Downloaded from www.ajronline.org by 24.49.69.7 on 03/27/15 from IP address 24.49.69.7. Copyright ARRS. For personal use only; all rights reserved
695
Diagnostic and Therapeutic Chest Sonography: Value in
Use of Critically
Ill Patients
Chong-Jen Yu1 Pan-Chyr Yang1 Dun-Bing Chang1 Kwen-Tay Luh2
OBJECTIVE. Portable chest radiography, used for critically ill patients, often falls to depict thoracic disease clearly. Chest sonography allows good characterization of pleural diseases, mediastinal lesions, and pulmonary consolidations, and provides accurate and safe guidance for interventional procedures. Accordingly, we evaluated its usefulness in the diagnosis and management of critically ill patients. SUBJECTS AND METhODS. Sonography was used prospectively to evaluate 41 critically ill patients, when portable chest radiographs were difficult to interpret and failed to explain the clinical findings. An initial diagnosis was made on the basis of clinical and radiologic information. The final diagnosis was confirmed by thoracentesis (n = 15), image-guided aspiration biopsy (n = 7), surgery (n = 3), and clinical follow-up (n = 16). All 41 sonographic examinations were performed by a sonographer who had no knowledge of the initial diagnosis. The sonographic findings were recorded and analyzed. The usefulness of sonography in diagnosis and management was then evaluated according to the following criteria: diagnostic value-(1) made diagnosis, (2) changed diagnosis, (3) additional information, (4) no benefit; and management aid-(1) affected decision, (2) aftected decision and guided thoracentesis, (3) affected diagnosis and guided aspiration, (4) no benefit. RESULTS. Chest sonography was helpful in diagnosis in 27 (66%) of 41 patients and in treatment in 37 (90%) of 41 patients. A significant influence on treatment planning occurred in 17 (41%) of 41 patients. A diagnostic aspiration biopsy under sonographic guidance was done without complications in seven patients with occult lung tumors detected sonographically. Thoracenteses were attempted in 25 of 29 patients with pleural effusion and were successful in 24 patients (96%); a minimal pneumothorax developed in one patient (4%). CONCLUSION. Chest sonography is a useful diagnostic tool for critically ill patients with chest diseases. This technique can be particularly helpful when CT is not available or when critically ill patients cannot be moved. AJR
Received February vision May 5, 1992. Supported LTD, Taipei,
24, 1992;
accepted
after re-
in part by Kwang Ten Trading Taiwan, Republic of China.
Co.,
1Department of Internal Medicine, National Taiwan University Hospital, No.7, Chung-Shan S. Rd., Taipei, Taiwan 100, Republic of China. Address reprint requests to P-C. Yang. 2Department wan University Taipei, Taiwan
of Clinical Pathology, National TaiHospital, No.7, Chung-Shan S. Rd., 100, Republic of China.
0361 -803X/92/1 594-0695 CAmerican Roentgen Ray
Society
159:695-701,
October
1992
Portable chest radiography is the most frequently used examination for critically ill patients in the intensive care unit or in the emergency service. These patients need an accurate and timely diagnostic procedure to provide appropriate information for subsequent management. However, portable chest radiographs have several shortcomings that make the interpretation of chest diseases difficult. One of the major problems is the difficulty in discerning the interface between pleural and parenchymal surfaces on portable chest radiographs [1, 2]; thus, it is not always possible to differentiate accurately between pleural and parenchymal opacities. Recent advances in sonographic technology have demonstrated the multipotential capabilities of chest sonography in the diagnosis of pleural diseases [3], peripheral parenchymal lesions [4], mediastinal lesions [5], consolidation, and atelectasis [6]. Transthoracic aspiration biopsy also can be performed safely under real-time sonographic guidance to obtain specimens adequate for histologic diag-
696
YU ET AL.
nosis [7]. The sonographic equipment is portable, and the examination can be done at the bedside if necessary. Previous study of sonography for critically ill patients has been re-
Downloaded from www.ajronline.org by 24.49.69.7 on 03/27/15 from IP address 24.49.69.7. Copyright ARRS. For personal use only; all rights reserved
stncted attempt
to its usefulness has been made
in guiding
thoracentesis
[3].
