Eur J Nucl Med (1983) 8:167-171 European M I It'~lr~,,-Jr Journal of I ~llI, J1k.#l~IE;l./

Medicine © Springer-Verlag 1983

Role of Radio-Aerosol and Perfusion Lung Imaging in Early Detection of Chronic Obstructive Lung Disease Abhimanyu Garg 1, P.G. Gopinath 2, J.N. Pande 1, and J.S. Guleria 1 1 Department of Medicine, 2 Department of Nuclear Medicine, All India Institute of Medical Sciences, New Delhi-110029, India

Abstract. The efficacy of radio-aerosol and perfusion lung imaging in the early detection of chronic obstructive lung disease was evaluated in 38 subjects. The subjects included 5 non-smokers, 21 smokers with minimal or no respiratory symptoms and 12 patients with chronic obstructive lung disease. Each subject consented to a respiratory question- naire, detailed physical examination, chest X-ray examina- tions, detailed puhnonary function tests and 99mTc-radio aerosol-inhalation lung imaging. Perfusion lung imaging with 99mTc-labelled macroaggregated albumin was per- formed in 22 subjects. A significant correlation (P < 0.001) was observed between the degree of abnormalities on radio- aerosol imaging and pulmonary function tests (PFTs) in- cluding forced expiratory volume in 1 s, maximum mid- expiratory flow rate and mean transit time analysis. Abnor- mal radio-aerosol patterns and deranged PFTs were ob- served in 21 subjects each. Of 21 subjects with abnormal radioaerosol pattern 8 had normal PFTs. Of 21 subjects with abnormal PFTs 8 had normal aerosol images. Aerosol lung images and PFTs were abnormal more frequently than perfusion lung images. The results suggest that radio- aerosol lung imaging is as sensitive an indicator as PFTs for early detection of chronic obstructive lung disease and can be usefully combined with PFTs for early detection of alteration in pulmonary physiology in smokers.

Introduction

For the prevention of the ever-increasing morbidity and mortality caused by chronic obstructive lung disease (COLD), early detection and prompt medical management are essential. Numerous indices of airways function such as forced expiratory volume in 1 s (FEV1), the measurement of frequency dependence of the dynamic compliance (Wool- cock et al. 1969), maximum expiratory flow rates at low lung volumes (Pride 1971) or the mean transit time (MTT) analysis of the forced expiratory spirogram (Tockman et al. 1976; Permutt and Menkes 1979; Jordanoglou et al. 1979) have been suggested for the early detection of peripheral airway disease. Radio-aerosol lung imaging has also been found to be a useful and sensitive index of airway obstruc- tion (Taplin et al. 1975; Ramanna et al. 1975; Taplin et al.

Offprints requests to." Dr. P.G. Gopinath, Department of Nuclear Medicine, All India Institute of Medical Sciences, Ansari Nagar, New Delhi-110029, India

1977). All these methods for the demonstration of peripher- al airway disease exploit the existence of varying time con- stants in different regions of the lung. Of the various param- eters mentioned, the measurement of F EV~, maximum mid- expiratory flow rate (MMFR) and MTT can be obtained from a record of the forced expiratory spirogram. The pres- ent study was undertaken to assess the sensitivity of the radio-aerosol inhalation and perfusion scans compared with the above mentioned spirometric indices in the early detection of peripheral airway disease.

Materials and Methods

The study involved 38 subjects including medical students and patients attending the chest clinic at the All India Insti- tute of Medical Sciences, New Delhi. All were males ranging in age from 18 to 75 years (mean 32 years). The subjects included 5 non-smokers, 21 smokers without any respirato- ry symptoms or minimal respiratory symptoms and 12 smokers with a history of chronic bronchitis as defined by World Health Organization. (WHO 1961).

Informed consent was obtained. Each subject answered a detailed respiratory questionnaire and findings on physi- cal examination were recorded. Radiological examination of the chest was performed along with electrocardiogram and routine haematological investigations. Sputum exami- nation for culture and acid-fast bacilli was carried out to exclude pulmonary tuberculosis, bronchiectasis and respira- tory infections.

Functional Assessment

This included the measurement of vital capacity (VC), forced expiratory volume in 1 s (FEV~) expressed as percent of forced vital capacity (FVC), maximum mid-expiratory flow rate (MMFR), mean transit time (MTT) and resting steady state pulmonary diffusion capacity of carbon mon- oxide (DLco).

