CHADS₂ score has a better predictive value than CHA₂DS₂-VASc score in elderly patients with atrial fibrillation

Background

Atrial fibrillation (AF) is a common cardiac rhythm disorder, which is responsible for substantial morbidity and mortality. The prevalence of nonvalvular atrial fibrillation (NVAF) increases with advancing age and is considered to be an important risk factor for ischemic stroke (IS) and thromboembolism (TE).¹ Anticoagulation is the cornerstone for AF management. However, various studies have reported the underuse of oral anticoagulation (OAC) among elderly patients with NVAF,^2–4 and the situation is more grim in the People’s Republic of China.⁵

Both CHADS₂ (defined as congestive heart failure, hypertension, age ≥75 years, type 2 diabetes mellitus [DM], previous stroke [doubled]) and CHA₂DS₂-VASc (defined as congestive heart failure, hypertension, age ≥75 years [two scores], type 2 diabetes mellitus, previous stroke, transient ischemic attack [TIA], or TE [doubled], vascular disease, age 65–74 years, and sex category) scores are well-validated tools for the estimation of stroke risk in patients with AF. CHA₂DS₂-VASc improves the precision of identifying “low-risk” patients.⁶ Age is a very important factor of stroke, and it is unclear which score is better suited for use in elderly patients.⁷ The goal of the present study was to compare the utility of CHA₂DS₂-VASc and CHADS₂ scores in predicting IS/TE for the patients with NVAF aged ≥65 years.

Methods

Ethical approval was obtained from the Hospital Ethical Committee of Beijing Friendship Hospital. A procedure-oriented informed consent form was signed by each patient. A retrospective study was conducted by collecting patient data available at Beijing Friendship Hospital for the period between January 1, 2011, and June 30, 2013. It was possible to retrieve the data pertaining to individual patients as all data at our hospital are linked to a unique, permanent, and personal registration number, which is assigned to every patient. Patients with NVAF aged ≥65 years, and not on OAC for the previous 6 months, were enrolled in the study. Diagnosis of AF was based on electrocardiography (12-lead electrocardiography) or 24-hour Holter monitoring. Patients with valvular AF, rheumatic mitral stenosis, mechanical or bioprosthetic heart valve, and mitral valve repair and those receiving hemodialysis or on OAC were excluded from the study.

The study consisted of baseline and follow-up periods. The date of the qualifying AF diagnosis made between January 1, 2011, and June 30, 2013, was designated as the index date. Data from the baseline period, which ended on the index date, were used to obtain information about each patient’s medical history. Follow-up was performed by going through medical records available in the hospital database. Data from the follow-up period, which started from the day after the index date and ended on March 1, 2015, were used to assess the risk of IS/TE. All patients who were lost to follow-up and those who took OAC during the study period were excluded.

The primary end point was the development of IS or TE events (ie, TIA or peripheral embolism). The secondary end point was all-cause death.

IS was defined as a new, sudden focal neurological deficit resulting from a presumed cerebrovascular cause that persisted >24 hours and was not attributable to other identifiable causes, such as tumor and seizure. Events that involved symptoms that lasted <24 hours were considered as TIA. Brain imaging was sought in each case to distinguish hemorrhagic from IS. Peripheral artery embolism was defined as abrupt vascular insufficiency associated with clinical or radiographic evidence of peripheral arterial occlusion in the absence of other likely causes. Presence of vascular disease was identified from previous diagnoses, including myocardial infarction (MI), peripheral artery disease, and complex aortic plaque.

Data were expressed as mean ± SD. The analyses were performed using SPSS 17.0 (SPSS, Inc., Chicago, IL, USA), except net reclassification improvement (NRI), which was analyzed using SAS9.2. Mean values and proportions of variables were compared using unpaired Student’s t-test, analysis of variance, and chi-square test. The IS/TE risk was assessed using Cox regression analysis. The cumulative incidence curve of IS/TE was plotted via the Kaplan–Meier method, with statistical significance examined using the log-rank test. We assessed the predictive accuracies of the CHADS₂ and CHA₂DS₂-VASc scores by calculating c-indices on the basis of receiver operating characteristic (ROC) curves and NRI. Areas under the ROC curves for these two scoring systems were compared using DeLong’s test. Statistical significance was defined as a P-value of <0.05.

Results

Characteristics of patients

Baseline characteristics of the study population are listed in Table 1. The mean age of patients was 80.82±7.34 years, with 70.9% being male. The median score of CHA₂DS₂-VASc and CHADS₂ was 4.77 and 2.95, respectively. Hypertension was the most prevalent comorbidity and was noted in 77.5% of patients. A total of 36.8% had a history of previous stroke or TIA. During the follow-up period of 1.99±1.29 years, 104 (25.2%) patients died and 59 (14.3%) patients had an IS/TE event.

Table 1 Baseline characteristics of patients with AF by CHADS2 scores Notes: CHADS2, congestive heart failure, hypertension, age ≥75 years, type 2 DM, previous stroke (doubled); CHA2DS2-VASc, congestive heart failure, hypertension, age ≥75 years (two scores), type 2 DM, previous stroke, TIA, or TE (doubled), vascular disease, age 65–74 years, and sex category. aCompared to patients with CHADS2 score 0, P<0.05. bCHA2DS2-VASc score ≥2 compared to patients with CHA2DS2-VASc score 1, P<0.05. Abbreviations: DM, diabetes mellitus; TE, thromboembolism; TIA, transient ischemic attack.

On the basis of the CHADS₂ score, 1.7%, 10.2%, and 86.3% of patients were classified as low risk (0 point), intermediate risk (1 point), and high risk (2–6 points), respectively.

Comparison between CHADS₂ and CHA₂DS₂-VASc scores

Both the CHADS₂ and CHA₂DS₂-VASc scores were the significant predictors of IS after adjusting for age and sex. Cox regression model improved from 1.286 (95% CI: 1.086–1.523) to 1.438 (95% CI: 1.187–1.743) when the CHADS₂ score was used for stroke risk categorization instead of the CHA₂DS₂-VASc score.

Of the components of CHADS₂ and CHA₂DS₂-VASc scores, cardiac failure and previous stroke/TIA were strongly associated with the primary end point (hazard ratio [HR] 2.253, 95% CI: 1.240–4.092; HR 2.555, 95% CI: 1.408–4.635, respectively). Age was also found to be associated with IS/TE during follow-up. However, hypertension, DM, vascular disease, and female sex were not found to be predictive of IS/TE (Tables 2 and 3). Among patients with the vascular disease, peripheral arterial disease significantly increased the risk of stroke by 2.71-fold. Previous MI was not a significant predictor of IS/TE.

Table 2 IS/TE risk of CHADS2 components from Cox regression analyses Note: CHADS2, congestive heart failure, hypertension, age ≥75 years, type 2 DM, previous stroke (doubled). Abbreviations: DM, diabetes mellitus; HR, hazard ratio; IS, ischemic stroke; TE, thromboembolism; TIA, transient ischemic attack; CI, confidence interval.

Table 3 IS/TE risk of CHA2DS2-VASc components from Cox regression analyses Note: CHA2DS2-VASc, congestive heart failure, hypertension, age ≥75 years (two scores), type 2 DM, previous stroke, TIA, or TE (doubled), vascular disease, age 65–74 years, and sex category. Abbreviations: DM, diabetes mellitus; HR, hazard ratio; IS, ischemic stroke; MI, myocardial infarction; PAD, peripheral artery disease; TE, thromboembolism; TIA, transient ischemic attack; CI, confidence interval.

Figure 1 shows the ROC curves of CHADS₂ and CHA₂DS₂-VASc scores in predicting IS/TE. The c-indices on the basis of area under the ROC curves for the CHADS₂ and CHA₂DS₂-VASc scores were 0.647 (95% CI: 0.599–0.693) and 0.615 (95% CI: 0.566–0.662), respectively. The difference was statistically significant in favor of the CHADS₂ score (DeLong’s test, P-value =0.0498–0.05, NRI =0.237). The cut-off value of CHADS₂ score was 2.5, with a specificity of 0.537 and a sensitivity of 0.780.

Figure 1 ROC curves of CHADS2 and CHA2DS2-VASc scores in predicting IS/TE. Notes: CHADS2, congestive heart failure, hypertension, age ≥75 years, type 2 DM, previous stroke (doubled); CHA2DS2-VASc, congestive heart failure, hypertension, age ≥75 years (two scores), type 2 DM, previous stroke, TIA, or TE (doubled), vascular disease, age 65–74 years, and sex category. Abbreviations: DM, diabetes mellitus; IS, ischemic stroke; ROC, receiver operating characteristic; TE, thromboembolism; TIA, transient ischemic attack.

CHADS₂ score of ≥3 identified a true high-risk cohort

The Kaplan–Meier curve of freedom from IS is shown in Figure 2. Patients with a CHADS₂ score of 0 had no stroke event. Patients with a CHADS₂ score of 2 had a similar event rate to those with a CHADS₂ score of 1 during the follow-up period (7.1% vs 7.8%; P=0.887). Compared with a CHADS₂ score of 1, patients with a CHADS₂ score of 3, 4, 5, or 6 had a higher event rate (14.4%, 23.1%, 24.4%, and 10%, respectively).

Figure 2 The Kaplan–Meier curve of freedom from ischemic stroke by CHADS2 score. Note: CHADS2, congestive heart failure, hypertension, age ≥75 years, type 2 DM, previous stroke (doubled). Abbreviation: DM, diabetes mellitus.

Using a CHADS₂ score of 1 as the reference in the Cox regression analysis model, the HRs associated with the CHADS₂ scores of 2, 3, 4, 5, and 6 were 1.09, 2.02, 3.32, 3.42, and 1.40, respectively (Figure 3). The event rates with the CHADS₂ scores of 1, 2, 3, 4, 5, and 6 were 7.14%, 7.81%, 14.44%, 23.08%, 20%, and 10%, respectively. These findings indicated that a CHADS₂ score of 2 had a similar event rate to a CHADS₂ score of 1, and CHADS₂ score ≥3 identified a cohort with a true high risk. The HR of the group with a CHADS₂ score of 6 was 1.40, perhaps because of its small size.

Figure 3 Risk of IS/TE based on CHADS2 scores. Note: CHADS2, congestive heart failure, hypertension, age ≥75 years, type 2 DM, previous stroke (doubled). Abbreviations: DM, diabetes mellitus; IS, ischemic stroke; TE, thromboembolism; M–H, Mantel–Haenszel test.

Discussion

CHA₂DS₂-VASc is reported to be better than the CHADS₂ score in identifying the true low-risk patients.^8–12 However, for the regions and population where OAC is frequently underused, it is more important to identify the true high-risk patients. The underuse of OAC among elderly patients with NVAF has been confirmed in different settings.^2–4 One of the most important reasons is that the treating physicians are not sure about which scoring system to follow to determine which patient requires the OAC the most. Therefore, it is necessary to compare the predictive value of two scores, CHA₂DS₂-VASc and CHADS₂, in predicting IS among patients diagnosed with NVAF aged ≥65 years, and to find patients with a true high risk.

In the present study, we only included patients aged 65 years or older. This implies that every patient was added at least 1 point by the CHA₂DS₂-VASc system, and that both CHADS₂ and CHA₂DS₂-VASc scores were useful parameters for predicting adverse events in patients with NVAF aged ≥65 years. However, the CHADS₂ score was found to be more appropriate for patients aged ≥65 years for the categorization of stroke risk when compared with the CHA₂DS₂-VASc score. A CHADS₂ score of ≥3 identified patients with a true high risk. Consistent with the findings of our study, Friberg et al¹³ found that the risk of IS in patients with a CHA₂DS₂-VASc score of 1 seemed to be lower than previously reported (0.1%–0.7%).

In the present study, cardiac failure, age, and history of previous stroke were found to be the independent predictors of IS/TE. Although vascular disease and female sex were not associated with IS/TE risk, both are the additional “non-CHADS₂” risk factors that are incorporated into the CHA₂DS₂-VASc score as per 2012 European Society of Cardiology guidelines.¹⁴

Several studies have been conducted to assess the impact of atherosclerotic vascular disease on stroke in patients with AF. Peripheral arterial disease significantly increased the risk of stroke in all observational studies with the reported risk ranging from 1.3-fold to 2.5-fold.^15,16 Complex aortic plaque in the descending aorta has also been reported as a significant risk factor.^17–19 However, there is no conclusive evidence that previous MI is a predictor of IS.²⁰ In our study, vascular disease included previous MI and peripheral arterial disease. We found peripheral arterial disease to significantly increase the risk of stroke by 2.71-fold. Previous MI was not a significant predictor of IS/TE, which is consistent with the findings of Lin et al.²¹

Though female sex is another “non-CHADS₂” risk factor and has been reported to be associated with IS/TE in patients with AF,^22,23 the said association is considered as controversial. Various studies have reported that female sex is associated with an increased risk of stroke in only those patients with AF aged ≥75 years, whereas female patients aged <65 years without other risk factors do not require anticoagulation therapy.^24,25 Moreover, most of the clinical trials supporting female sex as a risk factor are from the western countries. However, studies conducted in the eastern countries have not reported similar results.²⁶ It has been reported that female sex increases the risk for their comorbidities, such as heart and renal failures.²² In our study, which enrolled patients aged ≥65 years, there was no significant difference in the rate of hypertension, previous stroke/TIA, DM, and CHF in females when compared with males (73.4% vs 75.5%, 39.1% vs 37.7%, 37.5% vs 34.3%, and 23% vs 22.5%, respectively). In addition, the rate of IS/TE in females was not found to be significantly different from males (13.4% vs 14.6%; P>0.05). Therefore, in-line with other studies, the findings of our study indicate that female sex need not be considered when deciding on the antithrombotic therapy.²⁷

In our study, the cut-off value for a very high risk of stroke when using the CHADS₂ score was 3, which was determined by ROC curve analysis. In fact, the event rates during the follow-up period among patients with the CHADS₂ scores of 1 and 2 were almost the same (7%), thus indicating intermediate risk in the CHADS₂ score of 1. Both the CHADS₂ scores of 1 and 2 need OAC.

It is important to note the limitations of our study. Being a retrospective analysis, follow-up was performed by assessing medical records available in the hospital database only, hence some clinically relevant events may have been missed. The study had a limited number of patients, especially in the group of CHADS₂ scores 0 and 9. The HR associated with a CHADS₂ score of 6 (relative risk [RR] =1.4 [95% CI {0.16–12.09}]) is considerably lower than that with a CHADS₂ score of RR =5 (3.42, 95% CI [1.03–12.42]). for the size of sample. We will enlarge the sample in the future.

Conclusion

For patients with NVAF aged 65 years or older, both vascular disease and female sex were not the predictors of IS/TE risk. The use of the CHADS₂ score significantly improves the classification of patients with AF at high risk of stroke compared with the CHA₂DS₂-VASc score. Thus, future large-scale studies involving multiple centers are needed to further corroborate our findings.

Acknowledgments

This study was supported by Beijing Health Care Research Fund (J 09-04), Basic and Clinical Research Cooperation Fund, and Capital Medical University (13JL57).

Disclosure

The authors report no conflicts of interest in this work.