No
to use chest sonography as a diagnostic tool in critically ill patients. Therefore, a prospective study was performed to ascertain the value of sonography in assisting the diagnosis and management of critically ill patients whose chest radiographs were difficult to interpret. Subjects
and Methods
A prospective study to assess the value of chest sonography in the diagnosis and management of critically ill patients was performed at National Taiwan University Hospital from July 1 990 to April 1991. Patients were referred if: (1) their portable chest radiographs showed abnormal findings that failed to explain adequately the clinical conditions and showed radiographic opacities that were difficult to differentiate as parenchymal or pleural lesions or (2) timely diagnosis was necessary because of their critically ill condition. A total of 41 consecutive patients (31 male and 10 female; age range, 1 4-86 years; mean age, 56 years) were included. Eighteen patients had respiratory failure that necessitated mechanical ventilation (eight had pneumonia, four had empyema, two had massive pleural effusion, one had flail chest, one had total lung collapse, one had esophageal cancer with cervical spine metastasis, and one had mediastinal hemangiomatosis); seven patients had respiratory distress (four had lung tumors and three had pleural effusions), seven patients had aortic aneurysm (five had aortic dissections); four patients had congestive heart failure (one had hypertensive cardiovascular disease, one had dilated cardiomyopathy, one had ischemic heart disease, and one had infective endocarditis); two patients were in a coma (one had recurrent cerebrovascular accidents and one had head injury); one patient was in a postoperative state (Caroli’s disease-segmental hepatectomy); one patient had hepatic failure (fulminant hepatitis); and one patient had sepsis. Among these 41 patients, 1 9 (46%) were examined in the intensive care units, 1 6 (39%) were examined in the wards, and six (1 5%) were examined in the emergency department. The radiographic patterns found in these patients included total or nearly total opacification of a hemithorax in 15 patients, cavitary lesions in three patients, mass lesions in two patients, lobar or segmental consolidation in five patients, and pleural opacities in 1 6 patients. The attending physicians gave each patient a tentative diagnosis based on clinical information and findings on portable chest radiographs as interpreted by a radiologist before the sonographic examination. Final diagnoses were further confirmed by thoracentesis (n = 15), image-guided aspiration biopsy (n = 7), surgery (n = 3), and clinical follow-up (n = 16). In all patients, sonographic examination was done without knowledge of the tentative diagnosis. After viewing the chest radiographs of the patients, a sonographer performed the chest sonography with a portable sonographic machine equipped with a sector, a linear, and a convex transducer (Aloka SSD-256, and SSD-630, Aloka, Tokyo, Japan). The sonograms were recorded on Polaroid film (Polaroid, Cambridge, MA) and analyzed. The sonographic features were classified into pleural space processes (pleural effusion and pleural thickening), lung parenchymal lesions (consolidation, atelectasis, and tumor), and mediastinal lesions (tumor, cyst, and aneurysm). The sonographic patterns for various chest lesions have been described in prior studies [4-6, 8]. Consolidations are homogeneously hypoechoic, wedge-shaped parenchymal lesions, containing air or fluid bronchograms; they move with respiration. Atelectases are lesions with findings similar to those in consolidation, but with volume reduc-
AJR:1 59, October 1992
tion, mainly fluid bronchograms and no air bronchograms. Pleural effusions are spaces of various echogenicities, from anechoic to homogeneously echogenic, which may contain floating strands or complex septa, located between visceral and parietal pleuras. Pleural nodules are round, hypoechoic lesions with well-defined margins located in either the visceral or panetal pleura. Pleural thickenings are arbitrarily defined as thickness greater than 3 mm. When a fluid accumulation was detected sonographically and thoracentesis also was needed, sonography was used to select a favorable site. The skin-to-target depth was measured before thoracentesis. Thoracentesis was conducted with a 21 -gauge needle after local anesthesia. For those patients found to have occult tumor within the lung parenchyma, sonographically guided aspiration biopsy was performed by using a puncture probe, with a technique as described by lzumi et al. [7]. After sonographic examinations and the interventional procedures were completed, the usefulness of sonography in assisting the diagnosis of these critically ill patients was evaluated. Sonographic examinations that met one of the following criteria were considered to have had a positive role in the diagnosis: (1) made diagnosis: a tentative diagnosis was confirmed after sonographic examination, (2) changed diagnosis: a tentative diagnosis was changed after sonographic examination, (3) additional information: provided new information to prior diagnosis. Sonographic examinations were considered to be without benefit in assisting the diagnosis if they provided no additional information. The sonographic examinations were considered to have helped management if (1) a new decision on a patient’s management was made after sonographic examination; (2) thoracentesis, conducted for occult pleural effusion detected sonographically, gave rise to a new decision; (3) aspiration biopsy of the lesion detected by sonography gave rise to a new decision; (4) sonography was used to guide a thoracentesis alone. The sonographic examinations were considered to have no management benefit if the decisions concerning patients’ management, already made on the basis of clinical information and chest radiographic findings, did not change because of the sonographic findings.
