Sensor life time

When considering a lifetime, it is necessary to distinguish between the lifetime presented by the manufacturer and approved by various drug administration authorities and the physical lifetime of the device itself.

According to the User guide of the Medtronic’s continuous glucose monitoring system, the sensor should be removed after 72 hours of operation and replaced by the new one. The algorithms built in Paradigm 522 and Paradigm 722 pumps stop measuring after:

• 3 days (72 hours) since its first calibration with alarm Sensor end (internal code 107);
• 2 consequent calibrations when the calibration factor is higher than 20 mg/dl/nA or lower than 1 mg/dl/nA with alarm Calibration error (internal code 106) followed by alarm Bad sensor (internal code 108);
• 40 minutes of lost communication between the pump and sensor with alarm Weak signal (internal code 112) followed by alarm Lost sensor (internal code 113). This could be caused by distance between the pump and transmitter, environment blocking Bluetooth communication (like water or even hand or palm placed over the transmitter or interference from other electronic devices) or empty transmitter battery;
• Transmitter is disconnected from the sensor.

Physically, the sensor life time is determined by the time when its calibration factor is stable and the sensor gives reliable interstitial glucose values. There are two typical cases of sensor expiration that should initiate its replacement:

• Depletion of sensor active substance. This is indicated by steep growth of calibration factor as shown on Fig. Lifetime 1.

  Fig. Lifetime 1.    Evolution of the calibration factor
  

  The calibration factor on Fig. Lifetime 1. was stable (after initial period discussed in section "Behavior of the newly inserted sensor") for 7.75 days approximately. Its values varied between 6.929 and 9.042 mg/dl/nA. Then its values started to grow rapidly changing the value from 0.4669 to 1.1667 mmol/l/nA within 1.75 days.
  According to my experience, sensor calibration factor above 10.8 – 12.6 mg/dl/nA indicates that I should change the sensor to keep reliable interstitial fluid glucose readings. Firstly, because value of the ADD constant in formula {Profiles 1} becomes more important in comparison with value of ISIG needed to change SG by 1 mg/dl. This lowers precision of SG calculation. Secondly, the rapid increase of calibration factor between subsequent calibrations means that interstitial fluid glucose values shown by the pump are lower than the actual ones. As the calibration factor is continuously increasing, recalibrations to track the change lag behind, leaving all sensor readings lower than they should be, especially as the time since the previous calibration increases. This difference is furthermore amplified by calibration factor averaging described in section "Calibration process".

• Problems with interstitial fluid circulation or body reaction at the sensor insertion site. This is indicated by fluctuation of ISIG values. ISIG fluctuation reaches the level that cannot be suppressed by data smoothening algorithm described in section "Sensor instrumental delay". It results in fluctuation of interstitial glucose values and shivery graphs displayed by the pump. 24 hour graph is affected more than 3 hour graph. Typical example of ISIG fluctuation is shown on Fig. Lifetime 2.

  Fig. Lifetime 2.    Fluctuation of ISIG and interstitial fluid glucose values
  

  Two gaps on the red line of interstitial glucose (SG) values are caused by delayed calibration.
  Calibrations made around the time of ISIG pits cause incorrect calibration factors (due to the data smoothening algorithm) resulting in incorrect SG values until the next calibration.
  Some people say that reconnection of the transmitter together with warm bath or massage around the sensor insertion site may suppress ISIG fluctuation. I have tried it for couple of times but without any effect. So if the ISIG fluctuation appears, I replace the sensor.

Usually, I keep sensors inserted for 6 days. The best and most reliable cycle of changing sensors I have found in my case is:

• inserting the new sensor just before going to bed and still using the old one;
• removing the old sensor, charging the transmitter and connecting the new sensor just after I wake up. Then I take a shower;
• starting new sensor without disconnecting from transmitter and without transmitter charging after expiration of the first 3 days sensor cycle;
• inserting the new sensor in the evening before expiration of the second 3 day sensor cycle;
• removing the old sensor, charging the transmitter and connecting the new sensor and so on…