Answers are grounded in this paper and cite their sources.
Chronic arsenic poisoning is a rare but established cause of non-cirrhotic portal hypertension, first described in India in 1979 and characterized by prominent portosystemic collaterals and perisinusoidal fibrosis [1]. While long-term exposure is classically linked to arsenic-contaminated drinking water, non-water sources persist, including traditional medicines and occupational/environmental arsenicals, underscoring the ongoing global relevance of arsenic-related disease [2, 3]. We reported a patient in whom marked neurocutaneous toxicity and cachexia served as the dominant clues to arsenic-associated portal hypertension. A 45-year-old woman was admitted to the Hepatology Department with 3 months of progressive abdominal distension, bilateral lower-extremity edema, and periorbital swelling, along with 6 months of symmetric numbness and weakness in all four limbs, 1 year of bilateral eyelid ptosis without diurnal fluctuation, generalized skin hyperpigmentation, and long-standing irregular menses with intermittent amenorrhea. She denied tobacco, alcohol, or illicit drug use. There was no known history of liver disease and no family history of hepatic or neurologic disorders. On admission, she was profoundly cachectic (body mass index 11.3 kg/m2) and had diffuse mottled hyperpigmentation of the skin and gingiva, bilateral ptosis, symmetric muscle wasting with weakness in all limbs, glove-and-stocking sensory loss, and suspected intrinsic hand muscle atrophy (Fig. 1). Oral examination revealed the loss of six teeth and generalized loosening of the remaining dentition. The abdomen was distended with shifting dullness but without hepatosplenomegaly or stigmata of cirrhosis. Mild bilateral pitting edema was present. Multisystem manifestations of chronic arsenic toxicity. (A) Facial appearance demonstrating profound cachexia and diffuse hyperpigmentation (eyes masked). (B) Truncal skin showing diffuse mottled (“raindrop-like”) hyperpigmentation—a typical cutaneous feature of chronic arsenicosis. (C) Abdominal CT demonstrating massive ascites with compression and displacement of intra-abdominal organs. (D) Hands showing chronic arsenicosis-related cutaneous changes, including diffuse hyperpigmentation; palmar/plantar hyperkeratosis (arsenical keratosis) was present clinically. (E) Lower extremities demonstrating marked muscle wasting. (F) Upper endoscopy demonstrating esophageal varices without red wale signs or stigmata of recent bleeding. Laboratory evaluation demonstrated microcytic anemia (hemoglobin 95 g/L), hypoalbuminemia (3.33 g/dL), and low ceruloplasmin (13.3 mg/dL). Liver enzymes, bilirubin, renal function, thyroid function, and cortisol levels were normal. Autoimmune, viral, and metabolic serologies, as well as serum and urine immunofixation electrophoresis, were negative. Contrast-enhanced abdominal and pelvic computed tomography (CT) demonstrated a large volume of intraperitoneal fluid with the compression and displacement of multiple abdominal organs (Fig. 1). There was diffuse bowel wall thickening and edema with segmental luminal dilation, but no significant lymphadenopathy was observed in the abdominal or retroperitoneal cavities. Biochemical analysis of pleural and peritoneal fluid was consistent with a transudate and cytology was negative for malignant cells. Ascitic fluid analysis supported portal-hypertensive ascites (ascitic albumin 1.54 g/dL; serum-ascites albumin gradient 1.79 g/dL). Ultrasound elastography demonstrated a liver stiffness of 11.0 kPa and a spleen stiffness of 43.1 kPa. Upper endoscopy showed esophageal varices. Regarding the elevated D-dimer, at an outside-hospital D-dimer was 1,099 ng/mL and CT pulmonary angiography (CTPA) was reported as segmental pulmonary emboli in the right upper and right lower lobes; however, she had no chest pain, dyspnea, or hypoxemia, and anticoagulation was not initiated. Three months later at our institution, the D-dimer remained elevated (6.42 µg/mL, local assay) yet repeat CTPA showed no definite pulmonary embolism and bilateral lower-extremity Doppler ultrasound was negative for deep vein thrombosis. Nerve-conduction studies revealed severe mixed axonal and demyelinating sensorimotor polyneuropathy. Cervical spine magnetic resonance imaging showed multilevel disc protrusions and abnormal cord signal below C3. The combination of portal hypertension, progressive sensorimotor polyneuropathy, and mucocutaneous hyperpigmentation suggested a systemic disorder rather than isolated liver disease. The differential diagnosis included polyneuropathy, organomegaly, endocrinopathy, myeloma protein, and skin changes (POEMS) syndrome, infiltrative or malignant disorders, endocrinopathies, and nutritional or toxic etiologies. POEMS became unlikely because repeated serum and urine immunofixation showed no monoclonal gammopathy, imaging revealed no osteosclerotic lesions, and CT did not demonstrate convincing organomegaly or an occult plasma-cell disorder. Wilson disease was considered given the reduced ceruloplasmin; however, Kayser–Fleischer rings were absent and basal 24-hour urinary copper excretion was normal, making Wilson disease