Previous Next

MEASUREMENT OF BLOOD GAS TENSIONS

The most straightforward method of ascertaining PaO2 is to measure it directly from a blood sample, usually by using a Clark electrode, which incorporates platinum and reference electrodes in an electrolyte bath (see Chapter 41 ). When a negative polarizing voltage is applied to the platinum electrode, O2 molecules in solution are reduced by electrons from the electrode. The electrons from the cathode react as follows:

O2 + 2H2 O + 2e− → H2 O2 + 2OH−
H2 O2 + 2e− → 2OH− (15)

Each molecule of O2 is therefore reduced to two hydroxyl ions by four electrons. In the Clark electrode, the electrolyte solution is separated from the fluid being measured (e.g., blood) by a thin O2 -permeable membrane. The current generated is proportional to PO2 within the sample.

When CO2 is equilibrated with an aqueous solution, the concentration of carbonic acid is proportional to the PCO2 .

H2 CO3 = α PCO2 (16)

H2 CO3 ↔ H+ + HCO3 − (17)





In these equations, α is a constant, and k is a constant that equals 24 if the units are nanomoles/L for [H+ ], mm Hg for PCO2 , and mmol/L for [HCO3 - ]. PCO2 electrodes work by measuring the change in pH induced when blood equilibrates with a potassium chloride-sodium bicarbonate solution.[31]

The history of the development of blood gas analysis has been described by Severinghaus and Astrup[32] and that of O2 monitoring in general by Severinghaus.[33] To obviate the need to withdraw blood samples for PO2 analysis, intra-arterial PO2 monitors have been developed. Initial attempts to use a Clark-type electrode were confounded by problems of drift and blood coagulation. This has been largely replaced by a fiberoptic technique, based on the property of O2 to absorb energy from excited electrons in a fluorescent dye. Incident light is used to elevate electrons in the dye to a higher energy state. These excited electrons may then return to a lower energy level, emitting a photon in the process. Molecular O2 , by absorbing the energy, inhibits the photon emission. This process, called fluorescence quenching, is related to the PO2 . [34]

A system using this technique (Diametrics Medical, St. Paul, MN), which has a ruthenium indicator to measure PO2 , also incorporates an optical absorbance technique to measure PCO2 and pH. Blood pH is measured by continuous

1445
monitoring of the absorbance of a pH-sensitive dye (i.e., phenol red) in polyacrylamide gel. At low pH, the color of the dye is yellow-orange; at high pH, it is dark red. Pulses of a green monochromatic light beam (wavelength of 555 nm) are transmitted along an optical fiber through the dye and reflected by a microscopic mirror. Pulses of red light (660 nm) are synchronized between the green pulses. The red light, the absorption of which is unaffected by pH changes, serves as a reference. Blood pH is related to the ratio of the intensities of the green and red light returned through the fiberoptic beam to the analyzer. PCO2 measurement uses the same principle, in which the pH indicator with a fixed concentration of bicarbonate is sealed within a gas-permeable membrane that is impermeable to H+ ions. CO2 molecules diffusing into the cell liberate H+ ions from water. PCO2 can then be calculated from the measured pH (see Equation 18).

The fiberoptic probes of commercially available systems are somewhat fragile, and the probes can be subject to movement artifact. However, bias, the mean difference between instrument measurements and traditional blood gas analysis, is typically 5%, and imprecision,[35] the standard deviation of these differences, is about 10%, which are quite acceptable for clinical purposes. The 90% response time for pH is 70 to 80 seconds and 140 seconds for PO2 and PCO2 ; there is minimal drift over several days.[36] Use of continuous arterial blood gas monitoring systems has been described in several clinical settings[36] [37] [38] [39] and its use reviewed.[40] Umbilical artery insertion of a probe
TABLE 36-1 -- Algorithms for correction to body temperature of blood gas tensions measured at 37°C
pH

ΔpH/ΔT = - 0.0146 + 0.0065 (7.4 - pHm )

ΔpH/ΔT = - 0.015

ΔpH/ΔT = - 0.0147 + 0.0065 (7.4 - pHm ) *

ΔpH/ΔT = - 0.0146
PCO2

Δlog10 PCO2 /ΔT = 0.019 *

Δlog10 PCO2 /ΔT = 0.021
PO2




Δlog10 PO2 /ΔT = 0.0052 + 0.27 [1 - 10-0.13(100-SaO2 ) ]







