Cut-off value of FEV₁/FEV₆ as a surrogate for FEV₁/FVC for detecting airway obstruction in a Korean population

Abstract

Background

Forced expiratory volume in 1 second (FEV₁)/forced expiratory volume in 6 seconds (FEV₆) has been proposed as an alternative to FEV₁/forced vital capacity (FVC) for detecting airway obstruction. A fixed cut-off value for FEV₁/FEV₆ in a Korean population is lacking. We investigated a fixed cut-off for FEV₁/FEV₆ as a surrogate for FEV₁/FVC for detecting airway obstruction.

Materials and methods

We used data obtained in the 5 years of the Fifth and Sixth Korean National Health and Nutrition Examination Survey. A total of 14,978 participants aged ≥40 years who underwent spirometry adequately were the study cohort. “Airway obstruction” was a fixed cut-off FEV₁/FVC <70% according to the Global Initiative for Chronic Obstructive Lung Disease guidelines. We also used European Respiratory Society/Global Lung Initiative 2012 equations for the FEV₁/FVC lower limit of normal.

Results

Among the 14,978 participants (43.5% male, 56.5% female; mean age: 56.9 years for men and 57.0 years for women), 14.0% had obstructive lung function according to a fixed cut-off FEV₁/FVC <70%. Optimal FEV₁/FEV₆ cut-off for predicting FEV₁/FVC <70% was 75% using receiver operating characteristic curve analyses (area under receiver operating characteristic curve =0.989, 95% confidence interval 0.987–0.990). This fixed cut-off of FEV₁/FEV₆ showed 93.8% sensitivity, 94.8% specificity, 74.7% positive predictive value, 98.9% negative predictive value, and 0.8 Cohen’s kappa coefficient. When compared with FEV₁/FVC < lower limit of normal, FEV₁/FEV₆ <75% tended to over-diagnose airflow limitation (just like a fixed cut-off of FEV₁/FVC <70%). When grouped according to age and FEV₁ (%), FEV₁/FEV₆ <75% diagnosed more airway obstruction in older participants and mild–moderate stages compared with FEV₁/FVC <70%.

Conclusion

A valid fixed cut-off for detecting airway obstruction in a Korean population is FEV₁/FEV₆ of 75%, but should be used with caution in older individuals and those with mild–moderate airway obstruction.

Keywords:

• chronic obstructive pulmonary disease
• forced expiratory volume in 6 seconds
• pulmonary function test
• airway obstruction

Introduction

COPD is the fourth leading cause of death worldwide. However, the World Health Organization predicts that it will become the third leading cause of death by 2030. According to a recent report,¹ the priority of screening spirometry in asymptomatic patients has been reduced. However, a diagnostic pulmonary function test in symptomatic patients with wheezing, dyspnea, productive cough, or cough is needed (especially in primary care clinics).² Until now, expensive, labor-intensive, and inconvenient spirometry has been an obstacle in COPD diagnosis. In the United States, only 32% of COPD patients undergo spirometry within 2 years before, or 6 months after, the diagnosis has been made.³ Previously, we showed that only 27.5% of Korean patients with airway obstruction who used COPD-related health care facilities underwent spirometry.⁴

Several studies^5–18 have demonstrated that forced expiratory volume in 6 seconds of exhalation (FEV₆), and forced expiratory volume in 1 second (FEV₁) of exhalation ratio with FEV₆, can be used as convenient and accurate surrogates for forced vital capacity (FVC) and FEV₁/FVC, respectively. Owing to its use of a 6-second expiratory maneuver, FEV₆ makes spirometry in the office easier, faster, and safer¹⁹ than FVC measurement. This theoretical basis enables primary care physicians to use hand-held spirometric devices such as PiKo-6^® (nSpire Health, Inc. Longmont, CO, USA) or COPD-6™ (Vitalograph Ltd, Ennis, Co., Clare, Ireland) without having to employ expensive spirometry tests for the screening or diagnosis of COPD, and by easily providing the values of FEV at 1 and 6 seconds of exhalation.

