Flash glucose monitoring: objective, not self-referential, outcomes are needed. Reply to Warren RE [letter]
KeywordsClinical diabetes Devices Hypoglycaemia Insulin therapy
Novel Glucose-Sensing Technology and Hypoglycaemia in Type 1 Diabetes: a Multicentre Non-masked Randomised Controlled Trial
To the Editor: We thank Dr Roderick Warren for his letter  regarding our subgroup analysis of the Novel Glucose-Sensing Technology and Hypoglycaemia in Type 1 Diabetes: a Multicentre, Non-masked, Randomised Controlled Trial (IMPACT; NCT02232698) . In line with the accepted practice in continuous glucose monitoring studies, the trial was designed to compare individuals using the FreeStyle Libre system (Abbott Diabetes Care, Witney, UK) for glucose monitoring with those using the conventional technique, self-monitoring blood glucose (SMBG). The main glycaemic outcome measures with this type of study design require sensor glucose results to be visible to the intervention group but blinded in the control group. Warren reiterates a valid point regarding using the same device for the treatment and assessment of an outcome. This was indeed noted in our original study article ; however the general consensus is that there is no practical alternative to this approach .
With regard to the reliability of the sensor recordings, the accuracy of FreeStyle Libre in the low glucose range (capillary reference <4.4 mmol/l [80 mg/dl]), calculated from raw data collected by Bailey et al  had a mean bias of 0.2 mmol/l (2.9 mg/dl) and a mean absolute difference (MAD) of 0.6 mmol/l (10.4 mg/dl). In the IMPACT study baseline phase, mean bias was 0.0 mmol/l (0.3 mg/dl) and MAD 0.6 mmol/l (10.6 mg/dl) for capillary glucose <4.4 mmol/l. A recent head to head study supports the accuracy of FreeStyle Libre vs two continuous glucose monitoring (CGM) devices in the hypoglycaemic range .
Warren states that studies like this require externally valid outcomes such as HbA1c or severe hypoglycaemia but does not suggest a viable alternative outcome that could have been utilised for this study. HbA1c did not increase in the intervention group, which should be viewed positively in the context of a study population with extremely well-controlled blood glucose, reduced hypoglycaemia and a final mean HbA1c of 53.0 mmol/mol. Severe hypoglycaemia is indeed very rare, with six instances reported in the treatment phase of this study (two in the intervention group and four in the control group), and therefore much larger numbers of patients would need to be studied over long periods to detect a clinically relevant improvement. Notably, there was a 30% reduction in the frequency of participant-reported symptomatic hypoglycaemia in the final 2 weeks (p = 0.0063). Hence, in our view, this similar reduction in symptomatic hypoglycaemia is corroborating evidence for the validity of the primary outcome result.
The authors thank Z. Welsh (Statistics, Abbott Diabetes Care, Witney, UK).
JB wrote the first draft of the response letter and worked collaboratively with the co-authors to discuss and review this response. All authors reviewed and approved the final draft and approved the submission.
This study was funded by Abbott Diabetes Care and by grants from the Swedish Diabetes Association and the Diabetes Research Program at Karolinska Institutet.
Duality of interest
JB has received honoraria for consulting or lecture fees from Abbott Diabetes Care, AstraZeneca, Insulet Corporation, Integrity Applications, Novo Nordisk, and Sanofi-Aventis. RA has received honoraria for lecture fees for Abbott Diabetes Care. PG-D has received lecture honoraria from Abbott Diabetes Care and Medtronic and served on advisory boards for Medtronic. JK has received personal fees from Abbot Diabetes Care, personal fees from Abbott Diabetes Care during the conduct of this study; personal fees from Lilly, Novo Nordisk, Berlin Chemie, Medtronic, Sanofi, MSD and AstraZeneca. RW received lecture honoraria and serves on advisory boards for Abbott Diabetes Care, Boehringer-Ingelheim, Eli Lilly, Merck Sharp & Dohme, Novo Nordisk, Roche Diabetes Care, Sanofi and Servier; and has received unrestricted study grants from Eli Lilly, Novo Nordisk, and Sanofi. PO declares that there is no duality of interest associated with his contribution to this manuscript.
- 1.Warren RE (2018) Flash glucose monitoring: objective, not self-referential, outcomes are needed. Diabetologia. https://doi.org/10.1007/s00125-018-4608-0
- 2.Oskarsson P, Antuna R, Geelhoed-Duijvestijn P, Kroger J, Weitgasser R, Bolinder J (2018) Impact of flash glucose monitoring on hypoglycaemia in adults with type 1 diabetes managed with multiple daily injection therapy: a pre-specified subgroup analysis of the IMPACT randomised controlled trial. Diabetologia 61:539–550CrossRefPubMedGoogle Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.