- Email: [email protected]
Diagnostic Usefulness of Pericardial Fluid Cytology
Diapnostic Usefulness of Perlcardial Fluid Cytology· David G. Meyers, M.D., F.C.C.R; and DavidJ Bouska, B.S.
Although widely used, pericardial fluid cytology has not been adequately validated. We pooled 22 new cases with 71 previously reported cases having both fluid cytology and pericardial histopathologic examination. Cytology was correct in 87 of 93 cases (diagnostic accuracy, 94 percent). Sensitivity was 87 percent and speci6city was 100 percent. (Chelt 1989; 95:1142-43)
W
hile pericardial fluid cytology is widely utilized, surprisingly there are only three small studies which attempt to demonstrate its diagnostic accuracy. 1-3 To achieve adequate statistical power and to overcome possible selection bias, we pooled the results of the 71 previously reported patients with 22 additional patients. All patients in five community and two university hospitals having pericardial fluid sampled by either needle pericardiocentesis or surgical pericardiectomy plus histologic evaluation of pericardial tissue from either pericardial biopsy (15 cases) or autopsy (eight cases) during a five-year period were identified by review of medical records. The cause of the pericardial effusion in each case was determined from the composite of gross appearance, histologic findings and bacteriologic studies of pericardial tissue together with clinical information. Sensitivity, specificity, predictive values, and diagnostic accuracy plus their 95 percent confidence intervals (CI) were calculated from pooled data. Twenty-two patients were identified. The most frequent cause of effusion in our series as documented by histopathologic examination of pericardium was cancer (adenocarcinoma of the lung in five cases, squamous cell carcinoma of the lung in three, adenocarcinoma of the breast in one, and mesothelioma in one). Other causes included post-pericardiotomy syndrome (one case), postirradiation (four cases), collagen-vascular diseases (two cases), and infection (three cases). No cause was identified by examination of the pericardium in two cases. Cytologic examination of pericardial fluid identified malignant cells in nine of the ten patients with cancer (Table 1). Tissue examination confirmed 12 negative cytologic examinations. There were no false-positive tests but one false-negative test occurred in a patient with squamous cell carcinoma of the lung. *From the Department of Internal Medicine, University of Nebraska College of Medicine, Omaha. Reprint requests: Dr. Meyers, Section of Cardiology, University of
Nebraska Medical Center, Omaha 68105
Cases which were identified as having validation by histopathologic examination of pericardial tissue from three studies of pericardial fluid cytology are pooled with our data in the table. These studies were performed in a university hospital, 1 a Veterans Administration facility, 2 and in patients with cardiac tamponade seen at a cancer institute. 3 Included are 93 patients: 22 with lung cancer, nine with breast cancer, five with undifferentiated pericardial metastases, two with metastases from the colon, three with mesothelioma, four with non-Hodgkin's and two with Hodgkin's lymphoma, 14 with granulomatous pericarditis, four with renal failure, five with radiation pericarditis, three with infectious pericarditis, two vlith lupus erythematosus, one with postpericardiotomy syndrome, 14 with idiopathic pericarditis, and three patients whose diseases were not characterized. The prevalence of cancer involving the pericardium is 50 percent. Cytologic examination produced a sensitivity of 87 percent (95 percent CI = 77 to 96 percent) and a specificity of 100 percent (95 percent CI = 92 to 100 percent). The positive predictive value is 100 percent (95 percent CI = 91 to 100 percent) and the negative predictive value is 88 percent (95 percent CI = 78 to 97 percent). Cytologic studies correctly identified both the presence and absence of cancer (diagnostic accuracy) in 94 percent (95 percent CI = 88 to 97 percent). No false-positive tests occurred. A total of six patients with cancer had false-negative fluid cytology noted in which pericardial tissue examination demonstrated mesothelioma in two cases, undifferentiated carcinoma in two, and squamous cell carcinoma of the lung in one. The etiology in one case was not characterized. Data regarding clinical setting, amount of pericardial fluid, or degree of malignant pericardial involvement in these cases are not available. Malignant involvement of the pericardium and heart is found in from 0.1 to 6.4 percent of unselected autopsies and in up to 20.6 percent ofautopsies ofpatients dying ofcancer. 