The glucotoxicity paradigm
For as long as I have practiced medicine, the mantra of excellent diabetic care was tight blood glucose control. All the diabetes associations, the university professors, the endocrinologists, and diabetic educators agreed. The prime directive was “Get those blood sugars down into the normal range at all costs, soldier!” The only acceptable response was, “Sir! Yes, Sir!” Insubordination was not tolerated.
At first glance, lowering blood glucose as the primary therapeutic target seemed fairly logical. The underlying premise assumes that high blood glucose is the major cause of morbidity. But remember that high blood glucose is only the symptom. In Type 1 diabetes, insulin levels are very low and in type 2 diabetes insulin levels are very high. The symptom is the same, but the diseases are essentially opposites. So how could the exact same treatment be beneficial in both cases?
It’s hard to imagine that the same solution exists for opposite problems. For example, we don’t use the same treatment for both underactive and overactive thyroids. We don’t use the same treatment for both over-eating and under-eating. We don’t use the same treatment for both fever and hypothermia. We don’t wash clothes by soaking in water and then dry clothes by soaking in water.
Type 1 diabetes is caused by lack of insulin, so logically, the cornerstone of management is the replacement of the missing insulin. Type 2 diabetes, however is caused by excessive insulin, so logically the cornerstone of management should be the reduction of the high insulin. Further, being predominantly a dietary disease, the cure must be dietary rather than pharmaceutical. Drugs can’t cure a dietary disease. Only fixing the diet can. These inconvenient facts were simply ignored.
For most of the early and mid 20th century, much of the research focused on type 1 diabetes. Insulin replacement therapy effectively cured type 1 diabetes, the severe lack of insulin. Untreated patients, usually children, experienced unrelenting weight loss until they died, emaciated and skeletal. With exogenous insulin injections, weight stabilized and this formerly fatal disease became manageable. But injecting exogenous insulin came with its own complications.
In order to avoid too high or too low blood glucose, it is essential to match the insulin dose with the amount of food, predominantly carbohydrates, being eaten. Underdosing caused high blood glucose and overdosing of insulin caused low blood glucose, called hypoglycemic reactions. Patients would become sweaty, tremulous and if severe, could suffer from seizures, loss of consciousness and death.
These hypoglycemic reactions are immediately life threatening. Even over the longer term, they are strongly associated with a wide range of adverse outcomes including heart disease and death, particularly in older adults. In 2014, nearly 100,000 emergency room visits and 30,000 admissions to hospital are directly related to hypoglycaemia. From 1999-2011, an estimated 404,000 patients were hospitalized for low blood sugars compared to 280,000 hospitalized for high blood sugars. There’s also a steep fiscal cost associated with these reactions. Over 5 years, the estimated cost of this care is $600 million.
Because of the immediate severity of hypoglycemia, until the mid 1990s, researchers and physicians argued back and forth whether high blood sugars were even all that dangerous. Extremely high blood glucose could cause diabetic ketoacidosis in type 1 diabetes and non-ketotic hyperosmolar coma in type 2 diabetes, but these complications were relatively uncommon. On a typical blood glucose-monitoring device, the levels would literally need to be off the scale.
Moderately raised levels of raised blood glucose were not obviously detrimental to long-term human health. In the short-term, there were no symptoms whatsoever. Normal blood glucose levels vary depending upon what and when you last ate. Fasting blood glucose is considered normal below 6.1 mmol/L. It goes up after a meal and typically peaks approximately two hours after the meal at less than 10.0 mmol/L.
The renal threshold for glucose, the level where glucose begins to appear in the urine, is approximately 10 – 11 mmol/L. In the normal situation, glucose is completely reabsorbed by the proximate tubule of the kidney so that no glucose spills out into the urine. Above the renal threshold, this re-absorptive mechanism is overwhelmed and progressively more and more glucose spills out into the urine and symptoms may appear. Patients may notice increased urination as the excreted glucose pulls water along. Patients become dehydrated and then will also notice increased thirst. This excessive drinking and urination is a classic sign of diabetes mellitus.
