Risks and benefits of tight glycaemic control

Le résumé fourni par la source

We read with interest the article by Bone et al. [1] who described a useful and elegant method (the cumulative frequency plot) to display glycaemic control in the critically ill. They also highlight the difficulty in comparing the results from ICUs using different target ranges. Repeated audits in our unit between 2001 and 2006 show similar results, with compliance ranging from < 50% of values within a target range of 4.4–6.1 mmol.l−1 compared with > 75% within the range 4.4–8.3 mmol.l−1. This phenomenon is perhaps overstated by the authors and is to be expected with data approximating to a normal distribution. We would like to point out two other sources of error we identified when auditing glycaemic control in ICUs. Firstly, as glycaemic control tends to improve as a patient's condition is stabilised, the median length of stay needs to be comparable between units. Second, different blood glucose analysers may give significantly different values. Comparing the values obtained in our unit when identical blood samples were analysed using the laboratory analyser, an arterial blood gas machine, and a handheld blood glucose meter, we found up to a 4 mmol.l−1 difference, with the handheld meter performing least accurately, especially at lower blood glucose concentrations. Similar discrepancies have been reported by others [2–4]. Careful audit of glycaemic control is important and the points raised by Bone et al. are timely. Two recent multicentre, European trials of tight glycaemic control (4.4–6.1 mmol.l−1) [5] were stopped early because of an unacceptable incidence of severe hypoglycaemia in the tight control group (data from the GLUControl study presented by J. C. Preiser at the 19th Annual Congress of the European Society of Intensive Care Medicine, Barcelona, September 2006). Although subsequently lacking sufficient power, neither trial showed a trend towards improved survival with tight glycaemic control. This issue remains unresolved and we are not alone in hoping that the results of the NICE-SUGAR study will provide further information on the risks and benefits of tight glycaemic control in intensive care [6, 7]. In the meantime, raising the acceptable target range may improve both compliance and patient safety without a detrimental effect on survival.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.