Portal de Periódicos da CAPES

Elevated exhaled nitric oxide is a clinical indicator of future uncontrolled asthma in asthmatic patients on inhaled corticosteroids

2011; Elsevier BV; Volume: 128; Issue: 2 Linguagem: Inglês

10.1016/j.jaci.2011.06.008

1097-6825

Robert S. Zeiger, Michael Schätz, Feng Zhang, William Crawford, Michael Kaplan, Richard M. Roth, Wansu Chen,

Chronic Obstructive Pulmonary Disease (COPD) Research

To the Editor: Asthma is a chronic inflammatory process with airflow obstruction and hyperreactivity; however, diagnosis and monitoring rely on symptoms and lung function and do not routinely include inflammatory markers such as fractional exhaled nitric oxide (Feno). Feno levels >300% of predicted are associated with a high likelihood of a favorable response to inhaled corticosteroids (ICSs)1Smith A.D. Cowan J.O. Brassett K.P. Filsell S. McLachlan C. Monti-Sheehan G. et al.Exhaled nitric oxide: a predictor of steroid response.Am J Respir Crit Care Med. 2005; 172: 453-459Crossref PubMed Scopus (322) Google Scholar and also worsening asthma with ICS withdrawal.2Pijnenburg M.W. Hofhuis W. Hop W.C. de Jongste J.C. Exhaled nitric oxide predicts asthma relapse in children with clinical asthma remission.Thorax. 2005; 60: 215-218Crossref PubMed Scopus (254) Google Scholar We reported that highest versus lowest quartile Feno levels in asthmatic patients on ICSs were associated with excess short-acting β2-agonist (SABA) use and exacerbations requiring oral corticosteroid (OCS) courses in a prior year, independent of an asthma control tool and spirometry.3Zeiger R.S. Schatz M. Zhang F. Crawford W.W. Kaplan M.S. Roth R.M. et al.Association of exhaled nitric oxide to asthma burden in asthmatics on inhaled corticosteroids.J Asthma. 2011; 48: 8-17Crossref PubMed Scopus (25) Google Scholar We now test the hypothesis that Feno levels >300% predicted identify patients with future uncontrolled asthma, independent of an asthma control tool and spirometry. We conducted a multicenter, prospective, noninterventional study among 304 patients aged 12 to 56 years with allergist-diagnosed persistent asthma from 4 Kaiser Permanente Southern California (KPSC) allergy departments. KPSC Institutional Review Board approved the study. Additional inclusions were (1) aero-allergen sensitization by using skin/RAST testing, (2) monotherapy or combination ICS for ≥1 month before enrollment, and (3) continuous KPSC membership with 1-year pharmacy benefit. Exclusions included (1) smokers; (2) OCS, theophylline, or anticholinergic use within 2 weeks of enrollment; (3) FEV1 < 50% predicted; (4) chronic obstructive lung disease; (5) omalizumab use; or (6) participation in an asthma clinical trial. During a scheduled allergist visit, enrolled patients completed questionnaires on demographics, asthma/allergy history, Asthma Control Test (ACT), medication use, and asthma utilization. KPSC pharmacy dispensings and utilization were captured in electronic medical records and research data warehouses. The measurement of Feno levels was performed before or ≥20 minutes after spirometry by using the validated NIOX MINO hand-held device (Aerocrine AB, Solna, Sweden). FENO levels were masked by using an electronic chip downloaded at the data analysis site. Spirometry by American Thoracic Society standards was performed on a KOKO electronic pneumotach spirometer. Impairment was defined as SABA dispensings both as a count variable (number dispensed) and as a dichotomous variable (≥7 canisters dispensed: yes/no) since patients with ≥7 SABA canisters dispensed in a future year is a validated administrative data surrogate for asthma impairment.4Schatz M. Zeiger R.S. Vollmer W.M. Mosen D. Apter A.J. Stibolt T.B. et al.Validation of a beta-agonist long-term asthma control scale derived from computerized pharmacy data.J Allergy Clin Immunol. 2006; 117: 995-1000Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar Risk was defined as OCS courses dispensed within 2 days of an asthma exacerbation–coded provider visit or an asthma-related emergency department visit or hospitalization (defined as a primary asthma diagnosis [493.xx] or secondary asthma diagnosis when the primary diagnosis was related to a respiratory disease). Expert Panel Report-3–recommended step-care level was calculated by using an algorithm and formula based on asthma medication and amount dispensed.5Zeiger R.S. Schatz M. Li Q. Zhang F. Purdum A. Chen W. Step-up care improves impairment in uncontrolled asthma: administrative data study.Am J Manag Care. 2010; 16: 897-906PubMed Google Scholar Absolute Feno levels were transformed into percent predicted values by using reference equations for adolescents6Malmberg L.P. Petays T. Haahtela T. Laatikainen T. Jousilahti P. Vartiainen E. et al.Exhaled nitric oxide in healthy nonatopic school-age children: determinants and height-adjusted reference values.Pediatr Pulmonol. 