- Email: [email protected]
Performance and validation of R-CHA2DS2VASc score for thromboembolism in patients with hypertrophic cardiomyopathy
Journal Pre-proof Performance and Validation of R-CHA2DS2VASc Score for Thromboembolism in Patients with Hypertrophic Cardiomyopathy Ziqiong Wang, Hang Liao, Sen He, Xiaoping Chen PII:
S1109-9666(19)30200-3
DOI:
https://doi.org/10.1016/j.hjc.2019.08.001
Reference:
HJC 426
To appear in:
Hellenic Journal of Cardiology
Received Date: 4 May 2019 Revised Date:
1 July 2019
Accepted Date: 21 August 2019
Please cite this article as: Wang Z, Liao H, He S, Chen X, Performance and Validation of RCHA2DS2VASc Score for Thromboembolism in Patients with Hypertrophic Cardiomyopathy, Hellenic Journal of Cardiology, https://doi.org/10.1016/j.hjc.2019.08.001. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Hellenic Society of Cardiology. Publishing services by Elsevier B.V. All rights reserved.
Performance and Validation of R-CHA2DS2VASc Score for Thromboembolism in Patients with Hypertrophic Cardiomyopathy Ziqiong Wang, Hang Liao, Sen He*, Xiaoping Chen* Department of Cardiology, West China Hospital, Sichuan University, Chengdu 610041, China Ziqiong Wang and Hang Liao contributed equally to this work. *Correspondence to: Xiaoping Chen and Sen He, Department of Cardiology, West China Hospital, Sichuan University, 37 Guo Xue Xiang, Chengdu 610041, Sichuan Province, People’s Republic of China. Tel: (+86) 028 85422343, Fax: +86 02885422175,
E-mail:
[email protected]
[email protected] (for Sen He).
(for
Xiaoping
Chen),
Abstract: Aims: To validate the modified R-CHA2DS2VASc score as a predictor of thromboembolism in HCM patients. Methods: A total of 446 HCM patients were enrolled in our study, thirty-one (6.95%) patients experienced thromboembolic events during the follow-up time of 1786.7 person-years. The association between R-CHA2DS2VASc score and risk of thromboembolism was assessed by Cox’s proportional hazard analysis. The discriminatory power of R-CHA2DS2VASc score for thromboembolism prediction was assessed by Harrell’s C-statistic and validated internally by bootstrapping methods. Calibration curve was plotted by observed outcomes versus expected probabilities of thromboembolism. Results: The R-CHA2DS2VASc score was well calibrated with 0.84 thromboembolic events per 100 person-years in the predefined low risk (R-CHA2DS2VASc score ≤2) group, 1.84 in the low to moderate risk (R-CHA2DS2VASc score 3-4) group, 4.67 in the moderate to high risk (R-CHA2DS2VASc score 5-7) group, and 17.54 in the high risk (R-CHA2DS2VASc score
8) group. Hazard ratios for thromboembolism were
2.88 (95%CI: 1.06-7.82, P=0.038) for low to moderate versus low risk group, 5.30 (95%CI: 2.14-13.12, P=0.0003) for moderate to high versus low risk group, and 16.57 5.30 (95%CI: 4.96-55.33, P<0.0001) for high versus low risk group after adjusting left atria size. The Harrell’s C statistic was 0.7737 (95% CI: 0.65-0.89) for R-CHA2DS2VASc score. Conclusion: The R-CHA2DS2VASc score has shown good calibration and
discriminative power in the prediction of thromboembolism for HCM patients. It should be considered as a potential decision support tool for HCM patients during clinical practice. Key words: hypertrophic cardiomyopathy, thromboembolism, risk stratification, R-CHA2DS2VASc score 1. Introduction Hypertrophic cardiomyopathy is a common genetically transmitted disease, defined clinically by the presence of unexplained left ventricular hypertrophy [1]. Thromboembolism is known to occur as complication of HCM [2]. Given the fact that thromboprophylaxis, including anti-platelet and oral anticoagulation therapy could potentially prevent the occurrence of thromboembolism. The identification of subjects who are at high risk is of great clinical significance. Though previous studies provided important insights into risk factors contributing to thromboembolism, such as older age, atrial fibrillation (AF), left atrial diameter [3-4]. Comprehensive risk stratification of thromboembolism in HCM patients is still a tough issue and remains to be resolved. CHA2DS2-VASc score is a clinical tool intended for estimating overall stroke risk in patients with atrial fibrillation (AF) [5]. Previous studies have shown that it was not closely predictive of the clinical outcome in patients with HCM [6]. Barra’s team proposed a new model
R-CHA2DS2VASc score, which was derived
from the original CHA2DS2-VASc score, comprising its primordial variables plus two renal function parameters [glomerular filtration rate (GFR) and blood urea nitrogen (BUN)], AF status and performance of a revascularization procedure
to predict
stroke for patients with myocardial infarction and it showed good calibration and discriminative performance in the derived cohort[7]. The aim of this study is to validate the new R-CHA2DS2VASc score in the prediction of thromboembolism in our HCM population. 2. Methods 2.1 Study population From 12-05-2008 to 10-31-2018, 458 HCM patients were identified at inpatient department of West China Hospital, Sichuan University, a tertiary referral center. Diagnosis was based on the echocardiographic demonstration of an unexplained increase in wall thickness ≥15mm, in the absence of abnormal load conditions [8]. Eight patients who were under 18 years old and 4 patients with incomplete biochemistry data were excluded in the study. The primary endpoint was defined as the occurrence of thromboembolism including ischemic stroke and systemic embolism, which was confirmed by imaging studies (computed tomography scan or magnetic resonance imaging and vascular ultrasound) during the follow-up. Follow-up was conducted by clinical consultations, medical records review and telephone interviews. Informed consents were obtained from all individual participants included in the study. 2.2 R-CHA2DS2VASc score definition The factors comprising the CHA2DS2-VASc score are heart failure, hypertension, diabetes, prior thromboembolic event, vascular disease, female sex, and age ≥ 65 years; 2 points are accorded for age ≥ 75 and prior stroke event, and one point for
each of the other factors [5]. R-CHA2DS2VASc score was derived from the original CHA2DS2-VASc score, comprising all its primordial variables plus two renal function parameters (GFR by MDRD formula and BUN), atrial fibrillation and performance of a revascularization procedure. 0 point is assigned for GFR≥60 mL/min, 1 point for 30-59.9 mL/min, 2 points for GFR<30 mL/min. Urea
25mg/dl would add 1 point to
the score. Presence of AF adds another 2 points. Revascularization procedure would minus 1 point. Patients were divided into the following four risk categories: low risk (R-CHA2DS2VASc score
2), intermediate-low risk (R-CHA2DS2VASc score 3-4),
intermediate-high risk (R-CHA2DS2VASc score 5-7), high risk (R-CHA2DS2VASc score
8) [7].
2.3 Data analysis Baseline characteristics were described as mean ± standard deviation for continuous variables and counts and proportions for categorical variables. Differences of baseline characteristics between participants with hyperuricemia or normouricemia were tested by independent t-test for normally distributed variables and by the nonparametric Mann-Whitney U-test for skewed variables. Interactions between categorical variables were evaluated by Chi-square test. The thromboembolic events free survival curve was plotted using the Kaplan–Meier method with statistical significance examined by the log-rank. The association between risk factors and thromboembolism was assessed by univariate and multivariate Cox’s proportional hazard models. The multivariate analysis adjusted for potential confounding factor, left atria size. The discriminatory power of anthropometric measures was assessed by Harrell’s
C-statistic and validated internally through use of bootstrapping methods with 1000 replications.
Hosmer-Lemeshow
test
was
performed
to
evaluate
model
goodness-of-fit. Calibration curve was plotted by observed outcomes versus expected probabilities of thromboembolism. All data were analyzed using SPSS (version 25.0) and Microsoft Excel (version 2016). All statistical testings were 2-sided and statistical significances were set at p< 0.05. 3. Results 3.1 Basic characteristics A total of 446 HCM patients were enrolled in our study, thirty-one (6.95%) patients experienced thromboembolic events. Among them, four patients experienced peripheral embolisms, twenty-four patients experienced cerebral ischemic stroke (3 of them experience twice), the remaining three patients experienced both. There was 62.6%, 22.9%, 13%, 1.6% of patients assigning into low risk, low to moderate risk, moderate to high risk and very high risk group respectively. Detailed information about the study sample was described in Table 1. 3.2 Discrimination and calibration Kaplan-Meier curve illustrated the event free survival during follow-up according to risk stratification by R-CHA2DS2VASc model (P<0.0001) (Fig 1). Table 2 presented the results of cox proportional hazard analysis. HR of R-CHA2DS2VASc score for thromboembolism was 22.28 (P<0.001) for high risk group VS. low risk group, 5.63 (P=0.002) for moderate to high risk group VS. low risk group, and 2.97 (P=0.032) for low to moderate risk group VS. low risk group. After adjusting the potential
confounder, namely left atria size, hazard ratios were reduced but still remained significant. The Harrell’s C statistic was 0.7737 (95% CI: 0.65-0.89) for R-CHA2DS2VASc score. The internal validation results based on the bootstrapping method (1000 replicates) showed a bootstrap-corrected C- statistic of 0.7738 (optimism=0.0001). The Hosmer-Lemeshow test P value =0.435 (for R-CHA2DS2VASc score) demonstrated that there was no statistically significance between observed and predicted thromboembolic events, unveiling good calibration of the model (Fig 2). The R-CHA2DS2VASc score was well calibrated with 0.84 thromboembolic events per 100 person-years in the predefined low risk group, 1.84 in the low to moderate risk group, 4.67 in the moderate to high risk group, and 17.54 in the high risk group. 4. Discussion 4.1 Main findings In the present study, we found that R-CHA2DS2VASc score was well calibrated, showed good discriminative ability in the prediction of thromboembolism in HCM patients. Increasing R-CHA2DS2VASc score was associated with increasing thromboembolic risk for HCM patients, with strikingly 71.43% (5 out of 7) of patients with score 8 experiencing thromboembolism. Gradually increased HR of higher risk category in cox proportional analysis was also observed. The risk of developing thromboembolism was over 17 times higher for HCM in high risk category when compared to that of low risk category. Therefore, the R-CHA2DS2VASc score should be considered as a useful decision support tool in the care of patients with HCM.
4.2 AF and thromboembolism in HCM patients AF is the most frequent arrhythmia in HCM with the prevalence of 17.3% in our study cohort. The thromboembolic event rate was 20.78% in HCM with AF sub-cohort, which was 5-fold increased risk of that in HCM without AF sub-cohort (4.07%). Indeed, there was plenty of evidence supporting that AF was associated with overall disease deterioration and the incidence of stroke. For example, the frequency of ischemic stroke among HCM with AF patients was nearly 4 times as high as that for those without AF in a Japanese cohort [3]. In another study cohort with 900 HCM patients, 51 patients had stroke or other vascular events, 88% of them were complicated with AF [2]. The same phenomenon was observed in additional groups and a large meta-analysis [9-11]. Patients may even suffer from thromboembolism at the first episode of AF [4]. Moreover, anticoagulation for patients with AF has proven to be highly effective in reducing the incidence of thromboembolism. In that case, the new R-CHA2DS2VASc score, taking AF into consideration, objectively stressed the role of AF played in the development of thromboembolic events. What’s more, it could efficiently stratify thromboembolic risk in both full cohort and HCM-AF sub-cohort (P=0.029 by log rank in Kaplan-Meier analysis), which might contribute to the current unresolved clinical issue with regards to thromboembolic prediction of HCM patients since the original CHA2DS2VASc score was not well correlated with clinical outcomes for HCM patients. 4.3 Renal dysfunction and thromboembolism in HCM patients The association between renal dysfunction and stroke has been investigated in
previous studies. In a heavyweight meta-analysis, enrolling 83 studies (63 cohort studies and 20 RCTs) with 30 392 strokes from a pool of 2 253 741 participants in total, a linear relationship between GFR and risk of stroke was reported. The risk of stroke would increase 7% (relative ratio: 1.07, 95%CI: 1.04~1.09) for every 10 ml/min/1.73m2 decrease in GFR [12]. When further evaluate its impact for different stroke subtypes or in different populations, contradictory results may yield. In a general population of Swedish, gender-specific analyses showed that ischemic stroke was related to decreased GFR in both genders, however, hemorrhagic stroke was only related to renal dysfunction among women [13]. In the Rotterdam study, Bos et al demonstrated that no association was found between GFR and risk of overall stroke or risk of ischemic stroke, but only risk of hemorrhagic stroke with a significant dose-effect relationship [14]. In a Japanese cohort limited to type 2 diabetes, multivariate Cox analysis after adjusting albuminuria detected no significant association between GFR category and incident stroke and thus author suggested that the association between reduced GFR and stroke events may be mediated by albuminuria [15]. Yet according to the results of JDDM study 16, reduced GFR and albuminuria each act as an independent risk factor of cardiovascular diseases for patients with type 2 diabetes [16]. The correlation between renal dysfunction and stroke has been discussed in additional groups, including hypertensive population [17] and AF population [18]. Piccini’s team found that reduced creatinine clearance was a strong and independent predictor of stroke and systemic embolism for patients with nonvalvular AF and proposed another new renal function included stroke prediction
model-R2CHDS2 score [19], though external validations did not confirm the superiority of R2CHDS2 score over the old CHDS2 or CHA2DS2-VASc score [20]. However, little is known about its role in thromboembolic predictive ability for HCM patients. The HR of thromboembolism for patients with GFR<60 ml/min/1.73m2 was 4.198 (95%CI: 2.02~8.727, P=0.000) when compared to GFR>60 ml/min/1.73m2. The association was attenuated after adjusting confounding factors, including age, sex, hypertension, diabetes and AF, but statistical significance was still remained. (HR: 3.056, 95%CI: 1.206~7.746, P=0.019). In that case, inclusion of renal dysfunction as a predictor may be a new way to improve thromboembolic risk stratification for HCM population. BUN is a marker reflecting intravascular volume depletion or renal perfusion abnormality. Evidence of BUN’s predictive value for stroke is sparse. Wannamethee found that the risk of stroke significantly increased in the top decile of the ranked distribution of blood urea after adjusting potential confounders (RR: 1.5, 95%CI: 1.1~2,1) in their prospective study of middle-aged men [21]. Arnan’s team reported that postoperative BUN incremental was associated with stroke in cardiac surgical patients after excluding intraoperative influential variables. They suggested that the rise in BUN among patients with postoperative stroke signified renal vascular disease, and the possible coexistence of vascular disease in other arterial beds, such as the intracranial arteries explains stroke susceptibility [22]. In fact, BUN is more widely regarded as a prognostic biomarker for patients with heart failure (HF). It was highly associated with mortality, adverse outcomes and re-hospitalization [23-26] of HF
