Do you recognize this patient?  I have seen him in my office, or a person just like him. The Center for Disease Control (CDC, HHS) estimates that 57 million people in the U.S. have pre-diabetes. (1) If you look, you can probably recognize a person like this in every primary care office in the U.S. Unfortunately we still don’t have an evidence-based answer for the best therapeutic course for treatment of pre-diabetes.

In the absence of an established and proven treatment, therapy should be grounded on the best available evidence. In the responses to Case #1, three national experts have provided valuable insight into the best available evidence, and have given well-informed and differing perspectives on how to treat this particular presentation of pre-diabetes. Each description provides a reasonable and balanced approach that is consistent with available clinical evidence, and potentially represents an example of high quality best practice. When I read the discussions I have to admit that each author had me convinced that they had the best answer.  Lifestyle – of course! Metformin – sure enough! TZD – that’s for me!  Of course, from somewhere in my skeptical primary care heart I couldn’t help but wonder how we ever got to this point in the first place.

Implicit to the diagnoses of pre-diabetes is the idea you need to screen individuals at risk. This patient has a positive family history, is 45 years old, has a rising A1c, an elevated FPG, and elevated BMI, low HDL, high triglycerides, an elevated risk due to race, acanthosis nigracans, and has come into the office asking to be screened.  By American Diabetes Association criteria, this patient has a litany of risk factors that clearly makes him a candidate for screening (2). If the screening test is negative, the test should be repeated every one to two years.  But the evidence-based answer is not so clear.  With a systolic blood pressure under 135 mm Hg, the United States Preventive Services Task Force (USPSTF) still rates screening for diabetes with an “I” for insufficient evidence (3), and reports that a strict interpretation of the literature reveals  “There is inadequate evidence that early diabetes control as a result of screening provides an incremental benefit for microvascular clinical outcomes compared with initiating treatment after clinical diagnosis.”  Although the USPSTF may be the last group in town to be convinced of the value of many potential medical advances, the consequence is that many insurance companies still hesitate to consider diabetes screening as a legitimate “preventive” service. On the other hand, most insurance companies will pay for diagnostic blood tests if they are based upon the clinical suspicion of diabetes. Medicare will pay for up to 2 diabetes screens per year if a person has high blood pressure, dyslipidemia, obesity, or a history of high FPG, or if a person meets two of the following criteria: over 65, overweight, family history (first degree relative), or a history of gestational diabetes (4).  In the end, the ADA provides sound advice by recommending to patients, “It is best to consult your physician and health insurance representative with specific coverage questions.”(5)

So did we really need to screen this patient?  Despite a cautious USPSTF approach, this patient is clearly at high risk for diabetes and prediabetes. Although it is clear that we need to learn more about degrees of risk, population prevalence, and about the costs and benefits of screening specific populations, in this case it is equally clear that the best practice is to check for a diagnosis of diabetes or prediabetes and to treat the progressing glycemic abnormality. In my honest opinion, as long as the adverse event rate from treatment is low, early treatment has just got be better than late treatment.  After all, this patient came in because he is worried about ending up like his brother.  So this high risk patient got screened. Now what do we do with the result? 

With potentially millions of people needing to be treated, lifestyle modification initially seems the most acceptable. It would appear to be the best bet for having a low rate of adverse events, and to be a good investment for the average patient. Or is it?  Lifestyle modification has a relatively low success rate, and is resource intensive. Effective behavioral modification, at least as done by the DPP, requires educational support, staff, and provider time. Supporting an effective behavioral modification program is challenging in many primary care offices.   Although the DPP proved that lifestyle changes provide the greatest risk reduction, the DPP provided 16 intensive sessions for every patient. Most providers are not set up to provide this support, and many insurance companies don’t pay for intensive behavioral modification programs. As the first author says, "it is time that we take a closer look at reaching out to local community services, and helping to establish effective programs." I think my community YWCA would join in.  It is time that primary care practices better support behavioral change. Of course it is the right choice for this patient, but whether it is a practical solution depends on the person, the practice, and perhaps the community.  

