Global monitoring of hyperglycaemic emergencies in adults with diabetes: protocol for a systematic review of incidence and outcomes (2015–2024)

Background

Diabetes is a significant public health concern (Gregg et al., 2023). Worldwide, 529 million people were estimated to be living with the disease in 2021, with type 2 diabetes accounting for 96% of cases. By 2050, the number of people with diabetes is projected to exceed 1.3 billion, affecting low- and middle-income countries (LMICs) disproportionately (GBD 2021 Diabetes Collaborators, 2023). Sub-optimal diabetes management increases complication, hospitalisation and mortality risk (Fayfman et al., 2017; Umpierrez, 2020; Yan et al., 2018), while contributing to high healthcare costs; for instance, in the United States (US), diabetes-related expenses reached US$412.9 billion in 2022 (Parker et al., 2024). To reduce the global diabetes burden and improve care, the World Health Organization (WHO) has mandated more coherent monitoring of diabetes and related outcomes at a global level through the Global Diabetes Compact initiative (Gregg et al., 2023).

Acute diabetes-related emergencies include diabetic ketoacidosis (DKA) and hyperosmolar hyperglycaemic state (HHS) (Dhatariya et al., 2020; Umpierrez & Korytkowski, 2016). Mostly associated with type 1 diabetes, DKA occurs when insulin deficiency causes the body to break down fats too quickly, resulting in high ketone levels that can lead to severe complications such as dehydration, coma, and ultimately death. Alternatively, HHS, which predominantly affects individuals with type 2 diabetes, is characterized by extreme hyperglycaemia without significant ketoacidosis, leading to severe dehydration and altered mental status. The incidence of such hyperglycaemic emergencies (HGE) varies worldwide, likely influenced by differences in healthcare access, diabetes management, and study methodologies (Umpierrez et al., 2024a; Umpierrez et al., 2024b). For example, a 2017 systematic review found that DKA incidence significantly varied, with one study reporting the highest incidence in China at 263 per 1,000 person-years, whereas Israel and North America experienced lower incidence; as low as 0 per 1,000 person-years (Fazeli Farsani et al., 2017). Much of the epidemiological data available on DKA is derived from hospital discharge coding (Dhatariya et al., 2020). HHS incidence is less clearly documented (Umpierrez et al., 2024a; Umpierrez et al., 2024b), but remains a major contributor to diabetes-related emergencies, particularly among older adults (Pasquel & Umpierrez, 2014).

Both DKA and HHS require intensive treatment and hospitalisation, at high economic cost. In the US, the Centers for Disease Control and Prevention (CDC) have reported an increase in HGE-related hospitalisations, reaching an annual rate of 9.9 per 1,000 adults with diabetes in 2020 (CDC, 2024). The annual cost of DKA-related treatment in the US alone has been estimated to exceed US$1 billion (Desai et al., 2018; Kitabchi et al., 2001; Ramphul & Joynauth, 2020). Over the past decade, previously reported improvements in the mortality rates for HGE-related hospitalisations have seemingly plateaued in high-income countries, while the mortality rates reported in LMICs remain disproportionately higher, perhaps due to limited healthcare resources and delayed treatment (Umpierrez et al., 2024a; Umpierrez et al., 2024b).

The epidemiology, pathophysiology, clinical presentation, and recommendations for the diagnosis, treatment and prevention of DKA and HHS in adults has been the subject of a recent consensus report (Umpierrez et al., 2024a; Umpierrez et al., 2024b), an update of the American Diabetes Association (ADA) consensus statement on hyperglycaemic crises in adults with diabetes originally published in 2001 and last updated in 2009 (ADA, 2001; Kitabchi et al., 2009). While this report included a systematic examination of publications from 2009–2023, only English language studies indexed in one bibliographic database were considered, namely PubMed. Conducting a search of more than one database improves coverage and reduces the risk of missing potentially relevant data (Bramer et al., 2017; Ewald et al., 2022). Given the increasing incidence, severe clinical consequences and rising economic burden associated with DKA and HHS, plus the inclusion of DKA and HHS-related hospitalisation as part of WHO guidance on global monitoring for diabetes prevention and control (WHO, 2024), a more comprehensive assessment of the global burden of HGEs is imperative.

Aims and objectives

The objective of this systematic review is to examine the global variation in HGE (DKA and HHS) incidence and clinical outcomes in adults diagnosed with diabetes.

