Asian Cardiovasc Thorac Ann 1999;7:292-296
© 1999 Asia Publishing EXchange Pte Ltd
Techniques and Outcomes of Two Modes of Pericardial Drainage
Albert Leung Wai Suen, FRCP,
Ho Leung Sing, FRCS,1,
Chan Ngai Yin, MRCP,
Chan Mui Tong, FRCS,1,
Tsang Hing Hung, MRCP,
Kwok Miu Fong, MRCP
Cardiology Team, Department of Medicine and Geriatrics
1 Thoracic Surgery Unit, Department of Surgery, Princess Margaret Hospital, Hong Kong, People's Republic of China
|
|
For reprint information contact: Albert Leung Wai Suen, FRCP Tel: 852 2990 1111/3725 Fax: 852 2307 2270 Department of Medicine and Geriatrics, Princess Margaret Hospital, 1 Princess Margaret Hospital Road, Lai Chi Kok, Kowloon, Hong Kong, People's Republic of China.
|
 |
Abstract
|
|---|
The techniques and outcomes of two modes of pericardial drainage, percutaneous pericardiocentesis and surgical pericardiotomy, were analyzed. Percutaneous pericardiocentesis was performed using the Seldinger technique. The puncture site was determined by selected criteria. Surgical pericardiotomy was performed mainly through the subxiphoid route. There were 20 pericardiocenteses and 27 pericardiotomies performed in 39 patients; 19 pericardiocenteses were successful with no complications noted, all 27 pericardiotomies were successful with only minor complications. Bloodstained fluid was found in 27 of the 38 samples of drainage (71%). Cytology for malignancy was positive in 21% and culture for tuberculosis was positive in 1 case. Biopsies improved the diagnostic yield of either disease from 18% to 38%. Malignancy was the most common cause of effusion (41%), followed by uremia. No secondary causes were found in uremic patients. The causes in 7 patients (18%) were not identified. During the study period, 16 patients died, including 1 soon after surgical drainage. Both techniques were considered to be safe and effective. In view of a case of sudden death soon after surgical drainage, it is recommended that patients with severe tamponade should have controlled percutaneous drainage.
|
Introduction
|
|---|
Pericardial effusion can be drained by percutaneous pericardiocentesis (PP) or by surgical pericardiotomy (SP). In principle, PP should be chosen to relieve effusions severe enough to cause cardiac tamponade, while SP should be used to relieve less urgent clinical conditions because it is safer and can achieve a higher rate of diagnosis. Various techniques and instruments for PP have been developed.1–3 There are also several approaches that can be deployed for SP.4–6 However, each method of drainage carries its own set of precautions and potential complications. Unless there are complications arising from the procedure, the progress of the clinical condition of the patient after drainage usually depends on the natural course of the underlying disease. Pathological samples obtained during drainage can provide information on the etiology of the effusions but the diagnostic yields of such specimens are usually low.3,7,8
This retrospective study compared the two modes of pericardial drainage in patients treated between January 1995 and December 1997. The techniques, results, and complications of both drainage procedures were assessed. The pathology of the specimens obtained during drainage were analyzed in relation to the rate of diagnosis and the causes of pericardial effusions.
|
Patients and Methods
|
|---|
Thirty-nine patients underwent pericardial drainage by one or both techniques. PP was performed under 2-dimensional echocardiographic guidance. The Cook PCS-850 pericardiocentesis set (Cook, Inc., Bloomington, IN, USA) was selected in the early period. More recently, the Lock pericardiocentesis set (PCS-830-LOCK; Cook, Inc., Bloomington, IN, USA) was chosen. The Seldinger technique was employed, whereby the pericardium was entered with an 18-gauge needle, followed by insertion of a 0.035 inch (0.089 cm) J-tipped guidewire into the pericardial space.9 The needle was removed and a soft multihole catheter was positioned in the pericardial space over the guidewire. The pericardiocentesis set used in the initial period had a straight catheter, whereas the Lock set had a catheter with a pigtail shape, which was simpler to use (electrocardiographic monitoring was not employed). The patient was placed either in the supine position with the head slightly elevated or in a sloping position with the head and thorax tilted upward at 45 degrees (if the patient was dyspneic). The puncture site was determined by 2-dimensional echocardiography according to 3 criteria: a site where the fluid was closest to the transducer; entry in the region of maximal fluid volume; and a site where the needle track avoided the heart and other underlying vital structures. Of the 20 PP procedures, entry through an apical site was selected in 10 cases and a subxiphoid site was selected in 9. The approach was not recorded in one patient.
