P process or in association with progressive cardiac

Transcription

1 Chronic Cardiac Rejection Masking as Constrictive Pericarditis Thomas J. Hinkamp, MD, Henry J. Sullivan, MD, Alvaro Montoya, MD, Soon Park, MD, Linda Bartlett, RN, and Roque Pifarre, MD Loyola University Medical Center, Maywood, Illinois The hemodynamic changes consistent with constrictive pericarditis are often encountered in patients who have undergone cardiac transplantation. We describe here 4 patients who underwent pericardiectomy after cardiac transplantation. All were found to have evidence of a thickened and constricting peel of pericardium at surgical exploration. Their postoperative clinical courses were variable. One patient with primarily effusive constriction experienced marked improvement. Three patients failed to show clinical improvement and had persistently elevated atrial and ventricular end-diastolic pressures. A coexisting restrictive cardiomyopathy secondary to chronic rejection, coronary arteriopathy, or long-standing constriction may have been the cause of this poor outcome. Many patients with transplanted hearts exhibit evidence of poor diastolic ventricular compliance without evidence of classic constriction; some manifest both the restrictive and constrictive components. The careful selection of patients with constrictive pericarditis can optimize the outcome. (Ann Tkorac Surg 1994;57: ) ericarditis can occur either as a primarily effusive P process or in association with progressive cardiac constriction, and pericardiectomy is an effective surgical treatment. The overall incidence of constrictive pericarditis and the need for pericardiectomy have decreased since the advent of effective treatment for tuberculosis, but pericardiectomy is still frequently performed for the treatment of various other causes such as viral or bacterial infection as well as the effects of irradiation and of previous cardiac operation. The incidence of symptomatic constrictive pericarditis after open heart procedures is reported to be about 0.2% [l, 21 and a significant improvement in the functional status and relief of symptoms has been observed after pericardiectomy [3]. In 1986, a case report was published that described a patient who underwent pericardiectomy for the treatment of constrictive pericarditis after heart transplantation, with good results [4]. We report here on 4 patients who underwent pericardiectomy for the treatment of a constrictive pericarditis that developed after cardiac transplantation. This study was undertaken to improve our understanding of constrictive physiology post transplantation, as well as to identify the predictors of improved functional status after pericardiectomy in these patients. Material and Methods The medical records, x-ray studies, pathologic findings, and cardiac catheterization data from heart transplantation patients who underwent pericardiectomy at Loyola University Medical Center from January 1, 1986, through Accepted for publication Oct 18, Address reprint requests to Dr Hinkamp, Loyola University Medical Center, 2160 South First Ave, Maywood, IL November 1, 1992, were reviewed. Patients who had a pericardial window or who underwent pericardial biopsy or drainage of pericardial effusion were excluded. Classic postoperative acute tamponade from pericardial fluid developed in 18 other cardiac transplantation patients. Four patients fulfilled the study criteria and formed the basis of this report (Table 1). The preoperative symptoms; the findings from right and left heart catheterization, echocardiography, computed tomographic scans, right heart biopsy, and pathologic studies; and the data on the operative techniques and complications were recorded. The follow-up New York Heart Association (NYHA) functional class was available for 2 patients. The explanted heart was available for pathologic examination in 1 patient who underwent redo transplantation 2 days after pericardiectomy. Results Pericardiectomy was performed through a median sternotomy in all 4 patients. Standard cardiovascular monitoring devices were used, including a radial artery pressure line and Swan-Ganz catheter. The surgical dissection was begun where an early plane could be discerned, usually at the level of the diaphragm. The pericardium was removed from the diaphragmatic surface and both ventricular surfaces, and extended laterally to the phenic nerves. All patients underwent posterior dissection to remove adherent tissue around the pulmonary vein and posterior myocardium. One patient required the institution of cardiopulmonary bypass for the management of cardiac decompensation at pericardiectomy, and could not be weaned from the pump. He was supported with a Symbion (Salt Lake City, UT) biventricular assist device and an intraaortic balloon pump for 6 days without by The Society of Thoracic Surgeons /94/$7.00

2 1580 HINKAMP ET AL Ann Thorac Surg 1994; Table 1. Patient Profiles and Outcome No. of Time to Patient Rejection Pericardiectomy No. Age (y) Sex PPP Episodes (mo) Results 1 20 F Retransplantation 2 days later, lived 3 y, died 1989 of allograft failure 2 45 M BIVADIIABP, died 7 days later 3 58 M NYHA class IV, home dopamine treatment 4 59 M NYHA class I BIVAD = biventricular assist device; F = female; IABP = intraaortic balloon pump; M = male; NYHA = New York Heart,Association; PPP = previous pericardial procedure (drainage or reexploration) improvement, and died when the support was withdrawn. The overall incidence of constrictive pericardial disease in 295 patients who underwent heart transplantation was 1.3%. The onset of constriction averaged 9 months from the time of transplantation. Clinical presentations at the time of pericardiectomy included dyspnea on exertion, and congestive heart failure, which were present in all 4 patients. Three patients had evidence of hepatomegaly and peripheral edema. Echocardiography demonstrated some pericardial thickening in all patients and considerable pericardial effusion in 3. Computed tomographic scans demonstrated evidence of pericardial thickening in all but the last patient, who had a predominantly effusive disease. Myocardial biopsy specimens obtained at right heart catheterization before pericardiectomy showed no evidence of rejection in any of the 4 patients. The classic constrictive physiologic features were present in all 4 patients. Elevation and near equalization of diastolic pressures occurred in all patients. The classic ventricular diastolic pressure tracings consisting of rapid descent followed by an early pressure rise in ventricular filling (square root sign) existed in 2 patients. Despite the clearly demonstrated constrictive pericardial physiologic characteristics in all 4 patients, only 1 patient experienced substantial improvement in clinical course and hemodynamic variables after the pericardiectomy (Table 2). This patient presented with pericardial constriction and effusion within 2 months of cardiac transplantation. During operation, incidental Staphylococcus aureus was cultured in the effusion. Table 2. Right Heart Catheterization Data Before and After Pericardiectomv Patient No. Variables CI (L. min- * m- ) HR (beatslmin) SVI CI (L. min-. m- ) HR (beatslmin) SVI CI (L. min-. m- ) After Cardiac Before After Transplantation Pericardiectomy Pericardiectomy /15 21 W W /24 46/30 36 W /15 52/24 20 W /26 36 W /10 17 W /19 27 W a Postpericardiectomy data obtained 1 week postoperatively. Postpericardiectomy data obtained 1 month postoperatively. CI = cardiac index; HR = heart rate; LVEDP = left ventricular end-diastolic pressure; M = mean pulmonary artery pressure; PA = pulmonary artery pressure; RA = right atrial pressure; RV = right ventricular pressure; SVI = stroke volume index; W = wedge pressure.

3 Ann Thorac Surg 1994; HINKAMP ET AL 1581 Symptoms of pericardial constriction developed in our first patient 7 months postoperatively. At operation she was found to have pericardial thickening predominantly over the right heart. Despite the rather complete decortication, the underlying right ventricular contractility was extremely poor. She had exhibited no hemodynamic improvement and was scheduled for cardiac retransplantation. She underwent retransplantation 2 days later, and is alive and well as of the time of this report 3 years later. Review of 23 myocardial biopsy specimens obtained in this patient revealed recurrent acute rejection on six different occasions. She was treated at different times with antithymocyte globulin, antihuman mature T cells, and methotrexate. The explanted heart from the redo transplantation showed extensive myocardial fibrous vacuolization of the myocytes that was more prominent in the right heart, and a large right atrial thrombus. Also noted was extensive lymphocyte invasion suggesting chronic rejection. The second patient presented 2 years after cardiac transplantation with symptoms of cardiac failure in NYHA class IV. Cardiac catheterization revealed constrictive physiology. For approximately a year, he was found to have moderate pericardial effusion on repeated echocardiograms and a thickened pericardium on computed tomographic scans. He had suffered only one episode of rejection. At pericardiectomy a thickened constrictive pericardium was evident over both the right and left ventricles. During pericardiectomy profound cardiac failure developed and cardiopulmonary bypass had to be instituted. He could not be weaned from the bypass and was placed on Symbion ventricular support for 6 days with no myocardial improvement. His family did not wish for him to undergo retransplantation and the support was withdrawn after which the patient died. Long-standing constriction may have adversely affected his myocardial function. No autopsy was performed. His hemodynamic tracings are depicted in Figure 1. The third patient presented with constrictive pericardial disease 4 months after the transplantation. Unlike the usual transplantation patient, his pulmonary artery pressures never returned to normal after heart transplantation. Over the next several months equalization of the diastolic pressures developed, and a computed tomographic scan demonstrated a thickened pericardium. Despite the pericardiectomy, he experienced minimal clinical improvement. Some hemodynamic improvement was noted 1 month after pericardiectomy, but was short-lived, with persistent elevation of the right atrial, pulmonary artery, and wedge pressures to 29, 47/24, and 21 mm Hg, respectively, at 4 months (see Table 2). Cardiac catherization at 10 months revealed continued elevation of the diastolic pressures with a poor cardiac index. He is currently in NYHA class IV with mild renal insufficiency and is on intravenously administered dopamine therapy at home. Constrictive pericarditis with effusion developed in the fourth patient 2 months after transplantation. Repeated pericardial paracenteses were ineffective in draining the loculated effusions and eventually congestive heart failure Fig 1. (Patient 2.) Left ventricular (LV), right ventricular (RV), and central venous pressure tracings with prominent X and Y descents. (ECG = electrocardiogram.) developed. Dramatic improvements in the clinical and hemodynamic variables were noted after pericardiectomy. He is currently in NYHA class I. Comment Cicatricial thickening of the pericardium was described by Galen as early as 160 AD. Pericardiectomy for the management of constrictive pericarditis was first performed successfully in this country by Churchill in 1929 [5]. Early attempts at pericardiectomy were guided by an imperfect understanding of the pathophysiology of constrictive pericarditis. With subsequent clarification it is now known that impaired ventricular diastolic relaxation and filling is the key disturbance. A complete decortication of both ventricles, including the diaphragmatic surface has proved important, although aggressive decortication of the atria has not. Large series of patients reported in the literature have clearly demonstrated the beneficial effect of pericardiectomy in patients with constrictive pericarditis. Current left and right cardiac catheterization data, computed tomographic scanning, and echocardiography are essential in establishing the diagnosis of constrictive pericarditis. Our surgical technique included median sternotomy with aggressive pericardectomy extended from phrenic nerve to phrenic nerve anteriorly. We pursued fibrinous

4 1582 HINKAMP ET AL Ann Thorac Surg 1994;57: attachments posteriorly to completely dissect the posterior myocardium. We avoided using cardiopulmonary bypass whenever possible without compromising the extent of decortication, even though some groups in the literature believe that bypass provides added safety and promotes more complete decortication [6]. Because longstanding constriction has a detrimental effect on the myocardium [Z], we believe early surgical intervention is important to ensure a successful outcome. Previous reports, including one from our institution [8], have described the development of pericardial constriction after open heart operations [l, 9-11]. Pericardial irrigation with povidone-iodine solution, retained pericardial clot, and the postpericardiectomy syndrome are thought to foster the formation of a thickened constrictive pericardium. The mean interval to presentation in such patients is approximately 24 months. Its true incidence is uncertain, as mild cases may go undetected, but it has been approximately 0.2% in our experience. The reported operative mortality associated with pericardiectomy is variable (0 to 33%), and preoperative predictors of poor survival include a history of malignancy or radiation therapy, or a preoperative NYHA class IV. The mortality at our institution was 11% in 45 patients with symptomatic pericarditis after open heart operations [8]. Unlike the previously reported experience with pericardiectomy in nontransplantation patients [8], a significantly higher incidence of pericardial constriction (1.4%) was noted in our heart transplantation patients. The outcome after pericardiectomy has also been less than optimal. The usual causes of failure to improve after pericardiectomy are inadequate surgical decortication and a concomitant or predominant intramyocardial process resulting in restrictive cardiomyopathy. We believe that a substantial restrictive cardiomyopathy coexisted in 3 of our patients. In our first patient, severe chronic rejection with right ventricular failure was the predominant problem. Chronic pericardial constriction resulting in myocardial atrophy and secondary heart failure seemed to be the cause of death in our second patient. The quantification of restrictive versus constrictive dysfunction can pose a difficult diagnostic dilemma. Exercise, pharmacologic intervention, and fluid challenge may be used to distinguish between left and right ventricular diastolic pressures, but experience with these tests is limited and the sensitivity of their specificity is not know. Some groups have recently suggested that differentiation between constriction and restriction using Doppler echocardiography may be possible by comparing the respiratory changes in the transvalvular flow velocities [El. Heart transplantation patients are a challenging group, as diastolic function may be abnormal for many reasons in these patients. A recent study of the hemodynamic and Doppler echocardiographic findings relating to ventricular function in long-term cardiac transplantation patients identified evidence of restrictive and constrictive physiology in 15% of the patients [13]. Elevated diastolic pressures, an exaggerated rise in the early diastolic ventricular pressure, higher peak pressures, and early mitral flow velocities characterized this group of patients, and in- creased rejection episodes were observed. One long-term study of hemodynamic function in cardiac transplantation recipients demonstrated striking increases in the pulmonary artery, pulmonary artery capillary wedge, and right atrial pressures during exercise [14]. In fact, the response of the right atrial pressure waveform to volume loading and inspiration was remarkably similar to that reported by Bush and colleagues [15] in the setting of occult constrictive pericardial disease. Such an abnormal hernodynamic response of the transplanted heart is poorly understood. Cardiac denervation, physical deconditioning, ventricular afterload, poor atrial compliance, and myocardial fibrous disease may all play a role in this peculiar physiology. The link between abnormal diastolic function and immune-mediated damage of the allograft is complex. Postmortem examinations of the transplanted hearts have yielded findings which support the hypothesis that some ventricular diastolic dysfunction may be due to coronary artery disease of the graft [16]. In other studies, diastolic dysfunction has been used as a marker for allograft rejection [17]. Undoubtedly the etiology of diastolic dysfunction in allografts is multifactorial. The important fact, however, is that the allograft diastolic dysfunction exists in a fair percentage of patients and this may lead to potential confusion with the classic surgically correctable pericardial constriction. Moreover, it may often coexist with constrictive pericarditis. Clearly the surgical outcome of pericardiectomy depends on the nature of the underlying myocardial dysfunction. The clinical course