Effect of pericardial incision on left ventricular morphology and systolic function in patients during coronary artery bypass grafting

Patients who underwent elective off-pump beating heart CABG between January 2019 and June 2019 in our hospital were consecutive recruited for this study. Patients were excluded if they had a prior history of cardiac surgery, a combination of other cardiac surgery, pericardial disease, arrhythmia, or contraindication to TEE. All patients underwent coronary angiography before CABG and their results were assessed by experienced cardiologist. The SYNTAX score was calculated using the SYNTAX score algorithm (www.​syntaxscore.​com) [8]. All procedures performed in the studies involving human participants were in accordance with the ethical standards of the Ethics Committee of the China Medical University. Informed consent was obtained from all individual participants included in the study.

After induction of anesthesia, an intraoperative TEE examination was performed 5 min before and after pericardial incision during the surgery using the CX50 ultrasound system (Philips Healthcare, Andover, MA, USA) equipped with an X7-2t transducer. The TEE probe was inserted into the patient’s esophagus to obtain mid-esophageal 4-chamber (ME 4C), 2-chamber (ME 2C), and long-axis (ME LAX) views, and into the patient’s stomach to obtain transgastric basal, mid-papillary, and apical short-axis views. The basal level was identified by the mitral valve, and the apical level was defined as proximal to the level with ventricular cavity obliteration at end-systole. All manipulations of image acquisition were performed according to the recommendations from the American Society of Echocardiography [9]. Standard two-dimensional cine loops, including at least 3 consecutive cardiac cycles with frequency of 50–70 frames per second, and Doppler spectrum were stored in cine-loop format.

All data were transferred to a workstation for further offline analysis using the Qlab software package (Philips Healthcare). LV longitudinal and mid-cavity transversal diameters were measured at end-diastole in ME 4C view, and the LV sphericity index (LVSI) was defined as the ratio of the LV longitudinal to mid-cavity transversal diameter [10]. The LV internal diameter and posterior wall thickness were measured at end-diastole in transgastric mid-papillary short-axis view, and the relative wall thickness was calculated simultaneously [9]. The LV end-diastolic volume, end-systolic volume, stroke volume, and LVEF were measured using the biplane modified Simpson’s method with ME 4C and ME 2C views. Mitral inflow early (mitral E) and late (mitral A) diastolic velocity, mitral annular septal and lateral early diastolic velocity (septal e’ and lateral e’), and tricuspid regurgitation velocity were measured in the ME 4C view. Then mitral E/A, and mitral average E/e’ were calculated.

Deformation parameters, including LV GLS, global circumferential (GCS), and global radial strain (GRS), were measured using 2D STE. Endocardial contour was manually adjusted to optimize tracking when necessary, and patients with inadequate tracking, which was defined by the mismatch between the region of interest and the myocardium during dynamic loops, of more than two segments were excluded from the study. The degree of LV peak systolic apical and basal rotation was measured at the apical and basal levels, respectively. Twist was defined as the net difference between apical and basal rotation. The absolute change rates of LV GLS, GCS, GRS, twist, and LVSI were calculated as the percentage of the difference in these parameters before and after pericardial incision divided by that before pericardial incision. A significant reduction of GLS was defined as a decrease of GLS by > 15% immediately after pericardial incision [11].

The RV basal, mid-cavity transversal, and longitudinal diameters were measured at end-diastole in ME 4C view, and the RV sphericity index was defined as the ratio of the RV longitudinal to mid-cavity transversal diameter [10]. The RV end-diastolic and end-systolic areas were measured in ME 4C view, and the RV fractional area change was calculated simultaneously [1].

Hemodynamic monitors were supplemented by radial arterial and central venous catheters. Hemodynamic parameters including systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial blood pressure (MAP), central venous pressure (CVP), and heart rate (HR) were recorded simultaneously at the time of image acquisition.

Intra- and inter-observer agreement for global strain and rotation were examined in 10 randomly selected patients. The same observer, who was blinded to the initial measurements, repeated the measurements to assess intra-observer agreement after more than 4 weeks had elapsed. And another independent observer repeated the measurements twice to assess inter-observer agreement.

Statistical analyses were performed using SPSS 25.0 (IBM Corp., Armonk, NY, USA). Normality plots with tests were performed using the Shapiro–Wilk test. Categorical data are presented as frequencies and percentages, normally distributed data as mean ± SD, and non-normally distributed data as median and interquartile range. Comparisons of continuous variables before and after pericardial incision were made using the paired Student t-test or Wilcoxon test, as appropriate. Comparisons of SYNTAX score between patients with and without reduction of GLS by > 15% were made using the independent Student t-test. The Pearson coefficient was used for correlation analysis. The relationship between the change rate of GLS and continuous variables was analyzed using simple linear regression analysis. Intra-observer and inter-observer agreement analyses of global strain and rotation were assessed using Bland-Altman analysis in 10 randomly selected patients. A P-value < 0.05 was considered significant.

A total of 36 patients performed the intraoperative TEE. Among these patients, 4 patients were excluded because the standard transgastric short-axis view with optimal image quality were unable to be acquired, and 5 patients were excluded due to the inadequate speckle-tracking analyses from apical views (n = 1) and transgastric short-axis views (n = 4). Finally, 27 patients (18 men; mean age, 61.2 ± 6.7 years) were enrolled in the analyses (Fig. 1). During the perioperative period, there were 2 patients experienced prolonged mechanical ventilation due to poor cardiopulmonary function. The baseline characteristics of these patients are shown in Table 1.

Immediately after pericardial incision, there was no significant change in CVP, SBP, DBP, MAP, and HR (Table 2).

The LV mid-cavity transversal diameter increased (P < 0.001), while the LV sphericity index significantly decreased immediately after pericardial incision (P < 0.001), and the posterior and relative wall thicknesses increased (P < 0.05, Table 3). There were no significant differences in mitral E, A, E/A, septal e’, lateral e’ and mitral average E/e’, and tricuspid regurgitation velocity between before and after pericardial incision (Table 3).

In the 27 patients, there were 16 (9.9%) views were manually adjusted to optimize the tracking of LV myocardium. After pericardial incision, LV GLS and GCS decreased, and the GRS increased simultaneously (P < 0.001), while the LV volumes and LVEF remained unchanged. Meanwhile, the reduction of LV twist was accompanied by decreased apical and basal rotation (P < 0.001, Table 3 and Fig. 2). Furthermore, the change rate of GRS was correlated with the change rate of GCS (r = 0.44, P = 0.02), and the change rate of GLS was significantly and inversely correlation with the SYNTAX score (r = − 0.63, P < 0.001).

The RV mid-cavity transversal diameter significantly increased and the RV sphericity index decreased (P < 0.05), while the RV fractional area change remained unaltered, which indicated that the right ventricle became more round in shape (Table 4).

Results of the intra-observer and inter-observer agreement analyses of global strain and rotation are presented in Table 5.

This study has several limitations. First, the correlation between the lower SYNTAX score and the reduction of GLS > 15% was analyzed based on the limited sample size in the present study. Therefore, a further study with adequate samples from multicenter should be designed to verify these findings. In addition, the invasive LV filling pressure and pulmonary capillary wedge pressure were unable to be obtained during the off-pump beating heart CABG, which may provide the pathophysiological basis for LV modification. However, we provided the LV diastolic indices by noninvasive echocardiography, such as mitral E, mitral A, E/A, septal e’, lateral e’ and average E/e’, and tricuspid regurgitation velocity, which can indirectly reflect the LV filling pressure. Furthermore, we measured the tricuspid regurgitation velocity using ME 4C, and its value may be influenced by the angle of ultrasound beams and may be different from that measured by TTE; however, the present study highlighted its change after pericardial incision and the difference may not influence the research objective. Moreover, it is also worth mentioning that we obtained the results of this work with the Qlab software, which was originally developed for transthoracic images. The absolute values of strain and rotation in the present study may not be identical to the values by transthoracic echocardiography. Currently, there are no recognized standards for the definition of a reduction in LV global systolic function using GLS. We referred to the standard on cancer therapy from the American Society of Echocardiography, which recommended that a reduction in GLS > 15% from baseline is very likely to be abnormal during chemotherapy. The results should be further analyzed using special standards in the future.