Elizabeth M. C. Ashley
Elizabeth M. C. Ashley
Nina Brixey and Andrew Gratrix
Nina Brixey and Andrew Gratrix
Andrew Rhodes and Stephen Shepherd
A. Agatha Crerar-Gilbert and Nick Fletcher
Nick Fletcher and A. Agatha Crerar-Gilbert
Mark Hamilton and Jonny Price
Davide Bastoni, Hollmann D. Aya, and Maurizio Cecconi
Hollmann D. Aya, Simone Bazurro, and Maurizio Cecconi
Hollmann D. Aya, Maurizio Cecconi, and Andrew Rhodes
Jean-Louis Teboul and Olfa Hamzaoui
Electrocardiograph (ECG) monitoring is carried out to measure heart rate, rhythm and conduction disturbances, and to monitor pacemaker function. It is essential for the detection of myocardial ischaemia and may also give an indication of electrolyte imbalance.
An ECG ‘lead’ refers to the potential difference recorded between two defined points. The standard calibration for an ECG is a paper rate of 25 mm/s and 1 mV/cm (Figure 7.1). If the predominant direction of the electrical current is toward the positive electrode, then the deflection of the oscilloscope is upward; if it is away, the deflection is downward. The frequency range (cycles per second) of an ECG is 0.5–4 Hz, i.e. a heart rate of 30–240 bpm. The ECG can be broken down into sine waves or harmonics by Fourier analysis. A minimum of 10 harmonics has to be displayed to produce an ECG waveform, requiring a bandwidth between 0.05 and 40 Hz. The bandwidth has to be widened to 0.05–150 Hz to enable ST segment analysis. Low and high-frequency filters improve the waveform display; the low-frequency filter diminishes baseline drift caused by movement and breathing, but exaggerates ST segment changes. The high-frequency filter reduces wall power-source noise but prevents visualization of pacemaker spikes and interferes with QRS and J point recognition. Biological ECG signals are small and require amplification with a differential amplifier to minimize noise. Placing electrodes on bony prominences minimizes interference from muscle activity and the ECG. The signal is then transferred to an oscilloscope for continuous monitoring or to a printer for interpretation. Adjusting the gain on an ECG monitor may exaggerate or minimize ischaemic changes. It is therefore important to interpret ST elevation or depression when the standard gain of 10 mm/mV is selected.
Table 7.1 Anatomical location of the ECG leads
Augmented unipolar leadsa
Unipolar chest leadsb
I Between right arm and left arm
aVR right arm
V1 Fourth intercostal space, right sternal edge
II Between left leg and right arm
aVL left arm
V2 Fourth intercostal space, left sternal edge
III Between left leg and left arm
aVF left leg
V3 Midway between V2 and V4
V4 Fifth intercostal space, left midclavicular line
V5 Fifth intercostal space, left anterior axillary line
V6 Fifth intercostal space, left mid-axillary line
a, The reference electrode is formed by combining leads I, II, and III, i.e. the centre of Einthoven’s equilateral triangle, where the potential difference is zero.
b, The reference electrode is formed by the combined aV leads.
Continuous ECG monitoring: lead selection
In the operating theatre, a single lead ECG display is used, because theatre monitoring employs a three-lead ECG configuration. It is possible to switch between lead I, II, and III. During cardiac surgery or on the intensive therapy unit (ITU), a five-lead electrode cable is used. This allows the display of seven ECG leads at the same time, including I, II, III, aVR, aVL, aVF, and a single unipolar lead, e.g. the lateral chest lead V5.
Lead II or a modified V5 lead should be displayed for detection of arrhythmias, because it monitors down the axis of left ventricular (LV) depolarization in patients with a normal cardiac axis. If a preoperative ECG shows an axis deviation, it is best to display the lead that monitors most closely to the axis of depolarization, i.e. lead I in left axis deviation and lead III in right axis deviation. Coronary ischaemia is best monitored via a lateral chest lead. This is not possible with a three-electrode ECG, but a modified V5 lead can be obtained by placing the right arm lead on the manubrium sterni and the left leg electrode in the standard V5 position. Lead II must be selected on the monitor. This is known as the CM5 lead and is superior to an individual limb lead for the detection of coronary ischaemia.
Leads II and V4 or V5 should be selected for continuous monitoring. This combination has 90% sensitivity for the detection of inferior or anterior coronary ischaemia.
Epicardial ECG monitoring
After cardiac surgery, an atrial ECG can be monitored using the temporary epicardial atrial pacing wires that are placed on the heart. This is useful for the detection of atrial arrhythmias. Similarly, temporary ventricular pacing wires can be used to monitor the ventricular ECG. Epicardial ECG monitoring is also used during off-pump coronary artery surgery, during which the axis and amplitude of a conventional ECG may change owing to the positioning of the heart outside the pericardium, for inferior and lateral coronary grafting. Oesophageal or endotracheal electrodes have also been used during surgery. In the cardiac electrophysiology laboratory, a 15-lead ECG is continuously displayed to enable the mapping and ablation of cardiac arrhythmias. These include leads recording from the high right atrium (HRA), coronary sinus (CS), His bundle (HIS proximal and HIS distal), and right ventricular apex (RVA proximal and distal).
Telemetry is an automated communication process whereby data is collected remotely and transmitted to receiving equipment for monitoring. ECG monitoring via telemetry allows a patient to mobilize while being monitored for arrhythmias, or assessing response to antiarrhythmic drugs.
Subendocardial ischaemia is ‘demand ischaemia’ i.e. myocardial oxygen consumption is greater than supply in the subendocardial wall of the left ventricle, septum, or papillary muscles. It produces ST segment depression of >1 mm (0.1 mV), horizontal or down-sloping. It is measured 60 ms after the J point (Figure 7.3). It is non-localizing to coronary distribution, and leads V5 and V6 are the most sensitive for detection.
This is supply ischaemia and is associated with atherosclerosis involving a major coronary artery. It produces ST segment elevation. Significant ST elevation is considered to be elevation of 1 mm (0.1 mV) in two limb leads or 2 mm (0.2 mV) in two V leads. It is measured at the J point and localizes coronary artery distribution.
It has been shown that ECG monitoring on the ICU has a low sensitivity for detecting myocardial ischaemia, and frequent 12-lead ECGs are superior.
Interpretation and detection of arrhythmias
The PR interval represents the time for depolarization from the sinoatrial (SA) node to the ventricle. This is usually 0.2 s. A prolonged PR interval of >0.2 s signifies first-degree heart block. This may be a sign of coronary artery disease, rheumatic heart disease, digoxin toxicity, or electrolyte imbalance. It occurs postcardiac surgery, particularly after aortic valve replacement.
Second-degree heart block
There is progressive lengthening of the PR interval and then failure of conduction of an atrial beat. The following beat is conducted with a normal PR interval.
Mobitz type II
The PR interval of conducted beats is constant, but occasionally a P wave is not followed by a QRS complex, because the excitation fails to pass through the atrioventricular (AV) node and the bundle of His.
2:1 or 3:1 block
In 2:1 or 3:1 block, two or three P waves proceed each QRS complex. The PR interval is normal and constant in the conducted beats.
Wenckebach is usually a benign rhythm, but Mobitz type II, 2:1 or 3:1 block may herald complete heart block in the context of a myocardial infarction (MI).
Complete heart block
P waves march through the ECG at a constant rate, while the ventricular excitation is a slow ventricular escape rate. There is no relationship between the P waves and the QRS complexes, which are wide because of their ventricular origin. Complete heart block occurs acutely in the context of an MI or may be chronic due to fibrosis around the bundle of His. It is common postcardiac surgery, particularly valve surgery, and requires temporary pacing in the immediate postoperative period, prior to recovery or implantation of a permanent pacemaker.
Heart rate >100 bpm. Can precipitate myocardial ischaemia or LV failure owing to incomplete LV filling and decreased cardiac output (CO). It can be related to conditions such as anxiety, anaemia, hypovolaemia, fever, pain, hyperthyroidism, etc.
Paroxysmal atrial tachycardia
Originates above the ventricles from a site other than the SA node, but including the AV node. The rate is 150–250 bpm and is regular. There is one P wave per QRS complex, although the P wave may be hidden in the QRS complex or the T wave. It will compromise the patient if left untreated.
In atrial fibrillation (AF) there are no P waves and an irregularly irregular ventricular rate with normal QRS morphology. It can be associated with significant cardiac disease and can compromise CO (20% reduction in stroke volume [SV]) owing to inefficient ventricular filling. It can be diagnosed from the arterial waveform, which is irregular, with beat-to-beat variation in CO. If the AF has persisted for more than 48 h without therapeutic anticoagulation, a transoesophageal echo (TOE) should be performed prior to cardioversion to exclude thrombus in the left atrial appendage.
The atrial rate is between 250 and 350 bpm, with a ‘saw-toothed’ P wave appearance. The AV node cannot conduct at this rate so there is a variable block, i.e. 2:1, 3:1, or 4:1, resulting in a ventricular rate of 150, 100, or 75 bpm depending on the degree of block. A narrow complex tachycardia with a rate of 150 bpm on the ITU is likely to be flutter and is readily treated with a DC shock. If the arrhythmia has persisted for more than 48 h without therapeutic anticoagulation, a TOE must be performed prior to cardioversion to exclude thrombus in the left atrial appendage.
This arises from the area around the AV node. The P waves may not be seen, but the QRS complex is a normal shape because the ventricles are activated normally down the bundle of His.
Ventricular tachycardia (VT) is a rapid irregular depolarization from one or more ventricular foci. There are no P waves, the QRS complexes are wide, irregular, and vary slightly in shape. VT may or may not produce a CO, is life-threatening, and requires immediate treatment, according to the Advanced Life Support (ALS) protocol. The differential diagnosis of VT is supraventricular tachycardia (SVT) with aberrant conduction. This manifests as a wide complex tachycardia. Treatment with adenosine will differentiate between SVT with bundle branch block and VT. True VT will be unaffected by adenosine but SVT will be briefly interrupted.
A rapid irregular rhythm resulting from the discharge of impulses from one or more foci in the ventricles. There are no P waves, no QRS complexes, only bizarre erratic ventricular activity. There is no CO and immediate defibrillation according to the ALS protocol is required.
The radial artery is the most commonly used artery for arterial cannulation, as the hand receives collateral bloodflow via the ulnar artery. Direct cut-down onto an artery may sometimes be necessary in severely hypovolaemic patients. A non-invasive blood pressure monitor should be available to corroborate the mean arterial pressure (MAP). Mean pressures should be similar even if the systolic and diastolic values are different.
Allen’s test was originally described for assessing arterial bloodflow to the hand in thromboangiitis obliterans. The test is now used to assess ulnar arterial bloodflow prior to radial arterial cannulation or harvesting of the radial artery for coronary artery surgery; the ulnar and radial arteries are compressed at the wrist and the patient asked to clench tightly and open the hand. The hand appears white and blanched. Pressure over the ulnar artery is then released and the colour should return to the palm within 5–10 s. Delayed return of palmar bloodflow longer than 15 s is abnormal and may predict a risk of ischaemic changes if the radial artery is cannulated. The same test can be used prior to ulnar arterial cannulation but, instead, collateral circulation via the radial artery is assessed by releasing the pressure over the radial artery. However, the overall consensus is that Allen’s test is not discriminatory at a particular cut-off time. This does not imply that it should never be performed, but suggests that it should be replaced by more objective tests such as Doppler ultrasound.
• Awake or asleep: if the patient is awake, explain procedure and infiltrate with local anaesthetic.
• The radial artery can usually be found between the tendon of the flexorcarpi radialis and the head of the radius, but beware aberrant anatomy.
• Position the patient’s wrist in dorsiflexion, but do not overflex. An assistant can hold the wrist, or a roll of swabs and tape may be used.
• Insert the cannula slightly distal to the radial pulse at 45° to the skin (see Figure 7.4). When a flashback is obtained, advance the needle and the cannula further to compensate for the length of the bevel of the needle. Then carefully slide the cannula off the needle and advance it into the artery. Transfixion can be used, especially in babies or small children, and may have a higher success rate. However, the disadvantages include arterial damage and haematoma formation. The radial artery can also be cannulated using a Seldinger technique. This technique involves passing a Seldinger wire into the artery via a needle and then exchanging the needle for a plastic cannula prior to withdrawing the wire.
Alternative sites for arterial cannulation include:
• Brachial artery: this is an end artery supplying a large part of the forearm, therefore complications such as thrombosis will have severe consequences.
• Axillary artery: this is used in neonatal practice.
• Ulnar artery: the ulnar artery should not be used if the radial artery is occluded or has been damaged by previous cannulation attempts.
• Femoral artery: this is frequently used in babies and children, in patients undergoing bilateral radial artery harvest for coronary grafting, or in patients with arteriovenous shunts. A longer catheter is required, especially in obese patients. The catheter can easily become occluded if the hip is flexed and is more susceptible to infection.
• Dorsalis pedis artery: blood pressure will be 10–20 mmHg higher than in the central circulation.
Arterial pressure monitoring
To monitor arterial pressure, the pressure energy within the arterial cannula has to be transduced (changed or converted from one form of energy to another) to produce an electrical waveform. This requires a system that consists of the following components.
A flush system: this consists of a bag of fluid pressurized at 300 mmHg, a drip set, and a flow constrictor. This flushes the cannula with heparinized or normal 0.9% saline at a rate of 3–4 ml/h. The open system for blood gas sampling has one three-way tap for removal of dead space volume and then sampling. A closed system has a port for the removal of dead space volume beyond a separate sampling port. The closed system has been designed to reduce the risk of contamination of the sample by flush system fluid reduce the risk of bacteraemia, and minimize blood wastage because the dead space. Blood can be returned to the line without opening the system.
Excessive flushing of an arterial cannula should be avoided, especially in babies and children, to prevent air or debris entering into the arterial circulation. Flushing with a syringe may cause retrograde cerebral emboli.
Arterial cannula: 20 or 22G parallel-sided, stiff, Teflon cannula.
Connecting catheter: short and stiff saline-filled to reduce resonance. The number of three-way taps in the system should be kept to a minimum to decrease damping and minimize the risk of an accidental intra-arterial injection.
Transducer dome or diaphragm: this detects the small movement of saline to and fro along the catheter. The diaphragm receives this fluctuating pressure energy and converts it into an electrical signal, the amplitude of which depends on the degree of deformation of the dome. The transducer is placed at the level of the heart and zeroed in this position. Zeroing is carried out by exposing the transducer to atmospheric pressure through an open three-way tap and pressing ‘zero’ on the monitor.
Electrical monitor and connections: the input transducer leads to an amplifier and recorder. The frequency range of the arterial pressure waveform is between 0 and 40 Hz. The monitor must be able to respond adequately to this range of frequencies.
Normal arterial pressure waveform morphology
Digital readouts of systolic and diastolic blood pressure are displayed as a running average, which is updated at frequent intervals. The MAP is a calculated value (diastolic pressure + one-third of pulse pressure).
Resonance and damping
An arterial pressure waveform consists of a range of sine waves with different frequencies that are superimposed, producing the arterial pressure trace (Figure 7.5). The process of analysing complex wave patterns into a series of more simple sine waves is known as Fourier analysis.
The pressure-measuring system possesses a resonant frequency at which oscillations can occur. If this is <40 Hz, it falls within the range of the frequencies present in the blood pressure waveform and the sine wave is superimposed on the blood pressure waveform, producing a spiky, distorted, hyperresonant trace. If the resonant frequency of the system is outside the range of frequencies present in the blood pressure waveform, the problem is avoided. The resonant frequency can be raised by using a shorter stiffer arterial cannula. A hyperresonant trace will produce elevated systolic pressure readings and decreased diastolic pressure values, but the mean remains accurate.
A damped trace occurs if there is restriction of transmission of the blood pressure from the artery to the transducer diaphragm. This can be caused by clots, kinks, air bubbles, and excessively long or compliant tubing. The systolic pressure will be decreased, whereas the diastolic pressure will be elevated. Again, the mean pressure remains relatively accurate.
Difference between central and peripheral arterial pressure
The stiffness of the arterial tree increases with increasing distance from the aortic valve. The blood pressure wave becomes narrower and increases in amplitude in more peripheral arteries. Therefore, even in a supine patient the systolic pressure in the dorsalis pedis artery is higher than in the radial artery, which in turn is higher than in the aorta. The diastolic pressure similarly decreases peripherally and the pulse pressure widens. The effect of temperature and inotropes on the systemic vascular resistance can therefore influence the recording of arterial blood pressure. For these reasons, mean blood pressure values are frequently used in the ICU setting. Non-invasive blood pressure measurements do not always correlate with the invasive blood pressure reading, but the mean pressures should be similar. Beware, the ICU staff will favour the value that most closely approximates the target blood pressure!
Using the arterial pressure trace to estimate preload
Systolic pressure variation occurs in both mechanically ventilated and spontaneously breathing patients, where changes in intrathoracic pressure and lung volumes produce cyclical variations in blood pressure. In spontaneously breathing patients the blood pressure falls during inspiration. Pulsus paradoxus is an exaggeration of this occurring in severe asthma and cardiac tamponade. The reverse occurs in the ventilated patient, with the fall occurring during expiration. A systolic pressure variation of greater than 13–15% (approximately 10 mmHg) is an indicator of fluid responsiveness in mechanically ventilated patients in sinus rhythm with tidal volumes of >8 ml/kg. This ‘swing’ on the arterial pressure trace can be used to monitor the response to a fluid challenge. The position of a dicrotic notch on the downslope of the arterial waveform corresponds to closure of the aortic valve. The higher the notch, the better the cardiac contractility. Similarly, there are now many monitors that estimate CO by analysing and integrating the area under the curve of the pressure waveform (see section on ‘Non-invasive methods of CO monitoring’).
Arterial blood sampling
The erroneous use of glucose-containing solutions in arterial flush systems has led to falsely high glucose measurements, which have been treated with insulin, leading to hypoglycaemic brain injury and death. The Association of Anaesthetists of Great Britain and Ireland produced a safety guideline in 2014 with the following recommendations for prevention:
• Revision of hospital policies for arterial line sampling and staff education.
• Sodium chloride 0.9% with or without heparin should be the only fluids used in arterial line flush systems.
• Arterial infusion lines must be clearly identifiable. Lines should be colour-coded red with red caps on sampling ports. These should be of the unidirectional, needle-less variety to ensure a closed system.
• Arterial flush solutions must be prescribed and independently double-checked by a second practitioner when priming the arterial flush system, before it is attached to the arterial line.
• Pressurized flush bags should have a transparent front panel or netting as opaque bags can obscure the writing on fluid bags and prevent identification and checking of the flush solution. The flush infusion bag must be independently double-checked at each nursing handover.
• Sampling errors should be minimized by using a closed sampling system or minimizing the dead space in an open system. At least three times the dead space volume must be withdrawn from an open system prior to drawing the blood gas sample and all efforts made to minimize contamination of the sample with flush solution.
• When an arterial line is used to take blood samples, unexpected readings should provoke a review and resample from an alternative site before new therapy, e.g. insulin, is started.
The insertion of central venous catheters is a very common procedure on intensive care; however, to ensure minimal complications and the best patient outcomes, it is important that clinicians have a clear understanding of the indications, anatomy, and potential complications associated with these devices and their insertion.
Central venous access is almost universal in critical care patients.
• Monitoring of central venous pressure (CVP).
• Drug administration.
• Total parenteral nutrition.
• Fluid resuscitation.
• Insertion of temporary pacing wires.
• Insertion of pulmonary artery catheters (PACs).
• Renal replacement therapy.
• Lack of peripheral venous access.
These are relative, but include inability to identify landmarks, limited sites for access, previous difficulties or complications, severe coagulopathy, thrombocytopenia, and local sepsis. In addition, if an awake patient is unable to lie flat, central venous cannulation may be impractical without assisted ventilation.
Ultrasound guidance for vascular access
It is now recommended that ultrasound should be used to guide all central venous access. Ultrasound allows:
• Direct visualization of the vessels (artery and vein) and their associated structures.
• Identification of thrombosis, valve, or anatomical abnormalities.
• Identification of best target vessel.
• First-pass cannulation in the midline of a vessel directly avoiding other vital structures.
• Visualization of guidewire and cannulae entering vein.
• Reduction of puncture-related complications.
It is likely that the risk of catheter-related sepsis and thrombosis is reduced by limiting the number of needle passes with ultrasound.
Arteries can be distinguished from veins by their round cross-section, their non-compressibility, and their pulsatility. Veins, in contrast, show respiratory fluctuation and are easily compressible. To maintain sterility during vessel puncture, the ultrasound probe should be placed in a sterile plastic sheath. Sterile ultrasound gel is required both inside and outside the sheath. The use of ultrasound requires practice. Instruction should be sought before attempting to use it on a patient. Familiarity with the landmark approaches to the central veins is needed.
Internal jugular vein
Right internal jugular vein cannulation is associated with a lower incidence of procedural complications and a higher incidence of correct tip placement than other approaches. It is especially appropriate for patients with coagulopathy or those patients with lung disease in whom pneumothorax may be disastrous. It may be best avoided in those patients with carotid artery disease or those with raised intracranial pressure because of the risks of carotid puncture and impaired cerebral venous drainage. Internal jugular cannulation is associated with a higher incidence of catheter infection than subclavian cannulation, but both have a much lower infection rate than the femoral approach.
The internal jugular vein runs from the jugular foramen at the base of the skull (immediately behind the ear) to its termination behind the posterior border of the sternoclavicular joint where it combines with the subclavian vein to become the brachiocephalic vein. Throughout its length it lies lateral, first to the internal and then to the common carotid arteries, within the carotid sheath, behind the sternomastoid muscle.
Many approaches to the internal jugular vein have been described. Ultrasound will demonstrate the close association of the vein and carotid artery. Choose a site for puncture where the vein does not lie directly over the artery. A typical approach is from the apex of the triangle formed by the two heads of the sternomastoid:
• Slightly extend the neck.
• Turn the head slightly to the opposite side.
• Palpate the carotid artery at the level of the cricoid cartilage.
• Look for the internal jugular vein pulsation. If compressed, the internal jugular can usually be seen to empty and refill.
• To locate the vein, introduce the needle from the apex of the triangle at an angle of 30° and aim towards the ipsilateral nipple. The vein lies typically within 1.5–2 cm of the skin surface.
• Often when attempting to puncture the vein it collapses under the pressure of the needle and puncture is not recognized. The vessel may then be located by aspirating as the needle is slowly withdrawn. Blood will be aspirated as the needle tip passes back into the vein, which refills once the pressure has been removed.
External jugular vein
The external jugular vein lies superficially in the neck, running down from the region of the angle of the jaw, across the sternomastoid, before passing deep to drain into the subclavian vein. It can be used to provide central venous access, particularly in emergency situations, when a simple large-bore cannula can be used for the administration of drugs and resuscitation fluids. Longer central venous catheters can be sited via the external jugular, but the angle of entry to the subclavian vein often leads to inability to pass guidewires centrally and results in a high failure rate.
Subclavian vein cannulation is associated with a higher incidence of complications, particularly pneumothorax, and a higher incidence of incorrect line placement than internal jugular cannulation. It is, however, more comfortable for the patient long-term and the site can more easily be kept clean. The subclavian vein is a continuation of the axillary vein. It runs from the apex of the axilla behind the posterior border of the clavicle and across the first rib to join the internal jugular vein, forming the brachiocephalic vein behind the sternoclavicular joint.
• Position the patient supine (some people advocate placing a sandbag between the patient’s shoulder blades, which allows the shoulders to drop back out of the way).
• Identify the junction of the medial third and outer two-thirds of the clavicle.
• Introduce the needle just beneath the clavicle at this point and aim toward the clavicle until contact with bone is made.
• To locate the vein, redirect the needle closely behind the clavicle and toward the suprasternal notch.
• Ultrasound can be used to guide puncture of the vein using a more lateral approach. The axillary vein can be identified under the pectoral muscles at a depth of 3–4 cm in the average patient. Longer catheters (20 cm left and 25 cm right) are required by this approach. Supraclavicular approaches can also be used using both landmark- and ultrasound-guided techniques.
The femoral vein lies medial to the femoral artery immediately beneath the inguinal ligament. It is particularly useful for obtaining central access in small children and in patients with severe coagulopathy.
• Palpate the femoral artery.
• To locate the vein, introduce the needle 1 cm medial to the femoral artery close to the inguinal ligament. It is a common mistake to go too low, where the superficial femoral artery overlies the vein.
• Ultrasound should be used to identify the vessels (long saphenous vein, deep and superficial femoral arteries) and ensure that the vein is punctured near the inguinal ligament where the artery and vein lie side by side.
Central venous catheterization is almost universally achieved using a catheter over a guidewire (Seldinger) technique. This is associated with a lower incidence of incorrect line placement and complications than cannula-over-needle techniques.
• For internal jugular, external jugular, and subclavian veins, position the patient supine with 10–20° head-down tilt. This distends the vein to aid location and helps prevent air embolism.
