Rectifying COPD management: a critical analysis of the Global Initiative for Chronic Obstructive Lung Disease strategy

Abstract

The Global Initiative for Chronic Obstructive Lung Disease has greatly contributed to the management of COPD. However, several important issues persist. The initial conception of chronic bronchitis as a short-term illness caused by 1950s air pollution differs significantly from the current understanding of COPD. COPD and asthma are often diagnosed subjectively due to the lack of definitive diagnostic criteria. Asthma patients who lack classic features may be misclassified as having chronic bronchitis, contributing to diagnostic overlap and unnecessary heterogeneity. The inclusion of asthma patients in COPD studies may have introduced significant bias, leading to misguided management recommendations. Notably, studies that attempted to minimise asthma misclassification (such as by excluding patients with elevated blood eosinophil counts) demonstrated more favourable outcomes in the absence of inhaled corticosteroids compared to studies that applied no such exclusion criteria. COPD patients with blood eosinophilia respond to inhaled corticosteroids precisely because they probably have asthma. Bronchodilator reversibility and airway hyperresponsiveness tests have limited utility in ruling out asthma. Despite their limitations, lung function tests remain central to the assessment and management of COPD. Their clinical value is limited by measurement variability and controversies regarding their interpretation. The current “combined assessment” tool may have limited validity, particularly as treatment strategies have become increasingly standardised, emphasising early initiation of dual bronchodilators and avoidance of corticosteroids. Existing strategy documents remain subject to future revisions, as has occurred with previous versions. Realising their full potential requires critical engagement and thoughtful revision, particularly in addressing persistent areas of controversy.

Shareable abstract

In COPD studies, asthma patients were often included due to the lack of definitive diagnostic criteria, confounding treatment strategies. Lung function test results should be interpreted with flexibility, recognising their limited validity/utility. https://bit.ly/3VtuYCJ

Introduction

COPD is one of the greatest burdens and health threats worldwide [1]. COPD management is challenging due to its chronic, irreversible progression, frequent exacerbations and limited treatment responses [2]. The Global Initiative for Chronic Obstructive Lung Disease (GOLD) was established in 1997 to improve awareness and standardise COPD diagnosis and treatment [3].

The development of effective inhaled drugs and combinations thereof for COPD treatment is promising [4]. For nearly 30 years, the freely accessible, annually updated GOLD reports have guided clinicians [5]. Since their initial issuance, significant revisions have been made, particularly regarding severity grading for optimal treatment, combination drug strategies, and the role of corticosteroids [3, 5–9].

Given the frequent revisions, further changes are expected [5, 6, 9, 10]. The current strategy document should not be viewed as definitive. Ensuring effective future COPD management through the GOLD strategy will require ongoing constructive feedback and collaborative refinement. Although critical perspectives regarding these issues exist, these perspectives have not resulted in adequate changes to the GOLD reports [11–13]. A comprehensive evaluation of current and potential issues is necessary to maintain their relevance. This review does not intend to negate the importance of the GOLD reports. Rather, it aims simply to open a discussion on issues that could be considered to improve the understanding and management of COPD.

The COPD–asthma overlap: diagnostic challenges and therapeutic implications

Diagnosing COPD and asthma is challenging due to their overlapping clinical features and symptoms, and the lack of clearly defined diagnostic criteria for both [14, 15]. Despite extensive research on, and development of, inhaled drugs over the past 50 years [4, 5, 10], accurate differentiation between COPD and asthma remains a crucial prerequisite for valid clinical studies. Inclusion of undiagnosed asthma patients in COPD cohorts may significantly bias research outcomes.

While inhaled corticosteroids (ICS) are a cornerstone of asthma management, their role in COPD is controversial and has diminished [2, 16]. However, corticosteroids remain a significant component of COPD treatment, particularly during acute exacerbations [2]. This apparent inconsistency in corticosteroid efficacy may stem, in part, from the ambiguous diagnostic criteria for COPD and asthma.

