These pages are best viewed with Netscape version 3.0 or higher or Internet Explorer version 3.0 or higher. When viewed with other browsers, some characters or attributes may not be rendered correctly.

Debate: Does Frequent, Severe Hypoglycemia Cause Significant Cognitive Decrements?

On the Con Side . . .
Fear of Hypoglycemia Should Not Prevent Intensified Insulin Treatment!

Per Reichard, MD, PhD

Fear of hypoglycemia has been claimed to be a major obstacle to improved blood glucose control in patients with type I diabetes.¹ Hypoglycemia is, of course, a serious acute complication of diabetes treatment, and lower blood glucose levels result in more frequent hypoglycemia.² At the same time, it is now obvious and definitely proven that blood glucose levels closer to normal retard or halt the micro- and macrovascular complications of diabetes.^3-7 Unfortunately, the HbA_1c level required to do so⁸ leads to an almost threefold increase of severe hypoglycemic episodes.^4,6

The most important questions to be asked are the following:
• Does any degree of low blood glucose hurt the brain?
• Is the increased frequency of hypoglycemic episodes observed during intensified insulin treatment harmful to the brain, and does this increased frequency preclude intensified treatment?
• Can patients learn how to avoid severe hypoglycemia without jeopardizing metabolic control?

Hypoglycemia may exercise its effects on the brain in various ways.⁹ In rats, it has been shown that hypoglycemia must be prolonged and deep to cause permanent cell destruction, and the observed damage is correlated with the duration of iso-electric EEG recordings,¹⁰ a neurological finding indicating serious brain disturbance. Of course, experimental studies of this kind have never been performed in humans.

Studies using neuropsychological tests have shown that patients with type I diabetes who have had hypoglycemic comas manifest cognitive impairment.^11-14 In an early study by Bale,¹¹ blood glucose values were not obtained before the test, so some of the patients may have been hypoglycemic during testing. The two most cited studies,^12,14 discussed below, were more carefully designed.

In the study by Wredling and associates,^14,17 patients who experienced recurrent severe hypoglycemia (including unconsciousness) were compared with an equivalent number of patients who never experienced severe hypoglycemia. The median age across both groups was 49 years. The groups were matched for type of diabetes, sex, age, and duration of diabetes. Using the computerized automated psychological test (APT) system for patient asessment,¹⁵ the patients with a history of severe hypoglycemia were found to perform less well on a number of tests.

The APT consists of tests shown to the patient on a computer screen and solved in a predefined order. Subjects with recurrent hypoglycemia tended to perform more slowly on a number of measures, leading the authors to speculate that repeated severe hypoglycemia may result in permanent cognitive impairment. It is of interest that the patients with recurrent hypoglycemia actually had better values than the control group when the total number of correctly solved mazes in the perceptual maze test were counted. Taken together with other findings from this study, this superior performance seems to suggest a more careful strategy when solving complicated tasks. This might reflect a "pre-morbid" disposition to react less quickly to correct impending hypoglycemia. Wredling and associates entertained this alternative interpretation and concluded that they were unable to determine whether their subjects’ performance was the cause, or the consequence, of recurrent severe hypoglycemia.

Langan and associates¹² also used a cross-sectional approach. They studied 100 patients with adult-onset type I diabetes and found a weak association between recurrent severe hypoglycemia and some neuropsychological variables, especially measures of decision-making time. The patients had a mean age of 40 years and a mean duration of diabetes of 13 years. Those having at least five episodes of severe hypoglycemia did not have a lower HbA_1c level than those without severe hypoglycemia. Crucial to their argument are comparisons between measures of current cognitive functioning and measures of "pre-morbid intelligence." Yet pre-morbid functioning can never be measured directly, but can only be estimated, some would argue very crudely.

Cross-sectional studies show associations, never causality. Of special interest is the study by Wredling and associates¹⁴because it used the same battery of computerized neuropsychological tests as the prospective, randomized Stockholm Diabetes Intervention Study (SDIS),^5,6,16,17 described in more detail below. While the cross-sectional study showed signs of poorer neuropsychological test performance in patients who had experienced hypoglycemic coma,¹⁴ the longitudinal study of intensified insulin treatment showed no decrements on these tests in the intensively treated patients over a 10-year period, despite almost three times as many episodes of severe hypoglycemia in these patients.^16-19 Deary and associates criticize the methods used by the SDIS,¹³ but they do not mention that the same methods, when used in a cross-sectional study, showed results that were similar to their own.

