Impaired Glucose Tolerance, Obstructive Sleep Apnea
Conditions
Brief summary
The investigators performed a randomized, cross-over controlled clinical trial comparing 8 weeks of Continuous Positive Airway Pressure (CPAP) to 8 weeks of sham-CPAP in patients with moderate to severe Sleep Disordered Breathing (SDB) and impaired glucose tolerance. A rigorous assessment of metabolic responses to SDB treatment in this group is of great clinical significance because this sample is at high risk for developing diabetes. The paradigm shift of CPAP as a mode of prevention can affect clinical practice in the fields of both primary care and sleep medicine.
Detailed description
There are few controlled studies that address whether treatment of sleep apnea improves glucose tolerance. This is a randomized, double-blind cross-over study of subjects with sleep apnea (apnea hypopnea index \> 15) and impaired glucose tolerance. Subjects will be randomized to 8 weeks of Continuous Positive Airway Pressure (CPAP) or sham-CPAP, followed by the alternate therapy after a one month wash-out. After each treatment, subjects will undergo 2-hour oral glucose tolerance testing, polysomnography, actigraphy, and measurements of indices of glucose control. The investigators intend to analyze the changes in glucose metabolism and insulin sensitivity in patients with sleep apnea with CPAP intervention.
Interventions
DEVICEContinuous Positive Airway Pressure (CPAP) (Philips-Respironics RemStar Pro® CPAP)
Subjects will be randomized to 8 weeks of CPAP or sham-CPAP, followed by the alternate therapy after a one month wash-out.
DEVICESham-Continuous Positive Airway Pressure (CPAP) (Philips-Respironics)
Subjects will be randomized to 8 weeks of CPAP or sham-CPAP, followed by the alternate therapy after a one month wash-out.
Sponsors
National Institutes of Health (NIH)
National Center for Research Resources (NCRR)
Beth Israel Deaconess Medical Center
Case Western Reserve University
Brigham and Women's Hospital
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
TREATMENT
Masking
DOUBLE (Subject, Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to 75 Years
Healthy volunteers
No
Inclusion criteria
* moderate to severe Sleep Disordered Breathing (SDB) defined by an Apnea Hypopnea Index (AHI) \> 15 * had evidence of Impaired Glucose Tolerance (IGT) defined by the mean 2-hour Oral Glucose Tolerance Test (OGTT) glucose of \> 140 mg/dl calculated from the two 2-hour OGTTs performed within 3 days of each other
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Number of Subjects With Normalization of Impaired Glucose Tolerance (IGT) | 20 weeks | Number of subjects who experienced normalization of the mean 2-hour oral glucose tolerance test (OGTT) in the overall sample undergoing therapeutic CPAP vs. sham CPAP. (2-hour OGTT glucose\< 140 mg/dL) |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Mean and Standard Deviation of Glucose Indices After Therapeutic CPAP vs. Sham | 20 weeks | Reported values include: fasting glucose (mg/dL), 2 hour Oral Glucose Tolerance Test (OGTT) (mg/dL) |
| Mean and Standard Deviation of Insulin Indices After Therapeutic CPAP vs. Sham | 20 weeks | The data for fasting and 2 hour Insulin (iIU/dL) are presented according to therapeutic CPAP vs. Sham CPAP. |
| Mean and Standard Deviation of Indices of Insulin Resistance With Therapeutic CPAP vs. Sham | 20 weeks | Homeostasis Model Assessment-Insulin Resistance (HOMA-IR) with therapeutic CPAP vs. Sham CPAP |
| Mean and Standard Deviation of Insulin Sensitivity Index (ISI(0,120)) With Therapeutic CPAP vs. Sham | 20 Weeks | Insulin Sensitivity Index derived from the Gutt Index, uses the plasma glucose and insulin concentration from fasting (0 min) and 120-min samples from the OGTT, to calculate (Metabolic Clearance Rate)/log (Mean Serum Insulin). The range of possible values is based on the subset ranges of fasting and oral glucose tolerance test (OGTT) insulin and fasting and OGTT glucose, which calculate to be a range of 1.6 to 206.8. An increase in the ISI (0,120) indicates an improvement in the insulin sensitivity. |
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Continuous Positive Airway Pressure (CPAP) | 25 |
| Sham-Continuous Positive Airway Pressure | 25 |
| Total | 50 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Period 1, 2 Months | Physician Decision | 1 | 0 |
Baseline characteristics
| Characteristic | Sham-Continuous Positive Airway Pressure | Continuous Positive Airway Pressure (CPAP) | Total |
|---|---|---|---|
| Age, Categorical <=18 years | 0 Participants | 0 Participants | 0 Participants |
| Age, Categorical >=65 years | 2 Participants | 3 Participants | 5 Participants |
| Age, Categorical Between 18 and 65 years | 23 Participants | 22 Participants | 45 Participants |
| Age, Continuous | 53 years STANDARD_DEVIATION 8 | 54 years STANDARD_DEVIATION 10 | 53.6 years STANDARD_DEVIATION 9.9 |
| Region of Enrollment United States | 25 participants | 25 participants | 50 participants |
| Sex: Female, Male Female | 15 Participants | 14 Participants | 29 Participants |
| Sex: Female, Male Male | 10 Participants | 11 Participants | 21 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 12 / 50 | 7 / 50 |
| serious Total, serious adverse events | 1 / 50 | 0 / 50 |
Outcome results
Primary
Number of Subjects With Normalization of Impaired Glucose Tolerance (IGT)
Number of subjects who experienced normalization of the mean 2-hour oral glucose tolerance test (OGTT) in the overall sample undergoing therapeutic CPAP vs. sham CPAP. (2-hour OGTT glucose\< 140 mg/dL)
Time frame: 20 weeks
Population: All available oral glucose tolerance test data was analyzed for the total sample.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Therapeutic CPAP | Number of Subjects With Normalization of Impaired Glucose Tolerance (IGT) | 7 subjects |
| Sham CPAP | Number of Subjects With Normalization of Impaired Glucose Tolerance (IGT) | 5 subjects |
Comparison: Based on an intent to treat approach a Generalized Estimating Equation (GEE) was used to estimate the effect of therapy (CPAP or Sham) on the odds of normalization of Impaired Glucose Tolerance (IGT). This model provides an estimate of the odds ratio of normalizing the 2-hour oral glucose tolerance test (OGTT) with CPAP compared with Sham-CPAP.p-value: 0.5795% CI: [0.52, 3.24]Generalized Estimating Equation
Secondary
Mean and Standard Deviation of Glucose Indices After Therapeutic CPAP vs. Sham
Reported values include: fasting glucose (mg/dL), 2 hour Oral Glucose Tolerance Test (OGTT) (mg/dL)
Time frame: 20 weeks
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Therapeutic CPAP | Mean and Standard Deviation of Glucose Indices After Therapeutic CPAP vs. Sham | Fasting Glucose (mg/dL), CPAP | 106.4 mg/dL | Standard Deviation 12.4 |
| Therapeutic CPAP | Mean and Standard Deviation of Glucose Indices After Therapeutic CPAP vs. Sham | 2-Hour OGTT (mg/dL), CPAP | 171.1 mg/dL | Standard Deviation 43.5 |
| Sham CPAP | Mean and Standard Deviation of Glucose Indices After Therapeutic CPAP vs. Sham | Fasting Glucose (mg/dL), CPAP | 105.6 mg/dL | Standard Deviation 13.3 |
| Sham CPAP | Mean and Standard Deviation of Glucose Indices After Therapeutic CPAP vs. Sham | 2-Hour OGTT (mg/dL), CPAP | 178.2 mg/dL | Standard Deviation 48.5 |
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and measures of glucose indices, in this case fasting glucose, between CPAP and sham-CPAP.p-value: 0.3895% CI: [-1.2, 3]Regression, Linear
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between measures of glucose indices, in this case, 2 hour OGTT, and therapeutic CPAP vs. Sham.p-value: 0.1195% CI: [-16.3, 1.7]Regression, Linear
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and measures of glucose indices, in this case fasting glucose, between CPAP and sham-CPAP, stratified for the most severe sleep apnea group (Apnea-Hypopnea Index (AHI) \>=30) in order to assess whether the differences in the effect of CPAP on outcomes were observed according to sleep apnea severity. N=25.p-value: 0.8695% CI: [-3.3, 2.7]Regression, Linear
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and measures of glucose indices, in this case 2-hour oral glucose tolerance test (OGTT) glucose, between CPAP and sham-CPAP, stratified for the most severe sleep apnea group (Apnea-Hypopnea Index (AHI) \>=30) in order to assess whether the differences in the effect of CPAP on outcomes were observed according to sleep apnea severity. N= 25.p-value: 0.0895% CI: [-24.6, 1.3]Regression, Linear
Secondary
Mean and Standard Deviation of Indices of Insulin Resistance With Therapeutic CPAP vs. Sham
Homeostasis Model Assessment-Insulin Resistance (HOMA-IR) with therapeutic CPAP vs. Sham CPAP
Time frame: 20 weeks
Population: All subjects with available fasting insulin, glucose measurements
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Therapeutic CPAP | Mean and Standard Deviation of Indices of Insulin Resistance With Therapeutic CPAP vs. Sham | 4.4 percentage of beta cell function | Standard Deviation 2.6 |
| Sham CPAP | Mean and Standard Deviation of Indices of Insulin Resistance With Therapeutic CPAP vs. Sham | 4.6 percentage of beta cell function | Standard Deviation 3 |
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and Homeostasis Model Assessment-Insulin Resistance (HOMA-IR), between CPAP and sham-CPAP.p-value: 0.1895% CI: [-17.6, 3.8]Regression, Linear
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and measures of insulin resistance, in this case Homeostasis Model Assessment-Insulin Resistance (HOMA-IR), between CPAP and sham-CPAP, stratified for the most severe sleep apnea group (Apnea-Hypopnea Index (AHI) \>=30) in order to assess whether the differences in the effect of CPAP on outcomes were observed according to sleep apnea severity. N=23.p-value: 0.0895% CI: [-27.5, 1.8]Regression, Linear
Secondary
Mean and Standard Deviation of Insulin Indices After Therapeutic CPAP vs. Sham
The data for fasting and 2 hour Insulin (iIU/dL) are presented according to therapeutic CPAP vs. Sham CPAP.
Time frame: 20 weeks
Population: All subjects with available fasting and and 2-hour insulin measurements are included.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Therapeutic CPAP | Mean and Standard Deviation of Insulin Indices After Therapeutic CPAP vs. Sham | Fasting Insulin (iIU/dL) | 16.4 iIU/dL | Standard Deviation 9 |
| Therapeutic CPAP | Mean and Standard Deviation of Insulin Indices After Therapeutic CPAP vs. Sham | 2-Hour OGTT Insulin (iU/dL) | 110.1 iIU/dL | Standard Deviation 60.7 |
| Sham CPAP | Mean and Standard Deviation of Insulin Indices After Therapeutic CPAP vs. Sham | Fasting Insulin (iIU/dL) | 17.2 iIU/dL | Standard Deviation 10.1 |
| Sham CPAP | Mean and Standard Deviation of Insulin Indices After Therapeutic CPAP vs. Sham | 2-Hour OGTT Insulin (iU/dL) | 114.3 iIU/dL | Standard Deviation 68.1 |
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and measures of insulin indices, in this case, fasting insulin, between CPAP and sham-CPAP.p-value: 0.1495% CI: [-3.4, 0.5]Regression, Linear
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and measures of insulin indices, in this case, 2-hour oral glucose tolerance test (OGTT) insulin, between CPAP and sham-CPAP.p-value: 0.1295% CI: [-23.3, 2.8]Regression, Linear
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and measures of insulin indices, in this case fasting insulin, between CPAP and sham-CPAP, stratified for the most severe sleep apnea group (Apnea-Hypopnea Index (AHI) \>=30) in order to assess whether the differences in the effect of CPAP on outcomes were observed according to sleep apnea severity. N=23.p-value: 0.195% CI: [-5.2, 0.5]Regression, Linear
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and measures of insulin indices, in this case oral glucose tolerance test (OGTT) insulin, between CPAP and sham-CPAP, stratified for the most severe sleep apnea group (Apnea-Hypopnea Index (AHI) \>=30) in order to assess whether the differences in the effect of CPAP on outcomes were observed according to sleep apnea severity. N=23.p-value: 0.00295% CI: [-46.5, -10.9]Regression, Linear
Secondary
Mean and Standard Deviation of Insulin Sensitivity Index (ISI(0,120)) With Therapeutic CPAP vs. Sham
Insulin Sensitivity Index derived from the Gutt Index, uses the plasma glucose and insulin concentration from fasting (0 min) and 120-min samples from the OGTT, to calculate (Metabolic Clearance Rate)/log (Mean Serum Insulin). The range of possible values is based on the subset ranges of fasting and oral glucose tolerance test (OGTT) insulin and fasting and OGTT glucose, which calculate to be a range of 1.6 to 206.8. An increase in the ISI (0,120) indicates an improvement in the insulin sensitivity.
Time frame: 20 Weeks
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Therapeutic CPAP | Mean and Standard Deviation of Insulin Sensitivity Index (ISI(0,120)) With Therapeutic CPAP vs. Sham | 12 units on a scale | Standard Deviation 3.3 |
| Sham CPAP | Mean and Standard Deviation of Insulin Sensitivity Index (ISI(0,120)) With Therapeutic CPAP vs. Sham | 12 units on a scale | Standard Deviation 3.8 |
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and the Insulin Sensivity Index, between CPAP and sham-CPAP.p-value: 0.295% CI: [-2, 9.8]Regression, Linear
Comparison: Based on an intent-to-treat approach, a linear mixed effect model was used to estimate the association between therapy and the Insulin Sensitivity Index (ISI(0,120)), between CPAP and sham-CPAP, stratified for the most severe sleep apnea group (Apnea-Hypopnea Index (AHI) \>=30) in order to assess whether the differences in the effect of CPAP on outcomes were observed according to sleep apnea severity. N=23.p-value: 0.00295% CI: [5.2, 22.1]Regression, Linear