Dr
W Stephen Waring, Lecturer in Medicine, Clinical Pharmacology Unit, The
University Of Edinburgh
Dr William D Alexander, Consultant Physician, Department of Diabetes and Metabolic Medicine, Western General Hospital, Edinburgh
Correspondence to: s.waring@ed.ac.uk
SMJ 2004 49(3): 105-107
Abstract
We present the case of an elderly non-diabetic female who was admitted to hospital as an emergency due to loss of consciousness. Her clinical presentation was consistent with hypoglycaemia due to a massive insulin overdose. However, the patient refuted the possibility of insulin administration, and the circumstances were reported to the police for investigation. This case demonstrates the clinical and biochemical characteristics of insulin overdose. Furthermore, it serves to illustrate the sequence of events that may be created when foul play is suspected, and the factors related to patient confidentiality that require consideration by the responsible physician.
Case Report
A 71-year-old woman was found by her neighbour to be lying on the floor and unresponsive, and emergency medical services were summoned. On arrival at hospital she was found to have Glasgow Coma Scale of 3, and was moderately dehydrated. Rectal temperature was 34.0ºC, pulse rate 110 bpm, and blood pressure 130/70 mmHg. Respiratory rate was 25/min, and peripheral oxygen saturation was 98%. Tone and reflexes were symmetrical in all limbs, and there were no signs of trauma, raised intracranial pressure or sepsis. Medical history obtained from hospital case notes indicated breast carcinoma and left mastectomy 23 years earlier, mild asthma, and a transient ischaemic attack 3 years ago; regular medications were pulmicort and terbutaline metered-dose-inhalers. A BM strip blood glucose was 0.6 mmol/l and, therefore, pending results of a formal laboratory measurement, 50 ml 50% dextrose was administered intravenously followed by infusion of 10% dextrose at 125 ml/h. Biochemical and haematological investigations showed plasma glucose = 2.6 mmol/l, sodium = 131 mmol/l, potassium = 3.4 mmol/l, urea = 5.0 mmol/l, creatinine = 45 µmol/l, white cell count = 19.8 x109/l (neutrophils 17.5 x109/l), erythrocyte sedimentation rate = 6 mm/hour, and C-reactive protein < 0.7 mg/dl. Liver and thyroid biochemistry, serum calcium and magnesium, clotting screen, arterial blood gas analysis, electrocardiogram, chest radiograph, and cranial and abdominal computed tomography were found to be normal. A repeat BM strip blood glucose was < 2 mmol/l, despite administration of 10% dextrose by infusion. An additional bolus of 50 ml 50% dextrose was administered, which caused rapid improvement in her conscious state, and she became alert and orientated. Repeat clinical examination was normal, and no focal neurological abnormality was identified. Maintenance of normal plasma glucose concentrations proved difficult over the next 48 hours, such that 17 further doses of 50 ml 50% dextrose and continuous infusion of 20% dextrose at 125 ml/h were required to prevent hypoglycaemia.
Analysis of serum collected on admission showed insulin = 8430 mU/l (58.5 nmol/l; normal reference range < 15.6 mU/l) and C-peptide < 0.05 nmol/l. Sulphonylurea compounds were not detected in urine collected on admission, and serum cortisol concentration was 738 nmol/l. By the fourth day after admission to hospital, blood glucose and plasma insulin concentrations had normalised (Figures 1 and 2). On further examination, a small bruise had become apparent over the left flank, consistent with the site of a recent subcutaneous injection. These findings overwhelmingly suggested that hypoglycaemia had been caused by exogenous insulin administration1. We considered that the patient may have self-administered insulin, either inadvertently or as a deliberate act. However, our patient strongly rejected this possibility. She informed us that she lived alone and had no means of obtaining insulin, but her account of the events immediately preceding hospital admission was vague. We were concerned because of the possibility that, if insulin had not been self-administered, she may have been subject to a malicious act by a third party. In view of this, and with the patients consent after further discussion of the potential implications, we referred the matter to the police for further investigation. During an interview in hospital, our patient revealed to the police that she had sought the assistance of a third party to enable her suicide by means of massive insulin administration. She subsequently admitted to us that she had been feeling depressed, and cited financial problems as a major contributing factor. Psychiatric evaluation found evidence of underlying depression, but no on-going suicidal ideation. She was commenced on sertraline 50 mg daily and discharged home, with arrangements made for a Community Psychiatric Nurse to provide additional support at home. Shortly after, on review in the General Medicine outpatient department she was found to have made a good recovery with no residual neurological abnormality.
Discussion
Hypoglycaemia can be characterised as reactive (post-prandial), fasting, or drug-related2. Reactive hypoglycaemia can occur where there is rapid onset and fall-off of post-prandial plasma glucose concentrations, for example in patients with disordered gastrointestinal motility, or as a complication of gastrectomy. Fasting hypoglycaemia can arise in the setting of hepatic failure due to impaired glycogenolysis, or due to adrenocorticoid deficiency. The high random cortisol concentration obtained at the time of hospital admission was consistent with an appropriate stress response in this patient3. Relative hyperinsulinaemia can indicate prior administration of hypoglycaemic drugs, or the presence of a functional insulinoma4. In health, insulin and C-peptide are cleaved from pro-insulin and secreted in equimolar concentrations by pancreatic b-cells. Insulin undergoes rapid hepatic metabolism, whereas C-peptide is subject to slower renal clearance, and a typical physiological range for the plasma insulin: C-peptide molar ratio is 0.12 – 0.47. C-peptide is removed during the purification of pharmaceutical insulin preparations. Therefore, where hepatic and renal function are normal, a plasma insulin: C-peptide ratio > 1 strongly suggests exogenous insulin administration5. Sulphonylureas stimulate b-cells, and thereby release both insulin and C-peptide. They can be identified as a cause of hypoglycaemia by direct drug demonstration in urine6. Rarely, soft tissue sarcomata secrete insulin-like factors that can mimic the metabolic effects of insulin7. A discreet insulinoma, or soft tissue sarcoma were viewed as unlikely in view of the normal appearance of an abdominal computed tomography scan.
Glucose is the main metabolic substrate of the central nervous system, which is incapable of gluconeogenesis. Transient hypoglycaemia can cause impaired cognition, headache, and visual disturbance and, if persistent, may result in significant neurological sequelae or be fatal8,9. Under normal circumstances, hypoglycaemia is prevented by glycogenolysis and gluconeogenesis, which are stimulated by a variety of hormones, including glucagon, catecholamines, and cortisol. However, these compensatory mechanisms are completely suppressed in the presence of high circulating insulin concentrations. Therefore, availability of substrate is completely dependent on exogenous administration, and the goal of immediate management is to maintain adequate plasma glucose concentrations between 5 and 10 mmol/l to preserve central nervous system function10. The present case illustrates that hypoglycaemia may persist for several days after an insulin overdose depending on the type of insulin administered, in line with previous reports11,12. This appears to be the result of delayed absorption kinetics13. A total of 1745 g of dextrose was administered to our patient to correct hypoglycaemia, equivalent to 6980 kCal. Hepatic steatosis is a recognised complication of prolonged glucose administration that requires consideration after insulin overdose14.
Until the biochemical results were obtained, insulin overdose was considered an unlikely cause of unconsciousness at initial presentation in this elderly non-diabetic patient with established cerebrovascular disease. BM strip techniques are notoriously unreliable for measuring low glucose concentrations, and are particularly inaccurate in the setting of acute shock15. The elevated white blood cell count normalised soon after admission, and no source of infection was identified. We believe this was due to stress neutrophilia, which is characteristically caused by mobilisation of peripheral neutrophils in the setting of catecholamine excess, recent trauma or seizures16.
The care of this patient introduced several problems that were outside our normal professional training and experience as physicians. Our patient consistently denied self-administration of insulin, and we were then faced with the possibility that she may have unknowingly been subjected to administration by a third party with malicious intent. A number of such cases have previously been reported, both in a community17 and hospital environment18. We were concerned that she may be at continued risk if discharged from hospital care. We discussed our concerns with the patient, and sought her consent before reporting the matter to the police. The validity of informed consent depends on the patient understanding the potential consequences of providing or withholding consent, and having sufficient autonomy to make the decision freely. We recognised that the institutional setting may have had an important influence19,20, and were careful to give a detailed explanation of the clinical information we wished to disclose to the police, and indicated that an ensuing investigation could reveal incriminating evidence.
At present, doctors have a professional and moral responsibility to observe patient confidentiality, and no statutory law requires or permits doctors to report patients who make plausible allegations that a crime has been, or is likely to be committed. The General Medical Council indicates that, in certain situations, confidentiality is not an absolute obligation, and breeches are acceptable if there is wider public interest in preventing harm to others or preventing crime21. The decision to breach confidentiality can be difficult, and is often based on a variety of subjective and objective factors. We considered that breach of patient confidentiality to the police would have been justified in this case, because failing to do so may have put the patient, or another person, at a significant risk. In addition to the duty of care to individual patients, the importance of a wider societal responsibility is increasingly recognised22.
The source, dose and type of insulin, and means of administration in this patient remain unclear. The case serves to highlight the possibility that insulin overdose may be used as a means of attempted suicide, even in non-diabetic patients with no obvious access to insulin. We are also reminded that the possibility of assisted self-harm must not be overlooked in older patients.
References
1. Lebowitz MR, Blumenthal SA. The molar ratio of insulin to C-peptide. An aid to the diagnosis of hypoglycemia due to surreptitious (or inadvertent) insulin administration. Arch Intern Med 1993;153:650-5
2. Service FJ. Classification of hypoglycemic disorders. Endocrinol Metab Clin North Am 1999;28:501-17
3. Miller DB, O'Callaghan JP. Neuroendocrine aspects of the response to stress.
Metabolism 2002;51(Suppl 1):5-10
4. Field JB. Hypoglycemia. Definition, clinical presentations, classification, and laboratory tests. Endocrinol Metab Clin North Am 1989;18:27-43
5. Roberge RJ, Martin TG, Delbridge TR. Intentional massive insulin overdose: recognition and management. Ann Emerg Med 1993;22:228-34
6. Klonoff DC, Barrett BJ, Nolte MS, Cohen RM, Wyderski R. Hypoglycemia following inadvertent and factitious sulfonylurea overdosages. Diabetes Care 1995;18:563-7
7. Hunter SJ, Daughaday WH, Callender ME et al. A case of hepatoma associated with hypoglycaemia and overproduction of IGF-II (E-21): beneficial effects of treatment with growth hormone and intrahepatic adriamycin. Clin Endocrinol 1994;41:397-401
8. Flood JG, Williams TD, Bacon RC. Full recovery following massive overdose with insulin and thyroxine. Br J Clin Pract 1990;44:747-8
9. Cooper AJ. Attempted suicide using insulin by a non diabetic: a case study demonstrating the acute and chronic consequences of profound hypoglycemia. Can J Psychiatry 1994;39:103-7
10. Samuels MH, Eckel RH. Massive insulin overdose: detailed studies of free insulin levels and glucose requirements. J Toxicol Clin Toxicol 1989;27:157-68
11. Fasching P, Roden M, Stuhlinger HG et al. Estimated glucose requirement following massive insulin overdose in a patient with type 1 diabetes. Diabet Med 1994;11:323-5
12. Shibutani Y, Ogawa C. Suicidal insulin overdose in a type 1 diabetic patient: relation of serum insulin concentrations to the duration of hypoglycemia. J Diabetes Complications 2000;14:60-2
13. Bayly GR, Ferner RE. Persistent insulin secretion after insulin overdose in a non-diabetic patient. Lancet 1993;341:370
14. Jolliet P, Leverve X, Pichard C. Acute hepatic steatosis complicating massive insulin overdose and excessive glucose administration. Intensive Care Med 2001;27:313-6
15. Atkin SH, Dasmahapatra A, Jaker MA, Chorost MI, Reddy S. Fingerstick glucose determination in shock. Ann Intern Med 1991;114:1020-4
16. Stock W, Hoffman R. White blood cells 1: non-malignant disorders. Lancet 2000;355:1351-7
17. Bauman WA, Yalow RS. Hyperinsulinemic hypoglycemia. Differential diagnosis by determination of the species of circulating insulin. JAMA 1984;252:2730-4
18. Levy WJ, Gardner D, Moseley J, Dix J, Gaede SE. Unusual problems for the physician in managing a hospital patient who received a malicious insulin overdose. Neurosurgery 1985;17:992-6
19. Samuels A. Informed consent: the law. Med Sci Law 1992;32:35-42
20. Silverman WA. The myth of informed consent: in daily practice and in clinical trials. J Med Ethics 1989;15:6-11
21. General Medical Council. Confidentiality: Protecting and Providing Information General Medical Council, London, September 2000
22. Shepherd J. Doctors have a community responsibility. Br Med J 1997;315:1298