At blood glucose levels below 10 mmol/L, there are no noticeable symptoms of hyperglycemia. Insulin could tightly control blood glucose but inevitably resulted in more hypoglycemic episodes, and this tradeoff was not obviously beneficial. For many decades, the standard medical practice was to keep the blood glucose levels slightly high but below 10 mmol/L. This avoided both hypoglycemia and the symptoms of high blood glucose. Yes, the blood glucose level was higher than normal, but there were far fewer episodes of hypoglycemia. Nobody had yet found definitive proof that the slightly high blood glucose was injurious.
Things completely changed in 1993 with the publication of the Diabetes Control and Complications Trial (DCCT). This large randomized trial was designed to answer this very question by comparing intensive insulin therapy in type 1 diabetes to conventional treatment. The high dose group successfully lowered the A1C but, as expected, there was a price to be paid. The incidence of hypoglycemia was three times higher in the intensive control group. Was it worth it?
The results were nothing short of revolutionary. Diabetic eye disease decreased by 76%, kidney disease decreased by 50%, and nerve damage was reduced by 60%. Clearly, long-term hyperglycemia was causing end-organ damage. Insulin treatment to tightly control blood glucose could preserve organ function.
In 2005 the Epidemiology of Diabetes Interventions and Complications (EDIC) study was published. This followed the original DCCT patients out to seventeen years. The results were, once again, revolutionary. Intensive insulin treatment had reduced cardiovascular disease by an astonishing 42%. This clearly established the paradigm of glucotoxicity (toxicity resulting from high blood glucose) in type 1 diabetes.
Other than the increased hypoglycemic reactions, there is another cost to intensive insulin treatment. The incidence of substantial weight gain increased significantly. Over nine years, almost thirty percent of subjects developed ‘major weight gain’, far in excess of the conventional insulin group.
By the late 1990s, as the obesity epidemic gained momentum and the type 2 diabetes epidemic was starting its run, some inkling of concern percolated amongst some physicians about the excessive weight gain. One quarter of the intensive treatment group increased their Body Mass Index from 24 (normal weight) to 31 (obese). The well-known propensity of insulin to cause weight gain is not some trivial problem of trying to fit into your swimsuit, but had significant health consequences.
There were other danger signs, too. This weight gain concentrated in the abdominal area. Blood pressure and blood cholesterol increased. The combination of central adiposity, hypertension and dyslipidemia is the hallmark of the metabolic syndrome that is more typical of type 2 diabetes, characterized by insulin excess. This could not be good news, but there was worse to come.
Weight, waist circumference, and insulin dosage continued inexorably higher. Blood pressure and cholesterol followed the upward trend. Intensively treated type 1 diabetic patients were developing the metabolic syndrome, the problem of hyperinsulinemia.
The coronary artery calcification (CAC) score and carotid intimal medial thickness (CIMT) are both well accepted measures of atherosclerosis – the buildup of plaque in the artery that leads to heart attacks and strokes. Heavy calcification and thickened blood vessel walls are subclinical markers of atherosclerosis and sure signs of bad things to come. Type 1 diabetic patients with the most weight gain also developed higher CIMT and CAC scores. High insulin dosage reliably predicted advanced atherosclerosis. While type 1 diabetes is originally a disease of too little insulin, over time, these patients are all developing the problems of too much insulin.
There are two types of toxicity – glucotoxicity and insulin toxicity. At the time of diagnosis, type 1 diabetic patients have extremely low insulin levels, so the dominant problem in the short term (<10 years) is glucotoxicity. Taking insulin to reduce blood glucose improves clinical outcomes.
But over time, heavy handed dosing of insulin produces all the problems of excessive insulin – obesity, metabolic syndrome and atherosclerosis. Hyperinsulinemia leads to insulin resistance causing the same exact problems seen in type 2 diabetes. The heavy dosing of insulin in type 1 diabetes was creating type 2 diabetes!
The Dietary Guidelines for Americans had, since the late 1970s recommended a diet low in total fat and high in carbohydrates. Since refined carbohydrates such as bread raise blood glucose significantly more than dietary fat, this required higher insulin dosing in type 1 diabetics to keep blood glucose control. This wasn’t so bad. Was it?
It is well established that type 1 diabetics have a four fold higher risk of death compared to non-diabetics. Most believe this is caused by glucotoxicity, but the evidence points to the contrary. Multivariate analysis of the EuroDiab study revealed the stunning conclusion that glucotoxicity, as measured by Hemoglobin A1C was not a significant risk factor. The one factor everybody assumed was critically important turned out to be almost irrelevant. Instead, the most important modifiable risk factors were Waist to Hip Ratio (a measure of visceral fat), blood pressure, and cholesterol.
These are all the usual culprits found in metabolic syndrome, caused by insulin excess, not insulin deficiency. Type 1 diabetics patients suffered all the disease as type 2 diabetics, but high blood glucose was not the causal link, as had always been assumed.
Multiple other studies confirmed the EuroDiab results. The Golden Years Cohort Study followed 400 type 1 diabetic patients who lived over 50 years managing their disease. This group had beaten the odds and survived. What was their secret? One thing became clear. The secret was not tight blood glucose control.
The Golden Cohort’s average A1C was 7.6% – well over the standard recommended target of 7.0%. This level could only be described according to standard management as ‘poor’. Some had A1C measurements in the 8.5-9.0% range, yet still far out-lived the average. Strikingly, no Golden Cohort patient had an A1C in the normal range! Clearly, glucotoxicity was not the major player here. The entire Golden Cohort of survivors had ‘sub-optimal’ blood glucose control and it didn’t matter. Their health was excellent.
The common bond shared within this group was not low blood glucose, but rather a low insulin dosage. Obesity, high blood pressure, and the other manifestations of hyperinsulinemia were notably absent. Elevated HDL (good) cholesterol was associated with lower insulin doses. Since long-standing type 1 diabetics have no insulin production of their own, all insulin must be injected from prescribed doses of medication.
Another paper wonders, “Why do some patients with type 1 diabetes live so long?” Blood glucose control plays a minor role. Only a reasonable, but not necessarily ‘ideal’ blood glucose is needed for longevity. Glucotoxicity is a not a major player.
One again, low daily insulin dosage is the key to survival, leading to lower body weight, lower blood pressure, and improved cholesterol. High daily dosage of insulin leads to the metabolic syndrome, a disease of hyperinsulinemia. This puts them at the same risk of cardiovascular disease that all the type 2 diabetic patients develop. The over-exuberant treatment of type 1 diabetes leads, in essence to type 2 diabetes. They have what can only be called ‘double diabetes’.
The conventional dietary advice of the past 30 years to eat a low fat, high carbohydrate diet requires more insulin to keep blood glucose levels normal. This ‘carb up and shoot up’ strategy, requires decades of chronically high insulin levels, eventually leading to increasing insulin resistance and the precursor to type 2 diabetes as well as the rest of the metabolic syndrome. As type 1 diabetic patients develop double diabetes, factors like cholesterol, high blood pressure and central adiposity all become increasingly important.
There are two toxicities at work here. Early on, glucotoxicity is the main concern. But over time, insulin toxicity becomes increasingly important and the key determinant for survival. The optimal treatment strategy reduces both.
While we acknowledge the damage caused by glucotoxicity, we ignore the insulin toxicity. Raising insulin to lower blood glucose simply trades insulin toxicity for glucotoxicity. Sledgehammering patients with high doses of insulin to reduce blood glucose was not improving the health of patients. Blood glucose and insulin must be lowered simultaneously.