Mechanical ventilation itself and in particular high airway pressures interact with haemodynamics. The current concept of mechanical ventilation in acute respiratory distress syndrome (ARDS) is to prevent further lung injury by using high positive end-expiratory pressure (PEEP) levels to avoid cyclic end-expiratory lung collapse and by administering low tidal volumes to avoid over-distension of the lung . In this context, high-frequency oscillatory ventilation (HFOV) is a ventilatory mode in which a high fresh-gas flow produces an adjustable level of constant mean airway pressure in a closed respirator circuit. An electromagnetically controlled oscillating piston produces pneumatic pressure cycles via a diaphragm attached to a membrane of the closed patient system. The oscillations are superimposed to the constant fresh-gas flow in the patient system, and induce active inspiratory and active expiratory gas movement. This set-up results in high constant mean airway pressure and very low tidal volumes. All parameters (oscillatory frequency, 3-15 Hz; mean airway pressure, 3-55 cmH2O; fresh-gas flow, 0-60 L min−1 inspiratory time, 30-50% and pressure amplitudes up to 100 cmH2O) are continuously adjustable. The mechanisms of gas transport with HFOV are bulk axial flow, interregional gas mixing, axial and radial augmented dispersion (Taylor dispersion), convective dispersion and molecular diffusion. The carbon dioxide elimination occurs separately from oxygen delivery. Treatment with HFOV in patients with ARDS is coupled to a lung volume optimization protocol, which initially uses high mean airway pressures to achieve optimal lung volumes. HFOV was first used as a rescue technique in neonates with ARDS when conventional ventilation failed. In non-homogeneous lungs with very long inspiratory time constants, HFOV theoretically appears advantageous because the pressure swings are dampened during transmission to the alveoli, and the sustained high mean airway pressure may open slow-recruiting compartments, and keep open fast-collapsing portions of the lungs. HFOV is a safe and effective mode of respiratory support in the treatment of adult patients suffering ARDS, but only a few clinical studies with conflicting data focus on cardiovascular effects of HFOV in adults with abnormal respiratory function [2-6]. Cardiovascular effects of increases in airway pressure are well investigated with conventional ventilation modes. In patients with normal left ventricular and right ventricular function, administration of PEEP is usually accompanied by an increase in right atrial pressure and pulmonary artery occlusion pressure whereas venous return, left ventricular filling pressure, cardiac output and arterial pressure decrease [7,8]. The left ventricular afterload may decline with PEEP, especially in patients with chronic heart failure, and in ARDS patients systolic left ventricular function has been observed to be normal with PEEP [9-12]. It has also been reported that very high PEEP increases right ventricular afterload. Right ventricular enlargement can alter left ventricular filling and performance due to ventricular interdependence (i.e. leftward shift of the ventricular septum with decreased left ventricular compliance and disturbance of septal wall motion) . The aim of the present study was to evaluate early haemodynamic effects during the first 30 min of HFOV treatment with a standardized protocol upon data derived from pulmonary artery catheterization and transoesophageal echocardiography (TOE) measurements.
Study design and patient selection
The prospective observational study was approved by the Institutional Review Board and the State Ethics Committee. Patients with ARDS defined according to the American-European Consensus Conference on ARDS  were eligible after the decision for HFOV treatment was made.
The indications for HFOV were an established diagnosis of ARDS and PaO2/FiO2-ratio <200 despite an optimized pressure-controlled ventilation mode. Optimization included an tidal volume <8 mL kg−1 idealized body weight, inspiratory time of 50% of the respiratory cycle, PEEP ≥15 cmH2O and mean airway pressure ≥22 cmH2O over >4 h. Respiratory rate and inspiratory time during pressure-controlled ventilation were set such as to avoid intrinsic PEEP (monitored by expiratory flow tracings).
Patients were included if they met the following criteria: informed consent from legal representative, normal left and right systolic ventricular function.
Exclusion criteria were: concentric or eccentric ventricular hypertrophy, intracardiac shunt, oesophageal pathology, coronary heart disease, any degree of cardiac valvular stenosis, valvular insufficiency Grade II or more, regional wall motion abnormalities, left bundle branch block and/or atrial fibrillation, severe chronic pulmonary disease (restrictive and/or obstructive), and poor echocardiographic image quality precluding accurate measurements.
