|
|
||||||||
Eur J Cardiothorac Surg 2003;24:847-849
© 2003 Elsevier Science NL
Case report |
Cardiac Surgery Department, University of Amiens, South Hospital, Amiens Cedex 01, France
Received 2 April 2003; received in revised form 7 July 2003; accepted 14 July 2003.
* Corresponding author. Tel.: +33-3-2245-5925; fax: +33-3-2245-5331
e-mail: gtouati.hms{at}invivo.edu
| Abstract |
|---|
|
|
|---|
Key Words: Purulent pericarditis Esophago-pericardial fistula Carcinoma
| 1. Introduction |
|---|
|
|
|---|
Depending on the level of the lesion, oesophageal fistulas can present in the form of mediastinitis [13] or purulent pericarditis [4,6]. We report a case of purulent pericarditis with an unusual clinical presentation and a favourable outcome.
| 2. Case report |
|---|
|
|
|---|
Clinical interview revealed a history of weight loss of 4 kg over 10 days. Auscultation demonstrated a pericardial rub. The leukocyte count was elevated to 25 000/mm3. The electrocardiogram showed elevation of the PQ segment in leads II, III, V2, V3, and chest X-ray demonstrated cardiomegaly. Echocardiography showed a moderately large circumferential pericardial effusion with pericardial thickening.
Percutaneous aspiration was performed on the same day due to clinical and haemodynamic deterioration with severe dyspnoea, cyanosis and hypotension of 80/30 mmHg, and provided 550 ml of purulent fluid. Direct examination of the pericardial fluid showed several bacteria suggestive of salivary flora (Gram-positive and -negative bacteria, yeasts) and combination antibiotic therapy with amikacinticarcillin+clavulanic acid and fluconazole was commenced. Pericardial lavagedrainage with 0.9% NaCl solution and iodinated povidone was installed via a subxyphoid incision.
Thoracic computed tomography (CT) (Figs. 1 and 2) revealed:
|
|
The hypothesis of oesophago-pericardial fistula was proposed, although none of the various complementary investigations had demonstrated the fistula. Medical treatment with antibiotics and pericardial drainage was continued.
After a clinical improvement of the patient's status during 5 days, an aggravation of the pericardial compression was observed. Pericardial thickening was major leading to a true constrictive pericarditis, and it was subsequently decided to operate following the development of a peritoneal syndrome and cholestasis.
Operative strategy was difficult, but followed the obvious pattern. The first operative step was a phrenic-to-phrenic pericardiectomy via a median sternotomy. Calcifications and major thickening of the pericardium were found, with numerous pus sacs. After closure of the irrigated chest, the second step was a median laparotomy: surgical exploration revealed gangrenous cholecystitis, which was treated by cholecystectomy; a preparation of the stomach for gastroplasty was done. The third step was a right thoracotomy, after re-installation of the patient. After difficult dissection of inflammatory tissues, a thoracic oesophagectomy was done, with resection of the oesophago-pericardial fistula. After gastric rising, oesophago-gastric anastomosis was made in the chest, at the thoracic inlet. No tumour extension in relation to surrounding tissues other than the pericardium was observed. Histological examination of the resection specimen showed squamous carcinoma of the oesophagus with infiltration of the pericardium, but no lymph node invasion.
The postoperative course was marked by an episode of septic shock with anuria. Blood cultures revealed a Meth-R coagulase-negative Staphylococcus. Treatment then consisted of ticarcillin+clavulanic acid, amikacin and vancomycin, together with norepinephrine and continuous arteriovenous haemofiltration for 48 h. Mechanical ventilation was withdrawn on the 7th postoperative day and the patient was discharged from hospital on the 21st postoperative day. Adjuvant chemotherapy and radiotherapy was started on the 45th day.
With a follow-up of 8 years, the patient is still alive, in stable remission, eats normally and presents no clinical or CT signs of recurrence.
| 3. Discussion |
|---|
|
|
|---|
The initial symptoms are usually incomplete, before the development of tamponade. The earliest signs are electrocardiographic abnormalities and precordial discomfort. Pericardial rub is rare. Chest X-ray may sometimes visualize a fluid level in the pericardial cavity and cardiomegaly. The presence of pyopneumopericardium may induce a water-wheel murmur on auscultation of the precordial region.
Echocardiography demonstrates the pericardial effusion, evaluates its severity and its repercussions on the cardiac cavities, and guides pericardial aspiration which confirms the presence of purulent fluid. Surgical drainage is essential for lavage of the pericardial cavity, as tamponade rapidly recurred in all patients not initially drained.
Oesophageal endoscopy demonstrates the fistula and allows biopsies when a tumour is suspected. An upper gastrointestinal gastrograffin series may visualize the fistulous tract. It must be performed after pericardial drainage, as there is a real risk of inducing tamponade by passage of the contrast agent in the fistula. Computed tomography, performed in doubtful cases just after the upper gastrointestinal series, may visualize the presence of gastrograffin in the pericardium.
None of the complementary investigations performed in our patient demonstrated the fistulous tract, which was only discovered intraoperatively and which was responsible for this case of pericarditis. The main lesson to be drawn from this case is that the possibility of oesophago-pericardial fistula must be considered on the basis of clinical arguments, even in the absence of radiological evidence.
The clinical course of purulent pericarditis depends on rapid diagnosis, urgent treatment combining pericardial drainage, appropriate antibiotics and surgical repair of the fistula. Primary subxyphoid drainage is generally performed in patients presenting with a first episode of pericarditis. Following intraoperative discovery of purulent pericarditis, a lavagedrainage system must be installed, followed by pericardial resection. Anterior thoracotomy is recommended when the diagnosis of purulent pericarditis has been established preoperatively [1], as it provides good exposure and decreases the risk of sternal infection. However, in the case of constrictive pericarditis, midline sternotomy is required in order to perform phrenic-to-phrenic pericardiectomy.
Pericarditis secondary to oesophago-pericardial fistula has a poor prognosis. Among the 67 cases in the literature, generally caused by a foreign body in the oesophagus, only 13 patients survived (19%). None of the ten cases associated with oesophageal cancer survived for more than several weeks, as only palliative treatment with an oesophageal stent was proposed. However, the present case suggests that an aggressive therapeutic approach should be proposed in order to achieve long-term remission in these frequently young patients.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |