67 in ORBIT-AF, 0.66 in GARFIELD-AF). The discriminatory power of the ML models was similar or lower than the STEP models for most outcomes. The GB model had a higher AUC than STEP for death in GARFIELD-AF (0.76 vs. 0.75), but only nominally, and both performed similarly in ORBIT-AF. The multilayer NN had the lowest discriminatory power for all outcomes. The calibration of the STEP modelswere more aligned with the observed events for all outcomes. In the cross-registry models, the discriminatory power of the ML models was similar or lower than the STEP for most cases.
Conclusion
When developed from two large, community-based AF registries, ML techniques did not improve prediction modelling of death, major bleeding, or stroke.
Introduction
The Global Anticoagulant Registry in the FIELD–Atrial Fibrillation (GARFIELD-AF) aims to determine real-life treatment patterns and clinical outcomes of patients with newly diagnosed ...non-valvular atrial fibrillation (AF) and at least one investigator-determined risk factor for stroke. The registry includes a wide array of baseline characteristics and has a particular focus on: (1) bleeding and thromboembolic events; (2) international normalized ratio fluctuations; and (3) therapy compliance and persistence patterns.
Methods
Evolution in baseline treatment for patients enrolled in sequential cohorts showed an increase in prescribing of novel oral anticoagulants over time. Variability in novel oral anticoagulant prescription is primarily due to differences in availability of treatment and prescribing habits between countries and care settings. The GARFIELD-AF registry also provides insights into clinical management and related outcomes of AF in Middle East populations.
Results
A total of 1660 patients with non-valvular AF (median age 64.0 years, interquartile range 56.0–72.0), mostly diagnosed in cardiology settings from Egypt, the United Arab Emirates and Turkey, were recruited in cohorts 3–5. Data from patient populations in the Middle East related to the rates of stroke/systemic embolism, major bleeding and all-cause mortality 1 year after diagnosis of AF and treatment strategies, based on the stroke and bleeding risk, have been analysed and compared with the rest of the world. The use of antithrombotic treatment in the Middle East was generally higher than the non-Middle East, with increased prescription of antiplatelet therapy (AP) therapy. Appropriate use of Factor Xa inhibitors/direct thrombin inhibitors (DTIs) were 74.4% and Factor Xa/DTI + APs were 70.4% in the overall population, whereas they were 57.1% and 63.6%, respectively, in the Middle East.
Conclusion
We have found that rates of stroke and bleeding were lower, although mortality was higher, in the Middle East population. This paper describes the baseline characteristics, patterns of antithrombotic treatment and 1-year outcomes in Middle East AF patients.
Trial Registration
http://www.clinicaltrials.gov
. Identifier, NCT01090362.
Cancer patients undergoing surgery are at a high risk of venous thromboembolism, but few studies have described the rate of autopsy-confirmed fatal pulmonary embolism after heparin ...thromboprophylaxis. In a post hoc analysis of a randomized study (MC-4), which compared the efficacy and safety of certoparin (3000 anti-Xa IU, subcutaneously, once-daily) with unfractionated heparin (5000 IU, subcutaneously, three-times daily) in 23078 patients undergoing surgery lasting more than 30 min, the incidence of autopsy-confirmed fatal pulmonary embolism, death and bleeding in the cancer patients (n=6124) was compared with non-cancer patients (n=16954). Fatal pulmonary embolism was significantly more frequent in cancer patients (0.33% 20/6124) than in non-cancer patients (0.09% 15/16954, relative risk (RR), 3.7 95% confidence intervals (CI), 1.80, 7.77, p=0.0001) at 14 days post-prophylaxis. Perioperative mortality was also significantly higher in cancer patients than in noncancer patients (3.14% 192/6124 vs. 0.71% 120/16954, RR, 4.54 95% CI, 3.59, 5.76, p=0.0001), as were blood loss (p<0.0001), and transfusion requirements (p<0.0001). Prevention of venous thromboembolism in cancer surgical patients remains a clinical challenge.
In 2 double-blind studies, ambulatory patients with objectively proven, disseminated metastatic breast carcinoma (TOPIC-1) or stage III/IV non–small-cell lung carcinoma (TOPIC-2) were randomized to ...certoparin 3000 IU or placebo subcutaneously once daily, for 6 months. Primary efficacy outcome was objectively confirmed symptomatic or asymptomatic venous thromboembolism (VTE). Safety outcomes included bleeding (major and minor), and thrombocytopenia. TOPIC-1 was halted after an interim analysis. Venous thromboembolism occurrence was not different between treatment groups in TOPIC-1 (4% treated with certoparin, 7 of 174 vs 4% receiving placebo, 7 of 177, odds ratio OR 1.02; 95% confidence interval CI 0.30-3.48) and in TOPIC-2 (4.5%, 12 of 268) vs 8.3%, 22 of 264, respectively, OR 0.52; CI 0.23-1.12). Mortality was not different between groups. A post hoc analysis showed certoparin significantly reduced VTE in stage IV lung carcinoma (3.5% vs 10.2%; P = .032) without increased bleeding. In conclusion, thrombosis risk and prophylactic benefit was highest in stage IV lung carcinoma patients.
