International Federation of Clinical Chemistry (IFCC): Scientific Division, Committee on pH, Blood Gases and Electrolytes: Guidelines for Transcutaneouspo(2)andpco(2)Measurement.

This document provides guidelines for the terminology, methodology, and for the interpretation of data obtained from the use of skin (transcutaneous) po(2) and pco(2) electrodes. The transcutaneous technique has found special application in newborn infants. The causes of analytical bias with respect to arterial blood gas values, and imprecision obtained with transcutaneous pco(2) electrodes, are reviewed. Electrode temperatures above 44 degrees C should not be used routinely, and, at a measuring temperature of 44 degrees C, the measuring site should be changed at least every 4 h to avoid skin burns.


Introduction
Electrodes for transcutaneous measurements of oxygen and carbon dioxide are widely used to monitor continuously and non-invasively the oxygen and carbon dioxide status of patients with cardio-respiratory disorders. The transcutaneous technique has found special application in newborn infants, whose skin is thinner and has a greater density of capillaries, and in whom transcutaneous Po correlates closer to arterial blood Po than in adults. The purpose of this document is to provide guidelines for the terminology, methodology, and for the interpretation of data obtained from the use of skin electrodes. Arterial blood gas values are the reference points, and should be measured in connection with transcutaneous measurements. Recommendations covering the technical aspects of transcutaneous electrodes have been prepared by The American Society for Testing and Materials and The International Electrotechnical Commission [2].
Correspondence should be addressed to: P. D. Wimberley [8,9]. 5.1. Po2(tc) electrode Zero adjustment is usually performed automatically, assuming zero current at zero Po2, but may be checked with a 'zero' solution containing an oxygen consuming agent, such as sodium sulphite or with N2 gas.
Calibration is generally made with atmospheric air, or with a certified gas mixture containing known fractions of oxygen and carbon dioxide, balanced with nitrogen.
The electrode sensitivity (d Current/dpo2) should be within the manufacturer's specified limits, which depend on the cathode size and the membrane permeability and thickness. Furthermore, a stable calibration value should be attained within 3 min. Failure to satisfy these requirements should be indicated by an error message.

The monitor
The monitor should include: (a) Circuitry for Po2 and Pco2 measurement.  (1) Po2(tc) orpco2(tc) are outside pre-set limits selected by the clinician, or (2) electrode temperature deviates by more than 0"3 C from the selected temperature.

Calibration
Calibration should be performed at the electrode measuring temperature immediately prior to use and before each re-application of the electrode to the patient (i.e. normally every 4 h or less). Any fluid droplets on the outer side of the electrode membrane should be carefully removed before calibration. Calibration gas mixtures should have a relative inaccuracy of less than +0"5% of the stated value. The composition of the gas mixtures does not change with storage.

Pco2(tc) electrode
Two-point calibration is performed at least once a day with two dry gas mixtures with known CO2 fractions, for example 0"05 and 0"10 (Pco2 about 5 and 10 kPa respectively), with a one-point calibration (for example, with FcO2(G) 0"05) more frequently.
The electrode sensitivity (dE/dlgpco2) should be within 5% of the theoretical.
6. Measurement procedure A typical protocol is described, details of which may vary according to the electrode system and monitor used.
Detailed instructions should be found in the manuals produced by the manufacturers.

Selection of measurement site
Optimal measuring conditions are obtained in skin areas with high density of capillaries, ample capillary blood flow, thin epidermis, and small or no deposits of fat. The optimal places are the lateral sides of the abdomen and chest. The arms and legs are best avoided, as vasoconstriction, reducing skin blood flow, occurs earlier here if the patient becomes cold or hyp0tensive. In those newborn infants where there is a possibility ofright-to-left shunting through the ductus arteriosus, the electrodes must be placed on the right upper chest to detect/avoid high Po2 levels in the blood supplying the retina.
recommended calibration period of 4 h. The gas should have the same temperature and flow rate as that used for calibration.

Preparation of skin
Hair should be removed to ensure better adhesion.
Cleaning the skin with alcohol is recommended.

Fixation of the electrode
Fixation to the skin is facilitated by a self-adhesive ring.
6.9. Use of recorder A recorder is recommended because it has the advantage of providing retrospective analysis of changes in po2(tc and Peo2(tc). A recorder, however, increases the size of the monitoring system, and reduces its portability. Trends may also be observed by a visual display unit.

