Left ventricular diastolic dysfunction--an independent risk factor for weaning failure from mechanical ventilation

The objective of this study was to investigate the contribution of left ventricular (LV) diastolic dysfunction to weaning failure, along with the levels of the currently used cardiac biomarkers. Forty-two mechanically ventilated patients, who fulfilled criteria for weaning from mechanical ventilation (MV), underwent a two-hour spontaneous breathing trial (SBT). Transthoracic echocardiography (TTE) was performed before the start of the SBT. The grade of LV diastolic dysfunction was assessed by pulsed-wave Doppler and tissue Doppler imaging at the level of the mitral valve. Haemodynamic and respiratory parameters were recorded. Blood levels of B-type natriuretic peptide (BNP), troponin I, creatine kinase-MB, and myoglobin were measured on MV and at the end of the SBT. Weaning success was defined as the patient's ability to tolerate spontaneous breathing for more than 48 hours. Fifteen patients failed to wean. LV diastolic dysfunction was significantly associated with weaning failure (P<0.001). The grade of diastolic dysfunction was significantly correlated with BNP levels both on MV and at the end of the SBT (P<0.001, r=0.703 and P<0.001, r=0.709, respectively). BNP levels on MV were lower in patients who successfully weaned compared to those who did not (361±523 ng/l versus 643±382 ng/l respectively, P=0.008). The presence of diastolic dysfunction was independently associated with weaning failure (odds ratio [OR] 11.23, confidence interval [CI] 1.16-109.1, P=0.037) followed by respiratory frequency/tidal volume (OR 1.05, CI 1.00-1.10, P=0.048). Therefore, assessment of LV diastolic function before the start of weaning could be useful to identify patients at risk of weaning failure.