BackgroundThe extent to which differences in results from Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE) and Rivaroxaban Once Daily Oral Direct ...Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial (ROCKET) atrial fibrillation (AF)—the landmark trials for the approval of apixaban and rivaroxaban, respectively, for non-valvular AF—were influenced by differences in their protocols is debated. The potential influence of selection criteria on trial results was assessed by emulating these trials in data from the Global Anticoagulant Registry in the Field (GARFIELD)-AF registry.MethodsVitamin K antagonist (VKA) and non-vitamin K oral antagonist (NOAC) users from GARFIELD-AF were selected according to eligibility for the original ARISTOTLE or ROCKET AF trials. A propensity score overlap weighted Cox model was used to emulate trial randomisation between treatment groups. Adjusted HRs for stroke or systemic embolism (SE) within 2 years of enrolment were calculated for each NOAC versus VKA.ResultsAmong patients on apixaban, rivaroxaban and VKA, 2570, 3560 and 8005 were eligible for ARISTOTLE, respectively, and 1612, 2005 and 4368, respectively, for ROCKET AF. When selecting for ARISTOTLE criteria, apixaban users had significantly lower stroke/SE risk versus VKA (HR 0.57; 95% CI 0.34 to 0.94) while no reduction was observed with rivaroxaban (HR 0.98; 95% CI 0.68 to 1.40). When selecting for ROCKET AF criteria, safety and efficacy versus VKA were similar across the NOACs.ConclusionApixaban and rivaroxaban showed similar results versus VKA in high-risk patients selected according to ROCKET AF criteria, whereas differences emerged when selecting for the more inclusive ARISTOTLE criteria. Our results highlight the importance of trial selection criteria in interpreting trial results and underline the problems faced in comparing treatments across rather than within clinical trials.
The management of atrial fibrillation (AF), the most common sustained arrhythmia, impacts healthcare resource utilization (HCRU). This study aims to estimate global resource use in AF patients, using ...the GARFIELD-AF registry. A prospective cohort study was conducted to characterize HCRU in AF patients enrolled in sequential cohorts from 2012 to 2016 in 35 countries. Components of HCRU studied were hospital admissions, outpatient care visits, and diagnostic and interventional procedures occurring during follow-up. AF-related HCRU was reported as the percentage of patients demonstrating at least one event and was quantified as rate-per-patient-per-year (PPPY) over time. A total of 49,574 patients was analyzed, having an overall median follow-up of 719 days. Almost all patients (99.5%) had at least one outpatient care visit, while hospital admissions were the second most frequent medical contact, with similar proportions in North America (37.5%) and Europe (37.2%), and slightly higher in the other GARFIELD-AF countries (42.0%; namely Australia, Egypt, and South Africa). Asia and Latin America showed lower percentages of hospitalizations, outpatient care visits, and diagnostic and interventional procedures. Analyses of GARFIELD-AF highlighted the vast AF-related HCRU, underlying significant geographical differences in the type, quantity, and frequency of AF-related HCRU. These differences were likely attributable to health service availability and differing models of care.
Background
It is not always possible to verify whether a patient complaining of symptoms consistent with transient ischemic attack has had an actual cerebrovascular event.
Research question
To ...characterize the risk of cardiovascular events associated with a history of stroke/transient ischemic attack in patients with atrial fibrillation.
Study design and methods
This study investigated the clinical characteristics and outcomes of patients with a history of stroke/transient ischemic attack among 52,014 patients enrolled prospectively in GARFIELD-AF registry. The diagnosis of stroke or transient ischemic attack was not protocol defined but based on physicians’ assessment. Patients’ one-year risk of death, stroke/systemic embolism, and major bleeding was assessed by multivariable Cox regression.
Results
At enrollment, 5617 (10.9%) patients were reported to have a history of stroke or transient ischemic attack. Patients with stroke or transient ischemic attack were older and had a greater burden of diabetes, moderate-to-severe kidney disease, and atherothrombosis and higher median CHA2DS2-VASc and HAS-BLED scores than those without history of stroke or transient ischemic attack. After adjustment, prior stroke/transient ischemic attack was associated with significantly higher risk for all-cause mortality (hazard ratio (HR), 1.26; 95% confidence interval (CI), 1.12–1.42), cardiovascular death (HR, 1.22; 95% CI, 1.01–1.48), non-cardiovascular death (HR, 1.39; 95% CI, 1.15–1.68), and stroke/systemic embolism (HR, 2.17; 95% CI, 1.80–2.63) than patients without history of stroke/transient ischemic attack. In patients with a prior stroke alone higher risk was observed for all-cause mortality (HR, 1.29; 95% CI, 1.11–1.50), non-cardiovascular death (HR, 1.39; 95% CI, 1.10–1.77), and stroke/systemic embolism (HR, 2.29; 95% CI, 1.83–2.86). No significantly elevated risk of adverse events was seen for patients with history of transient ischemic attack alone.
