GOLD Science Committee recommendations for the use of pre- and post-bronchodilator spirometry for the diagnosis of COPD

Abstract

The Global Initiative for Chronic Obstructive Lung Disease (GOLD) report states that the diagnosis of COPD should be considered in individuals with chronic respiratory symptoms and/or exposure to risk factors. Forced spirometry demonstrating airflow obstruction after bronchodilation is required to confirm the diagnosis using a threshold of forced expiratory volume in 1 s (FEV₁)/forced vital capacity (FVC) ratio <0.7. This GOLD Science Committee review weighs the evidence for using pre- or post-bronchodilator (BD) spirometry to diagnose COPD. Cohort studies have shown that pre- and post-BD spirometry give concordant diagnostic results in most cases, although the prevalence of COPD is up to 36% lower with post-BD values. Discordant results may occur in “volume” or “flow” responders. Volume responders have reduced FVC due to gas trapping causing FEV₁/FVC ≥0.7 pre-BD, but a volume response occurs post-BD with a greater improvement in FVC relative to FEV₁ decreasing the ratio to <0.7. Flow responders show a greater FEV₁ improvement relative to FVC which may increase FEV₁/FVC from <0.7 pre-BD to ≥0.7 post-BD; these individuals have an increased likelihood of developing post-BD obstruction during follow-up and require monitoring longitudinally. GOLD 2025 recommends using pre-BD spirometry to rule out COPD and post-BD measurements to confirm the diagnosis. This will reduce clinical workload. Post-BD results close to the threshold should be repeated to ensure a correct diagnosis is made. Post-BD measurements ensure that volume responders are not overlooked and limit COPD overdiagnosis.

Shareable abstract

Pre-bronchodilator spirometry can exclude COPD, while post-bronchodilator measurement is needed to diagnose COPD https://bit.ly/4eKAbx8

Introduction

The Global Initiative for Chronic Obstructive Lung Disease (GOLD) report states that the diagnosis of COPD should be considered in individuals with chronic respiratory symptoms and/or exposure to risk factors, such as cigarette or biomass smoke exposure [1]. According to the GOLD report, forced spirometry demonstrating airflow obstruction after bronchodilation is required to confirm the diagnosis of COPD. The combination of clinical evaluation and spirometry reduces the potential for COPD overdiagnosis and overtreatment, which may occur if either step is omitted.

The threshold value which GOLD recommends to define airflow obstruction is a post-bronchodilator (BD) forced expiratory volume in 1 s (FEV₁)/forced vital capacity (FVC) ratio <0.7 [1]. This criterion was established in the first GOLD report in 2001 and, although regularly reconsidered, GOLD has maintained the recommendation [2, 3]. The concept that airflow obstruction in COPD is persistent and cannot be fully reversed provided the rationale for this recommendation, with post-BD FEV₁/FVC <0.7 ensuring that the diagnosis of COPD was confined to individuals with persistent airflow obstruction despite BD treatment [4]. The magnitude of BD responsiveness (BDR) was also believed to enable differentiation of COPD from asthma, although we now know that it has poor discriminative properties [5]. Other reasons why GOLD originally favoured post- over pre-BD measurements included the perception of greater reproducibility (although there is little evidence to support this), that post-BD values provide superior prognostic prediction, and that a pre-BD FEV₁/FVC ≥0.7 may be observed in COPD patients with significant gas trapping since an increased residual volume (RV) decreases inspiratory capacity (IC) and FVC leading to an increased FEV₁/FVC ratio [6, 7]. In such patients with significant gas trapping, the administration of a BD can reduce gas trapping and increase FVC (a volume response), sometimes to a greater extent than the effect on FEV₁, with the effect of reducing the FEV₁/FVC ratio. The use of post-BD values in clinical trials facilitated widespread acceptance of this measurement for COPD diagnosis.