• Monitor ECG in case of dysrhythmias.
• Universal precautions.
• Use aseptic technique, sterile gown, gloves, hat, and face mask.
• Prepare sterile field.
• Prepare all equipment.
• Check wire passes through the needle freely. Attach three-way taps to all open ports of the cannula. Flush the lumens with saline.
Prepare the skin using a single-use application 2% chlorhexidine in 70% alcohol preparation and allow to dry (povidone iodine-based if the patient is sensitive to chlorhexidine). Cover with a full sized sterile drape.
• Inject local anaesthetic to the entry site. Do not forget to anaesthetize suture sites as well.
• Identify the target vessel by ultrasound and/or landmark technique.
• Using a 10-ml syringe (partially filled with saline) and needle, enter the central vein by the chosen approach, maintaining suction on the syringe at all times.
• Pass the guidewire through the needle. This should pass freely and without any force into the vein. Watch for arrhythmias. Never pull the wire back through the needle once it has passed beyond the end of the bevel: it may shear off.
• Use a scalpel blade to make a small nick in the skin. Hold the blade up and cut away from the wire.
• If provided, pass the dilator over the wire into the vein. Then remove it, leaving the wire in situ.
• Pass the cannula over the wire into the vein. Make sure that, before you push the cannula forward, the wire is visible at the proximal end. Hold on to the wire at all times to prevent it being lost inside the patient.
• For an average adult patient the central venous cannula via the right internal jugular vein does not need to be inserted more than 12–15 cm. Check markings on the cannula. Many are 20 cm long and do not need to be inserted up to the hub.
• Draw back blood, check the colour, pulsatility, and the pressure of the back flow of blood, flush all the lumens of the line with saline, and attach appropriate antireflux needle-free access ports. At this point the patient can be levelled.
• Suture the line into place using the anchorage devices provided and cover with an adhesive sterile transparent dressing.
• If you appear to have missed the vein on the first pass, pull back slowly while maintaining suction on the syringe. You often find you have gone through the vein and can find it on withdrawal.
• Attach a transducer and display the waveform on the monitor.
• Dispose of your sharps and clear away your trolley.
• Obtain chest X-ray (CXR) to verify central position of the line and check for complications, including pneumothorax and haemothorax.
• Document the procedure in the patient’s notes.
Position on CXR
The catheter should lie along the long axis of the vessel and the tip should be in the superior vena cava (SVC) or at the junction of the SVC and right atrium, but ideally outside the pericardial reflection. The pericardium lies below the carina, so, ideally, catheter tips should be at or above the level of the carina. Catheters below this level may perforate the heart and cause cardiac tamponade. Catheters placed via the subclavian veins or left internal jugular vein must not be allowed to lie with the tip abutting the wall of the SVC. This may cause pain, perforation, and accelerated thrombus formation. Either advance the catheter to lie in the long axis of the SVC or pull it back to lie in the brachiocephalic vein. Bear in mind the limitations of CXR; it is useful to confirm central passage and no kinking. The close proximity of the SVC to the pleura, ascending aorta, and other structures means that confirmation of the true IV position cannot be inferred from a plain CXR.
Despite being a common procedure carried out on the ICU, there are several issues that can occur with central venous access, both during siting and in the longer term. Clinicians involved in the care of patients with central venous access need to be aware of both immediate complications but also longer-term ones and understand the possible solutions.
Cannot find the vein
Check position (ultrasound and/or landmarks) and try again. If unsuccessful do not persist with repeated passages of the needle in the hope of hitting the vein. You may have misinterpreted the landmarks or the vein may be absent, narrowed, or occluded (e.g. with thrombus).
Aspirating blood (needle in vein?) but cannot pass wire
Check needle position by drawing back on the syringe; good flow is essential. Adjust the angle of incidence of the needle to the vein to improve flow. Tip the patient further head down to expand the vein further. Try rotating the needle through 180° and draw back again. Remember the wire must pass easily without force. If this doesn’t work, repuncture the vein at a slightly different angle.
Is it arterial?
Occasionally, particularly if using a technique where the wire passes through the barrel of the syringe, it is difficult to know whether you have hit the artery or the vein. In this case, it is important to avoid passing a large central venous catheter into the vessel until you are sure. Consider the following:
• Remove the syringe from the needle and observe for pulsatile flow.
• Connect a transducer directly to the needle in the vessel and look at the waveform.
• Pass the wire into the vessel and remove the needle. Pass a 16/18G intravenous (IV) cannula over the wire into the vessel and remove the wire. Attach a transducer or manometer set directly to the cannula. When venous placement is confirmed, pass the wire back through the IV cannula and continue as before.
• Needle only, then simply remove and press.
• If large-bore cannula, then action depends on circumstances. If only in situ for a short period then it is usually safe to remove up to 8 French gauge (3 mm) and press until bleeding stops. In cases of a larger catheter in situ, carotid puncture in arteriopath, thrombus present, severe coagulopathy, or difficulty pressing (subclavian), leave in situ and give platelets and fresh frozen plasma before removing. Seek advice and consider the need for surgical exploration and removal under direct vision. Radiological stenting can also be used.
Complications of central venous cannulation depend in part on the route used but include those listed in Table 7.2.
Table 7.2 Complications of central venous cannulation
Erosion/perforation of vessels
Thoracic duct injury (chylothorax)
Embolization (including guidewire)
The management of pneumothorax depends upon the size of the pneumothorax and the patient’s condition, particularly whether ventilated or not. A small pneumothorax in an unventilated patient with good gas exchange may be observed or aspirated using a small-bore cannula and syringe with a three-way tap. Larger pneumothoraces, those that fail to resolve, or those that cause any impairment of gas exchange and/or haemodynamics require a formal chest drain. Any significant haemothorax should be formally drained as soon as possible. Once blood has clotted in the chest, drainage is difficult. Seek cardiothoracic/surgical opinion.
Bleeding around the puncture site can occasionally be a persistent problem. If this does not resolve with pressure, use a fine suture (e.g. 5/0 Prolene®) to tie a purse-string around the puncture site. This usually stops the bleeding.
Line colonization with bacteria and fungi is common, but there is no evidence that changing lines on a regular basis (e.g. every 5–7 days) is of benefit.
Changing catheters over a wire
If new central venous catheters are required, these should usually be placed at a clean site. Occasionally it may be necessary to change a catheter over a guidewire using an existing site. The technique is similar to that described earlier for placing any central venous catheter but should not be used in any patient with a catheter-associated bloodstream infection
The main problem is avoiding contamination of the new catheter.
• Cut sutures on the old line before scrubbing.
• Use universal precautions, aseptic technique, gown, and gloves.
• Clean and prepare the area.
• Pass the wire down the central lumen of the old central venous catheter. (Make sure that the new wire is longer than the old CVP line.)
• Remove the old catheter, leaving the wire in place, and send the tip of the old catheter for culture.
• Clean the puncture site with antiseptic solution.
• Use the wire to site the new line as required.
Removing central venous catheters
To remove central lines, ensure that all drugs and infusions have been stopped or relocated to other lines. If infection is suspected, send the tip of the line in a dry specimen pot for culture. Removal of central venous catheters can precipitate air embolism, pneumothorax, haemothorax, embolization of thrombus, and bacteraemia/sepsis. Make sure the puncture site is below the heart and apply pressure for at least 5 min; thereafter apply an occlusive waterproof dressing before sitting the patient up.
Choice of catheter
There are numerous devices on the market, including catheter-through-needle, catheter-over-needle, catheter-through-cannula, and catheter-over-wire. The choice of which to use should depend on the indication for its use, availability of equipment, and the skill of the operator.
Other choices include single or multilumen catheters, catheter material and long- or short-term use. The catheter with the minimum number of lumens essential for the patient’s management should be selected. Antimicrobial-impregnated devices are also available but their overall efficacy is still debated.
Intravascular catheter-related infections are a major cause of morbidity and mortality. Coagulase-negative staphylococci, Staphylococcus aureus, aerobic Gram-negative bacilli, and Candida albicans most commonly cause catheter-related bloodstream infection. Management of catheter-related infection varies according to the type of catheter involved. After appropriate cultures of blood and catheter samples, empirical IV antimicrobial therapy should be initiated based on clinical clues, the severity of the patient’s acute illness, underlying disease, and the potential pathogen involved. In most cases of non-tunnelled central venous catheter-related bacteraemia and fungaemia, the central venous catheter should be removed.
For management of bacteraemia and fungaemia from a tunnelled catheter or implantable device, such as a port, the decision to remove the catheter or device should be based on the severity of the patient’s illness, documentation that the vascular access device is infected, assessment of the specific pathogen involved, and the presence of complications such as endocarditis, septic thrombosis, tunnel infection, or metastatic seeding.
When a catheter-related infection is documented and a specific pathogen is identified, systemic antimicrobial therapy should be narrowed and consideration given for antibiotic lock therapy if the central venous catheter or implantable device is not removed.
Long-term central venous catheterization
There are many indications for patients to require longer-term central venous access. These include total parenteral nutrition, extended courses of IV antibiotics, repeated courses of chemotherapy and long-term conditions requiring repeated courses of IV therapy such as cystic fibrosis and pulmonary hypertension.
Depending on the indication and patient characteristics, there are several types of long-term central venous access devices.
• Hickman lines: one of the most common devices in use, these are tunnelled subcutaneously and have a Dacron® cuff which becomes coated in fibrin deposits, acting as an anchor.
• Groshong® catheters: tunnelled catheters similar to Hickman lines but with a valve device on the tip. This prevents air embolism and back-bleeding, negating the need for a heparinized solution to ‘lock’ the catheter with; however, there is a need to pressurize infusions.
• Peripherally inserted central catheters: these are fine-bore, flexible catheters placed in the central veins via the cubital, axillary, and subclavian veins. They can be placed in a ward or outpatient environment by appropriately trained nursing staff. However, they can be difficult to site centrally unless fluoroscopy is used, catheter tip site can vary significantly with arm movement, increasing risk of arrhythmia and thrombosis, and they have a higher thrombosis and premature failure rate than other forms of central access.
• Ports: fully implanted venous access devices, with a Hickman type catheter inserted into a central vessel attached to a subcutaneously implanted injection membrane that can be accessed using a non-coring needle. As these devices are fully implanted, the patient can bathe more easily, swim, and they have the lowest rate of infection.
Further long-term central venous access is used for patients undergoing chronic haemodialysis with several specially designed double-lumen lines that can be tunnelled for longer-term use.
The PAC is a haemodynamic tool able to measure intravascular and intracardiac pressures, CO, and mixed venous oxygen saturation (SvO2). It has a small (1.5 ml) balloon at the tip, just proximal to the distal lumen (Figure 7.6). When inserted into the right atrium this balloon is filled with air and carried by the flow of blood into the pulmonary vasculature, thus guiding the placement of the catheter. The insertion technique can be divided into two stages: 1) cannulation of a large vein with an introducer sheath utilizing the techniques described in the section on ‘Insertion of central venous catheters’; and 2) the passage of the catheter itself. Strict sterile technique should be maintained for the duration of the procedure.
Cannulation of a large vein
Selection of which vein to cannulate will be determined by operator familiarity, the presence of other indwelling vascular devices, and patient factors such as the necessity for cervical immobilization. As a general guide:
• The right internal jugular vein allows the shortest and straightest route to the right side of the heart.
• The left subclavian vein offers relatively unrestricted access to the heart.
• The right subclavian and left internal jugular veins require the catheter to navigate an acute angle to enter the heart.
• The femoral veins can be used if the other sites are unavailable, but passage of the catheter to the heart is technically difficult.
Most commercially available introducer sheaths are inserted by the modified Seldinger technique. Where appropriate, the insertion of the introducer sheath can be guided by ultrasound. It must be remembered that an introducer sheath is of large diameter (up to 8.5 French) and that the consequences of accidental arterial cannulation are likely to be more severe than would be the case with a standard central venous catheter.
A sterile sleeve is affixed to the introducer sheath through which the PAC will be passed. This will allow aseptic manipulation of the catheter once it is in situ.
Passage of the PAC
Before starting the insertion procedure, the PAC should be visually inspected for any obvious faults. The balloon should be inflated to test for integrity. Each of the lumens of the catheter should be flushed with saline to eliminate air bubbles. The markings on the catheter should be inspected to confirm understanding of the distance marking system used (Figure 7.7).
The passage of the PAC through the right side of the heart and into the correct position is monitored by observing the characteristic real-time pressure traces on a monitor. Thus the distal lumen of the catheter should be attached to a pressure transducer system, which allows continuous monitoring of the pressure waveform.
Knowledge of the normal pressures on the right side of the heart is invaluable when placing the PAC (Table 7.3).
