|Home | Articles | Forum | Glossary | Books|
The purpose of a Physiological Instrument is to provide information to a Physician as to the function and performance of an organ, group of organs or system within the body of a patient.
The performance may be a comparison of the function now to the function at some previous time or to the function of some other patient. The measurement may be specific to the individual or compared to a value defined for normal or abnormal populations.
Measurement -Signal Characteristics--Measurement Principle
Heart Rate Blood Pressure Arterial ECG Electrocardiogram
EMG Electromyogram EEG Electroencephalogram
[ max = 300 bpm min = 25 bpm normal range = 69-90 bpm min = 20 diastolic max = 300 systolic normal range = 120/80 bandwidth = 200 Hz Voltage 10 pV to 5 mV normal peak reading 1 mV bandwidth = 1 kHz max Voltage 20 pV to 500 pV bandwidth = 2 kHz max Voltage 2 pV to 200 pV bandwidth = 100 Hz max
[Measurement derived from blood pressure measurement, pulse oximeter, ECG measurement or from direct heart sounds Measured with sphygmomanometer and stethoscope or strain gauge transducers alternatively invasively measured by catheter tube and pressure transducer Measured by electrodes, placed on the arms and legs or from chest electrodes Measured by electrodes placed close to the muscle group under examination or by needle electrodes implanted into that muscle Measured by electrodes placed on the scalp above the region of the brain under investigation]
Isolation Front End Circuits
The measurement may concern an individual organ such as the heart or the collective function of a group of organs all contributing to a particular effect.
FIG. 1 shows a table of the typical size and type of signals
How can measurements be taken?
Internal measurements (blood pressure by transducers introduced into the arteries).
Surface measurements (electrical activity of the heart measured by potentials on the skin's surface).
At a distance using signals which are emitted from the body (infra red, including sensing of transmission of radiation through the body).
What can be measured?
Pressure, flow, velocity, force, acceleration, impedance, temperature, chemical concentration, gas concentrations.
2. Measurement Systems
The following sections are a brief introduction to the application of electronic measurement systems used in medical applications. This material is provided here only as a brief summary: the interested reader is referred to dedicated texts on these systems for a more thorough insight into current practice.
A transducer is the functional element that converts one form of energy to another. A physiological measurement instrument may consist of a multitude of transducers and electronic circuits to process their signals. Transducers convert a parameter from one form of energy to another form which is amenable to measurement. For example pressure measurement transducers consist of a membrane which converts a pressure difference into movement or strain, which in turn is transformed into a varying resistance by a strain gauge. The resultant varying resistance can be incorporated in a circuit to enable measurement of relative pressure. There is a vast and ever increasing number of transducers available to the engineer: many employ semi conductor technology. Examples of transducers are: piezo electric crystal, strain gauge, and photo diode.
2.2. Front End Circuits
Signals originating from transducers often require conditioning before they can be analyzed.
Many transducers require excitation, as is the case with strain gauges, or incorporation within a circuit so that their characteristics may be analyzed. The function of front end circuits then is to provide excitation of the primary transducer and to condition the detected signal. Signal conditioning may include pre-amplification and filtering.
Modern physiological instrumentation is designed to rigorous safety standards to provide patient protection from the instrument or other inter connected instruments. Therefore there is a requirement to isolate the transducer and front end circuits from the rest of the equipment, to reduce the possibility of dangerous currents and voltages coming into contact with the patient.
The isolation also serves to protect the patient from instrument faults. Isolation is provided as close to the patient as possible so that most circuits are separated: this reduces the demands on the power supply for the isolated section and the complexity of the isolation.
2.4. Intermediate Circuits
Intermediate circuits provide signal conditioning, filtering, amplification and analysis. For example the return signal from a Doppler blood flow detector is amplified in front end circuits, further amplified and filtered in the intermediate circuits. The signal is then modulated and the frequency deviation detected using either a zero crossing detector or phase locked loop. Thus the intermediate circuits provide signal conditioning, transformation and detection.
In modern medical equipment signal transformation and detection is often done digitally.
Increasingly the detected signal is intensively processed to extract the information which represents the required measurement.
2.5. Backend Circuits
Backend circuits provide analysis, display and interpretation. Traditionally the most common form of display was a paper chart recorder, although this has largely been replaced by Cathode Ray Tube (CRT) displays. However, most equipment manufactured within the last decade has incorporated computer generated displays and there is now a move from CRTs to Liquid Crystal Display (LCD) technology.
2.6. Future Developments
The future may see computerized physiological measurement instrumentation developed that is part of an 'Expert System'. That is a system which mimics the decisions and experience of a skilled physician, which may well 'make decisions' to alter treatment and implement them.
Section 7 provides a further description of their application in medicine.
Elsewhere in this section, we quantify the signal levels seen when measuring biopotentials such as in either Electrocardiography (ECG) or Electroencephalography (EEG) measurement systems. The signals involved were between a few V and several mV. The impedance required of the measurement system had to be sufficiently high as not to degrade the signals of interest.
The bandwidth of the signals in which we were interested was, however, limited.
The system gain required in order to drive either a CRT display or the motors used in a paper recorder must generate several volts from a relatively low impedance source. Gains of between 10^3 and 10^6 are therefore required. More exacting is the source impedance requirement which must be coupled with the need to provide safe isolation of the signals. The input impedance of the amplifiers may need to be between 10^6 and 10^9 ohms. This level of impedance coupled with the need to measure very low bandwidth signals leads to the use of either MOSFET or chopper stabilized amplifiers.
The main problem associated with high gain directly coupled amplifiers is that they are inherently liable to amplify equally both the required signal and any instabilities that may be present in their first stage. The instabilities are primarily due to the effects of thermal drift of their first stage components. As an alternative to DC coupling, chopper stabilized amplifiers are used in which the signal to be amplified is sampled at the chopper's frequency, and the sampled frequency is itself multiplied by an AC coupled amplifier which does not feed forward the drift from each stage of the amplifier. An outline of the chopping process is shown in FIG. 3 below.
The signals obtained from biopotential measurements are frequently very specific in their form and of limited bandwidth. Whilst the DC component itself carries no information and is of little interest, we are often interested in low frequency signals. Additionally, the source impedance, as we noted in the last section is often high and the leads connecting small signals between the patient and measurement apparatus are necessarily long. They are therefore susceptible to both capacitive and magnetic pickup of electrical noise. A major source of electrical noise is often the power supply at a frequency of 50 or 60 Hz. There are also noise sources operating at radio frequency, such as electro surgical apparatus, communication equipment and radiated noise from switching equipment including power supplies.
Filters used in modern circuits almost always rely on the use of operational amplifiers as active components. The use of amplifiers obviates the need for the use of inductors, so that accurate filtering characteristics may be obtained with RC circuits which do not suffer from manufacturing difficulties. They also typically have low output impedances from each filtering stage which permits their use in tandem without coupling between separate filtering elements.
Unwanted signals may be removed from the required information to a large extent by selectively amplifying only the frequencies in which we are interested. Filters, however, vary in their types and complexity. Simple filters remove either high frequencies, low frequencies or pass only a range of frequencies, as illustrated in FIG. 4: they are termed low pass, high pass and bandpass filters respectively.
The slope of a filter's characteristic attenuation as a function of frequency is controlled by the order of the filtering process. High order filters, whilst they provide sharp cut off, suffer from uneven transmission characteristics in their pass band.
Filters may either be implemented as analogue circuits or they can be built using digital components which analyze the frequency components of a digitized signal directly in the frequency domain and produce the desired cut off characteristic. The output of a digital filter may require to be reconstructed into its analogue form for later display.
2.9. Differentiation and Integration
Certain signals may require to be interpreted by either differentiation to examine their rate of change or be integrated, for instance to reduce certain noise characteristics. Both of these functions may be undertaken by analogue electronic circuits.
The integration process uses a circuit of the form shown in FIG. 5. In this circuit the integrating element is effectively the capacitor. The capacitor is charged by the current which is fed through the resistor. The amplifier's inverting input is a virtual earth, so the current flowing through the resistor in this configuration is independent of the voltage across the capacitor. The errors in this circuit are due to several factors. Firstly, the amplifier has a finite bias current which discharges the capacitor. Secondly, the capacitor has a leakage current, and finally there is a dependence on the amplifier's thermal drift which is exhibited as a change in its input offset voltage. The input offset voltage is the voltage required between its input terminals to maintain a zero output voltage.
2.10. Analogue to Digital and Digital to Analog Converters
Analogue to Digital converters accept an input signal usually in the form of voltage and produce a digital output which represents the analogue signal. These converters are either single integrated circuits or hybrid circuits. However, their implementation is independent of their circuit configuration.
A conversion process must be undertaken if signals are either to be stored or processed by computers. The representation of signals in a digital form also permits their transmission through systems in which there would otherwise be serious concern about signal distortion or deterioration. This is particularly important in a medical environment as we frequently require to provide electrical isolation between measurement and processing circuits for patient safety. The normal forms of coupling circuit are either transformer or optically based.
Transformer based circuits are normally difficult to build with adequate accuracy and are therefore expensive. Optical coupling methods do not readily lend themselves to the direct transmission of analogue signals as their transfer characteristics are normally far from linear.
Both of these isolation methods may be used to transfer digital information where either nonlinearity or the introduction of noise do not immediately compromise the signal accuracy.
The inverse conversion process clearly does not present us with these problems as it simply requires us to restore a digital representation of a signal into its analogue form.
2.10.1. Sampling Frequency or the Nyquist Rate
Several major factors characterize the digitization process. Firstly, we note that if a signal is to be completely characterized by sampling, then the sampling must be carried out at a rate which is sufficient not to cause the loss of information. This rate, called the Nyquist Rate, is twice the maximum frequency contained in the signal. If sampling is carried out at a lower rate, then the sampled signal contains information derived from signals at above the frequency for which the sampling was valid, and permits them to be represented in effect as lower frequency signals. If the sampled signal's frequency spectrum is then analyzed for its frequency components, it may reasonably contain only frequencies up to half the value of its Nyquist Frequency. The components of the signal at higher frequencies will still be represented there as energy, though their energy will falsely be indicated as lying within the permitted band. This error process is known as Aliasing. A mathematical presentation of the sampling process is given by Bracewell (1986).
2.10.2. Quantization Error
A further error in the sampling process is in converting the sampled signal from its analogue samples into the numbers required for representation in a digital form. Any converter has a limited precision and a limited range of voltages which it may represent. When you specify an analogue to digital converter for use in an application, you will firstly need to note the voltage range of the signal in which you are interested and match this, possibly by using amplifiers, to the available converters. Secondly, you will be concerned to ensure that the conversion of the signal to a digital form does not lose required information. Clearly there is little purpose in attempting to discriminate between signal levels which are separated by less than the noise level in the signal you are measuring. However, conventional converters normally provide for conversions on a linear scale so the precision of measurement is greater for large signals than smaller ones. You will require a conversion process that caters adequately in terms of the precision of all signals in which you are interested. To achieve an adequate dynamic range for the conversion process may require you to use additional circuits which alter the gain of a preamplification stage and provide separate indication of the scale factor employed separately to a data acquisition process. In all cases, the 'quantization error' introduced by the digitization process is equivalent to the voltage represented by half of the least significant bit of the converter.
2.10.3. Types of Converter
Several types of circuit exist for both digital to analogue and analogue to digital conversion.
Digital to analogue converters are normally relatively simple devices in which each bit of the digital signal is used to control whether a reference voltage is applied to a resistor. The value of the resistor is chosen to correspond with the magnitude of the bit with which it is concerned: the current which flows in each resistor is therefore dependent on the applied digital signal. The currents may therefore be summed and amplified by an operational amplifier to generate a voltage which represents the number presented to the converter. The performance of the converter is controlled by the precision of the components used, and in particular their susceptibilities to thermal drift. The speed of the conversion process is controlled by the device's switching time, the settling time of the amplifier to an impulse and the capacitance of its input stage.
Analogue to digital converters come in a greater variety, depending on their application. A simple way to build a converter is to use a digital to analogue converter, a clock and control logic. This is shown in FIG. 6. The major drawback in this case is that the conversion time is controlled by the clock rate, whose period must be greater than the converter's settling time times the precision of the conversion process. There are several variants on this arrangement which at the expense of additional circuit complexity accelerate the conversion time from that obtained by the simple process. The most straightforward of these uses the successive approximation technique in which attempts are made to estimate the magnitude of the analogue signal starting at the most significant bit of the converter. Each bit of the converter in turn is tested and the output is compared with the analogue signal. The bit is left set after the test if the generated signal remains below the level of the analogue signal. This conversion process can be carried out at each level at approximately the settling time of the converter, yielding a total conversion time of Setting time x number of levels To obtain very high conversion rates it is possible to convert all bits in parallel using high speed precision comparators together with a precision resistor network. The arrangement is shown in FIG. 6. Whilst this method allows for the conversion of signals at many MHz, its complexity increases quickly with additional precision and places ever greater demands on component accuracy.
3.1. Types of Transducers
The most useful form of transducer is one which converts energy to an electrical signal as this can be readily used and processed. Thus a thermocouple is more useful in instrumentation than a mercury in glass thermometer as the signal it produces is electrical and can be readily interfaced to other circuits for interpretation, display and recording.
Some transducers are used in a rather indirect manner. For example a displacement measurement transducer may be used to measure pressure. In this case a membrane is exposed to a pressure difference and its deformation is measured by a potentiometer.
Some measurements are made by active intervention to obtain a signal. As an example injecting a small quantity of ice cold saline into the blood stream enables measurement of flow in the circulation system. Monitoring the temperature change at a point downstream determines the flow rate.
Transducers vary greatly in complexity. Modern transducers may be complicated semiconductor devices which respond to chemical changes. Transducers should be designed to offer high reliability if their performance is safety critical.
3.2. Desired Attributes of a Measurement System
1. The amount of energy removed from a system by a measurement should be small. All systems are changed when a measurement is performed. A thermometer needs a finite amount of energy, in the form of heat, to function.
2. The system should be sensitive only to the desired signal. For example, ECG measurements should be sensitive to contractions of the heart and not to those of other muscles.
3. The measurement should be minimally invasive: the transducer should cause as little damage as possible.
As we saw in Section 2, the depolarization of muscles and nerves, which respectively cause contraction and the passage of information, is associated with the movement of chemicals across a semi-permeable membrane. This ionic movement causes the generation of an action potential. If two electrodes are placed near to an excitable cell then when the cell is depolarized a potential is developed between the two electrodes. Biopotential recording effectively measures the potential produced by cell depolarization or the electrical activity of nerves and muscles. Essentially, cell membrane depolarization is identical in nerves and muscles. The amplitude of the response is far greater in muscles: it is in the order of 1000 times greater from heart muscle contraction than through a neural nerve impulse.
The measurement of muscle contraction through the measurement of potential changes is called Electromyography (EMG), the measurement of the contraction of the heart, which can be viewed as a special type of muscle, is called Electrocardiography (ECG) and the measurement of nerve activity within the brain is termed Electroencephalography (EEG).
4.1. Biopotential Recording Systems
A block diagram of a typical biopotential recording system is shown in FIG. 7.
To perform a biopotential recording, electrodes or leads are attached to the patient and then connected to a high gain differential amplifier with good common mode rejection. The resulting signal is then displayed. The system also incorporates low pass filtering of the signals and perhaps notch filtering at the mains supply frequency. As biopotential recording necessitates that a low impedance connection is made to the patient, patient safety has to be assured. This is achieved by isolating the pre-amplifier stages.
We are surrounded by power lines and equipment operating from a mains supply. This is particularly true in hospitals where much equipment is gathered together. Power lines operate typically at levels of hundreds of volts. The patient standing in a room in FIG. 8 is capacitively coupled to the power lines. A biopotential measurement, such as an ECG, is then susceptible to a significant signal due to this coupling. As the biopotential is of the order of 1mV the capacitive interference may completely mask the biopotential signal. Early measurement systems were employed only in screened rooms in hospitals, and the recording equipment was battery operated. Now differential amplifiers which reject common mode signals facilitate this type of measurement in the presence of large interference signals.
Capacitive coupling of the power lines causes a signal at the first electrode which with respect to the common ground will be approximately the same as the signal at the second electrode.
The differential amplifier amplifies the difference between the signals at the two electrodes.
The ECG signal measured at one electrode is out of phase with the potential measured at the other electrode and therefore the difference signal is amplified. Hence the noise and power supply coupling are common to both electrodes and are therefore rejected. The ECG signal is different at the electrodes and therefore amplified. The ratio of the amplification of the required differential signal to the amplification of common mode signal is termed the common mode rejection ratio. For biopotential recording a high common mode rejection ratio is required. The common mode rejection ratio for biopotential recording apparatus is normally between 80 and 120 dB for ECG and EEG measurements. The impedance of the amplifier must be large so that the biopotential is developed across the input of the amplifier and not across the patient electrode interface. The input impedance is typically in excess of 5 MR.
The amplifier shown in the FIG. 8 is referred to ground: this used to be a common configuration. As ground referenced equipment incorporates a conduction path to ground, the patient is at risk of electric shock in the event of a fault from the biopotential amplifier, other medical instrumentation or from ancillary equipment such as lighting and radios. Therefore modern biopotential amplifiers are constructed with isolated differential input sections and earth-free patient connections.
The leads from biopotential measurement equipment form a loop which potentially generates a current in the presence of a varying magnetic field. To reduce this effect the leads to a biopotential recording machine are twisted and screened.
A patient's skin resistance can vary between 100 ohm and 2 M-ohm, depending on its condition. The resistance of thick areas of skin, like the soles of the feet, is higher than the resistance of upper parts of the body. Skin resistance is highly affected by moisture and, if the patient perspires, the resistance at the point of measurement will decrease drastically. The change in skin resistance with perspiration is made use of in the lie detector, where sudden changes in perspiration cause a drop in resistance, which is an indication of the patient's stress. Additionally, the patient's skin contains mineral salts. Therefore, to maintain a standardized low impedance interface with the patient, electrodes with a metal salt solution are used. This ensures that the skin is moist and the resistance stable and low. The metal salt sets up an ionically stable interface. If bare copper wires were used oxygen would be liberated at the electrode. With a silver chloride electrode, a balance redox reaction is maintained between chlorine and silver ions.
The epidermis is a semi-permeable membrane, so if a silver chloride electrode is placed upon the skin, chloride and silver ions can permeate the membrane. Thus a potential difference may be developed across the epidermis. Therefore, to prevent this happening, a patient's skin is prepared by rubbing it with an abrasive compound to strip the surface and stop a potential being set up across the semi-permeable membrane. A schematic of an electrode is shown in FIG. 9. Any movement of the electrode on the skin potentially causes signals to be generated. Therefore, the electrode is designed to maintain an even flat contact under which no movement will take place.
In addition to the electrodes described above, implantable electrodes are sometimes used.
These employ a needle with electrodes mounted at its tip. The skin's resistance is broken and therefore the amplitude of the signal received is larger than that from surface electrodes. The use of transcutaneous needle electrodes also allows measurement near to the desired excitable cell. For instance, if a signal is required to be detected from a muscle fiber in the biceps the needle electrode can be placed in that muscle group. Surface electrodes detect an average signal originating from a large volume of the patient whilst needle electrodes record from a specific small volume around the needle tip.
The assessment of the functional behavior of the heart by measurement of the potentials associated with cardiac muscle contraction is perhaps the most widely recognized biopotential recording.
The human heart can be considered as a large muscle whose beating is simply muscular contraction. Therefore contractions of the heart cause a potential to be developed. The measurement of the potential produced by cardiac muscle is called electrocardiology.
The depolarizing field in the heart is a vector which alters its direction and magnitude through the cardiac cycle. The placement of the electrodes on the surface of a patient determines the view which will be obtained of that vector as a function of time. The most commonly used electrode placement scheme is shown in FIG. 10. Here the differential potential is measured between the right and left arm, between the right arm and the left leg and between left arm and left leg. These three measurements are referred to as leads I, II, III respectively.
This measurement lead placement was developed by Einthoven who stated that through measurement of lead I and lead I1 the signal seen at lead I11 could be calculated. This is the most basic form of ECG lead placement: from this the various features of the heart's depolarization can be calculated. Clinically there is a range of lead placement schemes which incorporate limb leads and chest leads.
The heart consists of four chambers as we saw in Section 2, two atria and two ventricles. The atria contract together to force blood into the ventricles and then the ventricles contract.
A typical ECG trace is shown in FIG. 11; the peaks and troughs of the waveform are labeled in accordance with normal medical practice. The first positive peak, the P wave, originates from the depolarization of the atria. Following this there is a negative wave caused by the repolarization of the atria. However, this waveform is usually masked by the depolarization of the ventricles shown as the Q R S complex in FIG. 11. The repolarization of the ventricles causes the positive wave labeled T. Therefore the ECG waveform shows the clinician the electrical waveforms associated with the contraction of the atria and ventricles. From an ECG a clinician may determine the relative timing of the contractions of the atria and the ventricles and assess the relative amplitude of the atrial and ventricular depolarization and repolarization. This information may allow the identification of mild heart block. Following a heart attack a patient's ECG shows changes as the timing and shape of the waveform are dependent on the transmission of the waveform through the muscle tissue. This changes with ischemic muscle damage associated with heart attacks.
ECG measurement is the recording of the potential produced by cardiac muscle. Electromyography is the recording of the electrical activity of muscle: therefore ECG measurement is a type of EMG recording. EMG measurements are taken with essentially the same apparatus as is used for ECG measurement. However, in the majority of cases the signals received are smaller than those obtained in ECG recording as fewer muscle fibers are involved. The size of an EMG signal is directly related to the number of muscle fibers excited.
Clinically EMG is used to determine the function of muscle groups following trauma. It may also be used to assess muscle function following suspected neurological damage. EMG signals are larger than the signals associated with nerve depolarization. EMG is therefore used to detect the arrival of motor nerve information, that is sensing the contraction of a muscle group following its excitation (see section 4.5.3).
Electroencephalography or EEG is the measurement of neural activity within the brain. Each neuron in the brain receives and transmits information through the depolarization of its cell body sending an action potential along the nerve fiber. Within the human brain there is continuous activity and therefore continuous depolarization and re-polarization of neuron.
This results in the continuous generation of electrical signals. These signals can be detected by electrodes placed on the scalp (see FIG. 12). The measurement can be performed with instrumentation similar to that used for ECG measurement but with higher gain and common mode rejection. Signals are much smaller than ECG signals with an approximate amplitude of 2 pV. Usually surface electrodes are used of similar construction to ECG electrodes, with the metal-electrolyte interface being silver / silver chloride. During operations, needle electrodes may be implanted into various parts of the brain. This has allowed researchers to quantify which parts of the brain control which bodily functions.
We are continuously thinking and, therefore, EEG signals appear much like electrical noise.
In certain circumstances, the brain waves or EEG signal may have characteristics which can be identified. If a person is relaxed with the eyes closed lying in a prone position, the brain waves form into a regular low frequency amplitude modulated wave form characterized as an 'Alpha wave' (see FIG. 12). This occurs more readily in men than in women. The bandwidth of the signals is approximately 10 Hz. 'Beta waves' are classified as waveforms between 18 and 30 Hz. EEG is used in studying patterns of brain action during sleep to enable the quantification of a patient's sleep into classes such as rapid eye movement (REM approximately 30 second bursts of 8-12 Hz) and 'deep sleep' (frequencies of less than 4 Hz). During operations brain activity measured by EEG has been used to detect low oxygen and high carbon dioxide levels. Signals may also be recorded from electrodes implanted in the brain to assess the level of damage to a particular region.
A clinical use of EEG is in the diagnosis of epilepsy. Epilepsy is usually suffered by people during puberty or adolescence and may be triggered by fever or flashing lights. The EEG signals recorded during a fit are of higher frequency than normal. During an epileptic fit, EEG can be used to categories the seizure as grand-mal or petit-mal as the characteristic EEG patterns in each case are different.
4.5. Evoked Response Potentials
If you are sitting in a dark room, and suddenly there is a flash of light, then a signal is sent to the sensory cortex of your brain along the optic nerve. If your EEG is recorded then a small impulse is recorded, corresponding to the sensory cortex receiving the optical stimulation signal. In this instance the EEG electrodes are placed over the part of the skull above the visual cortex. This maximizes the signal received from this area. However, much of the signal recorded is due to other stimuli and impossible to distinguish from the specific visual stimulus. However, if the light in the room flashed a number of times, and each time the EEG waveform was recorded, averaging the recorded waveform would diminish signals uncorrelated with the visual event. In this way, if a patient experiences a visual stimulation and the EEG is measured, background interference or neural activity can be removed to leave a signal corresponding to the receipt by the visual cortex of the signal along the optic nerve from the eye. The resulting signal is called a visually evoked response potential.
The EEG signal needs to be captured after triggering by the flash or visual signal. The visual signal may be a simple light flash, a change of pattern on a checkerboard display or a complex video picture. Usually more than 64 individual signals need to be averaged to obtain an adequate display. However, to discriminate the signal properly, about 200 or 300 visual signals need to be applied. To avoid cyclic events in the activity of the brain becoming correlated with the sensory signal, the signal is applied with a random time spacing between impulses.
Visual evoked response potentials have been used to try to determine the way in which the visual cortex processes information. However, the most common application of visual evoked response potentials is in the assessment of the reception of visual signals when patients are either unco-operative or unable to communicate. Visually evoked responses are therefore used to assess babies' sight. Visually evoked response potentials may help to distinguish between physical disability and malingering.
During brain surgery the optic nerve may be stimulated by flashing a light into the patient's eye by mounting LEDs to the patient's eye lid. The signals are recorded from inside the brain using needle electrodes. If the surgeon cuts close to the optic nerve, the nerve shocks and stops transmitting signals and therefore the EEG signal disappears. The surgeon then knows that the cut is close to the optic nerve.
4.5.1. Auditory Evoked Response Potentials
In the same manner that a patient's visual sensory system may be stimulated with a flash, a patient's aural system may be stimulated with a noise. The patient is either stimulated with a click or a tone burst (a sine wave of finite duration). The patient's brain activity is recorded using an EEG and, as with the visual sensory system, a number of EEG waveforms are recorded correlated with tone bursts or clicks.
The auditory signal may be applied to the patient using either earphones, headphones or loudspeakers. Due to the amplitude of the EEG signals (approximately 2 pV) loudspeakers are sometimes preferred. Owing to their greater separation from the point of measurement, they produce less electromagnetically induced coupling from the moving coil into the biopotential recording apparatus than do headphones. As with visual evoked response potentials, auditory evoked response potentials are used to assess industrial injury claims and suspected malingerers. They are also useful if the patient is too young or too psychologically impaired to communicate.
4.5.2. Sumato Sensory Evoked Response Potentials
The third main sensory system that we have after sight and sound is our peripheral nervous system. The nerves are concentrated in areas such as our hands and feet. They respond to pain, pressure and temperature. If they are stimulated a signal is passed from a peripheral nerve to our spinal column and hence to our brain. Therefore, if a patient's limb is excited and the EEG measured an impulse may be obtained in the same manner as for visual and auditory response potentials. However, the sumato sensory response potentials signals may be recorded at many places along the route. If, for example, a patient's hand is stimulated, the sensory nervous signal may be detected at points on the patient's forearm, shoulder and where the nerve enters the spine, as well as being detected in the EEG waveform. Sumato sensory evoked response potentials can be used to determine the extent of sensory sensation that a patient has in a limb or other part of the body. This may be particularly useful following a trauma such as a motor cycle accident, where the patient's nervous system may be shocked so that it blocks all signals. The patient may then be unable to feel anything from a hand or arm. Using sumato sensory response potentials the integrity in the nerve pathway can be assessed.
4.5.3. Measurement of Peripheral Nerve Velocity
The nerves which control the movement of a limb are called motor nerves. The nerve travelling down next to the ulna (one of the bones in the forearm) is called the ulna nerve. This nerve may be stimulated at a point close to the elbow as the nerve is close to the surface (see FIG. 13). Using electrodes of essentially the same type as those used for ECG measurements, a current may be passed between the two electrodes through the patient. This excites the nerve by depolarizing its membrane and causing an action potential to be transmitted. The action potential travels along the nerve to a synapse, a junction between the nerve and muscle, in the thumb. An EMG then measures the electrical signal associated with the contraction of the muscle in the thumb. The time between nerve excitation and the muscle response as determined by the EMG is recorded as the latency. If the point of excitation is moved towards the wrist and the nerve excited again and the latency recorded as before, the difference between the two latencies corresponds to the time it takes for the nerve impulse to travel the distance between the two excitation points. In this way the peripheral nerve velocity can be measured. Measurement of peripheral nerve velocity may be useful in determining the extent of nervous damage following trauma.
4.5.4. Excitation of Nerves or Muscles
As we saw in Section 2, nerves and muscles are excitable cells. This means that their membranes' permeability changes if they are excited beyond a certain threshold. In a nerve, if a membrane is excited, then an action potential travels through the nerve and transmits information from one point to another. In a muscle, the travelling action potential causes a release of chemicals, which cause the contraction of the muscle. However, in both cases, the progression of information or the contraction of muscle is initiated by a change in membrane permeability. Change in membrane permeability can be measured using ECG, EMG or EEG methods as it is associated with a change in potential difference. Likewise, if a current passes through the semi-permeable membrane of a nerve or muscle, then an action potential can be generated. Thus the passage of current through the nervous tissue such as a muscle or a nerve, stimulates the tissue. Nerves and muscles may be electrically stimulated by placing electrodes on the skin and passing a current between them. The patient's skin is usually prepared in the same manner as for ECG electrodes and the electrodes are essentially the same as for that measurement. Electrical excitation of nerves and muscles can be extremely painful for the patient.
The duration of the nerve impulse is small. Potentials of up to 200 V may be used to generate the required current. Constant current sources are used rather than constant voltage as it is the passage of current through the nerve which causes its depolarization. As with measurement of EMG or recording from peripheral nerves, needle electrodes may be used for stimulation. A small area, a particular nerve or muscle fiber group may be excited more accurately than with a surface electrode.
When recording or stimulating, surface electrodes pick up an average signal from a large tissue area, corresponding to a large group of nerves or muscles, whereas needle electrodes either excite or record from a small area or group of fibers.
4.5.5. Electrical Activity of the Heart
The heart is essentially a large muscular bag, which beats continually to pump blood around our bodies. The heart consists of four chambers. The two atria contract simultaneously, as do the two ventricles. The rate at which a heart beats is determined by the Sino Atrial node or SA node. A depolarization pulse originating here spreads through both atria, causing contraction, and then spreads on to the AV node or Atrio Ventricular node. Depolarization passing through the atria does not spread to the ventricles except by the pathway through the AV node. When the depolarization pulse reaches the AV node it is delayed before passing to stimulate the ventricles. This allows ventricles maximally to fill with blood from the atria. The pulses then spread through the ventricles, causing their depolarization and subsequent contraction. Thus the atria and the ventricles contract in a co-ordinated fashion. The rate at which the SA node stimulates the atria is determined by the para-sympathetic and sympathetic nervous systems as described in Section 2.
The SA node is the heart's natural pacemaker as it determines the rate at which the heart beats.
Heart block is the condition where the electrical connection between the atria and the ventricles has been damaged, so that depolarization does not spread to the ventricles. In this instance, the atria beat at a rate determined by the SA node and governed by the body's sympathetic and para-sympathetic systems. However, the ventricular contraction is uncoordinated with the contraction of atria. The ventricles then beat at a rate of around about 40 beats per minute. Even though this beating is uncoordinated with the atrial contraction, the ventricular contraction is sufficient to supply a life sustaining flow of blood throughout the patient's body. However, the patient will have very little energy and may faint and lose consciousness.
A pacemaker is a device which takes over the timing control of the ventricular contraction from the body's natural system to ensure a rate fast enough to allow an active life for the patient. Pacemakers basically consist of a battery, a timing device and electrodes. The battery must be capable of supplying enough current to stimulate or excite the muscles in the ventricles and perhaps the atria for a number of years. As the heart beats approximately 70 times per minute, the battery life must be able to provide stimulation of 70 x 60 x 24 x 365 x the number of years it will survive. Therefore, the requirements for the battery are quite strict.
Battery life is expected to be in the range of 3 - 10 years. Usually 4 to 6 volt pulses are used to excite the heart with a duration of between 1.5 and 2.5 milliseconds. The electrical connection is made from the pacemaker to the patient's heart by leads. These must be rigorously designed as they must withstand a large number of contractions which result in flexing throughout their lifetime. The electrodes are either implanted in the heart or stitched to the surface of the heart.
As a temporary measure, electrodes may be floated into the ventricles of the heart in the same way as a catheter is introduced during blood pressure measurement. The pacemaker itself is situated underneath muscle in the chest. The devices are about half the size of an audio cassette.
The third main block of a pacemaker is the timing device, which regulates when or if the patient's heart is excited. The first type of pacemaker was based around an asynchronous mode timer (see FIG. 14). This constantly excites the patient's heart at a rate determined by the surgeon, but is approximately 70 beats per minute. The units are a simple design and therefore basic and cheap, but they suffered from an intrinsic problem. Since the ventricles sometimes spontaneously contract, if the pacemaker excites ventricles after their contraction, the patient's heart could fibrillate.
Ventricular Fibrillation is a mode of contraction in which the ventricles continue to be excited but in an uncoordinated fashion, causing the heart to flutter rather than to pump. In this state the patient receives no blood flow and therefore is in danger of dying. Once the patient's heart is in ventricular fibrillation, a large signal is required to depolarize the whole of the ventricles simultaneously to synchronize their depolarization. Asynchronous mode pacemakers were associated with heart attacks due to the competition between the pacemaker and the natural, spontaneous contraction of the ventricles themselves.
The battery life of a pacemaker may be monitored non-invasively. Asynchronous pacemakers were designed so that their rate would drop as the battery level fell. For example, when the batteries were fully charged the rate of the pacemaker may have been 70.9 beats per minute, whereas two to three years later, when the battery power dropped, it may have dropped to 70.1. In this way the patient's general practitioner could monitor the battery life of the pacemaker non-invasively during a normal clinic.
A more sophisticated type of pacemaker is the Demand Mode pacemaker (see FIG. 14). This type incorporates detection circuitry. Either separate leads, or those used to excite the ventricles, determine whether the ventricles have contracted. The pacemaker excites the patient's heart if the heart itself has not spontaneously contracted after a pre-determined period. Therefore, the demand mode pacemaker ensures a minimum heart rate but will not excite the patient's ventricles following a spontaneous contraction.
The third and most sophisticated type of pacemaker is the Atrial Synchronization pacemaker (see FIG. 14). In this type of pacemaker, separate electrodes are used to detect the contraction of the atria. The pacemaker introduces a small delay which models the delay in a healthy heart by the atrio ventricular node and then excites the ventricles. As with the demand mode pacemaker, the ventricles are not excited if spontaneous contraction occurs. However, in normal operation, contraction of the ventricles is synchronized with the contraction of the atria to ensure an efficient beating process. Since the excitation of the ventricles is determined by atrial contraction, itself controlled by the SA node, the body's own natural pacemaker, an atrial synchronization pacemaker ensures that the patient's heart rate responds to the demands of the body. Thus if the patient is exercising, the body releases chemicals which cause the heart rate to rise. The rising rate of atrial contraction causes a corresponding rising rate of ventricular contraction, so a patient may lead a normal active life.
Both the demand mode pacemaker and the atrial synchronized pacemaker offer improved performance and reduce the likelihood of patients suffering from ventricular fibrillation due to mistimed excitation pulses. However, both these types of pacemaker can suffer in the presence of noise. The reader will be aware that in the entrance to many libraries and some shops there are signs saying that people wearing pacemakers should contact the staff before entering the building. This is to avoid interference from electrical equipment disturbing the natural function of the pacemaker. Microwave ovens are another source of electrical interference which may disrupt pacemakers.
The most advanced pacemakers sense the level of background interference and, if this is high, they fall back to an asynchronous mode operation, returning to their atrial synchronization or demand mode when the noise level has dropped.
5. Blood Pressure Measurement
5.1. Medical Aspects of Blood Pressure Measurement
The measurement of blood pressure is one of the checks that your doctor routinely performs. Problems with the cardio-vascular system are often mirrored by abnormal blood pressure readings. More specifically malfunction of the heart can be directly diagnosed from blood pressure measurement. In the hospital environment blood pressure is monitored during operations and continuously recorded in intensive care units. Measurements of blood pressure within the heart may provide information about the integrity of the heart valves.
The normal resting patient will have a heart rate of approximately 70 beats per minute. Each stroke ejects about 70 cm^3 of blood from the left ventricle into the aorta (main artery leaving the left ventricle). At the hiatus of each stroke the pressure within the arteries reaches a maximum and then declines during the rest of the cardiac cycle. The peak blood pressure is referred to as systolic whilst the minimum value is the diastolic. The normal patient will have a systolic blood pressure of approximately 120 mm Hg and a diastolic blood pressure of approximately 80 mm Hg.
The body's circulation can be divided into two circuits: the Systemic circuit and the Pulmonary circuit. The pressure in the systemic circuit is high as the blood leaves the heart however, having passed through the body the pressure in major veins is as little as 4 mm Hg. The pulmonary circuit operates at relatively low pressure as the function of the lungs is to exchange oxygen. Pulmonary circuit pressure has a systolic peak of 25 mm Hg and a diastolic value of 10 mm Hg. Therefore blood pressure is pulseatile and dependent on the point of measurement. Clinically the doctor is interested in the systolic and diastolic reading and also the average pressure.
The heart rate is not directly related to the blood pressure as the impedance offered by the body is not constant. For example a sudden fright like finding that your alarm clock has broken on the morning of an exam will have a physiological effect known as the 'fight or flight' response. Following a sudden excitement the blood vessels near the skin's surface contract, therefore, reducing the flow of blood and minimizing the blood loss from a superficial wound. Blood is also diverted from circulating round the stomach and intestines and instead lies in enlarged arteries and veins. The composition of the blood also changes, increasing the clotting speed, with corresponding viscosity changes. Therefore, the blood pressure increases regardless of any change in the heart rate. Using the analogy of relating the blood pressure to voltage and the flow rate to current and the circulation system to impedance.
The blood pressure is clearly dependent on the heart rate, the flow rate and the impedance characteristics of the circulation system.
5.2. Sphygmomanometer Blood Pressure Measurement
The Sphygmomanometer is the most common method of blood pressure measurement. As it is noninvasive it can be performed by your General Practitioner.
The measurement is performed by wrapping an inflatable cuff around a patient's upper arm.
The cuff consists of a bag contained within a cloth sleeve and can be secured by a tie or Velcro strap. The cuff is connected to a hand bulb pump and a mercury manometer. When the cuff is inflated, the pressure increases and the tissues in the patient's upper arm become compressed occluding blood flow in the brachial artery (see Figure 2.24a). The physician listens to the flow of blood at a point below the bag with a stethoscope. The pressure in the cuff is increased above the point where the blood stops flowing in the artery.
Cuff pressure greater than diastolic and systolic.
Therefore, there is no blood flow during the cardiac cycle.
Cuff pressure greater than diastolic but less than systolic pressure. Therefore, blood flows during the systolic part of the cycle.
Cuff pressure greater than diastolic and less than systolic pressure. Therefore, there is some blood flow during the entire cardiac cycle. However, there is still noise associated with turbulence.
Cuff pressure less than systolic and diastolic blood pressure. Blood flows continuously.
The physician then gradually releases the pressure in the cuff; at a point when the pressure is equal to the systolic blood pressure blood begins to flow again. At this pressure blood squirts through the compressed artery during the systolic part of the cardiac cycle, but at the diastolic part of the cycle the artery closes again and no blood flows. Therefore, the artery opens and closes causing a characteristic knocking noise known as Korotkoff's sounds. At the point when these noises begin the clinician notes the manometer pressure as being equal to the systolic pressure.
The pressure within the cuff is then further reduced and consequently the artery remains open for an increasing portion of the cardiac cycle. When the cuff pressure is equal to the diastolic pressure the artery remains open during the whole cycle and the noises stop. The physician will record this cuff pressure as the diastolic pressure.
This method of measuring blood pressure suffers from several drawbacks.
The measurement is subjective, that is to say it is dependent on the opinion of the clinician when the Korotkoff's sounds begin and end.
The fidelity of the stethoscope used and of the hearing of the clinician directly control the measurement accuracy.
The sphygmomanometer is therefore a relatively inaccurate subjective measurement of blood pressure. However, it is non-invasive, easy to perform and the instrumentation required is cheap. This is the type of blood pressure measurement performed in the front line of medical practice, the fault finding level. Measurements with greater accuracy require invasive techniques.
There is a number of different methods of determining the point at which blood begins to flow again other than detecting the Korotkoff sounds. As before, pressure in a cuff cuts off the blood flow. As pressure is decreased the blood flow resumes when the cuff pressure is approximately equal to the systolic pressure. This flow may be detected by a Doppler ultrasound blood flow detector (see Section 9.3). The Doppler device produces an audible output which is proportional to the blood flow velocity.
A transducer may be used to measure vibrations in the cuff which recommence as the cuff pressure approximately equals the systolic blood pressure. The vibration in the cuff reaches a maximum at a point when the pressure in the cuff is equal to the average blood pressure, and decreases and disappears when the cuff pressure is equal to the diastolic pressure.
Instruments which perform non invasive measurement of blood pressure have been developed which use both the vibration principle and the detection of Korotkoff's sounds. In both cases the cuff is automatically inflated and the vibration signal detected by transducer located in the cuff. These machines provide measurements of the systolic, average and diastolic blood pressures. However, these machines perform badly in the presence of other vibration signals.
5.3. Invasive Measurement of Blood Pressure
Invasive blood pressure measurement allows more accurate blood pressure measurement, dynamic measurement and measurement of pressure at specific points in the circulation system.
5.3.1. Catheter Measurement
Invasive measurement of blood pressure is routinely performed using a catheter transducer.
An incision is made in the patient and a catheter is introduced into the circulation system. A catheter is an open ended tube used to couple the pressure inside the patient to an external transducer: in other circumstances catheters may be used for drug infusion or drainage.
The catheter may be introduced into either the arterial or the venous system depending on the point at which the measurement is to be performed. The most common points of incision are in the neck, arm or groin, as these sites have large veins and arteries which are close to the skin's surface.
The catheter is filled with saline solution and connected to an external pressure transducer (see FIG. 16). The saline filled catheter transmits the pressure at the catheter tip to the external transducer. The housing of the pressure transducer allows the catheter to be regularly flushed with saline solution to prevent blood from clotting at the open end of the tube (clots in the circulation system can obstruct blood flow and cause brain damage or damage to lung function). The transducer housing also allows the connection of a reference pressure source to allow for calibration of the transducer.
Once introduced into a blood vessel the catheter can be fed along the vein or artery reaching deep locations in the circulation system such as the heart or major arteries. Therefore the catheter tube has to be very compliant to bend around the structures within the body. The requirement for compliance contradicts the tube's primary function of transmitting pressure to the external transducer. If the tube is flexible and compliant then the pressure changes which are being measured may extend the tube and consequently be distorted when they arrive at the transducer. A further problem with catheters is that foreign matter or bubbles within the tube may also degrade performance. Bubbles within the catheter compress when pressure increases and distort the pressure measured at the transducer. These factors limit the dynamic performance (i.e. bandwidth) of the catheter tube blood pressure measuring system.
The bandwidth of such a system may be as little as 12 Hz.
However, the catheter system is relatively cheap, and the transducer is reusable and robust.
5.3.2. Catheter Tip Transducer
Another type of invasive blood pressure measuring system is the catheter tip blood pressure transducer. In this instance the transducer is mounted at the tip of a tube inserted into the patient and therefore directly measures the blood pressure at that point. The dynamic performance of this system depends on the transducer's bandwidth which may extend to as much as 1 kHz. Hence this kind of measurement is preferred when measuring the wide bandwidth signals found in the aorta and within the heart, particularly if associated with valve abnormalities. The intravenous tip transducer may be a little less compliant than the catheter system.
