- Email: [email protected]
Course and Prognosis of Patients with Chronic Airways Obstruction
Course and Prognosis of Patients with Chronic Airways Obstrudion Benjamin Bu"ows, M.D., F.C.C.P. •
I
was pleased to be asked by Dr. Petty to give this talk since it provides an opportunity to make a few comments on the general problem of obstructive diseases of the airways. As most of you know, in 1958 we began a study at the University of Chicago of a clinical entity that was then called "emphysema" by most physicians in the United States. At that time, we believed that patients with obstructive airways diseases could be divided into three broad categories: asthma, asthmatic bronchitis, and emphysema. It was our intention to restrict our study to patients in the emphysema category. Only after undertaking a comparison of our cases with cases in Britain of severely obstructed "chronic bronchitis" did we begin to use the more noncommittal term of "chronic obstructive lung disease" to delineate our patients. In fact, this term was introduced largely to resolve a semantic disagreement between British and American doctors. The important point is that the vast majority of the cases we studied did have frank emphysema and that patients with fluctuating degrees of ventilatory impairment whose forced expiratory volume (FEV) could eoet' exceed 60 percent of their predicted vital capacity (VC) were systematically excluded from our investigation. Indeed, some 26 subjects originally thought to be eligible showed too much improvement with therapy and, to my current dismay, were never followed-up. The course of the 200 remaining subjects who did meet our criteria has been the subject of numerous publications, and I will not review these in detail. Our most recent analyses 1 (using Richard Peto's log rank method of life table analysis) indicate that age does become a significant determinant of survival after about four years of follow-up. After accounting for age, the postbronchodilator percent predicted FEV1 is the best predictor of survival. Once this is taken into account, improvement with use of a bronchodilator per se does not relate to subsequent survival. In the yo!Jilger age group, cor pulmonale has an adverse effect on survival even after accounting for spirometric data. We have not emphasized enough the wide variability in mortality even when all initial characteristics are taken into consideration. It is this variability in course that I would like to emphasize today. The variability both between and within individuals is very obvious when one examines the course of spirometric tests. When we &rst looked at changes in FEV1 early in our study, there was very marked variability in results. When only two to three years of data points were 0
Professor of Internal Medicine and Director, Division of Respiratory Sciences, Arizona Health Sciences Center, Tucson. Supported by NHLBI SCOR Grant no. HL 14136.
Reprint requests: Dr. ButTOW8, Diolaiora of lfesplratorr/ Sci-
ences, Unroerrity of Ari2:ona CoUege of Medidne, Tucson
85724
250 22nd ASPEN LUNG CONFERENCE
available, there were decreases in FEV1 of as much as 400 ml per year and increases of as much as 300 ml per year. The range of FEV1 change appeared narrower when a few more years of follow-up were available, but there was still considerable variability. 2 Only when a data point was available after more than ten years of followup (and we have only 22 such cases) was a reasonably narrow range of declines observed. 8 Still one subject showed virtually no decline in FEV1 , while one declined over 100 ml per year. The problem in determining rate of decline is evident if one examines the courses of some individual subjects. There is so much variability in FEV1 over the first four to 6ve years that meaningful individual regression slopes cannot be calculated. Despite all of this, mean rates of decline are nearly identical regardless of the duration of follow-up. The problems may be even greater in examining the course of the FEV1 in subjects who are not clinically ill. Much of the information that we have in this regard is derived from the classic study by Fletcher and colleagues. • He emphasizes a relationship between the rate of decline and the absolute level of FEV1 and regards this relationship as proof that there are some subjects who are unusually susceptible to the effects of smoking. Such subjects are presumed to have a lifelong tendency to have a rapid decline in lung function which will gradually lead to severe airways obstruction. The important implication of this concept is that a slightly low FEV1 in middle life would predict the later development of severe clinical disease. But how well do these conclusions, based on mean data, reflect what is actually happening in the population? Examination of the raw data from Fletcher'r study shows enormous variability and leaves me with little assurance that one can predict much about subsequent course from a single FEV1 value. In interpreting Fletcher's data, it is important to remember that after nine years of follow-up and multiple FEV1 measurements, extensive statistical manipulations were still needed in order to demonstrate a low level correlation between rate of decline and level of FEV1 • In regard to the development of chronic airways obstruction, it may be worth mentioning a few anecdotes, such as the case of David Bates who maintained normal spirometric tests for many years, then showed some fluctuating values before quite abruptly. becoming persistently disabled by a chronic airways obstructive disorder. There is also a patient in the Chicago series who, unbeknownst to us, failed to take any prescribed medication during his 6rst year of follow-up. When simple bronchodilators were 6nally used, he showed marked improvement, enough improvement to have disquali6ed him from the study had it occurred at an earlier date. His subsequent rate of decline in lung function was then very rapid and by the end of his follow-up he would
CHEST, 77: 2, FEBRUARY, 1980 SUPPLEMENT
once again have been eligible for our original so-called "emphysema study." Finally, there seems to be a tendency to ignore an entity which in my current experience is even more common than what I would call "chronic obstructive lung disease." It is associated with some degree of chronic cough and wheeze but manifests extremely variable lung function. Even after maximum therapy, a mild degree of airways obstruction may persist, but the major problem is episodic bronchoconstriction. The long-range prognosis for this type of patient is totally unknown. It is evident, however, that prognosis would not be predictable from a single lung function test in such an individual. As exemplified by one of my current patients, the syndrome (let's call it "asthmatic bronchitis") can progress to severe irreversible disease which, at its end stage, cannot be distinguished from the kind of case we enrolled in our "emphysema study" in Chicago 20 years ago. There are really three major points of this talk. First, it is unreasonable to assume that every mild ventilatory abnormality discovered on a routine survey represents an early stage of "chronic obstructive lung disease." Secondly, one should be very cautious in interpreting individual rates of decline in lung function, especially over periods less than ten years. Finally, I am not convinced that we yet know the pre-clinical course of most patients who develop severe disabling chronic airways obstruction. While some may show a steady decline in lung function throughout the whole of their adult life, others may retain relatively normal function and then decline rapidly shortly before the onset of clinical symptoms. Still others may go through a stage of episodic bronchoconstriction which can last for many years before irreversible disease occurs.
1 Traver GA, Cline MG, Burrows B: Predictors of mortality in COPD: A 15-year follow-up study. Am Rev Respir Dis 119:895-902,1979 2 Burrows B, Earle RH: Course and prognosis of chronic obstructive lung disease, a prospective study of 200 patients. N Engl J Med 280:397-404, 1969 3 Burrows B: Course and prognosis in advanced disease (Chapter 2). In Chronic Obstructive Pulmonary Disease (T. Petty, ed) New York, Marcel Dekker, 1978, pp 23-33 4 Fletcher CM, Peto R, Tinker CM, et al: The Natural History of Chronic Bronchitis and Emphysema. Oxford; Oxford University Press, 1976, p 272
The Influence of Bronchitis and Asthma in Infancy and Childhood on Lung Function in Schoolchildren* Ann Woolcock, M.D., F.C.C.P.; S. Leetkr, Ph.D.; ]. Peat, B.Sc.; and C. Blackburn, M.D.
•From the Department of Medicine, University of Sydney, Australia.