Table 7.3 Normal right heart pressures (mmHg)
The PAC is inserted through the introducer sheath and advanced until a right atrial pressure trace is identified on the monitor. The distance to the right atrium varies depending on the insertion point of the introducer sheath. Typically the right atrial trace will be found at an insertion depth of 15–20 cm from the internal jugular veins, 10–15 cm from the subclavian veins, and 30–40 cm if the femoral route is used.
At this point the balloon is inflated with 1.0–1.5 ml of air and the inflation port locked. The distance marking on the catheter should be noted. The catheter is now slowly advanced whilst monitoring the pressure trace. The typical right ventricle pressure waveform should be seen after advancing the catheter approximately a further 10 cm, and this should change to the pulmonary artery waveform between 10 and 20 cm beyond that (Figure 7.8).
Note that the principal differences between the right ventricle and pulmonary artery waveforms are the higher diastolic pressures and the presence of a dicrotic notch in the pulmonary artery waveform. Both of these observations are because of the presence of elastic tissue in the walls of the pulmonary artery.
If the expected pressure changes are not seen after advancing the catheter the appropriate distance, it is possible that the catheter is coiling within the chamber and there is a risk of knotting. The balloon should be deflated and the catheter withdrawn slowly to the starting depth before inflating the balloon and trying again. The balloon should always be deflated before withdrawing the catheter to prevent damage to surrounding structures and minimize the risk of knotting.
Once in the pulmonary artery, the catheter should be advanced a further 10 cm or so until the typical wedge pressure trace is seen (Figure 7.8). This is the pulmonary artery occlusion pressure (PAOP). Once the PAOP has been measured the balloon should be deflated; if the typical pulmonary arterial waveform does not reappear then the catheter should be slowly pulled back until it does. The balloon should never be left inflated or the catheter in the wedged position because this can lead to erosion of the artery wall and subsequent artery rupture or to pulmonary infarction.
When the insertion procedure is complete, a CXR should be performed to confirm the correct position of the catheter and to identify complications of central venous access.
Pulmonary artery occlusion (wedge) pressure
When the PAC is in the wedged position there is an uninterrupted column of blood between the distal lumen of the catheter and the left atrium. At the end of diastole, when the mitral valve is open, this column of blood extends to the left ventricle, thus the pressure in the left ventricle at end-diastole is transmitted to the transducer system along an uninterrupted column of fluid. The PAOP can therefore be used as a marker of LV preload at the end of expiration when artificial intrathoracic pressure manipulations can be discounted.
There are certain circumstances in which the PAOP will not accurately reflect the LV end-diastolic pressure. These include:
• Mitral valve stenosis.
• Mitral valve regurgitation.
• Pulmonary venous obstruction, e.g. from pulmonary fibrosis.
• Tip of the catheter lying outside of Wests’ zone 3, i.e. the pulmonary capillary bed is compressed by the pressure within the alveoli at some point during the respiratory cycle.
Absolute contraindications to placement of a PAC include the presence of prosthetic tricuspid or pulmonary valves, endocarditis of tricuspid or pulmonary valves and the presence of right heart thrombus. Caution is advised in patients with recent cardiac arrhythmias and those with coagulopathy.
The complications of PAC placement can be divided into those caused by the initial vein puncture, which are common to all such procedures (see section on ‘Insertion of central venous catheters’), and those due to the passage or presence of the PAC itself (Table 7.4).
Table 7.4 Complications of PAC placement
Complications caused by the passage of the PAC
Complications caused by the presence of the PAC
Pulmonary artery rupture. The incidence of pulmonary artery rupture is 0.031%
Right bundle branch block occurs in up to 5% of insertions. Patients with pre-existing left bundle branch block are at risk of complete heart block
Pulmonary infarction may occur if the balloon is left inflated for prolonged periods of time or if the catheter migrates distally and occludes a small branch artery
Damage to the tricuspid and pulmonary valves
The catheter may knot if allowed to coil in one of the heart chambers during insertion
An echocardiography machine comprises two distinct modalities: ultrasound that is used to image cardiac structures and Doppler to examine aspects of blood flow. Transthoracic echocardiography (TTE) and transoesophageal echocardiography (TOE) have different advantages and indications in the evaluation of patients in the intensive care setting. TOE is a semi-invasive procedure with a small but definite risk of complications. In this chapter we describe use of echocardiography in a number of clinical situations, including pulmonary embolism (PE), valvular dysfunction, systemic embolization, infective endocarditis, unexplained hypoxaemia, pulmonary hypertension, and aortic dissection.
Physics of ultrasound
Sound is an example of a longitudinal wave oscillating back and forth through a transmitting medium at a fixed velocity, resulting in zones of compression and rarefaction. Ultrasound includes that portion of the sound spectrum above 20 kHz. Echo machines use frequencies of 2–10 MHz.
The wavelength (λ) is inversely related to the frequency (f) and directly related to the sound velocity (c) so that:
Sound velocity in a given material is constant but varies in different materials. Ultrasound propagates poorly in air. Sound velocity in blood is 1570 m/s, in soft tissue 1540 m/s, and in air 330 m/s.
Imaging by ultrasound
Ultrasound waves are generated by piezoelectric crystals in a transducer that vibrate when an alternating current is applied. Imaging is achieved by emitting ultrasound pulses from the transducer, which are reflected by a boundary between two tissue structures and received by the same transducer, generating a current that is processed to generate an image.
λ is a determinant of image quality as the spatial resolution is limited to approximately one λ. Therefore, shorter λs (obtained with higher fs as c is constant) produce better resolution. However, higher fs give reduced tissue penetration and thus a reduced image depth.
The pulsed ultrasound signal is described by the pulse repetition frequency. This must be set so that there is sufficient time for the pulsed wave to be transmitted, reflected, and received to display all objects uniquely within a typical 10-cm viewing window. There must only be a single pulse present between the transducer and the reflected object at any point in time to avoid range ambiguity. With a c in tissue of 1540 m/s this means that, at a depth of 10 cm, the pulse repetition frequency must be no more than 7.7 kHz.
The strength of reflection at an interface depends on the difference in acoustic impedance (AI) between two media. AI is the product of density and the c within the medium. There is a large AI mismatch between tissue and air, preventing imaging within the lung. This also occurs between the transducer and tissue, necessitating a layer of gel between tissue and transducer.
Doppler is used in echocardiography principally to look at aspects of blood flow. The Doppler effect is the apparent change in f that occurs when the source and observer are in motion relative to each other, with the f increasing when the source and observer approach and decreasing when they move apart. This shift in f from the transmitted to the received f is referred to as the Doppler shift (∆f) and is given by the Doppler equation:
where fo is the transmitting frequency; v is the velocity of blood flow; θ is the angle between the ultrasound beam and blood flow; and c is the ultrasound velocity in that medium.
The Doppler shift can therefore be used to measure the velocity of blood flow:
v is most accurately measured when θ is zero and the ultrasound beam is in line with the bloodflow. Doppler calculations cannot be made when the ultrasound beam is perpendicular to the flow of blood as the cosine of 90° is zero. As long as θ is <20° the measurement error is <6%.
Doppler-calculated velocities are displayed graphically on the echo machine with time on the x-axis and v on the y-axis. By convention velocities towards the transducer are displayed above the baseline of the vertical axis and velocities away from the transducer are displayed below the baseline.
Spectral Doppler comprises a continuous wave Doppler (CWD) and a pulsed wave Doppler (PWD). The CWD measures all velocities along the Doppler signal and is displayed as a filled envelope whereas PWD measures only selected velocities sampled from a particular chosen point on the Doppler pathway, resulting in display as an empty envelope.
Colour Doppler is a real-time colorized display of blood flow superimposed on a 2D image. This modality is used to identify turbulent flow within the heart that may occur, for example, with valvular pathology.
Indications for echocardiography in the ICU
• Haemodynamic instability: ventricular failure, hypovolaemia, vasodilation, acute valvular dysfunction, cardiac tamponade, PE.
• Aortic dissection.
• Infective endocarditis.
• Source of systemic embolus.
• Unexplained hypoxaemia.
Indications for TTE
• Imaging of the ascending aorta that is not visualized by TOE owing to interposition of the left main bronchus.
• Imaging of the left ventricle apex that is frequently foreshortened by TOE.
• When TOE is contraindicated.
Indications for TOE
• Conditions that prevent image acquisition with TTE such as in the presence of hyperinflated lungs, surgical dressings, wounds, drains, surgical emphysema, prone position, or excess fat tissue.
• Examination of posterior structures that are better visualized with TOE, such as thoracic aorta, pulmonary veins, left atrium, and mitral valve.
• Pathologies where TOE is more sensitive than TTE, e.g. left atrial appendage thrombus, interatrial septal defects, vegetations in infective endocarditis, cardiac tamponade, particularly with localized collections after cardiac surgery.
• In conditions where diagnosis and treatment require more detailed information, such as in suspected aortic dissection and complex mitral valve pathology.
Contraindications to TOE
These can be divided into relative and absolute (Table 7.5). Since there is no universal agreement as to which contraindications fall to the absolute and which to the relative category, the decision to perform TOE has to be based on individual benefit–risk assessment.
Table 7.5 Contraindications to TOE
Dislocation of the atlantoaxial joint
Oesophageal obstruction (stricture, mass)
Prior chest irradiation
Active upper gastrointestinal bleed
Clinical situations in the ICU
Echo can provide bedside diagnosis in the unstable ICU patient with massive PE when transport outside the ICU for angiography or computed tomography (CT) may be unsafe. TOE has a sensitivity of 80–92% and a specificity of almost 100%. Echocardiographic features of massive PE are:
• Acute cor pulmonale; specifically midcavity hypokinesia with preserved function at the apex, in contrast to right ventricle dysfunction owing to other causes in which wall motion is abnormal in all regions.
• Central emboli in the proximal pulmonary arteries; particularly the main and the right pulmonary artery (detection of left-sided emboli is limited by the poor propagation of ultrasound through air in the left bronchus).
Echo can be used to assess valvular lesions using 2D and 3D imaging, spectral Doppler, and colour flow mapping (CFM).
• To assess valvular appearance, whether bileaflet or trileaflet, leaflet calcification, and thickening (rheumatic).
• To assess secondary changes such as LV hypertrophy, LV dilatation and impairment, spontaneous echo contrast in the left atrium due to obstructed flow (± presence of thrombus) and atrial dilatation.
• Planimetry can be used to trace a valve area. This is particularly useful with aortic stenosis.
CFM detects turbulent flow through the restricted lesion and beyond, and gives an indication of its direction and extent.
It also detects the proximal isovelocity surface area (PISA), which is a flow convergence sphere that forms when a fluid passes through a small orifice. The radius of this sphere can be measured to quantify severity of stenosis. A radius greater than 1 cm signifies severe stenosis.
Vena contracta is the diameter of the narrowest part of the colour flow as it passes through the stenosis. Again, measurement of the width of vena contracta corresponds with severity of stenosis.
Measures the velocity (V) across a lesion, which can be used to calculate the pressure gradient.
Severe aortic stenosis has a mean ΔP 40–50 mmHg, and severe mitral stenosis has a mean ΔP >12 mmHg. Note that ΔP is flow-dependent and therefore underestimates stenotic severity when there is poor LV function.
This is used to assess valve appearance such as excess leaflet tissue in myxomatous mitral valve, to assess leaflet function (including chordal rupture and prolapse) and annular dilatation, and to look for evidence of endocarditis.
CFM is used to assess the size of the regurgitant jet. For mitral regurgitation a jet area >50% of the left atrial area is severe and for aortic regurgitation a jet width >40% of the LV outflow tract is severe. Note that this method tends to underestimate the severity of eccentric (wall hugging) jets of mitral regurgitation.
CFM is also used to measure PISA radius and vena contracta of the regurgitant jet.
TOE is the most sensitive and specific technique for determining the source and potential mechanism of systemic embolization for patients with cerebral ischaemic events or peripheral infarction.
TOE can identify cardiac sources of embolism, including intracardiac thrombi, vegetations, tumours, atrial septal defects, and atheromatous disease of the aorta.
The major criterion for the diagnosis of infective endocarditis is persistent bacteraemia with typical organisms with echocardiographic evidence of endocardial involvement.
Echocardiographic features of infective endocarditis are:
• An oscillating intracardiac mass which may be on a valve and/or supporting structure or in the path of a regurgitant jet or iatrogenic device.
• Intracardiac abscess.
• New dehiscence of a prosthetic valve.
• New valvular regurgitation.
TOE is more sensitive and specific than TTE for the detection of vegetations.
False-positive findings may occur from lesions that resemble vegetations, such as papillary fibroma, non-specific valve thickening or calcification, thrombus, Lambl’s excrescence or nodules of Arantius, and in patients with prosthetic valves: surgically severed chordae tendineae, fibrin strands, or periprosthetic material.
TOE can diagnose or rule out cardiac causes of hypoxaemia such as poor ventricular function, mitral regurgitation, pulmonary emboli, intracardiac shunts (patent foramen ovale or atrial septal defect) or even to detect pleural effusions.
Estimating pulmonary artery pressure
Systolic pulmonary artery pressures can be measured in patients with tricuspid regurgitation (TR).
The ΔP of the TR jet can be measured using CWD and, if the CVP is known:
Systolic right ventricle pressure = CVP + TR jet ΔP
Systolic right ventricle pressure = systolic pulmonary artery pressure (if no pulmonary valve pathology).
TOE provides high resolution real-time imaging of the aorta, resulting in high sensitivity (99%) for identifying dissection.
The unique advantage of bedside echo over CT or aortography lies in its portability, which is of particular value in unstable patients.
TOE can also identify complications of dissection such as extension of dissection into the coronaries, the presence of pericardial haematoma, the presence, mechanism, and severity of aortic regurgitation, the point of entry and exit between the true and false lumen, and LV function.
In suspected aortic dissection when TOE findings are equivocal or negative, aortography, CT, or magnetic resonance imaging should be performed in addition to TOE.
Echocardiography can be broadly categorized into basic and advanced, according to the indication, level of information required, and expertise of the practitioner. Basic echocardiography is a limited three-window 2D surface study. Advanced echocardiography would include in addition the transoesophageal route and all but the most basic 2D measurements—we shall refer to this as quantitative echocardiography for the purposes of this chapter.
Basic 2D echocardiography
The left ventricle (LV) is assessed in parasternal long and short-axis and apical views for end-diastolic diameter, shape, and gross abnormalities of contractility. The grading is a simple binary of normal or abnormal. This is integrated into the clinical context such as septic cardiomyopathy or MI.
The right ventricle (RV) is assessed in the apical and subcostal views and assessed for size and shape relative to the left ventricle, position of the interventricular septum, and free wall longitudinal contractility. In the context of acute cardiovascular collapse, gross right ventricle impairment is sought as an indicator of such conditions as pulmonary embolus.
The pericardium is inspected for evidence of significant pericardial fluid collection causing haemodynamic restriction. This may be evidenced by right atrial systolic or RV diastolic collapse.
Assessment of LV function
Measurement of both LV systolic and diastolic function that is physiologically relevant and practicable is essential and is increasingly linked to various critical care outcomes.
LV systolic function can be quantified by calculating an ejection fraction (EF) using: 1) 2D anatomical mode: and 2) Doppler modes.
These indices are dependent on good-quality views of the endocardial border, which may be difficult in TTE in critical care patients:
• Fractional shortening.
• Fractional area change.
• Ejection fraction (Simpsons disc method):
where EDV = end-diastolic volume and ESV = end-systolic volume.
Simpsons disc method is the current standard for EF. Ventricular volumes are obtained by tracing endocardial borders and modelling the ventricular cavity with a series of discs of uniform thickness. Views used in this method are the apical LV four- or two-chamber view using TTE or the midoesophageal four- or two-chamber views using TOE. A biplane LVEF measurement is the most accurate (Figure 7.9).
Other indices of LV systolic function
Doppler-derived measurements of LV function have the advantage of being less dependent on endocardial wall definition.
dP/dt. This utilizes the presence of any mitral regurgitation and measures the rate of increase in LV systolic pressure in early systole (between the 1-ms and 3-ms interval). It can be easily measured in apical TTE and mid-oesophageal TOE views and is relatively load-independent compared with the 2D measurements (Figure 7.10).
Tissue Doppler (TDI). This utilizes PWD to measure the velocity of longitudinal shortening of the LV wall just below the mitral annulus in the four-chamber view. This systolic component is known as S' and correlates well with EF. Septal and lateral velocities can be measured and averaged. It has limited value in the presence of basal segment systolic regional wall abnormalities and in mitral annular calcification.
Regional wall motion abnormality
It is important to recognize regional wall motion abnormality in critical care as acute coronary syndromes may be the reason for admission or may develop during the course of an admission.
Coronary lesions causing obstruction to flow can be mapped to areas of myocardium using a 17-segment model. Systolic thickening is graded as: i) normokinesia; ii) hypokinesia; iii) akinesia; iv) dyskinesia; and v) aneurysmal.
New regional wall motion abnormalities associated with haemodynamic instability are highly indicative of acute myocardial ischaemia.
Diastolic function essentially describes the ability of the left ventricle to relax following systole and allow the filling of the cavity from the left atrium. Diastolic function deteriorates with age and disease and is important in critical care as it is associated with failed weaning from the ventilator and the potential for fluid overload. There are two methods of determining diastolic function in critical care.
Mitral valve inflow PWD
The cursor is aligned with the mitral valve inflow in an apical four-chamber view and the marker is set at the tips of the mitral valve leaflets. The following measurements are made:
• E-wave—this corresponds to early diastolic rapid passive filling.
• A-wave—this corresponds to filling during atrial systole.
• DT—deceleration time, the rate of decease of passive filling.
• E/A ratio—this ratio is used to define mitral inflow diastolic function.
Left ventricle tissue Doppler
The echo machine is set to TDI mode and the cursor is aligned just below the medial and then the lateral mitral annulus as for S'. The velocity of the myocardial wall relaxation is recorded in the septal and lateral wall; this is known as E'. E' lateral and E' septal are averaged. An index E/E', which combines the pulse wave and TDI variables can be calculated, which gives a good estimation of left atrial pressure.
• Normal septal E' = normal lateral E', normal average E'.
• E/E' <8 corresponds to left atrial pressure >5–10 mmHg.
• E/E' >12 corresponds to left atrial pressure >12 mmHg (elevated).
SV and CO
A number of monitoring devices are available in the critical care unit to determine SV. This forms the basis of haemodynamic therapy to determine bloodflow and oxygen delivery to organs. Echocardiography is not a continuous output monitoring device but, because of its anatomical imaging capability, it can be a more accurate tool when other monitors are confounding. There are two methods of measurement:
1. Based on measuring LV volumes (EDV and ESV) using Simpsons disc method and using the patient’s heart rate (HR) in the equation:
2. Based on 2D and Doppler echocardiography; SV is given by the equation:
where VTILVOT is the velocity time integral through the LV outflow tract (LVOT), CSALVOT is the cross-sectional area of the LVOT, and VTILVOT is derived from Doppler analysis of the LVOT. This is obtained from the TTE apical three- or five-chamber view (Figure 7.11).
Assessment of RV function
RV dysfunction is frequently seen in critical care. RV function is altered by factors affecting RV afterload, such as high levels of intrathoracic pressure and increased pulmonary vascular resistance.
The most common causes of acute cor pulmonale are massive PE and acute respiratory distress syndrome (ARDS). Other causes of RV dysfunction include RV infarct, acute sickle cell crisis, air or fat embolism, and myocardial contusion.
RV assessment may alter treatment (fluid loading, vasopressors, thrombolytics) and is of prognostic value.
From an echocardiographic point of view, the RV cavity is flat in the four-chamber view or crescent-shaped in the short-axis view. Right ventricle walls are thinner than those of the left ventricle, and the interventricular septum acts as part of the left ventricle, moving towards it in systole.
In right ventricular failure the cavity enlarges, resulting in apical dilatation (in the four-chamber view) and rounder shape (in the short-axis view). RV enlargement is usually associated with inferior vena cava dilatation and loss of respiratory collapse, and tricuspid regurgitation with a jet velocity >2.5 m/s.
The LV and RV interact because of pericardial constraint. The sum of the diastolic ventricular dimensions has to remain constant. Any acute LV or RV dilatation is associated with proportional reduction in LV or RV diastolic dimension. RV dilatation can be quantified by measuring the ratio between the RV end-diastolic area (EDA) and LV EDA. Moderate RV dilatation corresponds to a diastolic ventricular ratio >0.6 and severe RV dilatation to a ratio >1.
Acute RV failure can lead to distortion of left ventricle size and geometry and as the right ventricle enlarges the septum is pushed towards the left ventricle, resulting in a small, ‘D-shaped’ LV cavity with compromised LV filling and function.
Longitudinal RV function (RV LAX) can be measured by tricuspid annular plane systolic excursion (TAPSE). TAPSE <1 cm = severe RV impairment.
Assessment of preload and volume responsiveness
Preload and volume responsiveness may be assessed by a number of echocardiographic methods. They can be broadly divided into static and dynamic indices.
Static indicators are determined under constant ventricular loading conditions.
Qualitative estimation of LV volume is only adequate at the extremes of volume status. Systolic obliteration of the LV cavity may be a sign of severe hypovolaemia, although it can be present in a number of other conditions. LV and RV EDA in general correlates poorly with filling pressure.
Right atrial pressure (RAP) can be estimated from assessment of the IVC. IVC diameter in expiration and response to an inspiratory sniff may give an indication of RAP. Volume responsiveness cannot be predicted accurately in the spontaneously breathing patient and respiratory disease and increased intra-abdominal pressure can confound IVC derived indices.
Dynamic echocardiographic indicators utilize varying loading conditions to determine where the patient is on the Starling ventricular function curve and thus predict volume responsiveness.
IVC distensibility. The IVC diameter is measured using M-Mode during fully synchronized mechanical ventilation and is defined as:
An index value of greater than 18% predicts fluid responsiveness. The usual considerations need to be taken into account when considering the IVC as an index: low tidal volume ventilation may render IVC and aortic VTI measurements less reliable.
In patients with preserved LV function, AV VTI respiratory variation can be used to predict fluid responsiveness. If there is >12% variation, then there is a 91% positive predictive value that the patient will be fluid-responsive (Figure 7.12).
This index can be used with a fluid challenge in mechanically ventilated patients. A change in AV VTI greater than 10% in response to 100 ml of fluid has a specificity of 95% for volume responsiveness.
The SVC collapsibility index can be used with TOE to predict fluid responsiveness. The SVC is measured in the upper oesophageal plane with M-Mode and a respiratory variation greater than 36% is significant. The index (Figure 7.13) is defined as:
In all of these dynamic indices it is essential that probe position does not vary with respect to the area of interest during respiratory excursion of the chest wall.
The Doppler effect was first described in 1842 by Christian Doppler, and describes the change in frequency as sound or light waves are reflected off a moving object. This relationship is described by the equation below. The Doppler effect can be utilized clinically to measure the velocity of bloodflow. An ultrasound beam of known frequency is directed at an angle to intersect the path of the bloodflow and reflected back by the red blood cells to an ultrasound detector. The change in the frequency of the reflected ultrasound waves is directly proportional to the velocity of the blood towards or away from the Doppler probe. Measurement of bloodflow using the Doppler effect has a variety of clinical applications, including the measurement of flow across heart valves and in the assessment of peripheral vascular disease. One of the most common applications of the Doppler effect in the intensive care setting is in the measurement of CO.
Doppler measurement of CO
The CO can be calculated by measuring either the velocity of blood through the descending aorta with a Doppler probe placed in the oesophagus or the velocity of blood in the ascending aorta with a probe in the suprasternal notch. The difficulties inherent in maintaining the position of the probe in the suprasternal notch have led to the dominance of oesophageal placement. Two commercially available oesophageal Doppler systems are the CardioQ™ from Deltex Medical and the HemoSonic TM 100 from Arrow International Inc.
Measurement of the velocity of blood (cm/s) in the descending aorta allows the flow of blood (ml/s) to be calculated provided that the cross-sectional area of the aorta is known. The CardioQ™ calculates the aortic cross-sectional area from a normogram based upon the patient’s age, height, and weight. The Hemosonic™ directly measures the diameter of the aorta by incorporating M-Mode ultrasound capability into the Doppler probe.
Calculation of the CO using the oesophageal Doppler method is dependent on five assumptions:
1. The distribution of blood caudally to the descending aorta and rostrally to the great vessels and coronary arteries maintains a constant ratio of 70% to 30%.
