- Email: [email protected]
Symptoms, spirometry, exhaled nitric oxide, and asthma exacerbations in clinical practice
Symptoms, spirometry, exhaled nitric oxide, and asthma exacerbations in clinical practice Daniel Menzies, MBChB*; Cathy Jackson, MD†; Catrina Mistry, RN*; Ruth Houston, RN‡; and Brian J. Lipworth, MD*
Background: Patient symptoms, spirometry measurements, exacerbation rates, and exhaled nitric oxide (FENO) levels have all been used to quantify asthma severity. Objective: To determine the relationships among these disease surrogates in clinical practice. Methods: Data were collected from 5 primary care asthma clinics on patient symptoms, reliever use, spirometry measurements, maintenance pharmacotherapy, disease severity (British Thoracic Society treatment step), and FENO level. Exacerbation data (asthma-related unscheduled health care contact or rescue oral corticosteroid therapy) for the 12 months before and 3 months after the clinic visit were then obtained. Results: A total of 267 adult asthmatic patients (mean [SEM] age, 51.6 [1.1] years; forced expiratory volume in 1 second, 86.3% [1.2%] of predicted) participated, and 157 exacerbations were captured. For the 12 months before the clinic visit, exacerbation rate was positively correlated with dose of inhaled corticosteroid (P ⬍ .001), treatment step (P ⬍ .001), reliever use (P ⫽ .002), and symptom score (P ⬍ .001) but was negatively correlated with FENO level (P ⫽ .04); only symptom scores correlated with exacerbation rate in the 3 months after the visit. Levels of FENO were significantly lower in frequently exacerbating patients receiving higher doses of maintenance inhaled corticosteroids compared with patients with mild disease who were corticosteroid naive (19.7 vs 40.4 ppb, P ⬍ .001). Measurement of FENO was an insensitive method (sensitivity, 66.7%; specificity, 51.9% at a cutoff value of 20 ppb) for identifying patients who subsequently exacerbated. Conclusion: Levels of FENO are paradoxically decreased in patients with more severe asthma and frequent exacerbations and may, therefore, be of limited utility in primary care. Ann Allergy Asthma Immunol. 2008;101:248–255.
INTRODUCTION Published asthma guidelines1,2 recommend that patients undergo regular, proactive monitoring in primary care to evaluate disease severity and symptom control, to review prescribed treatment, and to adjust therapy as appropriate. There is evidence that regular disease monitoring in this manner is beneficial. Improvements include better patient inhaler technique, increased compliance with prescribed medications, and a reduced disease exacerbation rate.3,4 Review of asthmatic patients at regular intervals is equally effective if undertaken by physicians or suitably trained nurses. In the United Kingdom, this is most commonly performed by the primary care practice nurse.4,5 Clinical outcome measures used to evaluate asthma control at these scheduled visits include objective symptom scores, recent reliever use, peak flow measurements, spirometry, and an estimate of disease Affiliations: * Asthma and Allergy Research Group, Department of Medicine and Therapeutics, Ninewells Hospital and Medical School, Dundee, Scotland; † Tayside Centre for General Practice, Dundee, Scotland; ‡ Yellow Practice, Drumhar Health Centre, Perth, Scotland. Disclosures: Authors have nothing to disclose. Funding Sources: This study was supported by a grant from the East of Scotland Primary Care Research Network; Aerocrine, the manufacturer of the portable nitric oxide analyzers, which donated the devices as an unrestricted educational gift; and the University of Dundee (acted as research sponsor). Received for publication February 14, 2008; Received in revised form May 2, 2008; Accepted for publication May 4, 2008.
248
severity indicated by the maintenance therapy required (ie, the current treatment “step”). Quantification of these severity indicators is used to formulate decisions regarding asthma treatment, with the aim of improving symptoms, reducing exacerbations, and minimizing the likelihood of medicationrelated adverse effects. Few published data from primary care clearly demonstrate relationships among these severity indicators, particularly regarding disease exacerbation rate, arguably one of the most important measures of disease activity. Recent data6,7 suggest that symptoms and spirometry may be relatively insensitive methods by which to monitor disease activity in asthma and that directly or indirectly measuring airway inflammation may be more useful. Quantification of exhaled nitric oxide (FENO) is now established as a reliable and reproducible method of determining airway inflammation in asthmatic patients that correlates with disease activity and may predict subsequent exacerbation.8 –11 Quantification of FENO can also differentiate between asthma and chronic obstructive pulmonary disease.12 Studies6,7,13–15 in adults and children have demonstrated that titrating asthma treatment against airway inflammation (airway hyperresponsiveness, sputum eosinophil count, and FENO level) can reduce the exacerbation rate or the dose of inhaled corticosteroid (ICS) required to maintain control compared with a “conventional” approach. However, these studies were undertaken in secondary care or specialist asthma clinics and stipulated frequent (every 3 or 4 months) evaluation of patients to allow treat-
ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY
ment titration. The recent availability of a portable nitric oxide analyzer (NIOX MINO; Aerocrine, Smidesva¨gen, Sweden) has allowed for such measurements to be performed in a primary care setting. Few data compare airway inflammation with conventional indices of asthma severity in primary care, where most patients are routinely treated. As such, it is unknown whether “inflammation-based” treatment approaches are applicable or relevant to patients who are not treated in specialist centers. We, therefore, prospectively gathered data from primary care asthma clinics to determine the relationships among symptoms, spirometric measurements, reliever use frequency, disease severity (guideline treatment step), airway inflammation (FENO level), and exacerbation rate. METHODS The Tayside Committee for Research Ethics approved the trial. Five general practice (GP) centers in Tayside, Scotland, participated. Adults with a physician diagnosis of asthma visiting their primary care practice nurse for a routine scheduled annual review were invited to participate and gave verbal consent. To be eligible for trial inclusion, participants had to have an established diagnosis of asthma for at least 1 year. The patients’ GP assigned the diagnosis in each case based on a combination of reported symptoms, spirometric reversibility, response to treatment, and clinical probability. This was intended to reflect everyday practice in primary care. Documented airway hyperresponsiveness was not a prerequisite, but this test is rarely performed in primary care, except in the case of diagnostic uncertainty. The nurses were trained to manage persistent asthma according to British Thoracic Society (BTS) guidelines and had autonomy to make treatment decisions, including recommending changes to prescribed pharmacotherapy. Current and former smokers were excluded to minimize the possibility of inadvertently including patients with chronic obstructive pulmonary disease. Measurements of Disease Severity The following measurements were undertaken for all participants in the order specified: Spirometry. This was performed in accordance with American Thoracic Society/European Respiratory Society guidelines to determine forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), and FEV1/FVC.16 Symptoms and reliever use. Participants estimated the number of times daily they required reliever medication. They were then specifically asked the Royal College of Physicians (RCP) “three questions”: whether in the last month they had experienced difficulty sleeping because of asthma, had experienced their usual daytime symptoms, and had asthma interference with their usual daily activities, including work. A total symptom score (possible range, 0 –3) was then calculated. Current prescribed medications. The participants’ usual asthma medications were recorded, including the dose of
VOLUME 101, SEPTEMBER, 2008
ICSs, allowing the BTS “treatment step” to be determined in each case (Table 1). The treatment step at the beginning of the consultation, before any changes to treatment were made, was used for the analysis. A record was made of any new changes to current prescribed treatment. FENO level. A single measurement of FENO was made using a portable chemiluminescence analyzer (NIOX MINO). A single measurement performed using this device has been shown to have accuracy comparable with 3 measurements using a laboratory-based analyzer.17,18 Except for measuring FENO, current BTS guidelines2 recommend quantifying all of the mentioned severity indices at the regular review. This study was, thus, designed to accurately reflect usual practice in the United Kingdom. Measurement of FENO was performed at the end of the consultation, after treatment decisions had been formulated, to minimize any potential bias associated with this additional procedure compared with conventional best practice. Exacerbation Rate Data Exacerbations were quantified primarily via unscheduled contact with a health care professional for asthma symptoms. Contact could be with the patients’ own GP, the emergency after-hours GP service, or either of the local accident or emergency departments. Patient contact with any of these services generates an immediate written or faxed letter, which is then permanently included in the patients’ primary GP record. The type of treatment initiated after these unscheduled health care contacts did not determine exacerbations. Data were extracted from the GP medical records at least 4 months after the scheduled nurse visit. Exacerbation data were obtained for the 12 months before and 3 months after the scheduled nurse visit. Statistical Analysis Statistical analysis was performed using SPSS version 15 (SPSS Inc, Chicago, Illinois). Spearman rank was used to determine correlations between variables, Wilcoxon rank sum to compare groups, 2 to compare proportions, and logistic regression to evaluate predictors of exacerbation. Tests were deemed significant at Bonferroni-corrected P ⬍ .05. Table 1. British Thoracic Society Guideline Treatment Steps Step 1 Mild intermittent asthma 2 Regular preventive therapy 3 Add-on therapy
4 Persistent poor control 5 Oral corticosteroids
Medication Short-acting bronchodilators ICS, 200–800 g/da A maximum of 800 g/d of an ICS and a LABA, an LRA, or theophyllinea A combination of ⱖ4 asthma medications Regular use of oral corticosteroids
Abbreviations: ICS, inhaled corticosteroid; LABA, long-acting -agonist; LRA, leukotriene receptor antagonist. a Doses are expressed as beclomethasone dipropionate equivalents.
