High patency of proximal splenorenal shunt: A myth or reality ? - A prospective cohort study.

<bold>Background: </bold>Proximal splenorenal shunt (PSRS) is a well-accepted surgical procedure for non-cirrhotic portal hypertension (NCPH). Though a patent shunt is important for good long term outcome, there are very few studies on patency of these shunts. We analysed shunt patency using dynamic computed tomographic (CT) portography and compared it with other modalities.<bold>Methods: </bold>From 2004 to 2014, 50 patients with PSRS were evaluated prospectively for shunt patency using dynamic CT portography, clinical parameters and ultrasound Doppler.<bold>Results: </bold>The causes of NCPH were extrahepatic portal vein obstruction (EHPVO) in 38 patients and non-cirrhotic portal fibrosis (NCPF) in 12 patients. The shunt patency rate using clinical parameters, ultrasound Doppler and dynamic CT portography were 70%, 40% and 60% respectively. Clinical parameters overestimated while ultrasound Doppler underestimated the shunt patency rate. Dynamic portography had 100% correlation with conventional angiography in the five patients when this was done. The site of shunt could be demonstrated convincingly by dynamic CT portography. The shunt patency rate decreased over time. It was 64%, 60% and 43% in <1 year, 1-5 years and >5 years respectively. Our NCPF patients had a greater shunt patency rate compared to EHPVO patients (9/12 vs. 21/38) though the difference was not significant. Only size of the splenic vein had a significant impact on the shunt patency rate on statistical analysis.<bold>Conclusions: </bold>Dynamic CT portography is useful for evaluation of shunt patency. Proximal splenorenal shunts have a high blockage rate which has hitherto not been reported. [ABSTRACT FROM AUTHOR]