SP was carried out through the subxiphoid approach in 23 (85%) of 27 procedures. General anesthesia was used in 15 cases, local anesthesia in 8, thoracoscopy in 3, and one was performed through an anterior intercostal route (due to a large liver blocking the subxiphoid region). For the subxiphoid approach, the patient was placed in a head-up tilt of 45 degrees. A 5-cm incision was made 1 cm below the left costal margin. The rectus sheath was incised and the rectus muscle and the diaphragm were detached from the costal margin. The pericardium was opened between stay stitches. A 3 x 3 cm window was created and a silicone drain was placed in the pericardial cavity through a separate skin incision.
|
Results
|
|---|
Table 1
shows the age and sex of the 39 patients according to the modes of pericardial drainage. Four patients underwent both techniques; 1 had a failed percutaneous puncture and the other 3 had repeat procedures because of recurrence of pericardial effusion. One of these 3 patients had 4 PP and 2 SP procedures, contributing to the total of 47 procedures (20 PP and 27 SP) in these 39 patients.
Nineteen (95%) of the 20 PP procedures were successful. In the unsuccessful case, the procedure was abandoned after two attempts failed to yield fluid; surgical drainage was immediately undertaken, which showed extensive fibrin inside the pericardial sac. No complications (including pneumothorax) resulted from any of the PP procedures. The drainage catheters were kept in position for 1 to 8 days (mean, 4.1 days). No infections or cardiac arrhythmias (except occasional ectopics) were induced. No mortality was directly related to the procedure.
All SP procedures were successful. The drainage catheters were kept in place for 1 to 14 days (mean, 6.3 days). One patient had a gaped wound at the drainage site after the operation and 1 had a wound infection after thoracoscopic drainage, which subsequently healed. No cardiac arrhyth-mias were induced in any of the surgical procedures. One death was probably related to the drainage procedure. This patient was a 23-year-old woman with known systemic lupus and membranous glomerulone-phritis. She had clinical features of extensive effusion and cardiac tamponade, approximately 1000 mL of fluid was drained successfully by the subxiphoid surgical approach. She developed bradycardia followed by cardiac asystole 30 minutes after drainage and responded initially to cardiopulmonary resuscitation but died one hour later because of asystole.
The amount of fluid drained ranged from 100 to 1100 mL. Thirty-nine samples of pericardial fluid and 26 pericardial biopsy specimens were examined. The pathological findings in pericardial fluid are shown in Table 2. The results were assessed according to whether the samples were collected from PP or SP. The biopsy specimens, all obtained from the surgical procedure, were checked for malignancy or tuberculosis. Fresh blood was drained from 1 patient who had developed pericardial effusion and tamponade as a result of perforation of a cardiac chamber by a guidewire during cardiac catheterization. The clinical condition improved markedly after PP and there was no recurrence of effusion. None of the other patients were trauma cases. Of the 38 nontraumatic fluid samples collected, 27 (71%) were bloodstained of which, 7 were heavily bloodstained. All recurrent cases had bloodstained fluid. There were 34 (87%) exudative fluid samples and 5 (13%) transudative. All the transudative samples were obtained by SP.
Cytology was positive for malignancy in only 8 (21%) samples and 1 was suspected to be malignant. Culture was positive for tuberculosis in only 1 case. Of the 26 pericardial biopsies obtained, diagnosis (malignancy or tuberculosis) was confirmed in 5, borderline in 3, and obscure in 18 (69%). When the results of fluid samples and biopsy specimens were analyzed together, the diagnosis was definite in 10 out of 26 patients, giving a diagnostic yield of 38%. When the borderline or suspicious cases were included, the rate of diagnosis was 54% (14 out of 26). However, in those who had no biopsy analysis, the diagnostic yield was only 18% (3 out of 17). Etiology of the pericardial effusions is listed in Table 3, based on definite pathological diagnosis or conclusions from clinical, laboratory, or radiological findings. Malignancy was the most frequent cause, followed by uremia, tuberculosis, and collagen diseases. The causes remained unidentified in 7 (18%) patients, 5 of whom had blood-stained fluid drainage.
