Physics & Monitoring in ICU – Part 1

Physics and Monitoring in ICU – Part 1

Written by – Dr Madhuri Anupindi

Describe the physical principles of ultrasound 

Ultrasound is sound whose frequency is above the range of human hearing. Sound is a mechanical wave produced by vibration of the medium and can be represented by a sinusoidal wave. The wavelength is the distance between two consecutive points of the same phase, the amplitude is the maximal distance covered by each particle during an oscillation and the frequency is the number of cycles occurring in 1 second measured in hertz.  Medical ultrasound uses high frequency sound ranging from 2 – 20MHz to aid in the diagnosis and treatment of patients. 

Ultrasound waves are generated from multiple small piezoelectric crystals within the transducer. When an alternating current is applied, these crystals deform and vibrate at high speeds, converting the electrical energy into an ultrasound wave. This process also works in reverse, where the ultrasound waves are received by the crystals and converted into electrical energy. The ultrasound waves are emitted as short pulses at regular intervals. Pulse duration is the length of the pulse measured in time, a short pulse duration results in better axial resolution, or the ability to distinguish two closely spaced objects along and parallel to the ultrasound path. The pulse is comprised of a range of frequencies and 2000 – 4000 pulses are emitted per second which is the pulse repetition frequency. These pulses propagate through different tissues and then return to the transducer as reflected echoes. These echoes are then converted back into electrical impulses by the transducer crystals and are amplified and processed to form an ultrasound image presented on the screen.

As ultrasound waves transverse through the body, several factors cause it to lose energy and undergo a reduction in amplitude. This is termed attenuation and is determined by the tissue involved, the distance travelled and the beam frequency. There are four main processes which result in attenuation:

• Absorption: part of the sound wave is absorbed by tissues and converted into heat. The amount absorbed depends on the frequency of the sound wave, scanning depth and the tissue itself. Higher frequency sound waves are absorbed more than lower frequency, therefore although higher frequency waves yield better image resolution they have lower penetration due to higher absorption. Higher frequency probes such as the linear probe have better resolution but poorer penetration compared to the curvilinear probe which has a lower frequency.
• Reflection: sound bounces back and returns to the probe. This occurs at interfaces between soft tissues of differing acoustic impedance or resistance. Acoustic impedance is proportional to tissue density, e.g. bone has very high acoustic impedance whereas fat has a much lower acoustic impedance. If two adjacent tissues have identical acoustic impedance, then no sound waves will be reflected back. Tissues with a large difference between acoustic impedance will be strong reflectors e.g. bone/tissue. Air and tissue have very large acoustic impedance with over 99% of sound waves being reflected back upon encountering air. Consequently, beyond this interface there is almost no ability to further delineate structures. This is why we use ultrasound gel, as it eliminates the air between the transducer and the skin and is formulated to have a similar acoustic impedance to body tissue, thereby allowing for the maximum amount of ultrasound wave transmission.
• Refraction: occurs when sound waves from one medium to another which has a different propagation velocity of sound. This change in velocity results in change in direction of the original wave which can result in double image artefacts. 
• Scattering: occurs when ultrasound waves encounter objects that are small compared to the size of the ultrasound wavelength causing the wave to scatter in all directions and most are not reflected back to the probe 

What is the Doppler effect and how does it relate to ultrasound?

The doppler effect is the observed change in frequency of a sound wave due to the relative movement between an emitting sound source and the receiver. This can be illustrated by thinking about the sound of an ambulance’s siren as it approaches and then passes. The pitch of the siren is higher as the ambulance approaches and lower as it drives by. The pitch of the siren has not actually changed, it just appears to have changed due to the relative position of the sound source and the receiver. This is because waves emitted by an object travelling towards an observer get compressed resulting in a higher frequency whereas when the source is moving away the waves are stretched out resulting in a lower frequency. The Doppler effect will not occur if the source and observer both move in the same direction at the same speed or if the source is at 90 degrees to the receiver.

The Doppler effect is used in ultrasound to detect and measure blood flow. When a transducer is placed over a blood vessel, the reflected wave will have a higher frequency compared to the initial wave if the flow of blood is towards the transducer. This is a positive Doppler frequency shift. Conversely, if the flow of blood is moving away from the transducer, a lower frequency will be reflected, that is a negative Doppler shift. When flow is perpendicular, no Doppler shift is detected. The Doppler shift frequency is proportional to the velocity of the blood flow and the cosine of the angle between the transducer and the blood flow. The optimal Doppler shift is when the transducer is parallel to the blood flow.

Outline the principles underlying pulse oximetry

Pulse oximetry uses the principles of light absorbance as described by the Beer-Lambert Law to calculate the oxygen saturation of haemoglobin. The Beer-Lambert law states that the measured absorbance for a single compound is directly proportional to the concentration of the compound and the length of the light path through the sample.  Therefore, the amount of light absorbed is proportional to the amount of haemoglobin present and the width of the artery that the light is travelling through. The second principle used in pulse oximetry is that oxyhaemoglobin and deoxyhaemoglobin absorb light of different wavelengths in different amounts. Deoxyhaemoglobin absorbs more of red light which has a wavelength of about 660nm whereas oxyhaemoglobin absorbs more infrared light which has a wavelength of around 940nm. Therefore, the amount of red and infrared light absorbed by the blood will depend on the proportion of deoxy and oxyhaemoglobin present. 

The third principle utilised is that pulse oximeters measures pulsatile blood. In the finger, light can also be absorbed by other tissue such as skin which would otherwise confound the measurement of oxygen saturation. Pulse oximeters compare the pulsatile signals, that is those that have changing absorbance levels over time with signals that are non-pulsatile such as skin and venous blood. A ratio or R is then created of the pulsatile and non pulsatile elements at the two different wavelengths measured. This R value is then calibrated by using a graph of empirical values derived measuring oxygen saturations of healthy volunteers who were asked to breathe lower and lower concentrations of oxygen. Measurements were made of the pao2 and oxygen saturation and a calibration graph created. It is important to note that the volunteers were not allowed to desaturate below 75% so the remaining calibration was calculated empirically. 

Pulse oximeters consist of an opaque probe which minimises ambient light and contains light emitting diodes that emits rapid sequences of red light at 660nm and infrared at 940nm. The pulse oximeter also contains a photodetector on the other side of the probe which records the amount of light received and can compare it to the amount of light received when the red light was on, the infrared light was on and when neither was on. In this way it can calculate the amount of ambient light (the light detected when neither of the LEDs were on) and thus negate its effect. The pulse oximeter also contains a signal amplifier and an integrated circuit controller and processor which can calibrate and analyse the data. This is then displaced on a user interface, ideally as a continuous trace. 

List the limitations of pulse oximetry

Limitations of pulse oximetry include difficulties with signal measurement, interference with absorbance, errors in calibration and limitations of technique. 

Errors and difficulty in signal measurement

• Excessive movement of patient e.g. seizures, agitation
• Poor perfusion e.g. hypovolemia, hypothermia, vasoconstriction
• Nail polish
• Excessive ambient light
• Non pulsatile flow e.g. bypass
• Arrhythmias 
• Venous pulsations e.g. severe TR
• Severe anaemia
• Pigmentation

Interference with absorbance

• Dyshaemoglobinaemias e.g. carboxyhaemoglobin or methaemoglobin
• Intravascular presence of dye e.g. methylene blue

Errors in calibration

• Table does not go down to the lowermost saturation – this was empirically calculated therefore it is less accurate at lower saturations

Limitations of technique

• Only one component of determinants of oxygen delivery 
• A normal oxygen saturation does not exclude respiratory failure 

What is the effect of an elevated level of carboxyhaemoglobin and methaemoglobin on pulse oximetry values. 

Carboxyhaemoglobin has similar absorbances of the red light (660nm) to oxygenated haemoglobin. This can result in the pulse oximeter seeing oxyhaemoglobin and carboxyhaemoglobin as the same when using the red light spectrum and therefore measure a combined absorbance. Carboxyhaemoglobin minimally absorbs infrared (940nm) light so this will not change the absorption. The red light absorption of the combined O2Hb and COHb will be maintained or slightly increased. However, as deoxygenated Hb still absorbs red light better than either COHb or O2Hb, the reduction in deoxygenated Hb in the presence of COHb results in a net effect of decreased red light absorption. Therefore when R, or the ratio of absorbance of the pulsatile and non-pulsatile elements at both wavelengths is calculated, it will be lower than in reality and a falsely elevated Sao2 is generated. 

