Guillermo Umpierrez, Professor of Medicine at Emory University School of Medicine in Atlanta, is the Chair of the Task Force for the 2023 Consensus Statement. It contains updated recommendations from the ADA, EASD, the Joint British Diabetes Societies for Inpatient Care, the American Association of Clinical Endocrinologists and the Diabetes Technology Society. 

Since the last consensus statement in 2009, there have been 6000 literature citations on hyperglycaemic crises, diabetic ketoacidosis (DKA) and hyperglycaemic hyperosmolar state (HHS), so it was clearly time for a review. “Our objective was to provide up-to-date knowledge of the epidemiology, pathophysiology, clinical presentation and recommendations for diagnosis, treatment and prevention of DKA and HHS in adult subjects,” explained ProfessorUmpierrez. “The statement is targeted at the full spectrum of healthcare professionals, including clinicians, diabetes care teams, diabetes educators and other stakeholders, as well as at people with diabetes.” 

Epidemiology

Dr Irl Hirsch from the University of Washington School of Medicine in Seattle presented data on the global trends, costs and mortality of DKA and HHS. Rates of both conditions rose in both the US and Europe up until 2015, when they levelled off, and recurrent episodes were a common theme. “Of course, the big difference since the previous guidelines were published is the introduction of SGLT-2 inhibitors,” said Dr Hirsch. “Systematic review and meta-analysis suggest that the hazard ratios for the risk of DKA on these medications is 3.38 in type 1 diabetes and 2.46 in type 2 diabetes.” 

The Task Force also looked at the characteristics of patients being readmitted with hyperglycaemic crisis within 30 days. “At a rate of 19.4%, this is really concerning,” he said. These patients usually have a high Charlson comorbidity index, with conditions like chronic kidney disease, cancer, chronic obstructive pulmonary disease, coronary artery disease, heart failure and hypertension. 

Meanwhile, the costs of managing hyperglycaemic crises are increasing while length of stay in hospital is decreasing. Length of stay is longer in the UK than in the US and costs are greater for type 2 diabetes than type 1, which is perhaps not surprising considering the comorbidities present in this generally older population.  

Mortality from DKA has reached a plateau in the last decade in the US and is 0.2% for type 1 and 1.02% for type 2. Meanwhile, mortality from HHS has decreased from 1.44% in 2008 to 0.77% in 2018. Mortality for both conditions is, however, much greater in low- to middle-income countries where it may be as high as 43%. Finally, data from the UK suggest that readmission is a factor in DKA mortality, being around six times higher for those with more than fiveadmissions.  

Clinical presentation

Dr Shivani Misra of Imperial Healthcare NHS Trust, London, reviewed the well-known symptoms of DKA and HHS and noted that, “one-third of hyperglycaemic emergencies have a hybrid DKA/HHS presentation and we are increasingly seeing this.” In the new Consensus Statement, the glucose cutoff for diagnosing DKA has been lowered from 250 mg/dl to 200 mg/dl and having a prior history of diabetes is also now included. “Both of these changes recognise the wide range of glucose levels in presentation, for euglycaemic DKA now accounts for 10% of cases,” said Dr Misra. 

There is also a new emphasis on hydroxybutyrate – the main ketone in DKA – measurement in serum or point of care, rather than urine ketones. And anion gap is no longer used in the main definition of DKA, although it may be of use in settings where there is no ketone testing.

“We’ve also introduced a mental status alert – alert or drowsy, stupor or coma – recognising that people who are drowsy or obtunded are more likely to have severe outcomes from DKA,” said Dr Misra. Of course, ketoacidosis doesn’t always mean DKA and there are multiple other causes including starvation, alcohol misuse and hyperemesis in pregnancy. “It is important to remember these other causes, particularly with the increase in euglycaemic ketoacidosis,” she continued.

For HHS, most recommendations are as before, save that the cutoff for serum osmolality has been lowered from 320 mOsm/l to 300 mOsm/l to account for the presence of dehydration. There is also recognition of hybrid DKA/HHS presentations. Finally, the Consensus Report has tried to introduce international harmonisation in cutoff values for diagnosis and severity. 

Management

Professor Ketan Dhatariya from Norfolk and Norwich Hospital, UK, said, “As we all know, the mainstay of management in both conditions is fluids, insulin and potassium.” He went on to summarise these requirements, highlighting what is new in the Consensus Statement. For instance, if glucose falls below 3.9 mmol/l, 5-10% dextrose should be added with saline and insulin to prevent hypoglycaemia in euglycaemic DKA. 

The Consensus Statement also includes guidance for treatment in different settings. “Where facilities are not available, you can treat mild DKA with subcutaneous insulin at 0.1 U/kg as a rapid-acting analogue and repeat every hour or give it as a fixed-rate infusion if the case is severe. With HHS, use regular insulin 0.05 U/kg,” he said. “We’ve discussed this and we leave it to you as the clinician – some say give fluid first and only give insulin when the glucose stops dropping because we don’t want to drop osmolality too quickly, whereas some give them together – there are different practices around the world.” 

When it comes to potassium, there needs to be some renal function, so ensure the patient is urinating at 0.5 ml/kg/h, otherwise it will send them into hyperkalaemia. “Then, in all cases, check everything every two to four hours until the patient is stable,” said Professor Dhatariya, adding that the Consensus Statement contains a useful downloadable slide with a definition of resolution of DKA and HHS, which contains some new guidance. 

