Management Of Antithrombotic Agents And Clinical Characteristics In Elderly Patients With Atrial Fibrillation

Part I The demographic characteristics, risk factors and antithrombotic agents management in elderly AF patients:A baseline level surveyObjective:To prospective investigate the management of antithrombotic drugs, clinical events and related risk factors in elderly patients with non-valvular atrial fibrillation.methods:The study is a prospective case-control research. We collected 200 elderly (aged 65 years and older)patients with non valvular atrial fibrillation in general hospital of Chinese people’s Liberation Army inpatient or outpatient as a group of atrial fibrillation from 2014 January 1 to 2015 December 31. At the same time, We collected 400 elderly(aged 65 years and older)patients without atrial fibrillation in general hospital of Chinese people’s Liberation Army inpatient or outpatient as a group of non atrial fibrillation. We use the t test or X2 test method to analyze all patients baseline demographic data, drug use of rhythm control and rate control, antithrombotic therapy, comorbidity and cardiovascular risk factors. We use the CHADS2 score and CHA2DS2-VASC score for all patients with thromboembolic risk stratification and the HAS-BLED score for bleeding risk stratification.Results:Of the group of 200 AF patients enrolled in this registry, the mean age was 86.6 ± 8.41,96% were men, BMI were 23.98±5.07kg/m2. Of the group of 400 non-AF patients, the mean age was 85.93 ±11.13,95% were men, BMI were 22.13±6.35kg/m2. Comparison of atrial fibrillation and non atrial fibrillation, comorbidity and cardiovascular risk factors included drinking history(45% VS 28.35%,P<0.01), hypertension(72% VS 63.5%,P<0.05), coronary heart disease (76.5% VS 45%,P<0.01), myocardial infarction (19.5% VS 7.25%,P<0.05), peripheral vascular diseases (51% VS 18.25%,P<0.01), heart failure (22% VS 2.25%,P<0.01), ischemic stroke (44% VS 16.75%,P<0.01), chronic idney disease (24.5% VS 7.5%,P<0.01), massive haemorrhage history (19.5% VS 1.5%,P<0.01), deep venous thrombosis(7.5% VS 1.25%,P<0.01)and so on. Comparison of atrial fibrillation and non atrial fibrillation, there were statistically significant differences in fasting blood glucose, urea, creatinine, eGFR, CRP, albumin, BNP, CK-MB and a number of coagulation indexes between the two groups. Comparison of atrial fibrillation and non atrial fibrillation, AF patients taking beta blockers, amiodarone, propafenone, digoxin, diuretics, nitrates and proton pump inhibitors were significantly increased (P<0.01). We analyzed the current situation of antithrombotic therapy in patients with atrial fibrillation, we found 25 (12.5%) patients taking rivaroxaban, 83(41.5%) patients taking clopidogrel or aspirin respectively,31 (15.5%) patients taking clopidogrel and aspirin, only 12 (6%) patients taking warfarin. Of the group of 200 elderly AF patients enrolled in this registry, there were 139 paroxysmal atrial fibrillation patients,14 persistent atrial fibrillation patients,47 permanent atrial fibrillation patients. Persistent and permanent atrial fibrillation patients were higher proportion than paroxysmal atrial fibrillation in the aspect of three comorbidities, heart failure (34.4% VS 16.55%, P<0.01)、ischemic stroke (55.7% VS 38.85%, P<0.05) and massive haemorrhage history (36.1% VS 12.23%, P<0.01). Persistent and permanent atrial fibrillation patients were a lower proportion than paroxysmal atrial fibrillation in the aspect of taking amiodarone (3.3% VS 29.5%, P< 0.01) and propafenone (1.6% VS 12.23%, P< 0.01). Persistent and permanent atrial fibrillation patients were a higher proportion than paroxysmal atrial fibrillation in the aspect of taking digoxin (32.8% VS 7.91%, P< 0.01) and beta blockers (45.9% VS 63.31%, P < 0.05). Persistent and permanent atrial fibrillation patients were a higher proportion than paroxysmal atrial fibrillation in the aspect of oral warfarin(11.5% VS 3.6%, P< 0.05). We use the CHADS2 score and CHA2DS2-VASC score for all patients with thromboembolic risk stratification and the HAS-BLED score for bleeding risk stratification, AF group and non AF group difference of three scores were statistically significant (P< 0.01), of which the CHADS2 score was 3.04±1.55 VS 1.88±1.34 and CHA2DS2-VASC score was 4.60±1.75 VS 3.12±1.47, the HAS-BLED score were 3.30±1.29 VS 2.23±0.88. At the same time, we used three score methods for risk stratification of all patients, just as low risk, medium risk and high risk, differences between groups were statistically significant (P< 0.01). Of the group of 200 AF patients enrolled in this registry,8 AF patients were CHADS2=0 score,29 AF patients were CHADS2=1 score,163 AF patients were CHADS2>2 score,6 AF patients were CHA2DS2-VASC=1 score,194 AF patients were CHA2DS2-VASC≥2 score,64 AF patients were HAS-BLED< 3 score,136 AF patients were HAS-BLED≥3 score. The antithrombotic agents of warfarin, rivaroxaban, aspirin and clopidogrel were not statistically significant difference in three groups of CHADS2=0,CHADS2=1 score, CHADS2>2 score and in two groups of HAS-BLED <3 score, HAS-BLED>3 score (P>0.05)Conclussion:1. Elderly AF patients are often associated with cardiovascular diseases or risk factors, including drinking history, hypertension, coronary heart disease, myocardial infarction, peripheral vascular disease, heart failure, ischemic stroke, chronic idney disease, massive haemorrhage history, deep venous thrombosis and so on; 2. Elderly AF patients taking ventricular rate control drugs were more than taking Antiarrhythmic drugs.3. Persistent and permanent atrial fibrillation patients were higher proportion than paroxysmal atrial fibrillation in the aspect of three comorbidities, heart failure、ischemic stroke and massive haemorrhage history. Persistent and permanent atrial fibrillation patients were a lower proportion than paroxysmal atrial fibrillation in the aspect of taking amiodarone and propafenone, were a higher proportion in the aspect of taking digoxin and beta blockers, were a higher proportion than paroxysmal atrial fibrillation in the aspect of oral warfarin.4. At present, the level of anticoagulation in elderly AF patients is obviously insufficient, does not comply with the guidelines.Part Ⅱ A study on the risk management of endpoint events and prognosis during the two years follow-up in elderly AF patientsObjective:We prospective studied the antithrombotic therapy schemes and clinical endpoint events, and analyzed the independent risk factors related to prognosis in elderly AF patients during the two years follow-up.methods:We collected conditions of these patients included rhythm control, rate control and antithrombotic therapy, analysis incidence of clinical endpoint events included all-cause death, thromboembolism and bleeding during the two years follow-up. We analyzed clinical endpoint events and independent risk factors of prognosis in elderly AF patients by Cox hazard proportional model. We evaluated the forecasting performance of thromboembolism risk score and bleeding risk score to clinical endpoint events in elderly AF patients by receiver operating curve. We studied the changes of CRP and PAF level and its clinical significance in elderly AF patients.Results:During the two years follow-up, these patients took the medicine and laboratory examination in stable condition. INR maintained at between 1.5-2 in elderly AF patients who took warfarin antithrombotic therapy. Clinical endpoint events in patients with AF were significantly higher than those of non AF patients, including clinical thromboembolism (27.6% vs.9.8%, P< 0.01), bleeding (12.7% vs 29.4%, P< 0.01), all-cause death (28.7% vs 7.5%, P< 0.01), ischemic stroke (5.7% vs 2.3%, P< 0.05), ACS (16.0% vs 6.7%, P< 0.01), other thrombosis (12.4% vs 1.8%, P< 0.01), major bleeding (14.4% vs 2.3%, P< 0.01), minor bleeding (20.6% vs 11.1%, P< 0.01). The median time of clinical endpoint events was as shown below, the ischemic stroke was 9 (6.5-18) months, ACS was 7.2 (2.2-11.8) months, other system thrombosis was 4.95 (2.775-9.55)months, major bleeding was 5.35 (2.175-9.45) months, minor bleeding was 4 (2-11.925) months, all-cause death was 6.85 (3.75-11) months. Different types of AF patients during follow-up of clinical endpoint events compared, paroxysmal AF patients with thromboembolism (20.6% vs 44.8%, P< 0.01), bleeding (43.1% vs 23.5%, P< 0.05) and all-cause death (39.7% vs 24.3%, P< 0.05) compared with persistent and permanent AF patients decreased significantly, the ACS (11.8% vs 25.7%, P< 0.01), other thrombosis (8.8% vs.20.7%, P< 0.01) and major bleeding (27.6% vs 8.8%, P< 0.01) clinical endpoint events in patients with paroxysmal AF was significantly less than that of the persistent and permanent AF patients. Patients with anticoagulation, antiplatelet and non antithrombotic therapy compared, the incidence of all-cause death clinical endpoint events in patients with non antithrombotic therapy increased significantly(16.2% VS 28% VS 46.9%, P< 0.01). The predictive ability of CHADS2 score and CHA2DS2-VASC score of clinical thromboembolic events:area under the ROC curve of CHADS2 score was 0.596 (P=0.038,95%CI:0.506-0.686), area under the ROC curve of CHA2DS2-VASC score was 0.620 (P=0.009,95%CI: 0.530-0.710).The thromboembolism was divided into ischemic stroke, ACS and other system thrombosis. The predictive ability of CHADS2 score of ischemic stroke was good, area under the ROC curve was 0.687 (P=0.037,95%CI:0.526-0.848). The predictive ability of CHA2DS2-VASC score of ischemic stroke was good, area under the ROC curve was 0.713 (P=0.018,95%CI:0.540-0.885). The predictive ability of clinical bleeding events:area under the ROC curve of HAS-BLED score was 0.621 (P=0.008,95%CI:0.532-0.709). The bleeding events was divided into major bleeding and minor bleeding. The predictive ability of HAS-BLED score of minor bleeding was good, area under the ROC curve was 0.664 (P=0.001,95%CI: 0.564-0.764). The predictive ability of three scores of all-cause death events:area under the ROC curve of CHADS2 score was 0.618 (P=0.010,95%CI:0.553-0.742), area under the ROC curve of CHA2DS2-VASC score was 0.600 (P=0.028,95%CI: 0.516-0.685), area under the ROC curve of HAS-BLED score was 0.531 (P=0.495, 95%CI:0.445-0.617). Multi-variate Cox regression analysis showed that age(HR for increased 1 year:1.061,95%CI:1.014-1.110, P=0.01), renal insufficiency history(HR:2.44,95%CI:1.41-4.221, P=0.003), major bleeding history(HR:2.546, 95% CI:1.403-4.620, P=0.002), whether oral digoxin (HR:2.155,95%CI: 1.188-3.906, P=0.011) were the independent risk factors of all-cause death two years in elderly AF patients. Whether oral ACEI/ARB(HR:0.503,95%CI:0.267-0.945, P=0.033), whether oral calcium channel blocker (HR:0.507,95%CI:0.281-0.915, P=0.024) and whether oral stains (HR:0.411,95% CI:0.237-0.714, P=0.002) were protective factors of all-cause death. BMI(HR for increased 1kg/m2:1.086, 95%CI:1.034-1.141, P=0.001)and whether oral digoxin (HR:1.952, 95%CI:1.056-3.607, P=0.033) were the independent risk factors of thromboembolism endpoint events in elderly AF patients, whether oral stains (HR:0.580,95%CI: 0.471-1.302, P=0.028) were protective factors of thromboembolism endpoint events in elderly AF patients. The thromboembolism was divided into ischemic stroke, ACS and other system thrombosis. The independent risk factors of ischemic stroke is whether there is a history of ischemic stroke (HR:1.544,95%CI:1.010-2.184, P=0.056) and whether there is a history of peripheral vascular disease (HR:4.236, 95%CI:0.915-5.620, P=0.045). The independent risk factors of ACS is whether oral digoxin (HR:2.268,95%CI:1.038-4.955, P=0.040). Age(HR for increased 1 year: 1.068,95%CI:1.017-1.122, P=0.09), major bleeding history(HR:2.573 95%CI:1.475-4.235, P=0.025) and whether oral digoxin (HR:1.925 95%CI:1.058-3.503, P=0.032) were the independent risk factors of bleeding endpoint events in elderly AF patients. Whether oral Beta blocker (HR:0.402, 95%CI:0.219-0.737, P=0.001) and whether oral calcium channel blocker(HR:0.413,95% CI:0.233-0.693,P=0.001) were protective factors of hemorrhage endpoint events in elderly AF patients. The bleeding events was divided into major bleeding and minor bleeding, heart failure history (HR:4.452,95%CI: 2.045-9.650, P=0.001) was the independent risk factors of major bleeding endpoint events, major bleeding history (HR:2.893,95%CI:1.544-5.419, P=0.01) was the independent risk factors of minor bleeding endpoint events, Whether oral Beta blocker(HR:0.645,95% CI:0.254-0.822, P=0.018) and whether oral calcium channel blocker(HR:0.228,95% CI:0.081-0.641, P=0.05)were protective factors of major bleeding endpoint events in elderly AF patients. During the two years follow-up, inflammation including CRP and PAF in elderly AF patients was significantly higher than those of non AF patients (P<0.05).CRP was the independent risk factors of clinical thromboembolic events (HR for increased 1mg/dl:1.254 95%CI:1.124-1.356, P=0.041), clinical bleeding events(HR for increased 1mg/dl: 1.148,95%CI:1.019-1.293, P=0.023)and clinical major bleeding events (HR for increased 1mg/dl:1.238,95%CI:1.082-1.416, P=0.023)in elderly AF patients. PAF was the independent risk factors of clinical thromboembolic events (HR for increased 1mg/dl:1.025,95%CI:1.012-1.204, P=0.026) and clinical bleeding events (HR for increased 1mg/dl:1.194,95%CI:1.025-1.345, P=0.014)in elderly AF patients. Conclussion:1.During the two years follow-up, these patients took the medicine and laboratory examination in stable condition. Compared with the elderly non AF patients, elderly AF patients had higher incidence of clinical endpoint events, including all-cause death events were 28.7%, thromboembolic events were 27.8%, bleeding events were 29.4%, ischemic stroke events were 5.7%, ACS events were 16%, the other thrombosis events were 12.4%, major bleeding events were 14.4% and 20.6% for minor bleeding events; 2.During the two years follow-up, oral anticoagulant or antiplatelet drugs in the treatment of elderly AF patients all-cause death events were less than patients without any antithrombotic therapy.INR maintained at between 1.5-2 in elderly AF patients who took warfarin antithrombotic therapy. Paroxysmal AF patients with thromboembolism events(20.6% VS 44.8%), bleeding events(23.5% VS 43.1%) and all-cause death events (24.3% VS 39.7%)outcomes than persistent and permanent AF patients significantly reduced. 3.Risk prediction ability of CHADS2 score and CHA2DS2-VASC score in elderly AF patients is good to clinical thromboembolism events and all-cause death events, risk prediction ability of HAS-BLED score is good to clinical bleeding events, CHA2DS2-VASC score of thromboembolism risk prediction ability is better than CHADS2 score.4.Including the traditional risk factors of age, female gender, hypertension, diabetes, heart failure and prior stroke/TIA history, the prognosis of the elderly AF patients related factors of BMI level, previous history of massive haemorrhage, chronic kidney disease, history of peripheral vascular disease history, baseline using ACEI/ARB, calcium channel blockers, statins, beta blockers and digoxin. We must attach much more importance to those phenomenas. 5.Inflammation including CRP and PAF is an important independent risk factors in elderly AF patients, and have a role in development and prognosis prediction.