Population and setting. We studied patients with asthma aged 13 years or older, of both sexes, in the city of São Paulo, Brazil. Patients were recruited from the emergency department of Hospital São Paulo (a teaching hospital of the Federal University of São Paulo) and from two public freestanding urgent care centers affiliated with the hospital: AMA Santa Cruz and AMA Sacomã.
We included patients with a previous diagnosis of asthma (dyspnea, cough, wheezing, chest tightness, associated with allergen exposure or cold air)20,21 who received follow-up at outpatient clinics within the catchment area of the Hospital São Paulo emergency department and had a peak flow <50% of predicted. All participants had a longstanding history of asthma, with repeated exacerbations and emergency room visits. Our included patients had at least 2 years of moderate or severe asthma, with a mean peak flow immediately before intervention of 163 L/min.
We excluded patients with body temperature ≥37.8°C, smokers, pregnant women, patients undergoing psychiatric treatment, patients with a history of heart, liver, kidney, or other disease that might contraindicate corticosteroid therapy, patients who had undergone lung resection, patients undergoing treatment for tuberculosis or mycotic infections of the lung, and patients with tracheotomy or mechanical obstruction of the trachea. We also excluded patients with myopathies or neurological conditions (such as sequelae of stroke or encephalopathies), as well as patients with BMI >40 kg/m².
Ethical Aspects. This study was approved by the Research Ethics Committee of the Federal University of São Paulo (judgment number 364240). All patients provided written informed consent for participation in accordance with international regulations for human subject research. When patients were underage, consent was obtained from their parents or legal guardians.
This study is registered in the Brazilian Registry of Clinical Trials (http://www.ensaiosclinicos.gov.br/ ) under accession number RBR-6XWC26.
Sample. We studied 31 patients in the ciclesonide group and 27 patients in the hydrocortisone group. We calculated sample size according to Greenberg22, considering a FEV1 improvement of 0.37±0.85 L after intervention, resulting in 65 patients for each group.
Study design. This is a double-blind, placebo-controlled, randomized clinical trial designed to compare the efficacy of inhaled ciclesonide versus intravenous hydrocortisone in the management of moderate or severe acute asthma in an emergency department setting.
Blinding. Both blinding and randomization were done centrally at Neuro-Sono Sleep Center, São Paulo, Brazil. Blinding of active ingredients and their respective placebos was achieved by random allocation of four letters – A, B, C, and D – to each of the following products: hydrocortisone, ciclesonide, hydrocortisone-identical placebo, and ciclesonide-identical placebo. After random allocation of letters to designate each product, we defined two product pairs, Inhaled Active Ingredient + Intravenous Placebo and Intravenous Active Ingredient + Inhaled Placebo, in a random combination that enhanced the safety of blinding. Both the intravenous placebo and the inhaled placebo were identical to their active counterparts.
Information about the intervention that each patient randomized to the study would receive was distributed in opaque, numbered envelopes, which were only opened at the time of use. The nursing staff prepared the medications for administration as instructed in the numbered envelopes. The staff who prepared the medications, the providers who administered them, and all researchers involved were blinded to the active pharmaceutical ingredients of interest.
Randomization. Patients included in the sample were recorded consecutively in a logbook and assigned a serial number.
The 58 patients were divided into 2 groups: study (Ciclesonide) and control (Hydrocortisone), according to two computer-generated random number tables. Each table contained an ascending sequence of numbers. Patients were allocated to one or the other according to the serial number attributed at the time of enrollment, which ensured that neither staff nor patients were aware of the intervention to which each patient would be allocated.
Ciclesonide Group. Patients received ciclesonide at a dose of 160 mcg/puff. The first dose was administered 5 minutes after inclusion in the trial, and consisted of 3 puffs (480 mcg); the second dose at 20 minutes (480 mcg); and the third dose at 40 minutes (480 mcg), for a total of 1440 mcg. Patients in this group also received hydrocortisone-identical placebo at 5 minutes.
Hydrocortisone Group. Patients in this group received 500mg of hydrocortisone intravenously and ciclesonide-identical placebo at 5, 20, and 40 minutes.
Both Groups. Both groups received short-acting bronchodilators (fenoterol hydrobromide and ipratropium bromide) at 0, 10, and 30 minutes.
