Hypertension

The Glucose-Insulin Tolerance Test and its Relevance to “Essential Hypertension” and HDL/LDL Cholesterol Abnormalities

By Jonathan V. Wright M.D.

Experienced practitioners have long made use of the work of Joseph Kraft1 From his series of 3650 glucose-insulin tolerance tests (referred for evaluation of suspected glucose intolerance), he distinguished 5 insulin-response patterns,4 of which were diagnostic or predictive of diabetes mellitus.

Of the 3,650 participants, 1,937 had abnormal glucose tolerance (Wilkerson point system, American Diabetes Association).

Of the remaining 1,713 individuals with perfectly normal glucose tolerance curves, 862 had abnormal insulin response patterns diagnostic of prediabetes (“diabetes in situ”), and an additional 240 were “borderline” predictive.

Thus 6,496 (1102/ 1713) of the “normal” individuals (judged on GTT alone) were actually abnormal when the additional insulin tolerance data was used.

Kraft went further to show that with appropriate diet control, improvement from abnormal to normal insulin tolerance response could be achieved.

Personally, I’ve found the “ITT” component of the “G-ITT” to be so helpful in finding refined-carbohydrate sensitive individuals that I always run the combined test. Unfortunately (as one might expect from Kraft’s results), I frequently must inform individuals that their previous advice of “no sugar trouble” (based on GTT alone) was quite erroneous. Too often, valuable time has been wasted and unnecessary symptoms (as well as disease progression) have been suffered because of this sort of incomplete testing.

THE G-ITT AND “ESSENTIAL” HYPERTENSION

Considerable recent investigation 2-5 has unearthed an unanticipated connection between “essential” hypertension and hyperinsulinaemic response to a glucose load. (The available literature-based evidence along with my clinical experience, in treatment as well as diagnosis, leads me to the firm conclusion that a very substantial proportion of “essential hypertension” is simply a symptomatic variant of our old nemesis, “refined carbohydrate disease”. My opinion, of course, but it appears there’s really something there.)

I can do no better than to quote a summary article #6: “The finding of insulin resistance and hyperinsulinaemia in nonobese patients with hypertension raises the possibility of a fundamental relation between insulin and blood pressure that transcends the narrower association of insulin with obesity-related hypertension.” The same article also notes: “… hyperinsulinaemia was observed in hypertensive subjects after they ingested glucose. Fasting insulin levels were normal. A similar situation prevails in obese patients with hypertension …”

This leads us back to Kraft’s glucose-insulin tolerance test, which I currently run for every “essential hypertensive” individual I see. (So far, I’ve found none who’ve had it done as part of a previous “hypertensive work-up”.) More than 50% are abnormal according to Kraft’s criteria. Treatment of these individuals as “refined carbohydrate sensitives” (who happen to have hypertension as one major symptom) is uniformly successful (usually over several months) in blood pressure control.

THE G-ITT AND ABNORMAL LIPOPROTEIN (HDL/LDL) RATIOS

Individuals with hypertension and an hyperinsulinaemic response to glucose appear to have the same fundamental mechanism affecting their lipoprotein distribution. Once again, direct quotes from a research report and its abstract (7) tell it best:

“Evidence has accumulated … documenting a significant correlation between the level of plasma glucose … and blood pressure in nondiabetic individuals. More recently, it has been shown that plasma insulin concentrations are also higher in patients with hypertension, independent of concomitant obesity or antihypertensive medication. In addition, abnormalities of lipoprotein metabolism also occur more frequently in patients with hypertension, either treated or untreated. Taken together, these observations suggest that high blood pressure is associated with a series of defects in glucose, insulin, and lipoprotein metabolism,”

“Plasma glucose, insulin and lipoprotein concentrations were determined in 20 (nonobese) men with hypertension, and compared with values in 20 (nonobese) normotensive men of comparable age and body mass index … a significant correlation existed between the plasma insulin response to oral glucose and both systolic and diastolic blood pressure … the greater the plasma glucose and insulin response to oral glucose, the lower the plasma HDL concentrations, and the higher the ratio of plasma LDL cholesterol to HDL cholesterol”

USING THE GLUCOSE-INSULIN TOLERANCE TEST (G-ITT) A SUMMARY

From the above, it should be obvious that a G-ITT (and not a GTT) is a rational and important part of the laboratory work-up for:
· suspected refined carbohydrate intolerance
· “essential” hypertension
· HDL/LDL cholesterol abnormalities

Especially since, as should also be obvious, the treatment for all three conditions may often turn out to be the same! (Or as our naturopathic colleagues might eruditely remark: tolle causam!)

JUST A WORD ABOUT TREATMENT

All of us have evolved our own system of treating what might well be called the “Cleave-Yudkin-Burkett refined carbohydrate syndrome.” A very significant recent contribution to the understanding of appropriate treatment of this syndrome has been made by John Cleary8 who emphasizes and explains the biochemical basis of the use of niacin and omega-3 fatty acids for the conditions listed above, as well as the syndrome in general.

REFERENCES

1. Kraft J.R. Detection of Diabetes Mellitus in Situ (Occult Diabetes). Lab Med 6(2), 10 (1975).
2. Singer P., et al. Postprandial hyperinsulinism in patients with mild essential hypertension. Hypertension 7, 182 (1948).
3. Ferrannini E., et al. Insulin resistance in essential hypertension, N Engl. J. Med. 317. 350 (1987).
4. DeFronzo RA, et al. The effect of insulin on renal handing of sodium, potassium, calcium and phosphate in man. J. Clin. Invest. 55, 845 (1987).
5. O’Hare J.A., et al. Insulin increases plasma norepinephrine and lowers plasma potassium equally in leean and obese men. Clin. Res. 33, 441A (abstract) (1985).
6. Landsberg L. Insulin and hypertension. N. Engl. J. Med> 317 (6). 378 (1987).
7. Fuh MM-T, et al. Abnormalities of carbohdrate and lipid metabolism in patients with hypertension. Arch. Int. Med. 147, 1035 (1987).
8. Cleary JP. The importance of oxidant injury as a cause of impaired mitochondrial oxidation in diabetes. J. Orth. Med. 3 (4), 164 (1988).

---

GLUCOSE-INSULIN TOLERANCE TEST

A summary of Dr. Kraft’s data

Dr. Kraft, a clinical pathologist, ran 3,650 glucose-insulin tolerance tests. Please keep in mind for the next part of the discussion that Dr. Kraft was looking for diabetes mellitus and was not particularly looking for hypoglycemia Very many of his tests were run for only three hours or four hours. For this reason, the new glucose-insulin procedure should not only yield more information as outlined below with regard to the diagnosis of diabetes mellitus and latent diabetes mellitus, but it also should enable us to do more for hypoglycemia Of these 3,650 glucose tolerance tests, 1,713 were defined as normal (for diabetes mellitus) according to the criteria of the American Diabetes Association.

Of these 1,713 tests, which would have been judged normal strictly on the basis of glucose tolerance, 50% were demonstrated abnormal on the basis of insulin tolerance and another 14% were borderline. 2.5% were low insulin response.

THIS MEANS THAT TWO-THIRDS OF THESE TESTS THAT WOULD HAVE BEEN JUDGED NORMAL BY THE GLUCOSE TOLERANCE TEST WERE FOUND TO BE LATENT DIABETIC WHEN THE INSULIN TOLERANCE TEST WAS TAKEN INTO CONSIDERATION.

Remember that all these people were referred because of suspicion of diabetes, so the high numbers should not be so surprising, but what the above data shows is that of the 1,713 cases, out of the 3,650 which would have been judged normal by only glucose tolerance test criteria, the insulin test showed that only one-third of them really were normal. Looking at the overall group of 3,600 people referred because of suspicion of diabetes, the usual glucose tolerance test criteria would have found that 1,713 were normal. Bringing to bear the insulin tolerance test, two thirds of these so-called “normal” were found to be latent diabetic. It appears obvious that the glucose-insulin tolerance test will detect many more latent diabetics.

Criteria for the insulin-tolerance test are as follows. This is once again as taken from the article by Dr. Kraft:

Pattern 1: Normal Fasting insulin 0-30 units. Peak insulin at 1/2-1 hour. The summary of the second and third hours insulin values less than 60.

Pattern 2: Normal fasting insulin. Peak at 1/2-1 hour with a delayed return to normal. “Second and third hour levels between 60-100 considered borderline for diabetes. Values of the second and third hour adding to greater than 100 considered definite diabetes.

Pattern 3: Normal fasting insulin. Peak at second or third hour instead of 1/2-1 hour. This is always considered diagnostic for diabetes.Pattern 4: High fasting insulin level. Above 30. This is considered positive for diabetes.

Pattern 5: Low insulin response. All tested values for insulin less than 30. This is considered to be insulinopenic pattern, this is associated with high levels of plasma glucose this is considered to be the “juvenile pattern” with high glucose and in effect insulin deficiency probably because of dead or near dead islet cells, Occasionally this pattern is with normal or borderline glucose tolerance. This is considered to be because of a low carbohydrate diet or carbohydrate preparation.”

Dr. Kraft’s article discusses only latent diabetes. In another article by Dr. William Hudspeth, associate professor of psychiatry from the University of Nevada School of Medicine, it is pointed out that frequently hypoglycemic symptoms do not correlate with the low point on a glucose tolerance test. It is pointed out that they also may have occurred despite a fairly normal looking glucose tolerance test. Dr. Hudspeth’s group have evaluated this phenomenon further by running insulin tolerance along with glucose tolerance and at the same time putting an EEG on each patient and observing for brain wave response.

Unfortunately, we’re not going to be able to do an EEG and pick up the abnormal brain wave response. What’s interesting is that Dr. Hudspeth found in some cases where the glucose was low and in other cases where it was not low at all, what seemed to correlate better with hypoglycemic symptomatology was either low glucose or normal glucose in combination with a distinct insulin peak. Although Dr. Hudspeth did not have a breakdown of the different types of patterns, just this information alone should enable us to better pick up hypoglycemic symptomatology in the absence of abnormal glucose tolerance test values.

REFERENCES:

1. Detection of Diabetes Mellitus, In Situ (occult diabetes), Karft, Joseph R., Laboratory Medicine, Volume VI, #2, pages 10-22, February 1975.
2. Neurobiology of the Hypoglycemia Syndrome, Hudspeth, W.J. Et Al, Journal of Holistic Medicine, Volume III. #1, pages 60-71, Spring/Summer, 1981.