2. That a flat velocity profile exists within the aorta.
3. The estimated cross-sectional area is close to the mean systolic diameter.
4. There is negligible diastolic bloodflow.
5. The velocity of bloodflow in the aorta is measured accurately.
The descending aorta runs parallel and immediately adjacent to the oesophagus at a depth of ~35–40 cm from the teeth. This allows uninterrupted passage of ultrasound waves between the probe and the aortic bloodstream. The Doppler probe is most commonly introduced orally in sedated or anaesthetized patients, but may also be introduced nasally in awake patients with mild sedation and topical anaesthesia. The probe should be well lubricated prior to insertion to prevent air between the probe and the oesophageal wall from attenuating the signal. When the probe is at the appropriate depth, it is rotated slowly so that the tip of the probe is facing posteriorly (toward the aorta). Once in position, the probe is manipulated to achieve the optimal Doppler waveform; this will be the clearest signal with the highest peak velocity. The characteristic waveform should be a dark/hollow centre surrounded by red and then white in the trailing edge of the waveform. The waveform should be discrete and well contrasted with respect to the background (Figure 7.14). The gain control alters the contrast between the Doppler waveform and background noise and should be adjusted to give a black background and sharply contrasted Doppler waveform.
The positioning technique and waveform recognition require a degree of operator training; however, it has been shown that an acceptable degree of competence can be achieved after the insertion of only 12 probes. Once inserted, Doppler probes have been used in unconscious patients for up to 14 days without complication. The use of an oesophageal Doppler probe is safe in most patients; however, contraindications include the presence of clotting abnormalities, oesophageal varices, and recent oesophageal surgery. Readings obtained from an oesophageal Doppler probe may be inaccurate in the following circumstances:
• Coarctation of the aorta.
• Thoracic aortic aneurysm (especially if aortic cross-section is calculated from nomogram).
• Presence of a working epidural/intrathecal anaesthetic with subsequent vasodilatation.
In these circumstances, while the absolute figures from the Doppler probe may be inaccurate, trends in the readings can still be used to guide therapy.
Interpretation of waveform and variables
Transoesophageal Doppler probes are capable of measuring a number of haemodynamic variables, the interpretation of which can then be used to guide treatment.
The peak velocity of blood in the aorta gives a good estimate of the contractility of the myocardium. Normal peak velocity varies with age; at the age of 20 normal peak velocity is 90–120 cm/s, falling to 50–80 cm/s by the age of 70.
Stroke distance is the area under the velocity–time waveform; when multiplied by the aortic diameter this gives a good estimate of the SV. Because of slight beat-to-beat variability in SV, the reading is usually averaged over several beats. The number of beats used for this calculation is the cycle length. A cycle length of five beats is usual, but can be increased to improve the accuracy of SV estimation when there is marked beat-to-beat variability, e.g. atrial fibrillation. The CO is calculated from the SV multiplied by the heart rate.
Corrected flow time
The flow time is the duration of forward flow of blood in the aorta, i.e. the width of the base of the velocity–time waveform. The flow time varies with heart rate and can be corrected to a heart rate of 60 bpm by dividing the flow time by the square root of the cardiac cycle time (analogous to correcting the Q-T interval in ECG interpretation); this value is the corrected flow time (FTc). The normal FTc is 330–360 ms. Anything that impedes filling or emptying of the left ventricle will cause a reduction in FTc. Most commonly this is seen in hypovolaemia, but may also be seen with mitral stenosis, PE, and excessive use of vasopressors. A prolonged FTc is seen in the vasodilated circulation, e.g. in sepsis. A prolonged FTc of up to 400 ms may be regarded as normal in an anaesthetized patient, especially in the presence of a working epidural.
In addition to the variables described, the appearance of the velocity–time waveform itself may be diagnostic, as described in Figure 7.15.
To date there has been no reported mortality associated with the use of oesophageal Doppler. Morbidity rates are also low; there are case reports of inadvertent insertion into the lower respiratory tract and of epistaxis after nasal insertion.
Single readings may be diagnostic, but observation of dynamic changes with therapy are often more useful.
• Low SV. Fluid is best given as boluses, e.g. 200 ml, to construct a Starling curve for that patient. If there is less than a 5–10% increase in SV with each bolus it suggests that the plateau point has been reached.
• Low FTc. Most commonly represents hypovolaemia. Fluid challenge as above to achieve an FTc of 330–360 ms. Other causes, e.g. PE, will not improve with fluids.
• Low peak velocity. Consider positive inotropes.
• Low peak velocity and low FTc. Consider moves to reduce afterload, e.g. peripheral warming, reduction of vasopressors, glyceryl trinitrate.
The use of oesophageal Doppler to measure CO in critically ill patients compares favourably with the use of the thermodilution method using a PAC.
In 2011 the National Institute for Health and Care Excellence (NICE) appraised the available evidence for using the CardioQ oesophageal Doppler monitor and concluded that the device should be considered for ‘all patients undergoing major or high risk surgery, or other surgical patients in whom a clinician would consider using invasive cardiovascular monitoring’, citing reductions in length of stay, perioperative complications, and cost. Subsequent widespread adoption of the device was driven by a financial reward (CQUIN payment) from the Department of Health. Debate exists on the strength of the evidence base accepted by NICE, with critics citing the relatively few, small trials with heterogeneous patient groups considered. More recently, the role of perioperative CO-guided haemodynamic therapy has been questioned by the multicentre OPTIMISE trial, which showed no reduction in mortality or major complications (albeit using pulse contour analysis rather than oesophageal Doppler in the intervention arm). There does, however, remain a considerable body of evidence to support flow-based technologies in helping to improve care for high-risk patients undergoing surgery as part of a package of care.
Measurement of CO is an important tool to guide therapy in critically ill patients. This is particularly relevant when evaluating the effect of fluid therapy or inotrope administration. Arterial waveform analysis uses pulse pressure algorithms to calculate SV and derived parameters. This requires invasive measurement of arterial pressure, which is considered minimally invasive in comparison to the pulmonary artery catheterization, which was historically considered the best way to measure CO.
Nowadays, the use of pulmonary artery catheterization is reserved to very particular patients, whereas in usual critical care and anaesthesia practice, minimally invasive techniques are preferred.
From compliance to arterial load
The aorta is the main arterial vessel of interest as it is the part of the arterial tree that directly receives the SV. With each heartbeat, an amount of volume fills the aorta, generating an increase in pressure. The arterial pressure waveform has a sharp rise until a maximal systolic pressure, then it is followed by a decrease until the dicrotic notch (aortic valve closure), and then a diastolic decline until a minimal diastolic pressure. The difference between systolic and diastolic pressures defines the pulse pressure. Two major factors affects the pulse pressure:
1. The SV, which in turns depends on the venous return.
2. Arterial compliance (C), which is defined by the relationship between the changes in aortic volume (ΔV) and aortic pressure (ΔP):
With each heartbeat, the walls of the aorta expand to accommodate the increase in blood volume. If it were not for the elastic properties of the arterial tree (compliance), the pulse pressure would be very high and the SV would have to flow instantaneously into the peripheral circulation in systole with no flow in diastole. The arterial compliance is a property of the arterial wall, mainly related to the proportion of elastic and muscular fibres, which is reduced with age owing to structural changes in the arterial wall.
Compliance in the arterial tree is not a linear relationship and so a constant change in pressure does not generate a constant change in volume. As a general rule, compliance is higher for lower pressures and lower for higher pressures. In addition, with each heartbeat, the SV injected into the aorta has to displace a similar amount of blood into the circulation. The impediment to bloodflow through the vascular tree is called vascular resistance and depends mainly on the vessel diameter and by extension on the level of vasoconstriction or vasodilation and on blood viscosity. Peripheral vascular resistance also affects the blood pressure and pulse pressure.
In summary, arterial load, defined as the net arterial opposition to bloodflow pumped by the left ventricle, is composed of:
• The vascular compliance.
• The effective elastance (the ratio between pressure variation and volume variation).
• The peripheral resistance.
• The systolic and diastolic time intervals.
Together with the LV SV, all these factors define the systemic arterial blood pressure profile.
Arterial pressure waveforms and reflected waves
With each heartbeat, at first only the most proximal part of the aorta is distended and then this wavefront travels farther along the aorta and other territories of the arterial tree. This is the transmission of the pressure pulse in the arteries. This transmission is faster in less compliant vessels. Consequently, the arterial pulse pressure profile changes across the arterial tree owing to the different elastic properties of the arterial walls. The systolic pressure increases and the diastolic decreases (increase in pulse pressure), although MAP does not change significantly.
Finally, the arterial tree is a complex system with bifurcations and changes of diameter. Any discontinuity (changes in area, bifurcations, local changes in elastic properties, etc.) in the properties of the aorta and other arteries will cause the pressure pulse wavefront to produce reflected and transmitted waves. These reflected waves will also modify the arterial pressure waveform recorded. The ideal pulse pressure algorithm needs to take into consideration all these forces when transforming a pressure waveform into a volume waveform to estimate SV accurately.
Summarizing, an ideal algorithm should:
• Work independently of the arterial site from where pressure is monitored despite changes in waveform shape and pressure through the arterial tree from the centre to the periphery.
• Correct for non-linear compliance and take account of individual variations in aortic characteristics, thus giving an absolute CO.
• Not be affected by changes in vascular resistance causing changes in reflected wave augmentation of the arterial pressure.
• Not rely on identifying details of wave morphology, but be capable of detecting variations in pulse pressure and SV as it happens in hypovolaemia states.
• Be only minimally affected by the damping often seen in arterial lines.
History of pulse pressure algorithms
The first algorithm used in clinical practice was the Wesseling algorithm in 1983. This algorithm is based on the hypothesis that the contour of the arterial pressure waveform is dependent on SV and that this can be estimated from the integral of the change in pressure over time, considering the interval between the end of diastole to the end of systole. This is the first algorithm of pulse contour analysis used to determine CO. From this point, new algorithms have been developed and applied to different devices.
Pulse-contour methods are based on the findings of Windkessel and Wesseling Modelflow. This approach is based on integrating the area of the systolic part of the calibrated and compliance corrected waveform.
Pulse power analysis uses the law of mass conservation and assumes that, within a fixed circulatory system, following correction for compliance and calibration, there is a linear relationship between the net power change across the whole beat (the rate of doing work) and the net flow. This method is not based on the morphology of the arterial curve. By taking into account the whole beat, not only the systolic portion, the method becomes independent of the position of the reflected wave and the effect of arterial damping is much more limited.
These devices require an external calibration to report patient-specific absolute CO values. The calibration is based in the infusion of a dye (temperature, lithium) at a known concentration in the venous circulation and then its detection in the arterial circulation. The dimensions of the heart and the transpulmonary circulatory system are estimated based on anthropometric measurements. The dilution rate (thermodilution or lithium dilution) equals the flow rate. Calibrated devices are more accurate but their performance decreases the longer the time from the last calibration.
In clinical practice the first pulse contour method using a Wesseling-based algorithm was the PiCCO system (Pulsion Munich, Germany). PiCCO is a calibrated cardiac monitor that measures CO and several volumes (intrathoracic blood volume, global end-diastolic volume, and extravascular lung water). The pulse pressure algorithm analyses both the systolic and the diastolic part of the arterial pressure to study and determine the non-linear compliance and the relationship between flow and pressure. The calibration should be repeated every time there is a significant haemodynamic change. To perform a thermodilution, a central venous catheter needs to be in place. Cold saline (15 ml) is injected through a sensor that detects the temperature and time of the injection. A specialized arterial catheter (usually a femoral artery) detects the changes in blood temperature, producing a thermodilution curve. The same arterial catheter is used to monitor blood pressure and the arterial waveform is analysed by the pulse pressure algorithm.
PiCCO continuous CO has been studied and validated against the PAC in several conditions and has proven to be a reliable device, requiring calibration more often only in cases of major haemodynamic changes.
The algorithm also provides the user with the analysis of the variation in either SV (SVV) or pulse pressure (PPV). SVV and PPV represent the variation of SV and of the pulse pressure during the respiratory cycle. In selected patients, these indices have proven to predict the response to a fluid challenge. A large variation (>10–14%) identifies responders with good sensitivity and specificity.
The LiDCO™plus system is a CO monitor that measures CO combining two technologies: lithium transpulmonary dilution (as calibration) and pulse power analysis. The calibration is obtained via a lithium dilution technique. To perform the calibration, a dose of 0.3 mmol of lithium chloride is injected using either central or peripheral venous access. A sensor connected to an arterial line (there is no need for a specialized catheter) makes it possible to generate a concentration–time curve and the CO is then calculated. This value is then used to calibrate the pulse pressure algorithm. Further calibrations should be performed in the case of major haemodynamic changes.
Continuous CO of LiDCO has already been validated in several studies. This new algorithm has so far proven to be reliable in both surgical and intensive care patients. The LiDCO algorithm also allows the analysis of the SVV or PPV and of the systolic pressure variation.
These devices use a model based on sex, age, height, and weight, assuming standard resistance and vascular compliance, to derive CO values.
FloTrac and Vigileo
FloTrac (Edwards Lifescience, Irvine, CA, USA) is the name of the specific transducer incorporated into the Vigileo monitor. The most interesting characteristic of this device is that it does not need to be calibrated and it needs just an arterial line to work. The hypothesis is that, to calculate the effects of compliance and peripheral resistance on flow, all the necessary information can be obtained by the analysis of the arterial pressure waveform. Age, weight, and sex of the patient are the only variables that the clinician needs to input into the Vigileo monitor. The transducer (FloTrac) can be connected to any functioning arterial access. The algorithm recalculates the compliance continuously, thus bypassing the need for calibration. The Vigileo algorithm also allows the analysis of SVV and PPV to be performed. Since its first release in 2005, the algorithms have been improved and their accuracy and precision, in the last version, are comparable to other devices in hypodynamic or normodynamic states, whereas in hyperdynamic states this is not adequate yet.
The LiDCOrapid uses the same algorithm (PulseCO) as the LiDCOplus but without lithium-dilution calibration. Sex, age, weight and pulse pressure values are required to estimate SV. CO is derived from heart rate. CO is derived, in the same monitor, with an independent technique. Its accuracy is not comparable to the calibrated LiDCOplus. The main advantages are that it is minimally invasive and it is able to track rapid changes of CO and SV, making this tool ideal for the operating theatre where rapid changes in instravascular volume or vascular tone are common.
The ProAQT/PulsioFlex and the MostCare
These two devices use different algorithms to derive the CO. The ProAQT/PulsioFlex still requires complete validation. The MostCare is the only device that is not only non-calibrated, but also independent of the patient’s characteristics. Its reliability appears to be sufficient in cardiac patients. Both have the advantage of being less invasive and requiring less setting time than the calibrated devices.
Pulmonary thermodilution is one of the most commonly used methods for CO measurement. However, this technique is very invasive because a PAC is required. For this reason, new technologies developed over the past few years aim to obtain haemodynamic parameters using a less invasive approach. Several devices have been developed, based on different assumptions to calculate CO.
Non-invasive devices do not need any arterial catheterization. They include:
• Thoracic bioreactance.
• Vascular unloading technique.
• Pulse wave transit time.
• Radial artery applanation tonometry.
Oesophageal Doppler and echocardiography are previously discussed in the relevant sections.
Biorectance represents the phase shift in voltage across the thorax. This technology assumes that the phase shift depends on pulsatile flow and that the bioreactance signal is closely related to aortic bloodflow. Moreover, it is less dependent on intravascular and extravascular lung water in comparison with bioimpedance, where the signal is limited by fluid in the thoracic compartiment and by changes in peripheral vascular resistance.
The NICOM (Cheetah Medical, Portland, OR, USA) system uses this technology. Using four electrode patches, CO is calculated separately for the right and left side of the body and the final CO is derived from the average of the two values.
In a multicentre study evaluating NICOM (in a mixed population of patients) in comparison with PAC-derived CO, the authors observed a low bias for ICU patients. Another study, however, conducted in the operating theatre showed poor agreement between bioreactance and transpulmonary thermodilution.
• The technology is able to recognize a phase shift in voltage across the thorax and than calculate the CO. It might to able to provide CO data in the presence of mild cardiac arrhythmias.
• Disadvantage: electrical interference can disturb bioreactance measurements. The system provides CO readings averaged over 60 s and is therefore not able to indicate very rapid changes in CO.
Vascular unloading technique
This technique uses a small cuff that applies pressure to the finger and is able to record the arterial pressure waveform non-invasively. The CO is then calculated analysing the arterial pulse contour.
The artery’s diameter is assessed using an infrared light sent through the finger and measuring the light absorption by the blood with a light detector integrated into the finger cuff. This system controls the finger cuff pressure so that the blood volume in the finger artery and the artery diameter are kept constant during the cardiac cycle. When the artery’s diameter is constant, the intra-arterial pressure must be equal to the cuff pressure. The arterial waveform is derived from the pressure needed to maintain a constant volume in the finger artery during the cardiac cycle.
When the cuff pressure equals the arterial pressure, the transmural pressure is zero and the artery may be defined as unloaded. Obviously, the vasomotor changes influence the unloading volume of the finger artery and the device has to react to these changes.
There are several devices on the market using different ways to assess the optimal unloading volume:
• In the CNAP (CNSystems Medizintechnik AG, Graz, Austria), the optimal unloading volume of the finger artery is evaluated by a technology called ‘interlocking control loops’ that includes a beat-to-beat vasomotor elimination mechanism called VERIFI. The continuous arterial pressure signal obtained at the level of the finger is elaborated to match systolic and diastolic arterial pressure values obtained by oscillometric arterial pressure using a proprietary protocol.
• ClearSight technology (Nexfin system), developed by Edwards Lifesciences (Irvine, CA, USA), uses an algorithm that analyses the plethysmography waveform to evaluate the optimal cuff pressure. The brachial arterial pressure is mathematically reconstructed based on the arterial pressure assessed with the finger cuff.
There are several studies, in different settings (operating theatre/ICU) showing contrasting results: the vascular unloading technology depends on a high-quality arterial pressure signal. In patients with finger oedema, the arterial pressure signal can be disturbed. Moreover, during a hypoperfusion state, low CO, and high systemic vascular resistance, the measurements could be unreliable.
• Analyse the pulse contour of a non-invasively recorded arterial pressure waveform.
• Disadvantage: in patients with finger oedema, hypoperfusion, low CO, and high systemic vascular resistance, the measurements could be unreliable.
Pulse wave transit time
The esCCO technology (Nihon Kohden, Tokyo, Japan) is based on pulse wave transit time, defined as the time between the R-wave in the electrocardiogram and the pulse wave rise-point, assessed by pulse oximeter. It is based on the assumption that an inverse correlation exists between the pulse wave transit time and SV. This technology provides non-invasive continuous CO reading, assessed by analysis of the electrocardiogram, the pulse oximeter-derived waveform, and arterial pressure.
esCCO needs an external reference CO value as a calibration reference to start the measurement and this is the major limitation. Moreover, the approach, based on demographic data, heart rate, pulse wave transit time, and pulse pressure, needs further improvements.
Few validation studies have currently been conducted using this technology. In a multicentre study on 213 surgical and critically ill patients in comparison with continuous pulmonary artery thermodilution, the correlation coefficient was 0.79 and the bias and precision were 0.13 l/min and 1.15 l/min, respectively. This study suggests that the accuracy of this technology is comparable with thermodilution. However, the result also indicated that the accuracy may be affected by systemic vascular resistance and the authors concluded that further study is needed to establish the efficacy of the technology.
In a recent study, conducted on 25 patients with septic shock, esCCO was compared to echocardiography and showed a strong correlation with clinically acceptable limits of agreement. In another study, conducted on 35 cardiac surgery patients, the authors concluded that esCCO is easy to use and provides continuous CO measurements, but it has large percentage errors with a consistently positive bias in comparison to thermodilution.
• Provides non-invasive continuous CO reading assessed by analysis of the electrocardiogram, the pulse oximeter-derived waveform, and arterial pressure.
• Disadvantage: need a reference CO value at the start of the measurement as calibration.
Radial artery applanation tonometry
Radial artery applanation tonometry allows continuous (beat-by-beat) non-invasive recording of the arterial pressure waveform.
The T-line system (Tensys Medical Inc., San Diego, CA, USA) uses a proprietary algorithm for continuous recording of arterial pressure waveform. A sensor is placed over the radial artery and the raw arterial pressure signal is obtained in the optimal ‘applanation’ position, i.e. in the position in which the artery’s transmural pressure is zero. MAP can then be measured and the systolic and diastolic pressures are derived using an algorithm.
In a recent pilot study, a new autocalibrating algorithm was described; CO was simultaneously recorded using radial artery applanation tonometry and pulse contour analysis calibrated by transpulmonary thermodilution. The results indicated a good level of accuracy and precision in a cohort of patients admitted to intensive care deliberately selected based on their impeccable arterial pressure waveform.
The limitation of this technology depends on the quality of the arterial pressure waveform; rapid movement of the patient or medical staff can disturb the recording of arterial pressure waveform and CO determination.
• Continuous (beat-by-beat) non-invasive recording of the arterial pressure waveform.
• Disadvantage: the data depend on the quality of the arterial waveform and movements can disturb the recording.
CO is the product of SV and heart rate. SV is determined by preload, contractility, and afterload. Preload, in turn, is defined as the end-diastolic myocardial stretch (sarcomere tension), which in clinical practice is impossible to measure. Hence, some indicators of preload have been suggested: RAP and its surrogate, CVP, are considered static measurements of RV preload. The problem with this cardiocentric approach is that clinical values of CVP do not accurately reflect preload. For example, when cardiac function decreases, CVP increases immediately without changes in preload. Therefore, to better understand haemodynamics at the bedside, it is essential to take into account the venous side of the circulation. In this section, we will review some basic concepts of cardiovascular physiology that provide useful tools to manage patients in intensive care.
The venous system
The venous system is not merely a conduit of blood to the heart. It is an adjustable blood reservoir able to modify bloodflow according to changing metabolic demands. Veins contain 70% of total blood volume, whereas arteries contain only 13–18%, and capillaries 7%. Venous walls have a much larger compliance compared to arterial walls. As in other parts of the vascular system, the veins are composed of three basic histological layers: the tunica intima, the tunica media, and the tunica adventitia. The tunica media contains a variable thick layer of vascular smooth muscle cells. These cells can be stimulated to contract by multiple mechanisms: nervous reflex signals, hormonal stimulation, by stretch of the muscle, and several local chemical conditions.
Nervous reflex signals
Arterial hypotension reduces baroreceptor activity and provokes an increased sympathetic discharge, which causes venoconstriction, vasoconstriction, and increased contractility and heart rate. Most of the change observed in vascular capacitance takes place in the splanchnic bed.
The capillaries and the smallest postcapillary venules have little or no innervation. Yet smooth muscle is highly contractile in response to local chemical conditions. Some of the specific control factors that allow venodilation are:
• Lack of oxygen in the local tissues.
• Excess CO2 causes smooth muscle relaxation.
• Increased hydrogen ion concentration (acidosis) causes smooth muscle relaxation.
Hyperkalaemia, hypocalcaemia, adenosine, lactic acid, and increased body temperature can cause local vasodilation.
Some circulating hormones in the blood can affect smooth muscle cell contraction, such as noradrenaline, adrenaline, acetylcholine, angiotensin, endothelin, vasopressin, oxytocin, serotonin, and histamine. Circulating catecholamines can induce contraction of venules and veins of skeletal muscle and the mesentery; this is probably by catecholamines released from the sympathetic nerve terminals of the arterial side, which pass through the capillary bed and affect the venous system. Splanchnic and cutaneous veins have a high population of α1- and α2-adrenergic receptors, so they are very sensitive to adrenergic stimulation, contrary to skeletal and muscle veins. Stimulation of the β-adrenergic receptors of arterioles cause vasodilation but have little effect on the veins.
The mean systemic filling pressure
The heart, under normal conditions, pumps blood continuously into the aorta so that the pressure in the aorta remains high, averaging 80–100 mmHg. As the blood flows into the systemic circulation, the mean pressure falls progressively as low as the level of the RAP. When the heart stops, the arterial pressure falls and the RAP progressively increases until a certain point when there is no blood motion. At this point, the pressure at every single point of the circulatory system will be the same. This is called the mean systemic filling pressure (Pmsf), and is the pressure related to the intravascular volume and the mean systemic capacity of the system. When the heart is beating, the pressure in the small veins (<1 mm) and venules is considered the ‘pivoting point’ of the system. The importance of this pivotal pressure is that it provides a quantitative measurement of the intravascular filling status independent from cardiac function, and its value is equal to the Pmsf.
As 70% of the blood is in the venous reservoir, let us imagine this ‘blood reservoir’ as a distensible compartment. The volume required to fill a distensible tube, such as a tyre or a blood vessel, with no pressure rise, is called the ‘unstressed’ volume. Further volume expansion will imply necessarily a pressure rise and an elastic distension of the wall of the tube, which depends on the compliance of the wall. This volume is the ‘stressed’ volume. Then, the Pmsf is only related to the stressed volume.
How does the Pmsf affect CO?
The rate of bloodflow is determined by the difference in pressure between two points of the cardiovascular system, and not by any single pressure at any point. Given that most blood is in the venous reservoir, the pressure at this point is particularly interesting. Venous return is defined by three parameters: the Pmsf, the RAP, and the resistance to venous return (RVR). This can be mathematically represented as follows:
The gradient of pressure between Pmsf and RAP is directly proportional to the venous return. Guyton described venous return curves changing the RAP under isovolumetric conditions (Figure 7.16). From these curves one can spot that the greater the RAP, the lower the venous return. As, during steady conditions, CO and venous return are equal, Pmsf plays an important role in the regulation of CO.
The CVP is a readily available measurement in any patient who has a central line. In a normal heart, with transducers correctly zeroed and levelled and at end of expiration, CVP could be a good estimate of RAP, which, during diastole, is the same as the RV pressure. CVP and RAP, however, are affected by many other factors that have nothing to do with the preload of the right ventricle. CVP (and RAP) is regulated by a balance between: a) the heart performance; and b) venous return. In turn, some of the factors that can increase venous return are: 1) increased blood volume; 2) increased venous tone (or decreased venous capacitance); 3) decreased peripheral resistance (relaxation of arterioles), which allows rapid flow from arterial to the venous side. These same factors also affect CO because CO is venous return.
• a wave: reflects the atrial contraction and can be identified just after the p wave of the ECG.
• z point: is the inflection point between the a wave and the c wave. It is not always present, but can be located at the end of the QRS complex.
• c wave: reflects the tricuspid valve elevation into the right atrium.
• x descent: corresponds to the relaxation of the atrium and a downward movement of the contracting right ventricle.
• y descent: triscuspid valve open in early ventricular diastole.
Table 7.6 Summary of the changes on CVP curve in some common pathological conditions
Lack of a wave
Prominent c wave
Cannon a wave
Tall systolic c-v wave
Loss of x descent
Tall a and v waves
Steep x and y descents
M or W configuration
Dominant x descent
Attenuated y descent
Monitoring Pmsf at bedside
It is not possible to measure Pmsf in patients with an intact circulation. However, some methods have recently been proposed to calculate or estimate this parameter.
• Construction of venous return curves observing the changes in CO and CVP during positive pressure recruitment manoeuvres, and then extrapolating the RAP value to zero CO.
• A rapid occlusion of the circulation in the arm (Pmsf-arm), also called the stop-flow arterial venous equilibrium method. Once the arterial (Pa) and venous pressures (Pv) in the arm equilibrate, the pressure measured would be Pmsf.
• A mathematical algorithm to build a cardiovascular model that uses the patient’s MAP, CVP, CO, and anthropometric data and estimates the Pmsf-analogue (Pmsa).
Since venous return equals CO, in clinical practice CO, along with CVP changes, can provide most of the information required to understand venous return. However, without understanding how venous tone works understanding how to use CVP can be difficult. Figure 7.16 provides a schematic idea about four possible changes in CO and CVP and possible explanations. Pmsf provides complementary information, which could help to discriminate changes in complex scenarios.
The venous system plays an important role in haemodynamic stability. Two variables are available to monitor the venous return: the CVP and the Pmsf. The Pmsf can now be estimated and is the pressure of the pivot point of the circulation, where the pressure is independent of bloodflow. This pressure is the driving pressure of the circulation and affects, along with cardiac function, venous return. The CVP is affected by the venous return and the cardiac function. Therefore, changes in CVP can only be understood together with CO monitoring.
Since only half of the patients—even with haemodynamic instability—are able to respond to fluid loading and since fluid overload may cause harm, predictors of fluid responsiveness are required to distinguish between patients who can benefit from fluid and those for whom fluid is useless and hence deleterious. The best method for predicting fluid responsiveness is to induce a transient change in cardiac preload and to observe the resulting effects on SV or CO. Observing the respiratory variation of haemodynamic signals has emerged as an excellent method for assessing fluid responsiveness without administering fluid. Among the different indices, PPV has become the most popular dynamic index of fluid responsiveness since it needs only an arterial catheter to be obtained and since numerous bedside monitors calculate it automatically and display its value in real-time. Respiratory variations of the subaortic flow, or the descending aortic bloodflow, or the pulse contour-derived SV, or the superior of IVC diameters are good alternatives.
Owing to several limitations of these dynamic indices to predict fluid responsiveness, other preload challenges such as end-expiratory occlusion and passive leg raising have been proposed and shown to be good alternatives. Recent randomized studies in the perioperative setting showed that using dynamic variables of preload responsiveness (PPV or SVV) is associated with fewer postoperative complications compared to standard care.
How to define fluid responsiveness?
Fluid responsiveness is defined as the ability of the heart to increase its SV significantly in response to volume expansion because of the presence of biventricular preload reserve.
Why and when to predict fluid responsiveness?
Three different scenarios must be distinguished.
1. Patients admitted with evidence of acute body fluid losses: the diagnosis of hypovolaemia is almost certain and the presence of clinical signs of haemodynamic instability (e.g. hypotension, low arterial pulse pressure, tachycardia, oliguria, mottled skin, increased capillary refill time, altered mental status) strongly suggests that a positive haemodynamic response to volume resuscitation will occur.
2. Patients admitted with early severe sepsis or septic shock: fluid resuscitation is necessary and there is no need for sophisticated parameters to predict volume responsiveness since a positive haemodynamic response is always expected at this early stage.
3. Patients who have been in the ICU for several hours or days and who experience haemodynamic instability that requires urgent therapy: only one-half of these patients are preload-responsive since they have frequently already been resuscitated. Further fluid infusion has the potential for promoting pulmonary oedema in cases of increased pulmonary permeability. Therefore, predictors of volume responsiveness are needed to distinguish between patients who can benefit from fluid resuscitation and those in whom fluid loading can be useless and even deleterious.
How to detect fluid responsiveness?
Because clinical evaluation is of poor value for detecting fluid responsiveness, advanced parameters based on physiological approach have been proposed to help the clinician in the therapeutic decision-making process.
Static markers of cardiac preload
Considering the Frank–Starling relationship (SV versus ventricular preload), the response to volume infusion is more likely to occur when the ventricular preload is low than when it is high (Figure 7.18). However, except for their lowest and highest ranges, static markers of ventricular preload such as filling pressures or cardiac chamber dimensions fail to predict fluid responsiveness reliably. Indeed, there is not one single curve but several relating SV to cardiac preload, depending on the ventricular contractile function. Thus, a given value of cardiac preload can be associated with preload responsiveness in case of normal cardiac function or with preload unresponsiveness in case of decreased contractility. Accordingly, CVP and other measures of preload have repeatedly been shown to be poor predictors of fluid responsiveness.
Dynamic markers of preload responsiveness
Heart–lung interaction indices
Mechanical insufflation may decrease RV filling through a decrease in venous return. This results in decreased RV SV when the right ventricle is preload-dependent. Because of the long pulmonary transit time, the filling of the left ventricle senses the decrease in RV SV two or three heartbeats later and thus generally during expiration. This results in decreased LV SV during expiration if the left ventricle is also preload-dependent. Thus, cyclical changes in LV SV occur in cases of biventricular preload dependence. Since fluid responsiveness occurs only in cases of biventricular preload dependence, it has been postulated that the magnitude of the cyclical changes in SV or of its surrogates such as arterial pulse pressure should correlate with the degree of fluid responsiveness. Thus, numerous dynamic indices using heart–lung interactions are proposed to assess fluid responsiveness in patients receiving controlled mechanical ventilation.
Respiratory variation of arterial pulse pressure
Since the arterial pulse pressure (PP, systolic pressure minus diastolic pressure) is proportional to LV SV, the magnitude of PPV has been proposed as a predictor of fluid responsiveness in patients equipped with an arterial catheter. The PPV is calculated as the difference between the maximal (PPmax) and the minimal (PPmin) value of PP over a single respiratory cycle divided by the average of the two values and expressed as a percentage:
In numerous categories of mechanically ventilated patients (septic shock, cardiac surgery, liver transplantation, etc.), PPV has been demonstrated to be an accurate marker of fluid responsiveness with a threshold value around 12%. Moreover, the fluid-induced decrease in PPV correlates well with the percentage changes in CO induced by 500 ml fluid infusion. Thus, PPV can be used not only to predict fluid responsiveness but also to assess the actual haemodynamic response to volume infusion.
Pulse contour-derived SV variation
Pulse contour CO monitoring devices can continuously measure and display SVV, which represents the variation of pulse contour SV over a floating period of a few seconds. In patients fully adapted to their ventilator, SVV accurately predicts fluid responsiveness. Threshold values of SVV range between 9.5% and 12.5%.
Respiratory variation of the peak Doppler aortic blood velocity
Doppler echocardiography allows aortic blood velocity to be measured at the level of aortic annulus (LVOT). A value of the respiratory variation of peak Doppler aortic blood velocity (ΔVpeak) >12% is predictive of fluid responsiveness.
Respiratory changes in IVC diameter
The IVC diameter can be measured using echocardiography from short-axis or long-axis subcostal views. A value of respiratory changes in IVC diameter (ΔDIVC) ([maximal diameter –minimal diameter)]/mean of the two values) >12% is predictive of fluid responsiveness.
The diameter of the SVC can be measured using TOE. A value of SVC collapsibility (difference between maximal and minimal diameter divided by maximal diameter) >36% allows excellent prediction of fluid responsiveness.
Respiratory changes in pulse oxymetry plethysmographic waveform amplitude
A value of >14–15% is predictive of fluid responsiveness in mechanically ventilated patients. Nevertheless, in patients receiving vasopressors, the predictive value of the variation of the plethysmographic signal is less predictive than PPV and SVV.
Limitations of using respiratory variability of haemodynamic indices
Volume responsiveness is a physiological phenomenon related to a normal preload reserve. Thus, detecting volume responsiveness must not systematically lead to the decision of infusing fluid, in particular in the absence of clinical or biological signs of tissue hypoperfusion.
The respiratory variation of surrogates of SV cannot be used to assess fluid responsiveness in patients receiving a tidal volume lower than 7 ml/kg, in patients with spontaneous breathing activity, in patients with arrhythmias (except for respiratory variation of vena cava diameter), in patients with low lung compliance, and during open-chest conditions.
The end-expiratory occlusion test
This test is an alternative to variability indices for detecting fluid responsiveness. During mechanical ventilation, each insufflation interrupts the venous return. In patients with cardiac arrhythmias or spontaneous triggering of the ventilator, an increase by more than 5% in arterial PP or in pulse contour-derived cardiac index during a 15-s end-expiratory occlusion enables prediction of fluid responsiveness with a good accuracy. The end-expiratory occlusion test can be used in patients with low lung compliance and in patients ventilated with positive end-expiratory pressure.
Passive leg raising
In cases where dynamic indices using heart–lung interaction cannot be used (see section on ‘Limitations of using respiratory variability of haemodynamic indices’), the passive leg raising (PLR) manoeuvre can be used as a test to detect fluid responsiveness. This test consists of moving the patient’s bed from a semi-recumbent position to a position where the legs are lifted by 45° and the trunk and head are horizontal. This induces a gravitational transfer of venous blood from the lower limbs and the abdominal cavity toward the intrathoracic compartment and thus increases cardiac preload. An increase in CO or SV in response to PLR is suggestive of the presence of cardiac preload responsiveness and hence predictive of fluid responsiveness. PLR mimics fluid challenge but, unlike fluid challenge, no drop of fluid is infused and its effects are rapidly reversible. The main advantage of PLR is that it can be used in patients with spontaneous breathing, arrhythmias, low tidal volume, and low lung compliance. Because of the short duration of this test (around 1 min), a real-time CO or SV measurement method is mandatory. An increase in CO >10% is highly predictive of fluid responsiveness.
The mini-fluid challenge: in rare situations where PLR may be inappropriate, such as in patients with trauma or after major surgery, a mini-fluid challenge can be proposed to predict fluid responsiveness. A 10% increase in subaortic VTI, measured by TTE after rapid infusion of 100 ml of fluid accurately predicted fluid responsiveness.
Minimally invasive as well as non-invasive dynamic parameters testing the sensitivity of the heart to changes in intrathoracic pressure during a mechanical breath can accurately discriminate between responders and non-responders to fluid infusion. In cases where these indices are not interpretable (inspiratory efforts, cardiac arrhythmias, low tidal volume, low lung compliance), the end-expiratory occlusion test and the response of SV (or its surrogates) to either PLR or to a mini-fluid challenge is helpful to predict fluid responsiveness. In the perioperative setting, the use of dynamic variables of preload responsiveness (PPV or SVV) was shown to be associated with fewer postoperative complications compared to standard care. In the context of septic shock, the use of PPV or PLR was shown to be associated with a decrease in blood lactate and a trend toward better outcome compared to standard care.
Multiple choice questions and further reading
Interactive multiple choice questions to test your knowledge on this chapter and additional further reading can be found in Appendix Chapter 7 Multiple choice questions and further reading