249
RESULTS Demographic details of the 267 enrolled patients are given in Table 2. Two hundred forty-nine participants had complete data; in the remaining 18 patients, it proved technically impossible to obtain a valid FENO measurement (ie, the patient could not perform the controlled exhalation maneuver required to provide a sample). There was no difference in FEV1 between patients in whom FENO was not quantified and the remaining study population (P ⫽ .31); however, the former were significantly older (61 vs 51 years, P ⫽ .01). Most patients were at BTS treatment step 2 or 3; only 2 patients were at step 5, so data from these patients were pooled with those from step 4. Seventy-two patients experienced 141 exacerbations in the 12 months before the visit, and 14 patients had 16 exacerbations in the 3 months subsequent, giving a total of 157 captured episodes. The dose of ICSs increased significantly with increasing treatment step, as did the overall exacerbation rate (Table 3 and Fig 1). In contrast, compared with step 1, FEV1 and FENO were significantly lower in patients at treatment steps 3 to 5 (Table 3 and Fig 1). There was no difference among groups for reliever use, but patients at treatment steps 1, 4, and 5 reported the greatest number of symptoms (Table 3). Level of FENO was significantly negatively correlated with ICS dose, treatment step, and exacerbation rate, whereas positive associations were found with symptom scores and reliever use (Table 4). Exacerbation rate was positively correlated with ICS dose (P ⬍ .001), treatment step (P ⬍ .001), reliever use (P ⬍ .002), and symptom score (P ⬍ .001), whereas the reverse was true for FENO (P ⫽ .04). Other correlations between severity indicators are detailed in Table 4. Direct comparison of patients who experienced disease exacerbation in the 12 months before the study visit and those who did not revealed greater reliever use (P ⫽ .005) and a higher dose of ICSs (P ⬍ .001), but this was not true for those who underwent disease exacerbation in the 3 months after the visit (P ⫽ .32 and P ⫽ .47, respectively) (Table 5). Proportionately more patients at treatment steps 3 to 5 experienced asthma exacerbations in the previous year than those at steps Table 2. Characteristics of the 267 Study Participants Characteristic Male sex, No. (%) Age, mean (SEM), y FEV1, mean (SEM), % of predicted FEV1/FVC, mean (SEM), % Dose of BDP, mean (SEM), g Treatment step, No. (%) Step 1 Step 2 Step 3 Step 4 Step 5
Value 122 (46) 51.6 (1.1) 86.3 (1.2) 77.6 (0.9) 502 (27) 57 (21.3) 99 (37.1) 66 (24.7) 43 (16.1) 2 (0.7)
Abbreviations: BDP, beclomethasone dipropionate; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity.
250
1 and 2 (44/111 [39.6%] vs 28/156 [17.9%], P ⬍ .001); this was also true for the 3 months immediately before the visit (18/111 [16.2%] vs 8/156 [5.1%], P ⫽ .003) but not for the 3 months after the visit (4/111 [3.6%] vs 9/156 [5.8%], P ⫽ .42) (Fig 2). A similar pattern was seen for symptom scores, with a lower proportion of patients exacerbating if they reported 0 or 1 RCP symptoms compared with 2 or 3 symptoms (41/207 [19.8%] vs 31/60 [51.2%], P ⬍ .001, and 14/207 [6.8%] vs 12/60 [20.0%], P ⫽ .002 for the 12 months and 3 months before the visit, respectively). Unlike treatment step, a significantly higher proportion of patients with 2 or more symptoms had a subsequent disease exacerbation in the following 3 months compared with those who had 1 or fewer (7/60 [11.7%] vs 6/207 [2.9%], P ⫽ .005) (Fig 2). Patients at treatment steps 4 and 5, who were receiving high doses of ICS and who exacerbated frequently, had lower FENO values than corticosteroid-naive patients at treatment step 1 (19.7 vs 40.4 ppb, P ⬍ .001) Despite the significant correlations between exacerbation rate and a variety of surrogate measures of asthma severity, logistic regression did not identify any of the quantified variables as significant positive predictors of disease exacerbation in the 12 months before or 3 months after the clinic visit. An RCP symptom score of 0 was identified as a significant negative predictor for exacerbations in the 12 months (P ⫽ .008) and 3 months (P ⫽ .005) before the clinic visit but not for the 3 months after the visit (P ⫽ .45) (ie, no symptoms increased the likelihood of no exacerbations, but the presence of symptoms was a poor predictor of subsequent or impending loss of asthma control). Logistic regression was also used to examine which factors were predictive of the nurses’ decision to step up prescribed therapy. No significant positive associations were found, but, again, there was a negative relationship with symptoms: patients reporting 0 or 1 RCP symptoms were less likely to have an increase to their maintenance treatment initiated by the nurse (P ⬍ .001 and P ⫽ .02, respectively). The nurse initiated changes in regular prescribed medications in 94 patients: 19 had their treatment decreased and 75 had an increase. A significantly greater proportion of those who had their medication increased experienced an exacerbation in the following 3 months compared with those who had no change or a dose reduction (9.3% vs 3.2%, P ⫽ .03). A post hoc analysis of the FENO data was performed, and receiver operating characteristic curves were constructed specifically to evaluate this tool as a predictor of previous and impending loss of disease control. This analysis revealed that for this unselected primary care patient group, FENO was a relatively poor predictor of disease exacerbation, with area under the receiver operating characteristic curves of 0.423 and 0.538 for any exacerbation in the year before and 3 months after the clinic visit, respectively. The sensitivity and specificity for detecting an impending exacerbation in the 3 months after the nurse consultation at a variety of cutoff points were 75.0% and 23.2% (10 ppb), 66.7% and 51.9% (20 ppb), and 33.3% and 69.6% (30 ppb). The corresponding
ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY
Table 3. Differences in Measured Variables by Asthma Treatment Stepa Step 1 FENO, ppb FEV1, % of predicted BDP, g Reliever use, puffs/d Exacerbations, No./y Symptom score
40.4 (4.5) 92.6 (1.9) NA 0.9 (0.2) 0.2 (0.1) 1.1 (0.1)
Step 2
Step 3 b
27.6 (2.8) 87.9 (1.6) 492 (24)b 0.8 (0.1) 0.3 (0.1) 0.7 (0.1)b
Steps 4 and 5 b
23.9 (2.8) 83.5 (3.0)b 533 (27)b 0.8 (0.5) 0.7 (0.1) 0.8 (0.1)
19.7 (2.8)b,c 80.5 (3.8)b 1129 (89)b,c,d 0.9 (0.1) 1.0 (0.2)b,c 1.1 (0.1)c,d
Abbreviations: BDP, beclomethasone dipropionate; FENO, exhaled nitric oxide; FEV1, forced expiratory volume in 1 second; NA, not applicable. a All the values are given as mean (SEM) and are Bonferroni corrected. b Significant difference from step 1, P ⬍ .05. c Significant difference from step 2, P ⬍ .05. d Significant difference from step 3, P ⬍ .05.
Figure 1. Differences in severity measurements stratified by British Thoracic Society treatment step. A, Inhaled corticosteroid (ICS) dose. B, Forced expiratory volume in 1 second (FEV1). C, Exhaled nitric oxide (FENO). D, Exacerbations. BDP indicates beclomethasone dipropionate. * Significant difference from step 1. † Significant difference from steps 1 and 2. ‡ Significant difference from steps 1, 2, and 3. Error bars represent SD.