2006; 41: 635-642Crossref PubMed Scopus (131) Google Scholar and adults7Olin A.C. Bake B. Toren K. Fraction of exhaled nitric oxide at 50 mL/s: reference values for adult lifelong never-smokers.Chest. 2007; 131: 1852-1856Crossref PubMed Scopus (151) Google Scholar to better generalize findings for clinical care as is done for spirometry. Feno was dichotomized by 300% predicted levels based on its clinical relevance1Smith A.D. Cowan J.O. Brassett K.P. Filsell S. McLachlan C. Monti-Sheehan G. et al.Exhaled nitric oxide: a predictor of steroid response.Am J Respir Crit Care Med. 2005; 172: 453-459Crossref PubMed Scopus (322) Google Scholar, 2Pijnenburg M.W. Hofhuis W. Hop W.C. de Jongste J.C. Exhaled nitric oxide predicts asthma relapse in children with clinical asthma remission.Thorax. 2005; 60: 215-218Crossref PubMed Scopus (254) Google Scholar as described earlier to provide clinically actionable information. Poisson regression models with a robust error variance were constructed to calculate the unadjusted and adjusted relative risk (RR) and 95% CI separately for patients with ≥7 SABA canisters dispensed and ≥2 exacerbations with OCS courses in the follow-up year. Negative binomial models using a robust error variance were used to calculate the unadjusted and adjusted incidence rate ratio (IRR) and 95% CI for number of SABA counts and number of exacerbations requiring OCS courses. Multivariate analyses were adjusted for ACT score and FEV1 % predicted. Analyses were 2-tailed with significance at .05 and were performed using SAS software (version 9.2 for Windows, SAS Institute, Cary, NC). Feno levels were elevated at 41.0 ± 36.2 ppb mean and 28 ppb median levels (range from 7 to 215 ppb) and 48% to 1855% predicted. Patients were predominately young adult, female, white, and highly atopic (Table I). Asthma burden was high in the follow-up year, with 20.4% dispensed ≥7 SABA canisters, 21.4% with an asthma exacerbation requiring OCS, and approximately 75% at a step-care level of 3 to 5.Table IDemographics, atopic history, and asthma features at baseline and during 1-y follow-up of total study patients and patients grouped by FENO status at baseline (% predicted ≤300% vs >300%)Demographics, atopy, asthma featuresTotal cohort (N = 304)Feno % predicted≤300% (n =199)>300% (n =105)Baseline Feno (ppb), median (range)28 (7-215)22 (7-47)61 (32-215) Age (y)35.9 (14.9)38.4 (14.3)31.0 (14.8)∗P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables. Females183 (60.2%)121 (60.8%)62 (59%) Geocoded median household income ($)71,991 (31,862)72,870 (32,898)70,297 (29,833)Ethnicity Hispanic80 (26.3%)42 (21.1%)38 (36.2%)∗P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables. White169 (55.6%)121 (60.8%)48 (45.7%)∗P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables. Black51 (16.8%)31 (15.6%)20 (19%)∗P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables. Asian/Pacific Islander39 (12.8%)20 (10.1%)19 (18.1%)∗P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables. Other races and combinations70 (23%)44 (22.1%)26 (24.8%)Atopy Allergic rhinitis177 (58.2%)114 (57.3%)63 (60%) Atopic dermatitis88 (28.9%)56 (28.1%)32 (30.5%) Aero-allergen sensitivities4.29 (2.30)4.12 (2.38)4.62 (2.11)∗P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables. ACT score19.6 (4.4)19.9 (4.4)19.2 (4.5)Spirometry FEV1 % predicted86.4 (14.9)87.2 (14.5)84.8 (15.4) FEV1 % predicted ≤80%108 (35.5%)63 (31.7%)45 (42.9%)1-y follow-up No. of SABA canisters dispensed4.3 (5.1)3.6 (4.1)5.7 (6.3)† 7 or more62 (20.4%)27 (13.6%)35 (33.3%)†Asthma exacerbations No. of OCS courses0.29 (0.66)0.23 (0.50)0.41 (0.87)∗P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables. ≥165 (21.4%)39 (19.6%)26 (24.8%) ≥214 (4.6%)5 (2.5%)9 (8.6%)∗P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables.Step-care level based on NAEPP guidelines 124 (7.9%)14 (7%)10 (9.5%) 253 (17.4%)33 (16.6%)20 (19%) 3109 (35.9%)74 (37.2%)35 (33.3%) 480 (26.3%)51 (25.6%)29 (27.6%) 538 (12.5%)27 (13.6%)11 (10.5%)Data presented as mean (SD) or frequency N (%).NAEPP, National Asthma Education and Prevention Program.