Monitoring oxygenation

Published on 27/02/2015 by admin

Filed under Anesthesiology

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 1424 times

Chapter 14 Monitoring oxygenation

Thomas John Morgan, Balasubramanian Venkatesh

THE ROLES OF OXYGEN IN AEROBIC ORGANISMS

Oxygen has several vital physiological roles:

1 Bioenergetics. Aerobic mitochondrial respiration accounts for 90% of oxygen consumption, generating adenosine triphosphate (ATP) by oxidative phosphorylation. Mitochondrial electron transfer oxidase systems provide the fundamental machinery, in particular the cytochrome oxidase of complex IV. Oxygen acts as terminal electron acceptor, combining with two protons to produce water.1
2 Biosynthetics. Oxygen transferase systems incorporate oxygen into substrates, such as prostanoids, catecholamines and some neurotransmitters.
3 Biodegradation and detoxification reactions. These mixed-function oxidase reactions require oxygen and a co-substrate (e.g. NADPH). The cytochrome P-450 hydroxylases are examples.
4 Generation of reactive oxygen species. These are essential antimicrobial defences deployed by neutrophils and macrophages.2

TISSUE HYPOXIA

As tissue PO2 falls, the biosynthetic and biodegradation systems are the first to succumb. Cell signalling by reactive oxygen species, released at mitochondrial complex III, activates hypoxia-inducible factor-1, a transcription factor upregulating genes important in hypoxic cell survival.3 Oxidative phosphorylation starts to fail at an intracellular PO2 of 0.1–1 mmHg, equivalent to an extracellular PO2 of around 5 mmHg. At this point oxygen-limited cytochrome turnover causes progressive ATP depletion, a process which has been termed ‘dysoxia’.4 Stopgap production continues by anaerobic glycolysis, but without correction of dysoxia there is progressive lactic acidosis and eventual cell death by apoptosis and necrosis. On reoxygenation, further release of reactive oxygen species causes oxidant stress, often overshadowing the hypoxic insult.2

CATEGORIES OF HYPOXIA

Classically, hypoxia has been categorised under three headings:1

1 stagnant hypoxia, where the primary abnormality is reduced tissue blood flow
2 hypoxaemic hypoxia, where the primary abnormality is a low arterial oxygen tension
3 anaemic hypoxia, where the primary abnormality is a low haemoglobin concentration

However in critical illness, tissue oxygenation can be disordered with a normal or even high tissue PO2:

1 cytopathic hypoxia, defined as abnormal cellular oxygen utilisation despite adequate oxygen delivery5
2 oxygen toxicity, defined as abnormal cell function due to high oxygen tensions6

THE OXYGEN CASCADE

In unicellular organisms, oxygen reaches the mitochondria across a short diffusion path with a steep partial pressure gradient. In multicellular animals the much longer diffusion path traverses a series of small partial pressure reductions known as the oxygen cascade. As a result, oxygen arrives at intracellular organelles at tensions above the anaerobic threshold.

Important steps in the oxygen cascade include:

1 inspired gas
2 alveolar gas
3 arterial blood
4 microcirculation
5 interstitium
6 mitochondria and other intracellular organelles

The cascade can be jeopardised at any step, with downstream oxygen deprivation of mitochondria and other intracellular organelles. In this chapter we will consider how oxygenation can be monitored at strategic points along the cascade.

INSPIRED GAS

Monitoring the fraction of inspired oxygen (FiO2) is necessary to prevent both hypoxaemia and the adverse effects of excess oxygen. The inspired oxygen tension (PiO2) of humidified gas is determined by the FiO2, the barometric pressure (BP) and the saturated vapour pressure of water (47 mmHg).

Equation 14.1 image

Gas supply pressures are monitored continuously. Ventilators incorporate input pressure alarms and oxygen analysers within the inspiratory module to identify oxygen source failure. Direct measurement of circuit oxygen concentration can be performed.

TRANSFER OF INSPIRED GAS TO ALVEOLI

Communication between oxygen delivery system and pulmonary alveoli is open if:

1 There are no signs of upper-airway obstruction (Chapter 25).
2 Expired tidal and minute volumes and airway pressures for the ventilated patient are within correctly set alarm limits (Chapter 27).
3 There is an appropriate end-tidal CO2 waveform (Chapter 34).

ALVEOLAR GAS

In a patient receiving 100% oxygen, alveolar PO2 in individual lung units can range from < 40 mmHg to > 600 mmHg. Consequently, end-tidal PO2 monitoring is of no value.

DISTRIBUTION OF ALVEOLAR VENTILATION

Clinicians routinely track chest movement, auscultate air entry and examine plain chest radiographs. Although computed tomography scanning can reveal occult overdistension,7 it has logistical disadvantages and is a significant radiation hazard. Electrical impedance tomography is under development as an alternative, and shows promise.8 Simple and non-invasive, it tracks lung volume changes in real time, with potential to optimise alveolar ventilation distribution while limiting overdistension.9

MATCHING OF VENTILATION AND PERFUSION

For efficient gas exchange, the majority of lung units must have well-matched alveolar ventilation and perfusion (V/Q ratios close to 1). Even in health there is a spread to lower and higher V/Q ratios, all clustered around unity. When lungs are diseased, V/Q scatter is much greater, with variable representation across the full spectrum from zero to infinity. Such complexity resists simple bedside quantification. The best method is the multiple inert gas technique (MIGET). MIGET generates a lung model with 50 compartments spanning the full range of V/Q ratios, by measuring the retention and elimination of six inert gases of varying solubility.10,11 Because MIGET is impractical at the bedside, simpler non-invasive modelling of shunt and V/Q mismatch is under evaluation, using FiO2 as a forcing function while tracking haemoglobin oxygen saturation.12 Predictive capacity13 and ease of application14 are encouraging.

THE IDEAL ALVEOLUS AND THE THREE-COMPARTMENT LUNG MODEL

Meanwhile the simplest model of all, a three-compartment model devised in the mid 20th century, is still in use. No manipulations such as FiO2 switching or inert gas infusions are needed. However, the trade-off is that the model is overly simplistic, with poor predictive capacity in many respiratory disorders. The three compartments are:

1 the ideal compartment, consisting of alveoli with perfectly matched perfusion and ventilation (V/Q = 1)
2 the venous admixture or shunt compartment, containing perfused non-ventilated alveoli (V/Q = 0)
3 the alveolar dead space compartment, made up of ventilated non-perfused alveoli. (V/Q = ∞)

The alveolar PO2 in the ideal compartment (PAO2) is calculated from the alveolar gas:

Equation 14.2 image

where R is the respiratory exchange ratio, either measured by indirect calorimetry or assumed to be 0.8. PiO2 is calculated as in Equation 14.1. PaCO2 is arterial PCO2.

Most clinicians use the following approximation:

image

It is important to remember that PAO2 is not a physiological reality. It is an artificial construct. Other parameters based on the three-compartment lung model, such as the A-a gradient and venous admixture (see below) must be regarded in the same light.

TRANSFER FROM ALVEOLI TO ARTERIAL BLOOD (PULMONARY OXYGEN TRANSFER)

The MIGET technique has identified V/Q mismatch and intrapulmonary right-to-left shunt as the two main causes of reduced pulmonary oxygen transfer in critical illness.15 Intrapulmonary shunt predominates in the acute respiratory distress syndrome (ARDS), in lobar pneumonia and after cardiopulmonary bypass, whereas V/Q mismatch without shunt is more prominent in chronic lung disease.16

BEDSIDE INDICES OF PULMONARY OXYGEN TRANSFER

These are either tension-based or content-based.

TENSION-BASED INDICES

A-a gradient

The A-a gradient is calculated as PAO2 – PaO2, where PAO2 is the ‘ideal’ compartment alveolar PO2 determined from the alveolar gas equation (Equation 14.2). Hypoxaemia can then be classified under two headings:

Normal A-a gradient

1 Alveolar hypoventilation (elevated PACO2)
2 Low PiO2 (FiO2 < 0.21, or barometric pressure < 760 mmHg)

Raised A-a gradient

1 diffusion defect (rare)
2 V/Q mismatch
3 right-to-left shunt (intrapulmonary or cardiac)
4 increased oxygen extraction (CaO2 – CvO2)

Although the A-a gradient forms part of the Acute Physiology, Age and Chronic Health Evaluation (APACHE 2) score, several drawbacks reduce its clinical usefulness. They include:

1 Normal values which vary with FiO2 and age. The normal A-a gradient breathing air is 7 mmHg in young adults and 14 mmHg in the elderly. On 100% oxygen, these values become 31 and 56 mmHg respectively.
2 An exaggerated FiO2 dependence in intrapulmonary shunt17 (Figure 14.1) and even more so in V/Q mismatch (Figure 14.2).
image

Figure 14.1 Effect of varying FiO2 (PAO2) on A-a gradient with different degrees of intrapulmonary shunt.

(Reproduced from Nunn JF. Oxygen. In: Nunn JF (ed.) Applied Respiratory Physiology, 4th edn. Oxford, UK: Butterworth-Heinemann; 1993: 264, with permission.)

image

Figure 14.2 Effect of varying FiO2 on A-a gradient with mild, moderate and severe V/Q mismatch. No allowance is made for absorption atelectasis or alterations in hypoxic pulmonary vasoconstriction.

(Reproduced from D’Alonzo GE, Dantzker DR. Respiratory failure, mechanisms of abnormal gas exchange, and oxygen delivery. Med Clin North Am 1983; 67: 557–71, with permission.)

Pao2/Fio2 ratio

The PaO2/FiO2 ratio is used to define acute lung injury and ARDS,18 and is an input variable in the Simplified Acute Physiology Score (SAPS II)19 and lung injury scoring systems.20 At sea level its normal value is ≥ 500 mmHg. In acute lung injury, PaO2/FiO2 < 300 mmHg, whilst in ARDS the ratio is < 200 mmHg.

Its only advantage is simplicity. There are several major disadvantages:

1 Barometric pressure alters the normal PaO2/FiO2 ratio. A ratio of 380 mmHg is unremarkable at 1600 metres elevation for a young adult breathing air, but not at sea level.
2 Unlike the A-a gradient, the PaO2/FiO2 ratio cannot distinguish hypoxaemia due to alveolar hypoventilation from other causes.
3 The ratio is markedly FiO2-dependent, both in right-to-left shunt (the predominant ARDS abnormality) and in lungs with a wide V/Q scatter (chronic obstructive pulmonary disease).17,21
4 The ratio is also highly dependent on CaO2 – CvO2,17 which tends to fluctuate markedly in sepsis.

CONTENT-BASED INDICES

Venous admixture (Qs/Qt)

Venous admixture, another construct based on the three-compartment lung model (see above), represents the proportion of mixed venous blood flowing through the shunt (V/Q = 0) compartment. It is determined according to the formula:

Equation 14.3 image

Cc’O2, CaO2 and CvO2 are the oxygen contents of pulmonary end-capillary, arterial and mixed venous blood respectively. CaO2 and CvO2 are calculated using data from arterial and mixed venous blood gas analysis and CO oximetry (see Table 14.4, below). Cc’O2 is derived differently, since pulmonary end-capillary blood cannot be sampled. image is assumed to equal PAO2 as derived from the alveolar gas equation (Equation 14.2). image (normally close to 1) can then be computed from an algorithm for the HbO2 dissociation curve.22

Table 14.4 Oxygen dynamics – measured and derived indices

image

Advantages of venous admixture

1 unaffected by barometric pressure
2 unaffected by alveolar hypoventilation
3 provided intrapulmonary right-to-left shunt is the dominant pathology (e.g. ARDS), venous admixture is stable across the entire FiO2 range, despite variations in CaO2 – CvO2.23

Disadvantages of venous admixture

1 Sampling mixed venous blood requires a pulmonary arterial (PA) catheter.
2 In V/Q mismatch without right-to-left shunt, venous admixture varies markedly with FiO2 and virtually disappears at FiO2 > 0.5 (Figure 14.3). Venous admixture is thus of little use in conditions where the dominant gas exchange abnormality is not intrapulmonary right-to-left shunt, such as chronic obstructive pulmonary disease.
image

Figure 14.3 Effect of varying FiO2 on venous admixture in various combinations of V/Q mismatch and shunt. No allowance has been made for absorption atelectasis or alterations in hypoxic pulmonary vasoconstriction.

(Reproduced from D’Alonzo GE, Dantzker DR. Respiratory failure, mechanisms of abnormal gas exchange, and oxygen delivery. Med Clin North Am 1983; 67: 557–71, with permission.)

When determined at FiO2 = 1, venous admixture is an accurate measure of right-to-left shunt. However, exposure to 100% oxygen causes absorption atelectasis of unstable low V/Q units and increases intrapulmonary shunt.

Estimated shunt fraction

If a fixed CaO2 – CvO2 can be assigned, no PA catheter is necessary. However, in critical illness CaO2 – CvO2 can range from 1.3 to 7.4 ml/dl. The inaccuracy from assigning a fixed value is such that even the direction of change of estimated shunt can be wrong.23

Non-invasive estimates of V/Q mismatch and shunt

Simpler lung models using varying FiO2 as a forcing function are under evaluation, as discussed above.12–14

ARTERIAL BLOOD

Indices of arterial oxygenation are PaO2 and SaO2. They are linked by the HbO2 dissociation curve (Figure 14.4).

image

Figure 14.4 Three HbO2 dissociation curves: normal (P50 = 26.7 mmHg), left-shifted (P50 = 17 mmHg) and right-shifted (P50 = 36 mmHg). The vertical line represents the normal oxygen loading tension (PO2 = 100 mmHg). The filled squares represent an oxygen extraction of 5 ml/dl blood, assuming a haemoglobin concentration of 150 g/l.

(Reproduced from Morgan TJ. The HbO2 dissociation curve in critical illness. Crit Care Resusc 1999; 1: 93–100, with permission.)

Clinically significant hypoxaemia is defined as PaO2 < 60 mmHg or SaO2 < 0.9. These values normally lie near the descending portion of the HbO2 dissociation curve, so that a further drop in PaO2 leads to a marked fall in SaO2 and thus CaO2.

BLOOD GAS ANALYSIS AND CO-OXIMETRY

Arterial blood is collected in a purpose-designed syringe containing lyophilised heparin to a final concentration of 20–50 U/mL. PaO2 measurements are made by a Clark electrode, and SaO2 by CO-oximetry. The Clark electrode works on polarographic principles, and CO-oximeters compute the concentrations of each of the four main haemoglobin species (HbO2, Hb, COHb, MetHb) from light absorbances of haemolysed blood over several wavelengths. SaO2 is functional saturation, determined from concentrations of HbO2 and Hb (see Table 14.4, below). Interference to CO-oximetry arises from substances with competing absorbance spectra, such as bilirubin, HbF, lipid emulsions and intravenous dyes. Newer multi-wavelength techniques reduce or eliminate this interference. SaO2 should always be measured rather than calculated.

ERRORS (Table 14.1)

Temperature correction

All measurements are made at 37°C. Temperature-corrected values can be calculated if the core temperature of the patient is entered into the device software. Most clinicians interpret blood gas data at 37°C, except when evaluating the A-a gradient.

Table 14.1 Preanalytic and analytic errors in PO2 measurement

Preanalytic Analytic
Oxygen diffusing into or out of air bubbles, according to the tension gradient Interanalyser variability. There is 7–8% measurement variation on the same sample
Contamination with flush solution. Discard volume should be 2–3 times the internal volume of cannula and tubing Inadequate anticoagulation, allowing protein deposition on the electrodes
Pseudohypoxaemia. Oxygen consumption in vitro from extreme leukocytosis Non-linearity at high PO2 (> 150 mmHg)
Artifactual PaO2 elevations. With polypropylene syringes stored on ice, the semipermeable plastic allows oxygen ingress, facilitated by the cold-induced increase in oxygen solubility Maintenance of electrode temperature within narrow limits (37 ± 0.1°C) is critical. PO2 changes by 7% for every degree Celsius temperature change
Interference by nitrous oxide and halothane is minimal, provided the polarizing voltage of the electrode does not exceed 600 mV
Quality control materials such as aqueous, perfluorocarbon and bovine haemoglobin solutions are used for convenience, but tonometry is the primary reference method
Arterial blood gas tensions fluctuate breath to breath Intermittent analysis is a snapshot

CONTINUOUS INTRA-ARTERIAL BLOOD GAS MONITORING (Table 14.2)24

Multiparameter fibreoptic sensors can be placed in the arterial stream. Fibreoptic sensors are called ‘optodes’, and those measuring PO2 normally operate by fluorescence quenching. They require calibration with precision gases or solutions before use. Typical sensors are 0.5 mm in diameter, and can be inserted through 20-G arterial cannulae. The 90% in vitro response time to a change in PO2 is 78 seconds. PO2 drift in vivo is 0.03 mmHg/hour. Recalibration in vivo can be performed against conventional blood gas analysis. Accuracy on in vitro and animal testing is good.

Table 14.2 Continuous intra-arterial PaO2 monitoring – advantages and disadvantages

Advantages Disadvantages
Eliminates preanalytic errors of intermittent blood gas analysis The ‘wall’ effect – a sudden decrease in measured Pao2 due to contact with the arterial wall, with averaging of arterial and wall oxygen tensions. The problem is reduced in larger arteries such as the femoral artery
More sensitive than pulse oximetry to changes in arterial oxygenation when PaO2 > 70 mmHg (the flat part of the HbO2 dissociation curve). The ‘flush’ effect. Unless the sensor is inserted a sufficient distance beyond the cannula tip, measured PaO2 can be altered by contamination with the continuous flush solution
Free from the sources of error of pulse oximetry (see Table 14.3) Damping of the arterial waveform
Near real-time PaO2 allows prompt tracking of responses to changed ventilator settings Large footprint of the free-standing monitor
Reduced exposure of personnel to potentially infected blood  
Reduced blood loss for diagnostic purposes  

Clinical trials evaluating the accuracy of these monitoring systems have revealed varying degrees of bias and imprecision. Trials demonstrating an improved outcome when therapeutic decisions are based on data from these devices are lacking. These factors, combined with the costs of these devices, have limited their bedside application.

Some problems encountered with intra-arterial sensors, particularly artefact due to flow and position, have prompted the development of extracorporeal monitors placed in line but ex vivo. These devices do not provide continuous real-time data. When a measurement is desired, a sample is drawn into the externally located cassette, and then returned. Results are available in 2 minutes. In preterm neonates this method has allowed significant reductions in red cell transfusions.25

TRANSCUTANEOUS PO2 and PCO2 MONITORING

Continuous non-invasive assessment of blood gas tensions is possible with transcutaneous PO2 and PCO2 monitoring. Available systems incorporate PO2 and PCO2 electrodes with integral thermistors and servo-controlled heaters. PO2 measurement utilises the principle of the Clark electrode, while the PCO2 device is a pH-sensitive glass electrode. To achieve good correlation with arterial values, the skin is warmed to a temperature of 42–44°C. Transcutaneous monitors generate reliable PCO2 values in adequately perfused patients, but PO2 measurements are more for trend analysis. Skin warming necessitates frequent site changes to prevent burns, and there is a need for regular recalibration. Monitoring in haemodynamically unstable patients is not recommended. The monitors have a role in the prevention of neonatal hyperoxia, a problem not reliably detected by pulse oximetry.26

PULSE OXIMETRY27,28

Pulse oximetry determines SpO2 from the absorbance of light at wavelengths 660 nm (red) and 940 nm (infrared) by tissue capillary beds such as fingers, earlobes and the nasal septum. Two light-emitting diodes cycle on and off at multiples of the mains frequency. A single photodiode detects the transmitted light, and a third interval allows correction for background ambient light. The emergent signal is pulsatile due to arterial volume fluctuations. Subtraction of the background signal (tissue, capillary blood and venous blood) isolates the arterial component.

For both wavelengths, absorbance (A) is determined as follows:

image

where I0 = incident light intensity, and I = emergent light intensity. For a given chromophore, A is proportional to its concentration (Beer’s law) and to the path length (Lambert’s law). From the pulsatile (AC) and background (DC) absorbance signals at both wavelengths, a ratio (R) is derived:

image

SpO2 is then computed from R, using software ‘look-up’ tables of empirically derived relationships between R-values and either SaO2 or fractional saturation (FHbO2) measured in the arterial blood of volunteers breathing hypoxic gas mixtures.

SpO2 is usually displayed as a percentage. Only two wavelengths are used, forcing the assumption that HbO2 and Hb are the only haemoglobin species in the light path. This is always incorrect, but the error is trivial with normal dyshaemoglobin concentrations. Some manufacturers calibrate R against FHbO2 rather than SaO2 (functional saturation). Because volunteers generating the data have normal dyshaemoglobin concentrations, differences between the two calibrations are small.

SPEED OF RESPONSE

SpO2 is averaged over 3–6 seconds, and updated every 0.5–1 second. With forehead probes a sudden reduction in FiO2 produces a response within 10–15 seconds, whereas with finger probes and peripheral vasoconstriction the delay can exceed 1 minute.

ACCURACY

In the 90–97% saturation range, SpO2 has a mean absolute bias of < 1%, and a precision (standard deviaition of bias) of < 3%. At SaO2 < 80% there is significant imprecision and a tendency towards negative bias. This is because very low SaO2 values are unsafe in volunteers, necessitating extrapolation from SpO2/R relationships at higher saturations.

ERROR

Causes of error are set out in Table 14.3. A falsely high SpO2 is of greatest concern.

