Adapted from MacCuish (36), with permission of Edward Arnold (Publishers)
Adapted from MacCuish (36), with permission of Edward Arnold (Publishers)
hypoglycemia, a continuous intravenous infusion of dextrose and frequent oral feeding will be required. When hypoglycemic coma does not respond to intravenous dextrose, neuroimaging must be performed urgently to look for cerebral edema, a recognized complication of severe hypoglycemia that is associated with a high mortality, and to exclude other intracranial pathology. Treatment of cerebral edema includes intravenous mannitol and high dose corticosteroids in addition to supportive measures such as high flow oxygen, sedation, and if indicated, anticonvulsant therapy (36). It may take several days to recover from profound hypoglycemic coma.
The practical management of hypoglycemia (particularly when severe) should also include some retrospective analysis of why the episode had occurred and how it can be avoided in future. Prevention of hypoglycemia is an important aspect of management because fear of hypoglycemia interferes with the long-term maintenance of glycemic control.
Severe hypoglycemia is potentially dangerous and has a significant mortality and morbidity, particularly in older people with insulin-treated diabetes who often have premature macrovascular disease. The hemo-dynamic effects of autonomic stimulation may provoke acute vascular events such as myocardial ischemia and infarction, cardiac failure, cerebral ischemia, and stroke (6). In clinical practice the cardiovascular and cerebrovascular consequences of hypoglycemia are frequently overlooked because the role of hypoglycemia in precipitating the vascular event is missed. The VADT (37) and the ACCORD study (38) have demonstrated the risks of very strict glycemic control in a population at risk of vascular events. These studies examined the effects of glycemic control in people with type 2 diabetes who had several vascular risk factors or had overt cardiovascular disease. The group with very strict glycemic control had greater exposure to severe hypoglycemia, which inevitably has been implicated as the cause of increased mortality, although a direct causal association cannot be proven. Unfortunately, the inherent difficulty of demonstrating a causal relationship between a severe hypoglycemic event and a serious vascular outcome has obscured this putative association, and clinical evidence for the vascular consequences of hypoglycemia is mainly anecdotal (6).
The profuse secretion of catecholamines in response to hypoglycemia provokes a fall in plasma potassium and causes electrocardiographic (ECG) changes, which in some individuals may provoke a cardiac arrhythmia, several of which have been described (39). A possible mechanism that has been observed with ECG recordings during hypoglycemia is prolongation of the QT interval, which is associated with an increased risk of cardiac conduction defects and dangerous ventricular arrhythmias (39). Hypoglycemia-induced arrhythmias during sleep have been implicated as the cause of the "dead in bed" syndrome that is recognized in young people with type 1 diabetes (40).
Total cerebral blood flow is increased during acute hypoglycemia while regional blood flow within the brain is altered acutely. Blood flow increases in the frontal cortex, presumably as a protective compensatory mechanism to enhance the supply of available glucose to the most vulnerable part of the brain. These regional vascular changes become permanent in people who are exposed to recurrent severe hypoglycemia and in those with impaired awareness of hypoglycemia, and are then present during normoglycemia (41). This probably represents an adaptive response of the brain to recurrent exposure to neuroglycopenia. However, these permanent hypoglycemia-induced changes in regional cerebral blood flow may encourage localized neuronal ischemia, particularly if the cerebral circulation is already compromised by the development of cerebrovascular disease associated with diabetes. This may increase the risk of localized cerebral ischemia when the individual is exposed to other forms of hemodynamic stress that affect the cerebrovas-cular circulation. Transient ischemic attacks and hemiplegia are recognized manifestations of hypoglycemia and in the elderly may be misdiagnosed as evidence of cerebrovascular disease.
The main neurological consequences of severe hypoglycemia are coma and seizures, which exposes the affected patient to potentially serious morbidity, including fracture-dislocations, soft tissue injuries, and head injury. Around one quarter of all episodes of severe hypoglycemia result in coma (42). When electroencephalography (EEG) has been performed during hypoglycemia in awake resting subjects, a decrease in alpha waves, an increase in theta waves, and increased bursts of delta waves are observed, occurring diffusely over the cerebral cortex, but the EEG changes are more pronounced over the anterior part of the brain. Studies in adolescents with type 1 diabetes have demonstrated more pronounced abnormalities than in non-diabetic controls, with a greater tendency to epileptiform activity, and theta wave changes persist after blood glucose has recovered (43). Hypoglycemia-induced EEG changes can persist for days or become permanent, particularly after recurrent severe hypoglycemia, and this may interfere with investigations to exclude idiopathic epilepsy.
