- Email: [email protected]
CT Diagnosis of Emphysema
CHEST
editorials VOLUME 103 I NUMBER 2 I FEBRUARY, 1993
CT Diagnosis of Emphysema It May Be Accurate, but Is It Relevant?
tomography (CT) allows direct demonC omputed stration of the presence, extent, and severity of emphysema, and CT findings correlate closely with pathologic findings. 1-3 Emphysema is characterized on CT by the presence of localized areas of abnormally low attenuation without surrounding walls or with very thin (:Sl-mm diameter) walls. Computed tomography is currently the best method of assessing gross lung morphology short of having the pathologic specimen in hand, and as shown by Bense and coworkers in this issue of Chest (see page 433), CT has the potential of greatly increasing our understanding of the pathogenesis and evolution of lung disease. It is known that the incidence of spontaneous pneumothorax is increased in smokers and that emphysema is found in the vast majority of smokers with spontaneous pneumothorax, suggesting a causal relationship between emphysema and spontaneous pneumothorax. 2 Bense et al provide evidence that emphysemalike changes are also seen on CT in the majority of nonsmoking patients with spontaneous pneumothorax. They compared the CT findings in 27 nonsmoking patients with radiologically verified spontaneous pneumothorax to the CT findings in 10 healthy subjects who had never smoked. Emphysemalike changes were seen on CT in 22 of 27 nonsmoking patients with spontaneous pneumothorax and in none of the 10 control subjects. In none of the cases was the emphysema detected on the chest radiograph. In the patients with spontaneous pneumothorax, the emphysemalike changes were seen mainly in the periphery of the upper lung zones. Why would subjects with spontaneous pneumothorax who had never smoked have these changes? The same authors have previously shown that nonsmoking patients with spontaneous pneumothorax have mild bronchial anomalies, especially bronchial disproportions. It is therefore tempting to postulate a causal link between bronchial anomalies and emphysemalike changes. Emphysema was not seen on CT in five of the patients with spontaneous pneumothorax, and the findings in the other patients were subtle. There was no pathologic proofof the CT finding. These limitations of the study raise the question of how accurate CT is
in the diagnosis of emphysema. Several studies have demonstrated that CT correlates better than pulmonary function tests with the presence and extent of emphysema and that CT may show evidence of emphysema in patients with normal chest radiographs. 4 •5 It should be pointed out, however, that the CT assessment of emphysema is influenced by a variety of factors, including the type of scanner, the thickness of the sections, and the window level and width at which the images are photographed. There is also significant intra- and interobserver variability of the interpretation. Using high-resolution CT and a state-of-the-art scanner, Hruban et al 1 were able to identify even mild emphysema when they scanned 20 postmortem lung specimens. The correlation between the in vitro CT emphysema score and the pathologic grade was at the level ofr=0.91. Although it is possible to detect even mild emphysema and to obtain a near 1:1 correlation in vitro, the correlations in vivo have not been quite as accurate. Using a GE 9800 scanner, we obtained a CT-pathologic correlation of 0.81 when using 10-mm-collimation scans and a correlation of 0.85 when using 1.5-mmcollimation scans. 2 The apparent only minimal improvement of the correlation coefficient using highresolution CT is misleading because even though emphysema can be seen on conventional CT, it is more conspicuous and more reliably diagnosed on high-resolution CT. In our series, 33 of38 patients had emphysema. Of these, four patients with mild centriacinar emphysema and two with mild panacinar emphysema had scans read as showing no emphysema on CT. Furthermore, even in patients in whom emphysema was diagnosed on CT, localized areas of destruction measuring less than 0.5 cm in diameter were often missed. Therefore, mild emphysema may be missed on CT and could have been present in the patients with normal CT findings in the study by Bense et al. Although Kuwano et al3 found no significant difference between the high-resolution CT and pathologic emphysema scores, their study did not include normal controls and therefore cannot be used to claim a high sensitivity of CT in the detection of mild emphysema. While not perfect, CT is currently the most accurate method for diagnosing emphysema in vivo. However, because of its cost it has a very limited role in the clinical assessment of emphysema. In most cases, the CHEST I 103 I 2 I FEBRUARY, 1993
329
