Enzyme Chemistries: GOx vs. GDH

Enzyme Chemistries: GOx vs. GDH

Every blood glucose test strip carries an enzyme, a protein that reacts with the glucose in your drop of blood, and the enzymes used on strips come from two families: glucose oxidase (GOx) and glucose dehydrogenase (GDH) [1]. The trade-off between them is easy to state. GOx is very picky about glucose, but it can also hand electrons to oxygen, so the oxygen level in a sample can nudge a GOx reading up or down [2]. GDH can't use oxygen at all. How picky a GDH strip is about other sugars depends on the helper molecule its enzyme carries [1].

What the enzyme does on a strip

When blood fills the strip, the enzyme pulls electrons off glucose. A mediator, a small molecule that ferries electrons, carries them to an electrode, and the meter turns the resulting current into a number [2]. Researchers group sensors by where those electrons go. First-generation designs gave them to oxygen; second-generation designs use a mediator instead, and third-generation designs pass them straight to the electrode [1]. We cover the mediator half of that story in Electron Mediator Technologies.

Glucose oxidase: specific, but sensitive to oxygen

A 2011 review called the mold Aspergillus niger the most common source of GOx for glucose testing, and that enzyme's very high specificity for glucose was documented back in the 1940s [1]. Its weak point is oxygen. On a strip, oxygen and the mediator compete for the enzyme's electrons, and only the mediator delivers them to the electrode. A sample rich in oxygen, such as arterial blood or blood from someone on supplemental oxygen, can read falsely low, and a sample low in oxygen can read high [2]. For that reason, GOx strips are generally calibrated for fingertip (capillary) blood. A 2009 review of meter errors added that GOx meters aren't approved for venous blood in most cases [2].

GOx hasn't disappeared from strips. The FDA review of the OneTouch Verio system, cleared in February 2011, describes the older OneTouch Ultra 2 system it was compared against as a glucose oxidase system [3]. The Prodigy Astro system, cleared in July 2016, uses glucose oxidase from Aspergillus niger [4].

Glucose dehydrogenase and its three cofactors

A cofactor is a non-protein helper molecule that accepts the electrons the enzyme strips from glucose. Strip GDH enzymes use one of three: pyrroloquinoline quinone (PQQ), nicotinamide adenine dinucleotide (NAD) or flavin adenine dinucleotide (FAD) [1]. None of them lets the enzyme pass electrons to oxygen, which is why GDH strips are less sensitive to oxygen changes than GOx strips [1][2]. The cofactor matters more than the family name, because it changes which other sugars the enzyme will react with.

GDH-PQQ: fast, but not picky

PQQ-GDH became a major strip enzyme because it is fast. Ferri and colleagues put its activity at more than 25 times that of GOx, and oxygen in the sample doesn't affect it [1]. The natural form, though, also reacts with other sugars, maltose among them [1]. In August 2009 the FDA warned that GDH-PQQ strips cannot tell glucose apart from maltose, xylose or galactose, sugars found in some drugs and biologic products or produced as the body breaks them down [5].

The numbers in that notice stopped us. From 1997 to 2009 the agency received 13 reports of deaths tied to this interference, ten of them in patients using Extraneal (icodextrin) peritoneal dialysis solution, and eight reports said the strip result ran 3 to 15 times higher than the laboratory value [5]. The same notice said strips using glucose oxidase, GDH-NAD or GDH-FAD were not affected by those sugars [5].

The FDA's August 2009 list of GDH-PQQ strips included Roche's Accu-Chek Aviva and Comfort Curve strips and Abbott's FreeStyle and FreeStyle Lite strips [5]. Both companies changed the chemistry afterward. FreeStyle Lite strips moved to a GDH-FAD enzyme, which the FDA cleared in May 2010 [6]. Roche took a different route: its Accu-Chek Aviva Plus strips, cleared in September 2011, use a modified ("mutant") GDH-PQQ [7]. Protein engineers had already shown that changing the PQQ enzyme could narrow the range of sugars it reacts with [1]. The change didn't remove every sugar effect, though. The Aviva Plus labeling still says blood galactose above 15 mg/dL will cause falsely high results [7].

GDH-NAD: specific, but the cofactor has to be added

NAD-dependent GDH is specific for glucose. Its NAD cofactor isn't attached to the enzyme, so strip makers have to supply it on the strip along with a mediator [1]. Abbott's FreeStyle Precision Neo strips use GDH-NAD [8]. Even so, their FDA review reports interference from xylose at low glucose levels and says not to test during a xylose absorption test, a diagnostic test that uses that sugar [8].

GDH-FAD: the most common enzyme in our check

FAD-GDH from fungi is specific for glucose and was studied largely because it doesn't react with maltose, but Ferri's review lists xylose in the sample as its known error source [1]. The labeling we read reflects that. In the FreeStyle Lite review, the highest xylose level with no significant interference was 6 mg/dL, and the strip insert warns against use during xylose absorption testing [6]. The Accu-Chek Guide labeling carries the same kind of warning [9].

Which enzyme is on your strip

Here is the enzyme each FDA review names for strip lines from the major US brands:

• Accu-Chek Guide: FAD-GDH [9]
• Accu-Chek Aviva Plus: mutant GDH-PQQ [7]
• OneTouch Verio: GDH-FAD [3]
• OneTouch Ultra Plus Flex system: FAD-GDH [10]
• OneTouch Ultra 2 system: glucose oxidase, as described in the Verio review [3]
• Contour Next strips: FAD-GDH [11]
• FreeStyle Lite: GDH-FAD [6]
• FreeStyle Precision Neo: GDH-NAD [8]
• True Metrix: GDH-FAD [12]

One note on True Metrix, which has nothing to do with its enzyme. As of October 2026, an FDA safety communication issued in April 2026 recommends that all True Metrix users, including people using store-brand versions, move to another way of testing when possible. The problem it describes is the meter's E-5 error code, which shows up both for very high glucose and for strip errors [13].

FreeStyle Lite is why we'd check the package insert rather than trust a strip's name. The strip kept its name when its enzyme changed [5][6], and the FDA's 2009 notice told providers and patients that makers of GDH-PQQ strips might switch methods and that labeling or the manufacturer should confirm which one a strip uses [5].

When the enzyme matters most

The sources point to specific situations rather than everyday fingertip testing. For GOx strips, the concern is unusual oxygen levels, such as arterial samples or supplemental oxygen [2]. For some GDH strips, it is other sugars: the FDA's list of interfering products included icodextrin dialysis solution and some immunoglobulin products [5], and several of the FDA reviews above warn against testing during a xylose absorption test [6][8][9]. If one of those applies to you, which meter and strip to use is a decision for your care team, and the insert lists the substances its maker tested.