BNP in Mitral Valve Disease: Too Good to be True?
BNP in Mitral Valve Disease: Too Good to be True?
Abstract & Commentary
By Michael H. Crawford, MD
Sources: Pizarro R, et al. Prospective validation of the prognostic usefulness of brain natriuretic peptide in asymptomatic patients with chronic severe mitral regurgitation. JACC. 2009;54:1099-1106. McCullough PA, Hanzel GS. B-type natriuretic peptide and echocardiography in the surveillance of severe mitral regurgitation prior to valve surgery. JACC. 2009;54:1107-1108.
The timing of mitral valve surgery in patients with severe organic mitral regurgitation (MR), but without symptoms, is controversial. Echocardiographic measurements of left ventricular (LV) size and function and severity of MR have shown predictive value in different studies. Also, it is known that brain natriuretic peptide (BNP) blood levels indicate increased wall stress and could be of prognostic value in valve disease patients. Thus, these investigators from Buenos Aires, Argentina, sought to determine the incremental prognostic value of BNP over echo parameters in 269 prospectively evaluated, asymptomatic patients with severe, organic MR and normal LV function (ejection fraction > 60%). The first 167 patients comprised the derivation set and the second 102 patients the validation set. Most of the patients had degenerative MR, were men, and had a mean age in their 60s. Other inclusion criteria included an echo-derived, effective regurgitant orifice area (EROA) > 40 mm2 and a regurgitant volume of > 60 mL/beat, and > 7 METs of exercise on a Bruce protocol treadmill test without symptoms, ventricular arrhythmias, hypotension, or ischemic ECG ST depression. Patients with other valve diseases, ischemic MR, previous cardiac surgery, or cardiomyopathies were excluded. Decisions regarding valve surgery were made by the primary physician without knowledge of the BNP values determined at entry and at one year. The primary combined endpoint was the occurrence of heart failure, LV dysfunction, or death. If a patient was sent to surgery before they developed symptoms or LV dysfunction, they were not considered to have reached the primary endpoint.
Results: Median BNP in the derivation and validation groups were 21 and 27 pg/mL, respectively. Receiver operating curves defined a BNP cutoff value of 105 pg/mL with regard to the primary endpoint. The endpoint was achieved in 21% of both cohorts. The cutoff value identified high-risk groups where the primary endpoint frequency was 76% and 66% in the two cohorts vs. 5% and 4% in the low-risk groups. Patients with a baseline BNP > 105 were more likely to have a new flail leaflet, a larger end-systolic diameter, a bigger left atrium, higher pulmonary pressures, and higher echo indices of the amount of regurgitation. By multivariate analysis, BNP was the strongest independent predictor of the endpoint. When BNP was added to the best echo model (EROA, end systolic diameter, and atrial volume), the ROC area significantly increased from 0.80 to 0.91 (p = 0.01). In the validation cohort, death occurred in 2%, new heart failure in 16%, and LV dysfunction in 4% over a 31-month average follow-up. Also, 4% developed NYHA class II symptoms, 5% new atrial fibrillation, and 10% pulmonary hypertension. Mitral valve surgery was performed in 29% (63% repair). The authors concluded that BNP has incremental prognostic value to echo measures in asymptomatic patients with severe MR and normal LV function.
Commentary
In asymptomatic patients with severe MR, the ACC/AHA and ESC guidelines recommend considering surgery if LV dysfunction, increased LV volume, pulmonary hypertension, or atrial fibrillation develops. This study confirms the predictive value of developing LV dysfunction (ESD/BSA > 22 mm/m2, OR = 3.4, CI 1.6-10.7, p = .01) and adds a new multivariate echo predictor reflecting the amount of mitral regurgitation (EROA > 55 mm2, OR 4.2, CI 2.1-11.4, p = .001). However, BNP was the strongest predictor of the primary outcome (> 105 pg/mL, OR = 4.6, CI 2.7-11.6, p = .0001). None of these three parameters are mentioned in the current guidelines. In addition, those with a large increase in BNP (> 25 pg/mL) over one year also had a higher incidence of the primary endpoint as compared to those with a lesser or no increase. It is hypothesized that BNP is released in these patients as a result of subclinical LV dysfunction or, perhaps, an increase in wall stress of the LV or atria.
Interestingly, other parameters in the guidelines, such as atrial fibrillation, ejection fraction, and pulmonary hypertension, were univariate predictors, but were not significant multivariate predictors. Other predictors mentioned in the literature, such as new flail leaflet, increased end-diastolic diameter, or enlarged left atrium, were univariate predictors, but also not significant in the multivariate analysis. Thus, this study would suggest that more appropriate criteria for surgery should be BNP > 105, ESD/BSA > 22, and EROS > 55. Clearly these parameters should be added to our decision-making process.
In the ACC/AHA guidelines, a IIa indication for surgery in those with severe MR and no symptoms is a likelihood of repair > 90%. The new significant parameters in this study may help to settle this controversial area, but should we operate solely on the basis of an elevated BNP? This is difficult to answer because there are many causes of an elevated BNP that would have to be excluded in a less selected patient population. Remember, a BNP measurement has been shown to be more useful in other patient populations if it is normal. I suspect the same will be true in valvular disease; a normal BNP will stay the surgeon's hand.
The timing of mitral valve surgery in patients with severe organic mitral regurgitation (MR), but without symptoms, is controversial.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.