At least five forced expiratory manoeuvres were per- formed on each subject in a standing position. Excursions were recorded on a fast-moving kymograph drum using a low resistance spirogram. Calculations of FEVj, M M F R and MTT were made from the best breath which was identi- fied as the one yielding the highest FEV1. All volumes were corrected to body temperature and pressure.

MTT was obtained by integrating the area under the spirogram curve of volume versus time and dividing by

168

FVC. The area of the FVC time curve was measured with a planimeter and the mean of five successive measurements was taken for study purposes. Pulmonary diffusing capacity of carbon monoxide was measured by modified steady state technique of Bates et al. (1955) utilizing a Rahn and Otis end tidal sampling device for obtaining the alveolar air.

Lung Imaging Procedures

Radio-aerosol lung imaging was performed in all the sub- jects, whereas perfusion lung scans were obtained in 22. 99mTc (technetium) in the pertechnetate form was utilized for radio-aerosol imaging. High specific activity solution containing 10 15 mCi/ml was used for conversion to fine aerosol.

A modified form of radio-aerosol administration system described by Hayes et al. (1979) was employed. A glass nebulizer was utilized for aerosol generation. Before inhal- ing the radioactive aerosol, a trial with a normal saline aerosol during tidal volume breathing was carried out. After radio-aerosol inhalation each subject was asked to rinse his mouth to reduce the amount of gastric deposition. In each subject anterior and posterior views of the lungs were obtained in the recumbent position. For each view 3 x 105 counts were accumulated. Total radioactivity retained in the lung was in the range of 2.0-3.0 mCi. Perfusion lung imaging was performed by injecting 2.0 mCi 99mTc-labelled macroaggregated albumin IV with the subject lying in a supine position and breathing evenly. Anterior and posteri- or views were taken accumulating 3 x 10 s counts for each view. Perfusion imaging was carried out 2 days after the radio-aerosol imaging procedure.

Method of Interpretation

Subjects were considered to be symptomatic if they had cough, expectoration, wheezing or breathlessness. Smoking history was considered significant if the subject had been smoking on an average more than five cigarettes or bidis per day for at least 1 year. Abnormal findings on chest examination included rhonchi, crepitations, hyperinflation, diminished breath sounds or signs of cor pulmonale. Roent- genograms of the chest were evaluated by a radiologist without prior knowledge of other findings. The presence of hyperinfiation, alteration in pulmonary vasculature and bullae in the roentgenograms were interpreted as abnormal.

The results of VC, M M F R and DLco were expressed as a percentage of the predicted values given by Goldman and Becklake (1959) and Bates et al. (1962). A FEV1 to FVC ratio of less than 75% was considered abnormal (mild 65%-75%; moderate 50%-65%; and severe less than 50%). M M F R was interpreted as abnormal when the values were less than 65% of the predicted values (mild 45%-65% ; moderate 25%-45% ; and severe less than 25%). MTT was considered abnormal if it was more than two standard de- viations above the predicted values derived from the regres- sion equation obtained for 81 healthy non-smokers in the laboratory (Shah et al. 1982; unpublished work).

The results of functional evaluation were classified as mildly abnormal ( + ) if either FEV1 or M M F R was mildly abnormal or if FEV1 and M M F R were normal but MTT was adjudged to be abnormal; as moderately abnormal

(+ + ) if either FEV 1 or M M F R was moderately abnormal; and severely abnormal (+ + + ) if either FEV 1 or M M F R were severely abnormal according to the criteria mentioned earlier.

The lung images were evaluated independently by two observers without prior information about other parame- ters. The extent of concordance between the two observers for defining the images as normal or abnormal was 89%. In case of disagreeement the final interpretation was based on re-evaluation and mutual consent. Lung image patterns were considered normal or abnormal according to the cri- teria used by Taplin et al. (1977).

Aerosol patterns were considered abnormal if the distri- bution was uneven, with discrete areas of increased and irregular regions of decreased radioactivity in the pulmo- nary parenchyma, or if there was excessive deposition in the regions of major airways. The aerosol pattern was clas- sifted as mildly abnormal ( + ) when there was slightly uneven distribution involving less than 25% of the pulmo- nary parenchyma. The images were considered moderately abnormal (+ + ) if there was either definite excessive central deposition or distinctly uneven deposition in the smaller airways involving 25%-50% of the lung. Aerosol images were considered severely abnormal (+ + + ) when there were gross abnormalities of deposition in either the major or minor airways together with absent or reduced deposi- tion in more than half of the pulmonary parenchyma (Taplin et al. 1977).

