Underdiagnosis of chronic obstructive pulmonary disease: should smokers be offered routine spirometry tests?

KEYWORDS:

• COPD
• underdiagnosis
• screening
• smoking
• mild chronic obstructive
• pulmonary disease

1. The problem of COPD underdiagnosis

Underdiagnosis of chronic obstructive pulmonary disease (COPD) has been found in all surveys and can reach >90% in spirometrically defined cases [1,2]. Even in developed countries, the number of spirometry tests performed in primary care has only increased marginally [3]. It is contradictory that this continues to occur despite that COPD is the third leading cause of death globally [4,5] and that there are several guidelines and recommendations for the diagnosis and treatment of COPD [5,6]. Patients under-recognize the significance of respiratory symptoms, and physicians tend to diagnose COPD based on clinical symptoms and exposures and frequently miss opportunities to request a spirometry [7]. Only 12.2% of patients with clinical symptoms suggestive of COPD have a spirometry to confirm the diagnosis [8]. But in countries where the majority of participants in surveys reported a previous spirometry, underdiagnosis remains common [2], which proves that spirometries require, similarly to other tests, quality control and proper interpretation, described in detail in international spirometry guidelines [9]. Of course, all facets of COPD cannot be identified with spirometry. Recent instruments to predict outcomes in COPD are multidimensional and include exercise capacity, health-related Quality of Life (QOL), and respiratory symptoms, especially dyspnea (e.g. the BODE index) [10]. In addition, smokers may have a variety of physiological abnormalities and exercise limitations without airflow obstruction, including chronic bronchitis and emphysema [11]. Computed tomography (CT) scanning diagnoses of emphysema can be present, and even in a substantial amount, with normal spirometry and lung function [12], and it predicts varied adverse outcomes, such as less physical activity, exacerbations, and accelerated lung-function decline [5,13,14].

Physicians, especially pulmonologists, have at their disposal additional tests that may provide important information on the patient, including CT scanning and other lung-function tests (Carbon monOxide Diffusion capacity of the Lung, DL_CO, lung volumes). A panel of experts utilizing all tools revealed differences in their diagnoses compared with spirometry [15]. Unfortunately, these tests require sophisticated equipment and training and are unavailable for the majority of patients, even in developed countries. However, spirometry, a simple test compared with those available to specialists remains a valuable tool for identifying exposed individuals with a higher risk of complications and adverse outcomes.

2. Avoiding underdiagnosis

Best strategy to avoid underdiagnosis is, curiously, undefined by scientific evidence. Proper and sufficiently large clinically controlled trials to define best procedures for case-finding of or screening for COPD, especially in high-risk groups, are lacking. Yet more importantly, it remains unknown whether or not screening leads to an improved relevant clinical outcome, such as QOL or survival [7].

Smoking, especially cumulative smoking (pack-years), is the most important risk factor for airflow obstruction [5]. Consequently, the presence of smoking is a common requisite for selecting individuals for screening or case-finding. But how much smoking is required to merit a spirometry, given that the higher the level of smoking, the higher the prevalence of COPD in the group? Also, the presence of respiratory symptoms and exacerbations, or a previous physician diagnosis of asthma or COPD, are associated with higher risk of airflow obstruction [16]. The risk for COPD comprises a continuum, and dichotomizing the risks for selecting individuals for screening leads to errors, both false-positive and -negative [13]. Several international organizations (American Thoracic Society, European Respiratory Society, American College of Chest Physicians, Global Initiative for Chronic Lung Disease [GOLD] [5]) have emitted recommendations for COPD screening, some meta-analyses and systematic reviews have analyzed this important issue, and the US Preventive Task Force (USPTF), the most comprehensive and detailed of these, has carried this out periodically [17,18].

In an efficient screening, the gold standard tends to be reserved for confirmation, especially if is expensive or risky, and the process can be step-wise, starting with a sensitive, inexpensive, and safe test and ending with the gold standard. Between these extremes, tests to increase specificity can be incorporated, aiming to reduce the need of the gold standard. This step-wise testing has been applied formally in the diagnosis of COPD [19] and has demonstrated efficiency. This strategy is applied routinely by the clinician in daily practice, selecting individuals for further testing in the presence of symptoms and exposures, that is, with a higher pretest probability of airflow obstruction group in which spirometry testing is more efficient and accompanied by fewer false-positive results.

3. Adverse consequences of underdiagnosis

Individuals with undiagnosed COPD likely lack access to available treatment and identification of relevant comorbidities. However, evidence for useful interventions in mild COPD, more likely found in a screening program, is limited: a reduction of symptoms and exacerbations with long-acting bronchodilators was described in one controlled clinical trial [20]. The majority of drug treatments for COPD have been tested in moderate-to-severe airflow obstruction. Fortunately, underdiagnosis is less common in moderate-to-severe airflow obstruction [1,2], where the majority of the interventions listed have proved their usefulness. However, even in severe airflow obstruction, underdiagnosis may be substantial [1,2].

