Abstract
Objective
To examine markers of improvement in left ventricular (LV) systolic function after coronary artery bypass grafting (CABG), assessed by changes (Δ) in LV ejection fraction (LVEF) and global longitudinal strain (GLS), and to evaluate the association between ΔLVEF/ΔGLS and all-cause mortality.
Methods
Patients who underwent isolated CABG from 2006 to 2011 at Gentofte Hospital with a preoperative echocardiography within 30 days before and postoperative echocardiography 30 to 365 days after surgery were included. Multivariable linear regression identified clinical and echocardiographic predictors of ΔLVEF and ΔGLS. Cox proportional hazards regression models related these changes to all-cause mortality.
Results
Among 226 patients (median age, 67 years, 14% women), LVEF improved from 48.5% ± 12.8% (pre-CABG) to 53.9% ± 9.8% (post-CABG) and GLS from 12.8% ± 4.4% to 13.6% ± 4.8% (absolute values, both P values < .001). Lower preoperative LVEF and GLS were the strongest independent determinants of postoperative improvement (β, −0.65; 95% CI, −0.75 to −0.55; and β, −0.51; 95% CI, −0.63 to −0.39; both P values < .001). Larger preoperative LV internal end-diastolic diameter was associated with greater LVEF improvement (P = .014), whereas higher atrial strain and tricuspid annular plane systolic excursion were associated with greater GLS improvement (both P values < .05). Chronic kidney disease (P = .040) and a history of heart failure (P = .024) were associated with less GLS improvement. Each 1% absolute-increase in ΔLVEF or ΔGLS was associated with lower mortality after multivariable adjustment (hazard ratio, 0.88; 95% CI, 0.83 to 0.93, and 0.66; 95% CI, 0.56 to 0.78, respectively; both P values < .001).
Conclusions
Postoperative improvements in LVEF and GLS were common, occurred most often in patients with lower pre-CABG LV function, and were independently associated with reduced mortality.
Key Words: coronary artery bypass grafting, left ventricular function, global longitudinal strain, echocardiography, long-term prognosis
Graphical Abstract
Schematic representation of the study overview.
Central Message.
Postoperative echocardiography after CABG reveals heterogeneous recovery of left ventricular systolic function and provides prognostically relevant information beyond preoperative assessment.
Perspective.
Routine postoperative echocardiography after CABG may provide clinically relevant information on recovery of left ventricular systolic function and long-term prognosis. Characterizing postoperative recovery trajectories, including strain, may support postoperative planning, particularly in patients with impaired preoperative function.
Coronary artery bypass grafting (CABG) remains among the most frequently performed cardiac surgical procedures worldwide, with approximately 800 000 surgeries conducted annually.1 Despite advances in surgical techniques and postoperative care, CABG candidates continue to have substantial cardiovascular risk due to advanced coronary artery disease and multiple comorbidities.2, 3, 4 Although preoperative imaging remains fundamental to risk stratification, systematic postoperative assessment might be equally important for evaluating cardiac recovery and long-term outcomes.5,6 Echocardiography plays a central role in this context, allowing detailed characterization of left ventricular (LV) function and detection of valvular abnormalities.7 Although preoperative echocardiographic predictors of prognosis after CABG have been well described,8 data based on paired pre- and postoperative echocardiography, and the prognostic relevance of postoperative recovery, remain limited in real-world populations. Previous studies suggest that CABG can lead to recovery of LV systolic function but results have been inconsistent.9, 10, 11 Beyond systolic function, CABG may influence diastolic filling, left atrial (LA) function, and right ventricular (RV) function,12,13 yet these aspects have received less systematic attention. A more comprehensive understanding of postoperative cardiac adaptation could refine patient selection, inform follow-up strategies, and improve long-term prognostic assessment.
The present study built upon a previously characterized CABG cohort in which preoperative echocardiographic parameters were investigated as prognostic markers.8 Here, we extended this work by analyzing paired pre- and postoperative echocardiograms to explore how cardiac structure and function evolved after CABG. The aim of this study was to investigate changes in cardiac structure and function from before to after CABG and to identify echocardiographic and clinical parameters associated with postoperative LV systolic improvement as well as all-cause mortality.
Methods
Study Design and Population
This was a retrospective longitudinal study of consecutively included patients undergoing isolated CABG at Copenhagen University Hospital, Gentofte, between January 2006 and May 2011. This study was derived from a previously characterized cohort in which preoperative echocardiographic predictors of outcome had been investigated.8 For the present study, patients were required to have both a preoperative transthoracic echocardiograph performed within 30 days before CABG and a postoperative echocardiogram obtained 30 to 365 days after surgery. If multiple postoperative examinations were available, the earliest eligible examination was selected. This 30- to 365-day interval was chosen to capture early-to-intermediate postoperative recovery, consistent with prior CABG imaging studies.14 Additionally, patients who had undergone rescue CABG, prior or concomitant valve surgery, or had at least moderate aortic or mitral valve disease were also excluded. Clinical, demographic, and perioperative data were obtained from electronic medical records. Echocardiographic data were obtained as part of the routine clinical care.
Ethics and Patient Consent
In Denmark, ethics committee approval is not required for retrospective studies. The study was approved by the Danish Data Protection Agency and the Danish Health and Medicines Authority. Patient consent was not required.
Echocardiographic Assessment
All transthoracic echocardiograms were acquired using GE Vivid ultrasound systems (GE Healthcare). Examinations were stored digitally and analyzed using EchoPAC software (GE Vingmed Ultrasound AS). Quantitative measurements were performed in accordance with clinical guidelines and recommendations.11 LV size and function, LA size, Doppler indices of diastolic function and RV function were assessed using standard approaches. LV ejection fraction (LVEF) was calculated by the modified Simpson's biplane method. Postoperative LVEF was assessed twice, and the average of the 2 measurements was used. Myocardial and atrial deformation were quantified using 2-dimensional speckle-tracking, including global longitudinal strain (GLS) and LA strain components, using standard speckle-tracking methods consistent with prior work in this cohort.15 All strain parameters are reported as absolute values and were analyzed using the same methodology in pre- and postoperative examinations. Detailed methods and definitions of echocardiographic measures of cardiac structure and function are provided in the Appendix E1.
Outcomes
The main end point was all-cause mortality, obtained from the Danish Civil Registration System and the National Causes of Death Registry. All end points were validated through review of medical records. Follow-up time was calculated from the date of postoperative echocardiography to the date of death or until end of follow-up (May 31, 2013).
Statistical Analyses
Absolute changes were calculated as: Δ = post-CABG value − pre-CABG value, with primary focus on changes in LVEF (ΔLVEF) and GLS (ΔGLS). For categorical comparisons, patients were stratified by ΔLVEF into 4 groups (<0%, 0%-10%, 10%-20%, and >20%). Continuous variables were presented as mean with SD or median with interquartile range (IQR), as appropriate, and categorical variables were presented as absolute counts with percentages (%). Within-patient changes from pre- to post-CABG were compared using paired t tests or Wilcoxon signed-rank tests. Trends across ordered ΔLVEF categories were assessed using linear regression (normally distributed variables), nonparametric trend tests (nonnormally distributed variables), and Cochran-Armitage tests (binary categorical variables). Associations between preoperative clinical/echocardiographic parameters and ΔLVEF/ΔGLS were examined using multivariable linear regression. Regression coefficients (β) were standardized, representing the change in outcome per 1-SD increase in each explanatory variable. In addition, multivariable logistic regression was used to assess parameters associated with clinically meaningful improvement (≥10% LVEF and ≥2% GLS), consistent with prior CABG literature.16 Three adjustment models were applied for multivariable regression analyses. Model 1: Preoperative value of the same variable (LVEF for ΔLVEF analyses and GLS for ΔGLS analyses). Model 2: Model 1 and additionally age, sex, heart rate, and follow-up time. Model 3: Model 2 and additionally major comorbidities, including history of heart failure, diabetes mellitus, and chronic kidney disease.17,18
All-cause mortality was assessed using Cox proportional-hazards regression with a separate, parsimonious adjustment due to the limited number of events: Model A was unadjusted; Model B adjusted for the preoperative value of the same variable; and Model C was additionally adjusted for age and sex. For survival analyses, ΔLVEF categories were defined as <0%, 0% to 10% and ≥10% to ensure adequate group sizes, and ΔGLS categories as <1.0%, 1.0% to 1.9%, and ≥2.0%. ΔLVEF and ΔGLS were examined both as continuous variables and across the predefined categories. The proportional hazards assumption was met for all Cox models. Kaplan-Meier curves were constructed by categories of ΔLVEF/ΔGLS, with long-rank tests for group differences.
Results were reported as standardized β, odds ratios (OR), and hazard ratios (HR) with 95% CI. All analyses were performed in R version 4.3.3 (R Foundation for Statistical Computing). Given the exploratory nature of the analyses, no formal adjustment for multiple comparisons was applied.
Results
Of 709 patients undergoing isolated CABG, the analyzed subgroup comprised those with a pre- and postoperative echocardiogram performed 30 to 365 days after surgery (n = 226) (Figure 1). Characteristics of excluded versus included patients are shown in Table E1. Median time from the preoperative echocardiogram to CABG was 14 days (IQR, 9-29 days), and from CABG to the postoperative echocardiography 115 days (IQR, 70-179 days) (Table 1). The cohort had a mean age of 67 ± 9 years and 14% were women, with a high prevalence of cardiovascular comorbidity, including hypertension (71%), and diabetes (30%) (Table 1). Most patients had 3-vessel disease, and 46% had acute myocardial infarction as the main indication for CABG. From pre- to postoperative echocardiography, LV systolic function improved, with LVEF increasing from 48.5% ± 12.8% to 53.9% ± 9.8% and GLS from 12.8% ± 4.4% to 13.6% ± 4.8% (both P values < .001) (Table 1).
Figure 1.
Recruitment and inclusion of patients undergoing coronary artery bypass grafting (CABG). Flowchart illustrating the selection process of patients undergoing isolated CABG between 2006 and 2011. Of 709 patients with preoperative echocardiography, 226 had both a pre- and postoperative echocardiogram performed 30 to 365 days after CABG and were included in the present analysis.
Table 1.
