A comparative study on the value of lower airway exhaled nitric oxide combined with small airway parameters for diagnosing cough-variant asthma

Abstract

Background:

The diagnosis of cough-variant asthma (CVA) is based on bronchial provocation test, which is challenging to be conducted. Most CVA patients have type 2 airway inflammation and small airway dysfunction. FeNO₂₀₀, reflecting small airway inflammation, may be used to diagnose CVA.

Objective:

This study aimed to explore and compare the value of lower airway exhaled nitric oxide (FeNO₅₀, FeNO₂₀₀, and CaNO) combined with small airway parameters for diagnosing CVA.

Methods:

Chronic cough patients who attended the clinic from September 2021 to August 2022 were enrolled and divided into CVA group (n = 71) and non-CVA (NCVA) group (n = 212). The diagnostic values of FeNO₅₀, FeNO₂₀₀, concentration of alveolar nitric oxide (CaNO), maximal mid-expiratory flow (MMEF), forced expiratory flow at 75% of forced vital capacity (FEF_75%) and forced expiratory flow at 50% of forced vital capacity (FEF_50%) for CVA were evaluated.

Results:

FeNO₅₀ [39(39) ppb versus 17(12) parts per billion (ppb), p < 0.01], FeNO₂₀₀ [17(14) ppb versus 8(5) ppb, p < 0.01] and CaNO [5.0(6.1) ppb versus 3.5(3.6) ppb, p < 0.01] in CVA group were significantly higher than those in NCVA group. The optimal cut-off values of FeNO₅₀, FeNO₂₀₀, and CaNO for diagnosis of CVA were 27.00 ppb [area under the curve (AUC) 0.88, sensitivity 78.87%, specificity 79.25%], 11.00 ppb (AUC 0.92, sensitivity 88.73%, specificity 81.60%) and 3.60 ppb (AUC 0.66, sensitivity 73.24%, specificity 52.36%), respectively. For diagnosing CVA, the value of FeNO₂₀₀ was better than FeNO₅₀ (p = 0.04). The optimal cut-off values of MMEF, FEF_75%, and FEF_50% for the diagnosis of CVA were 63.80% (AUC 0.75, sensitivity 53.52%, specificity 86.32%), 77.9% (AUC 0.74, sensitivity 57.75%, specificity 83.49%) and 73.50% (AUC 0.75, sensitivity 60.56%, specificity 80.19%), respectively. The AUCs of FeNO₅₀ combined with MMEF, FEF_75%, and FEF_50% for the diagnosis of CVA were all 0.89. The AUCs of FeNO₂₀₀ combined with MMEF, FEF_75%, and FEF_50% for the diagnosis of CVA were all 0.93.

Conclusion:

FeNO₂₀₀ > 11 ppb contributed strongly for differentiating CVA from chronic cough, especially in patients with small airway dysfunction.

Introduction

Chronic cough is defined as cough lasting no less than 8 weeks and with a normal chest X-ray. ¹ According to several domestic studies, cough-variant asthma (CVA) is one of the most common causes of chronic cough, accounting for nearly one-third of cases.^2,3 CVA is a special type of asthma, characterized by coughing as the sole or predominant symptom and airway hyperresponsiveness, without obvious wheezing or dyspnea. The positive bronchial provocation test indicating airway hyperresponsiveness is one of the diagnostic criteria for CVA. ¹ However, the bronchial provocation test is time-consuming and expensive. Besides, it has the risk of causing severe bronchospasm. Moreover, some hospitals are not able to conduct this test, especially primary hospitals. Therefore, a more convenient way to diagnose CVA is needed.

Fractional exhaled nitric oxide (FeNO) is a widely used noninvasive measurement in clinical practice. As a preliminary screening method for airway inflammation, FeNO can identify type 2 inflammation in the central airway, small airway, and alveoli by selecting different exhalation velocity, and also can be used to predict the responsiveness of patients with chronic cough to corticosteroid therapy. ⁴ FeNO₅₀ was measured at exhalation flow rates of 50 ml/s, reflecting the inflammation of central airway and bronchus. ⁵ FeNO₂₀₀ was measured at exhalation flow rates of 200 ml/s, reflecting the inflammation of peripheral small airway. ⁶ Concentration of alveolar nitric oxide (CaNO) was calculated by compartment model or linear regression, reflecting the inflammation of small airway in the alveolar or acinar area. ⁷

Asthma is a heterogeneous disease characterized by chronic airway inflammation and airway hyperresponsiveness. Traditionally, the inflammatory process of asthma was considered to mainly occur in the large airways, and the small airways were once considered as the ‘silent zone’. However, more and more studies have suggested that the small airway is the main site of airflow restriction, inflammatory infiltration, and structural remodeling. ⁸ According to a global multicenter study, small airway dysfunction (SAD) was found in 91% of asthma patients. ⁹ Most patients with CVA have normal pulmonary ventilation function, while small airway function is usually abnormal. A domestic study has shown that SAD existed in more than 60% of CVA patients when SAD was diagnosed on the basis of at least two of the following three indicators of lung function less than 65% of predicted: MMEF, FEF_50%, and FEF_75%. ¹⁰

It was reported that increased FeNO₅₀ combined with decreased small airway function indicated airway hyperresponsiveness, which had a good predictive value for positive bronchial excitation test. ¹¹ However, the value of FeNO₂₀₀ for predicting CVA remains unclear. This prospective study aimed to explore and compare the value of lower airway exhaled NO (FeNO₅₀, FeNO₂₀₀, and CaNO) alone or combined with small airway function for diagnosing CVA, and to determine the optimal cut-off value.

Methods

Subjects

Chronic cough patients who attended the Department of Pulmonary and Critical Care Medicine of Tongji Hospital from September 2021 to August 2022 were enrolled. After spirometry, histamine bronchial provocation test, FeNO, and other examinations, all of their clinical data were collected and analyzed (see Figure 1).

Figure 1.

CONSORT (Consolidated Standards of Reporting Trials) flow diagram of the study.

Patients who met all of the following criteria were eligible: (1) over 18 years old; (2) cough lasting no less than 8 weeks and with a normal chest radiograph; (3) forced expiratory volume in 1 s (FEV₁) / forced vital capacity (FVC) >70%, predicted FEV₁ >80%; (4) no corticosteroid was used in the past 4 weeks.

Patients were excluded if they met one of the following criteria: (1) current smokers or ex-smokers of ⩽2 years; (2) pregnant or lactating patients; (3) A history of acute upper respiratory tract infection within 8 weeks; (4) use of inhaled or oral corticosteroid within 4 weeks, or use of montelukast and long-acting β2-agonists in the previous week; (5) patients with severe cardiac insufficiency, severe liver and kidney insufficiency, mental and cognitive dysfunction, hearing and communication impairment; (6) multiple causes of chronic cough.

According to Chinese national guideline on diagnosis and management of cough (2021), ¹ the patient who met the following criteria was diagnosed with CVA: (1) chronic cough, often accompanied by a significant night cough; (2) positive bronchial provocation test; (3) positive to anti-asthma treatment.

Exhaled NO measurement

Exhaled NO was measured using the Nano Coulomb Breath Analyzer (Sunvou-CA2122, Wuxi, China), following the American Thoracic Society/European Respiratory Society (ATS/ERS) recommendations. ¹² Subjects were informed about inhaling NO-free air and exhaling via a mouthpiece at two constant flow rates: 50 ml/s, 200 ml/s. FeNO₅₀ and FeNO₂₀₀ were recorded. CaNO was calculated based on the linear model published by ERS: FeNO = CaNO + JawNO/VE. FeNO measurements were performed before spirometric assessments and bronchial provocation tests.

Spirometric measurement

Spirometric assessments were performed with a spirometer (Jaeger, Hoechberg, Germany) in accordance with the specifications and performance criteria recommended in the ATS/ ERS Standardization of Spirometry. ¹³

Histamine bronchial provocation test

Histamine bronchial provocation tests were performed with the Jaeger APS Pro system by using a Medic-Aid sidestream nebulizer, following the recommendations of the ATS/ERS Task force: Standardization of Lung Function Testing. ¹⁴ Provocative dose causing a 20% fall in FEV1 was recorded, and bronchial hyperresponsiveness (BHR) was defined as present if PD₂₀-FEV₁ <7.8 μmol.

Statistical analysis

The Kolmogorov–Smirnov method was used to test the normality of the distribution of all continuous variables. Normally distributed variables were presented as mean ± standard deviation and non-normally distributed variables were presented as median (interquartile range). The comparison of characteristic variables between CVA group and NCVA group was conducted by independent sample t-test, Chi-square test, and Mann–Whitney U test. Diagnostic values of single or combined measurements were calculated by constructing receiver operating characteristic (ROC) curves and measuring the area under the curves (AUC). The comparison between AUCs was performed using DeLong test. The value p < 0.05 was considered statistically significant. MedCalc 19.0.4 was used for ROC curve analysis and DeLong test, IBM SPSS Statistics 26.0 was used for other statistical analyses.

Results

Baseline characteristics

From September 2021 to August 2022, a total of 307 patients with chronic cough attended the Department of Respiratory and Critical Care Medicine of Tongji Hospital. Among them, 9 were excluded due to the absence of spirometry, bronchial provocation test, or exhaled NO measurement, and 15 were excluded due to multiple causes of chronic cough. Finally, 283 chronic cough patients with single cause were included in this study. In these 283 patients, 71 (25.1%) patients had a positive bronchial provocation test and were eventually diagnosed with CVA. The other 212 patients were enrolled in the NCVA group, including 34 (12.0%) patients with eosinophilic bronchitis (EB), 88 (31.1%) patients with gastroesophageal reflux-related chronic cough (GERC), 56 (19.8%) patients with upper airway cough syndrome (UACS), 18 (6.4%) patients with atopic cough (AC), and 16 (5.6%) patients with other causes [angiotensin-converting enzyme inhibitor (ACEI)-related cough, psychologic cough, postinfectious cough, and so on]. There were no significant differences in age, gender, body mass index (BMI), Leicester cough questionnaire (LCQ) score, and cough symptom score between the two groups (see Table 1).

Table 1.

Baseline characteristics of CVA and NCVA patients.

	CVA	NCVA	p-value
N (%)	71 (25.1)	212 (74.9)	–
Age, years	47.80 ± 15.92	44.58 ± 15.15	0.13
Female, n (%)	44 (62.0)	127 (59.9)	0.76
BMI, kg/m2	24.31 ± 3.11	24.87 ± 3.47	0.23
LCQ	14.50 ± 0.80	13.80 ± 1.00	0.57
Cough symptom score
Daytime	2 (1)	3 (1)	0.39
Nighttime	3 (1)	2(1)	0.09

BMI, body mass index; CVA, cough-variant asthma; IQR, interquartile range; LCQ, Leicester cough questionnaire; NCVA, non- cough-variant asthma; SD, standard deviation.

Data were presented as mean ± SD or median (IQR).

Comparison of spirometry and exhaled NO between CVA and NCVA

When the spirometry and exhaled NO were compared between the two groups, though the mean values of FEV₁ and FEV₁/FVC in CVA patients were within the normal range, they were all lower than those in NCVA patients (p < 0.01). In CVA patients, FEF_50%, FEF_75%, and MMEF significantly reduced, suggesting these patients may have SAD. FeNO₅₀, FeNO₂₀₀, and CaNO in CVA patients were significantly higher than those in NCVA patients (see Table 2).

