Contribution of Emphysema and Small Airways in COPD*

Transcription

1 Contribution of Emphysema and Small Airways in COPD* Arthur F. Gelb, MD, FCCP; James C. Hogg, MD; Nestor L. Miiller, MD, PhD, FCCP; Mark]. Schein, MD; Joseph Kuei, MD, FCCP; Donald P. Tashkin, MD, FCCP; Joel D. Epstein, MD; Jozef Kollin, MD; Robert H. Green, MD; Noe li.irnel, MD, FCCP; W. Mark Elliott, PhD; and Lida Hadjiaghai, MS Background: The contribution and role of emphysema and small airways disease in causing expiratory airflow limitation in COPD is controversial. Methods: We obtained high-resolution thin-section 2-mm CT scans of the lung for emphysema grading and lung function in 116 consecutively seen COPD outpatients with fixed expiratory airflow limitation. In this group, inflated whole lung(s) were subsequently obtained in 24 patients (23 autopsy, 1 surgery) for morphologic studies and results compared with lung CT. Airway histologic condition was studied in 17 of the 24 patients. Results: There was fair to weak negative correlation between CT emphysema score and either FEV 1/FVC percent (r=-0.51, p=0.001) or FEV1 percent predicted (r=-0.31, p=0.001). In only 24 of the 81 patients (30%) with FEV 1 less than 50% predicted, the CT emphysema score was 60 or more, indicating severe emphysema. In the 24 patients studied, there was a good correlation (r=0.86, p=0.001) between CT and pathologic grade of emphysema. While respiratory bronchioles (RBs) and membranous bronchioles (MBs) demonstrated marked morphologic abnormalities, there was a weak correlation with emphysema grade (for RB, r=0.36, p=0.16; for MB, r=0.41, p=0.10) or with FEV1 percent predicted (for RB, r=-0.21, p=0.42; for MB, r=-0.28, p=0.28). There was no correlation between emphysema and FEV1 percent predicted (r=-0.13, p=0.54). Conclusions: High-resolution CT lung scans are an in vivo surrogate to quantitate moderate to severe morphologic emphysema. Emphysema does not appear to be primarily responsible for severe expiratory airflow limitation in most patients with severe COPD. There was no correlation between severity of small airway histologic condition and emphysema or FEV 1 percent predicted. The causes of the lesions responsible for small airways obstruction need to be identified. (CHEST 1996; 109:353-59) Dco=diffusion of carbon monoxide; MB=membranous bronchiole; RB=respiratory bronchiole Key words: COPD; CT lung; diffusing capacity; emphysema; pulmonary function; small aiiways disease The pathophysiologic role and contribution of emphysema and small airways abnormalities in fixed expiratory airflow obstruction in COPD remain controversial.1 2 Previous studies correlating lung function with morphologic abnormalities were limited by the availability of either autopsy specimens or lobes or lungs obtained at surgery. Results indicated small airway morphologic lesions showed good 3-8 to poor 9-12 correlation with chronic expiratory airflow limitation. Additionally, investigators have concluded that emphysema was or was not the most important *From the Pulmonary Division, Department of Medicine (Drs. Gelb, Kuei, and Epstein) and Department of Radiology (Dr. Schein) and Pathology (Drs. Kuei and Green), Lakewood lfegional Medical Center, and University of California Los Angeles, S'chool of Medicine (Drs. Gelb and Tashkin); and Department of Biomathematics, UCLA School of Medicine (Dr. Hadjiaghai); and Pulmonary Division, Department of Medicine U nivers1ty of Toronto, Schoo[ of Medicine, Toronto, Canada (Dr. Zamel); and Departments of Radiology (Dr. Muller) and Pathology (Drs. Hogg and Elliott) and P u l mresearch o n ~ Laboratory, St. Paul's Hospital, University of British Columbia, Vancouver, British Columbia, Canada. Manuscript received May 18, 1995; revision accepted September 6. Reprint requests: Dr. Gelb, 3650 E South St, Sutte 308, Lakewood, CA lesion causing expiratory airflow limitation in COPD. High-resolution, thin-section