Portal de Periódicos da CAPES

Combination of external chest wall oscillation with continuous positive airway pressure

2001; Elsevier BV; Volume: 87; Issue: 3 Linguagem: Inglês

10.1093/bja/87.3.441

1471-6771

Stefan Scholz, J. Sticher, G. Ha ̈ufler, M. Mu ̈ller, Ning Ou, G. Hempelmann,

Chronic Obstructive Pulmonary Disease (COPD) Research

We studied the effects of continuous positive airway pressure (CPAP) on pulmonary gas exchange during external chest wall oscillation (ECWO), and the relationship with obesity, in nine patients with normal body weight (group ‘N’) and 10 obese patients (group ‘O’). During ECWO with CPAP 5, PaCO2 decreased in group ‘O’ (6.0 (sd 0.8) to 5.6 (0.5) kPa, P<0.05), whereas it increased in group ‘N’ at all levels (P<0.01). Arterial Po2 (P<0.001) was greater and PaCO2 (P<0.01) less in group ‘N’ during CPPV and ECWO plus CPAP. We also compared the haemodynamic effects of ECWO plus CPAP with those of continuous positive pressure ventilation (CPPV). ECWO plus CPAP and CPPV were applied for 30 min to 6 ASA III patients. Cardiac output (CI 2.7 (0.5) vs 2.1 (0.2) litre min−1 m−2, P<0.05) and stroke volume (SVI 49 (9) vs 32 (6) ml m−2, P<0.05) were greater during ECWO plus CPAP than with CPPV. ECWO is less effective in obese individuals than in those with normal body weight, and the effect of CPAP in overweight individuals is small. We studied the effects of continuous positive airway pressure (CPAP) on pulmonary gas exchange during external chest wall oscillation (ECWO), and the relationship with obesity, in nine patients with normal body weight (group ‘N’) and 10 obese patients (group ‘O’). During ECWO with CPAP 5, PaCO2 decreased in group ‘O’ (6.0 (sd 0.8) to 5.6 (0.5) kPa, P<0.05), whereas it increased in group ‘N’ at all levels (P<0.01). Arterial Po2 (P<0.001) was greater and PaCO2 (P<0.01) less in group ‘N’ during CPPV and ECWO plus CPAP. We also compared the haemodynamic effects of ECWO plus CPAP with those of continuous positive pressure ventilation (CPPV). ECWO plus CPAP and CPPV were applied for 30 min to 6 ASA III patients. Cardiac output (CI 2.7 (0.5) vs 2.1 (0.2) litre min−1 m−2, P<0.05) and stroke volume (SVI 49 (9) vs 32 (6) ml m−2, P<0.05) were greater during ECWO plus CPAP than with CPPV. ECWO is less effective in obese individuals than in those with normal body weight, and the effect of CPAP in overweight individuals is small. Although continuous positive pressure ventilation (CPPV) is widely used for patients requiring respiratory support, interest in negative pressure ventilation has returned in recent years,1Corrado A Gorini M Ginanni R Pelegatti C Villela G Buoncristiano U Guidi F Pagni E Peris A. De Paolo E. Negative pressure ventilation versus conventional mechanical ventilation in the treatment of acute respiratory failure in COPD patients.Eur Respir J. 1998; 12: 519-525Crossref PubMed Scopus (30) Google Scholar 2Heulitt MJ. The use of negative pressure ventilation in infants with acute respiratory failure: old technology, new idea.Respir Care. 2000; 45: 479PubMed Google Scholar partly because of a greater appreciation of the detrimental cardiovascular effects of CPPV. Compared with spontaneous ventilation, controlled positive pressure ventilation reduces cardiac output by reducing venous return.3Chihara E Hashimoto S Kinoshita T Hirose M Tanaka Y Morimoto T. Elevated mean systemic filling pressure due to intermittent positive-pressure ventilation.Am J Physiol. 1992; 262: H1116-H1121PubMed Google Scholar 4DeMaria EJ Burchard KW Carlson DE Gann DS. Continuous measurement of atrial volume with an impedance catheter during positive pressure ventilation and volume expansion.Surg Gynecol Obstet. 1990; 80: 501-509Google Scholar External chest wall oscillation (ECWO) by a Hayek oscillator is a recently developed mode that combines features of negative pressure ventilation and conventional high frequency oscillation.5Hayek Z Peliowski A Ryan CA Jones R Finer NN. External high frequency oscillation in cats. Experience in the normal lung and after saline lung lavage.Am Rev Respir Dis. 1986; 133: 630-634PubMed Google Scholar A cuirass is used to oscillate the anterolateral area of the chest and upper abdomen with an average subatmospheric pressure. In contrast to previous attempts at ECWO that relied upon the elastic recoil of the chest wall, both the inspiratory and expiratory phases of ventilation are actively controlled. By setting the end-expiratory chamber pressure at less than atmospheric, the decrease of functional residual capacity (FRC) could be reduced. However, in almost all studies carried out on anaesthetized humans, end-expiratory chamber pressure has been positive to achieve a sufficient pressure span for carbon dioxide removal.6Sidenoe B Vaage J. Ventilation by external high-frequency oscillations improves cardiac function after coronary artery bypass grafting.Eur J Cardiothorac Surg. 1997; 11: 248-257Crossref PubMed Scopus (15) Google Scholar 7Takeda S Nakanishi K Takano T Nejima J Takayama M Ishikawa G Ogawa R. The combination of external high-frequency oscillation and pressure support ventilation in acute respiratory failure.Acta Anaesthesiol Scand. 1997; 41: 670-674Crossref PubMed Scopus (13) Google Scholar In addition to this active compression of the chest, the required airtight seal of the cuirass itself may reduce FRC. These effects and the known decrease in FRC during anaesthesia may reduce the efficiency of ECWO by shifting the tidal range to the lower, flat portion of the volume-pressure curve of the respiratory system. Continuous positive airway pressure (CPAP) is an established ventilatory adjunct that increases FRC.8Wyche MQ Teichner RL Kallos T Marshall BE Smith TC. Effects of continuous positive pressure breathing on functional residual capacitiy and arterial oxygenation during intra-abdominal operations.Anesthesiology. 1973; 38: 68-74Crossref PubMed Scopus (44) Google Scholar Whether a certain value of CPAP improves alveolar gas exchange or over-distends the lung depends both on the initial level of the FRC and the compliance of the respiratory system. As reduction in FRC during anaesthesia is more marked in obese patients,9Wahba RWM. Perioperative functional residual capacity.Can J Anaesth. 1991; 38: 384-400Crossref PubMed Scopus (198) Google Scholar it can be assumed that the effects of CPAP differ between obese individuals and those with a normal body weight. This study was designed to test the hypotheses that (1) the application of a CPAP improves pulmonary gas exchange during external chest wall oscillation, (2) the effects of CPAP depend upon body status, and (3) cardiac output is less impeded during ECWO plus CPAP than with CPPV. The study protocol was approved by the ethical committee of Justus-Liebig-University Giessen, and written informed consent was obtained from each patient before the study began. The study consisted of two parts: in part one, the effects of a combination of four levels of CPAP with ECWO were investigated. Two groups of patients, with a body mass index of less and greater than 25, were compared. In part two, the haemodynamic effects of CPPV were compared with those of a combination of ECWO and CPAP. We chose the CPAP from the findings of part one of the study. Twenty patients scheduled for elective lower abdominal surgery were enrolled into the first part of the study, meeting the following criteria: American Society of Anesthesiologists (ASA) physical status I and II, aged between 18 and 60 yr, body height between 170 and 190 cm, and male gender. Exclusion criteria were: a history of restrictive or obstructive pulmonary disorders, cardiovascular diseases, and a body mass index outside the range of 20–35. Six patients classified as ASA physical status III and awaiting major abdominal surgery were included into the second part of the study. Patients were excluded if they had ventricular dysfunction classified as New York Heart Association (NYHA) function III or IV, unstable angina pectoris, or if pre-operative vital capacity (VC) or forced expiratory volume 1 s (FEV1) were less than 90 or 80%, respectively, of the predicted norm.10European Community for Steel and Coal Standardised lung function testing. Official statement of the European Respiratory Society.Eur Respir J. 1993; 6 (Suppl 16)Google Scholar Pre-medication consisted of 7.5 mg orally administered midazolam given 45 min before the patients were transferred to the operating room. After pre-oxygenation, anaesthesia was induced with fentanyl 3–4 μg kg−1 and propofol 2.0–2.5 mg kg−1. Vecuronium 0.1–0.12 mg kg−1 was given to facilitate intubation. Trachea was intubated with a 10.0 mm ID tracheal tube (Mallinckrodt, Athlone, Ireland) via direct laryngoscopy. Anaesthesia and relaxation were maintained with a continuous infusion of propofol 100–120 μg kg−1 min−1, fentanyl 2 μg kg−1 h−1 and vecuronium 0.2 mg kg−1 h−1. Patients were in the supine position and connected to a Servo Ventilator 900C (Siemens-Elema AB, Solna, Sweden) and the lungs were ventilated in the volume-controlled mode. Initial ventilator settings were: a respiratory rate of 12 min−1, an I:E ratio of 1:2 without an end-inspiratory pause, and a positive end-expiratory pressure of 5 cm H2O. Tidal volume was initially set at 8–10 ml kg−1 and then adjusted to achieve normocapnia. An inspired oxygen fraction of 0.6 was used for both studies. ECWO was delivered with a Hayek oscillator (Breasy Medical Equipment, London, UK). A cuirass of size 8 or 9 was placed over the anterior part of the chest and upper abdomen and sealed using Velcro straps placed around a bean filled pillow on which the patients were lying. The cuirass was connected via wide-bore tubing to a micro-processor-controlled power unit that contains a diaphragm pump. The initial oscillator settings were set as follows: frequency 60 min−1, end-inspiratory chamber pressure –25 cm H2O, end-expiratory chamber pressure 3 cm H2O, I:E ratio 1:1. After 10 min, end-inspiratory and end-expiratory chamber pressure were adjusted to obtain a tidal volume of at least 170 ml and a mean chamber pressure between –5 and – 10 cm H2O. The I:E ratio and frequency were kept constant during the study period. CPAP was delivered with a high flow CPAP system (Continuous Flow CPAP-System FDF 2/G.C.V., B+P, Neunkirchen, Germany). Oxygen and nitrogen were set to a flow rate of 20 litre min−1, resulting in a sum flow of 40 litre min−1. CPAP level was adjusted by a spring-loaded valve (Ambu International A/S, Brondby, Denmark). Standard monitors included a five-lead electrocardiograph (Sirecust 1280, Siemens Medical Electronics, Danvers, MA, USA) and a pulse oximeter (Nellcor Puritan Benett, Hayward, CA, USA). Each patient had a radial artery catheter (20 G, Vialon, Becton Dickinson, Heidelberg, Germany). A continuous intravascular blood gas system, consisting of fluorescent Po2, pH and Pco2 sensors and a thermocouple (Paratrend 7+, Diametrics, High Wycombe, Bucks, UK) was inserted after in vitro calibration via the radial artery catheter. Continuous blood–gas monitoring was used to allow rapid adjustment of the ventilator settings and to assess blood gas stability before an in vitro analysis was performed. After each study period, arterial blood was sampled for in vitro blood gas measurement (NOVA Stat 5, Nova Biomedical, Waltham, MA, USA). All patients of the second part of the study had a pulmonary artery catheter (Model 93A-434-7.5F G, Baxter Healthcare Corporation, Irvine, CA, USA) in place. Cardiac output was measured by the thermodilution technique, and the results of five measurements taken randomly through the respiratory cycle were averaged (Explorer RER, Baxter Healthcare Corporation, Irvine, CA, USA). Systemic and pulmonary artery pressure, pulmonary capillary wedge, and central venous pressure were recorded at end-expiration. Flow and airway pressure were measured via a variable-orifice pneumotachograph with an integrated differential pressure transducer (VarFlex, Bicore Monitoring Systems, Irvine, CA, USA), positioned between the tracheal tube and the T-piece. Tidal volume (VT), respiratory rate (RR) and expiratory minute ventilation (VE) were obtained from the flow signal. The first part of the study was conducted as a four-period crossover trial, with each combination of ECWO and CPAP studied for at least 15 min. Patients received in a randomized order ECWO with a CPAP of 0 (CPAP 0), 5 (CPAP 5), 10 (CPAP 10), and 15 (CPAP 15) cm H2O. When blood gases remained stable for 10 min in the respective combination of ECWO and CPAP, in vitro blood gas analysis was performed and circulatory values were recorded. In the second part of our investigation, ECWO plus CPAP and CPPV were randomly applied to each patient for 30 min. PEEP was set at 6 cm H2O. Attention was paid to match CPAP and PEEP level. When blood gases and circulation were stable for at least 10 min, in vitro blood gases, circulatory and respiratory values were recorded. Data were analysed using statistical software (SPSS release 9.01 for Windows; Chicago, IL, USA) running under Windows 98 (Microsoft Corporation, Redmond, WA, USA) on a personal computer. The two-sample t-test was used to compare biometric data between group ‘O’ with group ‘N’. The general linear model (GLM) approach was used to perform repeated measures analysis of variance.11McCullagh P Nelder JA. Generalized Linear Models. 2nd edn. Chapman and Hall, New York1989Crossref Google Scholar To assess the effect of CPAP on haemodynamics and gas exchange in part one of the study, a within-subject factor with four levels (CPAP 0, 5, 10, 15 cm H2O) was defined. To compare CPPV with ECWO plus CPAP, a separate analysis with five levels (CPPV, CPAP 0, 5, 10, 15 cm H2O) of the within-subject factor was performed. Differences between obese individuals and those with normal body weight were analysed by including a between-subject factor into the model. For any within-subjects effect, the Mauchly test of sphericity was performed. The Greenhouse–Geisser epsilon was used to correct the degrees of freedom if the assumption of sphericity was violated. When assessing the effects of CPAP, the four levels of the within-subjects factor were compared in each group by simple contrasts. Simple contrasts were also used to compare the effect of CPPV with those of ECWO plus CPAP. Bonferroni's correction was applied for double comparisons (contrasts in group ‘O’ and ‘N’). In part two, comparisons between ECWO plus CPAP vs CPPV in ASA III patients were made by the Wilcoxon matched pairs test. All values are mean (sd), and statistical significance was inferred if P<0.05. Among 20 patients who gave informed consent, one of group ‘N’ was excluded because of excessive pulmonary secretion during ECWO, which required repeated mechanical aspiration. The 19 remaining patients studied in part one of our investigation did not differ with respect to age and height. As required by the design, patients of the ‘obese’ group (group ‘O’) had a greater body mass index (Table 1).Table 1Characteristics of patients in study one. BMI=body mass index; †P<0.01; ‡P<0.001Group ‘O’ (n=10)Group ‘N’ (n=9)Age (yr)52 (39–66)46 (21–74)Height (cm)173 (11)175 (7)Weight (kg)90 (15)†71 (8)BMI29.9 (2.5)‡23.7 (1.1) Open table in a new tab Ventilator settings during ECWO and CPPV (Table 2) were comparable between the groups. During ECWO, measured CPAP levels were always higher than set at the PEEP valve.Table 2Ventilator settings during CPPV and ECWO. PIP, peak inspiratory pressure; PM, mean airway pressure; PEEP, positive end-expiratory pressure; Pinsp, peak inspiratory chamber pressure; Pexp, end-expiratory chamber pressure; MCP, mean chamber pressure; ΔPressure, chamber pressure spanGroup ‘O’ (n=10)Group ‘N’ (n=9)Statistical significanceCPPVPIP (cm H2O)20 (4)18 (4)n.s.PM (cm H2O)10 (1)9 (3)n.s.PEEP (cm H2O)7 (1)6 (1)n.s.ECWOFrequency (min−1)60 (0)60 (0)n.s.I:E ratio1:11:1n.s.Pinsp (cm H2O)–22.9 (4.8)–20.2 (3.9)n.s.Pexp (cm H2O)+5.9 (2.0)+3.4 (3.5)n.s.MCP (cm H2O)–6.3 (1.5)–6.7 (2.4)n.s.ΔPressure (cm H2O)28.8 (6.2)23.6 (6.0)n.s. Open table in a new tab Despite a decrease in tidal volume from 609 (125) to 193 (47) ml (P<0.001) after switching from CPPV to ECWO, blood gases remained constant in group ‘N’. With ECWO in group ‘O’, tidal volume decreased from 659 (98) to 175 (85) ml, P<0.001 and this was accompanied by decreases in PaO2 (24.3 (6.5) to 16.6 (4.4) kPa, P<0.01) and pH (7.43 (0.03) to 7.35 (0.05), P<0.01), whereas carbon dioxide tension increased from 4.7 (0.3) to 6.0 (0.8) kPa (P<0.01) (Table 3).Table 3Respiratory mechanics, arterial blood gases and haemodynamics during ECWO plus CPAP (group ‘N’: n=9, group ‘O’: n=10). PIP, peak inspiratory pressure; PM, mean airway pressure; CPAP, continuous positive airway pressure; VTe, expiratory tidal volume; AaGrad, alveolar-arterial oxygen gradient. *Difference between group ‘O’ and ‘N’ at P<0.05; †Difference between group ‘O’ and ‘N’ at P<0.01; ‡Difference between group ‘O’ and ‘N’ at P<0.001; §Difference from ‘ECWO+CPAP 0’ at P<0.05GroupECWO+CPAP 0ECWO+CPAP 5ECWO+CPAP 10ECWO+CPAP 15PIP (cm H2O)N5 (0)12 (1)16 (1)21 (2)O5 (1)11 (1)16 (2)21 (3)PM (cm H2O)N3 (1)9 (1)13 (1)19 (1)O3 (1)9 (1)13 (2)18 (2)CPAP (cm H2O)N3 (0)9 (1)12 (1)19 (1)O3 (1)9 (1)11 (1)19 (2)VTe (ml)N193 (47)170 (30)160 (26)150 (26)§O175 (85)179 (57)174 (57)163 (51)PHN7.46 (0.07)7.43 (0.07)7.40 (0.08)§7.36 (0.11)§O7.35 (0.05)*7.36 (0.04)§7.34 (0.06)7.31 (0.08)PaCO2 (kPa)N4.0 (0.6)4.5 (0.9)§4.9 (1.2)§5.7 (1.7)§O6.0 (0.8)†5.6 (0.5)§6.1 (1.0)6.6 (1.2)PaO2 (kPa)N31.7 (9.0)33.6 (7.2)34.0 (6.7)34.4 (7.3)O16.6 (4.4)‡18.2 (4.7)18.3 (4.9)20.2 (5.2)PaO2 – PaO2N24.9 (2.3)23.0 (2.0)22.5 (1.9)22.0 (2.1)O39.9 (2.2)‡38.3 (1.9)38.1 (1.8)36.2 (2.0)HR (beats min−1)N54 (11)56 (11)55 (11)59 (10)O59 (10)59 (10)58 (10)59 (12)MAP (mm Hg)N83 (19)84 (18)82 (19)81 (19)O79 (11)79 (12)80 (12)78 (12) Open table in a new tab Obese individuals and individuals with normal body weight responded differently when ECWO was combined with CPAP set at 5 cm H2O. Whereas PaCO2 decreased in obese subjects (6.0 (0.8) to 5.6 (0.5) kPa, P<0.05), it increased in normal subjects with normal body weight (4.0 (0.6) to 4.5 (0.9) kPa, P<0.05). In group ‘N’, a further increase in CPAP to levels of 10 and 15 cm H2O increased arterial Pco2 to 4.9 (1.2) kPa (P<0.05) and 5.7 (1.7) kPa (P<0.01, Table 3), respectively. Compared with CPAP 0, tidal volume decreased with CPAP 15 in group ‘N’ from 193 (47) to 150 (26) ml, P<0.05, and remained unchanged at all CPAP values in group ‘O’. Arterial Po2 (P<0.001) and pH (P<0.05) were greater, and PaCO2 (P<0.01) was less in subjects with normal body weight, irrespective of the ventilation used. In obese patients, the alveolar-arterial oxygen difference was greater during CPPV and ECWO plus CPAP (P<0.01). In part two of our investigation, seven patients agreed to participate, but one patient was excluded because of supraventricular arrhythmia immediately before the study began. Therefore, six patients (aged 60 (12) yr (range 41–75), height 178 (6) cm, weight 96 (13) kg, body mass index 30.5 (3.7), ASA physical status III (0)) were studied. Respiratory mechanics for CPPV and ECWO+CPAP are given in Table 4.Table 4Ventilator settings in six ASA III patientsCPPVECWO+CPAPStatistical significancePIP (cm H2O)19 (1)11 (2)<0.05PM (cm H2O)10 (1)8 (1)<0.05PEEP (cm H2O)6 (1)7 (1)n.s.VTe (ml)685 (72)193 (27)<0.05RR (min−1)12 (1)60 (0)–I:E1:21:1–Pinsp cuirass (cm H2O)––24.6 (6.8)–Pexp cuirass (cm H2O)–4.0 (3.1)–ΔP cuirass (cm H2O)–28.6 (4.1)–MCP (cm H2O)––8.3 (4.7)– Open table in a new tab PEEP and CPAP, respectively, were similar during both ventilatory modes. Peak inspiratory pressure (19 (1) vs 11 (2) cm H2O, P<0.05), mean airway pressure (10 (1) vs 8 (1) cm H2O, P<0.05) and tidal volume (685 (72) vs 193 (27) ml, P<0.05) were inherently greater during CPPV. The results of invasive circulatory and blood gas measurements are shown in Table 5. Cardiac index (2.7 (0.5) vs 2.1 (0.2) litre min−1 m−2, P<0.05), stroke volume (SVI 49 (9) vs 32 (6) ml m−2, P<0.05) and oxygen delivery (383 (84) vs 309 (29) ml min−1 m−2, P<0.05) were greater during ECWO+CPAP. A tendency for greater PaO2 and less PaCO2 during CPPV was statistically not significant.Table 5Cardiovascular measurements and blood gases in six ASA III patients, *P<0.05CPPVECWO+CPAPCI (litre min−1 m−1)2.1 (0.2)2.7 (0.5)*HR (beats min−1)66 (10)65 (9)MAP (mm Hg)77 (15)79 (5)MPAP (mm Hg)17 (6)20 (6)PCWP (mm Hg)10 (4)9 (4)CVP (mm Hg)7 (5)7 (4)SVI (ml m−2)32 (6)49 (9)*SVRI (dyn s cm−5 m2)2659 (808)2215 (892)PVRI (dyn s cm−5 m2)275 (85)330 (62)*PH7.45 (0.05)7.22 (0.39)*PaCO2 (kPa)4.6 (0.4)5.3 (1.0)PaO2 (kPa)25.9 (7.2)22.1 (7.7)SaO2 (%)100 (1)99 (1)SvO2 (%)73 (5)77 (4)*Do2 (ml min−1 m−2)309 (29)383 (64)*Vo2 (ml min−1 m−2)90 (15)88 (19)Qs/Qt (%)13 (2)19 (2)Oxygen extraction (%)29 (5)23 (2) Open table in a new tab In this study population of anaesthetized and paralysed patients, we have shown that (1) during ECWO, the effects of CPAP on pulmonary gas exchange are different in obese individuals and in those with normal body weight, and (2) that a combination of ECWO and CPAP allows a greater cardiac output than conventional positive pressure ventilation. In 10 subjects with normal body build, arterial Pco2 and Po2 did not differ between CPPV and ECWO. When ECWO was combined with CPAP, each increase in CPAP was accompanied by a decline in tidal volume and an increase in PaCO2. As the alveolar-arterial difference for oxygen remained constant, the increases in PaCO2 with increasing CPAP can be attributed to alveolar hypoventilation. Our obese individuals had a different response to CPAP during ECWO. We proposed that increasing the FRC with CPAP would improve carbon dioxide elimination by recruiting unventilated alveoli and by reducing the over-expansion of the remaining lung,12Flechter R Larsson A. Gas exchange during atelectasis.Anaesthesia. 1986; 41: 1135-1138Crossref PubMed Scopus (9) Google Scholar but the effect of CPAP 9 (2) cm H2O on PaCO2 was small. Because the chosen CPAP between 3 and 19 cm H2O should offset the FRC reduction in obese, but otherwise healthy subjects, we assume that the limited effect of CPAP on gas exchange was because of its distribution within the thorax. In the supine position, CPAP preferentially distends upper lung zones,13Hedenstierna G Santesson J Norlander O. Airway closure and distribution of inspired gas in the extremely obese, breathing spontaneously and during anaesthesia with intermittent positive pressure ventilation.Acta Anaesthesiol Scand. 1976; 20: 334-342Crossref PubMed Scopus (67) Google Scholar 14Bindslev L Hedenstierna G Santesson J Norlander O Gram I. Airway closure during anaesthesia, and its prevention by positive end expiratory pressure.Acta Anaesthesiol Scand. 1980; 24: 199-205Crossref PubMed Scopus (28) Google Scholar whereas the changes in FRC predominantly affect dependent lung zones.15Rehder K Sessler AD Rodarte JR. Regional intrapulmonary gas distribution in awake and anesthetized-paralyzed man.J Appl Physiol. 1977; 42: 391-402Crossref PubMed Scopus (114) Google Scholar The negative pressure from the cuirass was applied to the anterior part of the chest, which could have augmented the preferred distension of upper lung zones by CPAP. Compared with the subjects of normal body weight, arterial oxygenation in the obese individuals was already less during CPPV, and their oxygenation decreased when ECWO was started. Ventilation-perfusion inequality may contribute to their impaired oxygenation. Overweight patients develop more extensive compression atelectasis during anaesthesia.16Strandberg A Tokics L Brismar B Lundqvist H Hedenstierna G. Constitutional factors promoting development of atelectasis during anaesthesia.Acta Anaesthesiol Scand. 1987; 31: 21-24Crossref PubMed Scopus (117) Google Scholar Whether compression by the cuirass affects the development of atelectasis and is more marked in obese individuals, remains unclear. Despite greater use of ECWO, for instance in microlaryngeal surgery,17Dilkes MG McNeill JM Hill AC Monks PS McKelvie P Hollamby RG. The Hayek oscillator: a new method of ventilation in microlaryngeal surgery.Ann Otol Rhinol Laryngol. 1993; 102: 455-458Crossref PubMed Scopus (21) Google Scholar some problems regarding its efficiency in anaesthetized patients remain to be resolved. Normal subjects, awake or anaesthetized, can be ventilated efficiently with ECWO alone. To evaluate a prototype chest wall oscillator, Dolmage and co-workers18Dolmage TE De Rosie JA Avendano MA Goldstein RS Effect of external chest wall oscillation on gas exchange in healthy subjects.Chest. 1995; 107: 433-439Abstract Full Text Full Text PDF PubMed Scopus (6) Google Scholar studied seven volunteers with a body mass index of 24.3 (1.8). With a chamber pressure span of 25.1 (3.3) cm H2O, which was comparable to ours, their subjects ceased spontaneous breathing at oscillatory frequencies of 60–90 min−1. Five minutes of ECWO at a frequency of 60 min−1 and a tidal volume of 344 (34) ml decreased PaCO2 by 13 (1) mm Hg. In our study, ECWO with comparable oscillator setting was as effective as CPPV, although our subjects had been anaesthetized and ventilated in the supine position. In contrast to normal body subjects, our obese subjects were substantially less effective ventilated by ECWO. Despite careful setting of the airtight seal of the cuirass, we repeatedly failed to achieve normocapnia by increasing chamber pressure span. Problems in achieving effective ventilation by ECWO alone have been reported by Shekerdemian and colleagues.19Shekerdemian LS Bush A Lincoln C Shore DF Petros AJ Redington AN. Cardio-pulmonary interactions in healthy children and children after simple cardiac surgery: the effects of positive and negative pressure ventilation.Heart. 1997; 78: 587-593Crossref PubMed Scopus (65) Google Scholar In their study, pressure support was applied to overcome tube resistance. Pressure support ventilation (PSV) was also combined with ECWO by Takeda and co-workers,7Takeda S Nakanishi K Takano T Nejima J Takayama M Ishikawa G Ogawa R. The combination of external high-frequency oscillation and pressure support ventilation in acute respiratory failure.Acta Anaesthesiol Scand. 1997; 41: 670-674Crossref PubMed Scopus (13) Google Scholar although their patients were breathing spontaneously. Whether PSV or other ventilatory adjuncts in combination with ECWO are more effective in overweight patients or those with pulmonary disease, awaits further study. In our study, tidal volumes in the range of 150–200 ml raise the questions about the underlying mechanisms of gas exchange. Venegas and colleagues developed a mathematical model which describes the general relationship between gas transport, tidal volume, and respiratory rate. The transition from conventional ventilation to high-frequency ventilation occurred when the alveolar ventilation per breath was 20% of the volume of the conducting airways (VD), and the tidal volume is 1.2 times VD.20Venegas JG Hales CA Strieder DJ. A general dimensionless equation of gas transport by high-frequency ventilation.J Appl Physiol. 1986; 60: 1025-1030Crossref PubMed Scopus (20) Google Scholar The tidal volumes in our investigation were almost always more than 1.2 times the calculated VD and the calculated alveolar ventilation per breath exceeded 20% of VD, so gas exchange in our subjects was predominantly governed by conventional gas transport. In our six ASA III patients, cardiac output and stroke volume were greater with ECWO plus CPAP than with CPPV. The effects of positive pressure ventilation on cardiovascular function have been studied repeatedly.21Pinsky MR. The hemodynamic consequences of mechanical ventilation: an evolving story.Intensive Care Med. 1997; 23: 493-503Crossref PubMed Scopus (139) Google Scholar In brief, increasing intrathoracic pressure by CPPV decreases both venous return to the right ventricle (RV) and augments left ventricular (LV) ejection by decreasing LV afterload. Variations in right atrial pressure are the primary factor determining the fluctuation in pressure gradient for systemic venous return during ventilation. Increases in intrathoracic pressure (ITP) by CPPV decrease venous blood flow by increasing right atrial pressure.22Scharf SM Brown R Saunders N Green LH. Hemodynamic effects of positive pressure inflation.J Appl Physiol. 1980; 49: 124-131Crossref PubMed Scopus (86) Google Scholar The decrease in venous return is partially offset by an accompanying increase in abdominal pressure tending to maintain mean systemic pressure relative to right atrial pressure constant.23Takata M Robotham JL. Effects of inspiratory diaphragmatic descent on inferior vena caval venous return.J Appl Physiol. 1992; 72: 597-607PubMed Google Scholar Left ventricular afterload is determined by systolic wall tension which is proportional to the product of transmural LV pressure and the radius of curvature of the LV. Provided that LV inside pressure during ejection remains unaltered, increases in intrathoracic pressure by CPPV should decrease LV afterload by reducing its transmural pressure gradient. Besides the effects on transmural LV pressure, CPPV also decreases LV afterload by impeding venous return, thus decreasing the end-systolic LV radius of curvature. The net effect of an impeded venous return and the decrease in LV afterload on cardiac output is determined by the volume status and the cardiac function. When LV function is impaired, cardiac output increases in response to rises in ITP as the decrease in LV afterload has a greater effect than the decrease in LV filling pressure.24Pinsky MR Summer WR. Cardiac augmentation by phasic high intrathoracic support in man.Chest. 1983; 84: 370-375Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar When LV function is normal, increase in ITP will reduce cardiac output as the decrease in venous return has more effect than the increase in LV afterload.25Cassidy SA Wead WB Seibert GB Ramanathan M. Geometric left ventricular responses to interactions between the lung and left ventricle: positive pressure breathing.Ann Biomed Eng. 1987; 15: 373-383Crossref PubMed Scopus (5) Google Scholar In our study of patients without severe left ventricular dysfunction, cardiac output was greater during ECWO plus CPAP than with CPPV. Based on the assumption of a smaller ITP during ECWO, the increase in cardiac output can be explained by an improvement in venous return that offset the increase in LV afterload. This corresponds with the findings of an experimental study by Lockhat and co-workers, who found that devices, which confine negative pressure to the thorax and upper abdomen, increase cardiac output by raising the gradient for venous return.26Lockhat D Langleben D Zidulka A. Hemodynamic differences between continual positive and two types of negative pressure ventilation.Am Rev Respir Dis. 1992; 148: 677-680Crossref Scopus (26) Google Scholar The value of ECWO in cardiac failure remains to be clarified. The haemodynamic effects of ECWO and CPPV are similar when cardiac failure has been induced by a β-blocking agent.27Zobel G Dacar D Roedl S. Hemodynamic effects of different modes of mechanical ventilation in acute cardiac and pulmonary failure: An experimental study.Crit Care Med. 1994; 22: 1624-1630Crossref PubMed Scopus (33) Google Scholar Cardiac function improves with external high-frequency oscillation in patients with NYHA physical status I and II after coronary artery bypass grafting.6Sidenoe B Vaage J. Ventilation by external high-frequency oscillations improves cardiac function after coronary artery bypass grafting.Eur J Cardiothorac Surg. 1997; 11: 248-257Crossref PubMed Scopus (15) Google Scholar In patients with cardiac arrest coronary perfusion pressure was augmented by ECWO when compared with standard cardiopulmonary resuscitation. The increase in coronary perfusion pressure was attributed to an increased gradient between the aortic and the atrial pressure.28Smithline HA Rivers EP Rady MY Blake HC Nowak RM. Biphasic extrathoracic pressure CPR. A human pilot study.Chest. 1994; 105: 842-846Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar In conclusion, ECWO during anaesthesia and muscle relaxation is less effective in obese individuals than in normal subjects. CPAP only slightly improves pulmonary gas exchange in overweight individuals. Because of the favourable haemodynamic effects of ECWO, a systematic investigation of methods to increase the efficiency of this technique may be useful.