Upsala J Med Sci 105: 17-30,2000 Impact of Different Inspiratory Flow Patterns on Arterial C 0 2 - tension Agneta Markstrom,'. Anders Hedlund? Michael Lichtwarck-Aschoff,'. Anders Nordgren,' and Ulf Sjostrand' 'Department of Anesthesiology and Intensive Care, University Hospital, Uppsala, Sweden, 2Department of Otorhinolaryngology, University Hospital, Uppsala, Sweden, 31nstitute of Anesthesiology and Surgical Intensive Care Medicine, Zentralklinikum Augsburg, FRG, 4Department of Plastic Surgery-Burns Unit, University Hospital, Uppsala, Sweden ABSTRACT Ventilation with decelerating inspiratory flow is known to reduce the dead space fraction and to decrease PaCO2 . Constant inspiratory flow with an end-inspiratory pause (EIP) is also known to increase the removal of CO,. The aim of the study was to elucidate the effect of the pauseho-flow period while both the pattern and rate of inspiratory flow was unchanged, and when the lung was ventilated with sufficient PEEP to prevent end-expiratory collapse. Surfactant depleted piglets were assigned to decelerating or constant inspiratory flow with 24 breaths per minute (bpm) o r 1 2 bpm, or to constant flow, without and with an end-inspiratory pause of 25%. By adding an EIP the total time without active inspiratory flow of the respiratory cycle was kept unchanged. Gas exchange, airway pressures, functional residual capacity (using sulfurhexafluoride) and haemodynamics (thermo-dye indicator dilution technique) were measured. Irrespective of ventilatory frequency, PaCOz was lower and serial dead space reduced with decelerating flow, compared with constant inspiratory flow. With an end-inspiratory pause added to constant inspiratory flow, serial dead space was reduced but did not decrease PaCO, . The results of this study corroborate the assumption that total time without active inspiratory flow is important for arterial C0,-tension. INTRODUCTION Examples of newer means of mechanical ventilation are modified forms of pressure-controlled ventilation (PCV), High Frequency Positive Pressure Ventilation (HFPPV), Pressure Support Ventilation (PSV), and Airway Pressure Release Ventilation (APRV), (6,10,21,27). Under these modes a square wave of pressure is intermittently applied to the airway opening, and alveolar pressures and volumes are established during inspiration. Opinions differ as to whether improvements in gas exchange and lung mechanics can be specifically attributed to particular types 2-100126 17 of inspiratory flow pattern (1,3,5,28). In patients, the pathophysiology of acute respiratory failure is complex, having many causes and effects. This diversity is an important consideration when applying mechanical ventilation. Physiological and theoretical arguments favour strategies that avoid tidal alveolar collapse (12.26) and maintain transalveolar pressure within normal limits. Compared with constant flow inspiration, ventilation with decelerating inspiratory flow is known to reduce the dead-space fraction due to improved distribution of inspired gas (7,9). As the rapid gas inflation gives inspired gas a longer residence time, it presumably enhances intrapulmonary gas mixing (20) and improves Co;! exchange (11). Volume-controlled ventilation guarantees minute ventilation but allows airway pressure to increase as impedance increases. An end-inspiratory pause (EIP) is known to improve ventilation and is commonly used during volume-controlled ventilation (VC-EIP) (8,9,17). Several studies have shown improved gas distribution and increased removal of C 0 2 when using an end-inspiratory pause, but most studies have failed to show improved arterial oxygenation (28,11,17,). Modell and co-workers (20) claim that gas exchange impairment must exist before significant response to flow pattern can be detected. In this study the PEEP level was set sufficient high to prevent end-expiratory collapse, thereby no improvement in oxygenation was expected. The aim of the study was to elucidate the effect of the pauselno-flow period while both the pattern and rate of inspiratory flow were kept unchanged. In order to arrange similar conditions during the active inspiration we added an EIP to the constant inspiratory flow. The respiratory cycle is then divided into 2.5 seconds (s) active inspiration both with and without addition of an EIP. By choosing 12 breaths per minute (bpm), the period of no-flow with decelerating inspiratory flow corresponded to a pause of 25% with constant inspiratory flow MATERIAL AND METHODS Twelve Swedish Landrace piglets, with a mean body mass of 28 kg (+ 4), were subjected to lavage. Surfactant was removed (15,22). Anaesthesia (18,29) was induced with tiletamine 3 mg x kg-1+ zolazepam 3 mg x kg-1 and xylazine 2.2 mg x kg-l and atropine 0.04 mg x kg-' given intramuscularly, followed 5 min later by a ketamine infusion at 20 mg x kg-l x h - l . In addition, 20 mg of morphine was given iv. before the initial tracheotomy, preparation and introduction of intravascular catheters. Pancuronium bromide was given as relaxant as a continuous infusion at 0.26 mg x kg-1 x h - l . The animals' lungs were ventilated through an 8 mm diameter 18 Mallinckrodt endotracheal tube (Mallinckrodt Inc, Glens Falls, N.Y., USA) with a Servo 300 (Servo Ventilator 300, Siemens-Elema AB, Solna, Sweden). To maintain the animals body temperature at 38.0 C (k0.6) a thermostatically controlled heating pad was used. The animals were given 0.45% NaCl with 2.5% glucose (Rehydrex, Pharmacia Infusion AB, Uppsala, Sweden ) at 20 ml x kg-1 x h-1 and a bolus of 15 ml x kg-1 of dextran-70 (Macrodex 70, Pharmacia Infusion AB) to ensure normovolaemia. The investigations were performed at the Experimental Laboratories of the Department of Anaesthesiology and Intensive Care at University Hospital, Uppsala. The local medical ethics committee for animal experimentation reviewed and approved the protocol. Monitoring Intravascular catheters were surgically placed for the measurement of central venous, pulmonary arterial (via the external jugular vein) and aortic (via the carotid artery) pressures. The position of each catheter was confirmed by pressure tracing on a bedside monitor (Siemens Sirecust) and recorded with reference to the mid-thorax and at end-expiration level. Arterial and venous blood gases were measured (ABL 300/OSM 111, Radiometer A / S Denmark). Cardiac output (14,16.23) and extravascular lung water P T V ) , were estimated using a COLD System (Pulsion Medizintecknik KG, Germany). Airway pressures were obtained from the digital displays of the ventilator. Every morning a pre-use hnctional check was performed according to the schedule set out in the operating manual for the ventilator and the ventilator’s pressure and flow transducers had been calibrated with independent devices. The static chest-lung compliance (CLT) was calculated according to the formula CLT = Tidal volume x (end-inspiratory pressure - expiratory pressure)-1 , but with appropriate modifications to make allowance for the compressible volume (24,25). When the end-inspiratory occlusion pressure and the total end-expiratory pressure (PEEP) were measured, the ventilator’s hold function was used for 5 s before the equilibrium values were noted. To measure functional residual capacity (FRC), serial dead space (SDS), the SF6 tracer gas wash-idwash-out method was used (13). FRC was calculated as the total volume of SF6 washed-out divided by the alveolar concentration at the end of wash-in. SDS was obtained from the first few expirations during wash-out, and was defined as the volume expired when the SF6 concentration reached 50% of the alveolar concentration in that breath. In our laboratory the coefficient of variation for three sequential measurements in 9 animals for FRC, under a broad range of tidal volumes and different flow conditions, was 1 f 0.08%. 19 Recruitment Immediately after lavage, the surfactant depleted lungs were recruited with I:E 1: 1 and the external PEEP was set to 25 cmH,O and with a tidal volume to produce a peak inspiratory pressure (PIP) of 50 cmH20 during cyclic ventilation for 5 min. Ventilatory settings A Servo Ventilator 300 (Siemens-Elema AB, Solna, Sweden) provides both Volume Controlled ventilation (VC; constant inspiratory flow) and Pressure-Regulated Volume Controlled ventilation (PRVC; decelerating inspiratory flow). Experimental procedure Following anaesthesia and preparation, the animals were placed in prone position. Bronchoalveolar lavage was performed as described previously (16, 22). After lavage, a recruitment procedure was performed. Ventilation during the study was with inspiration-expiration ratio of 1:1, F I 0 2 0.3, and constant tidal volume. PEEP was set to 13 cm H,O. In this animal lung model several studies have proven that PEEP of 13 cm H,O is sufficient to prevent alveolar collapse. This PEEP level was maintained during the whole study. At 24 bpm the animals were normocapnic but hypercapnic at 12 bpm. To study how different inspiratory flow patterns utilized inspiration time, we used recordings from the inspiratory-expiratory flow (see Fig 1). The animals were then assigned to either decelerating or constant inspiratory flow without EIP at 24 or 12 bpm, or to constant inspiratory flow at 12 bpm with an E P of 25% added . The respiratory cycle with 12 bpm was divided into 2.5 s active inspiration both with and without EIP. By adding the pause of 1.25 s (25%) the total inspiration time was 3.75 s, resulting in an I:E ratio of 3:l. Each setting was applied for 15 min before measurements. Calculations and statistics All ventilatory volumes and derived parameters have been converted to BTPS conditions. Indexed values are related either to body mass, or to body surface area (BSA). Differences were evaluated with a non-parametric analysis of variance (Friedman test). If significant differences were detected, these differences were evaluated using the paired sign t test. A standard statistics package was used (Statview, Abacus Concepts, Berkeley, CA). Statistical significance was accepted at * p l 0.05. 20 RESULTS The results are presented in Tables 1 and 2 and in Figures 1 and 2. With all modes under study the tidal volumes remained constant, which produced normocapnia at 24 bpm, though at 12 bpm the animals were hypercapnic. Table 1 presents the results for decelerating versus constant inspiratory flow at 24 bpm and normocapnia. In Table 2 the results for decelerating versus constant inspiratory flow without and with 25% EIP and hypercapnia are presented. During a short period, postlavage with zero PEEP ventilation and F I 0 2 1.0, P a 0 2 was reduced from 98 a4 to 9 +2 kPa, while venous admixture (QvdQt) increased from 7 to 32.5%. During the postlavage studies with a PEEP level of 13 c m H 2 0 and I302 of 0.3, the (QvdQt) ranged between 6% and 14%. Airway pressure With 24 bpm and decelerating flow, peak inspiratory pressure (PIP) was 25 +3 cmH20 and with constant inspiratory flow PIP was 28 +3 cmH20. At 12 bpm and decelerating flow, PIP was 24+2 cmH20, with constant inspiratory flow 25 a3 cmH20, increasing to 27+3 cmH20 with an EIP. PaCO,, serial dead space, functional residual capacity Decelerating flow at 24 bpm yielded a PaCO2 of 6.3 +2 kPa, and with constant inspiratory flow, 6.9 +1 kPa. At 24 bpm and decelerating flow, serial dead space was 146 +13 ml, and for constant flow inspiration, 156 +14 ml. At 12 bpm and decelerating flow, PaCO2 was 8.4 +1 kPa, with constant inspiratory flow, 9.1+1 kPa, and 8.9 +1 kPa with an EIP. Serial dead space (SDS) was 141k8 ml for the decelerating flow and it was reduced to the same extent (141 +12) for constant inspiratory flow with EIP. For significances see Tables 1 and 2. DISCUSSION In a previous study (19) we found that PaCO, was lower with the decelerating inspiratory flow than with constant inspiratory flow. We assumed that this could be related to how the flow patterns distribute and redistribute the gas. The aim of the present study was to elucidate the effect of the pauselno-flow period while both the pattern and rate of inspiratory flow was unchanged, and when the lung was ventilated with a PEEP level set to prevent end-expiratory collapse. We found that arterial carbon dioxide tension (PaCOJ was lower and serial dead space reduced with decelerating flow, compared with constant inspiratory flow, irrespective of ventilatory frequency. With an end- 21 I Total PEEP IcmH201 I 1 3 + 1 Functional residual capacity [ml] Pa02 [kPa] I PIP[cmH201 I 25 k 3 *VC24 10632216 1075 +204 16 +4 18 2 2 Mean airway pressure 19+2 *VC24 Tidal volume [ml*kg-l] 12 +2 End-inspiratory hold pressure [cmH2O] Compliance [rnl*~mH20-~1 24 lt3 SDS [ml] s v 0 2 [%I ITBV [ml*kg-11 13 21 146lt13 *VC24 156 lt14 55 211 61 +12 19 +3 21 2 3 28 k3 17+2 12 21 24 23 35 lt6 Table 1. Results for decelerating versus constant inspiratory flow at 24 bprn under norrnocapnia Values are means i1 SD. In the paired t-test on mean difference of a given value "PC24" (decelerating flow at 24 bpm) denotes a significant difference vs "VC24" (constant inspiratory flow at 24 bpm). All ventilatory volumes and derived parameters have been converted to BTPS conditions. Indexed values are either vs body mass, or vs body surface area (BSA). Statistical significance was accepted at: 'p< 0.05. inspiratory pause (EIP) added to constant flow, serial dead space was reduced but no significant difference in PaCO, was seen. After comments on the inspiratory-expiratory flow recordings and the rationale behind the EIP, these findings will be discussed in the following paragraphs. Inspiratory-expiratoryflow recordings These pressurelflow recordings (Figure 1) illustrate that, with decelerating flow, the pre-set inspiratory pressure remains constant throughout inspiration. At 24 bpm the no-flow period is 0.3 s and at 12 bpm it is 1.25 s. During constant inspiratory flow there is a linear increase in pressure while the tidal volume is delivered. With constant inspiratory flow and an EIP, there is also a linear increase in airway pressure which decreases during the pause i.e. no-flow period. The respiratory 22 Post lavage Total PEEP [cmH20] PIP [cmH20] P C 1 2 *10.05 V C 1 2 V C i 2 E l P *20.05 13A 13k1 13+1 24 22 *VCU 25+3 27 +3 *VCi 3 EIP I 2210 1057 Functional residual capacity lmll Mean airway pressure End-inspiratory hold Tidal volume [ml*kg-l] pressure [cmH20] Compliance [rnl*~mH20-~1 I 10332177 I 1128 2200 *PC12 *vc12 1822 *vc12 1622 1922 *VC12,PC12 13 21 13 21 13 21 23 23 2323 2423 *vc 12 3828 42511 3627 ~ Pa02 [kPa] PaCO2 [kPa] SDS [ml] s v 0 2 [%] I 53212 I 53210 I 5529 14 21 13 22 14 21 8.4 21 * v c ~ 9.1 21 8.9 21 141 28 *VC12 154-cl3 141 212 *vc12 ITBV [mL*kg-11 CI [l*min-l*(m2)-1] QvdQt [%I DO;?I [ml*min-I*(m2)-'1 Table 2. Results for decelerating versus constant inspiratory flow at 12 bpm without and with EIP under hypercapnia. Values are means *1 SD. In the paired t-test on mean difference of a given value, "PC12'' (decelerating flowat 12 bpm), denotes significant difference vs "VC12" (constant inspiratory flow at 12 bpm), and vs ''VC12~1p" (constant flow at 12 bpm with end-inspiratory pause of 25%). All ventilatory volumes and derived parameters have been converted to BTPS conditions. Indexed values are either vs body mass, or vs body surface area (BSA). Statistical significance was accepted at three levels: *p< 0.05. 21 24 21 23 2023 6 20.8 6.0 20.8 6.0 20.8 12 2 7 14 26 1 1 24 563 2121 5692140 571 2117 cycle at 12 bpm is divided into 2.5 s active inspiration both with and without EIP. By adding the pause of 1.25 s (25%) the total inspiration time will be 3.75 s, resulting in an I:E ratio of 3:l. Notably: From the figure it is obvious that there was an ongoing flow at expiration i.e. there was an intrinsic PEEP when an EIP was added. Rationale for adding the end-inspiratory pause We wanted to elucidate the effect of the pauselno-flow period. For constant inspiratory flow this was only possible if the pause was added to an otherwise unchanged inspiratory phase. To obtain similar pauselno-flow periods, we chose 12 bpm. With this frequency the no-flow period with decelerating inspiratory flow was 1.25 s, corresponding to an EIP of 25% with the constant 23 inspiratory flow. By adding the 1.25 s pause after the active inspiration in the constant inspiratory mode, the active inspiration time of 2.5 s was unchanged, but the added pause time of 1.25 s increases the total inspiratory time to 3.75 s, reducing expiratory time. Consequently, the inspiratiodexpiration (1:E ratio) increases to 3:1, and also produces an intrinsic PEEP which increased FRC (see Table 2). That was why 12 bpm was chosen, as adding a 24 bpm pause had resulted in an unacceptably intrinsic PEEP. - 1 1 0 0 1 2 3 4 5 Time (seconds) Figure 1 25 20 0, I 15 5 10 f 5 t -0.5 - 1 j d o 0 1 2 3 4 5 20 0, I -0.5 -1 0 1 2 3 4 5 725 " 4 4 I I 4.5 - 1 1 0 0 1 2 3 4 5 Time (seconds) Recordings of ventilatory cycles in one of the piglets with a PEEP level of 13 cmH20 and tidal volume kept constant. Top; decelerating inspiratory flow at 12 bpm (PC,,) and constant flow at 12 bpm without EIP (VC,,). Centre; PC,, and VC,, with end-inspiratory pause (VC,,+EIP). Bottom: PC, and VC,,. These pressure/flow conditions illustrate that, with decelerating flow, the pre-set inspiratory pressure is constant throughout inspiration. With PC,,, the no-flow period is 1.25 s and with PC,,, 0.3 s. During constant inspiratory flow there is a linear increase in pressure while the tidal volume is delivered. With constant inspiratory flow and an EIP, there is also a linear increase in airway pressure which decreases during the pauseho-flow period. The respiratory cycle at 12 bpm is divided into 2.5s active inspiration, both with and without EIP. By adding the pause of 1.25 s (25%) total inspiration time will be 3.75 s, resulting in an I:E ratio of 3.1. 24 PC12 vc12 vc12 + EIP flow i Figure 2. Schematic drawing of inspiratory and expiratory flow curves with PC,, (top), VC,, (centre) and VC 12+EIP (bottom). Left side of the panel: inspiratory flow. Right side: expiratory flow. The time (in seconds) during which flow is active, near no-flow as well as pause time are indicated. The total inspiratory time, and the time during which neither active inspiratory flow nor an inspiratory near no-flow (PC,,) nor an inspiratory pause (VC,,+EIP) exists are also given and labelled as: '%without active inspiratory flow". 25 Mean airway pressure Mean airway pressure is closely related to certain ventilation settings, and influences haemo- dynamic function and ventilator-induced barotrauma. Ventilation with decelerating inspiratory flow has always a higher mean airway pressure, since a greater proportion of volume and flow is delivered earlier during the inspiration period than with constant inspiratory flow. Baker et al (2) hypothesised that the combination of increased mean airway pressure, better intrapulmonary gas distribution, and longer diffusion time all accounted for the improved gas exchange with decelerating inspiratory flow. Constant inspiratory flow with an EIP will also maintain the lung volume and alveolar pressures. The pause adds directly to the mean airway pressure and this will be further accentuated if the pause also causes intrinsic PEEP by shortening expiratory time. This study shows that the mean airway pressure is even higher for the constant flow with EIP. Constant inspiratoryflow with and without an EIP Irrespective of ventilatory frequency, with decelerating flow PaCOz was lower and serial dead space reduced, compared with constant inspiratory flow. This could be due either to early delivery of tidal volume, or to the period of no-flow, or to both. Adding an EIP to the constant inspiratory flow precluded assessment of the impact of the early delivery of tidal volume with the decelerating inspiratory flow. Instead, the intention was to obtain similar active inspiration conditions for both inspiratory patterns. Using constant inspiratory flow with EIP, serial dead space (SDS) was reduced but no statistically significant decrease in PaCO2 could be demonstrated, compared with constant flow without a pause. This finding was unexpected. One reason for the absence of decreased PaCO, when EIP was added could be related to the fact that the “total time without active inspiratory flow” remains the same as for constant flow without a pause, as illustrated in Fig. 2. With decelerating inspiratory flow, flow takes