CT scans of the lungs have been noted to correlate closely with the morphologic diagnosis of emphysema in patients with mild to moderate airflow limitation who have undergone lung resection. Our initial results in 56 patients concluded that CT-scored emphysema was not predominantly responsible for airflow limitation in COPD. 22 However, no morphologic corroboration was available in that study. The present study explores the relationship between moderate to severe expiratory airflow limitation and the presence and extent of morphologic and CT scored emphysema in 116 consecutively seen outpatients with COPD. MATERIALS AND METHODS We studied 116 patients (62 male and 54 female), aged 68::'::7 years (mean::'::1 SO) with fixed expiratory airflow limitation who were consecutively treated in an outpatient chest clinic at Lakewood Regional Medical Center from 1990 to All but five patients were long-term, regular cigarette smokers with a mean smoking history of 53::'::20 (mean::': 1 SD) pack-years. Our initial study2 2 in- CHEST 1109 I 2 I FEBRUARY,

2 Table!-Characteristics of the 116 Patients (62 Male, 54 Female) With Chronic Airflow Limitation Aged 68±7 Years (mean±l SD), Range 50 to 83 Years* Characteristic Mean::'::1 SD (Range) FVC, L, mean±1 SD 2.4::'::0.9 ( ) FVC, % predicted 68::+::19 (28-125) FEVb L 1.1::'::0.5 ( ) FEV1, %predicted 44::'::16 (15-103) FEV1/FVC,% 46::+::12 (21-75) FEF25-75, Us 0.5::'::0.3 ( ) FEF % predicted 18::'::9 (6-41 ) TLC, L 5.7::'::1.6 ( ) TLC, % predicted 111::'::14 (80-150) Alveolar volume 4.5::'::1.3 ( ) single breath, L Dco, muminlmm Hg 12::'::7 (3-36) Dco % predicted 57::+::31 (14-133) CT emphysema grade 32::'::27 (0-90) *FEF25-7s=expiratory flow between 25% and 75% of expired FVC; TLC=total lung capacity. eluded 56 of these patients. Consistent with our protocol, as approved by the Institutional Review Board for human investigation, we obtained pulmonary function studies and high-resolution thin-section 2-mm CT lung scans in all patients as previously described. 22 Only 12 patients met the American Thoracic Society criteria for chronic bronchitis. 23 Lung Function Studies As previously described, 22 all pulmonary function studies were obtained in the same laboratory using methods consistent with American Thoracic Society recommendations when the patients were in clinically stable conditions and compared with predicted values Tests were obtained initially at the time of obtaining the high-resolution CT lung scan and every 6 to 12 months thereafter. Lung CT Studies As previously described, high spatial frequency reconstruction algorithm, 2 mm collimation, high-resolution CT scans were obtained using a scanner (Picker 1200 SX) in the prone position with seven slices obtained from lung apex to base at 3-cm intervals at end inspiration. 22 Areas of low attenuation and vascular obliteration were graded for emphysema using a picture grading system 29 adapted for CT. 17.I All the CT lung scans were read by one radiologist (M.J.S.) and 78 were read independently by another (N.L.M.). Neither radiologist was aware of the clinical, anatomic, and physiologic results. Average scores for the two radiologists were used in the data analysis. Morphologic Studies During this study, 30 patients died and autopsy was performed (by J.K. and R.H.G.) in 23 patients; 3 deaths occurred from malignancy and 20 from respiratory failure, including 3 with focal pneumonia. One whole lung was obtained at surgery. The lungs were prepared (W.M.E.) and cut into multiple transaxial slices similar to CT from apex to base and graded by a pathologist (J.C.H.) for gross emphysema score using a modification 29 of the picture grading technique of Thurlbeck et at.3 In 17 patients, three to six stratified random blocks of lung tissue were cut from the medial and lateral parts of the lung slices without pneumonia using a template that allowed for correction for tissue shrinkage during