- Email: [email protected]
Low cardiac output complicating pericardiectomy for pericardial tamponade
228
CASE REPORT SUNDAY ET AL HEART FAILURE WITH PERICARDIECTOMY
Ann Thorac Surg 1999;67:228 –31
monitoring, 24 months after the operation revealed complete disappearance of both the mass in the right atrium and the tachyarrhythmia. Although tuberculous involvement of the heart is rare, it should be suspected in patients with a cardiac mass who have active pulmonary tuberculosis and cardiac arrythmia.
References
Fig 5. Multiplane transesophageal echocardiogram after 5 months of antituberculosis medication reveals complete disappearance of the mass and normal diameter of the superior vena cava. (LA 5 left atrium; RA 5 right atrium; SVC 5 superior vena cava.)
carried out before the introduction of specific antituberculous therapy assert that the myocardium is involved in less than 0.30% of patients dying of tuberculosis [2]. Cardiac tuberculosis is usually found at postmortem examination [1–3]. Organ involvement was classified as being either miliary (numerous small tubercles each less than 3 mm in diameter and resulting from blood-spread infection) or nodular (large rounded tuberculous lesions formed by confluent foci of tuberculous infection). Nodular myocardial tuberculosis might develop into a ventricular aneurysm [1]. True aneurysm formation due to myocardial tuberculosis is very rare. Halim and associates [4] reported a case of successful surgical treatment for a ruptured myocardial tuberculosis. The myocardium might be affected by direct spread from a mediastinal gland, by the lymphatic routes, or by the bloodstream [2]. This patient initially had miliary pattern pulmonary tuberculosis, and then it may have spread directly to the pericardium and myocardium. Preoperatively, cardiac tuberculoma could not be diagnosed because the results of two endomyocardial biopsies under transesophageal echocardiographic guidance suggested only intracardiac tumor with fibrosis. The presumptive diagnosis preoperatively was right atrial angiosarcoma or rhabdomyosarcoma. Clinical presentations of tuberculosis of the heart include pulmonary vein obstruction caused by left atrial mass lesions, right ventricular outflow tract obstruction, superior vena cava obstruction, aortic insufficiency, and cardiac arrythmia [2, 5, 6]. Arrythmia is a manifestation in patients with cardiac tuberculosis. Ventricular tachyarrythmia was found in a patient with right ventricular involvement of tuberculosis [5]. The treatment for cardiac tuberculosis is principally antituberculous medication. After histologic diagnosis with partial excision, the antituberculosis therapy was successful in this case. Close follow up with repeated two-dimensional echocardiography and 24-hour Holter © 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
1. Rose AG. Cardiac tuberculosis: a study of 19 patients. Arch Pathol Lab Med 1987;111:422– 6. 2. Horn H, Saphir O. The involvement of the myocardium in tuberculosis: a review the literature and report of three cases. Am Rev Tuberc 1935;32:492–506. 3. Kannangara DW, Salem FA, Rao BS, Thadepalli H. Cardiac tuberculosis: TB of the endocardium. Am J Med Sci 1984;287: 45–7. 4. Halim MA, Mercer EM, Guinn GA. Myocardial tuberculoma with rupture and pseudoaneurysm formation—successful surgical treatment. Br Heart J 1985;54:603– 4. 5. O’Neill PG, Rokey R, Greenburg S, Pacifico A. Resolution of ventricular tachycardia and endocardial tuberculoma following antituberculous therapy. Chest 1991;100:1467–9. 6. Soyer R, Brunet A, Chevallier B, Leroy J, Morere M, Redonnet M. Tuberculous aortic insufficiency: report of a case with successful surgical treatment. J Thorac Cardiovasc Surg 1981; 82:254– 6.
Low Cardiac Output Complicating Pericardiectomy for Pericardial Tamponade Renee Sunday, MD, Lary A. Robinson, MD, and Voytek Bosek, MD Department of Anesthesiology and Division of Cardiovascular and Thoracic Surgery, H. Lee Moffitt Cancer Center and Research Institute, University of South Florida, Tampa, Florida
Neoplastic involvement of the pericardium resulting in an effusion and subsequent tamponade is an emergency requiring prompt decompression, generally safely accomplished by subxiphoid pericardiectomy. However, the current case report describes a patient with florid pericardial tamponade who underwent surgical decompression with transient hemodynamic improvement, who then rapidly developed progressive, heart failure and death. This paradoxical response to pericardial decompression, similar to that seen occasionally with pericardiectomy in constrictive pericarditis, may be more frequent than currently appreciated. Its cause may relate to the sudden removal of the chronic external ventricular support from the effusion resulting in ventricular dilatation and failure. (Ann Thorac Surg 1999;67:228 –31) © 1999 by The Society of Thoracic Surgeons Accepted for publication June 9, 1998. Address reprint requests to Dr Robinson, Thoracic Oncology Program, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Dr, Tampa, FL 33612-9497.