place mainly during the first 0.8 s, thereafter followed by a no-flow period of 1.7 s before expiration for 2.5 s. The “total time without inspiratory flow” was 4.2 s, which was 2.5 s expiration +1.7 s no-flow period. For constant flow without a pause inspiratory flow was ongoing during 2.5 s , followed by a 2.5 s expiration, yielding a ”total time without inspiratory flow” of 2.5 s (expiration). When an EIP was added to the constant flow, the total inspiratory time was prolonged resulting in an I:E 3: 1, but the inspiratory flow continues for 2.5 s followed by a pause of 1.25 s during which flow ceases and time of expiration will be only 1.25 s. 26 Note that, “total time without inspiratory flow” was exactly the same as for constant flow without a pause, viz. 2.5 s expiration 1.25 s + 1.25 s pause. The above is related to what happens when gas is inspired into the lungs; part of it mixes with the resident gas, while some remains unmixed in the conducting airways (4). This unmixed portion con-stitutes the serial dead space (SDS) because it is in series with the mixed gas in the alveolar region and is only partly determined by anatomical factors. The interface of the distal boundary of the dead space moves up the airway during breathholding and down if flow is increased. The SDS is reduced when the flow rate at changeover from inspiration to expiration slows down and especially if the expiration phase is allowed to start slowly. Time is then allowed for mixing to occur and the interface of gas mixing is allowed to move up the airways. With decelerating inspiratory flow, initial flow is high and diminishes rapidly, the major part of the tidal volume being delivered after only 0.8 s. During the remaining 1.7 s of inspiratory time, convective flow ceases and hence the distal boundary at which convective flow equals diffusion moves up the airway, resulting in increased PaCO, elimination. This is in contrast to constant flow both with and without EIP, where flow is ongoing and tidal volume is not delivered until the end of 2.5 s. In summary, the results of this study corroborate the assumption that total time without active in- spiratory flow is important for arterial C0,- tension . ACKNOWLEDGMENTS This study was supported by the Swedish Medical Research Council (project 4252), Stockholm, and the Lions Cancer Foundation, Uppsala, Sweden; and the Laerdal Foundation for Acute Medicine, Stavanger, Norway 27 REFERENCES 1. Al-Saady, N. & Bennett, E. D: Decelerating inspiratory flow waveform improves lung mechanics and gas exchange in patients on intermittent positive-pressure ventilation. Intensive Care Med 11: 68-75. 1985. 2. Baker, A. B., Colliss, J. E. & Cowie, R. W: Effects of varying inspiratory flow waveform and time in intermittent positive pressure ventilation. 11: Various physiological variables. Br J Anaesth 49: 1221-1234, 1977. 3. Bergman, N. A: Effects of varying inspiratory waveforms on gas exchange. Anaesthesiology 4. Bowes, C . L., Richardsson, J. D., Cumming, G. & Horsfield: Effect of breathing pattern on gas mixing in a model with asymmetrical alveolar ducts. J Appl Physiol58: 18-26, 1985. 5. Dammann, J. F,. McAslan, T, C. & Maffeo, C.J: Optimal flow pattern for mechanical ventilation of the lungs. 2. The effect of a sine versus square wave flow pattern with and without an end-inspiratory pause on patients. Crit Care Med 6: 293-310, 1978. 6. Downs, J. B. & Stock, M. C:. Airway pressure release ventilation: a new concept in ventilatory support. Crit Care Med 15: 459-461, 1987. 7. Eriksson, I. & Sjostrand, U: Effects of high-frequency positive pressure ventilation (HFPPV) and general anaesthesia on intrapulmonary gas distribution in patients undergoing diagnostic bronchoscopy. Anesth and Analg 59: 585593,1980. 8. Fuleihan, S . , Wilson, R. & Pontoppidan, H: Effects of mechanical ventilation with end-inspiratory pause on blood gas exchange. Anaesthesia and Analgesia 55: 122-129, 1976. 9. Jansson, L. & Jonsson, B. A: theoretical study on flow patterns of ventilators. Scand J Respir Dis 53: 237-246, 1972. 10. Katz, J. A. & Marks, J. D: Inspiratory work with and without continuous positive airway pressure in patients with acute respiratory failure. Anaesthesiology 63: 598-607, 1985. 11. Knelson, J. H., Howatt, W. F. & DeMuth, GR: Effect of respiratory pattern on alveolar gas exchange. J Appl Physiol29: 329-331, 1970. 12. Lachmann, B: Open the lung and keep the lung open (Editorial). Int Care Med 18: 319-321, 1992. 13. Larsson, A., Linnarsson, D., Jonmarker, C., Jonsson, B., Larsson, H. & Werner, 0: Measurements of lung volume by sulphur hexafluoride washout during spontaneous and controlled ventilation. Further development of a method. Anesthesiology 67: 543-550, 1989. 14. Lichtwarck-Aschoff, M., Zeravik, J. & Pfeiffer; U: Intrathoracic Blood Volume accurately reflects circulatory volume status in critically ill patients with mechanical ventilation. Int Care Med 18: 142-147, 1992. 15. Lichtwarck-Aschoff, M., Nielsen, J. B., Sjostrand, U.H. & Edgren, E.L: An experimental randomized study of five different ventilatory modes in a piglet model of severe respiratory distress. Intensive Care Med 18: 339-347, 1992. 16. Lichtwarck-Aschoff, M., Leucht, S., Kisch, H., Zimmermann, G., Blumel, G. & Pfeiffer: U. Monitoring of right ventricular function using a conventional slow response thermistor catheter. Intensive Care Med 20: 348-353, 1994. 17. Lyager , S: Influence of flow pattern on the distribution of the respiratory air during intermittent positive-pressure ventilation. Acta Anaesthesiol Scand 12: 191-21 1, 1968. 18.Locher, W., Ganter, M. & Fassbender, C. P: Correlation between drug metabolite concentration in plasma and anesteic action of ketamine in swine. Am J Vet Res 51: 391-398, 1992. 19. Markstrom, A. M., Lichtwarck-Aschoff, M., Svensson, B. A., Nordgren, K.A. & Sjostrand, U. H: Ventilation with constant versus decelerating inspiratory flow in experimentally induced acute respiratory failure. Anesthesiology 84: 882-889, 1996. 1 103-07, 1979. 28: 390-395, 1967. 28 20. Modell, H. I. & Cheney, F. W: Effects of inspiratory flow pattern on gas exchange in normal and abnormal lungs. J Appl Physiol46: 1103-07, 1979. 21. Montgomery, AB,. Stager, M. A,. Carrico, C. J . & Hudson, L. D: Causes of mortality in patients with adult respiratory distress syndrome. Am Rev Respir Dis 132: 485-489, 1985. 22. Nielsen, J. B., Sjostrand, U. H., Edgren, E. L., Lichtwarck-Aschoff, M. & Svensson. B.A.: An experimental study of different ventilatory modes in piglets in severe respiratory distress induced by surfactant depletion. Intensive Care Med 17: 225-233, 1991. 23. Pfeiffer, U . J . , Backus, G . , Bliimel, G., Eckart, J . , Muller, J., Winkler, P., Zeravik, J . & Zimmermann, G : A fiberoptics based system for integrated monitoring of cardiac output, intrathoracic blood volume, extravascular lung water, 02-saturation, and a-v differences. In: Lewis, F. R.& Pfeiffer, U. J: eds. Fiberoptics in critical care monitoring. Springer, Berlin, Heidelberg, New York, London, Paris, Tokyo, Hong Kong: pp 114-125, 1990. 24. Pierson, D. J . & Kacmarek, R . M: Foundations of respiratory care. New York: Churchill Livingstone p. 976, 1992. 25. Scott, L.R., Benson, M. S. & Pierson, D. J: Effect of inspiratory flow rate and circuit compressible volume on auto-PEEP during mechanical ventilation. Respir Care 31: 1075- 1079, 1986. 26. Sjostrand, U. H., Lichtwarck-Aschoff, M., Nielsen, J. B., Markstrom, A. M., Larsson, A , , Svensson, B. A., Wegenius, G . A. & Nordgren, K. A: Different ventilatory approaches to keep the lung open. Intensive Care Med 21: 310-318, 1995.. 27. Sjostrand, U: High-frequency positive-pressure ventilation (HFPPV) a review. Crit Care Med 28. Sykes, M. & Lumley, J: The effect of varying inspiratory flow on gas exchange during 29. Wheland, G . & Flecknell, P. A: The assessment of depth of anaesthesia in animals and man. Address for correspondence: Agneta Markstrom, Department of Anesthesiology, University Hospital, S-751 85 UPPSALA, Sweden. Fax: + 46 18 32 18 09 8: 345-364, 1980 anaesthesia for open-heart surgery. Br J Anaesth 41: 374-380, 1969. Laboratory Animals 26: 153-162, 1992. 29