VOLUME 101, SEPTEMBER, 2008
251
Table 4. Correlations Among Measured Variables in the 12 Months Preceding Clinic Evaluationa FENO FENO FEV1 ICS dose Treatment step Reliever use Exacerbations Symptom score
FEV1
NA ⫺0.98 (.59) ⫺0.172 (.007)b ⫺0.249 (⬍.001)b 0.192 (.002)b ⫺0.131 (.04)b 0.136 (.03)b
ICS dose
Treatment step
Reliever use
Exacerbations
Symptom score
⫺0.98 (.59)b ⫺0.172 (.007)b ⫺0.249 (⬍.001)b 0.192 (.002)b ⫺0.131 (.04)b 0.136 (.03)b NA ⫺0.192 (.002)b ⫺0.214 (⬍.001)b ⫺0.116 (.06) ⫺0.081 (.19) ⫺0.096 (.12) ⫺0.192 (.002)b NA 0.745 (⬍.001) 0.126 (.04)b 0.263 (⬍.001)b 0.070 (.26) ⫺0.214 (⬍.001)b 0.745 (⬍.001) NA 0.039 (.53) 0.269 (⬍.001) 0.030 (.63) ⫺0.116 (.06) 0.126 (.04)b 0.039 (.53) NA 0.188 (.002) 0.377 (⬍.001)b ⫺0.081 (.19) 0.263 (⬍.001)b 0.269 (⬍.001)b 0.188 (.002)b NA 0.350 (⬍.001)b ⫺0.096 (.12) 0.070 (.26) 0.030 (.63) 0.377 (⬍.001)b 0.350 (⬍.001)b NA
Abbreviations: FENO, exhaled nitric oxide; FEV1, forced expiratory volume in 1 second; ICS, inhaled corticosteroid; NA, not applicable. a Values are given are Spearman rho (P value). b Significant correlation, P ⬍ .05. Table 5. Comparison of Measured Variables Between Patients With and Without Exacerbationsa 12 mo before visit Variable FENO, ppb FEV1, % of predicted ICS dose, g BDP Reliever use, puffs/d
3 mo before visit
3 mo after visit
No P value for Exacerbation No P value for No P value for Exacerbation exacerbation difference exacerbation difference Exacerbation exacerbation difference 31.3 (8.3) 85.0 (5.9) 668 (57) 1.4 (0.4)
28.0 (1.7) 86.7 (1.3) 440 (30) 0.8 (0.1)
.07 .11
25.4 (4.3) 83.4 (3.9)
⬍.001b b
.005
673 (118) 1.5 (0.3)
28.4 (1.8) 86.9 (1.3) 484 (27) 0.8 (0.1)
.47 .46 .11 b
.03
31.3 (8.3) 85.0 (5.9) 361 (82) 1.4 (0.4)
28.0 (1.7) 86.7 (1.3) 509 (28) 0.8 (0.1)
.66 .75 .47 .32
Abbreviations: BDP, beclomethasone dipropionate; FENO, exhaled nitric oxide; FEV1, forced expiratory volume in 1 second; ICS, inhaled corticosteroid. a All values are given as mean (SEM) and are Bonferroni corrected. b Significant difference, P ⬍ .05.
values for exacerbation in the 12 months before the visit were 68.2% and 16.9% (10 ppb), 40.9% and 48.1% (20 ppb), and 27.3% and 68.3% (30 ppb). DISCUSSION These data demonstrate a paradoxical relationship between airway inflammation and other measures of asthma severity, with airway caliber and symptoms and exacerbation rate disconnected from FENO measurement. Airway Inflammation The observation that symptoms and spirometry measurements are unrelated to exacerbation rate has been reported in several prospective studies in which measurements of airway inflammation were used to guide dose adjustment in secondary care. Sont et al7 and Green et al6 successfully improved asthma exacerbation rates compared with the conventional symptom and spirometry-based approach by titrating ICS dose against methacholine hyperresponsiveness and induced sputum eosinophil count, respectively. Quantification of FENO is a less invasive method of estimating asthmatic bronchial inflammation that can be performed in primary care using a portable analyzer. Level of FENO is correlated with a variety of other measures of airway inflammation and has been shown to predict loss of asthma control in a trial setting, particularly when combined with FEV1.8,9,19,20 It has also been shown to be as good as methacholine challenge (often con-
252
sidered the gold standard) for diagnosing asthma in a preselected corticosteroid-naive outpatient population.21 These data suggest that this method of detecting airway inflammation may not be clinically relevant in a primary care setting. Level of FENO was negatively correlated with ICS dose, treatment step, and exacerbation rate; did not accurately predict exacerbation in the bivariate model; and demonstrated poor receiver operating characteristic curve characteristics for exacerbation rate in the post hoc analysis. Although FENO measurement is a valuable method for assessing disease activity in a clinical trial, where it can help differentiate among subtle pharmacotherapeutic effects, the very steep relationship between ICS dose and attenuation of this inflammatory index renders it a comparatively blunt instrument for dose titration in clinical practice.22–24 This may explain why prospective clinical trials using this surrogate marker to make dose adjustments have not demonstrated a clear improvement in exacerbation rate over conventional symptom- and spirometry-based strategies.13,14,25 Levels of FENO are maximally suppressed in patients with more severe disease (receiving the highest doses of maintenance ICS) and, therefore, are of the least clinical utility in patients who potentially stand to gain the most from better disease control. Conventional Disease Severity Measurements The present data demonstrate that there are correlations among many of the guideline-recommended severity mea-
ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY
BTS treatment step, increased maintenance ICS requirement, increased symptoms, and more frequent reliever medication use), note that none of these factors were powerful enough to act as predictors of disease exacerbation. This suggests that these measurements are not precise enough tools on which to confidently base management decisions, particularly regarding dose and add-on adjustment of maintenance pharmacotherapy. Note that patients with a nurse-initiated increase in maintenance treatment were more likely to experience subsequent disease exacerbation than were those without. A possible explanation for this is that if a patient were experiencing symptoms heralding the beginning of a disease exacerbation at the time of consultation, then an increase in treatment would be initiated.
Figure 2. Proportionate number of patients experiencing disease exacerbation stratified by British Thoracic Society guideline treatment step (A) and symptom score (B). Percentages indicate the proportion in each group experiencing disease exacerbation.