However, its major drawbacks are price and operating life. The design requirements to provide an ultra miniature transducer make this kind of instrument expensive and it is easily damaged both in normal use and accidentally. The transducer's performance is also degraded by repeated sterilization at high temperature.
5.4. Design of Blood Pressure Transducers
The type of transducer most commonly used in both catheter tube and intravenous blood pressure measurement systems is the strain gauge transducer.
5.4.1. Strain Gauge Pressure Transducers
If a wire is stretched the resistance measured between its ends increases due to a number of factors. The diameter of the wire decreases, the length of the wire increases and the resistivity increases. These three factors are related to the resistance of a wire by the equation given below.
where A is the cross sectional area, 1 is the length of the conductor and p is the resistivity of the conductor.
In most strain gauges the changes in dimension are the dominant reason for the measured change in the resistance of a wire under tension. For small strains (those with changes of dimension of about 2%) the change of resistance with strain is approximately linear. If a wire is connected between two points monitoring the resistance of the wire correlates with the strain it experiences. This type of strain gauge is referred to as an un-bonded strain gauge and is uncommon in transducers today.
A bonded strain gauge consists of a plastic substrate upon which a layer of conducting material has been laid. The conducting material is laid out to maximize the length of the conductor in the direction of the required measurement. The bonded strain gauge may be glued to a material such as a block of steel. Any strain experienced by the block will be passed on to the bonded element. Bonded strain gauges can be fabricated from a variety of materials.
The temperature dependence of resistivity is possibly a limitation in the application of certain strain gauge transducers.
In a pressure transducer a membrane separates the fluid under test from a fluid at a reference pressure (normally air at atmospheric pressure); if the pressure of the test fluid exceeds that of the reference fluid then the membrane distorts in the manner shown in the FIG. 17a. There is therefore a strain in the membrane proportional to the membrane's displacement and thus the pressure difference.
Strain gauges are mounted directly to the reverse side of the membrane. The resistance change measured will be extremely small and so gauges are incorporated in a Wheatstone bridge circuit. To maximize the sensitivity of the transducer and also to reduce the effect of temperature changes four strain gauges are used in a full bridge configuration. The transducer mounted in the centre of the diaphragm (see FIG. 17b) experiences tension during membrane distortion and the gauges at the rim experience compression. This may at first seem rather surprising that the rim gauges will experience compression but if the situation is examined closely as in FIG. 17c, then the membrane at this point can be considered as a beam under load and the situation understood.
Both catheter tip transducers and the comparatively large transducers used for catheter blood pressure measurement are constructed from a flexible diaphragm mounted with a strain bridge.
There are other types of pressure transducer which use other physical properties such as the piezo electric effect, capacitive sensing of membrane displacement and optical fiber techniques.
5.4.2. Safety of Blood Pressure Transducers
Blood pressure transducers are electrically hazardous as they penetrate the skin and therefore form a low impedance connection to the patient. The most susceptible organ to electrical accidents is the heart; this is measured directly by blood pressure transducers. The fluid filled catheter method establishes a conducting path to the heart and the intravenous transducer actually incorporates an electrical circuit within the patient's heart. On the other hand, the optical method may be completely isolated. For this reason optical transducer methods for blood pressure measurement may become popular.
5.5. Measurement of Blood Pressure within the Heart (the Swan Ganz Catheter)
In certain clinical conditions such as heart valve failure it is necessary to measure the blood pressure within the heart. This is achieved by introducing a catheter transducer into a vein or artery and feeding the catheter up against the flow. The patient is examined radiologically to determine the position of the catheter. If the catheter is introduced into an artery then it enters the left hand side of the heart. If the pressure transducer is introduced into a vein then it enters the right hand side of the heart.
5.5.1. Heart Catheterization
To catheterize the right hand side of the heart a catheter is introduced into a vein and fed against the direction of blood flow. The position of the catheter is monitored by X rays as it is radio opaque. The transducer emerges at the entrance to the right atrium, and a pressure measurement may be taken at this point (see FIG. 18). The transducer can then be pushed through and into the atrium where again a pressure measurement may be performed. Careful manipulation may then position the catheter in the ventricle and the pressure can likewise be measured at this point. The catheter cannot be manipulated into the pulmonary artery and so to take measurements there a balloon is inflated close to the end of the catheter which is carried by the blood flow through the pulmonary valve and into the artery. Once in the artery the balloon is deflated and the pulmonary artery pressure taken. With the catheter in this position it is also possible indirectly to measure the pressure within the left atrium. This is achieved by re-inflating the balloon at the end of the catheter until blood flow in the artery stops. If there is no flow then the pressure within the pulmonary artery equals the pressure in the left ventricle as there is effectively a closed path.
This special process is performed using a Swan Ganz catheter. In special care wards patients are intensively monitored. Therefore patients are catheterized and the blood pressure within their hearts is continuously monitored. In addition the flow rate through the heart can be assessed by thermal dilution. This method comprises injecting a small quantity of ice cold saline solution into the heart and monitoring the length of time for this temperature fluctuation to be passed to a further part of the circulation system.
5.6. Measurement of Oxygen in the Blood
The function of the cardio-vascular system is to pass blood through the lungs so that hemoglobin can combine with oxygen, which can then be carried around the body. The measurement of the amount of oxygen in arterial blood is a direct measure of the performance of the cardio-vascular system in general. There are two clinically used methods of measuring the amount of oxygen in the blood: one relies on an optical measurement procedure, the other on a chemical reaction.
FIG. 19 Optical properties of hemoglobin and oxy-hemoglobin
Beer's Law relates the absorption of light passing through a solution to the concentration of the solution, the optical path length and the absorption coefficient. If the concentration of a solution is doubled, the absorption doubles and likewise an increase in the path length causes a corresponding increase in the absorption. This relationship is exploited in oximeter to measure the amount of oxygen in blood.
The active cells in blood which carry oxygen to tissue are hemoglobin. The spectral characteristic, the strength of absorption at a particular frequency, of hemoglobin changes when it is combined with oxygen. The spectral characteristics of oxygenated and deoxygenated hemoglobin are shown in FIG. 19. The characteristic absorption functions cross at the isobestic point. At this point, the absorption of hemoglobin and oxyhemoglobin are equal--this wavelength is 803 nm. The spectral absorption characteristics of blood account for the difference in the appearance of arterial and venous blood. Arterial blood (predominantly oxyhemoglobin) is red whilst venous return (hemoglobin) is blue.
The oximeter determines the percentage oxygenation of hemoglobin as defined below.
A sample of arterial blood taken from the patient is placed in a cuvette. A cuvette is a small rectangular glass vessel with known optical properties. The measurement is performed by measuring the absorption at two wavelengths, namely at the isobestic point and at 605 nm. As the path length (the dimensions of the cuvette) and the absorption coefficient are known the measurement at 803 nm allows the calculation of the total concentration of hemoglobin in the sample. When this is combined with a measurement at 605 nm equations can be solved to yield the percentage of oxygenated hemoglobin.
The advantage of this method of measuring the percentage saturation or the percentage oxygenated hemoglobin is that it is relatively easy to perform. The disadvantage is that a sample of blood has to be removed from the patient to perform the measurement. This means that it is relatively slow.
