Background

Asthma and chronic obstructive pulmonary disease (COPD) are heterogeneous lung diseases [1], and they can coexist in some given patients, namely asthma–COPD overlap (ACO). Although some consensus documents raise different criteria for diagnosing ACO, there is still a lack of widely accepted and simplified criteria [2,3,4]. The European Consensus for ACO in 2016 is one of the most recognized criteria [4]. The prevalence of ACO in asthma and COPD is similar, ranging from 20 to 30% [5]. ACO has a higher symptom burden and more frequent and severe exacerbations than asthma or COPD [6, 7]. Additionally, the treatment of ACO is different from that of COPD; patients with ACO are recommended to use inhaled corticosteroids (ICS) combined with inhaled bronchodilators [8, 9] and might benefit from biologics used in patients with severe asthma [10, 11]. Therefore, it is of great importance to determine a simplified and accurate method for differentiating ACO from COPD.

The bronchodilation test (BDT), which evaluates airway reversibility, is not only an important diagnostic base of asthma, but it also plays a critical role in differentiating between ACO and COPD [12]. However, nearly one-third of patients with COPD have a positive BDT result [13,14,15]. The Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2022 [16] also proposed that COPD alone often shows a positive BDT result when the baseline forced expiratory volume in 1 s (FEV1) is poor. As a major index for diagnosis of ACO according to the guidelines and consensus [4, 17,18,19], a strongly positive BDT result can be considered to differentiate ACO from COPD. Currently, there are three major criteria for a strongly positive BDT result: standard I [20,21,22], post-bronchodilator forced expiratory volume in 1 s response (ΔFEV1) > 400 mL + 15%; standard II [23], ΔFEV1 > 400 mL; and standard III [24], ΔFEV1 > 200 mL + 15%. However, it is unclear which one is more suitable in diagnosing ACO. Although the positive BDT result alone is limited in differentiating between asthma, COPD, and ACO, COPD has a lower ΔFEV1 (in mL) than asthma and ACO, and bronchodilator response (BDR) was helpful in the early screening of ACO [25]. Besides, BDR was vital in identifying different phenotypes of ACO [26]. Therefore, it is necessary to further explore the role of BDR in differentiating between asthma, COPD, and ACO in patients with a positive BDT result and determine whether there is a better predictive threshold.

The inflammatory biomarker fractional exhaled nitric oxide (FeNO) has been reported as an indicator for differentiating between ACO and COPD; one study reported its optimal cut-off value as 39.5 parts per billion (ppb) (sensitivity, 58.3%; specificity, 84.9%) [27], whereas another study reported its optimal cut-off value as 25.0 ppb (sensitivity, 60.6%; specificity, 87.7%) [28]. The two aforementioned studies showed a low sensitivity, indicating that FeNO alone is useful but difficult to use for differentiating between ACO and COPD. Recently, Wang reported that ΔFEV1 and FeNO were significantly different in ACO compared with COPD alone, which indicated that BDR combined with FeNO may be helpful in the early screening of ACO [25]. However, there is still a lack of an optimal value of ΔFEV1 for differentiating between asthma, COPD, and ACO in patients with a positive BDT result. It needs to be clarified whether ΔFEV1 combined with FeNO have advantages in differentiating between asthma, COPD, and ACO.

This study included patients with a positive BDT result to explore a simplified standard for a strongly positive BDT result and the value of ΔFEV1 in BDT and FeNO for differentiating between asthma, COPD, and ACO. We assumed that BDR in FEV1 and FeNO are helpful in differentiating between asthma, COPD, and ACO in patients with a positive BDT result, which contributes to the early screening of ACO.

