| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2005;28:783-787
© 2005 Elsevier Science NL
Erratum |
a Division of Cardiology, Loma Linda University Medical Center, Loma Linda, CA, USA
b Division of Cardiothoracic Surgery, Loma Linda University Medical Center, Loma Linda, CA, USA
* Corresponding author at: Division of Cardiology, University of Southern California, 1510 San Pablo Street, 322, Los Angeles, CA 90033, USA. Tel.: +1 323 442 6131; fax: +1 323 442 6133. (Email: rpai{at}usc.edu).
Abstract
OBJECTIVE: Fate of MR following CABG is variable. Predictors of MR regression following CABG alone are not known.
METHODS: From our surgical registry, CABG patients with both pre-operative and post-operative resting echocardiograms at our institution were screened. Of the 523 patients identified, 92 had 3+ (n = 65) or4+ (n = 27) MR on the pre-operative echocardiogram on a 0–4 scale, who had isolated CABG. MR regression was correlated with clinical, operative, electrocardiographic and echocardiographic variables.
RESULTS: Patient characteristics: age 68 ± 11 years, 62% male, and LVEF 37 ± 15%. MR grade decreased from 3.3 ± 0.5 to 2.3 ± 1.2 post-CABG. Residual 3 or 4+ MR post-CABG was present in 43 (47%) patients. Regression of MR (n = 49) was associated with reductions in LV end-diastolic (P = 0.006) and end-systolic (P = 0.0005) dimensions, improvement in LVEF (P = 0.01), longer cross-clamp time (P = 0.04), use of beta-blockers (P = 0.04) and lower presence of CVA as a possible marker of lower atherosclerotic burden (P = 0.03). There was a trend towards increased mortality (P = 0.3) with residual 3–4+ MR over a mean follow-up of 3.9 years.
CONCLUSIONS: In nearly half of patients with 3–4+ MR, MR does not regress with CABG alone. Residual MR may be associated with increased mortality. Regression of MR is related to LV size reduction and improvement in LV function. Presence of myocardial viability, adequate revascularization, lack of excessive atherosclerotic burden and therapy with beta-blockers and ace-inhibitors may be critical for MR regression following CABG alone.
Abbreviations: CABG = coronary artery bypass graft surgery • CVA = cerebral vascular accident • EKG = electrocardiogram • IMR = ischemic mitral regurgitation • LA = left atrium • LV = left ventricle • LVd = left ventricle end diastolic dimension • LVEF = left ventricle ejection fraction • LVs = left ventricle end systolic dimension • MR = mitral regurgitation • NSR = normal sinus rythum • NYHA = New York Heart Association Class
Key Words: CABG • Mitral regurgitation
1. Introduction
MR is common and present in 3–7% of patients undergoing CABG [1,2]. Optimal management of 3–4+ MR in patients undergoing CABG is unclear. While, there have been no prospective randomized studies to guide therapy of MR; many studies have contradictory results. Christenson et al. [3] demonstrated that five out of six patients with 3+ MR (LVEF 17%) treated with CABG alone, improved by one or more grade. Tolis et al. [4] showed all five patients with 3+ MR (mean LVEF 22.4%) had reduction in their MR grade with CABG alone. Mortality at 5 and 10 years follow-up was not different in 58 patients with moderate MR treated with CABG alone, in a study by Duarte et al. [5,6]. Harris et al. [7] advocated mitral valve surgery with CABG for 3–4+ MR because 36% out of 71 patients with moderate IMR (LVEF 38%) had residual 3–4+ IMR with CABG alone. Additionally, Aklog et al. [8] had 40% residual 3–4+ IMR out of 68 patients (LVEF 38%) treated with CABG alone. In a study by Lam et al. [10] 467 patients with moderate ischemic mitral regurgitation treated with CABG alone, 22% at 6 weeks post-CABG had severe MR. None of these studies explore the factors associated with improvement of 3–4+ MR with revascularization alone. We investigated the clinical, operative echocardiographic and electrocardiographic variables associated with regression of 3–4+ MR with CABG alone.
2. Methods
2.1 Study population
This retrospective observational cohort study was conducted at a tertiary medical center. Prospectively collected comprehensive computerized surgical registry was screened for adult patients who underwent CABG from July 1993 to December 2001 and had both pre-operative and post-operative echocardiograms at our institution. Of the 523 patients meeting these criteria, 92 had 3 or 4+ MR (3+ in 65 and 4+ in 27) on a 0–4 scale on the pre-operative echocardiograms. Echocardiograms were performed 41 ± 125 days before and 473 ± 611 days after CABG. Pre-EKGs were obtained 133 ± 125 days before and 708 ± 799 days after CABG.
2.2 Clinical data
The following patient characteristics were collected: (1) patient demographics: age and gender; (2) non-cardiac co-morbidities: presence of diabetes mellitus, renal insufficiency (serum creatinine > 2 mg/dl), renal replacement therapy with dialysis, CVA, COPD or peripheral vascular disease; (3) cardiac co-morbidities: extent of coronary artery disease (history of left-main disease and the number of diseased coronary vessels), history of myocardial infraction, congestive heart failure, cardiogenic shock and arrthymia; (4) prior cardiac surgical history: history of prior CABG; (5) surgical variables: perfusion time, cross-clamp time, number, type and location of grafts and use of intra-aortic balloon pump; and (6) medication use prior to surgery: use beta-blockers, ACE-inhibitor, digoxin and diurectics.
2.3 Echocardiogram data
Trans-thoracic or trans-esophageal echocardiographic examinations were performed with the use of standard techniques and commercially available equipment and anatomic measurements were made according to the American Society of Echocardiography guidelines [9]. The following parameters were collected on pre-CABG and post-CABG echocardiograms: (1) LVEF, (2) LV dimensions and wall thickness, (3) relative wall thickness calculated combined wall thickness/LVd, (4) MR grade as follows: 0 = none, 1 = mild, 2 = moderate, 3 = moderate to severe, 4 = severe based on jet size, size of vena contracta and area of flow acceleration and systolic flow reversal in the pulmonary veins; and (5) wall motion abnormalities: hypokinesis, dyskinesis or akinesis in the anterior wall and non-anterior wall (posterior and inferior) walls.
2.4 Electrocardiograms data
EKG data were collected from the MUSE system (digitally stored EKG database). Following parameters were collected on all EKGs: (1) rythum: sinus, atrial fibrillation (AF) or paced, (2) right and left bundle branch blocks (RBBB, LBBB), and (3) pathological Q-waves in: anterior wall and non-anterior distributions.
2.5 Mortality data
All cause mortality data was obtained from the National Death Index. The different causes of death were not identified.
2.6 Surgical procedures
This being an observational study, choice and number of grafts, and choice of cardioplegic technique, were made entirely by the operating surgeon.
2.7 Statistical analysis
Pre-CABG and post-CABG clinical, echocardiographic and EKG variables associated with regression of 3–4+ MR were identified. The impact on mortality of regression of 3–4+ MR post-CABG was also studied. Statview 5.01 (SAS Institute Inc., Cary, NC) program was used to assist in the statistical analysis. Groups were compared using the unpaired t-test or chi-square test. Kaplan-Meier method was used to produce survival curves and P-value was obtained using log rank analysis. The P-value < 0.05 was considered to be statistically significant.
3. Results
3.1 Baseline characteristics of the study group
This study group of 92 patients had a mean age of 68 ± 11 years with predominance of male gender (62%) and Caucasian race (83%). The mean LVEF was 37 ± 15%. History of prior CABG was present in 19 (21%) patients. AF was present in 13 (14%) patients. The mean NYHA class was 3.5 ± 0.6 and CCS class was 3.6 ± 0.6. IMR grade 3+ was present in 65 (71%) patients and grade 4+ in 27 (29%) patients. Over a mean duration of follow-up of 3.9 years, 38 (41%) patients died.
3.2 Distribution of MR (Fig. 1)
Post-CABG, the mean MR grade decreased from 3.3 ± 0.5 to 2.3 ± 1.2 (Fig. 1
). In patients with 3+ MR pre-CABG (n
= 65), MR grade improved to 0–2+ in 37 (57%) patients, remained 3+ in 20 (31%) patients and progressed to 4+ in 8 (12%) patients. In patients with. pre-CABG 4+ IMR (n
= 27), MR improved to 0–2+ in 12 (44%) patients, regressed to 3+ in 4 (15%) patients, and was unchanged in 11 (41%) patients. Hence, 3–4+ MR persisted in 43 (47%) patients, post-CABG.
|
|
3.5 Operative variables
In the study population, the mean number of bypass vessels grafted were 3.1 ± 1. In the groups with and without MR regression, there was no difference in the distribution of coronary artery disease, number of vessels grafted, whether arterial or venous graft were used. There was also no difference in the number of diseased vessels that were not grafted at all (Table 1).
3.6 Pre-operative echocardiographic variables (Table 1)
Patients with 3–4+ MR regression on pre-CABG echocardiogram had no significant difference in left ventricular ejection fraction (36 ± 16% vs 38 ± 14%, P
= ns), or LV size {LVd (57 ± 7 mm vs 56 ± 10 mm, P
= ns) and LVs (43 ± 8 mm vs 42 ± 12 mm, P
= ns)} (Table 1). There was a trend in patients with 3–4+ MR regression to have a smaller left atrium size (42 ± 5 mm vs 45 ± 7 mm, P
= 0.06). Relative wall thickness was not different in both groups.
3.7 Change in echocardiographic variables post-CABG
Post-CABG echocardiogram was done with a mean duration of follow-up of 1.3 years. The mean MR grade decreased from 3.3 ± 0.5 to 2.3 ± 1.2. Post-CABG, patients with 3–4+ MR regression had increase in LVEF (8.4 ± 18% vs –0.6 ± 14%, P
= 0.01), and decrease in left ventricular dimensions {LVd (–5+9 mm vs 1.8 ± 7 mm, P
= 0.006) and LVs (–6.5 ± 11 mm vs 4+7.7 mm, P
= 0.0005)}. New wall motion abnormality in the post-CABG echocardiogram in the anterior wall or in the non-anterior wall did not correlate with MR regression.
3.8 Electrocardiographic variables (Table 1)
Pre-CABG EKGs were available on 78 patients and post-CABG EKGs were available in 73 patients (Table 1). Pre-CABG EKGs were obtained within a mean of 133 ± 125 days and post-CABG EKGs. were done within 708 ± 799 days. Pre-CABG EKG variables associated with 3–4+ MR regression were (1) slightly higher incidence of sinus rhythm (88% vs 78%, P
= ns); (2) lower incidence of atrial arrthymias (10% vs 22%, P
= ns) like atrial fibrillation (2% vs 14%, P
= 0.06); (3) trend towards higher prevalence of LVH (17% vs 11%, P
= ns), lower prevalence of left atrium enlargement (17% vs 24%, P
= ns), RBBB (5% vs 11%, P
= ns), LBBB (2% vs 8%, P
= ns) and AV blocks (first degree AV block 2% vs 14%, P
= ns). Presence of new Q waves in the anterior and non-anterior leads post-CABG also did not correlate with IMR non-regression.
Subgroup analysis of MR associated with posterior wall motion abnormality: Regression of 3–4+ MR in patients with posterior wall motion abnormality as a function of presence or absence of inferior Q. waves was evaluated. There was no difference in 3–4+ MR in groups characterized by preserved myocardial wall thickness (n = 55) or those with no inferior Q, waves on EKG (n = 46) compared to those with thinned walls (n = 26) or those with pathological inferior Q waves (n = 2). MR regression was not related to inferior wall scarring, dyskinesis, or presence of inferior Q waves.
3.9 Independent predictors of post-operative 3–4+ MR
Stepwise multivariate regression analysis was done using pre-CABG history of CVA, use of beta-blockers, intra-operative cross-clamp time, post-CABG change in LVEF, LVd and LVs. The independent predictor of 3–4+ MR regression was absence of history of pre-CABG CVA.
3.10 Mortality (Fig. 2)
Kaplan-Meier survival curves of patients with 3–4+ MR regression (n
= 49) was compared with patients with 3–4+ MR residual (n
= 43) post-CABG (Fig. 2
). With a mean duration of follow-up of 3.9 years, there was a trend (P
= 0.3) towards higher mortality in patients with 3–4+ residual MR {1 year (23% vs 15%), 3 years (36% vs 27%), and 5 years (49% vs 39%)}.
|
Treatment of 3–4+ MR at the time of CABG is a clinical dilemma that we face routinely.
While no prospectively randomized study has been done to guide therapy, our study has 92 comprehensively characterized patients with 3–4+ MR treated with CABG alone with follow-up data and gives insights into mechanism of MR regression and risk factors for non-regression. We have shown that CABG alone for patients with 3–4+ MR leaves significant (47%) residual MR as also seen in studies by Harris et al. (36%) [7] and Aklog et al. (40%) [8]. Predictors of regression of 3–4+ MR have not previously been studied. In our study, pre-CABG, no significant difference was seen in 3–4+ regressors and non-regressors in the extent of coronary artery disease in terms of left-main and mean number of vessels involved, left ventricular function and size. Younger age and absence of indicators of excessive atherosclerotic disease burden like diabetes mellitus, renal insufficiency and cerebral vascular disease correlated with 3–4+ MR regression. In fact, the only independent predictor of 3–4+ MR regression with CABG alone was absence of history of cerebral vascular accident. Perhaps, patients with 3–4+ MR regression may have higher degree of viable myocardium and/or better re-vascularizability resulting in LV size regression. Patients with cerebral vascular accident history may have a diffuse vascular disease and small vessel disease and this may just be a marker of higher atherosclerotic burden. Their coronaries may have poor distal targets leading to residual ischemia and hence more prevalence of residual mitral regurgitation. There is also a suggestion that maintenance of sinus rhythm with normal intra-ventricular and AV conduction may be beneficial for MR regression post-CABG. It is known that intra-ventricular and atrio-ventricular synchrony are important for mitral valve function.
Regression of 3–4+ MR was seen in patients treated with beta-blockers possibly due to myocardial protection and promotion of reverse remodeling. Hence, we suggest that institution of beta-blocker and ace-inhibitor therapy pre-operatively and continuing long term may benefit in terms of improvement in LV size, function and regression of MR.
On subgroup analysis, there was no association of 3–4+ MR regression with posterior wall motion abnormalities with or without evidence of scarring. This may indicate that regression of LV size may be more important than improvement in regional posterior wall motion abnormality in the regression of functional MR. The number of patients in these groups, however, was small and we did not have any data on myocardial viability by PET scan or magnetic resonance techniques.
Having significant myocardium viable and having it revascularized adequately to reduce post-CABG left ventricular size and improve function correlated with 3–4+ MR regression. Significant three-dimensional change in left ventricle may account for 3–4+ MR regression since new wall motion abnormality in either anterior and non-anterior wall post-CABG did not cause 3–4+ MR residual. Also, wall motion abnormality in only the posterior wall did not correlate with 3–4+ MR regression. Additionally, new Q-wave myocardial infraction in anterior or non-anterior, or posterior wall distribution did not correlate with 3–4+ MR residual. Timek et al. [11] demonstrated that wall motion alone did not correlate with ischemic mitral regurgitation, but alteration in valvular and subvalvular three-dimensional geometry was necessary. Also, Liel-Cohen et al. [12] have shown that ischemic mitral regurgitation is related to change in three-dimensional mitral apparatus due to LV remodeling and dilation and not to LV contractile dysfunction alone. Messas et al. [13] have demonstrated decrease in IMR from inferobasal ischemia due to papillary muscle dysfunction; thus showing the significant role of change in the mitral apparatus as the cause of significant MR.
In contrary to the results of Arcidi et al. [6] and Duarte et al. [5] who reported no increase in mortality with moderate IMR, our study showed a trend towards higher mortality in patients with residual 3–4+ IMR. Duarte et al. did not study the effect on MR grade post-CABG. These patients with pre-CABG moderate MR may have had little to no residual MR post-CABG; hence there was no increase in 5- and 10-year mortality. However, Lam et al. showed statistically significant increase in mortality (P = 0.003) in patients with post-CABG MR. Despite a limited study size, our investigation illustrates an important trend since the effect on mortality of residual 3–4+ MR has not been studied before.
In conclusion, in nearly half of patients with 3–4+ MR, MR does not regress with CABG alone. Residual MR may be associated with increased mortality. Regression of MR is related to LV size reduction and improvement in LV function. Presence of myocardial viability, adequate revascularization, lack of excessive atherosclerotic burden and therapy with beta-blockers and ace-inhibitors may be critical for MR regression following CABG alone.
4.1 Study limitations
This is a retrospective observational study, hence has its inherited limitations. Patients selected for CABG alone versus CABG with mitral surgery at the time of surgery were based on operating surgeons. We have no specific viability study to compare pre-CABG and post-CABG wall motion abnormalities. No data is available on causes of death, and the quantitative details of medical therapy before and after surgery. Despite these limitations, the findings of this study give insights into regression of significant MR with CABG alone.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
