Elizabeth C.
McGeough and James D Walker*,
Department of Medicine,St John’s Hospital, Livingston, EH54 6PP.
SMJ 2006 51(1): 57
Abstract
Hypoglycaemia can be associated with a number of clinical conditions in patients without diabetes. It can be associated with severe sepsis and hypoglycaemia complicating septicaemia may not always be recognised. We report two cases of spontaneous hypoglycaemia complicating severe sepsis and summarise the literature.
Introduction
Hypoglycaemia may
complicate a number of medical conditions in patients without diabetes mellitus.
These include congestive cardiac failure, chronic renal failure, liver disease
and malignancy. Hypoglycaemia is also a recognised complication of severe
sepsis, although this is less commonly reported. We describe two cases of severe
hypoglycaemia in the context of septicaemia.
Case report 1
A 73-year-old woman presented with increasing dyspnoea, peripheral oedema, uncontrolled atrial fibrillation and congestive cardiac failure, secondary to ischaemic heart disease. She did not have diabetes. Treatment with diuretics, beta-blockers and anticoagulants resulted in a slow clinical improvement.
Ten days following admission she developed cellulitis of the right leg and was treated with oral flucloxacillin.
The following day she collapsed with hypotension, tachycardia and hypoxia with a reduced GCS. The cellulitis had worsened. She had a metabolic acidosis with a venous plasma glucose level of 0.5 mmol/l with a simultaneous serum cortisol of 10,680 nmol/l. Blood cultures grew gram-negative coliforms.
Her hypoglycaemia responded to boluses of intravenous 50% dextrose followed by a continual infusion of 5% dextrose. She was transferred to the High Dependency Unit. Despite full supportive management she developed multi-organ failure and died the following day.
Case
Report 2
A 56-year-old woman, with a long history of severe and deforming rheumatoid arthritis, was admitted for recuperation following the successful treatment of an MRSA infection of an olecranon bursa. She did not have diabetes.
On admission she was thin (BMI 17 Kg/m2), and had a large pressure sore over the sacrum. Over a 4-week period, her nutritional state improved and she began to mobilise.
In the early hours of her planned discharge day she was found in a collapsed state. She was cold, clammy, cyanosed and tachycardic. Her venous plasma glucose level was 0.2 mmol/l with a simultaneous serum cortisol of 794 nmol/l. Plasma insulin was appropriately suppressed with no sulphonyurea found in the urine. Blood cultures grew MRSA.
She
was resuscitated, treated with intravenous boluses of 50% dextrose and
transferred to the intensive care unit. Two days later she developed multi-organ
failure and died.
Discussion
Hypoglycaemia is recognised as a late complication of septic shock, and has been described in both gram-positive and gram-negative infections (1). The prevalence is difficult to quantitate however one study described hypoglycaemia in 18% of patients with sepsis (2). The association of hypoglycaemia and sepsis is associated with a poor prognosis with mortality rates of 67% and 90%, respectively with death often occurring within 24 hours of the hypoglycaemia (1,2).
It is unclear why some patients with septicaemia develop hypoglycaemia whilst others do not. The presence of liver disease and starvation have been found to be independent risk factors for the development of hypoglycaemia in sepsis (2) and this would certainly seem to hold true in our second case where the patient’s nutritional status was poor. In another study determining hypoglycaemia as a predictor of mortality in hospitalised patients, hypoglycaemia was found to occur more commonly in women, in those with a lower albumin, higher creatinine, or higher alkaline phosphatase level. Although hypoglycaemia was not an independent predictor of mortality, in this series, it did occur 10 times more frequently in those with sepsis (3).
The pathophysiology of hypoglycaemia in sepsis is likely to be due to a combination of mechanisms. Increased peripheral utilisation of glucose occurs in sepsis with reduced peripheral perfusion of tissues resulting in anaerobic metabolism. Gluconeogenesis is inhibited by metabolic acidosis (seen at the time of hypoglycaemia in the first case) and endotoxin has been shown to deplete glycogen store in the liver (1,2,4).
The two cases described above may not be exceptional. All patients acutely admitted to hospital must have a plasma glucose measurement made on admission and at other times when the clinical state demands it. This does not always occur (5). Septicaemia is not uncommon in acute medical and surgical wards and clinicians should be cognizant of this serious complication so as to deliver prompt treatment and avoid hypoglycaemia contributing to the already high mortality seen with septicaemia.
References
1. Miller SI, Wallace RJ, Musher DM, Septimus EJ, Kohl S, Baughn RE. Hypoglycaemia as a manifestation of sepsis. Am J Med 1980; 68: 649-659.
2. Rattarasarn C. Hypoglycaemia and Sepsis: Risk factors and Clinical Characteristics. J Med A Thailand 1997; 80: 760-766.
3. Kagansky N, Levy S, Rimon E, Cojocaru L, Fridman A, Ozer Z, Knobler H. Hypoglycaemia as a predictor of mortality in hospitalised elderly patients. Arch Intern Med 2003; 163: 1825 -1829
4. Walsh TD. Hypoglycaemia and Septicaemia. Postgrad Med J 1984; 60: 431- 432.
5. Hennessy AR, Reynolds RM, Walker JD. Blood glucose measurement in acute medicine: inadequate detection and management. Diabet Med 2002; 19: 698