Portal Hypertension Viva

Long Case Viva Defence

Candidate's opening statement (model answer)

"Mr Thompson is a 56-year-old painter who presents with acute haematemesis and melaena. He has a 15-year history of heavy alcohol consumption (approximately 80 grams per day) and was diagnosed with alcohol-related cirrhosis 2 years ago. He has known large oesophageal varices (last OGD 8 months ago, treated with carvedilol 12.5 mg daily for primary prophylaxis). [1]

His past history includes:

• Alcohol-related cirrhosis — Child-Pugh B (score 9), MELD-Na 17
• Previous variceal surveillance — large varices with red wale signs
• Mild ascites managed with spironolactone 100 mg daily
• No prior hepatic encephalopathy or hepatorenal syndrome
• No history of spontaneous bacterial peritonitis [1]

On arrival, his blood pressure was 88/55 with heart rate 112, consistent with hypovolaemic shock from acute variceal haemorrhage. His haemoglobin was 62 g/L. Urgent endoscopy confirmed active bleeding from a large distal oesophageal varix, treated with band ligation. [1]

His main problems are:

1. Acute variceal bleeding — now controlled, high-risk bleeder (Child-Pugh B with active bleeding at endoscopy)
2. Risk of early rebleeding and treatment failure
3. Alcohol-related cirrhosis, Child-Pugh B, MELD-Na 17 — transplant candidate if abstinent
4. Rising creatinine (90 to 145) — at risk of hepatorenal syndrome-AKI
5. SBP risk requiring antibiotic prophylaxis
6. Alcohol withdrawal risk
7. Malnutrition and sarcopenia [1]

My integrated plan is to secure haemostasis, assess for pre-emptive TIPS eligibility, commence vasoactive drugs and prophylactic antibiotics, monitor renal function closely for HRS-AKI, provide alcohol withdrawal prophylaxis with thiamine, and initiate liver transplant assessment." [1]

Examiner probing questions and model answers

Q1: "What is your immediate priority in managing this acute variceal bleed?" [1]

"Resuscitation is the first priority. I would secure the airway — intubating if he is encephalopathic or actively vomiting to protect against aspiration. I would establish two large-bore IV cannulae, cross-match blood, and commence a restrictive transfusion strategy targeting haemoglobin 70 to 80 g/L — over-transfusion raises portal pressure and worsens rebleeding. Simultaneously, I would start a vasoactive agent (terlipressin 2 mg IV every 4 hours, or octreotide 50 mcg bolus then 50 mcg/hour infusion) and prophylactic ceftriaxone 1 g IV daily. These should be commenced before endoscopy. Endoscopy within 12 hours for band ligation. I would not delay endoscopy for correction of INR — this is rarely fully achievable in cirrhosis and time is critical." [1]

Q2: "He is a Child-Pugh B with active bleeding at endoscopy. Does that change your management?" [1]

"Yes. This identifies him as a high-risk bleeder eligible for early pre-emptive TIPS. The Garcia-Pagan trial showed that patients with Child-Pugh C (10 to 13) or Child-Pugh B with active bleeding at endoscopy benefit from pre-emptive TIPS within 72 hours of admission — reducing treatment failure from 50 percent to 3 percent and improving one-year survival from 61 percent to 86 percent compared to standard therapy. I would urgently discuss this with the interventional radiology and hepatology teams. If he is not eligible or TIPS is declined, I would maintain vasoactive therapy for 2 to 5 days, ensure serial EVL for variceal eradication, and commence secondary prophylaxis with NSBB plus serial EVL once stable." [1]

Q3: "His creatinine rises to 190 micromol/L at 48 hours despite fluid resuscitation. How do you approach this?" [1]

"This meets ICA-AKI criteria — a rise from 90 to 190 is well above the 26.5 micromol/L threshold and more than doubles his baseline. My first step is the HRS-AKI diagnostic workup: stop all diuretics, give albumin 1 g/kg/day for 2 consecutive days, and simultaneously exclude alternative causes — check urinalysis for proteinuria and haematuria to exclude structural kidney disease, review medications for nephrotoxins (NSAIDs, aminoglycosides, contrast), and exclude shock. If there is no response to the albumin challenge and other criteria are met, I diagnose HRS-AKI and commence terlipressin 1 to 2 mg IV every 4 to 6 hours with albumin 20 to 40 g/day. I would monitor oxygen saturation and fluid balance closely — the CONFIRM trial showed terlipressin carries an 11 percent risk of respiratory failure compared to 2 percent with placebo. If terlipressin is contraindicated (hypoxia, volume overload, ischaemia), I would use noradrenaline in ICU. I would also check for SBP — 50 percent of HRS is triggered by SBP — with a diagnostic paracentesis." [1]

Q4: "Why ceftriaxone rather than norfloxacin for antibiotic prophylaxis?" [1]

"Ceftriaxone is superior to norfloxacin in high-risk cirrhotic patients with GI bleeding. The Fernandez trial demonstrated that ceftriaxone 1 g IV daily reduces proven bacterial infections compared to norfloxacin, largely due to the rising incidence of quinolone-resistant organisms. Ceftriaxone is preferred in patients with Child-Pugh C, prior quinolone prophylaxis, hospital-acquired bleeding, or in regions with high quinolone resistance — which includes most Australian hospitals. Norfloxacin 400 mg orally BID remains acceptable for low-risk patients. Antibiotic prophylaxis is not optional — it is one of the few interventions that reduces mortality (not just infection) in cirrhosis with GI bleeding, reducing mortality by approximately 10 percent." [1]

Q5: "He asks about his prognosis. How do you counsel him?" [1]

"I would be honest but measured. His cirrhosis is now decompensated — the variceal bleed is a decompensation event, and his one-year mortality without further intervention is 20 to 40 percent, depending on whether he abstains from alcohol and whether complications develop. His MELD-Na of 17 gives him a moderate estimated 90-day mortality. The best way to improve his prognosis is to achieve and maintain alcohol abstinence — this is the single most important prognostic factor in alcohol-related cirrhosis. Baveno VII describes the concept of recompensation, where removing the etiological cause can partially reverse portal hypertension. I would also initiate liver transplant assessment, as transplant is the definitive treatment for decompensated cirrhosis. For transplant candidacy, most centres require at least 6 months of documented abstinence and psychosocial support. I would involve the alcohol liaison team early and discuss advance care planning." [1]

Q6: "Six months later, he returns with refractory ascites despite maximal diuretics. Would you recommend TIPS?" [1]

"I would consider TIPS as a reasonable option but discuss the risks and benefits carefully. TIPS is most beneficial in patients with MELD below 18, no current or severe hepatic encephalopathy, and no significant cardiac or pulmonary disease. His MELD-Na of 17 is within the acceptable range. TIPS would reduce his need for paracentesis and may improve nutritional status. However, TIPS carries a 20 to 40 percent risk of new or worsened hepatic encephalopathy — I would assess his baseline cognitive function and prior HE history. I would also perform an echocardiogram to screen for cardiac dysfunction or pulmonary hypertension (mPAP above 45 is an absolute contraindication to TIPS). If he has no contraindications and understands the HE risk, I would refer for TIPS. Concurrently, I would escalate his transplant assessment — refractory ascites is itself a transplant indication, and TIPS is a bridge, not a cure. I would also address nutrition, as refractory ascites and frequent paracentesis accelerate protein loss and sarcopenia." [1]

Short Case Discussion

Scenario: "Examine this patient's abdominal system. You have 7 minutes for examination and 8 minutes for discussion."

Key signs the patient demonstrates (cirrhosis with portal hypertension)

• General: cachectic, spider naevi on chest and shoulders, palmar erythema, parotid enlargement, gynaecomastia, loss of axillary and pubic hair (in males), Dupuytren contracture
• Hands: asterixis (if encephalopathic), leukonychia, clubbing (in advanced disease)
• Abdomen — inspection: distended abdomen with everted umbilicus (ascites), visible dilated periumbilical veins radiating from the umbilicus (caput medusae — pathognomonic of portal hypertension via the recanalised umbilical vein), surgical scars from previous paracentesis
• Abdomen — palpation: firm, non-tender. Liver edge may be impalpable (shrunken cirrhotic liver) or enlarged and hard with a nodular surface. Splenomegaly is the key finding of portal hypertension — a palpable spleen in cirrhosis indicates portal hypertension (the spleen should not be palpable in health)
• Abdomen — percussion: shifting dullness and a fluid thrill confirm ascites. Percuss for the liver span — may be reduced in advanced cirrhosis
• Abdomen — auscultation: a Cruveilhier-Baumgarten murmur — a venous hum heard between the xiphisternum and umbilicus, produced by turbulent flow through the recanalised umbilical vein (portosystemic collateral)
• Other: examine for peripheral oedema (hypoalbuminaemia), test for asterixis (hepatic encephalopathy), assess conscious state (West Haven grade) [1]

