Our study shows the prognostic value of HAT₂CH₂ score for STE occurrence in older patients with CIEDs.

A variety of cardiac implantable electronic devices (CIED) are used in the elderly, including permanent pacemakers (PPM) (1), cardiac resynchronization therapy (CRT) (2), and implantable cardioverter defibrillators (ICD) (2). Atrial high-rate episodes (AHRE), commonly detected by CIED, are an important risk factor for new-onset atrial fibrillation (3) The latest European Society of Cardiology guidelines (3) regarding non-valvular atrial fibrillation (AF) state that CIED-detected AHRE > 5–6 min and > 180 bpm increase the risk for systemic thromboembolic events (STE). They recommend that AHRE ≥ 24 h should be closely monitored and treated. The CHA₂DS₂-VASc score is used for risk stratification of STE (4); however, its utility in non-valvular AF patients is controversial, primarily because this vascular scoring system does not include AF-related parameters (4). One meta-analysis showed that the discrimination power of the CHA₂DS₂-VASc score in predicting STE is modest and is similar in the presence or absence of non-valvular AF (4). Thus, more accurate STE prediction models are needed for assessing non-valvular AF and older patients with CIEDs.

The HAT₂CH₂ score, based on patient age and the presence of hypertension, stroke or transient ischemic attack, chronic obstructive pulmonary disease (COPD), and heart failure, was developed in 2010 for identifying patients who are likely to progress to sustained forms of AF in the near future (5). Studies have investigated the use of HAT₂CH₂ scores for predicting AF in cancer patients (6), in patients after coronary bypass surgery (7), and emergency-department patients (8). Other scoring systems for predicting new AF, including C₂HEST (9), mC₂HEST (10), and HAVOC (11, 12), have demonstrated acceptable discriminating power. For older patients with CIEDs, a small number of studies have investigated the performance of the HAT₂CH₂, C₂HEST, mC₂HEST, and HAVOC scoring systems for predicting new-onset atrial fibrillation and STE.

The present study aims to determine the performance of HAT₂CH₂ score for predicting STE and to compare this performance to that of AHRE ≥ 24 h and other scoring systems (CHA₂DS₂-VASc, C₂HEST, mC₂HEST, and HAVOC) in older patients with CIEDs and no history of AF.

Consecutive patients ≥18 years of age who underwent CIED implantation (Medtronic: dual chamber PPM, dual chamber ICD, CRTP, or CRTD) in the Cardiology Department of National Cheng Kung University Hospital from January 2015 to April 2021 were retrospectively included.

Between January 1, 2014 and April, 2021, a total of 453 consecutive patients who underwent Medtronic CIED implantation at National Cheng Kung University Hospital were recruited. Patients with previous AF (n = 105) and those age <65 years (n = 129) were excluded. The final analysis included 219 patients, of which 16 had experienced STE.

The median follow-up period was 35 months after CIED implantation. Table 1 shows patient baseline demographic and clinical characteristics according to the presence or absence of STE. The median age was 77 years, and 61.2% were men. Most patients were not obese. The types of CIED included dual chamber PPM (173; 79.0%), dual chamber ICD (66; 21.0%); CRTP (23; 7.3%); and CRTD (5; 1.6%). The most common indication for CIED implantation was sick sinus syndrome (53.9%), followed by atrioventricular block (25.1%) (Table 1). We observed an overall atrial pacing median of 33.5% and ventricular pacing median of 4.3%. High percentages of hypertension (92.2%), hyperlipidemia (86.3%), diabetes (52.1%), and CKD (39.7%) suggest a relatively high risk of STE for the entire study cohort. Ninety-two patients (42.0%) received antiplatelet agents and 21 patients (9.6%) received anticoagulants. Data regarding the type and incidence of STEs are reported in Table 2. The total number of STEs was 16 [incidence rate (IR), 2.51/100 patient-years; 95% CI, 1.65–5.48] (Table 2).

The main finding of this study is that the HAT₂CH₂ score is significantly and independently associated with STE in a population of older Taiwanese patients with CIED and no history of AF. The optimal HAT₂CH₂ score cutoff value for predicting subsequent STE was 3, with 100% sensitivity and 80% specificity. These results suggest that comprehensive assessment of older patients with CIED to determine the HAT₂CH₂ score may be warranted to allow for early, aggressive therapy to prevent STE.

The present study was conducted because the performance of several AF-predicting scoring systems (CHA₂DS₂-VASc, HAT₂CH₂, C₂HEST, mC₂HEST, and HAVOC) for predicting subsequent STE in older patients with CIED had not been confirmed. Of these scoring systems, CHA₂DS₂-VASc is recommended for predicting the risk of STE in patients with non-valvular AF (3). However, some studies report that the discrimination power of the CHA₂DS₂-VASc score for predicting STE in patients without non-valvular AF is insufficient (4, 13, 14). Melgaard et al. (15) showed the predictive accuracy of the CHA₂DS₂-VASc score for STE in patients with heart failure and without AF was modest (C-statistic, 0.64; 95% CI, 0.61–0.67). A meta-analysis by Siddiqi et al. (4) including 9 studies of patients (n = 453,747) with non-valvular AF and 10 studies of patients (n = 138262) without non-valvular AF reports that the discrimination power is modest (C-statistic, 0.67; 95% CI, 0.64–0.70) (4). Furthermore, Hu et al. used data from the Taiwan Health Insurance Research Database, reporting that for patients with venous thromboembolism (n = 56 996), the area under the curve of ROC for CHA₂DS₂-VASc score for predicting STE was 0.66, which also is modest (13). Therefore, more accurate models are needed for predicting STE in patients without AF.

To the best of our knowledge, the current study is the first to show that the HAT₂CH₂ score more accurately predicts the risk of STE in patients with CIED than does the CHA₂DS₂-VASc, C₂HEST, mCHEST, or HAVOC score (Table 1); the AUC of the ROC curve is 0.907, indicating excellent discrimination (Figure 1). We also proposed an STE risk stratification scheme based on the HAT₂CH₂ score (0–2, low risk; 3–5, medium risk; and 6–7, high risk) (Figure 3). Validation of this scheme in an external population is needed to confirm the accuracy of HAT₂CH₂ scores for predicting STE in patients with CIED.

We also observed that of the scoring systems investigated here, only the HAT₂CH₂ score could independently predict new-onset AF in this study population. The HAT₂CH₂ score includes COPD as one point instead of diabetes or vascular disease included as one point in the CHA₂DS₂-VASc score. Thus, different diseases have different effects on the risk of STE in patients with CIED. For example, systemic inflammation and oxidative stress are associated with COPD and promote platelet hyperactivity and cerebral vascular dysfunction (14), and COPD increases the risk of STE, independent of other shared risk factors of cardiovascular disease (16). Additional prospective studies are needed to elucidate the possible mechanisms underlying COPD-related STE risk and to identify effective preventive interventions.

Current guidelines recommend that patients with AHRE ≥ 24 h should be regarded as having the same risk of STE as those with clinical AF (3). A recent systemic review and meta-analysis also demonstrated that AHRE detected by CIED in patients without prior AF would increase the risks of stroke and clinical AF (17). Therefore, in the current study, we included AHRE ≥ 24 h (Table 1) and new AF (Table 1) in the multivariate Cox regression analysis for predicting STE. However, the HAT₂CH₂ score was still the only independent predictor of STE in our cohort. Although the proportion of STE associated with AF increases progressively with age (18), the cause of STE in a large proportion of patients is not cardioembolic. Non-cardioembolic risk factors for STE include atherosclerotic intracranial arterial stenosis (19), carotid atherosclerotic stenosis (20), and complex aortic plaque (21); these conditions share more risk factors with the HAT₂CH₂ score parameters than with the AHRE ≥ 24 h or AF. Therefore, the HAT₂CH₂ score provides a more comprehensive evaluation of STE risk factors in this population.

STEs are not uncommon in older patients after CIED implantation. The HAT₂CH₂ score is an independent predictor of STE risk in this population. Our results suggest that for patients with CIED and no history of AF, determination of the HAT₂CH₂ score may be warranted to allow for early, aggressive therapy to prevent STE, especially in those with HAT₂CH₂ score ≥ 3.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Ethics Committee of National Cheng Kung University Hospital. All patients provided signed informed consent at the time of CIED implantation for later use of their data in publications.

J-YC: conception and design, data analysis and interpretation, statistical analysis, drafting and finalizing the manuscript, and critical revision of the manuscript for important intellectual content. T-WC and W-DL: data acquisition. All authors contributed to the article and approved the submitted version.

The authors would like to thank the Ministry of Science and Technology of the Republic of China, Taiwan, for financially supporting this research under contract MOST 110-2218-E-006-017.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.