Interference and Hematocrit Correction

Interference and Hematocrit Correction in [Diabetic Test Strips](/sell-diabetic-supplies/)

A blood glucose strip doesn't measure glucose directly. It measures a tiny electrical current produced when an enzyme on the strip reacts with glucose, so anything else that changes that current can change your number [1]. The two biggest sources of this kind of error are interfering substances in the blood and hematocrit, the share of blood made up of red cells. Which substances matter depends mostly on the strip's enzyme. How well hematocrit is handled depends on how the meter reads the sample.

How a strip turns glucose into a number

Strip enzymes come in two families, glucose oxidase (GOx) and glucose dehydrogenase (GDH), and GDH is paired with one of three helper molecules called cofactors: NAD, FAD or PQQ [2]. The enzyme pulls electrons off glucose. A mediator, a chemical go-between, carries those electrons to an electrode, and the meter converts the resulting current into a reading [1]. FDA's review of the first OneTouch Verio system describes the chain in one line: a GDH-FAD enzyme and a mediator reaction "generate an electrical current proportional to the glucose concentration" [3].

Interference can break into that chain at three points. A different sugar can react with the enzyme, something can compete with the mediator for electrons, or a drug can react directly at the electrode [1]. Our enzyme chemistry comparison goes deeper into the enzymes themselves.

Which interferents affect which enzymes

Glucose oxidase: oxygen and electrode-active drugs

GOx strips are sensitive to oxygen, because oxygen competes with the mediator for the enzyme's electrons [1]. Strips are calibrated for capillary blood. High oxygen, as in arterial samples or in people on oxygen therapy, pushes readings falsely low, while the low oxygen of venous blood or severe chronic obstructive pulmonary disease can push them high [1]. On electrochemical GOx strips, acetaminophen, L-dopa, tolazamide and ascorbic acid (vitamin C) act at the electrode, and very high uric acid can be oxidized there too [1].

GDH-PQQ: other sugars

The original form of GDH-PQQ isn't glucose specific. It also reacts with maltose, galactose and xylose, so a strip can report glucose that isn't there [2]. Here's what surprised us when we read the safety record: a 2010 review of FDA adverse-event reports and published cases found 82 events tied to falsely high GDH-PQQ readings, and 16 were associated with death [2]. Most involved icodextrin peritoneal dialysis solution (78%) or maltose-containing intravenous immune globulin (13%) [2]. The harm came from insulin given in response to a number that was too high [2].

Manufacturers answered by changing the enzyme. Roche's Accu-Chek Aviva Plus strips use a modified (mutant) GDH-PQQ, and FDA's review lists just three substances that caused significant interference in testing: triglycerides above 1,800 mg/dL, galactose above 15 mg/dL, and intravenous vitamin C that raises blood levels above 3 mg/dL [4].

GDH-FAD and GDH-NAD: mostly xylose

GOx, GDH-NAD and GDH-FAD don't give falsely high readings with maltose or galactose, although GDH-FAD does react with xylose [2]. The labeling reflects that. The Accu-Chek Guide, a GDH-FAD system, says not to use it during a xylose absorption test and warns that vitamin C blood levels above 5 mg/dL may cause inaccurate results [5]. FreeStyle Precision Neo strips use GDH-NAD, yet their labeling still notes xylose interference at low glucose levels [6].

Interference doesn't always push a reading up. The original OneTouch Verio labeling says acetaminophen, uric acid and vitamin C at normal or high therapeutic levels don't significantly affect results, but abnormally high blood levels "may cause inaccurately low results" [3].

Why hematocrit skews readings

Red cells carry glucose too, at lower levels than plasma, so whole-blood glucose shifts with hematocrit even when plasma glucose stays put [1]. In Ginsberg's worked example, the same blood holds about 88 mg/dL of whole-blood glucose at a hematocrit of 44%, 80 at 66% and 94 at 22% [1]. Most meters measure something between the two and are calibrated against plasma glucose, which is why a hematocrit far from normal shows up as error [1].

The effect is easy to see in lab data. A 2000 study tested six handheld meters at hematocrits of about 19%, 39% and 58%, and most read high at low hematocrit and low at high hematocrit [7]. In another study Ginsberg describes, one meter was off by as much as 40% at very low hematocrit [1]. Hematocrit isn't fixed from person to person, either: normal values differ by 8 to 10% between men and women, and altitude, smoking and kidney failure can shift them [1].

How meters correct for hematocrit

The usual approach is to probe the blood electrically before measuring glucose. FDA's review of the Accu-Chek Aviva Plus system lays out the steps: the meter applies a small alternating current (AC) at four frequencies and reads the responses, then applies a direct-current (DC) voltage to measure glucose, and combines both into "a hematocrit and temperature compensated glucose result" [4]. The Accu-Chek Guide review says the sample and environmental conditions "are evaluated using AC and DC signals" [5]. We explain the signal side in AC impedance and multi-pulse technology.

Roche patented the general idea in US 7,407,811 B2, issued August 5, 2008. The patent claims measuring AC and DC responses, using the AC response to correct the DC response, and calculating the analyte concentration from the corrected value [8]. A patent records what was claimed. It doesn't tell you which meters use it.

Correction doesn't make hematocrit irrelevant. It gives each system a labeled hematocrit range, and the ranges differ. FDA's reviews list 10 to 65% for the Accu-Chek Aviva Plus [4] and the Accu-Chek Guide [5], 15 to 65% for FreeStyle Precision Neo strips read in a FreeStyle Libre reader's built-in meter [6], and 20 to 60% for the original OneTouch Verio system [3].

Where to look for your own strips

The limitations section of a strip's package insert is where the manufacturer lists the interferents and hematocrit range for that exact strip, and the lists vary by enzyme and brand [3][5][6]. Home meters also carry wider exclusions: the Accu-Chek Guide labeling rules out critically ill patients and people in shock, dehydrated or in a hyperosmolar state [5]. If you take high-dose vitamin C, receive icodextrin dialysis or maltose-containing infusions, or have a hematocrit outside the usual range, which meter and strip fit you is a decision to make with your care team.