Non-invasive monitoring

1. Skin electrodes detect the electrical activity of the heart (Fig. 10.5). Silver and silver chloride form a stable electrode combination. Both are held in a cup and separated from the skin by a foam pad soaked in conducting gel.

2. Colour-coded cables to transmit the signal from electrodes to the monitor. Cables are available in 3- and 5-lead versions as snap or grabber design and with a variety of lengths. All the cables of a particular set should have the same length to minimize the effect of electromagnetic interference.

3. The ECG signal is then boosted using an amplifier. The amplifier covers a frequency range of 0.05–150 Hz. It also filters out some of the frequencies considered to be noise. The amplifier has ECG filters that are used to remove the noise/artifacts from ECG and produce a ‘clean’ signal.

4. An oscilloscope that displays the amplified ECG signal. A high-resolution monochrome or colour monitor is used.

1. Proper attachment of ECG electrodes involves cleaning the skin, gently abrading the stratum corneum and ensuring adequate contact using conductive gel. Skin impedance varies at different sites and it is thought to be higher in females. The electrodes are best positioned on bony prominences to reduce artifacts from respiration.

2. Modern ECG monitors use multiple filters for signal processing. The filters used should be capable of removing the unwanted frequencies, leaving the signal intact (Fig. 10.6). Two types of filters are used for this purpose:

3. The ECG monitor can have two modes:

4. There are many ECG electrode configurations. Usually during anaesthesia, three skin electrodes are used (right arm, left arm and indifferent leads). The three limb leads used include two that are ‘active’ and one that is ‘inactive’ (earth). Sometimes five electrodes are used. Lead II is ideal for detecting arryhthmias. CM5 configuration is able to detect 89% of ST-segment changes due to left ventricular ischaemia. In CM5, the right arm electrode is positioned on the manubrium (chest lead from manubrium), the left arm electrode is on V5 position (fifth interspace in the left anterior axillary line) and the indifferent lead is on the left shoulder or any convenient position (Fig. 10.7).

5. The CB5 configuration is useful during thoracic anaesthesia. The right arm electrode is positioned over the centre of the right scapula and the left arm electrode is over V5.

6. A display speed of 25 mm/s and a sensitivity of 1 mV/cm are standard in the UK.

1. Incorrect placement of the ECG electrodes in relation to the heart is a common error, leading to false information.

2. Electrical interference can be a 50-Hz (in UK) mains line interference because of capacitance or inductive coupling effect. Any electrical device powered by AC can act as one plate of a capacitor and the patient acts as the other plate. Interference can also be because of high-frequency current interference from diathermy. Most modern monitors have the facilities to avoid interference. Shielding of cables and leads, differential amplifiers and electronic filters all help to produce an interference-free monitoring system. Differential amplifiers measure the difference between the potential from two different sources. If there is interference common to the two input terminals (e.g. mains frequency), it can be eliminated as only the differences between the two terminals is amplified. This is called common mode rejection ratio (CMRR). Amplifiers used in ECG monitoring should have a high CMRR of 100 000 : 1 to 1 000 000 : 1, which is a measurement of capability to reject the noise. They should also have a high input impedance (about 10 MΩ) to minimize the current taken from the electrodes. Table 10.1 shows the various types and sources of interference and how to reduce the interference.

3. Muscular activity, such as shivering, can produce artifacts. Positioning the electrodes over bony prominences and the use of low-pass filters can reduce these artifacts.

4. High and low ventricular rate alarms and an audible indicator of ventricular rate are standard on most designs. More advanced monitors have the facility to monitor the ST segment (Fig. 10.8). Continuous monitoring and measurement of the height of the ST segment allows early diagnosis of ischaemic changes.

5. Absence of or improperly positioned patient diathermy plate can cause burns at the site of ECG skin electrodes. This is because of the passage of the diathermy current via the electrodes causing a relatively high current density.

1. A cuff with a tube used for inflation and deflation. Some designs have an extra tube for transmitting pressure fluctuations to the pressure transducer.

2. The case where the microprocessor, pressure transducer and a solenoid valve which controls the deflation of the arm cuff are housed. It contains the display and a timing mechanism which adjusts the frequency of measurements. Alarm limits can be set for both high and low values.

1. The microprocessor is set to control the sequence of inflation and deflation.

2. The cuff is inflated to a pressure above the previous systolic pressure, then it is deflated incrementally. The return of blood flow causes oscillation in cuff pressure (Fig. 10.9).

