Development of Refractory Ascites during Amiodarone Therapy in a Hemodialysis Patient with Nephrogenic and Cardiogenic Ascites

Abstract

In this report, the clinical characteristics of a 65-year-old female patient with tricuspid regurgitation, ischemic cardiomyopathy, congestive heart failure, and chronic renal failure were retrospectively evaluated. Laboratory studies revealed cardiogenic ascites coincided with nephrogenic ascites and subclinical amiodarone-induced hypothyroidism. The ascites of the patient was responsive to management of congestive heart failure and therapeutic paracentesis during the first episode, add-on therapy with intensified hemodialysis during the second episode, and add-on therapy with low-dose eltroxin during the third episode. When nephrogenic ascites and cardiogenic ascites of maintenance hemodialysis patients become refractory, hypothyroidism should be examined in these patients.

• aminodarone
• cardiogenic ascites
• hemodialysis
• hypothyroidism
• nephrogenic ascites

INTRODUCTION

Episodes of iodine-induced hypothyroidism were described in chronic hemodialysis patients with high daily intake of iodine even in the absence of apparent underlying thyroid disease.¹ Furthermore, amiodarone-induced hypothyroidism was reported in chronic hemodialysis patients after an unusually short latency period.² Coincidences of cardiogenic ascites with nephrogenic ascites and subclinical amiodarone-induced hypothyroidism were rarely reported. The following case illustrates that a hemodialysis patient with both cardiogenic ascites and nephrogenic ascites may develop refractory ascites due to amiodarone-induced hypothyroidism. This unusual association could be explained by the observation that thyroid dysfunction in chronic renal failure was further aggravated due to the use of amiodarone.

Case Report

A 65-year-old female patient had a history of tricuspid regurgitation, congestive heart failure, hypertension, left-knee fracture s/p operation, acute myocardial infarction, triple vessel disease status after coronary artery bypass surgery, amputation right below knee, and diabetic nephropathy. She was treated with amiodarone for about 1 year to prevent ventricular arrhythmia. The results of laboratory investigations of three episodes of ascitic fluid are shown in Table 1. During the first episode of ascites, the diagnoses based on laboratory examinations were as follows: Non-enhanced computerized tomography demonstrated right-side pleural effusion, moderate ascites, slightly small-sized liver with slightly uneven contour, left renal cyst, some fluid around the portal vessels indicating periportal edema, and no pericardial effusion (Figure 1A and B). Abdominal ultrasound examination revealed hepatic venous congestion. Therapies of cardiogenic ascites with diuretics, inotropic agents, and therapeutic paracentesis were performed. Subsequently, she was admitted because of a second episode of ascites. Results of laboratory investigations were as follows: albumin, 3.5 g/dL (reference value 3.5–5.2 g/dL); blood urea nitrogen, 57.2 mg/dL (reference value 5.0–24.0 mg/dL); creatinine, 4.7 mg/dL (reference value 0.5–1.3 mg/dL). Renal sonography revealed parenchymal renal disease (left side 9.6 cm, right side 9.8 cm). The ascitic fluid urea level was 23.9 mg/dL after approximately 2 months of intensive hemodialysis. A vigorous search for esophageal varix, splenomegaly, portal vein thrombosis, pericardial disease, peritoneal malignancy, pancreatic pseudocyst, inferior vena cava obstruction, Budd–Chiari syndrome, urinary extravasation, and veno-occlusive disease, including non-enhanced computerized tomography and upper gastrointestinal scope, produced negative results. Hepatitis B surface antigenemia and hepatitis C viral serological markers were negative. In addition to cardiogenic ascites, the patient was believed to be suffering from nephrogenic ascites after exclusion of other causes of exudative ascites. Thus, rigid fluid control, intensive hemodialysis, high-protein diet, and intravenous albumin infusion were performed with the diminution of ascites. Later, refractory ascites occurred. Further laboratory studies produced the following results: The thyroid-stimulating hormone level was 17.8 mIU/L (reference value 0.3–5 mIU/L) with a free T4 (thyroxine) of 1.05 ng/dL (reference value 0.8–2 ng/dL). Antithyroid peroxidase antibody was 22 IU/mL (reference value <35 IU/mL). Antithyroglobulin antibody was <20.0 IU/mL (reference value <40 IU/mL). Thus, occurrence of amiodarone-induced subclinical hypothyroidism appeared to be likely. The follow-up thyroid-stimulating hormone level was 17.6 mIU/L with a free T4 (thyroxin) of 0.56 ng/dL. Thus, treatment with low-dose thyroxine sodium (25 mcg/day) was instituted, leading to diminution of ascites.