Right atrial pressure, mean pulmonary artery pressure, pulmonary artery occlusion pressure, cardiac index, stroke volume index, systemic vascular resistance index and pulmonary vascular resistance index were determined using a Swan-Ganz thermodilution pulmonary artery catheter. Heart rate (HR) was recorded from the electrocardiogram (ECG). Left ventricular end-diastolic area index (LVEDAI), left ventricular end-systolic area index (LVESAI), meridional left ventricular end-systolic wall stress (LVESWS), percent left ventricular fractional area change (FAC), ratio of left ventricular end-diastolic septum to lateral diameter and anterior to posterior diameter (LVED-ratio) and stroke area index (SAI) were determined using transoesophageal echo derived measurements. Early (MVE) diastolic transmitral flow velocity and during atrial contraction (peak velocity during atrial systole, MVA) as well as deceleration time of early transmitral inflow (MVDT) were determined using pulsed doppler measurements.
The following protocol was used for transition to HFOV, which is also the standard operating procedure of our intensive care unit (ICU) for HFOV initiation:
(a) In addition to standard invasive haemodynamic monitoring including pulmonary artery catheter, continuous on-line monitoring of PaO2, PaCO2 and arterial pH was established with an intravascular in-line multiparameter sensor via a femoral artery (Paratrend™ 7; Diametrics Medical Ltd, England).
(b) The mean airway pressure generated by current pressure-controlled ventilation settings was read directly from the ventilator, and the adjusted mean airway pressure of the HFOV was set 5 cmH2O above this value for the first 30 min after transition.
(c) Further HFOV settings were: FiO2 of 1.0, inspiratory time 33% of total respiratory cycle; oscillatory frequency 5 Hz; bias flow 30 L min−1; oscillatory amplitude selected such that PaCO2 remained stable at its level before HFOV.
(d) If, thereafter, PaO2/FiO2 ratio decreased below a value of 50 for >1 min, the patient was switched back to the conventional ventilator at his or her last pressure-controlled ventilation setting.
Protocol for vasoactive and inotropic drug treatment and intravascular volume replacement
The adjusted vasoactive and inotropic drug treatment and the calculated daily intravenous (i.v.) fluid substitution (35 mL kg−1 h−1) were maintained during the study procedure. If mean arterial pressure decreased <60 mm Hg over >1 min during HFOV, resuscitative fluid administration and changes in inotropic or vasopressor management were permitted at the discretion of the attending critical care physician, and recorded.
Routine monitoring was used to record HR, noninvasive oxygen saturation and invasive arterial pressure. Pressure transducers were calibrated and zeroed at the mid-axillary line to atmospheric pressure. The position of the pulmonary artery catheter was verified by transduction of typical waveforms. The pulmonary artery occlusion pressure was defined as the pressure immediately after the atrial contraction (a-wave) according to the ECG. The systemic vascular resistance index (dyn s cm−5 m2) was calculated according to:
Echocardiographic images were recorded on-line and stored using the Sonos 5500 system (Phillips Medical Systems, Germany) together with the ECG on videotape. All echocardiographic measurements were made off-line from the videotape by an experienced cardiologist-echocardiographer blinded to haemodynamic data and ventilator settings. Five cardiac cycles were analysed for each measurement. End-diastole was defined as the onset of the QRS complex and end-systole as the smallest intraventricular area. All area measurements were normalized to body surface area (71.84 × body height (cm)0.725 × body weight (kg)0.4025). Analysed standard view (2-D imaging) was the transgastric left ventricular short axis view at midpapillary level for LVEDAI, LVESAI and left ventricular end-systolic circumferential area index (LVESTAI). Both intracavity areas were determined by manual planimetry of endocardial borders (including the papillary muscles into the cavity). LVESTAI was measured by manual planimetry of epicardial borders in the same view. The LVESWS was determined according to Sutton's formula [15,16]:
The SAI was calculated as
The FAC was calculated as
The LVED ratio was also determined in the transgastric left ventricular short axis view at midpapillary level as
Measurements of MVE, MVA, and MVDT of MVE were obtained using pulsed-wave Doppler in the transoesophageal four-chamber view, with the sample volume of the pulsed Doppler at the level of the mitral valve leaflets.
Haemodynamic and echocardiographic parameters as well as ventilator settings were recorded during pressure-controlled ventilation (10 min before transition to HFOV, i.e. baseline), and at 5 and 30 min after initiation of HFOV. Haemodynamic and echocardiographic measurements during pressure-controlled ventilation were performed in end-expiration, whereas respective measurements during HFOV were performed with HFOV continuing at a frequency of 5 Hz.