subjects. Patients with HF are prone to develop stroke or systemic embolism with cardioembolic origin due to left atria stasis. From this point of view, it may be more appropriate to take BUN into account for thromboembolic risk stratification in the later phase of HCM. Since HCM with HF (EF%<40%) only make up a small portion of our study cohort (2.02%), the statistical power of BUN in predicting thromboembolism was limited. 4.4 Revascularization procedure and thromboembolism in HCM patients This modified R-CHA2DS2VASc model was designed for stroke stratification in patients with myocardial infarction (MI). As we know, revascularization procedure is a main therapeutic method for MI. In the derived cohort, the lack of revascularization procedure doubled the risk of an ischemic stroke and the odds ratio was 2.1 (6.1% VS. 2.9%, P=0.003) [7]. While in our study cohort of HCM patients, only 11 patients had coronary artery stenosis larger than 70% and had percutaneous transluminal coronary angioplasty during coronary angiography. Besides, adding the revascularization score did not change the risk category of the 11 patients and none of them experienced stroke or systemic embolism. On the basis of our results, it was hard to define the impact of revascularization procedure on thromboembolism prediction in HCM patients. 4.5 Anticoagulation for HCM patients Additional considerations regarding anticoagulation for HCM patients with AF need to be stated. In our study cohort, the intention to treat analysis demonstrated a relative risk reduction for thromboembolism albeit the absence of statistical significance,
which probably due to the inclusion of patients who had discontinued anticoagulation therapy or whose international normalized ratio (INR) was subtherapeutic at the time of a thromboembolic event. Only 45.5% patients with HCM and AF were discharged with warfarin, which indicated under treatment of anticoagulation, leading to high incidence of thromboembolism and catastrophic outcomes. The overall prescription rate of anticoagulation therapy in Asia is reported to be lower than Western countries. As in GARFIELD-AF registry, only 50% patients with AF in Asia received anticoagulation therapy and those who did receive therapy were in therapeutic range for a lower proportion of time compared to those in other regions of the world [27]. The importance of anticoagulation in HCM patients at high risk of thromboembolism should be emphasized. Though current guidelines recommended longtime therapeutic anticoagulation for HCM patients with AF, those without documented AF meanwhile at high risk of thromboembolism would be missed out during clinical practice. Besides, given the fact that anticoagulation is at the cost of potentially increasing bleeding risk, it is still essential to explore an appropriate method to comprehensively evaluate the thromboembolic risk for each HCM patient and R-CHA2DS2VASc has shown its certain clinical effectiveness in this aspect in the present study. 5. Conclusions In conclusion, R-CHA2DS2VASc score was capable of stratifying thromboembolic risk for patients with and well calibrated in our study cohort. R-CHA2DS2VASc score should be considered during clinical practice in care of HCM patients. Among the 4 new parameters, AF and GFR largely contributed to the improvement of risk
stratification. While impact of the remaining 2 factors, BUN and revascularization procedure warrant further investigation. Several limitations of this study need to be addressed here. This is a retrospective study with relatively small sample size. Besides, patients who had thromboembolism only account for 6.95%. But this was in line with the prevalence reported from other study cohorts. Second, AF patients taking anticoagulation therapy might decrease the incidence of thromboembolism and underestimated the strength of the results. Third, the predictive power of BUN and revascularization was weaker than AF and GFR, further studies are needed to delineate their role in thromboembolic risk stratification for HCM patients. Fourth, we directly followed the same cut-off values for classification of patients as the original study. Resetting the value might be more precise considering the different backgrounds of the two studies. But we should notice that the trend of thromboembolic incidence increasing along with R-CHA2DS2VASc score is clear and sound. Taken together, lager and prospective studies are encouraged to
explore
the
performance
of
R-CHA2DS2VASc
score
in
predicting
thromboembolism for HCM population. Acknowledgement This study was supported by the National Natural Science Foundation of China (grant number: 81600299). Declarations of interest none Reference:
1.
Elliott P, McKenna WJ. Hypertrophic cardiomyopathy. The Lancet 2004;
363:1881-1891. 2.
Maron BJ, Olivotto I, Bellone P, Conte MR, Cecchi F, Flygenring BP, et al.
Clinical profile of stroke in 900 patients with hypertrophic cardiomyopathy. Journal of the American College of Cardiology 2002; 39:301-307. 3.
Higashikawa M, Nakamura Y, Yoshida M, Kinoshita M. Incidence of ischemic
strokes in hypertrophic cardiomyopathy is markedly increased if complicated by atrial fibrillation. Japanese circulation journal 1997; 61:673-681. 4.
Haruki S, Minami Y, Hagiwara N. Stroke and Embolic Events in Hypertrophic
Cardiomyopathy: Risk Stratification in Patients Without Atrial Fibrillation. Stroke 2016; 47:936-942. 5.
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; 137:263-272. 6.
Guttmann OP, Pavlou M, O'Mahony C, Monserrat L, Anastasakis A, Rapezzi C,
et al. Prediction of thrombo-embolic risk in patients with hypertrophic cardiomyopathy (HCM Risk-CVA). Eur J Heart Fail 2015; 17:837-845. 7.
Barra S, Almeida I, Caetano F, Providencia R, Paiva L, Dinis P, et al. Stroke
prediction with an adjusted R-CHA2DS2VASc score in a cohort of patients with a Myocardial Infarction. Thromb Res 2013; 132:293-299. 8.
Authors/Task Force m, Elliott PM, Anastasakis A, Borger MA, Borggrefe M,
Cecchi F, et al. 2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic
Cardiomyopathy of the European Society of Cardiology (ESC). Eur Heart J 2014; 35:2733-2779. 9.
Olivotto I, Cecchi F, Casey SA, Dolara A, Traverse JH, Maron BJ. Impact of
atrial fibrillation on the clinical course of hypertrophic cardiomyopathy. Circulation 2001; 104:2517-2524. 10.
Kubo T, Kitaoka H, Okawa M, Hirota T, Hayato K, Yamasaki N, et al. Clinical
impact of atrial fibrillation in patients with hypertrophic cardiomyopathy. Circulation Journal 2009; 73:1599-1605. 11.
Guttmann OP, Rahman MS, O'Mahony C, Anastasakis A, Elliott PM. Atrial
fibrillation and thromboembolism in patients with hypertrophic cardiomyopathy: systematic review. Heart 2014; 100:465-472. 12.
Masson P, Webster AC, Hong M, Turner R, Lindley RI, Craig JC. Chronic
kidney disease and the risk of stroke: a systematic review and meta-analysis. Nephrol Dial Transplant 2015; 30:1162-1169. 13.
Holzmann MJ, Aastveit A, Hammar N, Jungner I, Walldius G, Holme I. Renal
dysfunction increases the risk of ischemic and hemorrhagic stroke in the general population. Ann Med 2012; 44:607-615. 14.