surements currently used in asthma management in primary care. However, it also demonstrates that none of these measurements were accurate methods by which to predict patients at risk for disease exacerbation. This is important because reduction in exacerbation rate is a primary goal of long-term asthma management because of not only the significant morbidity but also the socioeconomic burden associated with disease flares.26 However, we appreciate that the frequency of exacerbations will be relatively low for most patients in primary care, who have mild to moderate persistent asthma. The ICS dose was related to treatment step, as the BTS stepwise classification incorporates this as an index of severity. The FEV1 also declined with increased ICS dose and BTS treatment step, again reflecting disease severity. Although exacerbation rates were positively correlated with the quantified surrogate measurements of disease severity (higher
VOLUME 101, SEPTEMBER, 2008
Asthma Management in Primary Care Most of the 267 patients studied were at treatment step 2 or 3 of the BTS guidelines, yet examining only 15 months of data for each patient captured more than 150 exacerbations. This reflects previous trial data27 suggesting that most asthmatic patients have mild to moderate disease but nonetheless experience significant associated morbidity. Note that patients at BTS treatment step 1 had the highest levels of FENO and were relatively symptomatic, suggesting that these patients may be relatively undertreated. Asthma is a chronic disorder, and for most patients, nonspecialists in primary care successfully manage the disease without input from respiratory physicians or secondary care. In the United Kingdom, appropriately trained nurses undertake much of the long-term asthma management, with indirect supervision from GPs. The BTS asthma guidelines advocate regular disease monitoring in primary care using precisely this arrangement, yet few data are available to substantiate such an approach. The lack of data relevant to primary care may be partly compounded because most asthma studies are performed in select patients and usually in the context of a university research department or a secondary care facility. A recent study28 suggested that as little as 4% of asthmatic patients met the eligibility criteria for most randomized controlled trials on which current international guidelines are based. The present study is unique in that it gathered data on unselected asthmatic patients in the context of usual United Kingdom, guideline-based practice. Study Limitations Data were gathered only for severe exacerbations requiring treatment by a physician and not for mild to moderate exacerbations; the relationship between surrogates of disease severity and disease control could be explored more completely with this additional information. Follow-up was relatively short (3 months), and extending this may have strengthened the study. Note that exacerbations beyond 3 months would be unlikely to have any direct relationship with the clinic measurements and would, therefore, provide limited additional information. Most data in relation to FENO measurement in asthma relates specifically to atopic asthma, and these data
253
need to be interpreted without certainty about the allergic status of the participants. Also, note that the number of exacerbations in the prospective component of the trial was relatively small, and, therefore, the trial may be underpowered in this regard. Although FENO did not predict exacerbation rate, note that none of the other “conventional” asthma measures identified patients at risk for exacerbation either. It is conceivable that combining FENO with another surrogate of disease severity, or using the test in more select patients (eg, those with documented eosinophilic airway disease), might improve the accuracy and clinical value. These data do not address the potential utility of FENO measurement in other clinical scenarios. A single measurement of FENO may help discriminate between asthmatic and nonasthmatic individuals in a corticosteroid-naive population,21 and it may help determine which patients would benefit from ICS treatment.29 Longitudinal measurement of FENO may be valuable in assessing medication compliance and in determining whether treatment adjustments are required. In severe asthma, there may be a significant neutrophilic element to disease pathogenesis, and, as such, a low FENO level may be clinically important in distinguishing these patients from those with predominantly eosinophilic airway inflammation in whom FENO levels are elevated. CONCLUSION Some of the contrasts observed (particularly regarding FENO measurements) between the data presented and those previously reported may be explained by the differing context and patient demographics seen in clinical trials compared with clinical practice. Although new asthma management strategies aimed at reducing exacerbations by using surrogate measures of inflammation or any other index of disease severity to guide therapy may be warranted, it will be important to ensure that they are carefully validated in the target population. This study demonstrated that spirometry measurements, airway inflammation, and symptoms are poorly correlated with each other in a primary care cohort of asthmatic patients. In patients with more severe asthma (where better disease control would be most desirable), FENO is maximally suppressed secondary to concomitant use of relatively high-dose ICS and may, therefore, be of limited utility in guiding treatment decisions in a primary care population. ACKNOWLEDGMENTS We thank the GP nurses throughout Tayside who participated in this study.
4. 5. 6. 7.
8. 9. 10. 11. 12.
13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
REFERENCES 1. Global Initiative for Asthma (GINA). Global Strategy for Asthma Management and Prevention. Bethesda, MD: National Heart, Lung, and Blood Institute; January 1995 (updated 2002, 2003). NIH publication 02–3659. 2. British Thoracic Society Scottish Intercollegiate Guidelines Network. British Guideline on the Management of Asthma. Thorax. 2008;63(suppl 4):iv1-iv121. 3. Charlton I, Charlton G, Broomfield J, Mullee MA. Audit of the effect of
254
23.
24.
a nurse run asthma clinic on workload and patient morbidity in a general practice. Br J Gen Pract. 1991;41:227–231. Dickinson J, Hutton S, Atkin A, Jones K. Reducing asthma morbidity in the community: the effect of a targeted nurse-run asthma clinic in an English general practice. Respir Med. 1997;91:634 – 640. Lindberg M, Ahlner J, Ekstrom T, Jonsson D, Moller M. Asthma nurse practice improves outcomes and reduces costs in primary health care. Scand J Caring Sci. 2002;16:73–78. Green RH, Brightling CE, McKenna S, et al. Asthma exacerbations and sputum eosinophil counts: a randomised controlled trial. Lancet. 2002; 360:1715–1721. Sont JK, Willems LN, Bel EH, van Krieken JH, Vandenbroucke JP, Sterk PJ; AMPUL Study Group. Clinical control and histopathologic outcome of asthma when using airway hyperresponsiveness as an additional guide to long-term treatment. Am J Respir Crit Care Med. 1999; 159(pt 1):1043–1051. Jatakanon A, Lim S, Kharitonov SA, Chung KF, Barnes PJ. Correlation between exhaled nitric oxide, sputum eosinophils, and methacholine responsiveness in patients with mild asthma. Thorax. 1998;53:91–95. Jones SL, Kittelson J, Cowan JO, et al. The predictive value of exhaled nitric oxide measurements in assessing changes in asthma control. Am J Respir Crit Care Med. 2001;164:738 –743. Strunk RC, Szefler SJ, Phillips BR, et al. Relationship of exhaled nitric oxide to clinical and inflammatory markers of persistent asthma in children. J Allergy Clin Immunol. 2003;112:883– 892. Harkins MS, Fiato KL, Iwamoto GK. Exhaled nitric oxide predicts asthma exacerbation. J Asthma. 2004;41:471– 476. Fabbri LM, Romagnoli M, Corbetta L, et al. Differences in airway inflammation in patients with fixed airflow obstruction due to asthma or chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2003;167:418 – 424. Smith AD, Cowan JO, Brassett KP, Herbison GP, Taylor DR. Use of exhaled nitric oxide measurements to guide treatment in chronic asthma. N Engl J Med. 2005;352:2163–2173. Pijnenburg MW, Bakker EM, Hop WC, De Jongste JC. Titrating steroids on exhaled nitric oxide in children with asthma: a randomized controlled trial. Am J Respir Crit Care Med. 2005;172:831– 836. Jayaram L, Pizzichini MM, Cook RJ, et al. Determining asthma treatment by monitoring sputum cell counts: effect on exacerbations. Eur Respir J. 2006;27:483– 494. Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J. 2005;26:319 –338. Menzies D, Nair A, Lipworth BJ. Portable exhaled nitric oxide measurement: comparison with the “gold standard” technique. Chest. 2007;131:410 – 414. Alving K, Janson C, Nordvall L. Performance of a new hand-held device for exhaled nitric oxide measurement in adults and children. Respir Res. 2006;7:67. Gelb AF, Flynn Taylor C, Shinar CM, Gutierrez C, Zamel N. Role of spirometry and exhaled nitric oxide to predict exacerbations in treated asthmatics. Chest. 2006;129:1492–1499. Pijnenburg MW, Hofhuis W, Hop WC, De Jongste JC. Exhaled nitric oxide predicts asthma relapse in children with clinical asthma remission. Thorax. 2005;60:215–218. Berkman N, Avital A, Breuer R, Bardach E, Springer C, Godfrey S. Exhaled nitric oxide in the diagnosis of asthma: comparison with bronchial provocation tests. Thorax. 2005;60:383–388. Fardon TC, Burns P, Barnes ML, Lipworth BJ. A comparison of 2 extrafine hydrofluoroalkane-134a-beclomethasone formulations on methacholine hyperresponsiveness. Ann Allergy Asthma Immunol. 2006; 96:422– 430. Menzies D, Nair A, Fardon T, Barnes M, Burns P, Lipworth B. An in vivo and in vitro comparison of inhaled steroid delivery via a novel vortex actuator and a conventional valved holding chamber. Ann Allergy Asthma Immunol. 2007;98:471– 479. Menzies D, Nair A, Hopkinson P, McFarlane L, Lipworth BJ. Differential anti-inflammatory effects of large and small particle size inhaled corticosteroids in asthma. Allergy. 2007;62:661– 667.
ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY
25. Shaw DE, Berry MA, Thomas M, et al. The use of exhaled nitric oxide to guide asthma management: a randomized controlled trial. Am J Respir Crit Care Med. 2007;176:231–237. 26. Lloyd A, Price D, Brown R. The impact of asthma exacerbations on health-related quality of life in moderate to severe asthma patients in the UK. Prim Care Respir J. 2007;16:22–27. 27. Dusser D, Montani D, Chanez P, et al. Mild asthma: an expert review on epidemiology, clinical characteristics and treatment recommendations. Allergy. 2007;62:591– 604. 28. Travers J, Marsh S, Williams M, et al. External validity of randomised controlled trials in asthma: to whom do the results of the trials apply? Thorax. 2007;62:219 –223. 29. Smith AD, Cowan JO, Brassett KP, et al. Exhaled nitric oxide: a
VOLUME 101, SEPTEMBER, 2008
predictor of steroid response. Am J Respir Crit Care Med. 2005;172: 453– 459.
Requests for reprints should be addressed to: Brian J. Lipworth, MD Asthma and Allergy Research Group Department of Medicine and Therapeutics Ninewells Hospital and Medical School Dundee, Scotland DD1 9SY E-mail: [email protected]
255
Background: Patient symptoms, spirometry measurements, exacerbation rates, and exhaled nitric oxide (FENO) levels have all been used to quantify asthma severity. Objective: To determine the relationships among these disease surrogates in clinical practice. Methods: Data were collected from 5 primary care asthma clinics on patient symptoms, reliever use, spirometry measurements, maintenance pharmacotherapy, disease severity (British Thoracic Society treatment step), and FENO level. Exacerbation data (asthma-related unscheduled health care contact or rescue oral corticosteroid therapy) for the 12 months before and 3 months after the clinic visit were then obtained. Results: A total of 267 adult asthmatic patients (mean [SEM] age, 51.6 [1.1] years; forced expiratory volume in 1 second, 86.3% [1.2%] of predicted) participated, and 157 exacerbations were captured. For the 12 months before the clinic visit, exacerbation rate was positively correlated with dose of inhaled corticosteroid (P ⬍ .001), treatment step (P ⬍ .001), reliever use (P ⫽ .002), and symptom score (P ⬍ .001) but was negatively correlated with FENO level (P ⫽ .04); only symptom scores correlated with exacerbation rate in the 3 months after the visit. Levels of FENO were significantly lower in frequently exacerbating patients receiving higher doses of maintenance inhaled corticosteroids compared with patients with mild disease who were corticosteroid naive (19.7 vs 40.4 ppb, P ⬍ .001). Measurement of FENO was an insensitive method (sensitivity, 66.7%; specificity, 51.9% at a cutoff value of 20 ppb) for identifying patients who subsequently exacerbated. Conclusion: Levels of FENO are paradoxically decreased in patients with more severe asthma and frequent exacerbations and may, therefore, be of limited utility in primary care. Ann Allergy Asthma Immunol. 2008;101:248–255.
INTRODUCTION Published asthma guidelines1,2 recommend that patients undergo regular, proactive monitoring in primary care to evaluate disease severity and symptom control, to review prescribed treatment, and to adjust therapy as appropriate. There is evidence that regular disease monitoring in this manner is beneficial. Improvements include better patient inhaler technique, increased compliance with prescribed medications, and a reduced disease exacerbation rate.3,4 Review of asthmatic patients at regular intervals is equally effective if undertaken by physicians or suitably trained nurses. In the United Kingdom, this is most commonly performed by the primary care practice nurse.4,5 Clinical outcome measures used to evaluate asthma control at these scheduled visits include objective symptom scores, recent reliever use, peak flow measurements, spirometry, and an estimate of disease Affiliations: * Asthma and Allergy Research Group, Department of Medicine and Therapeutics, Ninewells Hospital and Medical School, Dundee, Scotland; † Tayside Centre for General Practice, Dundee, Scotland; ‡ Yellow Practice, Drumhar Health Centre, Perth, Scotland. Disclosures: Authors have nothing to disclose. Funding Sources: This study was supported by a grant from the East of Scotland Primary Care Research Network; Aerocrine, the manufacturer of the portable nitric oxide analyzers, which donated the devices as an unrestricted educational gift; and the University of Dundee (acted as research sponsor). Received for publication February 14, 2008; Received in revised form May 2, 2008; Accepted for publication May 4, 2008.
248
severity indicated by the maintenance therapy required (ie, the current treatment “step”). Quantification of these severity indicators is used to formulate decisions regarding asthma treatment, with the aim of improving symptoms, reducing exacerbations, and minimizing the likelihood of medicationrelated adverse effects. Few published data from primary care clearly demonstrate relationships among these severity indicators, particularly regarding disease exacerbation rate, arguably one of the most important measures of disease activity. Recent data6,7 suggest that symptoms and spirometry may be relatively insensitive methods by which to monitor disease activity in asthma and that directly or indirectly measuring airway inflammation may be more useful. Quantification of exhaled nitric oxide (FENO) is now established as a reliable and reproducible method of determining airway inflammation in asthmatic patients that correlates with disease activity and may predict subsequent exacerbation.8 –11 Quantification of FENO can also differentiate between asthma and chronic obstructive pulmonary disease.12 Studies6,7,13–15 in adults and children have demonstrated that titrating asthma treatment against airway inflammation (airway hyperresponsiveness, sputum eosinophil count, and FENO level) can reduce the exacerbation rate or the dose of inhaled corticosteroid (ICS) required to maintain control compared with a “conventional” approach. However, these studies were undertaken in secondary care or specialist asthma clinics and stipulated frequent (every 3 or 4 months) evaluation of patients to allow treat-
ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY
ment titration. The recent availability of a portable nitric oxide analyzer (NIOX MINO; Aerocrine, Smidesva¨gen, Sweden) has allowed for such measurements to be performed in a primary care setting. Few data compare airway inflammation with conventional indices of asthma severity in primary care, where most patients are routinely treated. As such, it is unknown whether “inflammation-based” treatment approaches are applicable or relevant to patients who are not treated in specialist centers. We, therefore, prospectively gathered data from primary care asthma clinics to determine the relationships among symptoms, spirometric measurements, reliever use frequency, disease severity (guideline treatment step), airway inflammation (FENO level), and exacerbation rate. METHODS The Tayside Committee for Research Ethics approved the trial. Five general practice (GP) centers in Tayside, Scotland, participated. Adults with a physician diagnosis of asthma visiting their primary care practice nurse for a routine scheduled annual review were invited to participate and gave verbal consent. To be eligible for trial inclusion, participants had to have an established diagnosis of asthma for at least 1 year. The patients’ GP assigned the diagnosis in each case based on a combination of reported symptoms, spirometric reversibility, response to treatment, and clinical probability. This was intended to reflect everyday practice in primary care. Documented airway hyperresponsiveness was not a prerequisite, but this test is rarely performed in primary care, except in the case of diagnostic uncertainty. The nurses were trained to manage persistent asthma according to British Thoracic Society (BTS) guidelines and had autonomy to make treatment decisions, including recommending changes to prescribed pharmacotherapy. Current and former smokers were excluded to minimize the possibility of inadvertently including patients with chronic obstructive pulmonary disease. Measurements of Disease Severity The following measurements were undertaken for all participants in the order specified: Spirometry. This was performed in accordance with American Thoracic Society/European Respiratory Society guidelines to determine forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), and FEV1/FVC.16 Symptoms and reliever use. Participants estimated the number of times daily they required reliever medication. They were then specifically asked the Royal College of Physicians (RCP) “three questions”: whether in the last month they had experienced difficulty sleeping because of asthma, had experienced their usual daytime symptoms, and had asthma interference with their usual daily activities, including work. A total symptom score (possible range, 0 –3) was then calculated. Current prescribed medications. The participants’ usual asthma medications were recorded, including the dose of
VOLUME 101, SEPTEMBER, 2008
ICSs, allowing the BTS “treatment step” to be determined in each case (Table 1). The treatment step at the beginning of the consultation, before any changes to treatment were made, was used for the analysis. A record was made of any new changes to current prescribed treatment. FENO level. A single measurement of FENO was made using a portable chemiluminescence analyzer (NIOX MINO). A single measurement performed using this device has been shown to have accuracy comparable with 3 measurements using a laboratory-based analyzer.17,18 Except for measuring FENO, current BTS guidelines2 recommend quantifying all of the mentioned severity indices at the regular review. This study was, thus, designed to accurately reflect usual practice in the United Kingdom. Measurement of FENO was performed at the end of the consultation, after treatment decisions had been formulated, to minimize any potential bias associated with this additional procedure compared with conventional best practice. Exacerbation Rate Data Exacerbations were quantified primarily via unscheduled contact with a health care professional for asthma symptoms. Contact could be with the patients’ own GP, the emergency after-hours GP service, or either of the local accident or emergency departments. Patient contact with any of these services generates an immediate written or faxed letter, which is then permanently included in the patients’ primary GP record. The type of treatment initiated after these unscheduled health care contacts did not determine exacerbations. Data were extracted from the GP medical records at least 4 months after the scheduled nurse visit. Exacerbation data were obtained for the 12 months before and 3 months after the scheduled nurse visit. Statistical Analysis Statistical analysis was performed using SPSS version 15 (SPSS Inc, Chicago, Illinois). Spearman rank was used to determine correlations between variables, Wilcoxon rank sum to compare groups, 2 to compare proportions, and logistic regression to evaluate predictors of exacerbation. Tests were deemed significant at Bonferroni-corrected P ⬍ .05. Table 1. British Thoracic Society Guideline Treatment Steps Step 1 Mild intermittent asthma 2 Regular preventive therapy 3 Add-on therapy
4 Persistent poor control 5 Oral corticosteroids
Medication Short-acting bronchodilators ICS, 200–800 g/da A maximum of 800 g/d of an ICS and a LABA, an LRA, or theophyllinea A combination of ⱖ4 asthma medications Regular use of oral corticosteroids
Abbreviations: ICS, inhaled corticosteroid; LABA, long-acting -agonist; LRA, leukotriene receptor antagonist. a Doses are expressed as beclomethasone dipropionate equivalents.