diagnosis can be easily made with the combination of clinical history, pulmonary function tests, and chest radiography. Occasionally er may be requested for symptomatic patients with abnormal gas transfer (decreased carbon monoxide diffusing capacity [DLCO]) without evidence of airway obstruction on pulmonary function tests. Klein et al11 reported the cases of ten patients with impaired gas transfer (single-breath DLCO <80 percent predicted), normal How rates (FEV/FVC >80 percent predicted), normal chest radiographs, and evidence of emphysema on highresolution er. All were smokers. In such patients, high-resolution er can be of value in clinical management by differentiating emphysema from pulmonary vascular disease. 6 Another indication for er is in the preoperative assessment of patients with large bullae being referred for bullectomy. 7 •8 Good outcome following surgical treatment is predicted by the presence of large bullae, a rapid onset of dyspnea, restrictive lung function due to compression of more normal areas of lung, and absence of generalized emphysema. The presence of crowding of vessels surrounding the bulla and the presence of emphysema surrounding the bulla or in the contralateral lung can be easily assessed. The less invasive nature of er compared with angiography makes er the examination of choice in the assessment of these patients. Another indication, perhaps, may be in the assessment of patients with recurrent spontaneous pneumothorax. However, while it would demonstrate emphysema in the vast majority of these patients, what difference would that make? It is unlikely that it would change patient management. More exciting at the moment is the use of er in research to increase our understanding of the pathogenesis and evolution of lung disease. There the information hopefully will be not only accurate but also relevant. Nestor L. Maller, M.D., Ph.D., F.C.C.P. Vancouver; Canada Department of Radiology, University of British Columbia, and Vancouver General Hospital. REFERENCES
1 Hruban RH, Meziane MA, Zerhouni EA, Khouri NF, Fishman EK, Wheeler PS, et al. High resolution computed tomography of inflation-fixed lungs: pathologic-radiologic correlation of centrilobular emphysema. Am Rev Respir Dis 1987; 136:935-40 2 Miller RR, Miiller NL, Vedal S, Morrison NJ, Staples CA. Limitations of computed tomography in the assessment of emphysema. Am Rev Respir Dis 1989; 139:980-83 3 Kuwano K, Matsuba K, Ikeda T, Murakami J, Araki A, Nishitani H, et al. The diagnosis of mild emphysema: correlation of computed tomography and pathology scores. Am Rev Respir Dis 1990; 141:169-78 4 Morrison NJ, Abboud RT, Muller NL, Ramadan F, Miller RR, Gibson NN, Evans KG. Comparison of single breath carbon monoxide diffusing capacity and pressure-volume curves in
330
detecting emphysema. Am Rev Respir Dis 1989; 141:1179-87 5 Klein JS, Gamsu G, Webb WR, Golden JA, Miiller NL. Highresolution CT diagnosis of emphysema in symptomatic patients
with normal chest radiographs and isolated low diffusing capacity. Radiology 1992; 182:817-21 6 Webb WR. High-resolution CT of the lung parenchyma. Radio) Clio North Am 1989; 27:1085-97 7 Gaensler EA, Jederlinic PJ, FitzGerald MY. Patient work-up fur bullectomy. J Thorac Imaging 1986; 1:75-93 8 Wade JF III, Mortenson T, Irvin CG. Physiologic evaluation of bullous emphysema. Chest 1991; 100:1151-54
Environmental and Asthma
Tobacco Smoke
cities now oblige smokers to leave enclosed Manyspaces, such as office buildings and bus and
train stations, if they wish to indulge their addiction. There can be little argument that the accumulation of stale tobacco smoke is unpleasant and annoying to both the nonsmoker and the ex-smoker, and this will undoubtedly remain the main reason why a smokefree environment is guaranteed in public places. Nevertheless, there is also some evidence that the concentrations of smoke likely in the absence ·of legislation have adverse consequences for health, including small but unacceptable increases in the risks of bronchial carcinoma, chronic obstructive lung disease, and coronary heart disease.u There has been considerable discussion as to whether exposure to tobacco smoke provokes asthma. 3-6 The tobacco plant undoubtedly produces powerful allergens, as witnessed by skin lesions in tobacco harvesters, and a proportion of subjects with a history of asthma demonstrate wheezing and signs of small-airway spasm with smoke exposure. 6 However, such responses have been inconsistent. Some authors have also argued that no specific allergens have been demonstrated, 7 although immune reactions to tobacco smoke have been observed in basophil leukocytes and immunoglobulin E antibodies. 8 One major source of difficulty in conducting more conclusive experiments is that the odor of tobacco smoke is well known, so that exposure can create psychological reactions including not only tachycardia, 9 •10 but possibly also bronchospasm. 11 Ifthe smoke cloud is dense, it is less easy to detect relative levels of exposure, and one potential way of distinguishing biological from psychological reactions is to demonstrate a doseresponse relationship. 12 The technique adopted by Danuser and associates and described in this issue of Chest (see page 353) is attractive in this regard: the major eye and nasal symptoms that could precipitate psychological reactions are avoided, and nicely graded doses of smoke can be administered to the airways. Like a number of previous investigators, Danuser
editorials VOLUME 103 I NUMBER 2 I FEBRUARY, 1993
CT Diagnosis of Emphysema It May Be Accurate, but Is It Relevant?
tomography (CT) allows direct demonC omputed stration of the presence, extent, and severity of emphysema, and CT findings correlate closely with pathologic findings. 1-3 Emphysema is characterized on CT by the presence of localized areas of abnormally low attenuation without surrounding walls or with very thin (:Sl-mm diameter) walls. Computed tomography is currently the best method of assessing gross lung morphology short of having the pathologic specimen in hand, and as shown by Bense and coworkers in this issue of Chest (see page 433), CT has the potential of greatly increasing our understanding of the pathogenesis and evolution of lung disease. It is known that the incidence of spontaneous pneumothorax is increased in smokers and that emphysema is found in the vast majority of smokers with spontaneous pneumothorax, suggesting a causal relationship between emphysema and spontaneous pneumothorax. 2 Bense et al provide evidence that emphysemalike changes are also seen on CT in the majority of nonsmoking patients with spontaneous pneumothorax. They compared the CT findings in 27 nonsmoking patients with radiologically verified spontaneous pneumothorax to the CT findings in 10 healthy subjects who had never smoked. Emphysemalike changes were seen on CT in 22 of 27 nonsmoking patients with spontaneous pneumothorax and in none of the 10 control subjects. In none of the cases was the emphysema detected on the chest radiograph. In the patients with spontaneous pneumothorax, the emphysemalike changes were seen mainly in the periphery of the upper lung zones. Why would subjects with spontaneous pneumothorax who had never smoked have these changes? The same authors have previously shown that nonsmoking patients with spontaneous pneumothorax have mild bronchial anomalies, especially bronchial disproportions. It is therefore tempting to postulate a causal link between bronchial anomalies and emphysemalike changes. Emphysema was not seen on CT in five of the patients with spontaneous pneumothorax, and the findings in the other patients were subtle. There was no pathologic proofof the CT finding. These limitations of the study raise the question of how accurate CT is
in the diagnosis of emphysema. Several studies have demonstrated that CT correlates better than pulmonary function tests with the presence and extent of emphysema and that CT may show evidence of emphysema in patients with normal chest radiographs. 4 •5 It should be pointed out, however, that the CT assessment of emphysema is influenced by a variety of factors, including the type of scanner, the thickness of the sections, and the window level and width at which the images are photographed. There is also significant intra- and interobserver variability of the interpretation. Using high-resolution CT and a state-of-the-art scanner, Hruban et al 1 were able to identify even mild emphysema when they scanned 20 postmortem lung specimens. The correlation between the in vitro CT emphysema score and the pathologic grade was at the level ofr=0.91. Although it is possible to detect even mild emphysema and to obtain a near 1:1 correlation in vitro, the correlations in vivo have not been quite as accurate. Using a GE 9800 scanner, we obtained a CT-pathologic correlation of 0.81 when using 10-mm-collimation scans and a correlation of 0.85 when using 1.5-mmcollimation scans. 2 The apparent only minimal improvement of the correlation coefficient using highresolution CT is misleading because even though emphysema can be seen on conventional CT, it is more conspicuous and more reliably diagnosed on high-resolution CT. In our series, 33 of38 patients had emphysema. Of these, four patients with mild centriacinar emphysema and two with mild panacinar emphysema had scans read as showing no emphysema on CT. Furthermore, even in patients in whom emphysema was diagnosed on CT, localized areas of destruction measuring less than 0.5 cm in diameter were often missed. Therefore, mild emphysema may be missed on CT and could have been present in the patients with normal CT findings in the study by Bense et al. Although Kuwano et al3 found no significant difference between the high-resolution CT and pathologic emphysema scores, their study did not include normal controls and therefore cannot be used to claim a high sensitivity of CT in the detection of mild emphysema. While not perfect, CT is currently the most accurate method for diagnosing emphysema in vivo. However, because of its cost it has a very limited role in the clinical assessment of emphysema. In most cases, the CHEST I 103 I 2 I FEBRUARY, 1993
329
