Br.J. Anaesth. (1978), 50, 3

CLINICAL ASSESSMENT OF RESPIRATORY FUNCTION J. R. A. RIGG AND N. L. JONES

The prime function of the lungs is to maintain oxygen uptake and carbon dioxide excretion. Clinical assessment of lung function before anaesthesia estimates the extent to which this function may be compromised by disease and the combined insults of anaesthesia and surgery. In this paper, the assessment of lung function will be emphasized mainly, although it is important to consider this function in the broader context of oxygen delivery and carbon dioxide removal from the tissues, in which the lung forms an important but not exclusive link. Broad clinical assessment is necessary to detect disorders that affect other links in the respiratory transport chain—disorders of the heart and central circulation, systemic regional circulation and blood volume, disorders that alter haemoglobinoxygen affinity, musculoskeletal disorders that affect the bellows functions of the chest wall and disorders of the regulation of breathing and of acid-base homeostasis. The importance of pulmonary assessment before anaesthesia and surgery is emphasized by two additional factors. First, lung function is a substantial determinant of the uptake and distribution characteristics of inhaled anaesthetic drugs (Eger, 1974a). Second, pulmonary complications are a recognized major source of postoperative morbidity and mortality and thorough assessment is essential to the prevention and optimum management of these problems (Bendixen et al., 1965; Tarhan and Moffitt, 1973).

Frequently, it may be necessary to postpone elective surgery to achieve improvement of airways obstruction by such therapy. The second reason is that many patients with chronic airways obstruction may also have impaired ventilatory control and this has important diagnostic and therapeutic implications. The combination of severe airways obstruction and chronic respiratory failure is a formidable challenge to the anaesthetist and surgeon. Alveolar collapse, consolidation and infection are a

common sequence and more likely to occur in patients with underlying lung disease. Gas exchange is impaired through the presence of non-ventilated, perfused regions in the lungs (West, 1970). Pulmonary oedema and cardiac failure impair gas

exchange by similar mechanisms. These problems are not uncommon and may be of rapid onset and lifethreatening, if not quickly diagnosed and treated. The concurrent existence of both heart and lung disease may make pulmonary oedema more difficult to diagnose and more life-threatening. A low cardiac output may aggravate an existing gas exchange problem because of the low mixed venous oxygen content which co-exists (Kelman et al., 1967; Philbin et al., 1970). Other factors which may contribute to these problems are hypoalbuminaemia (such as in renal failure, liver disease and gastroenteropathies), major trauma (shock lung) and overhydration. Acute bronchoconstriction is more likely to occur in patients with increased bronchial reactivity, particularly in patients with asthma or tobacco addiction. It CLINICAL RESPIRATORY DISORDERS may be triggered by drugs (inhalation or parenteral), The most common problems encountered are those surgical stimuli, aspiration, or instrumentation of the related to intrathoracic airways obstruction: bronchitis upper respiratory tract. emphysema and asthma. It may seem self-evident that Extrathoracic airway obstruction is not frequent but common disorders are the most important, but there may be acutely life-threatening. Important contriare two reasons to emphasize this point. The first is that airways obstruction often has a substantial buting factors are laryngeal spasm, obesity, drugcomponent of reversibility. It is important that this is induced depression of the central nervous system and recognized and that suitable therapy is given to ensure nasal obstruction. Typically, these conditions cause optimum respiratory function before anaesthesia. severe inspiratory airflow obstruction, in contrast to patients with intrapulmonary airways disease. Pneumothorax may occur in association with major J. R. A. RIGG, M.B., B.S., F.F.A.R.A.C.S., Department of trauma to the chest or following supraclavicular or Anaesthesia; N. L. JONES, M.D., F.R.C.P., F.R.C.P.(C), Department of Medicine; McMaster University, Hamilton, interscalene brachial plexus block. A subsequent general anaesthetic with nitrous oxide may lead to Ontario, Canada.

BRITISH JOURNAL OF ANAESTHESIA rapid expansion of the pneumothorax as the soluble anaesthetic replaces less soluble nitrogen in the pleural cavity (Eger, 1974b). Pleural effusion may be associated with infective disorders, pulmonary oedema or heart failure. Gas exchange is impaired by mechanisms similar to those described above — increased ventilation-perfusion mismatching, as a result of an increase in poorly ventilated, well-perfused lung regions (West, 1970).

FLOW-VOLUME RESERVE

10 in

0 out 10

FLOW (litres- 1 )

100(

PRESSURE-VOLUME RESERVE 7 TT 1 B.

PHYSIOLOGICAL ASSESSMENT

Mechanical reserve The simplest clinical method is to observe the patient perform a forced vital capacity manoeuvre; a useful measure of the degree of airways obstruction is obtained by measuring forced expired time. A forced expired time of more than 4 s is abnormal, and severe airways obstruction is usually present if the forced expired time is greater than 10 s. For greater precision, spirometry can be performed and repeated after administering a bronchodilator if airways obstruction is suspected. Respiratory muscle strength may be assessed by measuring maximum static inspiratory and expiratory pressures; however, these pressures are a function of lung volume and a wide range of values have been reported in both health and disease (Cook, Mead and Orzalesi, 1964; Black and Hyatt, 1969, 1971). In the period after anaesthesia, the adequacy of spontaneous ventilation may be assessed by the ability to sustain a tidal volume of 5 ml kg" 1 body weight. This is usually achieved if VC is greater than 15 ml kg" 1 (Pontoppidan, Geffin and Lowenstein, 1973). Failure to reach these values, particularly when associated with a progressive increase in breathing frequency, is an indication that alveolar hypoventilation may be present. The concept of functional mechanical reserve has been extended by Mead, using pressure-volume, flow-volume and relative motion diagrams of the respiratory system (fig. 1) (Mead, 1976). The pressure-volume diagram has also been used to display graphically the loss of this reserve during respiratory muscle weakness (Gal and Smith, 1976; Saunders et al., 1978). Alveolar ventilation The adequacy of alveolar ventilation may be determined by measuring the arterial carbon dioxide

200

-100 Pmu3 (cm H 2 0) CONFIGURATION RESERVE C. 100% VC 80

60 40/ < \ \

X
n / p \ cy\ readily applicable to studies in patients because it is 1 "O2 ~~ O2 \r "CCV ' \ ' (2) Physiological deadspace (ratio to total ventila- tedious, time-consuming and unpleasant for the tion) is computed from the EnghofF modification of subject. However, patient studies are more feasible with the rapid rebreathing method (Read, 1967). The the Bohr equation: rebreathing method also has an important, but less well-recognized physiological advantage—that of F D / F T = (/3aCO2-PECO2)/PaCO2 (3) "opening" the Pco2-ventilation control loop. Changes (3) Shunt fraction (this incorporates all pulmonary in ventilation during rebreathing do not influence venous admixture, that is blood from areas of low Pco 2 , which increases linearly with time, whatever VjQ ratios, poor diffusion and R-L shunting), the ventilation (Fowle and Campbell, 1964; Read expressed as a fraction of cardiac output, is computed and Leigh, 1967; Rigg, Rebuck and Campbell, 1974). from the equation: A reduction in the ventilatory response to carbon dioxide is interpreted in the light of the patient's Qa Cc' O2 -Ca Oa (4) ventilatory capacity in order to decide if there is Qt Cc' O2 -Cv O2 impaired ventilatory control; a low carbon dioxide response may reflect simply the mechanical limitation The unique shape of the oxyhaemoglobin dissocia- to ventilation imposed by airways obstruction tion curve confers a complex relationship between (Cherniack and Snidal, 1956; Eldridge and Davis, oxygen pressure and the content, and necessitates the 1959)—that is, the patient cannot breathe. On the use of contents rather than pressures to solve equation other hand, if a patient has a normal or near normal (4). The solution is obtained simply by a graphical VC and FEV13 it is clear that the patient can breathe, approach, using the standard oxygen pressure : con- and the finding probably indicates low central sensitent diagram (fig. 4). In most clinical circumstances, tivity to carbon dioxide. The problem of separating the position and shape of the oxyhaemoglobin mechanical and central influences that may contribute dissociation curve and mixed venous oxygen pressure to a low carbon dioxide response has led to iempts s ' /., ydtput as and content are assumed. Arterial Po2 is measured to establish other measures ' " alternatives to minute or alveolar ventilation. These and P A 0 2 computed from the alveolar air equation (2). These Po2 values may be graphically converted into include variables such as mouth occlusion pressure, corresponding arterial and pulmonary end-capillary the phrenic neurogram, and diaphragmatic and interoxygen contents, which together with mixed venous costal electromyography (Eldridge, 1975; Whitelaw, content (usually assumed) provide the solution to the Derenne and Milic-Emili, 1975; Kelsen, Altose and Cherniack, 1977). One technique which has attracted shunt equation (fig. 4). less attention is the breath-holding-rebreathing Although the errors inherent in the analysis are well recognized (Wagner, Saltzman and West, 1976), method. Stanley and his colleagues have used this this quantitative approach provides a degree of technique to differentiate patients with impaired precision that is quite adequate for the comprehensive central control of breathing from those with mechpreoperative assessment of patients with severe anical factors limiting the carbon dioxide response diminution of lung reserve and of patients who are (Stanley et al., 1975) (fig. 5). candidates for major cardiovascular or pulmonary Patients with chronic airways obstruction and a surgery. In such patients, serial estimates of physio- low carbon dioxide response are more likely to logical deadspace and shunt fraction in the period develop respiratory failure than patients who have a after operation may be of value in assessing continuing high response (Lane and Howell, 1970). Similarly, progress of ventilatory function, particularly in those asthmatic patients known to have low carbon dioxide who require artificial support of ventilation. responses when free from airways obstruction are more likely to develop hypercapnia during acute exacerbation of their illness than those patients who Control of breathing The most widely used clinical test of control of have high responses when they are well (Rebuck breathing is the ventilatory response to carbon dioxide and Read, 1971). These observations confirmed administered either as a series of constant inspired Lambertsen's suggestion that low carbon dioxide

Po2 is computed from the alveolar air equation, a simplified form of which is given below:

BRITISH JOURNAL OF ANAESTHESIA CHRONIC AIRWAYS OBSTRUCTION (A)

Pco, (kp«)

FIG. 5 (from Stanley et al., 1975, with permission). The relationship between breath-holding time (BHT) and />co2 plotted on semi-logarithmic paper in 17 patients with chronic airways obstruction and five cases of the idiopathic hypoventilation syndrome (IHS). The symbols indicate periods of breath-holds made at different Pco2 values by individual patients, and the shaded areas indicate the normal range of the BHT/Pco 2 relationship. The patients with chronic airways obstruction are in three groups: A were normocapnic; B were also normocapnic but had a previous record of hypercapnia; and C were hypercapnic at the time of study. Note the normal BHT/Pco 2 relationship in the obstructed patients, but the prolonged breath-holding time and flattening of the log BHT/Pco 2 slope in the cases of idiopathic hypoventilation syndrome. The breath-holding rebreathing procedure used in these studies was as follows: Two practice breath-holds were made breathing room air. The patient then began rebreathing from a bag containing 6 litre of 6-8% carbon dioxide in oxygen. Within a few breaths, equilibration occurred between carbon dioxide in the bag and the patient's lungs. Afterwards, the first test breath-hold was made. Following this, the patient rested for 5 min and a second breath-hold was made after a period of rebreathing had caused Phcot to increase to 0.6-0.8 kPa greater than Pco 2 at the start of the first breath-hold. The data were analysed by plotting breath-hold time against Pco 2 at the start of each breath-hold.

I

responsiveness may predispose patients to the development of carbon dioxide retention in various clinical circumstances (Lambertsen, 1960). However, studies of normal subjects given narcotics and thiopentone and of surgical patients, without cardiovascular or respiratory disease, given morphine suggest that low carbon dioxide sensitivity does not confer increased sensitivity of such patients' ventilatory control systems to the effects of central nervous depressant drugs (Rigg and Goldsmith, 1976; Rigg, 1978a, b). In recent years, the ventilatory response to hypoxia has also been used to assess control of breathing, but most investigators have studied normal subjects only (Weil et al., 1970; Godfrey et al., 1971; Rebuck and Campbell, 1974). The clinical utility of these methods remains to be established. Many patients with severe airways obstruction and hypercapnia may be primarily dependent on hypoxic drive to maintain alveolar ventilation during air breathing (Campbell, 1967). If such patients require general or regional anaesthesia, it is important to control inspired oxygen concentration to the lowest value that is clinically effective (usually 24%). High inspired oxygen concentrations may depress breathing

substantially in such patients and lead to progressive hypercapnia and the necessity for ventilatory support (Campbell, 1967). Exercise testing

Like control of breathing, clinical exercise testing has attracted attention in recent years as a means of evaluating both respiratory and cardiovascular reserve (Jones et al., 1975). In this regard, it is useful in patients scheduled for major surgery, in particular cardiovascular surgery, since many of these patients have lung problems associated with cardiac disease. An exercise test provides valuable information on the extent to which the heart and the lungs are capable of meeting the increased oxygen demands, in addition to establishing the importance of pulmonary and cardiac components of symptoms such as breathlessness. Appropriate exercise testing measures the functional reserve in the gas transport system and its main components, the heart and lungs. It may be helpful in assessing the fitness of patients for major surgery. A simple test, within the capability of any pulmonary or cardiac laboratory, studies the response to work loads which are progressively increased by a small amount every minute until the patient is stopped

ASSESSMENT OF RESPIRATORY FUNCTION

6

10 12

14

2

4

6

8

10

12 14

POWER OUTPUT (kW)

FIG. 6. Two examples of progressive exercise tests. Top panel: This patient, a 48-yr-old woman, was referred by a cardiac surgeon on account of a 12-month history of increasing breathlessness; a mitral valvulotomy for mitral stenosis had been performed 10 yr previously and the occurrence of signs of mitral incompetence raised a question of the need for a mitral valve prosthesis. She was a life-long smoker with chronic bronchitis, so that the clinical question posed was whether the increasing dyspnoea was a result of chronic airways obstruction or of cardiac insufficiency. There were clinical and radiological signs of slight cardiac enlargement and mitral incompetence; examination of the respiratory system showed the forced expired time to be increased to 8 s and expiratory rhonchi present with unforced expiration. Spirometry showed evidence of moderately severe airway obstruction (VC: 3.0 litre; FEV: 1.0 litre; predicted VC: 3.8 litre; FEV: 1.0 litre) and the estimated ventilatory capacity was reduced at 35 litre min~ l (FEVj x 35). The progressive exercise test showed a maximal oxygen uptake of 1.25 litre min" 1 (predicted for age and size: 1.9 litre min±0.4). Although cardiac frequency was increased at all power outputs, suggesting a low stroke volume, the usual maximum cardiac frequency (185 beat min" 1 ) was not reached, suggesting that some cardiac reserve was still available. Ventilation was within normal limits and carbon dioxide output was not excessive. However, the values for ventilation at the maximal power output were close to the calculated ventilatory capacity, suggesting that little or no respiratory reserve was left. Maximal tidal volume was 1340 ml, which should be compared with the FEV, of 1000 ml. The results showed that exercise limitation was mainly respiratory in origin; an improvement in cardiac function through mitral valve replacement would not be helpful without concomitant increase in ventilatory reserve; results allowed the clinicians to focus therapeutic attention on the airways obstruction. Bottom panel: Results of a progressive exercise test in a 50-yr-old patient studied before an operation for a large bulla in the left lung. He complained of mild dyspnoea but had little cough or sputum. Spirometry—FEV: 2.75 litre (predicted 3.80 litre); VC: 4.7 litre (predicted 5.0 litre). The results show a normal exercise capacity; maximum oxygen intake was 2.1 litre min" 1 , and heart rate and ventilatory responses were normal (predicted values shown as unbroken lines). The results established that cardiopulmonary reserve was excellent. by symptoms or by the observer if untoward signs appear. Measurements are made of heart rate and the electrocardiograph, arterial pressure, ventilation and breathing frequency. Maximal oxygen uptake, cardiac reserve, functional ventilatory reserve and maximum tidal volume can be assessed. The results provide a useful index of the reserve available in several systems and whether or not this reserve is sufficient to meet the demands of anaesthesia and surgery. Typical results obtained are illustrated in figure 6. Patients with severe or complex cardiovascular or respiratory problems, or both, or who may be candidates for cardiac or lung surgery may be more extensively investigated with different levels of

steady-state exercise. In these studies, a more precise quantitative estimate of cardiac function, ventilation and pulmonary gas exchange can be made by using arterial blood-gas analysis (Jones et al., 1975). GENERAL CLINICAL ASSESSMENT

In order to make rational decisions about patients with lung problems scheduled for anaesthesia, certain general features pertinent to surgery and anaesthesia must be assessed. Age. The very young and very old have a lower reserve of lung function. Weight. Obesity is frequently associated with a low lung volume and airway closure may occur within

BRITISH JOURNAL OF ANAESTHESIA

10 the tidal breathing range with resulting increased intrapulmonary shunting and arterial hypoxaemia (Pontoppidan, Geffin and Lowenstein, 1973). Site and magnitude of proposed surgery. Cardio-

vascular, lung and to a lesser extent abdominal surgery encroach substantially upon lung reserve and may profoundly influence management before and after anaesthesia. An assessment of the likely effect of anaesthesia and surgery should be made in an attempt to ensure that the patient will have adequate pulmonary reserve following surgery. Duration of surgery. The incidence of pulmonary complications increases substantially with duration of anaesthesia. This results from many interrelated factors, among the most important of which are greater anaesthetic dose requirement and slower recovery from anaesthetic effects. Non-pulmonary disorders. Many non-respiratory diseases influence profoundly anaesthetic and postanaesthetic respiratory management. Such disorders include brain diseases that disturb the central respiratory regulation, neuromuscular and musculoskeletal disorders, metabolic diseases, renal disorders affecting acid-base regulation, endocrine disorders and idiopathic hypoventilation syndromes.

Further investigations should follow the general plan given in the flow chart shown in figure 7. A functional classification may then be applied to guide management further (table I). TABLE I. Functional classification of patients with cardiopulmonary disorders (1) Normal cardiopulmonary reserve (2) Reduced cardiopulmonary reserve VC or FEVj, or both, less than 50% predicted Pco 2 normal Po 2 greater than 9.3 kPa and ()sIQt less than 10% (3) Severe reduction of cardiopulmonary reserve VC or FEVi 25-50% of predicted Pco 2 normal Po 2 less than 9.3 kPa 2s/0t m ° r e than 10% Exercise capacity below 75% of normal or Vo2 maximum less than 1.5 litre min" 1 (4) No cardiopulmonary reserve Preanaesthetic cardiac or respiratory failure, or both VC or FEVi less than 25% of predicted Arterial Pco 2 greater than 6.4 kPa Mixed venous Pco 2 greater than 8 kPa Arterial Po 2 less than 6.7 kPa 2s/6t greater than 25%

MANAGEMENT STRATEGIES

In assessing a patient's ability to undergo anaesthesia and the stress of the operative and postoperative periods, whether the operation has direct or only indirect effects on the lungs, the most important functions t o ' be considered are the mechanical characteristics of the respiratory system, and gas exchange. Any patient with respiratory symptoms or signs should have spirometry as a first investigation. SPIROMETRY ^ NORMAL CHEST X-RAY

\ ABNORMAL OBSTRUCTIVE /

NONOBSTRUCTIVE

X-RAY CHANGES

/

\ X-RAY NORMAL MIPS* MEPS

*MIPS=MAXIMUM STATIC INSPIRATORY PRESSURE AT FRC MEPS = MAXIMUM STATIC EXPIRATORY PRESSURE AT FRC

FlG. 7. Flow chart for investigation of the patient with respiratory symptoms and signs.

Spirometry (VC, FEVJ may be normal or reveal impairment of an obstructive (low FEV and FEV/VC ratio) or non-obstructive (low FEV and VC with normal FEV/VC ratio) pattern. An obstructive ventilatory defect is an indication for spirometry after a nebulized bronchodilator such as salbutamol. Normal spirometry in a patient complaining of dyspnoea may be an indication of a cardiac problem; an exercise test may be a help in quantifying the disability objectively. Normal spirometry may also be obtained in an asthmatic in remission; if this requires substantiation, a histamine challenge test will establish the diagnosis and quantify airway reactivity. Where a severe degree of obstruction is found, measurement of Pco 2 either by rebreathing or from an arterial sample will establish whether respiratory failure is present. A high P3iCOi is an indication for a ventilatory carbon dioxide response study to assess the contributions of central control and mechanical factors to alveolar hypoventilation. Non-obstructive (or restrictive) defects are usually associated with small lungs. If the chest x-ray is normal, the possibility of respiratory muscle weakness should be tested by measuring voluntary static

ASSESSMENT OF RESPIRATORY FUNCTION maximal inspiratory and expiratory pressures at FRC (MIPS, MEPS). Where a severe abnormality in pulmonary function is present or if the patient's ability to withstand a major operation is in question, an exercise test may help to define the risks more precisely. A capacity to maintain an oxygen intake of 1500 ml min" 1 indicates an adequate cardiorespiratory reserve: less than 800 ml min" 1 indicates a severe impairment such that a major operation constitutes a substantially increased risk. If an adult patient undergoes lung resection and is left with severely reduced ventilatory function— reflected in an FEV1 or less than 1.0 litre—serious disability with or without ventilatory failure may be expected after operation. Spirometry before operation should be considered in relation to the projected operation with the aim of preserving enough ventilatory function to keep the FEV greater than 1.0 litre or greater than 30% of predicted. If FEV after operation is less than 1.0 litre or 30% of predicted, then it is likely that ventilatory support may be required after anaesthesia, particularly in the first 24 h after operation. It should be emphasized that spirometry requires the full co-operation of the patient and that this may not be obtained in an emergency, in critically ill patients or those with an altered state of consciousness. In these patients, arterial blood-gas analysis is usually the most appropriate investigation. Conscious patients and unconsicous patients with an endotracheal tube in place can be assessed by measuring mixed venous Pco2 by rebreathing together with tidal volume and minute ventilation. Concurrent measurement of arterial and mixed venous Pco 2 enables computation of cardiac output by application of the Fick principle (Davis, Jones and Sealey, 1977). Spirometry, mixed venous Pco 2 and arterial blood-gas analysis are all simple, rapid and easy to perform. Their value in the assessment of ventilatory capacity, ventilation and gas exchange in surgical patients is largely a consequence of the fact that they can be measured frequently for continuing evaluation and reassessment. Patients scheduled for major elective cardiac, vascular, pulmonary or abdominal surgery may require more thorough investigation. Such assessment may include evaluation of the mechanical properties of the respiratory system with oesophageal pressure measurements and body plethysmography. Diffusing capacity, closing volume and estimation of lung volumes by the helium dilution technique may provide further information about lung function at

11 rest. Tests of control of breathing and of exercise responses are more useful, however, since they provide a measure of how the respiratory system responds when it is stressed. All of these tests have an important practical limitation. They are time-consuming and so logistically complex that they cannot be useful tools for continuing evaluation and reassessment during and after anaesthesia. How can the foregoing account provide a framework for simple and clear decisions that the anaesthetist can make with regard to investigation and anaesthetic and postanaesthetic management ? This is not an easy question but is obviously important. Accordingly, the scheme in table II is offered as a summary, and to provide broad guidelines for assessment and management. TABLE II. Summary of management strategies Functional classification

Anaesthetic management

Postanaesthetic management

(1)

Local, regional or general

Oxygen Physiotherapy after major cardiothoracic and abdominal surgery

(2)

Local, regional or general Assisted or controlled ventilation

Oxygen Physiotherapy ? Extradural block (indications discussed below)

(3)

Local, or regional Controlled ventilation technique preferred ? Positive end expiratory pressure If general indicated, (PEEP) controlled Controlled oxygen ventilation therapy ? Extradural block

(4)

Local or regional Controlled ventilation technique preferred PEEP Controlled oxygen General, controlled therapy ventilation Extradural block

SPECIFIC ASPECTS OF MANAGEMENT AND CLINICAL ASSESSMENT

Oxygen therapy after anaesthesia Arterial oxygen tension is known to decrease after anaesthesia, but it is not clear why this occurs in the patient with previously normal lung function (Nunn and Payne, 1962; Nunn, 1964,1965; Nunn, Bergman and Coleman, 1965). It is generally accepted that

12 most patients should be given oxygen in the recovery room. Patients with lung disease should always have controlled oxygen therapy to prevent alveolar hypoventilation resulting from unnecessary blunting of hypoxic drive to breathing (Campbell, 1967; Leigh, 1975). Extradural block Patients with lung disease requiring thoracic or abdominal surgery and who are elderly or obese or who may be unco-operative as a result of an altered conscious state are particularly helped by extradural block to control wound pain after anaesthesia, and to permit effective chest physiotherapy following surgery. Ventilatory support after anaesthesia A major consequence of recognizing the importance of maintaining normal alveolar ventilation during anaesthesia has been the increased use of controlled ventilation during and after anaesthesia (Pontoppidan, Geffin and Lowenstein, 1973). For many severely ill patients, particularly those who are obese or known to have atelectasis, pulmonary oedema or adult respiratory distress syndrome, the use of positive end expiratory pressure (PEEP) with controlled ventilation may improve gas exchange substantially (Pontoppidan, Geffin and Lowenstein, 1973). However, this introduces another dimension of clinical assessment of both respiratory and cardiovascular function, directly relating to important clinical limitations of PEEP. First, an excessive inspiratory distending pressure is easily reached in a patient with stiff lungs and a high tidal volume setting on the respirator. The result may be alveolar rupture, pneumothorax or pneumomediastinum, or both (Pontoppidan, Geffin and Lowenstein, 1973). Second, cardiac output may decrease as a result of a decrease of right atrial filling pressure. The patient must be monitored closely to detect such problems and management strategies changed accordingly (Suter, Fairley and Isenberg, 1975). Termination of ventilatory support How do we assess the patient from the point of view of weaning from a ventilator and establishing adequate spontaneous ventilation? Ability to wean the patient from a ventilator correlates poorly with most quantitative methods of estimating lung function (Pontoppidan, Geffin and Lowenstein, 1973). This may represent the inability of tests of lung function

BRITISH JOURNAL OF ANAESTHESIA to measure the influence of factors such as central neurogenic drive or its level of impairment by artificial ventilation (Eldridge, 1974, 1976). It is probable that such central neural factors are more important determinants of weaning success than is generally recognized. This review of clinical pulmonary assessment has been designed to help the practising anaesthetist decide what management strategies to adopt to achieve the goal of re-establishing adequate ventilation with spontaneous breathing as soon as is possible after the end of anaesthesia. It should be emphasized that a precise respiratory assessment is not always possible because of the uncertainties of intraanaesthetic management as a result of unanticipated surgical problems that may adversely affect respiratory outcome. REFERENCES

Bendixen, H. H., Egbert, E. D., Hedley-Whyte, J., Laver, M. B., and Pontoppidan, H. (1965). Respiratory Care, p. 34. St Louis: Mosby. Black, L. F., and Hyatt, R. E. (1969). Maximal respiratory pressures: Normal values and relationship to age and sex. Am. Rev. Respir. Dis., 99, 696. (1971). Maximal static respiratory pressures in generalized neuromuscular disease. Am. Rev. Respir. Dis., 103, 641. Campbell, E. J. M. (1967). The management of acute respiratory failure in chronic bronchitis and emphysema. Am. Rev. Respir. Dis., 96, 626. Cherniack, R. M., and Snidal, D. P. (1956). The effect of obstruction to breathing on the ventilatory response to CO2. J. Clin. Invest., 35, 1286. Cook, C. D., Mead, J., and Orzalesi, M. M. (1964). Static volume pressure characteristics of the respiratory system during maximal efforts. J. Appl. Physiol., 19, 1016. Davis, C. C , Jones, N. L. and Sealey, B., (1977). Measurement of cardiac output in seriously ill patients using indirect CO 2 rebreathing method. Chest (in press). Eger, E. I., II (1974a). Anaesthetic Uptake and Action, p. 77. Baltimore: Williams and Wilkins. (1974b). Anaesthetic Uptake and Action, p. 171. Baltimore: Williams and Wilkins. Eldridge, F. L. (1974). Central neural respiratory stimulatory effect of active respiration. J. Appl. Physiol., 37, 723. (1975). Relationship between respiratory nerve and muscle activity in muscle force output. J. Appl. Physiol., 39, 567. (1976). Central neural stimulation of respiration in unanesthetised decerebrate cats. J. Appl. Physiol., 40, 23. Davis, J. M. (1959). Effect of mechanical factors on respiratory work and ventilatory responses to CO2. J. Appl. Physiol., 14, 721. Fowle, A. S. E., and Campbell, E. J. M. (1964). The immediate carbon dioxide storage capacity of man. Clin. Sri., 27, 41.

ASSESSMENT OF RESPIRATORY FUNCTION Gal, T. J., and Smith, T. C. (1976). Partial paralysis with d-tubocurarine and the ventilatory response to CO,. Anesthesiology, 45, 22. Godfrey, S., Edwards, R. H. T., Copland, G. M., and Gross, P. L. (1971). Chemosensitivity in normal subjects, athletes, and patients with chronic airways obstruction. J. Appl. Physiol., 30, 193. Goldman, M. D., and Mead, J. (1973). Mechanical interaction between the diaphragm and rib cage. J. Appl. Physiol, 35, 197. Jones, N. L., Campbell, E. J. M., Edwards, R. H. T., and Robertson, D. G. (1975). Clinical Exercise Testing. Philadelphia: W. B. Saunders. Kelman, G. R., Nunn, J. F., Prys-Roberts, C , and Greenbaum, R. (1967). The influence of cardiac output on arterial oxygenation. Br. J. Anaesth., 39, 450. Kelsen, S. G., Altose, M. D., and Cherniack, N. S. (1977). Interaction of lung volume and chemical drive on respiratory muscle e.m.g. and respiratory timing. J. Appl. Physiol, 42, 287. Lambertsen, C. J. (1960). Carbon dioxide and respiration in acid-base hemostasis. Anesthesiology, 21, 642. Lane, D. J., and Howell, J. B. (1970). Relationship between sensitivity to carbon dioxideandclinicalfeatures in patients with chronic airways obstruction. Thorax, 25, 150. Leigh, J. (1975). Oxygen therapy at ambient pressure; in Scientific Foundations of Anaesthesia, p. 200. London: Heinemann. McEvoy, J. D. S., Jones, N. L., and Campbell, E. J. M., (1974). Mixed venous and arterial Pco 2 . Br. Med. J., 2 687. Mead, J. (1976). Of guinea pigs and men. Am. Rev. Respir. Dis., 114, 667. Nunn, J. F. (1964). Factors influencing the arterial oxygen tension during halothane anaesthesia with spontaneous respiration. Br. J. Anaesth., 36, 327. (1965). Influence of age and other factors on hypoxia in the postoperative period. Lancet, 2, 466. Bergman, N. A., and Coleman, A. J. (1965). Factors influencing the arterial oxygen tension during anaesthesia with artificial ventilation. Br. J. Anaesth., 37, 898. Payne, J. P. (1962). Hypoxemia after general anaesthesia. Lancet, 2, 631. Philbin, D. M., Sullivan, S. F., Bowman, F. O., Malm, J. R., and Papper, E. M. (1970). Postoperative hypoxemia: Contribution of the cardiac output. Anesthesiology, 32, 136. Pontoppidan, H., Geffin, B., and Lowenstein, E. (1973). Acute Respiratory Failure in the Adult, p. 60. Boston: Little Brown. Powles, A. C. P., and Campbell, E. J. M. (1978). An improved rebreathing method for measurement of mixed venous Pco 2 and its clinical application. Can. Med. Assoc. J., (in press).

13 Read, D. J. C. (1967). A clinical method for assessing the ventilatory response to carbon dioxide. Aust. Ann. Med., 1, 20. Leigh, J. (1967). Blood-brain tissue Pco 2 relationship and ventilation during rebreathing. J. Appl. Physiol, 23, 53. Rebuck, A. S., and Campbell, E. J. M. (1974). A clinical method for assessing the ventilatory response to hypoxia. Am. Rev. Respir. Dis., 109, 345. Read, J. (1971). Patterns of ventilatory response to CO 2 during recovery from severe asthma. Clin. Sci., 41,13. Rigg, J. R. A. (1978a). Ventilatory effects and plasma concentration of morphine in man. Br. J. Anaesth., 50 (in press). (1978b). Low ventilatory response to carbon dioxide is not associated with increased ventilatory depression by morphine. Br. J. Anaesth., 50 (in press). Goldsmith, C. H. (1976). Recovery of ventilatory responses to CO 2 after thiopentone, morphine and fentanyl in man. Can. Anaesth. Soc. J., 23, 370. Rebuck, A. S., and Campbell, E. J. M. (1974). Effect of posture on ventilatory response to CO2. J. Appl. Physiol, 37, 487. Riley, R. L., and Cournand, A. (1949). Ideal "alveolar air" and the analysis of ventilation-perfusion relationship in the lung. J. Appl. Physiol, 1, 825. Saunders, N. A., Rigg, J. R. A., Pengelly, L. D., and Campbell, E. J. M. (1978). Effect of curare on maximum static V-P relationships of the respiratory system. J. Appl. Physiol. (in press). Stanley, N. N., Cunningham, E. L., Altose, M. D., Kelsen, S. G., Levinson, R. A., and Cherniack, N. S. (1975). Evaluation of breath holding in hypercapnia as a simple clinical test of respiratory chemosensitivity. Thorax, 30, 337. Suter, P. M., Fairley, H. B., and Isenberg, M. D. (1975). Optimum end respiratory airway pressure in patients with acute pulmonary failure. N. Engl. J. Med., 292, 284. Sykes, M. K., McNicol, M. W., and Campbell, E. J. M. (1976). Respiratory Failure. Oxford: Blackwell Scientific Publications. Tarhan, S., and Morfitt, E. A. (1973). Principles of thoracic anesthesia. Surg. Clin. N. Am., 53, 825. Wagner, P. D., Saltzman, H. A., and West, J. B. (1974). Measurement of continuous distribution of ventilation perfusion ratios; theory. J. Appl. Physiol, 36, 585. Weil, J. V., Byrne-Quinn, E., Sodal, I., Friesen, W. O., Underhill, B., Filley, G. F., and Grover, R. F. (1970). Hypoxic ventilatory drive in normal man. J. Clin. Invest., 49, 1061. West, J. B. (1970). Ventilation/Blood Flow and Gas Exchange, 2nd edn. Oxford: Blackwell. Whitelaw, W. A., Derenne, J. P., and Milic-Emili, J. (1975). Occlusion pressure as a measure of respiratory centre output in conscious man. Respir. Physiol, 23, 181.