Preoperative characteristics and echocardiographic parameters before and after coronary artery bypass grafting (CABG)
| Group | Characteristic | n | All (n = 226) | P value | |
|---|---|---|---|---|---|
| Demographics | Age (y) | 226 | 67 ± 9 | ||
| Female | 226 | 31 (14) | |||
| BMI | 226 | 27.3 ± 4.2 | |||
| Time from preoperative echo to CABG (d) | 226 | 14 (9-29) | |||
| Time from CABG to postoperative echocardiogram (d) | 226 | 115 (70-179) | |||
| No. of diseased vessels | 226 | ||||
| 1 | 1 (0.4) | ||||
| 2 | 46 (20) | ||||
| 3 | 179 (79) | ||||
| AMI as indication for CABG | 226 | 104 (46) | |||
| Complications during surgery | 226 | 5 (2) | |||
| NYHA >II | 226 | 5 (2) | |||
| CCS class | 226 | ||||
| 1 | 18 (8.0) | ||||
| 2 | 179 (79) | ||||
| 3 | 28 (12) | ||||
| 4 | 1 (0.4) | ||||
| Current smoking | 226 | 179 (79) | |||
| Comorbidities | Atrial fibrillation | 226 | 8 (4) | ||
| Hypertension (%) | 226 | 161 (71) | |||
| Hypercholesterolemia (%) | 226 | 144 (64) | |||
| Diabetes mellitus (%) | 226 | 67 (30) | |||
| COPD (%) | 226 | 20 (9) | |||
| Peripheral artery disease (%) | 226 | 18 (8.0) | |||
| Chronic kidney disease (%) | 226 | 8 (4) | |||
| Peripheral artery disease | 226 | 18 (8.0%) | |||
| Prior stroke/TIA (%) | 226 | 29 (13) | |||
| Prior myocardial infarction (%) | 226 | 59 (26) | |||
| Biochemistry | Hemoglobin (g/L) | 226 | 8.4 ± 0.8 | ||
| Leukocytes (×109/L) | 226 | 8.2 (7.1-9.4) | |||
| CRP (mg/L) | 226 | 5.0 (2.0-9.0) | |||
| Creatinine (μmol/L) | 226 | 92 (81-107) | |||
| eGFR (mL/min/1.73 m2) | 226 | 73.7 ± 20.2 | |||
| LDH (U/L) | 224 | 179.3 ± 36.5 | |||
| CK-MB (μg/L) | 225 | 28 (21-38) | |||
| Troponin I (μg/L) | 153 | 6.4 (3.9-10.8) | |||
| Total cholesterol (mmol/L) | 185 | 4.6 ± 1.0 | |||
| LDL (mmol/L) | 184 | 2.6 (2.1-3.2) | |||
| HDL (mmol/L) | 185 | 1.1 (0.9-1.4) | |||
| Triglycerides (mmol/L) | 185 | 1.5 (1.-2.0) | |||
| Echocardiography | Preoperative | Postoperative | |||
| LV structure and function | |||||
| IVSd (cm) | 226 | 1.0 ± 0.2 | 1.2 ± 0.3 | <.001 | |
| LVIDd (cm) | 226 | 5.1 ± 0.7 | 5.1 ± 0.8 | .689 | |
| LVPWd (cm) | 226 | 1.0 ± 0.2 | 1.1 ± 0.4 | <.001 | |
| LVMi (g/m2) | 226 | 89.9 (74.6-112.0) | 112.1 (93.2-144.9) | <.001 | |
| LVEF (%) | 226 | 48.5 ± 12.8 | 53.9 ± 9.8 | <.001 | |
| LVEF <40% | 226 | 52 (23.0) | 20 (8.8) | <.001 | |
| GLS (%) | 201 | 12.8 ± 4.4 | 13.6 ± 4.8 | <.001 | |
| RV function | |||||
| TAPSE (cm) | 130 | 2.4 ± 0.5 | 1.6 ± 0.4 | <.001 | |
| TR max PG (mm Hg) | 117 | 23.6 (16.8-28.8) | 23.0 (17.0-27.5) | .671 | |
| LA volume and function | |||||
| LAVi (mL/m2) | 171 | 24.3 (18.5-30.1) | 18.0 (13.7-22.5) | <.001 | |
| PALS (%) | 155 | 26.3 ± 8.3 | 22.9 ± 8.9 | <.001 | |
| PACS (%) | 148 | 14.4 ± 5.3 | 13.4 ± 6.8 | .135 | |
| LACS (%) | 148 | 12.3 ± 5.0 | 9.6 ± 4.4 | <.001 | |
| Diastolic function | |||||
| MV E velocity (m/sec) | 191 | 0.7 (0.6-0.8) | 0.8 (0.6-0.9) | .003 | |
| MV A velocity (m/sec) | 191 | 0.8 (0.6-0.9) | 0.7 (0.6-0.8) | <.001 | |
| E/A (ratio) | 191 | 0.9 (0.7-1.1) | 1.1 (0.8-1.4) | <.001 | |
| E/e' mean (ratio) | 94 | 10.0 (8.1-12.2) | 8.8 (6.8-11.6) | <.001 | |
Values are presented as n (%), mean ± SD, or median (interquartile range) unless otherwise noted. Bold values indicate statistical significance (P < .05). BMI, Body mass index; AMI, acute myocardial infarction; NYHA, New York Heart Association; CCS, Canadian Cardiovascular Society (angina classification); COPD, chronic obstructive pulmonary disease; TIA, transient ischemic attack; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; LDH, lactate dehydrogenase; CK-MB, creatine kinase MB isoenzyme; LDL, low-density lipoprotein; HDL, high-density lipoprotein; LV, left ventricle; IVSd, interventricular septal thickness; LVIDd, left-ventricular end-diastolic diameter; LVPWd, LV posterior wall thickness; LVMi, left ventricular mass index; LVEF, left-ventricular ejection fraction; GLS, global longitudinal strain; RV, right ventricle; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation; PG, pressure gradient; LAVi, left-atrial volume index; PALS, peak atrial longitudinal strain; PACS, peak atrial contraction strain; LACS, left-atrial conduit strain; MV, mitral valve; MV E, mitral valve early diastolic inflow velocity; MV A, mitral valve late diastolic inflow velocity; E/A ratio, ratio of early (E) to late (A) transmitral flow velocity; E/e’, ratio of early transmitral flow velocity to early diastolic mitral annular velocity.
Preoperative Characteristics and Echocardiographic Parameters by ΔLVEF Group
Patients were stratified according to the degree of postoperative improvement in LVEF. Preoperative clinical characteristics are shown in Table 2, and echocardiographic parameters at pre- and postoperative assessment are shown in Table 3. Overall, 61 patients (27%) had a decline in LVEF (ΔLVEF <0%), 108 (48%) had a change of 0% to 10%, 35 (16%) improved by 10% to 20%, and 22 (10%) improved by >20%. The distribution of age and comorbidities was comparable across groups. Preoperative LVEF declined progressively across increasing ΔLVEF categories, from 55.0% in patients with ΔLVEF <0% to 28.4% in those with ΔLVEF >20% (Ptrend < .001) (Table 3). Preoperative GLS (absolute values) was also progressively lower across ΔLVEF groups, with the lowest values in patients with ΔLVEF >20% (14.0%, 13.6%, 10.4%, and 8.8%; Ptrend < .001). Markers of LV remodeling and filling pressure also differed across groups, with higher LV mass index, higher LA volume index, higher E/e’ (ratio of early transmitral flow velocity to early diastolic mitral annular velocity), lower tricuspid annular plane systolic excursion (TAPSE) and lower E/A ratio (ratio of early to late diastolic transmitral flow velocity) in groups with greater subsequent LVEF improvement (all P valuestrend < .05) (Table 3).
Table 2.
Preoperative characteristics by change in left ventricular ejection fraction (ΔLVEF) categories
| Group | Characteristic | ΔLVEF <0% | ΔLVEF 0%-10% | ΔLVEF 10%-20% | ΔLVEF >20% | Ptrend |
|---|---|---|---|---|---|---|
| Demographics | N | 61 | 108 | 35 | 22 | |
| Age (y) | 66 ± 10 | 68 ± 9 | 70 ± 8 | 65 ± 10 | .21 | |
| Female (%) | 12 (20) | 14 (13) | 3 (8.6) | 2 (9.1) | .11 | |
| BMI | 27.4 ± 3.5 | 27.1 ± 4.0 | 27.2 ± 5.3 | 28.2 ± 4.7 | .62 | |
| Time from preoperative echo to CABG (d) | 14 (10-27) | 14 (9-30) | 13 (8-23) | 17 (12-46) | .36 | |
| Time from CABG to postoperative echo (d) | 109 (65-183) | 107 (65-168) | 157 (98-195) | 118 (91-203) | .34 | |
| No. of diseased vessels | .34 | |||||
| 1 | 0 (0) | 1 (0.9) | 0 (0) | 0 (0) | ||
| 2 | 17 (28) | 23 (21) | 4 (11) | 2 (9.1) | ||
| 3 | 44 (72) | 84 (78) | 31 (89) | 20 (91) | ||
| AMI as indication for CABG | 24 (39) | 49 (45) | 20 (57) | 11 (50) | .15 | |
| NYHA > II | 1 (1.6) | 1 (0.9) | 1 (2.9) | 2 (9.1) | .07 | |
| CCS class | .19 | |||||
| 1 | 6 (9.8) | 9 (8.3) | 0 (0) | |||
| 2 | 43 (70) | 88 (81) | 33 (94) | |||
| 3 | 12 (20) | 10 (9.3) | 2 (5.7) | |||
| 4 | 0 (0) | 1 (0.9) | 0 (0) | |||
| Current smoking | 53 (87) | 83 (77) | 27 (77) | 16 (73) | .13 | |
| Comorbidities | Atrial fibrillation | 0 (0) | 4 (4) | 3 (9) | 1 (5) | .08 |
| Hypertension | 47 (77) | 74 (69) | 27 (77) | 13 (59) | .27 | |
| Hypercholesterolemia | 40 (66) | 68 (63) | 18 (51) | 18 (82) | .70 | |
| Diabetes mellitus | 20 (33) | 27 (25) | 10 (29) | 10 (45) | .45 | |
| COPD | 3 (5) | 11 (10) | 2 (6) | 4 (18) | .16 | |
| Peripheral artery disease | 3 (4.9) | 11 (10) | 2 (6) | 2 (9.1) | .67 | |
| Chronic kidney disease | 3 (5) | 4 (4) | 1 (3) | 0 (0) | .29 | |
| Prior stroke/TIA | 10 (16) | 15 (14) | 3 (9) | 1 (5) | .11 | |
| Prior myocardial infarction | 15 (25) | 30 (28) | 5 (14) | 9 (41) | .58 | |
| Biochemistry | Hemoglobin (g/L) | 8.4 ± 0.8 | 8.5 ± 0.7 | 8.4 ± 0.9 | 8.6 ± 0.9 | .42 |
| Leukocytes (×109/L) | 8.4 (7.2-9.6) | 8.0 (7.0-9.3) | 7.9 (6.8-9.6) | 8.6 (8.0-9.5) | .44 | |
| CRP (mg/L) | 2.0 (2.0-7.0) | 6.0 (2.0-10.0) | 6.0 (2.0-12.0) | 6.0 (2.0-18.0) | .07 | |
| Creatinine (μmol/L) | 89 (76-103) | 90 (81-105) | 102 (84-114) | 96 (83-115) | .26 | |
| eGFR (mL/min/1.73 m2) | 76.6 ± 20.5 | 74.3 ± 19.8 | 69.5 ± 23.0 | 69.7 ± 16.3 | .47 | |
| LDH (U/L) | 179.4 ± 28.3 | 176.2 ± 29.8 | 189.4 ± 57.1 | 178.3 ± 44.7 | .58 | |
| CK-MB (μg/L) | 30 (22-41) | 27 (21-39) | 26 (17-37) | 25 (21-31) | .37 | |
| Troponin I (μg/L) | 6.1 (3.9-11.3) | 6.9 (4.2-11.1) | 6.9 (4.4-10.9) | 3.0 (1.9-5.1) | .17 | |
| Total cholesterol (mmol/L) | 4.6 ± 1.1 | 4.5 ± 1.0 | 4.6 ± 1.1 | 4.7 ± 0.9 | .58 | |
| LDL (mmol/L) | 2.5 (1.9-3.2) | 2.6 (2.0-3.2) | 3.0 (2.3-3.4) | 2.6 (2.2-3.6) | .34 | |
| HDL (mmol/L) | 1.3 (1.0-1.5) | 1.1 (0.9-1.4) | 1.1 (0.9-1.4) | 1.1 (1.0-1.5) | .023 | |
| Triglycerides (mmol/L) | 1.6 (1.-2.0) | 1.4 (1.1-2.0) | 1.2 (1.1-1.7) | 1.5 (1.0-1.9) | .77 |
Values are presented as n (%), mean ± SD, or median (interquartile range) unless otherwise noted. Bold values indicate statistical significance (P < .05). BMI, Body mass index; CABG, coronary artery bypass grafting; AMI, acute myocardial infarction; NYHA, New York Heart Association; CCS, Canadian Cardiovascular Society (angina classification); COPD, chronic obstructive pulmonary disease; TIA, transient ischemic attack; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; LDH, lactate dehydrogenase; CK-MB, creatine kinase MB isoenzyme; LDL, low-density lipoprotein; HDL, high-density lipoprotein; LV, left ventricle; IVSd, interventricular septal thickness; LVIDd, left-ventricular end-diastolic diameter; LVPWd, LV posterior wall thickness; LVMi, left ventricular mass index; GLS, global longitudinal strain; RV, right ventricle; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation; PG, pressure gradient; LAVi, left-atrial volume index; PALS, peak atrial longitudinal strain; PACS, peak atrial contraction strain; LACS, left-atrial conduit strain; MV, mitral valve; MV E, mitral valve early diastolic inflow velocity; MV A, mitral valve late diastolic inflow velocity; E/A ratio, ratio of early (E) to late (A) transmitral flow velocity; E/e’, ratio of early transmitral flow velocity to early diastolic mitral annular velocity.