249
RESULTS Demographic details of the 267 enrolled patients are given in Table 2. Two hundred forty-nine participants had complete data; in the remaining 18 patients, it proved technically impossible to obtain a valid FENO measurement (ie, the patient could not perform the controlled exhalation maneuver required to provide a sample). There was no difference in FEV1 between patients in whom FENO was not quantified and the remaining study population (P ⫽ .31); however, the former were significantly older (61 vs 51 years, P ⫽ .01). Most patients were at BTS treatment step 2 or 3; only 2 patients were at step 5, so data from these patients were pooled with those from step 4. Seventy-two patients experienced 141 exacerbations in the 12 months before the visit, and 14 patients had 16 exacerbations in the 3 months subsequent, giving a total of 157 captured episodes. The dose of ICSs increased significantly with increasing treatment step, as did the overall exacerbation rate (Table 3 and Fig 1). In contrast, compared with step 1, FEV1 and FENO were significantly lower in patients at treatment steps 3 to 5 (Table 3 and Fig 1). There was no difference among groups for reliever use, but patients at treatment steps 1, 4, and 5 reported the greatest number of symptoms (Table 3). Level of FENO was significantly negatively correlated with ICS dose, treatment step, and exacerbation rate, whereas positive associations were found with symptom scores and reliever use (Table 4). Exacerbation rate was positively correlated with ICS dose (P ⬍ .001), treatment step (P ⬍ .001), reliever use (P ⬍ .002), and symptom score (P ⬍ .001), whereas the reverse was true for FENO (P ⫽ .04). Other correlations between severity indicators are detailed in Table 4. Direct comparison of patients who experienced disease exacerbation in the 12 months before the study visit and those who did not revealed greater reliever use (P ⫽ .005) and a higher dose of ICSs (P ⬍ .001), but this was not true for those who underwent disease exacerbation in the 3 months after the visit (P ⫽ .32 and P ⫽ .47, respectively) (Table 5). Proportionately more patients at treatment steps 3 to 5 experienced asthma exacerbations in the previous year than those at steps Table 2. Characteristics of the 267 Study Participants Characteristic Male sex, No. (%) Age, mean (SEM), y FEV1, mean (SEM), % of predicted FEV1/FVC, mean (SEM), % Dose of BDP, mean (SEM), g Treatment step, No. (%) Step 1 Step 2 Step 3 Step 4 Step 5
Value 122 (46) 51.6 (1.1) 86.3 (1.2) 77.6 (0.9) 502 (27) 57 (21.3) 99 (37.1) 66 (24.7) 43 (16.1) 2 (0.7)
Abbreviations: BDP, beclomethasone dipropionate; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity.
250
1 and 2 (44/111 [39.6%] vs 28/156 [17.9%], P ⬍ .001); this was also true for the 3 months immediately before the visit (18/111 [16.2%] vs 8/156 [5.1%], P ⫽ .003) but not for the 3 months after the visit (4/111 [3.6%] vs 9/156 [5.8%], P ⫽ .42) (Fig 2). A similar pattern was seen for symptom scores, with a lower proportion of patients exacerbating if they reported 0 or 1 RCP symptoms compared with 2 or 3 symptoms (41/207 [19.8%] vs 31/60 [51.2%], P ⬍ .001, and 14/207 [6.8%] vs 12/60 [20.0%], P ⫽ .002 for the 12 months and 3 months before the visit, respectively). Unlike treatment step, a significantly higher proportion of patients with 2 or more symptoms had a subsequent disease exacerbation in the following 3 months compared with those who had 1 or fewer (7/60 [11.7%] vs 6/207 [2.9%], P ⫽ .005) (Fig 2). Patients at treatment steps 4 and 5, who were receiving high doses of ICS and who exacerbated frequently, had lower FENO values than corticosteroid-naive patients at treatment step 1 (19.7 vs 40.4 ppb, P ⬍ .001) Despite the significant correlations between exacerbation rate and a variety of surrogate measures of asthma severity, logistic regression did not identify any of the quantified variables as significant positive predictors of disease exacerbation in the 12 months before or 3 months after the clinic visit. An RCP symptom score of 0 was identified as a significant negative predictor for exacerbations in the 12 months (P ⫽ .008) and 3 months (P ⫽ .005) before the clinic visit but not for the 3 months after the visit (P ⫽ .45) (ie, no symptoms increased the likelihood of no exacerbations, but the presence of symptoms was a poor predictor of subsequent or impending loss of asthma control). Logistic regression was also used to examine which factors were predictive of the nurses’ decision to step up prescribed therapy. No significant positive associations were found, but, again, there was a negative relationship with symptoms: patients reporting 0 or 1 RCP symptoms were less likely to have an increase to their maintenance treatment initiated by the nurse (P ⬍ .001 and P ⫽ .02, respectively). The nurse initiated changes in regular prescribed medications in 94 patients: 19 had their treatment decreased and 75 had an increase. A significantly greater proportion of those who had their medication increased experienced an exacerbation in the following 3 months compared with those who had no change or a dose reduction (9.3% vs 3.2%, P ⫽ .03). A post hoc analysis of the FENO data was performed, and receiver operating characteristic curves were constructed specifically to evaluate this tool as a predictor of previous and impending loss of disease control. This analysis revealed that for this unselected primary care patient group, FENO was a relatively poor predictor of disease exacerbation, with area under the receiver operating characteristic curves of 0.423 and 0.538 for any exacerbation in the year before and 3 months after the clinic visit, respectively. The sensitivity and specificity for detecting an impending exacerbation in the 3 months after the nurse consultation at a variety of cutoff points were 75.0% and 23.2% (10 ppb), 66.7% and 51.9% (20 ppb), and 33.3% and 69.6% (30 ppb). The corresponding
ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY
Table 3. Differences in Measured Variables by Asthma Treatment Stepa Step 1 FENO, ppb FEV1, % of predicted BDP, g Reliever use, puffs/d Exacerbations, No./y Symptom score
40.4 (4.5) 92.6 (1.9) NA 0.9 (0.2) 0.2 (0.1) 1.1 (0.1)
Step 2
Step 3 b
27.6 (2.8) 87.9 (1.6) 492 (24)b 0.8 (0.1) 0.3 (0.1) 0.7 (0.1)b
Steps 4 and 5 b
23.9 (2.8) 83.5 (3.0)b 533 (27)b 0.8 (0.5) 0.7 (0.1) 0.8 (0.1)
19.7 (2.8)b,c 80.5 (3.8)b 1129 (89)b,c,d 0.9 (0.1) 1.0 (0.2)b,c 1.1 (0.1)c,d
Abbreviations: BDP, beclomethasone dipropionate; FENO, exhaled nitric oxide; FEV1, forced expiratory volume in 1 second; NA, not applicable. a All the values are given as mean (SEM) and are Bonferroni corrected. b Significant difference from step 1, P ⬍ .05. c Significant difference from step 2, P ⬍ .05. d Significant difference from step 3, P ⬍ .05.
Figure 1. Differences in severity measurements stratified by British Thoracic Society treatment step. A, Inhaled corticosteroid (ICS) dose. B, Forced expiratory volume in 1 second (FEV1). C, Exhaled nitric oxide (FENO). D, Exacerbations. BDP indicates beclomethasone dipropionate. * Significant difference from step 1. † Significant difference from steps 1 and 2. ‡ Significant difference from steps 1, 2, and 3. Error bars represent SD.
