Abstract
Background
A reduction in right ventricular function commonly occurs in the early postoperative period after coronary artery bypass graft surgery (CABG). We sought to determine the longer-term effect of CABG on right ventricular function.
Methods
Cardiac magnetic resonance imaging was performed before and approximately 3 months after surgery in 28 patients undergoing elective CABG. Right ventricular (RV) ejection fraction was assessed by planimetry of electrocardiographically gated cine images.
Results
There was a statistically significant increase in left ventricular ejection fraction from 50% to 58% ( P =.003) after CABG. RV ejection fraction also increased from 54% to 60% ( P =.002). In patients with lower baseline RV ejection fraction (below the median, < 53%), this parameter improved from 47% to 57% ( P <.001). Both on-pump (47% vs. 62%, P =.003) as well as off-pump CABG (47% vs. 55%, P =.009) lead to an improvement in RV function in patients in the initial low RV ejection fraction group.
Conclusion
Long-term right ventricular function was not adversely affected by CABG. An improvement in RV function occurred after surgery in patients with low baseline RV ejection fraction and was similar in patients who underwent surgery with or without cardiopulmonary bypass.
1
Introduction
Right ventricular (RV) dysfunction is a recognized cause of hypotension early after coronary artery bypass graft surgery (CABG) . RV dysfunction may develop intraoperatively and persist into the early postoperative phase ; however, it is less clear whether CABG has a longer-term deleterious effect on the right ventricle. Traditional methods for quantifying RV function are challenging due to the unusual shape of the RV chamber and assessment post CABG is further complicated by adhesions to the chest wall . Cardiac magnetic resonance imaging (MRI) can provide accurate quantification of RV ejection fraction after CABG but has only been used in patients with normal preoperative RV function . Preoperative RV dysfunction increases the risks of CABG and is a predictor of late RV function . We sought to determine whether CABG has an adverse effect on the RV when assessed late postoperatively.
2
Methods
Patients undergoing elective CABG surgery between November 2005 and May 2007 at a single institution were offered enrollment. Screening took place after cardiac catheterization. Exclusion criteria included contra-indications to magnetic resonance imaging or gadolinium. Cardiac MRI was performed preoperatively and then 3–6 months following CABG surgery on a 1.5-T scanner with a five-channel phased-array coil (Intera, Philips Medical Systems, Best, Netherlands). A standard clinical cardiac MRI was carried out including a series of breath-hold cine Steady-State Free Precession images (typical imaging parameters matrix 256×256, field of view 350, slice thickness 6 mm, with 30 phases acquired during each cardiac cycle resulting in temporal resolution of 20–40 ms). In addition to four- and two-chamber views, a series of approximately 10 images in a left ventricular short axis orientation were acquired.
Image analysis was performed using commercially available software (ViewForum, Philips). Right and left ventricular ejection fractions were calculated by manual planimetry of a series of short axis images using the summation of discs method.
Statistical analysis was performed using Stata 10.0 for Macintosh (StataCorp, College Station, TX, USA). Continuous variables were expressed as means with standard deviations. The Wilcoxon signed rank test was used to assess change in ventricular function post cardiac surgery. The Wilcoxon rank sum test (Mann–Whitney U ) was used to assess the differences between on and off pump surgery groups. A level of .05 was deemed to be statistically significant.
2
Methods
Patients undergoing elective CABG surgery between November 2005 and May 2007 at a single institution were offered enrollment. Screening took place after cardiac catheterization. Exclusion criteria included contra-indications to magnetic resonance imaging or gadolinium. Cardiac MRI was performed preoperatively and then 3–6 months following CABG surgery on a 1.5-T scanner with a five-channel phased-array coil (Intera, Philips Medical Systems, Best, Netherlands). A standard clinical cardiac MRI was carried out including a series of breath-hold cine Steady-State Free Precession images (typical imaging parameters matrix 256×256, field of view 350, slice thickness 6 mm, with 30 phases acquired during each cardiac cycle resulting in temporal resolution of 20–40 ms). In addition to four- and two-chamber views, a series of approximately 10 images in a left ventricular short axis orientation were acquired.
Image analysis was performed using commercially available software (ViewForum, Philips). Right and left ventricular ejection fractions were calculated by manual planimetry of a series of short axis images using the summation of discs method.
Statistical analysis was performed using Stata 10.0 for Macintosh (StataCorp, College Station, TX, USA). Continuous variables were expressed as means with standard deviations. The Wilcoxon signed rank test was used to assess change in ventricular function post cardiac surgery. The Wilcoxon rank sum test (Mann–Whitney U ) was used to assess the differences between on and off pump surgery groups. A level of .05 was deemed to be statistically significant.
3
Results
Twenty-eight patients were studied pre and post CABG. The median age was 65 (±11) years and 25% were female. The baseline left ventricular ejection fraction (EF) and RVEF were 50% (±16%) and 54% (±9%), respectively ( Table 1 ). As expected, there was a statistically significant increase in left ventricular ejection fraction from 50% to 58% ( P =.003). RVEF also increased significantly from 54% to 60% ( P =.002). Fifteen patients underwent CABG without cardiopulmonary bypass (“off pump”). There was no difference found between the on- and off-pump groups with respect to change in RVEF ( P =.39 for comparison between groups) ( Fig. 1 ). In patients with baseline RVEF below the median (RVEF <53%), there was also a significant increase in RVEF from 47% to 57% ( P <.001). In this low baseline RVEF group, five and nine patients underwent on- and off-pump CABG, respectively. The improvement in RVEF was noted in both on-pump (47% vs. 62%, P =.04) as well as off-pump (47% vs. 55%, P =.01) CABG groups with low baseline RVEF ( Fig. 2 ).
