Background  A new inhalable insulin aerosol (Inh-Ins) was developed in China. The aim of this multicenter clinical study was to evaluate the efficacy and safety of this new Inh-Ins as a treatment of type 2 diabetes. Regular porcine insulin (RI) was used as a control.
Methods  This study is a prospective, randomized, open-label, parallel-group multicenter clinical trial in which 253 qualified patients with type 2 diabetes received the insulin Glargine daily at bedtime plus either a pre-meal Inh-Ins or a pre-meal subcutaneous RI for 12 weeks. HbA1c, fasting plasma glucose (FPG), the 1-hour-postprandial blood glucose (1hPBG) and the 2-hour-postprandial blood glucose (2hPBG) were measured. Events were monitored for adverse effects.
Results  After 12 weeks, the HbA1c decreased significantly from baseline in both treatment groups, with no significant difference between the two regimens. In the Inh-Ins group, FPG, both 1hPBG and 2hPBG significantly declined from baseline after the 8th- and 12th-weeks of treatment. The reduced values of FPG or 1hPBG between the two groups showed a more significant hypoglycemic effect with the Inh-Ins than the RI. After 12 weeks, the pulmonary carbon monoxide diffusing capacity (DLco) was significantly lower in Inh-Ins group than in the RI. The main side effects of Inh-Ins were coughing, excessive sputum, and hypoglycemia.
Conclusions  Inh-Ins was effective in decreasing HbA1c like the RI. It was better in lowering the FPG and the 1hPBG than the RI. Its main side effects were coughing, excessive sputum, and hypoglycemia. Also, Inh-Ins slightly impaired DLco.

·LogIn/LogOut

·Fulltext PDF(140K) Free

·Abstract download TXT | XML

·Articles in CMJ by LIAO Zhi-hong CHEN Ying-li

·Articles in PubMed by LIAO ZH CHEN YL

·Put into my bookshelf

·Email to Friend

·Email to author

·Visit:3638

·Download:1667

·Advanced Search

·Related Articles

·Change font size:

·Cannot read some characters

Inhalable insulin aerosol (Inh-Ins) originated from porcine insulin is manufactured by Chuangxinhui Biotech Venture Capital Co., Ltd, in Shenzhen, China, which is the first creator of aerosol-form inhalable insulin in China. The pharmacokinetics study in type 1 diabetic patients showed that the Inh-Ins exhibited a quicker onset and a higher peak of action as compared to regular pork insulin (RI) injected subcutaneously. The average relative bioavailability for Inh-Ins is 10.7%.¹ Inh-Ins, traveling through the respiratory tract, acts in a fashion similar to the physiological insulin secretion pattern after meal, so that postprandial hyperglycemia can be better addressed.^2,3 The present study was a phase II multicenter clinical study, which was approved by the State Drug Administration of China (No. 2002HL0212). The study was also registered at Current Controlled Trials (number: ISRCTN81772352- http://www.controlled-trials.com/ISRCTN81772352). The aims were as following: (1) to assess whether Inh-Ins combined with a single injection of insulin Glargine can provide a comparable glycemic control to that of a conventional subcutaneous insulin regimen for patients who previously used at least two daily subcutaneous injections of insulin to manage their type 2 diabetes, and (2) to assess the tolerability of the Inh-Ins for the subjects over a 3-month period.

Subjects
Numerous men and women (n=253) diagnosed with type 2 diabetes were screened out at 5 centers in China. The inclusion criteria were being of age 18 to 65 years, having a stable subcutaneous insulin schedule involving 2 to 3 injections daily for at least 2 months before study entry and having not received any oral antidiabetic agents for at least 1 month, having no more than a 13 mmol/L fasting plasma glucose (FPG) value for the screening and pre-randomization, having a body mass index (BMI) of 18–28 kg/m², and giving an written informed consent. The exclusion criteria included having asthma, chronic obstructive pulmonary disease or other significant respiratory diseases, having smoked within the last 6 months, having an abnormal screening chest X-ray, or an abnormal pulmonary function at screening (carbon monoxide diffusing capacity (DL_CO) <75%, total lung capacity (TLC) <80% or >120%, and forced expiratory volume in first second (FEV₁) <70% of predicted),⁴ having any major organ system disease, showing clinically significant abnormalities on laboratory screening, having partaken in concomitant therapy with systemic glucocorticoids, having participated in any inhalable insulin clinical trial previously, and having a daily insulin requirement of >1.0 U/kg.

