Index Hepatic Decompensation: Hepatic Hydrothorax Without Signs of Clinically Significant Portal Hypertension Diagnosed With Radioisotope Scintigraphy

ABSTRACT

Hepatic hydrothorax (HH) occurs in 4%–12% of patients with cirrhosis and rarely presents without accompanying evidence of clinically significant portal hypertension (CSPH). We report the case of a 65-year-old man with cirrhosis without prior decompensation, congestive heart failure, and recurrent right-sided pleural effusion. CSPH was not otherwise observed despite thorough laboratory, radiologic, and endoscopic evaluation. However, pleural fluid analysis revealed a serum effusion albumin gradient of 1.6, raising suspicion for a hepatic etiology. Suspected HH was diagnosed by technetium-99m sulfur colloid peritoneal cavity scintigraphy. As the index decompensating event, the patient's HH initiated a liver transplant evaluation in the absence of other evidence of CSPH.

INTRODUCTION

Hepatic hydrothorax (HH) is a transudative, usually right-sided pleural effusion not caused by primary cardiopulmonary disease, occurring in an estimated 4%–12% of patients with cirrhosis.^1–4 Typically, HH occurs when clinically significant portal hypertension (CSPH) causes ascitic fluid to migrate into the pleural space through diaphragmatic defects. Although ascites is usually present, up to 9% of HH cases may occur in its absence.⁴

HH diagnosis relies on pleural fluid studies demonstrating a serum effusion albumin gradient (SEAG) ≥1.1 and exclusion of other etiologies. However, in cases of diagnostic uncertainty involving other potentially confounding causes of transudative effusions, such as in congestive heart failure (CHF), technetium-99m-sulfur colloid peritoneal cavity scintigraphy (Tc-99m SCPCS) can support HH diagnosis by demonstrating direct evidence of pleuroperitoneal communication by tracking the migration of intraperitoneally injected radiotracers.⁵

HH is considered a hepatic decompensation event and, due to its associated morbidity and mortality, may warrant liver transplant (LT) evaluation, with Model for End-Stage Liver Disease (MELD) score exception being conferred in select cases.⁶ Management otherwise includes sodium restriction, diuresis, therapeutic thoracentesis, transjugular intrahepatic portosystemic shunt (TIPS), or repair of diaphragmatic defects.^6,7

We present a case of index hepatic decompensation due to HH in the absence of ascites or other evidence of CSPH diagnosed through Tc-99m SCPCS in a patient with alcohol-associated cirrhosis and underlying CHF.

CASE REPORT

A 65-year-old man with a history of compensated alcohol-associated cirrhosis, CHF, coronary artery bypass grafting (CABG) 12 years prior, and stage 4 chronic kidney disease presented with acute, progressive dyspnea. Three months prior, he had required a thoracentesis for a right-sided pleural effusion and had been taking torsemide 40 mg daily since that time with limited improvement.

His examination revealed a new supplemental oxygen requirement and diminished aeration and crackles of the right hemithorax. No orthopnea, edema, abdominal tenderness or distention, or fluid wave were appreciated. Thoracic computed tomography showed a massive right-sided pleural effusion with near-complete atelectasis of the right middle and lower lobes, indicating recurrence of his prior effusion. Transthoracic echocardiography showed a left ventricular ejection fraction of 44% consistent with his known baseline and >50% collapsibility of the inferior vena cava. Laboratory studies revealed a pro-brain natriuretic peptide (BNP) level of >20,000 pg/mL (decreased from >30,000 pg/mL 3 months prior), serum albumin of 2.7 g/dL, alkaline phosphatase of 153 U/L, platelets of 227 K/μL, international normalized ratio of 1.2, and normal bilirubin and transaminase levels. His MELD 3.0 score was 22. Abdominal ultrasonography demonstrated normal portal vein dimensions and hepatopetal blood flow and no ascites or splenomegaly. Esophagogastroduodenoscopy performed 1 month prior found no evidence of varices or portal hypertensive gastropathy. Pleural fluid studies obtained 1 day after admission and cessation of diuretics yielded a transudative effusion with a SEAG of 1.6 (Table 1).

Table 1.

