Accepted Manuscript Validation of ATS clinical practice guideline cut-points for FeNO in asthma Maria Jeppegaard, Sandra Veidal, Asger Sverrild, Vibeke Backer, Celeste Porsbjerg PII: S0954-6111(18)30296-8 DOI: 10.1016/j.rmed.2018.09.014 Reference: YRMED 5531 To appear in: Respiratory Medicine Received Date: 5 April 2018 Revised Date: 15 September 2018 Accepted Date: 19 September 2018 Please cite this article as: Jeppegaard M, Veidal S, Sverrild A, Backer V, Porsbjerg C, Validation of ATS clinical practice guideline cut-points for FeNO in asthma, Respiratory Medicine (2018), doi: https:// doi.org/10.1016/j.rmed.2018.09.014. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Validation of ATS clinical practice guideline cut-points for FeNO in asthma Maria Jeppegaard 1, MD; Sandra Veidal 1, MD; Asger Sverrild 1, MD, PhD; Vibeke Backer 1, MD, DMedSci, Celeste Porsbjerg 1, MD, PhD 1 Respiratory Research Unit, Department of Respiratory Medicine, Bispebjerg University Hospital, Copenhagen, Denmark Abstract: 248 words Word count: 3977 words (ekskl. Abstract, reference and tables) Corresponding author: Maria Jeppegaard, MD Respiratory Research Unit, Department of Respiratory Medicine, Bispebjerg University Hospital, Copenhagen, Denmark Mail: maria.jeppegaard@gmail.com
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 Abstract: Background: The American Thoracic society (ATS) has suggested using fractional exhaled nitric oxide (FeNO) to monitor the level of eosinophilic (EOS) airway inflammation in asthma, but validation of the proposed cut-points is required in real-life populations. Objective: To validate FeNO cut-points suggested by ATS in relation to sputum EOS count in a real life population of asthma patients. Methods: All patients referred consecutively over a 12-months period for specialist assessment of asthma, were examined with FeNO and induced sputum, and re-examined 12 months later. The predicted values of a positive and a negative test (PPV and NPV) for a cut off 3% EOS in sputum were calculated. Change in FeNO was defined in accordance with ATS (>20% or 10ppb if FeNO was <50ppb). Results: 144 adult asthma patients were examined (59% female). Low FeNO (< 25ppb) at baseline was found in 94 (65%), FeNO between 25-50ppb in 34 (24%) subjects and high FeNO >50ppb in 16 (11%) subjects. The PPV for FeNO >25ppb and >50ppb to predict EOS 3% was 45% and 77%, NPV was 88% and 83%. The sensitivity decreased from 70% to 37% at the >50ppb cut-off. A significant reduction in FeNO was associated with a reduction in sputum EOS (p=0.01). Conclusion: The findings support the validity of the FeNO cut-points suggested by ATS to monitor eosinophilic airway inflammation in asthma. However, in this real-life population, a large proportion of patients had intermediate FeNO values, which may limit the clinical usefulness of the ATS FeNO cutpoints. Word 248 Page 2 of 22
73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 Keywords: Nitric oxide, eosinophilic airway inflammation, asthma. Abbreviations used: FeNO: Fractionel exhaled nitric oxide. ACQ: Asthma Control Questionnaire. ICS: Inhaled corticosteroids. EOS: Eosinophilic airway inflammation measured in induced sputum. PPV: Positive predictive value. NPV: Negative predictive value. Page 3 of 22
88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 Background: Asthma is increasingly recognized as a heterogeneous disease, with a variable underlying pattern of airway inflammation. At present, monitoring airway inflammation is not included in current asthma guidelines (GINA) [1]. Thus, anti-inflammatory treatment in asthma is titrated according to the level of symptom control, but there is no clear correlation between symptoms and the level of airway inflammation [2]. There has been an increasing interest in inflammatory phenotyping of asthma in order to target asthma treatment more specifically, and hopefully achieve better asthma control using relevant doses of inhaled steroid (ICS). Induced sputum is considered the gold standard test for inflammatory phenotyping in asthma, but is time-consuming and not implemented outside specialist centers with special interest in difficult asthma. The American Thoracic society (ATS) has suggested using fractional exhaled nitric oxide (FeNO) to monitor the level of eosinophilic (EOS) airway inflammation [3, 4]. FeNO has the advantage of being standardized, quick, non-invasive, simple, and easy to reproduce [5]. Eosinophilic airway inflammation is associated with a treatment response to inhaled ICS, which leads to, that FeNO can be used as a predictor of ICS responsiveness, to monitor airway inflammation and to evaluate the adherence to anti-inflammatory medications [3, 4, 6]. Different optimal FeNO cut-points values have been suggested in different studies, but ATS recommendations suggest that a low FeNO < 25 ppb, measured at exhaled flow-rate of 50 ml/s, implies that there is no steroid-sensitive inflammation present, while a FeNO > 50 ppb indicates eosinophilic airway inflammation and responsiveness to corticosteroids [3, 7]. A FeNO between 25-50 ppb should be interpreted cautiously with reference to the clinical context. The ATS task force has, based on the normal diurnal variations in FeNO of asthma patients at approximately 20%, suggested that a change greater than 20% for FeNO>50 ppb or more than 10 ppb for FeNO<50 ppb determines a significant change in FeNO and thereby a significant change in the degree of eosinophilic airway inflammation. However, the ATS cut-points have not yet been validated in more general, unselected populations of asthma patients. Nor do we know whether these changes in FeNO will translate into significant changes in airway eosinophilia, or in the context of generally accepted clinical outcomes such as a change in asthma control scores. In this study we use the ATS cut-points in predicting sputum eosinophilia, and investigate whether a change of FeNO over 12 months corresponds with changes in sputum EOS, in order Page 4 of 22
120 121 to assess the ATS guidelines applicability in a real life setting. Page 5 of 22
122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 Material and methods: Design: The MAPOut II study is a prospective cohort study of patients referred over a 12-month period (March 2011 to April 2012) to the outpatient respiratory clinic at Bispebjerg University Hospital, Copenhagen, for specialist management of asthma or suspected asthma. The patients were followed-up 12 months later. Exclusion criteria were respiratory diseases other than asthma (e.g. sarcoidosis, COPD), age < 15 years, individuals older than 40 years with a smoking history of more than 10 pack-years, pregnancy and recent respiratory infection (< 6 weeks). The study was approved by the local ethical committee (H-3-2011-121) and all subjects signed an informed consent. Material: Baseline: A total of 256 patients were referred and 190 subjects were enrolled in the study. In 151 subjects with asthma (as defined below), FeNO and sputum were obtained in 144 and 110 subjects, respectively (Figure 1). 12-month follow-up: A total of 104 patients with asthma participated of which 90 subjects had a successful FeNO measurement both at baseline and at follow up. The subjects were divided into three groups: a group without a significant change in FeNO ( FeNO< 20%/10ppb), a group with a significant decrease in FeNO (( FeNO> 20%/-10ppb) and a group with a significant increase in FeNO ( FeNO>+20%/+10ppb). A representative sputum sample from both baseline and follow up were obtained in 54 subjects (Figure 1). Page 6 of 22
Enrollment ACCEPTED Assessed for MANUSCRIPT eligibility (n = 256) 145 146 147 148 Analysis: Validation of ATS cut-off points for FeNO Analysis: Validation of ATS cut-off points for changes in FeNO to determine significant changes in airway inflammation. Baseline Population n = 190 (96.4% of enrolled subjects) Subject with asthma* and a FeNO-value: n =144 (75.8%) *At least one positive diagnostic test (n=138 (95.8%)) or doctor s diagnosis + ICS (n=6 (4.5%) Follow up after 12 months Asthmatics who completed follow-up: n=90 (62.6%) Excluded: 21,5% (n=55) Not meeting inclusion criteria: 29,1% (n=16) Pregnant 7,3% (n=4) Age 45+ and >10 packyears 20% (n=11) >15 years old 1,8% (n=1) Declined to participate: 60% (n=33) Other reasons: 11% (n=6) o Urgent treatment 5,5% (n=3) o Language problems 5,5% (n=3) Dropped out during assessment visits: n = 11 (4.3%) Nonspecific respiratory symptoms n = 39 (20.5%) Failed to obtain FeNO n= 7 (3.7%) Dropped Out before Follow up: n=52 (36.1%) Failed to obtain FeNO at follow up: n=2 (1.4%) Subject with sputum at Baseline and Follow up n=54 (60%) Figure 1. Consort statement. Flow diagram describing the study population: all patients referred for specialist management of known or suspected asthma were screened and participating were offered to eligible subjects. In this present analysis, only subjects who were concluded to have asthma, and who had successfully obtained FeNO value were included. Page 7 of 22
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 Methods: Participants were assessed using a three-day asthma evaluation program over 2 weeks. The first visit included measurement of a spirometry with a reversibility test after 4 puffs of terbutaline (2 mg). At the second visit, a skin prick test with 10 standard allergens and a methacholine challenge test were performed. At the last visit, FeNO and responsiveness to inhaled mannitol were measured, and sputum induction was performed. All participants were interviewed using a questionnaire based on the GINA-guidelines, and answered a number of questionnaires including Asthma Control Questionnaire (ACQ) [8]. In the 2-week period between the first and the last visit, peak expiratory flow (PEF) was monitored (morning and evening). Prior to each visit, the participants were instructed to an ICS washout (12 