Table 3.
Echocardiographic parameters at pre- and postoperative assessment according to change in left ventricular ejection fraction (ΔLVEF) categories
| Category | Variable | ΔLVEF <0% | ΔLVEF 0%-10% | ΔLVEF 10%-20% | ΔLVEF >20% | Ptrend |
|---|---|---|---|---|---|---|
| Preoperative | ||||||
| LV structure and function | IVSd (cm) | 1.0 ± 0.2 | 1.0 ± 0.2 | 1.0 ± 0.2 | 1.0 ± 0.2 | .83 |
| LVIDd (cm) | 4.90 ± 0.59 | 5.02 ± 0.69 | 5.27 ± 0.74 | 5.52 ± 0.93 | <.001 | |
| LVPWd (cm) | 1.0 ± 0.2 | 1.0 ± 0.2 | 1.0 ± 0.2 | 1.1 ± 0.2 | .10 | |
| LVMi (g/m2) | 87.3 (73.1-103.4) | 90.7 (76.2-107.3) | 90.1 (80.1-112.9) | 117.1 (93.0-137.4) | .040 | |
| LVEF (%) | 55.0 ± 8.5 | 51.3 ± 10.7 | 40.9 ± 10.0 | 28.4 ± 10.2 | <.001 | |
| LVEF <40% | 4 (6.6) | 14 (13.0) | 16 (45.7) | 18 (81.8) | <.001 | |
| GLS (%) | 14.0 ± 4.1 | 13.6 ± 4.3 | 10.4 ± 3.6 | 8.8 ± 3.3 | <.001 | |
| RV function | TAPSE (cm) | 2.5 ± 0.4 | 2.5 ± 0.4 | 2.2 ± 0.4 | 2.1 ± 0.5 | <.001 |
| RV strain (%) | −19.55 ± 5.07 | −17.80 ± 4.47 | −18.55 ± 3.61 | −16.73 ± 2.92 | .13 | |
| TR max PG (mm Hg) | 23.6 (16.8-28.8) | 21.8 (14.8-28.1) | 23.8 (17.4-27.6) | 25.2 (18.4-31.4) | .08 | |
| LA volume and function | LAVi (mL/m2) | 24.2 (19.5-28.6) | 22.3 (17.3-28.1) | 25.5 (20.5-30.5) | 31.4 (25.5-37.0) | .034 |
| PALS (%) | 28.2 ± 8.1 | 26.7 ± 7.8 | 23.8 ± 8.6 | 23.3 ± 10.5 | .007 | |
| PACS (%) | 14.7 ± 5.1 | 14.7 ± 5.2 | 14.9 ± 6.0 | 12.1 ± 5.6 | .27 | |
| LACS (%) | 13.5 ± 5.4 | 12.4 ± 4.8 | 10.6 ± 3.3 | 11.6 ± 6.6 | .026 | |
| Diastolic function | MV E velocity (m/sec) | 0.7 (0.6-0.8) | 0.7 (0.6-0.8) | 0.8 (0.6-0.9) | 0.7 (0.6-0.9) | .42 |
| MV A velocity (m/sec) | 0.8 (0.7-0.9) | 0.8 (0.7-0.9) | 0.7 (0.6-0.9) | 0.7 (0.5-0.9) | .17 | |
| E/A ratio | 0.9 (0.7-1.1) | 0.9 (0.7-1.1) | 1.0 (0.8-1.4) | 0.9 (0.7-1.3) | .037 | |
| E/e' mean | 9.8 (8.3-11.1) | 9.6 (7.5-11.7) | 12.1 (9.1-15.0) | 11.2 (9.2-14.9) | <.001 | |
| Postoperative | ||||||
| LV structure and function | IVSd (cm) | 1.2 ± 0.3 | 1.2 ± 0.3 | 1.2 ± 0.3 | 1.3 ± 0.3 | .79 |
| LVIDd (cm) | 5.0 ± 0.7 | 5.0 ± 0.8 | 5.1 ± 0.9 | 5.5 ± 0.8 | .06 | |
| LVPWd (cm) | 1.0 ± 0.2 | 1.1 ± 0.3 | 1.2 ± 0.7 | 1.1 ± 0.3 | .20 | |
| LVMi (g/m2) | 115.4 (95.8-139.3) | 110.3 (93.0-136.9) | 115.9 (102.9-141.8) | 146.5 (98.0-165.1) | .045 | |
| LVEF (%) | 50.5 ± 9.5 | 55.0 ± 9.6 | 55.3 ± 10.6 | 55.6 ± 9.2 | .013 | |
| LVEF <40% | 9 (15) | 10 (9) | 1 (3) | 0 (0) | .012 | |
| GLS | 13.3 ± 5.7 | 14.2 ± 4.0 | 12.7 ± 6.4 | 12.9 ± 3.7 | .54 | |
| RV function | TAPSE (cm) | 1.53 ± 0.45 | 1.61 ± 0.38 | 1.63 ± 0.40 | 1.58 ± 0.49 | .563 |
| TR max PG (mm Hg) | 20.0 (16.0-24.0) | 24.0 (18.0-29.0) | 23.0 (17.0-27.0) | 24.0 (21.0-26.2) | .23 | |
| LA volume and function | LAVi (mL/m2) | 18.8 (15.3-24.7) | 16.9 (12.8-20.9) | 20.8 (14.3-26.4) | 19.6 (17.5-22.0) | .72 |
| PALS | 22.7 ± 8.8 | 23.7 ± 7.8 | 21.2 ± 8.2 | 24.9 ± 14.0 | .85 | |
| PACS | 13.4 ± 7.4 | 13.7 ± 6.3 | 12.1 ± 6.0 | 14.5 ± 8.2 | .93 | |
| LACS | 9.3 ± 4.5 | 10.2 ± 4.6 | 9.1 ± 4.1 | 10.5 ± 7.0 | .64 | |
| Diastolic function | MV E velocity (m/sec) | 0.8 (0.7-0.9) | 0.8 (0.6-0.9) | 0.8 (0.6-0.9) | 0.6 (0.5-0.8) | .43 |
| MV A velocity (m/sec) | 0.7 (0.6-0.8) | 0.7 (0.5-0.8) | 0.7 (0.5-0.8) | 0.8 (0.7-0.8) | .40 | |
| E/A (ratio) | 1.1 (0.9-1.4) | 1.1 (0.9-1.4) | 1.1 (0.8-1.8) | 0.9 (0.6-1.1) | .43 | |
| E/e' mean (ratio) | 8.4 (6.8-11.5) | 8.9 (6.6-10.2) | 8.6 (7.5-14.9) | 9.6 (7.4-11.6) | .33 | |
Values are presented as n (%), mean ± SD, or median (interquartile range) unless otherwise noted. Bold values indicate statistical significance (P < .05). BMI, Body mass index; CABG, coronary artery bypass grafting; AMI, acute myocardial infarction; NYHA, New York Heart Association; CCS, Canadian Cardiovascular Society (angina classification); COPD, chronic obstructive pulmonary disease; TIA, transient ischemic attack; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; LDH, lactate dehydrogenase; CK-MB, creatine kinase MB isoenzyme; LDL, low-density lipoprotein; HDL, high-density lipoprotein; LV, left ventricle; IVSd, interventricular septal thickness; LVIDd, left-ventricular end-diastolic diameter; LVPWd, LV posterior wall thickness; LVMi, left ventricular mass index; GLS, global longitudinal strain; RV, right ventricle; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation; PG, pressure gradient; LAVi, left-atrial volume index; PALS, peak atrial longitudinal strain; PACS, peak atrial contraction strain; LACS, left-atrial conduit strain; MV, mitral valve; MV E, mitral valve early diastolic inflow velocity; MV A, mitral valve late diastolic inflow velocity; E/A ratio, ratio of early (E) to late (A) transmitral flow velocity; E/e’, ratio of early transmitral flow velocity to early diastolic mitral annular velocity.
Predictors of Improvement in LV Systolic Function
Parameters associated with postoperative changes in LVEF and GLS are summarized in Table 4. In multivariable linear regression, lower preoperative LVEF was the strongest predictor of greater postoperative improvement in LVEF (β = −0.65 [95% CI, −0.75 to −0.55]; P < .001). Larger LV internal dimension at end-diastole (LVIDd) was associated with less LVEF recovery (β = −0.15 [95% CI, −0.27 to −0.03]; P = .014), whereas clinical variables showed no consistent associations. In multivariable logistic regression analyses evaluating marked LVEF improvement ≥10%, lower preoperative LVEF remained strongly associated with higher odds of improvement (Model 3, OR, 0.89 [95% CI, 0.85-0.93]; P < .001 (Table E2), with additional associations observed for longer time from CABG to postoperative echocardiography, smaller LVIDd, and higher preoperative E/e’ ratio (Table E2). Conversely, in logistic regression analyses examining lack of substantial LVEF improvement (<10%), higher baseline LVEF, larger LVIDd, increased LV mass index, and higher E/e’ ratio were associated with increased odds of limited functional recovery (Table E3).
Table 4.
