Blood glucose monitoring is fundamental in the management of diabetes and is essential to optimise glycaemic control. Achieving optimal glucose control is important in reducing the risk of significant long-term microvascular (nephropathy, retinopathy) and macrovascular (cardiovascular disease) complications, as well as neuropathy.
Blood glucose monitoring is fundamental in the management of diabetes and is essential to optimise glycaemic control. Achieving optimal glucose control is important in reducing the risk of significant long-term microvascular (nephropathy, retinopathy) and macrovascular (cardiovascular disease) complications, as well as neuropathy. Intensive insulin therapy and frequent blood glucose determinations are recommended to achieve glucose objectives in Type 1 diabetes patients.1 Self-monitoring of blood glucose (SMBG) is performed by obtaining a capillary blood sample by means of a lancing device and then measuring the blood glucose employing a glucose meter. The obtained result represents the blood glucose at the moment when the blood was drawn. This method provides an accurate determination the glucose levels; however, significant oscillations in blood glucose may be ignored, hindering the achievement of an optimal glycaemic control.2 Furthermore, SMBG entails a significant number of daily punctures that many patients find uncomfortable and painful.
Continuous glucose monitoring (CGM) systems measure interstitial fluid glucose levels providing continuous information reflecting blood glucose levels. This continuous monitoring may recognise glucose oscillations which may otherwise remain unidentified with SMBG alone. Currently, the use of CGM is not common practice.3 CGM is considered to be particularly useful for children (to reduce the often very high number of finger punctures in this group), for patients with poorly controlled diabetes, for pregnant women in whom tight glucose control is essential with respect to the outcome of pregnancyBlood glucose monitoring is fundamental in the management of diabetes and is essential to optimise glycaemic control. Achieving optimal glucose control is important in reducing the risk of significant long-term microvascular (nephropathy, retinopathy) and macrovascular (cardiovascular disease) complications, as well as neuropathy. Intensive insulin therapy and frequent blood glucose determinations are recommended to achieve glucose objectives in Type 1 diabetes patients.1 Self-monitoring of blood glucose (SMBG) is performed by obtaining a capillary blood sample by means of a lancing device and then measuring the blood glucose employing a glucose meter. The obtained result represents the blood glucose at the moment when the blood was drawn. This method provides an accurate determination the glucose levels; however, significant oscillations in blood glucose may be ignored, hindering the achievement of an optimal glycaemic control.2 Furthermore, SMBG entails a significant number of daily punctures that many patients find uncomfortable and painful.
Continuous glucose monitoring (CGM) systems measure interstitial fluid glucose levels providing continuous information reflecting blood glucose levels. This continuous monitoring may recognise glucose oscillations which may otherwise remain unidentified with SMBG alone. Currently, the use of CGM is not common practice.3 CGM is considered to be particularly useful for children (to reduce the often very high number of finger punctures in this group), for patients with poorly controlled diabetes, for pregnant women in whom tight glucose control is essential with respect to the outcome of pregnancy and for patients with hypoglycaemia unawareness (to prevent dangerous episodes of hypoglycaemia).4,5
A recent metanalysis by Langendam et al.4 shows that there is limited evidence for the effectiveness of real-time CGM (RT-CGM) on glycaemic control. However the reduction in HbA1c levels seems to be related to actual CGM use. After 12 months, those patients who used their CGM frequently had a significantly lower HbA1c level compared who patients who showed low or no sensor usage.4 Furthermore, a recent consensus statement from the European Society for Pediatric Endocrinology, the Pediatric Endocrine Society and the International Society for Pediatric and Adolescent Diabetes declared that the use of RT-CGM may be appropriate for motivated children and youth of all ages provided that appropriate support personnel are available.6
CGM therefore provides detailed information on glucose oscillations and trends. This allows patients to manage their diabetes more successfully. Several CGM systems are commercially available. Two types of CGM systems can be identified according to the way information is delivered:
- Retrospective systems that measure the glucose concentration during a certain time span: the information is stored in a monitor and can be downloaded in a second moment.
- RT systems that continuously provide the actual interstitial glucose concentration and trend on a display.
CGM devices can be further classified into three categories: invasive, minimally invasive, and non-invasive. Sensor placement/invasiveness depends on the its transduction mechanism.7,8
Current (Invasive) Continuous Glucose Monitoring Systems
There are four RT-CGM devices approved by the US Food and Drug Administration (FDA) and clinically used: DexCom® SEVEN® PLUS (San Diego, California), US, Medtronic MiniMed Paradigm® and Guardian® REAL-Time (Minneapolis, Minnesota, US) and Abbott Diabetes Care FreeStyle Navigator (Maidenhead, Berkshire, UK). Each system consists of a glucose oxidase-based electrochemical sensor, which is placed subcutaneously. Interstitial glucose measurements are then sent continuously from the sensor to a receiver through wireless technology. The DexCom SEVEN PLUS CGM is a wireless device with a sensor, approved for 7-day wear in 2010, inserted into the subcutaneous tissue of the abdomen. A transmitter connects to the sensor that sends the information to the receiver to display glucose measurements every five minutes. There is a 2-hour start-up period during which no glucose values are presented and requires calibrations by SMBG every 12 hours. This system also has alarms that can be set at specified levels to alert the patient of a hypoglycaemic or hyperglycaemic glucose level. The model also displays the glucose rate of change and trends of blood glucose over 1, 3, 6, 12 and 24 hours. One feature that distinguishes it from other models is the ability to enter times of meals, insulin administration and physical activity, providing a more complete picture of the potential causes of glucose excursions. However, acetaminophen interferes with the glucose measurements and should not be consumed during sensor wear. This device can be used for patients using (continuous subcutaneous insulin infusion) CSII in combination with the Animas® Vibe™ pump, which represents one of the two available CGM-enabled devices along with the Medtronic Paradigm pump. A new and more accurate G4 Platinum version was recently approved by the European Medicines Agency (EMEA).
The Medtronic MiniMed Enlite CGM consists of an external monitor and a subcutaneous sensor that must be calibrated by SMBG every 12 hours, approved for 6-day wear. This system also requires a 2-hour warm-up time before blood glucose values are displayed, after which readings are displayed every five minutes. This device can be used for patients using CSII and requires only one device that works as both the insulin pump and the CGM. Like the Animas Vibe, although the monitor is the same for both systems, it requires two separate insertion sites separated by at least a few centimetres: one for the CGM and one for the pump. The device presents glucose trend graphs over 3 and 24 hours as well as hypoglycaemic and hyperglycaemic alerts with trend arrows. These devices also have the predictive alarm feature that can alert the patient when the rate of change and current glucose value will lead to a hypoglycaemic or hyperglycaemic event within a specified time period. Meals and insulin administrations can be entered. The Medtronic Minimed VEO sensor-augmented pump is currently the only device available that features the ‘low glucose suspend’ (LGS) function, which suspends the insulin infusion for two hours when glucose values are below a pre-established glucose threshold. The goal of a threshold suspend device system is to help reduce the severity or reverse a dangerous drop in blood glucose level (hypoglycaemia) by temporarily suspending insulin delivery when the glucose level falls to or approaches a low-glucose threshold.
The Abbott Diabetes Care FreeStyle Navigator CGM was approved by the FDA in March 2008 and started distribution in 2011. This device also contained a subcutaneous sensor with an external monitor requiring calibrations at 10, 12, 24 and 72 hours. Although there are fewer total calibrations, the first blood glucose measurement is not displayed until after the first calibration is entered at 10 hours. Afterwards, it displays an updated glucose measurement every minute. A more recent version with a 1-hour warm-up period has been approved by the FDA, but is not currently available in the US. The Navigator alerts current and predictive lows and highs and displays 2, 4, 6, 12 and 24-hour trend graphs. Meals and insulin administrations can also be entered. The sensors are approved for 5-day continuous wear. A second-generation device is currently under study.
A different approach of invasive sensors includes microdialysis technology. In this context, a catheter housing a dialysis membrane inserted within the subcutaneous tissue to continuously pass glucosefree isotonic fluid across the skin. Upon passage through the skin, the isotonic fluid collects glucose that is assayd externally using optical or electrochemical techniques. Along these lines, catheter-shaped sensors have also been introduced, wherein the sensing element is located at the tip of the catheter, while the transmitter is located at its other end, which resides outside the skin. Similarly, a disposable, invasive optical fiber has also been introduced that is capable of percutaneous glucose monitoring via spectroscopic measurements.9
RT information by CGM devices can be used by patients to adjust their insulin doses and can be downloaded by physicians to provide an overall picture of glucose control and discuss it interactively with their patients. The adjustable hypo- and hyperglycaemia alerts may be valuable when patients usually do not check their SMBG (i.e. during sleep or while driving) helping to prevent potentially dangerous glucose excursions. CGM allows patients to actively evaluate the effects of lifestyle decisions on glucose excursions, allowing them to apply modifications to certain behaviors to ensure adequate glucose control in future similar circumstances, and progressively reduce HbA1c.10 These devices have been shown to help minimise time spent in hypoglycaemic and hyperglycaemic ranges and reduce glucose excursions.6
Presently, CGM use is approved only as a complementary tool alongside SMBG, requiring patients to confirm the CGM information with a fingerstick determination before making any therapeutic decisions.11 Furthermore, the sensors need to be calibrated when glucose values are most stable in order to display the most accurate glucose measurements. Approximately 5–20% of patients suffer from erythema, edema or skin irritation due to the sensor adhesive.10,12,13 Another disadvantage of CGM is the time lag between the blood glucose value and the interstitial glucose value. The time lag is related to errors in SMBG and is especially accentuated with a higher rate of change of glucose greater than 2 mg/dL/minute (0.1 mmol/l/ minute). Additionally, patients may be tend to overcorrect hyperglycaemia by repeated insulin boluses or overcorrect hypoglycaemia by multiple carbohydrate doses.14 This may ultimately lead to an increased risk of hypo- and hyperglycaemia. Consequently, patients should be adequately trained on the use of CGM devices to avoid misinterpretation of continuous data.
While CGM requires proper training and time from physicians to interpret and analyse the data, it has been shown to be advantageous to patients by increasing their time spent in euglycaemic range. This can improve the patient’s glycaemic control and can ultimately help reduce HbA1c.15–17
Despite the benefit of multiple information provided by RT-CGM (ie. glucose trends and trend alerts, alerts for hypo-/hyperglycaemia, predictive alerts) it has not been embraced. Several reasons such as complexity, inaccuracy, inappropriate expectations, invasiveness, cost, pain, discomfort, risk of infection and interference with daily activities remain significant drawbacks.18 These matters have hampered motivation to begin CGM and frequency of CGM use. Since clinical studies have shown a linear relationship between increased use of CGM and lowered HbA1c, lack of adoption and infrequent use are of significant concern and have spurred the necessity to develop less invasive technology.
Minimally Invasive Continuous Glucose Monitoring
Minimally invasive technology has been investigated in the past. The GlucoWatch was a near-continuous RT-CGM device shaped like a watch with hypo- and hyperglycaemia alerts. The glucose level was measured and displayed every 10 minutes for up to 13 hours.19 This technology seemed very appealing as there was no transdermal pricking. However, drawbacks were related to time lag between values, a cumbersome calibration procedure and poor accuracy in hypoglycaemic range. This device has therefore been removed from the market because of the development of more satisfying diabetes-management devices.20
Minimally invasive technology attempts to measure glucose concentration avoiding the continuous presence of a foreign object in the body. This is performed by measuring glucose from fluids (interstitial fluid or blood) obtained from the skin tissue. In this case both the sensor and controller are located outside the body and are connected to a fluiddrawing deice that is externally located.
Methods of such monitoring include iontophoresis, sonophoresis, micropore technology, microneedle technology and skin blister technique. Iontophoresis employs a low electrical current applied across the skin. This current induces a minute amount of interstitial fluid to be withdrawn and sampled by externally locate sensor. Sonophoresis is performed by using low-frequency ultrasound to increase skin permeability which in turn allows interstitial fluid extraction. The skin blister technique employs a minute local epidermal vacuum that creates an interstial fluid-filled blister from which glucose can be measured. Micropore technology creates multiple micropores in the stratum corneum by laser ablation through which interstitial fluid may be collected by application of a small vacuum. Microneedle technology is based on the use of disposable devices composed of a silicon microneedle and pouch that collects a minute blood sample.17
Non-invasive Continuous Glucose Monitoring
There is a need to minimise discomfort and the potential risk of infection from fluid-withdrawing probes penetrating the skin along with avoiding the foreign body response that otherwise can compromise accuracy. The lack of invasiveness would allow greater tolerability and wearability, therfore increasing sensor adoption rate and longterm adherence. This would ultimately favour reduction in HbA1c and glucose variability and consequently a significant decrease in diabetes acute and chronic complications.
Non-invasive devices include: transdermal sensors that pass near-infrared (NI) light across the stratum corneum to detect glucose concentrations under optical approaches, and external assays of body fluids (i.e. saliva, tears, breath) using various optical and electrochemical detection methods.
However, current non-invasive technologies have significant drawbacks mainly related to inaccuracies due to variable skin properties: pigmentation, body water content, hydration, non-specificity to glucose, temperature, poor correlation between blood glucose and glucose in body fluids (see Table 1).7 Multisensor systems are currently being studied as they may be able to achieve a broader biophysical characterisation of the multiple physical and chemical properties of the analysed tissue – mainly the skin – and improve accuracy under variable conditions.21
The development of non-invasive continuous glucose monitoring (NI-CGM) devices faces crucial challenges represented by the improvement of signal-to-noise ratio and sensitivity, development of wearable devices, development of procedures for precise blood glucose determination and reducing the time taken for glucose measurements. The signal-to-noise ratio and the sensibility of NI-CGM devices can be improved by employing next-generation transducers and methods that can perform parallel monitoring of multiple parameters. Retrieved sensor data can be further improved using digital filters and data treatment methods.7
Beyond Hardware
CGM implementation is still suboptimal because of several matters. The first concern is related to the uncertainty of CGM data because glucose readings suffer from interfence by noise that confounds their interpretation. Noise may result in false oscillations that could trigger artificial hypo- or hyperglycaemic alerts. Some denoising algorithms have been developed to resolve this potentially dangerous phenomenon.22–24
Another issue concerns accuracy. CGM data present delays that are mainly secondary to the blood-to-interstitium glucose transport and sensor processing time. Furthermore, systematic under- or over- estimations due to calibration problems may add to this inaccuracy.25 Several strategies have been proposed to compensate the inaccuracy and to enhance CGM data calibration.26–30 Furthermore, CGM sensors report glucose value with a lag time with respect to blood glucose.
Therefore, there is the necessity of generating predictive hypo- ad hyperglycaemic alerts by applying short-term glucose prediction algorithms. This would allow the patient to take action before an approaching glucose excursion. Newer insulin pump predictive shutoff algorithms are shown to be able to prevent hypoglycaemic events, especially during sleep.27
Finally, blood glucose meters are calibrated based on a laboratory reference, but current CGM devices are calibrated against a blood glucose meter, putting them one step behind a laboratory value. To overcome these problems, so-called ‘smart CGMs’ are being developed reduce uncertainty and inaccuracy of sensor collected data by applying RT algorithms.23
Conclusion
The frequent monitoring of glucose is a fundamental aspect of diabetes management as it is the only way by which blood glucose may be kept within the euglycaemic range. Blood glucose meters have already reached an advanced stage in terms of accuracy, cost-effectiveness, convenience and software-based data analysis and management. In fact, many companies have now started focusing in improving interface, mobile device compatibility and telemedicine. However, SMBG is still too invasive, time-consuming and cumbersome to be universally undertaken with sufficient frequency and at the same time be compatible with the daily activities of children and adults alike.31–33 Poor adherence to glucose monitoring determines an elevated risk of diabetes complications. Continuous and non-invasive technology is therefore warranted. The future of CGM relies not only on advances in hardware technology (lifetime, non-invasiveness, wearability, user interface, lag time, elimination of interference, accuracy, improved calibration, costeffectiveness, comfort, patient safety) but also by the way the stream of data is processed algorithmically. This will ultimately result in increased accuracy, biocompatibility and wearability, consequently leading to improved user compliance, health and quality of life. Furthermore there still are no universally accepted guidelines regarding how to apply diabetes management decisions using CGM trend information.34,35
Although technological improvements are crucial for the success of CGM, one must not neglect that any device, no matter how advanced, must be driven and assessed by a human mind. A trained care team and individualised treatment are necessary for effective blood glucose control in each patient. Adequate patient education on CGM data interpretation is fundamental to guarantee a successful outcome. and for patients with hypoglycaemia unawareness (to prevent dangerous episodes of hypoglycaemia).4,5
A recent metanalysis by Langendam et al.4 shows that there is limited evidence for the effectiveness of real-time CGM (RT-CGM) on glycaemic control. However the reduction in HbA1c levels seems to be related to actual CGM use. After 12 months, those patients who used their CGM frequently had a significantly lower HbA1c level compared who patients who showed low or no sensor usage.4 Furthermore, a recent consensus statement from the European Society for Pediatric Endocrinology, the Pediatric Endocrine Society and the International Society for Pediatric and Adolescent Diabetes declared that the use of RT-CGM may be appropriate for motivated children and youth of all ages provided that appropriate support personnel are available.6
CGM therefore provides detailed information on glucose oscillations and trends. This allows patients to manage their diabetes more successfully. Several CGM systems are commercially available. Two types of CGM systems can be identified according to the way information is delivered:
- Retrospective systems that measure the glucose concentration during a certain time span: the information is stored in a monitor and can be downloaded in a second moment.
- RT systems that continuously provide the actual interstitial glucose concentration and trend on a display.
CGM devices can be further classified into three categories: invasive, minimally invasive, and non-invasive. Sensor placement/invasiveness depends on the its transduction mechanism.7,8
Current (Invasive) Continuous Glucose Monitoring Systems
There are four RT-CGM devices approved by the US Food and Drug Administration (FDA) and clinically used: DexCom® SEVEN® PLUS (San Diego, California), US, Medtronic MiniMed Paradigm® and Guardian® REAL-Time (Minneapolis, Minnesota, US) and Abbott Diabetes Care FreeStyle Navigator (Maidenhead, Berkshire, UK). Each system consists of a glucose oxidase-based electrochemical sensor, which is placed subcutaneously. Interstitial glucose measurements are then sent continuously from the sensor to a receiver through wireless technology. The DexCom SEVEN PLUS CGM is a wireless device with a sensor, approved for 7-day wear in 2010, inserted into the subcutaneous tissue of the abdomen. A transmitter connects to the sensor that sends the information to the receiver to display glucose measurements every five minutes. There is a 2-hour start-up period during which no glucose values are presented and requires calibrations by SMBG every 12 hours. This system also has alarms that can be set at specified levels to alert the patient of a hypoglycaemic or hyperglycaemic glucose level. The model also displays the glucose rate of change and trends of blood glucose over 1, 3, 6, 12 and 24 hours. One feature that distinguishes it from other models is the ability to enter times of meals, insulin administration and physical activity, providing a more complete picture of the potential causes of glucose excursions. However, acetaminophen interferes with the glucose measurements and should not be consumed during sensor wear. This device can be used for patients using (continuous subcutaneous insulin infusion) CSII in combination with the Animas® Vibe™ pump, which represents one of the two available CGM-enabled devices along with the Medtronic Paradigm pump. A new and more accurate G4 Platinum version was recently approved by the European Medicines Agency (EMEA).
The Medtronic MiniMed Enlite CGM consists of an external monitor and a subcutaneous sensor that must be calibrated by SMBG every 12 hours, approved for 6-day wear. This system also requires a 2-hour warm-up time before blood glucose values are displayed, after which readings are displayed every five minutes. This device can be used for patients using CSII and requires only one device that works as both the insulin pump and the CGM. Like the Animas Vibe, although the monitor is the same for both systems, it requires two separate insertion sites separated by at least a few centimetres: one for the CGM and one for the pump. The device presents glucose trend graphs over 3 and 24 hours as well as hypoglycaemic and hyperglycaemic alerts with trend arrows. These devices also have the predictive alarm feature that can alert the patient when the rate of change and current glucose value will lead to a hypoglycaemic or hyperglycaemic event within a specified time period. Meals and insulin administrations can be entered. The Medtronic Minimed VEO sensor-augmented pump is currently the only device available that features the ‘low glucose suspend’ (LGS) function, which suspends the insulin infusion for two hours when glucose values are below a pre-established glucose threshold. The goal of a threshold suspend device system is to help reduce the severity or reverse a dangerous drop in blood glucose level (hypoglycaemia) by temporarily suspending insulin delivery when the glucose level falls to or approaches a low-glucose threshold.
The Abbott Diabetes Care FreeStyle Navigator CGM was approved by the FDA in March 2008 and started distribution in 2011. This device also contained a subcutaneous sensor with an external monitor requiring calibrations at 10, 12, 24 and 72 hours. Although there are fewer total calibrations, the first blood glucose measurement is not displayed until after the first calibration is entered at 10 hours. Afterwards, it displays an updated glucose measurement every minute. A more recent version with a 1-hour warm-up period has been approved by the FDA, but is not currently available in the US. The Navigator alerts current and predictive lows and highs and displays 2, 4, 6, 12 and 24-hour trend graphs. Meals and insulin administrations can also be entered. The sensors are approved for 5-day continuous wear. A second-generation device is currently under study.
A different approach of invasive sensors includes microdialysis technology. In this context, a catheter housing a dialysis membrane inserted within the subcutaneous tissue to continuously pass glucosefree isotonic fluid across the skin. Upon passage through the skin, the isotonic fluid collects glucose that is assayd externally using optical or electrochemical techniques. Along these lines, catheter-shaped sensors have also been introduced, wherein the sensing element is located at the tip of the catheter, while the transmitter is located at its other end, which resides outside the skin. Similarly, a disposable, invasive optical fiber has also been introduced that is capable of percutaneous glucose monitoring via spectroscopic measurements.9
RT information by CGM devices can be used by patients to adjust their insulin doses and can be downloaded by physicians to provide an overall picture of glucose control and discuss it interactively with their patients. The adjustable hypo- and hyperglycaemia alerts may be valuable when patients usually do not check their SMBG (i.e. during sleep or while driving) helping to prevent potentially dangerous glucose excursions. CGM allows patients to actively evaluate the effects of lifestyle decisions on glucose excursions, allowing them to apply modifications to certain behaviors to ensure adequate glucose control in future similar circumstances, and progressively reduce HbA1c.10 These devices have been shown to help minimise time spent in hypoglycaemic and hyperglycaemic ranges and reduce glucose excursions.6
Presently, CGM use is approved only as a complementary tool alongside SMBG, requiring patients to confirm the CGM information with a fingerstick determination before making any therapeutic decisions.11 Furthermore, the sensors need to be calibrated when glucose values are most stable in order to display the most accurate glucose measurements. Approximately 5–20% of patients suffer from erythema, edema or skin irritation due to the sensor adhesive.10,12,13 Another disadvantage of CGM is the time lag between the blood glucose value and the interstitial glucose value. The time lag is related to errors in SMBG and is especially accentuated with a higher rate of change of glucose greater than 2 mg/dL/minute (0.1 mmol/l/ minute). Additionally, patients may be tend to overcorrect hyperglycaemia by repeated insulin boluses or overcorrect hypoglycaemia by multiple carbohydrate doses.14 This may ultimately lead to an increased risk of hypo- and hyperglycaemia. Consequently, patients should be adequately trained on the use of CGM devices to avoid misinterpretation of continuous data.
While CGM requires proper training and time from physicians to interpret and analyse the data, it has been shown to be advantageous to patients by increasing their time spent in euglycaemic range. This can improve the patient’s glycaemic control and can ultimately help reduce HbA1c.15–17
Despite the benefit of multiple information provided by RT-CGM (ie. glucose trends and trend alerts, alerts for hypo-/hyperglycaemia, predictive alerts) it has not been embraced. Several reasons such as complexity, inaccuracy, inappropriate expectations, invasiveness, cost, pain, discomfort, risk of infection and interference with daily activities remain significant drawbacks.18 These matters have hampered motivation to begin CGM and frequency of CGM use. Since clinical studies have shown a linear relationship between increased use of CGM and lowered HbA1c, lack of adoption and infrequent use are of significant concern and have spurred the necessity to develop less invasive technology.
Minimally Invasive Continuous Glucose Monitoring
Minimally invasive technology has been investigated in the past. The GlucoWatch was a near-continuous RT-CGM device shaped like a watch with hypo- and hyperglycaemia alerts. The glucose level was measured and displayed every 10 minutes for up to 13 hours.19 This technology seemed very appealing as there was no transdermal pricking. However, drawbacks were related to time lag between values, a cumbersome calibration procedure and poor accuracy in hypoglycaemic range. This device has therefore been removed from the market because of the development of more satisfying diabetes-management devices.20
Minimally invasive technology attempts to measure glucose concentration avoiding the continuous presence of a foreign object in the body. This is performed by measuring glucose from fluids (interstitial fluid or blood) obtained from the skin tissue. In this case both the sensor and controller are located outside the body and are connected to a fluiddrawing deice that is externally located.
Methods of such monitoring include iontophoresis, sonophoresis, micropore technology, microneedle technology and skin blister technique. Iontophoresis employs a low electrical current applied across the skin. This current induces a minute amount of interstitial fluid to be withdrawn and sampled by externally locate sensor. Sonophoresis is performed by using low-frequency ultrasound to increase skin permeability which in turn allows interstitial fluid extraction. The skin blister technique employs a minute local epidermal vacuum that creates an interstial fluid-filled blister from which glucose can be measured. Micropore technology creates multiple micropores in the stratum corneum by laser ablation through which interstitial fluid may be collected by application of a small vacuum. Microneedle technology is based on the use of disposable devices composed of a silicon microneedle and pouch that collects a minute blood sample.17
Non-invasive Continuous Glucose Monitoring
There is a need to minimise discomfort and the potential risk of infection from fluid-withdrawing probes penetrating the skin along with avoiding the foreign body response that otherwise can compromise accuracy. The lack of invasiveness would allow greater tolerability and wearability, therfore increasing sensor adoption rate and longterm adherence. This would ultimately favour reduction in HbA1c and glucose variability and consequently a significant decrease in diabetes acute and chronic complications.
Non-invasive devices include: transdermal sensors that pass near-infrared (NI) light across the stratum corneum to detect glucose concentrations under optical approaches, and external assays of body fluids (i.e. saliva, tears, breath) using various optical and electrochemical detection methods.
However, current non-invasive technologies have significant drawbacks mainly related to inaccuracies due to variable skin properties: pigmentation, body water content, hydration, non-specificity to glucose, temperature, poor correlation between blood glucose and glucose in body fluids (see Table 1).7 Multisensor systems are currently being studied as they may be able to achieve a broader biophysical characterisation of the multiple physical and chemical properties of the analysed tissue – mainly the skin – and improve accuracy under variable conditions.21
The development of non-invasive continuous glucose monitoring (NI-CGM) devices faces crucial challenges represented by the improvement of signal-to-noise ratio and sensitivity, development of wearable devices, development of procedures for precise blood glucose determination and reducing the time taken for glucose measurements. The signal-to-noise ratio and the sensibility of NI-CGM devices can be improved by employing next-generation transducers and methods that can perform parallel monitoring of multiple parameters. Retrieved sensor data can be further improved using digital filters and data treatment methods.7
Beyond Hardware
CGM implementation is still suboptimal because of several matters. The first concern is related to the uncertainty of CGM data because glucose readings suffer from interfence by noise that confounds their interpretation. Noise may result in false oscillations that could trigger artificial hypo- or hyperglycaemic alerts. Some denoising algorithms have been developed to resolve this potentially dangerous phenomenon.22–24
Another issue concerns accuracy. CGM data present delays that are mainly secondary to the blood-to-interstitium glucose transport and sensor processing time. Furthermore, systematic under- or over- estimations due to calibration problems may add to this inaccuracy.25 Several strategies have been proposed to compensate the inaccuracy and to enhance CGM data calibration.26–30 Furthermore, CGM sensors report glucose value with a lag time with respect to blood glucose.
Therefore, there is the necessity of generating predictive hypo- ad hyperglycaemic alerts by applying short-term glucose prediction algorithms. This would allow the patient to take action before an approaching glucose excursion. Newer insulin pump predictive shutoff algorithms are shown to be able to prevent hypoglycaemic events, especially during sleep.27
Finally, blood glucose meters are calibrated based on a laboratory reference, but current CGM devices are calibrated against a blood glucose meter, putting them one step behind a laboratory value. To overcome these problems, so-called ‘smart CGMs’ are being developed reduce uncertainty and inaccuracy of sensor collected data by applying RT algorithms.23
Conclusion
The frequent monitoring of glucose is a fundamental aspect of diabetes management as it is the only way by which blood glucose may be kept within the euglycaemic range. Blood glucose meters have already reached an advanced stage in terms of accuracy, cost-effectiveness, convenience and software-based data analysis and management. In fact, many companies have now started focusing in improving interface, mobile device compatibility and telemedicine. However, SMBG is still too invasive, time-consuming and cumbersome to be universally undertaken with sufficient frequency and at the same time be compatible with the daily activities of children and adults alike.31–33 Poor adherence to glucose monitoring determines an elevated risk of diabetes complications. Continuous and non-invasive technology is therefore warranted. The future of CGM relies not only on advances in hardware technology (lifetime, non-invasiveness, wearability, user interface, lag time, elimination of interference, accuracy, improved calibration, costeffectiveness, comfort, patient safety) but also by the way the stream of data is processed algorithmically. This will ultimately result in increased accuracy, biocompatibility and wearability, consequently leading to improved user compliance, health and quality of life. Furthermore there still are no universally accepted guidelines regarding how to apply diabetes management decisions using CGM trend information.34,35
Although technological improvements are crucial for the success of CGM, one must not neglect that any device, no matter how advanced, must be driven and assessed by a human mind. A trained care team and individualised treatment are necessary for effective blood glucose control in each patient. Adequate patient education on CGM data interpretation is fundamental to guarantee a successful outcome.
