Evaluation of augmented pulse pressure variation using the Valsalva manoeuvre as a predictor of fluid responsiveness under open-chest conditions: A prospective observational study.

BACKGROUND Pulsepressure variation (PPV) is awell known dynamic preload indicator of fluid responsiveness. However, its usefulness in open-chest conditions remains controversial. OBJECTIVE We evaluated whether augmented PPV during a Valsalva manoeuvre can predict fluid responsiveness after sternotomy. DESIGN A prospective, observational study. SETTING Single-centre trial, study period from October 2014 to June 2015. PATIENTS Forty-nine adult patients who underwent off-pump coronary arterial bypass grafting. INTERVENTION After midline sternotomy, haemodynamic parameters were measured before and after volume expansion (6 ml kg(-1) of crystalloids). PPV was calculated both automatically (PPVauto) and manually (PPVmanual). For PPV augmentation, we performed Valsalva manoeuvres with manual holding of the rebreathing bag and constant airway pressure of 30cmH(2)O for 10 s before fluid loading and calculated PPV during the Valsalva manoeuvre (PPVVM). MAIN OUTCOME MEASURES The predictive ability of PPVVM for fluid responsiveness using receiver-operating characteristic curve analysis. Responders were identified when an increase in cardiac index of at least 12% occurred after fluid loading. RESULTS Twenty-one patients were responders and 28 were nonresponders. PPVVM successfully predicted fluid responsiveness with an area under the curve (AUC) of 0.88 [95% confidence interval (95% CI) 0.75 to 0.95; sensitivity 91%, specificity 79%, P<0.0001] and a threshold value of 55%. Baseline PPVauto and PPVmanual also predicted fluid responsiveness [AUC0.75 (0.62 to 0.88); sensitivity79%, specificity 75%; and 0.76 (0.61 to 0.87]; sensitivity 71%, specificity 71%, respectively). However, only PPVVM showed a significant AUC-difference from that of central venous pressure (P = 0.008) and correlated with the change of cardiac index induced by volume expansion (r = 0.6, P<0.001). CONCLUSION Augmented PPV using a Valsalva manoeuvre can be used as a clinically reliable predictor of fluid responsiveness under open-chest condition.