Parameters associated with improvement in left ventricular systolic function (change in left ventricular ejection fraction [ΔLVEF and change in global longitudinal strain [ΔGLS)
| Characteristic | ΔLVEF |
ΔGLS |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Model 1 |
Model 2 |
Model 3 |
Model 1 |
Model 2 |
Model 3 |
|||||||
| β (95% CI) | P value | β (95% CI) | P value | β (95% CI) | P value | β (95% CI) | P value | β (95% CI) | P value | β (95% CI) | P value | |
| Age (y) | 0.06 (−0.04 to 0.16) | .27 | 0.05 (−0.05 to 0.16) | .33 | 0.04 (−0.07 to 0.15) | .45 | 0.04 (−0.07 to 0.16) | .47 | 0.03 (−0.10 to 0.16) | .64 | 0.03 (−0.10 to 0.15) | .67 |
| Male sex | 0.01 (−0.09 to 0.11) | .88 | 0.01 (−0.09 to 0.11) | .80 | 0.02 (−0.09 to 0.12) | .77 | −0.06 (−0.18 to 0.07) | .37 | −0.05 (−0.18 to 0.08) | .45 | −0.05 (−0.18 to 0.07) | .39 |
| BMI | 0.01 (−0.09 to 0.11) | .86 | 0.05 (−0.06 to 0.16) | .38 | 0.06 (−0.06 to 0.17) | .32 | 0.05 (−0.07 to 0.17) | .44 | 0.05 (−0.08 to 0.19) | .44 | 0.04 (−0.09 to 0.18) | .51 |
| Heart rate (bpm) | −0.04 (−0.15 to 0.07) | .44 | −0.01 (−0.12 to 0.10) | .83 | −0.01 (−0.13 to 0.10) | .82 | −0.03 (−0.15 to 0.10) | .66 | −0.02 (−0.16 to 0.12) | .78 | −0.02 (−0.16 to 0.11) | .71 |
| Anemia | −0.04 (−0.14 to 0.06) | .43 | −0.06 (−0.16 to 0.05) | .28 | −0.05 (−0.15 to 0.06) | .38 | 0.06 (−0.05 to 0.18) | .28 | 0.06 (−0.06 to 0.18) | .34 | 0.07 (−0.05 to 0.19) | .27 |
| Diabetes | −0.06 (−0.16 to 0.04) | .24 | −0.05 (−0.16 to 0.05) | .32 | −0.05 (−0.15 to 0.06) | .36 | −0.02 (−0.14 to 0.10) | .75 | −0.02 (−0.15 to 0.11) | .77 | −0.02 (−0.15 to 0.10) | .73 |
| Hypertension | −0.08 (−0.18 to 0.02) | .10 | −0.09 (−0.19 to 0.02) | .10 | −0.08 (−0.19 to 0.03) | .15 | −0.08 (−0.20 to 0.03) | .16 | −0.11 (−0.23 to 0.01) | .080 | −0.10 (−0.23 to 0.02) | .11 |
| Chronic kidney disease | −0.08 (−0.18 to 0.02) | .12 | −0.08 (−0.18 to 0.03) | .15 | −0.07 (−0.17 to 0.03) | .18 | −0.13 (−0.25 to −0.02) | .022 | −0.13 (−0.25 to −0.01) | .031 | −0.12 (−0.24 to −0.01) | .040 |
| History of heart failure | −0.04 (−0.15 to 0.06) | .44 | −0.04 (−0.15 to 0.07) | .51 | −0.03 (−0.14 to 0.07) | .53 | −0.15 (−0.27 to −0.03) | .018 | −0.15 (−0.27 to −0.02) | .020 | −0.14 (−0.26 to −0.02) | .024 |
| AMI as indication for CABG | 0.05 (−0.05 to 0.15) | .31 | 0.07 (−0.03 to 0.18) | .17 | 0.06 (−0.04 to 0.17) | .24 | 0.13 (0.01 to 0.25) | .035 | 0.13 (0.00 to 0.26) | .043 | 0.12 (0.00 to 0.25) | .054 |
| Previous AMI | 0.00 (−0.10 to 0.10) | .94 | 0.01 (−0.09 to 0.11) | .88 | 0.02 (−0.08 to 0.12) | .71 | −0.06 (−0.18 to 0.06) | .36 | −0.06 (−0.18 to 0.07) | .37 | −0.04 (−0.17 to 0.08) | .51 |
| Peripheral artery disease | 0.01 (−0.09 to 0.12) | .79 | 0.01 (−0.10 to 0.11) | .91 | 0.00 (−0.11 to 0.11) | .99 | −0.02 (−0.14 to 0.10) | .74 | −0.02 (−0.16 to 0.11) | .72 | −0.04 (−0.17 to 0.09) | .52 |
| No. of diseased vessels | 0.03 (−0.07 to 0.14) | .53 | 0.03 (−0.08 to 0.13) | .59 | 0.03 (−0.07 to 0.14) | .54 | −0.04 (−0.16 to 0.08) | .47 | −0.06 (−0.19 to 0.06) | .32 | −0.06 (−0.19 to 0.06) | .33 |
| Time from CABG to postoperative echocardiography (y) | 0.01 (−0.10 to 0.11) | .92 | 0.01 (−0.09 to 0.11) | .82 | 0.01 (−0.10 to 0.11) | .89 | −0.09 (−0.21 to 0.03) | .16 | −0.09 (−0.22 to 0.04) | .17 | −0.08 (−0.20 to 0.05) | .23 |
| IVSd (cm) | −0.01 (−0.10 to 0.09) | .91 | −0.01 (−0.11 to 0.10) | .91 | 0.01 (−0.10 to 0.11) | .90 | −0.02 (−0.15 to 0.10) | .68 | −0.03 (−0.16 to 0.09) | .62 | −0.01 (−0.14 to 0.11) | .84 |
| LVIDd (cm) | −0.13 (−0.24 to −0.01) | .028 | −0.14 (−0.25 to −0.02) | .022 | −0.15 (−0.27 to −0.03) | .014 | −0.02 (−0.15 to 0.10) | .71 | −0.03 (−0.17 to 0.12) | .73 | −0.04 (−0.19 to 0.10) | .55 |
| LVPWd (cm) | 0.00 (−0.10 to 0.10) | .96 | 0.00 (−0.11 to 0.10) | >.99 | 0.02 (−0.09 to 0.12) | .78 | −0.06 (−0.18 to 0.06) | .30 | −0.06 (−0.19 to 0.06) | .33 | −0.04 (−0.17 to 0.08) | .49 |
| LVMi (g/m2) | −0.10 (−0.20 to 0.01) | .074 | −0.10 (−0.21 to 0.01) | .077 | −0.10 (−0.21 to 0.01) | .076 | −0.06 (−0.19 to 0.06) | .30 | −0.06 (−0.19 to 0.07) | .34 | −0.07 (−0.20 to 0.06) | .27 |
| LVEF (%) | −0.65 (−0.75 to −0.55) | <.001 | −0.65 (−0.75 to −0.55) | <.001 | −0.65 (−0.75 to −0.55) | <.001 | 0.07 (−0.10 to 0.24) | .44 | 0.05 (−0.13 to 0.22) | .61 | 0.03 (−0.15 to 0.20) | .76 |
| GLS (%) | 0.12 (−0.02 to 0.26) | .092 | 0.12 (−0.03 to 0.27) | .11 | 0.12 (−0.03 to 0.27) | .12 | −0.51 (−0.63 to −0.39) | <.001 | −0.51 (−0.63 to −0.39) | <.001 | −0.51 (−0.63 to −0.39) | <.001 |
| TAPSE (cm) | 0.02 (−0.09 to 0.14) | .68 | 0.01 (−0.11 to 0.13) | .88 | 0.00 (−0.12 to 0.12) | .97 | 0.11 (−0.02 to 0.25) | .11 | 0.11 (−0.04 to 0.25) | .15 | 0.07 (−0.07 to 0.22) | .31 |
| TR max PG (mm Hg) | 0.00 (−0.11 to 0.12) | .94 | 0.00 (−0.12 to 0.13) | .96 | 0.00 (−0.12 to 0.13) | .94 | −0.06 (−0.20 to 0.09) | .43 | −0.06 (−0.21 to 0.09) | .43 | −0.06 (−0.21 to 0.09) | .44 |
| LAVi (mL/m2) | 0.08 (−0.02 to 0.19) | .12 | 0.08 (−0.03 to 0.19) | .16 | 0.08 (−0.03 to 0.19) | .16 | 0.11 (−0.02 to 0.25) | .10 | 0.13 (−0.02 to 0.27) | .079 | 0.11 (−0.03 to 0.26) | .12 |
| PALS (%) | −0.04 (−0.15 to 0.07) | .48 | −0.04 (−0.15 to 0.07) | .48 | −0.07 (−0.18 to 0.05) | .27 | 0.11 (−0.02 to 0.24) | .11 | 0.11 (−0.03 to 0.25) | .12 | 0.13 (−0.02 to 0.28) | .10 |
| PACS (%) | −0.02 (−0.13 to 0.09) | .69 | −0.01 (−0.12 to 0.10) | .81 | −0.04 (−0.15 to 0.08) | .51 | 0.14 (0.01 to 0.26) | .030 | 0.14 (0.02 to 0.27) | .027 | 0.14 (0.01 to 0.28) | .038 |
| LACS (%) | −0.03 (−0.14 to 0.09) | .65 | −0.04 (−0.15 to 0.08) | .53 | −0.05 (−0.16 to 0.07) | .44 | 0.00 (−0.14 to 0.14) | .99 | −0.01 (−0.16 to 0.14) | .87 | 0.00 (−0.15 to 0.16) | >.99 |
| MV E velocity (m/sec) | 0.01 (−0.10 to 0.11) | .87 | 0.02 (−0.09 to 0.13) | .69 | 0.04 (−0.07 to 0.15) | .50 | −0.05 (−0.17 to 0.07) | .40 | −0.06 (−0.19 to 0.07) | .35 | −0.03 (−0.16 to 0.10) | .66 |
| MV A velocity (m/sec) | 0.01 (−0.09 to 0.12) | .79 | 0.02 (−0.09 to 0.13) | .74 | 0.02 (−0.09 to 0.13) | .69 | 0.05 (−0.07 to 0.17) | .43 | 0.03 (−0.11 to 0.17) | .67 | 0.02 (−0.12 to 0.15) | .78 |
| E/A ratio | 0.00 (−0.11 to 0.10) | .94 | 0.00 (−0.11 to 0.11) | .95 | 0.02 (−0.10 to 0.14) | .74 | −0.10 (−0.23 to 0.02) | .11 | −0.10 (−0.24 to 0.04) | .15 | −0.05 (−0.20 to 0.09) | .45 |
| E/e' mean ratio | 0.05 (−0.06 to 0.16) | .35 | 0.07 (−0.05 to 0.19) | .23 | 0.10 (−0.03 to 0.22) | .13 | −0.10 (−0.23 to 0.04) | .15 | −0.12 (−0.26 to 0.02) | .10 | −0.08 (−0.23 to 0.07) | .28 |
Values are presented as n (%), mean ± SD, or median (interquartile range) unless otherwise noted. Bold values indicate statistical significance (P < .05). BMI, Body mass index; CABG, coronary artery bypass grafting; AMI, acute myocardial infarction; NYHA, New York Heart Association; CCS, Canadian Cardiovascular Society (angina classification); COPD, chronic obstructive pulmonary disease; TIA, transient ischemic attack; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; LDH, lactate dehydrogenase; CK-MB, creatine kinase MB isoenzyme; LDL, low-density lipoprotein; HDL, high-density lipoprotein; LV, left ventricle; IVSd, interventricular septal thickness; LVIDd, left-ventricular end-diastolic diameter; LVPWd, LV posterior wall thickness; LVMi, left ventricular mass index; LVEF, left-ventricular ejection fraction; GLS, global longitudinal strain; RV, right ventricle; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation; PG, pressure gradient; LAVi, left-atrial volume index; PALS, peak atrial longitudinal strain; PACS, peak atrial contraction strain; LACS, left-atrial conduit strain; MV, mitral valve; MV E, mitral valve early diastolic inflow velocity; MV A, mitral valve late diastolic inflow velocity; E/A ratio, ratio of early (E) to late (A) transmitral flow velocity; E/e’, ratio of early transmitral flow velocity to early diastolic mitral annular velocity.
For GLS, lower preoperative GLS was the dominant determinant of postoperative improvement in multivariable linear regression (β = −0.51 [95% CI, −0.63 to −0.39]; P < .001). Chronic kidney disease and a history of heart failure were associated with less GLS recovery, whereas higher preoperative peak atrial contraction strain (PACS) were associated with greater improvement (Table 4). In logistic analyses of marked GLS improvement (ΔGLS ≥ 2%), lower preoperative GLS (OR, 0.80 [95% CI, 0.71-0.88]; P < .001) and both TAPSE and PACS remained independently associated with marked GLS recovery. Indices of less adverse diastolic filling [mitral valve (MV) early diastolic inflow velocity (E) and E/A ratio] were also related to greater GLS improvement (Table E4). In complementary analyses evaluating minimal GLS improvement (<2%), older age, worse baseline GLS, and lower TAPSE were associated with reduced likelihood of strain recovery (Table E5).
Association Between Changes in LVEF From Pre- to Post-CABG and All-cause Mortality
A total of 23 patients (9.3%) died during a median follow-up of 3.2 years (IQR, 2.5-4.1 years) from the postoperative echocardiogram. In the overall CABG cohort of 709 patients, 86 patients died during follow-up (11.9%), including 6 deaths within 30 days after surgery (0.83%), compared with 23 patients during the total follow-up among the patients in this analysis with pre-/post-CABG echocardiograms. When stratified by ΔLVEF, all-cause mortality incidence rates were 2.47 (95% CI, 0.80-5.76), 4.61 (95% CI, 2.63-7.48), and 0.93 (95% CI, 0.11-3.37) per 100 person-years across the <0%, 0% to 10% and ≥10% categories, respectively. In Cox regression (Model C), patients with ≥10% improvement in LVEF had a lower risk of all-cause mortality compared with those with ΔLVEF <0% (HR, 0.07 [95% CI, 0.01-0.42]; P = .004), whereas the 0% to 10% group did not differ from the <0% group (HR, 0.99 [95% CI, 0.34-2.86]; P = .99). Analyzed continuously, each 1% absolute increase in ΔLVEF was associated with lower mortality risk (Model C HR, 0.88 [95% CI, 0.84-0.93]; P < .001).
Across ΔGLS categories, incidence rates were 4.83 (95% CI, 2.50-8.44), 5.51 (95% CI, 2.22-11.36), and 0.99 (95% CI, 0.20-2.89) per 100 person-years for <1.0%, 1.0% to 1.9%, and ≥2.0% improvement, respectively. In Cox regression (Model C), each 1% absolute improvement in ΔGLS was associated with lower mortality risk (HR, 0.62 [95% CI, 0.51-0.76]; P < .001) (Table 5). Categorically, ≥2.0% GLS improvement was associated with substantially lower mortality risk compared with <1.0% (Model C HR, 0.10 [95% CI, 0.03-0.38]; P < .001), whereas 1.0% to 1.9% improvement was not (HR, 0.66 [95% CI, 0.24-1.80]; P = .41). Results were similar in a sensitivity analysis restricted to patients with preoperative LVEF <40% (Table E6). Kaplan-Meier analyses showed significant differences in survival across both ΔLVEF and ΔGLS categories (log-rank P = .047 and P = .007) (Figure 2).
Table 5.
Association between changes in systolic function (change in left ventricular ejection fraction [ΔLVEF] and change in global longitudinal strain [ΔGLS]) and all-cause mortality after postoperative echocardiography
| ΔLVEF | ||||||
|---|---|---|---|---|---|---|
| Model | ΔLVEF (per 1%) |
0%-10% vs <0% |
≥10% vs <0% |
|||
| Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P-value | Hazard ratio (95% CI) | P value | |
| Model A | 0.96 (0.92-1.01) | .090 | 1.87 (0.68-5.11) | .222 | 0.38 (0.07-1.97) | .250 |
| Model B | 0.90 (0.86-0.94) | <.001 | 1.15 (0.40-3.28) | .794 | 0.09 (0.01-0.53) | .008 |
| Model C | 0.88 (0.84-0.93) | <.001 | 0.99 (0.34-2.86) | .985 | 0.07 (0.01-0.42) | .004 |
| ΔGLS∗ | ||||||
|---|---|---|---|---|---|---|
| Model | ΔGLS (per 1%) | 1.0%-1.9% vs <1.0% | ≥2.0% vs <1.0% | |||
| Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P-value | Hazard ratio (95% CI) | P-value | |
| Model A | 0.83 (0.74-0.93) | .002 | 1.13 (0.45-2.89) | .791 | 0.20 (0.06-0.71) | .013 |
| Model B | 0.75 (0.66-0.85) | <.001 | 0.70 (0.26-1.91) | .489 | 0.11 (0.03-0.42) | .011 |
| Model C | 0.62 (0.51-0.76) | <.001 | 0.66 (0.24-1.80) | .412 | 0.10 (0.03-0.38) | <.001 |
Bold values indicate statistical significance (P < .05).
HR, Hazard ratio; CI, confidence intervals; LVEF, left ventricular ejection fraction; GLS, global longitudinal strain.
Values are presented as hazard ratios (HR) with 95% confidence intervals (CI).
Figure 2.
Survival by postoperative improvement in systolic function (change in left ventricular ejection fraction [ΔLVEF] and change in global longitudinal strain [ΔGLS]) Kaplan-Meier curves illustrating cumulative all-cause survival by categories of ΔLVEF (<0%, 0%-10%, and ≥10%) (A) and cumulative all-cause survival by categories of ΔGLS (<1.0%, 1.0%-1.9%, and ≥2.0%) (B) after CABG with 95% CI.
Discussion
In this CABG cohort, postoperative LV systolic function improved modestly but consistently from the preoperative to the postoperative echocardiographic assessment, reflected by an average increase of 6% in LVEF, and 1% in GLS. The magnitude of improvement varied substantially across patients and was greatest among those with the most impaired preoperative LV systolic function, consistent with a greater potential for functional improvement in this group. Importantly, postoperative improvements in both LVEF and GLS were associated with more favorable long-term outcomes, underscoring the clinical relevance of functional recovery after CABG.
Parameters Associated With LV Postoperative Recovery
In this cohort, lower preoperative LVEF showed the most consistent association with postoperative improvement, indicating that patients with greater preoperative systolic impairment achieved the largest absolute gains in LVEF after CABG. This aligns with findings from the Surgical Treatment for Ischemic Heart Failure (STITCH) trial and other revascularization cohorts, where lower preoperative LVEF was associated with greater postoperative improvement.14,16,19 However, unlike the STITCH trial, which enrolled patients with advanced ischemic cardiomyopathy under controlled trial conditions, our study reflects an unselected CABG population across a broad range of preoperative LV function. By evaluating postoperative recovery within a defined early-to-intermediate postoperative window, our study extends these observations across the full spectrum of LV dysfunction, including patients with mild-to-moderate impairment who are often underrepresented in prior studies. This study also broadens prior LVEF-centric assessments by incorporating GLS, a measure of myocardial deformation, as a complementary measure of systolic recovery. Notably, postoperative improvement in GLS was also greatest among patients with the most impaired preoperative GLS. Taken together, the pattern may partly reflect regression to the mean but may also be compatible with recovery of viable but dysfunctional myocardium after revascularization.
Beyond preoperative systolic impairment, structural remodeling and comorbidity burden appeared to influence postoperative LV systolic recovery. Greater LV dilatation was associated with less improvement in LVEF, suggesting that more advanced LV dilatation may reflect more chronic or irreversible remodeling with limited capacity for recovery after revascularization.20,21 In contrast, smaller improvements in GLS among patients with chronic kidney disease and a history of heart failure may be consistent with more advanced, diffuse, and less reversible myocardial disease, limiting recovery of longitudinal function.22,23 Measures reflecting atrial (eg, PACS) and diastolic function (MV E and E/A ratio) showed more modest associations with GLS recovery, suggesting that loading conditions and atrioventricular coupling may modulate recovery rather than serving as primary drivers.24,25 Higher preoperative TAPSE was also associated with marked GLS improvement, consistent with preserved RV systolic function accompanying a greater capacity for LV longitudinal recovery.26 Notably, changes in LVEF and GLS were not fully concordant, reinforcing that these indices capture distinct aspects of LV recovery and should be regarded as complementary in the postoperative assessment.27,28
Associations With All-cause Mortality
Postoperative improvements in both LVEF and GLS were associated with better long-term survival, with similar findings in the sensitivity analysis, supporting that functional recovery after CABG may carry clinically relevant information beyond preoperative status. The association with outcomes was most apparent among patients with more pronounced improvement, suggesting that meaningful reverse remodeling may be required to translate into better clinical outcomes.16,19 These observations are consistent with prior studies, including the STICH trial, demonstrating an association between postoperative improvement in LV systolic function and long-term outcomes after revascularization.16,19 In addition, several observational CABG cohorts have shown that GLS is associated with mortality, including in patients with preserved LVEF, supporting its role as a complementary marker of myocardial recovery.8,29
Pathophysiological Considerations
Variation in postoperative LV recovery may reflect differences in myocardial viability. Chronically under perfused but viable (so-called hibernating) myocardium may regain contractility after revascularization, which could contribute to greater improvement in patients with more impaired preoperative function.30,31 However, prior analyses have not consistently shown myocardial viability testing to predict subsequent improvement in LVEF after CABG, underscoring that recovery may be multifactorial.16,31 Conversely, advanced structural change and diffuse myocardial damage may limit reverse remodeling.31,32 Consistent with this, markers of more advanced remodeling and comorbidity burden in our study were associated with less systolic recovery. Measures of LA, RV, and diastolic function were also related to recovery, although associations differed between LVEF and GLS, and should be interpreted cautiously given loading dependence and potential residual confounding.19,25 Moreover, transient myocardial stunning in the peri-infarct and perioperative period could lower preoperative systolic measures and contribute to apparent recovery at follow-up, particularly among patients undergoing CABG in the setting of acute myocardial infarction.33 In addition, regression to the mean and a ceiling effect in patients with relatively preserved preoperative systolic function may partly contribute to the observed recovery patterns.34 Consistent with this, the proportion of patients with preoperative LVEF <40% were in the highest ΔLVEF category (Table 3). Overall, recovery appears influenced by preoperative function, structural remodeling, and loading conditions.
Clinical Implications
The observed associations between postoperative changes in LVEF and GLS and long-term survival suggest that echocardiographic assessment after CABG may provide clinically relevant information on recovery. Routine follow-up imaging may help characterize recovery trajectories, particularly in patients with preoperative LV dysfunction, and may support follow-up planning. Patients with no improvement in LV function (ΔLVEF <0%) or minimal strain recovery (ΔGLS <1.0%) had higher observed mortality, however the present observational design and postoperative echocardiography requirement do not allow identification of ventricles too sick for CABG. Overall, these findings indicate that longitudinal functional assessment, including strain, may be a useful component of postoperative evaluation after CABG.
Strengths and Limitations
This study has several strengths. It represents a well-characterized, real-world cohort of consecutively included patients undergoing isolated CABG, with paired pre- and postoperative echocardiography enabling detailed assessment of postoperative cardiac recovery. The inclusion of GLS and LA strain alongside conventional echocardiographic indices provided a comprehensive evaluation of systolic recovery, and remodeling.
However, certain limitations should be acknowledged. The retrospective, observational design precludes causal inference and renders the findings hypothesis-generating. Follow-up echocardiography was performed as part of routine clinical care rather than according to a standardized protocol and restricting the cohort to patients with postoperative echocardiography within 30 to 365 days may have introduced selection and survivorship bias. However, preoperative clinical characteristics were broadly similar between included and excluded patients without an echocardiography within the 30- to 365-day period. In addition, follow-up intervals varied, and patients with moderate LVEF improvement had slightly longer intervals between CABG and postoperative echocardiography, which may have contributed to greater observed recovery. Follow-up time was included as a covariate in the adjusted regression models, but residual confounding related to timing cannot be excluded. Due to Danish data protection regulations implemented in 2022,35 retrospective access to electronic medical records was not possible. This limited verification of indications for postoperative imaging, perioperative details (eg, viability testing), and changes in medical therapy between pre- and postoperative echocardiography, which may have influenced recovery. Finally, the moderate sample size limited subgroup analyses and generalizability to patients with severe LV dysfunction or concomitant procedures.
Conclusions
In patients undergoing isolated CABG, postoperative improvements in LVEF and GLS were frequent and were associated with all-cause mortality. Lower preoperative LV systolic function identified patients most likely to experience postoperative LVEF recovery, whereas greater LV dilatation appeared to identify patients with less LVEF improvement. GLS recovery was more closely related to lower preoperative GLS, clinical comorbidity, RV function, and LA strain. Postoperative increases in LVEF and GLS were associated with reduced mortality, underscoring the potential clinical relevance of reverse remodeling after CABG. Routine follow-up echocardiography, including strain assessment, may provide useful information on postoperative recovery and outcomes in this population.
Conflict of Interest Statement
Dr Pareek discloses the following relationships: advisory board: AstraZeneca, Janssen-Cilag; Novo Nordisk; grant support: Danish Cardiovascular Academy funded by the Novo Nordisk Foundation and the Danish Heart Foundation, the Danish Heart Foundation; and speaker honoraria: AstraZeneca, Bayer, Boehringer Ingelheim, Janssen-Cilag, Novo Nordisk. Dr Biering-Sørensen has received research grants from Moderna, Bayer, Novartis, Pfizer, Sanofi Pasteur, GSK, Novo Nordisk, AstraZeneca, Boston Scientific, and GE Healthcare; consulting fees from Novo Nordisk, IQVIA, Parexel, Amgen, CSL Seqirus, GSK, and Sanofi Pasteur; and lecture fees from Pfizer, AstraZeneca, Bayer, Novartis, Sanofi Pasteur, GE healthcare, and GSK. Dr Hardt-Lindberg is an employee of Novo Nordisk. All other reported no conflicts of interest.
The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.
Methods
Conventional Echocardiographic Measurements
Left ventricle (LV) wall thickness and chamber dimensions were assessed as interventricular septal thickness in diastole, LV posterior wall thickness in diastole, and LV internal dimensions using standard guideline-based methods.11 LV end-diastolic and end-systolic volumes were obtained by the modified Simpson’s biplane method to calculate LV ejection fraction. LV mass index was derived using the Devereux formula.11 Left atrial (LA) volume index was determined using the biplane area-length method and indexed to body surface area (Mosteller formula). Mitral inflow velocities (E, A, and E/A ratio) were measured by pulsed-wave Doppler, and septal and lateral early-diastolic annular velocities were obtained by tissue Doppler imaging. Right ventricle systolic function was assessed by tricuspid annular plane systolic excursion measured by M-mode through the lateral tricuspid annulus.