Table 2.

Comparison of spirometry and airway inflammation between CVA and NCVA.

Characteristic variables	CVA (n = 71)	NCVA (n = 212)	p-value
FEV1, %predicted	90.14 ± 18.17	101.69 ± 12.58	< 0.01*
FEV1/FVC, %	76.74 ± 9.63	83.66 ± 6.41	< 0.01*
FEF25%, %predicted	57.00 (38.29)	81.25(39.30)	< 0.01*
FEF50%, %predicted	67.41 ± 28.15	92.16 ± 22.67	< 0.01*
FEF75%, %predicted	75.04 ± 26.10	95.81 ± 18.40	< 0.01*
MMEF, %predicted	63.55 ± 28.58	89.42 ± 24.15	< 0.01*
FeNO50, ppb	39.00 (39.00)	17.00 (12.00)	< 0.01*
FeNO200, ppb	17.00 (14.00)	8.00 (5.00)	< 0.01*
CaNO, ppb	5.00 (6.10)	3.50 (3.60)	< 0.01*

CVA, cough-variant asthma; CaNO, concentration of alveolar nitric oxide; FEF_50%, forced expiratory flow at 50% of forced vital capacity; FEF_75%, forced expiratory flow at 75% of forced vital capacity; FENO, fractional exhaled nitric oxide; FEV₁, forced expiratory volume in 1 s; FVC, forced vital capacity; IQR, interquartile range; MMEF, maximal mid-expiratory flow; NCVA, non-cough-variant asthma; ppb, parts per billion; SD, standard deviation.

Data were presented as mean ± SD or median (IQR).

^*p < 0.05.

Diagnostic accuracy of single measurement for CVA

There were significant differences in small airway parameters and FeNO levels between CVA and NCVA patients, so they may be potential indicators for CVA in chronic cough patients. The ROC curve was constructed and the AUC was calculated to evaluate their diagnostic value for CVA. FeNO₅₀ and FeNO₂₀₀ both had high AUC, while the diagnostic value of FeNO₂₀₀ was better than FeNO₅₀ (p = 0.04), and the AUCs of FEF_50%, FEF_75%, and MMEF are similar (see Table 3 and Figure 3). The AUC of CaNO was significantly lower than FeNO₅₀ and FeNO₂₀₀ (p < 0.05), showing a low diagnostic value. Therefore, CaNO was not included in the subsequent combined diagnostic analysis.

Table 3.

Optimal cutoff values for the prediction of CVA.

Parameter	Cutoff	AUC	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	PLR	NLR
FEV1	87.50	0.71	47.89	88.68	58.60	83.60	4.23	0.59
FEV1/FVC	78.79	0.72	51.95	82.55	53.20	85.80	3.39	0.49
FEF25%	58.60	0.72	59.15	81.13	51.20	85.60	3.14	0.50
FEF50%	73.50	0.75	60.56	80.19	50.60	85.90	3.06	0.49
FEF75%	77.90	0.74	57.75	83.49	53.9	85.50	3.50	0.51
MMEF	63.80	0.75	53.52	86.32	56.70	84.70	3.91	0.54
FeNO50	27.00	0.88	78.87	79.25	56.00	91.80	3.80	0.27
FeNO200	11.00	0.92	88.73	81.60	61.80	95.60	4.82	0.14
CaNO	3.60	0.66	73.24	53.26	34.00	85.40	1.54	0.51

AUC, area under the curve; CVA, cough-variant asthma; CaNO, concentration of alveolar nitric oxide; FEF_25%, forced expiratory flow at 25% of forced vital capacity; FEF_50%, forced expiratory flow at 50% of forced vital capacity; FEF_75%, forced expiratory flow at 75% of forced vital capacity; FENO, fractional exhaled nitric oxide; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; MMEF, maximal mid-expiratory flow; NLR, negative likelihood ratio; NPV, negative predictive value; PLR, positive likelihood ratio; PPV, positive predictive value.

