|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
*
From the Pulmonary Division (Drs. Noseda and De Bruyne), Department of Internal Medicine, Centre Hospitalier Universitaire Brugmann, Brussels, Belgium; the Medical Statistical Unit (Dr. De Maertelaer), Institut de Recheiche Interdisciplinaire en Biologie Humaine et Nudéaire, Brussels, Belgium; and the Chest Department (Dr. Yernault), H
pital Erasme, Free University of Brussels, Brussels, Belgium.
Correspondence to: André Noseda, MD, PhD, Department of Medicine, Pulmonary Division, CHU Brugmann, Place A. Van Gehuchten 4, B-1020 Brussels, Belgium.
| Abstract |
|---|
|
|
|---|
Design: Randomized, placebo-controlled, double-blind crossover trial.
Setting: Medium-sized university general hospital.
Patients: Twenty-five asthma patients attending the chest
clinic with spontaneous complaints of increases in dyspnea and with a
Borg scale dyspnea rating
1 at rest.
Interventions: At 0 min, IV methylprednisolone (125 mg) vs saline solution; at 60 min, 5 x 500 µg terbutaline inhaled from an inhaler device.
Measurements and results: Change in dyspnea was assessed with bipolar visual analog scale (VAS) (much more short of breath, -100%; much less short of breath, + 100%), FEV1, and visual memory (using the Benton visual retention test). Eighteen subjects (mean age, 61 years) completed the study. At 5 min and 60 min, shortness of breath improved with no statistically significant difference between saline solution and methylprednisolone. The mean (SD) VAS rating at 60 min was 29% (39%) on the day that saline solution was administered and 36% (25%) on the day the steroid was administered. FEV1 and Benton score did not significantly change from baseline on either study day. Shortness of breath and FEV1 improved following terbutaline administration, with no significant difference between the days on which saline solution and the steroid were administered. In the seven subjects who were randomized to receive methylprednisolone on the first day, baseline dyspnea rated on the Borg scale was significantly lower on the second day (first day: median, 3; range, 3 to 4; second day: median, 2; range, 0.5 to 3; p = 0.040).
Conclusions: We conclude that in patients with an exacerbation of asthma, an IV bolus of methylprednisolone does not reduce dyspnea more than saline solution after 5 min and 60 min.
Key Words: asthma exacerbation dyspnea methylprednisolone visual analog scale
| Introduction |
|---|
|
|
|---|
The present study was designed to evaluate the early effects on dyspnea of an IV bolus of 125 mg methylprednisolone vs saline solution in a group of outpatients who were symptomatic because of an asthma exacerbation. The effect on dyspnea was assessed using a previously validated3 bipolar visual analog scale (VAS).
| Materials and Methods |
|---|
|
|
|---|
1 (ie, subject at
least "very slightly short of breath"). Symptoms such as chest pain
or purulent sputum were not present, and physical examination did not
disclose any sign suggestive of an etiology other than asthma
exacerbation. In all subjects, spirometry findings showed a sharp
decrease in FEV1 compared to the values in the
chart (Table 1
). Exclusion criteria were diabetes mellitus, a history of psychiatric
disorder, maintenance therapy with oral steroids, pregnancy, and severe
impairment of cognitive function (defined as a mini-mental
state5
of < 24/30).
|
Methylprednisolone
The solutions used in the present study were made up and coded
independently in the hospital pharmacy, using a randomization table.
The galenical form we used was a ready-to-use limpid solution of 125 mg
methylprednisolone that was indistinguishable from saline solution
(Promedrol; Pharmacia & Upjohn; Piscataway, NJ). The code
remained located in the hospital pharmacy during the whole study.
Dyspnea Rating
To refer to dyspnea, we always used the item "shortness of
breath" when talking with the patients since it is the most
frequently chosen and the most reproducible descriptor in patients with
asthma.6
The subjects were given neutral information about
the IV agent. They were told they would receive on each of the two
study days an IV agent that could modify the sensation of shortness of
breath. Questions about which IV agent would be used were not answered.
At the time of terbutaline inhalation, the subjects were told they
would inhale a bronchodilator agent that was expected to improve lung
function.
We rated baseline dyspnea at the time of inclusion, as well as before starting with the study protocol on each study day, using a modified Borg scale. The scale is a 20-cm vertical line labeled with numbers from 0 to 10 with descriptors ranging from "not short of breath at all" to "extremely short of breath." The Borg scale dyspnea rating (DR) aimed to rate the intensity of shortness of breath at the very moment the scale was shown to the subject.7 On each study day, we used a bipolar VAS to rate the change in dyspnea from baseline. The scale is a 40-cm horizontal line that has the midpoint that was labeled "no change," and the left end labeled "very much shorter of breath," and the right end labeled "very much less short of breath." The VAS DR, expressed as a percentage (range, -100% to + 100%), aimed to assess the change in shortness of breath from baseline at the very moment the scale was shown to the subject. Throughout the study, we carefully recommended to the subjects that they rate only shortness of breath and ignore other sensations such as cough, nasal irritation, or throat irritation.
Lung Function
A portable spirograph (Microspiro; Chest; Tokyo, Japan)
was used. In all subjects, two to three technically acceptable
maneuvers were obtained for the baseline evaluation as well as at each
step following the IV injection or the terbutaline inhalation. The
variable retained for analysis was the FEV1, with
the highest FEV1 value obtained being the one
reported. The FVC and the maximal expiratory flow at 50% of FVC
(MEF50) (measured, as recommended4
at 50% of the baseline FVC value) also were analyzed; however, as FVC
and MEF50 data yielded no additional information
compared to that obtained from FEV1 data, only
the latter are reported.
Visual Memory
Visual memory was assessed using the Benton visual retention
test.8
This test is based on a design reproduction, using
10 cards each showing one to three geometric figures. A 10-s exposure
of the card to the subject is followed by the removal of the card and
the drawing of its contents by the subject. Subjects scores are
derived from a standardized scoring procedure supplied with the test
(score range, 0 to 10). In subjects aged 55 to 65 years, the expected
scores range from 4 (poor visual memory level) to 8 (high level). The
test includes two parallel series of 10 cards; the first series was
used for a baseline assessment, at time zero, and the second was used
for a repeated assessment, at time 60 min.
Statistical Analysis
Samples statistics were reported as mean (SD) or median (range).
Confidence intervals (CIs) were calculated when appropriate. Values
from paired samples were compared using Students t test or
the Wilcoxon test when the variable was not normally distributed (Borg
scale DR). Correlations were evaluated using Pearson linear regression
or Spearman rank correlation when the variable was not normal. An
analysis of variance with time and drug as the repeated measures (time
at five levels and the drug at two levels) was performed to assess the
time effect and the effect of methylprednisolone or saline solution on
VAS DR and FEV1. Planned comparisons were
performed subsequently for comparing specific times.
The study was designed to give a 90% chance of detecting a 25% change in dyspnea with a 5% probability of a type I error. Taking into account a 25% between-subject coefficient of variation for the change in dyspnea recorded on the bipolar VAS following saline solution inhalations in a population with asthma,2 the sample size estimation indicated that at least 13 subjects were required.
| Results |
|---|
|
|
|---|
Baseline Values: Placebo Day vs Steroid Day
All the subjects, except one, had a baseline Borg scale DR
that was higher on at least one of the study days than at the time of
recruitment. The median baseline Borg scale DR was 3 (range, 0.5 to 7)
on the saline solution day, and 3.5 (range, 1.5 to 7) on the steroid
day (difference not statistically significant). The relationship (Fig 1
) of the baseline Borg scale DR did not reach significance with either
baseline FEV1 (r = -0.27 [ saline
solution day]; r = -0.24 [ steroid day]) or with the
drop in FEV1 (for best value minus baseline
value, see Table 1
) (r = 0.31 on both study days).
