A Double-Blind Crossover Study Comparing the Safety and Efficacy of Three Weeks of Flu/Sal 250/50 bid Plus Albuterol 180 ug prn q4 hours to Flu/Sal 250/50 bid Plus Albuterol/Ipratropium Bromide 2 puffs prn q4 hours in Patients with Chronic Obstructive Pulmonary Disease

Abstract

The Federal Drug Administration (FDA) approved the use of Fluticasone 250 μ g/Salmeterol 50 μ g 1 puff bid for maintenance therapy in patients with COPD associated with chronic bronchitis. Short-acting beta agonists (SABA) have been the recommended rescue medication; however, previous studies have shown that combination short-acting Albuterol (alb) /Ipratropium bromide (IB) has superior bronchodilator properties to albuterol alone in patients with COPD. The safety and efficacy of Albuterol compared to Albuterol/Ipratropium bromide as rescue medications for COPD patients on maintenance combination therapy of ICS/LABA has not been evaluated. Double-blind randomized crossover trial with COPD subjects receiving Fluticasone/ Salmeterol 500 μ g/50 μ g (Flu/Sal) 1 puff twice daily and 2 puffs of Albuterol Sulfate (90 μ g micrograms per inhalation) or 2 puffs of Albuterol (90 μ g/puff and Ipratropium Bromide 18 μ g/puff. Either Albuterol Sulfate (90 micrograms/puff) or Alb (90 micrograms/puff)/IB used prn for 3 weeks before crossing over to the other rescue formulation. This is a non-inferiority study where safety and efficacy outcomes were serially assessed, including adverse events, Baseline (BDI)/Transition Dyspnea Index (TDI), St. George Respiratory Questionnaire (SGRQ), SF36, diary cards, 24-hour cardiac monitoring, potassium and glucose levels and other adverse events. Twenty subjects completed the study. The mean age was 62.5 (± 14.5); 12 were males. The mean baseline FEV₁ (range) was 1.12 L (0.56–1.67) or 40.6 (21–65)% predicted. There were no statistically significant differences between either rescue inhaler formulation with regard to measures neither of lung function or dyspnea nor in terms of safety parameters of cardiac monitoring, glucose and potassium levels and other adverse events. SABA and combination SABA/Ipratropium bromide are equally safe and efficacious as rescue inhalers for patients on combination Fluticasone 500 μ g/Salmeterol 50 μ g.

Key words: :

• COPD
• fluticasone
• salmeterol
• albuterol
• atrouent
• combination therapy

View correction statement:

Erratum

INTRODUCTION

In the United States, the FDA has approved the use of salmeterol 50 μ g combined with fluticasone 250 μ g used twice a day as maintenance therapy for patients with COPD associated with chronic bronchitis. The only short-acting bronchodilators (SABD) that are available and appropriate as rapid relievers of dyspnea in COPD are beta-agonists such as albuterol sulfate and anticholinergics such as ipratropium bromide, although the latetr is not FDA approved for use as a prn agent for relief of dyspnea. These short-acting agents are available singly and in combination with each other, and the combination is widely used as a regularly scheduled medication. However, the safety and efficacy of SABD (albuterol alone or combined with ipratropium bromide) when added to a maintenance medication program of a LABA combined with an ICS have not been studied in COPD.

Most studies of LABA in COPD have used SABA for additional dyspnea relief, and LABA compared to placebo reduce the need for short-acting agents ([1], [2], [3], [4], [5], [6]). Short-acting anticholinergic bronchodilators are not FDA approved for prn use to reduce dyspnea for COPD. But, based upon their ability to improve airflow and to relieve dyspnea there is reason to believe that they should be effective for this purpose, although there have been no published reports examining this specific question. Based upon the findings that the combination of an anticholinergic and a short-acting beta agonist improve airflow better than with either agent alone, this combination may be more effective than short-acting beta-agonists alone in rapid relief of dyspnea ([7], [8]).

The safety of short-acting beta-agonists administered with LABA is of potential concern. Beta-agonists, particularly at high doses, may cause tachycardia, hypertension and potentiate arrhythmias and lower serum potassium. Patients with COPD frequently have co-morbidities and may have underlying cardiac disease because of their age and smoking history. Both COPD and cardiac disorders cause shortness of breath, and in patients with known COPD, dyspnea is usually ascribed to COPD. Thus, cardiac disease may be unrecognized. High doses of beta-agonists in COPD may also lead to a decrement in health status as compared to lower doses. For example, Jones et al. have shown that a lower dose of salmeterol improves health-related quality of life more than a higher dose ([2]). Thus, a concern can be raised that overuse of SABA along with regular LABA use may also impair health status.

Information regarding use of short-acting beta-agonists with or without anticholinergics and combination LABA/ICS on dyspnea and quality of life has not been published. Although all studies of long-acting agents have employed prn short-acting beta-agonists the purpose of these studies was to evaluate the long-acting beta-agonists ([1]) or long-acting anti-cholinergic ([9]), and we are not aware of any studies examining the safety and efficacy of short acting beta-agonist/anticholinergic with long-acting beta-agonists.

METHODS AND MATERIALS

Study subjects

Patients were recruited from the National Jewish Medical and Research Center Adult Outpatient Clinic and from the community. Inclusion criteria were subjects over the age of 40 years who were either current or ex-smokers with a greater than 10 pack- year history. The FEV₁ had to be less than or equal to 70% predicted post-bronchodilator, and greater than 0.6 liters or 35% of predicted (whichever was lower). The FEV₁/FVC ratio had to be less than or equal to 70%.

Patients were excluded if they had: a respiratory tract infection within 4 weeks of study entry, symptomatic coronary artery disease, severe hypertension, (malignancy not under optimal control), or significantly abnormal electrocardiograms (chronic but stable atrial fibrillation was allowed), abnormal baseline serum potassium or glucose, known hypersensitive reactions to beta-agonist or anticholinergic agents, or an allergy to soy or peanut. Patients needed to be willing and able to discontinue use of ipratropium (alone), and oral beta-agonists for the duration of the study. Patients taking chronic oral steroids were excluded from the study.

Study design

We conducted a double-blind randomized crossover trial with the intent to recruit 20 patients to complete the trial. Patients had to be stable at study entry; they could not have had an upper respiratory tract infection within 4 weeks of study entry; they could not require regular oral steroids; and, the FEV₁ at baseline needed to be greater than 0.6L. Further, they had to be stable enough to be willing to give up oral beta-agonists. After obtaining informed consent, baseline data were collected on day 0 that included history and physical exam, pre- and post-albuterol (2 puffs) spirometry, EKG, Baseline Dyspnea Index (BDI), SGRQ, SF-36 and a blood sample for potassium and fasting glucose (Figure 1). Subjects were then provided diary cards and entered a 2-week run-in on Fluticasone 500 μ g /Salmeterol 50 μ g (Flu/Sal) (Glaxo Smith Kline (Research Triangle Park/ North Carolina) one inhalation bid and rescue albuterol (Glaxo Smith Kline (Research Triangle Park/ North Carolina).

Figure 1 Study design: Timeline of study.

At the end of the 2-week run-in period, subjects had a 24-hour cardiac holter monitor on day 13. All Holter monitoring data was assessed for heart rate, supraventricular ectopics (SVEs), sustained and non-sustained ventricular tachycardia, and blood pressure changes or palpitations. On day 14, prior to being started on the first study drug, subjects had Transitional Dyspnea Index (TDI), SGRQ, and SF-36 measured. Pre-medication spirometry was measured more than 10 hours after last dose of Flu/Sal. Post-dose spirometry was measured 90 minutes after Flu/Sal and blinded bronchodilator. Subjects were then randomized to receive either albuterol sulfate (Alb) 2 puffs or combination Albuterol Sulfate 90 μ g /Ipratropium Bromide 18 μ g (Alb/IB) (Boehringer Ingleheim, Ridgefield, CT) 2 puffs as needed for rescue. Subjects had repeat measures of outcomes after 3-week trials on study rescue medications while on Fluticasone 500 μ g /Salmeterol 50 μ g (Flu/Sal) twice daily.

Adverse events and drug adherence were checked by telephone or face-to-face interviews every 1 to 2 weeks. Symptom and prn study medication use diary cards were changed out during face-to-face encounters with research team. On day 34 subjects had a repeat 24-hour Holter monitor and on day 35 finished the first treatment period. Blood samples were collected at that time for potassium and fasting glucose as well as clinical measures including Transitional Dyspnea Index (TDI), SGRQ, and SF-36.

Following a 2-week interim period on Flu/Sal and prn albuterol subjects were crossed over to the second treatment period beginning on day 49 with similar measurements as during the first treatment period including Transitional Dyspnea Index (TDI), SGRQ and SF-36. Second treatment Holter cardiac monitoring occurred on day 69 and the end of the second treatment regimen occurred on day 70. Blood samples for potassium and fasting glucose, pre and post spirometry as in treatment period 1 and Transitional Dyspnea Index (TDI), SGRQ, SF-36 were obtained.

ANALYSIS

The central aim of this study was to assess the safety and efficacy of using prn Albuterol versus prn Albuterol/Ipratropium Bromide in COPD subjects treated with Fluticasone 500 μ g/ Salmeterol 50 μ g twice per day. This was a crossover, repeated measures design with 2 groups of subjects. Comparisons between Albuterol versus Albuterol/Ipratropium Bromide were performed using matched pair analysis and ANOVA where appropriate. The primary outcome used for sample size calculations was the SGRQ. Statistical analysis used a mixed effects model for SGRQ, SF36, BDI/TDI, and spirometry.

With the SGRQ as the primary outcome, we assumed the group mean will change from 48.3 to 41.7 for the SGRQ after the administration of the treatment medicines (as in Dahl et al., 2001), then the between-subject standard deviation is 2.20 and the within-subject standard deviation is 4.40. At a 0.05 significance level, this design achieves 81% power when each group has 5 subjects, 92% power when each group has 6 subjects, and over 98% power when each group has 8 subjects. On the other hand, if we assumed that the group mean would only change from 48 to 44 (only a 4 unit decrease), which is the minimum decrease in the SGRQ that is considered to represent a clinically significant improvement, then we require slightly larger sample sizes. With these assumptions using a 0.05 significance level, this design achieves 79% power when each group has 10 subjects, 90% power when each group has 13 subjects, and over 98% power when each group has 20 subjects.

Sample size calculation

The central aim of this study was to examine the safety and efficacy of the two regimens using prn Albuterol versus prn Albuterol/Ipratropium Bromide. This was a cross-over repeated measures design with 2 groups of subjects each measured 4 times. Statistical analysis used a mixed-effects model for SGRQ, SF36, BDI/TDI, and spirometry. For sample size estimates SGRQ was used as the primary outcome variable of interest. At a 0.05 significance level, assuming that the group mean would only change from 48 to 44 (only a 4 unit decrease which is the minimum decrease in the SGRQ that is considered to represent a clinically significant improvement), then to achieve over 98% power we required 20 subjects in each group.

RESULTS

Twenty subjects were entered into the study. The mean age was 62.5 (± 14.5) years, and 12 were males, and there were 15 former smokers and 5 current smokers. As Table 1 demonstrates subjects had severe COPD with a mean baseline FEV₁ (range) of 1.12 (0.56–1.67) L or 40.6 (21–65)% predicted. Interestingly the mean bronchodilator response for FEV₁ was 21.9 (−4.6–71.43) (± 16.7) % however only 8 patients (42%) met criteria for a significant bronchodilator response with greater than 200 cc and greater than 12% improvement in FEV₁.

Table 1 Patient baseline characteristics

Baseline Assessment	N = 20
Age (range)	71.5 (60–83)
Male	12
Current Smokers	5
Atopic	5
FEV1 (L) mean (range)	1.12 (0.56–1.67)
FEV1 (%predicted) (range)	40.6 (21–65)
FVC (L) (range)	2.1 (1.3–3.6)
FVC % predicted (range)	60.7 (42–84)
FEV1/FVC (range)	0.50 (0.30–0.72)
% Reversibility (range)	21.9 (−4.6–71.4)
No. with significant reversibility * (% of patients)	8 (42%)
Short-acting beta-agonists (rescue)	17
Short-acting anticholinergics	2
Albuterol/Ipratropium Bromide	2
Inhaled corticosteroids	2
Long-acting beta-agonists	2
Long-acting beta-agonists/Inhaled corticosteroid use	12

> 12% and greater than 200 cc.

There were 5 patients who had seasonal allergies but no doctor diagnosis of asthma. Prior to study entry, 17 patients were on short-acting beta-agonists and 2 subjects on Ipratropium Bromide and 2 patients on Albuterol/Ipratropium Bromide combination. Two patients had been on ICS, 2 had been on long-acting beta-agonists alone and 1 patient had been on combined LABA/ICS prior to entering the study. Five patients were atopic with seasonal allergies. Eight patients had hypertension and 1 patient had a history of quadruple bypass surgery otherwise there was no pre-existing history of cardiac disease. Four patients had pre-existing problems with enlarged prostate and one had a history of prostate cancer. There were no other significant pre-existing medical conditions in the subjects studied.

Pre-dose spirometric measurements were not significantly different between treatments 1.16 L (44.9 % predicted) for Alb vs. 1.18 L (45.9% predicted) for Alb/IB (Table 2). Post-dose FEV₁ measurements were virtually identical 51.7% vs. 51.8% predicted respectively. FVC data were equally similar for both rescue inhalers. There was no significant difference in the change in FEV₁ or FVC (Figure 2). There were no significant differences between the two rescue medication regimens with regard to dyspnea indices. The mean change for the SF36 (± SD) scores after the albuterol treatment period were 2.1 (± 11.2) for general health, -0.8 (± 17.2) for physical function, 1.3 (± 25.2), for role physical and −1.3 (± 16.5) for social function compared to albuterol/ipratropium bromide scores of 2.0 (± 11.4), 3.2 (± 33.2), 0 (± 24.5), 5.5 (± 11.6) respectively (Figure 3).

Table 2 Spirometry after treatment trials

	Baseline	Albuterol	Albuterol/IB	p-value
Pre-am dose FEV1 (SD)	1.14 (0.38)	1.16 (0.35)	1.18 (0.44)	n.s.
Post-am dose FEV1(SD)	1.36 (0.50)	1.74 (1.98)	1.23 (0.46)	n.s.
Pre-am dose FVC	2.18 (0.58)	2.3 (0.57)	2.2 (0.61)	n.s.
Post-am dose FVC	2.59 (0.78)	2.4 (0.70)	2.3 (0.68)	n.s.
Rescue inhaler use (puffs per day) (SD)		4.1 (1.7)	3.9 (1.4)	n.s.

Figure 2 Change in am. pre-dose FEV₁ and FVC before vs after treatment: mean (± SD). No significant differences were noted between treatment groups.

Figure 3 Change in SF36 before vs after treatment mean and standard deviation. There were no significant differences with regard to any of the SF36 domains.

There was no difference with regard to the change in SGRQ scores for albuterol (–0.73 ± 6.7) versus albuterol/ipratropium bromide (–0.73 ± 7.7). TDI scores did not significantly differ either, albuterol (2.5 ± 0.79) versus albuterol/ipratropium bromide (2.2 ± 0.56). There was also no significant difference with regard to rescue inhaler use between albuterol 4.1 (± 1.7) puffs per day and albuterol/ipratropium bromide 3.9 (± 1.4) puffs per day for albuterol/ipratropium bromide.

Cardiac safety as assessed by Holter monitor data was no different between the two rescue treatments with regard to heart rate, supraventricular ectopics (SVEs), sustained and non-sustained ventricular tachycardia and blood pressure changes or palpitations (Table 3). There were no significant differences between albuterol alone and albuterol/ipratropium bromide for changes in serum potassium 0.15 (± 0.25) vs. 0.04 (± 0.24) meq/ml and glucose 1.5 (± 6.9) vs. 0.7 (± 6.7) mg/dl. There were two reports of bronchitis occurring during albuterol use and 1 report of headache and one of GI upset with Albuterol/Ipratropium bromide use. There were no reports of allergic reactions, upper respiratory tract infections, hoarseness, urinary difficulties or neurological sequelae with either rescue regimens.

Table 3 Assessment of Holter monitor data: change from baseline

	Baseline	Albuterol	Albuterol/Ipratropium Bromide	p-value
Mean Heart Rate ± SD	78.2 (9.5)	77.4 (9.5)	77.6 (8.8)	n.s.
Mean Blood pressure± SD	124 (18.8)/68 ([11])	129 (19.1)/74 (14.3)	124 (13.3)/71 (9.7)	n.s.
Supraventricular ectopics (SVEs)± SD	224 (442)	278.5 (589.6)	209 (556.7)	n.s.
Sustained and non-sustained ventricular tachycardia ± SD	0 (0)	0.45 (1.4)	4.75 (21.24)	n.s

DISCUSSION

Although previous studies of long-acting agents have employed rescue short-acting beta-agonists, the purpose of these studies was to evaluate the long-acting agents, and there have been no studies examining the safety and efficacy of Albuterol/Ipratropium bromide in combination with long-acting beta-agonists. The purpose of this study was primarily to assess the safety of the combination formulation Albuterol sulfate/Ipratropium bromide compared to Albuterol sulfate alone, as rescue bronchodilator inhalers in subjects diagnosed with COPD associated with chronic bronchitis on Fluticasone 500 μ g/Salmeterol 50 μ g twice daily.

This study was conceived and carried out before publication of the Hanania study ([10]) and the subsequent FDA indication given for use of the fluticasone propionate 250/Salmeterol 50 in patients with COPD. Nonetheless given the recent findings from the TORCH (Towards a Revolution in COPD Health) trial suggesting possible clinically significant reductions in mortality using the Fluticasone 500 μ g /Salmeterol 50 μ g. formulation where the p-value did not quite reach statistical significance (p = 0.052) suggests that there may be renewed interest in considering the use of the 500/50 formulation ([11]). For COPD patients. Further the 250/50 indication is an FDA derived standard and does not necessarily reflect the practice of physicians outside the United States and hence given the Torch results and practices of many physicians it is relevant to consider the effect with the Flu/Sal 500/50. It would have been interesting to see if there would have been a difference between the two rescue regimens given a less potent “maintenance” combination.

There were no differences between rescue inhalers in terms of various dyspnea scales and quality-of-life measures and physiological parameters. The formulations were equally safe as assessed by adverse events, potassium levels, glucose levels and Holter monitoring. Previous studies have demonstrated the superior bronchodilator effect of Albuterol/Ipratropium bromide compared to Albuterol alone ([8], [12]) in subjects with COPD yet our study did not demonstrate any advantage to using the combination of Albuterol/Ipratropium compared to Albuterol alone as a rescue inhaler in patients on combination Fluticasone/Salmeterol.

The explanation for the lack of difference in spirometry and the various dyspnea measures is likely multifactorial. The sample size for this study was rather small and one could argue it would be difficult to achieve statistical significance for FEV₁ when sample size was calculated for a change in SGRQ of 4 without a fairly large difference, yet our sample size calculations were based on changes that would be considered to be clinically relevant. There is also growing consensus that the FEV₁ is not the most sensitive parameter for detecting improvements in lung function for patients with COPD. Further, given the small sample size the study lacked significant racial and ethnic diversity and given previous concerns regarding the safety of salmeterol in certain subgroups such as African American asthmatic males this issue certainly warrants further study with inclusion of an adequate representation of such subgroups of concern.

The subjects in this study had a significant bronchodilator response and therefore it could be argued that perhaps these were asthma patients capable of significant improvement with albuterol alone with little added benefit from ipratropium bromide. It is important to note that although there was a significant reversibility in terms of the percent reversibility, a number of the patients had less than a 200 cc improvement and would not meet criteria for a significant BDR. In fact the 42% of patients demonstrating a significant bronchodilator response in our study is lower than that reported in the previously mentioned Hanania study ([10]), where over 50% of subjects in all treatment arms had a significant BDR as well as the Tiotropium study by Casaburi and colleagues ([13], [14], [15]). The subjects in our study were more likely to have COPD rather than asthma as the mean FEV₁ was extremely low (1.12 L or 40% of predicted) and the mean age was 62.5 years with significant smoking histories. None of these subjects carried a concomitant a physician diagnosis of asthma.

Finally, the lack of a significant difference between Albuterol and Albuterol/Ipratropium bromide may be that the use of Fluticasone/Salmeterol 500/50 as maintenance therapy provides optimum sustained bronchodilation such that there was little need or benefit of use of rescue inhalers. Several studies have now demonstrated that the combined use of long-acting beta-agonists and/or inhaled corticosteroids can reduce dyspnea compared to using these agents alone ([10], [16], [17]). We used the Fluticasone/Salmeterol 500/50 formulation in this study and this may have potentially provided additional bronchial-protection compared to the current FDA approved 250/50 formulation use in COPD in the United States.

The study by Mahler did not show any superior effect on the various dyspnea outcome measures comparing the 500/50 to the 250/50 formulation ([18]) suggesting that this may not be a significant factor. Hence, one might expect to see the same outcomes using the 250/50 formulation.

Further, the number of rescue puffs was relatively low in both groups as has been shown in previous studies that the use of long-acting beta-agonists in COPD patients reduces the need for rescue albuterol use ([10], [16], [19]). Further, while the Fluticasone/Salmeterol 250/50 formulation may be the FDA approved formulation in the United States at the current time the results of the TORCH trial by Calverly and colleagues ([11]) suggested potential benefits of using the 500/50 formulation in at least a subset of patients with COPD. In addition there are patients with COPD who demonstrate significant reversibility and/or presence of other features that suggest they fall into the asthma overlap sub-phenotype for whom the 500/50 formulation may provide greater benefit.

COPD patients have co-morbid illnesses (i.e., cardiovascular disease, osteopenia, cataracts) that increase our interest in minimizing complications and side effects from medications used for treating COPD. While there are possible cardiac side effects from both beta-agonists and anticholinergics this study did not demonstrate any significant side effects for either agent used as rescue for this cohort of COPD subjects treated with combination LABA/ICS as maintenance therapy. The subjects in this study had moderate-to-severe COPD and did not have significant rescue albuterol use even at baseline. Studies with more severe COPD patients that require greater rescue inhaler use could possibly identify cardiac and other side effects than were seen in this patient cohort. Further studies examining the use of these rescue inhaler options with maintenance Fluticasone/Salmeterol 250/50 formulation and with more diverse ethnic and racial populations will help to further clarify some of the concerns outlined above.

CONCLUSION

The results of this study indicate that Albuterol and Albuterol/Ipratropium bromide are equally safe and efficacious used as rescue medications in patients who are on maintenance therapy with Fluticasone/Salmeterol. Allowing for the small sample size there are no clear cut advantages in the use of Albuterol/Ipratropium bromide as rescue inhaler compared to Albuterol alone for rescue inhaler use for all patients on Fluticasone 500 ug / Salmeterol 50 ug. For those individual patients who do report superior dyspnea relief using the combination of Albuterol/Ipratropium bromide for rescue there would not appear to be any increased safety concerns compared to albuterol alone.