Cutoff points of FEV₁, FEF_25%, FEF_50%, FEF_75%, and MMEF are in percentage predicted; the cutoff point of FeNO₅₀, FeNO₂₀₀, CaNO is in ppb.

Figure 3.

ROC curves of (a) FeNO₂₀₀ combined with FeNO₅₀, (b) FeNO₂₀₀ combined with MMEF, (c) FeNO₂₀₀ combined with FEF_50%, (d) FeNO₂₀₀ combined with FEF_75%, (e) FeNO₅₀ combined with MMEF, (f) FeNO₅₀ combined with FEF_50%, and (g) FeNO₅₀ combined with FEF_75% for CVA diagnosis. (a) AUC_Combined = 0.91 (95% CI, 0.87–0.94); AUC_FeNO200 = 0.92 (95% CI, 0.88–0.95; p = 0.02 compared with the combined model); AUC_FeNO50 = 0.88 (95% CI, 0.84–0.92; p = 0.04 and p = 0.18, compared with FeNO₂₀₀ alone and the combined model, respectively). (b) AUC_Combined = 0.93 (95% CI, 0.90–0.96); AUC_FeNO200 = 0.92 (95% CI, 0.88–0.95; p = 0.29, compared with the combined model); AUC_MMEF = 0.75 (95% CI, 0.70–0.80; p < 0.01, compared with FENO₂₀₀ alone and the combined model, respectively). (c) AUC_Combined = 0.93 (95% CI, 0.90–0.96); AUC_FeNO200 = 0.92 (95% CI, 0.88–0.95; p = 0.43, compared with the combined model); AUC_FEF50%=0.75 (95% CI, 0.70–0.80; p < 0.01, compared with FeNO₂₀₀ alone and the combined model, respectively). (d) AUC_Combined = 0.93 (95% CI, 0.89–0.96); AUC_FeNO200 = 0.92 (95% CI, 0.88–0.95; p = 0.49, compared with the combined model); AUC_FEF75%=0.74 (95% CI, 0.68–0.79; p < 0.01, compared with FENO₂₀₀ alone and the combined model, respectively). (e) AUC_Combined = 0.89 (95% CI, 0.85–0.92); AUC_FeNO50 = 0.88 (95% CI, 0.84–0.92; p = 0.71, compared with the combined model); AUC_MMEF = 0.75 (95% CI, 0.70–0.80; p < 0.01, compared with FeNO₅₀ alone and the combined model, respectively). (f) AUC_Combined = 0.89 (95% CI, 0.85–0.92); AUC_FeNO50 = 0.88 (95% CI, 0.84–0.92; p = 0.73, compared with the combined model); AUC_FEF50% = 0.75 (95% CI, 0.70–0.80; p < 0.01, compared with FeNO₅₀ alone and the combined model, respectively). (g) AUC_Combined = 0.89 (95% CI, 0.85–0.93); AUC_FeNO50 = 0.88 (95% CI, 0.84–0.92; p = 0.65, compared with the combined model); AUC_FEF75%=0.74 (95% CI, 0.68–0.79; p < 0.01, compared with FeNO₅₀ alone and the combined model, respectively).

AUC, area under the curve; MMEF, maximal mid-expiratory flow; FENO, fractional exhaled nitric oxide; ROC, receiver operating characteristic.

Distribution of patients with FeNO₅₀ >27 and FeNO₂₀₀ >11 in CVA

There was a 73.2% overlap between FeNO₅₀ >27 and FeNO₂₀₀ >11 in CVA cases. About 5.6% and 15.5% of CVA cases were exclusively predicted by FeNO₅₀ and FeNO₂₀₀, respectively. FeNO₅₀ ⩽27 and FeNO₂₀₀ ⩽11 occurred in 5.6% CVA cases (see Figure 2).