However, an identical cut-off value between FEV₁/FEV₆ and FEV₁/FVC cannot be used for detecting airway obstruction because FEV₆ cannot be greater than FVC. Despite this controversy, some reports have argued that FEV₁/FEV₆ can reduce the sensitivity of detecting airway obstruction, especially in elderly patients with mild airway obstruction.^20,21 Therefore, other studies⁶ have submitted increased cut-off values (73%–76%) of FEV₁/FEV₆ for detecting airway obstruction to overcome these limitations.

Spirometry indices are influenced by age, height, sex, and ethnicity.²² Therefore, discovering the cut-off value of FEV₁/FEV₆ for detecting airway obstruction in a nationwide, representative population sample in Korea is worthwhile. The purpose of the present study was to ascertain the cut-off value of FEV₁/FEV₆ in place of FEV₁/FVC for detecting airway obstruction in a Korean population.

Materials and methods

Study population

Data for the present study were obtained from 5 years of the Fifth and Sixth Korean National Health and Nutrition Examination Survey (KNHANES V–VI), which was conducted between January 2010 and December 2014. KNHANES V–VI employed a stratified, multiple-stage, clustered-probability design to select a representative sample of non-institutionalized civilians among a Korean population. These data comprised information on age, ethnicity, sex, height, weight, and pre-bronchodilator spirometry values. A Vmax 2130 Dry Rolling-seal Spirometer (SensorMedics, Yorba Linda, CA, USA) was employed. Spirometry was undertaken by specially trained technicians according to recommendations set by the American Thoracic Society (ATS)/European Respiratory Society (ERS) in 2005.²³ Data from a total of 14,978 individuals aged >40 years who underwent complete pulmonary-function tests from KNHANES V–VI were selected for the present study. We analyzed data on FEV₁, FEV₆, and FVC from subjects with two or more acceptable spirometric performances. The study protocol was approved by the institutional review board of Severance Hospital (approval number: 4-2006-0101). Written informed consent was obtained from patients or their next of kin.

Definition of airway obstruction

All spirometric values came from pre-bronchodilator results in this study. “Airway obstruction” was defined as FEV₁/FVC <70% as a fixed cut-off point according to guidelines set by the Global Initiative for Chronic Obstructive Lung Disease (GOLD).²⁴ Severity of airway obstruction was classified as: “mild” (FEV₁ ≥80% predicted, GOLD stage I), “moderate” (50%≤ FEV₁ <80% predicted, GOLD stage II), “severe” (30%≤ FEV₁ <50% predicted, GOLD stage III), and “very severe” (30%≤ FEV₁ predicted, GOLD stage IV). In addition, “normal” lung function was defined as FEV₁/FVC ≥70% and FVC ≥80% predicted. A “restrictive spirometric pattern” was defined as FEV₁/FVC ≥70% and FVC <80% predicted.

ATS/ERS guidelines recommend use of reference equations derived from a representative sample of healthy individuals to determine the lower limit of normal (LLN), taking into account that spirometry indices are influenced by age, height, sex, and ethnicity. However, an officially recognized equation to use as a reference for pulmonary function in Korea is lacking. Hence, we used equations for FEV₁/FVC LLN set by the ESR/Global Lung Initiative (GLI) in 2012. These equations were announced by an ERS/GLI taskforce, and were derived from data collected from healthy nonsmokers 3 to 95 years old from 33 countries (including Korea). ERS/GLI 2012 equations provided multiple-ethnicity values and the LLN for spirometry.²⁵ “Airway obstruction” by FEV₁/FVC LLN was defined as being below its LLN in the presence of a normal FEV₁/FVC.

Statistical analyses

Data were analyzed using SAS v9.2 (SAS Institute Inc., Cary, NC, USA). Baseline characteristics were summarized using numbers and percentages to describe categorical variables, and compared using the chi-squared test. Continuous variables were compared using the Student’s t-test and are presented as the mean ± standard deviation (SD). FEV₁/FEV₆ was determined for the highest collective sensitivity and specificity using a receiver operating characteristic (ROC) curve for detecting airway obstruction. Cohen’s kappa coefficient (CKC) was used to evaluate agreement between FEV₁/FEV₆ and FEV₁/FVC. Positive predictive value (PPV) and negative predictive value (NPV) were calculated for the most suitable FEV₁/FEV₆ threshold for the diagnosis of FEV₁/FVC <70%.

Results

Study populations

Among the 14,978 participants, 6,515 (43.5%) were male and 2,104 (14.0%) had airway obstruction. A total of 1,018 (6.8%) participants were classified as having mild obstruction, 996 (6.6%) with moderate obstruction, 80 (0.5%) with severe obstruction, and ten (0.1%) with very severe obstruction according to spirometry values. Airway obstruction was more prevalent in male than in female participants (Table 1).

Table 1 Baseline characteristics of the study population

	Male	Female	Total
Subject, n (%)	6,515 (43.5)	8,463 (56.5)	14,978 (100)
Age (years)	56.9±10.9	57.0 ±10.6	57.0±10.7
Age group (years)
40–49	1,944 (29.8)	2,342 (27.7)	4,286 (28.6)
50–59	1,922 (29.5)	2,739 (32.4)	4,661 (31.1)
60–69	1,625 (24.9)	2,077 (24.5)	3,702 (24.7)
≥70	1,024 (15.7)	1,305 (15.4)	2,329 (15.5)
Height (cm)	168.4±6.0	155.4±5.8	161.0±8.7
Weight (Kg)	69.0±9.7	58.3±8.4	62.9±10.5
BMI (Kg/m^2)	24.3±2.8	24.1±3.2	24.2±3.0
Pulmonary function test
FEV1 (L)	3.11±0.66	2.33±0.45	2.67±0.67
FEV1 (%, pred)	89.79±13.69	94.40±12.95	92.40±13.47
FVC (L)	4.15±0.69	2.93±0.50	3.46±0.85
FVC (%, pred)	91.52±11.65	94.10±11.57	92.98±11.67
FEV1/FVC (%)	0.75±0.08	0.79±0.06	0.77±0.08
FEV6 (L)	3.98±0.71	2.87±0.50	3.35±0.82
FEV1/FEV6 (%)	0.78±0.07	0.81±0.05	0.80±0.06
Normal	4,291 (65.9)	7,145 (84.4)	11,436 (76.4)
Restricted	688 (10.6)	750 (8.9)	1,438 (9.6)
Obstructed	1,536 (23.6)	568 (6.7)	2,104 (14.1)
GOLD stage I	740 (11.4)	278 (3.3)	1,018 (6.8)
GOLD stage II	731 (11.2)	265 (3.1)	996 (6.6)
GOLD stage III	57 (0.9)	23 (0.3)	80 (0.5)
GOLD stage IV	8 (0.1)	2 (0.0)	10 (0.1)

Notes: Data are presented as numbers (percentages) for categorical variables. Continuous variables are presented as means ± standard deviations.

Abbreviations: FEV₁, forced expiratory volume in 1 second; FEV₆, forced expiratory volume in 6 seconds; FVC, forced vital capacity; BMI, body mass index; pred, predicted; GOLD, Global Initiative for Chronic Obstructive Lung Disease.

Spirometric diagnosis of airway obstruction

According to the airway obstruction with FEV₁/FVC <70%, an ROC curve was used to determine the best corresponding cut-off for FEV₁/FEV₆ (Figure 1). The area under the ROC curve was 0.989 (95% confidence interval [CI]: 0.987–0.990). The FEV₁/FEV₆ cut-off corresponding to the greatest sum of sensitivity and specificity was 75%. Table 2 shows the sensitivities and specificities of different thresholds of FEV₁/FEV₆ for prediction of FEV₁/FVC <70%. When choosing an FEV₁/FEV₆ cut-off of 70%, specificity reached 100% but the sensitivity dropped to 44.1%. A fixed cut-off of FEV₁/FEV₆ <76% showed high sensitivity (98.0%) and specificity (90.8%) but a low PPV (63.6%). When using a fixed cut-off FEV₁/FEV₆ of 75%, sensitivity of 93.8%, specificity of 94.8%, PPV of 74.7%, NPV of 98.9%, accuracy of 94.7%, and CKC of 0.8, (95% CI: 0.786–0.814) were obtained, suggesting very good agreement between FEV₁/FEV₆ and FEV₁/FVC. When using a fixed cut-off of FEV₁/FEV₆ <75%, the number of false-positive cases was 670 (5.2%) and false-negative cases was 130 (6.2%) (Table 3). In particular, false-negative cases were included in mild (n=102) and moderate (n=28) airway-obstruction groups (Table 4).

Table 2 Sensitivities and specificities of thresholds of FEV₁/FEV₆ for predicting FEV₁/FVC <70%

FEV1/FEV6 ratio	Sensitivity	Specificity	PPV	NPV	Accuracy	Kappa value (95% CI)
<70	44.1	100.0	100.0	91.6	92.2	0.576 (0.554–0.598)
<73	74.2	98.9	91.6	95.9	95.4	0.794 (0.778–0.810)
<75	93.8	94.8	74.7	98.9	94.7	0.800 (0.786–0.814)
<76	98.0	90.8	63.6	99.6	91.8	0.724 (0.710–0.738)

Note: Data are presented as percentages.

Abbreviations: FEV₁, forced expiratory volume in 1 second; FEV₆, forced expiratory volume in 6 seconds; FVC, forced vital capacity; PPV, positive predictive value; NPV, negative predictive value; CI, confidence interval.

Table 3 Comparison of FEV₁/FEV₆ with FEV₁/FVC for diagnosing airway obstruction

	FEV1/FEV6#		Totals
	No obstruction	Obstruction
FEV1/FVC§
No obstruction	12,204 (94.8)	670 (5.2)	12,874 (86.0)
Obstruction	130 (6.2)	1,974 (93.8)	2,104 (14.0)
Totals	12,334 (82.3)	2,644 (17.7)	14,978 (100)

Notes: Data are presented as numbers (percentages);

§ using FEV₁/FVC <70% as a fixed cut-off;

# using FEV₁/FEV₆ as a fixed cut-off <75%; prevalence of obstruction =17.7%, Kappa value =0.80, 95% CI 0.79–0.81, P<0.001.

Abbreviations: FEV₁, forced expiratory volume in 1 second; FEV₆, forced expiratory volume in 6 seconds; FVC, forced vital capacity; CI, confidence interval.

Table 4 Comparison of FEV₁/FEV₆ with FEV₁/FVC for the diagnosis of normal, restricted, and obstruction patterns

	FEV1/FEV6#		Totals
	No obstruction	Obstruction
FEV1/FVC§
Normal	10,862 (95.0)	574 (5.0)	11,436 (76.4)
Restricted	1,342 (93.3)	96 (6.7)	1,438 (9.6)
Obstructed	130 (6.2)	1,974 (93.8)	2,104 (14.1)
GOLD stage I	102 (10.0)	916 (90.0)	1,018 (6.8)
GOLD stage II	28 (2.8)	968 (97.2)	996 (6.6)
GOLD stage III	0 (0)	80 (100)	80 (0.5)
GOLD stage IV	0 (0)	10 (100)	10 (0.1)
Totals	12,334 (82.3)	2,644 (17.7)	14,978 (100)

Notes: Data are presented as numbers (percentages);

§ using FEV₁/FVC <70% as a fixed cut-off;

# using FEV₁/FEV₆ as a fixed cut-off <75%.

Abbreviations: FEV₁, forced expiratory volume in 1 second; FEV₆, forced expiratory volume in 6 seconds; FVC, forced vital capacity; GOLD, Global Initiative for Chronic Obstructive Lung Disease.

Figure 1 A receiver operating characteristic (ROC) curve showing the ability of FEV₁/FEV₆ to diagnose FEV₁/FVC <70%.

Notes: Area under the ROC curve is 0.989 (95% confidence interval, 0.987–0.990). The best cut-off value is 0.75 with 0.971% sensitivity and 0.927% specificity.
Abbreviations: FEV₁, forced expiratory volume in 1 second; FEV₆, forced expiratory volume in 6 seconds; FVC, forced vital capacity; AUC, area under the curve.

