Prognosis

In hospitalised patients, both hyperglycaemia and hypoglycaemia are independently associated with poorer outcomes:

Inpatient hyperglycaemia

  • Extensive evidence from observational studies and randomised trials in patients with and without diabetes (across critical and non-critical illness) shows a strong association between hyperglycaemia - particularly blood glucose levels above 11.0 mmol/L (200 mg/dL) - and worse outcomes, including increased mortality, infections, prolonged hospital stays, greater likelihood of intensive care unit admission, and a higher need for post-discharge transitional or nursing home care, compared with glucose levels below 5.6 mmol/L (100 mg/dL).[5][65][66][67][68][69][70][71][72][73]​​​​ This risk rises with the severity of hyperglycaemia before or during hospitalisation.[70][74][75][76]

  • Notably, patients without a prior diagnosis of diabetes who develop stress-induced hyperglycaemia appear to have a higher risk of complications and mortality than those with an established diabetes diagnosis.​[69]​​​[71][74][77]​​​[78]​​ While it remains unclear whether stress hyperglycaemia directly causes poor outcomes or simply reflects illness severity, data show that patients without known diabetes receive fewer capillary blood glucose checks than those with diabetes, even at similar glucose levels.[71][79]​​ Conversely, more frequent monitoring in patients with diabetes may allow earlier detection of complications, potentially reducing mortality.

Inpatient hypoglycaemia

  • Hypoglycaemia is common in hospitalised patients with diabetes and is associated with worse outcomes.[80]

  • One US study of over 107,000 hospital admissions involving people using insulin found that 20% experienced hypoglycaemia during their stay.[14]​ Those who developed hypoglycaemia had a significantly higher in-hospital mortality rate (6.5% vs. 3.8%) and stayed longer in hospital (median 8.2 vs. 5.2 days), with an adjusted odds ratio for inpatient mortality of 1.66.

  • One UK retrospective study of 6374 admissions found that even mild hypoglycaemia (blood glucose 2.3 to 3.9 mmol/L [41-70 mg/dL]) increased length of stay and in-hospital mortality, with higher risks for severe episodes.[81]

  • Supporting these findings, one systematic review and meta-analysis concluded that hospitalised adults with diabetes who experienced hypoglycaemia (blood glucose <4 mmol/L [<72 mg/dL]) had significantly longer hospital stays - by an average of 4.1 days - and more than twice the risk of in-hospital mortality compared with those without hypoglycaemia.[82]

In summary, hyperglycaemia and hypoglycaemia in hospitalised patients each serve as potent prognostic markers, independently correlating with increased hospital stays and mortality. Early intervention and careful glycaemic management are therefore essential components of inpatient care to improve outcomes.

Use of this content is subject to our disclaimer