Speckle-tracking Strain Analyses
Two-dimensional speckle-tracking echocardiography was used to quantify myocardial and atrial deformation, following the same analytical approach previously described for this cohort.8 For global longitudinal strain, endocardial borders were manually traced in apical 2-, 3-, and 4-chamber views, and the region of interest was adjusted to encompass the entire myocardial wall. Tracking quality was verified visually. global longitudinal strain was calculated as the mean peak systolic longitudinal strain. LA strain was assessed as peak atrial longitudinal strain (reservoir function), peak atrial contraction strain (contractile/booster function), and longitudinal atrial conduction strain conduit function) using standard speckle-tracking methods consistent with prior work in this cohort.15 All strain values were expressed as absolute values, and the same methodology was applied to both pre- and postoperative echocardiograms.
Appendix E1
Table E1.
Comparison of preoperative characteristics between excluded patients and the final study cohort
| Group | Characteristic | Excluded patients (n = 477) |
Current cohort patients (n = 226) |
P value |
|---|---|---|---|---|
| Demographics | Age (y) | 68.6 ± 8.3 | 67.4 ± 9.3 | .17 |
| Female (%) | 35 (20) | 31 (14) | .10 | |
| BMI | 27.0 ± 3.7 | 27.3 ± 4.2 | .33 | |
| No. of diseased vessels | .91 | |||
| 1 | 3 (0.6) | 1 (0.4) | ||
| 2 | 92 (19) | 46 (20) | ||
| 3 | 381 (80) | 179 (79) | ||
| AMI as indication for CABG | 253 (53) | 104 (46) | .09 | |
| NYHA >II | 10 (2.1) | 5 (2.2) | >.99 | |
| CCS class | .86 | |||
| 1 | 39 (8.2) | 18 (8.0) | ||
| 2 | 370 (78) | 179 (79) | ||
| 3 | 62 (13) | 28 (12) | ||
| 4 | 5 (1.1) | 1 (0.4) | ||
| Current smoking | 376 (79) | 179 (79) | >.99 | |
| Comorbidities | Atrial fibrillation | 11 (2.3) | 8 (3.5) | .49 |
| Hypertension | 320 (67) | 161 (71) | .33 | |
| Hypercholesterolemia | 310 (65) | 144 (64) | .78 | |
| Diabetes mellitus | 113 (24) | 67 (30) | .11 | |
| COPD | 32 (6.7) | 20 (8.8) | .40 | |
| Peripheral artery disease | 60 (13) | 18 (8.0) | .09 | |
| Chronic kidney disease | 12 (2.5) | 8 (3.5) | .61 | |
| Prior stroke/TIA | 45 (9.5) | 29 (13) | .22 | |
| Prior myocardial infarction | 107 (22) | 59 (26) | .34 | |
| Biochemistry | Hemoglobin (g/L) | 8.4 ± 0.9 | 8.4 ± 0.8 | .81 |
| Leukocytes (×10⁹/L) | 8.1 (6.9-9.7) | 8.2 (7.1-9.4) | .97 | |
| CRP (mg/L) | 3.0 (2.0-9.0) | 5.0 (2.0-9.0) | .13 | |
| Creatinine (μmol/L) | 97.0 (83.0-109.0) | 91.5 (81.0-107.0) | .037 | |
| CK-MB (μg/L) | 27.0 (20.0-39.0) | 28.0 (21.0-38.0) | .47 | |
| Troponin I (μg/L) | 6.4 (3.4-10.8) | 6.4 (3.9-10.8) | .63 | |
| Total cholesterol (mmol/L) | 4.8 ± 1.2 | 4.6 ± 1.0 | .05 | |
| LDL (mmol/L) | 2.7 (2.0-3.5) | 2.6 (2.1-3.2) | .77 | |
| HDL (mmol/L) | 1.2 (1.0-1.5) | 1.1 (0.9-1.4) | .09 | |
| Triglycerides (mmol/L) | 1.5 (1.0-2.1) | 1.5 (1.1-1.9) | .96 |
Values are presented as n (%), mean ± SD, or median (interquartile range) unless otherwise noted. Bold values indicate statistical significance (P < .05). BMI, Body mass index; CABG, coronary artery bypass grafting; AMI, acute myocardial infarction; NYHA, New York Heart Association; CCS, Canadian Cardiovascular Society (angina classification); COPD, chronic obstructive pulmonary disease; TIA, transient ischemic attack; CRP, C-reactive protein; eGFR, estimated glomerular filtration rate; LDH, lactate dehydrogenase; CK-MB, creatine kinase MB isoenzyme; LDL, low-density lipoprotein; HDL, high-density lipoprotein.
Table E2.
Parameters associated with improvement in left ventricular ejection fraction (LVEF) ≥10%
| Characteristic | Model 1 |
Model 2 |
Model 3 |
|||
|---|---|---|---|---|---|---|
| Odds ratio (95% CI) | P value | Odds ratio (95% CI) | P value | Odds ratio (95% CI) | P value | |
| Age (y) | 1.02 (0.98-1.06) | .36 | 1.02 (0.98-1.07) | .34 | 1.02 (0.98-1.07) | .30 |
| Male sex | 1.23 (0.41-4.28) | .73 | 1.23 (0.40-4.34) | .73 | 1.23 (0.40-4.39) | .73 |
| Diabetes | 1.11 (0.51-2.36) | .79 | 1.37 (0.61-3.03) | .44 | 1.39 (0.61-3.12) | .42 |
| Hypertension | 0.98 (0.45-2.20) | .96 | 0.93 (0.41-2.16) | .87 | 0.85 (0.36-2.08) | .72 |
| Chronic kidney disease | 0.31 (0.01-2.14) | .31 | 0.27 (0.01-2.07) | .28 | 0.26 (0.01-2.03) | .27 |
| AMI as indication for CABG | 1.42 (0.69-2.95) | .344 | 1.42 (0.67-3.05) | .363 | 1.47 (0.68-3.21) | .326 |
| Previous myocardial infarction | 0.68 (0.28-1.59) | .39 | 0.74 (0.29-1.76) | .51 | 0.71 (0.27-1.73) | .46 |
| Time from CABG to postoperative echocardiography (y) | 5.81 (1.06-32.94) | .043 | 5.73 (1.05-32.44) | .045 | 6.84 (1.23-39.78) | .029 |
| LVIDd (cm) | 0.27 (0.09-0.76) | .015 | 0.22 (0.07-0.66) | .009 | 0.20 (0.06-0.62) | .006 |
| LVPWd (cm) | 0.95 (0.10-8.46) | .96 | 0.80 (0.07-8.20) | .86 | 0.74 (0.06-7.97) | .81 |
| LVMi (g/m2) | 0.99 (0.98-1.00) | .09 | 0.99 (0.97-1.00) | .06 | 0.99 (0.97-1.00) | .05 |
| LVEF (%) | 0.89 (0.87-0.92) | <.001 | 0.89 (0.85-0.93) | <.001 | 0.89 (0.85-0.93) | <.001 |
| GLS (%) | 1.01 (0.89-1.15) | .87 | 1.02 (0.89-1.17) | .79 | 1.01 (0.88-1.16) | .86 |
| TAPSE (cm) | 0.95 (0.37-2.47) | .918 | 1.21 (0.44-3.42) | .709 | 1.18 (0.41-3.40) | .761 |
| LAVi (ml/ m2) | 1.03 (1.00-1.07) | .08 | 1.04 (1.00-1.08) | .06 | 1.03 (0.99-1.08) | .11 |
| PALS (%) | 0.99 (0.94-1.04) | .63 | 1.00 (0.95-1.05) | .88 | 1.00 (0.95-1.05) | .90 |
| PACS (%) | 1.01 (0.94-1.09) | .74 | 1.01 (0.94-1.09) | .77 | 1.02 (0.94-1.10) | .69 |
| LACS (%) | 0.96 (0.88-1.05) | .42 | 0.98 (0.89-1.08) | .69 | 0.98 (0.89-1.08) | .71 |
| MV E velocity (m/sec) | 3.16 (0.54-19.20) | .21 | 5.16 (0.75-37.77) | .10 | 5.37 (0.73-42.45) | .10 |
| MV A velocity (m/sec) | 0.90 (0.18-4.44) | .90 | 1.07 (0.19-5.92) | .94 | 1.01 (0.17-5.88) | .99 |
| E/A ratio | 1.32 (0.73-2.40) | .35 | 1.37 (0.72-2.59) | .33 | 1.42 (0.72-2.79) | .30 |
| E/e' mean ratio | 1.09 (1.00-1.19) | .05 | 1.13 (1.02-1.26) | .020 | 1.15 (1.03-1.30) | .015 |
Values are presented as odds ratios (OR) with 95% confidence intervals (CI). Bold values indicate statistical significance (P < .05). AMI, Acute myocardial infarction; CABG, coronary artery bypass grafting; LVIDd, left-ventricular end-diastolic diameter; LVPWd, LV posterior wall thickness; LVMi, left ventricular mass index; GLS, global longitudinal strain; TAPSE, tricuspid annular plane systolic excursion; LAVi, left-atrial volume index; PALS, peak atrial longitudinal strain; PACS, peak atrial contraction strain; LACS, left-atrial conduit strain; MV, mitral valve; MV E, mitral valve early diastolic inflow velocity; MV A, mitral valve late diastolic inflow velocity; E/A ratio, ratio of early (E) to late (A) transmitral flow velocity; E/e’, ratio of early transmitral flow velocity (E) to early diastolic mitral annular velocity (e’).
Table E3.
Parameters associated with lack of substantial left ventricular ejection fraction (LVEF) improvement <10%
| Characteristic | Model 1 |
Model 2 |
Model 3 |
|||
|---|---|---|---|---|---|---|
| Odds ratio (95% CI) | P value | Odds ratio (95% CI) | P value | Odds ratio (95% CI) | P value | |
| Age (y) | 0.98 (0.95-1.02) | .35 | 0.98 (0.94-1.02) | .25 | 0.98 (0.94-1.02) | .28 |
| Male sex | 0.57 (0.17-1.64) | .33 | 0.54 (0.15-1.58) | .29 | 0.52 (0.15-1.55) | .27 |
| Diabetes | 1.39 (0.68-2.96) | .38 | 1.22 (0.58-2.69) | .60 | 1.22 (0.57-2.70) | .62 |
| Hypertension | 1.35 (0.65-2.78) | .41 | 1.37 (0.64-2.92) | .42 | 1.32 (0.59-2.93) | .50 |
| Chronic kidney disease | 1.58 (0.31-12.21) | .61 | 1.72 (0.31-14.64) | .57 | 1.63 (0.28-14.25) | .61 |
| AMI as indication for CABG | 0.81 (0.42-1.59) | .55 | 0.78 (0.39-1.57) | .48 | 0.80 (0.39-1.62) | .53 |
| Previous myocardial infarction | 1.46 (0.66-3.35) | .36 | 1.35 (0.60-3.21) | .48 | 1.31 (0.57-3.12) | .54 |
| Time from CABG to postoperative echocardiography (years) | 0.18 (0.04-0.88) | .035 | 0.17 (0.03-0.84) | .031 | 0.18 (0.04-0.89) | .036 |
| LVIDd (cm) | 1.77 (1.04-3.06) | .037 | 1.94 (1.09-3.56) | .028 | 2.08 (1.14-3.91) | .019 |
| LVPWd (cm) | 1.24 (0.16-9.97) | .84 | 1.49 (0.18-13.72) | .72 | 1.31 (0.14-12.62) | .81 |
| LVMi (g/m2) | 1.01 (1.00-1.02) | .06 | 1.01 (1.00-1.03) | .044 | 1.01 (1.00-1.03) | .042 |
| LVEF (%) | 1.11 (1.08-1.14) | <.001 | 1.12 (1.08-1.18) | <.001 | 1.13 (1.08-1.18) | <.001 |
| GLS (%) | 0.93 (0.83-1.05) | .24 | 0.91 (0.80-1.03) | .16 | 0.91 (0.80-1.03) | .17 |
| TAPSE (cm) | 1.22 (0.51-2.90) | .65 | 1.12 (0.44-2.85) | .81 | 1.18 (0.45-3.11) | .73 |
| LAVi (mL/m2) | 0.98 (0.95-1.02) | .26 | 0.98 (0.94-1.01) | .20 | 0.98 (0.94-1.01) | .20 |
| PALS (%) | 1.00 (0.95-1.04) | .87 | 0.99 (0.94-1.04) | .63 | 1.00 (0.95-1.05) | .91 |
| PACS (%) | 0.99 (0.92-1.05) | .66 | 0.99 (0.93-1.06) | .82 | 1.00 (0.93-1.08) | .96 |
| LACS (%) | 1.00 (0.92-1.08) | .94 | 0.97 (0.90-1.06) | .52 | 0.98 (0.90-1.07) | .60 |
| MV E velocity (m/sec) | 0.36 (0.07-1.93) | .24 | 0.16 (0.02-1.03) | .06 | 0.14 (0.02-0.92) | .045 |
| MV A velocity (m/sec) | 0.88 (0.20-3.94) | .87 | 0.78 (0.15-3.97) | .77 | 0.75 (0.14-3.89) | .73 |
| E/A ratio | 0.85 (0.48-1.53) | .59 | 0.73 (0.39-1.41) | .33 | 0.70 (0.36-1.37) | .28 |
| E/e' mean ratio | 0.93 (0.86-1.01) | .11 | 0.88 (0.79-0.97) | .015 | 0.86 (0.77-0.96) | .010 |
Values are presented as odds ratios (OR) with 95% confidence intervals (CI). Bold values indicate statistical significance (P < .05). AMI, Acute myocardial infarction; CABG, coronary artery bypass grafting; LVIDd, left-ventricular end-diastolic diameter; LVPWd, LV posterior wall thickness; LVMi, left ventricular mass index; GLS, global longitudinal strain; TAPSE, tricuspid annular plane systolic excursion; LAVi, left-atrial volume index; PALS, peak atrial longitudinal strain; PACS, peak atrial contraction strain; LACS, left-atrial conduit strain; MV, mitral valve; MV E, mitral valve early diastolic inflow velocity; MV A, mitral valve late diastolic inflow velocity; E/A ratio, ratio of early (E) to late (A) transmitral flow velocity; E/e’, ratio of early transmitral flow velocity (E) to early diastolic mitral annular velocity (e’).