Comparison of FEV₁/FVC < LLN with FEV₁/FVC <70% or FEV₁/FEV₆ <75%

To detect airway obstruction, we compared FEV₁/FVC LLN with a fixed cut-off of FEV₁/FVC <70% or FEV₁/FEV₆ <75%. When using FEV₁/FVC LLN, the prevalence of obstruction was 7.2%. However, when choosing a fixed cut-off of FEV₁/FVC <70% and FEV₁/FEV₆ <75%, the prevalence of obstruction was 14.1% and 17.7%, respectively. Compared with FEV₁/FVC LLN, a fixed cut-off of FEV₁/FVC <70% showed 99.0% sensitivity, 92.6% specificity, 93.1% accuracy, 51.1% PPV, and 99.9% NPV. A fixed cut-off of FEV₁/FEV₆ <75% showed 99.9% sensitivity, 88.8% specificity, 89.6% accuracy, 41.0% PPV, and 99.9% NPV (Table 5). When grouped together according to age and FEV₁ (%), a fixed cut-off of FEV₁/FEV₆ <75% was used to diagnose more airway obstruction in older individuals and those with mild–moderate stages (Table 6). With increasing age and severity of obstruction, the mean difference between FEV₁/FEV₆ and FEV₁/FVC was increasingly larger because FEV₆ was increasingly smaller than FVC.

Table 5 Comparison of FEV₁/FVC LLN with FEV₁/FVC <70% or FEV₁/FEV₆ <75%

	FEV1/FVC LLN*		Totals
	No obstruction	Obstruction
FEV1/FVC§
No obstruction	12,857 (99.9)	11 (0.1)	12,868 (85.9)
Obstruction	1,030 (49.0)	1,074 (51.0)	2,104 (14.1)
FEV1/FEV6#
No obstruction	12,327 (100.0)	1 (0.0)	12,328 (82.3)
Obstruction	1,560 (59.0)	1,084 (41.0)	2,644 (17.7)
Totals	13,887 (92.8)	1,085 (7.2)	14,972 (100)

Notes: Data are presented as numbers (percentages);

* lower limit of normal using ERS/GLI 2012 equations;

§ using FEV₁/FVC <70% as a fixed cut-off;

# using FEV₁/FEV₆ as a fixed cut-off <75%.

Abbreviations: FEV₁, forced expiratory volume in 1 second; FEV₆, forced expiratory volume in 6 seconds; FVC, forced vital capacity; LLN, lower limit of normal; ERS, European Respiratory Society; GLI, Global Lung Initiative.

Table 6 Age and forced expiratory volume in 1 second (FEV₁) dependency of the lower limit of normal (LLN) for FEV₁/forced vital capacity (FVC) and FEV₁/forced expiratory volume in 6 seconds (FEV₆) in airway obstruction

	FEV1/FVC <LLN	FEV1/FVC <70%	FEV1/FEV6 <75%	ΔFEV1/FVC*	ΔFEV1/FEV6§	Δ(FEV1/FEV6−FEV1/FVC)#
	1,085 (7.2)	2,104 (14.1)	2,644 (17.7)	14,972 (100)	14,972 (100)	14,972 (100)
Age group (years)
40~49	153 (3.6)	151 (3.5)	310 (7.2)	10.5±5.2	11.8±4.6	1.3±1.2
50~59	278 (6.0)	417 (8.9)	594 (12.7)	9.9±6.0	12.0±4.9	2.1±1.7
60~69	361 (9.8)	766 (20.7)	888 (24.0)	9.0±8.4	12.3±5.7	3.1±2.3
70~	293 (12.6)	770 (33.1)	852 (36.6)	9.7±6.6	12.6±5.3	3.7±2.8
FEV1 (%)
>80	336 (2.6)	1,018 (8.0)	1,409 (11.1)	11.0±5.1	13.9±4.2	2.1±1.8
50~80	661 (30.3)	996 (45.6)	1,144 (52.4)	3.4±8.0	7.3±6.2	3.9±2.8
30~50	78 (89.7)	80 (92.0)	81 (93.1)	−15.7±10.9	−7.5±8.1	8.2±4.1
<30	10 (100.0)	10 (100.0)	10 (100.0)	−29.8±7.9	−19.9±5.4	9.9±3.3

Notes: Data are presented as numbers (percentage) or mean ± standard deviations;

* difference between LLN (%) and FEV₁/FVC (%),

§ difference between LLN (%) and FEV₁/FEV₆(%);

# difference between FEV₁/FEV₆ (%) and FEV₁/FVC (%).

However, with increasing age, the mean difference between FEV₁/FVC LLN and FEV₁/FVC or FEV₁/FEV₆ showed only a slight difference. With increasing severity of obstruction, the mean difference between FEV₁/FVC LLN and FEV₁/FEV₆ was larger than the mean difference between FEV₁/FVC LLN and FEV₁/FVC at mild–moderate stages of airway obstruction, but that was smaller than that at severe and very severe stages (Figures 2 and 3).