Measurements recorded from transoesophageal echo and pulmonary artery catheter were obtained simultaneously within an interval of 5 min by two investigators. Airway pressures and ventilator settings were read directly from the ventilators. Arterial blood gas status was recorded continuously, and PaO2/FiO2 ratio and oxygenation index (mean airway pressure × FiO2 × 100/PaO2) were analysed at the defined time points.
Data are expressed as median (range) and represented as box plots (median, interquartile range and minimum, maximum). Non-parametric testing (Wilcoxon signed rank sum test) with Bonferroni's correction for multiple testing was used to assess intraindividual differences in echocardiographic parameters, haemodynamics, blood gases between baseline (10 min prior to HFOV) and defined time points during HFOV (5 and 30 min of HFOV). P < 0.05 was considered statistically significant. Intra-observer variability for echocardiographic measurements was calculated as percent using the standard method:
where n1-n3 is the parameter of three repeated measurements and m is the mean of these measurements by one investigator.
Nine patients (five male, four female; median age 65 (42-70) years, median height 168 cm (150-185) and median weight 80 (50-100) kg) were studied with complete data sets. At inclusion, all patients were diagnosed with ARDS; 78% had a PaO2/FiO2 ratio of <150, and 22% of 150-200. Underlying diagnoses were in four patients gastric aspiration and in five patients bacterial pneumonia. The median acute physiology and chronic health evaluation (APACHE II) score at baseline was 27 (22-40).
At inclusion, all patients had normal global systolic left ventricular and right ventricular function according to transoesophageal echo without regional wall motion abnormalities.
During HFOV we observed no intervention-related adverse events (e.g. mucus plugging or pneumothorax), nor any arterial hypotension with mean arterial pressure <60 mmHg. The median administered intravenous fluid substitution was 117 (73-146) mL h−1 before and during the study period. At baseline, six patients were being treated with norepinephrine (0.05 (0.04-0.09) μg kg−1 min−1) and four of them additionally with dobutamine (4.1 (2.2-6.0) μg kg−1 min−1). No changes in vasoactive and inotrope medication nor resuscitative fluid replacement were required during 30 min after transition to HFOV. Of the nine patients, three died within 30 days after inclusion due to progressive multiple organ failure.
Mean arterial pressure and HR
Mean arterial pressure and HR showed no significant differences between baseline and measurements during HFOV (Table 1). Specifically, there was no hypotension (mean arterial pressure <60 mmHg) of >1 min duration during the examination (Table 1).
Pulmonary artery catheter data
Compared to measurements at baseline, right atrial pressure increased significantly during HFOV, whereas mean pulmonary arterial pressure showed no differences (Fig. 1). Pulmonary artery occlusion pressure remained unaffected after 5 min of HFOV but increased after 30 min (Fig. 1). The cardiac index decreased slightly after 30 min of HFOV whereas stroke volume index showed only a transient reduction after 5 min of HFOV (Table 1). Systemic and pulmonary vascular resistance indices varied widely but showed no differences between repeated measurements (Table 1).
Mean intraobserver variability was 6 ± 4% for 2-D echocardiographic measurements.
Compared to baseline, a subtle decrease of LVEDAI and LVESAI with HFOV reached statistical significance only after 30 min of HFOV (relative difference to baseline, −24%; Fig. 2). The LVESWS, which gives an estimate for left ventricular afterload, the SAI and the FAC remained unchanged during this period (Table 1). The LVED-ratio was <1.0 at baseline and remained so during HFOV (Table 1). Diastolic filling characterized by MVE and MVA as well as MVDT remained unaffected by transition to HFOV (Table 1).
Due to the transition protocol, adjusted mean airway pressure increased during HFOV (P = 0.004) compared to baseline. Transition to HFOV resulted in a significant increase of PaO2/FiO2 ratio after 30 min and oxygenation index increased after 5 min but was unchanged after 30 min. PaCO2 and arterial pH remained unchanged. A complete overview over the data is given in Table 2.
The main findings of this study are: transition to HFOV with a mean airway pressure adjusted to 5 cmH2O above that of the preceding conventional ventilator setting and subsequent improvement of PaO2/FiO2 ratio leads to statistically significant, but clinically only minor haemodynamic effects. Right atrial pressure and pulmonary artery occlusion pressure, when measured against atmospheric pressure, increased with HFOV, whereas left ventricular cavity areas as well as cardiac index and systemic vascular resistance index decreased, but remained well within normal limits.
Positive airway pressure ventilation interacts with haemodynamics and HFOV might be expected to do so as well. Some animal studies have demonstrated decreased cardiac function with commencement of HFOV at high mean airway pressures, whereas others did not show haemodynamic changes [17-20]. Conflicting data are reported about the impact of HFOV on haemodynamics and cardiac performance in adults; however, as yet, no study has employed TOE for evaluation of haemodynamic and cardiac effects of HFOV. First published haemodynamic data in adults with ARDS after transition to HFOV from conventional ventilation cover the first 4h of treatment, with widely different time points of measurement. Fort and colleagues reported, in 15 patients with ARDS, a brief but significant increase of pulmonary artery occlusion pressure after 3 h of HFOV at a mean airway pressure of 34 cmH2O, when compared to pressure-controlled ventilation at a mean airway pressure of 31 cmH2O . The MOAT trial described a slightly but significantly higher pulmonary artery occlusion pressure in the HFOV group after 2h compared with patients treated conventionally; at the same time, mean airway pressure was higher in the HFOV group . Andersen and colleagues found, retrospectively, no significant central venous pressure changes after 4h of HFOV in a mixed cohort of ARDS patients , with insufficient data on pulmonary artery occlusion pressure or cardiac index. Our study confirms the observations of the former reports that right and left atrial pressures, when measured against atmospheric pressure, increase on transition to HFOV. In our study as well as in those of Fort and colleagues  and Derdak and colleagues , transmural pressure measurements were not part of the protocol. It can be assumed, however, that intrapleural pressures were also higher when mean airway pressure had been increased by 5 cmH2O (3.75 mmHg) with HFOV. As a consequence, the absolute increase of left and right atrial pressures (by a median of 3-4 mmHg) after switching to HFOV does certainly not reflect a concomitant increase of transmural filling pressures by the same degree, but much less or even a net decrease, depending on the individual effective circulating volume of a patient. Since in this series, mean arterial pressure and HR remained remarkably stable without any hypotensive adverse event, a clinically relevant impairment of venous return or atrial filling due to intrathoracic pressure elevation appears unlikely. In another recent study of 42 ARDS patients, we also found no changes in mean arterial pressure after HFOV initiation at adjusted mean airway pressure which were also 5 cmH2O above those during pressure-controlled ventilation .
Despite stable mean arterial pressure, our data reflect a mild decrease of stroke volume index and cardiac index in response to HFOV. Traverse and colleagues found in a cat model that HFOV reduced cardiac output when increasing mean airway pressure. Studies of children and adult ARDS patients treated with HFOV reported both unchanged as well as reduced cardiac output after transition to HFOV [2-5,21,22]. Mehta and colleagues found, in a series of 12 ARDS patients, a 24% reduction of cardiac output immediately after transition to HFOV, when adjusted mean airway pressure had been increased from 24 to 31 cmH2O . In our series, this reduction amounted to only 13-16%, with cardiac index remaining within normal limits. Changes in left ventricular preload and/or systolic function, compression to the cardiac fossa, ventricular interdependence and changes in left ventricular afterload may all contribute to this reduction in stroke volume.
Preload decreases reproducibly during airway pressure elevation, and this mechanism is even utilized to obtain gradual preload reduction in studies of left ventricular systolic function [23-25]. The progressive reduction of LVEDAI after switching to HFOV may be an indicator of this effect. The parameter most widely used clinically to assess left ventricular preload is the pulmonary artery occlusion pressure, and it is known that its correlation to left ventricular end-diastolic volume is poor during continuous airway pressure elevations of >15 cmH2O. Such pulmonary artery occlusion pressure measurements may reflect intra-alveolar pressure rather than left ventricular filling pressure . The present data demonstrates after 30 min of HFOV with increased mean airway pressure, that simultaneously pulmonary artery occlusion pressure increased and left ventricular end-diastolic area and end-systolic area decreased. A similar response of end-diastolic dimensions has been described in newborns during HFOV . As a limitation, left ventricular end-diastolic cavity area is merely a representative of left ventricular end-diastolic volume. Several studies showed a close relationship between area and volume during volume loss and replacement . However, LVEDAI has never been validated as a preload parameter during mechanical ventilation at high airway pressures or even HFOV. Our findings may therefore be viewed as indicators, but not as proof for airway pressure-related preload reduction after transition to HFOV.