hours for ICS, 24 hours for ICS+LABA). Definition of asthma: Asthma was defined as symptoms suggestive of asthma, as well as either variable airflow obstruction (reversibility to beta-2-agonist ( 12% and at least 200 ml), peak flow variability (PEF ( 100 20%)), airway hyperresponsiveness to either mannitol (PD15 635 mg) or methacholine (PD20 8 µmol) or a doctors diagnosis of asthma at the time of referral and current use of inhaled corticosteroids, as these patients could have a negative test panel in spite of having asthma. Skin prick test: Skin prick test was performed a panel of allergen extracts including: birch, grass, mugwort, horse, cat, dog, dermatophagoides pteronyssinus and D. farinae, Alternaria, and Cladosporium herbarum [9]. A minimum of 3mm ((dia1+dia2)/2) defined a positive response. Fractional exhaled nitric oxide (FENO): FENO was analyzed with NioxMinor from Aerocrine and was performed following the recommendations of the ERS and ATS [3]. In accordance with the ERS/ATS recommendations, low FeNO was defined as a level <25 ppb, an intermediate level as 25-50 ppb and a high level as >50 ppb Two measurements was made in each subject and the average of the two were used in the analysis. Spirometry: Lung function was done using maximum expiratory flow volume according to the standards specified by the ERS and ATS [10], using an Easyone spirometer. A reversibility test was performed, using 4 puffs of 0.5 mg terbutaline, followed by lung function measurement after 15 minutes [11]. A significant reversibility was defined as a 12% increase in FEV1 (and minimum 200mL). Predicted normal values based on sex, height, weight and age for FEV1, FVC, FEV1/FVC ratio were calculated using NHANES reference values [12]. Page 8 of 22
180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 Peak-Flow variability (PEF): All patients were asked to measure PEF for 2 weeks, in the morning and in the evening. Variation was measured as the highest PEF during the two weeks minus the lowest PEF during the two weeks and divided by the highest PEF during the two weeks *100. A significant variability was defined as variation of at least 20%. PEF ( 100 20%) Mannitol bronchial provocation: The participants inhaled an empty capsule followed by capsules with increasing doses of mannitol until maximum cumulative dose (635mg) had been administered or a 15% reduction in FEV1 (PD15) had been reached [13]. Patients with a cumulative dose of mannitol of 635 mg mannitol and a decrease in FEV1 15% were defined as having a positive test (AHR). For subjects with a positive test, the sensitivity to mannitol was calculated as the dose resulting in a 15% decrease in FEV1 (PD15). For all subjects, the degree of responsiveness was calculated at the Response Dose Ratio (RDR): Percent fall in FEV1/pr. mg mannitol administered. Methacholine bronchial provocation: A dosimetric method was used to measure airway responsiveness to inhaled methacholine [14]. A Jaeger dosimeter generated the aerosols in the test solution. The patient inhaled a dose of isotonic saline, followed by five successively increasing doses of methacholine ranging from 0.06 to 7.8 µmol, until a cumulated dose of 7.8 µmol or a 20% decrease in FEV1 was reached. Patients with a cumulative dose of methacholine of 7.8 µmol and a decrease in FEV1 20% were defined as having a positive test (AHR). For patients with a positive test, the sensitivity to methacholine was calculated as the dose resulting in a 20% decrease in FEV1 (PD20). For all patients, the degree of responsiveness was calculated at the Response Dose Ratio (RDR): Percent fall in FEV1/pr. µmol methacholine. Sputum induction: Sputum induction was conducted according to ERS recommendations [15]. After inhalation of 1 mg terbutaline, sputum was induced by inhalation of hypertonic saline in increasing concentrations (3%, 4% and 5%) for three time periods each of 5 min (total duration 15 min). Sputum plugs were selected and processed and cytospins were prepared using standard methods and a differential cell count was performed [16]. Eosinophilic airway inflammation was defined as 3% eosinophils in the induced sputum [17]. Page 9 of 22
209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 Statistical analysis: Statistical analysis was performed with SPSS Statistics (21.0.0.0). Continuous variable are reported as mean (SD) for normally distributed data and median (min-max) for non-parametric data. ANOVA tests and unpaired t-tests were used for comparison of parametric continuous variables, Chi-2 analysis for categorical variables. Wilcoxon signed ranks test and Kruskal-Wallis test were used in non-parametric data. Sputum eosinophils 3% are the gold standard to define eosinophilic airway inflammation. Sensitivity is defined as the ability of a clinical test to correctly identify subjects with the disease, i.e. the proportion of subjects with sputum eosinophils 3%, who have a high FeNO (positive test). The specificity is defined as ability of the test to correctly identify subjects without the disease, i.e. the proportion of subjects with sputum eosinophils <3%, who have a low FeNO (negative test). The positive predictive value is defined as the probability that subjects with a positive test truly have sputum eosinophils 3%, whereas the negative predictive value is defined as the probability that subjects with a negative test truly have sputum eosinophils < 3% [18]. A p- value of 0.05 or less was defined as significant. To establish the optimal and most effective cut-point for FeNO, we identified the Youden index on the data from the receiver operating curve (ROC) and found the associating FeNO cut-point [19]. The Youden index, which is a used summary measure of the ROC curve, is defined as =!"#$+%#&#!"#$ 1' and is a way to evaluate the effectiveness of a test/biomarker. The index has a range between 1 (=complete separation of the distribution of the marker values for the diseased and thereby a perfect test) and 0 (=complete overlap and thereby a useless test). Furthermore we calculated the positive and negative predicting values for this cut-point as well as the sensitivity and specificity. To adjust for smoking, we applied a reduced FeNO cut-point to 18 ppb among smokers and calculated sensitivity, specificity, PPV and NPV to see if smoking had any impact on the results. The 18 ppb cut-point is based on our previous reports on the level of FeNO in smokers with asthma [20]. Page 10 of 22
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 Results: Validation of ATS cut-off points for FeNO In the 144 subjects with asthma who had FeNO measured, the median patient age was 30 years (15-63 years) with 59% (n=85) females. In total, 61% (n= 87) had a positive skin prick test. The mean ACQ score was 1.2 (0-4.6) and 40% (n=55) had an ACQ score >1.5 (Table 1). FeNO was < 25 ppb in 94 subjects (65%), > 50 ppb in 16 subjects (11%) and between 25-50 ppb in 34 subjects (24%) (Table 1). At baseline 65 (45%) subjects were using ICS/LABA+ICS. There was no difference in the ICS-use between subjects in the group with low, intermediate or high FeNO (p=0.89) (Table 1). Table 1. Characteristics of included subjects. All subjects with asthma FeNO<25 ppb FeNO: 25-50 ppb FeNO>50 ppb p-values No. Patients 144 94 34 16 - Age (years, median(min-max)) 30 (15-63) 28 (15-63) 32 (17-63) 30 (16-57) 0.16 Sex, Female (%) 85 (59%) 58 (62%) 20 (59%) 7 (44%) 0. 40 Height (cm, mean±sd) 174.7 ± 7.9 174.1 ± 8.2 176.7 ± 7.4 173.6 ± 7.1 0.23 Weight (kg, mean±sd) 72.9 ± 15.4 71.3 ± 13.3 79.2 ± 20.7 68.7 ± 9.4 0.02* Smoking (%) 12 (9%) 10 (11%) - 2 (13%) 0.15 FeNO (median(min-max)) 18.8 (3-184) 15.0 (3-24.5) 33 (25.5-49.5) 83.3 (54-184) 0.00* FEV1 (L, mean ± SD) 3.6 ± 0.8 3.54 ±0.8 3.71 ± 0.95 3.43 ± 0.7 0.53 FVC (L, mean ± SD) 4.6 ± 2.7 4.3 ± 1.0 4.76 ± 1.2 4.34 ± 0.8 0.32 Atopy positive (%) 87 (61%) 50 (53%) 26 (77%) 11 (73%) 0.04* ACQ score (median(min-max)) 1.2 (0-4.6) 1.2 (0-4.4) 1 (0-3.8) 1.3 (0.4-4.6) 0.34 ACQ > 1,5 (%) 55 (40%) 38 (42.7%) 9 (31%) 7 (44%) 0.51 ICS/combi use (%) 65 (45%) 42 (45%) 16 (47%) 6 (40%) 0.89 FeNO, fractional exhaled nitric oxide; Atopy positive, positive skin prick test to the standard inhalation panel; ACQ, Asthma Control Questionnaire; ICS, inhalation corticosteroid. *Significant, p-value <0.05. Among 110 (76%) asthma patients with sputum analysis, 27 subjects (25%) had eosinophilic airway inflammation (sputum eosinophilia 3%). Of the 110 subjects, 108 had a positive objective asthma test (either positive reversibility test, bronchial challenge test or PEF variation), and only two were included solely based on a doctor diagnosis of asthma and current use of ICS. Page 11 of 22
257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 At baseline, the proportion of subjects with eosinophilic airway inflammation was higher in the group of subjects with FeNO>50 ppb (76.9%) compared to subjects with FeNO<25 ppb (12%, p<0.01) or FeNO between 25-50 ppb (31%, p<0.01) (Table 2). Table 2. The distribution of sputum eosinophilia between FeNO-groups. All subjects with asthma FeNO<25 ppb FeNO: 25-50 ppb FeNO>50 ppb p-values No. Patients 110 68 29 13 0,41 EOS % **(median (min-max)) 0.5 (0-34) 1.0 (0.3-10.3) 1.5 (0-15.8) 12.4 (0-34) P<0.001* EOS 3% (%) 27 (25%) 8 (12%) 9 (31%) 10 (77%) P<0.001* EOS 3% + ICS 18 8 5 5 0.013* EOS 3% without ICS 9 0 4 5 0.00* **EOS: Eosinophilic airway inflammation in percent measured in induced sputum. *Significant, p-value <0.05. The sensitivity, specificity and positive and negative predictive values for identifying sputum eosinophils 3% were calculated for the respective cut-points for FeNO and are shown in Table 3a. The sensitivity of a cut-point for FeNO > 25 ppb was 70%, decreasing to 37% at the cut-point for FeNO > 50 ppb, whereas the specificity was 72% at the cut-point for FeNO >25 ppb, and increasing to 96% at the 50 ppb cut-point. The positive predictive value of > 25 ppb was 45%, and with a cut- point of 50 ppb, the PPV increased to 77%. The negative predictive value (NPV) of > 25 ppb and FeNO > 50 ppb were 88% and 83%. We made the same calculation were we differentiated between steroid naïve subject and subjects with ICS, these results are shown in table 3b. Table 3a. Predictive values for two FeNO cut-offs in identifying sputum eosinophilia 3%: Cut off FeNO>25 ppb Cut off FeNO>50 Sensitivity 70% 37% Specificity 72% 96% Positive Predictive Values 45% 77% Negative Predictive Values 88% 83% Page 12 of 22
281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 Table 3b. Predictive values for two FeNO cut-offs in identifying sputum eosinophilia 3% for subject with and without ICS treatment: Without ICS With ICS Cut off FeNO>25 ppb Cut off FeNO>50 Cut off FeNO>25 ppb Sensitivity 100% 71% 56% 28% Specitivity 76% 93% 71% 97% PPV 41% 56% 50% 83% NPV 100% 96% 75% 72% Cut off FeNO>50 A receiver operator characteristics (ROC) curve was constructed (figure 2a) to investigate the overall accuracy of FeNO to diagnose current eosinophilic inflammation 3%. The area under the curve (AUC) was 80% (95% CI: 69% to 90%). The same ROC curves was made for subjects without and with ICS treatment (Figure 2b and figure 2c), where the AUC where 96% (95% CI: 91% to 100%) and 72% (95% CI: 57% to 87%), respectively. By using the Youden index (=0.47) to find the most optimal FeNO cut-point, we identified a cut- point of 22 ppb and the associating sensitivity and PPV were 79% and 44.9%, respectively. The specificity and NPV were 68% and 90%. When adjusting for smoking, the sensitivity and positive predictive value remained similar at 81% and 34%. The specificity and negative predictive values were 52% and 90%. Page 13 of 22
297 298 299 300 307 317 c. 301 302 303 304 305 306 a. b. 308 309 310 311 312 313 314 315 316 Figure 2. a. ROC curve: FeNO as a predictor of EOS>3% (AUC: 0.798). b. ROC curves: FeNO as a predictor of EOS>3% in steroids naïve subjects (AUC: 0.958). c. ROC curves: FeNO as a predictor of EOS>3% in subjects using ICS (AUC: 0.717). Page 14 of 22
318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 Validation of ATS cut-points for changes in FeNO to determine significant changes in airway inflammation. A total of 90 asthmatic subjects participated in the 12-months follow-up (48 (53.3%) female, median patient age: 32 years (15-63 years)). When applying the cut-points for delta values suggested by ATS at >20% for FeNO>50 ppb or more than 10 ppb for FeNO<50 ppb, we found a non-significant change in FeNO in 64 subjects (71.1%), a significant decrease in FeNO (> 20% or 10 ppb) in 20 subjects (22.2%), and a significant increase (> 20% or 10 ppb) in FeNO in 6 subjects (6.7%) (Table 4). Table 4. Changes from baseline to follow up between subjects with a non-significant change, subjects with a significant fall and subjects with a significant increase in FeNO**. Non-significant change in FeNO (n= 64 ) Significant fall in FeNO (n= 20) Significant increase in FeNO (n= 6) Baseline Follow up p-values Baseline Follow up p-values Baseline Follow up p-values FeNO (median(min-max)) 17 (6-59) 17 (8-54.5) 0.37 45.3 (21.5-184) 19 (8-45.5) 0.00* 19.5 (11-49.5) 56.5 (25-129.5) 0.03* FEV1 (L,mean(SD)) 3.4 (0.8) 3.5 (0.8) 0.09 3.6 (1.0) 3.7 (1.0) 0.08 3.3 (0.6) 3.1 (0.5) 0.08 FVC (L,mean(SD)) 4.3 (0.8) 4.4 (0.9) 0.00 4.6 (1.2) 4.8 (1.3) 0.07 4.5 (1.2) 4.5 (1.2) 0.30 EOS % (median(min-max)) ACQ (median (min-max)) 0.3 (0-15.8) 1.1 (0-4.6) 0 (0-38) 0.37 1.3 (0-29.5) 0.4 (0-14) 0.01 0.7 (0-8.75) 4.0 (0.5-7.5) 0.18 0.6 (0-4.2) 0.01 1.2 (0-4.4) 0.5 (0-3) 0.049* 0.8 (0-2.2) 0.6 (0-1.8) 0.04* **Significant fall in FeNO: change 20% for FeNO>50 ppb or 10 ppb for FeNO<50 ppb. Significant rise in FeNO: change 20% for FeNO>50 ppb or 10 ppb for FeNO<50 ppb. * Significant, p-value <0.05. In subjects with a significant decrease in FeNO, a decrease in percent eosinophilia in induced sputum from 1.25% (0-29.5%) to 0.4% (0-14%), p=0.01) was observed, as opposed to patients without a decrease in FeNO. The spearman correlation coefficient between the change in FeNO and the change in percent eosinophils in induced sputum was r=0.4 (p=0.004). Overall, after 12 months specialist management, there was a significant change in the ACQ score within all three FeNO-groups (Table 4), although only clinically relevant (>0.5) in patients who had either a stable FeNO or a decrease in FeNO. Furthermore, when comparing the three groups, there was no difference between the change in ACQ, EOS or FEV1 between the groups, apart from a significant decrease in percent EOS in the subjects with a significant decrease in FeNO compared to the other two groups (Table 5). In the 32 subjects with a significant fall in ACQ > 0.5, 20 had a FeNO < 25 ppb at baseline and follow up. Furthermore, six had a FeNO at 25-50 ppb at baseline, where three had a significant fall Page 15 of 22
347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 in FeNO > 10 ppb. At last, six had a FeNO > 50 ppb at baseline, where five had a significant fall in FeNO > 20% to the intermediate group with FeNO between 25-50 ppb. Table 5. Comparison between the changes in FEV1, FVC, EOS and ACQ score between subjects with a nonsignificant change, subjects with a significant change and subjects with a significant rise in FeNO. Nonsignificant change Significant fall Significant increase P-values in FeNO (n=64) in FeNO (n=20) in FeNO (n=6) FeNO (ppb, median(min-max)) -0.5 (-10-24) -26.3 (- 140-(-12)) 19.8 (11.5-106) 0.000* FEV1 (L, median (min-max)) 0.03 (-0.58-0.94) 0.01 (-0.23-0.61) -0.14 (-0.57-0.03) 0.06 FVC (L, median (min-max)) 0.17 (-0.53-1.18) 0.11 (-0.28-0.68) 0.04 (-0.18-0.67) 0.73 EOS "%$(median (min-max))** 0 (-7.5-26.5) -1.16 (-22-1) 3.63 (0.5-6.75) 0.008* ACQ (median (min-max)) -0.4 (-4.6-2.2) -0.2 (-2.8-1.2) -0.2 (-0.4 0) 0.84 *Significant, p-value <0.05 **data on 54 subjects. An analysis was performed of potential differences in FEV1 percent of predicted, FeNO, eosinophilia in sputum and asthma control questionnaire (ACQ) score between subjects who participated in follow up and subjects who dropped out before follow up. Drop-outs were similar to subjects who completed the study, apart from the drop-outs had a significantly higher ACQ score (1.6 (0-4.4) vs. (1.2 (0-4.6) (p=0.02)). Page 16 of 22
364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 Discussion: In an asthma population of subjects referred to a tertiary outpatient clinic, a high FeNO (> 50 ppb) was associated with a high likelihood of sputum eosinophils 3% (PPV=77%), whereas eosinophilia was unlikely in asthma patients with a low FeNO (<25 ppb) (NPV=88%). However, the high cut-point of 50 ppb was associated with a very low sensitivity (37%), reflecting that in a group of asthma patients referred for specialist assessment, about one third (33%) of patients with increased sputum eosinophils were localized in the intermediate group between 25-50ppb. These results are supported by a previous retrospective study by Schleich et al. with 295 unselected asthmatics, where a cut-point of FeNO < 25 ppb had a negative predictive value of 79% to predict sputum eosinophils (sensitivity 78%, specificity 60%) and FeNO> 50 ppb had a positive predictive value of 75% (sensitivity 56%, specificity 86%). This study population differs from our study population with a median (min-max) FeNO of 31 (15-62) ppb, indicating that this population has more uncontrolled or more severe asthma than the present population [21]. The cut-points of 25 ppb and 50 ppb are supported by several studies: a review from Bjermer et al. suggests that a response to inhaled steroids is unlikely if FeNO 15-25 ppb and likely if FeNO > 25 ppb in steroid naïve subject and 35-50 ppb in subject already treated with ICS. [7]. Furthermore, Smith et al. found in an unselected group of 52 patients with undiagnosed respiratory symptoms that a cut-point at FeNO>47 ppb predicted a response to ICS and thereby indicated eosinophilic airway inflammation [6]. Thirtyfour of the subjects were in the intermediary FeNO-group (25-50 ppb) and only 31% of these subjects had sputum eosinophils 3%. This means that having only one cut-point at 22 ppb (according to the Youden index) or 25 ppb (according to ATS guidelines) we would have a high sensitivity and NPV but low specificity and PPV. Thereby a lot of patient would have an increased level of FeNO without sputum eosinophils 3%, and the benefits of using the test would then be limited. This supports that FeNO is most clinically useful in subjects with either a low FeNO (< 25ppb) or a high FeNO (>50 ppb). About 45% of our subjects received treatment with ICS at baseline, and our results showed, that the sensitivity, specificity and NPV were higher for ICS naïve subjects compared with subjects who already received ICS treatment. In our study only 9 steroid naive patients had eosinophils 3%, so these results are affected by power limitations. Page 17 of 22
395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 Our aim with this study was to investigate the usefulness of FeNO in an unselected population referred to a tertiary outpatient clinic for specialist assessment. Therefore, we were interested in analysis made on all subjects without considering ICS status. However, as would be expected, for ICS naïve patients, the sensitivity, specificity and NPV were clearly higher than on a subject treated with ICS. The AUC on the receiver operating curve was 80%, which indicates that FeNO has a reasonable ability to differentiate between patients with and without airway eosinophilia. For subjects who were ICS naïve, the AUC was 96%, and AUC was 72% for subjects who received ICS treatment, indicating that FeNO is most predicative in ICS naïve subjects. Our longitudinal analyses confirm that a significant decrease in FeNO reflects a parallel decrease in sputum eosinophils. Although patients with a decrease in FeNO