Double Valve Replacement
ABSTRACT & COMMENTARY
Synopsis: The best predictor of postoperative left ventricular function is preoperative left ventricular function.
Source: Skudicky D, et al. Circulation 1997;95:899-904.
Although the hemodynamic determinants of postoperative left ventricular function have been well worked out for single aortic and mitral valve replacements, little is known about this issue in double valve replacement. Thus, Skudicky et al evaluated 44 patients who underwent combined aortic and mitral valve replacement for severe pure-aortic regurgitation and predominant mitral regurgitation (valve area > 1.8 cm2) due to rheumatic heart disease and no coronary artery disease.Posterior mitral valve leaflet chordae were preserved in 27 cases, and nine also had tricuspid annuloplasty. Clinical and color Doppler echocardiographic follow-up was done serially for 4-73 months (mean, 40). Two patients died within 30 days of surgery, and three more died during follow-up, two of non-cardiac causes. All but one patient was in normal sinus rhythm during follow-up. NYHA class decreased significantly from 3.2 to 1.2 (P < 0.001). Left ventricular end diastolic size decreased significantly at three months without any change in systolic size, resulting in a fall in ejection fraction (60% to 48%; P < 0.001).
At one year, end systolic dimension decreased significantly, and ejection fraction was normalized (55%). In six patients, preoperative ejection fraction was less than 50% and dropped from 42% to 35% at three months, but returned to the preoperative value at one year.
The only predictors of postoperative ejection fraction on multivariate analysis of clinical and echocardiographic data were the preoperative end systolic size and ejection fraction. Chordal preservation and angiotensin converting enzyme inhibitor therapy did not influence postoperative ejection fraction.
Left atrial size also decreased (50 mm to 40 mm; P < 0.001). Two patients had mitral valve obstruction 2.0-2.5 years postoperatively, and both had low INRs. One patient was hospitalized for excessive anticoagulation- related bleeding. Five patients had mild ring leaks, but no other valve complications were noted during follow-up.
The authors conclude that in patients undergoing double left-sided valve replacement for rheumatic heart disease related to severe regurgitation, there was an expected early decrease in ejection fraction that normalized after one year. The best predictor of postoperative left ventricular function was preoperative left ventricular function.
COMMENT BY MICHAEL H. CRAWFORD, MD
This observational study makes several important points. First, operative mortality was 4.5%, which is encouraging for those of us who recommend surgery to patients. However, it must be pointed out that this was a young group of patients (mean age, 23) with relatively pure regurgitation from rheumatic heart disease.Second, the hemodynamics of aortic regurgitation dominated the clinical results. Left ventricular ejection fraction initially decreased and then returned to preoperative values at one year as has been observed in series of single valve replacement for aortic regurgitation. By contrast, mitral regurgitation valve surgery studies have generally shown no change or a reduction in ejection fraction depending on the type of mitral regurgitation, the severity of the preoperative left ventricular function, and the use of chordal preservation.
In support of the dominance of aortic regurgitation physiology was the lack of benefit from chordal preservation in this study. However, other studies of rheumatic heart disease patients have also shown less benefit of chordal preservation than series where mitral valve prolapse patients predominate. It may be that preserving rheumatic heart disease chordae is not beneficial because they are involved in the disease process in a way that reduces their value compared to mitral valve prolapse patients (for shortened vs. elongated).
Third, as most studies of patients with left sided valve regurgitation have shown, preoperative left ventricular function is the best predictor of postoperative left ventricular function. In this study as in others, left ventricular function is related to measurements of end systolic size and ejection fraction. End diastolic size is less predictive as is wall stress and other more complicated measurements. This suggests that operation must be performed before the ejection fraction falls below 55% and end systolic dimension increases above 50 mm, regardless of symptoms.
The latter point is not addressed in this study because almost all of their patients had symptoms (NYHA class III on average). Left atrial size, which one study showed was predictive of postoperative left ventricular function in mitral regurgitation patients, was not predictive in this study, again supporting the dominance of aortic regurgitation physiology. Pulmonary artery pressure was shown to be predictive in another mitral regurgitation surgery study but was not evaluated in this non-invasive study.
The explanation for the drop then increase in ejection fraction is not clear from this study. End systolic wall stress tended to parallel left ventricular function, suggesting that changes in loading conditions were related to left ventricular function, but changes in the contractile state could not be excluded. The one-year delay in recovery of left ventricular function could not be explained by changes in loading conditions. This study reaffirms the growing literature that supports early operation in left-sided valvular regurgitation patients before left ventricular dysfunction occurs. Fortunately, most patients with double valve leakage will be symptomatic, making this decision easier.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content