Modified CHADS2 shows best predictive performance for stroke across spectrum of kidney function
Validation of risk scores for ischaemic stroke in atrial fibrillation across the spectrumof kidney function
Literature - De Jong Y et al. - Eur Heart J. 2021 Mar 26;ehab059. doi: 10.1093/eurheartj/ehab059.Introduction and methods
The estimated prevalence of CKD is 10-15% in de general population and is increasing [1]. CKD is associated with an increased risk of ischemic stroke (IS) [2]. The increasing prevalence of CKD can thereby partly explain an increase in prevalence of IS [3,4]. For personalized anticoagulation therapy, risks scores for IS are important to estimate the risk of IS vs the risk of treatment-related bleeding. However, the predictive performance of commonly used risk scores for IS in patients with CKD is unclear. This study validated six risk scores for IS in patients with AF across the spectrum of kidney function.
This study used data from 36004 subjects with new-onset AF from the Stockholm CREAtinine Measurements (SCREAM) project [5]. The study outcome was hospitalization for IS or IS as main cause of death. The following risk scores were validated: AFI [6], CHADS2 [7], Modified CHADS2 [8], CHA2DS2-VASc [9], ATRIA [10], and GARFIELD-AF [11]. The predictive performance of these risk scores was evaluated by discrimination and calibration ability across three categories kidney function: Normal kidney function (eGFR >60mL/min/1.73 m²), mild CKD (eGFR 30–60mL/min/1.73 m²), and advanced CKD (eGFR<30mL/min/1.73 m²). Discrimination was assessed by c-statistic. The c-statistic lies between 0.5 and 1.0 and reflects the ability of the risk score to distinguish between patients with and without the outcome. A c-statistic <0.7 is considered poor to moderate, 0.8 good, and >0.9 excellent. Calibration ability reflects the agreement between the predicted and actual observed probabilities of the outcome.
Main results
- The highest and most consistent discriminatory ability in patients with AF across all categories of kidney function was observed with the Modified CHADS2 score (normal kidney function: c-statistic= 0.78, 95%CI 0.77-0.79; mild CKD: 0.73, 95%CI 0.71-0.74; advanced CKD: 0.74, 95%CI 0.69-0.79).
- The c-statistics for the CHADS2 score were relatively stable (normal kidney function: 0.78, 95%CI 0.77-0.80; mild CKD: 0.70, 95%CI 0.68-0.72; advanced CKD: 0.71, 95%CI 0.66-0.76). C-statistics for AFI score decreased across categories of kidney function (normal kidney function: 0.68, 95%CI 0.67-0.69; mild CKD: 0.58, 95%CI 0.57-0.59; advanced CKD: 0.55, 95%CI 0.51-0.59). Discrimination of the CHA2DS2-VASc was moderate in AF patients with normal kidney function, but poor in mild and advanced CKD (normal kidney function: 0.70, 95%CI 0.69-0.71; mild CKD: 0.60, 95%CI 0.58-0.62; advanced CKD: 0.58, 95%CI 0.52-0.64). Discrimination of ATRIA and GARFIELD-AF was good in AF patients with normal kidney function, but moderate in mild and advanced CKD (ATRIA: normal kidney function: 0.78, 95%CI 0.76-0.79; mild CKD: 0.68, 95%CI 0.66-0.70; advanced CKD: 0.66, 95%CI 0.60-0.72. GARFIELD-AF: normal kidney function: 0.76, 95%CI 0.75-0.77; mild CKD: 0.67, 95%CI 0.65-0.69; advanced CKD: 0.70, 95%CI 0.64-0.76).
- The Modified CHADS2 score showed good calibration in the normal eGFR and mild CKD category, but slightly overpredicted risks in the advanced CKD category. The CHADS2 score, CHA2DS2-VASc score and ATRIA score underpredicted the of risk of IS in all three kidney function categories, while the AFI score showed overprediction in all categories. GARFIELD-AF underpredicted the risk of IS in the normal eGFR category, but showed overprediction in the mild and advanced CKD categories.
