CLINICAL INVESTIGATIONS

Transcription

1 ACAD EMERG MED September 2002, Vol. 9, No CLINICAL INVESTIGATIONS A Randomized Controlled Trial of Mist in the Acute Treatment of Moderate Croup Gina M. Neto, MD, Osama Kentab, MD, Terry P. Klassen, MD, Martin H. Osmond, MD Abstract Objective: To determine whether the use of mist improves clinical symptoms in children presenting to the emergency department (ED) with moderate croup. Methods: Children 3 months to 6 years of age were eligible for the study if they presented to the ED with moderate croup. Moderate croup was defined as a croup score of 2 7. The patients were randomly assigned to receive either mist (humidified oxygen) via mist stick or no mist. The patients had croup scores measured at baseline and every 30 minutes for up to two hours. At these intervals the following parameters were also measured: heart rate, respiratory rate, oxygen saturation, and patient comfort score. The patients were treated until the croup score was less than 2 or until two hours had elapsed. All patients initially received a dose of oral dexamethasone (0.6 mg/kg). Other treatments, such as racemic epinephrine or inhaled budesonide, were given at the discretion of the treating physician. The research assistants were unaware of the assigned treatments. Results: There were 71 patients enrolled in the study; 35 received mist and 36 received no mist. The two treatment groups had similar characteristics at baseline. The median baseline croup score was 4 in both groups. The outcomes were measured as the change from baseline at 30, 60, 90, and 120 minutes. The change in the croup score from baseline in the mist group was not statistically different from the croup score change in the group that did not receive mist (p = 0.39). There was also no significant difference in improvement of oxygen saturation, heart rate, or respiratory rate at any of the assessment times. There was no adverse effect from the mist therapy. Conclusions: Mist therapy is not effective in improving clinical symptoms in children presenting to the ED with moderate croup. Key words: mist; humidified oxygen; croup; children; pediatrics. ACADEMIC EMERGENCY MEDICINE 2002; 9: Croup (acute laryngotracheobronchitis) is the most common form of upper airway obstruction in children. Children between the ages of 6 months and 6 years are primarily affected. The peak incidence is 4.6 cases per 100 in children aged 1 to 2 years, 1 and from 1.3% to 5% of children with croup are hospitalized. 1 3 Croup is often associated with a viral upper respiratory tract infection, most commonly, parainfluenza virus. The viral infection leads to inflammation of the nasopharynx and the upper airway, resulting in subglottic narrowing. Treatment consists of supportive care, systemic glucocorticoids, and inhaled racemic epinephrine (or L-epinephrine 1:1000) for more severe cases. Humidified oxygen (mist) is often used in children with croup, although its use remains controversial. From the Department of Pediatrics, University of Ottawa, Ottawa, Ontario, Canada (GMN, OK, MHO); and the Department of Pediatrics, University of Alberta, Edmonton, Alberta, Canada (TPK). Received December 31, 2001; revision received May 23, 2002; accepted May 23, Address for correspondence and reprints: Gina M. Neto, MD, Division of Emergency Medicine, Children s Hospital of Eastern Ontario, 401 Smyth Road, Ottawa, Ontario, Canada K1H 8L1. Fax: ; neto@cheo.on.ca. Humidified air has been used for many years in the treatment of croup. Croup kettles and, later, tents were in widespread use. It has been postulated that treatment with humidified oxygen works by two mechanisms. The mist soothes the inflamed laryngeal mucosa and decreases airway inflammation. It may also moisten secretions, making it easier for the child to clear them from the airway. This causes some relief of airway obstruction and the patient feels more comfortable. Conversely, it has been suggested that mist may actually make the child more uncomfortable. The sensation of mist being blown into the face can make some children anxious, leading to hyperventilation and a worsening of their respiratory obstruction. Many centers continue to routinely use humidified oxygen in the treatment of croup despite the fact that only two studies have previously investigated the effect of mist in croup. 4,5 Mist therapy may cause mild improvement in the clinical symptoms of croup, but there is lack of scientific evidence to support its use. We have designed a clinical trial to determine whether the use of mist as compared with no mist leads to clinical improvement in children presenting to the emergency department (ED) with moderate croup.

