Normal range of intraoperative three‐dimensionally derived right ventricular free‐wall strain in coronary artery bypass surgery patients

Data on intraoperative three‐dimensionally derived right ventricular free‐wall strain (3D‐RV FWS) is sparse.

was -37.1 to -12.8. There was no relevant correlation of 3D-RV FWS to postoperative outcome in this group of CABG patients.

Conclusion:
We present distribution values for intraoperative 3D-RV FWS and conventional parameters of RV function assessment in a healthy on-pump CABG patient population without serious perioperative complications. We observed no correlations of these parameters with any of the outcome parameters considered. Therefore, we consider these values to be intraoperative TEE-assessed normal values, which can be expected in on-pump CABG patients.

K E Y W O R D S
cardiac surgery, free wall strain, right ventricle, strain analysis, three-dimensional echocardiography

INTRODUCTION
Right ventricular (RV) function strongly predicts outcome after cardiac surgery. 1,2 The myocardial architecture and contraction pattern of the RV are complex and are often not reflected using two-dimensional (2D) echocardiographic measures. 3 Moreover, conventional echocardiographic parameters for RV function assessment have numerous limitations including incomplete imaging of the complex crescent-shaped RV and angle dependency of Doppler and M-mode techniques. 4 Strain analysis with speckle-tracking echocardiography (STE) has advanced as a technique to measure myocardial function. Myocardial strain analysis is less dependent on loading conditions, does not rely on geometric assumptions, has smaller interobserver variability, and is largely angle-independent. 5,6 This is of enormous values in the perioperative setting of cardiac surgery, given the dynamic changes of loading conditions and ventricular geometry. RV longitudinal strain is an independent predictor of RV function and has shown a high correlation with cardiac magnetic resonance imaging. 7 Moreover, intraoperative transesophageal echocardiography (TEE)-assessed RV longitudinal strain is independently associated with short-term outcome in patients undergoing cardiac surgery. 8 Therefore, RV strain analysis seems to be particularly well suited for the perioperative assessment of the RV.
Most data on RV strain analysis, however, comes from the evaluation of awake spontaneously breathing patients using transthoracic echocardiography (TTE). 4,9 Intraoperatively, evaluation of RV strain in anesthetized and ventilated patients using TEE is feasible and supports perioperative decision-making, but data are limited. 5

Study population
The institutional review board of the medical faculty of the Technical University of Dresden, Germany, approved this prospective observational study. After obtaining written informed consent, 150 adult patients were enrolled between January 2019 and April 2020. All patients had a preoperative sinus rhythm, a left ventricular (LV) ejection fraction ≥50%, and a preserved RV function, assessed according to recent guidelines. 4,13 Patients with a history of acute coronary syndrome and/or myocardial infarction, contraindications to TEE, redocardiac surgery, more than mild valvular heart disease, pulmonary hypertension (mean pulmonary artery pressure >25 mmHg), or insufficient quality of echocardiographic assessment preventing the analysis of 3D datasets were excluded. We also excluded patients with any inotropic or vasopressor therapy before or during echocardiography.
Only patients with isolated, stable coronary artery syndrome were included.

Anesthesia management
Following our institutional protocol, anesthesia was induced with fentanyl, propofol, and rocuronium and was maintained with sevoflurane and fentanyl. Lungs were ventilated mechanically with a tidal volume of 6 to 8 mL/kg with a positive end-expiratory pressure of

Surgical management
On-pump CABG surgery was performed under normothermic conditions via median sternotomy. Full myocardial revascularization was achieved using single internal thoracic artery grafting and/or aortocoronary vein grafts, according to the clinical demands. After aortic cross-clamping, cardiac arrest was achieved by blood cardioplegia. In all patients, the pericardium was closed at the end of surgery.

Echocardiographic assessment
TEE was performed after induction of anesthesia, before sternotomy and any CPB cannulation of the patient. All measurements were done with Philips Epiq 7 echocardiography machines and X8-2T-TEE probes.
Examinations were performed by a board certified echocardiographer according to recent recommendations for TEE examination. 14,15 Real-time 3D full-volume datasets of the entire RV were acquired from a mid-esophageal RV-focused four-chamber view. Four-beat acquisition during apnea was performed. The settings of the ultrasound machine were adapted for high temporal resolution. TomTec

Outcome measures
Postoperative outcome measures including MACE (major adverse cardiac events as composite of stroke, cardiovascular death, instable angina pectoris, heart failure), stroke, transitory ischemic attack, shock, TA B L E 1 Baseline patient characteristics (n = 142

RESULTS
3D-RV FWS analysis was feasible in 142 patients (95%). All patients had a complete myocardial revascularization at the end of surgery and pericardial closure. None was pacemaker dependent or on inotropic or vasopressor support during echocardiographic evaluation. Baseline characteristics are presented in Table 1. Essential perioperative data are presented in Table 2.   Table 3. The distribution of these parameters is presented using cumulative histograms in Figure 1.

DISCUSSION
3D-RV FWS analysis is an innovative instrument for assessment of longitudinal RV function. In the present study, we described intraoperative normal values for 3D-RV FWS that can be expected in on-pump CABG patients with preoperative preserved LV and RV function and an uneventful perioperative course, compared to conventional parameters assessing RV function. Although 3D-RV FWS assessment has already been deemed feasible, intraoperative data are sparse. 8,12,19 Intraoperative evaluation of RV strain and decision-making are ambitious in the absence of established reference values for anesthetized and ventilated cardiac surgery patients, which exist for awake spontaneously breathing patients in the non-operative setting. 4,9 Therefore, the present study provides important information in the intraoperative setting.
In about 95% our cohort of CABG patients with preserved LV and RV function and an uneventful intraoperative course, we were implantations in elective and urgent scenarios. Using a custom-made 3D mesh-derived software, they were able to measure 3D-RV FWS.
In their report, they present values of this 3D mesh-derived RV FWS of -17.7 ± 6.9, assessed by TEE after induction of anesthesia and before sternotomy. Their patient population was supported by vasopressors and/or inotropes as clinically necessary. The presented data for 3D-RV FWS of our study patients could be considered normal, because our patient cohort was less heterogeneous than those in the above-mentioned studies, was not supported by vasopressor and/or inotropes, and had no serious perioperative complications. Nonetheless, 3D-RV FWS in our study was more impaired than the normal TA B L E 3 Distributions of intraoperative values of right ventricular function.  Our study has limitations. It is a single-center observational study including 150 patients. Evaluation of 3D-RV FWS is an "offline" analysis and is generally not incorporated in the echo machines at the moment.

3D-RV FWS 3D-RV EF (%) RV FAC (%) RV S' (cm/s) TAPSE (mm)
However, we predict that this novel parameter will be included on the echo machines in the near future and will become available for intraoperative "online" evaluation. We did not analyze regional differences of 3D-RV FWS and did not evaluate for a correlation with localization of coronary heart disease, although most of our patients had triple vessel disease. We did not insert a pulmonary catheter, since it is uncommon to do this in CABG patients with preserved LV and RV function in our institution, as well as in most German heart centers. Therefore, we did not invasively measure pulmonary artery

F I G U R E 1 Continued
needed vasoactive support, and we restricted infusion to <500 mL balanced crystalloid until echocardiographic evaluation. Therefore, we are convinced that pulmonary artery pressure as well RV-pulmonary arterial coupling did not change significantly in our patients; but this remains a limitation of our study. Most of our patients were men aged >60 years, which might influence generalizability of the results. Finally, we did not evaluate intra-or interobserver variability. Since our trial is a prospective single center observational study, the results need to be confirmed in further well-designed multicenter trials. Our data might help to define reference values for intraoperative decision-making in on-pump CABG surgery patients and could serve as a basis for further research.

CONCLUSION
We

ACKNOWLEDGMENTS
The authors would like to thank Mrs. Allison Dwileski for editorial support. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sector.
Open access funding enabled and organized by Projekt DEAL.

CONFLICT OF INTEREST STATEMENT
The authors declare no conflicts of interest.

PATIENT CONSENSUS STATEMENT
All patients were enrolled between January 2019 and April 2020 after providing written informed consent.

DATA AVAILABILITY STATEMENT
All anonymized data are available upon your request. All authors confirm that the had full access to all data of the study and take responsibility for the integrity and accuracy for data analysis.