|
|
Effects of Methylprednisolone vs Saline Solution
The effect of the IV injection of saline solution or
methylprednisolone on dyspnea, assessed using the VAS DR, is shown in
Table 2
and Figure 3
. Five and 60 min after IV injection, an improvement in shortness of
breath occurred after either methylprednisolone or saline solution
injection, with no statistically significant difference between
them. Inhaled terbutaline produced a further decrease in dyspnea on
both study days. Comparing the mean VAS DR at times 65 and 90 with the
mean VAS DR at times 5 and 60, respectively, led to a mean difference
of 19% (significantly different from zero, p = 0.001; 95% CI, 10 to
27%) on the saline solution day and 15% (significantly different from
zero, p = 0.048; 95% CI, 0 to 30%) on the steroid day.
|
|
On the saline solution day, the score for visual memory on the Benton test was 4.4 (2.0) at baseline and 4.6 (2.2) at time 60. The corresponding scores on the steroid day were 5.2 (2.2) and 4.8 (1.8). No difference was statistically significant. The correlations between VAS DR at time 60 and the parallel changes in the Benton score for visual memory were not statistically significant (saline solution day, r = -0.38; steroid day, r = 0.07).
Patients Comments
One subject complained about a painful IV injection (steroid day),
and another developed skin erythema following IV injection (saline
solution day). Complaints following inhaled terbutaline administration
included tremor (n = 3), palpitations (n = 2), and nervousness
(n = 2) on the two study days. One subject mentioned headache in the
hours following the study (steroid day), and another mentioned insomnia
on the night following the study (steroid day). No serious effect was
observed, and the study code was not broken.
| Discussion |
|---|
|
|
|---|
It may be questioned whether our subjects were really asthmatic. First, in 4 of 18 subjects, the best FEV1 from the chart was < 50% of predicted, a pattern that is consistent with chronic severe obstruction. Second, the amount of terbutaline-induced bronchodilation was very modest in this study, although the 2,500-µg dose of terbutaline inhaled from the inhaler device was reported to be effective in acute asthma exacerbation.9 Such a small bronchodilation may look surprising in patients with asthma. However, the magnitude of the acute response to a given dose of a bronchodilator is known to vary over time in the same individuals.10 On the basis of the large variability in FEV1 over time, which was a selection criterion for this study, we consider that our subjects really had asthma. The patients we studied presented at the chest clinic with moderate exacerbation that was judged not to need an immediate increase in treatment; however, they were undoubtedly in a phase of aggravation of bronchial obstruction, with a FEV1 on the study days lower than their best FEV1 by 30% of predicted on average. In this setting, we think that the small response to terbutaline that we observed during the study reflected the severity of the subjects conditions at the time that they were studied. In particular, marked airway inflammation is known to cause a reversible decrease in bronchodilator responsiveness at the time of exacerbation.1
The relevance of this study depends on whether the chosen variables were appropriate. We used a Borg scale to rate baseline dyspnea and a bipolar VAS to rate the changes in dyspnea induced by the IV injection and/or by inhaled terbutaline. Indeed, the Borg scale is not well-suited to assess an acute change, as there is little room on the scale to rate any improvement in subjects with baseline dyspnea of low intensity.2 7 However, the bipolar VAS is specifically devised for interval estimation11 and has been validated in our laboratory to assess changes in dyspnea following acute interventions.2 To assess lung function, we used spirometry and focused on the FEV1. Several groups have shown that airway resistance,12 13 respiratory motor drive,14 and hyperinflation15 play a role in the sensation of dyspnea experienced by asthmatics. However, the same12 13 and other investigators16 have found that there is a close relationship among patients with asthma between the changes in dyspnea and the parallel changes in FEV1. In this study, FEV1 did not change from baseline within 60 min on either study day. The absence of effect of IV methylprednisolone on lung function after 60 min was not unexpected. In a group of eight patients with asthma, Klaustermeyer and Hale17 reported a mean increase in MEF50 of 0.29 L/s 2 h after methylprednisolone, 125 mg IV, administration. In a similar group of eight patients with asthma, Ellul-Micallef18 reported a mean increase in peak expiratory flow of 23% of baseline 1 h after hydrocortisone, 200 mg IV, administration. However, these studies are of limited relevance in view of the small numbers of subjects and the modest, although statistically significant, changes reported. Most studies involving large numbers of subjects concluded that 6 to 12 h are required before steroids exert their effect on lung function.1 The only effect of IV steroids that has been shown to occur within 60 min is restoration of ß2 mimetic responsiveness.19 In the present study, however, the amount of bronchodilation induced by terbutaline 60 min after the IV injection was very similar on both study days.
