Recent Surgical Experience in Chronic Constrictive Pericarditis

Abstract

We reviewed the cases of 42 consecutive patients who underwent surgery for chronic constrictive pericarditis between January 1994 and January 2002, to determine causes, surgical techniques, mortality and morbidity rates, and improvement of functional capacity. Thirty-one of our patients were men (73.8%) and 11 were women (26.2%); their average age was 39.3 years. The pericarditis was tuberculous in 24 patients (57.1%), idiopathic in 10 (23.8%), rheumatic in 4 (9.5%), uremic in 2 (4.8%), and neoplastic in 2 (4.8%).

In all cases, our approach was via median sternotomy. All anterior pericardium was resected from phrenic nerve to phrenic nerve. In 4 of 18 patients with calcific pericardial plaques, we could not develop a cleavage plane, so we left the plaques in place, reducing their size and relieving constriction by means of wedge incisions that reached the epicardium. No patient required cardiopulmonary bypass due to severe calcification or adhesion, but we did perform bypass with no difficulty in 5 patients who required additional cardiac surgery.

The early postoperative mortality rate was 11.9% (5 patients). In the 1st postoperative month, functional capacity improved dramatically: the number of patients in New York Heart Association functional class IV moved from 18 (preoperatively) to 1; in class III, from 20 to 6; in class II, from 4 to 13; and in class I, from 0 to 22. We recommend the median sternotomy approach for chronic constrictive pericarditis and consider cardiopulmonary bypass safe to use in indicated cases. (Tex Heart Inst J 2003;30:27–30)

The clinical table of chronic constrictive pericarditis (CCP) was first described in 1842 by Cheever in his Observations on the Diseases of the Orifice and Valves of the Aorta, ¹ in which he reported that dangerous symptoms arose chiefly from the compression of muscle tissue by sticky material that surrounded the heart. The causative factors for CCP have changed over time. ² For many years, most cases were due to tuberculosis, but currently the rates of uremic, postoperative, and neoplastic constriction are increasing, while those of tuberculous and idiopathic constriction are decreasing. ^3,4 Today, pericardiectomy is a safe procedure with available techniques. In this retrospective study of 42 consecutive cases at our hospital, we investigated the causal factors of CCP, surgical techniques, morbidity and mortality rates, and improvement in functional capacity after discharge.

Patients and Methods

During the 8-year period between January 1994 and January 2002, we operated on 42 patients with a diagnosis of CCP. Thirty-one of them were men (73.8%) and 11 were women (26.2%). The average age was 39.3 years (range, 17 to 63 yrs). The most frequent complaints were dyspnea, weakness, fatigue, chest pain, and fever. The cause of the CCP was tuberculosis in 24 patients (57.1%), idiopathy in 10 (23.8%), rheumatic fever in 4 (9.5%), uremia in 2 (4.8%), and cancer in 2 (4.8%).

In all patients, auscultation revealed a pericardial friction rub (a metallic 3rd voice with high frequency, just after the 2nd voice in diastole). Physical examination further revealed ascites in 25 patients (59.5%), peripheral edema in 14 (33.3%), and hepatomegaly in 13 (30.9%). Laboratory results showed hypoproteinemia and hypoalbuminemia in all patients. Two-dimensional, color-flow Doppler echocardiography revealed cardiomegaly in 12 patients (28.6%) and pericardial calcification in 18 (42.9%). Tuberculosis was associated with this calcification in 16 of the 18 patients (88.8%). Upon electrocardiography, low voltage was a finding common to all patients. The patient with concomitant coronary artery disease exhibited negative ST and T waves, and 2 of 4 patients with valve disease had atrial fibrillation. We performed cardiac catheterization in 34 of the 42 patients and found the mean left atrial pressure to be 16.3 mmHg (range, 9 to 28 mmHg), the mean right ventricular end-diastolic pressure to be 20.1 mmHg (range, 18 to 27 mmHg), and the mean left ventricular ejection fraction to be 0.40 (range, 0.30 to 0.55).

None of the 42 patients needed pericardiocentesis. Preoperatively, all patients were taking diuretics and digitalis at optimal doses, and those with tuberculosis or cancer were also receiving agents specific to those conditions. Four patients had concomitant cardiac valve disease and 1 patient had coronary artery disease.

Operative Technique

In all patients, we approached via median sternotomy. We freed the pericardium in this order: first from the aorta and pulmonary artery, including the left ventricular outflow tract; then from the left and right ventricles and the left pulmonary vein orifices; and finally from the superior and inferior venae cavae. During these steps, we set the amplitude of the cautery under 60 mV to avoid causing diathermal dysfunction of the right ventricle during electrocauterization. To avoid damage to the phrenic nerves, we surgically resected the entire anterior pericardium, within 3 or 4 cm of the phrenic nerves, then completed the dissection of the pericardium from the diaphragm. There was no major bleeding from the pericardial edges, nor was there evidence of phrenic nerve injury either perioperatively or postoperatively. None of the 42 patients experienced ventricular fibrillation, so there was no need for cardioversion.

In 4 of the 18 patients with calcific pericardial plaques, we left large plaques that did not permit the development of a cleavage plane. In these cases, we performed wedge incisions that reached the epicardium, which reduced the size of the plaques and relieved myocardial constriction.

Cardiopulmonary bypass (CPB) was performed only in the 5 patients who required additional cardiac surgery; it was not needed as a consequence of severe pericardial attachment or calcification. The mean duration of CPB was 38 min (range, 28 to 56 min), and the mean quantity of mediastinal drainage was 380 cc (range, 325 to 550 cc). None of the patients who underwent CPB needed mediastinal re-exploration due to hemorrhage.