Figure 2.

Predicted proportion of CVA by FeNO₅₀ >27 or using FeNO₂₀₀ >11.

Diagnostic accuracy of FeNO combined with small airway parameters for CVA

The combination of FeNO₂₀₀ and FeNO₅₀ failed to improve the diagnostic value. Compared with the single parameter of small airway parameters, the combination of FeNO₂₀₀ or FeNO₅₀ with FEF_50%, FEF_75% and MMEF significantly improved the diagnostic value (see Figure 3). There was no difference between the AUC of FeNO₂₀₀ and the AUC of FeNO₅₀ combined with FEF_50%, FEF_75% or MMEF (p > 0.05). FEF_25% is not generally considered to reflect small airway function so that it was not included in the combined analysis.

Lower airway exhaled NO levels and correlation with small airway parameters in chronic cough patients with SAD and normal small airway function

SAD was diagnosed when at least two of the following three indicators of lung function test were less than 65% of predicted: MMEF, FEF_50%, and FEF_75%. The incidence of SAD was higher in CVA patients (35 cases, 49.3%) than in NCVA patients (26 cases, 12.3%; p < 0.01). Sixty-one patients (21.6%) with chronic cough had SAD, FeNO₅₀ and FeNO₂₀₀ in patients with SAD were significantly higher than those in patients with normal small airway function, and there was no difference in CaNO levels between them (see Figure 4).

Figure 4.

Comparison of (a) FeNO₂₀₀, (b) FeNO₅₀, and (c) CaNO levels in patients with SAD and patients with normal small airway function (NSAF).

In patients with SAD, FeNO₂₀₀ had moderate negative correlation with MMEF (p < 0.01, r = −0.35), FEF_75% (p < 0.01, r = −0.37), and FEF_50% (p = 0.01, r = −0.33); FeNO₅₀ was associated with MMEF (p = 0.02, r = −0.29), FEF_75% (p = 0.02, r = −0.30), and FEF_50% (p = 0.047, r = −0.25) as well. In patients with normal small airway function, FeNO₂₀₀ had no correlation with MMEF (p = 0.15), FEF_75% (p = 0.39), and FEF_50% (p = 0.07); FeNO₅₀ did not correlate with FEF_75% (p = 0.67), FEF_50% (p = 0.06), and MMEF (p = 0.13). There was no correlation between CaNO and small airway parameters in both groups (p > 0.05).

Diagnostic value of FeNO₂₀₀, FeNO₅₀ alone or in combination with small airway parameters for CVA in chronic cough patients with SAD and patients with normal small airway function

Compared with FeNO₅₀, FeNO₂₀₀ was superior in SAD patients. In chronic cough patients with normal small airway function, FeNO₅₀ and FeNO₂₀₀ had similar diagnostic value for CVA, whether it was alone or in combination with small airway parameters (see Table 4).

Table 4.

Comparison between the AUCs of FeNO₅₀ and FeNO₂₀₀ for predicting CVA in patients with SAD and patients with normal small airway function.

	FeNO200	FeNO50	p-value
SAD patients
Single	0.95	0.88	0.02*
Combined with MMEF	0.96	0.91	0.04*
Combined with FEF75%	0.97	0.93	0.07
Combined with FEF50%	0.96	0.92	0.05
NSAF patients
Single	0.92	0.88	0.17
Combined with MMEF	0.92	0.88	0.11
Combined with FEF75%	0.91	0.87	0.10
Combined with FEF50%	0.92	0.87	0.11

AUC, area under the curve; CVA, cough-variant asthma; FEF_50%, forced expiratory flow at 50% of forced vital capacity; FEF_75%, forced expiratory flow at 75% of forced vital capacity; FENO, fractional exhaled nitric oxide; MMEF, maximal mid-expiratory flow; NSAF, normal small airway function; SAD, small airway dysfunction.

^*p < 0.05.

