Chapter 34 Respiratory monitoring

Clinical examination and the trend of vital signs such as respiratory rate (f), and quantity and nature of sputum are extremely important in managing patients with respiratory disease. In particular, clinical examination should look for evidence of excessive inspiratory and/or expiratory pleural pressure changes and effort such as accessory muscle use, tracheal tug, supraclavicular and intercostal indrawing, paradoxical abdominal movement (which is suggestive of diaphragmatic fatigue ¹) and pulsus paradoxus. During spontaneous ventilation an excessive fall in blood pressure during inspiration (> 10 mmHg) is found in a number of conditions such as cardiac tamponade, cardiogenic shock, pulmonary embolism, hypovolaemic shock and acute respiratory failure. A curvilinear relationship exists between the fall in blood pressure and the change in pleural pressure during inspiration; however, there is marked variation between individuals.² Consequently, pulsus paradoxus is most useful in following trends; a reduction in the degree of paradox may be due to improvement and a fall in the negative pleural pressure needed for ventilation, or due to respiratory muscle insufficiency, and an inability to generate the same negative pleural pressure.

Additional information can be gained from blood gases and pulse oximetry (Chapter 14), and capnography, ventilatory pressures and waveform analysis in patients receiving respiratory assistance. This chapter will focus on tests of respiratory function that are directly relevant to critically ill patients.

Monitoring gas exchange

Oxygenation

This is reviewed in Chapter 14, and is only briefly discussed here. Hypoxaemia may be due to a low partial pressure of inspired O₂ (rare), hypoventilation, diffusion impairment (rare), ventilation–perfusion mismatch and shunt. Inert gas analysis has been used to quantitate mismatch, and has demonstrated that hypoxaemia in acute respiratory distress syndrome (ARDS) is predominantly due to alveoli that are perfused but not ventilated (shunt),³ consistent with computed tomography (CT) scan evidence of increased dependent lung density. However, inert gas analysis remains a research tool, and less direct methods, such as the alveolar gas equation, are used to assess hypoxaemia:

where PAO₂ is the alveolar PO₂, and this is usually simplified to:

where 760 is atmospheric pressure in mmHg, and 47 is the saturated vapour pressure of water at 37 °C since gas at the alveolus is fully humidified. The normal PAO₂ to PaO₂ gradient is less than 7 mmHg, but increases to 14 mmHg in the elderly. This normal A–a gradient is due to some venous admixture through the lungs, and a small right-to-left shunt through both the bronchial veins and the thebesian veins of the coronary circulation. This equation removes hypercarbia as a direct cause of hypoxaemia, and an increase in the A–a gradient will usually be due to mismatch or right-to-left shunt. A commonly used alternative measure of hypoxaemia is the PaO₂/FiO₂ ratio. However, this does not account for the effect of a raised PaCO₂, and both measures are influenced by a number of factors (e.g. cardiac output, haemoglobin (Hb), FiO₂) in addition to the extent of venous admixture, which may be estimated from the intrapulmonary shunt equation:

where is the intrapulmonary shunt blood flow, is the total pulmonary blood flow, Cc′O₂ is the end-capillary O₂ content calculated from the PAO₂, and CaO₂ and CvO₂ are the O₂ contents of arterial and mixed venous blood respectively.

Carbon dioxide

PaCO₂ is determined by alveolar ventilation , and CO₂ production :

where is the minute ventilation minus the wasted or dead-space ventilation . The modified Bohr equation (assuming PACO₂ = PaCO₂) calculates the proportion of the V_T which is wasted ventilation (i.e. physiological dead space: V_Dphys):

where is the mixed expired , and V_Dphys is composed of anatmical dead space (V_Danat) and alveolar dead space (V_Dalv) – notionally due to alveoli that are ventilated but not perfused. Normally V_Dalv is minimal and V_Danat comprises 30% of V_T. Since the volume of an endotracheal tube is less than the mouth or nose, and pharynx, intubation may reduce V_Danat; however, when the connection from the endotracheal tube is taken into account there is little change in dead space. Positive-pressure ventilation increases dead space by distension of the airways increasing V_Danat, and through a tendency to increase alveoli that are ventilated but not perfused. In patients with ARDS, marked increases in V_Dalv lead to marked increases in the V_Dphys/V_T ratio (exceeding 0.6), which is an independent prognostic factor.⁴

