Discussion
3D-RV FWS analysis is an innovative instrument for assessment of longitudinal RV function. In the present study we described intraoperativ normal values for 3D-RV FWS that can be expected in onpump CABG patient with preoperative preserved LV and RV function and an uneventful perioperative course, and compared them to conventional parameters of RV function assessment. Although, 3D-RV FWS assessment was already described as feasible, intraoperativ data are sparse.11, 12, 18 Intraoperative evaluation RV strain and decision making is ambitious since there are no established reference values for anesthetized and ventilated cardiac surgery patients as there are for awake, spontaneously breathing patients in the non-operative setting.5, 9 Therefore, our study has important findings.
In our group of CABG patients, with preserves LV and RV function and an uneventful intraoperative course we were able to measure 3D-RV FWS after induction of anesthesia, under stable hemodynamics, a predefined fluid management and without the influence of vasoactive support or pacing, in about 95% of our patients. These measured values for RV FWS are notnormal values as it would be expected for healthy volunteers without cardiac disease, but normal values that can be expected in CABG patients with preserved LV and RV function, and a complication-free intraoperative course. In our opinion normal values from individuals without cardiac disease are not helpful in intraoperative decision making in cardiac surgery patients with different cardiac states (e.g., ischemia or subclinical myocardial dysfunction beside normal conventional parameters). Regarding existing literature, we are unaware of published reports on 3D-RV FWS in onpump CABG patients, beside our own pilot study.12 Other authors report about 3D-RV FWS in different patient populations and clinical scenarios. Keller et al.18 reported in a retrospective chart review about left-sided cardiac valve surgery patients with different grades of LV and RV function and different pulmonary pressure profiles. They described prebypass values of 3D-RV FWS of -19.4 ± 6 (95%-CI -20.9; -17.9) in 59 patients with normal pulmonary artery pressure profiles, but heterogenous LV function, and of -20.5 ± 7.3 (95%-CI -22.4; -18.6) in 60 patients with normal LV function, but different pulmonary pressure profiles. Patients in this group were treated with norepinephrine as clinical necessary. In another retrospective study Keller and colleagues11 observed presternotomy TEE-assessed 3D-RV FWS values of -19.6 ± 6.9 in twenty-three off-pump CABG patients and -20.1 ± 7.1 in twenty mitral valve surgery patients, both groups again treated with vasoactive therapy as clinical necessary. In a huge retrospective study including 496 patients, Keller et al.19 report about a heterogenous cardiac surgery patient population with different LV and RV function constellations including CABG and/or valve surgery procedure and even left ventricular assist device (LVAD) implantations in elective and urgent scenarios. Using a custom-made 3D mesh-derived software they were able to measure three-dimensional RV FWS. In their report they present values of this 3D mesh-derived RV FWS of -17.7 ± 6.9, assessed after induction of anesthesia and before sternotomy by TEE. 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 as normal because our patient collection was less heterogeneous than the above-mentioned studies, not supported by vasopressor and/or inotropes and had no serious perioperative complications. However, 3D-RV FWS in our study was more declined than the normal values of TTE-assessed 2D-RV FWS reported by ASE / EACVI guidelines on chamber quantification5 of -29±4.5, or normal values of 2D-RV FWS from World Alliance of Societies of Echocardiography Study20 of -28.3±4.3, or from a metanalysis from Fine et al.9 of -27±2, described for all ages and genders in healthy individuals. Although all of our patients were preoperatively evaluated having preserved RV function by conventional parameters, 3D-RV EF was reduced (<45%) in 56 patients. Therefore, we evaluated 3D-RV-FWS also in a subgroup of our patients with 3D-RV EF ≥45%.
Differences in the altered described normal values might be explained by the more declined RV function in heart valve surgery and mixed populations of the retrospective studies from Keller and colleagues compared to our relatively healthy CABG population with preserved LV and RV function and without more than mild heart valve disease and without elevated pulmonary pressure. Our data are comparable to observed values for 2D-RV FWS measured by TTE after induction of anesthesia in non-cardiac surgery patients without myocardial disease reported by Dalla21 et al. (-26.5±3.9). Prospective observations of 2D-RV FWS by Donauer et al.22 and Gronlykke et al.,23 each with 30 CABG patients included, revealed higher mean values, which means a more declined RV function, in their patient populations of -22±4 and -19.9±7.2, respectively. This may be explained by differences in used assessment technique, different vendors of echo machines and probes, or by initially more declined RV function beside stated normal or preserved RV function, as evaluated preoperatively by conventional echocardiographic parameters. Our own pilot trial12 showed a median 3D-RV FWS of -24.35 with an interquartile range (IQR) of -29.8 to -20.6. The fact that we measured RV function under stable hemodynamic conditions after induction of anesthesia, without influence of vasopressors or any other intravenous vasoactive therapy and in the absence of rhythm disorder may help to explain our data.
We have strengthened our trial in choosing a homogenous group of patients with preserved LV and RV function, homogenous procedures without influence of vasoactive support on echocardiographic evaluation, a predefined infusion regime and sinus rhythm, used only one vendor and type of echo machine and probes and used three-dimensional full-volume multi-beat acquisition that captures all essential structures through the entire cardiac cycle.
Our study has limitations. It is a single-center observational study including 150 patients. Evaluation of 3D-RV FWS is an “offline” analysis and mostly not incorporated in the echo machines at the moment. However, we suppose that this novel parameter will be included on the echo machines in the near future and 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. But most of our patients had triple vessel disease. Most of our patients were men older than 60 years, which might influence generalization 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 in to find reference values for intraoperative decision making in on-pump CABG surgery patients and could serve as a basis for further research.