📘 Continuous glucose monitoring

The continuous glucose monitoring (CGM) sensor is a small medical device that automatically measures glucose concentration in interstitial fluid every 1-5 minutes, 24 hours a day [1]. It uses a flexible filament with glucose oxidase enzyme, inserted under the skin. Upon interaction with local glucose, an electrical signal proportional to its level is generated. Unlike the classic glucometer, which gives you an instant snapshot of glucose, the sensor shows you a continuous movie of its evolution. The current value is accompanied by trend arrows.

Information is wirelessly transmitted to a dedicated receiver or smartphone, where you can see not only the current value, but also the graph of previous values, along with the direction of evolution [1]. You can set alarms for various glucose thresholds. It's like having a personal assistant who monitors your glucose non-stop and warns you before you have problems. The technology used varies between electrochemical (majority) and fluorescent (implanted system).

Modern sensors have excellent accuracy, with a mean absolute relative difference (MARD) of 9-11% compared to laboratory glucose, which means they are correct over 85% of the time [2]. In practice, you may see differences compared to glucometer of ±20 mg/dl (1 mmol/L) at normal glucose levels and up to ±20% at extreme values [3]. Accuracy is better in the range 70-180 mg/dl (5.6-10 mmol/L) and decreases at hypoglycemia or large hyperglycemia.

Differences also occur due to the 5-15 minute lag between blood changes and those in interstitial fluid [4]. The sensor shows where glucose was some time ago, not where it is now. Other factors that influence accuracy include the first 24 hours after application (stabilization period), dehydration, sensor compression during sleep or drug interferences. That's why it's good to confirm with the glucometer from time to time or before a major therapeutic decision.

The official approved site for most sensors is the back of the arm, at least 5 cm above the elbow, where the skin is thinner and muscle contractions somewhat smaller [5]. The area should be flat, without scars or tattoos and away from joints or areas with prominent muscles. Rotation between left and right arm at each sensor change can reduce the risk of local irritation.

Alternatively, the sensor can also be mounted on the abdomen (avoiding 5 cm around the navel and the horizontal line of a pants belt), lateral side of thighs or upper portion of buttocks [5]. Athletes prefer the abdomen for better stability, and very young children more often use buttocks, where they have more subcutaneous tissue. Important is to generally choose an area you don't sleep on and where clothes don't constantly rub the sensor.

The official operating duration varies between 7 and 15 days for the current generation of sensors [2]. The notable exception is the subcutaneously implanted system that operates 365 days [6]. After the approved period, the sensor automatically stops and should not be restarted for safety and accuracy reasons.

Progressive degradation of the enzyme used and formation of a fine fibrous capsule around the filament can reduce measurement accuracy in the last days of operation [6]. Some people try to extend use with unofficial applications, but this is not medically recommended. Monthly cost involves 2-4 sensors depending on the model, which can represent a significant cost if not reimbursed by the health system.

Most current sensors have IPX8 or IP67 certification, which means water resistance for normal activities such as shower, bath or surface swimming [7]. Specifications vary: Freestyle Libre supports immersion at 1 meter for 30 minutes, while Dexcom G6/G7 can be immersed at 2.4 meters for up to 24 hours. Water temperature should not exceed 38°C and shower jet pressure should not be directed directly at the sensor. No need to remove it or protect it specially for daily hygiene.

For prolonged swimming, dives beyond manufacturer specified limit or intense water sports, use additional waterproof transparent patches. Sauna is not recommended due to high temperature, which can damage the sensor's electronic component. After water activities, dry the area by light dabbing, not rubbing, so as not to weaken the adhesive.

The fundamental difference is that the sensor measures glucose concentration in interstitial fluid, which follows blood glucose with a 5-15 minute delay, depending on the rate of change and quality of tissue perfusion (hydration) [4]. During stable periods differences are minimal (under 10%), but after meals or during exercise when glucose changes rapidly, discrepancies of 50 mg/dl (2.8 mmol/L) or greater can appear [8].

Additional factors that increase differences include the first 24 hours after application (local tissue trauma), the last 24 hours before expiration (enzymatic degradation), significant dehydration (poor tissue perfusion), sensor compression during sleep (falsely low values) [8]. Medications such as paracetamol or vitamin C in large doses can interfere with electrochemical measurement. That's why, for critical decisions (severe hypoglycemia correction, large insulin dose), confirm with the glucometer.

Standard alarms include thresholds for hypoglycemia (customizable between 60-100 mg/dl), hyperglycemia (120-400 mg/dl) [9]. Some sensors also have predictive alarms, which warn you 30 minutes before reaching the set threshold, based on current trend [10]. You can also set alarms for rapid rate of change (when glucose rises or falls very fast). They are used less often, generally with the purpose of being able to make corrections in time. Most systems allow different programming for day and night.

Technical alarms include signal loss (when receiver no longer receives data), calibration need, sensor expiring or operating errors. Volume and alert type (sound, vibration, both) are configurable. Important is to find the balance between safety and excessive alarms that lead to "alert fatigue" and ignoring them [9].

Horizontal arrows (→) indicate stability, with variation under 1 mg/dl/minute, which means you can give standard insulin doses without adjustments for trend [11]. Oblique arrows (↗/↘) show moderate change. In 30 minutes glucose will be 30-60 mg/dl (1.7-3.3 mmol/L) higher or lower, requiring dose adjustment by ±20% [11]. Simple vertical arrows (↑/↓) indicate rapid change of 2-3 mg/dl/minute.

Double vertical arrows (↑↑/↓↓) signal very rapid change, over 3 mg/dl/minute. Glucose can change by over 90 mg/dl in 30 minutes! In this situation, postpone meal bolus if dropping fast or increase dose by 50% if it's a rapid rise [11]. Learning correct interpretation of arrows is as important as the current value [12].

Application uses an automatic spring-loaded device (applicator), which inserts the sensor in less than a second. Most people describe the sensation as a light pinch or pressure, less painful than a glucose prick [13]. The flexible filament of the sensor penetrates only the subcutaneous layer where there are few nerves, not the muscle where it would hurt.

After application, discomfort disappears in 10 minutes, and in normal use you don't feel the sensor's presence [13]. Persistent pain, continuous bleeding or burning sensation indicates incorrect positioning (probably in muscle) and requires removal and reapplication in another area. Anticipatory anxiety is usually greater than actual pain during application. After the first application most are surprised how simple and painless the process is.