Complications

Hyperglycaemic crises can lead to various complications and the Consensus Statement includes strategies to prevent or mitigate these. Gian Paolo Fadini, Professor of Endocrinology at the University of Padova, Italy, noted that hypoglycaemia affects 16-28% of those undergoing DKA treatment, which is severe in 2% of cases where it can increase mortality. Frequent blood glucose monitoring is necessary with reduction of insulin and addition of dextrose once glucose is less than 250 mg/dl.   

Another complication is development of a prothrombotic state, which is more likely in HHS than in DKA, although there is controversy over the actual cause. “There are some plausible mechanisms,” said Professor Fadini. “Hyperglycaemia causes endothelial damage, dehydration increases coagulability, while bed rest leads to venous blood stasis. More research is needed, but you can mitigate the risk with low-dose heparin and an anticoagulant if thrombosis is suspected.” 

Neurologic complications include cerebral oedema and osmotic demyelination syndrome. Cerebral oedema is rare in adults but may occur in children. It can only be diagnosed on imaging and mortality is 30%. Osmotic demyelination syndrome occurs when there is too rapid a correction of hyponatraemia and is mitigated by monitoring mental status, having a low threshold for imaging and ensuring a slow correction of hyperosmolality.  

Once the crisis is over, the patient is ready for discharge. “Discharge is an incredibly important moment in the management of these patients,” said Professor Fadini. In the US, 12.3% and 22% of patients are readmitted within 30 days or one year, respectively, following a hyperglycaemic crisis. “Close follow-up is needed within two to four weeks, looking into the reasons for the DKA or HHS, including mental health and social determinants,” he added. Risk factors for recurrent DKA include end-stage renal disease, a high number of comorbidities, discharge against medical advice (which is getting more common) and drug misuse. Recurrent DKA increases mortality four-fold. “On discharge, it is extremely important to offer appropriate education,” said Professor Fadini. “Focus on the current event and overall glycaemic control.”  

One study showed that starting flash glucose monitoring at this point can prevent recurrence – after 12 months, the rate of DKA went down by 52% and 56% in people with type 1 and type 2 diabetes, respectively. Finally, it’s important to supply adequate insulin and equipment along with healthcare professional contacts upon discharge, and contact should be made with social services if social determinants of health are an issue. 

Summary

Professor Robert Gabbay, Chief Scientific & Medical Officer at the ADA, and Professor Chantal Mathieu, EASD President, reviewed the Consensus Statement, saying that its aim was for everyone to take some practical tips back to their practice. The partner organisations had brought a broad and international perspective to this work. “Most of the time, it’s not the diabetologists and endocrinologists standing at the gate taking care of these people,” said Professor Mathieu. “It’s emergency doctors, general internists and, especially, it’s junior doctors that are on call because when a very bad thing happens, it’s during the weekend and during the night. So this is why it’s important to have this guidance.”  

“Capturing some of the data from Irl’s presentation, we see that although we’ve been making progress for some time in terms of hyperglycaemia emergencies, we’ve really plateaued and we’re not making much progress,” said Professor Gabbay. “People are still being admitted in large numbers and, if you look globally, even more so. Mortality continues to be a problem. It should be declining but it’s not. And in some low- and middle-income countries, mortality is 30% and higher, which should not be happening today, 102 years after the availability of insulin. There is also the recognition that 30% of these cases are mixed DKA/HHS and those people do a lot worse.”

Professor Dhatariya had earlier shown a paper looking at precipitating causes of DKA from 11 countries, which showed new-onset diabetes to be a significant cause – accounting for 6% of cases in the UK and 17-24% in the US. Infection was a leading cause in China, Indonesia and the UK, and omission of insulin a leading cause in Australia, Brazil and the US. However, many cases had an unknown cause. “I think this should be looked at,” said Professor Mathieu. “We need to find out why someone presents with DKA or HHS. It means something went wrong. We cannot say that in 18.7% (UK) or 35% (Nigeria) of cases, we don’t know what happened.” 

“So how do we go about preventing readmission?” asked Professor Gabbay. “Insulin supplies, healthcare contacts, social services – it sounds obvious, but is often not thought about. When they go home, will they be able to take their next dose of insulin? Do they have a way of assessing their blood glucose levels? Someone to call? Maybe the crisis could have been avoided if they’d been able to call somebody and work through sick-day rules and adjust their insulin. And an area that cannot be overemphasised enough is the social determinants of health – the social context they return to is going to be such a big predictor of whether they get readmitted – whether that’s homelessness or access to food and medical screening before discharge and referral to appropriate sources.” 

“The only way to avoid recurrence is education,” added Professor Mathieu. “I applaud that simple message.” The Task Force will leverage their partnering organisations’ reach and programmes to spread the word. They will identify professional stakeholders, including emergency junior clinicians and frontline staff. The ADA Institute of Learning – ‘ADA University’ – can help, with its modern techniques of adult learning. 

The Task Force has been increasingly interested in dissemination and implementation science, taking what is known and putting it into practice. “It’s about understanding the context of each organisation and health system, what’s available in the country to localise that change and implementation, so that it can be spread,” concluded Professor Mathieu. The session ended with a call for input from delegates to the Consensus Statement, as it is still a work in progress. 

To learn more, enrol on the EASD e-Learning courses ‘Management of hyperglycaemia in type 2 diabetes’, ‘Management of type 1 diabetes in adults’ and ‘Diabetic ketoacidosis’. 

Any opinions expressed in this article are the responsibility of the EASD e-Learning Programme Director, Dr Eleanor D Kennedy.