3. The transducer senses the pressure changes which are interpreted by the microprocessor. This transducer has an accuracy of ±2%.

4. The output signal from the transducer passes through a filter to an amplifier that amplifies the oscillations. The output from the amplifier passes to the microprocessor through the analogue digital converter (ADC). The microprocessor controls the pneumatic pump for inflation of the cuff and the solenoid valve for deflation of the cuff.

5. The mean arterial blood pressure corresponds to the maximum oscillation at the lowest cuff pressure. The systolic pressure corresponds to the onset of rapidly increasing oscillations.

6. The diastolic pressure corresponds to the onset of rapidly decreasing oscillations. In addition, it is mathematically computed from the systolic and mean pressure values (mean blood pressure = diastolic blood pressure + 1/3 pulse pressure).

7. The cuff must be of the correct size (Table 10.2). It should cover at least two-thirds of the upper arm. The width of the cuff’s bladder should be 40% of the mid-circumference of the limb. The middle of the cuff’s bladder should be positioned over the brachial artery.

8. Some designs have the ability to apply venous stasis to facilitate intravenous cannulation.

1. For the device to measure the arterial blood pressure accurately, it should have a fast cuff inflation and a slow cuff deflation (at a rate of 3 mmHg/s or 2 mmHg/beat). The former is to avoid venous congestion and the latter provides enough time to detect the arterial pulsation.

2. If the cuff is too small, the blood pressure is over-read, while it is under-read if the cuff is too large. The error is greater with too small than too large a cuff.

3. The systolic pressure is over-read at low pressures (systolic pressure less than 60 mmHg) and under-read at high systolic pressures.

4. Atrial fibrillation and other arrhythmias affect performance.

5. External pressure on the cuff or its tubing can cause inaccuracies.

6. Frequently repeated cuff inflations can cause ulnar nerve palsy and petechial haemorrhage of the skin under the cuff.

1. Two cuffs: the upper, occluding cuff (5 cm wide) overlaps a lower, sensing cuff (10 cm wide). An inflation bulb is attached.

2. The case which contains:

1. With the control lever at rest, air is pumped into both cuffs and the air-tight case of the instrument using the inflation bulb to a pressure exceeding systolic arterial pressure. By operating the control lever, the lower sensing cuff is isolated and the pressure in the upper cuff and instrument case is allowed to decrease slowly through an adjustable leak controlled by the release valve. As systolic pressure is reached, pulsation of the artery under the lower cuff results in pressure oscillations within the cuff and its bellows. The pressure oscillations are transmitted via a mechanical amplification system to the needle. As the pressure in the upper cuff decreases below diastolic pressure, the pulsation ceases.

2. The mean pressure is at the point of maximum oscillation.

3. This method is reliable at low pressures. It is useful to measure trends in blood pressure.

1. In order for the device to operate accurately, the cuffs must be correctly positioned and attached to their respective tubes.

2. The diastolic pressure is not measured accurately with this device.

1. A probe is positioned on the finger, toe, ear lobe or nose (Fig. 10.12). Two light-emitting diodes (LEDs) produce beams at red and infrared frequencies (660 nm and 940 nm respectively) on one side and there is a sensitive photodetector on the other side. The LEDs operate in sequence at a rate of about 30 times per second (Fig. 10.13).

2. The case houses the microprocessor. There is a display of the oxygen saturation, pulse rate and a plethysmographic waveform of the pulse. Alarm limits can be set for a low saturation value and for both high and low pulse rates.

1. The oxygen saturation is estimated by measuring the transmission of light, through a pulsatile vascular tissue bed (e.g. finger). This is based on Beer’s law (the relation between the light absorbed and the concentration of solute in the solution) and Lambert’s law (relation between absorption of light and the thickness of the absorbing layer).

2. The amount of light transmitted depends on many factors. The light absorbed by non-pulsatile tissues (e.g. skin, soft tissues, bone and venous blood) is constant (DC). The non-constant absorption (AC) is the result of arterial blood pulsations (Fig. 10.14). The sensitive photodetector generates a voltage proportional to the transmitted light. The AC component of the wave is about 1–5% of the total signal.

3. The high frequency of the LEDs allows the absorption to be sampled many times during each pulse beat. This is used to enable running averages of saturation to be calculated many times per second. This decreases the ‘noise’ (e.g. movement) effect on the signal.

