Validity of the COPD-6® Device for COPD Screening in the Primary Care Setting of China

Background: The use of simple and affordable screening tools for chronic obstructive pulmonary disease (COPD) is limited. We aimed to assess the validity of a handheld expiratory ow meter (COPD-6 ® , Vitalograph Ltd., Ireland) for COPD screening in Chinese primary care settings. Methods: In our cross-sectional study, subjects were randomly selected in eight primary care settings. Testing with the Vitalograph-COPD-6 ® and conventional spirometry were sequentially performed on subjects. The correlation between COPD-6 ® and conventional spirometry was determined. Validity was analyzed by the area under the receiver operator characteristic curve (AUC) of the forced expiratory volume in one second (FEV 1 ) / forced expiratory volume in six seconds (FEV 6 ) that used to detect airway obstruction. The sensitivity, specicity, predictive values, and likelihood ratio were calculated according to different FEV 1 /FEV 6 cut-off points. Results: 229 subjects (15.4%) were diagnosed with airow limitation by standard spirometry. FEV 1 , FEV 6 , and FEV 1 /FEV 6 measured by COPD-6 ® were correlated with FEV 1 , FVC, and FEV 1 /FVC measured by spirometry (r=0.889, 0.835and 0.647, p<0.001), respectively. AUC of the FEV 1 /FEV 6 to determine airow obstruction was 0.857 (95%CI: 0.826 to 0.888). No signicant difference of AUC was observed between the symptomatic group and the asymptomatic population (AUC=0.869 vs. 0.843, P=0.425). A similar phenomenon was found in the AUC of smokers and never-smokers (AUC=0.862 vs.0.840; P=0.515). The value of AUC was largest (i.e., 0.80) when the cut-off point for FEV 1 /FEV 6 was 0.77. Conclusions: The handheld COPD-6 ® could be used as a pre-screening device on early diagnosis of COPD in Chinese primary care settings.


Introduction
Characterized by persistent air ow limitation, chronic obstructive pulmonary disease (COPD) was a highprevalence disease with heavy mortality and morbidity burden [1,2]. It's reported that COPD caused 2.6% of global disability-adjusted life years (DALYs) and 3.2 million death worldwide in 2015 [3]. Air ow limitation is de ned as a post-bronchodilator forced expiratory volume in 1 second (FEV 1 ) / forced vital capacity (FVC) < 0.7 and regarded as the essential test for the diagnosis of COPD [4]. In COPD patients, persistent air ow limitation might lead to the substantially impaired quality of life and higher risk of premature death [5].
For the long-term of the asymptomatic phase, countless COPD patients remained undiagnosed until the onset of severe symptom [6,7]. Early symptoms of COPD are subtle and unrecognized for numbers of patients. The reduction of lung function was usually dramatic and irreversible when COPD was diagnosed for the rst time. What is more, the reduction of lung function could lead to poor health-related quality of life [7][8][9]. Although undiagnosed COPD patients usually have fewer exacerbations than severe COPD patients, they also require amount of medical care services for exacerbation events that should have been avoided [10]. Therefore, misdiagnosis of COPD could also bring considerable health burden. In this context, early screening for COPD was regarded as a potential method to reduce the burden of morbidity and mortality of patients [7]. However, the problem of underdiagnosis on COPD is obvious (ranging from 72-93%) [7]. There were increasing interests in improving the early detection of COPD in the primary care setting during the last decade. Spirometry is a well-established tool for quantifying air ow limitation and the diagnosis of patients with COPD [5]. However, there are several seasons for conventional spirometry in primary care practice. First, the expensive cost of the machine has limited technology extension. Second, shortness of professional training led to the unreliable quality of test and interpretation in primary care settings [11][12][13][14]. The U.S. Preventive Services Task Force and the American College of Physicians recommend that spirometry should not be used to screen for air ow limitation in individuals without respiratory symptoms. The use of conventional spirometry in primary care setting may result in a waste of medical resources and an overestimation of COPD burden [15,16].
Primary care physicians can obtain FEV 1 , FEV 1 % predicted, FEV 6 , FEV 6 %predicted, FEV 1 /FEV 6 , and lung age through COPD-6® test. It can also provide the diagnosis of air ow limitation and severity classi cation according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) guideline [5].
However, no study has examined the validity of COPD-6® in low or middle-income countries, including China. Furthermore, the best cut-off value using to de ne air ow limitation remains uncertain in China.
Therefore, we designed this study to assess the validity of COPD-6® device for COPD screening in primary care settings in China.

