Current consensus guidelines, such as the Global. Office Spirometry Significantly Improves Early Detection of COPD in General Practice*

Transcription

1 Office Spirometry Significantly Improves Early Detection of COPD in General Practice* The DIDASCO Study Johan Buffels, MD; Jan Degryse, MD, PhD; Jan Heyrman, MD, PhD; and Marc Decramer, MD, PhD Study objectives: To determine if spirometry is essential for the early detection of COPD in general practice, compared to the screening value of a short questionnaire. Methods: A prospective survey of the population aged 35 to 70 years visiting their general practitioner (GP) during a 12-week period, using a questionnaire on symptoms of obstructive lung disease (OLD). Spirometry was performed in all participants with positive answers and in a 10% random sample from the group without complaints. Twenty GPs were provided with a hand-held spirometer, and received training in performance and interpretation of lung function tests. All 35- to 70-year-old patients (n 3,408) were screened for current use of bronchodilators. The subgroup receiving bronchodilators (n 250, 7%) was assumed to have OLD, and was excluded. Airflow obstruction was defined according to the European Respiratory Society standards. Results: The positive predictive power of the questionnaire was low (sensitivity, 58%; specificity, 78%; likelihood ratio, 2.6). One hundred twenty-six cases of formerly unknown OLD were detected in the group of patients with complaints, vs an extrapolated number of 90 in the group without complaints. Despite a negative predictive value of 95% for the questionnaire used, 42% of the newly diagnosed cases of OLD would not have been detected without spirometry. Conclusions: The use of a spirometer is mandatory if early stages of OLD are to be detected in general practice. Screening for airflow obstruction almost doubles the number of known patients with OLD. (CHEST 2004; 125: ) Key words: asthma; COPD; diagnosis; family practice; spirometry Abbreviations: DIDASCO Differential Diagnosis Between Asthma and COPD; GOLD Global Initiative for Chronic Obstructive Lung Disease; GP general practitioner; OLD obstructive lung disease; YCG yearly patient contact group Current consensus guidelines, such as the Global Initiative for Chronic Obstructive Lung Disease (GOLD) program 1 emphasize the importance of early detection of the disease, even at a preclinical stage. It is commonly accepted that spirometry is the gold standard in the diagnosis of COPD. 1,2 Until now, only a few studies 3 have been performed on the *From the Departments of General Practice (Drs. Buffels, Degryse, and Heyrman) and Pulmonology (Dr. Decramer), Katholieke Universiteit, Leuven, Belgium. Financial support was provided by Boehringer Ingelheim Belgium. Manuscript received July 24, 2003; revision accepted November 24, Reproduction of this article is prohibited without written permission from the American College of Chest Physicians ( permissions@chestnet.org). Correspondence to: Johan Buffels, MD, Academisch Centrum voor Huisartsgeneeskunde Kapucijnenvoer 33 Blok J B-3000 Leuven, Belgium; johan.buffels@coditel.net diagnosis of COPD in general practice, although it appears quite probable that the majority of the patients with unidentified COPD first consult with a general practitioner (GP). Modern lightweight spirometry instruments that can be wired into a computer are now available. This makes spirometry technically feasible at the primary care level. Some authors 4 state that mild and even moderate COPD can occur without complaints or symptoms. Moreover, there seems to be a weak correlation between the severity of the complaints and the severity of airway obstruction. 5 In addition, there is a certain underreporting of complaints in subjects with obstructive lung disease (OLD), particularly shortness of breath. 6 8 Few data are available about the actual incidence and prevalence of the early stages of this disease. A major differential diagnosis of COPD is asthma Clinical Investigations

2 It is important to make a distinction between asthma and COPD, since the guidelines for assessment and management are quite different for the two diseases. 1,9 The aim of the DIDASCO (Differential Diagnosis Between Asthma and COPD) project was to study the ability of general practitioners to detect early stages of chronic OLD, and to make a distinction between asthma and COPD using relatively simple means. The first step in this larger project was to determine the accuracy of the results of office spirometry performed by GPs. Consequently, an algorithm was developed to determine whether airflow obstruction could be diagnosed most effectively by routine spirometry in the target population or by a questionnaire followed by spirometry in the patients with complaints. In this article, we report these results. Materials and Methods General Design of the Study The target population for early detection of COPD was defined as the age group from 35 to 70 years. The participating GPs were asked to screen every patient in this age group