An Extremely Rare Cause of Hyperammonemic Encephalopathy

Sir, Hyperammonemic encephalopathy continues to be an important treatable cause of the cognitive decline. Most of the cases with hyperammonemic encephalopathy are secondary to either drugs or liver dysfunction. Herein, we report a case of hyperammonemic encephalopathy without any evidence of liver dysfunction or drugs. Only after extensive evaluation, the cause was found to be portosystemic shunting secondary to extensive collateral formation induced by a gastrointestinal stromal tumor. A 56-year-old man was evaluated 3 years back for Melena and was diagnosed as a case of gastrointestinal stromal tumor (GIST) based on contrast-enhanced CT (computer tomography) scan of the abdomen. There was no evidence of metastases at that time. He underwent resection of the tumor. Postoperative histology revealed GIST with CD117 positivity. He was started on imatinib mesylate 400 mg OD. However, the tumor recurred within 2 years with whole-body PET (positron emission tomography) scan showing diffuse involvement of mesentery of small and large bowel [Figure 1a]. Imatinib was increased to 400 mg BD but had to be decreased to 400 mg OD as he developed a skin rash. Surgery was deferred due to poor performance status.Figure 1: Hyperammonemic encephalopathy due to the gastrointestinal stromal tumor. (a) Whole-body FDG PET showing FDG avid lesions suggestive of recurrence of tumor involving the mesentery of small and large intestine; (b) T1-weighted axial magnetic resonance imaging section shows T1 hyperintensities in bilateral GPi (globus pallidus interna) (blue arrows); (c) Contrast-enhanced computed tomographic scan of abdomen showing multiple intrahepatic collaterals between portal and hepatic veins (red arrow)He was presented to us with 3 months history of cognitive decline, difficulty in walking, and urinary incontinence. There was no history of fever, headache, seizures, limb weakness, sensory symptoms, jaundice, or Melena. On examination, he had poor attention, apathy and recent memory deficits, and marked symmetrical bradykinesia and rigidity as well as prominent asterixis. Due to the presence of asterixis, a possibility of metastatic liver disease with hepatic encephalopathy was considered and he was evaluated. Magnetic resonance imaging (MRI) scan of the brain [Figure 1b] revealed hyperintense signal changes in bilateral globus pallidus suggestive of hyperammonemic encephalopathy. Routine hematological (hemoglobin, total and differential leukocyte counts, platelet counts, erythrocyte sedimentation rate) and biochemical parameters (blood sugars, liver, renal and thyroid function tests, serum electrolytes, calcium and phosphorus) were normal. Arterial blood gases and toxic screen analysis were noncontributory. Arterial ammonia was elevated to 116 μmol/L (normal: 15.8–52.8 μmol/L); however, ultrasonography and fibrous scanning of liver were normal. Screening for autoimmune markers and hepatotropic viruses was negative. Finally, a possibility of hyperammonemic encephalopathy due to spontaneous portosystemic shunting secondary to GIST was considered. He underwent CT and MRI scan of the abdomen along with MR venography, which revealed a dilated portal vein (17 mm) with collaterals between hepatic and portal vein [Figure 1c] as well as collaterals in the perisplenic, perigastric, and periportal location, confirming the diagnosis of portosystemic hyperammonemic encephalopathy. He was started on antihepatic coma regimen including rifaximin 550 mg BD and lactulose following which he had complete resolution of symptoms. Follow-up arterial ammonia decreased to 40.1 μmol/L. He continues to do well on antihepatic coma measures till 1-year follow-up when he was lost to follow-up. Definitive treatment like embolization of portosystemic collaterals was deferred due to presence of widespread disease and poor performance status. Ammonia is a neurotoxic metabolic by-product of the urea cycle pathway. The causes for hyperammonemia are multiple including liver failure, portosystemic shunts, drugs like valproic acid and congenital urea cycle disorders like ornithine decarboxylase deficiency.[1,2] Noncirrhotic portosystemic shunting is a rare disease with most of the cases being iatrogenic secondary to the management of portal hypertension or rarely congenital portosystemic shunts. The resultant hyperammoninemia causes varied neurological manifestations including encephalopathy, dysarthria, ataxia, or extrapyramidal syndrome which were all noted in our index patient. The pathogenesis of neurological disorder is related to conversion of excess ammonia into glutamine by the astrocyte cells of brain.[2] GIST causing portosystemic shunting has previously been described in a single patient of metastatic GIST who underwent surgical debulking with regression of symptoms.[3] The absence of liver disease in our patient with classical signs and symptoms of hyperammonemic encephalopathy and characteristic MRI finding enforced us to probe further etiology of hyperammonemia in our patient. The pathogenesis of hyperammonemia in GIST is two-fold. Firstly, primary tumor burden may itself cause increased production of ammonia and in fact, transient hyperammonemia has been described with tumors such as multiple myeloma, lymphoma, and malignancies with high mitotic index because of high catabolism and cytoreductive chemotherapy.[4-6] Secondly, GIST may itself induce portosystemic shunting with resultant hyperammonemia. A mesenchymal tumor like GIST expresses significant levels of angiogenic factors such as VEGF (vascular endothelial growth factor), MCSF (macrophage colony-stimulating factor), and IL-2 (Interleukin 2) which can cause increased angiogenesis and localized increased portosystemic shunting.[7,8] Thus, the resultant portosystemic shunting can cause hyperammonemic encephalopathy as demonstrated in our case. Another important cause of hyperammonemic encephalopathy in patients with widespread tumors continues to be drugs, and indeed cases have been described in which treatment for GIST with sunitinib has been associated with hyperammonemia with reversal following withdrawal of the offending agent. However, no such association has been described with imatinib mesylate.[9] To conclude, our case highlights the need for keeping a possibility and evaluating the patient in detail till the diagnostic doubts are sorted out. Our case further highlights GIST as a potential cause of hyperammonemic encephalopathy in the absence of liver disease. Declaration of patient consent The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.