---

References

1.		Lin HJ, Wolf PA, Kelly-Hayes M, et al. Stroke severity in atrial fibrillation. The Framingham Study. Stroke. 1996;27(10):1760–1764.
2.		Marcucci M, Iorio A, Nobili A, et al. Factors affecting adherence to guidelines for antithrombotic therapy in elderly patients with atrial fibrillation admitted to internal medicine wards. Eur J Intern Med. 2010;21(6):516–523.
3.		Tulner LR, Van Campen JP, Kuper IM, et al. Reasons for undertreatment with oral anticoagulants in frail geriatric outpatients with atrial fibrillation: a prospective, descriptive study. Drugs Aging. 2010;27(1):39–50.
4.		Scowcroft AC, Lee S, Mant J. Thromboprophylaxis of elderly patients with AF in the UK: an analysis using the General Practice Research Database (GPRD) 2000–2009. Heart. 2013;99(2):127–132.
5.		Zhou Z, Hu D. An epidemiological study on the prevalence of atrial fibrillation in the Chinese population of mainland China. J Epidemiol. 2008;18(5):209–216.
6.		Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest. 2010;137(2):263–272.
7.		Zhu WG, Xiong QM, Hong K. Meta-analysis of CHADS2 versus CHA2DS2-VASc for predicting stroke and thromboembolism in atrial fibrillation patients independent of anticoagulation. Tex Heart Inst J. 2015;42(1):6–15.
8.		Olesen JB, Lip GY, Hansen ML, et al. Validation of risk stratification schemes for predicting stroke and thromboembolism in patients with atrial fibrillation: nationwide cohort study. BMJ. 2011;342:d124.
9.		Potpara TS, Polovina MM, Licina MM, Marinkovic JM, Prostran MS, Lip GY. Reliable identification of “truly low” thromboembolic risk in patients initially diagnosed with “lone” atrial fibrillation: the Belgrade atrial fibrillation study. Circ Arrhythm Electrophysiol. 2012;5(2):319–326.
10.		Chao TF, Liu CJ, Chen SJ, et al. Atrial fibrillation and the risk of ischemic stroke: does it still matter in patients with a CHA2DS2-VASc score of 0 or 1? Stroke. 2012;43(10):2551–2555.
11.		Lip GY, Frison L, Halperin JL, Lane DA. Identifying patients at high risk for stroke despite anticoagulation: a comparison of contemporary stroke risk stratification schemes in an anticoagulated atrial fibrillation cohort. Stroke. 2010;41(12):2731–2738.
12.		Chao TF, Lin YJ, Tsao HM, et al. CHADS(2) and CHA(2)DS(2)-VASc scores in the prediction of clinical outcomes in patients with atrial fibrillation after catheter ablation. J Am Coll Cardiol. 2011;58(23):2380–2385.
13.		Friberg L, Skeppholm M, Terent A. Benefit of anticoagulation unlikely in patients with atrial fibrillation and a CHA2DS2-VASc score of 1. J Am Coll Cardiol. 2015;65(3):225–232.
14.		Camm AJ, Lip GY, De Caterina R, et al. 2012 focused update of the ESC guidelines for the management of atrial fibrillation: an update of the 2010 ESC guidelines for the management of atrial fibrillation. Developed with the special contribution of the European Heart Rhythm Association. Eur Heart J. 2012;33(21):2719–2747.
15.		Goto S, Bhatt DL, Rother J, et al. Prevalence, clinical profile, and cardiovascular outcomes of atrial fibrillation patients with atherothrombosis. Am Heart J. 2008;156(5):855–863,863.e2.
16.		Conway DS, Lip GY. Comparison of outcomes of patients with symptomatic peripheral artery disease with and without atrial fibrillation (the West Birmingham Atrial Fibrillation Project). Am J Cardiol. 2004;93(11):1422–1425, A10.
17.		Blackshear JL, Pearce LA, Hart RG, et al. Aortic plaque in atrial fibrillation: prevalence, predictors, and thromboembolic implications. Stroke. 1999;30(4):834–840.
18.		Manning WJ, Douglas PS. Transesophageal echocardiography and atrial fibrillation: added value or expensive toy? Ann Intern Med. 1998;128(8):685–687.
19.		Zabalgoitia M, Halperin JL, Pearce LA, Blackshear JL, Asinger RW, Hart RG. Transesophageal echocardiographic correlates of clinical risk of thromboembolism in nonvalvular atrial fibrillation. Stroke Prevention in Atrial Fibrillation III Investigators. J Am Coll Cardiol. 1998;31(7):1622–1626.
20.		Anandasundaram B, Lane DA, Apostolakis S, Lip GY. The impact of atherosclerotic vascular disease in predicting a stroke, thromboembolism and mortality in atrial fibrillation patients: a systematic review. J Thromb Haemost. 2013;11(5):975–987.
21.		Lin LY, Lee CH, Yu CC, et al. Risk factors and incidence of ischemic stroke in Taiwanese with nonvalvular atrial fibrillation – a nation wide database analysis. Atherosclerosis. 2011;217(1):292–295.
22.		Dagres N, Nieuwlaat R, Vardas PE, et al. Gender-related differences in presentation, treatment, and outcome of patients with atrial fibrillation in Europe: a report from the Euro Heart Survey on atrial fibrillation. J Am Coll Cardiol. 2007;49(5):572–577.
23.		Fang MC, Singer DE, Chang Y, et al. Gender differences in the risk of ischemic stroke and peripheral embolism in atrial fibrillation: the AnTicoagulation and risk factors in atrial fibrillation (ATRIA) study. Circulation. 2005;112(12):1687–1691.
24.		Mikkelsen AP, Lindhardsen J, Lip GY, Gislason GH, Torp-Pedersen C, Olesen JB. Female sex as a risk factor for stroke in atrial fibrillation: a nationwide cohort study. J Thromb Haemost. 2012;10(9):1745–1751.
25.		Wagstaff AJ, Overvad TF, Lip GY, Lane DA. Is female sex a risk factor for stroke and thromboembolism in patients with atrial fibrillation? A systematic review and meta-analysis. QJM. 2014;107(12):955–967.
26.		Tomita H, Okumura K, Inoue H, et al. Validation of risk scoring system excluding female sex from CHA2DS2-VASc in Japanese patients with nonvalvular atrial fibrillation-subanalysis of the J-RHYTHM registry. Circ J. 2015;79(8):1719–1726.
27.		Verma A, Cairns JA, Mitchell LB, et al. 2014 focused update of the Canadian Cardiovascular Society Guidelines for the management of atrial fibrillation. Can J Cardiol. 2014;30(10):1114–1130.