∗ P < .05 and †P ≤ .001 were calculated using χ2 test or Fisher exact test for categorical variables and ANOVA or Kruskal-Wallis test for continuous variables and negative binomial model for count variables. Open table in a new tab Data presented as mean (SD) or frequency N (%). NAEPP, National Asthma Education and Prevention Program. Patients with Feno levels of >300% compared with those with ≤300% predicted were dispensed ≥7 SABA canisters more frequently (33% vs 13.6%) and also received more SABA canisters during the 1-year follow-up (both P < .001) (Table I). Moreover, both ≥2 and total number of asthma exacerbations requiring OCS occurred significantly more frequently in patients with Feno levels of >300% compared with ≤300% predicted (Table I). No association occurred between Feno levels and ACT score, and only a marginal one occurred with FEV1 ≤ 80% predicted (P = .052) (Table I). After adjusting for ACT score and FEV1 % predicted control categories, Feno levels at >300% compared with ≤300% predicted were associated with an RR of 2.26 (95% CI, 1.46-3.50; P < .001) for ≥7 SABA canisters dispensed (Table II) and an IRR of 1.45 (95% CI, 1.12-1.89, P = .005) for the total number of SABA canisters dispensed in the follow-up year. After similar adjustments, >300% compared with ≤300% predicted Feno levels was associated with ≥2 (RR, 3.26; 95% CI, 1.17-9.10; P = .024) (Table II) and marginally with the total number (IRR, 1.61, 95% CI, 0.99-2.62; P = .055) of exacerbations with OCS in the follow-up year. Similar findings were observed for adjustments with continuous ACT score and FEV1 % predicted levels or for adjustments with age, gender, and ethnicity (data not presented).Table IIUnadjusted and adjusted∗Variables adjusted in multivariate analyses were ACT scores, categorized by its 3 control categories: >19 (controlled), 16-19 (not well controlled), and 80% predicted (controlled) and ≤80% predicted (uncontrolled). RR (95% CI) by Poisson regression with a robust error variance for the listed outcomes during the follow-up yearOutcomes during follow-up yearFeno % predicted≤300%(n = 199)>300%(n = 105)P valuePatients with ≥7 SABA canisters dispensed Unadjusted1.002.46 (1.58-3.83)<.001 Adjusted1.002.26 (1.46-3.50)<.001Patients with ≥2 OCS courses with asthma exacerbations Unadjusted1.003.41 (1.17-9.92).024 Adjusted1.003.26 (1.17-9.10).024Significant differences noted in bold.∗ Variables adjusted in multivariate analyses were ACT scores, categorized by its 3 control categories: >19 (controlled), 16-19 (not well controlled), and 80% predicted (controlled) and ≤80% predicted (uncontrolled). Open table in a new tab Significant differences noted in bold. Feno is (1) an indicator of airway inflammation, with higher concentrations observed in ICS-naive patients, (2) sensitive to alterations in ICS treatment, (3) helpful in asthma diagnosis, and (4) predictive of loss of asthma control.8Taylor D.R. Nitric oxide as a clinical guide for asthma management.J Allergy Clin Immunol. 2006; 117: 259-262Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar Our study extends these relationships to real-world clinical care. Specifically, in patients with persistent asthma on ICS, Feno levels >300% compared with ≤300% predicted were significantly associated during a follow-up year, independent of an asthma control tool and spirometry, with (1) a 2.3-fold higher RR for ≥7 SABA canisters dispensed and about a 1.5 times higher IRR for the total number of SABA canisters dispensed and (2) a modest increase in patients with ≥2 asthma exacerbations requiring OCS. These findings support the addition of a measure of inflammation (Feno) to asthma control questionnaires and spirometry during routine asthma visits to identify patients at risk for future uncontrolled asthma. As such, Feno levels >300% of predicted identified patients, additional to those identified by ACT score and spirometry, who were dispensed excessive SABA or experienced ≥2 asthma exacerbations during a follow-up year. Studies conflict on the usefulness of Feno-based asthma management.9Petsky H.L. Cates C.J. Lasserson T.J. Li A.M. Turner C. Kynaston J.A. et al.A systematic review and meta-analysis: tailoring asthma treatment on eosinophilic markers (exhaled nitric oxide or sputum eosinophils).Thorax. 2010 Oct 11; ([Epub ahead of print] PMID: 20937641)Google Scholar The present study does not deal with management directly, but does provide novel information concerning future uncontrolled asthma and its relationship with high Feno levels in a real-world setting. The current study is limited by its single visit and determination of Feno. However, the independent relationship found between Feno >300% of predicted at a single point in time to excessive future SABA canister dispensings identifies its potential importance in assessing patients. In summary, our prospective findings highlight the independent relationship of Feno levels >300% of predicted to increased future 1-year asthma impairment and risk. As such, Feno determination appears clinically useful in identifying persistent atopic, nonsmoking asthmatic patients on ICS at risk for future uncontrolled asthma. Studies are needed to determine whether such Feno information will improve future asthma care and outcomes.