Table 14.3 Causes of error in SpO2 readings

Factor Comment
Carboxyhaemoglobin Measured as HbO2 – SpO2 may be falsely high – see text
Methaemoglobin Absorbs both wavelengths – see text
Low saturations Progressive inaccuracy below 70–80%, usually falsely low SpO2
Prominent venous signal Dependent limb, tricuspid regurgitation (venous pulsations) – falsely low SpO2
Non-pulsatile flow Cardiopulmonary bypass – poor signal
Vasoconstriction, limb ischaemia, shock states Low pulsatile signal
Motion artefact Tremor, voluntary movement – falsely low SpO2
Ambient light Strong sunlight, fluorescent and xenon lamps, flickering light– falsely low SpO2
Anaemia No effect
Dyes Methylene blue, indocyanine green, indigo carmine – falsely low SpO2
Black skin pigmentation Variable precision and bias. May require separate calibration
Nail polish Especially blue. Falsely low SpO2. Acrylic nails do not interfere
Optical shunting Due to inadequate probe contact – falsely low SpO2
Radiofrequency interference Reported with magnetic resonance imaging scanners – falsely high SpO2

Unlike CO-oximetry, pulse oximetry is not subject to interference from bilirubin, lipid emulsions and HbF.

Dyshaemoglobins and pulse oximetry

Pulse oximeters cannot distinguish COHb from HbO2. When [COHb] is elevated, SpO2 tends to overestimate SaO2. SpO2 can thus provide false reassurance when hypoxaemia is combined with a high [COHb], for example after an inhalational burn injury.

MetHb has more complex effects, since it absorbs both wavelengths. At normal saturations, increased [MetHb] causes underestimation of SaO2, but at very low oxygen tensions overestimation is possible. At [MetHb] ≥ 35%, the R-value becomes unity, which translates to SpO2 = 85%.

IMPORTANCE OF PULSE OXIMETRY

Pulse oximeters generate accurate real-time information without calibration, within moments of sensor placement. Their use is mandatory in patient transport, and in high-acuity areas such as operating rooms, recovery rooms and intensive care units. They are also useful screening tools. On the down side, pulse oximeters are the most common source of false alarms in intensive care. They are also insensitive to changes in arterial oxygenation at higher PaO2 values (> 70–100 mmHg).

MONITORING HAEMOGLOBIN–OXYGEN AFFINITY

Haemoglobin–oxygen affinity is the relationship between the oxygen tension of blood and its oxygen content, described by the sigmoid-shaped HbO2 dissociation curve (see Figure 14.4). The P50 is the oxygen tension at SO2 = 0.5. The normal value in humans is 26.7 mmHg. Factors which decrease haemoglobin–oxygen affinity increase the P50. They include acidaemia (the Bohr effect), hypercapnia, high levels of erythrocytic 2,3-diphosphoglycerate (2,3-DPG) and fever, whereas P50 is decreased (increased affinity) by alkalaemia, hypocapnia, low 2,3-DPG levels, hypothermia, COHb, MetHb and FHb.

In the intensive care unit, it is possible to calculate accurate P50 values from a single measurement of blood gases and SaO2 up to SaO2 = 0.97.29 However the impact of haemoglobin–oxygen affinity on tissue oxygenation in critical illness appears to be small,30 making routine monitoring unnecessary.

OXYGEN DYNAMICS

Common indices of oxygen dynamics are set out in Table 14.4.

DO2/VO2 RELATIONSHIPS

More than 30 years ago an association was reported between hyperdynamic oxygen flow patterns and survival after high-risk non-cardiac surgery.31 This led to the hypothesis that an induced perioperative hyperdynamic state is protective, subsequently supported by single-unit studies,32–35 but not by larger multicentre studies.36,37 Much of any benefit may be due to fluid loading, or even increased care and attention. Typical therapeutic goals have been cardiac index (CI) > 4.5 l/min per m2, DO2I > 600 ml/min per m2, VO2I > 170 ml/min per m2. More recently more emphasis has been placed on the DO2I target, since supranormal VO2I values are much harder to achieve. It is now also clear that in sepsis, aggressive pursuit of hyperdynamic goals is counterproductive.38

MEASURING DO2I

Although DO2I determinations require accurate measurements of CI and CaO2, a PA catheter is not essential (see Chapter 12). Normal ranges can be quoted (Table 14.4), but oxygen demand in critical illness is so variable that isolated DO2I measurements are difficult to interpret.

MEASURING VO2I

The two methods of measuring VO2I are the reverse Fick method (Table 14.4) and indirect calorimetry.

Reverse Fick method

The reverse Fick method requires a PA catheter, and has large random errors, ranging from 17% overestimation to 13% underestimation. Changes cannot be detected reliably unless they exceed 20%. The error is increased by lung inflammation, when up to 20% of VO2I can arise from the lungs alone.

Indirect calorimetry

Indirect calorimetry has better accuracy. VO2I is determined from the volumes and oxygen concentrations of inspired and expired gas. However, high FiO2 settings introduce error. Newer devices retain their accuracy up to FiO2 = 0.8, with relative errors of < 5%.39

MIXED VENOUS BLOOD

Mixed venous sampling is by gentle aspiration of blood from the distal port of an unwedged PA catheter. This ensures complete admixture of blood from superior and inferior venae cavae and coronary sinus. Mixed venous O2 and CO2 tensions and content are flow-weighted averages of the venous effluents of multiple tissues. The integrating process can conceal pockets of hypoxia and hypercapnia.

MIXED VENOUS PO2 (PvO2)

Venous gas tensions reflect postcapillary and tissue gas tensions. At a PvO2 of 26 mmHg, the average intracellular PO2 has fallen from 11 to 0.8 mmHg.40 A PvO2 below this value is highly suggestive of intracellular hypoxia. However, a normal or high PvO2 does not exclude regional tissue hypoxia, whether cytopathic hypoxia5 or hypoxia due to tissue shunting.41

MIXED VENOUS OXYGEN SATURATION (Svo2)

SvO2 is measured either intermittently by CO-oximetry on mixed venous samples or continuously by fibreoptic reflectance oximetry using a modified PA catheter. SvO2 measurements have a number of potential uses:

1 To calculate CvO2 (Table 14.4). CvO2 can then be used to determine Qs/Qt, VO2I by the reverse Fick method, the oxygen extraction ratio (Table 14.4) and cardiac output by the Fick method.
2 As an indirect index of tissue hypoxia. A SvO2 value of 0.5 corresponds to the theoretical critical PvO2 of 26 mmHg. Values between 0.7 and 0.8 represent a desirable balance between global oxygen supply and demand (Table 14.4), with lactic acidosis appearing between 0.3 and 0.5.42 Values exceeding 0.8 can be seen in high-flow states such as sepsis, hyperthyroidism and severe liver disease.

SvO2 as a therapeutic target (> 0.7) failed to improve survival in a multicentre trial.43 Only two-thirds of the treatment group achieved the SvO2 target. Like PvO2, SvO2 is insensitive to cytopathic hypoxia and tissue shunting. In chronic heart failure, low values can be surprisingly well tolerated.

CENTRAL VENOUS SATURATION (ScvO2)43–45

As with SvO2, ScvO2 can be measured either continuously using a central venous catheter modified for reflectance oximetry, or by intermittent sampling and CO-oximetry. ScvO2 is normally 2–3% lower than SvO2. However, in shock this difference can be reversed. Trends in SvO2 and ScvO2 in response to management usually run in parallel.

In an influential single-centre study of the early hypodynamic phase of severe sepsis and septic shock, resuscitation guided by a ScvO2 target of > 0.7, as well as by central venous pressure and mean arterial pressure values, appeared to reduce 28-day and 60-day mortality, and duration of hospitalization.45 It remains to be seen whether this success can be replicated in larger trials, and if so whether intermittent sampling is of equivalent benefit. Nevertheless, ScvO2 currently has equal billing with SvO2 in the management guidelines of the Surviving Sepsis Campaign.46

VENOARTERIAL PCO2 GRADIENT (ΔPCO2)

imagePCO2 (normally about 6 mmHg) is markedly increased during cardiac arrest and in experimental low-output states, but lacks sensitivity and specificity as a global index of tissue hypoxia. A sudden increase in the respiratory quotient (VCO2/VO2), or in the venoarterial CO2 tension difference/arteriovenous O2 content difference47 may be more reliable markers of the onset of anaerobic metabolism.

PLASMA LACTATE AND REDOX INDICES

See Chapter 15.

REGIONAL OXYGENATION INDICES

REGIONAL PCO248

Regional PCO2 reflects the balance between arterial blood CO2 content, tissue blood flow and tissue CO2 production. The CO2 gap, which is regional PCO2 – PaCO2, was devised to correct regional PCO2 for varying arterial CO2 content. As tissue blood flow falls, reduced CO2 clearance causes the CO2 gap to increase. With the onset of anaerobic metabolism, tissue CO2 production steadily decreases, although worsening regional metabolic acidosis generates some CO2 by proton titration of tissue and capillary image. A rising CO2 gap merely signals falling tissue blood flow. It cannot identify the onset of anaerobic metabolism.

GASTRIC TONOMETRY49

The impetus to develop regional gastric capnometry was the knowledge that splanchnic hypoperfusion occurs early in circulatory shock, tends to persist as ‘covert shock’, and is manifested by intramucosal hypercapnia and acidosis. A gastric tonometer was therefore developed as a modified nasogastric tube with a silicone balloon 11.4 cm from the tip. Gastric mucosal CO2 equilibrates with luminal CO2, which equilibrates with (and is measured in) fluid filling the balloon. Early on, this fluid was saline. Subsequently air was found to have more rapid equilibration characteristics, and automated cycling of air through an infrared CO2 analyser was far more efficient than measuring saline PCO2 intermittently and applying time-based correction factors.

The original concept involved calculation of intramucosal pH (pHi) via the Henderson–Hasselbalch equation, using gastric luminal PCO2 and arterial image. Intramucosal acidosis was defined as pHi < 7.3, and taken to indicate inadequate splanchnic perfusion. Changing the monitoring end-point from pHi to the mucosal–arterial CO2 gap (normally 8–10 mmHg) removed the most fundamental flaw – the use of arterial image as a surrogate for mucosal image. Automated air tonometry combined with simultaneous end-tidal CO2 measurement allows the regular calculation of a gastric to end-tidal CO2 gap, a parameter linked to outcome in high-risk surgery.50

Even with these improvements, the technique has not found widespread application, and is unlikely to do so for several reasons:

1 the need for a nasogastric tube
2 signal degradation by luminal contents, including feeds and blood. Feeds must be stopped 2 hours prior to measurements, jeopardising nutritional support
3 inability to identify a clear hypoxic threshold. Empiric recommendations have been to target a CO2 gap < 25 mmHg
4 a need to suppress gastric acidity to prevent luminal CO2 generation by HCl titration of duodenal bicarbonate
5 a lack of convincing evidence that tonometry-guided therapy improves outcome51

SUBLINGUAL CAPNOMETRY49,50

Sublingual capnometry measures the interstitial CO2 tension under the tongue, as its name suggests, using either electrode or optode technologies. Although the tongue vasculature is not part of the splanchnic circulation, it responds similarly to circulatory disturbances. It has the added advantage of accessibility with minimal invasion. Optode sensors in particular provide rapid-response PCO2 signals. It is also possible to perform simultaneous inspections of the microcirculation using orthogonal polarisation spectral imaging.

At this stage, all that can be said is that sublingual PCO2 seems to correlate with the severity of haemodynamic disturbances and with gastric mucosal PCO2, and that in a limited number of shocked patients the sublingual CO2 gap was correlated with survival. It remains to be seen whether the technique can be used to guide therapy.

CEREBROVENOUS OXYGEN SATURATION MONITORING

See Chapter 67.

DIRECT TISSUE PO2 MEASUREMENT

In patients this is largely impractical. Measurements have been recorded primarily in animal models in the brain, subcutaneous tissue, muscle and renal beds under a variety of perfusion insults.52 Tensions are commonly around 30–45 mmHg, but can range from < 10 mmHg in the renal medulla to > 70 mmHg in subcutaneous tissue.

OTHER REGIONAL TECHNIQUES

Orthogonal polarisation spectroscopy (OPS) allows real-time in vivo imaging of microcirculatory blood flow.42 Tissue beds visualised in intensive care have included the sublingual, rectal, oral and ileal (via stoma) microcirculations. OPS can be combined with near-infrared spectroscopy to monitor deeper regional mitochondrial redox status, and with reflectance spectrophotometry for measurement of superficial microcirculatory oxygen saturation. Combinations of this type provide uniquely integrated information on oxygen transport distribution during sepsis and septic shock.

Other techniques showing promise include in vivo magnetic resonance imaging53 and optical spectroscopy.54

REFERENCES

1 Schumacker PT. Cell metabolism and tissue hypoxia. In: Albert RK, Slutsky A, Ranieri M, et al, editors. Clinical Critical Care Medicine. Philadelphia: Mosby Elsevier; 2006:41-50.

2 Bayir H. Reactive oxygen species. Crit Care Med. 2005;33(Suppl.):S498-S501.

3 Schumacker PT. Hypoxia-inducible factor-1 (HIF-1). Crit Care Med. 2005;33(Suppl.):S423-S425.

4 Connett RJ, Honig CR, Gayeski TE, et al. Defining hypoxia: a systems view of VO2, glycolysis, energetics, and intracellular PO2. J Appl Physiol. 1990;68:833-842.

5 Fink MP. Bench to bedside review: cytopathic hypoxia. Crit Care. 2002;6:491-499.

6 Deby-Dupont G, Deby C, Lamy M. Oxygen therapy in intensive care patients: a vital poison? In: Vincent J-L, editor. Yearbook of Intensive Care and Emergency Medicine. Berlin: Springer-Verlag; 1999:417-432.

7 Terragni PP, Rosboch G, Tealdi A, et al. Tidal hyperinflation during low tidal volume ventilation in acute respiratory distress syndrome. Am J Respir Crit Care Med. 2007;175:160-166.

8 Dunlop S, Hough J, Riedel T, et al. Electrical impedance tomography in extremely prematurely born infants and during high frequency oscillatory ventilation analyzed in the frequency domain. Physiol Meas. 2006;27:1151-1165.

9 Wolf GK, Arnold JH. Noninvasive assessment of lung volume: respiratory inductance plethysmography and electrical impedance tomography. Crit Care Med. 2005;33(Suppl.):S163-S169.

10 West JB. Ventilation–perfusion relationships. Am Rev Respir Dis. 1977;116:919-943.

11 Yu G, Yang K, Baker AB, et al. The effect of bi-level positive airway pressure mechanical ventilation on gas exchange during general anaesthesia. Br J Anaesth. 2006;96:522-532.

12 Kjaergaard S, Rees S, Malczynski J, et al. Non-invasive estimation of shunt and ventilation–perfusion mismatch. Intens Care Med. 2003;29:727-734.

13 Rees SE, Kjaergaard S, Andreassen S, et al. Reproduction of MIGET retention and excretion data using a simple mathematical model of gas exchange in lung damage caused by oleic acid infusion. J Appl Physiol. 2006;101:826-832.

14 Kjaergaard S, Rees SE, Gronlund J, et al. Hypoxaemia after cardiac surgery: clinical application of a model of pulmonary gas exchange. Eur J Anaesthesiol. 2004;21:296-301.

15 D’Alonzo GE, Dantzker DR. Respiratory failure, mechanisms of abnormal gas exchange, and oxygen delivery. Med Clin North Am. 1983;67:557-571.

16 Rodriguez-Roisin R, Roca J. Mechanisms of hypoxaemia. Intens Care Med. 2005;31:1017-1019.

17 Nirmalan M, Willard T, Columb MO, et al. Effect of changes in arterial-mixed venous oxygen content difference (C(a-v)O2) on indices of pulmonary oxygen transfer in a model ARDS lung. Br J Anaesth. 2001;86:477-485.

18 Bernard GR, Artigas A, Brigham KL, et al. Report of the American–European consensus conference on ARDS: definitions, mechanisms, relevant outcomes and clinical trial coordination. Intens Care Med. 1994;20:225-232.

19 Le Gall J-R, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicentre study. JAMA. 1993;270:2957-2963.

20 Murray JF, Mathay MA, Luce JM, et al. An expanded definition of the adult respiratory distress syndrome. Am Rev Respir Dis. 1988;138:720-723.

21 Whiteley JP, Gavaghan DJ, Hahn CE. Variation of venous admixture, SF6 shunt, PaO2, and the PaO2/FiO2 ratio with FiO2. Br J Anaesth. 2002;88:771-778.

22 Siggaard-Andersen O, Siggaard-Andersen M. The oxygen status algorithm: a computer program for calculating and displaying pH and blood gas data. Scand J Clin Lab Invest. 1990;50(Suppl. 203):29-45.

23 Nirmalan M, Willard T, Khan A, et al. Changes in arterial–mixed venous oxygen content difference (CaO2– CvO2) and the effect on shunt calculations in critically ill patients. Br J Anaesth. 1998;80:829-831.

24 Venkatesh B, Hendry S-P. Continuous intra-arterial blood gas monitoring. Intens Care Med. 1996;22:818-828.

25 Widness JA, Madan A, Grindeanu LA, et al. Reduction in red blood cell transfusions among preterm infants: results of a randomized trial with an in-line blood gas and chemistry monitor. Pediatrics. 2005;115:1299-1306.

26 Rudiger M, Topfer K, Hammer H, et al. A survey of transcutaneous blood gas monitoring among European neonatal intensive care units. BMC Pediatr. 2005;5:30.

27 Jubran A. Pulse oximetry. In: Tobin MJ, editor. Principles and Practice of Intensive Care Monitoring. New York: McGraw Hill; 1998:261-287.

28 McMorrow RCN, Mythen MG. Pulse oximetry. Curr Opin Crit Care. 2006;12:269-271.

29 Morgan TJ, Koch D, Morris D, et al. Red cell 2,3-diphosphoglycerate concentrations are reduced in critical illness without net effect on in vivo P50. Anaesth Intens Care. 2001;29:479-483.

30 Gutierrez G. The unpretentious role of 2,3-diphosphoglycerate in critical illness. Crit Care Med. 2005;33:2423-2424.

31 Shoemaker WC, Montgomery ES, Kaplan E, et al. Physiologic patterns in surviving and nonsurviving shock patients. Arch Surg. 1973;106:630-636.

32 Boyd O, Grounds M, Bennett D. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA. 1993;270:2699-2707.

33 Wilson J, Woods I, Fawcett J, et al. Reducing the risk of major elective surgery: randomised controlled trial of preoperative optimisation of oxygen delivery. Br Med J. 1999;318:1099-1103.

34 Lobo SM, Salgoda PF, Castillo VGT, et al. Effects of maximizing oxygen delivery on morbidity and mortality in high-risk surgical patients. Crit Care Med. 2000;28:3396-3404.

35 Pearse R, Dawson D, Fawcett J, et al. Early goal-directed therapy after major surgery reduces complications and duration of hospital stay. A randomised, controlled trial. Crit Care. 2005;9:687-693.

36 Takala J, Meier-Hellmann A, Eddleston J, et al. Effect of dopexamine on outcome after major abdominal surgery: a prospective, randomized, controlled multicenter study. European Multicenter Study Group on Dopexamine in Major Abdominal Surgery. Crit Care Med. 2000;28:3417-3423.

37 Sandham JD, Hull RD, Brant RF, et alCanadian Critical Care Clinical Trials Group. A randomized, controlled trial of the use of pulmonary-artery catheters in high-risk surgical patients. N Engl J Med. 2003;348:5-14.

38 Gattinoni L, Brazzi L, Pelosi P. Does cardiovascular optimization reduce mortality? In: Vincent J-L, editor. Yearbook of Intensive Care and Emergency Medicine. Berlin: Springer-Verlag; 1996:308-318.

39 Walsh TS, Monaco F. Gas exchange measurement in the ICU. In: Vincent J-L, editor. Yearbook of Intensive Care and Emergency Medicine. Berlin: Springer-Verlag; 2005:632-643.

40 Siggaard-Andersen O, Fogh-Andersen N, Gøthgen IH, et al. Oxygen status of arterial and mixed venous blood. Crit Care Med. 1995;23:1284-1293.

41 Elbers PW, Ince C. Bench-to-bedside review: mechanisms of critical illness – classifying microcirculatory flow abnormalities in distributive shock. Crit Care. 2006;10:221.

42 Marx G, Reinhart K. Venous oximetry. Curr Opin Crit Care. 2006;12:263-268.

43 Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-orientated hemodynamic therapy in critically ill patients. N Engl J Med. 1995;333:1025-1032.

44 Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368-1377.

45 Pearse R, Dawson D, Fawcett J, et al. Changes in central venous saturation after major surgery, and association with outcome. Crit Care. 2005;9:R694-R699.

46 Dellinger RP, Carlet JM, Masur H, et al. Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock. Crit Care Med. 2004;32:858-873.

47 Mekontso-Dessap A, Castelain V, Anguel N, et al. Combination of venoarterial PCO2 difference with arteriovenous O2 content difference to detect anaerobic metabolism in patients. Intens Care Med. 2002;28:272-277.

48 Marik PE. Regional carbon dioxide monitoring to assess the adequacy of tissue perfusion. Curr Opin Crit Care. 2005;11:245-251.

49 Creteur J. Gastric and sublingual capnometry. Curr Opin Crit Care. 2006;12:272-277.

50 Lebuffe G, Vallet B, Takala J, et al. A European, multicenter, observational study to assess the value of gastric-to-end tidal PCO2 difference in predicting postoperative complications. Anesth Analg. 2004;99:166-172.

51 Miami Trauma Clinical Trials Group. Splanchnic hypoperfusion-directed therapies in trauma: a prospective, randomized trial. Am Surg. 2005;71:252-260.

52 Venkatesh B, Morgan TJ. Monitoring tissue gas tensions in critical illness. In: Vincent J-L, editor. Yearbook of Intensive Care and Emergency Medicine. Berlin: Springer-Verlag; 2001:251-265.

53 Zaharchuk G, Busse RF, Rosenthal G, et al. Noninvasive oxygen partial pressure measurement of human body fluids in vivo using magnetic resonance imaging. Acad Radiol. 2006;13:1016-1024.

54 Mayevsky A, Rogatsky G. Mitochondrial function in vivo evaluated by NADH fluorescence: From animal models to human studies. Am J Physiol Cell Physiol. 2007;292:C615-C640.

Related posts:

Physiotherapy in intensive care Default ThumbnailAbdominal and pelvic injuries Equipment for paediatric intensive care Haemodynamic monitoring Imaging the chest Default ThumbnailStatus epilepticus