Mechanical ventilation combined with positive end-expiratory pressure (PEEP) was first described in patients suffering from lung oedema caused by cardiac dysfunction . Since its introduction into clinical practice by Ashbaugh and colleagues , this ventilatory strategy is well accepted as part of the standard therapy for patients with acute respiratory failure requiring mechanical ventilation . Several authors showed that PEEP improves oxygenation by increasing lung volume [4-7]. This is caused by recruitment of non-ventilated lung regions, so-called atelectasis [8,9] or by stretching (hyperinflating) already opened alveoli . Today it is assumed, that an inappropriate PEEP may induce secondary lung injury . An increase of airway pressure above lung compliance causes micro-injuries of the alveolar membranes. In case of higher airway pressure alveolar rupture may occur, which leads to pneumothorax or pneumomediastinum.
Gattinoni showed an inhomogeneous distribution of tidal volume and recruitment in adult respiratory distress syndrome (ARDS) . And recently, differences in regional compliances were measured in a large series of ARDS patients . From this, an optimal PEEP should be based on a compromise between recruitment and lung over distension . For the treatment of patients with ARDS and acute lung injury (ALI) regional inhomogeneities of the diseased lung should be considered for optimizing ventilatory settings. In principle regional aeration can be obtained by computed tomography (CT). However, a CT-scanner is not available at the bedside and is associated with radiation load. Electrical impedance tomography (EIT) measures radiation-free at the bedside regional aeration and ventilation [15,16]. Recently, a number of investigators have focused on EIT applications in pulmonary imaging. Adler and colleagues [17,18] used an EIT system to generate dynamic images of impedance change in an uninjured animal lung. EIT described accurately lung volume changes during incremental increases of gas volume in conventional mechanical ventilation. Kunst and colleagues  used an EIT device to describe regional impedance changes in a saline lavaged lung injury model. Impedance changes correlated closely with whole lung pressure-volume relationships quantified by strain gauge plethysmography. The author observed marked differences in the pressure volume curves between the anterior and posterior part of the lung. The same group of investigators described the anterior-posterior lung impedance changes during recruitment and derecruitment in an animal model . This study shows the potential of EIT to provide dynamic, non-invasive information about differences in regional lung ventilation. The aim of the study was to investigate the influence of increasing PEEP on regional ventilation in mechanically ventilated patients at the bedside by EIT.
Patients and methods
After approval of the local Ethics Committee the study was performed in the intensive care unit of the Department of Anaesthesiology, Emergency and Intensive Care Medicine, University Hospital Göttingen, Germany. An informed written consent to participate in this study was obtained from the closest relatives of eight mechanically ventilated patients. For patient characteristics see Table 1.
Inclusion criteria were:
- Mechanically ventilated patient ≥24 h before onset of the study
- ALI (PaO2/FiO2 < 300 mmHg)
- Age ≥ 18 yr
- Clinically indicated arterial catheter.
Exclusion criteria were:
- Terminal illness
- Unstable haemodynamics
- Cardiac pacemakers
- Thoracic or cardiac surgery.
All patients were ventilated with a ventilator (Evita 2; Dräger AG, Lübeck, Germany) in a volume-controlled mode with constant tidal volume and respiratory rate. Tidal volume, inspiratory oxygen (FiO2) concentration and PEEP were set by the attending physician in order to achieve normocapnia (PaCO2 35-45 mmHg) and oxyhaemoglobin saturation (SaO2) >95%. At the beginning of the study PEEP was decreased to 0 mbar and occasionally the FiO2 was increased so that the SaO2 remained above 90%. Thereafter, PEEP was increased stepwise to 5, 10 and 15 mbar approximately after intervals of 40 min when steady-state condition had been achieved. At the end of each study period end-expiratory lung volume (EELV) was measured by an open circuit multibreath nitrogen washout manoeuvre (MBNW). Simultaneously, measurement of regional ventilation was performed by EIT. Additionally, we investigated oxygen saturation, carbon dioxide and arterial oxygen pressure by arterial blood gas samples (ABL 300 and OSM 3; Hemoximeter Radiometer, Copenhagen, Denmark).
End-expiratory lung volume
EELV was calculated by an modified open circuit MBNW [21,22]. Briefly, gas flow was measured with a pneumotachograph (Fleisch No. 2; Fleisch, Lausanne, Switzerland) and a differential pressure transducer (Huba Control, Würenlos, Switzerland) directly connected to a heat and moisture exchanger (Humid-Vent 2; Gibeck Respiration, Väsby, Sweden) at the endotracheal tube. Inspiratory and expiratory gas was continuously sampled via a capillary (length: 3.09 m, flow: 1 mLs−1) connected to the y-piece of the breathing circuit. Nitrogen (N2), oxygen (O2) and carbon dioxide (CO2) were measured by a mass spectrometer (MGA 1100; Perkin Elmer, Pomona, CA, USA). All data were sampled online by an analogue to digital converter (DT2801-A; Data Translation, Marlboro, MA, USA) at a rate of 40 Hz and processed by an IBM AT compatible personal computer. The data acquisition and processing software was programmed with a commercially available software program (Asyst® 4.0; Keithley Asyst, Taunton, MA, USA). Calculation of EELV was performed offline.
Electrical impedance tomography
Regional ventilation was studied by EIT  (APT System MK1, IBEES, Sheffield, UK). Briefly, 16 surface electrodes (Blue Sensor VL-50-K; Medicotest, Olstykke, Denmark) were placed equidistantly on the circumference of the thorax in a transversal plane at sixth intercostal space parasternal and one reference electrode at the abdomen (Fig. 1). For data collection an alternate current (5 mA p by p, 50 kHz) was injected between one pair of adjacent electrodes. The resulting surface potentials depend on the impedance inside the thorax and are measured between the remaining adjacent electrode pairs. All 16 electrode pairs were used, one pair after the other, as injecting electrodes. One EIT image was completed when all pairs of adjacent electrodes had been used once as injecting electrodes.
According to the method described by Barber and Brown the resulting 208 surface potentials were normalized to the mean surface potential (reference) during the measurement period. These normalized surface potentials were subsequently used for the reconstruction of regional impedance changes by a modified back-projection . This modified back-projection algorithm delivers regional impedance changes with respect to the reference in 912 regions-of-interest within a circular area in a 32 × 32 pixel matrix. Inspiration leads to an increase in regional impedance, while expiration leads to a decrease in regional impedance . Repeated measurements of 1000 EIT images with a sampling rate of 10 EIT images per second resulted in a study period of 100 s, in which regional ventilation was calculated.
In this investigation commercially available electrodes were used for EIT measurements. Hairy patients were shaved at the sites where the electrodes were applied to ensure optimal contact. The used software for data acquisition with the EIT system allows measuring continuously the quality of the contact between skin and electrodes. Thereby inadequate electrode contact can be detected.
Calculation of regional ventilation.
Functional electrical impedance tomography (f-EIT) images were calculated to determine regional lung ventilation . The f-EIT image is calculated from 1000 EIT images and shows the regional impedance amplitude of 32 × 32 regions between end inspiration and end expiration. Examples of an f-EIT image at different PEEP are shown in Figure 2. For further evaluation the regions of this f-EIT image were distributed with reference to the highest regional amplitude into four groups:
- None: regions were impedance amplitude was within 0-10% of the reference.
- Bad: regions were impedance amplitude was within 10-30% of the reference.
- Moderate: regions were impedance amplitude was within 30-70% of the reference.
- Well: regions were impedance amplitude was within 70-100% of the reference.
Calculations were performed using the STATISTICA software package (Statistica 5.1; StatSoft Inc, Tulsa, OK, USA) on a personal computer (Pentium III 800 MHz; Microsoft Windows XP). Since the data were not normally distributed, as tested by the Kolmogorov-Smirnov test, all data are presented as median and range (min-max (median)) unless stated otherwise. We subsequently analyse differences at different PEEP by Wilcoxon signed rank sum test. For all statistical tests P < 0.05 was considered to be significant.
Gas exchange and EELV
Arterial carbon dioxide pressure remained stable during all PEEPs (40-42 mmHg). The arterial oxygen saturations were 99-100% and remained also stable during all PEEPs. Oxygenation calculated by the arterial PaO2/FiO2 ratio (Horovitz index) improved significantly during increase of PEEP. Oxygenation varied from 215 mmHg at PEEP 0 to 286 mmHg at PEEP 15, which was an increase of 33%. After return to PEEP 0, oxygenation remained significantly higher (240 mmHg) than the initial oxygenation at PEEP 0 (215 mmHg) (Fig. 3).
EELV improved significantly during increasing PEEP. EELV ranged from 1316 mL at PEEP 0 to 2561 mL at PEEP 15, which was an enlargement of 95%. We found an EELV after return to PEEP 0 (1409 mL), which was significantly higher compared to the initial PEEP 0. All data are summarized in Figure 4.
The ‘none’ ventilated group contains atelectasis, chest wall and mediastinum. During increased PEEP the number of regions decreased by 25% significantly from 540 regions at PEEP 0 down to 406 regions at PEEP 15. Returning to PEEP 0 resulted in 455 regions not ventilated, which were 85 regions less than at the initial PEEP 0. The ‘bad’ ventilated group increased also significantly by 20% from 316 regions at PEEP 0 to 380 regions at PEEP 15. The ‘moderate’ group increased significantly 2.5-fold from 40 regions at PEEP 0 to 100 regions at PEEP 15. And, the ‘well’ group increased significantly from 0 region at PEEP 0 to 34 regions at PEEP 15. All data are summarized in Table 2, Figures 5 and 6.
The aim of the study was to investigate the effect of increasing PEEP levels on regional ventilation in mechanically ventilated patients at bedside with EIT. We found an improvement in oxygenation during stepwise increase of PEEP from 0 to 5, 10, 15 mbar. This improvement in oxygenation was accompanied by an increase of EELV. This increase of EELV was caused by a decrease of non-ventilated lung regions. Non-ventilated regions were recruited toward bad, moderate and well-ventilated regions.
Electrical impedance tomography.
EIT is an imaging technique offering the possibility of regional lung function studies . It was developed in the early 1980s by Barber and Brown and bases on an alternate current injection and voltage measurement via 16 surface electrodes placed around the thorax . All 16 adjacent electrode pairs are used, one pair after the other, as injecting electrodes while surface potentials are measured with the remaining 13 electrode pairs. One data collection cycle is completed when all pairs of adjacent electrodes have been used once as injecting electrodes. The sampling rate of the used EIT system is 10 cycles s−1. The resulting 208 surface potentials (16 current injections × 13 surface potentials) per cycle are normalized to the mean surface potential. These normalized potentials are used by a back-projection for reconstruction of regional impedance changes following changes of aeration in 912 regions within a thoracic plane . EIT was validated in pigs with the established techniques of CT and ventilation scintigraphy comparing regional ventilation during different tidal and lung volumes [24,27]. These studies confirmed that EIT is a suitable technique for detecting regional changes in air content (regional ventilation) and the monitoring of local effects of mechanical ventilation.
EIT is suitable for monitoring regional lung ventilation, regional tidal volume, regional lung volume and functional residual capacity non-invasively at the bedside [28-30]. In neonates, where radiation exposure is undesirable, EIT may even replace X-ray investigations in certain situations . However, EIT should not be applied to obtain morphological information similar to CT or magnetic resonance imaging, since the latter diagnostic methods provide anatomical information with a higher spatial and morphological resolution. The major advantages of EIT are that it is non-invasive, easy to use at the bedside and that data collection can be performed with a high time resolution and acceptable local resolution. According to Hahn and colleagues, the detectable lung volume by EIT ranges from 9 to 29 mL . Barber and Brown found a spatial resolution of approximately 8% of the thorax diameter, so that a resolution of 8 mL is achieved . Since its introduction the hardware and software have been improved continuously [34,35]. Hahn and colleagues developed more advanced algorithms for the analysis of dynamic physiological phenomena with low amplitudes. Thus, they introduced the f-EIT and averaging technique , which enables regional lung function studies.
Until now, regional lung function studies were performed by CT or ventilation scintigraphy of the lungs. However, both methods lead to radiation exposure and cannot be used at the bedside. The time resolution of both methods is rather poor. A fast speed CT is able to scan four CT images per second [36,37]. A ventilation scintigraphy  delivers one regional aeration image per 16 min. Therefore, following changes in regional ventilation during fast physiological events, like PEEP change cannot be detected by these methods.
One important methodological aspect is that it is limited to a single transverse thoracic plane. It has been demonstrated, that aeration gradients are found not only in the anterior-posterior but also in the cephalo-caudal direction [13,39]. During mechanical ventilation the lungs may displace in the caudal direction, so that EIT is measuring different lung planes. This is the same criticism against CT studies, which have been shown to be misleading in the presence of cranial to caudal heterogeneity. Therefore, analysing regional ventilation in a single transversal plane may underestimate heterogeneity of regional ventilation. In future the introduction of additionally EIT planes or an optimized current injection pattern may solve this problem .
An EIT image summarizes impedance variation from the thorax including lung, chest wall and mediastinum induced by varying air content as well as the pulsatile bloodflow  within tissues . In this study we assumed that impedance change above 10% the maximal impedance variation is caused by atelectasis which may lead to an underestimation . In future EIT systems with an improved signal-to-noise ratio may offers the possibility of new concept for the detection of atelectasis . One possible approaches may base on the fact that impedance change in the outer boundary of the EIT image is generated solely by the chest wall and not by atelectasis. Therefore, the limit of impedance change, defining lung regions will be defined for each patient individually.
Effects of PEEP on EELV.
In our patients, suffering from ALI, we found an increase of EELV by 95% if PEEP was raised from 0 to 15 mbar. Gattinoni and colleagues found a comparable increase of lung volume from 18 mL up to 34 mL by rising the PEEP from 0 to 20 mbar, in mechanically ventilated patients suffering from ARDS . The difference in lung recruitment (95% our study vs. 88% Gattinoni study) besides the higher PEEP used in the Gattinoni study might caused by the severity of lung injury (ALI our study vs. ARDS Gattinoni study) and the different methods of lung imaging. Katz and colleagues found a lower increase of lung volume (1.5-2.68 L; increase of 79%) in mechanically ventilated patients suffering from ALI during increase of PEEP 3 up to PEEP 18 compared to our study (1.3-2.6 L; increase of 95%) . Furthermore, they found time-dependent increase of lung volume within 1 min after increase of PEEP. In our study we found a similar time-dependent increase of lung volume, which occurs within 5-8 breaths after PEEP change.
Effects of PEEP on regional ventilation.
By highering PEEP we found an increase of ventilated regions, caused by redistribution from none ventilated regions to bad, moderate and well-ventilated regions (Table 2, Figs 5 and 6). The sustained improvement in regional ventilation seen on return to zero PEEP may be caused by the increase of PEEP in our study protocol similar to a PEEP trial, which is used as a manoeuvre for lung recruitment . In our study the ‘none’ ventilated regions consist of atelectasis, chest wall and mediastinum, because EIT does not discriminate between these tissues. We assumed that chest wall and mediastinum remain constant during the different PEEP levels. Therefore, changes in the number of ‘none’ ventilated regions are induced by a change in atelectasis.
In mechanically ventilated patients atelectasis occurs mainly in the dorsal part of the lungs. This is caused by the ventral-to-dorsal hydrostatic pressure gradient inside the lungs and the reduced movement of the dorsal part of the diaphragm . PEEP prevents the collapse of a given lung region when it is equal to or greater than the hydrostatic pressure superimposed , this is confirmed by our data, which shows a recruitment of non-ventilated regions and an improved gas exchange during higher PEEP levels. We found a shift of the centre of ventilation to the dorsal part of the lungs during stepwise increase of PEEP (Fig. 7). An increase of ventilation in the dorsal part of the lungs and a decrease in the ventral part of the lungs by increasing PEEP was also found in other studies [12,47].
A peak airway pressure (PEEP + inspiratory pressure) above lung compliance may cause stretch of already open alveoli (hyperinflation) and possible leads to micro injuries of the alveolar membranes. Pressure trauma (barotrauma) has been attributed to the peak airway pressure. Attention has also been focused on volume trauma (volutrauma) and on continuous reopening and collapsing of lung parenchyma through the generation of shear forces. Until now, the used techniques in our study do not offer the possibility to detect hyperinflation. Further studies using EIT might inaugurate more advanced algorithms for detection of hyperinflation.
The results show ventilation redistribution during mechanical ventilation from non-ventilated regions to bad ventilated regions as well as to moderately and well-ventilated regions by increasing the PEEP. These findings are consistent with results obtained by lung CT scan studies. However, EIT is a bedside technique and might be an alternative to CT scan to assess aerated lung regions.
This study was supported by departmental funds and by Dräger AG (Dräger AG, Lübeck, Germany).
1. Poulton EP. Left sided heart failure with pulmonary edema. Its treatment with the ‘pulmonary plus pressure machine’. Lancet
2. Ashbaugh DG, Petty TL, Bigelow DB, Harris TM. Continuous positive-pressure breathing (CPPB) in adult respiratory distress syndrome. J Thorac Cardiovasc Surg
3. Mancebo J. PEEP, ARDS, and alveolar recruitment. Intens Care Med
4. Suter PM, Fairley B, Isenberg MD. Optimum end-expiratory airway pressure in patients with acute pulmonary failure. New Engl J Med
5. Dammann JF, McAslan TC. PEEP: its use in young patients with apparently normal lungs. Crit Care Med
6. Kumar A, Falke KJ, Geffin B et al
. Continuous positive-pressure ventilation in acute respiratory failure. New Engl J Med
7. Chapin JC, Downs JB, Douglas ME et al
. Lung expansion, airway pressure transmission, and positive end-expiratory pressure
. Arch Surg
8. Moreci AP, Norman JC. Measurements of alveolar sac diameters by incident-light photomicrography. Effects of positive-pressure respiration. Ann Thorac Surg
9. McIntyre RW, Laws AK, Ramachandran PR. Positive expiratory pressure plateau: improved gas exchange during mechanical ventilation. Can Anaesth Soc J
10. Katz JA, Ozanne GM, Zinn SE, Fairley HB. Time course and mechanisms of lung-volume increase with PEEP in acute pulmonary failure. Anesthesiology
11. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies. Am J Respir Crit Care Med
12. Gattinoni L, Pelosi P, Crotti S, Valenza F. Effects of positive end-expiratory pressure
on regional distribution of tidal volume and recruitment in adult respiratory distress syndrome. Am J Respir Crit Care Med
13. Puybasset L, Gusman P, Muller JC et al
. Regional distribution of gas and tissue in acute respiratory distress syndrome. III. Consequences for the effects of positive end-expiratory pressure
. CT Scan ARDS Study Group. Adult Respiratory Distress Syndrome. Intens Care Med
14. Rouby JJ, Lu Q, Goldstein I. Selecting the right level of positive end-expiratory pressure
in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med
15. Barber DC, Brown BH. Applied potential tomography. J Phys E Sci Instrum
16. Victorino JA, Borges JB, Okamoto VN et al
. Imbalances in regional lung ventilation: a validation study on electrical impedance tomography. Am J Respir Crit Care Med
17. Adler A, Shinozuka N, Berthiaume Y et al
. Electrical impedance tomography can monitor dynamic hyperinflation in dogs. J Appl Physiol
18. Adler A, Amyot R, Guardo R et al
. Monitoring changes in lung air and liquid volumes with electrical impedance tomography. J Appl Physiol
19. Kunst PW, Bohm SH, de Vazquez A et al
. Regional pressure volume curves by electrical impedance tomography in a model of acute lung injury. Crit Care Med
20. Kunst PW, de Vazquez A, Bohm SH et al
. Monitoring of recruitment and derecruitment by electrical impedance tomography in a model of acute lung injury. Crit Care Med
21. Darling RC, Richards DW, Cournant A. Studies on intrapulmonary mixture of gases. Open circuit method for measuring residual air. J Clin Invest
22. Wrigge H, Sydow M, Zinserling J et al
. Determination of functional residual capacity (FRC) by multibreath nitrogen washout in a lung model and in mechanically ventilated patients. Accuracy depends on continuous dynamic compensation for changes of gas sampling delay time. Intens Care Med
23. Barber DC, Seagar AD. Fast reconstruction of resistance images. Clin Phys Physiol Meas
1987; 8(Suppl A):
24. Frerichs I, Hinz J, Herrmann P et al
. Detection of local lung air content by electrical impedance tomography compared with electron beam CT. J Appl Physiol
25. Hahn G, Sipinkova I, Baisch F, Hellige G. Changes in the thoracic impedance distribution under different ventilatory conditions. Physiol Meas
26. Frerichs I. Electrical impedance tomography (EIT) in applications related to lung and ventilation: a review of experimental and clinical activities. Physiol Meas
27. Hinz J, Neumann P, Dudykevych T et al
. Regional ventilation
by electrical impedance tomography - a comparison with ventilation scintigraphy in pigs. Chest
28. Arnold JH. Electrical impedance tomography: on the path to the holy grail. Crit Care Med
29. Hinz J, Hahn G, Neumann P et al
. End-expiratory lung impedance change enables bedside monitoring of end-expiratory lung volume change. Intens Care Med
30. Wolf GK, Arnold JH. Noninvasive assessment of lung volume: respiratory inductance plethysmography and electrical impedance tomography. Crit Care Med
31. Frerichs I, Schiffmann H, Oehler R et al
. Distribution of lung ventilation in spontaneously breathing neonates lying in different body positions. Intens Care Med
32. Hahn G, Hartung C, Hellige G (1998) Bestimmung der Grösse minimal erfassbarer Areale mit Ventilationsstörungen. p. 77
33. Brown BH, Barber DC. Electrical impedance tomography: the construction and application to physiological measurement of electrical impedance images. Med Prog Technol
34. Koukourlis CS, Kyriacou GA, Sahalos JN. A 32-electrode data collection system for electrical impedance tomography. IEEE Trans Biomed Eng
35. Li JH, Joppek C, Faust U. Fast EIT data acquisition system with active electrodes and its application to cardiac imaging. Physiol Meas
1996; 17(Suppl 4A):
36. Weisser G, Lehmann KJ, Scheck R et al
. Performance of electron-beam CT: continuous-volume-scan compared to spiral CT. Radiologe
37. Weisser G, Lehmann KJ, Scheck R et al
. Dose and image quality of electron-beam CT compared with spiral CT. Invest Radiol
38. Burch WM, Sullivan PJ, Lomas FE et al
. Lung ventilation studies with technetium-99 m pseudogas. J Nucl Med
39. Wrigge H, Zinserling J, Neumann P et al
. Spontaneous breathing improves lung aeration in oleic acid-induced lung injury. Anesthesiology
40. Faes TJ, van der Meij HA, de Munck JC, Heethaar RM. The electric resistivity of human tissues (100 Hz-10 MHz): a meta-analysis of review studies. Physiol Meas
41. Wtorek J, Polinski A. The contribution of blood-flow-induced conductivity changes to measured impedance. IEEE Trans Biomed Eng
42. Hahn G, Frerichs I, Kleyer M, Hellige G. Local mechanics of the lung tissue determined by functional EIT. Physiol Meas
1996; 17(Suppl 4A):
43. Hahn G, Thiel F, Dudykevych T et al
. Quantitative evaluation of the performance of different electrical tomography devices. Biomed Tech (Berlin)
44. Lim CM, Soon LS, Seoung LJ et al
. Morphometric effects of the recruitment manoeuvre on saline-lavaged canine lungs. A computed tomographic analysis. Anesthesiology
45. Hedenstierna G, Strandberg A, Brismar B et al
. Functional residual capacity, thoracoabdominal dimensions, and central blood volume during general anesthesia with muscle paralysis and mechanical ventilation. Anesthesiology
46. Gattinoni L, D'Andrea L, Pelosi P et al
. Regional effects and mechanism of positive end-expiratory pressure
in early adult respiratory distress syndrome. JAMA
47. Frerichs I, Hahn G, Golisch W et al
. Monitoring perioperative changes in distribution of pulmonary ventilation by functional electrical impedance tomography. Acta Anaesthesiol Scand