Methaemoglobin absorbs more light than either deoxy or oxyhaemoglobin at 940nm whereas at 660nm its absorption is similar to deoxyhaemoglobin. The pulse oximeter may therefore perceive an elevated concentration of deoxyhaemoglobin which can falsely lower the oxygen saturation. Furthermore, as methaemoglobin absorbs both red and infrared light well it results in an R value that approaches 1, which is equivalent to saturations of around 85%. Therefore, the higher the methaemoglobin concentration, the greater the tendency of the pulse oximeter to display a spo2 of 85% regardless of the actual oxygen saturation. If the actual Sao2 is less than 85% the Spo2 value will overestimate it, if it is more than 85% the Spo2 will underestimate it.  

How is ETCO2 measured using infrared spectroscopy and what are the potential sources of error?

Infrared spectroscopy also uses the Beer-Lambert law to calculate the concentration of carbon dioxide present in expired air. Infra-red light with a wavelength of 4.26um is passed through expired air and the amount absorbed is measured. The amount of infra-red light absorbed is proportional to the concentration of carbon dioxide. This wavelength was chosen as it fits well with the peak absorption of Co2 and is well separated from other absorption peaks of other gases. A light-emitting diode emits a narrow band of infrared light which passes through a sample chamber of fixed size which contains expired Co2. A detector is present on the opposite side which measures the absorption of light at this spectrum. A processor then calibrates this information and ideally produces a graph of CO2 concentration over time or a capnograph trace.

Capnograph analysers can be either main stream or side stream. Mainstream analysers place the detector directly in the path of the expired gas whereas sidestream monitors remove a small fraction of the expired gas via additional tubing and the analyser is located away from the patient. 

Sources of error include errors of sampling, calibration, interference and pathological.

• Sampling errors
  • Entrainment of atmospheric air if there is a leak in sidestream line
  • Occlusion of sidestream line causing loss of gas sampling
  • High sampling flow rate with sidestream sensor will produce a falsely low ETCO2
  • Water condensation can absorb light causing an erroneously high ETCO2
  • If sampling chamber too large it can result in erroneously low co2
• Calibration errors
  • Incorrect calibration of analyser
• Interference
  • Collision broadening effect: spectral absorption peaks of gas are broadened due to the collision or proximity of molecules of another gas. This can happen especially with excess nitrous oxide as its peak absorption spectrum is similar to carbon dioide. 

What are the causes of an increased difference in PaCO2 and ETCO2 levels?

The normal PaCO2 – ETCO2 difference is around 2 – 5 mmHg and is due to alveolar dead space. Alveoli which are ventilated but not perfused will therefore have an expired gas mixture which is identical to inspired gas. This would then dilute the expired CO2 concentration (as inspired gas does not contain CO2) and increase the difference between the PaCO2 and ETOC2. Therefore, anything that increases alveolar dead space will increase the PaCO2 – ETCO2 difference. Causes of an increased difference include:

• Decreased pulmonary perfusion
  • Embolic phenomenon: pulmonary embolism, air embolism, fat embolism
  • Decreased cardiac output: cardiac arrest, severe hypovolemia, right heart failure
  • Pulmonary hypertension
• Ventilation issues 
  • High PEEP causing increased VQ mismatch
  • High Fio2 causing shunt 
  • Increased alveolar dead space e.g emphysema
  • Oesophageal intubation
  • Inadequate expiration time or difficulties expiring resulting in incomplete alveolar expiration and falsely low ETCO2 e.g. asthma
• Measurement error
  • Issues with sampling line: excess water in trap, air leakage, line blockage
  • Loss of calibration of sampling unit
  • Presence of helium or nitrous oxide can sometimes incorrectly elevate ETCo2
  • Use of an inline HME filer can reduce ETCO2 concentration
  • Incorrect timing of measurement: needs to be end expiration