Early Parenteral Corticosteroid Administration in Acute Asthma ROBERT Y. LIN, MD, GENE R. PESOLA, MD, RICHARD E. WESTFAL, MD, LEONARD BAKALCHUK MD, CHRISTOPHER W. FREYBERG, MD, DAVID CATAQUET, MD, GREGORY T. HEYL, MD To test the hypothesis that early parenteral corticosteroid administration may be associated with a rapid improvement in airflow obstruction in adult asthmatic patients, a randomized, double-blind placebo-controlled study was carried out. Forty-five adult asthmatic patients, with initial peak expiratory flow rates (PEFRs) of <200 L/sec received an intravenous bolus of either 125 mg methylprednisolone (MP) or normal saline before any other emergency department treatments. This was immediately followed by 3 aerosol treatments of 2.5 mg of albuterol separated by 20-minute intervals. PEFRs and heart rates were measured over a 1-hour time frame. There was not a significantly higher rate of increase of PEFR in the MP group compared with the saline group. Similarly, the rate of increase in percent PEFR showed a trend to being higher in the saline group (P = .061). There was no significant difference in the proportion of hospitalizations and side effects between the two groups. Adjustment for other variables did not result in a model showing an enhanced PEFR improvement with MP treatment. This study does not support the concept that corticosteroid treatment effects are beneficial within the first hour after administration, Further studies of rapid-acting modalities to enhance bronchodilation are needed in treating acute asthmatics. (Am J Emerg Med 1997;15:621-625. Copyright © 1997 by W.B. Saunders Company) Despite the more widespread use of anti-inflammatory inhaled medication, exacerbations of asthma still occur regularly, and frequently progress to the point that emergency treatment is required. Although increased emergency department (ED) visits may be related in part to access to medical care, clearly asthma severity and the rapidity of deterioration also are important. The ED treatment of asthma has relied primarily on adrenergic agonist administration for many years. Presently, beta-2 specific agonists are used most frequently because of their rapid onset, ease of administration, and relatively low side-effect profile. There has been considerable interest in additional agents that can enhance From the Departments of Medicine and Emergency Medicine, St Vincent's Hospital and Medical Center of New York, New York Medical College, New York, NY. Received April 17, 1996, returned July 22, 1996; revision received September 10, 1996, accepted September 30, 1996. Address reprint requests to Dr Lin, Department of Medicine, St Vincent's Hospital, 153 W 11th St, New York, NY 10011. Key Words: Asthma, corticosteroids, adult, albuterol, peak flow, heart rate, randomized. Copyright © 1997 by W.B. Saunders Company 0735-6757/97/1507-0001 $5.00/0
the bronchodilatory response in severe asthma treated with beta agonists, and studies have examined the potentially additive effects of magnesium sulfate, 1,2 aminophylline, 3-7 and ipratropium bromide 8-I° in terms of ED and hospitalization courses. We have previously studied the use of higher doses of albuterol via continuous nebulization in acute asthmatics, and found no advantage for the continuous mode of nebulization over a 2-hour treatment time frame, u However, in examining this group further, 12 we found that a history of inhaled corticosteroids and acute parenteral corticosteroids was associated with different response patterns of forced expiratory volume/1 second (FEV1). There was higher response slope between the first and second hour of treatment in patients who received intravenous corticosteroids at the end of the first hour of treatment. This suggested to us that intravenous corticosteroids may have a beneficial effect in a shorter time frame than had been expected previously. A recent ED study 13 of asthmatic children also suggested that a rapid relative improvement resulted from dexamethasone administration, which allowed significantly more children to be discharged at 2 hours after initiation of treatment. These findings suggested to us that glucocorticoids may have a more rapid effect than previously postulated. If corticosteroid administration were shown to be rapidly effective, it would have implications in the emergent treatment sequence used for acute asthma. To test the hypothesis that early corticosteroids can induce a rapid beneficial response, we designed a controlled trial using intravenous methylprednisolone versus saline in adult asthmatic patients seen in the ED. MATERIALS AND METHODS This study was approved by the institutional review board for human research. Entry criteria consisted of a prior diagnosis of asthma by a physician with a history of improvement with bronchodilator therapy. This criteria took into consideration the 1982 American Thoracic Society definition of asthma 14as well as a desire to approximate the typical patient who presents with asthma to the emergency department. Additional inclusion criteria consisted of age greater than or equal to 18 years, and a peak expiratory flow meter measurement of less than 200 L/sec. Pregnant patients and patients with active peptic ulcer disease were excluded. Subjects who reported smoking more than 20 pack years of cigarettes or who had received nebulized beta agonists from the 621