in our first patient suggests moderately severe chronic rejection as the cause of her severe right ventricular dysfunction. Pericardial resection was, unfortunately, not helpful. We believe that the longstanding constriction, more than 1 year, as well as restrictive cardiomyopathy resulted in a poor cardiac output in the second patient. He had a substantial preoperative disability, which is a known predictor of poor outcome [17]. Redo transplantation would have been his only viable option. Patient 3 also had evidence of restrictive dysfunction, although early results at 4 weeks after pericardiectomy were favorable with improved symptoms. Long term he continues to struggle with elevated diastolic pressures, poor cardiac systolic performance, and mild renal insufficiency. Our fourth patient is a clear surgical success, with no recurrent symptoms after pericardiectomy. His constriction was primarily effusive and occurred early post transplantation. In conclusion, we believe the occurrence of pure constricting pericarditis in the heart transplantation patient is unusual, but pericardiectomy may bring about improvement in ventricular diastolic filling and relief of clinical symptoms, if considerable restrictive dysfunction can be ruled out. Rigid hemodynamic criteria, volume loading, and the use of Doppler flow studies are essential. However, even surgical exploration and removal of true peel formation may not ensure an optimal outcome in these patients. In the cardiac transplantation population, a more restrictive cardiac physiology may exist than has

5 Ann Thorac Surg 1994; HINKAMP ET AL 1583 been thought previously, and it may mimic the appearance of constriction with impaired ventricular filling. Certainly, acute rejection, coronary arteriopathy, and right ventricular dysfunction all play a role in this impaired ventricular filling seen in patients with high pulmonary vascular resistance. Predictions of substantial improvement in cardiac performance after pericardiectomy in heart transplantation patients must be tempered with this realization. Patients with an effusive characteristic to their pericardial constriction may prove to be the only subgroup worthy of surgical exploration. Redo transplantation for patients with considerable restrictive and constrictive physiology may provide the best solution to this difficult problem. References 1. Kutcher MA, King SB, Benjamin, et al. Constrictive pericarditis as a complication of cardiac surgery: recognition of an entity. Am J Cardiol 1982;50: Rice PL, Pifarre R, Montoya A. Constrictive pericarditis following cardiac surgery. Ann Thorac Surg 1981;31: McCaughan BC, Schaff HV, Piehler JM, et al. Early and late results of pericardectomy for constrictive pericarditis. J Thorac Cardiovasc Surg 1985;89: Copeland JG, Riley JE, Fuller J. Pericardectomy for effusive constrictive pericarditis after heart transplantation. J Heart Transplant 1986;5: Churchill ED. Decortication of the heart (Delorme) for adhesive pericarditis. Arch Surg 1929;19: Copeland JG, Stinton EB, Griepp RB, Shumway NE. Surgical treatment of chronic constrictive pericarditis using cardiopulmonary bypass. J Thorac Cardiovasc Surg 1975;2:23& Roberts JT, Beck CS. The effect of chronic cardiac compression on the size of the heart muscle fibers. Am Heart J 1941;22: Killian DM, Furiasse JG, Scanlon PJ, Loeb HS, Sullivan HJ. Constrictive pericarditis after cardiac surgery. Am Heart J 1989;118:3:56>8. 9. Cohen MV, Greenberger MA. Constrictive pericarditis: early and late complications of cardiac surgery. Am J Cardiol 1979;43: Ribiero P, Sapsford R, Evans T, et al. Constrictive pericarditis as a complication of coronary artery bypass surgery. Br Heart J 1984;51: Ng ASH, Dorosti K, Sheldon WC. Constrictive pericarditis following cardiac surgery-cleveland Clinic experience: report of 12 cases and review. Cleve Clin Q 1984;51: Hatle LK, Appleton CP, Popp RL. Differentiation of constrictive pericarditis and restrictive cardiomyopathy by Doppler echocardiography. Circulation 1989;79: Valentine HA, Appleton CP, Hatle LK, et al. A hemodynamic and Doppler echocardiographic study of ventricular function in long-term cardiac allograft recipients. Circulation 1989;79: Pflugfelder PW, McKenzie FN, Kostuk WJ. Hemodynamic profiles at rest and during supine exercise after orthotopic cardiac transplantation. Am J Cardiol 1988;61: Bush CA, Stanj JM, Wooley CF, Kilman JW. Occult constrictive pericardial disease. Diagnosis by rapid volume expansion and correction by pericardectomy. Circulation 1977;56: Bieher CP, Stinson EB, Shumway NE, et al. Cardiac transplantation in man: cardiac allograft pathology. Circulation 1970;41:75% Dawkins KD, Oldershaw PJ, Billingham ME, et al. Changes in diastolic function as a non-invasive marker of cardiac allograft rejection. Heart Transplant 1984;3:28&94.