Originally Published IVD Technology January 2002
COMMENTARY
Glucose sampling
from alternative anatomical sites
Carolyn D. Jones
Members
of the industry trade association AdvaMed (Washington, DC) include the major
manufacturers of blood glucose monitoring systems. Over time, these manufacturers
have developed a variety of new technologies that encourage good diabetes self-management.
Alternative anatomical site glucose sampling is a new tool that has been shown
to encourage patients to test more frequently.
The AdvaMed blood glucose monitoring working group strongly supports the development and use of new technologies that enable individuals with diabetes to test their blood sugar from anatomical sites other than their fingertips. It is well known that fingertip testing can be painful, and that many patients may not test as frequently as recommended because of the pain. A number of patients also prefer alternative sites to avoid potential injuries to the fingertips.
New technologies and protocols have been developed that use samples obtained from alternative anatomical sites such as the forearm, the palm of the hand, or the thigh. This approach relieves the pain associated with fingertip testing, thus encouraging more-frequent testing.
There is clinical
evidence to demonstrate that samples collected from alternative anatomical sites
provide blood glucose readings that can be accurate and compare well with fingertip-sample
results. Recent independent research and manufacturers' clinical studies
support this conclusion. However,
some of these studies also indicate that during times of rapid changes in glucose
concentrations, the results of samples taken from certain sites may occasionally
differ from the blood-glucose concentration measured at the fingertip. There
is also evidence indicating that this difference is not apparent in samples
taken from all alternative sites or seen with all test systems (devices and
test procedures).
Studies conducted by a number of manufacturers, and an interesting study by Koschinsky and Jungheim, have highlighted the need for manufacturers to study the conditions in which results obtained from anatomical sites other than the fingertip may differ significantly from results from sampling at the fingertip.1 This is relevant to situations in which the individual's glucose may be changing rapidly. For information on alternative-site testing to be of real benefit to patients, it is important that evaluations simulate real-life conditions as opposed to the extreme conditions used in the Koschinsky and Jungheim study. It is also important that all instructions in the manufacturer's labeling be followed.
Points to Consider
AdvaMed supports the development of a set of evaluation criteria that can be used for better assessment of device response in diabetic patients under typical use conditions. The assessment would include evaluation of any differences in results from samples taken from the fingertip versus other anatomical sites in non-steady-state patients. We recommend the following points to consider in evaluating blood glucose monitoring systems labeled for alternative-site testing.
Accuracy. When evaluating the significance of differences, if any, in the glucose results obtained using samples taken from the fingertip and other anatomical sites, the manufacturer should apply the same accuracy criteria as are used to assess differences between glucose-meter and laboratory results obtained using only fingertip capillary blood samples. Regardless of the anatomic site used for meter testing, the reference sample should be obtained at the fingertip, as this is the value commonly used for therapeutic decisions.
Conditions of
Use. Because studies have suggested the possibility of physiological differences
in blood glucose at various anatomical sites, studies done to confirm the acceptability
of an alternative site should gather data under the proposed conditions of use.
For instance, if a meter is labeled for alternative-site use at a time when
blood glucose may be changing rapidly (such as after a meal), testing should
be done at a time when blood glucose changes are likely to be occurring. If
testing is not done at such times, the meter labeling should indicate that the
alternative site should not be used in those cases.
In diabetic subjects showing the typical range of glycemic control, blood samples
drawn before and following a normal meal can be expected to span a range of
rates of glucose change from –2 to +2 mg/dl/min, with some values outside
this range. This physiological stimulus provides an adequate range of glucose
change to evaluate glucose-monitoring system performance in a non-steady-state
situation. Sponsors may propose other types of studies that look at the rate
of change.
Performance claims should be supported by a statistically sound study. The size
of the study (number of data points) should be determined by the nature of the
performance claims and the precision of a particular glucose-monitoring system.
Specific Labeling. Data specific to each proposed alternative site (arm,
palm, thigh, etc.) should be obtained, and the labeling claims for use limited
to the specific sites that have been evaluated.
Labeling Requirements
To demonstrate
system performance, each manufacturer would be responsible for submitting data
obtained in a manner consistent with the above points to consider. Labeling
claims must be supported by data. Where the system performance data demonstrate
a clinically significant lag, the manufacturer should establish a specific set
of precautionary statements to be included in the product labeling of the affected
device. It is the responsibility of each manufacturer to provide FDA with valid
scientific evidence to support its product labeling.
AdvaMed supports
the continued use of currently marketed test systems, as well as FDA clearance
of new systems, provided that manufacturers inform users about the possible
limitations of alternative-site testing, or that the manufacturer has demonstrated
or can demonstrate that its test system is not affected. As is the case with
any questions related to diabetes self-management, issues around alternative-site
sampling should be explained in clear and easily readable labeling that has
been assessed for clarity using readability evaluation procedures as suggested
in FDA guidelines for home-use device submissions.
Where a lag is
demonstrated, labeling statements and the proper education and training can
best be communicated to the healthcare community and consumers by the manufacturers.
Manufacturers would be responsible for disseminating the information to customers
through efficient communication channels. They should also provide supplemental
materials to healthcare professionals that explain, for example, specific study
results, the scientific basis for the results, and recommendations regarding
patients that may not be suitable candidates for use of some alternative-site
testing (e.g., patients with hypoglycemic unawareness).
AdvaMed does not believe that a requirement for prescription use is warranted for these products. We are convinced that prescription home use for alternative-site testing of blood glucose is not practical and, above all, will cause undue hardship for patients and healthcare providers. Because devices cleared for alternative-site testing are also cleared for finger testing, a prescription home-use requirement for a specific segment of the intended users would be impossible to administer, and would deprive the less-frequent user of alternative-site testing of important product-use information.
For test systems and body sites that demonstrate a clinically significant lag,
the following information, written in simple language so that all users can
understand its meaning, should be included in the labeling:
- [Site] results may differ from fingertip results when glucose levels are changing rapidly (e.g., after eating a meal, taking insulin, or during or after exercise).
- Do not use blood
samples from the [site], use your fingertips, if:
a. You think your blood sugar is low (hypoglycemia).
b. You have been diagnosed with hypoglycemic unawareness.
c. The [site] results do not agree with the way you feel.
Conclusion
AdvaMed strongly believes that the use of alternative-site testing technology
provides important patient benefits and is safe and effective when used with
the manufacturer's instructions and precautions.
REFERENCE
1. T Koschinsky and K Jungheim, "Risk Detection Delay of Fast Glucose Changes by Glucose Monitoring at the Arm," Diabetes Care 24, no. 7 (2001): 1303–1304.
Carolyn D. Jones, JD, is associate vice president for technology and regulatory affairs at AdvaMed (Washington, DC) and a member of the IVD Technology editorial advisory board. She can be reached via cjones@advamed.org. This article is based on testimony presented to FDA's Clinical Chemistry and Clinical Toxicology Devices Panel at its meeting in October 2001.
Copyright ©2002 IVD Technology
