In case you think you're confused . . .

I was as confused as hell today when I first read this in the New York Times, then I checked out this week's NEJM and read this (abstract only for you uncredentialed scum).

Item One:

For decades, researchers believed that if people with diabetes lowered their blood sugar to normal levels, they would no longer be at high risk of dying from heart disease. But a major federal study of more than 10,000 middle-aged and older people with Type 2 diabetes has found that lowering blood sugar actually increased their risk of death, researchers reported Wednesday. . . .

Among the study participants who were randomly assigned to get their blood sugar levels to nearly normal, there were 54 more deaths than in the group whose levels were less rigidly controlled. The patients were in the study for an average of four years when investigators called a halt to the intensive blood sugar lowering and put all of them on the less intense regimen.

Item Two:

Methods In the Steno-2 Study, we randomly assigned 160 patients with type 2 diabetes and persistent microalbuminuria to receive either intensive therapy or conventional therapy; the mean treatment period was 7.8 years. Patients were subsequently followed observationally for a mean of 5.5 years, until December 31, 2006. The primary end point at 13.3 years of follow-up was the time to death from any cause.

Results Twenty-four patients in the intensive-therapy group died, as compared with 40 in the conventional-therapy group (hazard ratio, 0.54; 95% confidence interval [CI], 0.32 to 0.89; P=0.02). Intensive therapy was associated with a lower risk of death from cardiovascular causes (hazard ratio, 0.43; 95% CI, 0.19 to 0.94; P=0.04) and of cardiovascular events (hazard ratio, 0.41; 95% CI, 0.25 to 0.67; P<0.001). One patient in the intensive-therapy group had progression to end-stage renal disease, as compared with six patients in the conventional-therapy group (P=0.04). Fewer patients in the intensive-therapy group required retinal photocoagulation (relative risk, 0.45; 95% CI, 0.23 to 0.86; P=0.02). Few major side effects were reported.

Conclusions In at-risk patients with type 2 diabetes, intensive intervention with multiple drug combinations and behavior modification had sustained beneficial effects with respect to vascular complications and on rates of death from any cause and from cardiovascular causes.

Before we retire to bedlam, I should note that there were differences between the two interventions. Although I don't have access to full information about the study that was suspended, it appears that the people in the intensive intervention group regularly injected insulin; the people in the steno-2 study did not. Maybe that had something to do with it. Also, the steno-2 study didn't make a fetish out of achieving a particular glycemic control target.

Anyway, even if we're left confused about how to manage Type 2 diabetes, we can draw an important lesson here: never depend on surrogate end points.

The false logic was that the definition of diabetes is high blood sugar, ergo, eliminate the high blood sugar and you eliminate the bad stuff that diabetes causes. Not necessarily so, it turns out. And this is the case with many drugs that are widely prescribed, that have received FDA approval not because they are shown to improve health or longevity, but because they influence some metabolic pathway or the level of some substance in the blood which is believed to be associated with a disease. We need a radical change in FDA policy to make approval of drugs based on surrogate endpoints much more difficult.

And, if you have diabetes, don't be terribly confused. It is still true, and absolutely supported and not controverted by anything we learned today, that losing weight, excercising more, and eating right will save your eyesight, your kidneys, your lower extremities, and your life. That is 100% operative.

Update -- Some Clarity: The intensive arm of the steno-2 study didn't just focus on blood sugar, it focused on blood pressure and cholesterol as well. The bad news: they weren't successful in getting people to make those all-important "lifestyle" changes, which undoubtedly would have benefited them even more. My main point -- that the surrogate endpoint of glycemic control turns out not to be a good marker of successful diabetes management -- is just reinforced by this analysis.