Video-Assisted Thoracic Surgery for Delayed Pericardial Effusion Post-CABG

and none had their conditions diagnosed preoperatively. The final diagnosis was based on direct microscopy or histologic examination of resected tissue. George et aJl 0 indicated that granulomatous pleuritis comprised approximately 10 percent of pleural effusions and that a differential between mycobacteria, fungi, and nocardia was usually required for differentiation. They emphasized the value of pleural biopsy and serologic techniques. In the case we are presenting, the patient had a productive cough and shortness of breath with subsequent finding of a pleural effusion. While the effusion was clearly an exudate, no organism was isolated from pleural fluid by smear or culture and pleural biopsy specimens were nonspecific. Bronchoscopy subsequently demonstrated the presence of sulfur granules and demonstration of the organism on Gram stain. Culture was subsequently positive. No granules were seen in the oral cavity and there was no evidence of contamination as a source. This case is unique in that the patient had been ill for many weeks with only pleural effusion and no formation of the expected sinuses or abscesses that are usually noted by this time in the course of actinomycosis involving the chest. Symptoms resolved quickly following institution of penicillin therapy. The purulent sputum also cleared with the penicillin therapy. Recommended therapy includes penicillin G, 10 to 20 million U/ d intravenously for 4 to 6 weeks, followed by penicillin V, 2 to 4 g/d orally for 6 to 12 months, or ampicillin, 50 mg/kg/d intravenously for 4 to 6 weeks, then 0.5 g of amoxicillin orally three times a day for 6 months.

Video-Assisted Thoracic Surgery for Delayed Pericardial Effusion Post-CABG* john P. Hurley, M.D.;f Konda Subarreddy, M.D.; jim McCarthy, M.D.; and Alfred E. Wood, M.D.

Delayed-onset pericardia! effusion following coronary artery bypass grafts can give rise to significant morbidity in its presentation and in its management by traditional surgical techniques. A video-assisted thoracoscopic technique to create a pericardia! window, with the advantage of a minimally invasive approach combined with excellent visualization in such a patient is (Chest 1994; 106:1617-19) described. CABG=coronary artery bypass graft; LIMA=left internal mammary artery; VATS=video-assisted thoracic surgery

was initially developed as a diagnostic tool T horacoscopy for intrathoracic disease; however, recent technical

advances in video camera technology have led to a resurgence of interest in thoracoscopy for both diagnostic and therapeutic uses. Video-assisted thoracic surgery (VATS) allows intrathoracic operations to be performed without formal thoracotomy_! It allows visualization and assessment of the pericardium. While surgical approaches to drainage of pericardia! effusions have been described since 1829, 2 the ideal surgical management of pericardia! effusions remains controversial in both approach and extent of resection. VATS may become another technique to deal with this condition of diverse etiologya

REFERENCES

CASE REPORT

1 Conant EF, Wechsler RJ, Actinomycosis and nocardiosis of the lung. J Thorac Imag 1992; 7:75-84 2 Hamer DH, Schwab LE, Gray R. Massive hemoptysis from thoracic actinomycosis successfully treated by embolization. Chest 1992; 101:1442-43 3 O'Sullivan RA, Armstrong JG, Rivers JT, Mitchell CA. Pulmonary actinomycosis complicated by effusive constrictive pericarditis. Aust N Z JMed 1991; 21:879-80 4 Coodley EL. Actinomycosis: clinical diagnosis and management. Postgrad Med 1969; 46:73-9. 5 Fife TD, Finegold SM, Grennan T. Pericardia! actinomycosis: case report and review. Rev Infect Dis 1991; 13:120-26 6 Ariel I, Breuer R, Kamal NS, Ben-Dov I, Mogel P, Rosenmann E. Endobronchial actinomycosis simulating bronchogenic carcinoma: diagnosis by bronchial biopsy. Chest 1991; 99:493-95 7 Severo LC, Kaemmerer A, Camargo JJ, Porto NS. Actinomycotic intracavitary lung colonization. Mycopathologia 1989; 108:1-4 8 Klapholz A, Talavera W, Rorat E, Salsitz E, Widrow C. Pulmonary actinomycosis in a patient with HIV infection. Mt Sinai J Med 1989; 56:300-03 9 Jensen BM, Kruse-Anderson S, Andersen K. Thoracic actinomycosis. Scand J Thorac Cardiovasc Surg 1989; 23:181-84 10 George RB, Penn RL, Kinasewitz GT. Mycobacterial, fungal, actinomycotic, and nocardia! infections of the pleura. Clin Chest Med 1985; 6:63-75

A 58-year-old man was admitted to the hospital with unstable angina. He was a nonsmoker but had a strong family history of coronary artery disease and a moderately raised cholesterol level. Coronary angiography demonstrated significant three-vessel coronary artery disease with minor dyskinesia of the inferior surface and the anterior free wall of the left ventricle. The patient was referred for coronary artery bypass graft (CABG) surgery. At operation the left internal mammary artery (LIMA) was harvested and both pleura were opened. The patient had a saphenous vein graft placed to the posterior descending branch of the right coronary artery and another to the first marginal branch of the circumflex system. The LIMA was used to bypass the critical proximal lesion in the left anterior descending artery. The pericardium, which had been opened through a vertical midline pericardiotomy, was not closed prior to closure of the sternum. Three chest drains were used, one in the pericardium and one in each of the pleural cavities. Postoperatively, the patient did well, remaining 1 day in intensive care; his chest drains were removed on the first postop*From the Department of Cardiothoracic Surgery, Mater Hospital, Dublin, Ireland. fCurrently at Dept. of Thoracic and Cardiovascular Surgery, The Cleveland Clinic Foundation. Reprint requests: Dr. Hurley, Department of Thoracic and Cardiovascular Surgery (Desk F25), The Cleveland Clinic Foundation, 4500 Euclid Avenue, Cleveland, OH 44195 CHEST /106/5/ NOVEMBER, 1994

1617

FIGURE l. Chest radiograph prior to VATS. erative day. Serial ECGs and daily estimates of cardiac isoenzymes showed no evidence of myocardial infarction and the patient was discharged home on the tenth postoperative day, taking aspirin, 300 mg daily, and acetaminophen (Paracetemol) for analgesia as required. He again presented 5 weeks later with vague symptoms of dyspnea on exertion and weakness. On examination he was apyrexial, had a sinus tachycardia of 100 beats/ min and an arterial blood pressure of 100/ 70 mm Hg. The jugular venous pulse was not elevated and the lung fields were clear. A chest radiograph revealed a globular-shaped heart with cardiomegaly (Fig 1). A significant pericardia! effusion with impaired diastolic function of the left ventricle was demonstrated by echocardiography. Results of hematologic investigations were unremarkable. Pericardia! drainage of 1 L of straw-colored fluid was performed under echocardiographic control on two separate occasions with initial symptomatic relief. Reaccumulation of the fluid and recurrence of the symptoms occurred within 1 week on both occasions despite the addition of hydrocortisone and indomethacin. The protein content of the fluid was 61 g/ L; cytologic study demonstrated reactive cells and no organisms were cultured. The patient underwent VATS to create a pericardia! window.

Technique The operation was performed under general anesthesia with double-lumen endotracheal intubation. Correct position of the endotracheal tube was confirmed with a flexible pediatric bronchoscope. The patient was placed in the left lateral position and draped in the standard fashion for formal thoracotomy. Mechanical ventilation of the right lung was discontinued. Even though preoperative echocardiography had revealed a large diffuse effusion, it was decided to use a right-sided approach. Caccavale, 4 in a previous description of VATS for a pericardia! window, advocated a right-sided incision because the pericardia! sac, when distended with fluid, tends to fill the left hemithorax making visualization difficult due to the proximity of the scope to the pericardium. He also noted that the pericardia) fat, which is well vascularized, tends to be more extensive over the bulging left

1618

ventricle. His series, however, did not include postcardiac surgical patients. In a post-CABG patient, the placement of the grafts at the original operation is a major factor to be considered before attempted drainage of the pericardium, and our main reason for adopting a right-sided incision was that the adhesions were likely to be more dense on the left side with the lung adherent to the chest wall following harvest of the LIMA graft. Video-optics consisted of a 10-mm 0° panoramic view diagnostic telescope, a camera, camera head, and video monitor. An initial2-cm incision was made in the fifth intercostal space in the midaxillary line for insertion of the thoracoscope and camera through a 10-mm thoracoport (cylindrical plastic tube). Exploratory thoracoscopy was then performed, which revealed adequate deflation of the right lung, a small pleural effusion, and pleural adhesions to a tense pericardium. A second thoracoport was placed in the fifth space anterior axillary line under direct vision. The right lung was easily separated from the pericardium with gentle traction with a grasping forceps passed through the anterior thoracoport. With the excellent view of the pericardium afforded by the video telescope, damage to an underlying coronary graft should be unlikely, particularly where the effusion is large, generalized, and the pericardium is tense. However, we advocate aspiration of a few milliliters of fluid with a long 18-gauge needle prior to making an initial incision in the pericardium at the site of the aspiration, as it is the initial incision in the pericardium that is most hazardous in the patient post-CABG. The presence of a pericardia! effusion was reconfirmed by aspiration with a long 18-gauge cannula, prior to creation of the window. Using a sharp-pointed scissors passed through the anterior thoracoport, and counter traction with a grasping forceps passed directly through the chest wall in the seventh intercostal space midaxillary line, an 8X1-cm incision was easily made in the pericardium anterior and parallel to the phrenic nerve. Bleeding from pericardia! vessels was minimal and easily controlled with diathermia. There was free drainage of pericardia! fluid through this window. Two chest drains were placed through the operating sites, one directly into the pericardia! cavity, and the camera was removed and its insertion site closed. Drainage gradually decreased from 350 ml/ day and the drains were removed after 1 week. Repeated echocardiography demonstrated a clear pericardium with no reaccumulation of fluid and good left ventricular function. The patient was discharged home receiving no medication and remains well 6 months later. DISCUSSION

The optimal surgical management of pericardia! effusions remains controversial. Approaches advocated include the subxiphoid incision and pericardia! window,5 left anterior thoracotomy or median sternotomy with creation of a pericardia! window, and varying degrees of pericardiectomy.6 Excellent results are achievable with all of these methods 7' 8 and the surgical option used is often dictated by the etiology of the effusion and the condition of the patient. Recently, with the resurgence of interest in thoracoscopy and the advent of VATS, drainage of pericardia! effusions by video-assisted thoracoscopic techniques has been described.4 However, this technique has not been reported previously for drainage of postoperative effusions following coronary artery bypass surgery. Although rare in comparison to the incidence of pericardia! effusions associated with the postpericardiotomy syndrome, the delayed appearance of a pericardia! effusion after cardiac surgery has been reported previously 9 and can lead to significant morbidity. 10 Management of the

Video-Assisted Thoracic Surgery for Delayed Pericardia! Effusion Post-CABG (Hurley eta/)

problem has followed the traditional surgical approaches of subxiphoid incision or thoracotomy, in addition to drainage under echocardiographic guidance. The options first considered in our patient were steroids to cover the possibility of a delayed postpericardiotomy syndrome and drainage under echocardiographic control. While this echocardiographic drainage was successful on both occasions, it was followed by rapid reaccumulation of fluid. It was decided to create a pericardia! window by VATS rather than place another percutaneous drain or follow one of the more traditional surgical approaches to the subxiphoid incision or thoracotomy. Advocates of the subxiphoid approach list as its chief advantages that it can be performed under local anesthetic and that it maintains dependent drainage of the pericardium . In our experience, placement of a subxiphoid drain under local anesthetic can be difficult in a patient following a recent sternotomy, and in addition, our patient was in good medical condition with no contraindication to general anesthesia. While dependent drainage may be a factor in decreasing the incidence of recurrence, VATS affords excellent visualization of the pericardium and allows accurate placement of the pericardia! drain. This accurate placement of an intrapericardial drain at thoracoscopy ensures continued decompression of the pericardium and facilitates the formation of intrapericardial adhesions, and these may reduce the recurrence rate. This factor may be of more importance than the dependent drainage facilitated by the subxiphoid approach. Proponents of thoracotomy drainage claim that the better visualization of the pericardia! sac at thoracotomy facilitates breakdown of intrapericardial adhesions and drainage of loculated fluid. Also, this together with the opportunity to widely excise pericardia! tissue should result in a lower reaccumulation rate. However, the extent of pericardia! resection required to prevent recurrence of the effusion is controversial. The diversity of causes of pericardia! disease may explain the wide range of reported recurrences with the transthoracic approaches in previous series. The extent of resection required probably depends largely on the etiology of the effusion and in cases where the etiology is in doubt, wider excision of the pericardium may correctly be the more prudent option. In the patient post-CABG, it is possible with VATS to safely create an adequate pericardia! window anterior to the phrenic nerve to decompress the pericardium without the morbidity of a thoracotomy. In addition, a left anterior thoracotomy following CABG entails dissection through adhesions formed following harvest of a LIMA graft. This difficulty is obviated by a right-sided VATS. While the development of thoracoscopy and VATS continues, at this early stage, we consider its chief advantages in the drainage of late pericardia! effusions following CABG over the more traditional surgical approaches to be the excellent visualization of the thoracic cavity and pericardium afforded without the morbidity of a thoracotomy. In addition, it allows the creation of an adequate pericardia! window to decompress the pericardia! sac and facilitates accurate placement of a pericardia! drain. However, we advise careful planning, taking particular note of the position of grafts and the extent of the effusion.

REFERENCES

1 Lewis RJ, Caccavale RJ, Sisler GE, MacKenzie JW. One hundred consecutive patients undergoing video-assisted thoracic operations. Ann Thorac Surg 1992; 54:421-26 2 Larrey DJ. New surgical procedure to open the pericardium in case of fluid in its cavity. Clin Chir 1829; 2:303-30, 315-22 3 BlakeS, Bonar S, O 'Neill H, Hanly P, Drury I, Flanagan M, et al. Aetiology of chronic constrictive pericarditis. Br Heart J 1983; 50:273 4 Caccavale RJ. Video assisted thoracic surgery for pericardia) disease. Chest Surg Clin North Am 1993; 3:271-81 5 Fontanelle L, Cuello L, Dooley B. Subxiphoid pericardia) window. Am J Surg 1970; 120:679-80 6 Piehler J, Pluth J, Schaff H, Danielson G, Orszulak T, Puga F. Surgical management of effusive pericardia! disease. J Thorac Cardiovasc Surg 1985; 90:506-16 7 Sugimoto J, Little A, Ferguson M, Borow K, Vallero D, Staszak R, et al. Pericardia! window: mechanism of efficacy. Ann Thorae Surg 1990; 50:442-45 8 Miller J, Mansour K, Hatcher C. Pericardiectomy: current indications, concepts and results in a university center. Ann Thorac Surg 1982; 34:40-5 9 Ofori-Krakye S, Tyberg T, Geha A, Hammond G, Cohen L, Langou R. Late cardiac tamponade after open heart surgery: incidence, role of anticoagulants in its pathogenesis and its relationship to the post pericardiotomy syndrome. Circulation 1981; 63:1323-28 10 Merrill W, Donahoo J, Brawley R, Taylor D. Late cardiac tamponade: a potentially lethal complication of open heart surgery. J Thorac Cardiovasc Surg 1976; 72:929-32

Massive Hemoptysis in Takayasu's Arteritis During Pregnancy* M. Patricia Rocha, M.D. ; Kalpalatha K. Guntupalli, M.D., F.C.C.P. ; Kenneth]. Moise , Jr. , M.D.; Lillian D. Lockett, M .D.; Fadi Khawli, M.D.; and Roxann Rokey, M.D.

Takayasu's arteritis is an uncommon condition affecting predominantly young women. Because the disorder affects women in childbearing age, it may be recognized the first time during pregnancy. Various cardiovascular events may occur in the perinatal period. We describe a patient with Takayasu's arteritis who presented with massive hemoptysis. To our knowledge, this manifestation has not been documented previously. (Chest 1994; 106:1619-22) TA=Takayasu's arteritis

*From the Departments of Internal Medicine, and Obstetrics and Gynecology, Baylor College of Medicine, Houston. Reprint requests: Dr. Guntupalli, Ben Taub General Hospital, 1504 Taub Loop, Pulmonary 6th Floor, Houston, TX 77030 CHEST 11 06 I 51 NOVEMBER, 1994

1619