Perfusion images were graded as mildly (+) , moderately (+ +), or severely (+ + + ) abnormal according to the visually estimated percentage of lung deprived of arterial blood flow, namely, less than 25% as mild, 25%-50% as moderate and more than 50% as severe impairment of the pulmonary blood flow (Taplin et al. 1977).

Results

Figure I (A-D) shows scans of a healthy non-smoker with a normal homogenous pattern of distribution of radio- aerosol and 99mTc-labelled macroaggregated albumin. Pul- monary function tests of the individual were classified as normal.

Figure 2(A, B) depicts a mildly abnormal ( + ) aerosol scan showing excessive central deposition and slightly uneven distribution. The perfusion scan (Fig. 2 C, D) shows a mildly abnormal pattern. Pulmonary function tests of the subject were graded as mildly abnormal.

Figure 3(A, B) shows a severely abnormal (+ + + ) aerosol scan with a central and spotty distribution pattern. Perfusion scan (Fig. 3C-E) shows moderate abnormality in distribution of 99~Tc-lahelled macroaggregated albumin. Functional evaluation suggested moderately abnormal pul- monary function tests.

A correlation matrix (Table 1) was calculated for the variables VC, FEVI, MMFR, DLco, MTT and the degree of abnormality on aerosol lung scanning. A highly signifi- cant correlation (P<0.001) was observed between the degree of abnormality on aerosol lung imaging and FEV1, M M F R and MTT. A significant correlation ( P < 0.01) was observed between degree of abnormality on aerosol scan and VC whereas no significant correlation was observed between degree of abnormality on aerosol scan and DLco. There was a highly significant correlation (P<0.001; r =

169

Fig. 1 A-D. Scintiscans of the lungs of a normal subject. A Aerosol scan, anterior view. B Aerosol scan, posterior view. C Perfusion scan, anterior view. D Perfusion scan, posterior view. Pulmonary function tests (normal) VC, 72.32% of predicted value; FEV1/FVC, 87.40% ; MMFR 71.15% of predicted value; DLco 62.37% of predicted value; MTT, 0.565 s

Fig. 2A-D. Scintiscans of the lungs showing mildly abnormal aerosol and perfusion scan. A Aerosol scan, anterior view. B Aerosol scan, posterior view. C Perfusion scan, anterior view. D Perfusion scan, posterior view. Pulmonary function tests (mildly abnormal) VC, 92.27% of predicted value; FEV1/FVC, 68.0%; MMFR, 49.78% of predicted value; DLco, 67.33% of predicted value; MTT, 0.947 s

0.688) between the abnormality in the aerosol scan and the overall grade of pulmonary dysfunction.

The perfusion scans were carried out in 22 subjects. Eight exhibited significant abnormalities on perfusion scans whereas aerosol lung scans and lung functions were abnor- mal in 10 and 13 respectively. There was a significant corre- lation between the degree of abnormality on functional evaluation and that on perfusion lung imaging ( r=0.561; p < 0.01). A highly significant correlation was observed be- tween the degree of abnormality on aerosol and perfusion lung imaging (r = 0.924; P < 0.001).

Inter-relationship amongst various abnormal findings in 38 subjects was studied (Table 2). There were 33 smok- ers; 22 had respiratory symptoms; 15 subjects had abnor- mal physical findings; 21 had abnormal PFTs (abnormal FEV1 was detected in 14 subjects, abnormal M M F R in 18 and abnormal MTT in 16 subjects) and 21 subjects had ab- normal aerosol lung imaging pattern. Of the 33 smokers, 45% had abnormalities in the physical examination, 58% had abnormalities in the results o f one or more of the pul- monary function tests (FEV1, M M F R , MTT), and 58% had abnormalities onaerosol lung imaging. Of the 22 sub-

170

Fig. 3A-E. Scintiscans of the lungs showing severely abnormal aerosol scan and moderately abnormal perfusion scan. A Aerosol scan, anterior view. B Aerosol scan, posterior view. C Perfusion scan, anterior view. D Perfusion scan, posterior view, left lung. E Perfusion scan, posterior view. right lung. Pulmonary function tests (moderately abnormal) VC 75.15% of predicted value; FEV1/FVC, 56.0% ; M M F R 64.40% of predicted value; DLco, 75.77% of predicted value; MTT, 1.635 s