Two prospective, randomized studies have examined the relationship between intensified insulin treatment and diabetic complications, the SDIS^5,6 and the Diabetes Control and Complications Trial (DCCT).^3,4 In both studies, neuropsychological tests were included from baseline. In the SDIS, computerized tests were used at entry to the study,¹⁶ after 3,¹⁶ 5,¹⁷ and 7.5 years,¹⁸ and then again after 10 years (no longer a randomized study).¹⁹

The SDIS used tests that had been validated in previous studies,^20-24 were only performed when the blood glucose levels were above 4.5 mmol/l, and were conducted twice, 1 or 2 days apart. The tests included measures of general intelligence, short-term memory, motor speed, and eye-hand coordination. This combination of tests has previously been found to detect even slight brain damage, and non-computerized versions of some of the tests have been used extensively in the past.²⁵

In their critique, Deary and associates claim that the treatment groups in the SDIS were not clearly separated on the basis of past history of severe hypoglycemia, that the tests used were not sensitive enough, and that the time of follow-up was too short.¹³ The subjects have now been followed for 10 years.¹⁹ During the first 7.5 years, there was an almost threefold increase in the frequency of severe hypoglycemia (requiring outside help, but not necessarily resulting in coma) in the more intensively treated group.^17,18 Clinically meaningful differences between groups with intensified or standard treatment would have been detected if there were any differences to be found, but there were none. This does not mean that hypoglycemia is never harmful to the brain, but rather that the increase of hypoglycemic episodes that require help seen during intensified treatment is not any more harmful than those seen in conventionally treated patients.

The neuropsychological tests results from the DCCT²⁶ generally support the results from the SDIS, but many more patients were studied for a mean of 6.5 years. In the DCCT, intensively treated patients performed as well as those receiving conventional therapy, and patients who experienced repeated episodes of hypoglycemia over the course of the study did not show more cognitive impairment.

Are there any particularly vulnerable groups? The developing brains of young children may be susceptible,²⁷ but the studies performed so far are few. Another potentially vulnerable group are adult patients with complications. Ryan and his colleagues studied the relationship between diabetic complications and cognitive function in adults with type I diabetes and found a correlation between some neuropsychological test results (psychomotor slowing) and measures of distal polyneuropathy.²⁸ They also found that cognitive impairment, primarily related to neuropathy, was aggravated by hypoglycemia,²⁹ findings corroborated by the SDIS. After 7.5 years of study, the interaction between clinical symptoms of neuropathy and the number of severe hypoglycemic attacks was significantly associated with some neuropsychological test results.¹⁸

This tendency was even more pronounced after 10 years of study and then applied to the interactions between hypoglycemia and both neuropathy and serious retinopathy (with photocoagulation), but not with nephropathy.¹⁹ I think that there may be a central nervous system neuropathy or a "diabetic encephalopathy" related to hyperglycemia,^9,30 and patients with this kind of presumably microvascular/neurological complication are perhaps vulnerable to severe hypoglycemia.

Is it possible for patients to learn how to avoid severe hypoglycemia? The results from the SDIS¹⁹ seem to show that the frequency of severe hypoglycemia decreases after some years of intensified treatment. Whether this is a general tendency and whether specific advice is of value to the patients must be studied separately.

Conclusions
Several cross-sectional studies have shown an association between severe hypoglycemia and cognitive impairment in patients with type I diabetes. It is striking, however, that the impairments that have been registered are modest and mostly concern decision-making speed. Cross-sectional studies can show correlations but never determine causality. As a consequence, it is conceivable that patients who find it difficult to make fast decisions develop severe hypoglycemia because they are unable to take adequate measures to ward off the danger in time.

It cannot be excluded, however, that hypoglycemia with unconsciousness in this group of patients who develop hypoglycemia in the absence of intensive insulin therapy causes neuropsychological deficits. Hypoglycemic coma causes brain damage in rats. Two long-term, prospective, randomized studies have shown that during intensified insulin treatment, the risk of hypoglycemia increases, but there is no increase in cognitive impairment. These studies have been criticized^13,31 but, in my opinion, generally without challenging the chief conclusion: during intensified treatment, there is an increase in the frequency of hypoglycemic attacks necessitating external help, but these attacks are not long or deep enough to cause cognitive impairment.

Patients with advanced neuropathy or serious retinopathy may be more vulnerable, and these patients should have glycemic goals that make it possible to avoid severe hypoglycemia. Children before puberty are also probably vulnerable to deleterious effects of hypoglycemia on the brain. Therefore, slightly higher glycemic goals should be set for these young patients. Generally, the present trend toward more intensive diabetes management and lower blood glucose levels in adults with type I diabetes should not be halted because of fear of hypoglycemia, but patients need experienced tutors.

Acknowledgments

Grants were received from the Swedish division of Novo Nordisk, Inc., and from Boehringer Mannheim Scand., Inc.

References

¹Cryer PE: Hypoglycemia: the limiting factor in the management of IDDM. Diabetes 43:1378-89, 1994.

²Thorsteinson B, Pramming S, Lauritzen T, Binder C: Frequency of daytime biochemical hypoglycemia in insulin-treated diabetic patients: relation to daily median blood glucose concentrations. Diabetic Med 3:147-51, 1986.

³The DCCT Research Group: The Diabetes Control and Complications Trial (DCCT): design and methodological considerations during the feasibility phase. Diabetes 35:530-45, 1986.

⁴The DCCT Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 329:977-86, 1993.

⁵Reichard P, Britz A, Cars L, Nilsson BY, Sobocinsky-Olsson B, Rosenqvist U: The Stockholm Diabetes Intervention Study (SDIS): 18 months’ results. Acta Med Scand 224:115-22, 1988.

⁶Reichard P, Nilsson BY, Rosenqvist U: The effect of long-term intensified insulin treatment on the development of microvascular complications of diabetes mellitus. N Engl J Med 329:304-09, 1993.

