DUPLEX-DOPPLER ULTRASONOGRAPHY IN THE ASSESSMENT OF PORTAL-HYPERTENSION - UTILITY OF THE MEASUREMENT OF MAXIMUM PORTAL FLOW VELOCITY

To assess the utility of Duplex-Doppler ultrasonography (DDUS) in the evaluation of portal haemodynamics we studied 52 patients with compensated liver cirrhosis (mean age 53.9 +/- 9.2 years, males 32, females 20) diagnosed by laparoscopy and biopsy. All patients underwent laparoscopy and oesophageal-gastro-duodenoscopy (OGDS): we postulated that haemodynamically significant portal hypertension (PH) was present if varices were detected at OGDS and/or if collateral veins were revealed at laparoscopy. DDUS was performed with a strictly standardized method and maximum portal flow velocity (PFV) was measured in all patients. Max-PFV ranged between 3.5 and 33.4 cm/s. Overall, 36 patients (69%) had a max-PFV lower than 20.3 cm/s (normal max-PFV range in our laboratory is 20.3-33.3 cm/s), while 16 patients (31%) had normal max-PFV values. Five patients (9.6%) had no sips of PH at laparoscopy and/or OGDS and all five had normal max-PFV values. The other 47 patients (90.4%) had collateral circuli at laparoscopy and 29/47 (61.7%) exhibited also varices at OGDS: max-PFV was lower than 20.3 cm/s in 36/47 patients (76.6%). The measurement of max-PFV demonstrated a 76.6% sensitivity and a 100% specificity in detecting PH, with 100% positive predictive value and 31% negative predictive value. Three patients with PH and apparently normal max-PFV values exhibited a recanalization of the umbilical vein. In conclusion, in cirrhotic patients: (1) a low max-PFV can be considered a sure sign of PH; (2) a normal max-PFV is less probative and may be interpreted in different ways: either (a) portal pressure is really normal or (b) portal pressure is high but collateral circuli originate distally from the sampling volume or (c) a hyperdynamic component exists.