processing into paraffin. Five-micrometer sections were cut into glass slides and stained with Hematoxylin-eosin, Masson trichrome, and periodic :a u =ij 80. I ~. Q t '1ii g o I o *oi > "o o o: oo 1 10 : I w II Ooo Emphysema Grade FIGURE 1. Relation between FEV1 percent predicted and CT emphysema grade in 116 patients with chronic airflow limitation (closed circles) and 24 patients with morphologic emphysema grade (open circles). acid Schiff stains. The extent and severity of the disease in the small membranous bronchioles ( <2 mm internal diameter) and respiratory bronchioles were evaluated by a pathologist (J.C.H.) using a standard set of photomicrographs 31 to grade epithelial changes, inflammatory cellular infiltration, connective tissue deposition, and carbon pigment deposition. The pathologist was not aware of the clinical, radiographic, and physiologic results. Statistical Analysis The significance of the difference between two group means was determined using a two-sample t test. Pooled variance estimates were used in the t tests, unless results of tests for homogeneity of variance indicated a significant difference between the variances of two groups, in which case separate variance estimates were used. The degree of linear association between two continuous measures was assessed using Pearson correlation coefficients. All tests were Table 2-Correlation Between Lung Function and CT Emphysema Score in 116 Patients With Chronic Airflow Limitation* Pearson Correlation Lung Function Coefficiant (r) p Value FVC, % predicted 0.09 <0.315 FEV1,% predicted FEF25.75, % predicted <0.001 FEV1/FVC,% TLC, % predicted 0.28 <0.002 Dco, % predicted when <0.001 FEV1 2:1 L Dco, % predicted when <0.001 FEV 1 2:50% predicted DcoNA when FEV 1 2: <0.001 L DcoN A when FEV < :50% predicted Dco, % predicted when <0.067 FEV1 <1 L Dco, % predicted when <0.001 FEV1 <50% predicted DeaNA when FEV <0.103 <ll DcoN A when FEV <0.001 <50% predicted *See Table 1 for abbreviations. VA=alveolar volume. Clinical Investigations

3 Table 3-Prevalence of CT-Scored Emphysema in Patients With Chronic Fixed Airflow Limitation Emphysema Score*(%) FEVt, % Predicted N : :60 < (43) 8 (10) 14 (17) 24 (30) (67) 3 (12) 5 (21) 0 (0) 2: (46) 2 (18) 2 (18) 2 (18) *:520=no or trivial emphysema; 21 to 39=mild emphysema; 40 to 59=moderate emphysema; and 2:60=severe emphysema. two tailed. Statistical significance was defined as a p value less than RESULTS Correlation Between Lung Function and CT Emphysema Score Physiologic characteristics of the patients appear in Table 1. The mean FEV1/FVC percent was 46± 12% (mean±1 SD). The mean FEV 1 percent predicted was 44± 16%. The 78 high-resolution thin-section CT lung scans that were read independently by two radiologists (M.J.S., N.L.M.) yielded grade scores that were comparable; r=0.87, p<o.ool. The mean CT emphysema score was 32±27 (mean±1 SD) in 116 patients. There was only a fair to poor correlation between lung volumes and forced expiratory flow rates and CT emphysema grade (Fig 1, Table 2). As shown in Table 3, in 35 of the 81 patients with severe expiratory airflow limitation, eg, FEV 1 less than 50% predicted, the emphysema CT grade was 20 or less, indicating no or trivial emphysema. Furthermore, only 14 of the 81 patients with FEV 1 less than 50% predicted had moderate to severe CT emphysema, ie, grade of 40 or more, and only 24 of 81 had CT score more than 60 indicating severe emphysema. Paradoxically, in 4 of 11 patients with only mild airflow limitation, ie, FEV1 of 70%, 76%, 83%, and 103% predicted, the corresponding CT emphysema grades were 70, 60, 40, and 45, respectively (Table 3) indicating moderate to severe gross emphysema. Furthermore, 2 of 26 patients with severe CT emphysema grade of 60 or more and 2 of 21 patients with moderate CT emphysema score 40 to 59 had FEV1 of 70% predicted or more. Correlation Between Diffusing Capacity and CT Emphysema Score The relationship between diffusing capacity of carbon monoxide (Dco) and lung CT emphysema score is described in Table 2. In patients with FEV1 of 1 L or more and 50% or more predicted and no or trivial CT emphysema (grade <20), the Dco was abnormal ( <76% predicted) in 6 of 27 patients. However, when the FEV 1 was less than 1 L, 18 of 46 patients had a reduced Dco despite CT lung scan emphysema grade of less than 20. As shown in Table 2, a strong correlation between Dco and CT emphysema grade was found only in the patients with an FEV 1 of 1 Lor more (r=-0.75, p<0.001), and FEV1 of 50% predicted or more (r=-0.76, p<0.001). The relationship between Dco/(V A) and CT emphysema grade was similar to that between Dco and CT emphysema grade (Table 2). Correlation Between Diffusing Capacity and Emphysema Morphology ( 16 Cases) The Dco was abnormal ( <76% predicted) in all eight patients with an FEV1less than 1 Land less than 50% predicted who had morphologic emphysema grade of 30 or more and in one of two patients with emphysema score of 25. However, it was also abnormal in two of four patients who had morphologic emphysema score of 10 or less. In the three patients studied with FEV1 more than 1 L, the Dco was abnormal when the corresponding emphysema score was 50 and 60 and remained normal when the emphysema score was 5. Correlation Between Lung CT and Morphology Score for Emphysema (24 Cases) The 24 CT lung scans that were read independently by two radiologists (M.J.S., N.L.M.) yielded comparable scores (r=0.90, p<0.001 ). Results of the 23 autopsied and 1 whole lung obtained at surgery scored for gross emphysema revealed strong morphologic correlation with the average CT emphysema score (r=0.86, p=0.001). The mean emphysema score±1 SD for Pathology FtcURE 2. Correlation between lung CT and pathology for emphysema scoring (n=24; r=0.86, p=o.ool ). CHEST I 109 I 2 I FEBRUARY,

4 Table 4-Comparison of Airway Scores (Mean±l SD) With Previous Preoperative Structure Function Studies Present Study (n=17) Total score MB* 251±38 Total score RB 159:!:42 Smoking history, pack-years 53±20 FEV/FVC, % 46±12 FEV 1, % preructed 44:!:16 Age, yr 68:!:7 Emphysema score 35:!:24 Cosio et al 8 Hogg et al 12 Wright et al 9 (n=ll) (n=42) (n=20) 238:!:72 244:!:31 174:!: 15 NR* 120:!:52 129:!: 12 41:!:5 47±28 50:!:6 56:!:4 69:!:8 NR NR <60 <60 63:!:2 63:!:11 65:!:2 37:!:31 30:!:16 20:!:6 *NR=not reported. morphology was 35±24 and for CT 38±28. There were three instances in which the average CT score overestimated the emphysema morphology score by a grade of 20, 25, and 30, and one case in which the average CT score underestimated the emphysema morphology score by 25 (Fig 2). Small Airway Histopathology Small airways scores were determined in 17 patients by a pathologist (J.C.H.) after analyzing 7±2 respiratory bronchioles (RBs) ( mean±1 SD) and 9±3 membranous bronchioles (MBs) per patient. The total pathologic score for the MBs was 251±38 (mean±1 SD ), and range was 171 to 315; and for the RBs, the score was 159±42 (mean± 1 SD ), and range was 94 to 236 (Table 4). Scores for age and smoking matched controls with normal lung function from Vancouver, British Columbia, Canada for MBs were 118±47 12 and for RBs were 83±37 12 and 77±12 9. The major abnormalities in the MBs included wall inflammation, fibrosis, muscle hypertrophy, and goblet cell metaplasia, and in the RB wall, inflammation, fibrosis, and muscle hypertrophy. Correlation Between Emphysema and Small Airways Analysis of the data failed to establish a significant linear correlation between airway scores for MB and RB and morphologic emphysema severity (for MB, r=0.41, p=0.10; and for RB, r=0.36, p=0.16, respectively). Comparison Between Lung Function and Morphology There was no correlation between MB and RB scores and FEV1 percent predicted (for MB, r=-0.28, p=0.28; for RB, r=-0.21, p=0.42), or between FEV 1 percent predicted and morphologic emphysema score (r=0.13, p=0.54). DISCUSSION Results of the present study indicate that most outpatients studied with severe COPD do not have moderate or severe CT-scored emphysema (grade 2>::40). 