⁷Jensen-Urstad K, Reichard P, Rosfors S, Lindblad LE, Jensen-Urstad M: Early atherosclerosis is retarded by long-term blood glucose control in patients with IDDM. Diabetes 45:1253-58, 1996.

⁸Reichard P: Are there any blood glucose thresholds for the deterioration of serious microvascular complications? J Diabetes Compl 9:25-30, 1995.

⁹Biessels GJ, Kappelle AC, Bravenboer B, Erkelens DW, Gispen WH: Cerebral function in diabetes mellitus. Diabetologia 37:643-50, 1994.

¹⁰Auer RN, Olsson Y, Siesjö B: Hypoglycemic brain injury in the rat. Correlation of density of brain damage with the EEG isoelectric time: a quantitative study. Diabetes 33:1090-98, 1984.

¹¹Bale RN: Brain damage in diabetes mellitus Brit J Psychiatry 122:337-41, 1973.

¹²Langan SJ, Deary IJ, Hepburn DA, Frier BM: Cumulative cognitive impairment following recurrent severe hypoglycaemia in adults with insulin-treated diabetes mellitus. Diabetologia 34:337-44, 1991.

¹³Deary IJ, Crawford JR, Hepburn DA, Langan SJ, Blackmore LM, Frier BM: Severe hypoglycemia and intelligence in adult patients with insulin-treated diabetes. Diabetes 42:341-44, 1993.

¹⁴Wredling R, Levander S, Adamson U, Lins PE: Permanent neuropsychological impairment after recurrent episodes of severe hypoglycaemia in man. Diabetologia 33:152-57, 1990.

¹⁵Levander S, Elithorn A: An Automated Psychological Test System, APT Manual: Version 1.0. Trondheim, Norway, Department of Psychiatry and Behavioral Medicine, University of Trondheim, 1987.

¹⁶Reichard P, Berglund B, Britz A, Levander S, Rosenqvist U: Hypoglycemic episodes during intensified insulin treatment: increased frequency but no effect on cognitive function. J Int Med 229:9-16, 1991.

¹⁷Reichard P, Britz A, Rosenqvist U: Intensified conventional treatment and neuropsychological impairment. Brit Med J 303:1439-42, 1991.

¹⁸Reichard P, Pihl M: Mortality and treatment side-effects during long term intensified conventional treatment in the Stockholm Diabetes Intervention Study (SDIS). Diabetes 43:313-17, 1994.

¹⁹Reichard P, Pihl M, Rosenqvist U, Sule J: Complications in insulin-dependent diabetes are caused by the elevated blood glucose level: The Stockholm Diabetes Intervention Study (SDIS) after 10 years. Diabetologia. In press

²⁰Levander S: Evaluation of cognitive impairment using a computerized neuropsychological test battery. Nord Psykiatr Tidskr 41:417-22, 1987.

²¹Elithorn A. Mornington S, Stavrou A: Automated psychological testing: some principles and practice. Int J Man-Machine Studies 17:247-63, 1982.

²²Levander SE, Bartfai A, Schalling D: Regional cortical dysfunction in schizophrenic patients studied by computerized neuropsychological methods. Perceptual and Motor Skills 61:479-95, 1985.

²³Levander S, Hagermark Ö, Ståhle M: Peripheral antihistamine and central sedative effects of three H1-receptor antagonists. Eur J Clin Pharmacol 28:523-29, 1985.

²⁴Levander M: Neuropsychology of Handedness. Thesis. Stockholm, Sweden, Department of Psychology, University of Stockholm, 1988.

²⁵Lezak M: Neuropsychological Assessment. 2nd ed. New York, Oxford University Press. 1983, p. 35, 55-56, 368-69, 417-19, 529, 556-59.

²⁶The DCCT Research Group: Effects of intensified diabetes therapy on neuropsychological function in adults in the Diabetes Control and Complica-tions Trial. Ann Intern Med 124:379-88, 1996.

²⁷Ryan C, Vega A, Drash A: Cognitive deficits in adolescents who developed diabetes early in life. Pediatrics 75:921-27, 1985.

²⁸Ryan CM, Williams TM, Orchard TJ, Finegold DN: Psychomotor slowing is associated with distal symmetrical polyneuropathy in adults with diabetes mellitus. Diabetes 41:107-13, 1992.

²⁹Ryan CM, Williams TM, Finegold DN, Orchard TJ: Cognitive dysfunction in adults with type I (insulin-dependent) diabetes mellitus of long duration: effects of recurrent hypoglycaemia and other chronic complications. Diabetologia 36:329-34, 1993.

³⁰Dejgaard A. Gade A, Larsson H, Balle V, Parving A, Parving H-H: Evidence for diabetic encephalopathy. Diabetic Med 8:162-67, 1991.

³¹Deary I, Frier B: Severe hypoglycaemia and cognitive impairment in diabetes. Brit Med J 313:767-68, 1996.

Per Reichard MD, PhD, practices in the Department of Internal Medicine at Södersjukhuset Medicinska Kliniken, in Stockholm, Sweden.