Measures. We adopted as a primary outcome measure the spirometric variables FEV1 and peak expiratory flow (PEF), as well as the clinical variables dyspnea, wheezing, and accessory muscle use during breathing (assessed by observation of the sternocleidomastoid muscle). As secondary outcomes, we evaluated the heart rate, respiratory rate, blood pressure, and pulse oximetry.
These parameters were measured every 30 minutes from the time of patient admission until the second hour and every 60 minutes thereafter until the fourth hour in the emergency department, for a total of 7 measurements, ensuring rigorous monitoring throughout the patient observation period. For the purposes of this study, we analyzed data from baseline and the fourth hour, as we felt these assessments were sufficient to represent the patients’ course among the 7 measurements obtained.
Procedures. The emergency room nurse applied the Manchester Triage System and measured oxygen saturation, blood pressure, and breathing pattern. The emergency room physician then confirmed the diagnosis of asthma exacerbation and notified the investigators, who performed an initial assessment by measuring peak flow and explaining the study to the patient. Patients with a peak flow less than 50% of predicted were invited to participate in the study (Figure 1), as the sample was designed to include only severe patients.
Once the patient was included, the investigators worked with the emergency department staff to provide all necessary care and perform the measurements required for the study.
Spirometric parameters were measured in an Easy One model 2009 spirometer (ndd Medizintechnik AG, Zurich, Switzerland). The best of three successive expiratory curves was considered valid and used for analysis, as recommended by the American Thoracic Society. Peak flow was estimated with the Mini-Wright Peak Flow meter (Clement Clarke, Hanlon, United Kingdom). Again, the highest of three measurements was considered for analysis. Dyspnea was assessed subjectively as perceived shortness of breath, using the Borg scale, a visual analogue scale of 0 to 10 where 0 is absence of dyspnea and 10 is maximal dyspnea. During the initial assessment and at each time point of reassessment, we evaluated wheezing and accessory muscle use. Wheezing was assessed through pulmonary auscultation and ranked from 0 to 3 on an ascending scale of severity (0: no wheezing; 1: slight wheezing; 2: moderate wheezing; 3: severe wheezing). Accessory muscle use was also measured on a scale of increasing intensity (0: no accessory muscle activity: 1: slight activity; 2: moderate activity; 3: marked accessory muscle activity). When there was little wheezing or a silent chest plus marked accessory muscle use or signs of muscle fatigue, dyspnea was classified as severe. Individual and pooled analyses were performed for all parameters.
Criteria for Improvement. Patients were evaluated for improvement at all time points of assessment, to ensure patient safety and detect possible need for additional interventions other than those provided for in the study protocol. For the purposes of this study, we considered the following definitions of improvement: 1) FEV1 and PEF ≥70% predicted for age, sex, weight, and height; 2) Improvement of dyspnea: a) Borg score <223; b) reduction of wheezing severity from baseline; and c) no accessory muscle use, as determined by observation of the sternocleidomastoid muscles.
Interim Analysis. We planned an interim analysis for when the number of patients included had reached approximately half the predicted sample size, to decide whether to continue or terminate inclusion. This analysis was carried out at the randomization and blinding center (Neuro-Sono Sleep Center) by a committee established specifically for this purpose. After inclusion of 58 patients, the Interim Analysis Committee suggested that the study be interrupted, since no difference between treatments was detected.
Adverse Events. We actively evaluated the more frequent adverse events, such as dry mouth, tremor, palpitations, anxiety, headache, and recorded any other patient-reported events24,25. These variables were evaluated by intention to treat (ITT).
Statistical Analysis. The sample size was calculated considering a change in FEV1 of 0.37 L, after treatment, as an indicator of improvement; a standard deviation of 0.85 L; a significance level of 5%; and a statistical power of 80%16,22, resulting in a sample size of n=130 patients, i.e., 65 patients in each group. As noted above, interim analyses were carried out as planned after enrollment of 30 patients in each group; at this time, in view of the results, the interim analysis committee recommended termination of enrollment.
Quantitative variables were expressed as mean ± SD, and categorical variables, as n (%). We used Student’s t-test for independent samples for normally distributed data, the Mann–Whitney U test for asymmetrically distributed data, and Pearson's chi-square test or Fisher’s exact test for categorical data26,27. Outcomes were assessed by ITT, considering the worst scenario, i.e., losses in the study group were considered treatment failures and losses in the control group as successful treatment. P-values < 0.05 were considered statistically significant.
Availability of Data and Materials. All data generated and analysed during this study are included in this published article (supplementary information files).