Viva · IR system

Portal Hypertension Viva Questions

Oral-exam style questions and answers on portal hypertension for MD, DNB and super-specialty practical examinations. Each answer links back to the playbook or Library entry it was written from.

Why does DIPS pass through the caudate lobe specifically, rather than any other segment?: Because the caudate lobe sits directly between the retrohepatic IVC posteriorly and the portal vein bifurcation anteriorly, giving the shortest path from cava to portal vein that stays entirely within liver parenchyma. Any other trajectory from the IVC to the portal vein leaves the liver and enters the peritoneum or retroperitoneum.
You have a Budd-Chiari patient with a compressed retrohepatic IVC and a measured gradient of 9 mmHg using caval pressure as the reference. Do you stop?: No — the reference is wrong. Caudate hypertrophy has raised the retrohepatic caval pressure, so subtracting it understates the true portosystemic gradient. Recalculate against right atrial pressure, and if caval compression is significant, treat the caval outflow with angioplasty or stenting, since the shunt cannot decompress into an obstructed cava.
Why is IVUS mandatory here when conventional TIPS can be done with wedged CO2 portography alone?: In TIPS, the hepatic vein origin guarantees the needle starts inside liver parenchyma, so a mapped fluoroscopic throw is reasonably safe. In DIPS the needle starts in the IVC, part of whose retrohepatic course is bare and not surrounded by liver, so only real-time imaging can confirm there is caudate parenchyma between the transducer and the target before the throw. Without that confirmation the puncture goes into the retroperitoneum.
Why is the throw shorter than in conventional TIPS, and what happens if you forget?: The caudate tract is typically only 2-4 cm, against a much longer hepatic vein to portal vein tract. A conventional-depth throw overshoots through the portal vein and out of the anterior caudate surface into the peritoneal cavity, causing intraperitoneal haemorrhage.
A DIPS patient is a transplant candidate. What device-related point must you document?: Whether the cranial stent end protrudes into the right atrium or beyond the caval-atrial junction. A transcaval device that extends into the atrium compromises caval reconstruction during hepatectomy, so the surgical team must know before listing, and this is why the covered length is chosen to the measured tract rather than by default.
Six months after DIPS for Budd-Chiari, the shunt occludes abruptly. What is the likely mechanism and how does it differ from post-TIPS dysfunction in cirrhosis?: Thrombosis driven by the underlying thrombophilia, rather than the gradual pseudointimal hyperplasia typical of cirrhotic TIPS dysfunction. It presents abruptly rather than as slowly recurring ascites, and management includes thrombectomy or thrombolysis plus a review of anticoagulation intensity, not angioplasty alone.
Two accesses are used. Justify each.: The jugular sheath carries the curved cannula and needle inline down the IVC, giving a workable anterior throw angle. The femoral sheath carries the IVUS catheter cranially so the transducer sits at the puncture level and images the target from below while the needle advances from above. Combining them into one access would put the imaging and the puncture on the same axis and lose real-time visualisation of the needle approaching the target.
When would you choose IVUS-guided conventional TIPS over DIPS?: Whenever a hepatic vein is patent and usably angled but the portal target is simply hard to hit fluoroscopically. IVUS solves the aiming problem without committing to a transcaval tract, which avoids caval wall injury and the atrial-protrusion and caval-reconstruction issues that DIPS introduces.
You open the right hepatic vein beautifully but the free hepatic vein pressure barely falls. What have you missed and what do you do?: An unrelieved obstruction downstream - most often a hepatic IVC web or extrinsic compression by a hypertrophied caudate lobe, giving combined-type disease. Pull back a pressure trace through the hepatic IVC and treat the caval lesion, because a recanalised vein draining into a hypertensive cava will re-thrombose.
Why is recanalising a single hepatic vein usually sufficient, and what does that let you do at the planning stage?: Intrahepatic veno-venous collaterals redistribute drainage from the whole parenchyma once one outflow channel is restored, which is why a single vein sufficed in the large majority of technically successful cases in published series. It means you select the single vein with the shortest occlusion and best calibre rather than attempting multiple veins.
Your patient desaturates during balloon inflation. What is happening and what is your immediate sequence?: Thrombus dislodged from the occluded segment has embolised to the pulmonary arteries. Stop dilatation, give oxygen and support the circulation at the table, continue heparin unless bleeding, and obtain CT pulmonary angiography once stable; subsegmental emboli in an anticoagulated patient are managed conservatively.
Justify the 15-minute delayed venogram. What are you deciding on the basis of it?: It detects elastic recoil that is invisible on the immediate run. Three findings mandate a stent rather than angioplasty alone: residual stenosis over 30%, a gradient over 15 cmH2O across the segment, or persistence of intrahepatic collaterals. Skipping it produces early symptomatic re-occlusion, classically in membranous lesions.
You cannot engage the hepatic vein ostium from the jugular route. Take me through your escalation and its price.: Escalate from angled hydrophilic wire with catheter support, to stiff wire, to the Rosch-Uchida needle, to the stiff back end of the wire advanced 3 mm at a time under road-map, and finally to ultrasound-guided transhepatic puncture with snaring of the wire into the jugular sheath. The price of the transhepatic route is capsular breach: haemorrhagic complications in reported series occurred exclusively in that group, which is why dilatation is still performed from the jugular side.
This patient has occlusion of all three main hepatic veins. Is there any endovascular target left before you commit to TIPS?: Yes - an obstructed but compensatory accessory or inferior right hepatic vein of 5 mm or more, if its occluded segment is shorter than that of any main vein. Roughly 86-89% of IVC-type and combined-type patients have a patent compensatory accessory vein, so it should be looked for on cavography and cross-sectional imaging before abandoning recanalisation.
Why is stent position at the hepatic vein ostium so unforgiving?: The ostia lie within 1-2 cm of the right atrium. A stent short of the ostium leaves the obstructing web untreated; one protruding far into the cava becomes a nidus for thrombus and can compromise the other hepatic vein ostia and any future TIPS; and an undersized device at that mobile junction can migrate into the right atrium.
Your patient is symptom-free with a patent vein at 6 months but returns at 10 months with ascites. Doppler shows retrograde flow. What is your line of management?: This is re-obstruction, the dominant late event given a 5-year primary patency around 58%. Check anticoagulation adequacy and the underlying thrombophilia treatment, then proceed to venography with a view to repeat angioplasty, additional stenting or thrombolysis; secondary patency remains around 75% at 5 years with such reintervention. Convert to TIPS or DIPS if the segment cannot be salvaged.
You are working from the middle hepatic vein rather than the right. Which way do you throw the needle, and why?: Posteriorly and to the right, because the portal branches lie posterior to the middle hepatic vein, the reverse of their relationship to the right hepatic vein. Throwing anteriorly from the middle hepatic vein drives the needle out through the anterior liver surface into the peritoneum.
Your post-TIPS gradient is 6 mmHg in a 68-year-old with refractory ascites and a past episode of encephalopathy. Are you satisfied?: No — this is over-decompression for an ascites indication in a high-risk patient. The target is a 20-50% reduction from baseline, not the sub-12 mmHg target used for variceal bleeding, because excessive portal diversion in an elderly patient with prior encephalopathy predicts disabling encephalopathy and hepatic insufficiency.
During a needle pass you aspirate bile. What do you do, and what are you now worried about downstream?: Abandon that tract and select a different puncture plane rather than dilating it. A biliary communication within a stented tract creates a biliary-shunt fistula, causing haemobilia and early shunt thrombosis because bile is strongly thrombogenic.
Why is the stent-graft covered, and why is the portal 2 cm left bare?: The PTFE covering excludes the parenchymal tract from bile ingress and from pseudointimal hyperplasia, which is what raised 1-year primary patency to around 80-90% compared with bare stents. The uncovered portal segment anchors the device in the portal vein without covering and occluding contralateral portal branches.
A patient with refractory ascites has severe tricuspid regurgitation on echocardiography. Defend your decision.: I would not proceed. The shunt abruptly increases venous return and right-sided preload, and a right heart that cannot accommodate it decompensates into pulmonary oedema and low-output failure. Severe right heart disease is an absolute contraindication rather than a caution.
Six months after TIPS, ascites recurs but Doppler ultrasound reports a patent shunt. Next step?: Proceed to shunt venography with direct portosystemic gradient measurement. Doppler is a screening test that misses stenoses within covered stents, and clinical recurrence of the index complication is itself an indication for venography irrespective of the ultrasound result.
Which patients with acute variceal bleeding get preemptive TIPS rather than TIPS after endoscopic failure?: High-risk bleeders — Child-Pugh C below 14 points, or Child-Pugh B with active bleeding at endoscopy — treated within 72 hours, ideally 24 hours. In this group early shunting improves both bleeding control and survival compared with waiting for endoscopic and pharmacological therapy to fail.
Your patient has a large spontaneous splenorenal shunt on CT. How does that change your plan?: It competes with the TIPS for portal flow, so the shunt may fail to decompress varices adequately and the patient is at higher risk of refractory encephalopathy. I would plan for possible concurrent or subsequent embolisation of the splenorenal shunt and counsel the patient specifically about encephalopathy.

Portal Hypertension Viva Questions

More viva questions