5.6.2. Oxygen Association
Oxygen in the blood is held there in two states:
1. Physically dissolved in the fluid.
2. Combined with hemoglobin.
FIG. 20 shows the oxygen dissociation graph for blood. It shows on the abscissa the percentage of oxygen combined with hemoglobin whilst the ordinate shows the total oxygen in the blood. Initially the graph is relatively linear however, once the percentage of combined hemoglobin exceeds 80%, the graph levels off and a large increase in total oxygen correlates with a small change in percentage oxygenation. For instance, the 20% increase in the total oxygen from 8 Pa to 10 Pa on the graph is equivalent to a 2% change in the percentage oxygenation. Therefore, when using an oximeter, the percentage hemoglobin is a good judge of the amount of hemoglobin which is combined with oxygen. However, it does not directly reflect the amount of oxygen in the blood; a small change, say from 98 to 96% in the hemoglobin concentration, can mask a change of 20% of the total oxygen within the blood.
A further problem is that some diseases or conditions cause a change in blood pH. This affects the oxygen dissociation curve, shifting it to the left or right, as shown in FIG. 20. In these instances the relationship between the percentage oxygenation of the blood or the percentage oxyhemoglobin to total oxygen in the blood is altered. Therefore, an oximeter measurement of blood oxygenation gives a good indication level of combined oxygen in the blood but not an accurate reading of the total oxygen present. Normally, the anaesthetist will expect a reading above 90% for the percentage saturation.
A further point of interest is that blood is composed of hemoglobin and abnormal hemoglobin, which has different spectral properties to normal hemoglobin. This may render the calculation of percentage oxygenation inaccurate. An oximeter measurement may also be disturbed if the patient is suffering from carbon monoxide poisoning, as this will alter the spectral characteristics of the blood.
5.6.3. The Pulse Oximeter
Oximeters rely on taking samples of blood, and placing them in external containers so that optical characteristics can be determined. This method is therefore slow and quite difficult to perform during an operation. The pulse oximeter uses essentially the same theory as the oximeter, but is a non-invasive measurement which can give a continuous reading of the percentage oxygenation of the hemoglobin.
A pulse oximeter takes a reading of the absorption for two wavelengths of light, as before, but through a portion of the patient's body containing arterial blood. The portions of the patient's body normally chosen are the ear lobe or the finger. A spring clip containing an optical source and detector is fastened around the patient's finger or ear lobe. The absorption through the patient is then dependent upon both the oxygenated and deoxygenated hemoglobin within the ear or finger, the tissues within the ear or finger and the skin, and in addition, the absorption of venous blood in the pathway. The light source emits pulses of light at both frequencies of light used in the measurement. A reading is also taken when the light source is off to determine the background noise or 'dark reading'. The pulses are emitted continually and so the absorption is continuously monitored. Due to the pulsing of the arterial blood, absorption is not constant with time as shown in FIG. 21. The relationship between the pulseatile component of the absorption and the percentage saturation of hemoglobin has been empirically determined. The pulse oximeter, therefore, performs a continuous in-vivo measurement of blood oxygen saturation. However, it suffers from the same limitations as the oximeter.
Measurements performed by oximeters can also be widely affected by electrical interference.
If a surgeon uses a diathermy in the operating theatre, then the pulse oximeter reading may be disrupted.
5.6.4. New Developments in Pulse Oximetry
To overcome the interference which pulse oximeters experience from other electrical equipment in the operating theatre, pulse oximeters are being built with the optical source and detector housed in a screened instrument case and the light fed to the patient by an optical fiber. A further modification to increase the accuracy of the calculation for measuring oxygenation is to include more than two wavelengths of light for measurement.
5.6.5. Chemical Measurement of Blood, Gas Content
The chemical oxygen sensor shown in FIG. 22 can be used to measure the level of oxygen in a sample of blood or gas. The cell consists of two electrodes in a solution and a semipermeable membrane. One of the electrodes is a silver / silver chloride electrode, whilst the other is platinum. The electrolyte within the cell is potassium chloride. The bottom of the cell is a semi-permeable membrane. There are two chemical reactions which can take place within the cell: one is silver combining with chlorine to form silver chloride and the other is oxygen combining with water to form hydroxyl ions.
The platinum electrode acts as a catalyst. The second reaction is oxygen dependent. For the water to dissociate to hydroxyl ions, oxygen needs to be present. The cell is operated by applying a potential of between 0.2 and 0.6 volts between the two electrodes. Initially, as the voltage is increased, the current increases between the two electrodes. However, it reaches a point when there is no increase in current with increased voltage. At this point the current that flows from one electrode to the other is dependent on the rate of the reaction and therefore the amount of oxygen entering the cell. The Clark Cell has specially designed permeable membranes to allow measurement of oxygen in blood. The reaction rate in the cell is determined by the amount of oxygen crossing the semi-permeable membrane from the blood.
The resulting current in the cell is proportional to the amount of oxygen in the blood.
This cell measures the total amount of oxygen in blood. The dissolved oxygen and that combined with both normal and abnormal hemoglobin are measured. However, the cell cannot be used non-invasively. Either a sample of blood must be removed from the patient and placed on the semi-permeable membrane, or the cell itself must be miniaturized so that it may be inserted into the patient.
However, a cell has been designed using the same procedure as the Clark electrode for noninvasive blood oxygen measurement. This cell is placed on the patient's skin and the skin is heated. Heating the skin increases peripheral arterial circulation and opens the capillary pathways. Therefore, using a semi-permeable membrane especially designed for skin contact, the rate of the reaction is controlled by oxygen diffusing through the patient's skin from the arterial circulation into the cell. In this way the cell can be used to measure in-vivo oxygen concentration of blood.
5.6.6. Gaseous Oxygen Measurement
A further measurement of the cardio-vascular system can be obtained by measuring the oxygen inhaled and exhaled by the patient. Oxygen is a paramagnetic substance. Therefore, the presence of oxygen increases a magnetic field. Diamagnetic materials experience a force in a non-uniform field. The gaseous-paramagnetic oxygen analyzer uses these two phenomena to measure the amount of oxygen in a sample of air.
The instrument is fabricated with two dumb-bells containing nitrogen, a diamagnetic substance, within a magnetic field (see FIG. 23). The dumb-bells are suspended from a wire so that the magnetic field causes a torque, twisting the wires attached to the dumb-bells. At a point there is a stable position reached when the torque acting on the dumb-bells due to the magnetic field equals the torque exerted through twisting the wire. If the magnetic field were increased then the dumb-bells would rotate further.
The instrument is calibrated by filling the chamber full of inert gas and marking the displacement of the dumb-bells. If the chamber is then completely filled with oxygen, the magnetic field increases and the dumb-bells rotate further. This new position is noted. There is a linear relationship between the dumb-bell rotation between the first point measured with inert gas and the second point, measured with 100% oxygen filling the cavity. The amount of oxygen in the gas can be determined. The measurement is practically performed in a similar way to a spot galvanometer, in that a minute mirror is placed in between th9 two dumb-bells at the point of intersection with the wire. Alight is shone onto the mirror and reflected onto a screen.
The position of the reflection corresponding to zero oxygen concentration in the chamber and 100% oxygen in the chamber are marked as being 0 and 100% oxygen. Then the position of the reflection between these two marks can be related to the percentage in the sample.
Modern gaseous oxygen analyzers incorporate a coil of wire around both dumb-bells so that when a current passes through the wire, an equal and opposite force to the magnetic torque due to the diamagnetic nitrogen is produced. In this way the current can be varied through the coils to maintain zero displacement of the dumb-bells. Hence, a graph of current through the dumb-bells to maintain zero position against oxygen concentration can be used to calibrate the device. The gaseous oxygen analyzer malfunctions in the presence of significant concentrations of other paramagnetic gases, such as carbon dioxide.