The Effectiveness of Continuous Glucose Monitoring in Patients with Type 2 Diabetes: A Systematic Review of Literature and Meta-analysis

Abstract

Background:

Continuous glucose monitoring (CGM) provides glucose trend information that can be used to guide treatment and motivate patients with diabetes. Currently, the evidence on effectiveness of CGM in patients with type 2 diabetes is debatable. We aim to provide a systematic review and meta-analysis to synthesize current evidence of effectiveness of CGM in adults with type 2 diabetes.

Materials and Methods:

Cochrane, Embase, PubMed, and Web of Science were searched to include all studies that reported effectiveness of CGM in terms of HbA1c in adults aged 18 and older, with type 2 diabetes, on any treatment for diabetes. Heterogeneity (I ²) was used to determine the variability between studies. All data were analyzed using Review Manager 5.3 software.

Results:

Seven randomized controlled trials and three cohort studies, involving 1384 patients for real-time CGM (RT-CGM) and professional CGM (P-CGM) and 4902 patients for flash glucose monitoring (FGM), were included. RT-CGM and P-CGM were associated with a modest but statistically significant reduction in HbA1c of 0.20% (95% confidence interval [CI] −0.31 to −0.09) compared with control. Patients who received FGM had a nonsignificant reduction of 0.02% (95% CI −0.07 to 0.04) compared with control. The meta-analysis indicated a low heterogeneity between studies.

Conclusion:

Our analysis of current evidence suggests that RT-CGM and P-CGM are effective in improving HbA1c in adults with type 2 diabetes. Due to insufficient evidence, it is premature to form conclusions on the effectiveness of FGM. Future multicenter trials accessing the effectiveness of CGM as well as safety and cost-effectiveness may be necessary.

Introduction

The global prevalence of diabetes in adults aged 18–99 years was 8.4% in 2017 and is expected to increase to 9.9% in 2045.¹ With the increasing prevalence, an effective treatment strategy is needed to improve glycemic control and prevent complications.² Self-monitoring blood glucose (SMBG) and HbA1c levels are traditionally used to assess control and guide treatment.³ However, SMBG and HbA1c fail to detect fluctuations in glucose as well as nocturnal and asymptomatic hypoglycemia.^3,4 The continuous glucose monitoring (CGM) system addresses these limitations by measuring patient's glucose levels continuously throughout the day, which provides glucose trend information and high and low glucose values.

The CGM system has a body sensor that is inserted subcutaneously to measure glucose levels in the interstitial fluid of the patient. Depending on the device, the transmitter wirelessly sends glucose values to a receiver as often as every 5 min. The CGM data can be used to treat and prevent hypoglycemic and hyperglycemic episodes and encourage lifestyle modifications in patients. The American Diabetes Association currently recommends CGM in patients with hypoglycemia unawareness and/or frequent hypoglycemic episodes.⁴

Professional CGM (P-CGM) and real-time CGM (RT-CGM) are available. RT-CGM provides real-time data with high and low glucose alerts to the patient.⁵ In contrast, real-time data are masked to the patient using P-CGM and are retrospectively evaluated by the clinician. Both devices require fingerstick calibration by the patient at least twice a day. Education and training are required to integrate information into treatment decisions and to prevent adverse events.

Recently in 2017, another variety of CGM system, FreeStyle^® Libre Flash (flash glucose monitoring [FGM]), was approved in the United States by the U.S. Food and Drug Administration (FDA) for adults older than 18. FGM is significantly different from other CGM because it provides glucose values for the previous 8-h period only on scanning, does not give glucose alerts, and is factory callibrated.^3,6 Since no fingerstick calibration is needed, FGM provides greater convenience to the patient and avoids errors related to the calibration. It is also lower in cost compared with other CGM.

Currently, the effectiveness of CGM in patients with type 2 diabetes on reduction of HbA1c level is not consistent. Meta-analysis studies have been conducted in the past, which reported a reduction in HbA1c in patients who received RT-CGM and P-CGM.^7,8 However, the results were based on older evidence from a few randomized controlled trials (RCTs). Since the time of publication, larger studies have been conducted using more advanced CGM technology, and not all studies reported significant improvement in glycemic control. Also, no previous meta-analysis studies have been conducted on effectiveness of FGM in improving HbA1c. Due to inconsistent results between studies and the recent approval of FGM, we performed a systematic review and meta-analysis to synthesize current evidence of effectiveness of the CGM systems in terms of improving HbA1c measures in adults with type 2 diabetes.

Materials and Methods

This systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) statement.⁹ All included studies received informed consents and ethics approval from their respective institutions.

Search strategy and study selection

A comprehensive search for literature was conducted through databases, including Cochrane, Embase, PubMed, and Web of Science. The controlled vocabulary and keywords used in the search were diabetes mellitus, glycated hemoglobin A, CGM, FGM, blood glucose monitoring, and ambulatory glucose profile. The detailed search strategy is reported in Supplementary Table S1.

Identified articles were initially screened for relevance to research question by title and abstract. Full text was retrieved for relevant studies and reviewed further for exclusion criteria (Fig. 1).

FIG. 1.

Study selection process.

Selection criteria

RCTs and observational studies that reported change in HbA1c in the CGM and control groups were considered for review and analysis. Included patient populations were adults aged 18 and older with type 2 diabetes on any treatment for diabetes such as lifestyle modifications, oral hypoglycemic agents, insulin therapy, and combination therapy. Exclusion criteria were duplicates, reviews, news, commentaries, and studies that included the following patient populations: type 1 diabetes or unable to separate from type 2 diabetes data, pregnant or gestational diabetes, neonatal and adolescents, hemodialysis, and hospitalized or critically ill (Fig. 1). No studies were excluded based on study size, publication year, and geographic location.

Data extraction

Data were collected from individual studies for sample size, study duration, country, device manufacturer, baseline patient characteristics (age, gender, and insulin use), inclusion and exclusion criteria, change in HbA1c from baseline, and other outcomes (adverse effects, blood pressure, body mass index [BMI], weight, exercise time, and changes in hypoglycemic agents).

Assessment of risk of bias

The Cochrane risk of bias tool was used to appraise the quality of research methods of RCTs based on random sequence generation, allocation concealment, blinding of participants, data collectors and outcome assessor, bias due to selective outcome reporting, percentage of total population who were lost to follow-up, intention-to-treat analysis, and source of study funding (Supplementary Table S2).¹⁰ The Newcastle–Ottawa Scale was used to appraise the quality of research methods of observational studies.¹¹ There were a maximum of four stars that can be earned in the category of selection, two stars in comparability, and three stars in outcome (Supplementary Table S3).

Statistical analysis

The meta-analysis was conducted for RT-CGM and P-CGM studies and for FGM studies. The outcome was change in HbA1c from baseline in treatment and control groups. Heterogeneity (I ²) was used to determine the variability between studies. I ² values of 25%, 50%, and 75% were assigned low, moderate, and high heterogeneity, respectively.¹² Potential publication bias was evaluated by funnel plot. All data were analyzed using Review Manager 5.3 software.¹³

Results

Search results

The initial search from databases identified 1011 potentially relevant studies. Of these, 912 articles were removed for being irrelevant to research question. Further excluded articles were reviews, news, and commentaries (37), duplicates (12), and studies that included the following patient populations: type 1 diabetes or unable to separate from type 2 diabetes data (22), pregnant or gestational diabetes (7), adolescents (2), hemodialysis (6), or hospitalized (3). Six RCTs and two cohort studies that used P-CGM and RT-CGM were included for review and analysis (Fig. 1). For the analysis of FGM, one RCT and one observational study were included.