Focal neurological lesions as a consequence of acute hypoglycemia are rare, as is severe brain damage, which probably requires several hours of exposure to profound hypoglycemia. These isolated events are usually associated with deliberate insulin overdose or excessive alcohol consumption. However, transient and reversible neurological deficits have increasingly been demonstrated in individual cases with sophisticated neuroimag-ing techniques (44, 45), indicating that functional changes within the brain may be commonplace, without leaving a permanent structural abnormality. Structural abnormalities of the brain that are observed in the survivors of exposure to profound hypoglycemia include cortical and hippocampal atrophies and ventricular dilatation. Such individuals either exist in a vegetative state or have evidence of profound cognitive damage. In patients with profound and protracted hypoglycemic coma, prognosis can be difficult to determine, but if serum markers of brain damage, neurone-specific enolase and the protein S-100, become elevated within 24-48 h of the onset of the coma, this usually indicates a poor outcome (46).
Hypoglycemia increases the risk of falls in the elderly, many of whom are frail and have osteoporosis, which increases their vulnerability to hypoglycemia-induced injury and bony fractures. In addition to the effects of trauma, when hypoglycemia occurs during driving it can cause road traffic accidents, with resultant morbidity.
The frequency of fatal episodes of hypoglycemia is difficult to determine because of the problem of determining whether low blood glucose was present before death and is compounded by the inaccuracy of death certification. As alluded to earlier, many fatal cases in elderly patients are attributed to a cardiovascular or cerebrovascular event, and the precipitating effect of acute hypoglycemia is not identified.
Most studies in populations with type 1 diabetes suggest that the proportion of deaths caused by hypoglycemia is between 2 and 6%, which is lower than the mortality associated with ketoacidosis. Death may result suddenly from a cardiac arrhythmia or more slowly from profound neuroglycopenia causing severe cerebral damage. In the large British Diabetic Association Cohort Study of people who had developed type 1 diabetes before the age of 30, acute metabolic complications of diabetes were the greatest single cause of excess death under the age of 30; hypoglycemia was the cause of death in 18% of males and 6% of females in the 20-49 age group (47).
Chronic exposure to hyperglycemia has long been recognized to be the major cause of vascular disease in diabetes and the severity of microangio-pathic complications is directly related to the quality of glycemic control (14). Conversely, recurrent exposure to hypoglycemia of any severity causes syndromes that are both hypoglycemia-related and impair the capacity of the individual to respond to this metabolic stress, further increasing the risk of severe hypoglycemia and effectively creating a vicious circle that is difficult to break. These syndromes of counterregulatory hormonal deficiencies and impaired awareness of hypoglycemia (IAH) develop over a period of years and ultimately affect a substantial proportion of people with type 1 diabetes and a lesser number with insulin-treated type 2 diabetes. They are considered to be components of hypoglycemia-associated autonomic failure (HAAF), through down-regulation of the central mechanisms within the brain that would normally activate glucoregulatory responses to hypoglycemia, including the release of counterregulatory hormones and the generation of warning symptoms (48).
Counterregulatory deficiencies: The glucagon secretory response to hypoglycemia becomes diminished or absent within a few years of the onset of insulin-deficient diabetes. With glucagon deficiency alone, blood glucose recovery from hypoglycemia is not noticeably affected because the secretion of epinephrine maintains counterregulation. However, almost half of those who have type 1 diabetes of 20 years duration have evidence of impairment of both glucagon and epinephrine in response to hypoglycemia (49); this seriously delays blood glucose recovery and allows progression to more severe and prolonged hypoglycemia when exposed to low blood glucose. People with type 1 diabetes who have these combined counterregulatory hormonal deficiencies have a 25-fold higher risk of experiencing severe hypoglycemia if they are subjected to intensive insulin therapy compared with those who have lost their glucagon response but have retained epinephrine secretion (50, 51). In people with type 1 diabetes, counterregu-latory deficiencies co-segregate with IAH (52), suggesting that they share a common pathogenetic mechanism within the brain.
Impaired awareness of hypoglycemia (IAH): The awareness of hypo-glycemia can be defined as the perception of the initial symptoms of falling blood glucose (11). This may become diminished or blunted in people with insulin-treated diabetes, particularly with type 1 diabetes, and is thought to result from diminished sympathoadrenal activation producing a resultant reduction in the autonomic symptom response to a given level of hypo-glycemia (11). Impaired awareness is not an "all or none" phenomenon. "Partial" impairment of awareness may develop, with the individual being aware of some episodes of hypoglycemia but not others (53). Alternatively, the intensity or number of symptoms may be reduced, and neuroglycopenic symptoms predominate. This gradually progresses to loss of awareness where the patient is no longer aware of the onset of hypoglycemia, although total absence of any symptoms, albeit subtle, is very uncommon, a fact that underpins the use of "Blood Glucose Awareness Training" that has been devised to help people with IAH (54).
Several mechanisms have been proposed as the cause of IAH, including chronic exposure to low blood glucose as in strict glycemic control (similar to the effects of insulinoma in non-diabetic patients), the effects of antecedent hypoglycemia, and HAAF (11). The glycemic thresholds for generation of symptoms, counterregulatory hormone secretion, and cognitive impairment are all re-set at lower blood glucose levels, as a manifestation of cerebral adaptation to repeated low blood glucose. Although affected individuals are able to function normally at low blood glucose levels, the window of opportunity to identify hypoglycemia and take avoiding action becomes very narrow, and severe neuroglycopenia can rapidly ensue (11).
IAH affects 20-25% of patients with type 1 diabetes (11, 55) and less than 10% with type 2 diabetes (24), becomes more prevalent with increasing duration of diabetes (12) (see Fig. 3), and predisposes the patient to a sixfold higher risk of severe hypoglycemia than people who retain normal awareness (56). When IAH is associated with strict glycemic control during intensive insulin therapy or has followed episodes of recurrent severe hypo-glycemia, it may be reversible by relaxing glycemic control or by avoiding further hypoglycemia (11), but in many patients with type 1 diabetes of long duration, it appears to be a permanent defect.
Central autonomic failure: Because counterregulatory deficiencies and IAH usually co-exist and are associated with an increased frequency of severe hypoglycemia, the concept of a "hypoglycemia-associated autonomic failure" (HAAF) (Fig. 4) was developed by Cryer (48), who argued that recurrent severe hypoglycemia is the primary problem which provokes these acquired abnormal responses, through cerebral adaptation to recurrent or chronic exposure to low blood glucose concentrations. It is reasoned that antecedent hypoglycemia in people with type 1 diabetes (48) and those with type 2 diabetes who have progressed to pancreatic beta-cell failure (57) causes defective glucose counterregulation in the absence of glucagon secretion because the epinephrine response is then markedly attenuated during exposure to subsequent hypoglycemia, while IAH develops through blunting of the sympathoadrenal response and reduced generation of autonomic symptoms (58).
Effects of Hypoglycemia on Everyday Life
For the person with insulin-treated diabetes, hypoglycemia is a fact of life and is indisputably the greatest barrier to maintaining good glycemic control. Because it is common, unpredictable, and potentially dangerous, it impinges on every aspect of daily existence and provokes fear and apprehension in those with diabetes and in their relatives. Fear of hypoglycemia causes anxiety and psychological distress and is so great in some people that it provokes behavioral change and may negatively influence their approach to self-management (59). This may underlie the resistance shown by some patients to therapeutic recommendations to intensify treatment to achieve strict glycemic control. Hypoglycemia can affect personal relationships (sometimes causing marital tension), employment prospects, driving, recreational activities including exercise and sport, travel, and holidays, and acceptance for insurance. Detailed accounts of the everyday problems of living with hypoglycemia, and how these may be addressed, are available elsewhere ( 60, 61).
Driving is a very common everyday activity that requires complex psychomotor abilities, which are particularly vulnerable to the effects of acute neuroglycopenia and merits further comment. Driving simulators have been used to examine the effects of hypoglycemia on the driving performance of people with type 1 diabetes and have shown that driving performance becomes impaired when arterialized blood glucose is lowered below 3.8 mmol/l (68 mg/dl) (62). Two thirds of the subjects did not recognize that their blood glucose was low while driving so took no corrective treatment, and fewer than 25% of participants were aware that their driving performance was impaired during hypoglycemia, typical manifestations of which were speeding and inappropriate braking, driving off the road or crossing the midline, ignoring "Stop" signs, and an increased number of "crashes."
Insulin-treated drivers should be informed about how to avoid hypoglycemia when driving, and what to do if this occurs. Typical recommendations are shown in Table 7, yet many drivers persistently ignore this advice. In a specialist diabetes center where information and education about driving and hypoglycemia are provided routinely to drivers when they commence treatment with insulin, a survey of 202 drivers with insulin-treated diabetes showed that around half never tested their blood glucose before driving, and only 14% did this with any regularity (mostly people with IAH) (63). Other deficiencies in safe practice for driving included not carrying carbohydrate for emergency use in the vehicle, not stopping the car if hypoglycemia developed, and believing that blood glucose values below 3.0 mmol/l (54 mg/dl) were safe for driving (63). A major education problem exists to persuade drivers to implement these simple practical measures for safe driving, and regular reinforcement is necessary.
The occurrence of hypoglycemia while driving is a recognized cause of road traffic accidents, although the frequency is difficult to determine. Dia-betologists are often required to provide medical reports about the likelihood of hypoglycemia having caused or contributed to a motor accident for legal and insurance purposes. Evidence is often circumstantial and it may be
Driving and insulin-treated diabetes: precautions to avoid and treat hypoglycemia
Prevention of hypoglycemia
• Test blood glucose before driving and every 2 h during long journeys
• Take regular snacks/meals
• Take regular breaks or rest periods
• Avoid drinking alcohol
• Keep an accessible emergency supply of fast-acting carbohydrate and a supply of food in the vehicle
• Stop the vehicle in a safe location
• Consume fast-acting carbohydrate, followed by a snack
• Defer driving for at least 45 min to allow cognitive recovery difficult to prove a causal relationship between suspected hypoglycemia and a driving accident. Taking a careful history, particularly of events preceding the incident, is essential when assessing the potential role of hypoglycemia in any individual case.
The modern management of diabetes strives to achieve strict glycemic control using intensive therapy to avoid or minimize the long-term complications of diabetes; this strategy tends to increase the risk of hypoglycemia and promotes development of the acquired hypoglycemia syndromes. Prevention of hypoglycemia therefore requires education of patients (and their relatives) when they commence treatment with insulin (and also those using sulfony-lureas) and regular review and reinforcement about its avoidance and treatment, with reference to various activities and circumstances. This includes setting realistic targets for blood glucose in individual patients, which, for pragmatic reasons, may have to be in a higher range in some groups of patient such as the very young, the elderly, and those with advanced complications or other co-morbidities such as ischemic heart disease, as suggested by recent trials (38, 37). A higher than desirable frequency of hypoglycemia may have to be tolerated in certain situations, such as pregnancy, where strict glycemic control is essential for fetal well-being.
Frequent monitoring of blood glucose is necessary to identify asymptomatic biochemical hypoglycemia, particularly when median blood glucose is within a normoglycemic range (28) and when patients have developed impaired awareness of hypoglycemia. There is a pressing need for an effective glucose sensor with an alarm system to detect (nocturnal) hypoglycemia during sleep (26). In addition, a long-acting or slowly absorbed carbohydrate supplement would be valuable, which can be ingested at bedtime and will act throughout the night. To be effective this would have to counteract nocturnal hyperinsulinemia and prevent a fall in blood glucose without compromising glycemic control or encouraging weight gain. Hypoglycemia that recurs at a particular time of day may be eradicated by changing the insulin regimen or the timing and dose of a particular insulin - such as changing the timing of the long-acting (basal) insulin to lower the risk of nocturnal hypoglycemia (26, 27). Claims that insulin analogs (rapid-acting and long-acting) have substantially lowered the risk of hypoglycemia (other than at night) have not withstood critical scrutiny (64), and hypoglycemia remains just as great a risk today with modern, multiple injection insulin regimens, as it was in the past. The increasing use of continuous subcutaneous insulin delivery with insulin pumps (which may be associated with a lower risk of severe hypoglycemia (65)) and real-time continuous glucose monitoring systems should improve hypoglycemia detection as glucose-sensing technology improves; more frequent information about prevailing blood glucose would then be available than can be provided by intermittent single-point monitoring (66).
Was this article helpful?