Glucose meters

General information

 

There are approximately 90 "Glucocard" meters in Christchurch Hospital. About another 60 reside at various locations: Christchurch Women's Hospital, Princess Margaret Hospital, Burwood Hospital and other smaller hospitals in the district under control of the CDHB.

 

Monthly aqueous quality control checks are made on meters by the POCT Coordinator. A database is maintained with records kept of location, meter serial numbers, Check Strip results, test strip lot numbers and calibration codes. In addition, a whole blood sample is analysed. Once all data is in, meters that show consistent outliers beyond 2SD's are more closely examined and replaced if necessary.

 

More sophisticated glucose meters connected through the hospital network are proposed: an electronic record of patient's results, quality control and operator identity can be maintained. A complete electronic audit trail of all records would satisfy IANZ requirements.

 

In the meantime, Glucocard meters offer reliable results on a minimum sample patient volume. Nevertheless, extremely high or low results must always be confirmed by the laboratory- send a heparin plasma sample to the lab. 

 

NOTE: Whole blood glucose meters should never be used to monitor glucose levels on unconscious, hyperosmolar, dehyrated patients with or without ketosis. 

 

Clinical staff members are advised to undergo competency testing and to keep an up to date record of their competency. There is no mandatory requirement for this at the current time.

 

Most POCT glucose samples are obtained by capillary sample. The quality of a capillary blood collect is very important- if the sample is poor, the answer will not be accurate. 

 
An electronic competency test for capillary blood collecting is also available on this site. How good is your capillary collecting skill?

 

All whole blood testing has its shortfalls. Glucose meters are no exception. For more details, please read below. 

 

Whole Blood Glucose measurement- limitations

Sources of false results when using Glucose meters

All whole blood glucoses less than 3.0 mmol/L and greater than 20 mmol/L must be confirmed by a plasma or serum sample analysed in the laboratory. The accuracy of whole blood glucose measurement is acceptable between these levels.

Less than 3 mmol/L

The reliability of any whole blood glucose meter result is dubious at levels below 3.0 mmol/L. Results below this level should be confirmed in the laboratory. This is because of analyser limitations- generally an accuracy of about ±10% or less is acceptable for normal levels of glucose on whole blood glucose meters. However at low levels, the inaccuracy can rise markedly.

 

Advances in POCT glucose meter technology continue. Modern meters generally have overcome accuracy and precision problems at low levels and can be regarded as reliable down to about 1.5 mmol/L. These modern types of meter are not yet used in the CDHB.

 

Greater than 20 mmol/L 

While results at this level are probably correct (most glucose meters are linear to about 30 mmol/L), a falsely low result may arise due to hyperglycaemic-hyperosmolar states (with or without ketosis) and/or severe dehydration, hypotension or shock.

 

This is due to a viscosity problem. Dehydration of the patient causes blood to become increasingly viscous. This results in slower absorption of blood into the test strip sampling area or chamber within a given time.  Incomplete absorption (or chamber fill, depending on the type of strip in use), can result in a falsely low result, due to insufficient sample arriving before the measuring process commences.

 

NOTE: capillary whole blood glucose meters should not be used in the above situations. Analyse your whole blood glucose sample using a blood gas analyser if available or send a STAT sample to the laboratory for analysis.

 

Oxygen tension

Most glucose test strips use a technology that relies on an enzyme called glucose oxidase (GOD): GOD reacts with glucose, releasing electrons in the process. These electrons create a very small electric current which is measured. The amount of current produced is proportional to the concentration of the glucose.  Dissolved oxygen in the sample will affect the efficiency of this reaction by competing for electrons. Hence a patient on oxygen therapy may give lower results than might be expected.

 

A newer glucose test strip technology which utilises a different enzyme, glucose dehydrogenase, is starting to be utilised by manufacturers of glucose meters. In fact there are new technologies on the horizon, including non-invasive continuous monitoring methods. 

Other Reasons Why Glucose Results May be Lower Than Expected

Reasons Why Glucose may be Higher than Expected