CHEST, 77: 2, FEBRUARY, 1980 SUPPLEMENT
Jnasthma order to document the effect of bronchitis and in infancy and early childhood on subsequent lung function, a prospective study of the lung function of a random sample of Sydney schoolchildren was begun in 1971. At the first study, a parental questionnaire was used to obtain a history of previous asthma and bronchitis, severe enough to require treatment by a medical practitioner. Maximal expiratory flow volume (MEFV) curves were recorded with a Fleisch no. 3 pneumotachograph onto magnetic tape. After digital conversion, the forced expiratory volume in 0.5 and 1.0 sec (FEVu and FEV1 }, the forced vital capacity ( FVC) and the maximal flow at 50 percent of the FVC (V50 ) were calculated. The lung function studies were repeated each year between 1972 and 1974, at which time a history of respiratory illness in the previous 12 months and a personal smoking history were recorded. In 1971, 4,639 children from primary schools (mean age 8.9 years) and 6,259 children from secondary schools (mean age 12.6 years} were studied. At this time the children were divided into five groups based on history: 1 ) bronchitis with first attack before age two; 2} bronchitis, first attack after age two; 3) asthma; 4) asthma and bronchitis; 5} neither asthma nor bronchitis. Mean values for lung function were calculated for each group excluding children with cough in the last six weeks or with current wheezing. Lower values for V50 (but not for other measurements} were found in groups 1, 2, 3 and 4 compared with group 5 for both age groups and sexes. The differences were highly significant for all groups and were still present in 1974. In 1974, the data from smokers were compared with those from nonsmokers. The secondary school boys in group 2 (mean age 15.6 years) who smoked had lower values for V50 than nonsmokers (P < 0.01}. Most other groups had too few smokers for meaningful analysis. The studies suggest that bronchitis as well as asthma in infancy and in childhood may affect lung function as children grow, that the abnormality may not be detected by the forced expiratory volume, that the abnormality persists, and that it is possible that it is further affected by smoking. Q: In young children, how do you separate bronchitis from asthma? A: We rely on diagnosis by the private physician or hospital records. Sometimes the conditions are difticult to differentiate.
The Role of Air Pollution, Smoking and Respiratory Illnesses in Childhood in the Development of Chronic Bronchitis* D. Irvine, M.Sc.; A. Brooks; and R. Waller, B.Sc. 0
Medical Research Council, St. Bartholomew's Hospital Medical College, London, EngJand.
Reprint reqtJeSta: Dr. Irvine, Mellical RssetJtCh Cooncil, Toz. icology Unit, Clinical Section, St. Bartholomew's Hosplltll MedicQl College, London EClM BBQ, England
CHRONIC OBSTRUCTIVE LUNG DISEASE 251
I
was pleased to be asked by Dr. Petty to give this talk since it provides an opportunity to make a few comments on the general problem of obstructive diseases of the airways. As most of you know, in 1958 we began a study at the University of Chicago of a clinical entity that was then called "emphysema" by most physicians in the United States. At that time, we believed that patients with obstructive airways diseases could be divided into three broad categories: asthma, asthmatic bronchitis, and emphysema. It was our intention to restrict our study to patients in the emphysema category. Only after undertaking a comparison of our cases with cases in Britain of severely obstructed "chronic bronchitis" did we begin to use the more noncommittal term of "chronic obstructive lung disease" to delineate our patients. In fact, this term was introduced largely to resolve a semantic disagreement between British and American doctors. The important point is that the vast majority of the cases we studied did have frank emphysema and that patients with fluctuating degrees of ventilatory impairment whose forced expiratory volume (FEV) could eoet' exceed 60 percent of their predicted vital capacity (VC) were systematically excluded from our investigation. Indeed, some 26 subjects originally thought to be eligible showed too much improvement with therapy and, to my current dismay, were never followed-up. The course of the 200 remaining subjects who did meet our criteria has been the subject of numerous publications, and I will not review these in detail. Our most recent analyses 1 (using Richard Peto's log rank method of life table analysis) indicate that age does become a significant determinant of survival after about four years of follow-up. After accounting for age, the postbronchodilator percent predicted FEV1 is the best predictor of survival. Once this is taken into account, improvement with use of a bronchodilator per se does not relate to subsequent survival. In the yo!Jilger age group, cor pulmonale has an adverse effect on survival even after accounting for spirometric data. We have not emphasized enough the wide variability in mortality even when all initial characteristics are taken into consideration. It is this variability in course that I would like to emphasize today. The variability both between and within individuals is very obvious when one examines the course of spirometric tests. When we &rst looked at changes in FEV1 early in our study, there was very marked variability in results. When only two to three years of data points were 0
Professor of Internal Medicine and Director, Division of Respiratory Sciences, Arizona Health Sciences Center, Tucson. Supported by NHLBI SCOR Grant no. HL 14136.
