Hematocrit Interference Mechanisms and Correction Algorithms

Hematocrit Interference Mechanisms and Correction Algorithms

Red blood cells change a test strip's signal in two ways. Physically, they get in the way: dissolved molecules have to travel around them and thicker blood slows everything down, so a high hematocrit pulls readings low and a low one lets them run high [1]. Chemically, red cells hold less glucose than plasma, so the glucose in a drop of whole blood shifts with the share of cells even when plasma glucose stays put [2]. Correction methods either measure something during the test that tracks hematocrit and adjust the math, or build the strip so red cells matter less [1].

Hematocrit is the share of blood volume made up of red cells. Our hematocrit interference and correction article covers which way the error runs and the hematocrit ranges on meter labeling; this one stays with the mechanisms.

What a strip is really measuring

A meter tests whole blood, a mix of plasma and cells, but it is calibrated to report a plasma-equivalent number: the value a lab would get from the liquid part of the sample [1]. LifeScan's researchers describe 42% hematocrit as the usual design point [1].

Part of the error is plain bookkeeping. Ginsberg's 2009 review notes that red cells, unlike most cells in the body, carry significant glucose inside them, about 70 mg/dL when plasma holds 100 [2]. At 44% hematocrit, that blood holds about 88 mg/dL of whole-blood glucose. Change only the hematocrit and the number moves, to about 80 at 66% and about 94 at 22% [2]. Most strips measure neither pure plasma nor whole blood but "a complex mixture" calibrated against plasma, which is how the cell fraction leaks into the result [2].

Why red cells change the current

They are obstacles

The best-described mechanism is crowding. LifeScan's 2012 paper treats red cells as impermeable barriers that dissolved molecules must detour around, which lengthens their path [1]. The enzyme reactions and the electrode reaction both depend on diffusion, so when diffusion slows, the signal drops [1].

The paper gives a formula for how fast molecules spread through blood: D(h) = D₀(1 − h)/(1 + μh), where h is hematocrit as a fraction and μ is a shape factor of about 0.952 for red cells [1]. We ran the numbers out of curiosity: by that formula, going from 42% to 60% hematocrit cuts the effective diffusion coefficient by close to 40%. That isn't a meter's reading error, since meters correct for it, but it shows why the effect is far from small.

They thicken the blood

Viscosity climbs faster than hematocrit does. The same paper puts blood's relative viscosity at about 4 at 40% hematocrit and about 8 at 60%, so a 50% rise in hematocrit roughly doubles it [1]. Using two standard relationships from physical chemistry (the Stokes–Einstein and Cottrell equations), the authors show the strip's current is inversely proportional to viscosity [1]. They point to polycythemia, a condition in which the blood carries too many red cells, as one place these swings happen [1].

They can crowd out plasma or clog a membrane

A 2013 lab study lists the hypotheses researchers have proposed: altered viscosity, cells keeping plasma from reaching the reaction surface, changed diffusion kinetics, and packed cells displacing so much plasma that too little is left for an accurate test [3]. LifeScan's paper adds pore blockage in strips built around a membrane [1], and Ginsberg notes that cells can block the electrode or the enzyme directly [2]. Here's what surprised us: the 2013 authors still wrote that the mechanism "is not entirely understood" [3].

How the correction methods work

LifeScan's researchers sort the fixes into two families. Compensating methods probe the sample and correct the math, while design methods build features into the strip that keep red cells away from the sensing chemistry [1].

Measuring electrical impedance

Red cells resist electric current more than the plasma around them, so more cells mean more resistance between two fixed electrodes [1]. A relationship worked out by Maxwell, and later adapted by Fricke, ties a suspension's resistivity to the volume fraction of particles in it, which turns a resistance reading into a hematocrit estimate [1]. Meters use alternating current (AC) to avoid driving chemical reactions at the electrodes, so strictly they measure impedance, the opposition to AC [1].

FDA's decision summary for the Accu-Chek Aviva Plus system spells out the sequence. The meter applies AC at four frequencies, and the responses "carry information about the sample type and environmental temperature." It then applies a small direct-current (DC) voltage, reads a current proportional to glucose, and combines both into "a hematocrit and temperature compensated glucose result" [4]. Our AC impedance and multi-pulse technology article goes further into the electronics.

In a 2026 lab study, researchers in China added impedance-based correction to a low-cost screen-printed strip without making its electrodes more complex [5]. Uncorrected, their readings were off by roughly −80% to +30% as hematocrit changed. Corrected, they stayed stable from 10% to 70% hematocrit [5].

Timing how fast the mediator travels

The OneTouch Verio strip, as LifeScan's paper describes it, measures diffusion instead of resistance. A gold electrode faces a palladium electrode across a thin spacer, and the glucose chemistry sits only on the palladium side [1]. For the first 4 seconds of the 5-second test, the meter holds the gold electrode at +300 millivolts to collect mediator (the chemical that carries electrons from the enzyme) diffusing across the gap. How much arrives depends on glucose and on how easily molecules move through that particular blood [1].

For the final second, the meter flips to −300 millivolts and records a second current. Its algorithm takes the ratio of the two currents, raised to a fitted power, and multiplies it by a background-corrected current that tracks glucose [1]. Both currents shrink as hematocrit rises, and the paper credits the combined measurement with results that are insensitive to hematocrit from 20% to 60% [1].

Dynamic electrochemistry

A third approach changes the stimulus as the test runs. The 2011 Musholt study describes dynamic electrochemistry as measuring under shifting frequencies and voltages and adjusting the input signal as the reaction progresses, producing a "fingerprint" the algorithm turns into correction factors [6]. Its model gives each oxidation reaction its own phase angle, so glucose's contribution can be separated from an interfering factor's [6]. The meter in that study that used it, sanofi-aventis's BGStar, kept its hematocrit spread at 8%, inside the study's 10% limit [6].

A dedicated hematocrit electrode

The most direct method adds an electrode whose only job is measuring hematocrit. Nova Biomedical's StatStrip, a point-of-care meter, works this way and feeds that reading into a correction algorithm [3]. In the 2011 study, it showed less than 2% bias across hematocrit levels [6].

Keeping cells away from the chemistry

Design methods change the strip rather than the math. LifeScan's paper cites patent filings for this route, including one for a strip with an integrated reagent and blood-separation layer [1]. A membrane has its own weak spot, though: the same paper lists pore blockage in membrane-based strips as a cause of hematocrit bias [1].

What correction leaves unsolved

Each correction is validated over a stated hematocrit range: 20% to 60% for the system in LifeScan's paper [1] and 10% to 70% in the 2026 lab study [5]. If your hematocrit runs well outside the usual range, which meter suits you is a decision for your care team.

The other gap is transparency. Of the papers and FDA summaries we read, only LifeScan's laid out its correction formula [1]. The Aviva Plus summary describes the steps but not the math [4].