Pulmonary gas exchange and the development of atelectasis were studied in eight essentially lung-healthy patients, awake and during halothane anaesthesia with mechanical ventilation. Gas exchange was evaluated by a multiple inert-gas elimination technique and conventional blood-gas analysis, and atelectasis was studied by computerized tomography (CT). Ventilation and lung perfusion were well matched in the majority of the patients when awake. In two patients there was low perfusion of poorly ventilated regions (low VA/Q). One patient had a shunt corresponding to 4% of cardiac output. None of the patients showed signs of atelectasis on the CT scans. After 15 min of anaesthesia, shunt had appeared in all patients, ranging from 1% in two patients (unchanged from the awake state) to 17%. The major VA/Q mode was widened and ventilation of poorly perfused regions (high VA/Q) was noted in seven patients. Densities in dependent lung regions (interpreted as atelectasis) were seen on the CT scans in six patients. The extent of atelectasis was significantly correlated both to the magnitude of shunt (r = 0.93, P less than 0.01) and to the impairment of arterial oxygenation (r = 0.99, P less than 0.001). The findings indicate that atelectasis in dependent lung regions during halothane anaesthesia creates shunting of blood flow and that atelectasis is the major or sole cause of impaired gas exchange in the lung-healthy, anaesthetized subject.

The extent of atelectasis was correlated to constitutional factors in 38 patients who underwent computed tomography prior to and during general anaesthesia with halothane. All patients but two developed atelectasis in dependent regions of both lungs immediately after induction of anaesthesia prior to surgery. The transverse area of the densities ranged from 0 to 27 cm2, and there were no significant differences between patients of different age or sex, or with different smoking habits. A significant linear regression was found between Broca's index weight (kg)/height (cm)-100 and the area of the densities, and also between an index describing the shape of the thorax and the density area. Thus, patients who were overweight and/or had a low and wide thorax tended to develop more extensive atelectasis during anaesthesia. This finding might partly explain why overweight patients develop postoperative pulmonary complications more often than non-obese patients.

The effects of atelectasis on pulmonary gas exchange were studied in eight supine, clinically lung-healthy patients. Atelectasis was studied by computerized tomography (CT), and gas exchange by blood gas analysis. The distribution of ventilation/perfusion ratios was assessed by a multiple inert gas elimination technique. No patient had any signs of atelectasis in the awake state, and gas exchange was normal. During ketamine anaesthesia and spontaneous breathing, lung ventilation and perfusion were well matched in most subjects. In one patient there was perfusion of poorly ventilated regions amounting to 14% of cardiac output, and in another there was a shunt of 4% of cardiac output; this patient was the only one who developed atelectasis in dependent lung regions. After muscular relaxation and commencement of mechanical ventilation, all patients but one developed both shunt (2-6% of cardiac output) and atelectasis. The shunt correlated to the size of atelectasis. It is concluded that the occurrence of shunt during anaesthesia is related to the development of atelectasis in dependent lung region, which is consistent with the hypothesis that it is changes in chest-wall mechanics that cause atelectasis.

The development of atelectasis and effects on gas exchange during enflurane anaesthesia in nitrogen/oxygen or nitrous oxide/oxygen (inspired oxygen fraction 0.4) were studied in 16 lung-healthy patients (mean age 49 years). Awake, no subject displayed atelectasis as assessed by computed x-ray tomography of the thorax. Pulmonary gas exchange, studied by multiple inert gas elimination technique, and blood gases were normal. After 10 min of enflurane anaesthesia in nitrogen/oxygen, 14 of 16 subjects had developed atelectasis. After 30 min of enflurane anaesthesia in nitrogen/oxygen or nitrous oxide/oxygen, all patients had developed atelectasis, and a further increase was observed after 90 min of anaesthesia to approximately 5% of the intrathoracic area. There was no difference between the two anaesthesia groups. In the nitrogen group, shunt rose to a maximum of 5.8% at 30 min of enflurane anaesthesia, with a significant reduction to the initial anaesthesia level after 90 min of anaesthesia (3.4%). Perfusion of poorly ventilated lung regions (low VA/Q) averaged 4-5% and did not vary significantly during the anaesthesia. In the nitrous oxide group, shunt increased to 6.3% after 90 min of anaesthesia, and there was a parallel decrease in perfusion of low VA/Q regions. The findings suggest that besides prompt collapse of lung tissue during induction of anaesthesia, absorption of gas from closed-off or poorly ventilated regions takes place and further increases the atelectatic area.

Twenty-one patients who underwent elective cholecystectomy were studied with regard to the effect of intrapleural administration of bupivacaine-adrenaline solution on postoperative pain and ventilatory capacity. Administration of 10 or 20 ml of 2.5 mg/ml or 5 mg/ml bupivacaine solution resulted in complete analgesia in 143 of 159 administrations. Most patients experienced the maximal pain-relieving effect within 1-2 min and analgesia persisted as a rule for 3-5 h. Forced vital capacity and forced expiratory volume in 1 s increased after intrapleural analgesia on average by 56% and 46%, respectively, on the first postoperative day and by 35% and 51%, respectively, on the second day. There was no significant difference in the analgesic effect or in the effect on the ventilatory capacity between the 2.5 mg/ml or the 5 mg/ml solution, in either the 10 ml or the 20 ml dose. Placebo (NaCl) given intrapleurally had no effect on pain or on the ventilatory capacity. The plasma concentration of bupivacaine after intrapleural administration showed a wide interindividual variation, with considerably higher average values when the 5 mg/ml solution had been used than for the 2.5 mg/ml solution. Although no toxic effects were noted, a 2.5 mg/ml solution, which can be given in an initial dose of 20 ml and top-up doses of 10 ml at 3-6 h intervals, is recommended. In four patients minor pneumothorax developed when the catheter was introduced. The pneumothorax was easily evacuated, but underlines the need for great care when introducing the catheter.

In previous studied with computed tomography (CT) prior to and during general anaesthesia, we found that densities developed in dependent parts of the lungs immediately after induction of anaesthesia in all examined patients. It was suggested that the densities were atelectases created by compression of lung tissue but an alternative explanation could be accumulation of extravascular fluid in the lung tissue and/or in the pleural space. In the present study the nature of the densities was analysed in further detail. Injections of contrast medium into the pleural space revealed that the densities were located in the lung tissue and not in the pleural space. By injecting contrast medium intravenously and repeating the CT scanning over a 2-min period the passage of contrast through the major vessels and the lung densities could be studied. The transit time of the contrast medium was of the same magnitude in the densities and the major lung vessels. This indicates that there were no regions with an increased amount of extravascular fluid to delay the contrast passage. These findings oppose the idea of fluid accumulation as the cause of the densities, while atelectasis remains the most plausible explanation.

Cardiac output, oxygen uptake and plasma catecholamines were studied in patients both awake and during anaesthesia prior to and during upper abdominal surgery. Two different forms of anaesthesia were used: neurolept-nitrous oxide (NLA) and halothane (HALO) anaesthesia. Oxygen uptake was determined by using a masspectrometer, and cardiac output was measured according to the Fick principle. Plasma catecholamines were analysed by high performance liquid chromatography. Cardiac output fell by 40% during NLA and by 30% during HALO. Concomitantly, the oxygen uptake fell by 40% and 35%, respectively. A linear relationship between cardiac output and oxygen uptake could be established both in the awake state and during anaesthesia, with no significant change in the slope or position of the regression line when anaesthesia was commenced. Ventricular filling pressures fell during both anaesthetic procedures. Adrenaline fell to half the plasma concentrations seen in normal subjects under resting conditions, while noradrenaline returned to normal from an initially 30-40% increased value. Surgery caused no significant changes in either cardiac output or oxygen uptake, whereas plasma adrenaline increased by 20 times and noradrenaline by 60-90%. The findings suggest that the reduced oxygen uptake during anaesthesia causes the fall in cardiac output rather than any cardiodepressant action of the anaesthetic. It is possible that the anaesthetic depresses whole-body metabolism by either blocking the effects of catecholamines or interfering with cellular metabolism.

Thoracic and abdominal lymph flow have been studied in 25 dogs. Thoracic lymph flow (TLF) was found to be (mean +/- s.e. mean) 6.1 +/- 1.4 ml/h before, and 29 +/- 4.6 ml/h after the induction of lung damage with oleic acid. TLF was depressed by 50% both before and after lung damage, when a positive end-expiratory pressure (PEEP) of 1.0 kPa (10 cmH2O) was applied. This suggests impeded drainage of the lung tissue. Spontaneous breathing, compared to mechanical ventilation, significantly increased TLF by approximately 70%. Abdominal lymph flow increased from 61 +/- 5.3 ml/h to 111 +/- 12.6 ml/h, when a PEEP of 1.0 kPa was applied. These findings demonstrate that PEEP may contribute to oedema in a surgical area. It is concluded that increased intrathoracic pressure reduces TLF, and spontaneous breathing increases TLF, as compared to mechanical ventilation without PEEP.