Spirometry is underutilised to diagnose COPD in clinical practice [8–10], with practical implementation and training issues contributing to low uptake. More widespread use must be encouraged. Studies in England, Wales, Denmark, Sweden, Korea and the USA have shown that spirometry had not been performed in 40–50% of patients diagnosed with COPD [8, 11–15] and post-BD measurements had not been performed in 25–50% of the cases with spirometry [8, 11]. The data from England also show that the proportion of patients who had spirometry at diagnosis has declined in recent years [12]. Even when spirometry has been performed there is evidence of incorrect interpretation resulting in significant misdiagnosis, e.g. 25% patients diagnosed with COPD in Wales having incompatible post-BD spirometry [11]. Compared to pre-BD spirometry, the addition of post-BD measurements is more time consuming and complicated to perform. This may impede efficient outpatient operations and contribute to suboptimal uptake in clinical practice.

Given these issues, it is important to ensure that the current recommendations for diagnostic spirometry are both supported by evidence and are as practical as possible. The GOLD Science Committee has conducted a re-evaluation of the evidence and rationale for using pre- or post-BD spirometry for the diagnosis of COPD. Here we discuss the evidence and describe changes to the GOLD 2025 report arising from the review. Of note, the Global Lung Function Initiative (GLI) has established reference ranges using pre-BD values in non-smokers, with age-corrected lower limit of normal (LLN) values used to define abnormal lung function [16]. There has been considerable debate concerning the merits of using a fixed ratio (0.7) versus LLN for the diagnosis of COPD [16, 17]. This GOLD Science Committee review focused on the use of pre- or post-BD measurements in the diagnostic process and the issues concerning fixed ratio versus LLN or the use of appropriate reference values were not within the current scope.

Effects of bronchodilators on lung physiology in COPD

The hallmark pathophysiological features of COPD include small airway luminal narrowing, caused by inflammation and remodelling, and small airway collapse due to the loss of supporting alveolar attachments [18, 19]. There is also destruction of peripheral airways and parenchyma, with the latter reducing the elastic recoil of the lungs during expiration [6]. Collectively, these features cause increased small airway resistance, reduced expiratory flow and gas trapping on expiration (figure 1) [6]. Expiratory flow limitation with small airway collapse during tidal breathing is common in COPD [20–23]. Increased airway resistance coupled with reduced elastic recoil prevents effort-related increases in expiratory flow rate. Therefore, gas trapping worsens when exercising as there is a hyperventilation-related reduction in expiratory time, causing an increase in operational lung volumes (dynamic lung hyperinflation) and work of breathing which is responsible for dyspnoea on exertion [24].

FIGURE 1.

Pathological changes in COPD which may cause flow or volume responses to a bronchodilator (BD).

FEV₁ reflects the expiratory flow rate, while, in contrast, FVC is not time constrained and represents the total volume expired. Lung destruction and gas trapping in COPD both reduce FVC, although there is a greater relative impairment of FEV₁ as this flow measurement is highly dependent on increased airway resistance caused by airway narrowing. BDs act by relaxing airway smooth muscle in the proximal and distal airways, which increases the airway luminal diameter leading to a reduction in airway resistance [25, 26]. Increases in FEV₁ or FVC after administration of an inhaled BD are often referred to as flow (rate) or volume responses, respectively (figure 1).

The change in FEV₁ after administration of a BD varies considerably between COPD patients and even within the same patient from test to test [27, 28]. In the Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) study (n=1831), the mean±sd FEV₁ increase in response to salbutamol was lower in more severe COPD patients: 160±170, 100±130 and 50±80 mL in GOLD grade 2, 3 and 4 patients, respectively [29]. A smaller FEV₁ response in COPD patients with lower pre-BD FEV₁ has also been observed when reported as percentage change [28], indicating that the smaller flow responses are not simply explained by lower lung volumes. Both the ECLIPSE study and the National Emphysema Treatment Trial (NETT) showed a smaller FEV₁ increase after BD in COPD patients with more severe emphysema [27, 29]. The presence of expiratory flow limitation due to small airway collapse can explain these observations, as this limits improvements in flow responses after administration of a BD [30].

