We reviewed the evidence regarding different drugs for managing cystic fibrosis-related diabetes (CFRD).
Cystic fibrosis (CF) is a common life-limiting genetic condition, mainly causing damage to the lungs and pancreas. The pancreas makes insulin, a hormone which the body needs to take sugar into the cells so it can be converted into energy. In up to 90% of people with CF, problems with the pancreas and digestive enzymes mean nutrients are not properly absorbed. People with CF need high-calorie and high-protein diets to provide enough energy to maintain muscle mass so as to support their breathing and make up for breathing difficulties from disease-damaged lungs. It is important for people with CF to turn sugar into energy efficiently. This is especially challenging for people with CFRD because damage to the pancreas affects insulin production and release. Inflammation seen in CF can reduce insulin production and lessen its effect by causing insulin resistance. Increased life expectancy for people with CF means that 50% of adults with CF are likely to develop CFRD.
We wanted to assess different treatments for CFRD to limit any related decline in health. These include artificial sources of insulin (like long-acting glargine or short-acting protamine insulin), the control of natural hormones that stimulate insulin release and medications that enhance a person’s own insulin release or affect insulin resistance.
The evidence is current to: 10 September 2020.
We included four trials (145 adults). Two trials compared insulin to either no treatment for two months (seven participants) or to a placebo (dummy drug with no active medication) for one year (61 participants). These trials, along with a third trial lasting two years (67 adults), also compared insulin (given via a syringe) to repaglinide tablets. The fourth trial (20 participants) compared the long-acting insulin glargine to short-term neutral protamine Hagedorn insulin over a 12-week period.
We found results for our primary outcomes of blood glucose levels, lung function and nutritional status and our secondary outcomes of pulmonary exacerbations (flare up of disease), complications and quality of life. We could only analyse limited results and did not find evidence that showed that any one treatment was better than another. Our analysis did not show that either insulin or repaglinide made any difference to any of our outcomes compared to placebo or no treatment at any time point. Similarly, most of our results did not show whether insulin or repaglinide was better, but the evidence was a little stronger that there was no difference in lung function, nutritional status or blood glucose levels at 12 months. Similarly, the evidence was not strong enough to show if there is any difference between glargine and neutral protamine Hagedorn insulin for any of our outcome measures. None of the trials reported if there were any deaths.
Given the treatment burden already experienced by people with CF, tablets may be a welcome alternative to injecting insulin. Longer-term trials with larger numbers of people are still needed to see how controlling CFRD affects lung function. Investigators should also assess how many people keep to their treatment plan when comparing insulin to tablets. Research should look at using the drugs by themselves or together with insulin to enhance insulin action, especially those agents with additional anti-inflammatory potential.
Quality of the evidence
We had some concerns about the quality of the evidence; mainly that there would be bias because clinicians probably knew in advance which treatment group the next person was going to be put into (e.g. healthier participants might be put into one group to show better results for that treatment) or bias from people taking part knowing which treatment group they were in (e.g. insulin via a syringe or repaglinide as a tablet). There may also be bias because results may only be selectively reported. We had some concerns that the analysis would not be precise due to small numbers of participants and low event rates. Finally, there may be bias in the results as the amounts of insulin and repaglinide given were not comparable.
This review has not found any conclusive evidence that any agent has a distinct advantage over another in controlling hyperglycemia or the clinical outcomes associated with CFRD. Given the treatment burden already experienced by people with cystic fibrosis, oral therapy may be a viable treatment option.
While some cystic fibrosis centers use oral medications to help control diabetes, the Cystic Fibrosis Foundation (USA) clinical practice guidelines support the use of insulin therapy and this remains the most widely-used treatment method. Randomized controlled trials specifically related to controlling diabetes and its impact on the course of pulmonary disease process in cystic fibrosis continue to be a high priority. Specifically, investigators should evaluate adherence to different therapies and also whether there is benefit in using additional hypoglycemic agents as well as the newly approved incretin mimics. Agents that potentiate insulin action, especially agents with additional anti-inflammatory potential should also be further investigated as adjuvant therapy to insulin.
The Cystic Fibrosis Foundation recommends both short-term and long-acting insulin therapy when cystic fibrosis-related diabetes (CFRD) has been diagnosed. Diagnosis is based on: an elevated fasting blood glucose level greater than 6.94 mmol/L (125 mg/dL); or oral glucose tolerance tests greater than 11.11 mmol/L (200 mg/dL) at two hours; or symptomatic diabetes for random glucose levels greater than 11.11 mmol/L (200 mg/dL); or glycated hemoglobin levels of at least 6.5%. This is an update of a previously published review.
To establish the effectiveness of insulin and oral agents for managing diabetes in people with cystic fibrosis in relation to blood sugar levels, lung function and weight management.
We searched the Cochrane Cystic Fibrosis and Genetic Disorders Group's Trials Register comprising references identified from comprehensive electronic database searches and handsearches of relevant journals and abstract books of conference proceedings. We also handsearched abstracts from pulmonary symposia and the North American Cystic Fibrosis Conferences.
Date of most recent register search: 10 September 2020.
We searched online trials registries; date of most recent searches: 21 March 2020.
Randomized controlled trials comparing all methods of pharmacological diabetes therapy in people with diagnosed CFRD.
Two authors independently extracted data and assessed the risk of bias in the included studies. Authors also used GRADE to assess the quality of the evidence.
The searches identified 29 trials (45 references). Four included trials provide results: one short-term single-center cross-over trial (seven adults) comparing insulin with oral repaglinide and no medication in adults with CFRD and normal fasting glucose; one long-term multicenter trial (61 adults with CFRD) comparing insulin with oral repaglinide and placebo; one long-term multicenter trial (67 adults) comparing insulin with oral repaglinide; and one 12-week single-center cross-over trial (20 adults) comparing the long-acting insulin glargine to short-term neutral protamine Hagedorn insulin. Two ongoing trials of newly approved incretin mimics have been noted for possible future inclusion.
Downgrading of the quality of the evidence was mainly due to risks of bias across all domains, but particularly due to concerns surrounding allocation concealment and selective reporting. There were also some concerns due to imprecision from small sample sizes and low event rates. Finally, there may be some bias due to the amounts of insulin and repaglinide given not being comparable.
Data from one trial comparing insulin to placebo (39 participants) did not show any difference between groups for the primary outcomes of blood glucose levels (very low-quality evidence), lung function (low-quality evidence) or nutritional status (low-quality evidence). Similarly, no differences between groups were seen for the secondary outcomes of number of hypoglycemic episodes (low-quality evidence), secondary infection complications or quality of life (QoL). These results were mirrored in the narrative reports for the second trial in this comparison (seven participants).
Data from the one-year trial comparing repaglinide to placebo (38 participants), showed no differences between groups for the primary outcomes of blood glucose levels (very low-quality evidence), lung function (low-quality evidence) and nutritional status (low-quality evidence). Also, no differences were seen between groups for the secondary outcomes of number of hypoglycemic episodes (low-quality evidence), secondary infection complications or QoL. These findings were mirrored in the narrative reports for the second trial (n = 7) in this comparison.
Three trials compared insulin to repaglinide (119 participants). Data from one trial (n = 67) showed no difference in blood glucose levels at either 12 months (high-quality evidence) or 24 months; narrative reports from one trial (45 participants) reported no difference between groups, but the second trial (7 participants) reported a beneficial effect of insulin over repaglinide. Two trials (112 participants) found no difference between insulin and repaglinide in lung function or nutritional status (moderate-quality evidence). Two trials (56 participants) reported no difference in the number of hypoglycemic episodes (low-quality evidence). One trial (45 participants) reported no difference between groups in secondary infections and cystic fibrosis QoL.
The single trial comparing glargine to neutral protamine Hagedorn insulin did not report directly on the review's primary outcomes, but did report no differences between groups in post-prandial glucose values and weight; neither group reported infectious complications. There was no difference in episodes of hypoglycemia (very low-quality evidence) and while there was no difference reported in QoL, all participants opted to continue treatment with glargine after the trial was completed.
Mortality was not reported by any trial in any comparison, but death was not given as a reason for withdrawal in any trial.