Methods

Study design and patients

To explore a simplified standard for a strongly positive BDT result and the value of ΔFEV1, this cross-sectional study included patients admitted to our hospital’s outpatient respiratory clinic from January 2019 to January 2021. The participants were diagnosed with asthma, COPD, or ACO. The diagnosis of asthma was defined by the Global Initiative for Asthma (GINA) guidelines [29], requiring: (1) a history of asthmatic symptoms (wheezing, shortness of breath, with or without chest tightness or cough) relieved spontaneously or by medication; (2) variable expiratory airflow limitation (BDR of FEV1 > 200 mL and 12%). The diagnosis of COPD was based on the GOLD guidelines, indicating a post-bronchodilator FEV1/forced vital capacity (FVC) < 0.70 [30]. The diagnosis of ACO was based on the European Consensus 2016 criteria, major criteria [4]: (1) post-bronchodilator FEV1/FVC < 0.70 in individuals > 40 years old; (2) at least 10 pack-years of tobacco smoking or equivalent exposure history; (3) history of asthma before age of 40 years or BDR of FEV1 > 400 mL. Minor criteria included: (1) history of atopy or allergic rhinitis; (2) positive BDT result on two or more visits; (3) blood eosinophil count (BEC) ≥ 300 cells/μL. Patients who met all three major criteria and at least one minor criterion were diagnosed with ACO. Patients included were newly diagnosed and without inhalant treatment, or those who did not receive inhalant treatment for at least 4 weeks prior to enrollment, including long-acting muscarinic antagonist (LAMA), long-acting beta-2 agonist (LABA) orICS. All patients had a positive BDT result. The exclusion criteria were as follows: (1) acute attack of respiratory system; (2) active pulmonary tuberculosis, interstitial pneumonia, fungal infection and lung tumors; (3) refuse to sign informed consent. This study was approved by the ethics committee (Code No. NFEC-2021-142).

Clinical information from electronic medical records, including demographic data, spirometry data, BDT result, FeNO value, BEC, and percentage were collected. To verify the results, a validation study included patients who were admitted to the respiratory clinic from June 2021 to December 2022.

Definition of the study groups

Three criteria were used to define a strongly positive BDT result in accordance with the GINA, National Institute for Health and Clinical Excellence, and American Thoracic Society (ATS) guidelines: standard I, ΔFEV1 > 400 mL + 15%; standard II, ΔFEV1 > 400 mL; and standard III, ΔFEV1 > 200 mL + 15%. Patients were grouped according to whether they had a strongly positive BDT result.

Spirometry, BDT, FeNO, and BEC

Spirometry was strictly measured by spirometers (Jaeger MasterScreen, Germany) with reference to the ATS criterion. A positive BDT result was defined as follows: ΔFEV1 ≥ 200 mL + 12% after inhaling 400 μg of salbutamol. The FeNO detection was measured by a NIOX VERO analyzer (Aerocrine AB, Solna, Sweden) with reference to the ATS/European Respiratory Society criterion. The count and percentage of blood eosinophil was read by the automatic hematology analyzer.

Statistical analysis

Statistical analysis was performed using SPSS statistics for Windows, version 24.0 (IBM Corp.). Data are presented as mean ± standard deviation for continuous variables and as median (first quartile, third quartile) for categorical variables. Comparisons between continuous variables were performed using the Student’s t-test or Mann–Whitney U test; the Chi-square test was used to analyze categorical variables. The factors of ΔFEV1 ≥ 400 mL in ACO patients were analyzed using the COX regression model. All variables detected in the univariate analyses (with a P-value less than 0.1) were included in the multivariate analysis. Predictive values of single or combined measurements were calculated by constructing receiver operating characteristic (ROC) curves and measuring areas under the curve (AUCs). A two-sided P-value < 0.05 was considered significant.

Results

Study participants

A total of 633 patients were enrolled from the hospital’s outpatient respiratory clinic. Finally, only 397 patients were eligible, including 192 (48.4%) with asthma, 135 (34.0%) with COPD, and 70 (17.6%) with ACO. The study flow chart is shown in Fig. 1.