4. The microprocessor is programmed to mathematically analyse both the DC and AC components at 660 and 940 nm calculating the ratio of absorption at these two frequencies (R/IR ratio). The result is related to the arterial saturation. The absorption of oxyhaemoglobin and deoxyhaemoglobin at these two wavelengths is very different. This allows these two wavelengths to provide good sensitivity. 805 nm is one of the isobestic points of oxyhaemoglobin and deoxyhaemoglobin. The OFF part allows a baseline measurement for any changes in ambient light.

5. A more recent design uses multiple wavelengths to eradicate false readings from carboxy haemoglobin and methaemoglobinaemia. Advanced oximeters use more than seven light wavelengths. This has enabled the measurement of haemoglobin value, oxygen content, carboxyhaemoglobin and methaemoglobin concentrations.

6. A variable pitch beep provides an audible signal of changes in saturation.

1. It is accurate (±2%) in the 70–100% range. Below the saturation of 70%, readings are extrapolated.

2. The absolute measurement of oxygen saturation may vary from one probe to another but with accurate trends. This is due to the variability of the centre wavelength of the LEDs.

3. Carbon monoxide poisoning (including smoking), coloured nail varnish, intravenous injections of certain dyes (e.g. methylene blue, indocyanine green) and drugs responsible for the production of methaemoglobinaemia are all sources of error (Table 10.3).

4. Hypoperfusion and severe peripheral vasoconstriction affect the performance of the pulse oximeter. This is because the AC signal sensed is about 1–5% of the DC signal when the pulse volume is normal. This makes it less accurate during vasoconstriction when the AC component is reduced.

5. The device monitors the oxygen saturation with no direct information regarding oxygen delivery to the tissues.

6. Pulse oximeters average their readings every 10–20 s. They cannot detect acute desaturation. The response time to desaturation is longer with the finger probe (more than 60 s) whereas the ear probe has a response time of 10–15 s.

7. Excessive movement or malposition of the probe is a source of error. Newer designs such as the Masimo oximeter claim more stability despite motion. External fluorescent light can be a source of interference.

8. Inaccurate measurement can be caused by venous pulsation. This can be because of high airway pressures, the Valsalva manoeuvre or other consequences of impaired venous return. Pulse oximeters assume that any pulsatile absorption is caused by arterial blood pulsation only.

9. The site of the application should be checked at regular intervals as the probe can cause pressure sores with continuous use. Some manufacturers recommend changing the site of application every 2 h especially in patients with impaired microcirculation. Burns in infants have been reported.

10. Pulse oximetry only gives information about a patient’s oxygenation. It does not give any indication of a patient’s ability to eliminate carbon dioxide.

1. The sampling chamber can either be positioned within the patient’s gas stream (main-stream version, Fig. 10.16) or connected to the distal end of the breathing system via a sampling tube (side-stream version, Fig. 10.17).

2. A photodetector measures light reaching it from a light source at the correct infrared wavelength (using optical filters) after passing through two chambers. One acts as a reference whereas the other one is the sampling chamber (Fig. 10.18).

1. Carbon dioxide absorbs the infrared radiation particularly at a wavelength of 4.3 µm.

2. The amount of infrared radiation absorbed is proportional to the number of carbon dioxide molecules (partial pressure of carbon dioxide) present in the chamber.

3. The remaining infrared radiation falls on the thermopile detector, which in turn produces heat. The heat is measured by a temperature sensor and is proportional to the partial pressure of carbon dioxide gas present in the mixture in the sample chamber. This produces an electrical output. This means that the amount of gas present is inversely proportional to the amount of infrared light present at the detector in the sample chamber (Fig 10.19).

4. In the same way, a beam of light passes through the reference chamber which contains room air. The absorption detected from the sample chamber is compared to that in the reference chamber. This allows the calculation of carbon dioxide values.

5. The inspired and exhaled carbon dioxide forms a square wave, with a zero baseline unless there is rebreathing (Fig. 10.20A).

6. A microprocessor-controlled infrared lamp is used. This produces a stable infrared source with a constant output. The current is measured with a current-sensing resistor, the voltage across which is proportional to the current flowing through it. The supply to the light source is controlled by the feedback from the sensing resistor maintaining a constant current of 150 mA.

7. Using the rise and fall of the carbon dioxide during the respiratory cycle, monitors are designed to measure the respiratory rate.

8. Alarm limits can be set for both high and low values.

9. To avoid drift, the monitor should be calibrated regularly with known concentrations of CO₂ to ensure accurate measurement.

1. The photo-acoustic technique is extremely stable and its calibration remains constant over much longer periods of time.

2. The very fast rise and fall times give a much more accurate representation of any change in CO₂ concentration.