 |
 |
|
Home |
Current Issue |
Past Issues |
In the Clinic |
ACP Journal Club |
CME |
Collections |
Audio/Video |
Mobile |
Subscribe |
Tools |
Help |
ACP Online
|
Rapid Responses to:
|
|
Electronic letters published:
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif){kind=icon}
The Authors Reply
Effect of inhaled insulin on fasting and postprandial plasma glucose
|
|
|||
|
Anastassios G Pittas, MD MSc Tufts-New England Medical Center, Lisa Ceglia and Joseph Lau
Send rapid response to journal:
apittas{at}tufts-nemc.org Anastassios G Pittas, et al.
|
The authors reply: We would like to thank Drs. Mikhail, Wali and Cope for their comments and we appreciate the opportunity to respond. In our analysis (1), we chose not to present data on fasting plasma glucose, because this measure reflects the effectiveness of a long-acting basal insulin regimen or oral medication rather than inhaled insulin therapy, which has a short time of action. Furthermore, fasting blood glucose is a less reliable measure of glycemia in open label trials because it can be affected by short-term changes in a variety of factors. After combining data for fasting blood glucose available from 8 out of the 16 trials included in our meta-analysis (original references 9, 17, 19-24), we found statistically significant differences in fasting blood glucose levels from baseline favoring the inhaled insulin arm (weighted mean difference -25.8 mg/dl [95% CI -41.9, -9.7] vs. subcutaneous insulin and -22.1 mg/dl [95% CI -44.0, -0.3] vs. oral agents) but with significant heterogeneity among studies. We agree with Dr. Mikhail and colleagues that the effect of inhaled insulin on postprandial hyperglycemia is of interest given the pre-meal indication for inhaled insulin therapy. In our meta-analysis, only 5 of the 16 trials (original references 19-23) reported postprandial blood glucose levels using the same standard meal – 16 oz. Boost liquid meal. None of the trials specified if insulin (inhaled or subcutaneous) was administered shortly prior to the liquid meal. When we combined the data from these trials, we found no significant differences in postprandial blood glucose levels (weighted mean difference -14.8 mg/dl [95% CI -33.2, 3.6] vs. subcutaneous insulin and -44.9 mg/dl [95% CI -92.2, 2.3] vs. oral agents). The results of the fasting and post-prandial glucose concentrations, which tend to favor inhaled insulin over subcutaneous insulin, are in contrast with the hemoglobin A1c results, which slightly favored subcutaneous insulin. In our analysis (1), we elected to present only the change in hemoglobin A1c concentration because it is a measure that captures both fasting and postprandial glycemia, and it is the most reliable and least biased glycemic measure (2). Hemoglobin A1c is also the best predictor of diabetic complications. Hemoglobin A1c is, therefore, the preferred outcome when evaluating the glycemic efficacy of new diabetes therapies (3). Lisa Ceglia, MD Joseph Lau, MD Anastassios G. Pittas, MD MSc Potential Financial Conflicts of Interest: None disclosed. References 1. Ceglia L, Lau J and Pittas AG. Meta-analysis: Efficacy and safety of inhaled insulin therapy in adults with diabetes mellitus. Ann Intern Med 2006; 145:665-675. 2. Goldstein DE, Little RR, Lorenz RA, Malone JI, Nathan D, Peterson CM et al. Tests of glycemic in diabetes. Diabetes Care 2005: 27:7:1761-1773. 3. Nathan DM. Thiazolidinediones for Initial Treatment of Type 2 Diabetes? N Eng J Med 2006; 355:2477-2480. Conflict of Interest:None declared |
|||
|
|
|||
|
Nasser E Mikhail, MD, MSc Endocrinology department, Olive View-UCLA Medical Center, Sylmar, CA 91342, Soma Wali, MD, and Dennis Cope MD
Send rapid response to journal:
nmikhail{at}ladhs.org Nasser E Mikhail, et al.
|
We read with interest the meta-analysis of Dr. Ceglia and colleagues (1) regarding the efficacy and safety of inhaled insulin. This analysis provides a valuable appraisal of this new form of insulin delivery because it represents one of the very few studies of inhaled insulin not sponsored by the manufacturer and therefore is less likely to be associated with bias. However, we disagree with the authors (1) in not collecting data on fasting or postprandial glucose values because they "were self-reported and therefore were less reliable than hemoglobin A1c levels". In fact, in at least 7 of the 16 trials included in the meta-analysis (1), both fasting and postprandial glucose values were obtained from plasma in the laboratory setting and considered as secondary efficacy endpoints. These 7 trials correspond to reference numbers 15, 17, 19, 20, 21, 22, and 23 of the meta-analysis (1). Results from these trials were mixed, with some studies showing superior (reference numbers 17,19,20,22,23), whereas others reporting similar (reference numbers 15,21) effects of inhaled insulin in improving one or both glucose parameters versus the comparison groups. It would be interesting if the authors (1) can analyze the levels of fasting and postprandial plasma glucose from all trials reporting these parameters to give an estimate of the global effect. The postprandial plasma glucose values are particularly important since inhaled insulin has a short duration of action designed to control postprandial hyperglycemia. Reference 1. Ceglia L, Lau J, Pittas AG. Meta-analysis: Efficacy and safety of inhaled insulin therapy in adults with diabetes mellitus. Ann Intern Med 2006; 145: 665-675. Conflict of Interest:None declared |
|||