Study and patient characteristics

The details of the included studies are provided in Table 1. Studies were published between 2008 and 2017 in North America (3), Europe (2), and Asia (5). Study duration ranged from 8 to 32 weeks. Devices used were Dexcom™ SEVEN^®, Dexcom™ G4 Platinum, iPro^®2, Medtronic™ MiniMed, GlucoDay^®, Guardian^® RT, and FreeStyle^® Libre. All trials reported change in HbA1c as a primary outcome.

Table 1.

Study Details

Study (publication year)	Total sample size	Study duration (weeks)	Country	Device	Mean age (SD) (years)	Gender, males (%)	Insulin use	Inclusion criteria	Exclusion criteria	Change in HbA1c from baseline (SD) (%)	Other outcomes	Details
Randomized controlled trials
Beck et al.¹⁴ (2017)	158	24	United States, Canada	Dexcom™ G4 Platinum	CGM 60 (11); control 60 (9)	44	156	Age ≥25 years; type 2 diabetes treated with multiple daily injections of insulin for ≥1 year; HbA1c levels of 7.5% to 10%; stable diabetes medication regimen ≥3 months	Use of RT-CGM past ≤3 months; plan to use CGM during study; pregnancy; renal disease	CGM 8.5 (0.6) to 7.7 (0.7); control 8.5 (0.7) to 8 (0.9); adjusted difference −0.3 (95% CI −0.5% to 0%); p = 0.022	No meaningful differences in severe hypoglycemia, diabetic ketoacidosis, or quality-of-life outcomes	CGM used daily; control group asked to monitor BG at least 4 times a day; therapy managed by the usual providers
Cosson et al.¹⁵ (2009)	25	24	France	GlucoDay^®	CGM 57.2 (4.4); control 57.3 (5.9)	68	9	Age 40–70 years; type 2 diabetes treated with oral antidiabetic agents with or without insulin at stable dosage ≥3 months; HbA1c levels of 8%–10.5%; SMBG ≥4 times a week; no previous experience with CGM; routine follow-up at the study center for ≥1 year	Pregnancy; acute disease with subsequent poor glycemic control; proliferative retinopathy; and renal failure	CGM 9.22 (0.99) to 8.59 (1.04), adjusted difference −0.63 (0.34), p = 0.05; control 9.07 (0.6) to 8.76 (1.43), adjusted difference −0.31 (0.29), p = 0.18	5.8% CGM group reported uncomplicated implantation site skin reactions; no significant differences in hypoglycemia	CGM implanted and removed after 48 h for both groups (repeated after 3 months); only intervention group had CGM results disclosed; therapy managed by their usual providers
Ehrhardt et al.²¹ (2011)	100	12	United States	Dexcom™ SEVEN^®	CGM 60 (11.9); control 55.5 (9.6)	55	33	Age ≥18 years; type 2 diabetes for ≥3 months; HbA1c levels of 7% to 12%; willing to perform SMBG 4 times daily	Use of prandial insulin; pregnant, lactating, or attempting pregnancy; use of glucocorticoids, amphetamines, anabolic or weight reducing medications	CGM 8.4 (1.3) to 7.4 (1); control 8.2 (1.1) to 7.7 (1.2); adjusted difference −0.6; p = 0.002	No differences in net change in number or dosage of hypoglycemic agents; no improvements in blood pressure and weight	Four 2-week cycles (2 weeks on/1 week off) of RT-CGM for a total of 12 weeks; therapy managed by their usual providers
Sato et al.²⁷ (2016)	34	32	Japan	iPro^®2	CGM 59.9 (9); control 63.1 (8.5)	59	34	Age ≥20 years; type 2 diabetes on insulin; HbA1c levels of 6.9% to 11% and fluctuations within ±5% prior 3 months; regular visit to clinic	Proliferative retinopathy; serious liver disease; serious kidney disease; acute heart failure; active malignancy; serious pancreatic disease; pregnancy	CGM 8.2 (1.2) to 8.2 (1.3), p ≥ 0.05; control 8.2 (0.9) to 7.9 (0.8), p ≥ 0.05	No significant difference in patient satisfaction (DTSQ score) between groups	2 Visits with 4- to 5-day continuous recording of retrospective CGM per visit for both groups; only intervention group had CGM results disclosed; guidance and treatment recommendations provided by study team
