Details

Autor(en) / Beteiligte

• Pickup, John C
• Freeman, Suzanne C
• Sutton, Alex J

Titel

Glycaemic control in type 1 diabetes during real time continuous glucose monitoring compared with self monitoring of blood glucose: meta-analysis of randomised controlled trials using individual patient data

Ist Teil von

• BMJ, 2011-07, Vol.343 (7815), p.d3805-d3805

Auflage

International edition

Ort / Verlag

England: BMJ Publishing Group

Erscheinungsjahr

2011

Quelle

Applied Social Sciences Index & Abstracts (ASSIA)

Beschreibungen/Notizen

• To determine the clinical effectiveness of real time continuous glucose monitoring compared with self monitoring of blood glucose in type 1 diabetes. Meta-analysis of randomised controlled trials. Cochrane database for randomised controlled trials, Ovid Medline, Embase, Google Scholar, lists of papers supplied by manufacturers of continuous glucose monitors, and cited literature in retrieved articles. Studies reviewed Randomised controlled trials of two or more months' duration in men and non-pregnant women with type 1 diabetes that compared real time continuous glucose monitoring with self monitoring of blood glucose and where insulin delivery was the same in both arms. Analysis Two step meta-analysis of individual patient data with the primary outcome of final glycated haemoglobin (HbA(1c)) percentage and area under the curve of hypoglycaemia (glucose concentration <3.9 mmol/L) during either treatment, followed by one step metaregression exploring patient level determinants of HbA(1c) and hypoglycaemia. Six trials were identified, consisting of 449 patients randomised to continuous glucose monitoring and 443 to self monitoring of blood glucose. The overall mean difference in HbA(1c) for continuous glucose monitoring versus self monitoring of blood glucose was -0.30% (95% confidence interval -0.43% to -0.17%) (-3.0, -4.3 to -1.7 mmol/mol). A best fit regression model of determinants of final HbA(1c) showed that for every one day increase of sensor usage per week the effect of continuous glucose monitoring versus self monitoring of blood glucose increased by 0.150% (95% credibility interval -0.194% to -0.106%) (1.5, -1.9 to -1.1 mmol/mol) and every 1% (10 mmol/mol) increase in baseline HbA(1c) increased the effect by 0.126% (-0.257% to 0.0007%) (1.3, -2.6 to 0.0 mmol/mol). The model estimates that, for example, a patient using the sensor continuously would experience a reduction in HbA(1c) of about 0.9% (9 mmol/mol) when the baseline HbA(1c) is 10% (86 mmol/mol). The overall reduction in area under the curve of hypoglycaemia was -0.28 (-0.46 to -0.09), corresponding to a reduction in median exposure to hypoglycaemia of 23% for continuous glucose monitoring compared with self monitoring of blood glucose. In a best fit regression model, baseline area under the curve of hypoglycaemia was only weakly related to the effect of continuous glucose monitoring compared with self monitoring of blood glucose on hypoglycaemia outcome, and sensor usage was unrelated to hypoglycaemia at outcome. Continuous glucose monitoring was associated with a significant reduction in HbA(1c) percentage, which was greatest in those with the highest HbA(1c) at baseline and who most frequently used the sensors. Exposure to hypoglycaemia was also reduced during continuous glucose monitoring. The most cost effective or appropriate use of continuous glucose monitoring is likely to be when targeted at people with type 1 diabetes who have continued poor control during intensified insulin therapy and who frequently use continuous glucose monitoring.