Static versus dynamic respiratory mechanics for setting the ventilator

of 10

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
10 pages
0 downs
British Journal of Anaesthesia 85 (4): 577±86 (2000) Static versus dynamic respiratory mechanics for setting the ventilator M. Lichtwarck-Aschoff 1*, V. Kessler 2, U. H. SjoÈstrand 3, A. Hedlund 4, G.
British Journal of Anaesthesia 85 (4): 577±86 (2000) Static versus dynamic respiratory mechanics for setting the ventilator M. Lichtwarck-Aschoff 1*, V. Kessler 2, U. H. SjoÈstrand 3, A. Hedlund 4, G. Mols 2, S. Rubertsson 3, A. M. MarkstroÈm 5 and J. Guttmann 2 1 Department of Anesthesiology and Intensive Care, University Hospital, Uppsala, Sweden and Zentralklinikum Augsburg, Germany. 2 Department of Anesthesiology and Critical Care Medicine, University of Freiburg, Freiburg, Germany. 3 Department of Anesthesiology and Intensive Care, University Hospital, Uppsala, Sweden. 4 Department of Plastic Surgery, University Hospital, Uppsala, Sweden. 5 Department of Anesthesiology and Intensive Care and Dept of Otorhinolaryngology University Hospital, Uppsala, Sweden *Corresponding author: Klinik fuèr Anaesthesiologie und Operative Intensivmedizin am Zentralklinikum Augsburg, Stenglinstr. 2, D Augsburg, Germany The lower in ection point (LIP) of the inspiratory limb of a static pressure-volume (PV) loop is assumed to indicate the pressure at which most lung units are recruited. The LIP is determined by a static manoeuvre with a PV-history that is different from the PV-history of the actual ventilation. In nine surfactant-de cient piglets, information to allow setting PEEP and V T was obtained, both from the PV-curve and also during ongoing ventilation from the dynamic compliance relationship. According to LIP, PEEP was set at 20 (95% con dence interval 17±22) cm H 2 O. Volume-dependent dynamic compliance suggested a PEEP reduction (to 15 (13±18) cm H 2 O). Pulmonary gas exchange remained satisfactory and this change resulted in reduced mechanical stress on the respiratory system, indirectly indicated by volume-dependent compliance being consistently great during the entire inspiration. Br J Anaesth 2000; 85: 577±86 Keywords: complications, acute respiratory failure; ventilation, positive end-expiratory pressure Accepted for publication: May 25, 2000 Traditionally, three distinct features of the static pressurevolume (PV) loop of the respiratory system 1 are used to assess the mechanical stress imposed by arti cial ventilation to the lung. The lower in ection point (LIP) of the inspiratory limb of the PV-loop is thought to indicate the pressure at which the majority of lung units are recruited. 2 The upper in ection point (UIP) indicates the pressure at which overdistension starts. The linear steep segment of the PV-curve between LIP and UIP is assumed to indicate the zone between end-expiratory collapse and overdistension although there is evidence that re-in ation of collapsed lung units continues above LIP and throughout the linear part of the PV-loop and perhaps even above UIP. Hence, it is hoped that lung injury from both shear stress and overdistension is avoided when tidal ventilation occurs on the linear segment of the PV-curve between LIP and UIP. This concept is intuitively appealing and it may improve ventilatory support in patients. 3 Recent studies have challenged these concepts, however, 4±10 as the mechanical properties of the respiratory system may depend on the pressure-volume history of the current tidal ventilation. 11 The assumption that information obtained from static manoeuvres like the PV-loop can be applied to tidal ventilation may be incorrect, given the pronounced differences in PV-history. Rather than studying the respiratory system under arti cial conditions, we have assessed dynamic mechanics (slice-compliance) 12 during uninterrupted ventilation, and, hence, without affecting the actual PV-history of the lung. We analysed compliance within the tidal breath, taking into account the endotrachealtube related ow-dependent resistance. We studied surfactant-de cient piglets, and obtained information for setting PEEP and V T both from the static PV-curve and from the slice-compliance technique. We compared settings of PEEP and V T derived from static mechanics with those derived from the slice-compliance curve, in terms of the mechanical stress imposed to the Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2000 Lichtwarck-Aschoff et al. respiratory system (end-inspiratory plateau pressure (P plat ) and the shape of the volume-dependent compliance curve), and gas exchange (arterial PO 2 ). Materials and methods The investigations were performed at the experimental laboratories of the Department of Anaesthesiology and Intensive Care Medicine at the University Hospital of Uppsala, Sweden, and were conducted in conformity with the Helsinki convention for the use and care of animals. The Local Ethics Committee for Animal Experimentation reviewed and approved the study. We studied nine healthy male and female piglets of Swedish landrace breed (25 (SD 2) kg). Anaesthesia and uid management Lavage Lavage was performed as previously described with 11 broncho-alveolar lavages (1.2±1.5 litres normal saline, corresponding to 50±60 ml kg ±1 ) which we have found causes a Pa O2 /FI O2 of 20 kpa. Between each lavage, the animals were ventilated for 5 min with pressure-controlled ventilation, FI O2 1.0, PEEP 15 cm H 2 O and V T 15 ml kg ±1. The effects of lavage on gas exchange, respiratory mechanics, and extravascular lung water (double-indicator dilution ) were compared between volume-controlled ventilation at ZEEP during healthy conditions and the surfactant-de cient conditions immediately after lavage. After lavage the animals were allowed to stabilize for 20 min. Study design After lavage, all animals underwent volume-controlled ventilation with the ventilator set according to both static (STAT) and dynamic (DYN) measurements, each setting being applied for 40 min (see Fig. 1). To compensate for time-related effects and to use the individual animals as their own controls, the animals were allocated randomly to settings by either method rst. The effects of STAT and the DYN settings were also compared with the healthy conditions before lavage, with PEEP set to 4 cm H 2 O (PEEP 4). Ventilator and acquisition of pressure and ow data The animals' lungs were ventilated through an endotracheal tube (#8, Mallinckrodt, Athlone, Ireland), connected by a 60 cm rigid tubing system to a Servo 300 ventilator (Siemens- Elema, Solna, Sweden). Airway pressure and ow were continuously measured with a pressure- ow transducer (BICORE CP100, Bicore Monitoring Systems, Irvine, CA, USA) placed between the tracheal tube and the ventilator circuit. The signals were sampled at 50 Hz and fed to a computer to be processed for the analysis of dynamic mechanics. Flow curves were continuously displayed and Fig 1 Schematic drawing of the study protocol. For further details please refer to the text. checked for ongoing ow at end-expiration as well as for any sign of respiratory muscle activity. Static measurements for ventilator adjustment Static mechanics were used to determine the LIP and the UIP for setting PEEP (at the level of LIP) and V T (below UIP). The lower in ection point of the static PV curve of the respiratory system was determined by a modi ed multiocclusion technique 17 as follows: a recruitment manoeuvre (see below) was performed to standardize the PV-history of the lung. Ventilation was started at zero PEEP, frequency of 16 min ±1, I:E 1:1 and with constant inspiratory ow. Seven breaths with a V T of 50 ml each were applied. At the end of the last breath, an end-inspiratory hold of 5 s was performed and the end-inspiratory airway pressure (P plat ) noted. For the next step, V T was increased to 100 ml at otherwise identical settings. The procedure was repeated with V T increased in steps of 50 ml up to 600 ml (corresponding to 24 ml kg ±1 ). The resulting values of plateau pressures after a 5 s end-inspiratory hold were used to construct the in ation limb of a static PV-loop (PV-curve) from which LIP and UIP were determined by tracing a straight line on the linear part as the best t by eye. Dynamic measurements for ventilation adjustment Dynamic measurements were made to adjust the DYN settings and also to estimate indirectly the mechanical stress on the respiratory system after 40 min of ventilation at DYN and at STAT settings. To detect non-linearities in dynamic respiratory system compliance within the tidal volume range, the slice method was used which measures volume-dependent dynamic compliance and resistance breath by breath. The method continuously calculates tracheal pressure (P trach ) 19 by subtracting the ow-dependent resistive pressure