Inpatient hyperglycaemia can be an indicator of diabetes mellitus: Study

UK: A new study published in Diabetic Medicine suggests that the number of people who need to be screened to identify an individual who may have diabetes reduces as the screening threshold based on the peak in-hospital glucose concentration increases.

When brought to the hospital, many people with undiagnosed diabetes develop hyperglycemia. Inpatient hyperglycemia can be a sign of diabetes, but it can also be a simple stress reaction. As a result, Andrew J. Farmer and colleagues' study describes the extent to which an in-hospital maximum recorded random glucose test predicts the necessity for an in-hospital (or subsequent) HbA1c assessment to look for undetected diabetes.

HbA1c, blood glucose, age, and gender were gathered from all people admitted to a UK NHS trust hospital between 1 January 2019 and 31 December 2020. The analysis was limited to patients who were registered with a GP practice that utilizes the trusted laboratory and had at least some tests ordered by those practices since 2008. Individuals were classified based on their maximum in-hospital glucose measurement, and the number of those with HbA1c levels of 48 mmol/mol (6.5 percent) prior to the index hospitalization, as well as during and after admission, was reported. An estimated proportion of people in each blood glucose stratum who did not have a follow-up HbA1c were found to have undiagnosed diabetes.

The key findings of this study were as follows:

1. In all, 764,241 glucose readings were taken for the 81,763 patients hospitalized at the Oxford University Hospitals Trust.

2. The median age was 70 (56, 81) years, with men accounting for 53%.

3. 70.7% of the population identified themselves to be of White ethnicity, 3.1% of Asian origin, and 1.1% of Black origin, with the remaining 23.1% unspecified.

4. Of those people, 22,375 (27.4%) had no prior HbA1c measurements.

5. A total of 1689 patients (2.5 percent) had a diabetes-range HbA1c throughout or after their hospitalization, and an additional 1496 (2.2%) may have undetected diabetes, with the majority of these having in-hospital glucose of 15 mmol/L.

In conclusion, there appears to be a lack of future HbA1c measurements among those with hyperglycemia and no previous HbA1c measurements in the diabetic range. The possibility of combining the testing and follow-up of people with seemingly unrecognized hospital hyperglycemia throughout main and secondary care is identified in this study.

Reference:

Farmer, A. J., Shine, B., Armitage, L. C., Murphy, N., James, T., Guha, N., & Rea, R. (2022). The potential for utilising in‐hospital glucose measurements to detect individuals at high risk of previously undiagnosed diabetes: Retrospective cohort study. In Diabetic Medicine. Wiley. https://doi.org/10.1111/dme.14918