Diagnosing asthma remains inherently subjective due to the absence of a definitive diagnostic gold standard. In COPD treatment research, excluding asthma is crucial, yet challenging. COPD clinical studies often use the criterion “prior history of asthma or current diagnosis of asthma” to exclude patients with asthma, and yet that diagnosis of asthma may depend on subjective physician assessment [17–21]. Asthma is frequently underdiagnosed and misdiagnosed due to various physician and patient factors [22, 23].

Although the diagnostic criteria for COPD are generally more detailed than those for asthma in clinical studies, comprehensive and standardised definitions are still lacking [24]. The definition of emphysema remains largely restricted to pathological findings, with limited attention to its clinical manifestations [25]. Similarly, the current definition of chronic bronchitis is impractical, and no clinically applicable diagnostic criteria have been established in nearly 70 years [3, 10, 25, 26]. Despite these limitations, COPD diagnosis primarily relies on spirometric thresholds [10].

Although the diagnostic limitations of both COPD and asthma are recognised, they are often frequently overlooked due to the perceived lack of immediate solutions. However, this “elephant in the room” situation can no longer be tolerated, given the seriousness of these diseases. The foundation of COPD management efforts will be threatened if these issues remain unaddressed. While immediate solutions may be elusive, these problems must be acknowledged and addressed. Clinical investigators and physicians have a responsibility to ensure that pharmaceutical advancements translate into improved patient outcomes.

Salient features of asthma are well-established and used for diagnosis. The Global Initiative for Asthma guideline describes characteristic symptoms, clinical history, and spirometry findings to support asthma diagnosis [27]. Although these criteria are designed to enhance diagnostic accuracy, they may paradoxically hinder it. Asthma is characterised by considerable clinical heterogeneity, and typical features or supportive test results are not consistently observed in all patients [28].

Symptom and obstructive lung function reversibility is a hallmark of asthma. However, reversibility tests have low sensitivity, and full reversibility may be complicated by factors such as long-standing, severe, poorly controlled and nonallergic asthma [29, 30]. Bronchodilator reversibility (BDR) tests yield positive results in <20% of asthma patients [31].

Notably, BDR rates of 18.4% have been reported in COPD patients, compared to 17.3% in subjects with asthma [32], and nearly two-thirds of patients with moderate-to-very-severe COPD tend to have positive BDR results [33].

Airway hyperreactivity (AHR) provocation tests are considered sensitive and specific for asthma, but in white Americans, sensitivity of 69% was reported, and was as low as 52% in nonatopic asthma [34]. Furthermore, >25% of COPD patients exhibit positive AHR [35]. Methacholine challenge test results vary with technical factors, and interpretation is controversial due to a wide borderline range [36]. BDR and AHR positivity rates in asthma may be even lower when underdiagnosed cases are taken into account [22]. Considering these limitations, the clinical utility of BDR and AHR assessments is probably overestimated [28].

As corticosteroids were recognised to be of limited utility and to have potential adverse effects, including pneumonia, researchers explored reducing or discontinuing their use, while emphasising the emerging dual bronchodilator combination of long-acting β-agonist (LABA) and long-acting muscarinic antagonist (LAMA) [19].

To clarify the role of ICS in COPD, numerous global studies have been conducted. However, the results were potentially influenced by the inclusion of asthma patients in COPD cohorts. Some studies attempted to exclude asthma based on participants’ blood eosinophil counts, revealing significant outcome differences. The landmark FLAME study, which directly compared LABA/LAMA and ICS/LABA for exacerbation prevention, surprisingly demonstrated superior outcomes without ICS [19]. That study excluded patients with a history of asthma and blood eosinophil counts >600 cells·µL⁻¹. A prior study, LANTERN, using the same blood eosinophil criteria, also showed LABA/LAMA superiority [37]. The SUNSET study found no exacerbation risk after ICS withdrawal from triple therapy in COPD patients with similar blood eosinophil exclusion criteria [38]. Only the AFFIRM study, which lacked statements regarding specific asthma exclusion strategies, showed comparable exacerbation-protective effects for LABA/LAMA without ICS [39].