Setting
With the socioeconomic differences between rural and urban regions, two urban streets and two rural communities were randomly selected from an urban region (Guangzhou, Guangdong Province) and a rural region (Lianping, Guangdong Province), respectively. Two primary care settings were selected from each of the street/community mentioned above. Finally, eight primary care settings were involved in our study.

Study Population
The sample size of each age group was calculated according to the percentage of the population aged ≥ 40 years reported in the latest census. In selected primary care settings, 200 residents from four different age groups (40-49, 50-59, 60-69, and ≥ 70) was required. People engaged in these projects should had given informed consent before conventional spirometry and COPD-6® testing.
There were several exclusion criteria for spirometry and COPD-6® testing to avoid : (1) medical history of thoracic, abdominal or eye surgery in previous three months; (2) medical history of acute heart events (e.g., angina, acute myocardial infarction, and malignant arrhythmia) in previous three months; (3) hospitalizations for heart diseases in previous one month; (4) patients with active pulmonary tuberculosis disease or taking anti-tuberculosis drugs; (5) patients with a history of retinal detachment; (6) patients with new tumor diagnosed or undergoing a tumor treatment; (7) patients with cognitive impairment or mental disorder; (8) high paraplegia or thoracic deformity; (9) women during pregnancy or lactation.

Data Collection
Procedures Unique ID number was assigned to each participant. A standardized questionnaire, COPD-6® testing, and conventional spirometry were conducted for each participant sequentially.
COPD-6® testing COPD-6® testing was executed by well-trained primary care physicians. At least three maneuvers were performed for each participant without the use of bronchodilator. Results should have met criteria for acceptability (forced expiration for at least 6 s) and reproducibility (at least three acceptable ow-volume curves and the second-highest FEV 6 and FEV 1 were within 0.2L or 10% of highest value). We selected the best value for the report.

Spirometry
Spirometry testing was performed independently by trained operators according to American Thoracic Society/European Respiratory Society guidelines [30]. Operators were blinded to the COPD-6® results. All study sites used the same model spirometer (JAEGER-Master Screen Pneumo®, Carefusion™, GER). Spirometers were calibrated before each day's testing. Lung function parameters were measured before and 15 ~ 25 minutes after inhaling a dose of 400 µg salbutamol through a 500 ml spacer. We determined a quality grade (A ~ F) based on acceptable maneuvers and repeatability of the FEV 1 and FVC [31].
Spirometry results with grades A, B, or C were considered acceptable for analysis.

Analysis
Standard validation measures, including sensitivity, speci city, positive predictive value (PPV), negative predictive value (NPV), and likelihood ratio (for a positive test, LR+) were calculated at different cut-off points of FEV 1 /FEV 6 . ROC curve will be used to facilitate the cut-off point. The correlations of FEV 1 , FEV 6 , FEV 1 /FEV 6 measured by the COPD-6® (pre-bronchodilator), with FEV 1 , FVC, FEV 1 /FVC measured by spirometry (post-bronchodilator), were examined by Pearson's correlation analysis and Bland-Altman plots 28 . The 95% con dence interval was presented for all variables.