The size of 126 subjects for each treatment group was calculated according to the main curative effects determined by the decreased mean hemoglobin A1c (HbA1c) from baseline, the validity of a test, and the rate of withdrawal.

Methods
This was a randomized, open-labeled, parallel, controlled study consisting of a screening visit, a 2-week baseline lead-in phase, and a 12-week treatment phase. All subjects received insulin Glargine at bedtime as the basal insulin supply during the two phases. During the baseline lead-in period, all subjects were injected subcutaneously with regular human insulin 30 minutes before each meal (3 times per day). Then, during the treatment phase, the patients were randomized to receive either a pre-meal Inh-Ins inhalation regimen or a pre-meal RI subcutaneous injection therapy (RI replaced regular human insulin of the lead-in period in RI regimen). Those who received Inh-Ins were given it 10 minutes before each meal (3 times per day); each capsular of the Inh-Ins contained 40 IU of RI. All patients followed the instructions about regulated diet, exercise, and self monitoring of blood glucose (SMBG). The dose of insulin was adjusted at the discretion of the investigator, based on the SMBG results, to achieve the target range of 4.4–7.8 mmol/L for fasting or pre-meal, <10 mmol/L for postprandial, and 5.6–8.9 mmol/L before bedtime. Hypoglycemia should be avoided as less as possible.

Assessments
The primary efficacy end point was the change in the HbA1c from baseline to week 12. All blood samples of the HbA1c were measured at the 301 Hospital, Beijing. The centralized measurement of HbA1c was proposed to preserve the good quality results. The secondary efficacy end point included measuring changes in the FPG and 1-hour and 2-hour postprandial plasma glucose (1hPBG and 2hPBG) responses after a standardized breakfast (1381 kJ of Glucerna-SR^®) load.

Hypoglycemia was defined by typical symptoms that resolved promptly with food intake without glucose measurements, typical symptoms include glucose concentrations of ≤3.3 mmol/L, or any glucose measurement of ≤2.7 mmol/L. For the classification of severe hypoglycemia, all of the following criteria had to be met: (1) the subject was unable to treat himself or herself, (2) they exhibited neurological symptoms (memory loss, confusion, uncontrollable or irrational behavior, difficulty in awakening, seizure, or coma), and (3) their blood glucose ≤2.7 mmol/L or, if not measured, the clinical manifestations were reversed by oral carbohydrates or intravenous glucose.

Laboratory tests (complete blood count, urinalysis, and blood chemistries), pulmonary function tests, chest X-rays, and electrocardiograms were performed at screening and during week 12. Physical examinations (heart rate, blood pressure, and pharynx and chest examination) were performed throughout the study. All adverse events were recorded by the investigators.

Statistical methods
This comparative trial was designed to test the "non-inferiority" of an Inh-Ins regimen relative to a subcutaneous insulin regimen with respect to the change in the HbA1c from baseline to week 12. An ANCOVA model, with baseline HbA1c as a continuous covariant and its indicator variables for the center and treatment group, was fitted to the week 12 change from the baseline HbA1c values. The 95% CI for the comparison of the inhalation and the subcutaneous insulin was derived from this model. Non-inferiority of the Inh-Ins regimen to the RI regimen was concluded if the upper limit of the 95% CI for the difference was < 0.5% HbA1c, as specified in the protocol. Analysis was performed for the full analysis set (FAS) population and per-protocol set (PPS) population separately. The hypoglycemic event rate was calculated by each subject in every exposure month. Its relative risk analysis was based on a generalized linear Poisson regression model. ANOVA and t-tests were carried out for measurement variables, and the Chi-square analysis was done for the enumeration data. The statistic significance was determined at the two-sided P value <0.05. All analysis was completed using the SAS 9.1.3 software package.

Characteristics of participants
Totally 253 patients were screened and randomized into the cohort study; 126 patients were allocated to the Inh-Ins regimen randomly, and 127 patients were on the RI regimen. Totally 241 patients fulfilled the trial, while the remaining 12 patients withdrew because of side effects (5 with cough, 3 with minor stoke), insufficient clinical response (1 case), unwillingness to continue (1 case), or lose of communication (2 cases). All the withdrawal subjects belonged to Inh-Ins group. The PPS population contained 238 patients; 113 patients followed the Inh-Ins regimen, and 125 the RI regimen. Characteristics of the participants noted at the study entry are given in Table 1.

view in a new window

Table 1. Demographic and clinical characteristics at study entry

view in a new window

Table 2. Mean changes in glucose control and the doses of insulin

In the Inh-Ins group, the mean levels of FPG, 1hPBG or 2hPBG decreased significantly after both the 8th and 12th week of the treatment when pair-compared to the baseline. On the other hand, in the RI group, the mean level of FPG, 1hPBG or 2hPBG had insignificant changes after the similar periods of treatment. The mean changes of the 12-week-treatment from the baseline differed significantly between the two regimens regarding the FPG, 1hPBG or 2hPBG.

An insignificant difference was shown for the doses of Glargine between the two groups. The doses of insulin before meals were different between the two groups. Compared to the RI pre-meal injection, the mean relative dose of Inh-Ins was about 11.3% of its original dose.

Serious adverse effect
Serious adverse effect (SAE) occurred in four subjects of Inh-Ins group (rate of 3.2%), none in RI group; however, the SAEs differences between the two regimens were statistically insignificant according to Fisher's exact test. The SAEs included three cases with minor ischemic strokes with short-term recovery. And one subject suffered sudden neuropathy deafness in both ears; the patient continued and completed the study. The investigators judged that the SAEs were treatment- unrelated adverse events (AE).

Pulmonary functions
The mean changes in Forced Vital Capacity, FEV₁, and TLC were small and comparable between the two treatment groups. The DLco was also comparable between two groups at baseline. It decreased significantly after the 12th week of the Inh-Ins treatment ((95.85±23.35)%), but not after the RI treatment ((99.93±22.67)%). The mean change of DLco from baseline differed significantly between the two regimens.

Cough
The predominant AEs were coughing and excessive sputum when the Inh-Ins was inhaled, which 42 cases (33.6%) exhibited. This rate was higher than that of the RI group. Four patients experienced coughing in the RI group (3.1%). Comparing in the different age subgroups of In-Ins, no age-related increase of coughing and sputum was found. The cough was mostly mild to moderate and decreased in incidence over the study period, but five patients discontinued the study because of either a serious cough or an infection of upper respiratory tract.

Hypoglycemia
In the Inh-Ins group, 83 patients (66.4%) experienced a total of 228 hypoglycemic events within the 12-week treatment: a crude rate of 0.55 events per subject every exposure month. The 59 patients (46.5%) of the RI group experienced a total of 126 events: a crude event rate of 0.23 events per subject-month. This represents a risk ratio (inhaled/subcutaneous) of 2.38 (95% CI, 1.57–3.61) for any hypoglycemic event, indicating that there is a higher risk of hypoglycemia associated with Inh-Ins. Within the Inh-Ins group, the hypoglycemic rates decreased significantly from 0.78 during the first month of treatment to 0.46 events per subject-month during the second month and 0.35 during the last month. No nocturnal or severe hypoglycemia events occurred in either treatment group.

Body weight
The mean body weights were (65.64±10.37), (65.63±10.3), (65.97±10.67), and (66.09±10.55) kg after 2 weeks, 4 weeks, 8 weeks, and 12 weeks of inhalation respectively. They were (65.02±10.3), (65.23±10.23), (65.48±10.55), and (65.7±10.55) kg for RI group. The mean body weight increased significantly for both regimens compared to the baselines. The difference of changes in body weight was insignificant between two groups.

Other adverse events
Other treatment-related AEs of Inh-Ins were dizziness/ headache (2 patients, 1.6%), heart-throbbing (2 patients, 1.6%), sweating (1 patient, 0.8%), chest pain (1 patient, 0.8%), vision abnormality (1 patient, 0.8%), and lip numbness (1 patient, 0.8%). The incidence of clinical laboratory abnormalities was rare, and no difference between the two treatment groups was shown.

Inhalable insulin is attractive for its convenience and efficacy.^5-7 Inh-Ins is a dry powder aerosol contained in a capsule and can be inhaled using a device different from that of Exubera^®.^8,9 Why should patients use regular pork insulin rather than human insulin? The originally invented powder was made from regular pork insulin; the manufacture company wanted to develop it, and its clinical effects were established in the phase I clinical study.¹ The phase I study only included subjects with type 1 diabetes. We conducted this phase II clinical study protocol. The protocol mimicked the AERx study¹⁰ and was similar to the efficacy and safety phase II and phase III study of Exubera^® inhalation in type 2 diabetic patients.^11,12 This multicenter study was carried out in five qualified hospitals for clinical trials in China. The results showed that Inh-Ins provides a glycemic control comparable to that of the RI pre-meal injection regimen in patients with type 2 diabetes, as assessed based upon the changes in the HbA1c from baseline to week 12. The decrease of the HbA1c in both treatment groups was modest because the inclusion criteria had no HbA1c limitation. This study only included subjects with FPG ≤ 13 mmol/L and who required an insulin supply. The baseline HbA1c levels were not high. In addition, the study was designed to demonstrate equivalence and was not target driven. As a result, we did not compare the proportion of patients who reached the American Diabetes Association's goal of HbA1c <7% in both groups. This efficacy result of Inh-Ins was similar to the Meta-analysis of other inhaled insulin therapies.¹³

This result showed that the addition of the pre-meal Inh-Ins reduced both FPG and PBG levels when compared to the mean levels of the FPG, 1hPBG, and 2hPBG between the baseline and week 12. Inh-Ins was better in PBG control than RI, especially for 1hPBG, which was well demonstrated by its rapid onset of action and higher peak insulin concentration, so that postprandial hyperglycemia can be better addressed.^1-3

Interestingly, the FPG decreased more in the Inh-Ins group than in the RI group. Similar results were reported by other studies.^10,11 This can be explained by the bedtime administration of Ultralente basal insulin in the inhalable insulin group that contributed to the lower FPG.¹¹ The parallel-control group for the Exubera^® study, a subcutaneous group, was given injections twice per day.¹¹ In our study, the same amount and the same injection time of the basal insulin Glargine were used for both groups. We thought one of the reasons for the lower FPG was due to the longer action of the Inh-Ins than that of the RI.^1,3 Devries gave the reasons.¹⁴ Many studies concluded that a mealtime inhalable insulin is better at lowering fasting glucose.^2,10,11,15

We also concluded that the relative dose of Inh-Ins was about 11.3% of its original dose, referring to the RI pre-meal injection. It was consistent with our phase I study among type 1 diabetic patients, which showed its relative bioavailability of 10.7%.¹ This study was not designed to find out the relative bioactivity of Inh-Ins among type 2 diabetes patients. We calculated the crude exchange rate of Inh-Ins to RI for the purpose of a clinical indication.

The risk of an hypoglycemic event was higher in the Inh-Ins group compared to the RI group, which is controversial to other studies.^3,11,16,17 We thought it might be a result of the inconvenience in adjusting the inhalation dose of Inh-Ins. As we know, each capsule contained 40 unit of insulin. We also found that the incidence of the hypoglycemic event decreased gradually, but significantly, from month to month, as the subjects adjusted to the dosage. That is a similar result to the other studies.¹⁸ We suggest that the initial dose of Inh-Ins should be lowered considerably when it is used for the first time, in order to avoid hypoglycemia.

The influence of the Inh-Ins on respiratory function has been a big concern since the invention of the inhalable drug delivery system. This study showed the impairment of DLco in Inh-Ins group. A long-term extension study of the Exubera^® Phase III program showed that the changes in lung function remained small and non-progressive.^11,18 The small reductions in the lung functions of FEV₁ and DLco occurred early (within 12 weeks), did not progress during the first 2 years of treatment, and was reversible after it was withdrawn.^19,20 Lung function should be measured every 6 to 12 months when the Exubera^® is administrated as indicated by the FDA. Hence, a long-term study of the Inh-Ins is required to assess any insidious effect more conclusively. As a clinical indication, respiratory disorders such as smoking, asthma, and chronic obstructive pulmonary disease, can influence the absorption of inhaled insulin.^21-23

Coughing is another problem for Inh-Ins use. The incidence of coughing was high, accounting for 33.6% cases in this study. Most of the coughing was not serious and was alleviated as the study progressed; however, five patients experienced a severe cough or an infection of upper respiratory tract caused by Inh-Ins. Therefore, the inhalation had to be stopped when the coughing became worse. Liu et al²⁴ reported that Exubera^® was not associated with any evidence of pulmonary inflammation during the 12 weeks of treatment. The treatment effects on lung function observed in Exubera^® trials were not caused by lung inflammation. In the present trial, no age-related increase of cough and sputum was found among the different age subgroups. The reason might be that the enrolled subjects did not have a respiratory disease and that they were not old enough. We agree with the idea of Allen et al²⁵ that in an older group of patients, the use of inhalable systemic treatments should be limited because of the physical weakness of muscle, the effects of lung aging, co-pathologies, and cognitive problems.