Figure 1.  (A) Nonenhanced computerized tomography showing right-side pleural effusion and no pericardial effusion. (B) Nonenhanced computerized tomography showing moderate ascites in both the right perihepatic space and the left perisplenic space, slightly diminished size and slightly uneven surface of liver, periportal edema, and no splenomegaly.

Table 1.  Laboratory analyses of three episodes of ascites.

	First (23 September 2008)	Second (8 July 2009)	Third (23 September 2009)
Appearance	Watery	Watery	Watery
Color	Bloody	Straw	Straw
Transparency	Turbid	Turbid	Turbid
Specific gravity	1.016	1.024	1.026
Serum total protein (g/dL)	6.6	7.4	8.9
Serum albumin (g/dL)	3.1	3.5	3.7
Ascites total protein (g/dL)	2.5	4.1	4.9
Ascites albumin (g/dL)	0.9	1.9	Omitted
SAAG	2.2	1.6	Omitted
Ascites RBC (/mm^3)	20,000	17,560	10,880
Ascites WBC (/mm^3)	700	54	200
Ascites neutrophil (%)	5	32	17
Ascites monocyte (%)	95	68	83
Cytology	Negative	Negative	Negative
Bacterial culture	Negative	Negative	Negative
Mycobacteria culture	Negative	NA	NA
Ascites glucose (mg/dL)	136	NA	103
Ascites LDH (IU/L)	NA	NA	250
Ascites fluid urea (mg/dL)	NA	NA	23.9
Ascites triglyceride (mg/dL)	NA	NA	20
Ascites amylase (IU/L)	NA	NA	12

Notes: SAAG, serum ascites albumin gradient; RBC, red blood cell; WBC, white blood cell, LDH, lactate dehydrogenase; NA, not available

DISCUSSION

Based on pathophysiology, it can be said that ascites occurs from the following five causes: (1) a disease in the peritoneum and increased capillary permeability of peritoneum, (2) portal hypertension with elevated hydrostatic pressure, (3) hypoalbuminemia and decreased colloid osmotic pressure, (4) leakage of fluid into peritoneal cavity, and (5) obstructed lymph drainage.^3–5 Serum ascites albumin gradient (SAAG) only offers an insight into the pathophysiology of ascitic fluid formation due to portal hypertension with 97% accuracy. SAAG has empirically been established to be ≥1.1 g/dL in the presence of portal hypertension and <1.1 g/dL in its absence.⁶

The typical indicators of cardiogenic ascites include bloody color, elevated red blood cell count, high lactate dehydrogenase, and an increased SAAG. Bloody color and elevated red blood cell counts are caused by elevated central venous pressure propagated from the right atrium to the hepatic veins, which leads to congestion of liver sinusoids and leakage of red blood cells into the ascites through the lymphatics.^7,8 Chronic diuretics usage may shift a transudate to an exudate. The lyses of intraperitoneal red blood cell resulted in an elevated lactate dehydrogenase. Although abdominal computerized tomography showed that the surface of liver is slightly uneven, portal hypertension is not due to cirrhosis of liver because there were no splenomegaly and esophageal varix. Increased SAAG is a manifestation of an elevated absolute portal pressure, which happens due to elevated inferior vena cava pressure. Passive hepatic congestion seen on abdominal sonography⁹ and periportal edema¹⁰ seen on abdominal computerized tomography support the hypothesis that the cause of the ascites was of cardiogenic origin.

Chronic renal failure patients with nephrogenic ascites typically present with hypertension, moderate to massive exudative ascites, minimal extremity edema, and cachexia. We propose the pathogenesis of nephrogenic ascites as follows: (1) alteration in peritoneal membrane vascular permeability induced by uremic toxins, circulating immune complexes, hemosiderin, prior exposure to dialysis solutions, and activation of the renin–angiotensin system; (2) obstruction of the small peritoneal lymphatics by inflammatory infiltrates induced by circulating immune complexes or peritoneal hemosiderosis. The contributing factors of nephrogenic ascites may include fluid overload, heart failure, elevated hepatic venous hydrostatic pressure, and hypoproteinemia. According to the pathogenesis and the contributing mechanisms, the various treatments of nephrogenic ascites comprise the following methods: (1) rigid fluid and salt restriction, (2) intensified dialysis, (3) high-protein diet, (4) intravenous albumin infusion, (5) intraperitoneal steroid injection, (6) therapeutic paracentesis, (7) peritoneovenous shunt, (8) peritoneal dialysis, (9) renal transplantation, and (10) reinfusion of ultrafiltrated ascites (extracorporeal ascites dialysis). Peritoneal venous shunt can be considered only after failure of other management methods.^11,12 Treatment of nephrogenic ascites should be aggressive because the development of sclerosing peritonitis is a poor complication of nephrogenic ascites.¹³ Based on the laboratory data on our patient, there was no pericardial disease, peritoneal infection, peritoneal malignancy, pancreatic pseudocyst, inferior vena cava obstruction, Budd–Chiari syndrome, esophageal varix, splenomegaly, or urinary extravasation.¹⁴ A comparison of simultaneous blood urea nitrogen with ascitic fluid urea (ascites fluid urea 23.9 mg/dL; blood urea nitrogen about 38.7 mg/dL) also suggested nephrogenic ascites instead of urinary extravasation.

Ascites is a well-documented but unusual occurrence in hypothyroid patients. The typical feature of ascites caused by hypothyroidism demonstrated high ascites total protein (>2.5 g/dL) and high cholesterol concentrations, moderate white blood cell counts, and a predominance of mononuclear cells.¹⁵ The mechanisms of ascites formation were proposed as follows: an increase in capillary permeability and lack of a compensatory increase in lymph flow and protein return rate.¹⁶

The uremic milieu could predispose patients with renal failure to thyroid dysfunction. Thyroid function may be directly influenced by many factors in renal failure. Protein caloric malnutrition could interfere directly with thyroid function. Fluoride derived from dialysate may also interfere with thyroid function. Uremic toxins could inhibit the peripheral conversion of T4 to triiodothyronine (T3). Uremic toxins may interfere indirectly with thyroid function through the pituitary–thyroid axis. Plasma inorganic iodide may be elevated in the blood of hemodialysis patients and could interfere with thyroid function.¹⁷ When the plasma iodide level elevates to a critical threshold, acute inhibition of iodine organization and subsequent thyroid hormone synthesis (acute Wolff–Chaikoff effect) occur. Adaptation to large doses of iodide is rapid, occurring in 48 h, which is associated with autoregulatory inhibition of iodide transporter across the intrathyroidal membrane. Failure to escape the Wolff–Chaikoff effect has been proposed to explain aminodarone-induced hypothyroidism.¹⁸

Amiodarone is eliminated primarily by hepatic metabolism and biliary excretion. Thus, the pharmacokinetics of amiodarone is not affected in chronic renal failure. The mechanisms of amiodarone-induced hypothyroidism were proposed to be the following: First, amiodarone inhibits the peripheral conversion of T4 to T3 by competitive inhibition of 5’-monodeiodinase and favors the generation of reverse T3, which has no T3 activity. Second, amiodarone is an iodinated compound of which 37.2% of the molecular weight is inorganic iodide. In our patient, about 9 mg/day of iodide released from dehalogenation of 300 mg of amiodarone.^19,20 As the daily iodide requirement is approximately 200 μg, the patient received many times her normal daily requirement.

In conclusion, chronic renal failure patients with unexplained and refractory ascites and anemia should be tested for hypothyroidism especially during amiodarone therapy. In this patient with a complex pathogenesis of ascites, SAAG can be omitted in the third round of diagnostic abdominal paracentesis. Uremic conditions and long-term therapy of aminodarone could complementarily cause hypothyroidism. We observed that amiodarone-induced hypothyroidism, even in the subclinical stage, could facilitate the accumulation of ascites in the patient. When the ascites of hemodialysis patients becomes refractory, hypothyroidism should be kept in mind. A low-dose T4 sodium was suggested initially for fear of atrial fibrillation.²¹

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.