A, Preoperative and postoperative left ventricular ejection fraction (LVEF). Vertical lines indicate standard deviation. B, Preoperative and postoperative left ventricular global longitudinal strain (LV-GLS). Vertical lines indicate standard deviation. C, χ2 for composite events demonstrating incremental prognostic utility of LV-GLS vs Society of Thoracic Surgeons (STS) score. D, Kaplan-Meier curves of the study sample, separated on the basis of normal vs abnormal LV-GLS during follow-up. AVR indicates aortic valve replacement.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Kafa R, Kusunose K, Goodman AL, et al. Association of Abnormal Postoperative Left Ventricular Global Longitudinal Strain With Outcomes in Severe Aortic Stenosis Following Aortic Valve Replacement. JAMA Cardiol. 2016;1(4):494–496. doi:https://doi.org/10.1001/jamacardio.2016.1132
Owing to improved survival, aortic valve replacement is a class I indication in patients with severe aortic stenosis and associated symptoms or cardiac dysfunction.1,2 However, a recent study has demonstrated that despite excellent early postoperative survival, almost 20% of patients who received aortic valve replacement for severe aortic stenosis were dead at 5 years.3 Because most patients with aortic stenosis have preserved left ventricular ejection fraction (LVEF), we hypothesized that preoperative and postoperative left ventricular global longitudinal strain (LV-GLS) may be a more sensitive way to detect subclinical left ventricular dysfunction and provide incremental prognostic value. We sought to study the effect of aortic valve replacement on postoperative LV-GLS in patients with severe aortic stenosis and preserved LVEF and to associate the change from preoperative to postoperative LV-GLS with outcomes after aortic valve replacement.
This was a retrospective observational study, approved by the Cleveland Clinic Institutional Review Board, conducted from January 1, 2003, to December 31, 2007, of 208 patients with severe symptomatic aortic stenosis (aortic valve area <1 cm2) and baseline LVEF of 50% or more who underwent subsequent aortic valve replacement during follow-up. Because of the nature of the study, the requirement for informed consent was waived. All patients had additional follow-up echocardiography performed 12 to 24 months after aortic valve replacement. Standard clinical and surgical data were obtained. Echocardiographic measurements and LV-GLS (measured offline; Velocity Vector Imaging4) were assessed before aortic valve replacement as well as 12 and 24 months after aortic valve replacement. We used a cutoff value of 14.5% or greater for LV-GLS (with lower numbers representing worse values) to define abnormal strain.5 A composite end point of death or admission for congestive heart failure (excluding admissions owing to rapid atrial fibrillation) was recorded. Statistical analysis was performed from October 1 to November 30, 2015, using SPSS, version 11.5 (SPSS Inc). P < .05 was considered significant.
The median time between aortic valve replacement and follow-up echocardiography was 14 months (interquartile range, 12-16). Correlation between baseline LVEF and LV-GLS was weak but significant (β, 0.23; P = .001). One hundred ninety-six patients (94.2%) had a bioprosthesis and 58 (27.9%) had obstructive coronary artery disease requiring concomitant bypass grafting (Table). Left ventricular ejection fraction was preserved with no postoperative deterioration (mean [SD]; preoperative, 59% [4%] vs postoperative, 59% [5%]; P = .60) (Figure, A), while left ventricular mass index regressed significantly (mean [SD]; preoperative, 118  vs postoperative, 101  g/m2; P < .01). However, despite a significant improvement in overall LV-GLS from baseline (mean [SD]; preoperative, –14.8% [4%] vs postoperative, –17.2% [3%]; P < .001) (Figure, B), 38 patients (18.3%) had abnormal LV-GLS values during follow-up. Of the 99 patients who had abnormal baseline LV-GLS values, the values normalized in 76 (76.8%), while values remained abnormal in 23 (23.2%) after aortic valve replacement.
At a mean (SD) of 3.9 (2.0) years, there were 38 end point events (15 deaths and 23 admissions for congestive heart failure). On multivariable Cox proportional hazards regression analysis, after adjustment for baseline LV-GLS and left ventricular stroke volume index, a worsening Society of Thoracic Surgeons score (hazard ratio [HR], 1.06; 95% CI, 1.02-1.10; P = .001) and abnormal follow-up LV-GLS value (time-dependent covariate HR, 2.76; 95% CI, 1.40-5.45; P = .003) were associated with higher composite end point events. Results were similar when coronary artery disease was excluded. Addition of postoperative LV-GLS values to the Society of Thoracic Surgeons score provided incremental prognostic utility (Figure, C). The C statistic for the Society of Thoracic Surgeons score was 0.61 (95% CI, 0.55-0.66), which increased to 0.69 (95% CI, 0.61-0.75) for the Society of Thoracic Surgeons score plus abnormal vs normal follow-up LV-GLS values (P < .01). Kaplan-Meier curves (normal vs abnormal follow-up LV-GLS values) are shown in Figure, D. An improvement of 1% in absolute LV-GLS value (from preoperative to postoperative echocardiography) was associated with fewer composite events (HR, 0.92; 95% CI, 0.84-0.97; P < .01). However, neither baseline LVEF (HR, 0.98; 95% CI, 0.95-1.06; P = .70) or absolute change in LVEF (HR, 0.99; 95% CI, 0.96-1.03; P = .40) were associated with composite events.
In patients with severe aortic stenosis, approximately 20% of patients who survived more than 1 year after aortic valve replacement had an abnormal LV-GLS value on postoperative echocardiography, despite a preserved postoperative LVEF and demonstrable left ventricular mass regression. This finding was independently associated with adverse events. Appropriately timed aortic valve replacement relieves left ventricular wall stress and prevents a decline in LVEF. However, despite a preserved LVEF, there may be irreversible impairment of LV-GLS postoperatively, with resultant adverse outcomes. Whether or not this is due to intrinsic myocardial disease (eg, myocardial fibrosis6) remains to be conclusively established. The data are hypothesis generating and need prospective validation.
Accepted for Publication: March 31, 2016.
Corresponding Author: Milind Y. Desai, MD, Center for Heart Valve Disease, Heart and Vascular Institute, Cleveland Clinic, 9500 Euclid Ave, Desk J1-5, Cleveland, OH 44195 (email@example.com).
Published Online: June 1, 2016. doi:10.1001/jamacardio.2016.1132.
Author Contributions: Dr Desai had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Kafa, Goodman, Griffin, Desai.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Kafa, Goodman, Desai.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Kafa, Kusunose, Goodman, Desai.
Administrative, technical, or material support: Svensson, Griffin, Desai.
Study supervision: Griffin, Desai.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Create a personal account or sign in to: