Expert Blog

Whose Diabetes Treatment Recommendations are Correct?

Irl B. Hirsch, MD
Primary care physicians are expected to be experts on dozens of different disease states despite the fact that each medical problem continues to evolve, with new understanding of the disease itself and its treatments. Frankly, I see this as a near impossible task since as an endocrinologist I can barely keep up with diabetes (in reality I can’t). The numbers regarding the epidemic of diabetes are thrown around at the beginning of every television or radio news story or diabetes CME and barely result in any type of response due to the fact we are all aware of the problem and have almost become immune to the statistics. The most recent Center of Disease Control figure that 8.3% of the US population (almost 26 million people) is only preparing us for a double-digit diabetes prevalence.
How Can Busy Physicians Keep Up With the Research?
So as physicians, whether we're specialists or primary care providers, how do we keep up with the research, particularly the treatments which impact every provider who manages patients with diabetes? I have (half) joked with my endocrine fellows that the best way to stay informed with the latest in diabetes is to subscribe to the Wall Street Journal. In the last six weeks (January 17, 2012 to March 3, 2012) there have been 128 articles regarding diabetes.  Over the last few years, between all of the new research and pharmaceutical controversies, I am unaware of a better source of information, except possibly the New York Times. I find it actually sad that our best source for diabetes information is an on-line newspaper.
Is this the best way to stay current with diabetes therapies? Before the introduction of metformin in 1995, decisions on which class of drug to use and when to use it were relatively simple. We used sulfonylureas early in the course of type 2 diabetes, and when required we turned to insulin therapy. The controversy had to do with which of the sulfonylureas to use (first generation debates in the 1960s and 1970s, second generations in the 1980s and 1990s). To be fair, it wasn’t quite as simple as this due to the fall-out from the University Group Diabetes Program suggesting the short-acting sulfonylurea tolbutamide increased the risk of cardiovascular mortality compared to insulin therapy (12.7% vs. 6.2% respectively).1 The study results were debated for years due to several major design flaws.2, but to this day there is a “black box warning” for sulfonylureas due to the concerns raised by this study.
Today's Choices
Today, of course, there are many choices for physicians deciding which drug to prescribe for patients with type 2 diabetes. Most recently, the American College of Physicians (ACP) published a clinical practice guideline for the oral pharmacologic treatment of type 2 diabetes.3 Based on a systematic evidence review, the ACP recommends clinicians prescribe monotherapy with metformin for initial pharmacologic therapy “to treat most patients with type 2 diabetes.”3 Based on a GRADE system, this was considered a strong recommendation based on high-quality evidence.
At first glance, this does not seem controversial at all. After all, metformin is generally safe, effective, well tolerated, and generic (cheap!). The impact on weight and lipids are at worst neutral but more often beneficial. Perhaps more importantly, the United Kingdom Prospective Diabetes Study reported that metformin monotherapy resulted in an improvement in cardiovascular disease outcomes,4 although this finding needs to be repeated. Nevertheless, based on the same data, the American Diabetes Association (ADA) recommends metformin monotherapy when type 2 diabetes is diagnosed, along with lifestyle modification, independent of HbA1c levels.5 In fact, based on the results of the Diabetes Prevention Program, a separate consensus statement from the ADA6 suggested metformin therapy for prediabetes with inadequate results from diet and exercise.
So while both the ACP and ADA agree that metformin should be the first drug for the treatment of type 2 diabetes, the disagreement is with the timing of initiating therapy, with the ADA starting metformin before diabetes is actually diagnosed.
And that’s just for the first drug. What about the second agent to treat type 2 diabetes? Not surprisingly, there is more disagreement and more complexity/confusion. ACP suggests “a second agent to metformin…” when “monotherapy with metformin fails to control hyperglycemia.”3 While six combination therapies were considered, sulfonylureas were singled out due to their low cost, although the increase in side effects was also acknowledged. The ADA in their consensus statement5 provides two tiers of options, each with two choices. For the “well-validated” options, clinicians can choose metformin with either sulfonylureas or basal insulin. For those “less-well-validated,” options include metformin with pioglitazone or a GLP-1 agonist.
So What Should the Confused Clinician do?
One option is to become more bewildered and look at another set of recommendations! In this case, we will choose the recommendations from the American Association of Clinical Endocrinologists (AACE).7 Here, pharmacologic therapy is based on initial HbA1c level. For HbA1c levels between 6.5-7.5%, one of four monotherapies (metformin, thiazolidinediones, DPP-4 inhibitors, or alpha-glucosidase inhibitors) is recommended.  For HbA1c levels between 7.6 and 9.0%, dual therapy by adding a second drug (GLP-1, DPP-4 inhibitor, thiazolidinedione, sulfonylurea, or glinide) is suggested. If, after 2-3 months, there is not an adequate response, one of six different triple therapy regimens is recommended.
So where does that leave us? I’m an endocrinologist and a card-carrying member of ACP, ADA, and AACE, yet even I am overwhelmed by all of these recommendations. Whose recommendations are right, or at the very least whose is more correct than the other? And what about cost? Should I use a different set of recommendations for different payers? And what happens to my thinking when ADA publishes their new set of recommendations? Two important points come to mind.
First, I’m not convinced there is any importance to any of these recommendations. In a recent report of almost 256,000 patients with type 2 diabetes initiating monotherapy between 2006-2008, 35% did not receive metformin as first-line monotherapy.8While the authors voice concern about not following the ADA recommendations, I would argue these 35% of patients were in compliance with the AACE recommendations. The study noted that over time, sulfonylureas and thiazolidinediones were used less frequently, while DPP-4 inhibitors were used more often. While no one can disagree that this adds to the economic burden of type 2 diabetes management, it also points to the power of a strong marketing presence in addition to the fact we know that expert recommendations have little impact on how physicians treat diabetes.9
But there is a more important point. As of now, we don’t have data about which drug or drug combination works best, let alone which agent(s) provides the best protection against the vascular complications of diabetes. The more cynical among us might ask which drug(s) cause the least risk of cancer. While these questions won’t be answered for many years, at the very least the US Food and Drug Administration has mandated new agents to market undergo cardiovascular risk assessment with long-term post-marketing follow-up. Furthermore, the National Institute of Diabetes and Digestive and Kidney Disease (NIDDK) will be funding a large randomized study to better understand the relative effectiveness of five different medications when used in combination with metformin.10 The goal is to learn if the traditional strategy of introducing the drugs sequentially or initially in combination is most effective in maintaining glycemic goals over time. Perhaps the results of this study with the acronym GRADE (Glycemia Reduction Approaches in Diabetes: A Comparative Effectiveness Study) will bring more consistency to our current guidelines.
For the time being, I don’t believe any set of recommendations are perfect, but as always, I am curious to your thoughts on this complex topic.


2. Davidson MB. Diabetes Mellitus: Diagnosis and Treatment, vol 1, 2nd ed. New York, NY: John Wiley & Sons, Inc.; 1986;198-205.