Interpretation
A history of prior stroke or transient ischemic attack is a strong independent risk factor for mortality and stroke/systemic embolism. This excess risk is mainly attributed to a history of stroke (with or without transient ischemic attack), whereas history of transient ischemic attack is a weaker predictor.
Clinical trial registration: NCT01090362.
Venous thromboembolism (VTE) has a long‐term risk of recurrence, dependent on the presence or absence of provoking risk factors at the time of the event.
To compare clinical characteristics, ...anticoagulant patterns, and 12‐month outcomes in patients with transient provoking factors, active cancer, and unprovoked VTE.
The Global Anticoagulant Registry in the FIELD (GARFIELD)‐VTE is a prospective, observational study that enrolled 10 207 patients with objectively diagnosed VTE from 415 sites in 28 countries.
Patients with transient provoking factors were younger (53.0 years) and more frequently women (61.2%) than patients with unprovoked VTE (60.3 years; 43.0% women) or active cancer (63.6 years; 51.7% women). After 6 months, 59.1% of patients with transient provoking factors remained on anticoagulation, compared to 71.3% with unprovoked VTE and 47.3% with active cancer. At 12 months, this decreased to 36.7%, 51.5%, and 25.4%, respectively. The risk of mortality (hazard ratio HR, 1.21; 95% confidence interval CI, 0.90‐1.62), recurrent VTE (HR, 0.84; 95% CI, 0.62‐1.14), and major bleeding (HR, 1.26; 95% CI, 0.86‐1.85) was comparable in patients with transient provoking factors and unprovoked VTE. Patients with minor and major transient provoking factors had a similar risk of recurrent VTE (HR, 0.99; 95% CI, 0.59‐1.66), but those with major transient risk factors had a lower risk of death (HR, 0.61; 95% CI, 0.38‐0.98).
At 1 year, nearly 40% of patients with transient provoking factors and slightly over half of patients with unprovoked VTE were on anticoagulant treatment. Event rates were comparable between the two groups. Risk of death was higher in patients with minor transient factors than in those with major transient factors.
Antithrombotic treatment may improve the disease course in non-critically ill, symptomatic COVID-19 outpatients.
We performed an individual patient-level analysis of the OVID and ETHIC randomized ...controlled trials, which compared enoxaparin thromboprophylaxis for either 14 (OVID) or 21 days (ETHIC) vs. no thromboprophylaxis for outpatients with symptomatic COVID-19 and at least one additional risk factor. The primary efficacy outcome included all-cause hospitalization and all-cause death within 30 days from randomization. Both studies were prematurely stopped for futility. Secondary efficacy outcomes were major symptomatic venous thromboembolic events, arterial cardiovascular events, or their composite occurring within 30 days from randomization. The same outcomes were assessed over a 90-day follow-up. The primary safety outcome was major bleeding (ISTH criteria).
A total of 691 patients were randomized: 339 to receive enoxaparin and 352 to the control group. Over 30-day follow-up, the primary efficacy outcome occurred in 6.0 % of patients in the enoxaparin group vs. 5.8 % of controls for a risk ratio (RR) of 1.05 (95%CI 0.57–1.92). The incidence of major symptomatic venous thromboembolic events and arterial cardiovascular events was 0.9 % vs. 1.8 %, respectively (RR 0.52; 95%CI 0.13–2.06). Most cardiovascular thromboembolic events were represented by symptomatic venous thromboembolic events, occurring in 0.6 % vs. 1.5 % of patients, respectively. A similar distribution of outcomes between the treatment groups was observed over 90 days. No major bleeding occurred in the enoxaparin group vs. one (0.3 %) in the control group.
We found no evidence for the clinical benefit of early administration of enoxaparin thromboprophylaxis in outpatients with symptomatic COVID-19. These results should be interpreted taking into consideration the relatively low occurrence of events.
•Pooled analysis of OVID and ETHIC trials investigating enoxaparin thromboprophylaxis in COVID-19 symptomatic outpatients.•No significant difference in primary outcome (untoward hospitalization/all-cause death) between enoxaparin and standard of care group (RR 1.05, 95% CI 0.57-1.92).•Secondary outcome (arterial and venous thromboembolism) lower in enoxaparin group, not statistically significant (RR 0.52, 95% CI 0.13-2.06).•Prophylactic enoxaparin lacks clinical advantage for symptomatic, but clinically stable COVID-19 outpatients.