Measurement
Blood Gas Analyzers: pH, pCO2, and pO2
Blood gas analyzers use electrodes (macroelectrochemical or microelectrochemical sensors)
as sensing devices to measure pO2, pCO2, and pH. The pO2 measurement is amperometric.
Amperometry is defined as measurement of amperes. Ampere is the unit of measure for
electric current. The reduction of oxygen produces a current that is proportional to the amount
of oxygen present in the sample. Conversely, the pCO2 and pH measurements are
potentiometric. Potentiometry measures the electric potential between two electrodes, in
which a change in voltage indicates the concentration of each analyte. Advances in microsensor
technology have greatly expanded the analytic menu of whole blood analyzers. In addition to pH
and blood gas measurements, many manufacturers include hemoglobin and/or hematocrit,
sodium, potassium, chloride, glucose, lactate, creatinine, blood urea nitrogen, and CO-oximetry
with their instrumentation. The blood gas analyzer can also calculate several additional
parameters: bicarbonate, total CO2, base excess, and SO2.
Measurement of pO2
The pO2 electrode is an amperometric electrode, often referred to as a Clarke electrode. An
oxygen-permeable membrane covers the tip of the electrode to selectively allow O2 to diffuse
into the electrolyte solution. A small, constant polarizing potential (approximately –0.65 V) is
applied between the anode and cathode. As a result, the electrons are drawn from the anode
to the cathode. The oxygen diffuses from the sample to the cathode, where it is subsequently
reduced. The current produced in this circuit is proportional to the amount of O2 reduced at the
cathode. A microammeter placed in the circuit between the anode and cathode measures the
movement of electrons (current). The semipermeable membrane will also allow other gases to
pass, such as CO2 and N2, but these gases will not be reduced at the cathode if the polarizing
voltage is tightly controlled.
The primary source of error for pO2 measurement is associated with the buildup of protein
material on the surface of the membrane. This buildup attenuates oxygen diffusion and slows
the electrode response.
It is of utmost importance to not to expose the sample to room air when collecting,
transporting, and making O2 measurements. Contamination of the sample with room air can
result in significant error and can cause spurious increases in the pO2 result. This is due to the
pO2 of room air being greater than 150 mm Hg. Another source of error in pO2 determinations
arises when a patient’s sample is not analyzed immediately. This is due to the leukocytes in the
sample continuing to use O2 in metabolic processes. In addition, if a patient has a markedly
increased leukocyte count, this reduction in pO2 can happen more rapidly.
Measurement of pH and pCO2
To measure pH, a glass membrane sensitive to H+ is placed around an internal Ag–AgCl
electrode to form a measuring electrode. Potential develops at the glass membrane as a result
of H+ diffusion from the unknown solution into the membrane’s surface. Hydrogen ions are
proportional to the difference in cH+ between the unknown sample and the buffer solution inside
the electrode. The developed potential at the glass membrane is measured by placing a
reference electrode in the solution, and both electrodes must be connected to a pH (volt)
meter. The indicator voltage is compared to the reference electrode (commonly either a
calomel [Hg–HgCl] or an Ag–AgCl half-cell) that provides a steady reference voltage. The pH
meter reflects the potential difference between the two electrodes. For the cell described, the
Nernst equation predicts that a change of +59.16 mV, at 25°C, results in a 10-fold increase in
H+ activity, equivalent to a decrease of one pH unit (e.g., pH 7.0 to 6.0). However, this
response is temperature specific. For example, at 37°C, a change of one pH unit elicits a 61.5
mV change.
pCO2 method is based on the pH electrode. It has an outer semipermeable membrane that
allows CO2 to diffuse and dissolve into an internal layer of electrolyte, usually a bicarbonate
buffer. The CO2 that diffuses across the membrane reacts with the buffer, forming carbonic
acid. Carbonic acid then dissociates into bicarbonate and hydrogen ions. The change in H+
activity is measured by the pH electrode and is related to the amount of CO2 present in the
sample.
As with any electrode, the buildup of protein material on the membrane will affect diffusion
and cause errors. pCO2 electrodes are slower to respond because of the chemical reaction
that must be completed. As previously mentioned, electrodes are also sensitive to
temperatures and barometric pressure. Other sources of error include erroneous calibration
caused by incorrect or contaminated calibration materials.
Types of Sensors
Macroelectrode sensors have been used in blood gas instruments since the beginning of the
clinical measurement of blood gases. These have been modified over time in an effort to
simplify their use and minimize the required sample volume and maintenance. Microelectrodes
basically are miniaturized macroelectrodes. Miniaturization became possible with better
manufacturing capabilities and with the development of the sophisticated electronics required to
handle minute changes in signal.
Thick and thin film technology is a further modification of electrochemical sensors. Although
the measurement principle is identical, the sensors are reduced to tiny wires embedded in a
printed circuit card. The special card has etched grooves to separate components. A special
paste material containing the required components (similar in function to the electrolytes of
macroelectrodes) is spread over the sensors. To reduce the required sample volume, several
sensors can be placed on a single small card. These sensors are disposable and less
expensive to manufacture, which reduces maintenance.
Optical sensors are another form of technology that is used for blood gas measurements.
Optical sensors (optodes) measure fluorescence or phosphorescence of organic dyes with O2,
CO2, and H+. The dye is separated from the sample by a membrane, as with electrodes, and
the analyte diffuses into the dye, causing either an increase in or a quenching of fluorescence
proportional to the amount of analyte. Calibration is used to establish the relationship between
concentration and fluorescence. Normally, a single calibration will suffice for long periods
because this technology is not subject to the drifts seen in electrochemical technology.
Optical technology has been applied to indwelling blood gas systems.3 Fiberoptic bundles
carry light to sensors positioned at the tip of catheters, and other bundles carry light back,
allowing changes in fluorescence to be measured in a catheter within the patient’s arterial
system. The commercial development of indwelling systems has been limited by the increased
probability of thrombogenesis and protein buildup on the membrane, separating the sample
from the fluorescing dyes. This buildup impedes free sample diffusion into the measuring
chamber.
Calibration
Calibration of a blood gas analyzer will vary among manufacturers. Normally, two different gas
mixtures with known pCO2 and pO2 levels are used. One of the mixtures is used to calibrate the
lower end, and the composition of this mixture is 0% O2 and 5% CO2, whereas the second
mixture is used to calibrate the gases on the upper end, and the composition of the mixture is
20% O2 and 10% CO2.
Most blood gas instruments will calibrate automatically at specified time intervals since
electrodes are drifted over time. Most electrodes will require calibration every 30 to 60
minutes. Instruments are programmed to prompt the operator if the values deviate from the
expected value. For example, if the value(s) obtained during calibration exceed(s) a
programmed tolerance limit, flagging of a drift error will occur at the time of calibration, and
corrective action will need to be taken before patient samples can be analyzed.
The pH electrode is calibrated against two primary traceable buffer solutions traceable to
standards that meet specifications set by the National Institute of Standards and Technology
(NIST). Typically, there are two calibrators, a low and high calibrator, with pH values of 6.8 and
7.38, respectively. The calibrators must be stored at the stated temperature and not exposed
to room air, since pH changes with the absorption of CO2.
pH and blood gas measurements are extremely sensitive to temperature. It is critical that the
electrode sample chamber be maintained at constant temperature for all measurements. All
blood gas analyzers have electrode chambers thermostatically controlled to 37°C ± 0.1°C, as
small temperature variations can drastically change the pH and blood gas values. Accordingly,
the Nernst equation is temperature specific; if the temperature of the measurement system
changes, the output (voltage) will change. The solubility of gases in a liquid medium also
depends on the temperature; as the temperature goes down, the solubility of the gas
increases.
Correction for Temperature
Values for pH, pCO2, and pO2 are temperature dependent. By convention, these measurements
are made at 37°C. But what happens when a patient’s actual body temperature differs from
37°C? After manual entry of the patient’s body temperature, the blood gas analyzer software
performs these corrections. However, the data can be difficult to interpret because appropriate
reference ranges for the patient’s temperature must be used, but these are not readily
available. When temperature-corrected results are reported, it is critical that results also be
reported with the 37°C (noncorrected) results.
Calculated Parameters
Several acid–base parameters can be calculated from measured pH and pCO2 values.
