Background/aims: The portal pressure response to propranolol administration is heterogeneous in patients with cirrhosis. The aim of this study was to examine the diagnostic accuracy of noninvasive hemodynamic parameters of superior mesenteric artery (SMA) and femoral artery (FA) in the prediction of portal pressure response to propranolol.
Methods: Twenty-six patients with cirrhosis were studied. Portal pressure was assessed by measurements of hepatic venous pressure gradient. Mean arterial pressure and heart rate were also recorded. Cardiac index, and flow velocity of SMA and FA, and pulsatility index of SMA and FA were then measured by means of Doppler ultrasonography. After intravenous propranolol administration (0.10 mg/kg), the above measurements were repeated.
Results: Propranolol significantly reduced cardiac index, heart rate, SMA flow velocity, and FA flow velocity and increased SMA pulsatility index and FA pulsatility index. Although propranolol significantly decreased hepatic venous pressure gradient, a reduction of > or =20% was seen in only 10 patients (good responders); the remaining 16 patients exhibited <20% reduction (poor responders). No significant differences in clinical and baseline hemodynamic data were found in the two groups. There were no also significant differences in changes in heart rate and cardiac index. However, reductions in SMA and FA flow velocity were significantly greater in good responders than in poor responders. Although there was no the increase in FA pulsatility index, the increase in SMA pulsatility index was significantly greater in good responders than in poor responders. When appropriate cut-off points were determined for these variables, overall predictive values of SMA flow velocity (-20%) and SMA pulsatility index (+15%) were 91% and 83%, whereas the overall predictive value of FA flow velocity (-25%) was only 69%.
Conclusions: These results suggest that SMA flow velocity and SMA pulsatility index, but not FA flow velocity and FA pulsatility index, are useful noninvasive parameters in the prediction of portal pressure response to propranolol administration.