Section 9: Metabolic Diseases

9.1 Diabetes Mellitus

Hypoglycemic Reactions

As noted earlier, of all the metabolic complications of diabetes, hypoglycemia represents the most acute risk for traffic safety concerns. Hypoglycemia is common in diabetic individuals treated with insulin and also can occur in individuals treated with oral hypoglycemic sulfonylurea agents. Severity of hypoglycemia can range from very mild lowering of glycemia (60-70 mg/dl) with minimal or no symptoms to severe hypoglycemia with very low glucose levels (<40 mg/dl) and neurologic impairment (Gerich, Mokan, Veneman, and Korytkowski, 1991).

A typical hierarchy of responses to decreases in plasma glucose concentrations has been described by Gerich et al. (1991). The initial response, occurring at approximately 70 mg/dl, involves an increase in the secretion of counter regulatory hormones (glucagon, epinephrine, growth hormone, and cortisol) and a concomitant increase in norepinephrine and acetylcholine release. If the initial responses are ineffective and further decreases in plasma glucose concentrations occur, the autonomic symptoms of sweating, tremor, hunger, anxiety, and palpitations occur, typically at blood glucose concentrations of 60 mg/dl. These autonomic symptoms usually act as a warning to the experienced individual to undertake protective measures (e.g., the intake of food) to ward off an impending hypoglycemic reaction. If the autonomic warning symptoms are ignored or unrecognized (hypoglycemic unawareness), with subsequent reductions in plasma glucose concentrations to around 50 mg/dl, symptoms of neuroglycopenia (weakness, lethargy, blurred vision, confusion, dizziness) and signs of cognitive dysfunction usually occur. Results from Pramming, Thorsteinsson, Bebdtson, and Bionder (1986) reveal deteriorations in cognitive performance in IDDM individuals at blood glucose concentrations just below subnormal levels (54 mg/dl). An important finding in this investigation is that for all but one of the neuropsychological tests (finger tapping), there was a gradual deterioration in cognitive performance with decreasing blood glucose concentrations. Based on outcomes, the authors concluded that performance on everyday tasks that entail planning and control would be adversely affected even at subnormal blood glucose concentrations, concentrations that are usually not considered to be hypoglycemic. Significant disruptions in simulated driving behaviors during moderate hypoglycemia (2.6 + 28 mM, ˜50 mg/dl) have been reported by Cox, Gonder-Frederick, and Clarke (1993). Disrupted behaviors included more swerving, spinning, time over midline, time off road, and apparent compensatory slowing with an increase in 'very slow' driving.

Despite the fact that hypoglycemia is the most common complication of insulin therapy in individuals with diabetes mellitus, the actual incidence of hypoglycemia is, however, difficult to ascertain. Ward, Stewart, Cutfield, et al. (1990) examined the prevalence of hypoglycemia in individuals randomly selected from outpatient clinics in Auckland, Australia. The authors found that the majority (98 percent) of those surveyed had experienced hypoglycemia, with 73 percent reporting having had at least one mild episode of mild hypoglycemia monthly. Episodes of minor hypoglycemia have been estimated to occur twice per week in individuals with IDDM (The DCCT Research Group, 1991). Although short-term effects of mild hypoglycemia are troublesome for the individual, it is unlikely that episodes of mild hypoglycemia, if circumvented, pose much of a danger for individuals operating a motor vehicle. This is because signs of cognitive dysfunction generally begin to occur at plasma glucose concentrations around 50 mg/dl, which are below plasma glucose concentrations that initiate warning signs (Blackman, Towle, Lewis, Spire, and Polobsky, 1990; Ipp and Forster, 1987; Mitarkou, Ryan, Veneman et al., 1991; Wideom and Simonson, 1990). Severe hypoglycemic reactions, on the other hand, represent the most significant short-term danger for the diabetic individual and particularly if the episode occurs during driving.

Definitions of severe hypoglycemia vary and include hypoglycemia resulting in a seizure or a coma, reactions that require the intervention of another person, or a reaction that requires the administration of intravenous glucose, intramuscular glucagon, or hospitalization. Table 25 presents a summary of studies that have investigated the incidence of severe hypoglycemia in insulin treated diabetics. As can be seen, there is considerable variation among studies in the reported frequency of severe hypoglycemia, with estimates ranging from 0.04 to 1.7 episodes per patient per year.

The disparity can be attributed to a number of factors, including differences in study population, criteria for severe hypoglycemia, and degree of metabolic control. Generally, those studies employing the most restrictive criteria for hypoglycemic reactions tend to report the lowest incidence (Casparie and Elving, 1985; Goldstein, England, Hess, Rawlings, and Walker, 1981; Mulhauser, Berger, Sonnenberg, et al., 1985; Nilsson, Tideholm, Kalen, and Katzman, 1988; The DCCT Research Group, 1987). The very low rate (0.04 episodes per patient per year) reported by Goldstein et al. is most likely attributable to their unusual criterion measure in which severe hypoglycemia was defined as those episodes of hypoglycemia characterized by altered central nervous system function or prolonged autonomic symptoms. The incidence of severe hypoglycemia reported in the MacLeod, Hepburn, and Frier study of 1.7 episodes/person/year is similar to that reported in the 1991 study by Pramming and collegeaues (1.6 episodes/person/ year). Both studies used similar definitions of severe hypoglycaemia, which included episodes that involved external help including the sole administration of oral carbohydrates. Excluding those individuals from the analysis, the incidence rate in the MacLeod et al. investigation was 0.46 episodes per patient per year, a rate similar to that reported by the DCCT Research Group. Many of the reported rates shown in Table 25 are based on individuals from medical centers or clinics. Thus, it may be that the rates overestimate the incidence of severe hypoglycaemia, given that individuals attending clinics or medical centres may have more problems with hypoglycemia than the general diabetic population. Results from Songer, Lave, and LaPorte (1993) support this assumption. Songer et al. (1993), using unpublished population-based data from the Pittsburgh Epidemiology Diabetes Complication Study, place the incidence of severe hypoglycemia in the insulin-dependent diabetic population at 0.31 episodes per person per year. Severe hypoglycemia in the Pittsburgh investigation was defined as loss of consciousness. Extrapolating from the available data, the best estimate for the incidence of severe hypoglycemia in insulin-dependent diabetes is around 0.30 incidences/person/year.

Table 25  Summary of Studies on the Incidence of Severe Hypoglycemic Reactions in Individuals with Diabetes Mellitus

It is important to note that differing types of treatment regimes also may affect incidence rates. In recent years, management of IDDM has included efforts to achieve near normal glucose levels as a means of controlling or delaying chronic complications. However, tighter glycemic control has not been without adverse consequences. As can be seen from Table 25, results from the DCCT Research Group's one year feasibility study (1987) revealed a three-fold increase in the occurrence of hypoglycemia in individuals receiving intensive therapy(an insulin pump or three of more insulin injections per day) compared with conventional therapy (one or two insulin injections per day). Analysis of data from the first 45 months of the DCCT revealed that the incidence of severe hypoglycemia ranged from two to six times that observed with conventional therapy (The DCCT Research Group, 1991). As noted by this group:

The substantially increased risk for severe hypoglycemia that accompanies intensive therapy re-emphasizes the importance of determining the potential benefits and risks of efforts to maintain blood glucose at near normal levels in persons with IDDM. Although long-term benefits of these intensive efforts remain unproven, the present study indicates that efforts to maintain pre-meal and bedtime glucose levels between 70 and 120 mg/dl, using treatment methods employed in the DCCT, will at least double the risk of hypoglycemia with temporary neurologic impairment sufficient to preclude self-treatment.

The DCCT Research Group, 1991, pp. 458-459.

If future trends are for stricter glycemic control, the present estimate rates may seriously underestimate the incidence for severe hypoglycemia in the diabetic population. More importantly, doubling the risk of hypoglycemic reactions as a result of tighter glycemic control could have important implications for traffic safety.

The effect of hypoglycemia on driving is an important issue. However, information on the frequency and severity of hypoglycemic reactions while driving is scarce. Clark, Knight, Wiles, et al. (1982), in a retrospective questionnaire of 94 insulin treated diabetic drivers, found that 49 percent of the men and 19 percent of the women interviewed had, at some time, experienced symptoms of hypoglycemia while driving. In a survey by Stevens et al. (1989), approximately 30 percent of diabetic drivers receiving insulin reported recognizing hypoglycemic symptoms while driving. Forty percent of randomly selected patients with IDDM attending outpatient clinics in Auckland, Australia reported experiencing hypoglycemia while driving, and 13 percent attributed a crash to hypoglycemia (Ward, Stewart, and Cutfield, 1990). Eadington and Frier (1989) estimate that 15 percent of crashes involving diabetic patients may be attributable to hypoglycemia. In 12 percent of the sample studied by Stevens and his colleagues, hypoglycemia was felt to be the cause of a crash. Forty-six percent of those drivers reported experiencing hypoglycemic events two to five times during the year, and 13 percent of the sample reported having had a hypoglycemic event more than five times. Importantly, the number of hypoglycemic episodes while driving during the past year was associated with the total number of crashes experienced by drivers during the past five years (p = 0.03). Drivers with two or more hypoglycemic events in the last year were almost twice as likely to incur one or more crashes during a five-year period as compared to those diabetic drivers without a hypoglycemic episode.

Frier, Matthews, Steel, and Duncan (1980) surveyed 250 individuals with IDDM currently licensed to drive. Thirty-eight percent of the individuals who admitted to being involved in a crash since starting insulin treatment attributed the causal factor to hypoglycemia. The role of hypoglycemia was reported to play a substantially greater role in motor vehicle crashes in a study by Chantelau (1991). His data revealed that 60 percent of severe car crashes, based on self-reports of diabetic individuals treated with insulin, were most probably related to hypoglycemia according to the patient's own monitoring of blood glucose levels, monitoring of the emergency department immediately after the crash, or both. Therefore, despite the lack of data from large samples indicating an increased risk for motor vehicle crashes as a result of hypoglycemia, the available evidence from smaller studies reveals a positive relationship between hypoglycemic reactions and motor vehicle crashes.