BDs can improve airway physiology and symptoms even when there is little or no flow response measured by FEV₁ [7]. BDs reduce end-expiratory lung volumes by allowing more volume to be exhaled, thereby improving IC [7]. These shifts in lung volumes can arise from small improvements in flow enabling larger and clinically relevant improvements in volumes, and reduced end-expiratory lung volumes improve the mechanical efficiency of diaphragm contraction thereby decreasing the work of breathing [31]. This leads to downstream effects on exercise tolerance particularly in patients reaching the limits of the ventilatory system during exercise [32]. BDs also improve the mechanical functioning of the small airways, reducing heterogeneity and preventing or delaying closure thus facilitating volumetric responses [33].

Clinical studies have demonstrated considerable interindividual variation in flow and volume responses in COPD patients, with volume responses being associated with greater gas trapping measured pre-BD. Walker and Calverley [34] reported that flow and volume responses alone (using thresholds of 160 and 330 mL, respectively) were present in 12% and 23% of COPD patients, respectively, after administration of salbutamol (n=266), with 24% displaying both flow and volume responses and 41% displaying neither. Volume responders were characterised by lower FEV₁ and FVC and higher RV. Volume responders also had greater dynamic airway collapse, defined as higher ratio of peak expiratory flow/forced expiratory flow at 50% of FVC. Newton et al. [35] retrospectively analysed the lung function results of 957 individuals (aged >55 years) with hyperinflation defined as total lung capacity (TLC) >115% predicted. The administration of salbutamol caused greater changes in FVC and RV in individuals with severe compared to moderate hyperinflation (TLC >133% predicted and 115–133% predicted, respectively). Furthermore, there was no association between volume and flow (FEV₁) changes.

Collectively, the evidence shows that flow responses decrease in more severe COPD patients, and that volume responses may occur in the absence of flow responses [7, 29, 33, 35, 36]. The relative magnitude of the flow and volume response to a BD can change the FEV₁/FVC ratio; in the ECLIPSE study, GOLD grade 2 patients showed a post-BD increase in this ratio, due to the greater flow responses, while grade 4 patients showed a decrease due to greater volume responses [29].

Forced or slow vital capacity

GOLD and the American Thoracic Society (ATS)/European Respiratory Society (ERS) guidelines on COPD both defined airflow obstruction by relating the FEV₁ to the FVC [2, 37]. In contrast, the 2005 ATS/ERS Task Force on Standardisation of Lung Function Testing recommended that FEV₁ should be related to vital capacity (VC) rather than just FVC [38]. They recommended that the largest available VC was used, whether obtained on inspiration (IVC), slow expiration (SVC) or forced expiration (FVC). FVC may underestimate the VC (i.e. be lower than IVC or SVC) in the presence of increased collapsibility of the small airways [39, 40] and in these circumstances the FEV₁/FVC ratio may underestimate or fail to detect airflow obstruction. One study has shown that 20% of patients referred for hospital pulmonary function tests had a preserved FEV₁/FVC ratio but low FEV₁/SVC ratio [41]. These patients had lower mid-expiratory flows, higher airway resistance, worse gas trapping and many had been diagnosed with an obstructive airway disease by a pulmonologist. Discordant results were more common in patients aged <60 years and those with a body mass index >30 kg·m⁻². It has been suggested that a slow manoeuvre should be routinely performed in symptomatic subjects with preserved FEV₁/FVC ratio who are <60 years old and obese to identify mild and more peripheral airflow obstruction. Performing additional SVC manoeuvres adds time to the test and the GOLD 2025 report stands by its recommendation that airflow obstruction should be identified using the FEV₁/FVC ratio. However, if there is a strong clinical suspicion of COPD and the FEV₁/FVC ratio is normal, GOLD 2025 recommends further follow-up and investigations, which could include assessment of the FEV₁/SVC ratio. This recommendation is in line with the 2022 ATS/ERS technical standard on interpretation of routine lung function tests which acknowledges that the use of SVC is more sensitive but less specific and adds complexity to lung function testing [16].

Variability in FEV₁/FVC results can arise due to variable expiratory times influencing the FVC values obtained. Using forced expiratory volume in 6 s (FEV₆) rather than FVC has been suggested as a means of overcoming variability in the VC. Measurement of FEV₁/FEV₆ has lower sensitivity for detecting airflow obstruction than FEV₁/FVC but similar specificity [42], and its test–retest intraclass correlation coefficient is not significantly better [43].