Fig. 1
figure 1

Study flowchart. BDT, bronchodilation test; FeNO, fractional exhaled nitric oxide; GINA, Global Initiative for Asthma; GOLD, Global Initiative for Chronic Obstructive Lung Disease; COPD, chronic obstructive pulmonary disease; ACO, asthma–chronic obstructive pulmonary disease overlap

Full size image

Demographic and clinical characteristics

The clinical characteristics of the participants are shown in Table 1. Patients with asthma were younger, female-dominant, had lower smoking pack-years, and better spirometric indices (including FEV1, FVC, %FEV1, %FVC, and FEV1/FVC) than those with COPD or ACO. FEV1, FVC, %FEV1, and FEV1/FVC values of patients with COPD were lower than those of patients with ACO. Patients with ACO and asthma had a higher BDR in ΔFEV1 (mL) than those with COPD; the BDR in ΔFEV1 was highest in patients with ACO. Patients with asthma had a lower ΔFVC (mL) value than those with COPD and ACO; however, there was no difference in ΔFVC (mL) between patients with COPD and those with ACO. FeNO levels were lower in patients with COPD than in other patients, but there was no difference in FeNO levels between patients with asthma and those with ACO. The BEC was higher in patients with asthma than in those with COPD; however, there was no significant difference between patients with ACO and those with asthma or COPD. There was also no statistical difference in blood eosinophil percentage between patients with asthma, ACO, and COPD.

Table 1 Patient characteristics
Full size table

Difference analysis of a strongly positive BDT rate under different standards

The strongly positive BDT rates in patients with asthma, COPD, and ACO under different standards are shown in Fig. 2 and Additional file 1. In the asthma group, 66 (34.4%), 74 (38.5%), and 135 (70.3%) patients had a strongly positive BDT rate under standards I, II, and III, respectively; those respective values were 4 (3.0%), 4 (3.0%), and 103 (76.3%) in the COPD group and 40 (57.1%), 42 (60.0%), and 61 (87.1%) in the ACO group. Under standards I and II, the ACO group had the highest strongly positive BDT rate, followed by the asthma group, and there were statistical differences between the three diseases (P < 0.05). Under standard III, the ACO group had a higher strongly positive BDT rate than the asthma group (P = 0.005), but there was no statistical difference between the COPD group and the other two groups. All three diseases had a higher strongly positive BDT rate in standard III than the other two standards (P < 0.001). However, there was no statistical difference in the strongly positive BDT rate between standards I and II among the three diseases (asthma, COPD, and ACO: P = 0.396, P = 1.000, and P = 0.731, respectively).

Fig. 2
figure 2

Strongly positive BDT rates in the asthma, COPD, and ACO groups under three different standards. COPD, chronic obstructive pulmonary disease; ACO, asthma–chronic obstructive pulmonary disease overlap; BDT, bronchodilation test

Full size image

Comparison of clinical characteristics under different standards

The comparison of clinical characteristics of asthma, COPD, and ACO under standards I, II, and III is shown in Additional files 2, 3 and 4, respectively. The clinical characteristics between the two groups with a strongly positive BDT result or not were consistent between standards I and II; however, clinical characteristics in standard III were different from those in standard I or II. Under standards I and II, asthma patients with a strongly positive BDT result were younger, more male-dominant, and had higher smoking pack-years and higher FEV1 and FVC values than those with COPD or ACO. Under any standard, there was no between-group difference in FeNO levels. However, under standard III, all patients with a strongly positive BDT result had poor pulmonary function, including FEV1, FVC, %FEV1, %FVC, and FEV1/FVC.

Factors associated with ΔFEV1 ≥ 400 mL in ACO patients

The univariate and multivariate analysis with ΔFEV1 ≥ 400 mL in ACO patients is shown in Table 2. In the univariate Cox regression analysis, only FVC and ICS/LABA/LAMA (yes vs. no) were an independent predictor of ΔFEV1 ≥ 400 mL in ACO patients [FVC: HR = 1.97, 95% CI 1.04–3.71, P = 0.037; ICS/LABA/LAMA (yes vs. no): HR = 0.12, 95% CI 0.01–0.94, P = 0.044)]. The multivariate Cox regression analysis found that FVC was significantly correlated with ΔFEV1 ≥ 400 mL in ACO patients (HR = 2.71, 95% CI 1.31–5.63, P = 0.007). And the inhalation therapy of ICS/LABA/LAMA (yes vs. no) was also correlated with ΔFEV1 ≥ 400 mL in ACO patients (P = 0.013).

