Author + information
- Published online October 1, 2018.
- Hong-Mi Choi, MD,
- Yeonyee E. Yoon, MD,
- Il-Young Oh, MD,
- Youngjin Cho, MD and
- Goo-Yeong Cho, MD, PhD∗ ()
- ↵∗Department of Internal Medicine, Seoul National University College of Medicine, Cardiovascular Center, Seoul National University Bundang Hospital, 82, Gumi-ro 173 Beon-gil, Bundang-gu, Seongnam, Gyeonggi-do, 13620, Republic of Korea
Silent atrial fibrillation (SAF) detected by cardiac implantable electronic device (CIED) is known to increase the risk of death, stroke, and heart failure (HF), which is comparable to overt atrial fibrillation (AF) (1). Left atrial (LA) strain, an emerging parameter of LA function, was found to predict the development and progression of AF in previous reports (2,3). We investigated whether global LA strain (GLAS) can predict the development of SAF and clinical outcomes after CIED implantation in patients without AF.
We retrospectively analyzed a consecutive, single-center cohort of patients with sinus rhythm who received a dual-chamber pacemaker or implantable cardioverter-defibrillator from January 2007 to December 2014. Among 235 patients, we excluded patients with a history of AF, moderate to severe mitral stenosis or prosthetic mitral valve, absence or sensing failure of atrial lead, and patients prescribed amiodarone. A total of 127 patients were finally analyzed, after exclusion of patients without echocardiography appropriate for strain measurement and electronic analysis of pacemaker system within 6 months.
Two-dimensional echocardiography was performed before CIED implantation. Two independent authors blinded to the patients’ clinical data measured GLAS by speckle-tracking echocardiography using an off-line workstation (Image-Arena version 4.6, Tomtec Imaging System, Unterschleissheim, Germany). GLAS was calculated by averaging peak segmental LA strain of the 8 segments from apical 4- and 2-chamber views after excluding 4 segments representing LA roof. Reference point was the onset of QRS complex. SAF was detected by more than 5 min of atrial high rate episodes, defined as atrial rate over 171 or 180 beats/min. The primary endpoint was development of SAF at 6 months. All-cause death, stroke, and admission for HF were defined as secondary composite endpoint.
Of 127 patients (68.8 ± 13.7 years of age, 53 men), SAF was detected in 17 patients (13.4%) during 6 months. During follow-up of 1,471.5 ± 747.9 days, death occurred in 5 patients (3.9%), stroke in 4 (3.1%), and admission for HF in 6 (4.7%). All 6 patients who were admitted because of HF had lower GLAS but normal left ventricular ejection fraction (LVEF), and 2 of them had SAF. Of 4 patients who developed stroke, 2 had SAF during follow-up. Baseline characteristics were well-balanced between the patients with and without primary and secondary endpoints, except for higher age and CHA2DS2-VASc score in patients with secondary endpoint. Among echocardiographic parameters, patients with primary endpoint showed significantly larger left atrial volume index (LAVI) and lower GLAS (49.1 ± 18.5 ml/m2 vs. 39.3 ± 15.7 ml/m2, p = 0.002 for LAVI; 31.1 ± 11.7% vs. 38.9 ± 14.3%; p = 0.035 for GLAS).
Univariate Cox proportional hazard analysis demonstrated that LAVI and GLAS were associated with the development of the primary endpoint. We used the cutoff value of GLAS 37.4%, drawn from the receiver-operating characteristic curve for prediction of primary endpoint with the highest sum of sensitivity and specificity (sensitivity 88.2%, specificity 55.2%). The area under the curve was 0.660 (p = 0.034). After adjustment for age, LVEF, and LAVI, GLAS <37.4% was an independent predictor of SAF during 6 months after device implantation (hazard ratio [HR]: 8.82; 95% confidence interval [CI]: 1.82 to 42.74; p = 0.004) (Figure 1A). GLAS <37.4% significantly improved continuous net reclassification improvement (0.402; p = 0.027) and integrated discrimination improvement (0.066; p = 0.020), compared with the model using traditional risk factors. Two-variable model was generated, and the same result was drawn that GLAS <37.4% was an independent predictor of SAF after adjustment of LAVI (HR: 7.382; 95% CI: 1.64 to 33.27; p = 0.009). GLAS <37.4% also independently predicted the secondary composite endpoint after adjustment for CHA2DS2-VASc score (HR: 5.43; 95% CI: 1.14 to 25.87; p = 0.034) (Figure 1B).
We found that: 1) low GLAS was a good predictor of early SAF after CIED implantation in patients without previous AF; and 2) low GLAS, as well as CHA2DS2-VASc score, could predict the composite endpoint after adjustment for LVEF and LAVI. A small number of events and over-fitting problem are limitations. Notwithstanding some technical issues in measuring LA strain and inherent weakness from retrospective study, our results suggests the value of GLAS as a predictor of SAF and that frequent follow-up of electronic analysis is warranted in patients with lower baseline GLAS.
Please note: The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Hoit B.D.
- Kosmala W.,
- Saito M.,
- Kaye G.,
- et al.