Bos MJ, Koudstaal PJ, Hofman A, Breteler MM. Decreased glomerular filtration
rate is a risk factor for hemorrhagic but not for ischemic stroke: the Rotterdam Study. Stroke 2007; 38:3127-3132. 15.
Bouchi R, Babazono T, Nyumura I, Toya K, Hayashi T, Ohta M, et al. Is a
reduced estimated glomerular filtration rate a risk factor for stroke in patients with type 2 diabetes? Hypertens Res 2009; 32:381-386. 16.
Yokoyama H, Oishi M, Kawai K, Sone H, Japan Diabetes Clinical Data
Management Study G. Reduced GFR and microalbuminuria are independently associated with prevalent cardiovascular disease in Type 2 diabetes: JDDM study 16. Diabet Med 2008; 25:1426-1432. 17.
Zheng L, Sun Z, Zhang X, Li J, Hu D, Sun Y. The association between
glomerular filtration rate and stroke in hypertensive patients in rural areas of China. J Hypertens 2012; 30:901-907.
18.
Fanikos J, Burnett AE, Mahan CE, Dobesh PP. Renal Function Considerations
for Stroke Prevention in Atrial Fibrillation. Am J Med 2017; 130:1015-1023. 19.
Piccini JP, Stevens SR, Chang Y, Singer DE, Lokhnygina Y, Go AS, et al. Renal
dysfunction as a predictor of stroke and systemic embolism in patients with nonvalvular atrial fibrillation: validation of the R(2)CHADS(2) index in 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) and ATRIA (AnTicoagulation and Risk factors In Atrial fibrillation) study cohorts. Circulation 2013; 127:224-232. 20.
O'Brien EC, Holmes DN, Thomas L, Singer DE, Fonarow GC, Mahaffey KW, et
al. Incremental prognostic value of renal function for stroke prediction in atrial fibrillation. Int J Cardiol 2019; 274:152-157. 21.
Wannamethee SG, Shaper AG, Perry IJ. Serum creatinine concentration and risk
of cardiovascular disease: a possible marker for increased risk of stroke. Stroke 1997; 28:557-563. 22.
Arnan MK, Hsieh TC, Yeboah J, Bertoni AG, Burke GL, Bahrainwala Z, et al.
Postoperative blood urea nitrogen is associated with stroke in cardiac surgical patients. Ann Thorac Surg 2015; 99:1314-1320. 23.
Filippatos G, Rossi J, Lloyd-Jones DM, Stough WG, Ouyang J, Shin DD, et al.
Prognostic value of blood urea nitrogen in patients hospitalized with worsening heart failure: insights from the Acute and Chronic Therapeutic Impact of a Vasopressin Antagonist in Chronic Heart Failure (ACTIV in CHF) study. J Card Fail 2007; 13:360-364. 24.
Giamouzis G, Kalogeropoulos AP, Georgiopoulou VV, Agha SA, Rashad MA,
Laskar SR, et al. Incremental value of renal function in risk prediction with the Seattle Heart Failure Model. Am Heart J 2009; 157:299-305. 25.
Klein L, Massie BM, Leimberger JD, O'Connor CM, Pina IL, Adams KF, Jr., et
al. Admission or changes in renal function during hospitalization for worsening heart failure predict postdischarge survival: results from the Outcomes of a Prospective Trial of Intravenous Milrinone for Exacerbations of Chronic Heart Failure
(OPTIME-CHF). Circ Heart Fail 2008; 1:25-33. 26.
Cauthen CA, Lipinski MJ, Abbate A, Appleton D, Nusca A, Varma A, et al.
Relation of blood urea nitrogen to long-term mortality in patients with heart failure. Am J Cardiol 2008; 101:1643-1647. 27. Guttmann OP. Atrial fibrillation and thromboembolism in hypertrophic cardiomyopathy - An underestimated risk. Int J Cardiol 2018; 273:187-188.
Table 1: basic characteristics of study cohort. overall sample (N=446) Age (year) Male gender History of stroke HTN DM History of vascular disease congestive heart failure Revascularization AF discharged on wafarin discharged on aspirin discharged on clopidegrol LA (mm) LVOTO EF(%) GFR (mL/min/1.73 m2) GFR (mL/min/1.73 m2) ≥60/30-59/<30 BUN (mg/dl) BUN (mg/dl)≥25 TG (mmol/L) TC (mmol/L) HDLC (mmol/L) LDLC (mmol/L) TE during follow-up sroke-related mortality mean R-CHA2DS2-VASc score low risk category low to moderate category moderate to high category high category
56.1±15.2 251 22 141
56.3% 4.9% 31.6%
37
8.3%
32
7.2%
9 (2%) 11 (2.5%) 77 (17.3) 40 (51.9%) 82 (18.4%) 24 (5.4%) 40.4±7.2 176 (41.9%) 66.5±8.9 86.4±117.7 80.9%/16.1%/2.9% 19.6±12.1 72 (16.1%) 1.6±1.1 4.3±1 1.3±0.4 2.4±0.8 31 (6.95%) 11 (2.5%) 2.17±2.05 279 (62.56%) 102 (22.87%) 58 (13.00%) 7 (1.57%)
Abbreviations: HTN: hypertension, DM: diabetes mellitus, AF: atrial fibrillation, LA: left atria size, LVOTO: left ventricular outflow tract obstruction, EF: ejection fraction, GFR: glomerular filtration rate, BUN: blood urea nitrogen, TG: triglycerides, TC: total cholesterol,
HDLC: high density lipoprotein cholesterol, LDLC: low density lipoprotein cholesterol, TE: thromboembolism.
Table 2: univariate and multivariate cox proportional hazard models for thromboembolism in HCM patients. univariable analysis
Multivariable analysis
predictors
HR
95% CI
P-value
Age (year) sex prior TE vascular disease HF HTN DM AF LA (mm) LVOTO TC (mmol/L) TG (mmol/L) GFR (mL/min/1.73m2)
1.04 1.91 6.51 4.38 3.61 1.48 2.02 5.40 1.06 0.63 1.13 0.77
1.00-1.06 0.93-3.94 2.76-15.36 1.61-11.94 0.48-27.02 0.72-3.06 0.77-5.28 2.67-10.93 1.01-1.11 0.29-1.37 0.78-1.64 0.49-1.20
0.012 0.079 <0.001 0.004 0.211 0.291 0.152 <0.001 0.015 0.243 0.525 0.243
reference 4.90 NA 1.10
2.35-10.21 NA 0.39-3.15
<0.001 NA 0.857
60 30-59.9 <30
BUN (mg/dl) R-CHA2DS2VASc group low risk low to moderate risk moderate to high risk
reference 2.97 5.63
1.10-8.05 2.28-13.95
0.032 0.002
predictors
HR
95% CI
R-CHA2DS2VASc group low risk low to moderate risk moderate to high risk
reference 2.88 1.06-7.82 5.30 2.14-13.12
P-value
0.038 <0.001
high risk
22.28
7.46-66.56
<0.001
high risk
16.57 4.96-55.33
<0.001
Abbreviations: HR: hazard ratio, CI: confidence interval, HF: heart failure, for other abbreviations, see table 1. Left atria size was adjusted for multivariate analysis.
