Efficacy of Intermittently Scanned Continuous Glucose Monitoring in the Prevention of Recurrent Severe Hypoglycemia

Abstract

Background:

People with diabetes experiencing hypoglycemia are at increased risk of recurrence because of attenuated autonomic warning. We assessed the efficacy of intermittently scanned continuous glucose monitoring (isCGM; FreeStyle Libre™, Abbott) compared with usual-care self-monitoring of blood glucose (SMBG) in reducing this risk in type 1 and insulin-treated type 2 diabetes.

Methods:

Insulin-treated adults with diabetes and an episode of clinically significant biochemical hypoglycemia (blood glucose [BG] <3.0 mM) or symptomatic hypoglycemia and BG <4.0 mM were randomized to 6 months of isCGM (intensive group) or SMBG (control group) against a background of usual care. The primary outcome was hypoglycemia requiring second-party assistance for recovery. Prespecified secondary outcomes included other hypoglycemic episodes (self-reported, and BG <3.0, 3.0–3.9, <4.0 mM) and change in HbA_1c at 24 weeks.

Results:

Of 59 participants (mean age 53.6 years, 44.1% males, median HbA_1c 61.8 mmol/mol or 7.8%), 30 were allocated to isCGM and 29 to SMBG. The incidence of severe hypoglycemia was not significantly different between the two groups (incident rate ratio [95% confidence interval]: 1.49 [0.46–5.56], P = 0.47). The incidence of other recorded hypoglycemic episodes in the intervention group was double that in the control group (P < 0.001). There was no difference in the change in HbA_1c between the two groups (P = 0.74). There were seven serious adverse events and none was considered related to the intervention.

Conclusions:

Although isCGM is safe, it does not appear to have a role in preventing recurrent severe hypoglycemia in at-risk individuals with diabetes.

Introduction

Hypoglycemia, even when nonsevere, can attenuate the sympathoadrenal response to subsequent episodes, and this state of hypoglycemia unawareness substantially increases the risk of subsequent severe hypoglycemia.¹ Continuous glucose monitoring (CGM) reduces hypoglycemia in type 1^2,3 and insulin-treated type 2⁴ diabetes. It has been recommended for all people who have hypoglycemia unawareness,⁵ but the data supporting this recommendation are sparse. In a crossover study of 52 participants with type 1 diabetes and confirmed hypoglycemia unawareness whose insulin regimens were actively managed during the trial, there were fewer episodes of hypoglycemia requiring second-party assistance as a secondary endpoint during unmasked CGM compared with self-monitoring of blood glucose (SMBG) with capillary strip testing in combination with masked CGM.⁶ In a retrospective audit of 35 people with type 1 diabetes and hypoglycemia unawareness, CGM reduced severe hypoglycemia and improved HbA_1c but failed to restore hypoglycemia awareness.⁷ There have been no prospective comparative trials of CGM against utilization of SMBG testing as part of usual management in the community in either type 1 or at-risk type 2 diabetes with severe hypoglycemia as a primary endpoint.

Conventional CGM measures interstitial glucose levels through insertion of a subcutaneous sensor, typically into the anterior abdominal wall, with or without calibration with SMBG, and lasts up to 10 days.⁸ An alternative is the recently developed intermittently scanned continuous glucose monitoring (isCGM; FreeStyle Libre™, Abbott Diabetes Care, Doncaster, Victoria, Australia) that comprises a precalibrated subcutaneous sensor worn on the back of the upper arm, which lasts up to 14 days. This sensor automatically stores glucose data every 15 min, but a real-time glucose concentration can be accessed up to every minute by scanning the sensor with a dedicated reader with a display or another device with radio frequency identification capability. The reader stores data for 90 days and the data can be uploaded to a computer for subsequent review.⁹ Although there is some evidence that conventional CGM may be superior to isCGM in preventing hypoglycemia in people with type 1 diabetes and hypoglycemia unawareness,¹⁰ the isCGM system is less expensive and easier to use.¹¹

The aim of this study was to assess the efficacy of the FreeStyle Libre™ system compared with usual-care SMBG in reducing the incidence of hypoglycemia requiring second-party assistance in insulin-treated adults with a recently confirmed episode of clinically significant hypoglycemia.

Research Design and Methods

Study design, participants, and approvals

The present study was a 24-week, prospective, open-label, two-arm randomized controlled trial conducted between September 2017 and December 2018. Participants aged at least 18 years with insulin-treated diabetes who had experienced a confirmed symptomatic (self- or laboratory-measured blood glucose (BG) <4.0 mM) or clinically significant biochemical (self- or laboratory-measured BG <3.0 mM) hypoglycemic episode requiring health service attendance or prompt review by a diabetes health professional in the 2 weeks before recruitment were eligible. Inclusion criteria comprised no cognitive impairment or other disability such as visual impairment that would interfere with the participant's ability to comply with all study requirements, including providing informed consent, completion of documentation, and operation of the FreeStyle Libre™ system, and no known allergy to adhesive dressings that might result in a local reaction when the sensor device was in place. Participants were recruited from the Emergency Department or clinic attendances at Fiona Stanley Hospital, Fremantle Hospital, and/or local specialist clinics, or through community advertisements. Ethical approval was obtained from the Southern Metropolitan Health Service Human Research Ethics Committee. Written informed consent was obtained from all participants. The trial was registered with the Australia New Zealand Clinical Trials Registry (ACTRN12616001695493).

Methods

Participants were randomized 1:1 to 6 months of FreeStyle Libre™ isCGM (intervention group) or to usual-care SMBG (control group) without CGM according to a computer-generated list in random permuted blocks of 2–6 stratified by type of qualifying episode (clinically significant biochemical or symptomatic). Participants, investigators, and study staff were not masked to group allocation.

At the baseline visit, participants completed a detailed questionnaire covering sociodemographic and lifestyle details, history of diabetes, including current use of SMBG or CGM and BG-lowering therapies, and other medical history and medications. A detailed history of the qualifying hypoglycemic episode was taken. A brief physical examination was performed, and blood and urine collected for routine biochemical tests in a single nationally accredited pathology laboratory.

Following randomization, education was provided to all participants, including detailed advice regarding avoidance and management of hypoglycemia in those who had not received this from a credentialed diabetes health professional as part of the management of their qualifying episode, and a diary was issued for recording SMBG results and to document other events, including hypoglycemia. Adjustments to insulin doses and other BG-lowering therapies, as well as decisions as to the frequency of SMBG in those allocated to this form of monitoring, were left to the participant and/or his/her usual-care health professionals as part of usual care.

Those randomized to isCGM were given full device-specific education, including making the participant aware of the delay between the actual BG level and FreeStyle Libre™ system result and the need for confirmatory SMBG if an isCGM result was within the hypoglycemic range, and they were provided with two sensors. The frequency of scanning of the sensor was, as per the manufacturer's user's manual, recommended to be a minimum of every 8 h to facilitate complete data collection, but no less than that through prior SMBG.

Follow-up visits for all participants were scheduled 4 weekly at which SMBG diaries and isCGM data downloads were reviewed, and any medication changes and adverse events were documented. For those randomized to isCGM, two new sensors were provided. At the final 24-week visit, the physical examination and biochemical testing were repeated.

Outcomes

The primary outcome was the incidence of severe hypoglycemia requiring second-party assistance for recovery during the 6-month follow-up period. Prespecified secondary outcomes included the incidence of any self-reported significant hypoglycemia, recorded hypoglycemic episodes (BG or subcutaneous glucose <3.0, 3.0–3.9, <4.0 mM), the change in HbA_1c between entry and last visit, and the change in SMBG during follow-up. Safety outcomes incorporated all adverse events, including sensor insertion or sensor wear-related issues, diabetic ketoacidosis or hyperosmolar hyperglycemic states, and cardiovascular disease events, but excluded hypoglycemic events since these were primary/secondary outcomes.

Statistical analysis

The incidence of severe hypoglycemia requiring health service use in people with type 1 or insulin-treated type 2 diabetes has been quantified at 0.12 episodes per person-year.¹² This rate, a conservative estimate given that some severe hypoglycemia requiring second-party assistance will not utilize health services, is increased at least sixfold in those with hypoglycemia unawareness,¹ or to ≥0.72 episodes per person-year or ∼0.4 episodes per 6 months of follow-up. We assumed, therefore, that 40% of our participants in the usual care group would experience the primary outcome. A comparative study in type 1 diabetes has shown that use of the FreeStyle Libre™ system was associated with a graded reduction in hypoglycemic events, with a 33% lower rate versus SMBG for those <3.9 mM and up to a 55% lower rate for those <2.2 mM.³ Given that this latter study was part of usual care in clinic-based participants without hypoglycemia as an outcome,³ and since the present study was designed specifically to allow high-risk participants to use isCGM to anticipate such episodes, we assumed that there would be a 75% reduction in the primary outcome in the intervention group (or a rate of 10% over 6 months). The study had 80% power with a one-sided α of 0.05 to detect this 30% between-group absolute difference with 29 participants per group required after allowing for 15% attrition postrandomization. Results presented are for the intention-to-treat (ITT) population utilizing the full analysis set supplemented by a per protocol analysis that included only those participants who completed the study.

Data are presented as mean ± standard deviation or, in the case of variables that did not conform to a normal distribution, median and interquartile range (IQR). Incidence rates (95% confidence intervals [CI]) for severe hypoglycemia requiring second-party assistance and recorded hypoglycemic episodes were calculated as the total number of each event experienced during follow-up divided by the total follow-up time. These were compared by intervention status, and incident rate ratios (IRR) and 95% CI were calculated. The prespecified secondary endpoint of change in HbA_1c at 24 weeks was analyzed based on the between-intervention difference (95% CI) in estimated marginal mean (EMM) change (95% CI) from baseline at α of 0.05 using an analysis of covariance with intervention as a fixed factor and baseline HbA_1c as a covariate to adjust for any chance imbalance between groups in baseline HbA_1c. Missing HbA_1c values at 24 weeks were imputed using the last-observation-carried-forward for the ITT analysis.

Results

Participant characteristics

Fifty-nine participants were randomized between September 2017 and July 2018 and were included in the ITT population. Of these, 30 were in the intervention group and 29 were in the control group. Consistent with the fact that CGM (including isCGM) was not government-subsidized for adults with diabetes at the time of recruitment, only one participant was using CGM (isCGM) at study entry and only five of the participants (all with type 1 diabetes) had used CGM previously. The median BG recorded during the qualifying hypoglycemic episode was 2.8 mM (median [IQR]: 2.7 [2.5–3.3] mM in the control group and 2.8 [2.4–3.5] mM in the intervention group). There were 34 participants (57.6%) who had clinically significant biochemical hypoglycemia (documented BG <3.0 mM; 58.6% of the control group and 56.7% of the intervention group), the remainder experiencing a BG <4.0 mM with autonomic symptoms. The baseline demographic, disease-specific, and clinical characteristics of the participants were similar in the two groups (Table 1).

Table 1.

Baseline Characteristics of the Trial Participants by Treatment Assignment (Intention-to-Treat Population)

	Control (n = 29)	Intervention (n = 30)
Age (years)	51.9 ± 14.8	55.3 ± 14.2
Male (%)	51.7	36.7
Anglo-Celt ethnicity (%)	82.8	90.0
Education beyond secondary level (%)	58.6	50.0
Alcohol consumption (units/day)	1 (1–8)	3 (0–8)
Current smoking (%)	6.9	10.0
Age at diabetes diagnosis (years)	27.7 ± 15.9	30.3 ± 13.6
Diabetes duration (%)	24.7 (13.4–32.7)	24.4 (12.2–32.2)
Diabetes type (1/LADA/2) (%)	62.1/3.4/34.5	73.3/3.3/23.3
Insulin treatment (injections only/injections+others^a/pump) (%)	51.7/27.6/20.7	50.0/26.7/23.3
Body mass index (kg/m²)	29.7 ± 5.5	27.7 ± 5.0
Systolic blood pressure (mmHg)	126 ± 18	128 ± 18
Diastolic blood pressure (mmHg)	72 ± 10	75 ± 9
HbA_1c (mmol/mol)	62 (50–69)^b	63 (57–71)
HbA_1c (%)	7.8 (6.7–8.5)^b	7.9 (7.4–8.6)
SMBG frequency (times/day)	4 (2–6)	4 (3–7)
Antihypertensive therapy (%)	51.7	43.3
Lipid-modifying therapy (%)	51.7	46.7
Aspirin (%)	27.6	13.3
Myocardial infarction and/or coronary revascularization (%)	20.7	13.1
Stroke/transient ischemic attack (%)	6.9	0
Heart failure (%)	6.9	0
Chronic kidney disease (%)	6.9	0
Claudication (%)	3.4	0
Peripheral sensory neuropathy (%)	66.7	33.3
Retinopathy/prior laser treatment	24.1	16.7
Cataracts/intraocular lens replacement	24.1	23.3
Cancer	3.4	13.3
Asthma/chronic lung disease	10.3	23.3

Oral glucose-lowering therapies and/or glucagon-like peptide 1 analogues.

n = 27.

LADA, latent autoimmune diabetes of adults; SMBG, self-monitoring of blood glucose.

Primary and secondary outcomes

Forty-eight participants (81.4%) completed the trial, with a greater number of withdrawals in the control group (10 vs. 1 in the intervention group, P = 0.002). Four of these 11 could not be contacted, 6 had personal or family issues, which precluded further participation (including the 1 withdrawal from the intervention group), and 1 control participant started isCGM during the trial. The total duration of follow-up was 3719 days in the control group and 4991 in the intervention group (median [IQR]: 168 [52–181] and 168 [166–175], respectively; P = 0.59).

The primary and secondary outcomes by intervention status for the full analysis set are shown in Table 2. There were 10 primary outcomes in the intervention group and 5 primary outcomes in the control group with IRs that were not significantly different between the two groups [IRR for intervention vs. control (95% CI): 1.49 (0.46–5.56), P = 0.47]. The incidence rate of any self-reported significant hypoglycemia was 47% higher in the intervention than control group (Table 2; P < 0.001). The incidence rate of recorded hypoglycemic episodes was twice as high in the intervention group versus the control group for all definitions of hypoglycemia. For BG or isCGM glucose concentrations <3.0 mM, the IRR was 2.05 (1.80–2.33), for 3.0–3.9 mM, the IRR was 2.16 (1.97–2.36), and for <4.0 mM, the IRR was 2.12 (1.97–2.28) (all P < 0.001). Per protocol analysis of primary and secondary endpoints was also performed and this showed results that were consistent with the ITT analysis (Supplementary Table S1).

Table 2.

Primary and Secondary Outcomes by Intervention Status for the Full Analysis Set

	Control (n = 25)	Intervention (n = 30)	P
Primary outcome
Hypoglycemia requiring second-party assistance
Number (median [IQR])	5 (0 [0–0])	10 (0 [0–1])	0.49
Incidence rate (CI)/person-year	0.49 (0.16–1.15)	0.73 (0.35–1.35)	0.47
Secondary outcomes
Any self-reported hypoglycemic episode (%)	92.0	100	0.20
Number (median [IQR])	786 (14 [8–64])	1562 (39 [16–68])	0.08
Incidence rate (CI)/person-year	78 (73–84)	114 (109–120)	<0.001
Blood/subcutaneous glucose <3.0 mM
Number (median [IQR])	335 (5 [2–27])	920 (22 [10–43])	0.003
Incidence rate (CI)/person-year	33 (29–37)	67 (63–72)	<0.001
Blood/subcutaneous glucose 3.0–3.9 mM
Number (median [IQR])	648 (21 [9–39])	1877 (52 [29–79])	<0.001
Incidence rate (CI)/person-year	64 (59–69)	137 (131–144)	<0.001
Blood/subcutaneous glucose <4.0 mM
Number (median [IQR])	983 (26 [14–66])	2797 (75 [45–115])	<0.001
Incidence rate (CI)/person-year	96 (91–103)	205 (197–212)	<0.001
Change in HbA_1c (mmol/mol)	−0.7 ± 5.5^a	−2.3 ± 8.0	0.40
Change in HbA_1c (%)	−0.06 ± 0.51^a	−0.21 ± 0.73	0.40

n = 27.

CI, confidence interval; IQR, interquartile range.

The EMM change in HbA_1c at 24 weeks was −1.2 mmol/mol (−3.7 to 1.3 mmol/mol) (−0.11% [−0.34% to 0.12%]) in the control group and −1.8 mmol/mol (−4.2 to 0.6 mmol/mol) (−0.16% [−0.38% to 0.05%]) in the intervention group after adjusting for baseline HbA_1c. The difference (95% CI) in the EMM change between the control and intervention was 0.6 mmol/mol (−2.9 to 4.1 mmol/mol) (0.05% [−0.27% to 0.37%]) after adjusting for baseline HbA_1c; F(1, 54) = 0.11, P = 0.74).

At baseline, all 59 study participants were self-monitoring their BG levels a median of 4 times/day. At the final visit at 24 weeks, over half (53.6%) of the intervention group had stopped SMBG completely, but most were accessing isCGM at least 4 times a day (range 1 to >30 times/day). The 11 participants in the intervention group who continued to self-monitor had reduced the median frequency to 1/day. All continuing participants in the control group continued to self-monitor a median of 4 times/day.

Safety outcomes

Six participants reported seven serious adverse events during follow-up (Table 3 and Supplementary Table S2), but there was no statistical difference in the proportion or incidence rate by intervention status (P ≥ 0.51) and none of these events led to withdrawal from the trial. Thirty-four participants had 64 nonserious adverse events with no statistical difference in the proportion or incidence rate by intervention status (P ≥ 0.31). There were no intervention-related nonserious adverse events and none led to discontinuation.

Table 3.

Summary of Adverse Events by Intervention Status for the Full Analysis Set

	Control	Intervention	P
Number of participants in full analysis set	25	30
Number (%) of participants reporting serious adverse events	2 (8.0)	4 (13.3)	0.68
Number (median [IQR]) of serious adverse events	2 (0 [0–0])	5 (0 [0–0])	0.51
Number (%) of participants reporting nonserious adverse events	14 (56.0)	20 (66.7)	0.58
Number (median [IQR]) of nonserious adverse events	23 (1 [0–2])	41 (1 [0–2])	0.31

The serious adverse events included two episodes of diabetic ketoacidosis that required inpatient management but resolved without complications, one uncomplicated miscarriage, and three admissions to hospital for 2 or more days for nonglycemic medical complaints, two of these hospital admissions being for a single participant. There were no sensor-wear issues among intervention participants, but two reported problems with sensor retention at the 4-week follow-up visit.

Discussion

The present study was designed to pragmatically assess the utility of the FreeStyle Libre™ system as a way of reducing the most clinically severe form of hypoglycemia (that requiring second-party assistance) in some of the highest risk people with diabetes, namely those with a recent previous significant episode and thus at substantial risk of hypoglycemia unawareness.¹ Although a positive outcome would make a strong argument for adoption of isCGM as a routine adjunct to the acute management of such individuals, especially given the high cost of health service utilization due to hypoglycemia,¹³ there was no significant difference in subsequent severe hypoglycemia between those allocated to the FreeStyle Libre™ and those continuing SMBG over the 6-month follow-up period in our trial. As has been reported in previous studies,¹⁴ as seen in our own secondary endpoints, and as would be expected from ready availability of the results of more frequent interstitial glucose measurements compared with SMBG, other manifestations of hypoglycemia were detected significantly more often by isCGM than SMBG in our participants.

Previous studies of CGM in people with hypoglycemia unawareness that have suggested benefits for severe hypoglycemia were conducted in relatively small numbers of clinic-based individuals with type 1, whose insulin regimens were under clinical supervision, did not utilize isCGM, and were either retrospective⁷ or hypothesis-generating given that severe hypoglycemia was a secondary outcome.⁶ In addition, the use of masked CGM during the SMBG phase of the crossover study⁶ may have had a significant impact on participant behavior that influenced glycemic outcomes. Our trial was designed to minimize the influence of study procedures on usual-care self-management. We used monthly study visits to ensure regular data collection, including self-reported hypoglycemic events, and in-date supply of FreeStyle Libre™ sensors to participants in the intervention group.

The FreeStyle Libre™ system was safe and most of our participants who used it did not have any problems with the technology or its application. Although all, but one of the intervention group, completed all study procedures compared with only 19 of 29 in the control group, suggesting a high degree of participant satisfaction with the device, there was no difference in the change in HbA_1c between the two groups over the 6-month follow-up. In two 6-month randomized studies of the FreeStyle Libre™ system versus SMBG in insulin-treated type 2 diabetes, there was similarly no effect on HbA_1c but reduced hypoglycemia with FreeStyle Libre™ use in one,¹⁵ and a lower HbA_1c but no effect on hypoglycemia in the other.¹⁶ Differences between study design in key aspects such as participant selection criteria (documented hypoglycemia from the community in the present trial vs. clinic-based participants in the two previous studies^15,16) and baseline glycemic control (median HbA_1c 63 mmol/mol or 7.9% in the present study vs. 73 mmol/mol or 8.8%¹⁵ and 71 mM or 8.6%¹⁶) together with the discordant findings suggest that the benefits of application of technology such as the FreeStyle Libre™ system are highly context dependent.

Of relevance to this issue, we restricted our participants to those with a recent single hypoglycemic episode because there are compelling data demonstrating a blunted autonomic response in this situation,¹⁷ which only recovers after at least 2 weeks of strict hypoglycemia avoidance.¹⁸ Supportive evidence for our participants having significant hypoglycemia unawareness comes from the fact that the incidence of self-reported episodes of hypoglycemia during follow-up was lower than biochemically significant episodes (those <4.0 mM, reflecting thresholds for detection of typical autonomic symptoms ranging from <3.6 mM in healthy volunteers up to <4.3 mM in people with poorly controlled diabetes^19,20) captured by SMBG or especially isCGM (Table 2). However, there could be other specific categories of hypoglycemia-prone insulin-treated individuals, such as those with very frequent episodes²¹ or those with chronic kidney disease,²² who might benefit from access to isCGM for hypoglycemia prevention.

The lack of efficacy of isCGM in preventing recurrence of hypoglycemia as hypothesized in our participants may reflect a number of factors. First, the greater quantity and more ready availability of isCGM versus SMBG data may have increased the likelihood of insulin dose adjustments that led to an increased risk of hypoglycemia in the intervention group and, in turn, an attenuation of isCGM benefits for hypoglycemia prevention. We did not collect detailed insulin dosage data and so an examination of this potential temporal relationship was beyond the scope of the present study. Second, isCGM has relatively low accuracy at sub-normal BG levels and can produce lower values than SMBG in this situation.²³ Although this may have led to earlier preventive action than necessary in those allocated isCGM, it may also have progressively and inappropriately lessened concerns regarding the risk of progression to severe hypoglycemia when isCGM values were low. Indeed, there were 11 participants allocated isCGM who had stopped doing SMBG during the trial despite most having episodes of hypoglycemia on isCGM profiles that should have prompted confirmation by SMBG. Third, despite the detailed education regarding the delay in subcutaneous relative to BG concentrations in isCGM users that was provided at the time of randomization, some of the participants allocated to isCGM may have forgotten or ignored this advice, thus putting them at increased risk of a severe episode. Notwithstanding accuracy issues at low BG levels, it is possible that the introduction of an alarm system on more recent versions of the FreeStyle Libre™ system may mean that it is more effective than SMBG in preventing recurrent hypoglycemia, but this awaits a definitive study.

The present study had limitations. We estimated our sample size on available published data relating to health service utilization for severe hypoglycemia in the context of hypoglycemia unawareness that might not have been applicable to a contemporary urban Australian setting.^1,12 In addition, we did not formally assess hypoglycemia unawareness as it was not a criterion for the qualifying hypoglycemia episode. Nevertheless, our overall incidence of severe hypoglycemia requiring second-party assistance (0.63 [95% CI: 0.41–0.80] episodes per person-year) was close to the 0.72 episodes per person-year on which we based our power calculations. In addition, and notwithstanding differences in the study design, our sample size was greater than those used in previous studies assessing the impact of CGM on severe hypoglycemia risk in people with type 1 diabetes and hypoglycemia unawareness.^6,7

In conclusion, the present 6-month randomized unmasked trial did not show that the FreeStyle Libre™ system was better than conventional SMBG in preventing severe hypoglycemia requiring second-party assistance in community-based insulin-treated people who had experienced a recent hypoglycemic episode. However, the rapid development of CGM, including isCGM technology, and its potential for integration with other devices with potential benefits for hypoglycemia detection such as activity trackers and other wearable physiological sensors,⁸ offers future hope for people at high risk of severe hypoglycemia.

Footnotes

Acknowledgments

We thank the participants for their participation, the staff from Fiona Stanley Hospital, Fremantle Hospital, and Diabetes Western Australia for assistance in participant identification and recruitment, and PathWest Laboratory Medicine for biochemical tests.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was funded by a grant from the Spinnaker Health Research Foundation.

Supplementary Material

Research Protocol

Supplementary Table S1

Supplementary Table S2

References

Cryer

: Mechanisms of hypoglycemia-associated autonomic failure in diabetes. N Engl J Med, 2013; 369:362–372.

Battelino

, Phillip

, Bratina

, et al.: Effect of continuous glucose monitoring on hypoglycemia in type 1 diabetes. Diabetes Care, 2011; 34:795–800.

Bolinder

, Antuna

, Geelhoed-Duijvestijn

, et al.: Novel glucose-sensing technology and hypoglycaemia in type 1 diabetes: a multicentre, non-masked, randomised controlled trial. Lancet, 2016; 388:2254–2263.

Haak

, Hanaire

, Ajjan

, et al.: Use of flash glucose-sensing technology for 12 months as a replacement for blood glucose monitoring in insulin-treated type 2 diabetes. Diabetes Ther, 2017; 8:573–586.

Grunberger

, Bailey

, Camacho

, et al.: Proceedings from the American Association of Clinical Endocrinologists and American College of Endocrinology consensus conference on glucose monitoring. Endocr Pract, 2015; 21:522–533.

van Beers

, DeVries

, Kleijer

, et al.: Continuous glucose monitoring for patients with type 1 diabetes and impaired awareness of hypoglycaemia (IN CONTROL): a randomised, open-label, crossover trial. Lancet Diabetes Endocrinol, 2016; 4:893–902.

Choudhary

, Ramasamy

, Green

, et al.: Real-time continuous glucose monitoring significantly reduces severe hypoglycemia in hypoglycemia-unaware patients with type 1 diabetes. Diabetes Care, 2013; 36:4160–4162.

Cappon

, Vettoretti

, Sparacino

, Facchinetti

: Continuous glucose monitoring sensors for diabetes management: a review of technologies and applications. Diabetes Metab J, 2019; 43:383–397.

Abbott Diabetes Care: FreeStyle Libre Software. www.FreeStyleLibre.com.au (accessed September, 2019).

10.

Reddy

, Jugnee

, El Laboudi

, et al.: A randomized controlled pilot study of continuous glucose monitoring and flash glucose monitoring in people with type 1 diabetes and impaired awareness of hypoglycaemia. Diabet Med, 2018; 35:483–490.

11.

Hirsch

: Professional flash continuous glucose monitoring as a supplement to A1C in primary care. Postgrad Med, 2017; 129:781–790.

12.

Leese

, Wang

, Broomhall

, et al.: Frequency of severe hypoglycemia requiring emergency treatment in type 1 and type 2 diabetes—a population-based study of health service resource use. Diabetes Care, 2003; 26:1176–1180.

13.

de Groot

, Enters-Weijnen

, Geelhoed-Duijvestijn

, Kanters

: A cost of illness study of hypoglycaemic events in insulin-treated diabetes in the Netherlands. BMJ Open, 2018; 8:e019864.

14.

Blumer

: The contemporary role of masked continuous glucose monitoring in a real-time world. J Diabetes Sci Technol, 2016; 10:790–792.

15.

Haak

, Hanaire

, Ajjan

, et al.: Flash glucose-sensing technology as a replacement for blood glucose monitoring for the management of insulin-treated type 2 diabetes: a multicenter, open-label randomized controlled trial. Diabetes Ther, 2017; 8:55–73.

16.

Ajjan

, Jackson

, Thomson

: Reduction in HbA1c using professional flash glucose monitoring in insulin-treated type 2 diabetes patients managed in primary and secondary care settings: a pilot, multicentre, randomised controlled trial. Diab Vasc Dis Res, 2019; 16:385–395.

17.

Widom

, Simonson

: Intermittent hypoglycemia impairs glucose counterregulation. Diabetes, 1992; 41:1597–1602.

18.

Fanelli

, Epifano

, Rambotti

, et al.: Meticulous prevention of hypoglycemia normalizes the glycemic thresholds and magnitude of most of neuroendocrine responses to, symptoms of, and cognitive function during hypoglycemia in intensively treated patients with short-term IDDM. Diabetes, 1993; 42:1683–1689.

19.

Mitrakou

, Ryan

, Veneman

, et al.: Hierarchy of glycemic thresholds for counterregulatory hormone secretion, symptoms, and cerebral dysfunction. Am J Physiol, 1991; 260(1 Pt 1):E67–E74.

20.

Boyle

, Schwartz

, Shah

, et al.: Plasma glucose concentrations at the onset of hypoglycemic symptoms in patients with poorly controlled diabetes and in nondiabetics. N Engl J Med, 1988; 318:1487–1492.

21.

Flatt

AJS

, Little

, Speight

, et al.: Predictors of recurrent severe hypoglycemia in adults with type 1 diabetes and impaired awareness of hypoglycemia during the HypoCOMPaSS study. Diabetes Care, 2019 [Epub ahead of print]; DOI: 10.2337/dc19-0630.

22.

Ahmad

, Zelnick

, Batacchi

, et al.: Hypoglycemia in people with type 2 diabetes and CKD. Clin J Am Soc Nephrol, 2019; 14:844–853.

23.

Mancini

, Berioli

, Santi

, et al.: Flash glucose monitoring: a review of the literature with a special focus on type 1 diabetes. Nutrients, 2018; 10:E992.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.29 MB

0.01 MB