Bronchodilator responsiveness

A positive acute BDR test was historically advocated to differentiate asthma from COPD, with a positive response favouring the former. However, many COPD patients demonstrate excellent flow and/or volume responses signalling the poor discriminative properties of BDR testing [5]. The usefulness of BDR testing in clinical practice is hampered by practical issues such as non-standardisation of which BD should be used and at what doses, as well as variable criteria to define a positive response. In patients with an established diagnosis of COPD, there is no consistent evidence that BDR is associated with clinical characteristics or prognosis. In fact, the evidence suggests it is the post-BD value achieved that is of relevance rather than the magnitude of change in response to the BD [7, 44].

The Subpopulations and Intermediate Outcome Measures in COPD Study (SPIROMICS) reported outcomes over 4 years in 1481 smokers who did not meet spirometric criteria for the diagnosis of COPD [45]. BDR was defined using a threshold of 12% and 200 mL (compared to baseline pre-BD value) to define flow (FEV₁) or volume (FVC) responders. Individuals with consistent BDR (at every visit) had the lowest pre-BD FEV₁ (61% predicted) and FEV₁/FVC ratio, while these lung function parameters were also lower in individuals with inconsistent BDR responses versus those never showing a BDR response. Computed tomography (CT) scanning showed that BDR was associated with gas trapping and hence greater small airway dysfunction after adjusting for the post-BD FEV₁, irrespective of whether FEV₁ or FVC was used to define BDR. BDR defined using either FEV₁ or FVC was associated with greater progression to a diagnosis of COPD after adjusting for the post-BD FEV₁. In the context of diagnostic testing for COPD, these findings suggest that the magnitude of BDR might offer clinically useful information in smokers with post-BD FEV₁/FVC ≥0.7, as those individuals with greater and more consistent BDR are more likely to progress to COPD. Additionally, some of these smokers would be labelled as preserved ratio impaired spirometry (PRISm), which is defined as low FEV₁ values but normal FEV₁/FVC ratio post-BD; individuals with PRISm sometimes (but not always) progress to develop COPD [46, 47], particularly in individuals who continue smoking or have FEV₁/FVC values close to 0.7.

Pre- versus post-bronchodilator: COPD prevalence and clinical outcomes

Studies have consistently shown that the prevalence of COPD is lower when using post-BD spirometry (table 1). The Lung Health Study enrolled current smokers with FEV₁/FVC <0.7 and FEV₁ 55–90% predicted [48]. The lung function results after 1 year (n=5307) showed a prevalence of airflow obstruction (FEV₁/FVC <0.7) that was 16.1% lower using post- versus pre-BD spirometry (77.7% and 91.4%, respectively). The National Health and Nutrition Examination Survey (NHANES) 2007–2010 conducted spirometry in a US adult population (n=5477), with approximately half being never-smokers [49]. The prevalence of obstructed spirometry (FEV₁/FVC <0.7) was ∼33% lower when post-BD values were used compared to pre-BD values. In the Latin American PLATINO study (n=5183), the prevalence of airflow obstruction was 35% lower using post-BD values compared to pre-BD values, decreasing from 21.7% to 14% [50]. In high-risk individuals, defined as having either respiratory symptoms and/or significant environmental exposure, the prevalence was 33.6% lower when using post-BD values (17.4% versus 26.2%). A Norwegian study (n=2235) showed 27.1% lower prevalence in a population of smokers and never-smokers, and 31% lower in current smokers with a post-BD prevalence of 8.5% [51]. These large studies recruited diverse populations with different smoking histories, but all showed a consistent pattern of lower rates of obstructed spirometry post- compared to pre-BD, with the reduction up to 33.6% in current and ex-smokers.

TABLE 1.

