Inpatient Capillary Glucose Monitoring Systems Less Accurate, FDA Adcomm Says

Alicia Ault

April 02, 2018

New data suggest that blood glucose monitoring systems that rely on capillary samples may not be as accurate as arterial or venous sampling when used in patients receiving intensive medical intervention. However, the benefits of the devices still outweigh any risk for a potential adverse event related to erroneous results, a US Food and Drug Administration (FDA) advisory committee concluded on Friday.

"I think the benefits overwhelmingly outweigh the minimal risks," said panel member Hugh A. Cassiere, MD, chief of the Division of Medicine, Pulmonary, Critical Care and Sleep Medicine at North Shore University Hospital, Manhasset, New York.

The FDA asked the panel to consider whether the glucometers should be subject to perhaps less regulation than currently required when used in critical care patients. Agency officials presented data from three studies it says it came upon recently, conducted by blood glucose monitoring systems manufacturers in inpatients. The agency was given permission to share the data, but would not divulge the companies' names, said Courtney Lias, PhD, director of the FDA's division of chemistry and toxicology devices in the Office of In Vitro Diagnostics and Radiological Health.

Overall, glucometers using capillary samples produced an accurate result, defined by the FDA as within plus or minus 12 mg/dL, for glucose concentrations below 75 mg/dL in 85% to 91% of the samples. For glucose concentrations above 75 mg/dL, the device readings were accurate in from 85% to 86% of the samples.

The manufacturers' studies showed that standard testing methods were significantly more accurate, including samples obtained by venous or arterial puncture, plasma samples from a healthy population, and arterial and venous samples collected during studies conducted for the Nova StatStrip Glucose Hospital Meter System (Nova Biomedical), which is the only blood glucose monitoring system approved for use in critically ill patients.

Specifically, the comparators were 97% to 100% accurate for the below 75 mg/dL cutpoint and 95% to 97% accurate for the above 75 mg/dL cutpoint.

Panelists, even those who specialize in critical care, said the lower accuracy of the capillary samples was a surprise. They said the FDA should make an effort to inform clinicians of the newfound difference.

"There's a need for increased education, or promotion of the potential limitations of interpreting blood glucose values in the hospitalized patient," said panel chairman Andrew Bremer, MD, PhD, program director of the Division of Diabetes, Endocrinology, and Metabolic Diseases at the National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, Maryland.

Given the still-limited data, it's not possible to determine with certainty which critically ill patients should or should not be tested using the capillary samples, Bremer said.

Panelist Kathleen Wyne, MD, from the Ohio State University's Wexner Medical Center, said her institution adopted the BRAVE protocol as a means of identifying patients who should not receive testing via capillary samples. That protocol prescribes measures for blood pressure, reduced capillary refill rate, acidosis, vasopressors, and edema.

Other committee members suggested that perhaps the FDA could consider the BRAVE criteria in its next guidance on capillary sampling devices.

The agency also asked for discussion on whether using the capillary tests in the critically ill should be considered to be "waived" under the Clinical Laboratory Improvement Amendments (CLIA) regulations enforced by the Centers for Medicare & Medicaid Services. Only the Nova StatStrip system has been both approved by the FDA and given a waiver from CLIA requirements for use in the critically ill. All other glucometers in critical care units that rely on capillary samples are being used off-label and without CLIA waivers.

Thus, by law, these glucometers are considered high-complexity tests under CLIA. That triggers a host of requirements, which panelists said could make use of the bedside monitors more difficult, costly, and time-consuming.

Cassiere said he favored a waiver, but said that special controls might still be needed, including performance monitoring for the tests and more education for the test-givers, and studies that correlate capillary samples with venous samples. "I'm recommending free use of the device with controls around it, but not arduous controls," and controls "that do not place a burden on the healthcare provider," he said.

Bremer noted that there did not seem to be a strong consensus among panelists on whether to waive the CLIA requirements.

David Klonoff, MD, medical director of the Dorothy L. and James E. Frank Diabetes Research Institute of Mills-Peninsula Health Services in San Mateo, California, and a clinical professor at the University of California San Francisco, who spoke in the public hearing portion of the meeting, suggested that on the basis of the data presented by the FDA, capillary sampling for concentrations above 75 mg/dL were not a high-risk proposition. But for those lower than 75 mg/dL, the device was less accurate, demanding more scrutiny, especially given the risks of hypoglycemia.

For more diabetes and endocrinology news, follow us on Twitter and on Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.