Table 1. Correlation matrix between pulmonary function tests and degree of abnormality of aerosol scans

Variable VC FEV1/FVC M M F R DLco MTT Aerosol lung imaging

VC 1.000 0.207 0.324" 0.364" -0 .334 a -0 .456 b FEV1/FVC 1.000 0.883 c 0.379" -0 .863 c -0.741 c M M F R 1.000 0.497b -0 .783 c -0 .609 c DLco 1.000 - 0.373" - 0.240 MTT 1.000 0.674 c Aerosol lung imaging 1.000

a Significant P<0.05 b Significant P<0.01 c Significant P<0.001

Table 2. Inter-relationship among various abnormal findings in 38 subjects

variables Number of" Percent u subjects a

Smoking Symptoms Physical Pulmonary Aerosol lung findings function tests imaging

Smoking 33 ... 67 45 58 58 Symptoms 22 100 .. 67 68 77 Physical findings 15 100 100 .. 80 87 Spirometric data 21 90 71 57 .. 62 Aerosol lung imaging 21 90 81 62 62 ..

a Total number of subjects with abnormal results for each variable listed at left side of Table b Coprevalence of pairs of abnormalities (expressed as percentage), given that subjects have abnormal results for

variables listed at left side of the Table

jects wi th resp i ra tory s y m p t o m s 77% had a b n o r m a l aerosol scans whereas 68% had abnormal i t i e s o f PFTs . O f the 15 subjects wi th physical f indings suggest ive o f resp i ra tory disease 80% had abnormal i t i e s on PFTs , whereas 87% had a b n o r m a l aerosol lung Scans.

Discussion

A highly s ignif icant cor re la t ion be tween the degree o f ab- normal i t ies on aerosol lung imaging, F E V 1 and M M F R and M T T suggests tha t the m e t h o d o f in t e rp re ta t ion o f

171

aerosol scans employed in the present study provides a fair assessment of the functional abnormalities of peripheral airways.

A good correlation between the degree of abnormality on pulmonary function tests and aerosol lung imaging seems to substantiate the usefulness of aerosol lung imaging in the assessment of overall lung function. In a study of 100 subjects, Taplin et al. (1977) have documented a good cor- relation between the above mentioned parameters (r = 0.59). Our observations are similar to that of Taplin et al. (1977).

A significant correlation between the results of perfusion lung imaging and pulmonary function tests and a high cor- relation between perfusion and aerosol lung imaging may be attributed to the regional disturbances in perfusion pro- duced in areas of diminished ventilation. Perfusion lung imaging does not compare with pulmonary function tests or aerosol lung imaging as far as early detection of peripher- al airways disease is concerned (Ramanna etal. 1975; Taplin et al. 1977). The present study also supports this contention.

The frequency of abnormalities in aerosol lung images and pulmonary function tests was equal; each being abnor- mal in 21 subjects. Aerosol lung imaging was more sensitive in detecting early airway disease than FEV 1, MMFR and MTT when considered separately. Of the 21 subjects with abnormalities detected on aerosol images, 8 had normal PFTs. On the other hand, 8 of 21 subjects with abnormal PFTs had normal aerosol images. The observed difference in the results of PFTs and aerosol lung scans may have several explanations. The two procedures may be looking into slightly different aspects of derangements of pulmo- nary physiology. Aerosol scan is influenced by parameters other than regional ventilation, e.g. velocity of air flow, turbulence, particle size and impaction. Many of these pa- rameters may be deranged enough to produce abnormalities in the aerosol scan without affecting the lung mechanics as evaluated by tests like FEV1, MMFR and MTT.

Another plausible explanation is the production of false positive images on aerosol scan because of excessive central deposition of aerosol particles greater than 5.0 pm in the larger airways. The introduction of a large reservoir bag between the nebulizer and the mouthpiece in the method of aerosol generation used in the present study helps in the settling of larger particles of aerosol in the bag. Thus the particles delivered to the patient were less than 2 p (Hayes et al. 1979). A practice session of breathing an aerosol of normal saline solution at tidal volume at a normal respiratory rate of 12-16 times/min minimized the central aerosol deposition by abnormally rapid or shallow breathing patterns. A study of delayed aerosol images by Ramanna et al. (1975) and Taplin et al. (1977) revealed that only 10% 16% of the subjects having excessive central de- position initially showed normal images 1-2 h later. Since the central deposition pattern was observed in only 3 of 38 subjects, the possibility of false positive images remains low in the light of earlier observations. The specificity of isolated abnormalities detected by lung imaging in the pres- ence of normal pulmonary function tests can be confirmed only by further longitudinal studies.

The observation of abnormal pulmonary function tests with normal aerosol images may be due to the fact that PFTs are truly quantitative indices of overall functions whereas the radio-nuclide study is only a semi-quantitative measure of regional pulmonary function.

Taplin et al. (1977) found aerosol lung imaging to be more sensitive than the detailed evaluation of PFT by spiro- merry, flow volume loop, closing volume and PaO 2. In our study aerosol lung imaging was found to be as sensitive a procedure for early detection of airway disease as the PFTs including MTT analysis. Aerosol lung imaging in con- junction with pulmonary function tests certainly enhance the overall sensitivity for the early detection of alterations in pulmonary physiology due to smoking.

It is concluded that aerosol lung scanning is a sensitive indicator for early detection of obstructive lung disease and can be gainfully combined with various pulmonary function tests. It also helps in localization of regional abnormalities in the airways of patients with respiratory disease.

Acknowledgements. We thank the medical students and the patients who consented to participate in the study, Dr. Rakesh K. Rustagi (Senior Research Officer, Institute for Research in Medical Statis- tics) for help in statistical analysis and Dr. AK Padhy (Senior Resident, Department of Nuclear Medicine, AIIMS) for invaluable assistance during the project.

References

Bates DV, Boucot NG, Doemer AE (1955) The pulmonary diffus- ing capacity in normal subjects. J Physiol (London) 129:237

Bates DV, Woolf CR, Paul GI (1962) A report on the first two stages of the coordinated study of chronic bronchitis. In the Department of Veterans Affairs, Canada. Med Serv J Canada 18 : 211-303

Goldman HI, Becklake MR (1959) Respiratory function tests, normal values at median altitudes and prediction of normal results. Am Rev Tuberc 79:457

Hayes M, Taplin GV, Chopra SK, Knox DE, Elam D (1979) Im- proved radioaerosol administration system for routine inhala- tion lung imaging. Radiology 131:256-258

Jordanoglou J, Koursouba E, Lalenis C, Gotsis T, Kontos J, Gar- dikas C (1979) Effective time of the forced expiratory spirogram in health and airways obstruction. Thorax 34:187-193

Permutt S, Menkes HA (1979) Spirometry: analysis of forced expi- ration within the time domain. In: Macklem P, Permutt S (ed) The lung in the transition between health and disease. Chapter 6 (Lung biology in health and disease series, Vol 12). Marcel Dekker, New York

Pride NB (1971) The assessment of airflow obstruction. Br J Dis Chest 65:135 169

Ramanna L, Tashkin DP, Taplin GV, Elam D, Detels R, Coulson A, Rokaw SN (1975) Radioaerosol lung imaging in chronic obstructive pulmonary disease - Comparison with pulmonary function tests and roentgenography. Chest 68 : 634-640

Taplin GV, Tashkin DP, Chopra SK, Anselmi OE, Elam D, Cal- varese B, Coulson A, Detels R, Rokaw SN (1977) Early detec- tion of chronic obstructive pulmonary disease using radionu- clide lung-imaging procedures. Chest 71 : 567-575

Taplin GV, Tashkin DP, Ramanna L, Detels R, Anselemi OE (1975) Radioaerosol lung scintigraphy - An efficient indicator of obstructive airways disease (ABS) JNM 16:574

Tockman M, Menkes H, Cohen B, Permutt S, Benjamin J, Bail WC Jr, Tonascia J (1976) A comparison of pulmonary function in male smokers and non-smokers. Am Rev Resp Dis 114:711 722

WHO Technical Report Series (1961) Definition and diagnosis of pulmonary diseases with special reference to chronic bronchitis and emphysema, in 'Chronic cor-pulmonale', Report of an Expert Committee. 213:14-19

Woolcock AJ, Vincent NJ, Macklem PT (1969) Frequency depen- dence of compliance as a test for obstruction in the small airways. J Clin Invest 48:1097-1106

Received October 30, 1982