Reprint requests: Dr. ButTOW8, Diolaiora of lfesplratorr/ Sci-
ences, Unroerrity of Ari2:ona CoUege of Medidne, Tucson
85724
250 22nd ASPEN LUNG CONFERENCE
available, there were decreases in FEV1 of as much as 400 ml per year and increases of as much as 300 ml per year. The range of FEV1 change appeared narrower when a few more years of follow-up were available, but there was still considerable variability. 2 Only when a data point was available after more than ten years of followup (and we have only 22 such cases) was a reasonably narrow range of declines observed. 8 Still one subject showed virtually no decline in FEV1 , while one declined over 100 ml per year. The problem in determining rate of decline is evident if one examines the courses of some individual subjects. There is so much variability in FEV1 over the first four to 6ve years that meaningful individual regression slopes cannot be calculated. Despite all of this, mean rates of decline are nearly identical regardless of the duration of follow-up. The problems may be even greater in examining the course of the FEV1 in subjects who are not clinically ill. Much of the information that we have in this regard is derived from the classic study by Fletcher and colleagues. • He emphasizes a relationship between the rate of decline and the absolute level of FEV1 and regards this relationship as proof that there are some subjects who are unusually susceptible to the effects of smoking. Such subjects are presumed to have a lifelong tendency to have a rapid decline in lung function which will gradually lead to severe airways obstruction. The important implication of this concept is that a slightly low FEV1 in middle life would predict the later development of severe clinical disease. But how well do these conclusions, based on mean data, reflect what is actually happening in the population? Examination of the raw data from Fletcher'r study shows enormous variability and leaves me with little assurance that one can predict much about subsequent course from a single FEV1 value. In interpreting Fletcher's data, it is important to remember that after nine years of follow-up and multiple FEV1 measurements, extensive statistical manipulations were still needed in order to demonstrate a low level correlation between rate of decline and level of FEV1 • In regard to the development of chronic airways obstruction, it may be worth mentioning a few anecdotes, such as the case of David Bates who maintained normal spirometric tests for many years, then showed some fluctuating values before quite abruptly. becoming persistently disabled by a chronic airways obstructive disorder. There is also a patient in the Chicago series who, unbeknownst to us, failed to take any prescribed medication during his 6rst year of follow-up. When simple bronchodilators were 6nally used, he showed marked improvement, enough improvement to have disquali6ed him from the study had it occurred at an earlier date. His subsequent rate of decline in lung function was then very rapid and by the end of his follow-up he would
CHEST, 77: 2, FEBRUARY, 1980 SUPPLEMENT
once again have been eligible for our original so-called "emphysema study." Finally, there seems to be a tendency to ignore an entity which in my current experience is even more common than what I would call "chronic obstructive lung disease." It is associated with some degree of chronic cough and wheeze but manifests extremely variable lung function. Even after maximum therapy, a mild degree of airways obstruction may persist, but the major problem is episodic bronchoconstriction. The long-range prognosis for this type of patient is totally unknown. It is evident, however, that prognosis would not be predictable from a single lung function test in such an individual. As exemplified by one of my current patients, the syndrome (let's call it "asthmatic bronchitis") can progress to severe irreversible disease which, at its end stage, cannot be distinguished from the kind of case we enrolled in our "emphysema study" in Chicago 20 years ago. There are really three major points of this talk. First, it is unreasonable to assume that every mild ventilatory abnormality discovered on a routine survey represents an early stage of "chronic obstructive lung disease." Secondly, one should be very cautious in interpreting individual rates of decline in lung function, especially over periods less than ten years. Finally, I am not convinced that we yet know the pre-clinical course of most patients who develop severe disabling chronic airways obstruction. While some may show a steady decline in lung function throughout the whole of their adult life, others may retain relatively normal function and then decline rapidly shortly before the onset of clinical symptoms. Still others may go through a stage of episodic bronchoconstriction which can last for many years before irreversible disease occurs.