Prevalence of COPD using pre- or post-bronchodilator (BD) spirometry in cohort studies

First author, year [ref.]	Sample size (n)	Age (years)	Smoking criteria	Smokers# (%)	Prevalence (%)¶		Reduction pre- versus post-BD (%)+	Smokers only: reduction pre- versus post-BD (%)+
					Pre-BD	Post-BD
Mannino, 2011 [48] Lung Health Study	5307	35–65	Current smokers§	100	91.4	77.7	16.1	16.1
Tilert, 2013 [49] NHANES	5477ƒ	40–79		50.6	20.9	14.0	33.0	NC
Pérez-Padilla, 2007 [50] PLATINO	5183	>40		63.5##	21.7	14	35.5	33.6
Johannessen, 2005 [51]	2225	15–70		61	9.6	7	27.1	20.3
Fortis, 2017 [52] COPDGene	10 000	45–80	>10 pack-years	100	50.2	44.5	11.3	11.3

FEV₁: forced expiratory volume in 1 s; FVC: forced vital capacity; NC: not calculated (data not available). ^#: includes current and former smokers; ^¶: prevalence of FEV₁/FVC <0.7; ⁺: reduction (%) pre- versus post-BD calculated by 100×(pre-BD−post-BD)/pre-BD; ^§: also required to have post-BD FEV₁/FVC <0.7 at study visit 1 year before; ^ƒ: post-BD conducted in 564 individuals; ^##: includes significant exposure history (to cigarette smoke, biomass or occupational dusts) or respiratory symptoms.

The COPDGene study was conducted in the USA, enrolling individuals with a smoking history >10 pack-years [52]. Pre- and post-BD obstruction defined by FEV₁/FVC <0.7 were termed PREO and POSTO, respectively, while no obstruction (hence normal) was termed PREN and POSTN, respectively. Figure 2 shows that concordant results were obtained in the majority of the cohort (n=10 000), either PREO–POSTO (n=4150) or PREN–POSTN (n=4683). The discordant groups comprised 11.7% of the cohort, PREO–POSTN (n=866) or PREN–POSTO (n=301), with lower post-BD FEV₁ in both groups compared to PREN–POSTN. PREO–POSTN was characterised by greater flow (FEV₁; 7.7%) than volume (FVC; −2.1%) responses, while the opposite occurred in PREN–POSTO (1.5% and 17.7%, respectively). PREN–POSTO individuals (volume responders) had greater dyspnoea and reduced exercise capacity compared to PREO–POSTN, although this finding disappeared in modelling analysis which adjusted for FEV₁. These findings confirm that using post-BD spirometry results in fewer diagnosed COPD cases, while also ensuring that the discordant PREN–POSTO individuals characterised by greater volume responses with more severe disease characteristics are not overlooked.

FIGURE 2.

Discordant diagnostic spirometry in COPDGene [52] and SPIROMICS [53] cohorts. COPDGene participants had >10 pack-year smoking history, while SPIROMICS participants had ≥20 pack-year smoking history. Change calculated using pre-bronchodilator (BD) value. FEV₁: forced expiratory volume in 1 s; FVC: forced vital capacity; PREO: pre-BD obstruction; PREN: pre-BD no obstruction; POSTO: post-BD obstruction; POSTN: post-BD no obstruction.

In COPDGene, both PREO and POSTO measurements were associated with clinical characteristics and outcomes including dyspnoea, exercise capacity, CT measurements of emphysema and gas trapping, exacerbations and mortality, although post-BD measurements performed marginally better in some models including mortality. Similar numbers of PREN–POSTO and PREO–POSTN developed PREO–POSTO after 5 years of follow-up (nearly 50% in both groups).

A similar analysis was performed in the SPIROMICS study (figure 2); 603 smokers were identified without pre- or post-BD airflow obstruction (PREN–POSTN) and 175 smokers with pre-BD obstruction which resolved post-BD and was equivalent to the PREO–POSTN group in COPDGene [53]. PREO–POSTN had lower FEV₁ predicted compared to PREN–POSTN and increased flow but not volume response to BD. The PREO–POSTN group was also characterised by greater small airway dysfunction and emphysema quantified by CT scan. During longitudinal follow-up (mean 7 years) there was a 6.2-fold increased risk of developing COPD (defined by POSTO) in the PREO–POSTN versus PREN–POSTN group (61% versus 14%, respectively). Overall, these results mirror COPDGene for the PREO–POSTN group with regard to increased flow responses and increased future risk of post-BD obstruction.