Table 2 Univariate and multivariate associations with ΔFEV1 ≥ 400 mL in ACO patients
Full size table

Predictive value of ΔFEV1 alone or combined with FeNO for the diagnosis of ACO or asthma in patients with a positive BDT result

The predictive value of ΔFEV1 alone or combined with FeNO was evaluated using ROC curves, adjusted by covariates. Only ΔFEV1 could predict the diagnosis of ACO in COPD patients with a positive BDT result, with a cut-off value of 345 mL (AUC: 0.881; 95% CI 0.83–0.94) (Fig. 3). Table 3 shows the sensitivity, specificity, positive predictive value, negative predictive value, and Youden index of each cut-off value for ΔFEV1.

Fig. 3
figure 3

ROC curves for ΔFEV1 in predicting ACO diagnosis for COPD with a positive BDT result. ROC, receiver operating characteristic; ΔFEV1, post-bronchodilator forced expiratory volume in 1 s response; COPD, chronic obstructive pulmonary disease; ACO, asthma–chronic obstructive pulmonary disease overlap; BDT, bronchodilation test

Full size image
Table 3 Predictive values for predicting ACO diagnosis from COPD in patients with a positive BDT result
Full size table

After excluding patients with ACO, ΔFEV1 (AUC: 0.613; 95% CI 0.55–0.68) and FeNO (AUC: 0.765; 95% CI 0.71–0.82) could predict the diagnosis of asthma in patients with a positive BDT result (Table 4). The AUC for ΔFEV1 combined with FeNO was 0.774 (95% CI 0.72–0.83), which was significantly higher than that of ΔFEV1 or FeNO alone; the cut-off values for ΔFEV1 and FeNO were 315 mL and 28.5 ppb, respectively (Table 4 and Fig. 4).

Table 4 Predictive values for predicting asthma diagnosis in patients with a positive BDT result
Full size table
Fig. 4
figure 4

ROC curves for the model of ΔFEV1 combined with FeNO in predicting the diagnosis of asthma. ROC, receiver operating characteristic; ΔFEV1, post-bronchodilator forced expiratory volume in 1 s response; FeNO, fractional exhaled nitric oxide; BDT, bronchodilation test. \({\text{AUC}}_{{\text{FeNO}}+\Delta{\text{FEV}}_1} = 0.774\) (95% CI 0.72–0.83); AUCFeNO = 0.613 (95% CI 0.55–0.68); \({\text{AUC}}_{\Delta{\text{FEV}}_1} = 0.765\) (95% CI 0.71–0.82)

Full size image

Validation study

To verify the reliability of the prediction model, we continued to recruit 209 patients from June 2021 to December 2022, including 132 with asthma, 57 with COPD, and 20 with ACO. The clinical characteristics of the three diseases were broadly in line with our original study (Additional file 5). Surprisingly, ΔFEV1 ≥ 345 mL could predict the diagnosis of ACO in COPD patients with a positive BDT result, with a great sensitivity and specificity of 90.0% and 91.2%, respectively, in external validation (Additional file 6). Additionally, ΔFEV1 < 315 mL combined with FeNO < 28.5 ppb could eliminate asthma diagnosis in patients with a positive BDT result, with a high specificity of 87.0% but a low sensitivity of 54.2% (Additional file 7).

Discussion

This is the first study to compare three criteria of a strongly positive BDT result in chronic airway disease patients with a positive BDT result. This study demonstrated that standard II (ΔFEV1 > 400 mL) can effectively replace standard I (ΔFEV1 > 400 mL + 15%).