attending clinic at least once during 12 successive weeks for use of bronchodilators and/or inhaled steroids, since they were assumed to have known asthma or COPD. All other subjects were to complete a short questionnaire pointing to signs and symptoms of OLD (Table 1). Patients with at least one positive answer were given an appointment for a spirometry within 1 week. In the group without complaints, an electronic randomization procedure assigned 10% to a control group, which underwent a spirometry within 1 month of the questionnaire being completed. The study flow diagram is shown in Figure 1. Spirometry in General Practice Twenty GPs in the region of Brussels were equipped with a Spirobank spirometer (MIR; Rome, Italy). This is a hand-held instrument for lung function tests that has been wired into a computer. Winspiro software (MIR) compares the measured values with reference tables and automatically calculates the reproducibility of the performed spirometry in accordance with European Respiratory Society guidelines. Twenty-five GPs followed an intensive 12-h course, including aspects of diagnosis and management of asthma and COPD, as well as training in Table 1 Questionnaire Do you have one of the following complaints: Cough, lasting for at least 2 weeks? Breathing difficulties during mild exercise, or at night? Wheezing? Any kind of nasal allergy or hay fever? Have you suffered from one or more of these complaints during the past year? Have you ever had to visit your doctor for a wheezing or longlasting cough? performance and interpretation of spirometry. Continuous technical and methodologic support was made available to the participating GPs through the whole registration period. Airflow obstruction was defined according to the European Respiratory Society standards for optimal management of COPD. 10 Values of the FEV 1 /FVC ratio 88.5% of the predicted value for men, and 89.3% of the predicted value for women were considered to be consistent with airflow obstruction. Questionnaire Figure 1. Study flow diagram. The questionnaire was short and simple, because the screening took place during the normal consultation activity of the GPs, including their home visits. The version that was used (Table 1) was derived from the European Community Respiratory Health Study questionnaire. 11 The patients were asked if they were troubled by one of the following complaints either currently or during the past year: a cough lasting for at least 2 weeks, breathing difficulties during mild exercise or during the night, wheezing, or any kind of nasal allergy or hay fever. They were also asked if they ever visited the doctor for wheezing or a long-lasting cough. If one of these questions received a positive answer, the patients were asked to undergo spirometry. The questionnaire focused on signs and symptoms of possible OLD, not on risk factors. There was no separate question about cigarette smoking, because there is existing evidence that spirometric testing should be offered to all current smokers 45 years of age, and GPs usually already know the smoking habits of their patients. An extra question regarding sputum production might have added to the sensitivity of the questionnaire. However, the European Community Respiratory Health Study 11 found a high correlation coefficient between the factors cough and phlegm, and it was important to make the questionnaire as concise as possible. Accuracy of the Spirometric Values The accuracy of the spirometric values obtained by the participating GP was examined as follows. In a separate event, each GP CHEST / 125 / 4/ APRIL,

3 invited a patient with airflow obstruction to a joint meeting in the lung function laboratory of the University Hospital in Leuven. Each of the patients underwent five successive spirometric tests: four tests by different GPs, using their own spirometer, and one test performed by a professional laboratory technician with his usual equipment. A reliability analysis was performed using the generalizability theory (unpublished data). Differentiation Between Asthma and COPD A second part of the study focused on the ability of GPs to distinguish between asthma and COPD. Every patient with newly found airway obstruction was offered a bronchodilator reversibility test within 1 week of the initial spirometry. Reversibility was defined as an increase of at least 12% for the FEV 1 percentage of the predicted value or 200 ml after administration of 400 g salbutamol (or 80 mg ipratropium bromide for patients at risk). Patients with reversible OLD were considered to have asthma. Patients who failed to show reversible airway obstruction after the use of bronchodilators were asked to undergo a steroid challenge test by inhalation for a period of 12 weeks. At the end of that period, lung function tests were repeated. The results of this trial will be presented in a separate article. Table 2 Comparison of the Group With and Without Complaints* Variables Group With Complaints Group Without Complaints Spirometric tests, No Male gender 308 (44) 109 (49) Female gender 395 (56) 113 (51) Smokers 240 (34) 44 (20) Ex-smokers 116 (17) 51 (23) Smokers plus ex-smokers 356 (51) 95 (43) Nonsmokers 336 (49) 127 (57) Airflow obstruction 126 (18) 9 (4) Mean FEV 1 /FVC % in 64.3 (8.3) 63.9 (3.9) obstructive subjects Mean FEV 1 /FVC % in 80.4 (6.7) 80.7 (6.0) normal subjects Mean FEV 1 % predicted in obstructive subjects 73.7 (18.4) 79.4 (15.0) *Data are presented as No. (%) unless otherwise indicated. Data are presented as % (SD). Estimated. Results Demographic Data During a period of 12 weeks from April to July 1999, 22 GPs applied the case-finding algorithm to all of their visiting patients (n 3,408) aged 35 to 70 years. Two hundred fifty patients (7.3%) used bronchodilators and/or inhaled steroids, and were assumed to have asthma or COPD. The other 3,158 subjects were submitted to the short questionnaire. Seven hundred twenty-eight responders (23%) gave at least one positive answer, indicating that they had a higher probability of OLD. From this group, 703 spirometric tests were obtained. Consequently, 126 cases of formerly unknown OLD were identified, ie, 18% of the group with complaints. Ten percent of the 2,430 remaining patients without complaints were selected at random to form a control group. From the 222 spirometric tests performed in this group, nine results (4%) revealed airflow obstruction. Extrapolation to the total group refers to 90 expected new cases. In total, approximately 216 new cases were detected. This represents 7.4% of the population aged 35 to 70 years who spontaneously visit their GP. The newly diagnosed cases with airflow obstruction were spread equally over the different age categories, while patients with known OLD were more obviously present in the older age groups. Fifty-two percent of the population studied were women. In the group with newly diagnosed OLD, 45% were women, whereas in the group with known obstructive airway disease only 15% were women. As expected, the percentage of smokers in the group with newly diagnosed OLD (48%) exceeds by far the number of smokers in the group with normal lung function test results (28%). A less important difference was seen for the number of ex-smokers (23% vs 19%). In the population with formerly unknown airway obstruction, the mean FEV 1 was 2.57 L (SD 0.75), and the mean FEV 1 predicted was 74.09% (SD 18.2). Table 2 shows a comparison between the groups with and without complaints. Table 3 shows the staging of the newly found airflow obstruction. A large majority of subjects presented with mild-tomoderate airflow obstruction: 90% of these patients had a FEV 1 50% of the predicted value. Only one subject had a FEV 1 of 30% predicted at the time of detection. Diagnostic Value of the Questionnaire Using the results of spirometry, the diagnostic value of the questionnaire was computed. The sensitivity of the questionnaire to detect OLD was 58%, and the specificity 78%, which gives a positive predictive value of only 18% (Table 4). The negative predictive value of 96% is also rather weak. With a prior negative chance of 93%, answering negatively to the questionnaire only raises this negative preva- Table 3 FEV 1 in Newly Found OLD FEV 1, % Predicted Stage Subjects, No. (%) 80 GOLD I 53 (39) 50 to 80 GOLD II 69 (51) 30 to 50 GOLD III 12 (9) 30 GOLD IV 1 ( 1) 1396 Clinical Investigations

4 Table 4 Diagnostic Value of the Questionnaire* Variables Obstructive Normal Total Prediction Complaints 126 (58) 577 (22) 703 PPV 18% (126 of 703 cases) No complaints 90 (42) 2,130 (78) 2,220 NPV 96% (2,130 of 2,220 cases) Total 216 (prior chance, 7%) 2,707 (NegPC, 93%) 2,923 PLR, 2,6 (58%/22%); NLR, 0,5 (42%/78%) *Data are presented as No. (%) unless otherwise indicated. PPV positive predictive value; NPV negative predictive value. NegPC negative prior chance; PLR positive likelihood ratio; NLR negative likelihood ratio. lence by 3%, and gives at the same time 4% of false-negative results (90 of 2,230 tests). Consequently, 42% (90 of 216 newly diagnosed cases) with OLD would not have been detected without lung function tests. A regression analysis was performed to determine the significance of each separate question (Table 5). The most relevant item appeared to be prior visit for wheezing or a long lasting cough. Accuracy of the Spirometric Values Both the FEV 1 and the FVC measurements by different GPs were compared to each other and to the values obtained by a professional in the lung function laboratory. The mean difference in FEV 1 / FVC between the values of the professional equipment and the Spirobank device was 2.16% (SD 0.58). The mean difference in FVC appeared to be 3.52% (SD 0.76). The SEM by one observer was calculated to reach L. The interobserver variance was smaller than the mean expected intraindividual variance for both measurements. Discussion The present study demonstrates that spirometry in general practice is feasible and sufficiently accurate. In the target population aged 35 to 70 years, we found 7.4% of patients with formerly unknown OLD. Of these patients, 42% would not have been found without the screening with spirometry. The degree of airflow obstruction was mild to moderate. Epidemiologic Considerations In the DIDASCO project, the GPs managed to screen all of their visiting patients of the selected age group for this prospective survey. Although this study was not designed as a population survey, some epidemiologic extrapolations can be attempted, in the form of rough estimations. Belgian primary health care has no patient census list. Prior evidence has shown that the number of patients visiting the GP at least once during a 12-week period equals a 60% fraction of the yearly patient contact group (YCG). This total YCG is relatively stable in a single practice and is equivalent to approximately 70% of the total practice population. 