Why the New FDA Glucose Meter POCT Guidance Is Disappointing

Glucose meters differ from many other analytes since for some time, glucose meters have had accuracy performance standards for patient samples. The ISO 15197 standard of 2003, while for home glucose meters, was adopted by the Clinical Laboratory Standards Institute (CLSI) C30-A2, to specify glucose meter performance for point-of-care testing (POCT) meters. I critiqued these standards^1-3 since, among other things, they left 5% of the results as unspecified.

In 2013, CLSI issued an updated standard which had more narrow limits but still left 2% of the results as unspecified. ⁴ This prompted another critique, ⁵ and in 2014 the Food and Drug Administration (FDA) issued a draft guidance that had a remarkable statement: “FDA believes that the criteria set forth in the ISO 15197 standard do not adequately protect patients using BGMS devices in professional settings, and does not recommend using these criteria.” FDA also issued performance limits for 100% of the data.

But now, in 2016, FDA has released an updated guidance, ⁶ which is no longer a draft, which again leaves 2% of the results as unspecified!

A summary of the performance limits is listed in Table 1.

Table 1.

History of POCT Glucose Meter Performance Standards.

Year	Source	Percentage of data specified	Details
2003	ISO	95	95% of results must be < ±15 mg/dL below 75 mg/dL and < ±20% above 75 mg/dL
2013	CLSI	98	95% of results must be < ±12 mg/dL below 100 mg/dL and < ±12.5% above 100 mg/dLNo more than 2% > 15 mg/dL below 75 mg/dL and > 20% above 75 mg/dL
2014	FDA	100	99% of results must be < ±7 mg/dL below 70 mg/dL and < ±10% above 70 mg/dL0% > 15 mg/dL below 70 mg/dL and > 20% above 70 mg/dL
2016	FDA	98	95% of results must be < ±12 mg/dL below 75 mg/dL and < ±12% above 75 mg/dL98% of results must be < ±15 mg/dL below 75 mg/dL and < ±15% above 75 mg/dL

One argument that has surfaced is that one can never prove that 100% of the results will fall within limits. While this is true, it is illogical to use this fact as a reason to exclude a percentage of results from being specified. Nor does one see this in other medical arenas. Thus, one does not hear that 98% of surgeries should be performed on the correct site.

A simple remedy is to use an error grid for 100% of the data.