Case Reports Ascites Secondary to Indolent Pacemaker Infection in a Fontan Patient C. Tania Condurache, MD Christopher L. Johnsrude, MD Kristina Bryant, MD Michael R. Recto, MD This report describes what we believe is the 1st reported case of indolent pacemaker infection’s causing abdominal ascites in a patient who has undergone a Fontan operation. Abdominal ascites in a Fontan patient is commonly due to protein-losing enteropathy, systemic venous thrombosis, myocardial dysfunction, chylous ascites, liver cirrhosis, or pancreatitis. Our patient, who had a functional single ventricle, presented with ascites 3 years after undergoing the Fontan operation and pacemaker implantation. After extensive testing and evaluation, we attributed the ascites to indolent infection of the abdominal pacemaker. (Tex Heart Inst J 2008;35(3):342-4) T he Fontan operation is the definitive palliative surgical approach for patients who have univentricular heart physiology. Improvements in surgical technique and medical management have resulted in improved patient outcomes; however, these patients remain at risk for developing arrhythmias, thromboembolic events, conduit obstruction, and heart failure. The development of abdominal ascites is a worrisome finding that occurs in 2.5% to 10% of patients. Ascites is usually secondary to the development of protein-losing enteropathy (PLE), elevated pulmonary arteriolar resistance, or systemic venous thrombosis.1 Determination of the exact cause can be difficult and generally requires extensive evaluation. Herein, we report the case of a patient who presented with ascites 3 years after undergoing the Fontan operation and pacemaker implantation. Key words: Ascites; Fontan procedure; heart defects, congenital/surgery; pacemaker, artificial/ adverse effects; staphylococcal infections; surgical wound infection From: Department of Pediatrics (Dr. Condurache), and Divisions of Infectious Disease (Dr. Bryant) and Pediatric Cardiology (Drs. Johnsrude and Recto), Department of Pediatrics, University of Louisville, Louisville, Kentucky 40202 Address for reprints: Michael R. Recto, MD, Division of Pediatric Cardiology, University of Louisville, 571 S. Floyd Street, Suite 334, Louisville, KY 40202 E-mail: mitch.recto@ louisville.edu © 2008 by the Texas Heart ® Institute, Houston 342 Case Report In April 2000, a 2-week-old female infant with double-inlet left ventricle, L-transposition of the great arteries, and advanced 2nd-degree atrioventricular block had undergone implantation of an epicardial dual-chamber pacemaker. This had been followed at 4 months of age by a Damus-Kaye-Stansel operation with separation of her pulmonary arteries into a right-sided Glenn anastomosis and a left-sided Blalock-Taussig shunt. The pulmonary arteries were separated because of pulmonary hypertension (the mean pressure in both pulmonary arteries was 17 mmHg with a corresponding left atrial mean pressure of 5 mmHg) and low oxygen saturation (systemic saturation, 60%–65%). The patient then underwent re-anastomosis of the right and left pulmonary arteries with takedown of the Blalock-Taussig shunt, at 12 months of age. Cardiac catheterization before the Fontan operation showed a mean pressure of 12 mmHg in both pulmonary arteries, with a corresponding left atrial pressure of 7 mmHg. The extracardiac autologous pericardial Fontan operation and pacemaker revision were performed at approximately 3 years of age. The generator was placed in an abdominal pocket under the left rectus abdominis muscle. Three years later, at 6 years of age, the patient presented to us with a 6-week history of progressive abdominal distention, poor appetite, diminished activity, and vomiting. On physical examination, the abdomen was distended without evidence of jugular venous distention or development of superficial abdominal veins. There was no evidence of facial or pedal edema. The liver was palpable 2 cm below the right costal margin, and the pacemaker pocket was not erythematous or tender to the touch. A complete metabolic panel, including liver and pancreatic enzymes, was within normal limits Ascites Secondary to Pacemaker Infection Volume 35, Number 3, 2008 except for the initial serum albumin level, which was 3.2 gm/dL (range, 3.8–5.4 gm/dL). Urinalysis was normal, without evidence of proteinuria. These studies essentially eliminated nephrotic syndrome, cirrhosis, and pancreatitis as potential causes of the ascites. An abdominal computed tomographic scan showed moderate ascites. The pacemaker pocket was poorly seen due to artifact. Because of fever and progressive abdominal distention secondary to ascites, the patient was admitted to the hospital for further evaluation, which included paracentesis to help determine the cause of the ascites. Inflammatory markers, including erythrocyte sedimentation rate (ESR) (81 mm/hr; range, 0–10 mm/hr) and Creactive protein (CRP) (228 mg/dL; range, 0–9 mg/ dL) were elevated. The repeat serum albumin level was 4.1 gm/dL (range, 3.8–5.4 gm/dL) and the total protein level was 7 gm/dL (range, 6.3–8.2 gm/dL). Blood, urine, and stool cultures were negative for bacteria. Serologic studies were negative for cytomegalovirus, Epstein-Barr virus, blastomyces, histoplasma, Aspergillus, and Coccidioides. Because of persistent abdominal distention, repeat paracentesis was performed. The fluid was described as cloudy and amber in color. The protein level in the peritoneal fluid was elevated (5.9 g/dL) and the white blood cell count was 3,100/mm3 (52% of which were neutrophils). There were no malignant cells identified, and flow cytometry was negative. Because the ascites was considered potentially infectious, the patient was empirically treated with broad-spectrum antibiotics (vancomycin and meropenem) for 14 days. This resulted in mild clinical improvement with decrease in acute-phase reactants (ESR, 70 mm/hr; CRP, 149 mg/dL). Of note, abdominal ultrasonography did not show portal hypertension or thrombosis within the hepatic, splenic, and portal circulations, and pelvic ultrasonography showed a normal uterus and ovaries. Serum markers for peritoneal and ovarian malignancy were negative. Antinuclear antibody (ANA) and ribonucleic protein (RNP) antibody screens were normal, which made the diagnosis of connective-tissue–induced serositis unlikely. Because protein-losing enteropathy could not be completely ruled out as the cause of the ascites, the decision was made to proceed with cardiac catheterization to rule out systemic venous hypertension. The mean pressure recorded within the Fontan circuit was normal (13–14 mmHg). The mean pulmonary capillary wedge pressure (7 mmHg) and left ventricular end-diastolic pressure (7 mmHg) were also normal. Pulmonary vascular resistance was 2.4 Wood units, and there was no evidence of obstruction within the Fontan baffle. Pacemaker function also remained normal. The patient was subsequently discharged from the hospital with mild symptomatic improvement. Over the next 2 months, the patient remained asymptomatic except for persistent mild-to-moderate abdomTexas Heart Institute Journal inal distention. Serum total protein and albumin levels remained normal, and CRP had decreased to 14 mg/ dL (range, 0–9 mg/dL) 2 weeks after hospital discharge. Serial echocardiograms and pacemaker interrogations showed stable findings. Skin-test results of the 13 components contained in her pacemaker were all negative. Two months later, the patient complained of an intermittent “bee-stinging” sensation around her pacemaker, and her pacemaker pocket seemed full upon palpation. Outpatient pacemaker interrogation now showed elevated ventricular capture threshold, and ultrasonography of the pacemaker pocket revealed increased fluid around the generator. The decision was then made to take the patient to the operating room, where a large amount of cloudy, purulent, but odorless material was found in the pacemaker pocket. The generator and all the epicardial leads were removed, and temporary pacing leads were placed. Tissue histology was consistent with chronic inflammation. Bacterial culture from the pacemaker pocket grew Staphylococcus epidermidis. A new dual-chamber epicardial pacemaker was then inserted in a different location after 14 days of intravenous antibiotic therapy with clindamycin. She was eventually discharged from the hospital, and serial follow-up evaluations showed complete resolution of her ascites and good function of her new pacing system. Discussion We describe what we believe to be the 1st reported case of indolent pacemaker infection as a cause of abdominal ascites in a Fontan patient. Ascites in Fontan patients can occur secondary to several conditions, and is often associated with a poor prognosis. The most common cause of ascites in patients who have undergone the Fontan operation is PLE. Protein-losing enteropathy causes ascites from excessive protein loss via the gastrointestinal tract, presumably due to mesenteric hypoperfusion (venous congestion impairing the intestinal mucosal barrier), with subsequent protein leakage into the lumen of the gut.1-4 Protein-losing enteropathy is usually diagnosed by