Elsevier

The Lancet

Volume 397, Issue 10291, 12–18 June 2021, Pages 2275-2283
The Lancet

Articles
Comparing real-time and intermittently scanned continuous glucose monitoring in adults with type 1 diabetes (ALERTT1): a 6-month, prospective, multicentre, randomised controlled trial

https://doi.org/10.1016/S0140-6736(21)00789-3Get rights and content

Summary

Background

People with type 1 diabetes can continuously monitor their glucose levels on demand (intermittently scanned continuous glucose monitoring [isCGM]), or in real time (real-time continuous glucose monitoring [rtCGM]). However, it is unclear whether switching from isCGM to rtCGM with alert functionality offers additional benefits. Therefore, we did a trial comparing rtCGM and isCGM in adults with type 1 diabetes (ALERTT1).

Methods

We did a prospective, double-arm, parallel-group, multicentre, randomised controlled trial in six hospitals in Belgium. Adults with type 1 diabetes who previously used isCGM were randomly assigned (1:1) to rtCGM (intervention) or isCGM (control). Randomisation was done centrally using minimisation dependent on study centre, age, gender, glycated haemoglobin (HbA1c), time in range (sensor glucose 3·9–10·0 mmol/L), insulin administration method, and hypoglycaemia awareness. Participants, investigators, and study teams were not masked to group allocation. Primary endpoint was mean between-group difference in time in range after 6 months assessed in the intention-to-treat sample. This trial is registered with ClinicalTrials.gov, NCT03772600.

Findings

Between Jan 29 and Jul 30, 2019, 269 participants were recruited, of whom 254 were randomly assigned to rtCGM (n=127) or isCGM (n=127); 124 and 122 participants completed the study, respectively. After 6 months, time in range was higher with rtCGM than with isCGM (59·6% vs 51·9%; mean difference 6·85 percentage points [95% CI 4·36–9·34]; p<0·0001). After 6 months HbA1c was lower (7·1% vs 7·4%; p<0·0001), as was time <3·0 mmol/L (0·47% vs 0·84%; p=0·0070), and Hypoglycaemia Fear Survey version II worry subscale score (15·4 vs 18·0; p=0·0071). Fewer participants on rtCGM experienced severe hypoglycaemia (n=3 vs n=13; p=0·0082). Skin reaction was more frequently observed with isCGM and bleeding after sensor insertion was more frequently reported by rtCGM users.

Interpretation

In an unselected adult type 1 diabetes population, switching from isCGM to rtCGM significantly improved time in range after 6 months of treatment, implying that clinicians should consider rtCGM instead of isCGM to improve the health and quality of life of people with type 1 diabetes.

Funding

Dexcom.

Introduction

The majority of people with type 1 diabetes do not achieve glycated haemoglobin (HbA1c) below 7% (53 mmol/mol)1, 2 and spend a considerable part of the day in hypoglycaemia and hyperglycaemia, exposing them to risk of hypoglycaemic coma, ketoacidosis, and chronic microvascular and macrovascular disease. In recent years, progress has been made in the field of home glucose self-monitoring with the advent of subcutaneous sensors capable of reporting glycaemic levels on demand (intermittently scanned continuous glucose monitoring [isCGM]) or in real time (real-time continuous glucose monitoring [rtCGM]). People on isCGM can check their glucose values by scanning the sensor transmitter with a receiver or smartphone, whereas the transmitter of rtCGM automatically sends a new value to a receiver, smart watch, or smartphone every 1–5 min. In contrast to isCGM, rtCGM has the option of (predictive) alerts for high and low blood glucose levels, but is generally more expensive than isCGM.3 Several randomised controlled trials comparing the use of CGM with capillary blood glucose measurements, showed that isCGM and rtCGM have a beneficial effect on glycaemic outcomes and quality of life in people with type 1 diabetes treated with multiple daily injections or insulin pump therapy.4, 5, 6, 7, 8, 9 To address the question whether switching from isCGM to rtCGM offers additional benefits, we did a multicentre, non-masked, randomised controlled trial comparing 6-month rtCGM use with 6-month isCGM use in adults with type 1 diabetes who were previously using isCGM.

Research in context

Evidence before this study

We searched PubMed for trial reports published in English up to Feb 2, 2021, comparing real-time continuous glucose monitoring (rtCGM) with intermittently scanned continuous glucose monitoring (isCGM) in non-pregnant adults with type 1 diabetes. We used search terms “intermittent” or “flash”, “real time”, “continuous glucose monitoring”, and “type 1 diabetes”. We identified six manuscripts of four trials. Three manuscripts were based on the randomised controlled I HART trial, one was the randomised controlled CORRIDA trial, and two described observational trials, evaluating real-world data from France, Germany, and Austria.

The I HART trial by Reddy and colleagues was an 8-week trial, followed by an 8-week extension phase, done in 40 adults with type 1 diabetes and impaired awareness of hypoglycaemia, treated with multiple daily insulin injections. Median glycated haemoglobin (HbA1c) was 7·3% (56 mmol/mol). The CORRIDA trial by Hásková and colleagues studied 60 adults with type 1 diabetes and normal hypoglycaemia awareness during a 4-day exercise and 4-week home phase. Mean HbA1c was 7·8% (62 mmol/mol). Both trials included CGM naive people. Of note, outcomes regarding glycaemic control were not measured with the same CGM device in the I HART trial, which is an important limitation of the study design.

The French observational trial by Préau and colleagues assessed the effect of switching from isCGM to rtCGM in 18 adults with type 1 diabetes and a high risk of hypoglycaemia (n=8), or an elevated HbA1c (n=9), or both (n=2) after 3 and 6 months. Mean HbA1c was 8·07% (65 mmol/mol). The German and Austrian trial by Sandig and colleagues included 233 adults with type 1 diabetes using isCGM or rtCGM, with a median HbA1c of 7·3% (56 mmol/mol), in a cross-sectional analysis. In both observational studies, most participants used insulin pump therapy.

I HART, CORRIDA, and the observational trials concluded that rtCGM was superior to isCGM with regard to glycaemic control based on CGM metrics; time in range (sensor-glucose 3·9–10·0 mmol/L [70 to 180 mg/dL]) was higher, while time in hypoglycaemia (definitions ranged from sensor-glucose <3·0 mmol/L [54 mg/dL] to <3·9 mmol/L [70 mg/dL]) was lower with rtCGM than with isCGM. Additionally, the I HART trial showed less worry about hypoglycaemia with the use of rtCGM. Only two trials evaluated HbA1c and observed no benefit with rtCGM versus isCGM. However, as trials were limited in terms of study population (small and prespecified), design, and duration (short follow-up), it is still not clear whether switching from isCGM to rtCGM with alert functionality offers additional benefits in a large and unselected population with type 1 diabetes over a longer period of time.

Added value of this study

The ALERTT1 trial is the first 6-month, multicentre, prospective, randomised controlled trial comparing rtCGM with isCGM in 254 adults with type 1 diabetes, who previously used isCGM. Mean HbA1c was 7·4% (58 mmol/mol) and a minority of the study population was hypoglycaemia unaware (44 [17%] people) or had a history of severe hypoglycaemia (29 [11%]). Most (205 [81%]) were treated with multiple daily injections. Findings showed that in an unselected group of people with type 1 diabetes, 6-month use of rtCGM with alert functionality improved time in range (sensor-glucose 3·9–10·0 mmol/L [70–180 mg/dL]), while HbA1c, time in clinically significant hypoglycaemia (sensor-glucose <3·0 mmol/L [54 mg/dL]), and time in hyperglycaemia (sensor-glucose >10·0 mmol/L [180 mg/dL]) were reduced. Aditionally, more people on rtCGM achieved glycaemic targets as defined by international consensus guidelines, and had less frequently severe hypoglycaemia.

Besides glycaemic control, the ALERTT1 trial also evaluated patient reported outcomes through various validated questionnaires. Despite relatively high quality of life of all participants at baseline, rtCGM users experienced less hypoglycaemia worry and higher treatment satisfaction at the end of study.

Implications of all the available evidence

People using rtCGM showed significant benefits over 6 months compared with people using isCGM in terms of both glycaemic control and patient reported outcomes. This implies that clinicians should consider rtCGM to improve the health and quality of life of people living with type 1 diabetes.

Section snippets

Study design and participants

Comparing Continuous With Flash Glucose Monitoring in Adults With Type 1 Diabetes (ALERTT1) was a 6-month double-arm, parallel-group, non-masked, randomised controlled trial comparing rtCGM (intervention group) with isCGM (control group), and was done in the diabetes clinics of three regional and three university medical centres in Belgium. People aged 18 years or older with diagnosis of type 1 diabetes for 6 months or more were eligible for inclusion. Additional inclusion criteria were

Results

Between Jan 29 and July 30, 2019, 269 people with type 1 diabetes were recruited. Subsequently, 254 participants were randomly assigned to the intervention group (rtCGM; n=127) or control group (isCGM; n=127), of which 124 in the rtCGM and 122 in the isCGM group completed the study at 6 months (figure 1).

The rtCGM and isCGM groups had similar baseline characteristics (table 1). Average age of participants was 42·9 years (SD 14·1; range 18–76 years). 239 (94%) were White and most were highly

Discussion

To our knowledge, ALERTT1 is the first long-term, multicentre, prospective, randomised, controlled trial comparing rtCGM with isCGM in adults with type 1 diabetes and a history of previous use of isCGM. The trial showed that 6-month use of rtCGM with alert functionality improved time in range by 6·85 percentage points, and reduced time in hyperglycaemia >10·0 mmol/L (180 mg/dL) by 6·27% points and time in hypoglycaemia <3·0 mmol/L (54 mg/dL) by 0·35% points. The proportion of participants who

Data sharing

Anonymous data are shared with centres participating in the ALERTT1 trial, based on research questions mentioned in the ALERTT1 study protocol or based on a new study protocol approved by the relevant ethical committees. Selected anonymous data collected in the study and additional documents can be made available to others not involved in the ALERTT1 trial, on the basis of a reasonable request. Please contact Dr Pieter Gillard: [email protected].

Declaration of interests

UZ Leuven received non-financial support for travel from Novo Nordisk for MMV. KU Leuven received non-financial support for travel from Medtronic, and financial support for travel from Roche for SC. CDB reports consulting fees and honoraria for speaking for Abbott, AstraZeneca, Boehringer Ingelheim, A Menarini Diagnostics, Eli Lilly, Medtronic, Novo Nordisk, and Roche. RH serves or has served on the advisory panel for Merck Sharp and Dohme, Boehringer Ingelheim, and Eli Lilly. LVH reports

References (29)

  • T Danne et al.

    International consensus on use of continuous glucose monitoring

    Diabetes Care

    (2017)
  • T Battelino et al.

    Clinical targets for continuous glucose monitoring data interpretation: recommendations from the International Consensus on Time in Range

    Diabetes Care

    (2019)
  • CJ Coffman et al.

    To condition or not condition? Analysing ‘change’ in longitudinal randomised controlled trials

    BMJ Open

    (2016)
  • M Reddy et al.

    Switching from flash glucose monitoring to continuous glucose monitoring on hypoglycemia in adults with type 1 diabetes at high hypoglycemia risk: the extension phase of the I HART CGM study

    Diabetes Technol Ther

    (2018)
  • Cited by (0)

    View full text