1. Home>
2. Specialties>
3. General Medicine>
4. Feature

How Much Evidence Do We Need to Change Practices in Which We Firmly Believe?

Enough already! Randomized trials show that tight glucose control in patients with long-standing type 2 diabetes isn't beneficial.

Should the glycosylated hemoglobin (HbA_1c) level goal in patients with long-standing type 2 diabetes be 7%? 6.5%? Lower? Although many clinicians believe in tight control for patients with type 2 diabetes, recent studies suggest that this practice is not beneficial. Several recently published commentaries cite evidence that challenges current beliefs and practices.

In the first major trial (done in the 1960s) of tight glucose control in patients with type 2 diabetes, oral glucose-lowering agents were associated with higher cardiovascular mortality and no differences in microvascular complications compared with placebo.¹ Insulin also was not associated with clinical benefit.

In three recent large randomized trials (ACCORD,² ADVANCE,³ and VADT⁴), tight control in patients with long-standing type 2 diabetes did not lower overall mortality, cardiovascular-related mortality, stroke, amputations, or even clinical (as opposed to surrogate) microvascular endpoints. Differences in specific outcomes in these trials might be related to different treatments or to duration of diabetes in participants. In some studies, fewer intensively treated patients reached composite outcomes (such as "any diabetes complications"), but the bulk of improvement was in nonclinical outcomes (e.g., incident albuminuria). Tight control was associated with severe hypoglycemia and weight gain. In the UKPDS study,⁵ published a decade ago, nonobese intensively treated participants with newly diagnosed type 2 diabetes were less likely to reach microvascular endpoints (including "need for photocoagulation," but not visual loss) but showed no difference in mortality (cardiovascular, diabetes-related, or all-cause) compared with nonobese control patients. Among obese participants, metformin alone lowered long-term mortality and myocardial infarction rate, but sulfonylureas and insulin did not; tight control did not lessen risk for microvascular complications. Metformin and sulfonylureas in combination were associated with excess diabetes-related deaths and all-cause mortality.

Because trials do not support tight control and because of the cost, burden, and harms associated with tight control, we should be emphasizing cardiovascular risk reduction (particularly control of blood pressure and cholesterol levels) and healthy lifestyles for patients with type 2 diabetes.⁶ Several groups of editorialists suggest aiming for HbA_1c levels of 7.0% or 7.5% in patients in whom this goal is achievable with one medication and adjusting this target for others based on symptoms, side effects, treatment burden, and patient values and preferences.^6,7,8 Commentary authors suggest that the HbA_1c goals for practice guidelines should not be <7% and that, to encourage individualized treatment, performance measures should set an upper limit (e.g., 9%) rather than a lower limit (e.g., <7%).⁷

Randomized trial results often are not available to answer important clinical questions. In this case, they are. We shouldn't ignore them. Many clinical trials are completed that show benefits, and much time passes, before new treatments are adopted; similarly, many trials that show lack of benefit, or even harm, might be required before clinicians abandon ineffective practices that have become routine. Haynes and Haynes ask, "What does it take to put an ugly fact through the heart of a beautiful hypothesis?" and they quote poetry: "The chains of habit are too weak to be felt until they are too strong to be broken."⁹ Social psychology literature suggests that people cling to belief even in the face of mountains of evidence to the contrary. But, as physicians and scientists, we should embrace change when new evidence consistently contradicts our prior beliefs and clinical practice.

— Richard Saitz, MD, MPH, FACP, FASAM

Published in Journal Watch General Medicine July 30, 2009

Citation(s):

1. Meinert CL et al. A study of the effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes: II. Mortality results. Diabetes 1970; 19:Suppl:789.

• Medline abstract (Free)

2. Gerstein HC et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008 Jun 12; 358:2545. (http://dx.doi.org/10.1056/NEJMoa0802743)

• Original article (Subscription may be required)
• Medline abstract (Free)

3. Patel A et al. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med 2008 Jun 12; 358:2560. (http://dx.doi.org/10.1056/NEJMoa0802987)

• Original article (Subscription may be required)
• Medline abstract (Free)

4. Duckworth W et al. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med 2009 Jan 8; 360:129. (http://dx.doi.org/10.1056/NEJMoa0808431)

• Original article (Subscription may be required)
• Medline abstract (Free)

5. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998 Sep 12; 352:837.

• Medline abstract (Free)

6. Montori VM and Fernández-Balsells M. Glycemic control in type 2 diabetes: Time for an evidence-based about-face? Ann Intern Med 2009 Jun 2; 150:803.

• Original article (Subscription may be required)
• Medline abstract (Free)

7. Lehman R and Krumholz HM. Tight control of blood glucose in long standing type 2 diabetes. BMJ 2009 Apr 18; 338:901. (http://dx.doi.org/10.1136/bmj.b800)

8. Havas S. The ACCORD trial and control of blood glucose level in type 2 diabetes mellitus: Time to challenge conventional wisdom. Arch Intern Med 2009 Jan 26; 169:150.

• Original article (Subscription may be required)
• Medline abstract (Free)

9. Haynes RB and Haynes GA. What does it take to put an ugly fact through the heart of a beautiful hypothesis? Evid Based Med 2009 Jun; 14:68.

• Original article (Subscription may be required)
• Medline abstract (Free)

Reader Remarks:

Read all Reader Remarks on this article

• Tight Glucose Control? The Glucose Control Memory!
  Dr.Ajay Kumar Khandal, 3 Aug 2009 2:55 AM EST
  Glycemic bar to be pressed to as low as possible to normal, around 6.5 and the same is also to... [more]
• Tight glucose control
  Helen E. Hilts, MD, 2 Aug 2009 4:40 PM EST
  All of those studies were done with a fairly high carbohydrate diet (130+ grams/day) which causes high glucose variablity, increases... [more]
• “Tight glycaemic control in people with long-established diabetes is not beneficial”; from single individual studies to pooled results.
  P Chamnan, MRC Epidemiology Unit, Cambridge, 1 Aug 2009 9:48 AM EST
  This is not the first time that evidence from observational and intervention studies shows conflicting results. Observational studies have repeatedly... [more]
• Tight blood glucose control in the type 2 diabetes management
  Davide M Carvalho, Dept. of Endocrinology Faculdade Medicina do Porto, 1 Aug 2009 4:44 AM EST
  I disagree with the statement that tight blood control is not a target in the type 2 diabetes treatment: The... [more]
• diabetes control
  Steven A. Crane, Grand River Gastroenterology, 31 Jul 2009 7:55 PM EST
  It is all about obesity, isn't it? Care to tell Congress (again)?
• Don't look to deep into these studies!
  RT Thrush, 31 Jul 2009 3:22 PM EST
  The reason why tight control is made out to be the bad guy is becuse of the side effects of... [more]
• Glycemic control and the Pulse Mass Index
  Prof. Enrique J. Sánchez-Delgado, MD, Hospital Metropolitano Vivian Pellas, Managua, Nicaragua, 1 Aug 2009 2:58 PM EST
  Glycemic Control, Drugs, Goals, and Pulse Mass Index in T2 Diabetes
  
  When trying to balance the pros and cons of... [more]
• Tight diabetes control
  Ronald E. Habros M.D., 31 Jul 2009 12:01 AM EST
  Hallalujah ! At last some common sense. Achieving HbA1c's of less than 7.0 is so difficult, frought with dreadful hypoglycemic... [more]
• Glucose control for type II diabetics
  Fred S Hahn, Serious Strength Studio, 31 Jul 2009 8:24 AM EST
  The reason why some research (and may I stress some) indicates that tight glucose control is not beneficial is because... [more]
• Diabetes control
  Neil Hall, 31 Jul 2009 9:16 AM EST
  At last someone has spoken the truth about the evidence on diabetes control. In this world emphasizing evidence based practice,... [more]
• Type 2 diabetes
  TS Ch'ang, 31 Jul 2009 10:12 AM EST
  It is about time to consider Long_Term (say >20 years) and geriatric (say 80-years or older) Diabetic patients as a... [more]
• Adequate control
  J.J. Wilhelm, 31 Jul 2009 1:48 PM EST
  Too tight of blood sugar control is bad too loose is worse.
  
  Perhaps we should be reminded of all the... [more]
• You tell 'em!
  Ronald L Hirsch, 30 Jul 2009 10:08 PM EST
  It is amazing how we ignore good data to continue to do what we "believe" is correct, spending huge amounts... [more]
• Scientific Logic
  Richard Feinman, SUNY Downstate Medical Center, 31 Jul 2009 8:54 AM EST
  If you give a drug to target HbA1c and there are side-effects, do you assume that you shouldn't target A1c?... [more]
• Tight glucose Control
  Roberta J Lilly, 4 Aug 2009 9:48 AM EST
  It is not the tight glucose control in Type II diabetes that is the enemy, it is trying to attain... [more]
• You First!
  Chris Trinkwasser, 10 Aug 2009 4:49 PM EST
  Whenever I read someone with an A1c below 5 stating that "diabetics" do not deserve an A1b below 7 I... [more]
• Reply from Journal Watch Author
  Richard Saitz, MD, Boston Medical Center; Journal Watch Deputy Editor, 12 Aug 2009 4:27 PM EST
  The many thoughtful comments written in response to my commentary (prompted by several recent editorials and trials) suggest that tight... [more]
• Dispassionate Discourse
  Anthony Musci, Intermountain Medical Center, 13 Aug 2009 1:14 PM EST
  I think that Dr. Saitz' perspective is fundamentally sound, but I wonder if the reaction might have been different had... [more]
• Glycemic control should be individualized
  Gauranga C. Dhar, Teacher, Bangladesh Institute of Family Medicine and Research, USTC., 17 Aug 2009 1:37 PM EST
  Key cornerstone strategy in the treatment of hyperglycemia is to avoid hypoglycemia. It is not hyperglycemia but hypoglycemia is the... [more]
• Tight control in elderly DM patients
  Hayward Zwerling, M.D., FACP, FACE, Lowell Diabetes and Endocrine Center, 19 Aug 2009 9:11 AM EST
  For diabetic patients >70-75 yo, I adjust DM Rx so that the HBA1c is approximately = age/10. This is moderated... [more]
• Cardiovascular Disease
  Chris Trinkwasser, 19 Aug 2009 8:26 AM EST
  I don't recall the source, but a cardiologist described Type 2 as "a cardiovascular disease sometimes associated with high blood... [more]
• Looking at the bigger picture while treating type 2 diabetes mellitus
  Sumeet Gupta, M M College Of Pharmacy,MULLANA(india), 25 Aug 2009 11:14 AM EST
  We congratulate the authors for bringing forth the bold observations.Truly the facts were already there,but were unnoticed and we were... [more]
• Tight control begs for tight individualization
  Ruy O Pantoja Filho, UNICEUMA - UNIVERSITY - BRAZIL, 8 Nov 2009 8:11 AM EST
  WGAS research is underscoring the obvious - people are different. Rigid conclusions in the face of ample pharmacogenomics differences call... [more]