A Prospective Study to Assess the Impact of Diastolic Dysfunction on Post-Operative Outcomes after Coronary Artery Bypass Graft Surgery

INTRODUCTION

Heart failure (HF) is a growing global health problem associated with high morbidity, mortality, and impaired quality of life. In India, HF affects about 1% of the population – nearly 8–10 million individuals – with an annual mortality of 0.1–0.16 million.^[1] Although medical therapy has improved outcomes, many patients with advanced coronary artery disease (CAD) still require surgical revascularization, most commonly coronary artery bypass grafting (CABG).

Pre-operative anesthetic evaluation is crucial for optimizing outcomes in cardiac surgery. Established risk models such as the Parsonnet score,^[2,3] EURO score,^[4] and Society of Thoracic Surgeons score^[5] emphasize left ventricular ejection fraction (LVEF) as a marker of systolic function but often neglect diastolic dysfunction (DD), a recognized independent contributor to perioperative risk. Similarly, indices such as the Goldman Cardiac Risk Index,^[6] Detsky Modified Index,^[7] and Revised Cardiac Risk Index^[8] consider myocardial infarction and HF but lack diastolic assessment.

DD, defined by impaired ventricular relaxation and increased stiffness, may occur despite preserved LVEF (heart failure with preserved ejection fraction) and is commonly linked with CAD, hypertension, diabetes, and ischemia.^[9-11] These factors elevate left ventricular filling pressures and cause wall motion abnormalities.^[12] DD independently predicts major cardiac events, prolonged ventilation, extended hospitalization, and mortality.^[13,14]

Earlier, invasive methods like cardiac catheterization were used to evaluate diastolic properties, but their routine application is limited.^[11,15] Doppler echocardiography has since become a reliable, non-invasive alternative for assessing diastolic filling and ventricular compliance.^[16] Echocardiography also helps detect ischemic wall motion and previously unrecognized myocardial dysfunction.^[17] The 2025 American Society of Echocardiography (ASE)/European Association of Cardiovascular Imaging (EACVI) guidelines have refined DD evaluation by integrating parameters such as mitral inflow velocity, tissue Doppler e’ velocity, E/e’ ratio, and left atrial volume, providing greater diagnostic accuracy and clinical relevance.^[18]

While systolic dysfunction has been well studied in CABG outcomes, the influence of DD remains underexplored, particularly in the intraoperative setting. Most studies rely on pre-operative transthoracic echocardiography, leaving limited evidence on intraoperative transesophageal echocardiography (TEE) findings.^[8] This prospective study aims to assess DD intraoperatively using TEE and to examine its association with early post-operative outcomes in elective CABG patients. The primary outcomes include duration of mechanical ventilation, new-onset atrial fibrillation (AF), and hospital length of stay. Secondary outcomes comprise re-intubation, respiratory complications, acute kidney injury (AKI), inotropic and vasopressor requirements, intensive care unit (ICU) readmission, and 7-day and 30-day mortality.

MATERIAL AND METHODS

TEE assessment

Left ventricular diastolic function (DF) was evaluated intraoperatively using standard TEE views, primarily the mid-esophageal four-chamber (ME4C) view at a 0° angle. Mitral inflow velocities – peak early (E) and late (A) diastolic velocities – were obtained using pulsed-wave Doppler with the sample volume placed at the mitral leaflet tips [Figures 1a-c]. The E/A ratio and E-wave deceleration time (DT) were calculated, with DT measured as the interval from peak E velocity to baseline on the deceleration slope.

Close

Figure 1:

Transesophageal image of transmitral flow velocity (a) Normal diastolic function (b) Grade I diastolic dysfunction (c) Grade II diastolic dysfunction.

Export to PPT

Tissue Doppler imaging (TDI) was used to measure lateral e’ and septal e’ velocities [Figures 2a-c and 3a-c], with the sample volume positioned just below the mitral annulus in the ME4C view. These velocities, expressed in cm/s, were used to calculate the E/e’ ratio, reflecting left ventricular (LV) filling pressures. The tricuspid regurgitation (TR) jet velocity was recorded in m/s using continuous-wave Doppler from the mid-esophageal right ventricular inflow–outflow or modified bicaval views [Figures 4a-c].

Close

Figure 2:

Pulsed-wave tissue Doppler imaging (PW-TDI) lateral e’ velocity (a) Normal diastolic function (b) Grade I diastolic dysfunction (c) Grade II diastolic dysfunction.

Export to PPT

Close

Figure 3:

Pulsed-wave tissue Doppler imaging (PW-TDI) septal e’ velocity (a) Normal diastolic function (b) Grade I diastolic dysfunction (c) Grade II diastolic dysfunction.

Export to PPT

Close

Figure 4:

Tricuspid regurgitation systolic jet velocity (a) Normal diastolic function (b) Grade I diastolic dysfunction (c) Grade II diastolic dysfunction.

Export to PPT

DD was defined when one or more of the following abnormalities were present: Septal e’ <7 cm/s, lateral e’ <10 cm/s, or TR velocity >2.8 m/s. Grading of DD was based on abnormalities in two or more parameters (E/A ratio, average E/e’, and DT) as follows: Grade I (impaired relaxation), Grade II (pseudonormal filling), and Grade III (restrictive filling pattern).

Grouping and data collection

A total of 75 patients were included and divided into two groups: Group I (n = 16) with normal DF and Group II (n = 59) with DD [Figure 5]. Intraoperative variables such as total surgical duration, cardiopulmonary bypass time, and number of grafts were documented. Postoperatively, patients were managed in the cardiac ICU following standard protocols. Continuous monitoring included heart rate, MAP, CVP, urine output, and serial arterial blood gases.

Close

Figure 5:

Consort diagram.

Export to PPT

Patients were extubated once institutional weaning criteria were fulfilled. Post-extubation, oxygen was administered through face mask, and patients were monitored for respiratory and hemodynamic stability.

RESULTS

In total, 80 patients were initially enrolled. Three were excluded due to intraoperative conversion to cardiopulmonary bypass and two for re-exploration. After intraoperative TEE evaluation, patients were categorized into two groups: Group I with normal DF (n = 16) and Group II with DD (n = 59). The final analysis included 75 patients who successfully underwent off-pump CABG with complete intraoperative and post-operative assessment.

Table 1 compares baseline demographic parameters between patients with normal and abnormal DF. Patients with abnormal DF were significantly older (64.47 ± 7.67 vs. 55.56 ± 7.02 years; P < 0.001). No significant differences were noted in weight, height, or BMI (P > 0.05), indicating comparable baseline characteristics.

Table 2 summarizes echocardiographic findings for both groups. There were no significant differences in E wave (P = 0.684), A wave (P = 0.862), E/A ratio (P = 0.840), or DT (P = 0.912). However, patients with abnormal DF showed significantly reduced septal e’ (6.05 ± 1.14 vs. 8.53 ± 1.42; P < 0.001), lateral e’ (7.35 ± 1.43 vs. 11.3 ± 1.42; P < 0.001), and average e’ (6.70 ± 1.2 vs. 9.91 ± 0.93; P < 0.001) [Figure 6]. In addition, the E/e’ ratio was markedly higher in the abnormal DF group (9.77 ± 2.46 vs. 6.74 ± 2.12; P < 0.001), indicating elevated filling pressures [Figure 7]. A significant increase in TR jet velocity (1.7 ± 0.7 vs. 1.28 ± 0.72; P = 0.038) was also observed [Figure 8].

Close

Figure 6:

Comparison of echocardiographic measurements (velocities).

Export to PPT

Close

Figure 7:

Comparison of echocardiographic measurements (tricuspid regurgitation jet velocity).

Export to PPT

Close

Figure 8:

Comparison of echocardiographic measurements (ratios).

Export to PPT

Table 3 compares perioperative and post-operative parameters between patients with normal and abnormal DF. Patients with abnormal DF had significantly longer intubation times (16.84 ± 2.98 vs. 6.07 ± 1.23 h; P < 0.001 [Figure 9]), higher VISs (5.87 ± 4.68 vs. 2.40 ± 2.32; P < 0.001, [Figure 10]), and prolonged ICU stays (3.37 ± 1.48 vs. 2.06 ± 0.56 days; P < 0.001 [Figure 11]). The duration of surgery did not differ significantly (P = 0.486). Although hospital stay and packed red blood cells transfusion requirements were higher in the abnormal DF group, they were not statistically significant (P = 0.101 and P = 0.058, respectively).

Close

Figure 9:

Duration of intubation (h).

Export to PPT

Close

Figure 10:

Vasoactive inotropic scores.

Export to PPT

Close

Figure 11:

Length of intensive care unit stay (days).

Export to PPT

Table 4 shows comparable graft distribution between groups, with most patients receiving three or four grafts (P = 1.000). Table 5 demonstrates significantly higher rates of pleural effusion (P = 0.002) and new-onset AF (P = 0.032) in the abnormal DF group, while no differences were found in HOS, reintubation, AKI, or ICU readmission [Figure 12]. No mortality was observed in either group at 7-day or 30-day follow-up [Table 6].

Close

Figure 12:

Incidence of post-operative complications.

Export to PPT

DISCUSSION

This prospective study evaluated the impact of DD on post-operative outcomes in patients undergoing CABG. A high prevalence of abnormal DD (78.67%) was observed, underscoring its clinical relevance. Similar findings were reported by Kyle et al.,^[14] and Ashes et al.,^[19] who noted a significant incidence of pre-operative DD among cardiac surgery patients. These results highlight the importance of incorporating DD assessment into pre-operative evaluation and perioperative management. Metkus et al.^[20] further demonstrated that moderate to severe DD (Grade II–III) independently predicts higher mortality and prolonged mechanical ventilation, emphasizing the value of early identification and intervention.

Our analysis revealed significantly higher age in patients with abnormal DD, consistent with Kyle et al.^[14] and Mondal et al.,^[13] indicating age as a strong determinant of DD. Although Ashes et al.^[19] did not report a significant association, the overall evidence supports advancing age as a key risk factor. No gender-based differences were observed, contrasting with Mondal et al.^[13] and Ludden et al.,^[21] who found higher DD prevalence and poorer outcomes among females.

Echocardiographic parameters, including septal e’, lateral e’, and the E/e’ ratio, were markedly impaired in patients with DD, aligning with the 2025 ASE/EACVI recommendations,^[18] which emphasize these measures as reliable indices of LV relaxation and filling pressure. Previous studies by Metkus et al.,^[20] Kaw et al.,^[22] and Cho et al.^[23] similarly demonstrated that elevated E/e’ ratio correlates with increased post-operative morbidity, prolonged mechanical ventilation, and higher mortality. This reinforces the utility of integrating TEE-based diastolic assessment into perioperative echocardiographic protocols.

Intraoperative parameters, including surgery duration and number of grafts, were comparable between groups, indicating that DD does not directly influence the technical aspects of CABG but may affect post-operative recovery. Patients with DD required significantly longer mechanical ventilation, consistent with findings by Kyle et al.,^[14] Mondal et al.,^[13] and Metkus et al.,^[20] who associated impaired DF with delayed weaning due to reduced cardiac performance and pulmonary congestion.

VISs were also significantly higher in DD patients, suggesting greater cardiovascular instability. Similar observations by Kyle et al.^[14] and the VIS classification proposed by Koponen et al.^[24] link higher VIS with an increase in-hospital mortality, reinforcing its prognostic relevance.

Post-operative complications such as pleural effusion and new-onset AF were more frequent in the DD group, aligning with previous studies indicating elevated left atrial pressures and structural remodeling as contributing factors. The incidence of AKI was comparable between groups, consistent with Mondal et al.,^[13] though Kyle et al.,^[14] reported a higher rate, likely reflecting population or management differences. This variability may reflect differences in patient demographics, surgical techniques, or perioperative management practices.

The requirement for post-operative oxygen support and non-invasive ventilation was higher among patients with DD, though not statistically significant. Mondal et al.^[13] reported a longer oxygen support duration in similar patients. Reintubation occurred only in those with DD, consistent with findings by Kyle et al.^[14] and Mondal et al.,^[13] suggesting that DD may contribute to post-operative respiratory compromise.

No mortality was observed in either group, similar to reports by Kyle et al.^[14] and Ashes et al.,^[19] indicating that DD may not significantly affect short-term survival. However, Kaw et al.^[22] and Cho et al.^[23] identified DD as a predictor of mortality, implying that our study’s limited sample size may have precluded the detection of such differences.

Patients with DD experienced significantly longer ICU and hospital stays, consistent with Kyle et al.,^[14] Metkus et al.,^[20] and Ludden et al.^[21] These findings reflect greater postoperative complications and prolonged supportive care needs in this population.

Diastolic dysfunction has emerged as an important determinant of adverse postoperative outcomes following CABG surgery. The review on prolonged ICU stay after cardiac surgery highlights ventricular dysfunction— including impaired filling and relaxation—as a contributor to low cardiac output states, prolonged ventilation, and extended ICU length of stay, even in patients with preserved ejection fraction.^[25] Perioperative heart failure literature further emphasizes that diastolic dysfunction leads to elevated filling pressures and poor tolerance to fluid shifts, predisposing CABG patients to pulmonary congestion, atrial fibrillation, and delayed recovery.^[26] Collectively, these findings align with complex cardiac surgical experiences showing that underlying myocardial dysfunction, even when subtle, amplifies perioperative risk and negatively impacts postoperative outcomes after coronary revascularization.^[27]

This study presents important insights but also has some limitations that warrant consideration for future research. This single-center study with a modest sample size may limit the generalizability of findings. DF was assessed intraoperatively using a simplified TEE-based approach; however, more advanced modalities such as left atrial strain, myocardial work index, and 3D echocardiography, as recommended in the 2025 ASE/EACVI guidelines, could provide greater diagnostic precision. Exclusion of high-risk patients with severe cardiac disease further restricts applicability to broader populations. The study focused only on short-term post-operative outcomes, without long-term follow-up. In addition, as an observational design, it identifies associations rather than causality, and factors such as preload, left atrial compliance, and hemodynamic variability may have influenced results. Future multicentric studies with larger cohorts and advanced echocardiographic evaluation are needed to validate these findings and explore long-term prognostic implications.

CONCLUSION

In summary, DD is highly prevalent among patients undergoing CABG and is associated with adverse early post-operative outcomes, including prolonged mechanical ventilation, higher inotropic requirements, and increased complications such as AF and pleural effusion. Early recognition allows for optimized perioperative management and individualized hemodynamic strategies, which may improve recovery and reduce ICU and hospital stay durations.