Yoo et al.¹⁶ (2008)	57	12	South Korea	Guardian^® RT	CGM 54.6 (6.8); control 57.5 (9)	42	32	Age 20–80 years; type 2 diabetes treated with insulin or oral hypoglycemic agents for ≥1 year; HbA1c levels of 8% to 10%; stable diabetes medication regimen ≥2 months	Severe diabetic complications; corticosteroid use past ≤3 months; liver disease; renal insufficiency	CGM 9.1 (1) to 8 (1.2); control 8.7 (0.7) to 8.3 (1.1); p = 0.004	Significant increase in total exercise time per week in CGM group compared with control group; no significant differences in dropout rates, reports of skin reactions, or hypoglycemic events weight, and BMI, waist circumference, FBG, PBG, lipid profiles, and total calorie intake	3-Day RT-CGM once a month for 3 months, control group instructed to check SMBG at least 4 times a week
Allen et al.²² (2008)	52	8	United States	Medtronic™ MiniMed	CGM 57 (14.56); control 57 (12.47)	48	0	Age ≥18 years; type 2 diabetes not receiving insulin; HbA1c levels >7.5%; not engaged in physical activity more than 2 days/week	Inability to walk 0.25 miles in 10 min; taking glucocorticoids; uncontrolled BP; neuropathy	CGM-adjusted difference −1.16 (1.04), p < 0.05; control- adjusted difference −0.32 (1.02), p < 0.05	Significant increase in physical activity and significant decrease in BMI and systolic BP in CGM group compared with control	CGM used for 3 days; therapy managed by their usual providers
Haak et al.¹⁷ (2017)	224	24	France, Germany, United Kingdom	FreeStyle Libre™	CGM 59.0 (9.9); control 59.5 (11)	67	224	Age ≥18 years; type 2 diabetes treated with insulin for ≥6 months; HbA1c levels of 7.5% to 12%; current diabetes medication regimen ≥3 months; SMBG >10 per week for ≥2 months	A total daily insulin dose of ≥1.75 units/kg; severe hypoglycemia, diabetic ketoacidosis or hyperosmolar-hyperglycemic state past ≤6 months; use of CGM past ≤4 months; pregnant or planning pregnancy; receiving steroid therapy	CGM 8.65 (1.01) to 8.37 (0.83); control 8.75 (0.98) to 8.34 (1.14); 0 = 0.82	Significant reduction of hypoglycemia in CGM group; no difference in hyperglycemia, insulin dose, and weight; significant increase in patient satisfaction (DTSQ and DQOL); no serious adverse events related to CGM; 4% CGM group reported skin reactions	FGM used daily; control group managed by SMBG
Observational studies
Kesavadev et al.¹⁸ (2017)	592	24	India	iPro2, Medtronic™	CGM 54 (13); control 54.47 (11.66)	73	545	Age ≥21 years; type 2 diabetes treated with insulin or oral agents; had baseline P-CGM for 6–7 days and 6-month follow-up evaluation at comprehensive diabetes clinic from June 2010 to August 2014	History of bleeding disorder or on warfarin/heparin; recent history of steroid or alcohol abuse; pregnant or lactating; noncompliant with instructions from diabetes care team	CGM 7.5 (1.4) to 7 (0.9), p < 0.0001; control 7.7 (1.1) to 7.4 (1.1), p = 0.0663	Significant increase in frequency of performing SMBG from the baseline in CGM group compared with control	7-Day P-CGM data collected from patient charts; therapy managed by diabetes care team
Kim et al.¹⁹ (2014)	366	24	South Korea	Medtronic MiniMed	CGM 59 (10); control 59.1 (11)	71	98	Type 2 diabetes; CGM use between December 2008 and November 2009 for CGM group; clinic visit over 3 months starting December 2008 for insulin-treated control group or over 2 weeks for noninsulin-treated control group	Lack of HbA1c measurements before CGM and at 3 months post-CGM use; cancer or severe illness; renal insufficiency;	CGM 7.9 (1.5) to 7.3 (1.1); control 7.9 (1.6) to 7.7 (1.6); p = 0.010	Significantly more changes in the treatment modality of CGM group compared with control group	3-Day retrospective CGM data collected from patient charts; therapy managed by study team; propensity score matched
Anjana et al.²⁰ (2017)	4678	24	India	FreeStyle Libre	CGM 57.3 (12.1); control 57.1 (12.2)	61.1	3164	Type 1 or 2 diabetes; FGM-based ambulatory glucose profiling use between March 2015 and October 2016	Lack of HbA1c measurements; HbA1c <7%	CGM 9.2–8.3; control 9.2–8.5; p < 0.001	30.4% patients had no insulin dose adjustment following AGP, while 33.2% patients had insulin dose increase	14-Day FGM data collected from patient charts; therapy managed by their usual providers