In this study, patients with COPD were older, male-dominant, smoker-dominant, and had poorer baseline lung function than patients with asthma, which is consistent with previous studies’ findings [31, 32]. Compared with patients with COPD, those with ACO were younger and had better baseline lung function. A previous study reported that patients with ACO were younger than those with COPD, but they had a lower FEV1 [7]. Herein, patients with ACO and those with asthma had a higher BDR (mL in FEV1) than those with COPD, which is in line with previous studies’ findings [32, 33]. We found a significant difference in BECs between patients with asthma and those with COPD, but there was no difference between patients with ACO and those with COPD or asthma. We thought the reason for this phenomenon was that the positive BDT result and exposure to tobacco smoke reduced the difference in BECs between ACO and COPD or asthma [34,35,36]. Peng reported that patients with ACO had higher BECs than those with COPD [37]. However, there was no difference in BECs among the asthma, COPD, and ACO groups in the real-world study cohort, NOVELTY [6]. Additionally, patients with asthma and those with ACO had higher FeNO levels than those with COPD, which is similar to previous studies’ results [38, 39]. Therefore, BDR combined with the biomarkers of type 2 airway inflammation may be a useful tool in distinguishing between COPD and ACO or asthma.

A large multicenter study [40] reported that in 1106 participants with low FEV1 values (mL), the FEV1 increased by 12–44.7% relative to the baseline but < 200 mL, and ΔFEV1% increased with the level of airflow obstruction but decreased with severe obstruction, indicating that patients with severe obstruction rarely meet standard I. Thus, our result that standard II can replace standard I is clinically significant. However, there is still a lack of relevant research on the specific value of BDR in distinguishing between ACO and COPD with a positive BDT result.

Here, we firstly showed that ΔFEV1 ≥ 345 mL could help physicians to distinguish ACO from COPD in patients with a positive BDT result. To verify the accuracy of this conclusion, we strictly recruited 20 ACO, 57 COPD, and 132 asthma patients with a positive BDT result; we found that ΔFEV1 ≥ 345 mL was an excellent marker in distinguishing ACO from COPD. Similarly, a previous study showed that patients with ACO had a significantly higher ΔFEV1 value (mL) than those with COPD [25]. Moreover, some guidelines have suggested BDR ≥ 400 mL as the basis for distinguishing ACO from COPD, but in clinical practice, very few patients with COPD meet this criterion. It is very important to determine whether ΔFEV1 ≥ 345 mL could distinguish ACO from COPD in patients with a positive BDT result. Another study revealed that lung function parameters are potentially important tools in discriminating between asthma, ACO, and COPD [41]. COPD and asthma are characterized by incompletely reversible and reversible airflow obstruction, respectively [22, 42]. ACO shares the airflow obstruction characteristics of both asthma and COPD [22]. Thus, BDR is a key differential tool for distinguishing between COPD and ACO, especially in patients with a positive BDT result.

Alcázar-Navarrete [43] reported that an FeNO level of ≥ 19 ppb could distinguish ACO from COPD. Takayama [28] showed that COPD patients without treatment can be diagnosed as having ACO when the FeNO level is ≥ 25 ppb. In this study, patients with ACO had higher FeNO levels than those with COPD, but FeNO had no value in predicting ACO from COPD, which was inconsistent with previous studies’ findings [38, 44]. This result is likely due to the fact that patients had a positive BDT result, which can weaken the difference of the FeNO level between ACO and COPD. In the present study, most patients with a positive BDT result and airway limitation were diagnosed with COPD instead of asthma when ΔFEV1 < 315 mL was combined with an FeNO level < 28.5 ppb, which was verified by the validation study. This indicates that our prediction model is more meaningful in excluding a diagnosis.

This study has a few potential limitations, which should be considered. First, this is a single-center design, so multicenter studies are needed to confirm our findings. Secondly, all patients in this study are with a positive BDT result.

Conclusions

Our study showed that the simplified standard II could replace the common standard I as the criterion of a strongly positive BDT result. Additionally, ΔFEV1 alone or combined with FeNO are helpful in differentiating between asthma, COPD, and ACO in patients with a positive BDT result.