INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Salbutamol is the most commonly used selective ₂-receptor agonist, and is available for inhalation via pressurized metered dose inhalers (pMDI) in more than 90 countries. Most patients are comfortable with this device. However, some asthmatics have difficulties in coordinating the inspiration with activation of the inhaler, especially children and the elderly. This problem can to some degree be overcome by the use of breath-actuated powder inhalers (1).

Examples of commonly used powder inhalers include the Turbuhaler, Diskhaler, and Diskus, all of which are more or less dependent upon the patient achieving enough inspiratory effort to aerosolize the powder. The Turbuhaler is one of the most commonly used powder inhalers, and recently salbutamol has been added to the range of drugs available in this device. However, only limited data comparing the efficacy and safety of salbutamol given via Turbuhaler compared with pMDI have been presented (2, 3). Some of these recent studies suggest that when salbutamol is given via the Turbuhaler, only half the dose is required compared with drug given by the pMDI (2). These studies have compared the efficacy of single doses of 50 to 200 µg salbutamol Turbuhaler with twice the dose from 100 to 400 µg pMDI or Diskhaler, or cumulative doses of salbutamol Turbuhaler up to 2,000 µg, with twice the dose of salbutamol pMDI (2, 4). The aim of the present study was therefore to compare the topical and systemic effects of salbutamol given as the same cumulative doses (µg for µg) via pMDI or Turbuhaler.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The study was performed in accordance with the principles stated in the Declaration of Helsinki, and was approved by the Ethics Committee of Göteborg and the Swedish Drug Agency (Uppsala, Sweden). Furthermore, Good Clinical Trial Practice principles were applied, and the study was monitored by the sponsor (GlaxoWellcome).

Patients

Twenty-two adult patients of either sex (11 females), all with reversible airflow obstruction, were included. The mean age was 48 yr (range, 21 to 68 yr), mean height 173 cm (range, 161 to 188 cm) and mean weight 82 kg (range, 55 to 108 kg). All patients were using inhaled glucocorticoids regularly, and short-acting ₂-agonists on an "as needed" basis. If on a constant dose, the patients were permitted to use inhaled glucocorticoids up to 1,000 µg twice a day, sodium cromoglycate, nedocromil sodium, oral antihistamines, and ₂-agonists prior to entry into and throughout the study. Seven patients were previous smokers, and two patients were ongoing smokers.

Study Design

The study was designed as a single center, randomized, double-blind, double-dummy, cross-over evaluation of the effects of salbutamol Turbuhaler and pMDI versus placebo. The study objective was to compare the topical and systemic effects in cumulative doses of salbutamol delivered by either device. The parameters measured were FEV₁, serum potassium, blood glucose, heart rate, and systolic and diastolic blood pressure. Each patient visited the clinic a total of five times. At the prestudy visit, eligibility was assessed. The following study days were performed at 2- to 10-d intervals.

On the prestudy screening visit, reversibility of FEV₁ was tested, using a stepwise reversibility with salbutamol 200 plus 1,600 µg. The improvement should be at least 10% after the lower dose of salbutamol, and at least another 5% in FEV₁ after the higher dose of salbutamol, reaching at least a total of 15% improvement in FEV₁ over baseline. The effect of the higher dose of salbutamol was required to be at least 50% of the effect observed 15 min after the lower dose.

The patients were not allowed to take inhaled short-acting ₂-agonist 6 h, long-acting ₂-agonist 12 h, or an oral/sustained-release ₂- agonist 24 h prior to prestudy day, or the three study days.

On the study days, each patient was given the following treatments in randomized order: (1) salbutamol via Turbuhaler and placebo via pMDI, or (2) salbutamol via pMDI and placebo via Turbuhaler, or (3) placebo via pMDI and placebo via Turbuhaler. Salbutamol was given in a cumulative dose fashion at approximately 30-min intervals (time = 0, 30, 60, 90, and 120 min). The doses were 200, 200, 400, 800, and 1,600 µg, resulting in the cumulative doses of 200, 400, 800, 1,600, and 3,200 µg. The nominal dose per actuation was 100 µg from both pMDI and Turbuhaler (2 + 2 + 4 + 8 + 16 = 32 inhalations on the dosing day per device per day). During the placebo day, exactly the same protocol was followed.

The Turbuhaler was primed vertically, and several times. Drug was from the Turbuhaler inhaled with a sharp inspiration, and from the pMDI with a steady deep inhalation. The pMDI was actuated in the early part of the inhalation. The patients were trained in inhalation technique prior to study, and every inhalation of study drug was visually monitored by the research staff.

Blood sample for assessment of serum potassium was taken before the cumulative dosing regimen (baseline). After that, blood samples for serum potassium and blood glucose, pulse rate, blood pressure, and FEV₁ measurements were taken 20 to 25 min after each cumulative dose.

Statistical Methods

The efficacy endpoints were based on the FEV₁, heart rate, plasma potassium, and glucose responses following the cumulative dose of salbutamol 3,200 µg given at each study day. Each of the variables was summarized in three ways: the mean value obtained after cumulative dose; area under the dose-response curve; the slope of the linear regression of the response against cumulative dose of each treatment. The comparison of mean values at specified time points or after specified cumulative doses was performed by analyses of covariance using SAS proc GLM (Statistical Analysis System, Institute, NC). The model includes terms for patient, visit, baseline (predosing) level, as well as treatment group.

The comparison of mean area under the response-time curve (AUC) and mean slope of regression of response on (log-) cumulative dose (SLOPE) was carried out using the same procedure as above, though with a change of response variable (AUC or SLOPE) as appropriate. Statistics were performed by the sponsor (GlaxoWellcome, Hockley Park, UK). Data are shown as mean ± SEM.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Twenty-two patients were included in the study, and 20 completed all study days. Baseline FEV₁ for each study day was 2.42 L on the Turbuhaler day, 2.37 L on the pMDI day, and 2.35 L on the placebo day, respectively (p = NS). Mean baseline S-K⁺ for each study day was approximately 4.1 mmol/L on all study days (p = NS). Salbutamol given via both Turbuhaler and pMDI caused a substantial and statistically significant improvement in FEV₁ compared with placebo (Figure 1A). The AUC for FEV₁ for the whole dose-response curve during the Turbuhaler day was 415 ± 29, the pMDI day, 406 ± 27, and the placebo day, 346 ± 25. There were no significant differences in the improvement in FEV₁ between salbutamol given via Turbuhaler or pMDI at any given dose (Figure 1A). The 95% confidence interval for the difference in AUC between pMDI and Turbuhaler was 36 to +44 (p = NS).

View larger version (14K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   Figure 1.   (A) Mean percent improvement in FEV1 after increasing cumulative doses of salbutamol given via either pMDI or Turbuhaler, compared with placebo. Data are shown as mean for each dose of salbutamol. The maximal standard error of the mean was 2.3 after pMDI, 2.9 after Turbuhaler, and 1.8 after placebo. (B) Mean changes in serum potassium after increasing cumulative doses of salbutamol given via either pMDI or Turbuhaler, compared with placebo. Blood samples were taken 20 to 25 min after each dose of salbutamol, or at the corresponding time point after placebo. Data are shown as mean for each dose of salbutamol. The maximal standard error of the mean change in S-K+ was 0.06 after all treatments.

There was no tendency of difference in the induced changes in serum potassium (Figure 1B), or in heart rate, diastolic blood pressure, or blood glucose (data not shown), and after total cumulative dosing, these variables were all within the defined region for equivalence comparing the two active treatments (95% confidence intervals for the true difference).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study does not find any evidence of difference in bronchodilatory effect of salbutamol given either via Turbuhaler or via pMDI, strongly arguing for clinical equivalence of these two devices. Furthermore, we have not been able to detect any tendency of differences between these treatments on systemic safety parameters, defined as serum potassium, heart rate, blood glucose, and diastolic blood pressure. Importantly, a retrospective power calculation based on our own data (the standard deviation of percent FEV₁ at the lowest dose of salbutamol) showed that we had 80% power to detect a difference in improvement of FEV₁ of 5% at this dose (20 patients required), whereas the measured difference was approximately 0.5%.

Earlier studies comparing terbutaline inhaled via Turbuhaler and terbutaline via pMDI suggest clinical equivalence (5, 6). However, two recent reports suggest a higher potency of terbutaline given via the Turbuhaler versus pMDI (7, 8), associated with a higher lung deposition of the drug (8, 9). This discrepancy of data between studies comparing Turbuhaler and pMDI is maintained also when salbutamol is studied.

In contrast to our present study, a recently published study by Löfdahl and coworkers (2) suggested that salbutamol given via the Turbuhaler is twice as potent as salbutamol given by the pMDI, likely due to increased lung delivery from the device (8, 9). Our study, and the study by Löfdahl and coworkers, are similar in some respects, but fundamentally different in other ways. Both studies were careful to select patients with dose-related improvements in FEV₁ within the doses studied. The previous study (2) detected mean efficacy differences (peak FEV₁) between the least and highest effective dose of 0.15 L, whereas our study showed dose differences of 0.31 L. Therefore, the lack of differences between the devices in the present study does not depend on lack of a slope of the dose- response curve of FEV₁.

The study by Löfdahl and coworkers (2) was extremely rigorous to control the range of inhalation flow from each device, whereas the present study has included only the normal training of inhalation technique, used in regular clinical practice. We believe that the method of controlling the inhalation flow in the previous study may have decreased the efficacy mainly of salbutamol pMDI. The patients in that study were looking at an on-line, real life recording of their inhalation flow, and the slow inhalation from the pMDI was often disrupted by the patient when the rapid peak of flow was recorded when the pMDI was activated, and the aerosol and gas was leaving the device (Arvidsson and Lötvall, unpublished observation). By contrast, when the inhalation through the Turbuhaler was recorded, patients were asked to inhale as rapidly as they could, which seemed to be much easier for the patients. We were careful to avoid this procedure in the present study, to avoid the unnatural inhalation patterns, mainly from the pMDI, that we had observed previously. In addition, all the patients in the present study were experienced in using the Turbuhaler device in their everyday treatment, whereas no patient used pMDI on a regular basis. Thus, it could be argued that patient selection, if anything, was in favor of the Turbuhaler.

Most importantly, we have in the present study been particular to compare the same nominal doses of salbutamol from each device, to further avoid any study design advantage for either of the two devices. In the previous study (2), comparisons were made in principle between salbutamol given via Turbuhaler 50 µg and via pMDI 100 µg, and Turbuhaler 200 µg and pMDI 400 µg, which may have been unfair to the pMDI. For example, it may not be fair to compare the effect of a lower dose of salbutamol via Turbuhaler, with a higher dose from the pMDI, because the initial bronchodilation may be similar regardless of dose of a ₂-agonist (4). To show difference in efficacy of the same drug from different devices, in a strict scientific way, we therefore suggest that the same nominal doses of the drug from either device should be used.

In summary, we suggest that the overall evidence argues against any clinically important potency difference between salbutamol inhaled via Turbuhaler or via pMDI in patients with mild asthma. However, further studies are required to fully understand the dose equivalence of lower doses of salbutamol given via different inhaler devices in different populations of asthmatic patients.