4. Contact liquid
Contact between the electrode and the skin is best ensured by a thin layer of fluid (for example, water or glycerol). It is important to avoid the presence of air between the electrode and the skin, and not to moisten the adhesive tape. In order to ensure the correct interpretation of transcutaneous values, at the start of each patient monitoring period, the relationship between Po(tc) and Po2(aB), and between Pco(tc) and PCO2(aB), should be established. Furthermore, it should be remembered that any change in Po(tc) may be due to either a change in PO2(aB), or due to a change in skin blood flow (see section 8.2.1). Po(tc) and Pco(tc) only reflect PO2(aB) and PCO(aB), respectively, at relatively high local blood flow.
Capillary blood cannot replace arterial blood as the reference system because the presence of even very small amounts of venous blood, which commonly contaminate the specimen, may result in a marked decrease in capillary blood Po2 compared to arterial blood Po. 6 At an electrode temperature of 44C, the electrode measuring site must be changed at least every 4 h, in order to avoid blister formation. Even after shorter measuring times, the skin will show erythema when ,the electrode is removed, though this erythema disappears after one or two days. The risk ofskin burning is increased in patients with peripheral circulatory failure, and in very small, premature infants and therefore the measuring site should be changed more frequently, for example every 2-3 h. An alarm should be included to ensure that the device automatically turns off within a few seconds, should the electrode core temperature rise above 46 C.
Two thermistors are recommended to ensure back up in the event of failure.
See also section 9.1 on electrode temperature control.

Electrical safety
All equipment used should be certified by its manufacturer to comply with a national or international electrical safety standard. The equipment must not interfere electrically with other equipment (for example an electrocardiogram or a heart pacemaker) attached to the patient.
8. Performance characteristics 8.1. Imprecision Imprecision may be due to the electrode drift, variations in pressure on the electrode, and the variable properties of the skin. Coefficients of variation (CV) determined from duplicate measurements in the same individual may be expected to be about 10% for Po2(tc and 5% for Pco2(tc) [10,11]. These figures include both biological and analytical variation. The same electrode should have a CV of only 1-2%, when duplicate measurements are made in dry gases [10]. 8 [4,12,13]. However, large biases in Po2(tc) are common [10], especially in adults [7,14] due to the following causes: (a) Variation in the Po2 temperature coefficient with increasing Po2 level, gives a negative bias in Po2(tc) at high Po2 levels, i.e. Po2(tc) reads too low. Other factors, for example abnormal total haemoglobin concentration and an altered haemoglobin oxygen affinity, also give a bias, but these effects are smaller [15]. (b) Variation in skin capillary blood flow: decrease in skin capillary blood flow causes a decrease in Po2(tc)" Changes in blood flow may be due to sepsis, changes in blood Pco2, administ.ration of drugs, for example tolazoline and catecholamines, and furthermore may be general or localized to the area of skin under the electrode.
(c) Variations in skin anatomy: increasing skin thickness and decreasing skin vascularization, which, for example occur with increasing age, result in low Po2(tc values. Differences in skin anatomy between various parts of the body also cause bias. (d) Interference from anaesthetic gases: halothane and nitrous oxide may give falsely elevated Po2(tc) values [16][17][18][19]. Pco2(te) value measured at an electrode temperature 0(E) is divided by exp (0.04 ((0(E)/C) 37)) and then reduced by 0"8 kPa. The temperature coefficient of 0"04 used here is slightly lower than the anaerobic temperature coefficient for Pco2 in blood (dlnpco/dO 0"048 C-1 [23]) because the skin capillary temperature is slightly lower than the electrode temperature [24,25]. 9. Quality assurance 9.1. Electrode temperature control Temperature control should be accurate to within 0'3 C.
This should be checked periodically by placing the electrode in a temperature-controlled environment. With the heater power off, the electrode monitor's temperature measurement should be within +0"3 C of the environmental temperature [1]. With heater power on, and monitor temperature set to the environmental temperature, the monitor heat output should read zero.

Electrode drift rate
When placed in a calibration gas, Po and Pco should not drift more than +1% per hour. For drift during patient monitoring, see section 6.8.

Changes in heat consumption during patient monitoring
A decrease in Po2(tc value, together with decreased heat dissipated by the heating element, indicates reduced blood flow in the skin capillaries and in the deeper and larger vessels under the electrode, and not necessarily reduction in PO2(aB). However, changes in heat consumption also occur with changes in the environmental and the patient's temperature.