after treatment also reported a lower ACQ score, the improvement in ACQ score was similar among patients with and without a decrease in FeNO, suggesting that their improvement in symptoms after treatment was not related to a reduction in eosinophilic airway inflammation. Several studies have tried to show improvement in asthma control with an asthma treatment strategy based on the measurement of FeNO. Shaw et al. found, that treatment based on FeNO and ACQ neither resulted in a reduction in asthma exacerbation nor a reduction in the total amount of inhaled ICS [22]. However, they found that FeNO-guided management was most effective in subjects with EOS>3%. Szefler et al. showed, that addition of FeNO to conventional management as an indicator of asthma control, resulted in a higher dose ICS and no clinically improvement in asthma control compared to the conventional asthma management [23]. Opposed to this, Smith et al. showed that treatment managed by FeNO led to significant reduced use of ICS without compromising asthma control [24]. A general problem with these studies and the use of FeNO is they all had a FeNO cut-point between 15 and 26 ppb. A large part of patients in these studies will, as demonstrated in Shaw et al. and our study, have a FeNO in the intermediate group (FeNO 25-50 ppb), which most likely are patients with non-eosinophilic asthma, in whom FeNO is less likely to be of valuable for titrating anti- inflammatory treatment [22]. In contrast, Powell et al. found that management of asthma in pregnancy based on FeNO and symptoms measured by the asthma control questionnaire (ACQ) led to a significant reduction in asthma exacerbations as well as a more frequent use of ICS, but a lower total daily dose and earlier introduction of long-acting B2-agonist [25]. Page 18 of 22
427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 In our study only 31% of the intermediate FeNO group had eosinophilic airway inflammation, which means, that the use of FeNO to monitor asthma treatment without a selected population with a high FeNO and symptoms, will probably result in a non-significant effect on asthma control. On the other hand, if we choose a higher cut-point for FeNO in treatment-algorithms, we risk that some subjects with sputum eosinophils 3% (who have a FeNO-value in the intermediate FeNO-group) fail to be included. The main issue with FeNO in an unselected population in clinical specialist settings is that few (approximately 25%) of the referred subjects have eosinophilic airway inflammation. This may affect the negative and positive predictive values, with an increasing likelihood of false positive test results. Furthermore, in this real life setting, 65 % (n=94) of the subjects had a low FeNO (< 25 ppb) at baseline, which resulted in a limited amount of subjects in whom it would be possible to measure a significant decrease in FeNO. One third of subjects with eosinophilic airway inflammation (sputum eosinophils 3%) were in the intermediary FeNO-group, and it is unclear how these individuals are best detected. From the present study population, we recently reported that combining a positive result of the mannitol test with a FeNO value > 25ppb, a PPV of 73% can be achieved [26]. This indicates it might be beneficial to investigate this group with FeNO combined with other tests e.g. a bronchial challenge. The strength of this study is that it describes a complete clinical population, as all patients referred over 12 months were asked to participate and thereby the study-population was representative for the patients in a specialist clinical setting. However, a potential limitation is that in this population, FeNO may be affected by external factors such as cigarette smoking ( FeNO), exposure to allergens ( FeNO), diet [27] and virus infections ( FeNO) [3, 20, 28]. However, when we adjusted for smoking by reducing the cut off for FeNO, we did not see any significant changes in the specificity and positive predictive value of FeNO. A limitation in this study is a high dropout rate, which might affect the follow-up data with selection bias. In this case, it might reduce the overall change in ACQ at follow-up, as subjects who completed the study had less room for improvement in their ACQ score. In conclusion, our results support the cut-points for FeNO suggested by ATS: FeNO may be used to monitor eosinophilic airway inflammation in patients with high FeNO at baseline, as a significant decrease in FeNO correlates with a significant decrease in eosinophilia in induced sputum. Page 19 of 22
459 460 461 462 463 464 465 466 467 468 469 470 471 Furthermore, in patients with low FeNO, eosinophilic airway inflammation is unlikely. However, one third of patients with eosinophilic airway inflammation had a FeNO-value between 25-50 ppb, where the diagnostic validity for determining airway eosinophilia is poor and it is important to look at the clinical context in this group. Page 20 of 22
472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 References: 1. Global Initiative for Asthma (GINA). Global Strategy for Asthma Management and Prevention.. Availble from: http://www.ginaasthma.org/. Accessed March 1, 2010 2010. 2. Tillie-Leblond, I., et al., Relation between inflammation and symptoms in asthma. Allergy, 2009. 64(3): p. 354-67. 3. Dweik, R.A., et al., An official ATS clinical practice guideline: interpretation of exhaled nitric oxide levels (FENO) for clinical applications. Am J Respir Crit Care Med, 2011. 184(5): p. 602-15. 4. Berry, M.A., et al., The use of exhaled nitric oxide concentration to identify eosinophilic airway inflammation: an observational study in adults with asthma. Clin Exp Allergy, 2005. 35(9): p. 1175-9. 5. Sandrini, A., et al., Fractional exhaled nitric oxide in asthma: an update. Respirology, 2010. 15(1): p. 57-70. 6. Smith, A.D., et al., Exhaled nitric oxide: a predictor of steroid response. Am J Respir Crit Care Med, 2005. 172(4): p. 453-9. 7. Bjermer, L., et al., Current evidence and future research needs for FeNO measurement in respiratory diseases. Respir Med, 2014. 108(6): p. 830-41. 8. Pinnock, H., E.F. Juniper, and A. Sheikh, Concordance between supervised and postal administration of the Mini Asthma Quality of Life Questionnaire (MiniAQLQ) and Asthma Control Questionnaire (ACQ) was very high. J Clin Epidemiol, 2005. 58(8): p. 809-14. 9. Dreborg, S., et al., Results of biological standardization with standardized allergen preparations. Allergy, 1987. 42(2): p. 109-16. 10. Swanney, M.P., et al., Using the lower limit of normal for the FEV1/FVC ratio reduces the misclassification of airway obstruction. Thorax, 2008. 63(12): p. 1046-51. 11. Miller, M.R., et al., Standardisation of spirometry. Eur Respir J, 2005. 26(2): p. 319-38. 12. Brazzale, D.J., A.L. Upward, and J.J. Pretto, Effects of changing reference values and definition of the normal range on interpretation of spirometry. Respirology, 2010. 15(7): p. 1098-103. 13. Anderson, S.D., et al., A new method for bronchial-provocation testing in asthmatic subjects using a dry powder of mannitol. Am J Respir Crit Care Med, 1997. 156(3 Pt 1): p. 758-65. 14. Yan, K., C. Salome, and A.J. Woolcock, Rapid method for measurement of bronchial responsiveness. Thorax, 1983. 38(10): p. 760-5. 15. Paggiaro, P.L., et al., Sputum induction. Eur Respir J Suppl, 2002. 37: p. 3s-8s. 16. Efthimiadis, A., et al., Methods of sputum processing for cell counts, immunocytochemistry and in situ hybridisation. Eur Respir J Suppl, 2002. 37: p. 19s-23s. 17. Jatakanon, A., S. Lim, and P.J. Barnes, Changes in sputum eosinophils predict loss of asthma control. Am J Respir Crit Care Med, 2000. 161(1): p. 64-72. 18. Ghaaliq Lalkhen, A. and A. McCluskey, Clinical tests: sensitivity and specificity. 2008. 19. Perkins, N.J. and E.F. Schisterman, The inconsistency of "optimal" cutpoints obtained using two criteria based on the receiver operating characteristic curve. Am J Epidemiol, 2006. 163(7): p. 670-5. 20. Malinovschi, A., et al., The value of exhaled nitric oxide to identify asthma in smoking patients with asthma-like symptoms. Respir Med, 2012. 106(6): p. 794-801. Page 21 of 22
518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 21. Schleich, F.N., et al., Exhaled nitric oxide thresholds associated with a sputum eosinophil count >/=3% in a cohort of unselected patients with asthma. Thorax, 2010. 65(12): p. 1039-44. 22. Shaw, D.E., et al., The use of exhaled nitric oxide to guide asthma management: a randomized controlled trial. Am J Respir Crit Care Med, 2007. 176(3): p. 231-7. 23. Szefler, S.J., et al., Management of asthma based on exhaled nitric oxide in addition to guideline-based treatment for inner-city adolescents and young adults: a randomised controlled trial. Lancet, 2008. 372(9643): p. 1065-72. 24. Smith, A.D., et al., Use of exhaled nitric oxide measurements to guide treatment in chronic asthma. N Engl J Med, 2005. 352(21): p. 2163-73. 25. Powell, H., et al., Management of asthma in pregnancy guided by measurement of fraction of exhaled nitric oxide: a double-blind, randomised controlled trial. Lancet, 2011. 378(9795): p. 983-90. 26. Porsbjerg, C., A. Sverrild, and V. Backer, Combining the Mannitol Test and FeNO in the Assessment of Poorly Controlled Asthma. J Allergy Clin Immunol Pract, 2015. 27. Olin, A.C., et al., Increased nitric oxide in exhaled air after intake of a nitrate-rich meal. Respir Med, 2001. 95(2): p. 153-8. 28. Olin, A.C., et al., Height, age, and atopy are associated with fraction of exhaled nitric oxide in a large adult general population sample. Chest, 2006. 130(5): p. 1319-25. Page 22 of 22
Highligts: Validation of the ATS cut points for FeNO to monitor the level of EOS. Investigated in a real life-population Easy to use the findings in a clinical setting.
本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 (www.xuebalib.com) 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具