0003-4975/99/$20.00 PII S0003-4975(98)01143-6
Ann Thorac Surg 1999;67:228 –31
CASE REPORT SUNDAY ET AL HEART FAILURE WITH PERICARDIECTOMY
229
I
nvolvement of the pericardium or heart in patients with disseminated cancer is surprisingly common, occurring in up to 21% of cases [1]. The most common tumors causing pericardial metastases are lung, breast, leukemia, and lymphoma, usually resulting in a pericardial effusion. The effusion may remain asymptomatic or it may rapidly progress to cause dyspnea, weakness, hypotension, and shock. The progression of the pericardial effusion to cause cardiac tamponade is one of the true oncologic emergencies. Although some controversy exists as to the ideal treatment of tamponade, the most common current approach is for prompt surgical decompression of the pericardium by subxiphoid pericardiectomy. This highly effective technique generally may be performed rapidly with high diagnostic accuracy, a low recurrence rate, and minimal morbidity [2, 3]. However, there have been scattered disturbing reports over the past few years of fatal or near-fatal systolic dysfunction occurring soon after decompression of pericardial tamponade accomplished by either pericardiocentesis [4] or subxiphoid pericardiectomy [5–7]. We now report a fatal case of postpericardiectomy cardiac failure similar to previous patients, and this patient may serve to illustrate an uncommon yet definite clinical entity that clinicians must be aware of in treating the tamponade patient.
Case Report A 60-year-old female smoker presented with a 3-month history of cough and weakness, a 5-lb weight loss, and recently a 3-day history of dyspnea and weakness. She had no previous significant illnesses. Her physical examination was remarkable for a resting tachycardia 120 per minute and a paradoxical pulse of 18 mm Hg, although she still had a blood pressure 110/74 mm Hg. She was pale and peripherally cool and constricted to the extent that a pulse oximeter would not register a signal. Her chest radiograph (Fig 1) demonstrated an enlarged cardiac silhouette, a 2.5-cm right upper lobe mass, and otherwise clear lung fields. Computed tomography of the chest demonstrated a right upper lobe mass with hilar adenopathy, liver lesions suspicious for metastases, and a very large pericardial effusion surrounding the entire heart (Fig 2). A brain computed tomographic scan revealed a possible small cerebellar metastasis. One month earlier, a needle biopsy of a superficial enlarged cervical lymph node demonstrated metastatic papillary adenocarcinoma. Her CA-125 (334 U/mL) and carcinoembryonic antigen (472 ng/mL) levels were both markedly elevated. An echocardiogram revealed a very large pericardial effusion with diastolic right atrial and right ventricular collapse but with well-preserved left ventricular function (ejection fraction estimated 65%). She was believed to have florid cardiac tamponade that was best treated with emergent subxiphoid pericardiectomy. When the patient arrived in the operating room, the radial artery catheter pressure was 150/90 mm Hg and the pulse was 110 per minute. She was peripherally very constricted and the pulse oximeter still did not register. Anesthesia and muscle relaxation were induced with
Fig 1. Posteroanterior chest radiograph demonstrating a large globular cardiac silhouette, 2.5 cm in diameter right upper lobe lung mass, and otherwise clear lung fields with no significant pleural effusion.
etomidate and rocuronium. After tracheal intubation, anesthesia was maintained with isoflurane, air, and 50% oxygen. With general anesthesia, the blood pressure decreased to 140/90 mm Hg and the heart rate increased to 120 beats/min. Through a midline subxiphoid incision, a tense pericardium was sharply opened with a resultant geyser of dark bloody fluid totaling 700 mL. Immediately with pericardial decompression, the pulse oximeter then began to register 99% saturation, the patient’s color
Fig 2. Computed tomographic scan of the chest at the level of the right and left ventricles demonstrating a very large circumferential pericardial effusion (white arrow), which reaches as much as 3.0 cm in width. The interventricular septum is readily visible separating the two contrast-filled ventricles.
230
CASE REPORT SUNDAY ET AL HEART FAILURE WITH PERICARDIECTOMY
improved, the pulse rate declined to 90 beats/min, and the blood pressure decreased to 120/80 mm Hg. After removing a 4-cm by 4-cm portion of thickened pericardium, the right ventricle was visualized contracting poorly and was noted to be covered by whitish nodules. The right atrium was distended and was not contracting. After leaving in two 24F intrapericardial drainage catheters and closing the wound, the patient was returned to the recovery room after this 25-minute operation. Within minutes of returning to the recovery room, the patient’s pulse increased to 140 beats/min, her blood pressure decreased to 100 mm Hg systolic, and she again became ashen with a pulse oximeter that would not register. She obviously was developing heart failure. A pulmonary artery catheter was inserted revealing pressures of 39/26 mm Hg but no reliable pulmonary capillary wedge pressure or cardiac output could be obtained. By this time, she was receiving high doses of dobutamine, dopamine, and epinephrine. A transthoracic echocardiogram demonstrated no pericardial fluid and a very hypocontractile right heart initially, which progressed to a noncontractile right heart. The left ventricle also was poorly contractile with a 30% estimated ejection fraction. She developed progressive hypotension, anuria, and metabolic acidosis with fulminant pulmonary edema. Despite heroic efforts, the refractory heart failure with pulmonary edema progressed to her death 6 hours after operation. The autopsy demonstrated bilateral, multifocal bronchiolar–alveolar carcinoma of the lung with the predominant 3-cm diameter tumor mass in the right upper lobe. No pulmonary embolus or pulmonary vascular disease was found. Metastases were widespread involving the paraesophageal soft tissues and lymph nodes, the liver, both adrenal glands, the left ovary, the cerebellum and frontal cortex, and the leptomeninges in multiple areas. The pericardium and the epicardium of both the right and left ventricles contained extensive metastases. There was focal extension of the metastases into the regional myocardial blood vessels and the lymphatics were distended by tumor. No other cardiac abnormalities were found.
Comment The 30-day operative mortality after subxiphoid pericardiectomy is stated to be 20% but all deaths in these recent series [2, 3] are reported to be unrelated to the surgical procedure. However, the details of the deaths were not discussed. Gregory and associates [8] in 1985 reported a series of 49 patients undergoing pericardiectomy mostly by left anterior pericardiectomy and they also attributed none of their 12 (24%) early 30-day postoperative deaths to the surgical procedure. Nevertheless, two of their patients (4.1%) developed severe refractory hypotension in the immediate postoperative period after improving initially at operation. Both patients died in much the same manner as the present patient. Wolfe and Edelman [4] first reported in 1993 two cases of global
Ann Thorac Surg 1999;67:228 –31
systolic dysfunction after pericardiocentesis to relieve tamponade. Both patients had metastatic breast cancer and with vigorous support eventually recovered completely normal cardiac function in 1 to 2 weeks. Braverman and Sundaresan [5] subsequently described another case of transient, severe left ventricular systolic dysfunction after surgical creation of a pericardial window to relieve tamponade caused by benign inflammatory pericarditis. This patient also recovered normal cardiac function over the next 3 weeks after operation. But unlike the prior cases, this patient had some systolic dysfunction noted by echocardiogram immediately before pericardial decompression. Neelakandan and associates [6] in 1996 reported a young patient with purulent pericarditis and tamponade who developed low cardiac output after partial pericardiectomy that progressed to death 6 hours after operation in much the same manner as the present case report. Another patient in their report was described with an inflammatory pericardial tamponade who developed reversible low cardiac output syndrome after subxiphoid pericardiectomy. Most recently, Dosios and Angouras [7] reviewed their experience with subxiphoid pericardiectomy in 53 patients and found 4 patients (7.5%) who developed this low cardiac output syndrome within the first 24 hours of pericardiectomy. Although all had marked, sudden hemodynamic improvement intraoperatively with pericardial drainage, 3 of the 4 patients (75%) died postoperatively of heart failure within a few hours of the onset of this problem. Only one of the patients with a fatal outcome had malignant pericardial disease. The present case report is similar in many respects to these prior cases in that there was initial intraoperative hemodynamic improvement followed by rapid cardiac failure postoperatively. In those fatal cases, death occurred within hours of operation. Generally, it would be reasonable to expect that after relief of tamponade the decompressed heart would demonstrate marked, sustained hemodynamic improvement as cardiac filling and stroke volume are increased. However, it is paradoxical that a small group of patients rapidly develop worsening cardiac function, often leading to death. The cause of this relatively sudden heart failure has been variably attributed to prior myocardial damage from antineoplastic drugs, direct myocardial involvement by tumor, or decreased coronary blood flow during tamponade with resultant myocardial ischemia, stunning, and reversible myocardial hibernation [4, 5]. Another possibility is that occult systolic dysfunction may already be present during tamponade but it may be masked by reduced chamber sizes and the tachycardia. The dysfunction is then brought out by pericardial decompression. Alternatively, high levels of sympathetic tone and endogenous catecholamines during tamponade may mask preexisting myocardial dysfunction, which is then accentuated after pericardial decompression. However, chronic preexisting impaired cardiac function is unlikely as patients who survive this syndrome subsequently regain normal function.
Ann Thorac Surg 1999;67:231–3
A more plausible explanation for this early and rapid cardiac failure is related to the low cardiac output syndrome, which is seen in up to 28% of patients with chronic constrictive pericarditis after undergoing pericardiectomy [9, 10]. It is believed that the postoperative myocardial dysfunction in this setting is related to myocardial atrophy. The chronic external support of the heart by the tight pericardium and fluid when released may allow the heart to overdilate rapidly, leading to systolic dysfunction and heart failure. If this pathophysiologic mechanism also occurs in some patients with acute tamponade, then the suggestion by Neelakandan and associates [6] for gradual decompression of the pericardial effusion may have some merit, although the practical implementation of this suggestion may be difficult. Presumably pericardiocentesis with gradual removal of pericardial fluid with an indwelling catheter might be a feasible approach. However, this syndrome, no matter how lethal, is relatively rare and recommendations on prevention will await characterization of the patients at risk. Therefore, at present until more information is available, we recommend only that anesthesiologists, cardiologists, and thoracic surgeons be aware of the possibility that their patient might develop transient or even fatal heart failure after relief of a benign or malignant pericardial tamponade, similar to that seen in our patient. Treatment for this problem should be supportive with appropriate invasive monitoring and inotropic support, which will result in recovery in some patients.
References 1. Robinson LA, Ruckdeschel JC. Management of pleural and pericardial effusions. In: Berger A, Portenoy RK, Weissman DE, eds. Principles and practices of supportive oncology, 1st ed. Philadelphia: Lippincott-Raven, 1998:327–52. 2. Palatianos GM, Thurer RJ, Pompeo MQ, Kaiser GA. Clinical experience with subxiphoid drainage of pericardial effusions. Ann Thorac Surg 1989;48:381–5. 3. Moores DWO, Allen KB, Faber LP, et al. Subxiphoid pericardial drainage for pericardial tamponade. J Thorac Cardiovasc Surg 1995;109:546–52. 4. Wolfe MW, Edelman ER. Transient systolic dysfunction after relief of cardiac tamponade. Ann Intern Med 1993;119:42–3. 5. Braverman AC, Sundaresan S. Cardiac tamponade and severe ventricular dysfunction. Ann Intern Med 1994;120: 442. 6. Neelakandan B, Jayanthi N, Kanthimathi R. Subxiphoid drainage for pericardial tamponade. J Thorac Cardiovasc Surg 1996;111:489. 7. Dosios T, Angouras D. Low cardiac output syndrome complicating subxiphoid pericardiostomy for pericardial effusion. J Thorac Cardiovasc Surg 1997;113:220. 8. Gregory JR, McMurtrey MJ, Mountain CF. A surgical approach to the treatment of pericardial effusion in cancer patients. Amer J Clin Oncol 1985;8:319–23. 9. McCaughan BC, Schaff HV, Piehler JM, et al. Early and late results of pericardiectomy for constrictive pericarditis. J Thorac Cardiovasc Surg 1985;89:340–50. 10. Douglas JM Jr. The pericardium. In: Sabiston DC Jr, Spencer FC, eds. Surgery of the chest, 6th ed. Philadelphia: WB Saunders, 1995:1365– 86. © 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
CASE REPORT GAISSERT ET AL ESOPHAGECTOMY FOR FUNGAL NECROSIS
231
Surgical Management of Necrotizing Candida Esophagitis Henning A. Gaissert, MD, Christopher K. Breuer, MD, Alan Weissburg, MD, and Leonard Mermel, DO Section of General Thoracic Surgery, Rhode Island Hospital, and Departments of Surgery, Anesthesiology, and Internal Medicine, Brown University, Providence, Rhode Island
Invasive esophageal candidiasis produced transmural necrosis leading to perforation in 2 patients aged 10 and 27 years. Both patients survived after esophageal resection and complete diversion. One patient with acute leukemia and neutropenia experienced systemic candidiasis, which resolved after esophagectomy. Esophagectomy and diversion for yeast-induced necrosis may lead to complete recovery and resolution of disseminated candidiasis when combined with systemic antifungal therapy. (Ann Thorac Surg 1999;67:231–3) © 1999 by The Society of Thoracic Surgeons
C
andida colonization of the gastrointestinal tract is the most important source of opportunistic dissemination. The esophagus is the most common site of colonization [1]. Although cases of ulceration, perforation [2], and stricture attest to the potential for panmural involvement, recent evidence suggest that ulceration and deep tissue invasion are rare in patients without neutropenia [3]. To our knowledge, no one has correlated the depth of esophageal invasion with the likelihood of disseminated candidiasis; however, normally functioning polymorphonuclear leukocytes likely limit invasion and concomitant dissemination [3]. Systemic antifungal therapy remains the best available treatment, and more aggressive intervention, including operation, is rarely considered. The prognosis is poor in disseminated candidiasis, and half of all neutropenic patients with disseminated disease die [4]. We report 2 patients who required surgical intervention for complications of panmural esophageal candidiasis. Immediate resection, indicated for perforation of the esophagus in one and total necrosis in the other, resulted in complete recovery and freedom of fungal colonization during a 1-year period of observation. These cases demonstrate that esophageal candidiasis may progress insidiously to necrosis, perforation, and mediastinitis, and they support the notion of invasive esophageal candidiasis as a nidus for systemic infection. Heightened awareness of fungal esophagitis as a risk factor for necrosis and prompt intervention may improve the outcome of patients with this condition. Accepted for publication June 10, 1998. Address reprint requests to Dr Gaissert, 2 Dudley St, Suite 470, Providence, RI 02905 (e-mail: [email protected]).
0003-4975/99/$20.00 PII S0003-4975(98)01144-8
CASE REPORT SUNDAY ET AL HEART FAILURE WITH PERICARDIECTOMY
Ann Thorac Surg 1999;67:228 –31
monitoring, 24 months after the operation revealed complete disappearance of both the mass in the right atrium and the tachyarrhythmia. Although tuberculous involvement of the heart is rare, it should be suspected in patients with a cardiac mass who have active pulmonary tuberculosis and cardiac arrythmia.
References
Fig 5. Multiplane transesophageal echocardiogram after 5 months of antituberculosis medication reveals complete disappearance of the mass and normal diameter of the superior vena cava. (LA 5 left atrium; RA 5 right atrium; SVC 5 superior vena cava.)
carried out before the introduction of specific antituberculous therapy assert that the myocardium is involved in less than 0.30% of patients dying of tuberculosis [2]. Cardiac tuberculosis is usually found at postmortem examination [1–3]. Organ involvement was classified as being either miliary (numerous small tubercles each less than 3 mm in diameter and resulting from blood-spread infection) or nodular (large rounded tuberculous lesions formed by confluent foci of tuberculous infection). Nodular myocardial tuberculosis might develop into a ventricular aneurysm [1]. True aneurysm formation due to myocardial tuberculosis is very rare. Halim and associates [4] reported a case of successful surgical treatment for a ruptured myocardial tuberculosis. The myocardium might be affected by direct spread from a mediastinal gland, by the lymphatic routes, or by the bloodstream [2]. This patient initially had miliary pattern pulmonary tuberculosis, and then it may have spread directly to the pericardium and myocardium. Preoperatively, cardiac tuberculoma could not be diagnosed because the results of two endomyocardial biopsies under transesophageal echocardiographic guidance suggested only intracardiac tumor with fibrosis. The presumptive diagnosis preoperatively was right atrial angiosarcoma or rhabdomyosarcoma. Clinical presentations of tuberculosis of the heart include pulmonary vein obstruction caused by left atrial mass lesions, right ventricular outflow tract obstruction, superior vena cava obstruction, aortic insufficiency, and cardiac arrythmia [2, 5, 6]. Arrythmia is a manifestation in patients with cardiac tuberculosis. Ventricular tachyarrythmia was found in a patient with right ventricular involvement of tuberculosis [5]. The treatment for cardiac tuberculosis is principally antituberculous medication. After histologic diagnosis with partial excision, the antituberculosis therapy was successful in this case. Close follow up with repeated two-dimensional echocardiography and 24-hour Holter © 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
1. Rose AG. Cardiac tuberculosis: a study of 19 patients. Arch Pathol Lab Med 1987;111:422– 6. 2. Horn H, Saphir O. The involvement of the myocardium in tuberculosis: a review the literature and report of three cases. Am Rev Tuberc 1935;32:492–506. 3. Kannangara DW, Salem FA, Rao BS, Thadepalli H. Cardiac tuberculosis: TB of the endocardium. Am J Med Sci 1984;287: 45–7. 4. Halim MA, Mercer EM, Guinn GA. Myocardial tuberculoma with rupture and pseudoaneurysm formation—successful surgical treatment. Br Heart J 1985;54:603– 4. 5. O’Neill PG, Rokey R, Greenburg S, Pacifico A. Resolution of ventricular tachycardia and endocardial tuberculoma following antituberculous therapy. Chest 1991;100:1467–9. 6. Soyer R, Brunet A, Chevallier B, Leroy J, Morere M, Redonnet M. Tuberculous aortic insufficiency: report of a case with successful surgical treatment. J Thorac Cardiovasc Surg 1981; 82:254– 6.
Low Cardiac Output Complicating Pericardiectomy for Pericardial Tamponade Renee Sunday, MD, Lary A. Robinson, MD, and Voytek Bosek, MD Department of Anesthesiology and Division of Cardiovascular and Thoracic Surgery, H. Lee Moffitt Cancer Center and Research Institute, University of South Florida, Tampa, Florida
Neoplastic involvement of the pericardium resulting in an effusion and subsequent tamponade is an emergency requiring prompt decompression, generally safely accomplished by subxiphoid pericardiectomy. However, the current case report describes a patient with florid pericardial tamponade who underwent surgical decompression with transient hemodynamic improvement, who then rapidly developed progressive, heart failure and death. This paradoxical response to pericardial decompression, similar to that seen occasionally with pericardiectomy in constrictive pericarditis, may be more frequent than currently appreciated. Its cause may relate to the sudden removal of the chronic external ventricular support from the effusion resulting in ventricular dilatation and failure. (Ann Thorac Surg 1999;67:228 –31) © 1999 by The Society of Thoracic Surgeons Accepted for publication June 9, 1998. Address reprint requests to Dr Robinson, Thoracic Oncology Program, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Dr, Tampa, FL 33612-9497.
0003-4975/99/$20.00 PII S0003-4975(98)01143-6
Ann Thorac Surg 1999;67:228 –31
CASE REPORT SUNDAY ET AL HEART FAILURE WITH PERICARDIECTOMY
229
I
nvolvement of the pericardium or heart in patients with disseminated cancer is surprisingly common, occurring in up to 21% of cases [1]. The most common tumors causing pericardial metastases are lung, breast, leukemia, and lymphoma, usually resulting in a pericardial effusion. The effusion may remain asymptomatic or it may rapidly progress to cause dyspnea, weakness, hypotension, and shock. The progression of the pericardial effusion to cause cardiac tamponade is one of the true oncologic emergencies. Although some controversy exists as to the ideal treatment of tamponade, the most common current approach is for prompt surgical decompression of the pericardium by subxiphoid pericardiectomy. This highly effective technique generally may be performed rapidly with high diagnostic accuracy, a low recurrence rate, and minimal morbidity [2, 3]. However, there have been scattered disturbing reports over the past few years of fatal or near-fatal systolic dysfunction occurring soon after decompression of pericardial tamponade accomplished by either pericardiocentesis [4] or subxiphoid pericardiectomy [5–7]. We now report a fatal case of postpericardiectomy cardiac failure similar to previous patients, and this patient may serve to illustrate an uncommon yet definite clinical entity that clinicians must be aware of in treating the tamponade patient.
Case Report A 60-year-old female smoker presented with a 3-month history of cough and weakness, a 5-lb weight loss, and recently a 3-day history of dyspnea and weakness. She had no previous significant illnesses. Her physical examination was remarkable for a resting tachycardia 120 per minute and a paradoxical pulse of 18 mm Hg, although she still had a blood pressure 110/74 mm Hg. She was pale and peripherally cool and constricted to the extent that a pulse oximeter would not register a signal. Her chest radiograph (Fig 1) demonstrated an enlarged cardiac silhouette, a 2.5-cm right upper lobe mass, and otherwise clear lung fields. Computed tomography of the chest demonstrated a right upper lobe mass with hilar adenopathy, liver lesions suspicious for metastases, and a very large pericardial effusion surrounding the entire heart (Fig 2). A brain computed tomographic scan revealed a possible small cerebellar metastasis. One month earlier, a needle biopsy of a superficial enlarged cervical lymph node demonstrated metastatic papillary adenocarcinoma. Her CA-125 (334 U/mL) and carcinoembryonic antigen (472 ng/mL) levels were both markedly elevated. An echocardiogram revealed a very large pericardial effusion with diastolic right atrial and right ventricular collapse but with well-preserved left ventricular function (ejection fraction estimated 65%). She was believed to have florid cardiac tamponade that was best treated with emergent subxiphoid pericardiectomy. When the patient arrived in the operating room, the radial artery catheter pressure was 150/90 mm Hg and the pulse was 110 per minute. She was peripherally very constricted and the pulse oximeter still did not register. Anesthesia and muscle relaxation were induced with
Fig 1. Posteroanterior chest radiograph demonstrating a large globular cardiac silhouette, 2.5 cm in diameter right upper lobe lung mass, and otherwise clear lung fields with no significant pleural effusion.
etomidate and rocuronium. After tracheal intubation, anesthesia was maintained with isoflurane, air, and 50% oxygen. With general anesthesia, the blood pressure decreased to 140/90 mm Hg and the heart rate increased to 120 beats/min. Through a midline subxiphoid incision, a tense pericardium was sharply opened with a resultant geyser of dark bloody fluid totaling 700 mL. Immediately with pericardial decompression, the pulse oximeter then began to register 99% saturation, the patient’s color
Fig 2. Computed tomographic scan of the chest at the level of the right and left ventricles demonstrating a very large circumferential pericardial effusion (white arrow), which reaches as much as 3.0 cm in width. The interventricular septum is readily visible separating the two contrast-filled ventricles.
230
CASE REPORT SUNDAY ET AL HEART FAILURE WITH PERICARDIECTOMY
improved, the pulse rate declined to 90 beats/min, and the blood pressure decreased to 120/80 mm Hg. After removing a 4-cm by 4-cm portion of thickened pericardium, the right ventricle was visualized contracting poorly and was noted to be covered by whitish nodules. The right atrium was distended and was not contracting. After leaving in two 24F intrapericardial drainage catheters and closing the wound, the patient was returned to the recovery room after this 25-minute operation. Within minutes of returning to the recovery room, the patient’s pulse increased to 140 beats/min, her blood pressure decreased to 100 mm Hg systolic, and she again became ashen with a pulse oximeter that would not register. She obviously was developing heart failure. A pulmonary artery catheter was inserted revealing pressures of 39/26 mm Hg but no reliable pulmonary capillary wedge pressure or cardiac output could be obtained. By this time, she was receiving high doses of dobutamine, dopamine, and epinephrine. A transthoracic echocardiogram demonstrated no pericardial fluid and a very hypocontractile right heart initially, which progressed to a noncontractile right heart. The left ventricle also was poorly contractile with a 30% estimated ejection fraction. She developed progressive hypotension, anuria, and metabolic acidosis with fulminant pulmonary edema. Despite heroic efforts, the refractory heart failure with pulmonary edema progressed to her death 6 hours after operation. The autopsy demonstrated bilateral, multifocal bronchiolar–alveolar carcinoma of the lung with the predominant 3-cm diameter tumor mass in the right upper lobe. No pulmonary embolus or pulmonary vascular disease was found. Metastases were widespread involving the paraesophageal soft tissues and lymph nodes, the liver, both adrenal glands, the left ovary, the cerebellum and frontal cortex, and the leptomeninges in multiple areas. The pericardium and the epicardium of both the right and left ventricles contained extensive metastases. There was focal extension of the metastases into the regional myocardial blood vessels and the lymphatics were distended by tumor. No other cardiac abnormalities were found.
Comment The 30-day operative mortality after subxiphoid pericardiectomy is stated to be 20% but all deaths in these recent series [2, 3] are reported to be unrelated to the surgical procedure. However, the details of the deaths were not discussed. Gregory and associates [8] in 1985 reported a series of 49 patients undergoing pericardiectomy mostly by left anterior pericardiectomy and they also attributed none of their 12 (24%) early 30-day postoperative deaths to the surgical procedure. Nevertheless, two of their patients (4.1%) developed severe refractory hypotension in the immediate postoperative period after improving initially at operation. Both patients died in much the same manner as the present patient. Wolfe and Edelman [4] first reported in 1993 two cases of global
Ann Thorac Surg 1999;67:228 –31
systolic dysfunction after pericardiocentesis to relieve tamponade. Both patients had metastatic breast cancer and with vigorous support eventually recovered completely normal cardiac function in 1 to 2 weeks. Braverman and Sundaresan [5] subsequently described another case of transient, severe left ventricular systolic dysfunction after surgical creation of a pericardial window to relieve tamponade caused by benign inflammatory pericarditis. This patient also recovered normal cardiac function over the next 3 weeks after operation. But unlike the prior cases, this patient had some systolic dysfunction noted by echocardiogram immediately before pericardial decompression. Neelakandan and associates [6] in 1996 reported a young patient with purulent pericarditis and tamponade who developed low cardiac output after partial pericardiectomy that progressed to death 6 hours after operation in much the same manner as the present case report. Another patient in their report was described with an inflammatory pericardial tamponade who developed reversible low cardiac output syndrome after subxiphoid pericardiectomy. Most recently, Dosios and Angouras [7] reviewed their experience with subxiphoid pericardiectomy in 53 patients and found 4 patients (7.5%) who developed this low cardiac output syndrome within the first 24 hours of pericardiectomy. Although all had marked, sudden hemodynamic improvement intraoperatively with pericardial drainage, 3 of the 4 patients (75%) died postoperatively of heart failure within a few hours of the onset of this problem. Only one of the patients with a fatal outcome had malignant pericardial disease. The present case report is similar in many respects to these prior cases in that there was initial intraoperative hemodynamic improvement followed by rapid cardiac failure postoperatively. In those fatal cases, death occurred within hours of operation. Generally, it would be reasonable to expect that after relief of tamponade the decompressed heart would demonstrate marked, sustained hemodynamic improvement as cardiac filling and stroke volume are increased. However, it is paradoxical that a small group of patients rapidly develop worsening cardiac function, often leading to death. The cause of this relatively sudden heart failure has been variably attributed to prior myocardial damage from antineoplastic drugs, direct myocardial involvement by tumor, or decreased coronary blood flow during tamponade with resultant myocardial ischemia, stunning, and reversible myocardial hibernation [4, 5]. Another possibility is that occult systolic dysfunction may already be present during tamponade but it may be masked by reduced chamber sizes and the tachycardia. The dysfunction is then brought out by pericardial decompression. Alternatively, high levels of sympathetic tone and endogenous catecholamines during tamponade may mask preexisting myocardial dysfunction, which is then accentuated after pericardial decompression. However, chronic preexisting impaired cardiac function is unlikely as patients who survive this syndrome subsequently regain normal function.
Ann Thorac Surg 1999;67:231–3
A more plausible explanation for this early and rapid cardiac failure is related to the low cardiac output syndrome, which is seen in up to 28% of patients with chronic constrictive pericarditis after undergoing pericardiectomy [9, 10]. It is believed that the postoperative myocardial dysfunction in this setting is related to myocardial atrophy. The chronic external support of the heart by the tight pericardium and fluid when released may allow the heart to overdilate rapidly, leading to systolic dysfunction and heart failure. If this pathophysiologic mechanism also occurs in some patients with acute tamponade, then the suggestion by Neelakandan and associates [6] for gradual decompression of the pericardial effusion may have some merit, although the practical implementation of this suggestion may be difficult. Presumably pericardiocentesis with gradual removal of pericardial fluid with an indwelling catheter might be a feasible approach. However, this syndrome, no matter how lethal, is relatively rare and recommendations on prevention will await characterization of the patients at risk. Therefore, at present until more information is available, we recommend only that anesthesiologists, cardiologists, and thoracic surgeons be aware of the possibility that their patient might develop transient or even fatal heart failure after relief of a benign or malignant pericardial tamponade, similar to that seen in our patient. Treatment for this problem should be supportive with appropriate invasive monitoring and inotropic support, which will result in recovery in some patients.
References 1. Robinson LA, Ruckdeschel JC. Management of pleural and pericardial effusions. In: Berger A, Portenoy RK, Weissman DE, eds. Principles and practices of supportive oncology, 1st ed. Philadelphia: Lippincott-Raven, 1998:327–52. 2. Palatianos GM, Thurer RJ, Pompeo MQ, Kaiser GA. Clinical experience with subxiphoid drainage of pericardial effusions. Ann Thorac Surg 1989;48:381–5. 3. Moores DWO, Allen KB, Faber LP, et al. Subxiphoid pericardial drainage for pericardial tamponade. J Thorac Cardiovasc Surg 1995;109:546–52. 4. Wolfe MW, Edelman ER. Transient systolic dysfunction after relief of cardiac tamponade. Ann Intern Med 1993;119:42–3. 5. Braverman AC, Sundaresan S. Cardiac tamponade and severe ventricular dysfunction. Ann Intern Med 1994;120: 442. 6. Neelakandan B, Jayanthi N, Kanthimathi R. Subxiphoid drainage for pericardial tamponade. J Thorac Cardiovasc Surg 1996;111:489. 7. Dosios T, Angouras D. Low cardiac output syndrome complicating subxiphoid pericardiostomy for pericardial effusion. J Thorac Cardiovasc Surg 1997;113:220. 8. Gregory JR, McMurtrey MJ, Mountain CF. A surgical approach to the treatment of pericardial effusion in cancer patients. Amer J Clin Oncol 1985;8:319–23. 9. McCaughan BC, Schaff HV, Piehler JM, et al. Early and late results of pericardiectomy for constrictive pericarditis. J Thorac Cardiovasc Surg 1985;89:340–50. 10. Douglas JM Jr. The pericardium. In: Sabiston DC Jr, Spencer FC, eds. Surgery of the chest, 6th ed. Philadelphia: WB Saunders, 1995:1365– 86. © 1999 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
CASE REPORT GAISSERT ET AL ESOPHAGECTOMY FOR FUNGAL NECROSIS
231
Surgical Management of Necrotizing Candida Esophagitis Henning A. Gaissert, MD, Christopher K. Breuer, MD, Alan Weissburg, MD, and Leonard Mermel, DO Section of General Thoracic Surgery, Rhode Island Hospital, and Departments of Surgery, Anesthesiology, and Internal Medicine, Brown University, Providence, Rhode Island
Invasive esophageal candidiasis produced transmural necrosis leading to perforation in 2 patients aged 10 and 27 years. Both patients survived after esophageal resection and complete diversion. One patient with acute leukemia and neutropenia experienced systemic candidiasis, which resolved after esophagectomy. Esophagectomy and diversion for yeast-induced necrosis may lead to complete recovery and resolution of disseminated candidiasis when combined with systemic antifungal therapy. (Ann Thorac Surg 1999;67:231–3) © 1999 by The Society of Thoracic Surgeons
C
andida colonization of the gastrointestinal tract is the most important source of opportunistic dissemination. The esophagus is the most common site of colonization [1]. Although cases of ulceration, perforation [2], and stricture attest to the potential for panmural involvement, recent evidence suggest that ulceration and deep tissue invasion are rare in patients without neutropenia [3]. To our knowledge, no one has correlated the depth of esophageal invasion with the likelihood of disseminated candidiasis; however, normally functioning polymorphonuclear leukocytes likely limit invasion and concomitant dissemination [3]. Systemic antifungal therapy remains the best available treatment, and more aggressive intervention, including operation, is rarely considered. The prognosis is poor in disseminated candidiasis, and half of all neutropenic patients with disseminated disease die [4]. We report 2 patients who required surgical intervention for complications of panmural esophageal candidiasis. Immediate resection, indicated for perforation of the esophagus in one and total necrosis in the other, resulted in complete recovery and freedom of fungal colonization during a 1-year period of observation. These cases demonstrate that esophageal candidiasis may progress insidiously to necrosis, perforation, and mediastinitis, and they support the notion of invasive esophageal candidiasis as a nidus for systemic infection. Heightened awareness of fungal esophagitis as a risk factor for necrosis and prompt intervention may improve the outcome of patients with this condition. Accepted for publication June 10, 1998. Address reprint requests to Dr Gaissert, 2 Dudley St, Suite 470, Providence, RI 02905 (e-mail: [email protected]).
0003-4975/99/$20.00 PII S0003-4975(98)01144-8