Common errors of glucose estimation in a laboratory
A blood sample is composed mainly of two components, cellular (containing red and white blood cells) and plasma (containing water in addition to many constituents like proteins and lipids). Glucose is more concentrated in the plasma part of the blood, so plasma blood glucose is 10-12 % higher compared to whole blood. Handling of blood sample and timing of analysis has a great impact on the accuracy of the test. If the sample analysis is delayed, it will falsely report a lower blood glucose level and may give a false assurance to the patient. At room temperature, glucose is consumed by red blood cells (by a process of glycolysis) at the rate of 5-7% per hour (decrease of around 36 mg/liter per hour). Some chemicals like fluoride can inhibit glycolysis process (hence theoretically prevent red blood cells from consuming glucose), but it takes at least one hour for the fluoride action to start and fall in blood glucose can continue for up to four hours. To avoid such a fall in glucose, it is recommended to separate the blood from cells immediately (within 30 minutes) and refrigerate before analysis.
Home glucose monitoring and Glucometers
As the prevalence of diabetes is increasing, its management is equally improving. Discovery of newer and safer drugs (including insulin) and use of modern technology for monitoring of blood glucose have helped a long way in prevention of complications and prolonging longevity in these patients. The development of self-monitoring blood glucose (SMBG) is the most important advancement in controlling diabetes since the discovery of insulin in 1920. Concept of glucometers started in 1960s when paper strip was used which would develop blue color upon contact with glucose (in either urine or blood); this color was then compared with a color concentration chart. Subsequent improvements technology has resulted in the development of glucometers that use a minimal blood sample to test for glucose and the results are available instantly with reasonable accuracy. Originally developed as a tool for self-monitoring of blood glucose, glucometers are now used in medical camps, hospitals, clinics, emergency rooms, ambulances, helicopters, cruise and ships and so on.
How do glucometers work?
Essentially glucometers have two parts: reaction part (glucostrips) and detection part (glucometer proper). Glucostrips (test strips) contain glucose oxidase, an enzyme that reacts to (patient’s) glucose in the blood droplet. This reaction generates an electric current that is detected by the glucometer. The glucometer is calibrated so the number appearing in its digital readout corresponds to the strength of the electrical current, which is proportional to concentration of glucose in blood sample. Meters with glucose oxidase strips require oxygen and water for their reaction so are susceptible to extremes of water and oxygenation. As discussed above, glucose is concentrated more in water content of blood (plasma) and less in red blood cells. Therefore, blood glucose concentration will be low if sample contains more red blood cells, high protein and lipid content. Modern day glucometers use auto correction so that the interference of red blood cells, lipids and proteins is taken care of.
Glucometer is primarily aimed at measuring glucose in capillary blood (usually from the fingertips). It measures blood glucose immediately, so the issue of loss of glucose through red blood cells does not arise. In normal situations, there are some differences between venous and capillary blood. In the fasting state, capillary glucose is about five mgs/dl higher than venous blood; after meals, capillary glucose may be 25% greater than venous blood. Glucometers measure blood in whole blood (capillary blood) and internally convert it to plasma values.
How accurate a glucometer should be?
The journey in search of most accurate glucometer has not ended. Clinical and laboratory standards institute proposed that a difference between glucometer and laboratory measurement of glucose should not exceed >20% at abnormal blood glucose levels (blood glucose of >100 mgs/dl) and the difference should not be more than 15 mgs/dl at blood glucose level of <100 mgs/dl. Overall, in an average glucometer a variation of ± 20% is expected and is a normal phenomenon. American Diabetes Association has desired a standard of variation of <5% which is not achieved as of now.
Common factors affecting the accuracy
Operator factors
The ability of the patient to use their meter properly can have a great influence on the accuracy of blood glucose meter. The most common reasons for discrepancies could be: applying too much mechanical pressure on the strip, dirty meters, failure to clean the meter and site of testing. In addition, sample issues like clots, bubbles and failure to apply an adequate amount of blood to the strip can result in significant disparity in the results. Too much pressure applied to strips can break the connection between the receiving part and the testing part. Some patients generally do not wash their hands before testing; this contamination can be a major source of inaccuracy. Dirty meters and unclean hands can greatly influence the results. Even a small amount of contamination can raise the blood sugar reading by 300mgs/dl.
Coding errors
Some older meters need to be calibrated before the use of new lot of strips. Mis-calibrated meters can have a variation of between 30 to 60 %. With new Code Free or No Code meters, such errors are not seen.