Discussion

As a special type of asthma, CVA is one of the most common causes of chronic cough. In this study, CVA patients accounted for 25.1% of all patients with chronic cough, which is consistent with our previous study. ³ In this prospective study, both FeNO₂₀₀ and FeNO₅₀ had high diagnostic value for CVA, whether alone or in combination with small airway parameters. FeNO₂₀₀ >11 parts per billion (ppb) contributed strongly for differentiating CVA from chronic cough, especially in SAD patients.

Small airway refers to the bronchioles with the internal diameter less than 2 mm. Airway resistance is inversely proportional to the cross-sectional area of the airway. Cross-sectional area of the small airway is much larger than that of the airway with a diameter >2 mm (more than 100 cm²). However, small airway resistance only accounts for less than 20% of the total. Thus, there can be no clinical symptoms and signs when lesions occur. In recent years, more and more studies have proved that the small airway is no longer the ‘silent zone of lung’. The inflammatory reaction of asthma is not limited in the proximal airway, but also exists in the distal small airway. The tissue cells with positive interleukin-5 (IL-5) and IL-4 messenger RN (mRNA) in the distal airway of asthma patients are significantly increased compared with those in non-asthma patients, and the expression of IL-5 mRNA in the distal airway is higher than that in the central airway, and more activated eosinophils were found in the small airways of asthma patients than in airways with an inner diameter of >2 mm, suggesting a similar but more severe inflammatory process in the distal airways than proximal airways. ¹⁵ Animal experiment has confirmed that SAD was an important feature of asthma and was associated with airway inflammation, mucus hypersecretion, and airway hyperresponsiveness. ¹⁶ Global multicenter prospective studies also showed that SAD is widespread in patients with all severities of asthma, especially in severe asthma. ⁹ SAD including airway wall thickening, airway narrowing, and hyperresponsiveness can lead to poorly controlled asthma, frequent acute attacks, nocturnal asthma attacks, exercise-induced asthma attacks and worse airway hyperresponsiveness.^17,18 The presence of SAD in patients with CVA has been confirmed, which was less severe than that in patients with asthma. ¹⁹ CVA patients with SAD had a higher recurrence rate. ²⁰ SAD was diagnosed when at least two of the following three indicators of lung function were less than 65% of predicted: MMEF, FEF_50%, and FEF_75%. ²¹ We found that about half of the CVA patients had SAD, and the incidence was significantly higher than that of NCVA patients, which was consistent with the study of Yi et al. ¹⁰ FEV₁ and FEV₁/FVC, which reflect the function of the large airways, were still within the normal range in CVA patients, but were lower than those in NCVA patients. The FEF_50%, FEF_75%, and MMEF of CVA patients were significantly reduced, suggesting that the whole airway function of CVA patients was worse than that of NCVA patients. Moreover, CVA patients have SAD, the clues of CVA may be found earlier according to small airway parameters. While small airway parameters such as FEF_50%, FEF_75%, and MMEF were used alone for the diagnosis of CVA, the AUCs were all around 0.75, indicating that only small airway parameters were insufficient to diagnose CVA.