diagnosis can be easily made with the combination of clinical history, pulmonary function tests, and chest radiography. Occasionally er may be requested for symptomatic patients with abnormal gas transfer (decreased carbon monoxide diffusing capacity [DLCO]) without evidence of airway obstruction on pulmonary function tests. Klein et al11 reported the cases of ten patients with impaired gas transfer (single-breath DLCO <80 percent predicted), normal How rates (FEV/FVC >80 percent predicted), normal chest radiographs, and evidence of emphysema on highresolution er. All were smokers. In such patients, high-resolution er can be of value in clinical management by differentiating emphysema from pulmonary vascular disease. 6 Another indication for er is in the preoperative assessment of patients with large bullae being referred for bullectomy. 7 •8 Good outcome following surgical treatment is predicted by the presence of large bullae, a rapid onset of dyspnea, restrictive lung function due to compression of more normal areas of lung, and absence of generalized emphysema. The presence of crowding of vessels surrounding the bulla and the presence of emphysema surrounding the bulla or in the contralateral lung can be easily assessed. The less invasive nature of er compared with angiography makes er the examination of choice in the assessment of these patients. Another indication, perhaps, may be in the assessment of patients with recurrent spontaneous pneumothorax. However, while it would demonstrate emphysema in the vast majority of these patients, what difference would that make? It is unlikely that it would change patient management. More exciting at the moment is the use of er in research to increase our understanding of the pathogenesis and evolution of lung disease. There the information hopefully will be not only accurate but also relevant. Nestor L. Maller, M.D., Ph.D., F.C.C.P. Vancouver; Canada Department of Radiology, University of British Columbia, and Vancouver General Hospital. REFERENCES
1 Hruban RH, Meziane MA, Zerhouni EA, Khouri NF, Fishman EK, Wheeler PS, et al. High resolution computed tomography of inflation-fixed lungs: pathologic-radiologic correlation of centrilobular emphysema. Am Rev Respir Dis 1987; 136:935-40 2 Miller RR, Miiller NL, Vedal S, Morrison NJ, Staples CA. Limitations of computed tomography in the assessment of emphysema. Am Rev Respir Dis 1989; 139:980-83 3 Kuwano K, Matsuba K, Ikeda T, Murakami J, Araki A, Nishitani H, et al. The diagnosis of mild emphysema: correlation of computed tomography and pathology scores. Am Rev Respir Dis 1990; 141:169-78 4 Morrison NJ, Abboud RT, Muller NL, Ramadan F, Miller RR, Gibson NN, Evans KG. Comparison of single breath carbon monoxide diffusing capacity and pressure-volume curves in
330
detecting emphysema. Am Rev Respir Dis 1989; 141:1179-87 5 Klein JS, Gamsu G, Webb WR, Golden JA, Miiller NL. Highresolution CT diagnosis of emphysema in symptomatic patients
with normal chest radiographs and isolated low diffusing capacity. Radiology 1992; 182:817-21 6 Webb WR. High-resolution CT of the lung parenchyma. Radio) Clio North Am 1989; 27:1085-97 7 Gaensler EA, Jederlinic PJ, FitzGerald MY. Patient work-up fur bullectomy. J Thorac Imaging 1986; 1:75-93 8 Wade JF III, Mortenson T, Irvin CG. Physiologic evaluation of bullous emphysema. Chest 1991; 100:1151-54
Environmental and Asthma
Tobacco Smoke
cities now oblige smokers to leave enclosed Manyspaces, such as office buildings and bus and
train stations, if they wish to indulge their addiction. There can be little argument that the accumulation of stale tobacco smoke is unpleasant and annoying to both the nonsmoker and the ex-smoker, and this will undoubtedly remain the main reason why a smokefree environment is guaranteed in public places. Nevertheless, there is also some evidence that the concentrations of smoke likely in the absence ·of legislation have adverse consequences for health, including small but unacceptable increases in the risks of bronchial carcinoma, chronic obstructive lung disease, and coronary heart disease.u There has been considerable discussion as to whether exposure to tobacco smoke provokes asthma. 3-6 The tobacco plant undoubtedly produces powerful allergens, as witnessed by skin lesions in tobacco harvesters, and a proportion of subjects with a history of asthma demonstrate wheezing and signs of small-airway spasm with smoke exposure. 6 However, such responses have been inconsistent. Some authors have also argued that no specific allergens have been demonstrated, 7 although immune reactions to tobacco smoke have been observed in basophil leukocytes and immunoglobulin E antibodies. 8 One major source of difficulty in conducting more conclusive experiments is that the odor of tobacco smoke is well known, so that exposure can create psychological reactions including not only tachycardia, 9 •10 but possibly also bronchospasm. 11 Ifthe smoke cloud is dense, it is less easy to detect relative levels of exposure, and one potential way of distinguishing biological from psychological reactions is to demonstrate a doseresponse relationship. 12 The technique adopted by Danuser and associates and described in this issue of Chest (see page 353) is attractive in this regard: the major eye and nasal symptoms that could precipitate psychological reactions are avoided, and nicely graded doses of smoke can be administered to the airways. Like a number of previous investigators, Danuser