Costs involved in widespread spirometry would be high and wasteful, as COPD prevalence in individuals 40 years of age or older in many countries is relatively low, especially in asymptomatic individuals lacking risky exposures. Case-finding in higher-risk groups considerably increases efficiency.

All smokers should receive the recommendation to quit, but quitting is more urgent in individuals with lung-function abnormalities, respiratory symptoms, or comorbidities. Increasing rates of tobacco abstinence have been described after spirometry results, but this finding has not been consistent.

4. Risks of screening for COPD

Screening also possesses potential risks and disadvantages in addition to costs, as individuals labeled with potential COPD require further tests and are subject to anxiety and fears. Screening at the primary-care level may be focused on individuals with respiratory symptoms and high-risk exposure. Mass screening of asymptomatic smokers, with a low prevalence of disease, would lead to abundant false-positive results, especially in the elderly [13,20], and moreover if GOLD fixed-ratio criteria are utilized for diagnosis instead of the age-adjusted lower limit of normal [13,21]. The diagnosis of airflow obstruction is currently based on the postbronchodilator spirometric FEV₁/FVC ratio, known to decrease progressively with age. However, the GOLD Committee continues to recommend, as a criterion of airflow obstruction, a fixed post-bronchodilator FEV₁/FVC<0.7, which leads to a considerable false-positive rate in elderly, asymptomatic never-smokers, a disadvantageous situation for a widespread systematic campaign attempting to identify undiagnosed COPD. It is clear, however, that COPD increases with aging regardless of the spirometric criteria, due to cumulative exposure to tobacco smoke and other pollutants. In addition, diagnosis offers more difficulties due to the frequent comorbidities of the elderly and the less-than-optimal performance of spirometry in older age.

5. Comment

Unfortunately, the most relevant clinical trials to demonstrate a proper strategy to screen or case-find individuals with COPD are lacking and, most importantly, trials that demonstrate measurable benefits from screening are also lacking [16–18]. Despite recent evidence of a variety of functional abnormalities in individuals with mild COPD or in smokers without airflow obstruction, systematic efforts should concentrate on identifying moderate-to-severe COPD in which clinical consequences are important and a variety of treatments are available.

Primary care should provide the main health recommendations, including those on healthy habits, diet, exercise, and quitting smoking.

Best screening strategy likely would vary with the country, region, and overall characteristics of the population and the health system. Symptoms and exposures can be explored quickly with questionnaires, selecting individuals considered at high risk for spirometry. An intermediate step with a simplified lung function test (Peak Expiratory Flow Rate, PEFR, or 6-s spirometry) may increase availability and reduce spirometry tests, a strategy especially efficient if the aim is to identify moderate-to-severe airflow obstruction.

Risks for COPD can be considered a continuum that includes age, gender, familial predisposition, tobacco smoking, and exposure to other types of smoke or occupational dusts and fumes, a previous diagnosis of asthma, chronic bronchitis, or COPD. The greater the number of risk factors, the greater the need for spirometry testing. Although screening programs target individuals without a previous diagnosis of asthma or COPD [18], a physician’s diagnosis of COPD is often based on clinical symptoms and is erroneous. A previous diagnosis of COPD or asthma requires spirometry testing if not conducted previously. Current questionnaires to identify COPD or individuals at risk for COPD (e.g. the Clinical COPD Questionnaire score CDQ, or the Lung Function Questionnaire [LFQ]), do not include all possible risky exposures that may be relevant under specific conditions or in specific places, such as exposures to occupational dust and fumes, or biomass smoke, essential in some populations, and questionnaires should not substitute for clinical judgment. The lower the COPD risk, the more unnecessary spirometry testing is and the higher the risk of false-positive results. Where to draw the dividing line at this moment or which level of risk should trigger the measurement of PEFR or spirometry must await more clinician-oriented trials and, at this time, must be decided by the physician, especially in the presence of uncommon exposures. Current guidelines recommend spirometry for individuals ranging from those above 35 years of age with any smoking with or without respiratory symptoms (usually cough, phlegm, wheezing, or dyspnea), to individuals over 45 years of age with at least moderate smoking (>20 pack-years), or equivalent exposures in symptomatic individuals [5,6,22].

In the community setting, again, the advantage of early identification of COPD cases must be demonstrated, as well as the best strategy to perform this. The latter could be conducted step-wise, reserving spirometry for individuals with a higher risk of COPD identified by a questionnaire and a simplified pulmonary-function test [19]. New GOLD guidelines emphasize respiratory symptoms and exacerbations and the importance of the efforts to reduce them, even in individuals with mild airflow obstruction [5]. This is a healthy alert for clinicians, but also an appeal for much needed trials defining the best way to improve symptoms and outcomes, in addition to stopping exposure to smoke and pollutants in individuals with mild airflow obstruction or no airflow obstruction.

In summary, currently, collective efforts should be focused on identifying moderate-to-severe COPD, targeting, for spirometry, symptomatic smokers (or individuals with exposure to smoke or dust) preferably if abnormal results in PEFR or 6s spirometry were obtained. However, all smokers should be advised to quit and offered help to accomplish this, as recommended by all organizations and dictated by common sense.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This manuscript has not been funded.