Dual Portal Hypertension in Hepatic Tuberculosis: A Case Report Open Access

Tuberculosis (TB) affects pulmonary and extrapulmonary locations [1], but hepatic TB is a rare form of extrapulmonary TB. Moreover, hepatic TB presenting with portal hypertension is even rarer. The liver can be affected with TB mainly in three ways: the miliary form, accounting for 50–80% of cases, granulomatous hepatitis, and the localized form, which is the rarest (<1%) and includes tuberculomas, nodules, or biliary tract disease [2]. Hepatic granulomas are typically found near the portal tract, with most patients being minimally symptomatic or asymptomatic [3]. To date, only eight cases of portal hypertension due to TB have been reported. The low oxygen tension in the liver makes it an unfavorable environment for mycobacterial proliferation [3]. This case is unique as it describes dual portal hypertension, involving both sinusoidal intrahepatic portal hypertension and an extrahepatic component due to tubercular periportal lymphadenopathy.

A 64-year-old woman presented to the Accident and Emergency department with a 2-month history of weight loss and fatigue, along with recent-onset hematemesis and melena. Laboratory evaluation pointed toward chronic liver disease: total bilirubin – 2.5 mg/dL, aspartate aminotransferase – 123 U/L, alanine transaminase – 136 U/L, alkaline phosphatase – 150 U/L, gamma-glutamyl transferase – 68 U/L, low albumin of 2 g/dL, platelet count of 120,000/µL, and hemoglobin (Hb) of 7.5 mg/dL. She was resuscitated with intravenous fluids and received one unit of packed red blood cells before undergoing an upper gastrointestinal endoscopy. grade 2 esophageal varices with active oozing were identified and endoscopic variceal band ligation was performed. Abdominal ultrasound was done and the findings were suggestive of cirrhosis to some extent in terms of surface nodularity, but the ultrasound did not document volume redistribution changes, nor did it pick up any lymph nodes. Given the possibility of chronic liver disease with portal hypertension, an etiological workup was performed; autoimmune hepatitis serology, hepatitis B and hepatitis C, workup for Wilson’s disease and hemochromatosis were negative. Liver stiffness measurement revealed a value of 14 kPa suggestive of F4 fibrosis. A working diagnosis of cryptogenic chronic liver disease with portal hypertension and variceal bleeding was arrived at. Four months later, she developed progressive dyspnea. High-resolution computerized tomography chest showed multiple lymph nodes in the axilla and mediastinum with features suggestive of bronchiectasis. Contrast enhanced CT revealed hepatomegaly with intra-abdominal lymphadenopathy in the porta hepatis and perisplenic collaterals (shown in Fig. 1). The liver surface had subtle nodularity, but there were no significant volume redistribution changes, suggesting an alternative cause of portal hypertension other than cirrhosis. These CT findings were not typical of cirrhosis of the liver. Due to the risk of vascular injury with a retro-portal venous node biopsy, we opted for laparoscopic lymph node sampling and liver biopsy. Moreover, excisional biopsy was preferred to better demonstrate caseating necrosis and to rule out lymphoma. During laparoscopy, the liver was soft on probing, which was not typical of cirrhosis. Biopsies obtained from axillary and periportal nodes revealed necrotizing granulomatous lymphadenitis. Liver biopsy was suggestive of chronic granulomatous hepatitis with portal to central bridging fibrosis (shown in Fig. 2, 3). TB interferon-gamma assay was positive and TB polymerase chain reaction from the axillary node turned out to be positive. Thus, a diagnosis of chronic granulomatous hepatitis with extrahepatic periportal venous nodal compression causing variceal bleeding was made. The hepatic venous pressure gradient (HVPG) measured via the femoral route, was 7 mm Hg, indicating sinusoidal portal hypertension. Given that clinically significant portal hypertension is defined as HVPG of more than 10 mm Hg, variceal formation was primarily attributed to extrahepatic lymph node compression of the portal vein along with a smaller contribution from the sinusoidal component. Based on the final diagnosis of abdominal TB along with chest TB, antitubercular therapy was initiated according to National Tuberculosis Elimination Programme (NTEP) guidelines: a 2-month intensive phase (isoniazid, rifampicin, pyrazinamide, and ethambutol) followed by a 4-month continuation phase (isoniazid and rifampicin). While on follow-up the patient showed significant clinical improvement with weight gain of 10 kg and improvement in serum albumin from 2 g/dL to 3.4 g/dL. At 6 months, repeat imaging demonstrated the resolution of lymphadenopathy, and a follow-up esophagoscopy revealed only minimal residual varices.

Hepatic TB remains a diagnostic challenge due to its nonspecific presentation and ability to mimic other liver pathologies; hence, hepatic TB is a challenge even in South Asian and Eastern European regions where TB is prevalent. Clinically as well as radiologically, the disease is considered a masquerader [1]; there are documented cases where hepatic TB was misdiagnosed as hepatocellular carcinoma (HCC) because it was fluorodeoxyglucose (FDG)-avid during an F-18 FDG positron emission tomography/CT [3]. There are reported cases of hepatic TB causing variceal bleeding due to extrahepatic portal venous obstruction and portal vein thrombosis [4‒6]. However, dual portal hypertension, involving both sinusoidal and prehepatic components due to hepatic TB with hilar lymph node involvement is a rare phenomenon. HVPG is the best method for categorizing the type of portal hypertension and quantifying the severity of sinusoidal portal hypertension. When HVPG is more than 10 mm Hg, it is clinically significant portal hypertension and results in variceal bleeding and other decompensation events [7]. In our case, the HVPG of 7 mm Hg suggested a predominant extrahepatic etiology, emphasizing the importance of considering alternative causes of variceal bleeding beyond cirrhosis. This case calls for a better understanding of manifestations of hepatic TB, particularly its potential to present with portal hypertension and variceal bleeding [8, 9]. As per the literature review, the combination of hepatic parenchymal TB with periportal lymphadenitis causing dual portal hypertension and consequent variceal bleeding has not been previously reported. Not all cases of variceal bleeding are due to cirrhosis and a high index of suspicion is needed to diagnose hepatic TB.

This study was approved by the Ethics Committee of Apollo Adlux Hospital, Ernakulam, Kerala, India (Ethical Approval reference No.: EC/NEW/AAH/2022/2956). Written informed consent was obtained from the patient for publication of the case details and accompanying images.

The authors have completed the CARE checklist to ensure comprehensive and transparent reporting of this case report. The completed checklist has been provided as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000545522).

The authors have no conflicts of interest to declare.

This study was not supported by any sponsor or funder.

Dr. Amal Joseph: draft preparation and literature collection; Dr. Prajob Geevarghese Prasad: submission draft preparation; Dr. Jeby Jacob: literature review and manuscript refinement; and Dr. Harikumar Nair: supervision and manuscript editing.

The data supporting the findings of this study are available from the corresponding author upon reasonable request.