1 Traver GA, Cline MG, Burrows B: Predictors of mortality in COPD: A 15-year follow-up study. Am Rev Respir Dis 119:895-902,1979 2 Burrows B, Earle RH: Course and prognosis of chronic obstructive lung disease, a prospective study of 200 patients. N Engl J Med 280:397-404, 1969 3 Burrows B: Course and prognosis in advanced disease (Chapter 2). In Chronic Obstructive Pulmonary Disease (T. Petty, ed) New York, Marcel Dekker, 1978, pp 23-33 4 Fletcher CM, Peto R, Tinker CM, et al: The Natural History of Chronic Bronchitis and Emphysema. Oxford; Oxford University Press, 1976, p 272
The Influence of Bronchitis and Asthma in Infancy and Childhood on Lung Function in Schoolchildren* Ann Woolcock, M.D., F.C.C.P.; S. Leetkr, Ph.D.; ]. Peat, B.Sc.; and C. Blackburn, M.D.
•From the Department of Medicine, University of Sydney, Australia.
CHEST, 77: 2, FEBRUARY, 1980 SUPPLEMENT
Jnasthma order to document the effect of bronchitis and in infancy and early childhood on subsequent lung function, a prospective study of the lung function of a random sample of Sydney schoolchildren was begun in 1971. At the first study, a parental questionnaire was used to obtain a history of previous asthma and bronchitis, severe enough to require treatment by a medical practitioner. Maximal expiratory flow volume (MEFV) curves were recorded with a Fleisch no. 3 pneumotachograph onto magnetic tape. After digital conversion, the forced expiratory volume in 0.5 and 1.0 sec (FEVu and FEV1 }, the forced vital capacity ( FVC) and the maximal flow at 50 percent of the FVC (V50 ) were calculated. The lung function studies were repeated each year between 1972 and 1974, at which time a history of respiratory illness in the previous 12 months and a personal smoking history were recorded. In 1971, 4,639 children from primary schools (mean age 8.9 years) and 6,259 children from secondary schools (mean age 12.6 years} were studied. At this time the children were divided into five groups based on history: 1 ) bronchitis with first attack before age two; 2} bronchitis, first attack after age two; 3) asthma; 4) asthma and bronchitis; 5} neither asthma nor bronchitis. Mean values for lung function were calculated for each group excluding children with cough in the last six weeks or with current wheezing. Lower values for V50 (but not for other measurements} were found in groups 1, 2, 3 and 4 compared with group 5 for both age groups and sexes. The differences were highly significant for all groups and were still present in 1974. In 1974, the data from smokers were compared with those from nonsmokers. The secondary school boys in group 2 (mean age 15.6 years) who smoked had lower values for V50 than nonsmokers (P < 0.01}. Most other groups had too few smokers for meaningful analysis. The studies suggest that bronchitis as well as asthma in infancy and in childhood may affect lung function as children grow, that the abnormality may not be detected by the forced expiratory volume, that the abnormality persists, and that it is possible that it is further affected by smoking. Q: In young children, how do you separate bronchitis from asthma? A: We rely on diagnosis by the private physician or hospital records. Sometimes the conditions are difticult to differentiate.
The Role of Air Pollution, Smoking and Respiratory Illnesses in Childhood in the Development of Chronic Bronchitis* D. Irvine, M.Sc.; A. Brooks; and R. Waller, B.Sc. 0
Medical Research Council, St. Bartholomew's Hospital Medical College, London, EngJand.
Reprint reqtJeSta: Dr. Irvine, Mellical RssetJtCh Cooncil, Toz. icology Unit, Clinical Section, St. Bartholomew's Hosplltll MedicQl College, London EClM BBQ, England
CHRONIC OBSTRUCTIVE LUNG DISEASE 251