Reviewed and revised 3 February 2013
- Routine management in ICU involves avoiding complications of hyperglycaemia (infections) and hypoglycaemia (arrhythmias, neurological damage, cardiac events)
- Traditional goals have varied
- Tight glucose control was in vogue following the Leuven trials led by Greet Van Den Berghe, but this was over-turned by the subsequent definitive NICE-SUGAR trial
- In Australasia the ‘standard’ glucose control target is usually < 10mmol/L
Glucose variability is associated with worse morbidity and mortality, as shown by Badawai et al, 2012:
- retrospective observational study
- n = 194,772 ICU patients with LOS >48 hours
- risk of mortality progressively increased with severity and duration of deviation from euglycemia and with increased variability in a ‘dose-dependent’ manner
- severe intensive care unit-acquired hyperglycemia, hypoglycemia, and variability are associated with similar risks of mortality
EXAMPLE OF A PROTOCOL FOR GLUCOSE CONTROL IN THE ICU
EVIDENCE FOR GLUCOSE CONTROL
- See Glucose Control Literature Summaries for details on the individual papers
- The 1995 DIGAMI study found that intensive blood glucose control in diabetic patients with acute myocardial infarctions improved mortality, though the subsequent DIGAMI-II trial did not favour any particular glucose control strategy
- The 2001 Leuven Surgical Trial led by Greet Van Den Berghe was a landmark study that signalled the trend to intensive glucose control in ICU (Paul Marik has called November 8th 2001, the day this paper was publsihed along with the Rivers EGDT trial, ‘the darkest day in the history of intensive care’). This study found that ICU mortality decreased from 8 to 4.6% with tight control (BSL 4.5-6.5).
- In 2006 the Leuven Medical Trial was published, it found improved morbidity but not mortality in medical ICU pateints receiving intensive insulin therapy
- The 2008 VISEP study showed that intensive insulin therapy in patients with severe sepsis did not affect mortality or oragan failure, but raised concerns due to higher rates of severe hypoglycaemia and serious adverse events.
Although considered landmark studies, the Leuven trials have many problems:
- single centre
- open label
- trials stopped early because of observed benefit, which may have exaggerated treatment effect
- difference found through sub-group analysis
- observed difference may have been due to chance
- the studies are not externally valid, except to ICUs with similar high glucose loads on day 1 and high rates of supplemental or total parental nutritiono on day 2
- A definitive MC RCT of over 6000 ICU patients found increased mortality and more severe hypoglycemias in the intensive control group (glucose 4.6-6 mM) compared to the conventional therapy group
- this study contradicted, and superseded the Leuven trials
- A 2012 post-hoc analysis found a dose-response relationship between severe hypoglycemia and mortality, and that intensive control was associated with more episodes severe hypoglycemia
- intensive insulin therapy could conceivably come back into vogue with the development of accurate, continuous glucose monitoring +/- computerised feedback systems
- determining how diabetic hyperglycemia and stress-induced hyperglycemia should be managed as separate entities in the critically ill
- role of glucose variability control
References and Links
- Badawi O, Waite MD, Fuhrman SA, Zuckerman IH. Association between intensive care unit-acquired dysglycemia and in-hospital mortality. Crit Care Med. 2012 Dec;40(12):3180-8. doi: 10.1097/CCM.0b013e3182656ae5. PubMed PMID: 22971590.
- Egi M, Finfer S, Bellomo R. Glycemic control in the ICU. Chest. 2011 Jul;140(1):212-20. doi: 10.1378/chest.10-1478. Review. PubMed PMID: 21729892. [Free Full Text]
- Finfer S, et al. Clinical review: Consensus recommendations on measurement of blood glucose and reporting glycemic control in critically ill adults. Crit Care. 2013 Jun 14;17(3):229. [Epub ahead of print] PubMed PMID: 23767816; PubMed Central PMCID: PMC3706766.
- Flower O, Finfer S. Glucose control in critically ill patients. Intern Med J. 2012 Jan;42(1):4-6. doi: 10.1111/j.1445-5994.2011.02631.x. PubMed PMID: 22276558.
- Henderson WR, Finfer S. Differences in outcome between the NICE-SUGAR and Leuven trials: possible methodological explanations. Crit Care Resusc. 2009Sep;11(3):175-7. PubMed PMID: 19737117.
- Myburgh JA, Chittock DR. Differences in outcome between the NICE-SUGAR and Leuven trials: biological mechanisms of intensive glucose control in critically ill patients. Crit Care Resusc. 2009 Sep;11(3):178-9. PubMed PMID: 19737118.
- Smith FG, Sheehy AM, Vincent JL, Coursin DB. Critical illness-induced dysglycaemia: diabetes and beyond. Crit Care. 2010;14(6):327. doi: 10.1186/cc9266. Epub 2010 Nov 5. Review. PubMed PMID: 21067560; PubMed Central PMCID: PMC3220014.
FOAM and web resources