Asian Cardiovasc Thorac Ann 2000;8:372-374
© 2000 Asia Publishing EXchange Pte Ltd
Pericardial and Pleural Effusion After Central Venous Line Insertion
Ehud Deviri, MD,
Oleg Avrutis, MD, PhD,1,
Shalom Friedman, MD,1,
Jack Meshoulam, MD,1,
Ohn Sibirsky, MD,1,
George Blinder, MD,2,
Joseph B Borman, FRCS
Department of Cardiothoracic Surgery
1 Department of Surgery
2 Medical Imaging Center
Bikur-Cholim Hospital
Jerusalem, Israel
|
|
For reprint information contact: Ehud Deviri, MD Tel: 972 2 646 4342 Fax: 972 2 646 4372 Department of Cardiothoracic Surgery, Bikur-Cholim Hospital, 5 Strauss Street, P.O. Box 492, Jerusalem 91004, Israel.
|
 |
Abstract
|
|---|
A large volume of fluid collected in the pericardium and left pleural space following central venous catheter insertion for total parenteral nutrition in a 35-year-old man. This did not cause fatal hemodynamic compromise and was successfully diagnosed and treated.
|
Introduction
|
|---|
Central venous catheters are those with the distal tip located in the vena cava. Such catheters are used widely for the administration of fluids, medications, and parenteral nutrition, as well as for hemodynamic monitoring, blood sampling, and hemodialysis. Perforation of the heart with a central venous catheter is a well-known complication associated with a high mortality.1–3 In most reported cases, the signs of tamponade were misleading and in many, death was immediate before a diagnosis could be made.1,2 A case is described in which fluid collected in the pericardium and left pleural cavity following insertion of a central line for total parenteral nutrition (TPN).
|
Case Report
|
|---|
A 35-year-old man was admitted for elective cholecystec-tomy and repair of recurrent postoperative ventral hernia. Seven years earlier, he had undergone exploration by laparotomy for an abdominal stab wound, followed 3 years later by repair of an abdominal wall hernia with intraperitoneal nonabsorbable synthetic mesh. The hernia reoperation was complicated by tight intraperitoneal adhesions between the mesh, the stomach, and the small and large bowel loops. After dissection of the adhesions, open cholecystectomy and repair of the abdominal wall hernia was performed with tension-free onlay Prolene mesh (Johnson & Johnson, Brussels, Belgium). Ten weeks later, a further reoperation was needed due to Escherichia coli wound infection unresponsive to antimicrobial treatment. The infected mesh was removed, the wound was debrided and drained, and antimicrobial treatment was administered. On the second postoperative day, an enteric cutaneous fistula developed. TPN was initiated via a right subclavian catheter that was located above the right atrium by radiographic guidance. Nine days later, the right subclavian catheter was replaced over a guide wire (Centra-Line; Biometrix, Jerusalem, Israel). A chest radiograph after the procedure showed the catheter in the right atrium; however, it was not relocated. On the following day, the patient complained of malaise and diaphoresis. His blood pressure was 115/45 mm Hg. He was afebrile, his hemoglobin level was 155 g•L–1, and his white blood cell count was 24,000/mm3. Within a few minutes, he had a short syncopal episode. His blood pressure dropped to 70/40 mm Hg with a pulse rate of 120 beats•min–1. Septic shock due to TPN line infection was suspected and the infusion was discontinued. Blood cultures were taken and preparations to replace the line began. However, the patient improved spontaneously within a few minutes and became hemodynamically normal without any complaints. There was no longer any sign suggesting line sepsis so TPN was restarted. During the next few hours, the patient developed progressive dyspnea. His blood pressure was 100/60 mm Hg, his pulse rate was 120 beats•min–1 without pulsus paradoxus, and his temperature was 37.4°C. The TPN catheter was relocated above the right atrium. Since a pulmonary embolism was suspected, a computed tomography scan was performed. This revealed a thrombus over the tip of the TPN catheter and evidence of emboli in the tributaries of the left pulmonary artery to the upper segment, as well as extensive left pleural and pericardial effusions. A transthoracic echocardiogram confirmed the diagnosis of a large pericardial effusion with compression of the right atrium. Under general anesthesia, the patient underwent subxiphoid pericardiotomy and 300 mL of milky fluid was drained from the pericardial space. A 28F drain was left in the pericardial cavity and the wound was closed. Another 36F drain was inserted into the left pleural cavity through the fifth intercostal space (midaxillary line), which drained more than 2 L of a similar milky fluid. A new central line was inserted via the left subclavian vein. The postoperative course was uneventful. Immediately following surgery, heparin was administered for 7 days. Thirty-six hours after the operation, the drains were removed because drainage was minimal. Computed tomography 7 days postoperatively showed neither signs of pulmonary emboli nor pericardial or pleural effusions. Since the enterocutaneous fistula failed to close after 5 weeks of conservative treatment, the patient underwent laparotomy with removal of the entire intraperitoneal mesh inserted 4 years previously, resection and anastomosis of the ileum involved in the fistula, and debridement and primary closure of the abdomen. The postoperative course was uneventful and when seen 6 months later, he was well and symptom-free.
|
Discussion
|
|---|
Brown and Kent4 first reported perforation of the right atrium by a central venous catheter in 1956. Since then, over 100 cases of heart perforation have been reported.3 Catheter manufacturers enclose literature in catheter kits, warning of the risk of cardiac tamponade. Cardiac per-foration by a central venous catheter is a preventable complication. In all the reported cases of tamponade that followed insertion of a central venous catheter, chest radiography showed that the catheters were within the cardiac silhouette.3 Krog and colleagues5 stressed that central venous catheters should be located 2 cm above the cardiac silhouette.
Tamponade caused by a central venous catheter can be disastrous. Mortality rates vary from 65% to 91%.1–3,6,7 Collier and Goodman3 reviewed 11 cases of cardiac tamponade in which 10 patients died; the only survivor sustained extensive brain damage and remained in a vegetative state. Early diagnosis of cardiac tamponade caused by a central venous catheter is difficult. Of 67 cases reviewed by Chabanier and colleagues,2 23% presented with circulatory arrest, 33% with rapid deterioration into circulatory collapse, and 39% with a slower progressive deterioration. In 65% of the patients who died, the diagnosis was made at autopsy.2 Several patients presented with signs suggesting pulmonary emboli.1 The site of perforation was the right atrium in 53%, right ventricle in 33%, superior vena cava in 6.6%, and inferior vena cava in 6.6%.1 The fluid found within the pericardial cavity was the intravenous perfusate in 68% and hemorrhagic fluid in 29%.1,2 The time from insertion of the central venous catheter to perforation was less than 4 hours in 15% of cases; in 46%, the perforation occurred between 4 and 48 hours from catheter insertion and in 37% it was between 2 and 22 days.2 Simultaneous perforation into the pericardial and contralateral pleural cavity after right subclavian vein catheterization is extremely rare and has been reported only once in the English-language medical literature.8 In that case, the complication was diagnosed and treated successfully.
The patient reported here was probably unusually fortunate in that the majority of the TPN fluid was drained from the pericardium into the left pleural cavity (Figure 1
). It was presumed that during or after insertion of the central line, there was perforation of the heart, pericardium and mediastinal pleura, but it is not known whether the catheter perforated the heart spontaneously or if it was inserted erroneously into the pericardial cavity directly. However, the TPN fluid drained into the pleural cavity most likely saved the patient's life since the course of the tamponade was more gradual and there was sufficient time to make the diagnosis and treat the complication. As reported by others, a diagnosis of pulmonary emboli was also suspected in this case. The source of pulmonary emboli demonstrated by computed tomography presumably stemmed from thrombi at the site of the heart perforation or the tip of the catheter. Probably, if the catheter tip had been properly placed above the cardiac silhouette, tamponade could have been prevented.
View larger version (45K):
[in this window]
[in a new window]
|
Figure 1. The presumed course of the guidewire and catheter that probably perforated both the heart and pericardium, enabling drainage of the total parenteral nutrition fluid into the left pleural cavity.
|
|
It is essential in every instance of insertion of a central line that the catheter tip be located proximal to the cardiac silhouette. In the event of hemodynamic deterioration in a patient with a central line, cardiac tamponade should be suspected. Only strict adherence to these principles will decrease the incidence of this malignant complication.
|
References
|
|---|
-
Collier PE, Ryan JJ, Diamond DL. Cardiac tamponade from central venous catheters. Report of case and review of the English literature. Angiology 1984;35:595–600.
-
Chabanier A, Dany F, Brutus P, Vergnoux H. Iatrogenic cardiac tamponade after central venous catheter. Clin Cardiol 1988;11:91–9.[Medline]
-
Collier PE, Goodman GB. Cardiac tamponade caused by central venous catheter perforation of the heart: a preventable complication. J Am Coll Surg 1995;181:459–63.[Medline]
-
Brown CA, Kent A. Perforation of right ventricle by polyethylene catheter. South Med J 1956;49:466–7.
-
Krog M, Berggren L, Brodin M, Wickbom G. Pericardial tamponade caused by a central venous catheter. World J Surg 1982;6:138–43.[Medline]
-
McGoon MD, Benedetto PW, Greene BM. Complications of percutaneous central venous catheterization: a report of two cases and review of the literature. Johns Hopkins Med J 1979;145:1–6.[Medline]
-
Defalque RJ, Campbell C. Cardiac tamponade from central venous catheters. Anaesthesiology 1979;50:249–52.[Medline]
-
Krauss D, Schmidt GA. Cardiac tamponade and contralateral hemothorax after subclavian vein cathe-terization. Chest 1991;99:517–8.[Abstract/Free Full Text]
TOP
Abstract
Introduction