Some other features of the design of our study need a critical evaluation. Because a double-blind procedure could not be guaranteed with the commercially available galenical form (lyophilized powder), a methylprednisolone solution that was characterized by its limpidity was used. The dose of methylprednisolone was fixed at a sufficiently high level (125 mg) to avoid the possibility that a negative result could be ascribed to the use of too low a dose. Indeed, although the optimal dosage of glucocorticoid treatment in asthma exacerbation is still disputed,20 21 the 125-mg dose lies within current dose recommendations and practices.21 Finally, a crossover design may not be optimal for a comparison between a steroid agent and placebo. To account for any carryover effect of methylprednisolone in the seven subjects randomized to receive the active agent on the first study day, we analyzed this subgroup separately and found that baseline dyspnea significantly decreased from the first to the second day. This decrease may be due to several factors. In asthmatic children, steroid therapy has been shown to affect cognitive performance, including visual memory assessed with the Benton visual retention test, as soon as 6 h after oral ingestion.22 In our study, however, reduced dyspnea, as rated on the Borg scale, is very unlikely to result from a nonspecific effect on cognitive function, since changes in Borg scale DRs and changes in Benton scores were unrelated. The parallel change in FEV1 (ie, from the first to the second day) was very small (mean, 0.08 L [3% of predicted]); as the changes in Borg scale DR were unrelated with those in FEV1, the reduction in dyspnea cannot be ascribed to an improvement in FEV1. The most attractive explanation is that shortness of breath improved as a consequence of steroid-reduced inflammation within the airways. There is some experimental support to the concept that dyspnea and inflammation may be linked in patients with asthma. It has been shown, for example, that in asthma patients with normal baseline FEV1, dyspnea is higher when a given degree of bronchoconstriction is induced by some agent (eg, by hypertonic saline solution), which is known to induce inflammatory mechanisms, rather than by methacholine.23
In conclusion, this placebo-controlled study shows that in patients with asthma who attended the chest clinic and who had increased dyspnea and decreased FEV1, IV methylprednisolone, at a dose of 125 mg, has no more effect on dyspnea than saline solution, within a time course of 60 min. Doctors should be aware that an IV bolus has a placebo effect on dyspnea in patients experiencing an exacerbation of asthma, but they should no longer think that the administration of an IV corticosteroid has any early specific effect on shortness of breath in such patients.
| Appendix 1 |
|---|
|
|
|---|
Of 193 doctors, 147 chose the expected answer, while 14 chose IV theophylline, 6 chose nebulized sympathomimetic, 4 chose subcutaneous adrenaline, and 19 answered that all four agents have more rapid effects than placebo. Three did not answer.
| Acknowledgements |
|---|
| Footnotes |
|---|
Received for publication December 3, 1999. Accepted for publication May 25, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
More Evidence That IV Steroids Have No Immediate Effect in Acute Asthma Journal Watch Emergency Medicine, February 28, 2001; 2001(228): 2 - 2. [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |