Continuous glucose monitoring: current clinical use


  • Hun-Sung Kim,

    1. Department of Endocrinology and Metabolism, Seoul St. Mary’s Hospital, The Catholic University of Korea, Seoul, Korea
    Search for more papers by this author
  • Jeong-Ah Shin,

    1. Department of Endocrinology and Metabolism, Seoul St. Mary’s Hospital, The Catholic University of Korea, Seoul, Korea
    Search for more papers by this author
  • Jin-Sun Chang,

    1. Department of Endocrinology and Metabolism, Seoul St. Mary’s Hospital, The Catholic University of Korea, Seoul, Korea
    Search for more papers by this author
  • Jae-Hyoung Cho,

    1. Department of Endocrinology and Metabolism, Seoul St. Mary’s Hospital, The Catholic University of Korea, Seoul, Korea
    Search for more papers by this author
  • Ho-Young Son,

    1. Department of Endocrinology and Metabolism, Seoul St. Mary’s Hospital, The Catholic University of Korea, Seoul, Korea
    Search for more papers by this author
  • Kun-Ho Yoon

    Corresponding author
      *Correspondence to:
      Kun-Ho Yoon
      Department of Internal Medicine,
      The Catholic University of Korea,
      Seoul St. Mary’s Hospital,
      #505, Banpo-Dong, Seocho-Ku,
      137-040 Seoul, Korea.
    Search for more papers by this author

*Correspondence to:
Kun-Ho Yoon
Department of Internal Medicine,
The Catholic University of Korea,
Seoul St. Mary’s Hospital,
#505, Banpo-Dong, Seocho-Ku,
137-040 Seoul, Korea.


Four kinds of subcutaneous continuous glucose monitoring (CGM) machines have been currently introduced in clinical practice. These machines exhibit real-time glucose on the monitor every 5 minutes and have alarms to indicate hypoglycaemia and hyperglycaemia. However, thus far, there is no clear consensus about the clinical indications for CGM in actual clinical practice. CGM should be an ideal and powerful tool for monitoring glucose variability. Glycaemic variability has become a major concern over the years with growing evidence on its detrimental impact with respect to the risk of diabetic complications. Although the HbA1c level is ubiquitously measures in clinical practice, this level does not adequately represent glycaemic variability. Currently available evidence indicates that CGM aids in lowering the HbA1c level without increasing the incidence of severe hypoglycaemic episodes in patients with type 1 diabetes. Thus far, CGM has not been indicated for preventing severe hypoglycaemia or for treating type 2 diabetes because sufficient supporting evidence has not been obtained. Promising results have been obtained for the use of CGM for pregnant women with diabetes and for patients with hospital hyperglycaemia. Predictions regarding the feasibility of the closed-loop system have proven to be optimistic. CGM-integrated communication systems using information technology such as smart phone help controlling blood glucose more easily and effectively.

Unmet needs of current diabetes management

There is a consensus about the importance of blood glucose management in patients with diabetes. The goal of blood glucose management is generally determined by using the HbA1c levels as an index. The American Diabetes Association (ADA) presented with a mean HbA1c level of <7%, and this level was achieved under strict blood glucose management, keeping the results of the Diabetes Control and Complications Trial (DCCT) and the United Kingdom Prospective Diabetes Study (UKPDS), as the target blood glucose levels [1,2]. A consensus statement from the ADA and the European Association for the Study of Diabetes (EASD) [3,4], first enacted in 2006, included the characteristics of the treatment method, synergy effect and economical aspects. The statement aimed at achieving HbA1c levels of <7%. The UKPDS, which was based on patients with type 2 diabetes, proved that strict blood glucose management that resulted in a 1% decrease in HbA1c levels reduced the occurrence of microvascular complications by 37% [5]. Thus, strict blood glucose management is important for decreasing the occurrence and retarding the progress of diabetes-associated complications.

As mentioned above, HbA1c has been used as an index for the glucose status of patients with diabetes. However, differences in the occurrence of diabetes-associated complications have been observed in patients with the same level of HbA1c [6]. Thus, the HbA1c level alone does not adequately represent the true glycaemic status of the patient. The drastic rise in the number of patients with diabetes [7–10] calls for the urgent development of a new and effective blood glucose controlling system, which differs from the outpatient treatment system and from the patients’ self-monitoring of blood glucose (SMBG). The target of blood glucose management should be individualized to each patient on the basis of the long-term advantages of lowering HbA1c levels along with the side effects of drugs, patients’ compliance, costs and effects of medications besides changes in the blood glucose level. Continuous glucose monitoring (CGM) is a technique that meets the requirements and overcomes the limitations mentioned above. Therefore, the aim of this study was to determine efficacy and range of application of continuous glucose monitoring system (CGMS). The second objective was to justify a current condition of CGMS and its future direction and prospects. We reviewed all the full-text articles about CGMS and categorized it according to significance, results of clinical studies and discussion of the current CGM system. We examine the components of CGM, assess the results of clinical studies performed by using CGM and discuss the current CGM system.

Significance of continuous glucose monitoring

In cases when insulin injections fail to control the glucose level and when the patients have just started diabetes management, it is imperative that the patients completely understand glucose variability, and thus, CGM is recommended.

Continuous glucose monitoring device can measure the glucose level continuously for 3 consecutive days in patients with type 1 or type 2 diabetes. Patients can objectively identify maximum and minimum blood glucose levels and glucose variability, which are otherwise difficult to measure. CGM devices can detect hypoglycaemia or hyperglycaemia, which cannot be detected by patients or physicians, and can thus help improve the control of blood glucose in patients with diabetes. Thus, we can not only know current treatment and effectiveness/problems of the revised treatment but also predict the glucose patterns of the patients’, such as dawn phenomenon or somogy-phenomenon, and consider these predictions while prescribing further treatment.

Since the Food and Drug Administration (FDA) approved CGM devices manufactured by MiniMed in 1999, overall glucose variability of patients with diabetes can be detected, and it made strict blood glucose management possible [11]. A CGM device is largely composed of three parts: an electrochemical sensor that is placed in the subcutaneous fat and measures glucose levels in the interstitial fluid, a receiver/monitor that shows real-time blood glucose levels, and a transmitter that is connected to the sensor and performs wireless transfer of data (blood glucose levels) from the sensor to the monitor every 5 min. The sensor is inserted in the patients’ subcutaneous fat and measures the blood glucose level once every 10 s. The average of those values is automatically saved in the device once every 5 min for 72 h. The sensor is detached from the patients after 3 days, and the results are obtained. These results are slightly different from those of a portable blood glucose measuring device, which measures glucose levels in the plasma and uses capillary blood collection along with identifying the increase or decrease in blood glucose concentration.

Currently, four real-time CGM devices are available:

  • 1Freestyle Navigator (Abbott Diabetes Care, Alameda, CA)
  • 2Guardian Real-Time (Medtronic MiniMed, Northridge, CA)
  • 3Dexcom SEVEN (Dexcom, San Diego, CA)
  • 4GlucoDay (Menarini Diagnostics)

With the advancement of technology, a system that includes an insulin pump and CGM has been developed, making intensive insulin therapy possible. This system consists of a CGM device that measures the concentration of blood glucose and a drug delivery system that injects insulin. The CGM device is attached to the body or is injected inside the body to measure the blood glucose levels. The information obtained is transferred to the insulin pump and controls the dose of insulin.

Who will benefit from CGM?

Many studies compared the portable blood glucose measuring device and the CGM device. In a study by the Juvenile Diabetes Research Foundation, a total of 322 adults and infants with type 1 diabetes were assigned to either the CGM group or the control group, which performed home glucose monitoring with a blood glucose metre. The results obtained showed that when CGM was used with intensive insulin therapy, it was effective in lowering the HbA1c level in patients (>25 years of age) with type 1 diabetes [12]. On the basis of the results of that study, the ADA diabetes treatment guidelines 2009 suggested that CGM is an effective method to decrease HbA1c levels and is useful in treating type 1 diabetic patients (>25 years of age) using intensive insulin therapy [13]. Presently available evidences show that CGM lowers HbA1c levels without increasing the incidence of severe hypoglycaemia episodes in patients with type 1 diabetes [14–16].

Currently developed treatments for type 1 diabetes include the use of various types of insulin pumps and insulin pumps with a glucose sensor. The expertise of medical practitioners, patients’ careful decision-making, close supervision and detailed education about the pump are prerequisites for using an insulin pump: thus, detailed counselling is highly recommended before choosing this treatment method. We expect that this kind of insulin pump and CGM along with the development of medical devices will help diabetic patients manage their glucose levels. In a multi-institutional and randomized study [16], the authors compared the yearlong effects of sensor-augmented pump therapy to that of multiple injection therapy in 485 patients (329 adults and 156 infants) with type 1 diabetes. Sensor-augmented pump therapy significantly improved HbA1c levels, whereas insulin injection therapy did not in type 1 diabetic patients (both adults and infants) whose blood glucose level was well controlled. The ratio of adults to infants who achieved the target HbA1c level was significantly higher in the sensor-augmented pump treatment group than in the injection therapy group. There is a recent report on continuous subcutaneous insulin infusion (CSII) in inpatient setting [17] and its effect. Several studies have proven that CGM with CSII is effective [14–19]. Therefore, once CGM with CSII is applied to patients who need to receive treatment in a hospital, it is expected to be efficient in blood glucose control. A summary of previous CGM with CSII trials in patients with type 1 diabetes is presented in Table 1.

Table 1.   CGM with CSII trials in type 1 diabetes
StudiesAge (years)Study durationIntervention n HbA1c reduction from baselineSevere Hypoglycaemia
  1. CGM, continuous glucose monitoring; CSII, continuous subcutaneous insulin infusion; GDM, gestational diabetes mellitus; MDI, multiple daily insulin injections; SMBG, self-monitoring of blood glucose.

Hirsch et al. [18]12–726 monthsCGM and CSII72−0.71% = 0.373 events = 0.04
SMBG and CSII66−0.56%11 events
Raccah et al. [14]2–656 monthsCGM and CSII55−0.81% = 0.091 episode due to improperly calibrated
SMBG and CSII60−0.57%No episodes
O’Connell MA et al., [15]13–403 monthsCGM and CSII31−0.2% < 0.05No episodes
SMBG and CSII31+0.3%No episodes
Bergenstal et al. [16]7–7012 monthsCGM and CSII244−0.8% < 0.00113.3 per 100 person-years = 0.58
SMBG and MDI241−0.2%13.5 per 100 person-years
Hermanides et al. [19]18–6526 weeksCGM and CSII41−1.23% < 0.0014 episodes = 0.21
SMBG and MDI36−0.13%1 episode

Continuous glucose monitoring is also considered useful for those who are unaware of hypoglycaemia or experience hypoglycaemia frequently [20] and for those who are pregnant [21].

There is little evidence-based report on the effects of CGM in type 2 diabetes mellitus, but it is considered useful for patients with frequent hypoglycaemia, patients unaware of hypoglycaemia and patients who experience nocturnal hypoglycaemia/dawn phenomenon. Patients who have good compliance but cannot control glucose levels were considered wearing the CGM device. When CGM was applied to gestational diabetic patients, 31% of the participants started to receive insulin treatment and it is stated that CGM is potentially beneficial in finding the necessity of insulin treatment in the early stages of diabetes [22,23]. CGM during pregnancy was presented in Table 2 [23–25]. There is a report on the application of CGM in patients of the intensive care unit (ICU) in 2004 [26], but no randomized control clinical study has been performed; thus, additional data are needed. Maintaining the blood glucose level within the normal range without hypoglycaemia after major diseases like myocardial infarction and septicaemia and after the surgery of large vessels is considered important in the treatment process. Thus, if real-time CGM, which is equipped with an alarm, is well utilized, it may facilitate strict blood glucose management without hypoglycaemia in ICU patients [27].

Table 2.   Continuous glucose monitoring during pregnancy
StudiesSample sizeResults
  1. CGM, continuous glucose monitoring; CGMS, continuous glucose monitoring system; CSII, continuous subcutaneous insulin infusion; GDM, gestational diabetes mellitus; SMBG, self-monitoring of blood glucose; T1DM, type 1 diabetes; T2DM, type 2 diabetes.

Kestila et al. [23] Detection of needing antihyperglycemic drug therapy
73 GDM CGMS (n = 36)31% = 0.0149
 SMBG (n = 37)8%
McLachlan et al. [24]37 GDM
10 T2DM
8 T1DM
62% detected hyperglycaemia
providing additional information
altered clinical management decisions.
Yogev et al. [25] Mean total time of hyperglycaemia
6 T1DM
 Fingerstick152 ± 33 min/day < 0.03
 CGM89 ± 17 min/day

Limitations and future prospects of CGM

Cost and coverage policies

Continuous glucose monitoring is a useful device for glucose monitoring and diabetes treatment. However, the glucose sensor is expensive and does not last for more than 7 days; hence, it is difficult to generalize the examination. Medical care expenses need to be re-examined via future frequency survey. Annual cost or initial cost of CGMS is high, but CGMS can help physicians and patients not only maintaining intensive glucose control but also changing their communication by graphical analysis of glucose data. Well-controlled blood glucose levels reduce the costs which are associated with the treatment of diabetic complications [28]. For that reason, it is reasonable to assume that CGMS will provide a cost-effective benefit for patients. The previous study, CGMS is expected increase in cost of $23 552 and showed to be cost-effective compared with SMBG in type 1 diabetes. Based on this, the incremental cost-effectiveness ratio of approximately $45 033/QALY was reported. Therefore, cost-effectiveness of CGMS has some validity [29]. The problem is an expensive application of CGMS. It is considered that cost-sharing insurance policy should be accompanied [30].

Technological improvements

The following technological improvements are needed to globally commercialize CGM.


Many studies reported that there is no significant difference between glucose concentration in the blood vessels and in the interstitial fluid if the blood glucose concentration is stable. Even when the blood glucose concentration varies drastically, it has no clinical insignificant [31]. One study stated that the accuracy and sensitivity of CGM decrease with time [32]. Therefore, new CGM devices that can sustain accuracy are needed.

Physiological lag time and lack of a practical algorithm

Due to the physiological lag time, the glucose values on the CGM screen appear 5–15 min later than the actual SMBG values [33]. Therefore, the latest SMBG levels have to recheck before changing the insulin dose or the hypoglycaemia/hyperglycaemia alarms or before making any therapeutic decision. Considering this fact, the inclusion of a clinical algorithm in the CGM device is necessary.

Implantable devices

If a new algorithm that does not need this process is developed, it is expected that the ultimate goal of CGM is the completion of a closed-loop system (artificial pancreas), which is connected to the insulin pump. FDA approved and confined the current closed-loop system to automatically stop the insulin pump if there is no response to the hypoglycaemia warning alarm.

Integration with other diabetes management tools

Interactive communication systems using information technology, such as the smart phone or personal digital assistant devices, were shown to be effective in diabetes management (Figure 1).

Figure 1.

 Components of a CGM device and CGM-integrated system, CGM-integrated communication systems using information technology (IT) such as smart phone or personal digital assistant (PDA) devices that help controlling blood glucose levels more easily and effectively. CGM, continuous glucose monitoring

Trained care team for CGM

In order to successfully attach a CGM device, it is necessary to educate the patients about the practices involve. Patients should know how to operate the CGM device by themselves, and effective dietary therapy, exercise and drug treatment should be performed concurrently. For educating patients, professionally trained-medical personnel, such as physicians, nurse practitioners, or physician’s assistants, are required. A trained care team and individualized treatment are necessary for effective blood glucose control in each patient.

Need more clinical evidences

The studies on the effectiveness of real-time CGM in patients with type 1 diabetes have been performed by using multiple insulin injection therapy. There are many studies on the effectiveness of CGM in patients with type 1 diabetes; however, very few such studies have focused on patients with type 2 diabetes. Therefore, clinical studies for the effectiveness of CGM in patients with type 2 diabetes should be carried out.

Need of CGM coordinating centre

As mentioned above, it is necessary to have trained-medical team for CGM to be well used and analysed. However, there is short number of professionals who could analyse CGM data in local clinics. Therefore, there needs to be CGM coordinating centre which is connected with local clinics.

It is expected to control blood glucose level more effectively if CGM combines information technology (IT). In 2010, Cho et al. [34], interactive communication systems using ITs such as the internet or telecommunications devices were shown to be effective in diabetes management. Glucose data, recoded in CGM, are uploaded to CGM centre in real time via smart phone. One CGM-trained nurse in CGM coordinating centre communicates with patients in local clinics (Figure 2). She will give feedbacks or direct calls to patients if the patients were problem such as poorly controlled glucose levels. She notifies to doctors in the centre when patients has serious condition such as severe hypoglycaemia. The doctors in the centre analyse CGM data and explain to doctors in clinics.

Figure 2.

 The role of the CGM coordinating centre. CGM, continuous glucose monitoring; CSII, Continuous subcutaneous insulin infusion

Continuous glucose monitoring centre also needs automatic glucose analysis programme, which will process auto-filtering uploaded data. If glucose level is well-controlled, auto-recommendation is provided through smart phone. However, when poorly controlled, difficult to determine the result or there are risk of hypoglycaemia, the result is reported to the nurse. If there is a high possibility of severe hypoglycaemia, the result is passed directly to the doctor, and he or she will call the patient immediately (Figure 1). This algorithm-based GGM-integrated system will allow patients to control glucose level effectively as well as to reduce physicians’ labour time and improve the cost-effectiveness for diabetes management [35]. There are many researches to come ahead, which will help diabetic patients with various glucose management programme based on IT.


Continuous glucose monitoring has the potential to improve the lifestyle of patients with type 2 diabetes. It not only helps patients who do not recognize major hypoglycaemia but also helps diabetic patients in various situations, such those with unstable diabetes, wherein the glucose level cannot be well controlled, and those with gestational diabetes, those who plan to start diabetes management programme, and those who plan to start insulin pump therapy.

The CGM device with a sensor is a new and improved user-friendly technological device. The data obtained and saved in the device can be transferred onto a computer. The information is analysed and drawn in curve according to date. Therefore, anyone can easily see the variability of blood glucose levels and the patients are self-motivated to control the glucose levels. The physicians can also detect continuous blood glucose variability for 72 h and prescribe accurate medicine to provide the required treatment.

Thus far, it cannot be said that glucose variability should be measured in every patient with diabetes. However, if patients experience frequent hypoglycaemia, show high levels of HbA1c that cannot be explained by an ordinary SMBG and cannot achieve the target HbA1c level, CGM should be considered. The algorithm-based CGM-integrated communication systems using information technology, automatic glucose analysis programme and other diabetes management tools with trained care team also help controlling blood glucose more easily and effectively in patients with uncontrolled blood sugar.

Conflicts of interest

The authors have no conflicts of interest.