Sixteen of the patients were known to have died during the study period (Table 4) but some may have been lost to follow-up. Nine of the recorded deaths were in patients diagnosed with malignancy and all had bloodstained drainage. Three of those with pericardial effusion of unidentified cause also had bloodstained fluid and died during follow-up. One of these (case 5) was found to have carcinoma of the lung one year later, and died soon after. Two patients died on the 2nd day after SP. They were critically ill and had suffered cardiac arrest before drainage was attempted; the surgical procedures were successful with no complications and death was due to widespread malignancy in both patients. In addition to the patient with systemic lupus erythematosus who died immediately after SP, another patient with systemic lupus died 3 days after a successful and uncomplicated PP, due to multiple organ failure related to systemic lupus. The other known deaths occurred at least 2 weeks after the drainage procedure.
|
Discussion
|
|---|
The use of 2-dimensional echocardiography to select the optimal site of entry for PP was suggested by various authors and the criteria used to judge the best site were selected from these previous reports.2,3,10 The use of a needle attached directly to a transducer for continuous monitoring was abandoned because it was found to be time-consuming and unnecessary.2 Attachment of a clip to the needle for electrocardiographic monitoring had the same problems so it was not used. Drainage of fluid was monitored intermittently by 2-dimensional echocardio-graphy. The procedures were generally successful; only 1 patient had a failed puncture, probably because of marked dense deposits of fibrin. No complications or mortality could be attributed to PP. Pneumothorax was reported by Callahan and colleagues2 in a case where the apical approach was used. In our series, the apical approach was selected only in patients with effusions located predominantly in the apical region; in such cases, the effusion may displace lung tissue away from the entry site, avoiding this complication. The use of a pigtail catheter rather than a straight catheter was less traumatic and may be more advantageous.
Previously, an anterior thoracotomy and creation of a pericardial window into the pleural cavity was used as the standard approach for SP. In recent years, more reports favored the subxiphoid approach because of the dependent drainage and less chance of adhesion by lung tissue.5,6 It is also simple to perform and can be carried out under local anesthesia with sedation. Thus, it was frequently selected in this series. The thoracoscopic approach has no added advantage, it requires general anesthesia and double-lumen intubation that takes a much longer time. However, its use is worthwhile when the cause of a pleural effusion needs to be dealt with at the same time.
Most of the patients had bloodstained fluid and the majority were found to have a malignancy. In all those who had malignancy and later died, the fluid samples were bloodstained. In the recurrent case requiring 6 draining procedures, the fluid was bloodstained but the diagnosis (carcinoma of the lung) was ascertained only from biopsy of a coin lesion noted in a chest radiograph. Bloodstained fluid was also found in 3 patients who had unidentified causes of effusion and subsequently died. Therefore, it seems appropriate to assume bloodstained fluid to be of malignant origin until proven otherwise. Malignancy was the most common cause of pericardial effusion with a frequency ranging from 24% to 58% in previous reports.10–12
The second most common cause of pericardial effusion in this study was uremia. Most of the uremic patients (4 out of 5) underwent SP. No additional causes of effusion were found and it is likely that uremia was the primary cause in these patients, rather than secondary infections. The incidence of tuberculosis was more frequent than in Western studies, reflecting its prevalence in this region. Irradiation was noted among the causes of effusion in other studies but was not found in this series.11,12 This might be due to the small sample size or less frequent use of radiation therapy. No patient with effusion secondary to a bleeding tendency was encountered, in agreement with other studies.10–12
The diagnostic yield of samples of pericardial fluid in cases of malignancy and tuberculosis was low, in agreement with other reports. The rate of diagnosis improved when considered together with biopsy results. Thus, if the cause is unknown before drainage, biopsy should be considered. Malignancy accounted for 56% of the known mortality. Surgical pericardiotomy was performed in most of these cases, which explains the higher mortality associated with SP and also indicates that outcome after successful drainage depends on the natural course of the disease.