Results Sonographic
Features
Pleural effusion was the only finding in 1 9 patients. Ten patients had pleural effusions associated with other parenchymal abnormalities (three with pulmonary consolidation, four with tumors, two with consolidation and tumor, and one with
atelectasis).
included
The
1 7 anechoic
images lesions,
of these five
complex
29
pleural
effusions
nonseptated,
five
complex-septated, and two homogeneously echogenic. Sonography was very useful in the detection of pleural effusions. For two patients, originally considered on the basis of the findings on portable chest radiographs to have lobar consolidation with pleural effusion, parenchymal consolidation was the only finding detected sonographically. In one patient originally considered to have bilateral pulmonary contusions, a moderate pleural effusion was found unexpectedly with sonography (Fig. 1). Eight patients were found to have occult tumors with sonography. Among them, seven patients had hemithoracic opacification on the chest radiographs and two patients had tumor hidden in the consolidated lungs (Figs. 2 and 3). Atelectasis was noted in five cases. The sonogram revealed a picture of consolidation with fluid bronchograms. No air
Downloaded from www.ajronline.org by 24.49.69.7 on 03/27/15 from IP address 24.49.69.7. Copyright ARRS. For personal use only; all rights reserved
AJR:159, October 1992
CHEST SONOGRAPHY
IN CRITICALLY
ILL PATIENTS
697
Fig. 1.-A and B, A 14-year-old boy had multIple fractures from an accident. Initial chest radiograph (not shown) showed possible contusion In rIght lung. Admitted to surgical Intensive care unit, he was put on a ventilator. Follow-up chest radiograph (A) 4 days later showed total opaclficatlon of both lungs. Chest sonogram (B) disclosed bIlateral small to moderate pleural effusions (only left-sided pleural effusion is shown), which were then drained. The follow-up chest radlographs showed pattern of cardiogenic lung edema, and a traumatic ventricular septal defect was dIagnosed surgically.
Fig. 2.-A 71-year-old woman who had had diabetes and hypertension for decades was ad. mifted to the general ward for treatment of pneumonia with heart failure. A, Chest radiograph taken I month after admission shows bilateral pleural effuslons with pneumonia in right lower lobe. B, Chest sonogram shows a central tumor within consolidated lung. Sonographically guided aspIration biopsy confirmed the diagnosIs of adenocarcinoma.