An insignificant difference was found between the changes of body weight of the two groups, although the body weight increased slightly and significantly from baseline in both regimens. That is the result of the intensive glucose control, which is consistent with the results of other studies.²⁶ The increase of insulin antibodies with Inh-Ins has previously been reported, but has not been clinically meaningful.^19,27 Because we have not measured antibody levels in this study, we did not discuss them here.

Unfortunately, four cases of SAEs occurred in four patients following the Inh-Ins regimen, even though the SAEs were not related to Inh-Ins usage based on the judgment of the investigators. We should be aware of the potential role of hypoglycemic reactions in inducing a stroke, even though no such evidence was proved before the events of SAE. The investigator could not present on the sites when the SAEs occurred. Statistically, an insignificant difference of SAE was shown between the two regimens. In searching through published papers about inhalable insulin, we found that no drug-related SAE had been observed hitherto.^19,28

Overall, compared to a subcutaneous RI regimen, the Inh-Ins showed the similar efficacy in glycemic control. Also, it was better in lowering FPG and PBG than RI group. Inh-Ins was well tolerated. Coughing and excessive sputum were the main side effects. Inh-Ins impaired the DLco of the lung slightly; thus, a long-term safety study is necessary. A lower initial dose is recommended because of its hypoglycemic effect.

Competing interests: None declared.
Acknowledgments: We thank all of the investigators and coordinators who took part in this study. The Contract Research Organization, Excel Pharma Studies Inc., Beijing, monitored and performed the statistic analysis for the study.

1. Liao ZH, Liao Y, Li YB, Hu GL, Wang XD, Huang M, et al. Phase I clinical study of inhaled insulin Powder. Natl Med J China (Chin) 2006; 86: 2005-2007.

2. Rave KM, Nosek L, de la Peña A, Seger M, Ernest CS 2nd, Heinemann L, et al. Dose response of inhaled dry-powder insulin and dose equivalence to subcutaneous insulin lispro. Diabetes Care 2005; 28: 2400-2405.

3. Liao ZH, Meng DD, Weng JP. The update researches for inhaled insulin. Chin J Diabetes (Chin) 2006; 14: 313-314.

4. Zheng JP, Zhong NS. Normative values of pulmonary function testing in Chinese adults. Chin Med J 2002; 115: 50-54.

5. Chancellor J, Aballéa S, Lawrence A, Sheldon R, Cure S, Plun-Favreau J, et al. Preferences of patients with diabetes mellitus for inhaled versus injectable insulin regimens. Pharmacoeconomics 2008; 26: 217-234.

6. Testa MA, Simonson DC. Satisfaction and quality of life with premeal inhaled versus injected insulin in adolescents and adults with type 1 diabetes. Diabetes Care 2007; 30: 1399-1405.

7. Hayes RP, Muchmore D, Schmitke J. Effect of inhaled insulin on patient-reported outcomes and treatment preference in patients with type 1 diabetes. Curr Med Res Opin 2007; 23: 435-442.

8. Chen XJ, Zhu JB, Wang GJ, Zhou MX, Xin FM, Zhang N, et al. Hypoglycemic efficacy of pulmonary delivered insulin dry power aerosol in rats. Acta Phamacol Sin (Chin) 2002; 23: 467-470.

9. Chen XJ, Zhu JB, Wang GJ, Zhou MX, Xu YN. HPLC determination of insulin and its related substances in insulin powder for inhalation. Acta Pharmaceutica Sinica (Chin) 2002; 37: 131-133.

10. Hermansen K, Ronnemaa T, Petersen AH, Bellaire S, Adamson U. Intensive therapy with inhaled insulin via the AERx insulin diabetes management system: a 12-week proof-of-concept trial in patients with type 2 diabetes. Diabetes Care 2004; 27: 162-167.

11. Hollander PA, Blonde L, Rowe R, Mehta AE, Milburn JL, Hershon KS, et al, for the Exubera Phase III Study Group. Efficacy and safety of inhaled insulin (exubera) compared with subcutaneous insulin therapy in patients with type 2 diabetes: results of a 6-month, randomized, comparative trial. Diabetes Care 2004; 27: 2356-2362.

12. Cefalu WT, Skyler JS, Kourides IA, Landschulz WH, Balagtas CC, Cheng S, et al, for the Inhaled Insulin Study Group. Inhaled human insulin treatment in patients with type 2 diabetes mellitus. Ann Intern Med 2001; 134: 203-207.

13. 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.

14. Devries JH. Mealtime inhaled insulin lowers fasting glucose: a look at possible explanations. Diabetologia 2005; 48: 2682-2683.

15. Edgerton DS, Cherrington AD, Williams P, Neal DW, Scott M, Bowen L, et al. Inhalation of human insulin (exubera) augments the efficiency of muscle glucose uptake in vivo. Diabetes 2006; 55: 3604-3610.

16. Heise T, Bott S, Tusek C, Stephan JA, Kawabata T, Finco-Kent D, et al. The effect of insulin antibodies on the metabolic action of inhaled and subcutaneous insulin: a prospective randomized pharmacodynamic study. Diabetes Care 2005; 28: 2161-2169.

17. Dailey G. Developing a pulmonary insulin delivery system for patients with diabetes. Clin Ther 2007; 29 Spec No: 1271-1283.

18. Skyler JS, Jovanovic L, Klioze S, Reis J, Duggan W, Inhaled Human Insulin Type 1 Diabetes Study Group. Two-year safety and efficacy of inhaled human insulin (Exubera) in adult patients with type 1 diabetes. Diabetes Care 2007; 30: 579-585.

19. Barnett AH, Lange P, Dreyer M, Serdarevic-Pehar M, Exubera Phase 3 Study Group. Long-term tolerability of inhaled human insulin (Exubera) in patients with poorly controlled type 2 diabetes. Int J Clin Pract 2007; 61: 1614-1625.

20. Hegewald M, Crapo RO, Jensen RL. Pulmonary function changes related to acute and chronic administration of inhaled insulin. Diabetes Technol Ther 2007; 9 Suppl 1: S93-S101.

21. Pan AX, de la Peña A, Yeo KP, Chan C, Loh MT, Wise SD, et al. Effects of smoking cessation, acute re-exposure and nicotine replacement in smokers on AIR inhaled insulin pharmacokinetics and glucodynamics. Br J Clin Pharmacol 2008; 65: 480-487.

22. Wolzt M, de la Peña A, Berclaz PY, Tibaldi FS, Gates JR, Muchmore DB. AIR inhaled insulin versus subcutaneous insulin: pharmacokinetics, glucodynamics, and pulmonary function in asthma. Diabetes Care 2008; 31: 735-740.

23. Rave K, de la Peña A, Tibaldi FS, Zhang L, Silverman B, Hausmann M, et al. AIR inhaled insulin in subjects with chronic obstructive pulmonary disease: pharmacokinetics, glucodynamics, safety, and tolerability. Diabetes Care 2007; 30: 1777-1782.

24. Liu MC, Riese RJ, Van Gundy K, Norwood P, Sullivan BE, Schwartz PF, et al. Effects of inhaled human insulin on airway lining fluid composition in adults with diabetes. Eur Respir J 2008; 32: 180-188.

25. Allen S. Are inhaled systemic therapies a viable option for the treatment of the elderly patient? Drugs Aging 2008; 25: 89-94.

26. Hollander PA, Krasner A, Klioze S, Schwartz P, Duggan W. Body weight changes associated with insulin therapy: a retrospective pooled analysis of inhaled human insulin (Exubera) versus subcutaneous insulin in five controlled Phase III trials. Diabetes Care 2007; 30: 2508-2510.

27. Davidson MB, Mehta AE, Siraj ES. Inhaled human insulin: an inspiration for patients with diabetes mellitus? Cleve Clin J Med 2006; 73: 569-578.

28. Harsch IA. Inhaled insulins: their potential in the treatment of diabetes mellitus. Treat Endocrinol 2005; 4: 131-138.