356 Only 14 of 81 patients (17%) with severe airflow limitation, FEV 1 less than 50% predicted, had CT moderate emphysema grade of 40 or more and 24 of the 81 ( 30%) had severe emphysema CT scores of 60 or more, while 35 patients ( 43%) had no or trivial CT emphysema scores of 20 or less. A relatively lower prevalence of CT moderate to severe emphysema score of 40 or more was also noted in patients with mild to moderate respiratory airflow limitation (Table 3). Paradoxically, 10% of patients with moderately severe emphysema CT scores of 40 to 59 and 8% with severe emphysema CT scores of 60 or more had FEV 1 of 70% predicted or more. We believe these results based on lung CT scoring are valid since there was a significant correlation for emphysema between lung CT and whole lung morphology in 24 patients (r=0.86, p<0.001). These findings extend our original observations in 56 patients without morphologic corroboration 22 and suggest that emphysema is not the major cause of chronic expiratory airflow limitation in COPD. However, despite marked small airway histopathologic features, there was no significant correlation of these variables with FEV1 percent predicted (r=-0.28, p=0.28) or emphysema score (r=-0.13, p=0.54). Correlation Between Lung CT and Morphology for Emphysema The good corroboration between lung CT and whole lung morphology for emphysema scoring in the present study extends previous results in patients with less severe emphysema and airflow limitation who underwent surgery or postmortem CT. Previous studies reported good concordance between lung CT and surgical morphology for emphysema scoring using 5- to 13-mm collimation (r=0.74 to 0.90, p<o.ool) or 1.5- to 2-mm collimation (r=0.85 to 0.91, p<0.001). Muller et al 33 and Morrison et al, using 10-mm collimation, and Miller et al, 37 using 1.5-mm collimation and a picture grading system similar to the present study, noted lung CT tended to underestimate mild morphologic emphysema score of 10 or less. In the present study, despite good corroboration between CT and morphologic emphysema score, Clinical Investigations

5 there remained sufficient variation such that CT score could underestimate or overestimate the morphologic extent and severity of emphysema (Fig 2). Lung CT and Diffusing Capacity in Emphysema We have previously reported that airflow limitation particularly at low lung volumes (normal or borderline FEV 1) and decreased diffusing capacity, reflecting a reduction in alveolar capillary surface area, could detect clinically unsuspected yet significant emphysema grade of20 or more as assessed by morphologic studies of inflated lobes or lungs surgically resected However, in the present study, we noted a decreased diffusing capacity in one of two patients with emphysema grade 25. Alternatively, our original observation 39 in surgical specimens may have underestimated the extent and severity of emphysema scored. The spurious low diffusing capacity obtained in two of the four patients with no or trivial morphologic emphysema confirms our original observation 22 without anatomic corroboration. Diffusing capacity cannot be relied on to detect emphysema when the FEV 1 is less than 1L. In severe obstructive lung disease, inhomogeneity of ventilation and increased physiologic dead-space may lead to spurious diffusing capacity measurements by unevenly s a m p l i units n ~ lwith u n g washout time constants As previously noted, 2 analysis of patient tracings failed to detect differences in breath-holding time, washout volume, or alveolar sample size to account for unexplained diffusing capacity values. In patients undergoing lung resection, Morrison et v a r y i n ~ al 20 noted in the 18 patients who had CT emphysema grade of 20 