movement within the involved bronchus or respiratory movement of the involved segment was visible on real-time sonography (Fig. 4). Aortic aneurysms were diagnosed on the basis of chest sonography in two patients with chest pain. On the basis of findings on chest radiographs and clinical information, an anterior mediastinal tumor had been diagnosed in one and a lung tumor in the left upper lobe had been diagnosed in the other. Aneurysm had not been diagnosed in either patient before sonography. Sonography showed pulsating anechoic cystic lesions communicating with the aorta in both patients. The diagnoses were confirmed by aortography after the condition stabilized. Sonography failed to provide satisfactory diagnosis for two patients. Both had bronchopleural fistulas with pleural opacities. Sonographic examination did not show whether these patients had a lung abscess or empyema with bronchopleural
fistula. The diagnosis chest CT examination. Interventional Examinations
Diagnostic
was based
Procedures
on results
of subsequent
After Sonographic
Of 29 patients with pleural effusions, 25 underwent thoracentesis. Twenty-four patients had successful thoracentesis (success rate, 96%); thoracentesis failed in one who had only a minimal effusion. All 14 patients undergoing thoracentesis under mechanical ventilation had successful tapping. In one patient (7%) with severe hypoalbuminemia (serum albumin level
695
Diagnostic and Therapeutic Chest Sonography: Value in
Use of Critically
Ill Patients
Chong-Jen Yu1 Pan-Chyr Yang1 Dun-Bing Chang1 Kwen-Tay Luh2
OBJECTIVE. Portable chest radiography, used for critically ill patients, often falls to depict thoracic disease clearly. Chest sonography allows good characterization of pleural diseases, mediastinal lesions, and pulmonary consolidations, and provides accurate and safe guidance for interventional procedures. Accordingly, we evaluated its usefulness in the diagnosis and management of critically ill patients. SUBJECTS AND METhODS. Sonography was used prospectively to evaluate 41 critically ill patients, when portable chest radiographs were difficult to interpret and failed to explain the clinical findings. An initial diagnosis was made on the basis of clinical and radiologic information. The final diagnosis was confirmed by thoracentesis (n = 15), image-guided aspiration biopsy (n = 7), surgery (n = 3), and clinical follow-up (n = 16). All 41 sonographic examinations were performed by a sonographer who had no knowledge of the initial diagnosis. The sonographic findings were recorded and analyzed. The usefulness of sonography in diagnosis and management was then evaluated according to the following criteria: diagnostic value-(1) made diagnosis, (2) changed diagnosis, (3) additional information, (4) no benefit; and management aid-(1) affected decision, (2) aftected decision and guided thoracentesis, (3) affected diagnosis and guided aspiration, (4) no benefit. RESULTS. Chest sonography was helpful in diagnosis in 27 (66%) of 41 patients and in treatment in 37 (90%) of 41 patients. A significant influence on treatment planning occurred in 17 (41%) of 41 patients. A diagnostic aspiration biopsy under sonographic guidance was done without complications in seven patients with occult lung tumors detected sonographically. Thoracenteses were attempted in 25 of 29 patients with pleural effusion and were successful in 24 patients (96%); a minimal pneumothorax developed in one patient (4%). CONCLUSION. Chest sonography is a useful diagnostic tool for critically ill patients with chest diseases. This technique can be particularly helpful when CT is not available or when critically ill patients cannot be moved. AJR
Received February vision May 5, 1992. Supported LTD, Taipei,
24, 1992;
accepted
after re-
in part by Kwang Ten Trading Taiwan, Republic of China.
Co.,
1Department of Internal Medicine, National Taiwan University Hospital, No.7, Chung-Shan S. Rd., Taipei, Taiwan 100, Republic of China. Address reprint requests to P-C. Yang. 2Department wan University Taipei, Taiwan
of Clinical Pathology, National TaiHospital, No.7, Chung-Shan S. Rd., 100, Republic of China.
0361 -803X/92/1 594-0695 CAmerican Roentgen Ray
Society
159:695-701,
October
1992
Portable chest radiography is the most frequently used examination for critically ill patients in the intensive care unit or in the emergency service. These patients need an accurate and timely diagnostic procedure to provide appropriate information for subsequent management. However, portable chest radiographs have several shortcomings that make the interpretation of chest diseases difficult. One of the major problems is the difficulty in discerning the interface between pleural and parenchymal surfaces on portable chest radiographs [1, 2]; thus, it is not always possible to differentiate accurately between pleural and parenchymal opacities. Recent advances in sonographic technology have demonstrated the multipotential capabilities of chest sonography in the diagnosis of pleural diseases [3], peripheral parenchymal lesions [4], mediastinal lesions [5], consolidation, and atelectasis [6]. Transthoracic aspiration biopsy also can be performed safely under real-time sonographic guidance to obtain specimens adequate for histologic diag-
696
YU ET AL.
nosis [7]. The sonographic equipment is portable, and the examination can be done at the bedside if necessary. Previous study of sonography for critically ill patients has been re-
Downloaded from www.ajronline.org by 24.49.69.7 on 03/27/15 from IP address 24.49.69.7. Copyright ARRS. For personal use only; all rights reserved
stncted attempt
to its usefulness has been made
in guiding
thoracentesis
[3].
No
to use chest sonography as a diagnostic tool in critically ill patients. Therefore, a prospective study was performed to ascertain the value of sonography in assisting the diagnosis and management of critically ill patients whose chest radiographs were difficult to interpret. Subjects
and Methods
A prospective study to assess the value of chest sonography in the diagnosis and management of critically ill patients was performed at National Taiwan University Hospital from July 1 990 to April 1991. Patients were referred if: (1) their portable chest radiographs showed abnormal findings that failed to explain adequately the clinical conditions and showed radiographic opacities that were difficult to differentiate as parenchymal or pleural lesions or (2) timely diagnosis was necessary because of their critically ill condition. A total of 41 consecutive patients (31 male and 10 female; age range, 1 4-86 years; mean age, 56 years) were included. Eighteen patients had respiratory failure that necessitated mechanical ventilation (eight had pneumonia, four had empyema, two had massive pleural effusion, one had flail chest, one had total lung collapse, one had esophageal cancer with cervical spine metastasis, and one had mediastinal hemangiomatosis); seven patients had respiratory distress (four had lung tumors and three had pleural effusions), seven patients had aortic aneurysm (five had aortic dissections); four patients had congestive heart failure (one had hypertensive cardiovascular disease, one had dilated cardiomyopathy, one had ischemic heart disease, and one had infective endocarditis); two patients were in a coma (one had recurrent cerebrovascular accidents and one had head injury); one patient was in a postoperative state (Caroli’s disease-segmental hepatectomy); one patient had hepatic failure (fulminant hepatitis); and one patient had sepsis. Among these 41 patients, 1 9 (46%) were examined in the intensive care units, 1 6 (39%) were examined in the wards, and six (1 5%) were examined in the emergency department. The radiographic patterns found in these patients included total or nearly total opacification of a hemithorax in 15 patients, cavitary lesions in three patients, mass lesions in two patients, lobar or segmental consolidation in five patients, and pleural opacities in 1 6 patients. The attending physicians gave each patient a tentative diagnosis based on clinical information and findings on portable chest radiographs as interpreted by a radiologist before the sonographic examination. Final diagnoses were further confirmed by thoracentesis (n = 15), image-guided aspiration biopsy (n = 7), surgery (n = 3), and clinical follow-up (n = 16). In all patients, sonographic examination was done without knowledge of the tentative diagnosis. After viewing the chest radiographs of the patients, a sonographer performed the chest sonography with a portable sonographic machine equipped with a sector, a linear, and a convex transducer (Aloka SSD-256, and SSD-630, Aloka, Tokyo, Japan). The sonograms were recorded on Polaroid film (Polaroid, Cambridge, MA) and analyzed. The sonographic features were classified into pleural space processes (pleural effusion and pleural thickening), lung parenchymal lesions (consolidation, atelectasis, and tumor), and mediastinal lesions (tumor, cyst, and aneurysm). The sonographic patterns for various chest lesions have been described in prior studies [4-6, 8]. Consolidations are homogeneously hypoechoic, wedge-shaped parenchymal lesions, containing air or fluid bronchograms; they move with respiration. Atelectases are lesions with findings similar to those in consolidation, but with volume reduc-
AJR:1 59, October 1992
tion, mainly fluid bronchograms and no air bronchograms. Pleural effusions are spaces of various echogenicities, from anechoic to homogeneously echogenic, which may contain floating strands or complex septa, located between visceral and parietal pleuras. Pleural nodules are round, hypoechoic lesions with well-defined margins located in either the visceral or panetal pleura. Pleural thickenings are arbitrarily defined as thickness greater than 3 mm. When a fluid accumulation was detected sonographically and thoracentesis also was needed, sonography was used to select a favorable site. The skin-to-target depth was measured before thoracentesis. Thoracentesis was conducted with a 21 -gauge needle after local anesthesia. For those patients found to have occult tumor within the lung parenchyma, sonographically guided aspiration biopsy was performed by using a puncture probe, with a technique as described by lzumi et al. [7]. After sonographic examinations and the interventional procedures were completed, the usefulness of sonography in assisting the diagnosis of these critically ill patients was evaluated. Sonographic examinations that met one of the following criteria were considered to have had a positive role in the diagnosis: (1) made diagnosis: a tentative diagnosis was confirmed after sonographic examination, (2) changed diagnosis: a tentative diagnosis was changed after sonographic examination, (3) additional information: provided new information to prior diagnosis. Sonographic examinations were considered to be without benefit in assisting the diagnosis if they provided no additional information. The sonographic examinations were considered to have helped management if (1) a new decision on a patient’s management was made after sonographic examination; (2) thoracentesis, conducted for occult pleural effusion detected sonographically, gave rise to a new decision; (3) aspiration biopsy of the lesion detected by sonography gave rise to a new decision; (4) sonography was used to guide a thoracentesis alone. The sonographic examinations were considered to have no management benefit if the decisions concerning patients’ management, already made on the basis of clinical information and chest radiographic findings, did not change because of the sonographic findings.
Results Sonographic
Features
Pleural effusion was the only finding in 1 9 patients. Ten patients had pleural effusions associated with other parenchymal abnormalities (three with pulmonary consolidation, four with tumors, two with consolidation and tumor, and one with
atelectasis).
included
The
1 7 anechoic
images lesions,
of these five
complex
29
pleural
effusions
nonseptated,
five
complex-septated, and two homogeneously echogenic. Sonography was very useful in the detection of pleural effusions. For two patients, originally considered on the basis of the findings on portable chest radiographs to have lobar consolidation with pleural effusion, parenchymal consolidation was the only finding detected sonographically. In one patient originally considered to have bilateral pulmonary contusions, a moderate pleural effusion was found unexpectedly with sonography (Fig. 1). Eight patients were found to have occult tumors with sonography. Among them, seven patients had hemithoracic opacification on the chest radiographs and two patients had tumor hidden in the consolidated lungs (Figs. 2 and 3). Atelectasis was noted in five cases. The sonogram revealed a picture of consolidation with fluid bronchograms. No air
Downloaded from www.ajronline.org by 24.49.69.7 on 03/27/15 from IP address 24.49.69.7. Copyright ARRS. For personal use only; all rights reserved
AJR:159, October 1992
CHEST SONOGRAPHY
IN CRITICALLY
ILL PATIENTS
697
Fig. 1.-A and B, A 14-year-old boy had multIple fractures from an accident. Initial chest radiograph (not shown) showed possible contusion In rIght lung. Admitted to surgical Intensive care unit, he was put on a ventilator. Follow-up chest radiograph (A) 4 days later showed total opaclficatlon of both lungs. Chest sonogram (B) disclosed bIlateral small to moderate pleural effusions (only left-sided pleural effusion is shown), which were then drained. The follow-up chest radlographs showed pattern of cardiogenic lung edema, and a traumatic ventricular septal defect was dIagnosed surgically.
Fig. 2.-A 71-year-old woman who had had diabetes and hypertension for decades was ad. mifted to the general ward for treatment of pneumonia with heart failure. A, Chest radiograph taken I month after admission shows bilateral pleural effuslons with pneumonia in right lower lobe. B, Chest sonogram shows a central tumor within consolidated lung. Sonographically guided aspIration biopsy confirmed the diagnosIs of adenocarcinoma.
movement within the involved bronchus or respiratory movement of the involved segment was visible on real-time sonography (Fig. 4). Aortic aneurysms were diagnosed on the basis of chest sonography in two patients with chest pain. On the basis of findings on chest radiographs and clinical information, an anterior mediastinal tumor had been diagnosed in one and a lung tumor in the left upper lobe had been diagnosed in the other. Aneurysm had not been diagnosed in either patient before sonography. Sonography showed pulsating anechoic cystic lesions communicating with the aorta in both patients. The diagnoses were confirmed by aortography after the condition stabilized. Sonography failed to provide satisfactory diagnosis for two patients. Both had bronchopleural fistulas with pleural opacities. Sonographic examination did not show whether these patients had a lung abscess or empyema with bronchopleural
fistula. The diagnosis chest CT examination. Interventional Examinations
Diagnostic
was based
Procedures
on results
of subsequent
After Sonographic
Of 29 patients with pleural effusions, 25 underwent thoracentesis. Twenty-four patients had successful thoracentesis (success rate, 96%); thoracentesis failed in one who had only a minimal effusion. All 14 patients undergoing thoracentesis under mechanical ventilation had successful tapping. In one patient (7%) with severe hypoalbuminemia (serum albumin level