Figure legends: Fig 1: Event free survival rate stratified by R-CHA2DS2VASc risk categories. Fig 2: Calibration plot of R-CHA2DS2VASc score for HCM patients.
S1109-9666(19)30200-3
DOI:
https://doi.org/10.1016/j.hjc.2019.08.001
Reference:
HJC 426
To appear in:
Hellenic Journal of Cardiology
Received Date: 4 May 2019 Revised Date:
1 July 2019
Accepted Date: 21 August 2019
Please cite this article as: Wang Z, Liao H, He S, Chen X, Performance and Validation of RCHA2DS2VASc Score for Thromboembolism in Patients with Hypertrophic Cardiomyopathy, Hellenic Journal of Cardiology, https://doi.org/10.1016/j.hjc.2019.08.001. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Hellenic Society of Cardiology. Publishing services by Elsevier B.V. All rights reserved.
Performance and Validation of R-CHA2DS2VASc Score for Thromboembolism in Patients with Hypertrophic Cardiomyopathy Ziqiong Wang, Hang Liao, Sen He*, Xiaoping Chen* Department of Cardiology, West China Hospital, Sichuan University, Chengdu 610041, China Ziqiong Wang and Hang Liao contributed equally to this work. *Correspondence to: Xiaoping Chen and Sen He, Department of Cardiology, West China Hospital, Sichuan University, 37 Guo Xue Xiang, Chengdu 610041, Sichuan Province, People’s Republic of China. Tel: (+86) 028 85422343, Fax: +86 02885422175,
E-mail:
[email protected]
[email protected] (for Sen He).
(for
Xiaoping
Chen),
Abstract: Aims: To validate the modified R-CHA2DS2VASc score as a predictor of thromboembolism in HCM patients. Methods: A total of 446 HCM patients were enrolled in our study, thirty-one (6.95%) patients experienced thromboembolic events during the follow-up time of 1786.7 person-years. The association between R-CHA2DS2VASc score and risk of thromboembolism was assessed by Cox’s proportional hazard analysis. The discriminatory power of R-CHA2DS2VASc score for thromboembolism prediction was assessed by Harrell’s C-statistic and validated internally by bootstrapping methods. Calibration curve was plotted by observed outcomes versus expected probabilities of thromboembolism. Results: The R-CHA2DS2VASc score was well calibrated with 0.84 thromboembolic events per 100 person-years in the predefined low risk (R-CHA2DS2VASc score ≤2) group, 1.84 in the low to moderate risk (R-CHA2DS2VASc score 3-4) group, 4.67 in the moderate to high risk (R-CHA2DS2VASc score 5-7) group, and 17.54 in the high risk (R-CHA2DS2VASc score
8) group. Hazard ratios for thromboembolism were
2.88 (95%CI: 1.06-7.82, P=0.038) for low to moderate versus low risk group, 5.30 (95%CI: 2.14-13.12, P=0.0003) for moderate to high versus low risk group, and 16.57 5.30 (95%CI: 4.96-55.33, P<0.0001) for high versus low risk group after adjusting left atria size. The Harrell’s C statistic was 0.7737 (95% CI: 0.65-0.89) for R-CHA2DS2VASc score. Conclusion: The R-CHA2DS2VASc score has shown good calibration and
discriminative power in the prediction of thromboembolism for HCM patients. It should be considered as a potential decision support tool for HCM patients during clinical practice. Key words: hypertrophic cardiomyopathy, thromboembolism, risk stratification, R-CHA2DS2VASc score 1. Introduction Hypertrophic cardiomyopathy is a common genetically transmitted disease, defined clinically by the presence of unexplained left ventricular hypertrophy [1]. Thromboembolism is known to occur as complication of HCM [2]. Given the fact that thromboprophylaxis, including anti-platelet and oral anticoagulation therapy could potentially prevent the occurrence of thromboembolism. The identification of subjects who are at high risk is of great clinical significance. Though previous studies provided important insights into risk factors contributing to thromboembolism, such as older age, atrial fibrillation (AF), left atrial diameter [3-4]. Comprehensive risk stratification of thromboembolism in HCM patients is still a tough issue and remains to be resolved. CHA2DS2-VASc score is a clinical tool intended for estimating overall stroke risk in patients with atrial fibrillation (AF) [5]. Previous studies have shown that it was not closely predictive of the clinical outcome in patients with HCM [6]. Barra’s team proposed a new model
R-CHA2DS2VASc score, which was derived
from the original CHA2DS2-VASc score, comprising its primordial variables plus two renal function parameters [glomerular filtration rate (GFR) and blood urea nitrogen (BUN)], AF status and performance of a revascularization procedure
to predict
stroke for patients with myocardial infarction and it showed good calibration and discriminative performance in the derived cohort[7]. The aim of this study is to validate the new R-CHA2DS2VASc score in the prediction of thromboembolism in our HCM population. 2. Methods 2.1 Study population From 12-05-2008 to 10-31-2018, 458 HCM patients were identified at inpatient department of West China Hospital, Sichuan University, a tertiary referral center. Diagnosis was based on the echocardiographic demonstration of an unexplained increase in wall thickness ≥15mm, in the absence of abnormal load conditions [8]. Eight patients who were under 18 years old and 4 patients with incomplete biochemistry data were excluded in the study. The primary endpoint was defined as the occurrence of thromboembolism including ischemic stroke and systemic embolism, which was confirmed by imaging studies (computed tomography scan or magnetic resonance imaging and vascular ultrasound) during the follow-up. Follow-up was conducted by clinical consultations, medical records review and telephone interviews. Informed consents were obtained from all individual participants included in the study. 2.2 R-CHA2DS2VASc score definition The factors comprising the CHA2DS2-VASc score are heart failure, hypertension, diabetes, prior thromboembolic event, vascular disease, female sex, and age ≥ 65 years; 2 points are accorded for age ≥ 75 and prior stroke event, and one point for
each of the other factors [5]. R-CHA2DS2VASc score was derived from the original CHA2DS2-VASc score, comprising all its primordial variables plus two renal function parameters (GFR by MDRD formula and BUN), atrial fibrillation and performance of a revascularization procedure. 0 point is assigned for GFR≥60 mL/min, 1 point for 30-59.9 mL/min, 2 points for GFR<30 mL/min. Urea
25mg/dl would add 1 point to
the score. Presence of AF adds another 2 points. Revascularization procedure would minus 1 point. Patients were divided into the following four risk categories: low risk (R-CHA2DS2VASc score
2), intermediate-low risk (R-CHA2DS2VASc score 3-4),
intermediate-high risk (R-CHA2DS2VASc score 5-7), high risk (R-CHA2DS2VASc score
8) [7].
2.3 Data analysis Baseline characteristics were described as mean ± standard deviation for continuous variables and counts and proportions for categorical variables. Differences of baseline characteristics between participants with hyperuricemia or normouricemia were tested by independent t-test for normally distributed variables and by the nonparametric Mann-Whitney U-test for skewed variables. Interactions between categorical variables were evaluated by Chi-square test. The thromboembolic events free survival curve was plotted using the Kaplan–Meier method with statistical significance examined by the log-rank. The association between risk factors and thromboembolism was assessed by univariate and multivariate Cox’s proportional hazard models. The multivariate analysis adjusted for potential confounding factor, left atria size. The discriminatory power of anthropometric measures was assessed by Harrell’s
C-statistic and validated internally through use of bootstrapping methods with 1000 replications.