4 Pericardiocentesis and fluid cytology is a standard diagnostic tool in the evaluation of pericardial effusion. Yet, validation ofpericardial fluid cytology has been lacking. The usefulness of the three previous studies has been limited by their small sample sizes and potential selection bias stemming from their particular case mixes. 1-3 These limitations are overcome by analysis of pooled data from these and the present study, all of which used similar techniques. S The results demonstrate that pericardial fluid cytology is highly accurate. Positive tests results are completely reliable whereas occasional false-negative tests unaccountably occur. REFERENCES 1 Zipf RE, Johnston ww. The role of cytology in the evaluation of pericardial effusions. Chest 1972; 62:593-96 2 Posner MR, Cohen GI, Skarin AT. Pericardial disease in patients
Table 1- Cytologic Study Results Source
TruePositive
FalsePositive
TrueNegative
FalseNegative
Total
Zipf and Johnston l Posner et all Reyes et al3 Present study Total
12 11 9 9 41
0 0 0 0 0
32 2 0 12 46
3 2 0 1 6
47 15 9 22 93
1142
Diagnostic Usefulness of PericardiaJ Fluid Cytology (Meyers, Bouska)
with cancer: the differentiation of malignant from idiopathic and radiation-induced pericarditis. Am J Med 1981; 71:407-13 3 Reyes C~ Strinden C, Banerji M. The role of cytology in neoplastic cardiac tamponade. Acta Cytologica 1982; ,26:299-302 4 King DT, Nieberg RK. The use of cytology to evaluate pericardial effusions. Ann Clin Lab Sci 1979; 9:18-23 5 L'Abbe KA, Detsky AS, O'Rorke K. Meta-analysis in clinical research. Ann Intern Med 1987; 107:224-33
Massive Hemoptysis during Catheterization of the Internal Jugular Vein* Yaakov Friedman, M.D.; Elena nuy, M.D.; Eric C. Rackow, M.D., F.C.C.R; and Max Harry Weil, M.D., Ph.D., F.C.C.R
Percutaneous cannulation of the internal jugular vein is commonly performed to obtain central venous access. We report the first case of massive hemoptysis occurring during cannulation of the internal jugular vein. (Chest 1989; 95:1143)
C
annulation of the internal jugular vein (IJV) is currently accepted as a relatively safe method of obtaining central venous access. 1 The most common complication is carotid artery puncture, which is usually benign. 2 We report a case of massive hemoptysis, cervical hematoma and airway obstruction occurring during IJV catheterization. CASE REPORT
A 51-year-old woman with a history of intravenous drug abuse was being treated for osteomyelitis with a six-week course of piperacillin, oxacillin, and gentamicin. Due to the need for prolonged antibiotic administration, catheterization of the right IJV for venous access was performed. An 18-gauge needle was inserted via the anterior approach with immediate return of venous blood. A .035 mm guidewire was threaded, the vessel dilated with a 7 French dilator, and a 7 French triple lumen catheter was inserted without difficulty. Suddenly, the patient complained of severe chest pain and turned her neck dislodging the catheter. She bled vigorously from the puncture site and developed a large cervical hematoma despite the application of pressure. She rapidly became hypotensive and diaphoretic, and was treated with fluids and infusion of intravenous norepinephrine to maintain blood pressure. She complained of shortness of breath, developed stridor, and had 300 to 400 ml of bright-red hemoptysis. The patient was intubated with some difficulty because of the amount of hemoptysis and transferred to the intensive care unit. The norepinephrine infusion was rapidly tapered and discontinued. There was continued mild bleeding through the endotracheal tube which resolved over two hours. The patient's hematocrit value initially decreased from 40 to 30 percent, then stabilized at 32 percent after transfusion of three units of packed red blood cells. The patient's PT, PTT, and platelet count were within normal limits. The patient was never noted to have a carotid broit or thrill.