Early death was defined as death before discharge or during the 1st postoperative month. All statistical results are presented as mean values, without standard deviation. A P value of <0.05 was considered statistically significant and was calculated by the Cox proportional hazard method. ⁵

Results

Postoperative complications were arrhythmia, low cardiac output, acute renal failure, and acute respiratory distress syndrome (Table I). Our mortality rate for the early postoperative period was 11.9% (5 patients). The causes of early death were low cardiac output (2 patients), acute renal failure (2 patients), and acute respiratory distress syndrome (1 patient). Other factors contributing to early death and serious complications are shown in Table II. Many of the patients whom we characterized as being in “poor general condition before operation” were in New York Heart Association (NYHA) functional class III or IV. During the 1st postoperative month, the functional capacity of our patients improved dramatically (Table III).

TABLE I. Postoperative Complications and Their Rates in Our 42 Patients

TABLE II. Statistically Significant Factors Causing Severe Complications and Early Death

TABLE III. Comparison of Preoperative and Postoperative (within 30 Days) Functional Capacity in Our 42 Patients

In the 4 patients whose large calcific plaques had to be left in place, the wedge incisions successfully relieved myocardial constriction, and there was no late recurrence of CCP. Indeed, we found no postoperative recurrence of CCP in any of our patients during late follow-up (after 6 postoperative months).

Discussion

In chronic constrictive pericarditis, there are changes in the volume elasticity slopes of both ventricles. Secondary to constriction, there is an increase in left and right ventricular end-diastolic pressures and a decrease in outflow volume. ⁶ Moreover, thickened and calcified pericardium frequently comes into direct contact with the myocardium, decreasing the heart muscle's contractility and disrupting the coordination of diastolic filling of the ventricles. ⁷

In our series, tuberculosis is the most frequent cause of CCP, but in developed western countries cases of uremic, postoperative, and neoplastic pericarditis are increasing overall. ⁸ For both the effusive and constrictive forms of pericarditis associated with tuberculosis, early surgical intervention is indicated. ³ Fibrosis and calcification can extend very severely in tuberculosis patients, and this is the primary reason to perform the operation as early as possible. Chronic constrictive pericarditis of more than a year in duration decreases the success rate of surgery and increases the mortality rate. ⁷ Due to early diagnosis and surgery, severe cases have become rare. In patients with CCP, performing cardiopulmonary bypass is a big problem. We used CPB in 5 of 42 patients, and then only because it was necessary for the correction of concomitant cardiovascular disease.

Another difficulty in cases of CCP is hypoproteinemia as a consequence of enterogenous protein loss. Portal hypertension ruptures capillaries, returning protein to the lumen of the gut. As blood colloidal pressure decreases and hydrostatic venous pressure increases, peripheral edema develops. Protein loss, especially albumin loss, increases lymph production and causes secondary dilatation in the lymphatic system. High pressure in the dilated lymph vessels increases fat and protein loss through transudation or outright tearing of the vascular wall. All of these conditions are reversible if pericardial constriction is relieved. ⁹ Every patient in our series had hypoproteinemia and hypoalbuminemia preoperatively. Their rates of ascites, peripheral edema, and hepatomegaly were 59.5%, 33.3%, and 30.9%, respectively.

Whether pericardiectomy is indicated in cancer patients with CCP depends, of course, on the nature and extent of the underlying disease. ¹⁰ Often, pericardial or cardiac infiltration is inoperable. The chance of a cure is high with breast cancer and lymphoma, but CCP with lung cancer is generally inoperable even at diagnosis. We performed pericardiectomy in 2 CCP patients with cancer, but the disease was primary lymphoma.

Currently, the incidence of idiopathic pericarditis has decreased, due to the improvement of diagnostic techniques. Pericardiectomy for CCP has very low morbidity and mortality rates in cases of idiopathic, traumatic, and rheumatic CCP. ¹¹ In our series, CCP cases with these causes were associated with no operative or postoperative morbidity or death.

Chronic constrictive pericarditis after open-heart surgery was reported first in 1972 by Kendall. ¹² The development period for constrictive pericarditis after coronary artery bypass averages 2 years. ¹³ Fewer than 5% of CCP cases are caused by surgery, and in our series we didn't encounter a CCP case with this origin. The number of cases of CCP caused by mediastinal radiotherapy has increased in recent years, and the development period is between 8 and 16 months; in these cases, cardiopulmonary bypass is necessary because of severe attachments. ¹⁴

Bilateral thoracotomy, left anterior thoracotomy, and median sternotomy are all approaches that can be used in treating CCP. ¹⁵ The sternotomy incision enables exploration of the left ventricle and right part of the heart and direct vision of the great vessels. It is possible to perform an extensive pericardiectomy with minimal cardiac manipulation. The left ventricle can be decorticated easily via sternotomy, and better cardiac hemodynamics can be achieved through a sternotomy than through a thoracotomy. Further advantages of the median sternotomy are the ease and comfort of this approach for the surgeon, the ability to initiate CPB if necessary, and the postoperative comfort of the patient; for these reasons, we used this technique in all our cases. On the basis of information in the literature, ¹⁴ we excised the whole of the anterior pericardium between the phrenic nerves.

In some cases of CCP surgery, the importance of CPB is evident. If there is a calcific and thickened pericardial fragment without a cleavage plane, it can be left in place; but if the surgeon wants to excise the thickened part, CPB should be performed after cannulation of the femoral artery and vein under normothermic conditions (37 °C). Large series ^3,4,10 have shown that CPB does not place the patient at serious risk for early death or severe complications. If there is low cardiac output before the operation, primary treatment with catecholamine infusions should be started 24 to 48 hours before the operation. In unresponsive cases, an intraaortic balloon pump can be tried, as in our 3 patients.