Presentation template (model answer)

"I examined Mr Thompson's abdominal system. He is cachectic with spider naevi on the anterior chest wall, palmar erythema, and bilateral parotid enlargement. He has a fine tremor consistent with asterixis. [1]

The abdomen is distended with an everted umbilicus. There are visible dilated periumbilical veins radiating from the umbilicus in a caput medusae distribution. The abdomen is soft and non-tender. The liver edge is not palpable but percussion confirms a reduced liver span of approximately 8 cm. The spleen is palpable 3 cm below the left costal margin. There is shifting dullness and a fluid thrill, confirming ascites. On auscultation, bowel sounds are present and there is a faint venous hum in the epigastrium consistent with a Cruveilhier-Baumgarten murmur. [1]

There is bilateral pitting oedema to the mid-shin. There is no evidence of hepatic fetor. [1]

In summary, these findings are consistent with decompensated chronic liver disease with portal hypertension — evidenced by splenomegaly, caput medusae, ascites, and the Cruveilhier-Baumgarten murmur — complicated by hepatic encephalopathy." [1]

Discussion questions

Examiner: "What is the significance of splenomegaly in this patient?" [1]

"Splenomegaly in the context of cirrhosis is a direct consequence of portal hypertension — the spleen enlarges because portal venous congestion increases splenic venous pressure. A palpable spleen in a patient with cirrhosis is one of the most reliable clinical indicators of portal hypertension, and it correlates with the presence of clinically significant portal hypertension (HVPG at or above 10 mmHg). Hypersplenism may also develop, causing thrombocytopenia and leucopenia from splenic sequestration. This thrombocytopenia can make the platelet count a useful non-invasive marker — Baveno VII uses platelet count above 150 to help rule out varices needing treatment." [1]

Examiner: "Explain the pathophysiology of caput medusae." [1]

"Caput medusae represents the recanalisation of the umbilical vein, which in fetal life carries oxygenated blood from the placenta through the umbilicus to the portal venous system. After birth, the umbilical vein normally obliterates and becomes the ligamentum teres. In portal hypertension, the pressure gradient between the portal system and the systemic venous system causes the umbilical vein to recanalise, creating a portosystemic collateral through the periumbilical veins (the paraumbilical veins of Sappey). The dilated, tortuous periumbilical veins radiating from the umbilicus resemble the snakes on Medusa's head — hence the name. The Cruveilhier-Baumgarten murmur is the auscultatory correlate of turbulent flow through these dilated veins." [1]

Examiner: "How would you confirm the presence of portal hypertension non-invasively?" [1]

"The gold standard is hepatic venous pressure gradient (HVPG) measurement via hepatic venous catheterisation — a HVPG greater than 5 mmHg confirms portal hypertension, at or above 10 mmHg defines clinically significant portal hypertension, and above 12 mmHg is the threshold for variceal bleeding. However, HVPG is invasive and not routinely performed outside specialist centres. The non-invasive approach per Baveno VII uses liver stiffness measurement (LSM) by transient elastography: LSM at or above 25 kPa rules in CSPH with high probability, while LSM below 15 kPa combined with a platelet count above 150 rules out varices needing treatment. Ultrasound Doppler can demonstrate portal vein patency and flow direction, splenomegaly, ascites, and portosystemic collaterals. Endoscopy directly visualises varices." [1]

Examiner: "What is the window hypothesis regarding non-selective beta-blockers?" [1]

"The window hypothesis describes the concept that NSBB are beneficial within a therapeutic window in cirrhosis but become harmful when the patient is too sick. In compensated cirrhosis and early decompensation, NSBB reduce portal pressure and prevent first decompensation, variceal bleeding, and ascites — the PREDESCI trial confirmed this benefit. However, in advanced decompensation — specifically refractory ascites or HRS-AKI with systolic blood pressure below 90 mmHg — NSBB may worsen renal perfusion by reducing cardiac output in a patient who is already vasodilated and hypotensive. The NSBB-induced reduction in mean arterial pressure can precipitate or worsen renal failure. In these situations, the NSBB dose should be reduced or ceased. This is a nuanced area — NSBB should not be stopped in every decompensated patient, but careful attention to blood pressure and renal function is essential." [1]