AGP, ambulatory glucose profile; BG, blood glucose; BMI, body mass index; BP, blood pressure; CGM, continuous glucose monitoring; CI, confidence interval; DQOL, Diabetes Quality-of-Life Questionnaire; DTSQ, Diabetes Treatment Satisfaction Questionnaire; FBG, fasting blood glucose; FGM, flash glucose monitoring; PBG, postprandial blood glucose; P-CGM, professional CGM; RT-CGM, real-time CGM; SD, standard deviation; SMBG, self-monitoring blood glucose.

The pooled sample size of all included studies that used RT-CGM and P-CGM was 1384, of which 574 patients received the intervention and 810 patients were in the control. The mean age was 56.8 years, and 65.2% of subjects were males. Patients on insulin therapy with or without other hypoglycemic agents were 65.5%. The pooled sample size of RCTs only was 426 patients with 213 patients who received CGM and 213 patients for the control. The mean age was 58.5 years, and 49.4% of subjects were males. Patients on insulin therapy with or without other hypoglycemic agents were 62.0%.

For FGM, the pooled sample size of all included studies was 4902, of which 2488 patients received the intervention and 2414 patients were in the control. The mean age was 57.3 years, and 61.4% of subjects were males. Patients on insulin therapy with or without other hypoglycemic agents were 69.1%.

Glycemic control

Patients who received RT-CGM and P-CGM had a significantly greater reduction in HbA1c compared with the control group in all studies and in RCTs only. The standardized mean difference (95% confidence interval [CI]) was −0.20 (−0.31 to −0.09) in all studies and −0.33 (−0.52 to −0.13) in RCTs (Figs. 2 and 3). There was a low heterogeneity between studies as measured by I ². The I ² value for all studies was 8% and for RCTs was 3%. The funnel plot did not show asymmetry (Fig. 4).

FIG. 2.

Analysis of all studies. The forest plot demonstrating a significantly increased reduction in HbA1c from baseline associated with the RT-CGM and P-CGM use and a nonsignificant reduction for FGM. CGM, continuous glucose monitoring; FGM, flash glucose monitoring; P-CGM, professional CGM; RT-CGM, real-time CGM.

FIG. 3.

Analysis of randomized controlled trials only. The forest plot demonstrating increased reduction in HbA1c from baseline associated with the RT-CGM and P-CGM use and a nonsignificant reduction for FGM.

FIG. 4.

The funnel plot of RT-CGM and P-CGM studies demonstrating no evidence of publication bias.

Patients who received FGM had a nonsignificant reduction in HbA1c compared with the control group. The standardized mean difference (95% CI) was −0.02% (95% CI −0.07 to 0.04) in all studies (Fig. 2). There was no statistical evidence for difference between studies as measured by I ² (0%).

Due to insufficient data and heterogeneity of outcome assessment and reporting, the meta-analysis of other criteria for assessing the quality of glycemic control was not conducted. Therefore, these outcomes, including time in range, time in hypoglycemia, time in hyperglycemia, and glycemic variability, are provided in Supplementary Table S4.