Manufacturers of blood gas instruments include algorithms to perform these calculations. No
calculated parameter is universally used; many physicians have preferred parameters for
identifying various pathologies.
measurement is based on the Henderson-Hasselbalch equation and can be calculated
when pH and pCO2 are known. The Henderson-Hasselbalch equation assumes that the pKa of
the bicarbonate buffer system in plasma at 37°C is 6.1.
Carbonic acid (H2CO3) concentration can be calculated using the solubility coefficient of CO2
in plasma at 37°C. The solubility constant to convert pCO2 to mmol/L of H2CO3 is 0.0307. If the
temperature or the composition of plasma changes (e.g., an increase in lipids, in which gases
are more soluble), the constant will also change.
Total carbon dioxide content (ctCO2) is the bicarbonate plus the dCO2 (carbonic acid) plus
the associated CO2 with proteins (carbamates). A blood gas analyzer approximates ctCO2 by
adding the bicarbonate and carbonic acid values (ctCO2 = c + [0.0307 pCO2]).
Some clinicians use base excess to assess the metabolic component. Base excess is
calculated from an algorithm that uses the patient’s pH, pCO2, and hemoglobin. A positive base
excess value indicates an excess of bicarbonate and suggests metabolic alkalosis. Conversely,
a negative value (base deficit) indicates a deficit of bicarbonate and suggests metabolic
acidosis. Because the nonrespiratory alkalosis or acidosis may be a result of primary
disturbances or compensatory mechanisms, base excess values should not be used alone in
assessing a patient’s acid–base status.
Spectrophotometric Determination of Oxygen Saturation (Co-Oximetry)
The actual percent oxyhemoglobin (O2Hb) can be determined spectrophotometrically using a
CO-oximeter designed to directly measure the various hemoglobin derivatives. Each
hemoglobin derivative has a characteristic absorbance curve (Figure 12.5). The number of
hemoglobin derivatives measured will depend on the number and specific wavelengths
incorporated in the instrumentation. For example, two-wavelength instrument systems can
measure only two hemoglobin species (i.e., O2Hb and HHb), which are expressed as a fraction
or percentage of the total hemoglobin.
Figure 12.5 Optical absorption of hemoglobin fractions.
Reproduced from Clin Chem News 1990 [January], with permission.
Description
Ideally, CO-oximeters should have four wavelengths for measurements of HHb, O2Hb, COHb,
and MetHb. Instruments with five wavelengths can also measure sulfhemoglobin as well as
recognize dyes and pigments, turbidity, and abnormal proteins. Some CO-oximeters employ
hundreds of wavelengths, which has reduced measurement interferences. Microprocessors
control the sequencing of multiple wavelengths of light through the sample and apply the
necessary matrix equations after absorbance readings are made to calculate the percentage of
the individual hemoglobin derivative:
(Eq. 12.15)
Description
where a1, an, bn, etc., are coefficients that are analogues of the absorption constant (a) that
are derived from established methods and A1, A2 . . . An are the absorbances of the sample.
The matrix equations will change depending on the number of wavelengths of light (which is
manufacturer specific) passed through the sample.
The results from these instruments should not be confused with a calculated SO2 from a
blood gas analyzer, which estimates the SO2 value from a measured pO2 and an empirical
equation based on the oxygen–hemoglobin dissociation curve. Only measured O2Hb values
reflect the patient’s true status because calculated SO2 and O2Hb values will be different in the
presence of dyshemoglobins. For example, in carbon monoxide poisoning, the SO2 will likely be
normal with a significantly decreased O2Hb level.
As with any spectrophotometric measurement, potential sources of error exist, including
faulty calibration of the instrument and spectral-interfering substances. The presence of any
substances absorbing light at the same wavelengths used in the measurement is a potential
source of error. Product claims for specific instruments must be consulted for interferences.
The primary purpose of determining O2Hb is to assess oxygen transport from the lungs.
Before blood sample collection, it is best to stabilize the patient’s ventilation status. Following
changes in supplemental O2 or mechanical ventilation, wait an appropriate period of time before
the sample is drawn. Heparinized blood samples should be collected under anaerobic
conditions and mixed immediately. All samples should be analyzed promptly to avoid changes in
saturation resulting from the consumption of oxygen by metabolizing cells.