Hosmer-Lemeshow
test
was
performed
to
evaluate
model
goodness-of-fit. Calibration curve was plotted by observed outcomes versus expected probabilities of thromboembolism. All data were analyzed using SPSS (version 25.0) and Microsoft Excel (version 2016). All statistical testings were 2-sided and statistical significances were set at p< 0.05. 3. Results 3.1 Basic characteristics A total of 446 HCM patients were enrolled in our study, thirty-one (6.95%) patients experienced thromboembolic events. Among them, four patients experienced peripheral embolisms, twenty-four patients experienced cerebral ischemic stroke (3 of them experience twice), the remaining three patients experienced both. There was 62.6%, 22.9%, 13%, 1.6% of patients assigning into low risk, low to moderate risk, moderate to high risk and very high risk group respectively. Detailed information about the study sample was described in Table 1. 3.2 Discrimination and calibration Kaplan-Meier curve illustrated the event free survival during follow-up according to risk stratification by R-CHA2DS2VASc model (P<0.0001) (Fig 1). Table 2 presented the results of cox proportional hazard analysis. HR of R-CHA2DS2VASc score for thromboembolism was 22.28 (P<0.001) for high risk group VS. low risk group, 5.63 (P=0.002) for moderate to high risk group VS. low risk group, and 2.97 (P=0.032) for low to moderate risk group VS. low risk group. After adjusting the potential
confounder, namely left atria size, hazard ratios were reduced but still remained significant. The Harrell’s C statistic was 0.7737 (95% CI: 0.65-0.89) for R-CHA2DS2VASc score. The internal validation results based on the bootstrapping method (1000 replicates) showed a bootstrap-corrected C- statistic of 0.7738 (optimism=0.0001). The Hosmer-Lemeshow test P value =0.435 (for R-CHA2DS2VASc score) demonstrated that there was no statistically significance between observed and predicted thromboembolic events, unveiling good calibration of the model (Fig 2). The R-CHA2DS2VASc score was well calibrated with 0.84 thromboembolic events per 100 person-years in the predefined low risk group, 1.84 in the low to moderate risk group, 4.67 in the moderate to high risk group, and 17.54 in the high risk group. 4. Discussion 4.1 Main findings In the present study, we found that R-CHA2DS2VASc score was well calibrated, showed good discriminative ability in the prediction of thromboembolism in HCM patients. Increasing R-CHA2DS2VASc score was associated with increasing thromboembolic risk for HCM patients, with strikingly 71.43% (5 out of 7) of patients with score 8 experiencing thromboembolism. Gradually increased HR of higher risk category in cox proportional analysis was also observed. The risk of developing thromboembolism was over 17 times higher for HCM in high risk category when compared to that of low risk category. Therefore, the R-CHA2DS2VASc score should be considered as a useful decision support tool in the care of patients with HCM.
4.2 AF and thromboembolism in HCM patients AF is the most frequent arrhythmia in HCM with the prevalence of 17.3% in our study cohort. The thromboembolic event rate was 20.78% in HCM with AF sub-cohort, which was 5-fold increased risk of that in HCM without AF sub-cohort (4.07%). Indeed, there was plenty of evidence supporting that AF was associated with overall disease deterioration and the incidence of stroke. For example, the frequency of ischemic stroke among HCM with AF patients was nearly 4 times as high as that for those without AF in a Japanese cohort [3]. In another study cohort with 900 HCM patients, 51 patients had stroke or other vascular events, 88% of them were complicated with AF [2]. The same phenomenon was observed in additional groups and a large meta-analysis [9-11]. Patients may even suffer from thromboembolism at the first episode of AF [4]. Moreover, anticoagulation for patients with AF has proven to be highly effective in reducing the incidence of thromboembolism. In that case, the new R-CHA2DS2VASc score, taking AF into consideration, objectively stressed the role of AF played in the development of thromboembolic events. What’s more, it could efficiently stratify thromboembolic risk in both full cohort and HCM-AF sub-cohort (P=0.029 by log rank in Kaplan-Meier analysis), which might contribute to the current unresolved clinical issue with regards to thromboembolic prediction of HCM patients since the original CHA2DS2VASc score was not well correlated with clinical outcomes for HCM patients. 4.3 Renal dysfunction and thromboembolism in HCM patients The association between renal dysfunction and stroke has been investigated in
previous studies. In a heavyweight meta-analysis, enrolling 83 studies (63 cohort studies and 20 RCTs) with 30 392 strokes from a pool of 2 253 741 participants in total, a linear relationship between GFR and risk of stroke was reported. The risk of stroke would increase 7% (relative ratio: 1.07, 95%CI: 1.04~1.09) for every 10 ml/min/1.73m2 decrease in GFR [12]. When further evaluate its impact for different stroke subtypes or in different populations, contradictory results may yield. In a general population of Swedish, gender-specific analyses showed that ischemic stroke was related to decreased GFR in both genders, however, hemorrhagic stroke was only related to renal dysfunction among women [13]. In the Rotterdam study, Bos et al demonstrated that no association was found between GFR and risk of overall stroke or risk of ischemic stroke, but only risk of hemorrhagic stroke with a significant dose-effect relationship [14]. In a Japanese cohort limited to type 2 diabetes, multivariate Cox analysis after adjusting albuminuria detected no significant association between GFR category and incident stroke and thus author suggested that the association between reduced GFR and stroke events may be mediated by albuminuria [15]. Yet according to the results of JDDM study 16, reduced GFR and albuminuria each act as an independent risk factor of cardiovascular diseases for patients with type 2 diabetes [16]. The correlation between renal dysfunction and stroke has been discussed in additional groups, including hypertensive population [17] and AF population [18]. Piccini’s team found that reduced creatinine clearance was a strong and independent predictor of stroke and systemic embolism for patients with nonvalvular AF and proposed another new renal function included stroke prediction
model-R2CHDS2 score [19], though external validations did not confirm the superiority of R2CHDS2 score over the old CHDS2 or CHA2DS2-VASc score [20]. However, little is known about its role in thromboembolic predictive ability for HCM patients. The HR of thromboembolism for patients with GFR<60 ml/min/1.73m2 was 4.198 (95%CI: 2.02~8.727, P=0.000) when compared to GFR>60 ml/min/1.73m2. The association was attenuated after adjusting confounding factors, including age, sex, hypertension, diabetes and AF, but statistical significance was still remained. (HR: 3.056, 95%CI: 1.206~7.746, P=0.019). In that case, inclusion of renal dysfunction as a predictor may be a new way to improve thromboembolic risk stratification for HCM population. BUN is a marker reflecting intravascular volume depletion or renal perfusion abnormality. Evidence of BUN’s predictive value for stroke is sparse. Wannamethee found that the risk of stroke significantly increased in the top decile of the ranked distribution of blood urea after adjusting potential confounders (RR: 1.5, 95%CI: 1.1~2,1) in their prospective study of middle-aged men [21]. Arnan’s team reported that postoperative BUN incremental was associated with stroke in cardiac surgical patients after excluding intraoperative influential variables. They suggested that the rise in BUN among patients with postoperative stroke signified renal vascular disease, and the possible coexistence of vascular disease in other arterial beds, such as the intracranial arteries explains stroke susceptibility [22]. In fact, BUN is more widely regarded as a prognostic biomarker for patients with heart failure (HF). It was highly associated with mortality, adverse outcomes and re-hospitalization [23-26] of HF
