Two older diabetics suffer permanent brain damage, one from human insulin, one from glipizide caused hypoglycemia in hospital.

Iatrogenic Hypoglycemia and Malpractice Claims


ABSTRACT: Iatrogenic neuroglycopenia represents a medical emergency for which accepted and usually effective treatment exists. Treating physicians face a malpractice risk if there is permanent neurologic damage and there is a perception of failure to act prudently both before and after the damaging episode. We report two cases to illustrate the medical and legal issues that may be raised in such claims. [Strangely, the known Humulin/human insulin problems aren't even raised in defense in either of these two cases, but then far more of the elderly are forced to take synthetic insulins than any of the various oral anti-diabetic agents such as the glipizide in case 2 We refer interested readers to the article on the bad record of human insulin with older diabetics. Naturally, both settled out of court to prevent publicity. Ed]

A REVIEW of Indiana malpractice claims involving care of diabetic patients during a 13-year period (35 claims) found none based directly on hypoglycemic brain damage.1 We report two malpractice claims brought against internists in the southeastern United States after insulin- and glipizide-induced neuroglycopenia with permanent sequelae occurred.


Case 1. A 75-year-old man with insulin-treated [ diabetes was hospitalized after an episode of disorientation. He was taking 50 units of NPH insulin at 8 AM and 30 units at 7 PM. He had a 10-year history of hypertension controlled by a low-salt diet and 250 mg of chlorothiazide daily. There was no history of neurologic or ophthalmologic abnormality. On admission, he was described as malnourished. His vital signs were normal. He was alert and in no distress. Neurologic and ophthalmologic examinations were unremarkable apart from absent ankle reflexes. There were no sensory abnormalities. No carotid bruits or cardiac murmurs were heard. Laboratory studies showed a fasting blood glucose level of 90 mg/dL, serum albumin value of 2.6 mg/dL, BUN value of 35 mg/dL, and creatinine value of 2.4 mg/dL, with 1+ albuminuria.

The next day computed tomography (CT) of the brain showed an old right temporal cerebral infarct. That afternoon the patient had an episode of slurred speech, which cleared with the administration of orange juice. The blood glucose concentration was not measured. The next evening at 8 o'clock he became lethargic, speech was again slurred, and his face was drawn to one side. A blood glucose value of 32 mg/dL was obtained before the administration of a bolus of 50% glucose. His mentation cleared immediately and neurologic signs disappeared. An order was written to continue 5% glucose in water at 100 mL/hour overnight. However, the intravenous infusion infiltrated at midnight and was discontinued by the nursing staff. The physician was not called.

At 6:30 the next morning, the patient could not be aroused. The blood glucose value was 45 mg/dL, and a bolus of 50% glucose was given intravenously, followed by a constant infusion of 5% glucose. During the next 30 minutes, he gradually became alert and after 1 hour was able to eat breakfast. He complained of partial loss of vision bilaterally and color blindness. These losses were permanent. An ophthalmologist later documented abnormal visual evoked potentials and suggested neuronal degeneration due to hypoglycemia as the cause.2 Repeated CT of the brain was unchanged at 6 weeks. A claim citing failure to protect against recurrent hypoglycemia and substandard treatment after it occurred was made against the attending physician and the hospital. The case was settled before trial.

Case 2. A thin 69-year-old woman required nursing home placement due to Alzheimerís disease. She had a history of diabetes controlled by diet, mild hypertension, and stable angina. She had renal insufficiency with a BUN value of 40 mg/dL and creatinine value of 2.2 mg/dL. During her second year in the home, her fasting blood glucose values rose to 150 to 180 mg/dL, and 2.5 mg daily of glipizide was prescribed. Two weeks later, she had a hypoglycemic episode manifested by confusion and upper respiratory distress. She did not lose consciousness. The blood glucose value was found to be 19 mg/dL, and an intravenous bolus of 50% glucose was administered, followed by an infusion of glucose for 4 hours. She responded immediately to the IV bolus of glucose with a return to normal mental state. Glipizide therapy was discontinued, and she remained in the nursing home. Two days later, she was hospitalized because of a fecal impaction with rectal bleeding. At midnight of the next day, she was found to be confused with slurred speech. Blood glucose measured 30 mg/dL. She responded to a bolus of 50% glucose followed by 5% glucose in water through the night. There were no recognized sequelae to these episodes.

Four months later, a random blood glucose value of 278 mg/dL was reported to a physician providing weekend call to the nursing home. He was unaware of the earlier hypoglycemic episodes and gave a telephone order for 5 mg of glipizide daily, to be followed by a fasting blood glucose measurement in 1 week. In the late afternoon 3 days later, the patient became unresponsive to verbal stimuli. She was given 1 unit of glucagon subcutaneously and became alert. She ate dinner and a late snack, but at 6 oíclock the next morning she was again unresponsive. An infusion of glucose in half-normal saline was begun, which led to pulmonary edema. She was hospitalized, and the admitting chest film showed pulmonary congestion and a pneumonic infiltrate at the left base. She never regained consciousness and died 5 days later. The autopsy showed patchy areas of necrosis in the brain with the distribution characteristic of neuroglycopenic damage.3,4 There was evidence of coronary artery disease, with a healed myocardial infarction, and bilateral bronchopneumonia. A claim of malpractice was made against the attending physician, the nursing home director, and the nursing home. It cited the manner in which glipizide therapy was reinstituted despite the patient's history of vulnerability, and the failure to provide adequate treatment, including timely hospitalization. The case was settled during the trial.


Both cases raised legal issues involving alleged negligence and causation. Negligence was cited first in the failure to protect these patients from neuroglycopenia. Juries are easily educated about the lesser importance of a hormonal response to hypoglycemia as contrasted with one eliciting neurologic signs, and the plaintiff experts cited the general vulnerability of both patients to neuroglycopenia due to their age, poor nutritional status, and renal insufficiency. Equally harmful to the defense were the episodes of hypoglycemia with neurologic signs that preceded the permanent damage. Physicians are not expected to anticipate every complication, but once harbingers are present, as in these cases, physicians have difficulty in proving that they continued to provide prudent care. Both cases also cited substandard care after the appearance of neurologic signs in the final episodes. Substandard care is readily recognized when a physicianís orders are not followed (Case 1). Substandard treatment in Case 2 was the failure to provide an immediate bolus of 50% glucose and provide the conditions for a constant glucose infusion given that the patient had already shown a tendency to relapse 3 days after discontinuance of glipizide therapy. Pertinently, the court exhibits included a page from a standard textbook,5 which stated that all cases of sulfonylurea-induced severe hypoglycemia require hospital admission with blood glucose monitoring for at least 3 days.

It has been known for decades that neuroglycopenia has specific deleterious effects on the central nervous system after administration of insulin6 and sulfonylurea.3,7 The retina (Case 2) is vulnerable to hypoglycemia because it has an unusually active metabolism, particularly on the part of the photoreceptor cells. As in the brain, these needs are met by the circulating blood glucose, with only a limited quantity of glycogen available (most of it in the Müller cells). The defense in both cases conceded the importance of adequate cerebral access to glucose but emphasized the element of causation and the possibility that the iatrogenic falls in blood glucose played no role in the adverse outcomes. Proximate cause requires reasonable foreseeability and cause in fact, and the possibility that the damages were due to ischemia rather than hypoglycemia was raised in both patients. This defense was supported by the patientsí histories of hypertension and cardiovascular disease, as well as the CT evidence of a past stroke in Case 1. However, this defense was weakened by the temporal relationship between the neurologic damage and the hypoglycemia and by the findings on ophthalmologic and autopsy examinations. Brain damage due to neuroglycopenia shows a distribution pattern that differs from that of ischemia, with greatest neuronal loss in the superficial and paraventricular layers of the cerebral cortex and hippocampus, especially the dentate gyrus of the latter. Ischemia, on the other hand, causes selective loss of one type of neuron depending on the brain region involved and also affects glial and endothelial cells.8 Although not pertinent in our cases, it has been pointed out that the presence of ischemic change does not in itself rule out hypoglycemic brain damage; infarction may well develop secondarily in the recovery period after hypoglycemia due to tissue lactic acidosis after rapid glucose administration.4,8

Physicians who treat patients with diabetes may face claims that involve hypoglycemia indirectly. The Indiana claims review1 included three patients who had indirect complications (eg, falls) attributed to hypoglycemia. An interesting example of indirect legal exposure is Pittman v Upjohn C.9 Without his grandmotherís knowledge, 26-year-old Pittman mistook her prescription Micronase (glyburide) tablets for aspirin and had neuroglycopenic brain damage as a result. The patientís representative sued the prescribing physician for failure to warn the grandmother of the danger of Micronase if it were to be consumed by someone other than herself.

We see narrow limits to the risk management steps physicians can take for these indirect claims. However, our cases do illustrate the risk management value of a prompt, complete, and well-documented medical response when a physician is called to treat neuroglycopenia. In the cases we have reported, neither party made reference to current guidelines for the management of hypoglycemia,10 and it remains to be seen what role guidelines will play in future suits.


1. Clark CM Jr, Kinney ED: The potential role of diabetes guidelines in the reduction of medical injury and malpractice claims involving diabetes. Diabetes Care 1994; 17:155-159

2. Harrad RA, Cockram CS, Plumb AP, et al: The effect of hypoglycaemia ( hypoglycemia , insulin shock, insulin reaction ) on visual function: a clinical and electrophysiological study. Clin Sci 1985; 69:673-679

3. Kalimo H, Olsson Y: Effects of severe hypoglycemia on the human brain-neuropathological case reports. Acta Neurol Scand 1980; 62:345-356

4. Auer RN: Progress review: hypoglycemic brain damage. Stroke 1986; 17:699-708

5. Rifkin H, Porte D: Diabetes Mellitus. New York, Elsevier Press, 1990, 4th Ed, p 569

6. Lawrence RD, Meyer A, Nevin S: The pathological changes in the brain in fatal hypoglycaemia ( hypoglycemia , insulin shock, insulin reaction ). Q J Med 1942; 11:181-201

7. Vital CI, Picard J, Arne L, et al: Pathological study of three cases of hypoglycemic encephalopathy (one of which occurred after sulfamidotherapy). Le Diabete 1967; 12F:291-296

8. Helgason CM: Blood glucose and stroke. Stroke 1988; 19:1049-1053

9. Basanta WE: Physician liability to persons other than patients. Tort Insurance Law J 1996; 31:357-361 (890 SW2d 425 Tenn 1994).

10. Herman W (ed): The Prevention and Treatment of Complications of Diabetes Mellitus. A Guide for Primary Care Practitioners. Atlanta, Ga, Centers for Disease Control, Department of Health and Human Services, Public Health Service, 1991, pp 16-17

Back to February INDEX