VOLUME 101, SEPTEMBER, 2008
251
Table 4. Correlations Among Measured Variables in the 12 Months Preceding Clinic Evaluationa FENO FENO FEV1 ICS dose Treatment step Reliever use Exacerbations Symptom score
FEV1
NA ⫺0.98 (.59) ⫺0.172 (.007)b ⫺0.249 (⬍.001)b 0.192 (.002)b ⫺0.131 (.04)b 0.136 (.03)b
ICS dose
Treatment step
Reliever use
Exacerbations
Symptom score
⫺0.98 (.59)b ⫺0.172 (.007)b ⫺0.249 (⬍.001)b 0.192 (.002)b ⫺0.131 (.04)b 0.136 (.03)b NA ⫺0.192 (.002)b ⫺0.214 (⬍.001)b ⫺0.116 (.06) ⫺0.081 (.19) ⫺0.096 (.12) ⫺0.192 (.002)b NA 0.745 (⬍.001) 0.126 (.04)b 0.263 (⬍.001)b 0.070 (.26) ⫺0.214 (⬍.001)b 0.745 (⬍.001) NA 0.039 (.53) 0.269 (⬍.001) 0.030 (.63) ⫺0.116 (.06) 0.126 (.04)b 0.039 (.53) NA 0.188 (.002) 0.377 (⬍.001)b ⫺0.081 (.19) 0.263 (⬍.001)b 0.269 (⬍.001)b 0.188 (.002)b NA 0.350 (⬍.001)b ⫺0.096 (.12) 0.070 (.26) 0.030 (.63) 0.377 (⬍.001)b 0.350 (⬍.001)b NA
Abbreviations: FENO, exhaled nitric oxide; FEV1, forced expiratory volume in 1 second; ICS, inhaled corticosteroid; NA, not applicable. a Values are given are Spearman rho (P value). b Significant correlation, P ⬍ .05. Table 5. Comparison of Measured Variables Between Patients With and Without Exacerbationsa 12 mo before visit Variable FENO, ppb FEV1, % of predicted ICS dose, g BDP Reliever use, puffs/d
3 mo before visit
3 mo after visit
No P value for Exacerbation No P value for No P value for Exacerbation exacerbation difference exacerbation difference Exacerbation exacerbation difference 31.3 (8.3) 85.0 (5.9) 668 (57) 1.4 (0.4)
28.0 (1.7) 86.7 (1.3) 440 (30) 0.8 (0.1)
.07 .11
25.4 (4.3) 83.4 (3.9)
⬍.001b b
.005
673 (118) 1.5 (0.3)
28.4 (1.8) 86.9 (1.3) 484 (27) 0.8 (0.1)
.47 .46 .11 b
.03
31.3 (8.3) 85.0 (5.9) 361 (82) 1.4 (0.4)
28.0 (1.7) 86.7 (1.3) 509 (28) 0.8 (0.1)
.66 .75 .47 .32
Abbreviations: BDP, beclomethasone dipropionate; FENO, exhaled nitric oxide; FEV1, forced expiratory volume in 1 second; ICS, inhaled corticosteroid. a All values are given as mean (SEM) and are Bonferroni corrected. b Significant difference, P ⬍ .05.
values for exacerbation in the 12 months before the visit were 68.2% and 16.9% (10 ppb), 40.9% and 48.1% (20 ppb), and 27.3% and 68.3% (30 ppb). DISCUSSION These data demonstrate a paradoxical relationship between airway inflammation and other measures of asthma severity, with airway caliber and symptoms and exacerbation rate disconnected from FENO measurement. Airway Inflammation The observation that symptoms and spirometry measurements are unrelated to exacerbation rate has been reported in several prospective studies in which measurements of airway inflammation were used to guide dose adjustment in secondary care. Sont et al7 and Green et al6 successfully improved asthma exacerbation rates compared with the conventional symptom and spirometry-based approach by titrating ICS dose against methacholine hyperresponsiveness and induced sputum eosinophil count, respectively. Quantification of FENO is a less invasive method of estimating asthmatic bronchial inflammation that can be performed in primary care using a portable analyzer. Level of FENO is correlated with a variety of other measures of airway inflammation and has been shown to predict loss of asthma control in a trial setting, particularly when combined with FEV1.8,9,19,20 It has also been shown to be as good as methacholine challenge (often con-
252
sidered the gold standard) for diagnosing asthma in a preselected corticosteroid-naive outpatient population.21 These data suggest that this method of detecting airway inflammation may not be clinically relevant in a primary care setting. Level of FENO was negatively correlated with ICS dose, treatment step, and exacerbation rate; did not accurately predict exacerbation in the bivariate model; and demonstrated poor receiver operating characteristic curve characteristics for exacerbation rate in the post hoc analysis. Although FENO measurement is a valuable method for assessing disease activity in a clinical trial, where it can help differentiate among subtle pharmacotherapeutic effects, the very steep relationship between ICS dose and attenuation of this inflammatory index renders it a comparatively blunt instrument for dose titration in clinical practice.22–24 This may explain why prospective clinical trials using this surrogate marker to make dose adjustments have not demonstrated a clear improvement in exacerbation rate over conventional symptom- and spirometry-based strategies.13,14,25 Levels of FENO are maximally suppressed in patients with more severe disease (receiving the highest doses of maintenance ICS) and, therefore, are of the least clinical utility in patients who potentially stand to gain the most from better disease control. Conventional Disease Severity Measurements The present data demonstrate that there are correlations among many of the guideline-recommended severity mea-
ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY
BTS treatment step, increased maintenance ICS requirement, increased symptoms, and more frequent reliever medication use), note that none of these factors were powerful enough to act as predictors of disease exacerbation. This suggests that these measurements are not precise enough tools on which to confidently base management decisions, particularly regarding dose and add-on adjustment of maintenance pharmacotherapy. Note that patients with a nurse-initiated increase in maintenance treatment were more likely to experience subsequent disease exacerbation than were those without. A possible explanation for this is that if a patient were experiencing symptoms heralding the beginning of a disease exacerbation at the time of consultation, then an increase in treatment would be initiated.
Figure 2. Proportionate number of patients experiencing disease exacerbation stratified by British Thoracic Society guideline treatment step (A) and symptom score (B). Percentages indicate the proportion in each group experiencing disease exacerbation.
surements currently used in asthma management in primary care. However, it also demonstrates that none of these measurements were accurate methods by which to predict patients at risk for disease exacerbation. This is important because reduction in exacerbation rate is a primary goal of long-term asthma management because of not only the significant morbidity but also the socioeconomic burden associated with disease flares.26 However, we appreciate that the frequency of exacerbations will be relatively low for most patients in primary care, who have mild to moderate persistent asthma. The ICS dose was related to treatment step, as the BTS stepwise classification incorporates this as an index of severity. The FEV1 also declined with increased ICS dose and BTS treatment step, again reflecting disease severity. Although exacerbation rates were positively correlated with the quantified surrogate measurements of disease severity (higher
VOLUME 101, SEPTEMBER, 2008
Asthma Management in Primary Care Most of the 267 patients studied were at treatment step 2 or 3 of the BTS guidelines, yet examining only 15 months of data for each patient captured more than 150 exacerbations. This reflects previous trial data27 suggesting that most asthmatic patients have mild to moderate disease but nonetheless experience significant associated morbidity. Note that patients at BTS treatment step 1 had the highest levels of FENO and were relatively symptomatic, suggesting that these patients may be relatively undertreated. Asthma is a chronic disorder, and for most patients, nonspecialists in primary care successfully manage the disease without input from respiratory physicians or secondary care. In the United Kingdom, appropriately trained nurses undertake much of the long-term asthma management, with indirect supervision from GPs. The BTS asthma guidelines advocate regular disease monitoring in primary care using precisely this arrangement, yet few data are available to substantiate such an approach. The lack of data relevant to primary care may be partly compounded because most asthma studies are performed in select patients and usually in the context of a university research department or a secondary care facility. A recent study28 suggested that as little as 4% of asthmatic patients met the eligibility criteria for most randomized controlled trials on which current international guidelines are based. The present study is unique in that it gathered data on unselected asthmatic patients in the context of usual United Kingdom, guideline-based practice. Study Limitations Data were gathered only for severe exacerbations requiring treatment by a physician and not for mild to moderate exacerbations; the relationship between surrogates of disease severity and disease control could be explored more completely with this additional information. Follow-up was relatively short (3 months), and extending this may have strengthened the study. Note that exacerbations beyond 3 months would be unlikely to have any direct relationship with the clinic measurements and would, therefore, provide limited additional information. Most data in relation to FENO measurement in asthma relates specifically to atopic asthma, and these data
253
need to be interpreted without certainty about the allergic status of the participants. Also, note that the number of exacerbations in the prospective component of the trial was relatively small, and, therefore, the trial may be underpowered in this regard. Although FENO did not predict exacerbation rate, note that none of the other “conventional” asthma measures identified patients at risk for exacerbation either. It is conceivable that combining FENO with another surrogate of disease severity, or using the test in more select patients (eg, those with documented eosinophilic airway disease), might improve the accuracy and clinical value. These data do not address the potential utility of FENO measurement in other clinical scenarios. A single measurement of FENO may help discriminate between asthmatic and nonasthmatic individuals in a corticosteroid-naive population,21 and it may help determine which patients would benefit from ICS treatment.29 Longitudinal measurement of FENO may be valuable in assessing medication compliance and in determining whether treatment adjustments are required. In severe asthma, there may be a significant neutrophilic element to disease pathogenesis, and, as such, a low FENO level may be clinically important in distinguishing these patients from those with predominantly eosinophilic airway inflammation in whom FENO levels are elevated. CONCLUSION Some of the contrasts observed (particularly regarding FENO measurements) between the data presented and those previously reported may be explained by the differing context and patient demographics seen in clinical trials compared with clinical practice. Although new asthma management strategies aimed at reducing exacerbations by using surrogate measures of inflammation or any other index of disease severity to guide therapy may be warranted, it will be important to ensure that they are carefully validated in the target population. This study demonstrated that spirometry measurements, airway inflammation, and symptoms are poorly correlated with each other in a primary care cohort of asthmatic patients. In patients with more severe asthma (where better disease control would be most desirable), FENO is maximally suppressed secondary to concomitant use of relatively high-dose ICS and may, therefore, be of limited utility in guiding treatment decisions in a primary care population. ACKNOWLEDGMENTS We thank the GP nurses throughout Tayside who participated in this study.