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 15, Number 7 • November 1997
emergency medical services or other emergency staff before study physician evaluation were also not recruited. The design consisted of a double-blind placebo-controlled parallel track study. Five of the investigators (GR LB, CF, DC, GH) administered the study protocol to the patients over a 10-month period. After verifying entry criteria and obtaining written informed consent, baseline measurements of heart rate and peak expiratory flow rates were obtained. All peak flow rates consisted of the best of three efforts observed by the study physician. A Wright peak flow meter (London N183JD, Ferraris Medical Ltd) was used for peak flow measurements. A nurse then drew into a syringe either 125 mg of freshly dissolved methylprednisolone sodium succinate (Solumedrol; Upjohn, Kalamazoo, MI) or an equal volume of normal saline based on a computer-generated random number assignment list. This syringe was then given to the study physician, unaware of its contents, who then administered the contents by bolus injection to the subject. Immediately thereafter, a 2.5-rag aibuterol nebulization was administered to the subject. This was repeated for two additional doses at 20-minute intervals. Heart rates and repeat peak flow measurements were obtained before each additional albuterol treatment and at the end of 1 hour of total protocol treatment. The presence or absence of agitation, tremors, and accessory respiratory muscle retractions were noted at the same time points in which heart rate and peak flow measurements were obtained. Past medical histories were obtained from the patients, including past and current asthma treatment and complications. Admitting criteria agreed on by the study physicians included the presence, after treatment, of any of the following: (1) accessory muscle use, (2) respiratory rate of >24 breaths/rain, (3) arterial blood CO, greater than 44, (4) arterial blood O2 less than 70, (5) associated diseases such as pneumonia or febrile illness greater than 102°F, and (6) failure to show improvement after 5 to 6 hours of observation with associated fatigue and shortness of breath with exertion. Demographic information and the final disposition of the patient were also noted (admission, discharge, or leaving against medical advice). A target sample size of 24 for each group was chosen based on power calculations that used airflow indices taken from a prior study, u An 80% power of finding a 100 L/sec greater improvement in final peak flow rates in the steroid group compared with placebo group was calculated. Statistical methods used Fisher's exact test for categorical data, student's t-tests or analysis of variance for simple continuous variables, and repeated measures analysis of variance for the primary outcome variables of heart rate and peak flow over time. The Greenhouse-Geisser adjustment for nonsphericity was employed. For duration of acute asthma symptoms, the Wilcoxon rank sum test for nonparametric distributions was used to compare the two primary treatment groups. The SAS software program (SAS for Windows 6.1; SAS Institute, Cary, NC) was used for all analyses. A significant difference was stated if the P value was less than or equal to .05. Percent predicted peak flow rates were also calculated using the Quanjer prediction equations 15 for smokers/ nonsmokers/exsmokers, which is based solely on age (years), sex, and height (meters). For men the equation for predicted peak flow (L/sec) was 5.48 × meters - .041 × years + 1.58. For women predicted peak flow was 3.72 × meters - .03 × years + 2.24. The percent predicted value was also examined as a primary outcome variable in a similar manner as peak flow and heart rate. Covariate effects of baseline heart rates (for peak flow outcomes only), duration of acute symptoms, and prior medication were also examined using either analysis of covariance or multiway analysis of variance for repeated measures (respectively). Means are shown with standard deviations except in the figures where standard errors are shown for ease of viewing.
RESULTS Forty-eight protocols were administered. Inadvertently, three subjects were studied twice. Only the first study for each of these repeaters was used for analysis to avoid a learning effect. This resulted in 23 methylprednisolone subjects and 22 saline patients being used for the final analyses. Similar baseline characteristics were observed for both groups of patients (Table 1), except for a tendency for the methylprednisolone group to have a shorter onset of asthma exacerbation (.01 < P < .05) and a somewhat higher baseline heart rate (.1 < P < .15. Accessory muscle use (ie, retractions) was noted in seven methylprednisolone patients and five saline patients (not significantly different). Both treatment groups had a significant increase in peak flow rates with time (Figure 1A). However, there was not a significant difference between the rate of improvement of peak flow rates between the methylprednisolone group and the saline group (P = .072, Figure 1A). Using calculated percent predicted peak flow, a similar improvement slope with greater bronchodilation for the saline group was observed, at a statistical significance level of P = .061) (Figure 1B). There was an overall decrease in heart rates over time (P = .009). The proportion o f admitted patients in the two groups also did not differ significantly (Table 1). Baseline heart rates were examined for a covariate effect on sequential peak flow responses. Although there was a significant covariate effect observed for baseline heart rates (.01 < P < .05), there was no difference between the two treatment groups with regards to this effect. The use of TABLE 1, Characteristics of Study Patients
Number studied Age (years) Females/males Ethnicity (AA/H/W)* Initial peak flow (L/sec) Initial percent predicted peak flow Initial heart rate inhaled steroid use Oral steroid use Theophylline use Admitted to hospital Discharged against medical advice Required steroid during study Comorbidities Hospitalizations in past year Emergency visits in the past month Acute symptom duration (days) Asthma duration (years) Cigarette smoking history
23 41.4 +_ 13.3 12/11 7/9/7 138 + 40
22 39.0 + 13.4 11/11 5/8/9 134 _+ 32
NS NS NS NS
29.6 + 9.6 104 ± 20 8 1 10 7
27.9 -+ 8.6 95 + 18 7 2 9 4
NS NS NS NS NS NS
0.48 _+ 0.73
0.45 -+ 0.8
0.48 + 0.95
0.64 -+ 1.00
2.0 _ 1,3
5.9 --- 7.1
25,9 - 12.7
21.5 +- 14.4
*AA, African-American; W, Caucasian; H, Hispanic.
LIN ET AL • CORTICOSTEROIDS IN ACUTE ASTHMA
the patients who had received saline. This difference in proportions was significant (X2, P = .038).
200 / / /
150 vo --o--
.... o ....... SALINE
M < UJ 50 I:L £3 LU I'-O 40 uA
,-""?T 'f'I"t" ...,.~f,,,.O!
Z/I / t"
z 30 u3 (D rr LU 13.. 20
P:2 .L ' 0
FIGURE 1. (A) Peak flow rates at different time points in the saline and methylprednisolone groups depicted as means and standard errors. (t3) Percent predicted peak flow rates at different time points in the saline and methylprednisolone groups depicted as means and standard errors.
inhaled and/or oral steroids did not have a significant independent effect with regard to the peak flow or heart rates. The proportion of patients with agitation and tremor were similar in the two groups at baseline and at later time points. Baseline tremor was observed in two methylprednisolone patients and one saline patient, whereas late tremor was observed in seven and nine patients in the respective groups. Baseline agitation was observed in three methylprednisolone patients and one saline patient, whereas late tremor was observed in one and two patients in the respective groups. Two patients in each group showed late accessory muscle of respiration use. No discernible side effects were reported with either saline or methylprednisolone treatment. Parenteral corticosteroids were administered during the time course of the study (-<60 minutes) to eight of the patients who had received initial methylprednisolone and to two of
This study shows no early benefit in using parenteral corticosteroids as the initial treatment in adult asthmatic patients who present to the ED with acute severe bronchospasm. Although a previous study 12 by our group suggested an early benefit (after 50 minutes) from using methylprednisolone, that study allowed discretionary corticosteroid use only after initial albuterol treatment. The present study intentionally chose patients with significant initial airflow impairment (mean of 24% percent predicted peak flow), but because no benefit was observed with methylprednisolone, it is possible that initial airflow impairment alone does not select for a rapidly steroid-responsive group of patients treated with up-front corticosteroids. An alternative explanation is that there is no early effect of corticosteroid administration in asthma (ie, during the first hour of treatment). A study of asthmatic children by Wolfson et all6 also showed no benefit of early methylprednisolone treatment over placebo in the ED (within 45 minutes of arrival). The mean observation time frame of that study was 2.9 hours. A randomized but not placebo-controlled study by Mort'ell et a l : examined showed no increase in the improvement of FEV1 or peak flow within 48 hours in adult asthmatics given methylprednisolone compared with those given no corticosteroids. Scarfone et a113 compared inhaled dexamethasone with oral prednisone in emergency treatments of asthmatic children, and observed that the former group had significantly more discharges from the ED at 2 hours time. Our study did not examine 2-hour responses, and thus could not exclude a later corticosteroid enhanced response. As reviewed by Engel and Heinig, is most but not all placebocontrolled studies of corticosteroids in acute asthma have shown an overall benefit for corticosteroids with respect to pulmonary function, blood gases, or hospital admission. Taken together these data suggest that there is variability in early responses to corticosteroid treatment, with possible selection, patient population, and study design differences as factors. Early treatment responses have physiological plausibility in that some corticosteroid effects to not require protein synthesis and thus may be relatively rapid, such as certain membrane effects and interactions with adrenergic function. 19,20 The possibility of differences in corticosteroid responsiveness has been suggested by a study by Sher et al, 2~ which showed that glucoeorticoid receptor number and affinity can be abnormal in corticosteroid-resistant asthmatic patients. If even a minority of asthmatics could be shown to have rapid corticosteroid responsiveness, the possibility that a few patients could benefit from early corticosteroids may justify further exploration of this approach. Identifying corticosteroid-responsive patients for future clinical studies would, however, be a challenge because clinical parameters in this study, such as baseline heart rate, initial peak flow rates, and the presence of accessory respiratory muscle use, did not allow for the formation of a statistical model showing a corticosteroid effect. Studies that relate patients' corticosteroid use and clinical responses to cellular glucocorticoste-
AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 15, Number 7 • November 1997
roid receptor status may be required to explore this possibility further. The finding that peak flow rate improvements tended to be less in the methylprednisolone group suggests a detrimental effect from this treatment. The observed increased discretionary corticosteroid administration used in this group is another parameter pointing toward a poorer short-term outcome related to this treatment. The possible reasons for these observations are not clear. A recent report by Enomoto et a122identified a small number of aspirin-sensitive asthmatics who also had apparent reactions to parenterally administered corticosteroids (6 out of 850 asthmatic patients studied). Because our questionnaire did not query histories of aspirin hypersensitivity, it is conceivable that the parenteral corticosteroid administration may have compounded the asthma severity acutely in predisposed patients. This intriguing possibility cannot be excluded as a factor in our study, but based on the relatively low reported incidence of this phenomenon 22,23 and the randomized allocation used, it seems unlikely as an explanation for the overall findings of our study. Another possibility is that the preservative (benzyl alcohol) used in the methylprednisolone preparation may have had an unpredictable effect that impaired patients' bronchodilator responses. We are not aware of reports that relate bronchospasm to this compound. Finally, the mode in which the methylprednisolone preparation was given (bolus injection) may have somehow resulted in proinflammatory mediator release, thus opposing the bronchodilator therapy administered to these patients. Severe asthma continues to regularly result in hospitalization and intubation despite present treatment modalities. In another study24 of intubation for adult asthma at this institution, we observed no decrease in either the number of patients intubated per year or in the number of asthma admissions per year since 1988 despite an overall trend toward decreasing hospital stay for asthma. This implies that despite the national2s and international 26 promotion of treatment guidelines for asthma stressing more use of anti-inflammatory medication, severe acute asthma will continue to be a regular and common ED condition. Indeed, an economic evaluation by Weiss et al27 suggests that ED visits along with hospitalization and death constituted 43% of the national economic impact of asthma. To address this medical usage pattern and expense, some have suggested the need for structured teaching and/or treatment programs 2831 both on an inpatient and outpatient basis for reducing asthma-related hospital encounters. It is likely that many patients with severe acute asthma have crossed the threshold for effective rapid bronchodilation by the time they have arrived at the ED. Earlier intervention must be made available to asthmatic patients undergoing an exacerbation. Moreover, asthmatic patients need to understand when and how to seek more urgent medical treatment. 3a In summary, we conclude that severe acute asthma does not preferentially respond to intravenous methylprednisolone treatment within the first hour of presentation to the ED. This result is consistent with previous concepts about the time course of steroid therapy in acute asthma. More effective emergency treatment regimens need to be explored for severe acute asthma.
We thank the St. Vincent's Hospital Emergency Department nursing staff for their help and support during this study.
REFERENCES 1. Tiffany BR, Berk WA, Todd IK, et al: Magnesium bolus or infusion fails to improve expiratory flow in acute asthma exacerbations. Chest 1993; 103:831-834 2. Green SM, Rothrock SG: Intravenous magnesium for acute asthma: failure to decrease emergency treatment duration or need for hospitalization. Ann Emerg Med 1992;21:260-265 3. Wrenn K, Slovis CM, Murphy F, et al: Aminophylline therapy for acute bronchospastic disease in the emergency room. Ann Intern Med 1991; 115:241-247 4. DiGiulio GA, Kercsmar CM, Krug SE, et al: Hospital treatment of asthma: lack of benefit from theophylline given in addition to nebulized albuterol and intravenously administered corticosteroid. J Pediatr 1993;122:464-469 5. Needleman JP, Kaifer MC, Nold JT, et ai: Theophylline does not shorten hospital stay for children admitted for asthma. Arch Pediatr Adolescent Med 1995;149:206-209 6. Strauss RE, Wertheim DL, Bonagura VR, et al: Aminophylline therapy does not improve outcome and increases adverse effects in children hospitalized with acute asthmatic exacerbations. Pediatrics 1994;93:205-210 7. Huang D, O'Brien RG, Harman E, et al: Does aminophylline benefit adults admitted to the hospital for an acute exacerbation of asthma? Ann Intern Med 1993; 119:1155-1160 8. Schuh S, Johnson DW, Callahan S, et al: Efficacy of frequent nebulized ipratropium bromide added to frequent high-dose albuterol therapy in severe childhood asthma. J Pediatr 1995;126:639-645 9. O'Driscoll BR, Taylor RJ, Horsley MG: Nebulized salbutamol with and without ipratropium bromide in acute airflow obstruction. Lancet 1989;24:1418-1420 10. Rebuck AS, Chapmen KR, Abboud R, et al: Nebulized anticholinergic and sympathomimetic treatment of asthma and chronic obstructive airways disease in the emergency room. Am J Med 1987;82:59-64 11. Lin RY, Sauter D, Newman T, et al Intermittent versus continuous albuterol administration in acute asthma. Ann Emerg Med 1993;22:1847-1853 12. Lin RY, Sauter D, Newman T, et al: An association between the reported use of inhaled triamcinolone and differential short term responses to aerosolized albuterol in emergency department asthmatics. Chest 1994;106:452-457 13. Scarfone RJ, Loiselle JM, Wiley JF, et al: Nebulized dexamethasone versus oral prednisone in the emergency treatment of asthmatic children. Ann Emerg Med 1995;26:480-486 14. Specter SL: Definition of asthma. ATS News 1982;8:5 15. Quanjer PH, van der Maas LJ, Van Hartevelt JH, et al: Epidemiological follow-up investigations of lung function. SEPCR working group meeting Goethenburg Sweden, 11th/12th June, 1976 (proceedings). Scand J Resp Dis 1976;57:309-310 16. Wolfson DH, Nypaver MM, Blaser M, et al: A controlled trial of methyprednisolone in the early emergency department treatment of acute asthma in children. Pediat Emerg Care 1994; 10:335-338 17. Morell F, Orriols R, de Gracia J, et al: Controlled trial of intravenous corticosteroids in severe acute asthma. Thorax 1992;47: 588-591 18. Engel T, Heinig JH: Glucocorticoid therapy in acute severe asthma--A critical review. Eur Respir J 1991 ;4:881-889 19, Duval D, Durant S, Homo-Delatche F: Interactions of steroid molecultes with membrane structures and functions. Biochim Biophys Acta 1983;737:409-442 20. Miller-Larrson A, Brattsand R: Topical anti-inflammatory activity of the glucocorticoid budesonide on airway mucosa: Evidence for a "hit and run" type of activity. Agents Actions 1990;29:127-129 21. Sher ER, Leung DYM, Surs W, et al: Steroid resistant asthma: Cellular mechanisms contributing to inadequate response to glucocorticoid therapy. J Clin Invest 1994;93:33-39 22. Enomoto T, Okada T, Ichihashi K, et al: Examination on aspirin-induced asthma and hypersensitivity to steroids--A questionnaire to 850 asthma on hypersensitivity to non-steroidal anti-
LIN ET AL • CORTICOSTEROIDS IN ACUTE ASTHMA
inflammatory drugs and hypersensitivity to steroids. Arerugi 1995;44: 534-539 23. Feigenbaum BA, Stevenson DD, Simon RA: Hydrocortisone sodium succinae does not cross react with aspirin in aspirin-sensitive patients with asthma. J Allergy Clin Immuno11995;96:545-548 24. Lin RY, Rehman A, Ahmed S: Clinical characteristics of adult asthmatics requiring intubation. J Medicine 1995;26:261-277 25. National Heart, Lung, and Blood Institute--National Asthma Education Program Expert Panel: Guidelines for the diagnosis and management of asthma. J Allergy Clin Immunol 1991 ;68:425-534 26. Sheffer AL, Bartal M, Bousquet J, et al Global Initiative for Asthma: Global Strategy for Asthma Management and Prevention; NHLBI/WHO Workshop Report. Bethesda, MD, National Institutes of Health, 1995 (Publication No. NIH 95-3659) 27. Weiss KB, Gergen PJ, Hodgson TA: An economic evaluation of asthma in the United States. N Engl J Med 1992;326:862-866
28. Trautner C, Richter B, Berger M: Cost-effectiveness of a structured treatment and teaching programme on asthma. Eur Respir J 1993;6:1485-1491 29. Yoon R, McKenzie DK, Bauman A, et al: Controlled trial evaluation of an asthma education programme for adults. Thorax 1993;48:1110-1116 30. Kuo E, Kesten S: A retrospective comparative study of inhospital management of acute severe asthma: 1984 vs 1989. Chest 1993;103:1655-1661 31. Muhlhauser I, Richter B, Kraut D, et al: Evaluation of a structured treatment and teaching programme on asthma, J Intern Med 1991;230:157-164 32. Becker G, Janson-Bjerklie S, Benner P, et al: The dilemma of seeking urgent care: Asthma episodes and emergency service use. Soc Sci Med 1993;37:305-313