Research questions

This systematic review is guided by the following research questions:

Protocol registration

The review protocol is informed by guidance from the PERSyst group (Prevalence Estimates Reviews – Systematic Review Methodology Group) for conducting systematic reviews of prevalence and incidence (Barker et al., 2021), and is reported in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement for protocols (PRISMA-P; Moher et al., 2015). The completed review will be reported using the PRISMA 2020 Statement (Page et al., 2021). The protocol has been registered on the International Prospective Register of Systematic Reviews (PROSPERO; CRD42024588719), and review materials will be available on the Open Science Framework (OSF; https://doi.org/10.17605/OSF.IO/6HBVQ).

Eligibility criteria

Population-based observational studies conducted in any geographic setting, involving adults (aged ≥18 years) with diagnosed type 1 or type 2 diabetes, will be included, regardless of sample size. In line with the global effort to monitor non-communicable diseases (NCDs) under the WHO NCD Global Monitoring Framework (WHO, 2014; WHO, 2015), studies published between 2015–2024 with data collection periods that started in 2015 or later will be included. This will also enable us to capture pre- and post-COVID-19 pandemic periods. Studies will be included regardless of language and translated to English language through use of translation software. Experimental research studies (e.g., randomised controlled trials, intervention studies), case reports and series, systematic and non-systematic reviews, editorials, letters, commentaries, perspectives, opinions, preprints, study protocols, book chapters, conference abstracts and other grey literature will be excluded.

Information sources

The search strategy was developed with an information specialist (KW) for the selected databases: Ovid MEDLINE, Embase, Scopus, EBSCO CINAHL, and WHO Global Index Medicus. As an example, the search strategy for Embase is presented in Table 1. The complete search strategy is available on OSF (https://doi.org/10.17605/OSF.IO/6HBVQ; see Supplementary File 1).

Data management and study selection

All study records will be uploaded to Covidence to manage de-duplication, title and abstract screening, full-text review, and data extraction. Two reviewers will independently examine retrieved records against eligibility criteria at each stage of the review process. A third reviewer will be consulted to resolve any disagreement among reviewers to reach consensus. The reference lists of included studies will be searched for other potentially eligible studies; and the citationchaser software tool will be utilised (Haddaway et al., 2022). A PRISMA flow diagram will be generated to show the total number of records identified, included and excluded, and the reasons for exclusion.

Quality assessment

A quality assessment of included studies will be conducted by two independent reviewers, using the JBI Critical Appraisal tool for studies reporting on prevalence data (Munn et al., 2015). The checklist includes nine questions addressing study design, conduct and data analysis, with three potential ratings for assessing methodological quality (i.e., yes, no, unclear). A third reviewer will be consulted to resolve any disagreement between reviewers.

Data extraction and synthesis

A standardised data extraction form will be developed in Covidence. Two reviewers will independently extract study population, methodological, and outcome data, including lengths of hospital stay, intensive care unit admissions, and mortality rates. A sample of the data extraction template is provided in Table 2. A third reviewer will be consulted to resolve any disagreement among reviewers.

Data will be synthesised using descriptive statistics, such as proportions and counts for categorical variables, and means, standard deviations and ranges for continuous data. Tables and graphs will be used to present descriptive data on study characteristics, participant characteristics, incidence estimates, and quality assessment ratings. Depending on data availability, potential differences in DKA and HHS incidence, and outcomes will be explored between regions and countries. If feasible, data analyses will be stratified for diabetes type, sex, or other population characteristics. The potential for inequities will be explored and summarised narratively.

If feasible, a meta-analysis using a random-effects model will be conducted. This will include pooling incidence rates and measures of association such as odds ratios and relative risks. This requires considering the various methodological challenges related to the conduct and reporting of systematic reviews of proportions (Barker et al., 2021). Heterogeneity can be a concern in a meta-analysis of proportional data. The I² statistic will be reported despite its limitations, and reporting will be supplemented by prediction interval estimation and sensitivity analyses (Barker et al., 2021; Migliavaca et al., 2022). If conducting a meta-analysis is not feasible, reporting will follow the Synthesis Without Meta-analysis guideline (SWiM; Campbell et al., 2020).

Dissemination

Upon completion of this systematic review, findings will be submitted for publication in a peer-reviewed journal, and presented at scientific conferences both nationally and internationally.

Study status

Between 2 October 2024 and 9 December 2024, the initial search strategy was developed, piloted, and refined, in consultation with an information specialist. Subsequently, from 6–9 January 2025, amendments were made to the search strategy, with the inclusion of additional search terms, alternative spellings for key terms, and added truncations. The last search was conducted on 13 January 2025. Title and abstract screening commenced on 13 January 2025 and is ongoing at the time of submission.