- The effect of the prediction timeframe on the performance of risk scores was assessed and showed that c-statistics were relatively stable over time. For the optimal calibration in the large, the optimal prediction timeframe was shorter than the timeframe in de development studies for CHADS2 (optimal at 6 months, developed for 12 months), CHA2DS2-VASc (optimal at 1 month, developed for 12 months), ATRIA (optimal at 17 months, validated at 29 months) and GARFIELD-AF (optimal at 9 months, developed for 12 months), and longer for AFI (optimal at 49 months, validated at 28 months). The Modified CHADS2 score was developed for 60 months, but did not reach an optimal prediction timeframe within 72 months.
Conclusion
Most studied risk scores showed moderate to good discrimination in AF patients with normal eGFR, but discrimination decreased in patients with mild or advanced CKD. Calibration was largely independent of eGFR. The Modified CHADS2 score showed good performance for discrimination and calibration in all three kidney function categories. The authors of the article therefore suggest that the Modified CHADS2 score is the preferred risk score in clinical practice.
References
1. Turakhia MP, Blankestijn PJ, Carrero J-J, Clase CM, Deo R, Herzog CA, Kasner SE, Passman RS, Pecoits-Filho R, Reinecke H, Shroff GR, Zareba W, Cheung M, Wheeler DC, Winkelmayer WC, Wanner C, Amann K, Banerjee D, Bansal N, Boriani G, Bunch J, Chan CT, Charytan DM, Conen D, Friedman AN, Genovesi S, Holden RM, House AA, Jadoul M, Jardine AG, Johnson DW, Jun M, Labriola L, Mark PB, McCullough PA, Nolin TD, Potpara TS, Pun PH, Ribeiro ALP, Rossignol P, Shen JI, Sood MM, Tsukamoto Y, Wang AY-M, Weir MR, Wetmore JB, Wranicz JK, Yamasaki H, Conference Participants. Chronic kidney disease and arrhythmias: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) controversies conference. Eur Heart J 2018;39:2314–2325.
2. Kumar S, Lim E, Covic A, Verhamme P, Gale CP, Camm AJ, Goldsmith D. Anticoagulation in concomitant chronic kidney disease and atrial fibrillation: JACC review topic of the week. J Am Coll Cardiol 2019;74:2204–2215.
3. Cherng Y-G, Lin C-S, Shih C-C, Hsu Y-H, Yeh C-C, Hu C-J, Chen T-L, Liao C-C. Risk and outcomes in patients with chronic kidney disease or end-stage renal disease: two nationwide studies. PLoS One 2018;13:e0191155.
4. 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.
5. Runesson B, Gasparini A, Qureshi AR, Norin O, Evans M, Barany P, Wettermark B, Elinder CG, Carrero JJ. The Stockholm CREAtinine Measurements (SCREAM) project: protocol overview and regional representativeness. Clin Kidney J 2016;9:119–127.
6. Atrial Fibrillation Investigators. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch Intern Med 1994;154:1449–1457.
7. Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001;285:2864–2870.
8. Rietbrock S, Heeley E, Plumb J, van Staa T. Chronic atrial fibrillation: incidence, prevalence, and prediction of stroke using the Congestive heart failure, Hypertension, Age >75, Diabetes mellitus, and prior Stroke or transient ischemic attack (CHADS2) risk stratification scheme. Am Heart J 2008;156:57–64.
9. 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.
10. Singer DE, Chang Y, Borowsky LH, Fang MC, Pomernacki NK, Udaltsova N, Reynolds K, Go AS. A new risk scheme to predict ischemic stroke and other thromboembolism in atrial fibrillation: the ATRIA study stroke risk score. J Am Heart Assoc 2013;2:e000250.
11. Fox KAA, Lucas JE, Pieper KS, Bassand JP, Camm AJ, Fitzmaurice DA, Goldhaber SZ, Goto S, Haas S, Hacke W, Kayani G, Oto A, Mantovani LG, Misselwitz F, Piccini JP, Turpie AGG, Verheugt FWA, Kakkar AK. Improved risk stratification of patients with atrial fibrillation: an integrated GARFIELD-AF tool for the prediction of mortality, stroke and bleed in patients with and without anticoagulation. BMJ Open 2017;7:e017157.
Facebook Comments