Discussion
This is the rst study to con rm the validity of a handheld expiratory device (COPD-6®) for COPD screening for primary care settings in China. We also came out with result that the appropriate cut-off value for FEV 1 /FEV 6 to determine air ow limitation was 0.77 in Chinese primary care settings, including both rural and urban area.
It has already been demonstrated that FEV 6 is a reliable alternative for FVC to detect airway obstruction and restriction [20]. There are two types of handheld tools for measuring FEV 6 and FEV 1 /FEV 6 : the Piko-6 (Ferraris Co., UK) and the COPD-6® (Vitalograph Ltd., Ireland). Previous studies have demonstrated that these devices could be useful in detecting pulmonary obstructive pathologies [23][24][25][26][27][28]32]. However, the best cut off point to use for de ning air ow obstruction remained uncertain. Vandevoorde et al. [33]and Melbye et al. [34] reported that FEV 1 /FEV 6 < 70-73% can be used as a valid alternative to FEV 1 /FVC < 70% for the detection of obstruction using conventional spirometers. Rosa et al. reported that the best cut-off point for the FEV 1 /FEV 6 ratio was 0.75 in subjects aged 40 years or over [35]. However, these studies were performed with conventional spirometers. Since we used the COPD-6®, a handheld spirometer, we cannot blindly adapt these values to our study directly.
In our study, we use FEV 1 /FVC < 70% as the "gold standard" to detect air ow obstruction. The AUC for FEV 1 /FEV 6 to identify air ow limitation was 0.857. The best cut-off point for FEV 1 /FEV 6 was 77.15% with a sensitivity of 71.2% and speci city of 89.8%. Besides, almost all of the discordant cases were close to the cut-off value. Our results in line with previous study which determined 73% as the cut-off value with greatest sensitivity and speci city [27]. The latest study from United Kingdom also support our result with closely cutoff (e.g., 78%) [36].
There are several seasons lead to the heterogeneity of studies: methodological measures, different prevalence of airway limitation and the cut off points used to de ne air ow obstruction. Previous results of multiple meta-regression presented that the prevalence of airway limitation may have an effect on diagnostic-odds ratio [21]. According to previous study, sensitivity and speci city was dependent on the prevalence of moderate-to-severe airway obstruction. Low prevalence of severe airway obstruction may reduce the sensitivity of FEV 6 , and Low prevalence of mild airway obstruction reduced the speci city of FEV 6 [37]. In our study, subjects were randomly selected and representative of the real world. Subjects included smokers and non-smokers, rural and urban residents, previous diagnosed and never diagnosed COPD patients. In this study, the prevalence of obstruction was 15.3% (229/1487). The best cut-off point for FEV 1 /FEV 6 was 77.15% with a sensitivity of 71.2% and speci city of 89.8%. In previous study, (population aged 45-85 years and with smoking history of > 15 pack-years), sensitivity and speci city were 79.2% and 80.3% when cut-off value was set as 73% [27]. Despite the inconstant result of cut-off point of FEV 1 /FEV 6 , our ndings show that COPD-6® device was effective in detecting previously undiagnosed COPD and ought to be used as a tool for COPD screening in Chinese primary care settings.
We came out with a result that about one of two patients whose FEV 1 /FEV 6 < 0.77 will exhibit air ow limitation with spirometry.
Furthermore, similar AUC values were obtained in the symptomatic population (AUC = 0.87) and asymptomatic population (AUC = 0.84). No signi cant difference was observed between smoking group (including smokers and ex-smokers) (AUC = 0.86) and non-smokers (AUC = 0.84). The results remind us that the COPD-6® device was effective in detecting air ow limitation for population with diverse characteristics, especially in non-smokers and asymptomatic patients.
To our best knowledge, handheld spirometric measurements (i.e., COPD-6®) are not identical to conventional spirometry. The most limitation of this approach was handheld spirometric measurements may not be appropriate to be used for determining the grade of air ow limitation. We assessed the correlation of several measures by COPD-6® and spirometry in different stages of COPD. No signi cant correlations were observed between FVC measured by spirometry and FEV 6 measured by COPD-6® in GOLD stage IV. FEV 1 /FVC measured by spirometry was also not correlated with FEV 1 /FEV 6 measured by COPD-6® in GOLD stage IV.
These phenomena indicated the fact that COPD-6® may not be appropriate to be used for determining the grade of air ow limitation.
There are several potential reasons for differences and inconsistency mentioned above: (1) FEV 1 /FVC and FVC are more dependent on the FET(Forced expiratory time) than FEV 1 /FEV 6 and FEV 6 [38]; (2) Instead of measuring the whole FVC, COPD-6® testing stops measuring after 6 s and results in the risk of overstating FEV 1 /FEV 6 ratio; (3) COPD-6® does not provide graph analysis of the volume/time or ow/volume curves that are essential (especially the later ones) in quality control; (4)In this study, COPD-6® were performed before the use of bronchodilation, and the conventional spirometry were adapted after the process of post-bronchodilation; (5) In our study, only ve patients were GOLD stage IV, which might cause statistical bias.
However, limitations mentioned above is irrelevant for the COPD screening in primary care settings. Utilization of handheld expiratory ow meter (COPD-6®) was aimed to reduce misdiagnosis rate and avoid the waste of medical resources at the same time. Our study, including the handheld expiratory ow meter and its cutoff value, can be widely recommended for the practice of COPD screening in Chinese communities.

Conclusions
The handheld Vitalograph COPD-6® meter could be used as a pre-screening device in early diagnosis of COPD in Chinese primary care settings. Furthermore, it should be noted that the cut-off value for FEV 1 /FEV 6 to determine air ow limitation was 0.77.

List Of Abbreviations
COPD, chronic obstructive pulmonary disease; FEV 1 , forced expiratory volume in one second; FEV 6 , forced expiratory volume in six seconds; FVC, forced vital capacity; AUC, area under the receiver operator characteristic curve; GOLD, global initiative for chronic obstructive lung disease; LoA, limit of Agreement.

Declarations
Ethics approval and Consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
All data which were generated or analysed are included in this published article and also its supplementary information les.