4. 5. 6. 7.
8. 9. 10. 11. 12.
13. 14. 15. 16. 17. 18. 19. 20. 21. 22.
REFERENCES 1. Global Initiative for Asthma (GINA). Global Strategy for Asthma Management and Prevention. Bethesda, MD: National Heart, Lung, and Blood Institute; January 1995 (updated 2002, 2003). NIH publication 02–3659. 2. British Thoracic Society Scottish Intercollegiate Guidelines Network. British Guideline on the Management of Asthma. Thorax. 2008;63(suppl 4):iv1-iv121. 3. Charlton I, Charlton G, Broomfield J, Mullee MA. Audit of the effect of
254
23.
24.
a nurse run asthma clinic on workload and patient morbidity in a general practice. Br J Gen Pract. 1991;41:227–231. Dickinson J, Hutton S, Atkin A, Jones K. Reducing asthma morbidity in the community: the effect of a targeted nurse-run asthma clinic in an English general practice. Respir Med. 1997;91:634 – 640. Lindberg M, Ahlner J, Ekstrom T, Jonsson D, Moller M. Asthma nurse practice improves outcomes and reduces costs in primary health care. Scand J Caring Sci. 2002;16:73–78. Green RH, Brightling CE, McKenna S, et al. Asthma exacerbations and sputum eosinophil counts: a randomised controlled trial. Lancet. 2002; 360:1715–1721. Sont JK, Willems LN, Bel EH, van Krieken JH, Vandenbroucke JP, Sterk PJ; AMPUL Study Group. Clinical control and histopathologic outcome of asthma when using airway hyperresponsiveness as an additional guide to long-term treatment. Am J Respir Crit Care Med. 1999; 159(pt 1):1043–1051. Jatakanon A, Lim S, Kharitonov SA, Chung KF, Barnes PJ. Correlation between exhaled nitric oxide, sputum eosinophils, and methacholine responsiveness in patients with mild asthma. Thorax. 1998;53:91–95. Jones SL, Kittelson J, Cowan JO, et al. The predictive value of exhaled nitric oxide measurements in assessing changes in asthma control. Am J Respir Crit Care Med. 2001;164:738 –743. Strunk RC, Szefler SJ, Phillips BR, et al. Relationship of exhaled nitric oxide to clinical and inflammatory markers of persistent asthma in children. J Allergy Clin Immunol. 2003;112:883– 892. Harkins MS, Fiato KL, Iwamoto GK. Exhaled nitric oxide predicts asthma exacerbation. J Asthma. 2004;41:471– 476. Fabbri LM, Romagnoli M, Corbetta L, et al. Differences in airway inflammation in patients with fixed airflow obstruction due to asthma or chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2003;167:418 – 424. Smith AD, Cowan JO, Brassett KP, Herbison GP, Taylor DR. Use of exhaled nitric oxide measurements to guide treatment in chronic asthma. N Engl J Med. 2005;352:2163–2173. Pijnenburg MW, Bakker EM, Hop WC, De Jongste JC. Titrating steroids on exhaled nitric oxide in children with asthma: a randomized controlled trial. Am J Respir Crit Care Med. 2005;172:831– 836. Jayaram L, Pizzichini MM, Cook RJ, et al. Determining asthma treatment by monitoring sputum cell counts: effect on exacerbations. Eur Respir J. 2006;27:483– 494. Miller MR, Hankinson J, Brusasco V, et al. Standardisation of spirometry. Eur Respir J. 2005;26:319 –338. Menzies D, Nair A, Lipworth BJ. Portable exhaled nitric oxide measurement: comparison with the “gold standard” technique. Chest. 2007;131:410 – 414. Alving K, Janson C, Nordvall L. Performance of a new hand-held device for exhaled nitric oxide measurement in adults and children. Respir Res. 2006;7:67. Gelb AF, Flynn Taylor C, Shinar CM, Gutierrez C, Zamel N. Role of spirometry and exhaled nitric oxide to predict exacerbations in treated asthmatics. Chest. 2006;129:1492–1499. Pijnenburg MW, Hofhuis W, Hop WC, De Jongste JC. Exhaled nitric oxide predicts asthma relapse in children with clinical asthma remission. Thorax. 2005;60:215–218. Berkman N, Avital A, Breuer R, Bardach E, Springer C, Godfrey S. Exhaled nitric oxide in the diagnosis of asthma: comparison with bronchial provocation tests. Thorax. 2005;60:383–388. Fardon TC, Burns P, Barnes ML, Lipworth BJ. A comparison of 2 extrafine hydrofluoroalkane-134a-beclomethasone formulations on methacholine hyperresponsiveness. Ann Allergy Asthma Immunol. 2006; 96:422– 430. Menzies D, Nair A, Fardon T, Barnes M, Burns P, Lipworth B. An in vivo and in vitro comparison of inhaled steroid delivery via a novel vortex actuator and a conventional valved holding chamber. Ann Allergy Asthma Immunol. 2007;98:471– 479. Menzies D, Nair A, Hopkinson P, McFarlane L, Lipworth BJ. Differential anti-inflammatory effects of large and small particle size inhaled corticosteroids in asthma. Allergy. 2007;62:661– 667.
ANNALS OF ALLERGY, ASTHMA & IMMUNOLOGY
25. Shaw DE, Berry MA, Thomas M, et al. The use of exhaled nitric oxide to guide asthma management: a randomized controlled trial. Am J Respir Crit Care Med. 2007;176:231–237. 26. Lloyd A, Price D, Brown R. The impact of asthma exacerbations on health-related quality of life in moderate to severe asthma patients in the UK. Prim Care Respir J. 2007;16:22–27. 27. Dusser D, Montani D, Chanez P, et al. Mild asthma: an expert review on epidemiology, clinical characteristics and treatment recommendations. Allergy. 2007;62:591– 604. 28. Travers J, Marsh S, Williams M, et al. External validity of randomised controlled trials in asthma: to whom do the results of the trials apply? Thorax. 2007;62:219 –223. 29. Smith AD, Cowan JO, Brassett KP, et al. Exhaled nitric oxide: a
VOLUME 101, SEPTEMBER, 2008
predictor of steroid response. Am J Respir Crit Care Med. 2005;172: 453– 459.
Requests for reprints should be addressed to: Brian J. Lipworth, MD Asthma and Allergy Research Group Department of Medicine and Therapeutics Ninewells Hospital and Medical School Dundee, Scotland DD1 9SY E-mail: [email protected]
255