clinical presentation, careful hemodynamic assessment, evaluation of hypoalbuminemia and elevated stool α-1 antitrypsin levels.3 Our patient had isolated ascites without peripheral edema, and except for the initial serum albumin level, which was low, all succeeding serum albumin levels were within normal limits. In addition, the stool α-1 antitrypsin level was normal, which made the diagnosis of PLE unlikely. Ascites can also occur in Fontan patients from hemodynamic abnormalities, including systolic or diastolic dysfunction, atrioventricular valve insufficiency, or systemic venous hypertension. Multiple imaging studies performed in our patient excluded other general causes of ascites, such as thrombosis within the hepatic, portal, or mesenteric circulation.5 Ascites Secondary to Pacemaker Infection 343 Patients with pacemaker infections usually present with skin inflammation over the pacemaker pocket, pacemaker malfunction, or sepsis.6 At presentation, our patient showed none of these features. Presumably, chronic inflammation in the pacemaker pocket caused by indolent S. epidermidis infection led to inflammation of the adjacent peritoneum, with subsequent ascites. Of note, serial computed tomographic scans did not reveal fluid in the pacemaker pocket or associated focal peritoneal inflammation. Antibiotic treatment and replacement of the pacemaker resolved the long-standing ascites, and it did not recur. Despite improvements in perioperative techniques, pacemaker infection occurs in 0.5% to 5.1% of patients.7-11 Some cases are due to bacterial contamination acquired during surgery. In 1 prospective study, the organism that caused the pacemaker infection was present in the operative sample at surgery in 50% of the patients who manifested pacemaker infections between 6 weeks and 16 months later.7 While virulent organisms (like S. aureus) tend to cause infection within 60 days after pacemaker implantation, indolent organisms like coagulase-negative staphylococci are usually responsible for delayed infections. Pacemaker infection may also develop via hematogenous spread. S. epidermidis, an opportunistic skin pathogen, is the primary pathogen in infections of indwelling medical devices. These infections are indolent and clinically silent; furthermore, when isolated, S. epidermidis is often misidentified as a contaminant. Treatment of pacemaker infections typically requires antibiotic therapy, together with removal of the entire pacing system, as was done in this patient.6,7 In conclusion, this case demonstrates that pacemaker infection should be excluded when Fontan patients present with ascites, particularly when common causes such as PLE and abnormal hemodynamics are excluded. 344 Ascites Secondary to Pacemaker Infection References 1. Giannico S, Hammad F, Amodeo A, Michielon G, Drago F, Turchetta A, et al. Clinical outcome of 193 extracardiac Fontan patients: the first 15 years. J Am Coll Cardiol 2006;47(10): 2065-73. 2. Marino BS. Outcomes after the Fontan procedure. Curr Opin Pediatr 2002;14(5):620-6. 3. Rychik J, Spray TL. Strategies to treat protein-losing enteropathy. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu 2002;5:3-11. 4. Stiller B. Protein-losing enteropathy in patients with fontan operations. J Pediatr Gastroenterol Nutr 2003;37(5):528-30. 5. Cochran WJ. Ascites. In: McMillan JA, Feigin RD, DeAngelis C, Jones MD, editors. Oski’s Pediatrics: principles & practice. 4th ed. Philadelphia: Lippincott Williams & Wilkins; 2006. p. 2002-9. 6. Chua JD, Wilkoff BL, Lee I, Juratli N, Longworth DL, Gordon SM. Diagnosis and management of infections involving implantable electrophysiologic cardiac devices. Ann Intern Med 2000;133(8):604-8. 7. Da Costa A, Lelievre H, Kirkorian G, Celard M, Chevalier P, Vandenesch F, et al. Role of the preaxillary flora in pacemaker infections: a prospective study. Circulation 1998;97(18): 1791-5. 8. Bluhm G, Jacobson B, Julander I, Levander-Lindgren M, Olin C. Antibiotic prophylaxis in pacemaker surgery--a prospective study. Scand J Thorac Cardiovasc Surg 1984;18(3):227-34. 9. Cohen MI, Bush DM, Gaynor JW, Vetter VL, Tanel RE, Rhodes LA. Pediatric pacemaker infections: twenty years of experience. J Thorac Cardiovasc Surg 2002;124(4):821-7. 10. Mokaddem A, Bachraoui K, Sdiri W, Kachboura S, Boujnah MR. Pacemaker infections [in French]. Tunis Med 2002;80 (9):509-14. 11. Dheedene M, Klug D, Jarwe M, Kouakam C, Marquie C, Kacet S. Infections secondary to the implantation of a pacemaker: update [in French]. Ann Cardiol Angeiol (Paris) 2000; 49(4):230-7. Volume 35, Number 3, 2008