The large-scale WISDOM study evaluated the consequences of gradual ICS withdrawal from triple therapy in COPD patients [20]. While not employing blood eosinophil criteria, it found no increased exacerbation risk after de-escalation to LABA/LAMA. However, a rescue procedure that allowed ICS addition at the physician's discretion may have filtered out some asthma patients from the final analysis. The unexpected significant decrease in forced expiratory volume in 1 s (FEV_1­) after ICS withdrawal may reflect the influence of remaining asthma patients. Single-inhaler triple therapy (SITT) with ICS/LABA/LAMA was developed and popularised for both COPD and asthma management. Several large studies have evaluated SITT against other drug combinations, comparing inhaler regimens with and without ICS [17, 18, 21, 40]. However, those studies did not specifically implement asthma-exclusion procedures, such as blood eosinophil criteria, and consistently failed to replicate the exacerbation prevention efficacy of dual bronchodilators without ICS seen in earlier studies such as FLAME. Recent real-world studies have evaluated SITT in COPD [41, 42]. In these settings, asthma-patient inclusion is less of a concern, as clinicians are more likely to prescribe ICS-containing inhaler combinations. Additionally, patients who experience inappropriate or undesirable drug effects may withdraw from the cohort.

Notably, both real-world and controlled SITT studies have consistently demonstrated that ICS-responsive patients exhibit blood eosinophilia. However, such COPD patients may in fact have undiagnosed asthma. Their exclusion based solely on a lack of a prior asthma diagnosis, one that often relies on subjective assessment, is questionable. Blood eosinophilia and corticosteroid responsiveness are characteristic features of asthma, while COPD, including emphysema, lacks a clear pathogenic or pathophysiological explanation for eosinophilia [43]. Some ambiguous conditions such as cough-variant asthma, asthma–COPD overlap syndrome, and eosinophilic bronchitis, which may be asthma, are often misclassified due to an incomplete presentation of classic asthma features. Although the above analyses suggest that the inclusion of asthma may have influenced study outcomes, further studies are needed to distinguish definitively between misdiagnosed asthma and the eosinophilic phenotype of COPD.

Given the issues outlined herein, the potential inclusion of asthma patients in COPD trial populations may undermine the accurate evaluation of ICS in COPD. Additionally, biologics, which have become integral to asthma management [44], have shown promise in selected COPD patients, particularly those with eosinophilia [45–47]. However, their apparent effectiveness in COPD may be confounded by the inclusion of asthma patients within COPD cohorts. Therefore, concerted efforts are needed to derive more robust criteria for differentiating asthma from COPD, thereby ensuring the validity of appropriate therapeutic strategies.

Despite declining smoking rates, the incidence of COPD has not decreased [1]. This phenomenon may be partly attributed to the long latency period of COPD, which typically requires decades of exposure, as well as to other contributing risk factors, including air pollution [2]. Furthermore, the inclusion of the asthma patients lacking classic features, misdiagnosed as having chronic bronchitis, may also account for this observation. Regardless, if factors unrelated to smoking are increasingly contributing to COPD development, current COPD studies, which typically use a “>10 pack-years of smoking history” inclusion criterion, should be reconsidered so that, in future studies, the COPD population may be represented more accurately.

Questionable origins of chronic bronchitis

The identification and characterisation of chronic bronchitis primarily originated in the United Kingdom during the 1950s, remaining largely unknown elsewhere. This phenomenon has been linked to severe air pollution associated with industrialisation, excessive coal use, and specific meteorological conditions, all occurring in the absence of effective environmental controls, particularly during the 1952 Great Smog [48–51]. At that time, chronic bronchitis was considered a short-term illness characterised by toxic inflammation of the respiratory tract or exacerbation of pre-existing respiratory conditions. The event resulted in widespread respiratory distress, with >12 000 deaths and 150 000 hospitalisations. The British Medical Research Council (MRC) established the Air Pollution Unit to assess and manage this public health crisis [26]. The MRC played a pivotal role in developing and administering comprehensive questionnaires to evaluate the affected population [52]. This effort led to the widely accepted definition of chronic bronchitis as “the presence of a chronic productive cough and sputum production for ≥3 months out of a year for a 2-year period” [52, 53]. However, this traditional definition is difficult to apply in clinical practice, and a more practical, clinically applicable definition remains to be established.