Environmental factors
Temperature and humidity
Changes in the temperature outside the recommended range have an adverse effect on accuracy of blood glucose measurement. Most of the modern glucometers have temperature sensors and report error if testing is done at low or high temperatures. Despite this, ambient temperature and temperature changes have a profound effect on blood glucose (variation can go to the extent of 30 to 40 mgs/dl). Usually high temperature overestimates and low temperature underestimates actual blood glucose level. Exposure or storage of glucose strips to extremes of temperature will affect the accuracy of testing. It is recommended to store the test strips and glucometer at room temperature avoiding storage at minus temperature or hot temperature. Avoid keeping the meter and strips in cars for prolonged time. While testing for blood glucose, bring the meter, strips to room temperature for 15 to 20 minutes, and then do the testing. Similarly, excessive humidity also overestimates the measured blood glucose (up to 30 mgs/dl). So it is advised to seal the container of the strip pack tightly after taking out a strip, keep the strips away from direct sunlight, not to store in areas of high humidity like kitchen and washrooms. After pulling from the container, the test strips should be used within three minutes.
Altitude
Most of the glucometers available have been tested up to the altitude of 3000 meters above sea level. However, at higher altitudes there is decrease in oxygen pressure, temperature and humidity, which will cause a wide variation.
Mobile phones
Mobile phones emit electromagnetic radiation. An error of ±10-13 mgs/dl either up or down can occur if mobile phones are nearby. It is recommended to keep mobile phone at least half a meter away from glucose meters.
Other factors
Hemoglobin, oxygen, uric acid and triglyceride levels in blood can alter blood glucose readings. A low hemoglobin results in high readings and high hemoglobin may result in lower readings. Some modern glucometers auto correct the effect of hematocrit and are not affected by level of hemoglobin. Low blood glucose reading may be seen in presence of high bilirubin level. Oxygen level in blood also affect glucometer reading with high readings at low oxygen levels (pO2 <45 mm Hg) and lower readings at high oxygen levels (pO2 >150 mm Hg). High uric acid in blood (usually > 20 mgs/dl) can cause falsely high blood glucose readings. The combination of anemia and high uric acid is seen in many patients with malignancy or on chemotherapy could result in falsely normal readings and patient may actually be in hypoglycemia. Some drugs like paracetamol, vitamin C, aspirin, heparin, warfarin and some antibiotics can have variable effect on blood glucose measurements. Paracetamol, a commonly used drug causes inaccurately high blood glucose readings. People on peritoneal dialysis may falsely show high blood glucose readings.
Precautions while doing a blood test
Wash your hands with soap and water and dry. Prick in the finger and put the first drop on the strip. If hand washing is not possible, discard the first and take the second drop. After the drop is out, put it quickly (in <20 seconds) on the test strip. Use the test strip within 3 minutes after taking out from the bottle.
Summary Points
1. Appropriate training and good performance of blood glucose testing have impact on reliability of blood glucose measurement
2. Adequate handling and storage of blood glucose testing system (including meter and test strips) are requirements for a precise glucose testing
3. Meter and strips should be kept away from heat, cold and excess humidity, bring the meter and strips to room temperature for 20 minutes before testing. Do not keep the strips in car, kitchen or in washroom.
4. Some meters must be coded to each container of test strips. Make sure the code number in the device matches the code number on the test strip container. Code free meters (without need for coding) are preferred over coding meters.
5. Proper hand washing and applying a proper amount of blood are essential requirements for a precise blood glucose measurements.
6. Some conditions like dehydration, anemia or polycythemia, level of oxygen in body, levels of uric acid and triglycerides affect blood glucose readings
7. Drugs like paracetamol and ascorbic acid also affect the glucose levels to a minor extent
8. Usually glucometer readings are more reliable in the normal range (deviation of around 15 mgs/dl with blood glucose of < 100 mgs/dl) and results can deviate around 15-20% in the abnormal range (blood glucose of > 100 mgs/dl).
9. Final comments
10. Glucometer is an important tool for home glucose monitoring. With the knowledge about proper procedure, most of the glucometers are accurate and sufficiently reliable. A margin of 15 to 20 % error is accepted in a glucometer. The variation is expected more at abnormal glucose levels; however, as the glucose level reaches towards normal, variations are insignificant. A proper knowledge and wisdom on the part of physician goes a long way in interpreting the accuracy of a test result.
Dr. Bashir Ahmad Laway is Professor & Head, Department of Endocrinology, Sher-I-Kashmir Institute Of Medical Sciences (SKIMS), Soura, Srinagar, Jammu & Kashmir
drlaway@gmail.com