Conventional echocardiographic parameters are currently used in determining the timing for surgery in patients with mitral regurgitation. Since brain natriuretic peptide (BNP) rises in response to ventricular muscle stretch, and is to detect early heart failure, we hypothesized that BNP would be activated in patients with regurgitant valvular heart disease and concomitant left ventricular dilatation. Aim/Objectives: We therefore studied the pattern of changes in NT-pro BNP in patients with chronic severe rheumatic mitral regurgitation who were undergoing mitral valve replacement and compared this with the newer modality of tissue Doppler imaging (TDI). Setting: Patients submitted to surgery were prospectively evaluated over 8 months at Inkosi Albert Luthuli Central Hospital, Department of Cardiology. Controls were obtained from the outpatients' follow-up clinic. Methods: Simultaneous quantification of the severity of mitral regurgitation (MR), left ventricular (LV) end systolic volume (ESV), left atrial (LA) volume and Doppler filling ratios (mitral (E)/annulus (Ea)) were performed at baseline in all patients and was repeated at 1-week and at the six-week follow-up visit in surgical patients. Results: Both groups were similar for age and gender and echo-Doppler parameters in all patients preoperatively except LA size (p< 0.01) and volume (p<0.004) which were more elevated in the surgical group. Mean NT-pro BNP levels were markedly elevated preoperatively (262 pmolll) in all surgical cases compared to controls (57 pmol/l; p=0.0001). NT-pro BNP levels increased further at one week post surgery (395 pmol/l) and subsided at the six week follow-up visit (94 pmol/I). These changes were accompanied by significant reduction in LA (p= 0.003) and LV chamber dimensions (EDD = 0.004) with an increase in the ejection fraction from 42% at one week to 52 % at six weeks. Four patients had abnormally elevated NT-pro BNP levels (>53pmol/l) at the 6-week follow-up visit. A ROC curve was constructed for all variables to separate surgical cases from controls. The area under the curve was highest for NT-pro BNP (sensitivity= 96%, specificity 45 %). Conclusion: 1. There was a significant difference in the left atrial chamber size and volume, as well as Em/Ea (TDI) and NT-proBNP levels preoperatively between the two groups. The lack of a significant difference in the LV parameters between surgical and control groups suggest an almost total reliance on symptoms in deciding the timing of surgery which was reflected by markedly elevated NT-pro BNP in all surgical patients. 2. Postoperatively, there was a significant reduction in LA and LV dimensions. 3. The high false positivity rate for NT-pro BNP suggests that the test is most likely reflecting early LV decompensation in the less symptomatic control patients who rightly need surgery. 4. Tissue Doppler indices had similar sensitivity but low specificity compared to NT-proBNP. 5. Serial estimations of NT-pro BNP may prove useful in selecting patients for surgery.

Thesis (M.Med.Sc.)-University of KwaZulu-Natal, 2007.