At the 1958 CIBA guest symposium, efforts were made to define and classify chronic obstructive respiratory diseases [25]. British representatives played a pivotal role in establishing chronic bronchitis as a major component of what was then termed as chronic nonspecific lung disease, alongside asthma and emphysema. At the time, participants from other countries were unfamiliar with the clinical utility of chronic bronchitis [25, 54]. Subsequently, emphysema and chronic bronchitis were combined to form COPD, whereas asthma was categorised separately. This historical progression has probably had a lasting influence on current concepts of disease classification and clinical characterisation. However, the chronic bronchitis of that era differed significantly from the current understanding of COPD. While air pollution can contribute to COPD, the underlying pathophysiology is distinct from the long-term, irreversible structural damage now associated with COPD [2]. A study that used questionnaires to evaluate and validate the definition of chronic bronchitis focused on postal workers exposed to high levels of air pollution. However, it did not report specific measures to exclude asthma and emphysema, nor did it adequately address smoking as a contributing factor [52]. The historical context remains poorly understood within the respiratory community. When chronic asthma patients lack classic features, they may be misdiagnosed as having chronic bronchitis, leading to diagnostic controversies. It is necessary to reconsider and redefine chronic bronchitis, taking into account the potential for diagnostic overlap with asthma and the disconnect between its historical background and current clinical understanding.

COPD is currently considered an “umbrella term” that encompasses “diverse phenotypes” due to the combination of emphysema and chronic bronchitis, two structurally and clinically distinct diseases, into a heterogeneous disease entity [55]. From a historical perspective, this heterogeneity may be, to some extent, an artificial construct. However, the terminology of chronic bronchitis is increasingly used in contemporary clinical phenotyping efforts, extending beyond its origin as a historical artefact. It should be recognised that a productive cough remains a relevant clinical feature in a subset of patients with COPD and has been associated with outcomes such as exacerbation risk. Further studies are needed to establish a clinically meaningful and historically informed definition of chronic bronchitis.

Emphysema, currently defined solely as “destruction of alveolar septa distal to terminal bronchioles”, neglects inflammatory changes [56]. Expanding asthma criteria to include diverse clinical manifestations would resolve confusion regarding corticosteroid responsiveness and eosinophilia, which are better explained by asthma. Steroid responsiveness and blood eosinophilia could even serve as adjunctive diagnostic criteria for asthma. Based on these concepts, the diagnostic boundaries of emphysema and asthma may be broadened to include certain features of chronic bronchitis. However, current studies suggest that steroid-responsive COPD cannot be fully attributed to undiagnosed asthma alone, without considering the possibility of a distinct eosinophilic COPD phenotype.

The evolving landscape of GOLD patient stratification

Throughout the GOLD era, significant changes have occurred in patient stratification. Initially, disease severity was classified into four stages based on FEV₁ in 2007 [57]. In 2011, symptom severity and exacerbation history were incorporated to create a two-dimensional “combined initial assessment” (ABCD grades), as FEV₁ alone did not adequately reflect symptom severity, health status, functional limitation, quality of life or outcome [58, 59]. Exacerbation history also emerged as a strong predictor of future exacerbation risk [60]. By 2017, FEV₁ was removed from the scheme, and the assessment tool was solely on symptom scores and exacerbation risks [9]. In 2023, the four-sector design was simplified to a three-sector ABE grade system [10].