Table E4.
Parameters associated with improvement in global longitudinal strain (GLS) ≥2%
| Characteristic | Model 1 |
Model 2 |
Model 3 |
|||
|---|---|---|---|---|---|---|
| Odds ratio (95% CI) | P value | Odds ratio (95% CI) | P value | Odds ratio (95% CI) | P value | |
| Age (y) | 1.01 (0.98-1.04) | .59 | 1.01 (0.97-1.04) | .68 | 1.00 (0.97-1.04) | .78 |
| Male sex | 0.74 (0.32-1.78) | .49 | 0.82 (0.34-2.01) | .65 | 0.85 (0.35-2.10) | .71 |
| Diabetes | 0.75 (0.39-1.41) | .38 | 0.77 (0.39-1.50) | .45 | 0.78 (0.39-1.53) | .48 |
| Hypertension | 0.61 (0.33-1.14) | .12 | 0.57 (0.29-1.09) | .09 | 0.57 (0.29-1.12) | .11 |
| Chronic kidney disease | 0.51 (0.07-2.28) | .42 | 0.49 (0.07-2.23) | .40 | 0.51 (0.07-2.31) | .42 |
| AMI as indication for CABG | 1.59 (0.89-2.85) | .116 | 1.74 (0.95-3.21) | .071 | 1.68 (0.91-3.12) | .094 |
| Previous myocardial infarction | 1.31 (0.68-2.50) | .42 | 1.39 (0.70-2.74) | .35 | 1.43 (0.71-2.86) | .31 |
| Time from CABG-postoperative echocardiography (y) | 0.38 (0.09-1.49) | .17 | 0.36 (0.08-1.46) | .16 | 0.34 (0.08-1.39) | .14 |
| LVIDd (cm) | 1.12 (0.51-2.48) | .79 | 1.10 (0.45-2.70) | .84 | 1.09 (0.45-2.70) | .84 |
| LVPWd (cm) | 0.12 (0.02-0.78) | .030 | 0.12 (0.02-0.79) | .032 | 0.12 (0.02-0.86) | .039 |
| LVMi (g/m2) | 0.99 (0.98-1.00) | .12 | 0.99 (0.98-1.00) | .12 | 0.99 (0.98-1.00) | .12 |
| LVEF (%) | 1.00 (0.97-1.04) | .81 | 1.00 (0.97-1.04) | .88 | 1.00 (0.97-1.03) | .97 |
| GLS (%) | 0.81 (0.75-0.87) | <.001 | 0.80 (0.71-0.88) | <.001 | 0.80 (0.71-0.88) | <.001 |
| TAPSE (mm) | 1.08 (1.01-1.17) | .032 | 1.10 (1.01-1.19) | .029 | 1.10 (1.01-1.19) | .032 |
| LAVi (mL/m2) | 1.00 (0.97-1.03) | .82 | 1.01 (0.98-1.04) | .63 | 1.01 (0.98-1.04) | .63 |
| PALS (%) | 1.02 (0.98-1.06) | .32 | 1.02 (0.98-1.06) | .45 | 1.02 (0.98-1.06) | .41 |
| PACS (%) | 1.07 (1.01-1.14) | .032 | 1.07 (1.00-1.14) | .050 | 1.07 (1.01-1.15) | .040 |
| LACS (%) | 0.97 (0.90-1.04) | .40 | 0.96 (0.89-1.03) | .27 | 0.96 (0.88-1.03) | .26 |
| MV E velocity (m/s) | 0.16 (0.03-0.78) | .025 | 0.14 (0.03-0.75) | .024 | 0.15 (0.03-0.78) | .027 |
| MV A velocity (m/s) | 1.75 (0.46-6.71) | .41 | 1.57 (0.37-6.60) | .54 | 1.71 (0.40-7.34) | .47 |
| E/A ratio | 0.48 (0.25-0.84) | .017 | 0.47 (0.24-0.85) | .019 | 0.46 (0.23-0.85) | .020 |
| E/e' mean ratio | 0.93 (0.86-1.00) | .045 | 0.93 (0.85-1.00) | .06 | 0.92 (0.85-1.00) | .06 |
Values are presented as odds ratios (OR) with 95% confidence intervals (CI). Bold values indicate statistical significance (P < .05). AMI, Acute myocardial infarction; CABG, coronary artery bypass grafting; LVIDd, left-ventricular end-diastolic diameter; LVPWd, LV posterior wall thickness; LVMi, left ventricular mass index; GLS, global longitudinal strain; TAPSE, tricuspid annular plane systolic excursion; LAVi, left-atrial volume index; PALS, peak atrial longitudinal strain; PACS, peak atrial contraction strain; LACS, left-atrial conduit strain; MV, mitral valve; MV E, mitral valve early diastolic inflow velocity; MV A, mitral valve late diastolic inflow velocity; E/A ratio, ratio of early (E) to late (A) transmitral flow velocity; E/e’, ratio of early transmitral flow velocity (E) to early diastolic mitral annular velocity (e’).
Table E5.
Parameters associated with improvement in global longitudinal strain (GLS) <2%
| Characteristic | Model 1 |
Model 2 |
Model 3 |
|||
|---|---|---|---|---|---|---|
| Odds ratio (95% CI) | P value | Odds ratio (95% CI) | P value | Odds ratio (95% CI) | P value | |
| Age (y) | 0.96 (0.94-0.99) | .010 | 0.97 (0.94-0.99) | .021 | 0.96 (0.93-0.99) | .019 |
| Male sex | 1.36 (0.63-2.96) | .43 | 1.22 (0.55-2.77) | .62 | 1.22 (0.54-2.79) | .63 |
| Diabetes | 1.15 (0.65-2.03) | .63 | 1.03 (0.56-1.90) | .92 | 1.05 (0.57-1.96) | .87 |
| Hypertension | 0.99 (0.57-1.74) | .99 | 1.22 (0.67-2.22) | .52 | 1.17 (0.63-2.22) | .62 |
| Chronic kidney disease | 2.36 (0.60-11.49) | .24 | 2.13 (0.52-10.72) | .31 | 2.05 (0.49-10.39) | .34 |
| AMI as indication for CABG | 0.87 (0.52-1.45) | .59 | 0.77 (0.45-1.33) | .35 | 0.78 (0.45-1.36) | .38 |
| Previous myocardial infarction | 0.80 (0.45-1.44) | .46 | 0.80 (0.43-1.49) | .49 | 0.76 (0.40-1.42) | .40 |
| Time from CABG to postoperative echocardiography (years) | 1.05 (0.32-3.41) | .93 | 1.00 (0.29-3.44) | .99 | 0.96 (0.27-3.37) | .95 |
| LVIDd (cm) | 0.81 (0.54-1.20) | .29 | 0.70 (0.44-1.12) | .14 | 0.74 (0.46-1.18) | .21 |
| LVPWd (cm) | 2.80 (0.57-14.26) | .21 | 3.94 (0.70-23.13) | .12 | 3.58 (0.62-21.64) | .16 |
| LVMi (g/m2) | 1.00 (1.00-1.01) | .33 | 1.01 (1.00-1.01) | .20 | 1.01 (1.00-1.02) | .17 |
| LVEF (%) | 1.00 (0.97-1.03) | .95 | 1.00 (0.97-1.03) | .78 | 1.00 (0.97-1.03) | .88 |
| GLS (%) | 1.09 (1.03-1.16) | .004 | 1.13 (1.03-1.24) | .012 | 1.14 (1.04-1.25) | .008 |
| TAPSE (mm) | 0.43 (0.22-0.83) | .013 | 0.33 (0.15-0.69) | .004 | 0.35 (0.16-0.75) | .008 |
| LAVi (mL/m2) | 0.98 (0.95-1.00) | .07 | 0.98 (0.95-1.01) | .15 | 0.98 (0.95-1.01) | .16 |
| PALS (%) | 0.99 (0.95-1.02) | .43 | 0.97 (0.93-1.01) | .10 | 0.97 (0.93-1.01) | .09 |
| PACS (%) | 0.97 (0.91-1.02) | .21 | 0.95 (0.90-1.01) | .09 | 0.95 (0.89-1.01) | .10 |
| LACS (%) | 0.99 (0.94-1.05) | .84 | 0.97 (0.90-1.03) | .29 | 0.96 (0.90-1.03) | .24 |
| MV E velocity (m/sec) | 1.07 (0.27-4.32) | .92 | 1.03 (0.23-4.57) | .97 | 0.75 (0.16-3.55) | .72 |
| MV A velocity (m/sec) | 0.65 (0.20-2.16) | .48 | 0.65 (0.17-2.43) | .52 | 0.71 (0.18-2.73) | .62 |
| E/A ratio | 1.23 (0.76-2.02) | .39 | 1.31 (0.76-2.26) | .32 | 1.14 (0.65-2.03) | .64 |
| E/e' mean ratio | 1.00 (0.94-1.07) | .92 | 1.04 (0.96-1.12) | .32 | 1.02 (0.94-1.10) | .69 |
Values are presented as odds ratios (OR) with 95% confidence intervals (CI). Bold values indicate statistical significance (P < .05). AMI, Acute myocardial infarction; CABG, coronary artery bypass grafting; LVIDd, left-ventricular end-diastolic diameter; LVPWd, LV posterior wall thickness; LVMi, left ventricular mass index; GLS, global longitudinal strain; TAPSE, tricuspid annular plane systolic excursion; LAVi, left-atrial volume index; PALS, peak atrial longitudinal strain; PACS, peak atrial contraction strain; LACS, left-atrial conduit strain; MV, mitral valve; MV E, mitral valve early diastolic inflow velocity; MV A, mitral valve late diastolic inflow velocity; E/A ratio, ratio of early (E) to late (A) transmitral flow velocity; E/e’, ratio of early transmitral flow velocity (E) to early diastolic mitral annular velocity (e’).
Table E6.