or more, 15 had an abnormal diffusing capacity, and Kuwano et af 8 reported that in the 13 patients who had CT emphysema grades of 20 or more, only 4 had a low diffusing capacity. Klein et al 44 reported a close inverse correlation between high-resolution, thin-section CT emphysema score and DCO (r=-0.643) in symptomatic patients with normal expiratory airflow rates. Correlation Between Airflow Limitation and Emphysema The present study noted no correlation between FEV 1 percent predicted and either CT or morphologic scored emphysema. Similar observations were noted by Gould et al, 19 Morrison et al, 32 and Kuwano et al 18 using CT with good morphologic corroboration for emphysema in surgical specimens. A recent structurefunction study by Hogg et al 12 in 407 patients who underwent lung resection demonstrated that despite progressive deterioration of FEV 1 from 70 to 79% to less than 50% predicted, the prevalence (50%) and mean severity score 27±15 (mean±1 SD) of emphysema (scores >8) remained unchanged. Furthermore, there was no significant (p>0.05) measurable loss of static lung elastic recoil pressure at total lung capacity and at functional residual capacity to explain the observed decrease in FEV1. Conversely, 10 of 54 patients with FEV 1 more than 100% predicted had gross emphysema scores of 24±17 (mean±1 SD).l 2 These observations 12 reinforce similar results in the present study that progressive expiratory airflow limitation is independent of extent and severity of emphysema. Discrepancy with previous reports that suggest emphysema was primarily responsible for airflow limitation in COPD is not easily resolved, but may in part be related to patient selection limited by the availability of autopsy specimens or lobes or lungs obtained at surgery, patient sample size, and characteristics, including but not limited to smoking history, lung function and exposure to environmental pollutants, and interpretation of morphologic studies. Correlation Between Airflow Limitation and Small Airways Abnormalities In contrast to the results of Cosio et al, 8 it has been reported by Wright et al 9 and Hogg et al 12 that a poor correlation exists between decreasing FEV 1 percent predicted and an increase in the pathologic scores of either the RBs or MBs after matching for age and smoking history despite the absence of progressive emphysema. Total airway scores ranged from 118±47 (mean±1 SD) to 155±60 in the MBs and 83±37 to 102±43 in the RBs, as FEV1 declined from more than 110 to less than 50% predicted.l 2 However, the apearent discrepancy between the findings of Hogg et al, 2 Wright et al, 9 and Cosio et al 8 appear to be resolved when age and smoking history are better matched and progressive small airway scores are not associated with a decline in lung function. However, the observation that small airway abnormalities identified by the current scoring system do not correlate with decreasing FEV 1 despite the absence of progressive emphysema suggests that airway narrowing and/or closure may be responsible. Hogg et al 12 emphasized the inability of current grading methods using light microscopy to measure small airway narrowing, partial closure, and/or obliteration where there is a decline in FEV 1 that could not be accounted for by progressive morphologic emphysema or physiologic loss of lung elastic recoil. Furthermore, Bosken et al 10 and Kazuyoshi et al 11 reported that small airways in patients with COPD are only very slightly thicker and their respective lumens only slightly narrower when compared with matched controls for age and smoking but without moderate airflow limitation. In contrast, the increased MB and RB airway scores noted in the present and previous surgical studies with less airflow limitation, when compared with the age and smoking, matched Vancouver study 12 with normal airflow (Table CHEST I 109 I 2 I FEBRUARY,

6 4), suggest marked structural abnormalities could partially contribute to greater airflow limitation whether emphysematous destruction was present. Macklem and Permutt 45 have suggested that lesions responsible for airflow limitation in COPD may be heterogeneously distributed along different parallel pathways that would be difficult to detect and score using present small airway grading techniques. While there was no correlation among MB, RB scores, and emphysema, similar to previous studies, this may be in part related to the narrow range of FEV 1 observed in the present study. In conclusion, we have demonstrated that CT lung scans are able to closely corroborate the extent and severity of whole lung morphologic emphysema in patients with severe airflow limitation. The poor correlation between expiratory airflow limitation and CT as well as morphologic scored emphysema suggests that emphysema is not the major cause of airflow limitation in COPD. Despite marked abnormalities in small airways, there was no correlation with extent and severity of emphysema or with airflow limitation. The cause of the lesions responsible for small airway obstruction needs to be identified. REFERENCES 1 Snider GL. Emphysema: the first two centuries and beyond. Am Rev Respir Dis 1992; 146: , Wright JL. Small aiiways disease: its role in chronic airflow obstruction. Semin Respir Med 1992; 13: Hogg JC, Macklem IT, Thurlbeck WM. Site and nature of airwayobstruction in chronic obstructive lung disease. N Eng! J Med 1968; 278: Macklem IT, Thurlbeck WM, Fraser RG. Chronic obstructive disease of small aiiways. Ann Intern Med 1971; 74: Berend N, Woolcock AJ, Marlin GE. Correlation between the function and structure of the lung in smokers. Am Rev Respir Dis 1979; 119: Berend N, Wright JL, Thurlbeck WM, et al. Small aiiways disease: reproducibility of measurements and correlation with lung function. Chest 1981; 79: Petty TL, Silvers GW, Stanford RE, et al. Small aiiway pathology is related to increased closing capacity and abnormal slope of phase III in excised human lungs. Am Rev Respir Dis 1980; 121: Cosio M, Ghezw H, Hogg JC, et al. The relations between structural changes in small aijways and pulmonary function tests. N Eng! J Med 1978; 298: Wright JL, Lawson LM, Pare PD, et al. The detection of small aiiways disease. Am Rev Respir Dis 1984; 129: Bosken CH, Wiggs BR, Pare PO, et al. Small ai!ways dimension in smokers with obstruction to airflow. Am Rev Respir Dis 1990; 142: Kazuyoshi K, Bosken CH, Pare PD, et al. Small aiiways dimensions in asthma and in chronic obstructive pulmonary disease. Am Rev Respir Dis 1993; 148: Hogg JC, Wright JL, Wiggs BL, et al. Lung structure and function in cigarette smokers. Thorax 1994; 49: Mitchell RS, Stanford RE, Johnson JM, et al. The morphologic features of the bronchi, bronchioles, and alveoli in chrome aiiway obstruction: a clinicopathologic study. Am Rev Respir Dis 1976; 114: Hale KA, Ewing SL, Gosnell BA, et al. Lung disease in long term cigarette smokers with and without chronic air-flow obstruction. Am Rev Respir Dis 1984; 130: Nagai A, West WW, Thurlbeck WM. The National Institute of Health Intermittent Positive-Pressure Breathing Trial: pathology studies: II. Correlation between morphologic findings, clinical findings and evidence of expiratory airflow obstruction. Am Rev Respir Dis 1985; 132: Srtider GL. Chronic obstructive pulmonary disease: a definition and implications of structural determinants of airflow obstruction for epidemiology. Am Rev Respir Dis 1989; 140:S Jamal K, Cooney TP, Fleetham JA, et al. Chronic bronchitis: correlation of morphologic findings to sputum production and flow rates. Am Rev Respir Dis 1984; 129: Kuwano K, Matsuba K, Ikeda T, et al. The diagnosis of mild emphysema: correlation of computed tomography and pathology scores. Am Rev Respir Dis 1990; 141: Gould GA, MacNee W, McLean A, et al. CT measurements of lung density in life can quantitate distal airspace enlargement -an essential defining characteristic of human emphysema. Am Rev Respir Dis 1988; 137: Morrison NJ, Abboud RT, Ramadan F, et al. Comparison of single breath diffusing carbon monoxide diffusing capacity and pressure-volume curves in detecting emphysema. Am Rev Respir Dis 1989; 139: Hruban RH, Meziane MA, Zerhourti EA, et al. High resolution computed tomography of inflation-hxed lungs: pathologic-radiologic correlation of centrilobular emphysema. Am Rev Respir Dis 1987; 136: Gelb AF, Schein M, Kuei J, et al. Limited contribution of emphysema in advanced chronic obstructive pulmonary disease. Am Rev Respir Dis 1993; 147: American Thoracic Society. Definitions and classification of chronic bronchitis, asthma, and pulmonary emphysema. Am Rev Respir Dis 1962; 85: American Thoracic Society. Lung function testing: selection of reference values and interpretative strategies. Am Rev Respir Dis 1991; 144: American Thoracic Society. Single breath carbon monoxide diffusing capacity (transfer factor): recommendation for a standard technique. Am Rev Respir Dis 1987; 136: Goldman HI, Becklake MR. Respiratory function tests: normal values at median altitudes and the prediction of normal results. Am Rev Tuberc Pulm Dis 1959; 79: Morris JF, Koski A, Johnson LC. Spirometric standards for healthy nonsmoking adults. Am Rev Respir Dis 1971; 103: Gaensler EA, Smith AA. Attachment for automated single breath diffusin); capacity measurement. Chest 1 07.'3: fi:l: 1.' Wright JL, Wiggs B, Pare PD, et al..h.anking the severity of emphysema on whole lung slices: concordance of upper lobe, lower lobe and entire lung ranks. Am Rev Respir Dis 1986; 133: Thurlbeck WM, Dunnell MS, Hartung W, et al. A comparision of three methods of measuring emphysema. Hum Pathol 1970; 1: Wright JL, Cosio M, Wiggs B, et al. A morphologic grading system for membranes and respiratory bronchioles. Arch Pathol Lab Med 1985; 109: Morrison NJ, Abboud RT, Muller NL, et al. Pulmonary capillary blood volume in emphysema. Am Rev Respir Dis 1990; 141: Muller NL, Staples CA, Miller RR, et al. Density mask: an objective method to quantitate emphysema using computed tomography. Chest 1988; 94: Hayhurst MD, Flenley DC, McLean A, et al. Diagnosis of pulmonary emphysema by computerized tomography. Lancet 1984; 2: Foster WL Jr, Pratt PC, Roggli VL, et al. Centrilobular emphy- 358 Clinical Investigations

7 sema: CT-pathologic correlation. Radiology 1986; 159: Bergin C, Muller NL, Nichols DM, eta!. The diagnosis of emphysema: a computed tomographic-pathologic correlation. Am Rev Respir Dis 1986; 133: Miller RR, Muller NL, Veda! S, et a!. Limitations of computed tomography in the assessment of emphysema. Am Rev Respir Dis 1989; 139: Gelb AF, Gold WM, Wright RR, e t a!. Physiologic diagnosis of subclinical emphysema. Am Rev Respir Dis 1973; 107: Zamel N, Hogg J, Gelb AF. Mechanisms of maximal expiratory flow limitation in clinically unsuspected emphysema and obstruction of the peripheral airways. Am Rev Respir Dis 1976; 113: Forster RE. Exchange of gases between alveolar air and pulmonary capillary blood: pulmonary diffusing capacity. Physiol Rev 1957; 37: Piiper J, Sikand RS. Determination of DCO by the single breath method in inhomogeneous lungs: theory. Respir Physiol 1966; 1: Graham BL, Mink JT, Cotton OJ. Overestimation of the singlebreath carbon monoxide diffusing capacity in patients with airflow obstruction. Am Rev Respir Dis 1984; 129: Cotton OJ, Graham BL. Effect of ventilation and diffusion nonuniformity on Dl.co (exhaled) in a lung model. J Appl Physiol 1980; 48: Klein JS, Gamsu G, Webb WR, e t a!. High-resolution CT diagnosis of emphysema in symptomatic patients with normal chest radiographs and isolated low diffusing capacity. Radiology 1992; 182: Macklem PT, Permutt S. The lung in transition between health and disease. In: Lenfant C, ed. Lung biology in health and disease. New York: Marcel Dekker, 1979; CHEST / 109 / 2 / FEBRUARY,