The COPDGene analysis showed no or only minor differences between pre- and post-BD measurements in terms of prognosis. The Lung Health Study showed that both measurements offered similar prediction for mortality [48], while a smaller study of COPD patients in China (n=300) favoured post-BD measurements [54].

GOLD recommendations

Figure 3 shows the GOLD 2025 recommendations for the use of spirometry in the diagnosis of COPD. A single normal pre-BD measurement can be used to exclude COPD, but post-BD measurements are required to confirm the diagnosis of COPD and demonstrate without doubt that the individual has persistent airflow obstruction.

FIGURE 3.

Global Initiative for Chronic Obstructive Lung Disease (GOLD) diagnostic pathway for spirometry using forced expiratory volume in 1 s (FEV₁)/forced vital capacity (FVC) ratio. BD: bronchodilator.

A recommendation to use a diagnostic test should be based primarily on sensitivity and specificity, but also consider issues such as data complexity, practicality and cost. For the diagnosis of COPD, pre-BD spirometry detects more people with airflow obstruction so can be viewed as more sensitive overall. However, a significant proportion of patients with pre-BD airflow obstruction are not obstructed using post-BD values (17% of the patients in COPDGene). In parallel, pre-BD measurements miss a small number of cases (∼3% of the screened population in the COPDGene study [52]). These missed cases had gas trapping and were volume responders requiring post-BD measurements to reveal an obstructed ratio. The discordant groups in the COPDGene study, ∼11% of the screened population, represent individuals who would change diagnosis if one measurement was used rather than another. Mirroring this figure, almost 90% of measurements were concordant, showing that both measurements give similar diagnostic outcomes in most cases.

Large cohort studies have shown no or only minor differences between pre- and post-BD measurements in terms of prognosis [48, 52]. The main reason to favour post- over pre-BD measurements is to ensure that volume responders are not overlooked in the diagnostic process and to limit COPD overdiagnosis, which is estimated at ∼11–35% from various studies in smokers (table 1) [29, 48, 49, 51, 52]. COPD overdiagnosis inappropriately increases the overall COPD population and burden on health resources.

While there are concerns that using FEV₁/FVC <0.7 may overdiagnose airflow obstruction in older individuals in comparison to the use of LLN [16], this is limited by the requirement to only consider COPD diagnosis in individuals with exposure risk factors and symptoms. The use of post-BD testing for COPD diagnosis further reduces the potential for overdiagnosis.

Most GOLD stage 1 individuals have post-BD FEV₁/FVC values close to the diagnostic threshold with 80% within the range 0.6–0.7, falling to <10% within this range for GOLD stage 3 and 4 [55]. The PREO–POSTN group in the COPDGene study demonstrated that relatively modest flow responses (FEV₁ change 7.7% after BD) changed the diagnostic conclusion in these potentially GOLD stage 1 individuals (post-BD FEV₁ 84.7% predicted). As nearly 50% of this PREO–POSTN group progressed to PREO–POSTO during follow-up, GOLD 2025 recommends these individuals (with PREO–POSTN) should be closely followed up, including repeat testing as clinically indicated with a suggested time frame of 3–6 months, given the high likelihood of the development of post-BD obstruction over time (figure 3) particularly in individuals who continue smoking or have FEV₁/FVC values close to 0.7 [47]. A larger BD response, in either FEV₁ or FVC, is also an indicator of greater risk for subsequent COPD development in individuals with post-BD FEV₁/FVC ≥0.7 [45].

It has been commented upon that removing a COPD diagnosis after a positive treatment response in flow responders (PREO–POSTN) is illogical [56]. The counter-argument (for using post-BD values for diagnostic purposes) is that volume responders (PREN–POSTO) with more severe disease characteristics are diagnosed [35, 52]. Also, the current GOLD 2025 recommendation to closely follow-up individuals with PREO–POSTN ensures that individuals with progressive airflow obstruction are not missed in the diagnostic process.

