Implementing a multicentre surveillance system to standardise hyperosmolar hyperglycaemic state (HHS) care: development, outcomes and lessons learned
Background: Hyperosmolar hyperglycaemic state (HHS) is a life-threatening emergency of diabetes mellitus (DM) associated with significant mortality. Despite this, there is no established framework to monitor national trends and adherence to guidelines. This limits insights into performance, which ar...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-07-01
|
| Series: | Clinical Medicine |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1470211825001332 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Background: Hyperosmolar hyperglycaemic state (HHS) is a life-threatening emergency of diabetes mellitus (DM) associated with significant mortality. Despite this, there is no established framework to monitor national trends and adherence to guidelines. This limits insights into performance, which are essential to enhancing patient outcomes. A multi-centre surveillance system was developed with the following objectives to address these challenges:• Monitoring trends in HHS presentation, outcomes, complications and management practices;• Assessing adherence to national HHS guidelines and providing evidence-based insights;• Identifying barriers and facilitators to adopting international HHS guidelines to enhance clarity and consistency of management. Methods: The framework of the HHS surveillance system is based on the DEKODE DKA model as developed by the DEVI collaboration at the University of BirminghamData parameters to be collected by the surveillance system were decided based on an expert consultation with the authors of the JBDS HHS guidelines. Data are collected in the following categories: patient demographics; aetiology of HHS; biochemistry on diagnosis; management; patient outcome and complications; and factors affecting length of hospital stay. A standardised data collection form was created incorporating these parameters. This can be accessed by a web-based form allowing multiple users to input data simultaneously. The form was piloted in Hospital A and is now used in 12 hospitals. HHS cases are identified in each hospital based on a health informatics search. A designated team member at each hospital then assigns an anonymous identifier to each case for input into the form. Cases are screened to confirm that they fulfil the diagnostic criteria for HHS as per JBDS guidelines; mixed DKA-HHS is excluded. Results: Between January 2021 and November 2024, 218 episodes of HHS were identified with a median age of 77 years (IQR 64–85). Most patients (84.4%) had type 2 DM (T2DM), while 8.7% of episodes were first-time presentations of DM. Intercurrent illness was the primary precipitating cause of HHS in 49.5% of episodes. Over a median HHS duration of 48.2 h (IQR 24.9–74.15), an average of 6.5 L (IQR 4.0–9.7) of fluids and 69.0 units (IQR 30.8–116.0) of insulin were required for resolution. Adherence to ketone monitoring was low 28.9% (IQR 14.9–49.7). A comparative analysis between two demographically similar hospitals (A and B) showed that Hospital A had 86.4% adherence to glucose monitoring guidelines and hospital B 64.9%. Mortality rates at hospital A were 2.3% vs 16.3% in hospital B. Conclusion: Our analysis provides insights into UK trends in HHS and underscores the need for a standardised management approach. Comparative analysis reveals differences in HHS management and outcomes between hospitals. The most common precipitating causes of HHS were intercurrent illnesses, which highlights the importance of identifying precipitating aetiology and the complexity of managing HHS along with concurrent other serious illness. For a significant minority of cases, HHS was also their first presentation of diabetes, highlighting the importance of being vigilant for HHS in patients without known diabetes. |
|---|---|
| ISSN: | 1470-2118 |