Percutaneous Balloon Pericardiotomy in Patients with Recurrent Pericardial Effusion

INTRODUCTION

Recurrent pericardial effusion is often caused by pericardial metastasis of extracardiac tumors. These effusions are generally copious hematic effusions that often lead to cardiac tamponade, which requires emergency drainage by pericardiocentesis. The recurrence rate of pericardial effusion after drainage is high, particularly in effusions of tumoral origin (13%-50%).¹ In patients with recurrent pericardial effusion, the treatment traditionally recommended includes, pericardiocentesis with instillation of sclerosing substances or chemotherapeutic agents,^1,2 opening a surgical pericardial window by a subxiphoid approach,³ or pericardiotomy.⁴ However, since patients generally have serious clinical conditions and impaired general status, it would preferable to avoid the added risk of surgical intervention and general anesthesia. This motivated Palacios et al⁵ to develop the percutaneous balloon pericardiotomy technique.

PATIENTS AND METHOD

Population

To date, we have performed percutaneous balloon pericardiotomy in 5 patients. The clinical characteristics of the patients are summarized in Table 1. In every case, pericardiotomy was performed in patients with severe pericardial effusion and clinical and echocardiographic signs of cardiac tamponade in which at least one recurrence had taken place after previous drainage. In 4 patients, the cause of the effusion was tumoral (Table 1), but in 1 patients with severe hematic effusion the results of cytology and a smear were negative.

Procedure protocol

The procedure was carried out in the hemodynamics laboratory. After local anesthesia of the skin and subcutaneous cellular tissue, pericardial puncture was performed in the subxiphoid region under radiological control. A 0.035-inch vascular guide was inserted and the pericardial puncture needle was removed. A dilator was then passed and a drainage catheter was inserted over the guide. Samples were obtained for laboratory and histology studies. Part of the pericardial fluid was drained if necessary to stabilize the patient. Contrast was injected into the drainage catheter to highlight the parietal pericardium (Figure 1). The drainage catheter was replaced by an introducer (9-11 Fr), then a pediatric valvuloplasty balloon (diameter 20 mm, length 30 mm) was advanced through a 0.038-inch vascular guide (Figure 2). The balloon was placed in the parietal pericardium under radiological control and was slowly inflated, confirming that the pericardium made a notch in the center of the balloon (hourglass image). Inflation pressure was then increased until the notch disappeared (Figure 3). The procedure was guided exclusively by fluoroscopy. Echography was performed at the end of the procedure to confirm the complete drainage of the pericardial effusion. After draining the pericardial fluid, a drain was left in place until the next day (as long as less than 100 ml was drained over the next 24 h). All patients received antibiotic prophylaxis with cloxacillin.

Fig. 1. Introducer in the pericardium through which contrast has been injected to locate the parietal pericardium.

Fig. 2. Balloon inserted in the parietal pericardium through a 0.035-inch guide. Before inflating, the introducer is removed slightly.

Fig. 3. Progressive inflation of the balloon until the notch disappears, observing the typical hourglass image.

RESULTS

In the 5 patients treated, the procedure was successful in a single session with one pericardial puncture. In 3 patients, a sufficient opening was obtained by inflating the balloon once. One patient required 3 inflations for correct balloon placement and pericardial opening. In one patient, the balloon could not be expanded completely with 4 inflations, but there was no later recurrence of pericardial effusion. The procedure took 80 min in the first case and 55-65 min in the following cases. In every case, the pericardial fluid was hematic and varied in volume from 700 to 1700 ml. Inflation of the balloon was accompanied in every patient by mild, transitory discomfort that disappeared immediately after deflating the balloon. Most patients had some degree of cardiac tamponade that immediately improved with drainage. Almost complete evacuation of the pericardial effusion was achieved in all patients. There were no post-procedure complications in any patient. Pleural effusion occurred in all the patients in the next 24-48 h, and resolved spontaneously in all cases (Figure 4).

Fig. 4. Patient with post-pericardiotomy pleural effusion that resolved spontaneously in a week.