12,13 During the 12 registration weeks, 250 patients in 18 practices were found to use bronchodilators. After simple linear extrapolation for the YCG, this should refer to 416 cases with known OLD. However, it might be argued that the majority of the population with asthma or COPD will visit the GP within a 12-week period anyway, either for follow-up of the disease and prescriptions, or for intercurrent diseases or exacerbations. Using this minimal hypothesis, the total number of known cases with airway obstruction would hardly exceed the initial 250 cases. However, several studies 7 report a low compliance to therapy in a significant number of patients with obstructive airway disease, especially in the group with mild signs and symptoms. This could lead to underreporting in the present study, even in the maximal hypothesis. Table 5 Coefficients* Unstandardized Standardized Model SE t p Value Constant Breathing difficulty 8.75E Consult wheezing/cough Wheezing 9.98E Cough 2.19E Nasal allergy 1.240E *Obstruction is dependent variable. CHEST / 125 / 4/ APRIL,

5 On the denominator side, a total number of 3,408 patients from 35 to 70 years old underwent screening. Extrapolation to the YCG leads to 5,680 subjects, or a total practice population of 8,114 persons for this age group. The prevalence of known airway obstruction in this study can be calculated as from 250 to 416 cases for the total population of 8,114, or from 30 to and 51 cases per 1,000. This is quite comparable with prevalence estimates obtained in other studies eg, the United States Third National Health & Nutrition Examination Survey 14 and the Confronting COPD International Survey. 15 The number of patients that can be found during 1 year using the active screening strategy adopted in this study is estimated to total 360 cases (extrapolation to the YCG). Comparing this figure with the number of cases with known airway disease, we conclude that there is about one patient with unknown asthma or COPD for every known patient in primary care (360 cases vs 250 to 416 estimated cases). These findings confirm the conclusions of earlier studies that OLD in general practice can be compared to an iceberg: only the tip is visible. The combined prevalence of new and known cases of OLD leads to an estimated global prevalence between 75 and 96 per 1,000. Of course, as we stated above, several causes of bias must be taken into account since this was not a population study. The figures of newly diagnosed obstructive airway disease in the present study are lower than those of a Dutch population study, the Detection, Intervention and Monitoring of COPD and Asthma project. 16 It should be noted that the design of this included several measurements of the lung function, while the DIDASCO study is based essentially on only one spirometry as a screening tool. Probably a larger number of asthma patients can be detected if spirometry is repeated during a period of several months. This could be considered as a methodologic weakness of the study design, but the primary focus was rather early detection of COPD. Furthermore, the DIDASCO study is not to be considered as a study about the broad general population. Only patients visiting a GP were screened. It is difficult to estimate the differences between this selection and the general population. More patients with general morbidity will probably have been included in the present study. However, patients with low self-care and low health-care consumer profile will be less represented. In comparison to a Polish study 18 on early detection of COPD, we found lower rates for OLD; this can be explained by the different inclusion criteria because the Polish study focused on smokers only. Is Early Detection of Obstructive Airway Disease Indicated? Early detection of a disease only makes sense if an appropriate treatment is available. 