Pleural fluid studies from current and prior hospitalization

Pleural fluid (PF) and serum (S) analysis studies	Current hospitalization	Previous hospitalization (3 mo prior)
PF pH	7.44	7.52
PF glucose (mg/dL)	76	133
PF total protein (g/dL)	2.0	2.7
S total protein (g/dL)	6.0	6.2
PF albumin (g/dL)	1.1	1.7
S albumin (g/dL)	2.7	2.9
PF LDH (U/L)	88	125
S LDH (U/L)	184	236
PF total nucleated cell count/uL	80	74
PF neutrophils (%)	0	5
PF lymphocytes (%)	44	43
PF eosinophils (%)	0	2
PF monocytes (%)	56	48
PF mesothelial cells (%)	0	2
Serum effusion albumin gradient (SEAG)	1.6	1.2
Light's criteria	Not met	Not met

Given recurrence of the patient's effusion with limited evidence of CHF exacerbation or an alternative etiology, suspicion for index hepatic decompensation due to HH arose due to his history of cirrhosis. However, given the absence of apparent CSPH and the patient's underlying, albeit stable CHF, Tc-99m SCPCS was pursued, which demonstrated increased Tc-99m-sulfur colloid uptake along the right pleura at 1 hour and 3 hours postintraperitoneal injection, indicating the presence of peritoneopleural communication and thus HH (Figure 1).

Figure 1.

Technetium-99m-sulfur colloid peritoneal cavity scintigraphy demonstrating scattered foci of increased uptake along the chest with increase in number and intensity between 1 and 3 h post-injection in (A) transverse, (B) coronal, and (C) sagittal views.

With index decompensation confirmed, LT evaluation was pursued; however, the patient did not qualify for treatment with either TIPS or LT due to his underlying cardiovascular comorbidities. Following a therapeutic thoracentesis, the patient was discharged and instructed to return for repeat thoracenteses as needed. He was maintained on his prehospital torsemide regimen only, given his advanced kidney disease, with plans for close outpatient follow-up for further characterization of his disease and optimization of therapy; however, he was lost to follow-up after discharge.

DISCUSSION

This case demonstrates a rare presentation of HH as the only sign of CSPH diagnosed through Tc-99m SCPCS. HH has historically been associated with accompanying signs of CSPH.⁷ Although numerous case reports have been published on HH in the absence of evident ascites, all besides 1 have noted the presence of at least 1 additional CSPH sign, and none have described HH in a patient with concurrent cardiopulmonary disease.^7–16 Although not routinely used, Tc-99m SCPCS is indicated for confirming HH when diagnostic uncertainty exists. It is considered highly specific for HH due to direct visualization of transdiaphragmatic movement of radiotracer-labeled ascitic fluid. Relative contraindications to Tc-99m SCPCS include recent abdominal surgery, active peritonitis, or tracer allergy. In addition, the requirement for patient cooperation during prolonged imaging may limit its use in critically ill patients.^5,17

In the case presented, the patient's underlying CHF and lack of accompanying CSPH signs initially confounded the diagnosis. There was no thrombocytopenia, hypersplenism, varices, hepatic encephalopathy, ascites, or portovenous collateral vessels. However, in the absence of clinical, laboratory, and echocardiographic signs of CHF exacerbation or valvular disease, suspicion for HH arose due to the patient's cirrhosis history and transudative effusion with SEAG of 1.6.¹⁸ The uncommon nature of HH in the absence of other CSPH signs, as well as the presence of underlying CHF necessitated the use of Tc-99m SCPCS to confirm HH diagnosis.¹⁹ Although we suspect a possible role of the patient's prior CABG in precipitating pleuroperitoneal communication by postsurgical microscopic fibrotic scarring of the diaphragm, a primary postoperative effusion was considered unlikely given that these effusions are most commonly exudative, and that the patient developed cirrhosis in the interim between his CABG and initial presentation for recurrent, right-sided pleural effusions.²⁰

Diagnosis of HH and thus index decompensation was crucial in this setting, as this prompted LT evaluation, as HH can be considered a MELD exception for LT listing for patients with recurrent, refractory effusions.⁶ Although the patient's cardiovascular comorbidities precluded LT or TIPS, this case highlights the importance of recognizing HH, even in the absence of classic CSPH signs, and the role of Tc-99m SCPCS as a valuable diagnostic tool in complex cases.

DISCLOSURES

Author contributions: AC Gunderson, B Andres, JR Gunderson, and B. Ainapurapu contributed to chart review, literature review, and manuscript production. B. Savir-Baruch performed the nuclear medicine study and C. Sabottke and B. Savir-Baruch provided the figure and its interpretation. B. Ainapurapu is the article guarantor.

Financial disclosure: None to report.

Previous presentation: The American College of Gastroenterology Annual Scientific Meeting & Postgraduate Course; October 27, 2024; Philadelphia, Pennsylvania.

Informed consent was obtained for this case report.

ABBREVIATIONS:

CABG

coronary artery bypass grafting

CHF

congestive heart failure

CSPH

clinically significant portal hypertension

HH

hepatic hydrothorax

LT

liver transplant

MELD

model for end-stage liver disease

SEAG

serum effusion albumin gradient

Tc-99m SCPCS

Technetium-99m-sulfur colloid peritoneal cavity scintigraphy

TIPS

transjugular intrahepatic portosystemic shunt