The patient with systemic lupus who died soon after PP, was found at postmortem examination to have hypertropic cardiomyopathy. It was postulated that diastolic function was probably impaired before the procedure and the heart had compensated for the existing cardiomyopathy and pericardial effusion but could not withstand the sudden decrease in diastolic pressure when 1000 mL of fluid was drained rapidly. This may be similar to the reported occurrence of pulmonary edema secondary to excessive fluid drainage.13 We have now adopted controlled drainage of pericardial fluid during PP (less than 300 mL at one time). This is not applicable to SP because once the parietal pericardium is cut, it is not easy to stop the fluid coming out rapidly. Therefore, it is recommended that in cases of severe tamponade, controlled percutaneous drainage should be chosen as the initial treatment rather than surgical pericardiotomy. It was concluded that drainage of pericardial effusion was effective and safe using the percutaneous and surgical techniques as described.
|
References
|
|---|
-
Lorell BH, Grossman W. Profiles in constrictive pericarditis, restrictive cardiomyopathy, and cardiac tamponade. In: Grossman W, Baim DS, editors. Cardiac catheterization, angiography, and intervention. 5th ed. Philadelphia: Lea & Febiger, 1996;787–800.
-
Callahan JA, Seward JB, Nishimura RA, Miller FA, Reeder GS, Shub C, et al. Two-dimensional echocardiographically guided pericardiocentesis: experience in 117 consecutive patients. Am J Cardiol 1985;55:476–9.[Medline]
-
Krikorian JG, Hancock EW. Pericardiocentesis. Am J Med 1978;65:808–14.[Medline]
-
Girardi LN, Ginsberg RJ, Burt ME. Pericardiocentesis and intrapericardial sclerosis: effective therapy for malignant pericardial effusions. Ann Thorac Surg 1997;64:1422–8.[Abstract/Free Full Text]
-
Campbell PT, van Trigt P, Wall TC, Kenney RT, O'Connor CM, Sheikh KH, et al. Subxiphoid pericardiotomy in the diagnosis and management of large pericardial effusions associated with malignancy. Chest 1992;101:938–43.[Abstract/Free Full Text]
-
Moores DWO, Allen KB, Faber LP, Dziuban SW, Gillman DJ, Warren WH, et al. Subxiphoid pericardial drainage for pericardial tamponade. J Thorac Cardiovasc Surg 1995;109:546–52.[Abstract/Free Full Text]
-
Permanyer-Miralda G, Sagrista-Sauleda F, Soler-Soler J. Primary acute pericardial disease: a prospective series of 231 consecutive patients. Am J Cardiol 1985;56:623–30.[Medline]
-
Merce J, Sagrista-Sauleda J, Permanyer-Miralda G, Soler-Soler J. Should pericardial drainage be performed routinely in patients who have a large pericardial effusion without tamponade? Am J Med 1998;105:106–9.[Medline]
-
Seldinger SI. Catheter replacement of the needle in percutaneous arteriography: a new technique. Acta Radiol (Stockh) 1953;39:368.
-
Levine MJ, Lorell BH, Diver DJ, Come PC. Implications of echocardiographically assisted diagnosis of pericardial tamponade in contemporary medical patients: detection before hemodynamic embarrassment. J Am Coll Cardiol 1991;17:59–65.[Abstract]
-
Guberman BA, Fowler NO, Engel PJ, Gueron M, Allen JM. Cardiac tamponade in medical patients. Circulation 1981;64:633–40.[Abstract/Free Full Text]
-
Wall TC, Campbell PT, O'Connor CM, Trigt P, Kenney RT, Sheikh KH, et al. Diagnosis and management (by subxiphoid pericardiotomy) of large pericardial effusions causing cardiac tamponade. Am J Cardiol 1992;69:1075–8.[Medline]
-
Glasser F, Fein AM, Feinsilver SH, Cotton E, Niederman MS. Non-cardiogenic pulmonary edema after pericardial drainage for cardiac tamponade. Chest 1988;94:869–70.[Abstract/Free Full Text]
TOP
Abstract
Introduction