Capnography

Capnography measures and displays exhaled CO₂ throughout the respiratory cycle, with sampling usually by a mainstream sensor since sidestream systems tend to become blocked by secretions. However, when capnography is used in non-intubated patients, sidestream sampling is commonly used (e.g. modified nasal cannulae). Infrared spectroscopy measures the fraction of energy absorbed and converts this to a percentage of CO₂ exhaled. During expiration the capnogram initially reads no CO₂, but as anatomical dead space is exhaled there is a rise in the exhaled CO₂ to a plateau which falls to 0% CO₂ with the onset of inspiration. In patients with significant respiratory disease a plateau may never be achieved. The end-tidal CO₂ (PetCO₂) is the value at the end of the plateau, and is normally only slightly less than the PaCO₂. However, this gradient will increase when alveolar dead space (V_Dalv) increases, such as low cardiac output, pulmonary embolism and elevated alveolar pressure. Consequently the PetCO₂ may not reflect PaCO₂ in critically ill patients. Nevertheless, in a stable patient the gradient will be fairly constant, and can be used to guide during transport,⁵ and when other factors including the adequacy of minute ventilation are unchanged, sudden changes in the PetO₂ may provide an early signal. Indeed, PetCO₂ directly correlates with cardiac output, and PetCO₂ monitoring has been used to assess adequacy of cardiopulmonary resuscitation, and its prognosis.⁶

The presence of exhaled CO₂ is secondary confirmation of endotracheal tube placement, and is commonly recommended even when the tube is seen to pass through the vocal cords,⁷ since clinical assessment is not always reliable. Simple colorimetric devices may be used for this purpose. However, detection of expired CO₂ is not infallible⁷ as false positives can rarely occur following ingestion of carbonated liquids, and false negatives may be due to extremely low pulmonary blood flow, or very large alveolar dead space such as pulmonary embolus or severe asthma. Monitoring with capnography has also been recommended for transport⁸ and respiratory monitoring⁹ in critically ill patients, and should be available for every anaesthetised patient.¹⁰

Gas transfer (diffusing capacity)

This is a test of the transfer of a gas, typically carbon monoxide (CO), across the alveolocapillary barrier. The transfer factor is calculated as:

and, since CO is so completely taken up by Hb, Pc CO is taken as zero. This test is usually performed as an outpatient, is not usually measured in the intensive care unit (ICU), and diffusion abnormality is rarely a cause of hypoxaemia. The transfer factor is often corrected for lung volume since diseases such as emphysema and pulmonary fibrosis may affect both lung volume and diffusion.

Lung volume and capacities (Figure 34.1)

The tidal volume (V_T) is the volume of gas inspired and expired with each breath, with the volume at end-expiration termed the functional residual capacity (FRC). If a forced expiration is performed the expiratory reserve volume (ERV) is expired down to the residual volume (RV). If a maximum inspiratory effort is made from FRC, this is termed a vital capacity (VC) manoeuvre when the total lung capacity (TLC) is reached. Clinically, the most important of these are the FRC, V_T and VC, and the latter two are easily measured using a spirometer or integrated from flow.

Figure 34.1 Lung volumes and capacities. TLC, total lung capacity; IRV, inspiratory reserve volume; TV, tidal volume; ERV, expiratory reserve volume; RV, residual volume; IC, inspiratory capacity; FRC, functional residual capacity; VC, vital capacity.

Tidal volume

Minute volume is composed of f and V_T – normally ∼17 breaths/min and ∼400 ml respectively in adults.¹¹ Rapid shallow breathing is common in patients with respiratory distress, and in those failing weaning. Although a proposed index, an f/V_T ratio > 100 was initially shown to be highly predictive of weaning failure;¹² subsequent studies have reported varying results.

Vital capacity

At TLC the forces due to the inspiratory muscles are counterbalanced by elastic recoil of the lung and chest wall. Consequently, the TLC is determined by the strength of the inspiratory muscles, the mechanics of the lung and chest wall and the size of the lung, which varies with body size and gender (Table 34.1). Since the VC is the difference between TLC and FRC, factors that reduce FRC, such as increased abdominal chest wall elastance and premature airway closure in chronic obstructive pulmonary disease (COPD), will also reduce it. The normal VC is ∼70 ml/kg and reduction to 12–15 ml/kg has previously indicated a probable need for mechanical ventilation. However, many other factors need to be considered, including the patient’s general condition, the strength of the patient’s expiratory muscles, glottic function and the use of non-invasive ventilation. Indeed, many chronically weak patients are able to manage at home with extremely low VC with the assistance of non-invasive ventilation.

Table 34.1 Factors that decrease vital capacity

Decreased muscle strength

• Myopathy

• Neuropathy

• Spinal cord injury

Increased lung elastance

• Pulmonary oedema

• Atelectasis

• Pulmonary fibrosis

• Loss of lung tissue

Increased chest wall elastance

Reduced functional residual capacity

• Atelectasis

• Premature airway closure (e.g. chronic obstructive pulmonary disease)

Functional residual capacity

Direct measurement of FRC is rarely measured in ICU; however, techniques such as nitrogen wash-in and wash-out ¹³ to estimate FRC are becoming available on modern ventilators. When FRC is less than the closing volume, the lung volume at which airway closure collapse is present during expiration, there is a marked increase in mismatch. Consequently, positive end-expiratory pressure (PEEP) is commonly used to elevate FRC. Increases in lung volume above resting lung volume can be directly measured from a prolonged expiration to atmospheric pressure using either a spirometer or integration of flow,¹⁴ or by repeated FRC measurements. FRC is decreased in ARDS and pulmonary oedema, in patients with abdominal distension and following abdominal and thoracic surgery. An increase in FRC places the diaphragm at a mechanical disadvantage, and is seen with severe airflow limitation and dynamic overinflation, and when there is loss of elastic recoil (e.g. emphysema).

Measurement of lung mechanics

The forces the respiratory muscles must overcome during breathing are the elastic recoil of the lung and chest wall, and airway and tissue resistance. During controlled mechanical ventilation the ventilatory pressures reflect the work done to overcome these forces; however during partial ventilatory support the pressure at the airway opening reflects both these forces and those generated by the respiratory muscles. Estimates of respiratory mechanics are often readily available, and can assist titration of ventilatory support.

Elastic properties of lung and chest wall

The respiratory system (RS) is composed of the lung (L) and chest wall (CW), which is comprised of the ribcage and abdomen. Although it is often convenient to consider respiratory system mechanics as implying information about the lung, abnormal chest wall compliance can markedly influence these measurements.¹⁵^–¹⁸

The pressure gradient across the lung (P_L) that generates gas flow is equal to the difference between the pressure at the airway opening (P_ao) and the mean pleural pressure (P_pl) which is estimated as the oesophageal pressure (P_es):

Although the P_es is not always an accurate measure of the absolute P_pl, the change in P_es reflects the change in P_pl. However, this requires an appropriately positioned and functioning oesophageal balloon. In spontaneously breathing subjects a thin latex balloon sealed over a catheter is introduced into the lower third of the esophagus and P_es and P_ao are measured simultaneously during an end-expiratory airway occlusion. A well-positioned oesophageal balloon will have a ratio of ΔP_es/ΔP_ao of ∼1.¹⁹ This technique is reliable in supine, intubated spontaneously breathing patients,²⁰ and in paralysed subjects it appears that a similar pressure change, induced by manual ribcage pressure,²¹ can be used to verify oesophageal balloon function.

Chest wall mechanics are derived from P_es referenced to atmospheric pressure, and in ventilated relaxed subjects respiratory system mechanics are derived from P_ao referenced to atmospheric pressure. It is not surprising then that P_RS = P_L + P_CW. Finally, abdominal mechanics can be estimated as either intravesical or intragastric pressure. However, despite these provisos, useful information can be obtained from respiratory system mechanics.

Measuring the slope of the V–P relationship of the lung or respiratory system allows a simple estimate of the elastic properties of the lung. This is termed the elastance, which is the inverse of the compliance. The E_RS is directly related to its components (E_RS = E_L + E_CW)_, and 1/C_RS = 1/C_L + 1/C_CW. The normal E_RS is 10–15 cmH₂O/l and the normal C_RS is 60–100 ml/cmH₂O in ventilated patients. Since elastance directly refers to the elastic properties of the lung and respiratory system, mechanics will be described in terms of elastance rather than compliance.

Measurement of elastance

Elastance and resistance are frequency-dependent, and respiratory mechanics depend upon the volume and volume history of the lung.²² With increasing frequency of breathing, total respiratory system resistance falls and elastance increases, and this is particularly obvious in patients with airflow obstruction.^23,²⁴ Consequently these factors must be taken into account when interpreting respiratory mechanics. In a passively ventilated subject P_ao is the sum of: (1) the pressure required to overcome airway, endotracheal tube and circuit resistance (P_res); (2) the elastic pressure required to expand the lung and chest wall (P_el); (3) the elastic recoil pressure at end-expiration or total PEEP (P_o); and (4) the inertial pressure required to generate gas flow (P_inert):

Since the elastance (E) is equal to ΔP/ΔV, with the resistance (R) equal to ΔP/Δ, and ignoring the inertance,²⁵ this can be rewritten as the single-compartment equation of motion:

Elastance can then be measured using either static techniques, where cessation of gas flow allows dissipation of P_res, or using dynamic techniques where flow is not interrupted.

END-INSPIRATORY OCCLUSION METHOD

The simplest estimate of E_RS can be made using a rapid end-inspiratory airway occlusion during a constant-flow breath, provided that the respiratory muscles are relaxed (Figure 34.2). If a plateau is introduced at end-inspiration there is a sudden initial pressure drop due to dissipation of flow resistance (P_pk – P₁) followed by a slower, secondary pressure drop to a plateau (P_dif = P₁ – P₂) due to stress relaxation. At least 1–2 seconds are taken for this plateau to be achieved, and P₂ is often called the plateau pressure; however, if P_plat is measured too soon it will lie somewhere between P₁ and P₂.