2 874 Neto et al. MIST FOR MODERATE CROUP METHODS Study Design. This was a randomized, controlled trial conducted in the ED of a tertiary care children s hospital that serves as the pediatric referral center for the region. Prior to enrollment of the child in the study, informed written consent was obtained from all parents. The study protocol was approved by the hospital ethics committee. Study Setting and Population. Children aged 3 months to 6 years presenting to the ED of the Children s Hospital of Eastern Ontario from October to December in three consecutive years, , were eligible for the study if they presented with a syndrome consistent with croup (hoarseness, stridor, and barking cough) and had a croup score of 2 or more (Table 1). Patients were excluded if they had epiglottitis, chronic respiratory disease, tuberculosis, chickenpox or exposure to chickenpox in the past three weeks, or oxygen saturation less than 92%. Patients with asthma were excluded if wheezing was present during the croup episode. Study Protocol. Eligible patients were randomly assigned to receive either humidified oxygen delivered by a mist stick or no mist. The mist group received humidified oxygen via a mist stick, which is a setup consisting of an Adjustable Entrainment Nebulizer Adapter and a plastic bottle of 750 ml Sterile Water for Inhalation (Automatic Liquid Packaging, Inc., Woodstock, IL) through which 15 L/min oxygen is humidified and then delivered to the patient by a corrugated plastic tubing (Hudson RCI, Temecula, CA). The parent is instructed to hold the tubing within 15 cm of the face throughout the treatment. The no-mist group was placed in an assessment room that contained the mist stick apparatus, but this was not turned on by the emergency nurse. The parents were informed that no mist would be used for their child. All patients received an oral dose of dexamethasone (0.6 mg/kg). The patient was given inhaled epinephrine or budesonide at the discretion of the treating physician. Randomization was performed in blocks of 10 by using a computer-generated table. The randomization list was generated by the pharmacy department, which was not involved in the study. The treatment assignments were concealed in sequentially numbered opaque envelopes. The research assistant and the treating physicians were blinded to the treatment assignment. The parents were informed not to tell the research assistant or treating physician whether the child was receiving mist. Prior to each assessment the child was brought to TABLE 1. Croup Score Clinical Scoring System* Stridor None 0 Audible with stethoscope (at rest) 1 Audible without stethoscope (at rest) 2 Retractions None 0 Mild 1 Moderate 2 Severe 3 Air entry Normal 0 Decreased 1 Severely decreased 2 Cyanosis None 0 With agitation 1 At rest 2 Level of consciousness Normal 0 Altered 5 *Reproduced with permission from: Westley CR, Cotton EK, Brooks JG. Nebulized racemic epinephrine by IPPB for the treatment of croup. Am J Dis Child. 1978; 132: Copyrighted (1978), American Medical Association. a separate assessment area by an emergency nurse, not involved with the study. The child then returned to the room to resume treatment. We tested adequacy of blinding by asking both the research assistant and the treating physician to state to which treatment group they thought the child belonged. Measures. The primary outcome measure was the croup score developed by Westley et al. 6 This has been used in other clinical trials of patients with croup and has been shown to be valid, reliable, and sensitive to change. 7,8 Secondary outcome measures included respiratory rate, heart rate, oxygen saturation, length of stay in the ED, and admission to the hospital. The research assistants, treating physicians, and parents were asked to rate the change in the patient s condition compared with when they arrived. The global assessment of change was measured using a 15-point Likert scale ( 7 = a very great deal worse to 7 = a very great deal better ). We also assessed the level of patient comfort by asking the parents to assess the patient s level of comfort at each of the assessment times compared with their level of comfort when they arrived in the ED. The patient s level of comfort was assessed using a seven-point scale (1 = extremely comfortable to 7 = extremely uncomfortable ). The uses of racemic

3 ACAD EMERG MED September 2002, Vol. 9, No epinephrine and inhaled budesonide were viewed as cointerventions. The outcome measures were assessed at baseline and every 30 minutes until two hours had elapsed, or the croup score had decreased to 1 or 0. The trial was stopped if the patient required supplemental oxygen or if the respiratory status deteriorated such that the patient required intubation or admission to the intensive care unit. The baseline measurement was obtained prior to the administration of the dexamethasone or, if given, any inhaled treatment. Data Analysis. The sample size was calculated based on observing a difference in croup score of 1 between the two groups. This difference of 1 is the minimally clinical important difference and is based on previously published data. 7,8 Assuming a type I (two-sided) error of 0.05 and type II error of 0.20 (i.e., statistical power of 80%) resulted in a sample size of 35 patients per group. Baseline demographic data were analyzed using descriptive statistics. Changes in an outcome measure were related to treatment, time, baseline measure, and treatment by time interaction using an analysis of covariance (ANCOVA) model. The model accounted for repeated measures and included two sources of variation: a random variation due to patient and a variation due to chance error. Differences in the final croup scores were compared using the baseline scores as the covariate. Multiple linear regression and logistic regression were used to determine whether any of the historical variables listed were predictive of responsiveness to humidified oxygen. Differences between groups in admission rates were assessed using contingency table analysis with 2 degrees of freedom. Data were analyzed using SPSS 8.0 data entry and statistical package (SPSS Inc., Chicago, IL). RESULTS Patient Characteristics. During the study period, 648 patients had a diagnosis of croup. Of these patients, 514 did not meet the eligibility criteria. There were 473 patients with croup symptoms that were too mild (croup score < 2 ), 19 patients had mist treatment started upon arrival, 13 patients had asthma with wheezing, and nine patients had oxygen saturation < 92%. There were 61 patients who were not recruited because the research assistant was unavailable or the ED failed to contact them. There were two patients whose parents refused to participate in the study (Fig. 1). There were 71 patients enrolled in the study; 35 received mist and 36 received no mist. Table 2 shows the patient characteristics at baseline. The two treatment groups were similar at baseline. The median baseline croup score was 4 in both groups. Oxygen saturations, heart rates, and respiratory rates were also similar in the two groups at baseline. Both racemic epinephrine and budesonide are given with humidified oxygen and, therefore, if a child received either of these he or she received a mist-like treatment. A separate analysis was conducted with these patients removed. There were 20 patients who each received one treatment of inhaled racemic epinephrine (mist n = 15, no mist n = 5). There were six patients who each received one treatment of inhaled budesonide (mist n = 1,no mist n = 5). At baseline, the groups were similar with a median croup score of 4 in both groups and no difference in heart rate, respiratory rate, and oxygen saturation. The results from this separate analysis did not lead to any different conclusions. There continued to be no significant difference between the two groups in any of the outcome measures. Outcome Variables. The outcomes were measured Figure 1. Patient flowchart.

4 876 Neto et al. MIST FOR MODERATE CROUP TABLE 2. Patient Characteristics at Baseline Mist (n = 35) No Mist (n = 36) p-value* Patient characteristics Age (months) 18 (12 24) 22.5 ( ) 0.47 No. (%) of males 24 (69) 23 (64) 0.80 No. (%) with upper respiratory tract infection in past 5 days 27 (77) 29 (81) 0.78 No. (%) with croup symptoms < 1 day 17 (49) 12 (33) 0.23 No. (%) with asthma 1 (3) 4 (11) 0.36 No. (%) with family history of asthma 11 (31) 8 (22) 0.43 No. (%) with steroid use in past week 1 (3) 5 (14) 0.20 No. (%) who used mist at home 13 (37) 12 (33) 0.81 No. (%) treated with racemic epinephrine 15 (42) 5 (14) 0.01 No. (%) treated with budesonide 1 (3) 5 (14) 0.08 Baseline measures Croup score 4 (3 4) 4 (3 5) 0.38 Oxygen saturation (%) 97 (96 99) 96 (95 98) 0.09 Heart rate (beats/min) 144 ( ) 150 ( ) 0.55 Respiratory rate (breaths/min) 31 (26 42) 32 (26 39) 0.85 *The p-value is for descriptive purposes only. Data are given as median (25th 75th percentile). TABLE 3. Treatment Results Outcome Variable Time (Min) Patients (n) Mist Change from Baseline* No Mist Change from Baseline* Treatment Difference* p-value Croup score ( 1.6, 0.6) n = ( 1.3, 0.4) n = ( 0.9, 0.3) ( 2.2, 1.3) n = ( 1.9, 1.0) n = ( 0.9, 0.4) ( 3.0, 1.9) n = ( 2.7, 1.7) n = ( 1.0, 0.5) ( 3.5, 2.3) n = ( 3.1, 2.0) n = ( 1.1, 0.6) 0.58 Overall p = 0.39 Oxygen saturation (%) ( 0.1, 1.1) 0.1 ( 0.5, 0.6) 0.5 ( 0.4, 1.3) (0.2, 1.5) 0.1 ( 0.6, 0.5) 0.9 (0.1, 1.8) ( 0.3, 1.3) 0.1 ( 0.7, 0.9) 0.4 ( 0.7, 1.5) (0.1, 1.9) 0.8 ( 0.1, 1.6) 0.2 ( 1.0, 1.5) 0.14 Overall p = 0.17 Heart rate ( 11.1, 1.5) 3.2 ( 9.3, 2.9) 1.5 ( 10.3, 7.2) ( 12.7, 0.7) 6.7 ( 13.0, 0.4) 0.7 ( 8.5, 9.9) ( 18.4, 2.2) 4.8 ( 12.5, 2.9) 5.5 ( 16.7, 5.7) ( 16.8, 1.7) 10.6 ( 18.8, 2.4) 3.0 ( 9.3, 15.4) 0.40 Overall p = 0.84 Respiratory rate ( 4.0, 0.4) 2.7 (4.8, 0.5) 0.9 ( 2.2, 3.9) ( 6.0, 1.3) 3.9 ( 6.1, 1.7) 0.2 ( 3.0, 3.4) ( 8.0, 2.2) 4.1 ( 6.8, 1.4) 1.0 ( 5.0, 3.0) ( 8.3, 1.6) 6.1 ( 9.0, 3.2) 1.2 ( 3.3, 5.6) 0.64 Overall p = 0.82 *Change was defined as the difference between baseline minus treatment values. Changes in an outcome measure were related to treatment, time, baseline measure, and treatment by time interaction using an analysis of covariance model. The model accounted for repeated measures and included two sources of variation: a random variation due to patient and a variation due to chance error. Least-square estimates of the mean changes over time and treatment effects were derived from the fitted models. as the change from baseline at 30, 60, 90, and 120 minutes. All patients improved over time. The change in the croup score from baseline in the mist group was not statistically different from the croup score change in the group that did not receive mist. There was also no significant difference in improvement of oxygen saturation, heart rate, or respiratory rate at any of the assessment times (Table 3). The global assessment of change scores by the parents, physicians, and research assistants were

5 ACAD EMERG MED September 2002, Vol. 9, No TABLE 4. Parental Assessment of Comfort Time (Min) Mist (n = 35) Change from Baseline* No Mist (n = 36) Change from Baseline* Treatment Difference p-value (2.5, 3.3) 3.2 (2.8, 3.6) 0.3 ( 0.8, 0.3) (2.3, 3.2) 2.8 (2.4, 3.2) 0.1 ( 0.6, 0.6) (2.0, 3.0) 2.1 (1.6, 2.6) 0.4 ( 0.3, 1.1) (1.3, 2.5) 2.0 (1.5, 2.6) 0.1 ( 0.9, 0.7) 0.86 Overall p = 0.96 *The question was phrased so that the assessment given was in comparison with when the child arrived (baseline measurement). not different between the two groups. The parental assessment of patient comfort is shown in Table 4. There was no difference in the comfort score between the two groups at any of the assessment times. There was no difference in the lengths of stay in the ED. At the end of the study period, 17 patients required further observation and treatment (mist group = 12, no mist = 5). One patient required hospital admission (mist group). There were no adverse outcomes in either group. No patient had clinical deterioration during the study period. Blinding. We assessed adequacy of blinding by asking the research assistant and the physician to guess to which treatment group the patient was assigned. There was no evidence that either the research assistant or the physicians were more successful at guessing which patients received mist than would be expected by chance. The levels of agreement (kappa) between the guess and the actual treatment were 0.44 for the research assistants and 0.48 for the physicians. DISCUSSION Treatment with mist does not improve the clinical outcome of children treated in the ED with moderate croup. The two groups had similar improvements in croup scores over time, with or without mist. There was no difference in oxygen saturation, respiratory rate, and heart rate in the mist group and the no-mist group. The global assessment of change and comfort scores were also similar in the two groups. The use of mist in the emergency treatment of croup has been a traditional practice of unproven benefit There have been various methods of delivery of humidified air throughout the years. In the beginning of the 19th century, hospitals used croup kettles to deliver steam. These were associated with scalding injuries. 