Adverse effects

No severe hypoglycemic episodes were observed in patients who received any type of CGM device.^14

–17 Patients who experienced implantation site skin reactions using RT-CGM were 0% and 5.8% in the study by Yoo et al. and Cosson et al., respectively.^15,16 Reported skin reactions required no specific treatment and resolved without complications. Four percent of patients who received FGM reported sensor-adhesive reactions that were treated with topical medications.¹⁷

Other outcomes

The meta-analysis of the secondary outcomes was not conducted due to heterogeneity of outcome assessment and reporting.

Two observational studies reported significantly more changes in the treatment modality in the P-CGM group compared with control.^18,19 The authors suggested that the CGM data allowed clinicians to make more changes in the patient's therapy, which resulted in better glycemic control. However, three studies, including one that used FGM, did not show significant changes in hypoglycemic agents in the intervention group compared with usual care.^20
–22

In two RCTs, clinicians used the CGM data to counsel patients on the benefits of physical activity in glycemic control. A significant increase in the physical activity time was observed in the intervention group compared with the control.^16,22 Yoo et al. reported a significant increase in exercise time from 188.2 to 346.6 min/week in the intervention group (p < 0.01) and a nonsignificant change in the control.¹⁶ Allen et al. reported an increase of 5.48 min/day in the intervention group (p < 0.05) and a nonsignificant change in the control.²²

Studies reported inconsistent results in the reduction of blood pressure, BMI, and weight associated with the CGM use.^16,21,22

Discussion

Summary of results

The effectiveness of the CGM system was evaluated by 10 studies, of which seven were RCTs. Each study reported a reduction in HbA1c from baseline as a primary outcome. The meta-analysis of all participants showed a modest but statistically significant reduction in HbA1c in the CGM group compared with the control group. However, the patients who received FGM did not have a statically significant reduction in HbA1c. Majority of the participants who received CGM experienced no severe hypoglycemic events and minor skin reactions related to the device.

Strengths and limitations

The current meta-analysis has important strengths, including high consistency between studies as measured by I ² and an analysis of all relevant RCTs, which provide the highest level of evidence. Multiple databases were searched to identify all studies relevant to research question. Three of the included studies were observational. However, the overall analysis of all studies reported similar results. According to Shrier et al., including observational studies in addition to RCTs in a meta-analysis can increase precision.²³

Some of the limitations were that included RCTs had small sample size and a short-term follow-up. Different CGM devices were used in the studies, but the subjects were instructed to calibrate using a fingerstick to prevent inaccurate readings. Also, the patients and clinicians were not blinded. However, this was necessary to assess the effect of disclosing CGM data to guide treatment and motivate patients. Not all studies reported tolerability and adverse effects of CGM to assess its safety.

In our analysis, we combined RT-CGM and P-CGM data. Although they are different types of CGM, RT-CGM was used for a limited period ranging from 48 h to 14 days per cycle in all studies except Beck et al.¹⁴ After intervention, the CGM data generated by P-CGM were used by clinicians to adjust therapy and motivate patients throughout the study duration of 8 to 32 weeks. When used for a short period of time, the differences between RT-CGM and P-CGM would not be likely significant. In addition, separating P-CGM from RT-CGM data in all studies did not significantly change the results in our analysis. The standardized mean difference (95% CI) was −0.39 (−0.65 to −0.12) in RT-CGM studies and −0.29 (−0.5 to −0.08) in P-CGM studies (Supplementary Figs. S1 and S2).

There are other criteria for assessing the quality of glycemic control, including time in range, time in hypoglycemia, time in hyperglycemia, and glycemic variability.²⁴ Due to insufficient data and heterogeneity of outcome assessment and reporting, our analysis was limited to the reduction in HbA1c.