The GOLD 2025 report recommends that the presence or absence of airflow obstruction should be confirmed by repeat spirometry on a separate occasion if the post-BD FEV₁/FVC ratio is between 0.60 and 0.80 as in some cases the ratio may change as a result of biological variation when measured at a later interval [57, 58]. If the initial post-BD FEV₁/FVC ratio is <0.60 it is very unlikely to rise spontaneously above 0.7 [57]. Repeat spirometry should be guided by clinical status, with a suggested time frame of 3–6 months.

ECLIPSE and other studies have demonstrated that flow responses decrease in more severe COPD, where volume responses can drive a fall in post-BD FEV₁/FVC ratio [29]. This indicates that the potential importance of conducting post-BD spirometry is greater in individuals with lower baseline FEV₁, e.g. <80% predicted. The GOLD 2025 report states that pre-BD spirometry provides useful information as an initial test to investigate airflow obstruction in symptomatic patients. If pre-BD spirometry does not show obstruction, then performing post-BD spirometry is not required in the majority of cases. The exception to this is when the clinical suspicion of COPD is high. Importantly, the pre-BD FEV₁ can be used as a guide to the probability that the individual may be a volume responder, as this is more likely with lower FEV₁ measurements, e.g. <80% predicted.

The ATS/ERS technical statement on the standardisation of spirometry contains a grading system for the quality of the test session to provide a level of confidence that the results represent the best performance of the individual [59]. Ideally, grade A results which require at least three acceptable measurements within repeatability criteria would be obtained. The guidance states that “some maneuvers may be usable at grading levels lower than A”, which includes grade E where only one acceptable test is achieved or even grade U with at least one usable but not acceptable measurement. We agree with the ATS/ERS recommendation that even grade E or U readings can be used for diagnostic purposes when carefully considered alongside clinical information, particularly when used to rule out the diagnosis of COPD. In such circumstances, it is preferable to repeat spirometry to confirm initial findings.

The increasing use of CT scanning, for lung cancer screening and for assessment of respiratory symptoms or abnormal spirometry findings, can provide information relevant to the diagnosis of COPD such as the presence of emphysema. However, the availability and cost of CT scanning currently preclude this from being routinely used to diagnose COPD. Finally, most of the studies that inform current management recommendations have used post-BD values as inclusion criteria. The results of these studies cannot be assumed to apply in a population defined by pre-BD values.

Conclusion

Spirometry remains essential to confirm the presence of airflow obstruction and, therefore, to diagnose COPD. The GOLD Science Committee concludes that using pre-BD values alone would lead to a significant increase in the number of patients labelled as having COPD, putting additional pressures on health services and payors. Post-BD measurements remain the optimum diagnostic methodology for the reasons discussed above. The GOLD 2025 report recognises that measuring post-BD values is more time consuming, so GOLD advocates the use of pre-BD spirometry to rule out COPD and post-BD measurements to confirm the diagnosis. This will reduce workload in practice.

Key take-home messages regarding pre- and post-BD spirometry are shown in table 2. There are clinical situations where the exposure history, burden of symptoms and information from other tests (such as CT scans) raise a high suspicion of COPD but the diagnostic strategy outlined here fails to diagnose COPD. The GOLD 2025 report recommends further follow-up and investigations including repeating the spirometry after an interval in these individuals.

TABLE 2.

Take-home messages

Pre-BD FEV1/FVC

≥0.7 rules out COPD in most cases; no further testing needed

Post-BD FEV1/FVC

Perform post-BD spirometry to confirm diagnosis if pre-BD FEV1/FVC <0.7

Volume responders

Gas trapping can reduce FVC; BD can increase FVC

Volume responders can have discordant pre- (≥0.7) and post- (<0.7) BD ratios

Flow responders

Flow responses can increase FEV1/FVC ratio from <0.7 (pre-BD) to ≥0.7 (post-BD)

Increased chance of developing COPD

Follow-up with repeat testing

BD: bronchodilator; FEV₁: forced expiratory volume in 1 s; FVC: forced vital capacity.

Shareable PDF

This PDF extract can be shared freely online.

Shareable PDF ERJ-01603-2024.Shareable ^{(587.2KB, pdf)}