These revisions raise several concerns. First, FEV₁ remains an important objective measure for treatment guidance. Numerous studies have demonstrated the significance of FEV₁ for outcome prediction [61, 62]. The new combined assessment system has failed to outperform the older staging system in terms of mortality prediction [63, 64]. Furthermore, strong correlations exist between FEV₁ changes and patient-reported health status [65–67]. The GOLD report's assertion that “FEV₁ may be useful for population levels, but not individual evaluation” is contradicted by the individual-level studies that have found superior performance of lung function tests over the ABCD assessment [7, 63, 64, 68]. Statistical analysis of grouped populations aims to elucidate the significance for individual application, not solely for population characterisation.

Second, the combined assessment grouping may be limited in terms of clinical distinction, as early dual bronchodilator (LABA/LAMA) therapy is widely recommended, despite the GOLD report's advocacy for the use of single bronchodilators in group A. Treatment options have increasingly converged across groups, favouring dual bronchodilators and discouraging ICS use.

Third, the division of a wide scoring range into only two sectors is arbitrary. The COPD Assessment Test (CAT), with scores ranging from 0 to 40, assigns patients with scores of 9 and 10 to different groups, despite similar symptom severity. Fourth, the horizontal axis, based on CAT or modified MRC scores, yields significantly different patient groupings near the border [69]. Finally, the vertical axis, based on exacerbation frequency, lacks universally defined acute exacerbation and frequent exacerbator criteria [12, 70, 71]. Exacerbation frequency varies yearly in individuals [72, 73]. Dividing exacerbation risk into two sectors oversimplifies the significance of each incremental exacerbation [74].

Finding better lung function test criteria for COPD

The 2025 GOLD report addresses several key issues related to spirometry interpretation [75]. BDR testing is no longer routinely recommended, reflecting its limited clinical utility. Airflow obstruction is often defined as an FEV₁/forced vital capacity (FVC) ratio of <0.7, particularly in COPD diagnosis. However, this fixed ratio should not be considered an absolute criterion, as it naturally decreases with age and can be influenced by various conditions and diseases [76–79]. Minor restrictive conditions, such as interstitial lung abnormalities, senile pulmonary fibrosis, pleural thickening, obesity, muscular weakness, and insufficient effort during testing, can also affect the ratio by decreasing FVC [77, 80].

The normal FEV₁ range, traditionally 80–120% of predicted values based on the coefficient of variation, applies primarily to young, healthy adults in their twenties. Younger and older individuals exhibit wider ranges [76, 81]. Furthermore, predicted normal spirometric values vary significantly depending on race and the reference standards applied [81]. Preserved ratio impaired spirometry, a recently discussed phenomenon, may be a product of variability rather than a distinct entity, given the aforementioned lung function test limitations [82]. The American Thoracic Society (ATS)/European Respiratory Society (ERS) technical standards advise against using a fixed FEV₁/FVC ratio <0.7 for airflow obstruction and an FEV₁ >80% pred value for normality [79]. However, GOLD continues to advocate for the fixed ratio. While the 80% predicted normality limit is not explicitly mentioned in the GOLD report, it remains common in clinical practice and research. The ATS/ERS technical standards recommend z-scores as a statistically robust methodology [83]. However, the GOLD report has not fully endorsed the use of z-scores, citing a lack of sufficient supporting evidence [75].

Establishing normal criteria for human performance measurements, including lung function, is challenging. More important than reference selection is the understanding that results should be interpreted cautiously, acknowledging the potential for misjudgement, regardless of the chosen reference and criteria. However, spirometry results are often applied too rigidly in COPD diagnosis and assessment, despite existing controversies.

Conclusion

The GOLD strategy played a pivotal role in shaping COPD management worldwide through sustained global collaboration. Proper COPD understanding and management are long-term endeavours that require reasonable and prudent approaches for future generations. For this significant disease, objective evaluation and critical analysis of controversial issues are essential.

Collaborative efforts that promote open discussion and constructive criticism are essential for advancing the understanding of COPD. Clear and distinct definitions of asthma and COPD, particularly chronic bronchitis, are critical prerequisites for refining treatment principles and strategies. In addition, clinical diagnostic criteria and lung function testing require thorough re-evaluation to improve the accuracy of COPD assessment and optimise patient management.