unlikely. The hypoceruloplasminemia was therefore considered secondary to severe malnutrition [4]. As routine evaluations remained unrevealing, a repeated nonjudgmental review of non-prescription products identified the long-term intermittent use of Niuhuang Jiedu Pian—an over-the-counter traditional Chinese medicine containing realgar (arsenic disulfide) that the patient had intentionally used for weight control since 2019; she described a recurring pattern of 3 days of fasting followed by 1 “purgation day” during which she ingested ∼30 tablets, far exceeding the labeled dose. She reported no other traditional Chinese medicines, herbal products, or dietary supplements during the relevant period. Based on the reported regimen and the labeled composition, the estimated arsenic exposure on each dosing day approached 1.05 g. Even under a conservative assumption of frequency, the cumulative elemental arsenic exposure over 5 years would have been substantial. Targeted toxicology testing obtained ∼3 months after cessation showed a mildly elevated spot urine arsenic level (0.188 µg/L; laboratory ULN < 0.1 µg/L) and a blood arsenic level within the laboratory reference range (16.5 µg/L; reference 2.5–190 µg/L). Hair and nail arsenic measurements were not available. Given the delay after cessation and the intermittent nature of the exposure, these results may underestimate the prior arsenic burden; therefore, the exposure history and classic neurocutaneous manifestations were central to the diagnosis. Taken together, the multisystem phenotype and exposure history were most consistent with chronic arsenic toxicity, with arsenic-associated non-cirrhotic portal hypertension (porto-sinusoidal vascular disease) as the most plausible explanation for the varices and ascites. In addition, repeat CTPA showed no definite pulmonary embolism and lower-extremity Doppler ultrasound was negative; in the absence of cardiopulmonary symptoms, ongoing venous thromboembolism was considered unlikely. Persistent D-dimer elevation was interpreted as nonspecific in portal-hypertensive ascites rather than attributable to arsenic exposure [5]. Management focused on permanent discontinuation of the offending agent, high-calorie nutritional rehabilitation, diuretics for ascites, and neurotrophic support (vitamins B1, B12, folate). Chelation therapy was withheld, given the limited benefit in chronic, intermittent exposure with stable symptoms. Non-selective beta-blocker prophylaxis was deferred because of the poor-tolerability risk in the setting of profound cachexia and borderline hypotension (blood pressure ∼100/60 mmHg); she also had no prior variceal bleeding and endoscopy showed varices without red wale signs. Reassessment for prophylaxis was planned after nutritional and hemodynamic improvement. At the 3-month follow-up, the ascites and edema had markedly improved, weight had increased by 5 kg, and ptosis and neuropathy had partially recovered. The patient remains under long-term surveillance for potential late effects, including cutaneous malignancies, cardiovascular disease, and the progression of portal hypertension. This case emphasizes that portal hypertension accompanied by neurocutaneous and neurologic clues should prompt evaluation for systemic toxic exposures rather than isolated liver disease. Pathophysiologically, chronic inorganic arsenic exposure can cause endothelial injury and microvascular remodeling in the portal venous system (obliterative portal venopathy/porto-sinusoidal vascular disease spectrum), increasing intrahepatic vascular resistance and leading to non-cirrhotic portal hypertension [6, 7]. In parallel, arsenic exerts systemic toxicity through sulfhydryl binding, mitochondrial dysfunction, and oxidative stress, with the preferential involvement of keratin-rich tissues (hyperpigmentation/hyperkeratotic changes) and peripheral nerves (distal symmetric polyneuropathy), which may secondarily contribute to muscle wasting and functional decline [8, 9]. Clinically, the diagnostic value of this case lies in the exposure history: a widely accessible over-the-counter traditional medicine containing realgar was self-administered in a markedly exaggerated regimen for weight control. Repeated, nonjudgmental medication/supplement reconciliation was pivotal in identifying the causative exposure. Management centers on permanent discontinuation of the exposure source and standard care for portal-hypertension complications, together with nutritional rehabilitation and supportive management of neurotoxicity. Overall, this case underscores that arsenic poisoning remains an important and potentially overlooked cause of non-cirrhotic portal hypertension, and that neurocutaneous manifestations can be the most salient clues to guiding timely diagnosis and appropriate management. Conceptualization: Y.D., G.Z., W.D., X.Z.; data curation: Y.D., G.Z.; writing original draft: Y.D., G.Z., W.D.; writing review and editing: X.Z.; final approval: Y.D., G.Z., X.Z. The study was funded by the National Natural Science Foundation of China [No.82500714]. None declared
No takes yet. Share an insight, caveat, or question.
You et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: