|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original Research |
John L. Faul, MD, Department of Respiratory Medicine Connolly Hospital, Dublin, Ireland
Sandra R. Wilson, PhD, Palo Alto Medical Foundation, Research Institute, Palo Alto, California, USA and Division of Pulmonary and Critical Care Medicine, Stanford University, Palo Alto, California, USA
James W. Chu, MD, Division of Endocrinology and Metabolism, Stanford University School of Medicine, Palo Alto, California, USA
James Canfield, BS, Pulmonary Section, Veterans Affairs, Palo Alto Health Care System, Palo Alto, California, USA
Ware G. Kuschner, MD, Division of Pulmonary and Critical Care Medicine, Stanford University, Palo Alto, California, USA and Pulmonary Section, Veterans Affairs, Palo Alto Health Care System, Palo Alto, California, USA
Corresponding Author: Ware G. Kuschner, MD, U.S. Department of Veterans Affairs, Palo Alto Health Care System, 3801 Miranda Avenue, Mail Code: 111P, Palo Alto, CA, 94304, Tel: 650-493-5000, ext. 63544, Fax: 650-852-3276, Email: kuschner{at}stanford.edu
Received: September 11, 2008.
Revised: January 15, 2009.
Accepted: January 21, 2009.
| Abstract |
|---|
|
|
|---|
Design: A prospective randomized, double-blind, double-dummy placebo-controlled, crossover investigation of inhaled steroids and oral leukotriene blockers.
Setting: A United States Department of Veterans Affairs Health Care System outpatient setting.
Participants: Adults with type 2 diabetes and asthma or COPD.
Methods: Subjects (n=12) were randomized to receive either inhaled fluticasone propionate (440 µg twice daily) and oral placebo, or inhaled placebo and oral montelukast (10 mg/day). After 6 weeks, subjects were switched to the opposite therapy for 6 weeks. The primary outcome measure was the change in the percentage of glycosylated hemoglobin (%HbA1c) at 6 weeks relative to the baseline value.
Results: Ten patients completed the study. The difference between the mean within-subject changes in %HbA1c associated with 6-week periods of fluticasone and the mean changes associated with montelukast therapy was small but statistically significant (mean difference=0.25; P<0.025). Neither fluticasone nor oral montelukast therapy for 6 weeks led to a significantly different mean % HbA1c compared with the relevant baseline (mean differences=0.11 and –0.14, respectively).
Conclusion: The absence of a clinically significant within-subject difference in the changes in %HbA1c associated with fluticasone versus oral montelukast therapy, or between either therapy or baseline does not warrant recommending changes in therapy for asthma or diabetes in patients with these co-morbid conditions. However, we suggest that clinicians carefully monitor blood glucose control when diabetic patients initiate ICS, especially with higher dosages.
Key Words: Asthma COPD Diabetes mellitus Inhaled steroid
Type 2 diabetes mellitus, chronic obstructive pulmonary disease (COPD), and asthma are increasing in prevalence worldwide. Many patients with type 2 diabetes have an indication to use inhaled corticosteroids (ICS) for coexisting asthma or COPD. ICS have been shown to have systemic effects, but their effect on glucose metabolism in patients with type 2 diabetes has not been well defined. Although considered a safe therapy, there are concerns about the systemic effects of ICS, including bone formation in children1,2 and cataract development in adults.3–5 Additionally, several studies suggest ICS can suppress the function of the hypothalamic pituitary adrenal axis.6–10 While such adverse effects have been detected in research studies, they are thought not to be clinically important. A significant increase (1.0%) in glycosylated hemoglobin (%HbA1c) and persistent glucosuria has been reported previously in a patient with type 2 diabetes mellitus and asthma who used inhaled fluticasone propionate (FP) at a high dose (2 mg/day).11 It is unclear whether lower doses of ICS might disturb glucose metabolism.
In order to assess the effect of anti-asthma therapy on glucose control in patients with type 2 diabetes, we performed a double-blind, double-dummy, randomized, placebo-controlled, cross-over investigation analyzing patients treated with inhaled FP, 440 µg twice daily, and the oral leukotriene receptor antagonist, montelukast. The purpose of this investigation was to determine whether moderate doses of inhaled FP (<1 mg/ day) adversely affected glucose metabolism in patients with type 2 diabetes.
We compared effects of inhaled FP on glucose metabolism with montelukast, a leukotriene receptor antagonist with no known effects on glucose metabolism, as the control condition. Leukotriene receptor antagonists are widely used in asthma. They are not indicated in the treatment of COPD, but appear to lead to improved lung function in some patients.12,13 Our primary outcome measure was the within-subject difference in %HbA1c after 6 weeks use of ICS compared with 6 weeks use of oral montelukast. Percent HbA1c is a reliable and clinically important marker of glucose control in diabetes mellitus.14–17
| METHODS |
|---|
|
|
|---|
|
Inclusion Criteria
Patients were included if they (1) were age 18 years or greater, (2) had type 2 diabetes confirmed by a fasting plasma glucose >126 mg/dl on a screening visit, and (3) had a physician diagnosis of either asthma or COPD confirmed by electronic medical record review.
Exclusion Criteria
Patients were excluded if they had (1) used tobacco within 6 months of enrollment, (2) a history of exacerbation of asthma or COPD within 3 months of enrollment, (3) current insulin use, (4) use of systemic corticosteroids or ICS, leukotriene receptor antagonists, or theophylline within 1 month of enrollment, (5) inability to read and complete a diary card, (6) inability to perform spirometry, peak expiratory flow testing, and (7) inability to use a metered dose inhaler with spacer.
Interventions
After enrollment, participants completed a 3-week run-in period during which they took a placebo inhaler, two puffs twice daily. On day 21, the initial baseline data were obtained including a blood draw for %HbA1c testing and the randomization was performed in coordination with the research pharmacy at VAPAHCS. A random number generator was used to randomize subjects, in a double-blinded fashion, to receive either inhaled FP (880 µg/day or 2 puffs twice daily of a 220 µg/actuation metered dose inhaler) with a spacer and oral placebo tablets, or inhaled placebo, two puffs twice daily, and oral montelukast (10 mg/day) for 6 weeks. After 6 weeks, subjects crossed over to the opposite treatment limb. After the second 6-week treatment period, the study was completed (figure 1
). At the conclusion of both treatment periods, the research pharmacy disclosed to the investigators the treatment assignment order for each of the study participants.
The placebo medications were supplied by Merck & Co., USA (tablets) and GlaxoSmithKline, USA (metered-dose inhalers) and were identical in appearance to montelukast and FP, respectively. All subjects were instructed in the use of a spacer device in combination with their metered dose inhalers. Albuterol (GlaxoSmithKline, USA) via metered dose inhaler was prescribed on an as-needed basis. Other prescribed medications, including cholesterol lowering agents and oral hypoglycemic agents, were not changed during the study. At the time of randomization and after each 6-week treatment period, participants underwent spirometry (Sensor Medics, Yorba Linda, CA) both before and again 20 minutes after the administration of nebulized albuterol. Subjects also had a fasting blood draw for %HbA1c. Six weeks was chosen as a treatment period because clinically significant increases in %HbA1c after 6 weeks of therapy with very high-dose (2 mg/ day) inhaled corticosteroids had been previously reported.11 Six subjects who also took oral hypoglycemic medication recorded once daily self-monitoring of blood glucose (SMBG) measured after an overnight fast. The others who controlled their diabetes with diet only did not perform daily SMBG. Fasting serum insulin was measured in all subjects.
Participants were contacted by telephone once weekly during the study to ensure drug compliance and to assess for adverse events. Subjects were instructed to measure and record their daily peak expiratory flow rate (PEFR) using a hand-held Mini-Wright flow meter (Clement Clarke International, Essex, UK). The study safety limit was specified to be an acute fall in PEFR by more than 20% or a SMBG measurement >300 mg/dl.
Objectives
The primary objective was to determine whether inhaled FP at a dose of 440 µg twice daily affects glucose control in patients with type 2 diabetes who have coexistent asthma or COPD. The primary outcome measure was within-subject difference in %HbA1c after 6 weeks of fluticasone versus 6 weeks of oral montelukast therapy. The secondary outcome was the within-subject change in %HbA1c from baseline after each type of therapy.
Outcome Measures
The study was conducted as a crossover design with no washout period between therapies. Percent HbA1c was measured on days 63 (at the end of treatment period 1), and on day 105 (at the end of treatment period 2) by high performance liquid chromatography (Bio-Rad Variant, Bio-Rad Laboratories, USA). Baseline %HbA1c for period 1 was the day 21 value; baseline for period 2 was the day 63 value.
Since only six subjects measured SMBG, we used SMBG as a safety rather than an outcome measure in those patients. Weekly mean SMBG was determined by averaging the daily values within each 7-day period. SMBG was measured using the Accu-chek (Roche Diagnostics, USA) glucose meter system. Other safety measures included lung function (forced expiratory volume in one second [FEV1], the forced expiratory volume in 1 second to forced vital capacity ratio [FEV1/FVC], the forced expiratory flow during the middle 50% of the forced expiratory maneuver [FEF25–75], and the presence of a bronchodilator response defined as an increase in FEV1 over baseline spirometry 15 minutes after administration of 2 puffs [180 µg] of albuterol by metered dose inhaler) and change in fasting serum insulin. Weight was measured with clothes on and shoes off at each study visit.
Sample Size
We anticipated a mean increase of 1.0 in %HbA1c after fluticasone treatment based on a previous report.11 The standard deviation (SD) of %HbA1c in patients with type 2 diabetes ranges from 0.27 among those with good glucose control and 0.44 among patients with fair control to 1.73 with poor control.18 Since no previous investigations have studied the effect of ICS on glucose control, we estimated the SD of the change in %HbA1c from baseline would be 1.0, with no change as a result of oral montelukast therapy. Using these values, we estimated a sample size of 16 patients with a crossover study design would be necessary to detect a statistically significant difference between the therapies utilizing a two-tailed t test (
=0.05, β=0.20).
Data Collection
Baseline demographic, lung function, and %HbA1c and fasting blood glucose data were obtained just prior to randomization (table 1
) and at 6 and 12 weeks after initiation of therapy. Subjects subsequently recorded daily PEFR in diary cards; those subjects performing SMBG also recorded results in diary cards. Investigators were blinded to the treatment limb throughout the study.
|
| RESULTS |
|---|
|
|
|---|
The mean within-subject baseline to 6-week difference in %HbA1c was significantly greater after treatment with inhaled FP than after treatment with oral montelukast therapy (mean within-subject difference=0.25; P<0.025) (table 2
), but the mean difference was smaller than the 0.5% difference that would be considered clinically relevant and smaller than the 1.0% value the sample size was originally designed to detect. The fact that this difference was statistically significant with only 5 subjects per group is due to the lower-than-anticipated variance in the within-subject, within-treatment differences.
|
|
Post hoc analysis revealed the presence of a carryover effect (P <0.02) that resulted in an underestimate of the change in %HbA1c associated with inhaled FP therapy. There was no significant difference between therapies in terms of weight or serum insulin (table 2
).
| DISCUSSION |
|---|
|
|
|---|
The observed treatment-related changes in %HbA1c in this study are consistent with our prior report of hyperglycemia and glucosuria in an asthmatic patient with type 2 diabetes mellitus who took very high doses of inhaled FP at a dose of 2 mg/day; however, the mean increase resulting from FP therapy, relative to the individuals own baseline, is substantially smaller than in that individual case.11 Reduced insulin sensitivity, without hyperglycemia, has been demonstrated in non-diabetic, non-asthmatic adults who took inhaled beclomethasone dipropionate (1 mg/day for 4 weeks)19 and asthmatic (non-diabetic) children who took inhaled budesonide (0.8 mg/m2/day for 4 weeks).20 In one uncontrolled study, reductions in serum insulin concentrations were observed in non-diabetic patients who took high-dose inhaled steroids for unstable asthma.21 The authors concluded that this effect was more likely due to a fall in endogenous catecholamines that accompanied the resolution of unstable disease, rather than an effect of inhaled steroid therapy. In another study, 14 normal subjects and 10 non-asthmatic patients with diet-controlled type 2 diabetes mellitus took inhaled placebo for 2 weeks followed by inhaled beclomethasone dipropionate (2 mg/day) for 2 weeks.22 Inhaled glucocorticoid therapy did not significantly alter glucose tolerance tests and insulin levels after 2 weeks. However, %HbA1c and SMBG were not measured, and the authors comment that an effect might have been observed with prolonged therapy, which also was not explored in the present study. In addition, beclomethasone is less potent than fluticasone as assessed by gene transactivation and transrepression.23
Large intervention studies in patients with type 1 and type 2 diabetes mellitus have demonstrated that intensive glycemic control reduces the onset and delays the progression of microvascular complications, including retinopathy, nephropathy, and neuropathy.16,24–28 Risk reductions in various outcomes ranged between 25% to 75%, and these reductions appeared to be related to the duration and severity of hyperglycemia.24,27,28 The United Kingdom Prospective Diabetes Study demonstrated a continuous relationship between glycemic control and various complications, such that for a reduction in %HbA1c of 1.0, there was a 35% reduction in the risk of complications, a 25% reduction in diabetes-related deaths, a 7% decrease in all-cause mortality and an 18% reduction in combined fatal and non-fatal myocardial infarction.24,27 Accordingly, modest deterioration in glucose control attributable to anti-asthma therapy, even on the order of the differences observed in this study, could contribute to the development of diabetic complications in at-risk populations.
| CONCLUSION |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
|
|
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Laghi, N. Adiguzel, and M. J. Tobin Endocrinological derangements in COPD Eur. Respir. J., October 1, 2009; 34(4): 975 - 996. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |