Hypos in the News: More Bad Science

Two studies in this week's issue of JAMA looked at the relationship of hypos with mortality and dementia.

The first study received no play in the medical press, the other got a lot of coverage. The major difference between these studies as far as I can see is that the one that was ignored was much better designed and the one that got all the press is yet another poorly conceptualized study that will be used by insurance companies to deny patients the tools they need to normalize blood sugars.

Let's look at the studies.

The first was:

Relationship Between Spontaneous and Iatrogenic Hypoglycemia and Mortality in Patients Hospitalized With Acute Myocardial Infarction. Mikhail Kosiborod et al. JAMA. 2009;301(15):1556-1564.

What it did was look at

7820 patients were hospitalized with AMI [heart attack] and were hyperglycemic on admission (glucose level ≥140 mg/dL). Patients were stratified based on whether they developed a hypoglycemic event (random glucose level <60 mg/dL) during subsequent hospitalization. Logistic regression models were used to evaluate the association between hypoglycemia and in-hospital mortality within subgroups of patients who were and were not treated with insulin therapy.

What they found was this:

Among patients treated or not treated with insulin, those with hypoglycemia were older and had more comorbidity [i.e. other medical conditions]. Hypoglycemia was associated with increased mortality in patients not treated with insulin (18.4% [25/136] ... but not in those treated with insulin (10.4% [36/346]... Hypoglycemia was a predictor of higher mortality in patients who were not treated with insulin (odds ratio, 2.32 [95% confidence interval, 1.31-4.12] vs patients without hypoglycemia), but not in patients treated with insulin (odds ratio, 0.92 [95% confidence interval, 0.58-1.45] vs patients without hypoglycemia).

In short, this study finds that very sick older people experience serious hypos when hospitalized for heart attacks when not taking insulin. Even more importantly, the study found that though they are more likely to hypo if they are given insulin, these hypos do NOT make them more likely to die unlike the non-insulin-related hypos that do seem associated with higher mortality.

So now we know something new about hypos that most of us didn't know--very sick older people hypo when they're in poor shape.

Now on to the next study, the one that got all the press:

Hypoglycemic Episodes and Risk of Dementia in Older Patients With Type 2 Diabetes Mellitus. Rachel A. Whitmer, et al. JAMA. 2009;301(15):1565-1572.

Researchers in this study tracked hypoglycemic events from 1980-2002 using hospital discharge and emergency department diagnoses. "Cohort members with no prior diagnoses of dementia, mild cognitive impairment, or general memory complaints as of January 1, 2003, were followed up for a dementia diagnosis through January 15, 2007."

The researchers found:

At least 1 episode of hypoglycemia was diagnosed in 1465 patients (8.8%) and dementia was diagnosed in 1822 patients (11%) during follow-up; 250 patients had both dementia and at least 1 episode of hypoglycemia (16.95%).

Compared with patients with no hypoglycemia, patients with single or multiple episodes had a graded increase in risk with fully adjusted hazard ratios (HRs): for 1 episode (HR, 1.26; 95% confidence interval [CI], 1.10-1.49); 2 episodes (HR, 1.80; 95% CI, 1.37-2.36); and 3 or more episodes (HR, 1.94; 95% CI, 1.42-2.64). The attributable risk of dementia between individuals with and without a history of hypoglycemia was 2.39% per year (95% CI, 1.72%-3.01%).

The researchers conclude: Among older patients with type 2 diabetes, a history of severe hypoglycemic episodes was associated with a greater risk of dementia.

The press reported this study with headlines like this one from USA Today: Type 2 Diabetics more susceptible to dementia. Or the NYTimes's Study Finds Risk of Dementia Increases After Hypoglycemia.

But what is missed here is that the study found association, not causation. And this is an important point.

Owning a yacht is associated with earning a high income. Studies could easily reveal that yacht owners have a higher "risk" of earning a higher income than people who do not own yachts (HR, 6.94; 95% CI, 5.42-8.64). But buying a yacht will not increase your paycheck. Rich people are more likely to own yachts because being rich means they can afford them.

So when we find an association between severe hypos and dementia we have to ask whether that hypoglycemia causes dementia or whether it is caused by the brain deterioration that occurs as people develop dementia.

As the first study cited above suggests, old, sick people who were more likely to die were prone hypo even without using insulin, which may suggest this is the case. It is possible that very early changes in a brain that is suffering mini-strokes or growing plaques and tangles, changes too subtle to lead to a dementia diagnosis, affect the blood sugar "thermostat" functions of the brain.

In addition, we all know that people in the early stages of the deterioration that leads to dementia are forgetful and likely to a) forget they took their insulin and shoot a second dose or b) confuse their basal and fast acting insulin. Both of these can lead to severe hypos. Because of this, severe hypos may be a very early symptom of dementia not its cause.

But even without that explanation, people who a decade later may develop dementia may also be more prone to hypo long before their diagnosis because the family doctors who treat them (as is the case with most older people with Type 2 diabetes) may put them on "one size fits all" insulin regimens that don't match their fast acting insulin to their carb intake. Typically they will tell the patient to use X units per meal, without any attempt to explain that insulin dose should vary with carbohydrate intake.

These set insulin regimens require no effort on the part of the doctor which makes them attractive to overworked family doctors who do not get reimbursed for helping people with diabetes adjust their insulin. These regimens always prescribe intentionally low insulin doses to avoid hypo risk since insulin isn't matched to meals. As a result, patients using these inadequate insulin regimens usually run very high blood sugars--many studies show their A1cs ranging from 8-10%. So their long term exposure to high blood sugars may be contributing to the progress of their dementia. Even with the too-low doses, they are prone to hypo if they become sick, can't eat, and continue to inject their set insulin doses as doctors often do not explain to them that they should not use insulin if they can't eat. So it is possible that some of the hypos in this study are a marker of poor insulin prescribing practices leading to poor blood sugar control, and that this poor control is causing the dementia, not the hypos.

This isn't just my conclusion. It's the conclusion of Dr. Nir Barzilai, a distinguished endocrinologist from Albert Einstein Medical School (Dr. Bernstein's Alma Mater) who was quoted in Forbes' coverage of this story:

... the study looks at association only, and doesn't actually prove any cause-and-effect link between the two conditions, cautioned Dr. Nir Barzilai, director of the Institute for Aging Research at Albert Einstein College of Medicine and the Montefiore Hospital Diabetes Clinic in New York City.

"It could be fluctuation of glucose. We know that hyperglycemia [high blood sugar] is also very toxic to the cells. All those things cannot be dissected on a study like this," he said.

Further on Dr. Barzilai added:

But the whole picture is likely to be much more complicated, Barzilai said.

"The glucose concentrations in the brain are much, much lower than in the [rest of the body], and it takes it a long time to actually adjust if you change the peripheral glucose for the brain to have lower glucose," he explained. "Not only that, but the neurons in the brain are really not fed by glucose but by other metabolites. The rest of the body, when glucose goes down, will feel it. The brain is totally different story.

More Ammo for Those Who Don't Want to Pay for Insulin?

The worrisome thing about this study is that it comes hard on the heels of several other poorly designed and badly reported studies which are being interpreted (erroneously) as "proving" that tight control achieved using insulin is dangerous for people with Type 2 diabetes.

Most notably, too, all the studies showing problems with insulin are studies done in populations of elderly Type 2s with years of poor control behind them and advanced cardiovascular disease resulting from those years of poor control, but the results in this population are being extrapolated to apply to the huge and growing population of young people with Type 2 whose ages range from their teens on up.

Armed with these studies, insurers can argue that "evidence based medicine" suggests people with Type 2 should not use insulin. They are already arguing that it is "dangerous" to strive for an A1c lower than 7.0%, even though there are almost two decades of research showing that lowering blood sugar prevents the classic, horrifying diabetic complications: amputation, blindness and kidney failure.

Now the opponents of tight control can claim that "using insulin leads to hypos and we 'know' hypos cause dementia." And they will. Because insulin is expensive and insurers could boost their bottom line enormously by refusing to supply insulin to people with Type 2.

I don't know what we people with diabetes can do to stem the tide of poor research being used to deny us lifesaving treatment. It's getting worse with each passing week, and with the emphasis on "cost cutting" in medicine rising world-wide, it is quite possible that society will just let older Type 2s deteriorate "safely" because of the imagined risks of tight control.

But I do have to wonder what's wrong with JAMA and NEJM that their "peers" are reviewing and recommending publication of studies like these without demanding that the summaries make it clear--and that the media report--the real findings of these studies, rather than the misrepresentations that may cause family doctors to provide "care" that will blind and cripple their patients.

The same doctors who find no fault with the oral drugs that can promote cancer (Januvia) and blindness (Actos) are knocking themselves out to find problems with insulin that can only be demonstrated in studies with significant methodological flaws.

One thing is for certain: hypos attributable to insulin are far less likely to cause life-ruining damage than are prolonged exposure to high blood sugars.