FOLLOW-UP

During follow-up (4 to 20 months), 2 patients died from their malignancy. The time from pericardiotomy to death was 1 and 9 months, respectively. Pericardial effusion did not recur in any of the 5 patients treated and no patient has had cardiac tamponade. The 3 patients who remain alive, some of who have been followed-up for almost 2 years, have not been readmitted to the hospital for pericardial problems and show no signs of constrictive pericarditis.

DISCUSSION

Pericardial effusion associated with pericardial metastasis of extracardiac tumors often leads to cardiac tamponade. Recurrence after pericardiocentesis is frequent.¹ The therapeutic options for the treatment of these patients, such as instillation of sclerosing substances or chemotherapeutic agents in the pericardium, are associated with high recurrence rates and patient discomfort.^2,6 Another option for these patients is surgery to create a pericardial window. Some of these techniques have the drawback of requiring general anesthesia, as in the case of thoracotomy to open a pleuropericardial window or pericardiectomy. This is an added risk in patients who are generally gravely ill. Using a subxiphoid approach to create a pericardial window has the advantage of requiring only local anesthesia, a low complication rate, and it allows the pericardium to be sampled.⁷ For these reasons, it has been the preferred technique in recent years. Percutaneous balloon pericardiotomy is another therapeutic alternative in these patients. It can be viewed as a percutaneous variant of the subxiphoid approach to creating a surgical window, but it is even less traumatic. Our experience confirms that it is a simple procedure that can be performed in seriously ill patients (cardiac tamponade) with a low rate of complications and little discomfort to the patient.

The mechanism of pericardial drainage is not entirely clear. All our patients had left pleural effusion after the procedure, so we assume that pericardial effusion does not accumulate because it drains into the pleural cavity. Drainage may also occur to the peritoneum or the parietal and visceral pericardium may fuse as a result of an inflammatory reaction to pericardiotomy.⁸

The complications of percutaneous pericardiotomy are similar to those of pericardiocentesis in general. Pericardial puncture is usually simple in these patients, with little risk of cardiac perforation due to the abundant pericardial effusion. Fever did not appear after the procedure in any of our patients, in contrast with other series,⁵ probably because they were treated prophylactically with cloxacillin. Pleural effusion was the most frequent complication, with all our patients presenting it to some degree. The degree of pleural effusion is probably related with the quality of pleural drainage in these patients, which depends, in turn, on whether or not they have received thoracic radiotherapy or have tumoral pleural involvement and impaired lymphatic drainage. In some series, pleural drainage had to be performed to manage pleural effusion.⁵ Pleural effusion resolved spontaneously in our patients.

Pericardiotomy may be performed as the initial therapy at the same time as the first pericardiocentesis. The advantages of this approach are that it eliminates the risk of a second pericardial effusion and eventual cardiac tamponade, which usually requires emergency pericardiocentesis. Up until now, we have always performed pericardiotomy as a separate procedure after at least one recurrence of pericardial effusion. The reason for this was that the diagnosis of tumoral pericardial effusion had not been established at the time of pericardiocentesis in some patients. By deferring pericardiocentesis, we avoided pericardiotomy in 3 cases (all of tumoral origin) who had no recurrence of effusion after the initial pericardiocentesis. In addition, due to close echocardiographic follow-up of our patients after pericardiocentesis, pericardial effusion occurred, but not cardiac tamponade. This allowed pericardiotomy to be performed electively in a stable patient.

CONCLUSIONS

We found percutaneous balloon pericardiotomy to be a simple and safe procedure. This procedure allows recurrent pericardial effusion to be treated with a high rate of success and scant complications. In addition, pericardial effusion did not recur within a relatively long time. The long-term prognosis of these patients depends mainly on their underlying disease.

Correspondencia: Dr. A. Iñíguez Romo.
Unidad de Hemodinámica y Cardiología Intervencionista.
Servicio de Cardiología. Fundación Jiménez Díaz.
Avda. Reyes Católicos, 2. 28040 Madrid. Spain
rreo electrónico: [e-mail]