Ann Thorac Surg 2006;81:356-358
© 2006 The Society of Thoracic Surgeons
Case report
Gastropericardial Fistula, Purulent Pericarditis, and Cardiac Tamponade After Laparoscopic Nissen Fundoplication
Eero I. T. Sihvo, MD, PhD,
Jari V. Räsänen, MD,
Marja Hynninen, MD, PhD,
Tuomo K. Rantanen, MD, PhD,
Jarmo A. Salo, MD, PhD
*
Department of Cardiothoracic Surgery, Helsinki University Central Hospital, Helsinki, Finland
Accepted for publication August 23, 2004.
* Address correspondence to Dr Salo, Department of Cardiothoracic Surgery, Helsinki University Central Hospital, Haartmaninkatu 4, PO Box 340, Helsinki, FIN-00029 HUS Finland (Email: jarmo.salo{at}hus.fi).
 |
Abstract
|
|---|
Gastropericardial fistula, purulent pericarditis, and cardiac tamponade developed 7 years after laparoscopic Nissen fundoplication. The patient was successfully managed through a thoracotomy by open drainage of the pericardium, excision and closure of the fistula, and an omentum flap.
|
Introduction
|
|---|
Although laparoscopic antireflux surgery is considered a safe and durable procedure, the long-term complications related to recurrent hiatal hernia such as obstruction, incarceration, or perforation can cause substantial morbidity and mortality [1]. We report on a patient in which a gastropericardial fistula 7 years after a laparoscopic Nissen procedure caused purulent pericarditis and cardiac tamponade.
A 54-year-old man was hospitalized for acute chest pain. His medical history included untreated hypertension. Because of gastroesophageal reflux disease, 7 years earlier he had undergone a laparosopic Nissen fundoplication. The short gastric vessels were divided, a 360-degree wrap was performed around an intraluminal bougie, and the hiatus was closed behind the wrap. His reflux symptoms had returned during the year before admission.
At admission the patient complained of right-sided chest pain. His blood pressure was 60/40 mm Hg and his heart sounds were weak. Electrocardiography revealed low-voltage and marginal inferior segment elevations. Hypotension was corrected with fluid resuscitation and inotropic medication. Suspicion of myocardial infarction led to uncomplicated thrombolysis. Later, cardiac enzymes were normal. Liver enzymes and serum creatinine were elevated, and the patient had microcellular anemia. A chest roentgenogram revealed air in the pericardium. Computed tomography of the chest and upper abdomen showed a 2-cm layer of air and liquid in the pericardium and a small paraesophageal hernia without signs of extravasation of contrast material within the pericardium (Fig 1). Because cardiac tamponade was suspected, the patient was transferred to Helsinki University Central Hospital. In the emergency room, electrocardiography showed tachycardia (123/min) and low voltage without segment elevation changes. Gastroscopy revealed a small paraesophageal hernia with two ulcers in the herniated part of the fundus. The wrap was a little loose, but it was underneath the diaphragm. On the operating table, his face and neck were bluish.
View larger version (70K):
[in this window]
[in a new window]
|
Fig 1. Computed tomography of the chest and upper abdomen after oral administration of contrast material. Air and liquid present in the pericardium (white arrow) (left). Contrast material present in small paraesophageal hernia without signs of extravasation within the pericardium (dotted arrow) (right).
|
|
Left posterolateral thoracotomy through the sixth intercostal space was performed to relieve the tamponade and close the suspected fistula between the stomach and pericardium. After opening of the pericardium, the systolic blood pressure doubled from 100 to 200 mm Hg, and the diagnosis was purulent pericarditis and fistula between the pericardium and the egg-sized paraesophageal part of the fundus. Cultures grew Streptococcus viridans. The left diaphragm was opened, and the herniated part of the fundus was brought to the abdomen. The fistula was excised, and the healthy gastric wall around the fistula was closed in two layers. Because of dense inflammation and scarring around the lower esophagus and hiatal area, primary closure of the diaphragmatic defect (3 cm in size) or the use of mesh was considered impossible. Therefore an omentum flap was used to cover the suture line and infected area in the lower mediastinum and to obstruct the defect in the diaphragm.
Postoperatively, delirium prolonged the patient's stay in the intensive care unit for 12 days. Broad-spectrum antibiotics and antifungal therapy were continued, and he was discharged on postoperative day 27. At follow-up 2 months postoperatively, infection measurements were normalized, and he reported normal swallowing. At his 4-month postoperative follow-up, the gastroscopy showed that the Nissen wrap was beneath the diaphragm and no paraesophageal hernia was detectable.
|
Comment
|
|---|
Fistula between either the esophagus or stomach and the pericardium is a rarity. For benign fistulas, the main predisposing factor is previous surgery near the esophagogastric junction [2]. After antireflux surgery, fistulas have been reported from stomach to aorta, diaphragm, bronchi, pleural space, mediastinum, and pericardium [1, 3–5]. Only two reported cases exist of gastropericardial fistulas presenting as a late complication of laparoscopic Nissen fundoplication [5, 6].
In a review of gastropericardial fistula and of esophagopericardial fistula, mortality was 85% (44 of 52) [7] and 83% (24 of 29) [8]. The predominant pathophysiological features accompanying these fistulas were purulent pericarditis and tamponade (tamponade in 34%) [2]. None of the reported patients with gastropericardial fistula and purulent pericarditis have survived to date. The only reported cases featuring tamponade as a late complication of antireflux surgery date from the era of open surgery [4]. These two patients had sterile hydropericardium, and their favorable outcome according to the authors was related to absence of infection.
Repeated surgery and subphrenic abscess formation after antireflux surgery seem to be factors for early gastropericardial fistulas [9, 10]. Late fistulas, months or years after open Nissen fundoplication, are most commonly related to ulcer formation and recurrent hernias [5]. Our patient had both: two ulcers and a new paraesophageal hernia. The cause of these ulcers could be chronic ischemia at the tip of the hernia. The two earlier cases with gastropericardial fistulas after laparoscopic Nissen fundoplication had herniated wraps. Laparoscopic fundoplications are increasing; improper closure of the hiatus or dehiscence of hiatal closure may predispose these patients to severe late complications.
In case a fistula between the upper gastrointesinal tract and pericardium is suspected, the rationale for early gastroscopy can be questioned. Due to air insufflation, tamponade may increase and the patient acutely deteriorates. The mainstay of treatment of late fistulas is prompt surgery. A successful treatment includes early pericardial drainage, usually by pericardial window and de-functioning esophagostomy or gastrostomy, or corrective surgery, or both [2]. In recent cases, intrathoracic rupture of herniated Nissen wraps healed spontaneously after replacement of the wrap into the abdomen, and creation of a controlled gastric fistula [1]. In the two cases of late gastropericardial fistulas after laparosopic Nissen, the fistula in the stomach was resected or merely closed, the wrap was reduced beneath the diaphragm, and the diaphragm with healthy muscle then served as a barrier between the closure of the stomach and pericardium [5, 6]. In our case, because primary closure of the small diaphragmatic defect was impossible because of intense inflammation and scarring, our principle was to open drainage of the pericardium and excise the diseased area of the gastric wall, to close the healthy wall with two-layer suturing, and then to cover the area of infection and sutures in the upper abdomen and lower mediastinum with a well-vascularized omentum flap, as is done in high-risk airway or esophageal anastomoses.
|
Acknowledgments
|
|---|
We express our gratitude for the secretarial assistance of Yvonne Sundström.
|
References
|
|---|
- Mansour KA, Sharma J. Delayed intrathoracic rupture of herniated Nissen fundoplicationreport of two cases. Ann Thorac Surg 2003;75:1957-1959.[Abstract/Free Full Text]
- West AB, Nolan N, O'Brian DS. Benign peptic ulcers penetrating pericardium and heartclinicopathological features and factors favoring survival. Gastroenterology 1988;94:1478-1487.[Medline]
- Hauters P, de Canniere L, Collard JM, Buysschaert M, Michel LA. Gastrodiaphragmatic fistula after transabdominal Nissen fundoplicationAn unusual complication. Jun J Clin Gastroenterol 1990;12(3):313-315.[Medline]
- de Bruyne B, Dugernier T, Goncette L, Reynaert M, Otte JB, Col J. Hydropneumopericardium with tamponade as a late complication of surgical repair of hiatus hernia Am Heart J 1987;114(2):444-446.[Medline]
- Murthy S, Looney J, Jaklitsch MT. Gastropericardial fistula after laparoscopic surgery for reflux disease N Engl J Med 2002;346:328-332.[Free Full Text]
- Su-Gandarilla J, Guillem P, Herjean M, Triboulet JP. Gastropericardial fistula after failure of laparoscopic hiatal repair Dis Esophagus 2000;13:262-264.[Medline]
- Letoquart JP, Fasquel JL, L'Huillier JP, et al. Gastropericardial fistulaReview of the literature apropos of an original case. J Chir 1990;127:6-12.
- Konttinen MP, Pitkaranta PP, Heikkinen LO, Talja MT, Ala-Kulju KV. Esophago-pericardial fistulaA case report and review of the literature. Thorac Cardiovasc Surg 1985;33:341-343.[Medline]
- Ikard RW, Jacobs JK. Gastropericardial fistula and pericardial abscessan unusual complication of subphrenic abscess following Nissen fundoplication. S Med J 1974;67:17-19.
- Herrington JL, Ibrahim A. Complications of repeated operations to control esophageal reflux Am Surg 1977;43:203-207.[Medline]
This article has been cited by other articles:
|
|
|
|
 
D. Pop, N. Venissac, L. Rami, and J. Mouroux
Gastropericardial fistula after laparoscopic surgery for gastroesophageal reflux disease
J. Thorac. Cardiovasc. Surg.,
June 1, 2007;
133(6):
1676 - 1677.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