19 For early stages of COPD, the only approach that has proven useful on modifying the course of the disease is smoking cessation. However, the usefulness of medical treatment for mild and early asthma is well documented. Early detection of obstructive airway disease should be promoted if an important fraction of the newly diagnosed cases appears to be asthma. Moreover, general measures such as avoidance of other etiologic factors, physiotherapy, anti-infectious prophylaxis, and adequate physical activity can be promoted earlier in the course of COPD. 1,20 Conclusion The present study shows that spirometry in general practice is feasible provided that the GPs are instructed and trained in the use of a hand-held spirometer. The accuracy of their measurements was found to be highly acceptable. The GPs were able to incorporate spirometry in their daily routines. If early detection of OLD is to be realized in general practice, the use of a spirometer is essential. Detection based on the use of a questionnaire appeared to be insufficient. These conclusions could change the policy for early detection of OLD. 1,15,16,21,22 Efforts should be made to make hand-held spirometers available in primary care, and to provide GPs with skills and training in this specific domain. Programs for routine office spirometry in general practice are to be encouraged. ACKNOWLEDGMENT: We thank the GPs who participated in the registration: P. Bastaerts, M. Biervliet, P. Boeykens, B. Bruynbroeck, L. De Moor, P. De Ridder, A.M. Deleenheer, L. Frisch, R. Gors, M. Joossens, H. Marechal, L. Martens, C. Micholt, G. Opsomer, W. Renier, A. Roex, M. Roex, I. Tambuyzer, P. Tarpataki, D. Van Achter, D. Van de Sijpe, and L. Van Parijs. We are very grateful to P. Vermeire, C. Van Weel, and C. P. Van Schayck for reviewing the study design. References 1 U. S. Department of Health and Human Services, Public Health Service, National Institutes of Health, National Heart, Lung, and Blood Institute. Global initiative for chronic obstructive lung disease: pocket guide to COPD diagnosis, management, and prevention. NIH Publication No. 2701B, Available at: Accessed March 16, Calverley PM. COPD: early detection and intervention. Chest 2000; 117(Suppl):365S 371S 3 Nihlen U, Montnemery P, Lindholm LH, et al. Detection of chronic obstructive pulmonary disease (COPD) in primary health care: role of spirometry and respiratory symptoms. Scand J Prim Health Care 1999; 17: Clinical Investigations

6 4 Van den Boom G, Rutten-van Molken MPMH, Tirimanna PRS, et al. Association between health-related quality of life and consultation for respiratory symptoms: results from the DIMCA programme. Eur Respir J 1998; 11: Brand PLP, Rijcken B, Schouten JP, et al. Perception of airways obstruction in a random population sample: relationship to airways hyperresponsiveness in the absence of respiratory symptoms. Am Rev Respir Dis 1992; 146: Barnes PJ. Poorly perceived asthma. Thorax 1992; 47: Kendrick AH, Hoggs CMB, Whitfield MJ, et al. Accuracy of perception of severity of asthma: patients treated in general practice. BMJ 1993; 307: Van den Boom G, Tirimanna PRS, Kaptein AA, et al. Under presentation of shortness of breath in the general population: results of the DIMCA programme. Asthma Gen Pract 1999; 7:3 7 9 Global strategy for asthma management and prevention: update report NIH Publication No Available at Accessed March 16, Siafakas NM, Vermeire P, Pride NB, et al. Optimal assessment and management of chronic obstructive pulmonary disease (COPD): The European Respiratory Society Task Force. Eur Respir J 1995; 8: Sunyer J, Basagan X, Burney P, et al. International assessment of the internal consistency of respiratory symptoms: European Community Respiratory Health Study (ECRHS). Am J Respir Crit Care Med 2000; 162: De Loof J, Heyrman J, De Jaarlijkse Contact Groep [in Dutch]. Huisarts Nu 1979; 8: Kilpatrick SJ. The distribution of episodes of illness: a research tool in general practice? J R Coll Gen Pract 1975; 25: Mannino D, Gagnon R, Petty T, et al. Obstructive lung disease and low lung function in adults in the United States: data from the National Health and Nutrition Examination Survey, Arch Intern Med 2000; 160: Rennard S, Decramer M, Calverley P, et al. The impact of COPD in North America and Europe in 2000: the subjects perspective of the Confronting COPD International Survey. Eur Respir J 2002; 20: Van den Boom G, van Schayck CP, Rutten-van Molken, MPMH et al. Active detection of chronic obstructive pulmonary disease and asthma in the general population. Am J Respir Crit Care Med 1998; 158: Coultas D, Mapel D, Gagnon R, et al. The health impact of undiagnosed airflow obstruction in a national sample of United States adults. Am J Respir Crit Care Med 2001; 164, Zielinski J, Bednarek M, et al. Early detection in a high-risk population using spirometric screening, Chest 2001; 119: Wilson JMG, Jungner G. Principles and practice of screening for disease. Public Health Paper No. 34. Geneva, Switzerland: WHO, Troosters T, Gosselinck R, Decramer M. Short- and longterm effects of outpatient rehabilitation in patients with chronic obstructive pulmonary disease: a randomised trial. Am J Med 2000; 109: Van den Boom G, Rutten-Van Mölken MPMH, Den Otter JJ, et al. Economic evaluation of early detection of asthma/ COPD: does screening lead to additional use of health care resources? Am J Respir Crit Care Med 1994; 149:A Pauwels RA, National and international guidelines for COPD: the need for evidence. Chest 2000; 117(2 Suppl):20S 22S CHEST / 125 / 4/ APRIL,