Systematic review
Effect of new anti-diabetic medications in combination with metformin compared to sulfonylurea in combination with metformin in patients with type 2 diabetes
Systematic review
|Updated
This overview of reviews updates evidence comparing the effects of the currently recommended treatment for type 2 diabetes.
Key message
Type 2 diabetes is a serious, chronic disease that can increase the risk of death, heart attack, stroke and kidney disease. This overview of reviews updates evidence comparing the effects of the currently recommended treatment (sulfonylurea + metformin) to treatment with newer anti-diabetic mediations (DPP-4 inhibitors or GLP-1 analogues) added to metformin when neither life-style changes nor metformin alone are effective in reducing blood sugar levels.
- No conclusions can be drawn about the effect on mortality of adding DPP-4 inhibitor versus sulfonylurea to metformin because of few events and low quality evidence. There was no documentation for the effect on mortality of adding GLP-1 analogues versus sulfonylurea to metformin.
- We found no evidence on the effects on micro and macrovascular complications for either comparison.
- Results for intermediate outcomes (HbA1c change, weight, hypoglycaemic incidents) are mixed. Reduction in HbA1c is larger with sulfonylurea, but DPP-4 inhibitors and GLP-1 analogues result in more weight loss and a lower risk of a hypoglycaemic episode.
- All differences are small in magnitude and may be clinically unimportant.
- Sulfonylurea + metformin is substantially less costly than either DPP-4 inhibitors or GLP-1 analogues + metformin.
Summary
Background
Diabetes is a serious chronic disease, characterized by higher than normal levels of sugar (glucose) in the blood resulting from inadequate insulin production and/or ineffective use of insulin by the body (insulin resistance). Insulin is the hormone that regulates blood sugar. It enables cells in the body to absorb glucose from the blood and it reduces endogenous glucose production. Over time, an above normal blood glucose level (hyperglycaemia) can lead to increased risk of mortality and micro and macrovascular complications, including damage to the heart, blood vessels, kidneys, eyes and nerves. The costs to the individual and society of uncontrolled diabetes are substantial.
Type 2 diabetes, which is associated with excessive weight, poor diet and limited physical activity, accounts for approximately 80-90% of all diabetes and is a growing problem in Norway. Recent estimates suggest that from 300 000 – 350 000 individuals suffer from type 2 diabetes, approximately half of them undiagnosed.
Norwegian national diabetes treatment guidelines from 2009 recommend life-style changes, followed by the addition of metformin (an oral anti-diabetic medication). If metformin therapy is ineffective, the guidelines further recommend adding sulfonylurea, another anti-diabetic drug, or insulin to metformin. Two newer types of drugs, DPP-4 inhibitors and GPL-1 analogues, are only recommended if treatment with metformin + sulfonylurea is not tolerated, because at the time the guidelines were issued there was little clinical evidence about their effects. Recently, the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD) issued recommendations that suggested equal preference be given to sulfonylurea, pioglitazone, insulin, DPP-4 inhibitors and GLP-1 analogues in second-line treatment decisions.
Objective
To upgrade the evidence about the effect of DPP-4 inhibitors or GLP-1 analogues in combination with metformin compared to sulfonylurea in combination with metformin, among adults with type 2 diabetes inadequately controlled with metformin monotherapy.
Method
We have conducted this overview of reviews in accordance with the Handbook for the Norwegian Knowledge Center for the Health Services.
A systematic literature search of several databases was conducted in April 2013 and updated in February 2014. Two review authors reviewed all citations to identify relevant systematic reviews according to pre-specified criteria. Full text publications of potentially eligible references were retrieved, and we assessed the quality of the included systematic reviews. We extracted data from the included references using a pre-designed data recording form. Article selection was completed independently by two review authors and then compared. Data extraction and quality assessment was conducted by one of the review authors and then checked by one of the others.
We used the effect estimates reported in the included reviews. For assessment of individual trials included in the systematic reviews, we relied on assessments performed by the review authors. When such assessments were conducted using tools other than those recommended in the Handbook, we performed our own risk of bias assessments.
Results
Results are based on two recent, high quality systematic reviews. No conclusions can be drawn about differences in mortality for the comparison DPP-4 inhibitors + metformin versus sulfonylurea + metformin because of few events and low quality evidence. No mortality evidence was available for the comparison GLP-1 analogues + metformin versus sulfonylurea + metformin. There was no evidence on micro or macrovascular complications for either comparison.
Sulfonylurea + metformin results in greater reductions in HBA1c than DPP-4 inhibitors + metformin and similar reductions compared to GLP-1 analogues + metformin, but also leads to an increased risk of experiencing at least one hypoglycaemic incident and results in weight gain rather than weight loss when compared to metformin combined with either DPP-4 inhibitors or GLP-1 analogues. The evidence was generally of moderate to low quality.
Direct treatment costs differ substantially among the three treatment options. Medication costs are lowest for treatment with sulfonylurea + metformin (NOK 1217). DPP-4 inhibitors + metformin are approximately three to four times more and GLP-1 analogues + metformin approximately ten to fourteen times more expensive than sulfonylurea + metformin.
Discussion
We conducted a systematic search of the literature to find the most current systematic reviews comparing the effects of adding DPP-4 inhibitors or GPL-4 analogues versus sulfonylurea to metformin for treating type 2 diabetes inadequately controlled by metformin alone.
The results are limited by the lack of long-term data on important clinical outcomes (mortality, micro and macrovascular complications). To the extent that evidence exists, it is of low quality. Effect data for changes in HbA1c and weight is of moderate quality, but the quality of evidence on hypoglycaemic episodes is low because relatively few episodes occur and studies often fail to distinguish between serious and minor events. Although differences in outcomes exist among the comparisons, they do not uniformly support one treatment. In addition, the differences are relatively small in size and may not be clinically relevant.
Our estimates of economic costs only include the costs of medications since we assume that other direct costs, such as doctor visits and blood sugar testing materials, would be equivalent across treatment groups. To the extent that sulfonylurea + metformin results in a higher risk of hypoglycaemic incidents than DPP-4 inhibitors or GLP-1 analogues + metformin, the associated direct costs may increase. We did not have enough information to estimate the size of this potential effect, but it is not likely to reverse the substantial cost advantage of sulfonylurea.
Conclusion
The most recent systematic reviews still provide either no evidence or evidence of low quality about differences in important outcomes (mortality, micro and macrovascular complications) between DPP-4 inhibitors or GLP-1 analogues + metformin and sulfonylurea + metformin. Differences in intermediate outcomes (HbA1c change, weight, hypoglycaemic incidents) exist but do not uniformly support one treatment. The direct cost of sulfonylurea is substantially lower than either DPP-4 inhibitors or GPL-1 analogues when added to metformin.
Continuing research is needed on important long-term outcomes and safety.