Comparison with other studies

The results of this meta-analysis, which supports the use of RT-CGM and P-CGM, were consistent with previous studies that were published in the past.^7,8,25 In 2011, Gandhi et al. included only three studies and reported a significant reduction in HbA1c of 0.70% in adults with type 2 diabetes using CGM compared with SMBG.⁸ Another meta-analysis of four studies reported 0.31% reduction in HbA1c in adults with type 2 diabetes who received CGM.⁷ Both of these studies did not include FGM. More recently in 2017, Taylor et al. published a systematic review of nine RCTs and three observational studies that assessed for HbA1c reduction as well as lifestyle and behavioral changes with CGM.²⁵ The study supported the use of CGM to promote glycemic and weight control and lifestyle modifications in adults with type 2 diabetes. However, they did not perform a meta-analysis due to high heterogeneity of interventions and outcomes of the included studies. With the inclusion of recent studies on RT-CGM and P-CGM that assessed glycemic control, our analysis showed a significant reduction in HbA1c of 0.20% in all studies and 0.33% in RCTs only.

No previous meta-analysis studies have been conducted on effectiveness of FGM in improving HbA1c. Our analysis of two studies showed a nonsignificant reduction in HbA1c of 0.02%. In the past, trials have shown that FGM is effective in improving short-term glycemic measures.^17,26 Bolinder et al.²⁶ conducted an RCT in 239 adults with well-controlled type 1 diabetes in Europe. Compared with control, the intervention group that used FGM daily had a significant reduction in hypoglycemia, which was defined as glucose levels below 70 mg/dL. At 6 months, the difference in adjusted means was −1.24 h/day, which was 38% reduction in the intervention group (p < 0.0001). This result was consistent in another study that included patients with type 2 diabetes.¹⁷ According to our analysis, however, the long-term impact on HbA1c was not improved with FGM use. This may be due to lack of real-time data and alerts in FGM or due to the reduction in hypoglycemia that can affect HbA1c levels.

Clinical implications and future studies

Our analysis of current evidence suggests that RT-CGM and P-CGM are effective in improving HbA1c in adults with type 2 diabetes. Due to insufficient evidence, it is premature to form conclusions on the effectiveness of FGM in lowering HbA1c. CGM data provide glucose trend information as well as high and low values that can be used to guide treatment and motivate patients. The findings of this systematic review and meta-analysis can be used to assist clinicians in optimizing therapy for patients with type 2 diabetes.

Currently, not many high-quality studies reporting the effectiveness of CGM in reducing HbA1c are available. Also, the data on safety and cost-effectiveness of CGM are limited. Therefore, future multicenter trials accessing these outcomes may be necessary.

Footnotes

Acknowledgment

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author Disclosure Statement

No competing financial interests exist.

References

Cho

, Shaw

, Karuranga

, et al.: IDF Diabetes Atlas: global estimates of diabetes prevalence for 2017 and projections for 2045. Diabetes Res Clin Pract, 2018; 138:271–281.

Mohan

, Shah

, Saboo

: Current glycemic status and diabetes related complications among type 2 diabetes patients in India: data from the A1chieve study. J Assoc Physicians India, 2013; 61(Suppl. 1):12–15.

Bolli

: Hypoglycaemia unawareness. Diabetes Metab, 1997; 23(Suppl. 3):29–35.

American Diabetes Association: 6. Glycemic targets: Standards of Medical Care in Diabetes—2018 . Diabetes Care, 2018; 41(Suppl. 1): S55–S64.

Klonoff

, Ahn

, Drincic

: Continuous glucose monitoring: a review of the technology and clinical use. Diabetes Res Clin Pract, 2017; 133:178–192.

Heinemann

, Freckmann

: CGM versus FGM; or, continuous glucose monitoring is not flash glucose monitoring. J Diabetes Sci Technol, 2015; 9:947–950.

Poolsup

, Suksomboon

, Kyaw

, et al.: Systematic review and meta-analysis of the effectiveness of continuous glucose monitoring (CGM) on glucose control in diabetes. Diabetol Metab Syndr, 2013; 5:39.