Metformin seems a good compromise. Of course, the DPP was an efficacy study, and was never intended to evaluate the effectiveness of large scale metformin therapy across a primary care population.  Demonstrating that prescribing metformin lowers risk of developing diabetes significantly more than doing nothing does not mean that metformin should be the treatment of choice.  In the community, the frequency of preventing a diabetes complication must be balanced against the frequency of serious adverse events caused by the medication. In addition, the risk and benefits of one solution should be weighed against those of other available solutions. Although metformin can delay or prevent the onset of diabetes, how does it compare to the safety and effectiveness of other potential treatments?  Most adverse events from metformin are mild, but serious events do occur. When prescribed across a large and relatively low risk population, the frequency of potential problems following unplanned procedures, cardiac events, or among patients with liver or renal dysfunction provides reason for a cautious approach.  Still, metformin is pretty easy to take, and gets my vote as second best or when lifestyle modification clearly isn’t effective.

The TZD option is the most modern approach and may be the most enlightened.    But what about those adverse side effects (6)? [ref] Pedal edema, weight gain, fluid retention, evolving congestive heart failure, and then there is that vague but persistent nagging about an association with myocardial infarcts. (7)   Although I am an advocate for early treatment of glycemic dysfunction, the potential problems with adverse events seem real and substantial.    Although TZD or even basal insulin may be the best therapy for some individuals, the people in whom risk increases more than benefit have not been well defined. Low dosages may help avoid adverse events.   Of course, if I am comfortable prescribing an expensive drug for a non-FDA approved indication, then I would more likely consider the incretins because of the better side effect profile.  With animal studies suggesting enhanced beta cell preservation,  DPP-4 inhibitors are both well tolerated and target the postprandial hyperglycemia often seen with early disease (8).

Providing the best medical care requires a balance between rapidly adopting promising new treatments and cautiously waiting for enough evidence to ensure long term safety.  In the end, good ideas sometimes turn out badly, and we are reminded of the importance of new comparative effectiveness research. Until community-based, patient-centered clinical research provides us with better comparisons of available treatments, decisions about the best treatment for prediabetes remain somewhere between what we think and what we know. Those decisions are still best addressed individually in a discussion between a patient and their doctor.

REFERENCES

1. Centers for Disease Control and Prevention. National diabetes fact sheet: general information and national estimates on diabetes in the United States, 2007. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, 2008.
2. American Diabetes Association, Standards of Medical Care in Diabetes , Diabetes Care, v32 (supp 1)S13-S61, 2009
3. Norris SL, Kansagara D, Bougatsos C, et al. Screening Adults for Type 2 Diabetes: A Review of the Evidence for the U.S. Preventive Services Task Force. AHRQ Publication No. 08-05116-EF-4, June 2008. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/clinic/uspstf08/type2/type2art.htm Medicare payment diabetes screening
4. Center for Medicare and Medicaid Services, Medicare and You 2010, US Dept Health and Human Services, page 29.( http://www.medicare.gov/Publications/Pubs/pdf/10050.pdf)
5. http://www.diabetes.org/diabetes-basics/prevention/pre-diabetes/pre-diab...
6. Norris SL, Carson S. Drug class review: Thiazolidinediones. 2008. http://www.ohsu.edu/drugeffectiveness/reports/final.cfmCardiac effects of tzd
7. David J. Graham; Rita Ouellet-Hellstrom; Thomas E. MaCurdy; Farzana Ali; Christopher Sholley; Christopher Worrall; Jeffrey A. Kelman, Risk of Acute Myocardial Infarction, Stroke, Heart Failure, and Death in Elderly Medicare Patients Treated With Rosiglitazone or Pioglitazone. JAMA. 2010;304(4):411-418.
8. Incretin references on the website to the supportive work presented by basic scientists