SO2 ≤ 95%: Δlog10 PCO2 /ΔT = 0.31

SO2 > 95%: Δlog10 PCO2 /ΔT = 0.032 - 0.0268e(0.3SO2 - 30)
Hb, blood hemoglobin concentration in g/dL; pHm and PO2m , pH and PO2 values measured at an electrode temperature of 37°C; PO2 , partial pressure of oxygen in mm Hg; SO2 , hemoglobin-oxygen (Hb-O2 ) saturation in percent; T, temperature in degrees centigrade (°C).
Data from Ashwood ER, Kost G, Kenny M: Temperature correction of blood-gas and pH measurements. Clin Chem 29:1877, 1983, and Siggaard-Andersen O, Wimberley PD, Gothgen I, Siggaard-Andersen M: A mathematical model of the hemoglobin-oxygen dissociation curve of human blood and of the oxygen partial pressure as a function of temperature. Clin Chem 30:1646, 1984.)
*National Committee for Clinical Laboratory Standards (NCCLS) approved standard.[41]




has been successfully used to monitor blood gas parameters in neonates.

When arterial puncture cannot be achieved or may be technically difficult (e.g., in neonates), capillary PO2 may approximate the arterial value, particularly if the sampling site (e.g., heel) is prewarmed, causing an abundance of local blood flow relative to local tissue O2 consumption. Although capillary PO2 tends to be lower than the arterial value because of the shallow slope of the upper part of the Hb-O2 dissociation curve, capillary O2 saturation approximates arterial O2 saturation (SaO2 ) (capillary value is usually slightly lower).

Temperature Correction of Blood Gases

Conventionally, the temperature of the electrodes in blood gas analyzers is maintained at 37°C, but rarely is a patient's body temperature exactly this temperature (see Chapter 41 ). A blood gas sample from a patient is warmed or cooled to 37°C before analysis. Temperature changes cause alterations in gas solubility in plasma and O2 affinity for hemoglobin. Because solubility and Hb-O2 affinity decrease as temperature rises (and vice versa), if a syringe of blood is gas tight and no bubbles are present, heating the sample will elevate blood gas tensions and reduce the pH; cooling the sample will cause the reverse. Several algorithms exist for temperature correction, allowing the true partial pressure, predictive of physical and chemical activity at the tissue, to be calculated ( Table 36-1 ).

1446


Figure 36-8 Temperature correction of blood PO2 : percentage error of PO2 measurement at various body temperatures with the blood gas electrode at 37°C. If the patient's temperature is 20°C, a PO2 measurement performed at 37°C would overestimate the true value by more than 200%. The curves demonstrate that the percentage of error is greater for venous values (PO2 < 100 mm Hg) than for arterial measurements (PO2 > 300 mm Hg). (Adapted from Camporesi EM, Moon RE: Arterial blood gas values should be corrected for body temperature during hypothermia. In Fyman PN, Gotta AW [eds]: Controversies in Cardiovascular Anesthesia. Boston, Kluwer Academic Publishers, 1988, p 35.)

The algorithms approved by the National Committee for Clinical Laboratory Standards [41] and most commonly used in clinical instruments sold in the United States are provided in Table 36-1 .

Application of the equation for PO2 is shown in Figure 36-6 . The PO2 derived from a 37°C electrode is overestimated if the patient is hypothermic and underestimated if the patient is febrile. At high PO2 values (>400 mm Hg) the effect is small, because hemoglobin is fully saturated in this region. At PO2 values below 100 mm Hg, however, the degree of overestimation may be significant. For example, at a patient temperature of 30°C and a PO2 below 80 mm Hg, the true PO2 is overestimated by about 60% unless temperature correction is applied ( Fig. 36-8 ).

Available evidence indicates that, during hypothermia, corrected arterial pH should be maintained in the alkalotic range (i.e., maintain uncorrected pH close to 7.4) (see Chapter 41 ). In contrast, the analysis previously presented suggests that PO2 values should be corrected for temperature.[42]

Previous Next