Cox regression sensitivity analyses for association between changes in systolic function (change in left ventricular ejection fraction [ΔLVEF] and change in global longitudinal strain [ΔGLS]) and all-cause mortality after postoperative echocardiography in patients with preoperative LVEF <40%
|
ΔLVEF | ||||||
|---|---|---|---|---|---|---|
| Model |
ΔLVEF (per 1%) |
0-10% vs <0% |
≥10% vs <0% |
|||
| Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P value | |
| Model A | 0.82 (0.73-0.91) | <.001 | 0.37 (0.09-1.57) | .178 | 0.02 (0.00-0.21) | .001 |
| Model B | 0.81 (0.72-0.91) | <.001 | 0.40 (0.09-1.79) | .232 | 0.02 (0.00-0.24) | .002 |
| Model C | 0.83 (0.74-0.93) | .001 | 0.57 (0.09-3.78) | .563 | 0.02 (0.00-0.27) | .003 |
|
ΔGLS∗ | ||||||
|---|---|---|---|---|---|---|
| Model | ΔGLS (per 1%) |
1.0-1.9% vs <1.0% |
≥2.0% vs <1.0% |
|||
| Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P value | |
| Model A | 0.69 (0.55-0.87) | .002 | 0.55 (0.13-2.33) | .415 | 0.10 (0.02-0.53) | .007 |
| Model B | 0.68 (0.52-0.89) | .006 | 0.59 (0.11-3.05) | .528 | 0.11 (0.02-0.60) | .011 |
| Model C | 0.68 (0.52-0.87) | .003 | 0.92 (0.15-5.50) | .928 | 0.08 (0.01-0.49) | .006 |
Bold values indicate statistical significance (P < .05).
HR, Hazard ratio; CI, confidence intervals; LVEF, left ventricular ejection fraction; GLS, global longitudinal strain.
Values are presented as hazard ratios (HR) with 95% confidence intervals (CI).
References
- 1.Nalysnyk L., Fahrbach K., Reynolds M.W., Zhao S.Z., Ross S. Adverse events in coronary artery bypass graft (CABG) trials: a systematic review and analysis. Heart. 2003;89(7):767–772. doi: 10.1136/heart.89.7.767. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Benjamin E.J., Muntner P., Alonso A., et al. Heart disease and stroke statistics-2019 update: a report from the American Heart Association. Circulation. 2019;139(10):e56–e528. doi: 10.1161/CIR.0000000000000659. [DOI] [PubMed] [Google Scholar]
- 3.Glance L.G., Osler T.M., Mukamel D.B., Dick A.W. Effect of complications on mortality after coronary artery bypass grafting surgery: evidence from New York State. J Thorac Cardiovasc Surg. 2007;134(1):53–58. doi: 10.1016/j.jtcvs.2007.02.037. [DOI] [PubMed] [Google Scholar]
- 4.van Domburg R.T., Kappetein A.P., Bogers A.J.J.C. The clinical outcome after coronary bypass surgery: a 30-year follow-up study. Eur Heart J. 2009;30(4):453–458. doi: 10.1093/eurheartj/ehn530. [DOI] [PubMed] [Google Scholar]
- 5.Kang D.K., Lim S.H., Park J.S., Sun J.S., Ha T., Kim T.H. Clinical utility of early postoperative cardiac multidetector computed tomography after coronary artery bypass grafting. Sci Rep. 2020;10(1):9186. doi: 10.1038/s41598-020-66176-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gaudino M., Dangas G.D., Angiolillo D.J., et al. Considerations on the management of acute postoperative ischemia after cardiac surgery: a scientific statement from the American Heart Association. Circulation. 2023;148(5):442–454. doi: 10.1161/CIR.0000000000001154. [DOI] [PubMed] [Google Scholar]
- 7.Montalescot G., Sechtem U., Achenbach S., et al. Task Force Members 2013 ESC guidelines on the management of stable coronary artery disease: the Task Force on the Management of Stable Coronary Artery Disease of the European Society of Cardiology. Eur Heart J. 2013;34(38):2949–3003. doi: 10.1093/eurheartj/eht296. [DOI] [PubMed] [Google Scholar]
- 8.Olsen F.J., Lindberg S., Pedersen S., et al. Global longitudinal strain predicts cardiovascular events after coronary artery bypass grafting. Heart Br Card Soc. 2021;107(10):814–821. doi: 10.1136/heartjnl-2020-318462. [DOI] [PubMed] [Google Scholar]
- 9.Haxhibeqiri-Karabdic I., Hasanovic A., Kabil E., Straus S. Improvement of ejection fraction after coronary artery bypass grafting surgery in patients with impaired left ventricular function. Med Arch. 2014;68(5):332–334. doi: 10.5455/medarh.2014.68.332-334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Papestiev V., Jovev S., Risteski P., et al. Myocardial function after coronary artery bypass grafting in patients with preoperative preserved left ventricular ejection fraction-the role of the left ventricular longitudinal strain. Medicina (Kaunas) 2023;59(5):932. doi: 10.3390/medicina59050932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lang R.M., Badano L.P., Mor-Avi V., et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr Off Publ Am Soc Echocardiogr. 2015;28(1):1–39.e14. doi: 10.1016/j.echo.2014.10.003. [DOI] [PubMed] [Google Scholar]
- 12.Diller G.P., Wasan B.S., Kyriacou A., et al. Effect of coronary artery bypass surgery on myocardial function as assessed by tissue Doppler echocardiography. Eur J Cardiothorac Surg. 2008;34(5):995–999. doi: 10.1016/j.ejcts.2008.08.008. [DOI] [PubMed] [Google Scholar]
- 13.Koene R.J., Kealhofer J.V., Adabag S., Vakil K., Florea V.G. Effect of coronary artery bypass graft surgery on left ventricular systolic function. J Thorac Dis. 2017;9(2):262–270. doi: 10.21037/jtd.2017.02.09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wrobel K., Stevens S.R., Jones R.H., et al. Influence of baseline characteristics, operative conduct, and postoperative course on 30-day outcomes of coronary artery bypass grafting among patients with left ventricular dysfunction: results from the surgical treatment for ischemic heart failure (STICH) trial. Circulation. 2015;132(8):720–730. doi: 10.1161/CIRCULATIONAHA.114.014932. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Duus L.S., Lindberg S., Olsen F.J., et al. Left atrial strain predicts heart failure and cardiovascular death in patients undergoing coronary artery bypass grafting. JACC Cardiovasc Imaging. 2021;14(1):295–296. doi: 10.1016/j.jcmg.2020.07.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Downey M.C., Hooks M., Gravely A., et al. Perioperative changes in left ventricular systolic function following surgical revascularization. PLoS One. 2022;17(11) doi: 10.1371/journal.pone.0277454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zafrir B., Leviner D.B., Saliba W., Sharoni E. Prognostic interplay of chronic kidney disease, anemia, and diabetes in coronary bypass surgery. Ann Thorac Surg. 2021;111(1):94–101. doi: 10.1016/j.athoracsur.2020.04.124. [DOI] [PubMed] [Google Scholar]
- 18.Ohira S., Doi K., Numata S., Yamazaki S., Kawajiri H., Yaku H. Impact of chronic kidney disease on long-term outcome of coronary artery bypass grafting in patients with diabetes mellitus. Circ J. 2016;80(1):110–117. doi: 10.1253/circj.CJ-15-0776. [DOI] [PubMed] [Google Scholar]
- 19.Patlolla S.H., Crestanello J.A., Schaff H.V., et al. Changes in ejection fraction after coronary artery bypass grafting for ischemic cardiomyopathy: impact on survival and determinants of improvement. J Thorac Cardiovasc Surg. 2026;171(4):944–953.e3. doi: 10.1016/j.jtcvs.2025.10.029. [DOI] [PubMed] [Google Scholar]
- 20.Bakaeen F.G., Gaudino M., Whitman G., et al. 2021: the American Association for Thoracic Surgery Expert Consensus Document: coronary artery bypass grafting in patients with ischemic cardiomyopathy and heart failure. J Thorac Cardiovasc Surg. 2021;162(3):829–850.e1. doi: 10.1016/j.jtcvs.2021.04.052. [DOI] [PubMed] [Google Scholar]
- 21.Truong D.T., Dionne A., Muniz J.C., et al. Clinically suspected myocarditis temporally related COVID-19 vaccination in adolescents and young adults: suspected myocarditis after COVID-19 vaccination. Circulation. 2022;145(5):345–356. doi: 10.1161/CIRCULATIONAHA.121.056583. [DOI] [PubMed] [Google Scholar]
- 22.Krishnasamy R., Hawley C.M., Stanton T., et al. Left ventricular global longitudinal strain is associated with cardiovascular risk factors and arterial stiffness in chronic kidney disease. BMC Nephrol. 2015;16:106. doi: 10.1186/s12882-015-0098-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Tröbs S.O., Prochaska J.H., Schwuchow-Thonke S., et al. Association of global longitudinal strain with clinical status and mortality in patients with chronic heart failure. JAMA Cardiol. 2021;6(4):448–456. doi: 10.1001/jamacardio.2020.7184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lassen M.C.H., Lindberg S., Olsen F.J., et al. Early diastolic strain rate in relation to long term prognosis following isolated coronary artery bypass grafting. Int J Cardiol. 2021;345:137–142. doi: 10.1016/j.ijcard.2021.10.022. [DOI] [PubMed] [Google Scholar]
- 25.Ródenas-Alesina E., Lozano-Torres J., Badia-Molins C., et al. Left atrial strain in non-ischemic dilated cardiomyopathy in sinus rhythm: prognostic implications and interrelationship with left ventricular deformation. Int J Cardiol. 2025;420 doi: 10.1016/j.ijcard.2024.132753. [DOI] [PubMed] [Google Scholar]
- 26.Taha S., Taha A., Refaat M., et al. Assessment of right ventricular function in patients after coronary artery bypass graft: a single center study. Sci Rep. 2025;15(1) doi: 10.1038/s41598-025-25795-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Labus J., Winata J., Schmidt T., et al. Perioperative two-dimensional left ventricular global longitudinal strain in coronary artery bypass surgery: a prospective observational pilot study. J Cardiothorac Vasc Anesth. 2022;36(1):166–174. doi: 10.1053/j.jvca.2021.08.004. [DOI] [PubMed] [Google Scholar]
- 28.Mihos C.G., Liu J.E., Anderson K.M., et al. Speckle-tracking strain echocardiography for the assessment of left ventricular structure and function: a scientific statement from the American Heart Association. Circulation. 2025;152(10):e96–e109. doi: 10.1161/CIR.0000000000001354. [DOI] [PubMed] [Google Scholar]
- 29.Kang H.U., Nam J.S., Kim J.H., Chin J.H., Choi I.C. Incremental prognostic value of left ventricular longitudinal strain over ejection fraction in coronary artery bypass grafting. J Cardiothorac Vasc Anesth. 2022;36(12):4305–4312. doi: 10.1053/j.jvca.2022.08.011. [DOI] [PubMed] [Google Scholar]
- 30.Schinkel A.F.L., Bax J.J., Delgado V., Poldermans D., Rahimtoola S.H. Clinical relevance of hibernating myocardium in ischemic left ventricular dysfunction. Am J Med. 2010;123(11):978–986. doi: 10.1016/j.amjmed.2010.03.025. [DOI] [PubMed] [Google Scholar]
- 31.Panza J.A., Chrzanowski L., Bonow R.O. Myocardial viability assessment before surgical revascularization in ischemic cardiomyopathy: JACC review topic of the week. J Am Coll Cardiol. 2021;78(10):1068–1077. doi: 10.1016/j.jacc.2021.07.004. [DOI] [PubMed] [Google Scholar]
- 32.Panza J.A., Ellis A.M., Al-Khalidi H.R., et al. Myocardial viability and long-term outcomes in ischemic cardiomyopathy. N Engl J Med. 2019;381(8):739–748. doi: 10.1056/NEJMoa1807365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kloner R.A., Jennings R.B. Consequences of brief ischemia: stunning, preconditioning, and their clinical implications: part 2. Circulation. 2001;104(25):3158–3167. doi: 10.1161/hc5001.100039. [DOI] [PubMed] [Google Scholar]
- 34.Nakae M., Kainuma S., Toda K., et al. Incidence, determinants and clinical impact of left ventricular function recovery after surgical treatments for ischaemic cardiomyopathy. Eur J Cardiothorac Surg. 2021;60(3):689–696. doi: 10.1093/ejcts/ezab122. [DOI] [PubMed] [Google Scholar]
- 35.Forskning og statistik. https://www.datatilsynet.dk/regler-og-vejledning/forskning-og-statistik