Gandhi

, Kovalaske

, Kudva

: Efficacy of continuous glucose monitoring in improving glycemic control and reducing hypoglycemia: a systematic review and meta-analysis of randomized trials. J Diabetes Sci Technol, 2011; 5:952–965.

Liberati

, Altman

, Tetzlaff

: The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ, 2009; 339:b2700.

10.

Higgins

, Altman

, Gotzsche

, et al.: The Cochrane Collaboration's tool for assessing risk of bias in randomized trials. BMJ, 2011; 343:d5928.

11.

Cook

, Reed

: Appraising the quality of medical education research methods: the Medical Education Research Study Quality Instrument and the Newcastle-Ottawa Scale-Education. Acad Med, 2015; 90:1067–1076.

12.

Higgins

, Thompson

, Deeks

, et al.: Measuring inconsistency in meta-analysis. BMJ, 2003; 327:557–560.

13.

Review Manager (RevMan) [Computer program]: Version 5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014.

14.

Beck

, Riddlesworth

, Ruedy

, et al.: Continuous glucose monitoring versus usual care in patients with type 2 diabetes receiving multiple daily insulin injections: a randomized trial. Ann Intern Med, 2017; 167:365–374.

15.

Cosson

, Hamo-Tchatchouang

, Dufaitre-Patouraux

, et al.: Multicentre, randomised, controlled study of the impact of continuous sub-cutaneous glucose monitoring (GlucoDay) on glycaemic control in type 1 and type 2 diabetes patients. Diabetes Metab, 2009; 35:312–318.

16.

Yoo

, An

, Park

, et al.: Use of a real time continuous glucose monitoring system as a motivational device for poorly controlled type 2 diabetes. Diabetes Res Clin Pract, 2008; 82:73–79.

17.

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.

18.

Kesavadev

, Vigersky

, Shin

, et al.: Assessing the therapeutic utility of professional continuous glucose monitoring in type 2 diabetes across various therapies: a retrospective evaluation. Adv Ther, 2017; 34:1918–1927.

19.

Kim

, Kim

, et al.: Effectiveness of 3-day continuous glucose monitoring for improving glucose control in type 2 diabetic patients in clinical practice. Diabetes Metab J, 2014; 38:449–455.

20.

Anjana

, Kesavadev

, Neeta

, et al.: A multicenter real-life study on the effect of flash glucose monitoring on glycemic control in patients with type 1 and type 2 diabetes. Diabetes Technol Ther, 2017; 19:533–540.

21.

Ehrhardt

, Chellappa

, Walker

, et al.: The effect of real-time continuous glucose monitoring on glycemic control in patients with type 2 diabetes mellitus. J Diabetes Sci Technol, 2011; 5:668–675.

22.

Allen

, Fain

, Braun

, et al.: Continuous glucose monitoring counseling improves physical activity behaviors of individuals with type 2 diabetes: a randomized clinical trial. Diabetes Res Clin Pract, 2008; 80:371–379.

23.

Shrier

, Boivin

, Steele

: Should meta-analyses of interventions include observational studies in addition to randomized controlled trials? A critical examination of underlying principles. Am J Epidemiol, 2007; 166:1203–1209.

24.

Rodbard

: Evaluating quality of glycemic control: graphical displays of hypo- and hyperglycemia, time in target range, and mean glucose. J Diabetes Sci Technol, 2015; 9:56–62.

25.

Taylor

, Thompson

, Brinkworth

: Effectiveness and acceptability of continuous glucose monitoring for type 2 diabetes management: a narrative review. J Diabetes Investig, 2018; 9:713–725.

26.

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.

27.

Sato

, Kanazawa

, Ikeda

, et al.: Effect of treatment guidance using a retrospective continuous glucose monitoring system on glycaemic control in outpatients with type 2 diabetes mellitus: a randomized controlled trial. J Int Med Res, 2016; 44:109–121.

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

0.24 MB