A supportive explanation for reductions both of LVEDAI and LVESAI during HFOV would be a direct mechanical compression of the cardiac fossa due to increased lung volume. Lung volume recruitment with HFOV can be inferred from the significant increase in PaO2/FiO2 ratio after 30 min, but there were no chest CT data to illustrate this mechanism.
TOE assessment of left ventricular systolic function was performed by measurement of fractional shortening, ejection fraction and FAC. So far, only one study in pediatric patients used echocardiography for measurements of left ventricular systolic function during HFOV. The authors described a slight but significant decrease in left ventricular fractional shortening . In adults with ARDS, we found no influence of HFOV on FAC as a representative of global left ventricular systolic function.
Furthermore, impairment of left ventricular relaxation due to increased ventricular interdependence has to be taken into account. Administrations of high PEEP levels with increased right ventricular afterload, as well as a left-side shifting of the interventricular septum are described in the literature [9,13]. Fellahi and colleagues reported that PEEP elevation in patients with normal or impaired systolic left ventricular function decreased end-diastolic and end-systolic cavity areas measured by TOE; they observed that the major decrease of left ventricular area was due to a reduction in septum-to-lateral left ventricular diameter . In partial agreement to this, our data show that during pressure-controlled ventilation at high PEEP (baseline), the left ventricular end-diastolic septum-to-lateral diameters were already shorter than anteroposterior diameters; however, we observed no further decrease after transition to HFOV. Although right atrial pressure increased by the step-up in mean airway pressure, mean pulmonary arterial pressure as well as pulmonary vascular resistance index remained unchanged and there was no further leftward shifting of the interventricular septum. Since also transmitral flow velocities, and as markers of left ventricular diastolic function, the ratio of the early and atrial peak velocities, and the deceleration time of the early flow velocity were unaffected by HFOV. Impairment of left ventricular relaxation due to increased ventricular interdependence appears less likely.
The same applies to increases in left ventricular afterload: mean arterial pressure, systemic vascular resistance index and LVESWS did not change significantly, the latter with a trend to decrease rather than increase during 30 min HFOV. Left ventricular afterload of children also decreased or remained unchanged after transition to HFOV with higher adjusted mean airway pressure than during pressure-controlled ventilation [22,27]. Another echocardiographic study also found LVESWS unchanged after initiation of HFOV .
Finally, the major limitations of the present study are due to its observational design and are first, that intrathoracic and pericardial pressure measurements were not available during transition to HFOV; therefore, the effect of our intervention on transmural pressures of the heart could not be quantified. Second, since in all of these patients with severe oxygenation failure, PaO2/FiO2 ratio improved on HFOV without procedure-related complications, we were unable to return the patients at this stage once more to conventional ventilation for experimental reasons only. Thus the relationship between initiation of HFOV in the described manner and changes in the various parameters has to be viewed, at this stage, as descriptive rather than causative.
This study in ARDS patients with normal left ventricular function presents the first simultaneous haemodynamic and echocardiographic assessment during transition to HFOV treatment. Initiation of HFOV with adjusted man airway pressure exceeding those during preceding pressure-controlled ventilation by 5 cmH2O increased right atrial pressure and pulmonary artery occlusion pressure, but reduced end-diastolic and end-systolic left ventricular cross-sectional area indices. Also, cardiac index and stroke volume index decreased but still remained within normal limits. Our findings support the contention that transition to HFOV can induce airway pressure-related preload reduction. This is, however, of minor clinical relevance if preload conditions are optimized already for the conventional ventilation pattern prior to HFOV initiation.
1. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome
. The acute respiratory distress syndrome
network. New Engl J Med
2. Mehta S, Lapinsky SE, Hallett DC, et al.
Prospective trial of high-frequency oscillation in adults with acute respiratory distress syndrome
. Crit Care Med
3. Fort P, Farmer C, Westerman J, et al.
High frequency oscillatory ventilation for adult respiratory distress syndrome-a pilot study. Crit Care Med
4. Derdak S, Mehta S, Stewart TE, et al.,
and the Multicenter Oscillatory Ventilation for Acute Respiratory Distress Syndrome
Trial (MOAT) study Investigators. High-frequency oscillatory ventilation for acute respiratory distress syndrome
in adults. Am J Respir Crit Care Med
5. Andersen FA, Guttormsen AB, Flaatten HK. High frequency oscillatory ventilation in adult patients with acute respiratory distress syndrome
- a retrospective study. Acta Anaesthesiol Scand
6. David M, Weiler N, Heinrichs W, et al.
High-frequency oscillatory ventilation in adult acute respiratory distress syndrome
. Intensive Care Med
7. Downs JA, Douglas ME. Assessment of cardiac filling pressure during continuous positive pressure ventilation. Crit Care Med
8. Pinsky M, Vincent JL, De Smet JM. Estimating left ventricular filling pressure during positive end-expiratory pressure in humans. Am Rev Resp Dis
9. Fellahi JL, Valtier B, Beauchet A, Bourdarias JP, Jardin F. Does positive end-expiratory pressure ventilation improve left ventricular function? A comparative study by transesophageal echocardiography in cardiac and noncardiac patients. Chest
10. Viquerat CE, Righetti A, Suter PM. Biventricular volumes and function in patients with adult respiratory distress syndrome ventilated with PEEP. Chest
11. Terai C, Uenishi M, Sugimoto H, Shimazu T, Yoshioka T, Sugimoto T. Transesophageal echocardiographic dimensional analysis of four cardiac chambers during positive end-expiratory pressure. Anesthesiology
12. Potkin RT, Hudson LD, Weaver LJ, Trobaugh G. Effect of positive end-expiratory pressure on right and left ventricular function in patients with the adult repiratory distress syndrome. Am Rev Respir Dis
13. Jardin F, Farcot JC, Boisante L, Curien N, Margairaz A, Bourdarias JP. Influence of positive end-expiratory pressure on left ventricular performance. New Engl J Med
14. Bernard GR, Artigas A, Brigham KL, et al.
The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med
15. St John Sutton MG, Plappert TA, Hirshfeld JW, Reichek N. Assessment of left ventricular mechanics in patients with asymptomatic aortic regurgitation: a two-dimensional echocardiographic study. Circulation
16. Reichek N, Wilson J, Sutton MGSJ, Plappert TA, Goldberg S, Hirshfeld JW. Noninvasive determination of left ventricular end-systolic stress: validation of the method and initial application. Circulation
17. Osiovich HC, Suguihara C, Goldberg RN, Hehre D, Martinez O, Bancalari E. Hemodynamic effects of conventional and high frequency oscillatory ventilation in normal and septic piglets. Biol Neonate
18. Traverse JH, Korvenranta H, Adams EM, Goldthwait DA, Carlo WA. Impairment of hemodynamics with increasing mean airway pressure during high-frequency oscillatory ventilation. Pediatr Res
19. Lucking SE, Fields AI, Mahfood S, Kassir MM, Midgley FM. High-frequency ventilation versus conventional ventilation in dogs with right ventricular dysfunction. Crit Care Med
20. Zobel G, Dacar D, Rodl S. Hemodynamic effects of different modes of mechanical ventilation in acute cardiac and pulmonary failure: an experimental study. Crit Care Med
21. Simma B, Fritz M, Fink C, Hammerer I. Conventional ventilation versus high-frequency oscillation: hemodynamic effects in newborn babies. Crit Care Med
22. Gutierrez JA, Levine DL, Toro-Figueroa LO. Hemodynamic effects of high frequency oscillatory ventilation in severe pediatric respiratory failure. Intensive Care Med
23. Haney MF, Johansson G, Haggmark S, Biber B. Heart-lung interactions during positive pressure ventilation: left ventricular pressure-volume momentary response to airway pressure elevation. Acta Anaesthesiol Scand
24. Haney MF, Johansson G, Haggmark S, Biber B. Method of preload reduction during LVPVR analysis of systolic function: airway pressure elevation and vena cava occlusion. Anesthesiology
25. Weismann IM, Rinaldo JE, Rogers RM. Current concepts: positive end-expiratory pressure in adult respiratory failure. New Engl J Med
26. Cheung AT, Savino JS, Weiss SJ, Auckburg SJ, Berlin JA. Echocardiographic and hemodynamic indexes of left ventricular preload in patients with normal and abnormal ventricular function. Anesthesiology
27. Goodman AM, Pollack MM. Hemodynamic effects of high-frequency oscillatory ventilation in children. Pediatr Pulmonol