Endocrine Abstracts (2012) 29 P1148

The incidence of hypoglycaemia in non-diabetic hospital patients outside critical care estimated by a capture-recapture technique: retrospective analysis

K. Nirantharakumar1,2, T. Marshall1, J. Hodson2, P. Narendran1,2, J. Deeks1, J. Coleman1,2 & R. Ferner1,3


1University of Birmingham, Birmingham, United Kingdom; 2University Hospital Birmingham, Birmingham, United Kingdom; 3City Hospital, Birmingham, United Kingdom.


Introduction: Unexplained or a cluster of non diabetic hypoglycaemia has at times been the only clue to malicious use of Insulin in hospitals. We wished to establish the incidence of hypoglycaemia in non-diabetic adult patients outside the intensive care unit (ICU) in a 1200-bed university hospital.

Methods: We retrospectively analysed data for 2010 from three distinct sources to identify patients: bedside and laboratory blood glucose measurements; medication records for those treatments (high-strength glucose solution and glucagon) commonly given to reverse hypoglycaemia; and diagnostic codes for hypoglycaemia. We excluded from the denominator admissions of patients with a diagnosis of diabetes or prescribed diabetic medication. The denominator included patients admitted to ICU, as they almost always have a period of stay outside ICU (susceptible population).Case notes of all patients identified were reviewed to ascertain hypoglycaemia and where it occurred. We used capture-recapture methods based on log linear models to establish the likely true rate of hypoglycaemia in non-diabetic in-patients outside ICU. Analysis was carried out at different cut-off points for hypoglycaemia. We also recorded co-morbidities that might have given rise to hypoglycaemia.

Results: Among the 37 898 admissions we identified a total of 71 admissions of non-diabetic patients with hypoglycaemia at or below 3.3 mmol/l and 37 admissions at or below 2.7 mmol/l. Using capture-recapture methods the estimated incidence was 50 (95% CI 33–93) and 13 (95% CI 11–19) per 10 000 admissions at 3.3 mmol/l and 2.7 mmol/l cut-off values. Admissions of patients aged above 65 years were 50% more likely to have an episode of hypoglycaemia. Commonest relevant co-morbidities linked to hypoglycaemia were sepsis, renal disease, alcohol dependence, pneumonia, liver disease, cancer and self harm with hypoglycaemic agents.

Conclusion: Hypoglycaemia is rare in non-diabetic patients, and usually associated with important co-morbidity. Our study suggests a way to monitor an unusual increase in hypoglycaemia in hospital.

Declaration of interest: The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research project.

Funding: This work was supported however funding details unavailable.

Article tools

My recent searches

No recent searches.