This article is more than 5 years old. Information may no longer be current.
Continuous glucose monitoring: Past, present and future
A recent device approval marks another landmark in the progress of this field and its application to the treatment of type 1 diabetes.
Click Here to Manage Email Alerts
The recent announcement that the new Dexcom STS continuous glucose monitor has been approved by the FDA is another landmark in the progress of this field and its application to treatment of type 1 diabetes mellitus. A brief overview of the past, present and future applications of this technology is in order.
The past
Type 1 diabetes mellitus is a relentless, onerous and burdensome disease. Management requires eternal vigilance in monitoring blood glucose, delivery of a variety of available insulin types by manual or automated delivery devices and integration of food intake with or without exercise in an overall plan to achieve near-normal glycemia.
The desirability of near-normal glycemia is important for the quality of life and the avoidance of micro- and macrovascular complications as conclusively demonstrated in the Diabetes Control and Complication Trial (DCCT) and its follow-up, the Epidemiology of Diabetes Intervention and Complications (EDIC).
Short of a cure for type 1 diabetes, the best way to achieve these goals is to provide insulin linked with an auto-feedback loop to glucose concentration that mimics, albeit imperfectly, normal insulin secretion. In humans, this precision of insulin secretion is modulated minute by minute, not only by glucose concentration but also by the nutrient mixture of amino acids and fatty acids and a variety of neurohumoral signals.
Early attempts to mimic these events used continuous blood withdrawal from one vein for measurement of glucose coupled to glucose–insulin delivery from two separate bottles joined by a Y connector and fed into a second distinct vein. Highly promising algorithms were developed that took into account not only the desirable limits of high and low glucose concentration, but also the rate of change in either direction. This enabled the provision of more exogenous glucose as endogenous glucose concentration fell and more insulin as the endogenous glucose concentration rose.
Similar features were incorporated into the Biostator (Miles Laboratories), a device that was bulky and large — about the size of a small portable refrigerator — and used primarily in clinical research centers to provide important information regarding continuous IV insulin delivery.
However, it soon became apparent that the general applicability was severely limited by the need for constant blood withdrawal for continuous measurement of glucose; IV probes for glucose measurement fared no better, with a usable survival of about three days. Nevertheless, lessons learned were incorporated and improved upon into ever more sophisticated pumps, providing algorithms for subcutaneous continuous insulin delivery at basal and bolus rates. These rates were based on only five to 10 blood glucose measurements per day to guide adjustments.
Blood glucose values were measured by portable devices using small amounts of capillary blood obtained by skin puncture. An entire industry grew up around various glucose meters, abbreviated to glucometers. Their precision was considered acceptable within +/– 20% of actual laboratory reference values.
Still, finger pricking to obtain more than 10 capillary blood samples per day was unreasonable in theory and uncommon with practice. Moreover, it was not possible to know the pattern of blood glucose in the minutes or hours between the 10 or so samples per day. This was particularly important to know at two times: the two to four hours after eating, when glucose tends to be high, and after approximately eight hours of fasting during sleep.
The present
To tackle the problems of continuous glucose monitoring, clinicians have found two approaches that seemed promising. The first was iontrophoresis of tissue fluid with measurement of glucose in a device termed “the Glucowatch.” Initially promising, the device was brought to market, but a series of problems in wearer comfort, skin reactions and precision led to a significant decline in its use.
A more successful approach was the use of subcutaneous probes measuring tissue fluid glucose that could be calibrated against simultaneously obtained blood glucose standards. This was essential, given the time delay in transition of fluid from the intravascular compartment to the tissue fluid compartment — a lag of approximately 10 to 20 minutes.
The probes were left in situ for three days, measuring glucose every minute and recording the glucose values every five minutes. Thus, 288 values were recorded daily for three days, generating some 864 values that could be graphed over three days. The information so obtained could be used to adjust insulin pump rates and, with repeat iterations, reduce the fluctuations of hyperglycemia post-prandially and hypoglycemia nocturnally or post-exercise.
A major drawback was that the information obtained needed to be downloaded to graphic form after completion of the study, so that information obtained was recorded but not available in real time. However, progress has been rapid, and now two separate continuous glucose monitoring systems are approved for use, and a third is pending.
The MiniMed Paradigm REAL-Time system for continuous glucose monitoring (Medtronic) is approved for adults, teens and children. It features a sensor the size of the nickel and a transmitter the size of a quarter linked via radio frequency to communicate with the Paradigm pump so as to display glucose values in real time. This is one step beyond the MiniMed Guardian (Medtronic), also approved, which used similar sensors and transmitters but displays real-time information in a separate monitor not directly on the pump. Both have specific calibrations guidelines plus various sounds and volumes for different levels of alarms at both the high and low levels of glucose.
More recently, the DexCom STS system (DexCom) has been approved; its sensor probe is the smallest gauge (25 gauge) vs. 22 gauge for MiniMed. It requires a separate monitor similar in size to the MiniMed Guardian system. Pending approval is the Abbott Freestyle Navigator system. Its features are similar to the MiniMed Guardian and DexCom STS systems; all require calibration, all display rates of change, and directional trends are displayed on all types except for the Guardian, which is now superseded by the REAL-Time system. The MiniMed REAL-Time, Abbott Freestyle and DexCom STS systems include graphic displays at various times, along with the directional rate of change plus display of glucose every minute for the Abbot system and every five minutes for the other systems.
The rapid progress in these devices is a tribute to brilliant technological innovation as part of the revolution in microelectronics. It is also a tribute to the cooperation between for-profit corporations in industry and academia funded in part by the public through the National Institutes of Health and partly by the private corporations themselves. DIRECNET is a consortium of academic pediatric institutions funded by the NIH and cooperating with industry for testing and improving the devices in the ultimate beneficiaries: patients with type 1 diabetes.
The future
The future is not difficult to envision: linking the information as it is being obtained with computer-driven algorithms that determine the rates and timing of insulin infusion subcutaneously. Such trials are already under way, some via DIRECNET, with overnight insulin infusion driven by nearby laptop computers with wireless or infrared reception of data and transmittal of this information to pump function based on algorithms repeatedly tested and refined.
Given the remarkable progress in the past decade, the next decade has the potential to be even more exciting; a closed-loop system for treating for type 1 diabetes with maintenance of near-normal glycemia could be created. Further refinements may come from the inclusion of substances of such amylin or GLP-1 analogs along with the insulin in the delivery systems to delay gastric emptying and food absorption, benefiting both post-prandial hyperglycemia and nocturnal hypoglycemia.
Those of us who take care of children with type 1 diabetes truly can’t wait. Patients and parents of children with type 1 diabetes are even more impatient. All recognize that we are on the cusp of a new era in managing for type 1 diabetes that should result in overall better metabolic control. It will still not be entirely perfect because delivery of insulin subcutaneously is not the same as natural delivery via the portal vein to the liver at first pass, and it will lack the endogenous control systems. However, it will be a long way from once- or twice-daily insulin injections with which many of us treated patients a generation ago.
For more information:
- Mark A. Sperling, MD, is a Professor in the Department of Pediatrics at the University of Pittsburgh and the Children’s Hospital of Pittsburgh. He is also a member of Endocrine Today’s Editorial Board.