subjects. Patients with HF are prone to develop stroke or systemic embolism with cardioembolic origin due to left atria stasis. From this point of view, it may be more appropriate to take BUN into account for thromboembolic risk stratification in the later phase of HCM. Since HCM with HF (EF%<40%) only make up a small portion of our study cohort (2.02%), the statistical power of BUN in predicting thromboembolism was limited. 4.4 Revascularization procedure and thromboembolism in HCM patients This modified R-CHA2DS2VASc model was designed for stroke stratification in patients with myocardial infarction (MI). As we know, revascularization procedure is a main therapeutic method for MI. In the derived cohort, the lack of revascularization procedure doubled the risk of an ischemic stroke and the odds ratio was 2.1 (6.1% VS. 2.9%, P=0.003) [7]. While in our study cohort of HCM patients, only 11 patients had coronary artery stenosis larger than 70% and had percutaneous transluminal coronary angioplasty during coronary angiography. Besides, adding the revascularization score did not change the risk category of the 11 patients and none of them experienced stroke or systemic embolism. On the basis of our results, it was hard to define the impact of revascularization procedure on thromboembolism prediction in HCM patients. 4.5 Anticoagulation for HCM patients Additional considerations regarding anticoagulation for HCM patients with AF need to be stated. In our study cohort, the intention to treat analysis demonstrated a relative risk reduction for thromboembolism albeit the absence of statistical significance,
which probably due to the inclusion of patients who had discontinued anticoagulation therapy or whose international normalized ratio (INR) was subtherapeutic at the time of a thromboembolic event. Only 45.5% patients with HCM and AF were discharged with warfarin, which indicated under treatment of anticoagulation, leading to high incidence of thromboembolism and catastrophic outcomes. The overall prescription rate of anticoagulation therapy in Asia is reported to be lower than Western countries. As in GARFIELD-AF registry, only 50% patients with AF in Asia received anticoagulation therapy and those who did receive therapy were in therapeutic range for a lower proportion of time compared to those in other regions of the world [27]. The importance of anticoagulation in HCM patients at high risk of thromboembolism should be emphasized. Though current guidelines recommended longtime therapeutic anticoagulation for HCM patients with AF, those without documented AF meanwhile at high risk of thromboembolism would be missed out during clinical practice. Besides, given the fact that anticoagulation is at the cost of potentially increasing bleeding risk, it is still essential to explore an appropriate method to comprehensively evaluate the thromboembolic risk for each HCM patient and R-CHA2DS2VASc has shown its certain clinical effectiveness in this aspect in the present study. 5. Conclusions In conclusion, R-CHA2DS2VASc score was capable of stratifying thromboembolic risk for patients with and well calibrated in our study cohort. R-CHA2DS2VASc score should be considered during clinical practice in care of HCM patients. Among the 4 new parameters, AF and GFR largely contributed to the improvement of risk
stratification. While impact of the remaining 2 factors, BUN and revascularization procedure warrant further investigation. Several limitations of this study need to be addressed here. This is a retrospective study with relatively small sample size. Besides, patients who had thromboembolism only account for 6.95%. But this was in line with the prevalence reported from other study cohorts. Second, AF patients taking anticoagulation therapy might decrease the incidence of thromboembolism and underestimated the strength of the results. Third, the predictive power of BUN and revascularization was weaker than AF and GFR, further studies are needed to delineate their role in thromboembolic risk stratification for HCM patients. Fourth, we directly followed the same cut-off values for classification of patients as the original study. Resetting the value might be more precise considering the different backgrounds of the two studies. But we should notice that the trend of thromboembolic incidence increasing along with R-CHA2DS2VASc score is clear and sound. Taken together, lager and prospective studies are encouraged to
explore
the
performance
of
R-CHA2DS2VASc
score
in
predicting
thromboembolism for HCM population. Acknowledgement This study was supported by the National Natural Science Foundation of China (grant number: 81600299). Declarations of interest none Reference:
1.
Elliott P, McKenna WJ. Hypertrophic cardiomyopathy. The Lancet 2004;
363:1881-1891. 2.
Maron BJ, Olivotto I, Bellone P, Conte MR, Cecchi F, Flygenring BP, et al.
Clinical profile of stroke in 900 patients with hypertrophic cardiomyopathy. Journal of the American College of Cardiology 2002; 39:301-307. 3.
Higashikawa M, Nakamura Y, Yoshida M, Kinoshita M. Incidence of ischemic
strokes in hypertrophic cardiomyopathy is markedly increased if complicated by atrial fibrillation. Japanese circulation journal 1997; 61:673-681. 4.
Haruki S, Minami Y, Hagiwara N. Stroke and Embolic Events in Hypertrophic
Cardiomyopathy: Risk Stratification in Patients Without Atrial Fibrillation. Stroke 2016; 47:936-942. 5.
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; 137:263-272. 6.
Guttmann OP, Pavlou M, O'Mahony C, Monserrat L, Anastasakis A, Rapezzi C,
et al. Prediction of thrombo-embolic risk in patients with hypertrophic cardiomyopathy (HCM Risk-CVA). Eur J Heart Fail 2015; 17:837-845. 7.
Barra S, Almeida I, Caetano F, Providencia R, Paiva L, Dinis P, et al. Stroke
prediction with an adjusted R-CHA2DS2VASc score in a cohort of patients with a Myocardial Infarction. Thromb Res 2013; 132:293-299. 8.
Authors/Task Force m, Elliott PM, Anastasakis A, Borger MA, Borggrefe M,
Cecchi F, et al. 2014 ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic
Cardiomyopathy of the European Society of Cardiology (ESC). Eur Heart J 2014; 35:2733-2779. 9.
Olivotto I, Cecchi F, Casey SA, Dolara A, Traverse JH, Maron BJ. Impact of
atrial fibrillation on the clinical course of hypertrophic cardiomyopathy. Circulation 2001; 104:2517-2524. 10.
Kubo T, Kitaoka H, Okawa M, Hirota T, Hayato K, Yamasaki N, et al. Clinical
impact of atrial fibrillation in patients with hypertrophic cardiomyopathy. Circulation Journal 2009; 73:1599-1605. 11.
Guttmann OP, Rahman MS, O'Mahony C, Anastasakis A, Elliott PM. Atrial
fibrillation and thromboembolism in patients with hypertrophic cardiomyopathy: systematic review. Heart 2014; 100:465-472. 12.
Masson P, Webster AC, Hong M, Turner R, Lindley RI, Craig JC. Chronic
kidney disease and the risk of stroke: a systematic review and meta-analysis. Nephrol Dial Transplant 2015; 30:1162-1169. 13.
Holzmann MJ, Aastveit A, Hammar N, Jungner I, Walldius G, Holme I. Renal
dysfunction increases the risk of ischemic and hemorrhagic stroke in the general population. Ann Med 2012; 44:607-615. 14.