12 Methods were then developed that used cool mist. These include croup tents, face masks, and mist sticks. Humidity has been presumed to improve airway inflammation and moisten secretions, thereby improving the child s respiratory status. It has been shown that when sputum is exposed to humidity and nebulized water, there is a significant decrease in viscosity. The thinner secretions may be easier to clear from the upper airway. It has been suggested that the narrowing of the subglottic space causes turbulent air flow, which leads to drying and further inflammation of the mucous membranes. The moist air may prevent the drying and thickening of the secretions, thereby improving the flow of air through the trachea In order for the aerosolized water droplets to reach the subglottic area, the particle size should be 5 10 m in diameter. Particles greater than 10 m will reach the oropharynx but not further into the respiratory tract. The water particles from steam in the shower or home humidifiers are in this range. 17 Although we did not measure the actual particle size generated by our mist stick apparatus, it has been described that nebulizers can generate particles < 5 m. 18 Despite the widespread use of mist for many years, only three studies have examined the effectiveness of moist air in the treatment of croup. Lenney and Milner 4 conducted a study where nebulized sterile water was given to five children with croup. Total respiratory resistance was measured before and after treatment. There was no improvement noted. In 1984, Bourchier and Fergusson 5 reported a controlled trial of humidification in viral croup. There were 16 patients, admitted with croup, who were randomly assigned to receive either humidified air in an enclosed bed covering ( croup tent ) or room air. During the 12-hour study period, the two groups had similar rates of recovery as measured by heart rate, respiratory rate, oxygen saturation, transcutaneous carbon dioxide level, and clinical rating. There was no clinical benefit of humidified air noted. Recently, a study has examined the effect of humidified air in croup using respiratory inductance plethysmography. These authors prospectively

6 878 Neto et al. MIST FOR MODERATE CROUP evaluated 46 patients with mild to moderate croup using humidified air versus no treatment. The rib cage/abdominal wall (RC/ABD) asynchrony, determined by respiratory inductance plethysmography, was used to measure chest wall retractions. The patients were treated only for 20 minutes. The authors noted an improvement in both croup scores and RC/ABD asynchrony in the treated group. At the same time interval there had been no significant change in the control group. 19 To the best of our knowledge, our study is the largest randomized, controlled trial to date on the use of humidified air in the treatment of croup. Although the total number of patients is relatively small, the study was designed to detect a difference in croup score of 1 between the two groups (with a statistical power of 80%). The Westley croup score has been shown to be valid, reliable, and responsive to change in clinical status. 7 A croup score change of 1 is the minimally important clinical difference. There were improvements of croup scores in both the mist and no-mist groups over time ( 2.9 and 2.6, respectively), but they were not statistically different. We believe this study is the first to compare mist therapy with no mist in a blinded fashion. The patient was placed in a treatment room away from the assessment area. The research assistant did not approach the patient s room as the administration of mist could be heard outside of the room. The patient was brought to the assessment area by an emergency nurse not involved in the study. In this separate area, both the research nurse and the physician assessed the patient. The adequacy of blinding was tested by comparing the levels of agreement (kappa) between the guessed treatment and the actual treatment. The kappa scores of 0.44 (research assistants) and 0.48 (physicians) indicate that our blinding appears to have been adequate. LIMITATIONS There are several limitations to this study. The sample size is small but, despite this, the study has the power to detect a change in croup score of 1, which was believed to be the minimally important clinical difference between the two groups. Of all of the eligible patients, only 53% were enrolled. This was due to the unavailability of the research assistant at those times. There were 26 patients who received racemic epinephrine or inhaled budesonide. These treatments were given prior to randomization and after the baseline measurement. These were distributed fairly evenly in the two groups (mist = 16, no mist = 10). Because these are delivered in humidified oxygen, they may have affected the study results. Although the separate analysis with these patients removed did not show a difference, this subgroup analysis has a small number of patients. However, if the results were affected by the administration of the inhaled medication, one would have expected a final result favoring mist because there were more patients in the mist group who received an inhalation treatment, and there was no difference in the baseline severity of symptoms between the groups. It is unclear how well the mist therapy was administered by the parents, as the research assistant was not observing the treatment. This, however, reflects the reality of using mist treatment in the ED where parents are initially advised on its use but not continuously observed throughout the treatment. Many physicians advise parents to initially treat the child at home with humidity. They are instructed to use a humidifier or to take the child into a steamy bathroom. In our study, 25 patients had been treated in this manner before coming to the ED. Theoretically, the water particle size generated by home humidifiers and by a steamy shower is too large to reach the subglottic area. 17 The results of our study may not be generalizable to this population as we studied ED patients who may be a subgroup of children who do not respond to the mist treatment at home. Humidified air may be effective in some children, and these are the ones who do not need to come to the ED. Our study excluded patients with severe croup. We thought it was not ethical to withhold humidified oxygen therapy from these patients if a possible treatment effect exists. It is not known whether mist therapy may have an adjunctive role in this group. We also excluded patients with mild croup and do not know whether a treatment effect exists in this population. CONCLUSIONS We found no difference in any of the outcome measures in children treated with or without mist for moderate croup. Based on the results of the trial, we conclude that mist therapy is not effective in improving clinical symptoms in the ED treatment of children with moderate croup. References* 1. Denny FW, Murphy TF, Clyde WA Jr, Collier AM, Henderson FW. Croup: an 11-year study in a pediatric practice. Pediatrics. 1983; 71: Henrickson KJ, Kuhn SM, Savatski LL. Epidemiology *A list of further suggested reading material is available from the authors.

7 ACAD EMERG MED September 2002, Vol. 9, No and cost of infection with human parainfluenza virus types 1 and 2 in children. Clin Infect Dis. 1994; 18: Marx A, Torok TJ, Holman RC, Clarke MJ, Anderson LJ. Pediatric hospitalizations for croup (laryngotracheitis): biennial increases associated with human parainfluenza virus epidemics. J Infect Dis. 1997; 176: Lenney W, Milner AD. Treatment of acute viral croup. Arch Dis Child. 1978; 53: Bourchier D, Fergusson DM. Humidification in viral croup: a controlled trial. Aust Paediatr J. 1984; 20: Westley CR, Cotton EK, Brooks JG. Nebulized racemic epinephrine by IPPB for the treatment of croup. Am J Dis Child. 1978; 132: Klassen TP, Feldman ME, Watters LK, Sutcliffe T, Rowe PC. Nebulized budesonide for children with mild to moderate croup. N Engl J Med. 1994; 331: Klassen TP, Craig WR, Moher D, et al. Nebulized budesonide and oral dexamethasone for treatment of croup: a randomized controlled trial. JAMA. 1998; 279: Henry R. Moist air in the treatment of laryngotracheitis. Arch Dis Child. 1983; 58: Couriel JM. Management of croup. Arch Dis Child. 1988; 63: Skolnik NS. Treatment of croup: a critical review. Am J Dis Child. 1989; 143: Greally P, Cheng K, Tanner MS, Field DJ. Children with croup presenting with scalds. BMJ. 1990; 301: Dulfano MJ, Adler K, Wooten K. Physical properties of sputum: effects of 100 percent humidity and water mist. Am Rev Respir Dis. 1973; 107: Sasaki CT, Suzuki M. The respiratory mechanism of aerosol inhalation in the treatment of partial airway obstruction. Pediatrics. 1977; 50: Ophir D, Elad Y. Effects of steam inhalation on nasal patency and nasal symptoms in patients with the common cold. Am J Otolaryngol. 1987; 8: Szilagyi PG. Humidifiers and other symptomatic therapy for children with respiratory tract infections. Pediatr Infect Dis J. 1991; 10: O Callaghan C, Barry PW. The science of nebulized drug delivery. Thorax. 1997; 52(suppl 2):S31 S Ho SL, Coates AL. The effect of dead volume on the efficiency and the cost to deliver medications in cystic fibrosis with four disposable nebulizers. Can Respir J. 1999; 6: Jamshidi PB, Kemp JS, Peter JR, et al. The effect of humidified air in mild to moderate croup: evaluation using croup scores and respiratory inductance plethysmography [abstract]. Acad Emerg Med. 2001; 8:417.