The most important pathological change of bronchial asthma is chronic airway inflammation, and the degree of airway inflammation is closely related to airway hyperresponsiveness, which is the root cause of clinical symptoms. ²² FeNO is a marker of type 2 airway inflammation, ⁷ and has been strongly recommended in the diagnosis of adult asthma. ²³ Type 2 inflammation in the central airway, small airway, and alveoli can be reflected by FeNO at different exhalation velocity: FeNO₅₀ was measured at exhalation flow rates of 50 ml/s, reflecting the inflammation of central airway and bronchus; FeNO₂₀₀ was measured at exhalation flow rates of 200 ml/s, reflecting the inflammation of peripheral small airway. The application of FeNO₅₀ in the diagnosis of CVA has been explored. Zhang et al. ²⁴ completed a meta-analysis involving 12 studies and 1968 patients, confirming that FeNO₅₀ can be used as a diagnostic indicator for CVA, this study also showed that FeNO₅₀ had good diagnostic value for CVA. It is known that CVA patients have small airway inflammation and SAD, so FeNO₂₀₀ can be used for the diagnosis of CVA in theory. This study confirmed that FeNO₂₀₀ indeed had a good diagnostic value for CVA and was superior to FeNO₅₀. We found that FeNO₅₀ >27 ppb occurred in 5.63% of CVA patients with a normal FeNO₂₀₀, and FeNO₂₀₀ >11 ppb occurred in 15.5% of CVA patients with a normal FeNO₅₀, most CVA patients were diagnosed using FeNO₅₀ and FeNO₂₀₀ together. Therefore, it is possible for clinicians to use FeNO₅₀ and FeNO₂₀₀ together to improve the diagnostic rate of CVA and avoid missed diagnosis caused by single usage of FeNO₅₀ or FeNO₂₀₀. In addition, more than one-fifth of patients with chronic cough had SAD. FeNO₂₀₀ in chronic cough patients with SAD are moderately and negatively correlated with small airway parameters. FeNO₅₀ and FeNO₂₀₀ in SAD patients are significantly higher than those in patients with normal small airway function. What’s more, FeNO₅₀ and FeNO₂₀₀ had no correlation with small airway parameters in chronic cough patients with normal small airway function, suggesting the important role of SAD in chronic cough and the close relationship between airway inflammation and SAD, especially in CVA patients.

Bao et al. ¹¹ found that increased FeNO₅₀ combined with decreased small airway function indicated airway hyperresponsiveness, which had a good predictive value for positive bronchial provocation test. Therefore, increased FeNO₅₀ combined with small airway parameters may replace bronchial provocation test to diagnose CVA. Different studies showed that the diagnostic value of FeNO₅₀ combined with MMEF or FEF_50% in CVA is significantly higher than that of FeNO₅₀ or MMEF alone.^25,26 In this study, FeNO₅₀ combined with small airway parameters also had a good diagnostic value for CVA. We further focused on small airway inflammation and dysfunction, finding the diagnostic value of FeNO₂₀₀ was consistent with that of FeNO₅₀ combined with small airway parameters. From this point, FeNO₂₀₀ could play an important role in the diagnosis of patients who have difficulty in completing spirometry.

Generally, the calculation of CaNO requires three kinds of exhalation flow rates: low, medium, and high, ⁵ which is challenging in practice. Patients need to maintain accurate movements in the three operations, and the error rate will increase, which may affect the final result. We found the diagnostic value of CaNO for CVA was inferior to FeNO₅₀ and FeNO₂₀₀. There was no difference between the CaNO in chronic cough patients with SAD and in patients with normal small airway function. In addition, CaNO had no correlation with small airway parameters whether patients had SAD or not. Compared with the complex operation and calculation of CaNO, FeNO₂₀₀ can reflect peripheral airway inflammation more directly and accurately, which has been confirmed by Paredi et al. ⁶ Besides, FeNO₂₀₀ was moderately correlated with small airway parameters in patients with SAD. Taken together, we are more inclined to use FeNO₂₀₀ instead of CaNO to evaluate small airway inflammation in patients with chronic cough.

There are some limitations in this study. First of all, the experimental design of this study is relatively simple. Although we have met the requirements of sample size estimation, it is still not large enough. Second, some patients with chronic cough who did not complete spirometry, bronchial provocation test, or exhaled NO measurement and patients with multiple causes of chronic cough were not included in this study, which may result in some biases.

In conclusion, small airway inflammation and SAD play an important role in chronic cough. CVA patients have chronic small airway inflammation and decreased small airway function. FeNO₂₀₀ >11 ppb contributed strongly for differentiating CVA from chronic cough, especially in patients with SAD.