Bos MJ, Koudstaal PJ, Hofman A, Breteler MM. Decreased glomerular filtration
rate is a risk factor for hemorrhagic but not for ischemic stroke: the Rotterdam Study. Stroke 2007; 38:3127-3132. 15.
Bouchi R, Babazono T, Nyumura I, Toya K, Hayashi T, Ohta M, et al. Is a
reduced estimated glomerular filtration rate a risk factor for stroke in patients with type 2 diabetes? Hypertens Res 2009; 32:381-386. 16.
Yokoyama H, Oishi M, Kawai K, Sone H, Japan Diabetes Clinical Data
Management Study G. Reduced GFR and microalbuminuria are independently associated with prevalent cardiovascular disease in Type 2 diabetes: JDDM study 16. Diabet Med 2008; 25:1426-1432. 17.
Zheng L, Sun Z, Zhang X, Li J, Hu D, Sun Y. The association between
glomerular filtration rate and stroke in hypertensive patients in rural areas of China. J Hypertens 2012; 30:901-907.
18.
Fanikos J, Burnett AE, Mahan CE, Dobesh PP. Renal Function Considerations
for Stroke Prevention in Atrial Fibrillation. Am J Med 2017; 130:1015-1023. 19.
Piccini JP, Stevens SR, Chang Y, Singer DE, Lokhnygina Y, Go AS, et al. Renal
dysfunction as a predictor of stroke and systemic embolism in patients with nonvalvular atrial fibrillation: validation of the R(2)CHADS(2) index in 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) and ATRIA (AnTicoagulation and Risk factors In Atrial fibrillation) study cohorts. Circulation 2013; 127:224-232. 20.
O'Brien EC, Holmes DN, Thomas L, Singer DE, Fonarow GC, Mahaffey KW, et
al. Incremental prognostic value of renal function for stroke prediction in atrial fibrillation. Int J Cardiol 2019; 274:152-157. 21.
Wannamethee SG, Shaper AG, Perry IJ. Serum creatinine concentration and risk
of cardiovascular disease: a possible marker for increased risk of stroke. Stroke 1997; 28:557-563. 22.
Arnan MK, Hsieh TC, Yeboah J, Bertoni AG, Burke GL, Bahrainwala Z, et al.
Postoperative blood urea nitrogen is associated with stroke in cardiac surgical patients. Ann Thorac Surg 2015; 99:1314-1320. 23.
Filippatos G, Rossi J, Lloyd-Jones DM, Stough WG, Ouyang J, Shin DD, et al.
Prognostic value of blood urea nitrogen in patients hospitalized with worsening heart failure: insights from the Acute and Chronic Therapeutic Impact of a Vasopressin Antagonist in Chronic Heart Failure (ACTIV in CHF) study. J Card Fail 2007; 13:360-364. 24.
Giamouzis G, Kalogeropoulos AP, Georgiopoulou VV, Agha SA, Rashad MA,
Laskar SR, et al. Incremental value of renal function in risk prediction with the Seattle Heart Failure Model. Am Heart J 2009; 157:299-305. 25.
Klein L, Massie BM, Leimberger JD, O'Connor CM, Pina IL, Adams KF, Jr., et
al. Admission or changes in renal function during hospitalization for worsening heart failure predict postdischarge survival: results from the Outcomes of a Prospective Trial of Intravenous Milrinone for Exacerbations of Chronic Heart Failure
(OPTIME-CHF). Circ Heart Fail 2008; 1:25-33. 26.
Cauthen CA, Lipinski MJ, Abbate A, Appleton D, Nusca A, Varma A, et al.
Relation of blood urea nitrogen to long-term mortality in patients with heart failure. Am J Cardiol 2008; 101:1643-1647. 27. Guttmann OP. Atrial fibrillation and thromboembolism in hypertrophic cardiomyopathy - An underestimated risk. Int J Cardiol 2018; 273:187-188.
Table 1: basic characteristics of study cohort. overall sample (N=446) Age (year) Male gender History of stroke HTN DM History of vascular disease congestive heart failure Revascularization AF discharged on wafarin discharged on aspirin discharged on clopidegrol LA (mm) LVOTO EF(%) GFR (mL/min/1.73 m2) GFR (mL/min/1.73 m2) ≥60/30-59/<30 BUN (mg/dl) BUN (mg/dl)≥25 TG (mmol/L) TC (mmol/L) HDLC (mmol/L) LDLC (mmol/L) TE during follow-up sroke-related mortality mean R-CHA2DS2-VASc score low risk category low to moderate category moderate to high category high category
56.1±15.2 251 22 141
56.3% 4.9% 31.6%
37
8.3%
32
7.2%
9 (2%) 11 (2.5%) 77 (17.3) 40 (51.9%) 82 (18.4%) 24 (5.4%) 40.4±7.2 176 (41.9%) 66.5±8.9 86.4±117.7 80.9%/16.1%/2.9% 19.6±12.1 72 (16.1%) 1.6±1.1 4.3±1 1.3±0.4 2.4±0.8 31 (6.95%) 11 (2.5%) 2.17±2.05 279 (62.56%) 102 (22.87%) 58 (13.00%) 7 (1.57%)
Abbreviations: HTN: hypertension, DM: diabetes mellitus, AF: atrial fibrillation, LA: left atria size, LVOTO: left ventricular outflow tract obstruction, EF: ejection fraction, GFR: glomerular filtration rate, BUN: blood urea nitrogen, TG: triglycerides, TC: total cholesterol,
HDLC: high density lipoprotein cholesterol, LDLC: low density lipoprotein cholesterol, TE: thromboembolism.
Table 2: univariate and multivariate cox proportional hazard models for thromboembolism in HCM patients. univariable analysis
Multivariable analysis
predictors
HR
95% CI
P-value
Age (year) sex prior TE vascular disease HF HTN DM AF LA (mm) LVOTO TC (mmol/L) TG (mmol/L) GFR (mL/min/1.73m2)
1.04 1.91 6.51 4.38 3.61 1.48 2.02 5.40 1.06 0.63 1.13 0.77
1.00-1.06 0.93-3.94 2.76-15.36 1.61-11.94 0.48-27.02 0.72-3.06 0.77-5.28 2.67-10.93 1.01-1.11 0.29-1.37 0.78-1.64 0.49-1.20
0.012 0.079 <0.001 0.004 0.211 0.291 0.152 <0.001 0.015 0.243 0.525 0.243
reference 4.90 NA 1.10
2.35-10.21 NA 0.39-3.15
<0.001 NA 0.857
60 30-59.9 <30
BUN (mg/dl) R-CHA2DS2VASc group low risk low to moderate risk moderate to high risk
reference 2.97 5.63
1.10-8.05 2.28-13.95
0.032 0.002
predictors
HR
95% CI
R-CHA2DS2VASc group low risk low to moderate risk moderate to high risk
reference 2.88 1.06-7.82 5.30 2.14-13.12
P-value
0.038 <0.001
high risk
22.28
7.46-66.56
<0.001
high risk
16.57 4.96-55.33
<0.001
Abbreviations: HR: hazard ratio, CI: confidence interval, HF: heart failure, for other abbreviations, see table 1. Left atria size was adjusted for multivariate analysis.
Figure legends: Fig 1: Event free survival rate stratified by R-CHA2DS2VASc risk categories. Fig 2: Calibration plot of R-CHA2DS2VASc score for HCM patients.