Figure 2 FEV₁/FVC LLN, FEV₁/FVC, FEV₁/FEV₆, and the mean difference between FEV₁/FEV₆ and FEV₁/FVC according to the age group.

Notes: With increasing age, the mean difference between FEV₁/FEV₆ and FEV₁/FVC was increasingly larger. However, the mean difference between FEV₁/FVC LLN and FEV₁/FVC or FEV₁/FEV₆ showed only a slight difference.

Abbreviations: FEV₁, forced expiratory volume in 1 second; FEV₆, forced expiratory volume in 6 seconds; FVC, forced vital capacity; LLN, lower limit of normal.

Figure 3 FEV₁/FVC LLN, FEV₁/FVC, FEV₁/FEV₆, and the mean difference between FEV₁/FEV₆ and FEV₁/FVC according to the FEV₁ (%) group.

Notes: With increasing severity of obstruction, the mean difference between FEV₁/FEV₆ and FEV₁/FVC was increasingly larger. The mean difference between FEV₁/FVC LLN and FEV₁/FEV₆ was larger than the mean difference between FEV₁/FVC LLN and FEV₁/FVC at mild–moderate stages of airway obstruction, but was smaller at severe and very severe stages.
Abbreviations: FEV₁, forced expiratory volume in 1 second; FEV₆, forced expiratory volume in 6 seconds; FVC, forced vital capacity; LLN, lower limit of normal.

Discussion

FEV₁/FEV₆ and FEV₆ are reliable surrogates for FEV₁/FVC and FVC to identify obstruction and restriction using spirometry. However, in previous studies, the cut-off values of FEV₁/FEV₆ for detecting airway obstruction were different among LLN or fixed ratios (70%–76%).⁶ LLN for FEV₁/FEV₆ is the most accurate for detecting airway obstruction, but many countries do not have pulmonary function reference equations of LLN for FEV₁/FEV₆ for local ethnic groups. Nevertheless, primary care clinicians have used handheld spirometric devices such as PiKo-6^® or COPD-6™ without proper reference values for FEV₁/FEV₆. These two devices are inexpensive, and require less quality control of equipment, and no experienced personnel to carry out tests and analyses. The main purpose of the present study was to determine a fixed cut-off value for FEV₁/FEV₆ as a surrogate for FEV₁/FVC for detecting airway obstruction in a Korean population.

The present study elicited two main findings. First, a fixed cut-off for FEV₁/FEV₆ as a good surrogate for FEV₁/FVC for detecting airway obstruction in a Korean population was 75% with 93.8% sensitivity, 94.8% specificity, 74.7% PPV, 98.9 NPV, and good CKC (0.8). Second, a raised fixed cut-off value of FEV₁/FEV₆ <75% to improve sensitivity showed the tendency to over-diagnosis in the mild–moderate airflow limitation group compared with FEV₁/FVC<70% were observed in the mild–moderate airflow limitation group compared with FEV₁/FVC. Therefore, a raised and fixed cut-off value of FEV₁/FEV₆ <75% should be used with caution in older individuals or those with mild–moderate airflow limitation because this cut-off value can lead to over-diagnosis rather than under-diagnosis for airway obstruction screening. A fixed ratio of cut-off led to greater over-diagnosis in patients with mild airflow limitation than LLN, but it was less-time consuming, easier, and safer for patients. Additionally, recording FEV₆ was more reproducible than FVC, less physically demanding for patients, and the results were more clear at the end of the test.²²

In a study in Belgium, a fixed cut-off of FEV₁/FEV₆ of <73% showed 94.4% sensitivity, 93.3% specificity, 92.2% PPV, and 95.2% NPV.¹¹ Jing et al showed, in a meta-analysis, that the efficacy of FEV₁/FEV₆ is not affected adversely by the choice of cut-off point,⁶ but an appropriate clinical guideline for using FEV₁/FEV₆ in place of FEV₁/FVC is needed. According to one study,²⁶ the cut-off value of pre-bronchodilator FEV₁/FEV₆ <70% had 91.4% sensitivity, 100% specificity but a 91.4% PPV and 87.4% NPV in 353 Koreans aged >65 years. However, this cut-off value of pre-bronchodilator FEV₁/FEV₆ <70% cannot be relied upon because the data were from a small study cohort with only older participants; thus the results were not applicable to a general population. Further, a cutoff value of a fixed ratio of FEV₁/FVC <70% cannot be applied because FEV₆ cannot be greater than FVC – otherwise, airway obstruction may be underestimated. Because an appropriate study on the cut-off of FEV₁/FEV₆ for airway obstruction in a Korean population had not been carried out, a revised clinical guideline²⁷ for COPD in Korea published in 2014 recommended a cut-off of FEV₁/FEV₆ of 73% for detecting airway obstruction based on studies undertaken outside Korea.^24,27

Our study showed that 75% as a fixed value of FEV₁/FEV₆ in analyses of ROC curves was the cut-off point with the best combination of sensitivity and specificity and a good surrogate for FEV₁/FVC <70%. Prevalence of airway obstruction was 17.7% when using FEV₁/FEV₆ <75% and 14.1% with FEV₁/FVC <70%. Rosa et al found that FEV₁/FEV₆ <75% showed 89.0% sensitivity, 95.4% specificity, 85.2% PPV, and 96.7% NPV in pre-bronchodilator curves, results that were similar to our research.²⁸

A total of 620 (5.2%) false-positive values and 130 (6.2%) false-negative values were observed (Table 3). In the subgroup analysis, 130 false-negative cases showed a mean difference of FEV₁/FVC and FEV₁/FEV₆, with a respective LLN of 4.4% (SD =2.0) and 11.4% (SD =2.1). They were classified as GOLD stages I and II because reduced end-expiratory flow and expiratory time >6 seconds might have occurred in the initial stage of airflow limitation (Table 4). A total of 620 false-positive cases showed a mean difference of FEV₁/FVC and FEV₁/FEV₆ with a respective LLN of 4.8% (SD =2.9) and 6.8% (SD =2.8). This finding might have been due to different fixed cut-off values between FEV₁/FVC <70% and FEV₁/FEV₆ <75%.

The main strength of our study was that we obtained results from a representative sample of noninstitutionalized civilians among a Korean population using a stratified, multiple-stage, clustered-probability design. However, our study also had several limitations. First, lung function data from prebronchodilator spirometric values might have inadvertently included asthma patients, and hence be skewed. Carrying out postbronchodilator spirometry in a large, nationwide sample from the Korean National Health and Nutritional Examination Survey is difficult. Second, the absence of data on smoking history, previously diagnosed diseases (eg, COPD, asthma, tuberculosis, and bronchiectasis) in our study could have resulted in abnormal values for airway obstruction. So our results for detecting of airway obstruction were not limited to COPD, asthma, or other obstructive disease. Third, there were no available data on FEV₁/FEV₆ LLN because a pulmonary reference equation for FEV₆ for a Korean population is lacking. Hence, we could not obtain a cut-off value of FEV₆ in place of FVC for detecting restrictive pulmonary patterns. Finally, the fixed cut-off value of FEV₁/FEV₆ <75% should be verified with handheld spirometric devices with good repeatability, as in other studies,¹⁸ and not only with office spirometry.

Conclusion

A fixed cut-off for FEV₁/FEV₆ as a good surrogate for FEV₁/FVC for detecting airway obstruction was 75% in a Korean population. However, a fixed cut-off value of FEV₁/FEV₆ <75% should be used with caution in older individuals and in those with mild–moderate airflow obstruction.

Author contributions

Kyung Soo Chung contributed to the study design, analysis and interpretation of data, and writing of the manuscript. Ji Ye Jung, Moo Suk Park, Young Sam Kim, Se Kyu Kim, and Joon Chang contributed to the study design, analyses and interpretation of data, and critically revised the manuscript. Joo Han Song, contributed to the study design, analyses and interpretation of data, and writing of the manuscript.

Disclosure

The authors report no conflicts of interest in this work.