Angiography of the cervical and thoracic vessels done two days later was normal and did not demonstrate an arteriovenous fistula. The chest roentgenogram revealed bilateral alveolar infiltrates consistent with blood aspiration, which cleared over eight days. There was no evidence of air in the soft tissues of neck, pneumothorax, or pneumomediastinum at anyone time. The cervical hematoma allowing the patient to be extubated after seven resolved slowl~ days, and resolving completely in three weeks. *From the Division of Critical Care Medicine, Cook County Hospital, Chicago, and University of Health Sciences/fhe Chicago Medical School, North Chicago, IL.
DISCUSSION
Central venous catheterization of the IJV has a low complication rate. l The most common complication is carotid artery puncture, occurring in 4.2 percent of cases, usually without significant consequences. 2 Other complications include pneumothorax, thrombosis of the IoN air embolus, and cerebrovascular accidents. There have been reports of less frequent, but more serious, complications, such as massive hemorrhage from carotid arteriotomy after attempted IJV cannulation in patients on cardiopulmonary bypass. 3 Wiseheart et al4 reported a death from massive hemorrhage after IJV cannulation secondary to puncture ofthe ascending cervical artery with tears in the mediastinal and apical pleura and resultant hemothorax. Hansbrough et allS reported two cases of presumed IJV-earotid artery fistulae manifested by cervical bruits. Cervical hematoma that caused aiIway obstruction requiring intubation was reported by Knoblanche. 8 Two cases of tracheal puncture during IJV cannulation diagnosed by sudden leak in the endotracheal tube cuff have been reported. 7 Our patient is the first described to have massive hemoptysis associated with a cervical hematoma and airway obstruction following IJV cannulation. Although the actual mechanism of hemoptysis cannot be established, it is possible that the needle created a fistulous tract between the IJ~ carotid artery, and trachea. This laceration could have provided a route for the development of hemoptysis and a cervical hematoma leading to acute airway obstruction. Alternativel~ hemoptysis may have been caused by lung puncture, as reported during transthoracic needle biops~ However, this is unlikely as the patient did not develop a pneumothorax, and hemoptysis in needle biopsy is usually insignificant. 8 The dramatic, sudden occurrence ofa cervical hematoma, airway obstruction, and massive hemoptysis during insertion of an IJV catheter emphasizes the potential for life-threatening complications during IJV cannulation. REFERENCES 1 English IC~ Frew RM, Pigot JF, Zaki M. Percutaneous catheterization of the internal jugular vein. Anaesthesia 1969; 24:52131 2 Schwartz AJ, Jobes DR, Greenhow DE, Stephenson L~ Ellison N. Carotid artery puncture with internal jugular cannulation. Anesthesiology 1979; 51:160 3 McEnany MT, Austen WG. Life-threatening hemorrhage from Ann Thorac Surg 1977; 24:233inadvertent cervical arteriotom~ 36
4 Wiseheart JD, Hasson MA, Jackson JW A complication ofinternal jugular cannulation. Anaesthesia 1979; 34:1035-37 5 Hansbrough JF, Narrod JA, Rutherford R. Arteriovenous fistulas following central venous catheterization. Inten Care Med 1983; 9:287-89 6 Knoblanche GE. Respiratory obstruction due to hematoma following internal jugular vein cannulation. Anaesth Inten Care 1979; 7:286 7 Konichezky S, Saguib S, Soroker D. Tracheal puncture - a complication of percutaneous internal jugular vein cannulation. Anaesthesia 1983; 38:572-74 8 Berquist TH, Bailey PB, Cortese DA, Miller WE. Transthoracic needle biopsy: accuracy and complications in relation to location and type of lesion. Mayo Clio Proc 1980; 55:475-81 CHEST I 95 I 5 I MA~
1888
1143
Although widely used, pericardial fluid cytology has not been adequately validated. We pooled 22 new cases with 71 previously reported cases having both fluid cytology and pericardial histopathologic examination. Cytology was correct in 87 of 93 cases (diagnostic accuracy, 94 percent). Sensitivity was 87 percent and speci6city was 100 percent. (Chelt 1989; 95:1142-43)
W
hile pericardial fluid cytology is widely utilized, surprisingly there are only three small studies which attempt to demonstrate its diagnostic accuracy. 1-3 To achieve adequate statistical power and to overcome possible selection bias, we pooled the results of the 71 previously reported patients with 22 additional patients. All patients in five community and two university hospitals having pericardial fluid sampled by either needle pericardiocentesis or surgical pericardiectomy plus histologic evaluation of pericardial tissue from either pericardial biopsy (15 cases) or autopsy (eight cases) during a five-year period were identified by review of medical records. The cause of the pericardial effusion in each case was determined from the composite of gross appearance, histologic findings and bacteriologic studies of pericardial tissue together with clinical information. Sensitivity, specificity, predictive values, and diagnostic accuracy plus their 95 percent confidence intervals (CI) were calculated from pooled data. Twenty-two patients were identified. The most frequent cause of effusion in our series as documented by histopathologic examination of pericardium was cancer (adenocarcinoma of the lung in five cases, squamous cell carcinoma of the lung in three, adenocarcinoma of the breast in one, and mesothelioma in one). Other causes included post-pericardiotomy syndrome (one case), postirradiation (four cases), collagen-vascular diseases (two cases), and infection (three cases). No cause was identified by examination of the pericardium in two cases. Cytologic examination of pericardial fluid identified malignant cells in nine of the ten patients with cancer (Table 1). Tissue examination confirmed 12 negative cytologic examinations. There were no false-positive tests but one false-negative test occurred in a patient with squamous cell carcinoma of the lung. *From the Department of Internal Medicine, University of Nebraska College of Medicine, Omaha. Reprint requests: Dr. Meyers, Section of Cardiology, University of
Nebraska Medical Center, Omaha 68105
Cases which were identified as having validation by histopathologic examination of pericardial tissue from three studies of pericardial fluid cytology are pooled with our data in the table. These studies were performed in a university hospital, 1 a Veterans Administration facility, 2 and in patients with cardiac tamponade seen at a cancer institute. 3 Included are 93 patients: 22 with lung cancer, nine with breast cancer, five with undifferentiated pericardial metastases, two with metastases from the colon, three with mesothelioma, four with non-Hodgkin's and two with Hodgkin's lymphoma, 14 with granulomatous pericarditis, four with renal failure, five with radiation pericarditis, three with infectious pericarditis, two vlith lupus erythematosus, one with postpericardiotomy syndrome, 14 with idiopathic pericarditis, and three patients whose diseases were not characterized. The prevalence of cancer involving the pericardium is 50 percent. Cytologic examination produced a sensitivity of 87 percent (95 percent CI = 77 to 96 percent) and a specificity of 100 percent (95 percent CI = 92 to 100 percent). The positive predictive value is 100 percent (95 percent CI = 91 to 100 percent) and the negative predictive value is 88 percent (95 percent CI = 78 to 97 percent). Cytologic studies correctly identified both the presence and absence of cancer (diagnostic accuracy) in 94 percent (95 percent CI = 88 to 97 percent). No false-positive tests occurred. A total of six patients with cancer had false-negative fluid cytology noted in which pericardial tissue examination demonstrated mesothelioma in two cases, undifferentiated carcinoma in two, and squamous cell carcinoma of the lung in one. The etiology in one case was not characterized. Data regarding clinical setting, amount of pericardial fluid, or degree of malignant pericardial involvement in these cases are not available. Malignant involvement of the pericardium and heart is found in from 0.1 to 6.4 percent of unselected autopsies and in up to 20.6 percent ofautopsies ofpatients dying ofcancer. 4 Pericardiocentesis and fluid cytology is a standard diagnostic tool in the evaluation of pericardial effusion. Yet, validation ofpericardial fluid cytology has been lacking. The usefulness of the three previous studies has been limited by their small sample sizes and potential selection bias stemming from their particular case mixes. 1-3 These limitations are overcome by analysis of pooled data from these and the present study, all of which used similar techniques. S The results demonstrate that pericardial fluid cytology is highly accurate. Positive tests results are completely reliable whereas occasional false-negative tests unaccountably occur. REFERENCES 1 Zipf RE, Johnston ww. The role of cytology in the evaluation of pericardial effusions. Chest 1972; 62:593-96 2 Posner MR, Cohen GI, Skarin AT. Pericardial disease in patients
Table 1- Cytologic Study Results Source
TruePositive
FalsePositive
TrueNegative
FalseNegative
Total
Zipf and Johnston l Posner et all Reyes et al3 Present study Total
12 11 9 9 41
0 0 0 0 0
32 2 0 12 46
3 2 0 1 6
47 15 9 22 93
1142
Diagnostic Usefulness of PericardiaJ Fluid Cytology (Meyers, Bouska)
with cancer: the differentiation of malignant from idiopathic and radiation-induced pericarditis. Am J Med 1981; 71:407-13 3 Reyes C~ Strinden C, Banerji M. The role of cytology in neoplastic cardiac tamponade. Acta Cytologica 1982; ,26:299-302 4 King DT, Nieberg RK. The use of cytology to evaluate pericardial effusions. Ann Clin Lab Sci 1979; 9:18-23 5 L'Abbe KA, Detsky AS, O'Rorke K. Meta-analysis in clinical research. Ann Intern Med 1987; 107:224-33
Massive Hemoptysis during Catheterization of the Internal Jugular Vein* Yaakov Friedman, M.D.; Elena nuy, M.D.; Eric C. Rackow, M.D., F.C.C.R; and Max Harry Weil, M.D., Ph.D., F.C.C.R
Percutaneous cannulation of the internal jugular vein is commonly performed to obtain central venous access. We report the first case of massive hemoptysis occurring during cannulation of the internal jugular vein. (Chest 1989; 95:1143)
C
annulation of the internal jugular vein (IJV) is currently accepted as a relatively safe method of obtaining central venous access. 1 The most common complication is carotid artery puncture, which is usually benign. 2 We report a case of massive hemoptysis, cervical hematoma and airway obstruction occurring during IJV catheterization. CASE REPORT
A 51-year-old woman with a history of intravenous drug abuse was being treated for osteomyelitis with a six-week course of piperacillin, oxacillin, and gentamicin. Due to the need for prolonged antibiotic administration, catheterization of the right IJV for venous access was performed. An 18-gauge needle was inserted via the anterior approach with immediate return of venous blood. A .035 mm guidewire was threaded, the vessel dilated with a 7 French dilator, and a 7 French triple lumen catheter was inserted without difficulty. Suddenly, the patient complained of severe chest pain and turned her neck dislodging the catheter. She bled vigorously from the puncture site and developed a large cervical hematoma despite the application of pressure. She rapidly became hypotensive and diaphoretic, and was treated with fluids and infusion of intravenous norepinephrine to maintain blood pressure. She complained of shortness of breath, developed stridor, and had 300 to 400 ml of bright-red hemoptysis. The patient was intubated with some difficulty because of the amount of hemoptysis and transferred to the intensive care unit. The norepinephrine infusion was rapidly tapered and discontinued. There was continued mild bleeding through the endotracheal tube which resolved over two hours. The patient's hematocrit value initially decreased from 40 to 30 percent, then stabilized at 32 percent after transfusion of three units of packed red blood cells. The patient's PT, PTT, and platelet count were within normal limits. The patient was never noted to have a carotid broit or thrill.
Angiography of the cervical and thoracic vessels done two days later was normal and did not demonstrate an arteriovenous fistula. The chest roentgenogram revealed bilateral alveolar infiltrates consistent with blood aspiration, which cleared over eight days. There was no evidence of air in the soft tissues of neck, pneumothorax, or pneumomediastinum at anyone time. The cervical hematoma allowing the patient to be extubated after seven resolved slowl~ days, and resolving completely in three weeks. *From the Division of Critical Care Medicine, Cook County Hospital, Chicago, and University of Health Sciences/fhe Chicago Medical School, North Chicago, IL.
DISCUSSION
Central venous catheterization of the IJV has a low complication rate. l The most common complication is carotid artery puncture, occurring in 4.2 percent of cases, usually without significant consequences. 2 Other complications include pneumothorax, thrombosis of the IoN air embolus, and cerebrovascular accidents. There have been reports of less frequent, but more serious, complications, such as massive hemorrhage from carotid arteriotomy after attempted IJV cannulation in patients on cardiopulmonary bypass. 3 Wiseheart et al4 reported a death from massive hemorrhage after IJV cannulation secondary to puncture ofthe ascending cervical artery with tears in the mediastinal and apical pleura and resultant hemothorax. Hansbrough et allS reported two cases of presumed IJV-earotid artery fistulae manifested by cervical bruits. Cervical hematoma that caused aiIway obstruction requiring intubation was reported by Knoblanche. 8 Two cases of tracheal puncture during IJV cannulation diagnosed by sudden leak in the endotracheal tube cuff have been reported. 7 Our patient is the first described to have massive hemoptysis associated with a cervical hematoma and airway obstruction following IJV cannulation. Although the actual mechanism of hemoptysis cannot be established, it is possible that the needle created a fistulous tract between the IJ~ carotid artery, and trachea. This laceration could have provided a route for the development of hemoptysis and a cervical hematoma leading to acute airway obstruction. Alternativel~ hemoptysis may have been caused by lung puncture, as reported during transthoracic needle biops~ However, this is unlikely as the patient did not develop a pneumothorax, and hemoptysis in needle biopsy is usually insignificant. 8 The dramatic, sudden occurrence ofa cervical hematoma, airway obstruction, and massive hemoptysis during insertion of an IJV catheter emphasizes the potential for life-threatening complications during IJV cannulation. REFERENCES 1 English IC~ Frew RM, Pigot JF, Zaki M. Percutaneous catheterization of the internal jugular vein. Anaesthesia 1969; 24:52131 2 Schwartz AJ, Jobes DR, Greenhow DE, Stephenson L~ Ellison N. Carotid artery puncture with internal jugular cannulation. Anesthesiology 1979; 51:160 3 McEnany MT, Austen WG. Life-threatening hemorrhage from Ann Thorac Surg 1977; 24:233inadvertent cervical arteriotom~ 36
4 Wiseheart JD, Hasson MA, Jackson JW A complication ofinternal jugular cannulation. Anaesthesia 1979; 34:1035-37 5 Hansbrough JF, Narrod JA, Rutherford R. Arteriovenous fistulas following central venous catheterization. Inten Care Med 1983; 9:287-89 6 Knoblanche GE. Respiratory obstruction due to hematoma following internal jugular vein cannulation. Anaesth Inten Care 1979; 7:286 7 Konichezky S, Saguib S, Soroker D. Tracheal puncture - a complication of percutaneous internal jugular vein cannulation. Anaesthesia 1983; 38:572-74 8 Berquist TH, Bailey PB, Cortese DA, Miller WE. Transthoracic needle biopsy: accuracy and complications in relation to location and type of lesion. Mayo Clio Proc 1980; 55:475-81 CHEST I 95 I 5 I MA~
1888
1143