drop caused by the resistance of the endotracheal tube (ETT) from the pressure measured at the airway opening. The resulting P trach ±V T loop is divided into consecutive volume slices and mean compliance (intrinsic PEEP considered) and mean resistance (ETT resistance excluded) is calculated for each slice by repeated application of the linear 578 Dynamic mechanics for setting the ventilator Fig 2 The slice-method (illustrated in one representative animal with surfactant de ciency). (A) Pressure as measured at the airway opening is plotted against volume (PAW/V-loop). Upward arrow denotes inspiration, downward arrow expiration. (B) The tracheal pressure P trach (inner loop, dashed area) is calculated point by point according to reference 19. The pressure difference between the outer P AW /V-loop and the inner P trach /V-loop mainly represents the ow-dependent resistive pressure drop across the endotracheal tube (ETT). The pressure- ow relationship of the ETT used had been determined in the laboratory beforehand. The P trach /V-loop is subdivided into eight slices (indicated by horizontal lines) and respiratory mechanics are analysed separately for each slice. The upper and lower 5% of the tidal volume (V T ) (i.e. slices 1 and 8, respectively) are excluded from analysis because of interference due to the ventilator's valve and the large volume acceleration. The remaining 90% of V T are divided into 6 slices, each comprising 15% of V T. One volume-dependent dynamic compliance and resistance of the respiratory system (C rs,dyn and R rs,dyn ) are calculated per slice. (C) Quality check: P trach is recalculated point by point for each slice, using the calculated values for C rs,dyn and R rs,dyn and the measured volume and ow. This recalculated P trach /V-loop is superimposed on the measured P trach /V-loop, and the pressure difference between both loops re ects the accuracy of the calculated mechanical parameters. The measured and recalculated loops in C are barely distinguishable. (D) Plot of C rs,dyn over the slices of the tidal volume. The shape of this particular plot indicates overdistension after about one third of the tidal volume has been delivered (i.e. beginning with the third slice), reduction of V T should be considered. (Values are mean of 20 consecutive breaths; SD omitted for the sake of readability.) resistance-compliance model (RC-model). To reduce the effect of cardiac oscillations, the slice values of 20 consecutive breaths were averaged and used for the analysis (see Fig. 3; for the sake of clarity, the SD for each slicecompliance value is omitted in Figs 2 and 5). Combining the compliance and resistance values of all the slices gives the course of compliance and resistance within V T (i.e. the volume-dependent compliance and resistance within one breath). The interpretation of the slice-compliance over V T plots was as follows: the compliance curve represents the slope, i.e. the rst derivative of the S-shaped PV-curve. The derivative of an S-shaped curve results in a trapeziform curve. 7 The lower bow of the S-curve is the ascending part and the upper bow is the descending part of the trapeziform compliance curve. The steepest middle segment of the S- curve is the horizontal and highest part of the compliance curve, which was the target of our DYN settings. The advantage of considering the derivative (compliance) curve rather than the original (PV) curve is that compliance, as a differential value, is very sensitive to changes in the shape of the PV-curve. A horizontal course of the dynamic slicecompliance within a single tidal volume suggests that a constant volume change per pressure change be obtained. If an ascending shape was present, PEEP was increased to see whether this resulted in a greater initial compliance level and a longer horizontal course of the compliance over the V T. If a descending shape was found, an inappropriately high PEEP and/or V T was assumed and adjustments were made (see below) to obtain a horizontal slice-compliance curve at a high absolute compliance level. Before each study mode the animals' lungs were thoroughly suctioned. To make sure that any potential partial tube obstruction due to secretions or kinking had not in uenced the results, the expiratory ow curves were 579 Lichtwarck-Aschoff et al. Fig 3 Representative plot (animal #8) of slice-compliance over tidal volume to illustrate the procedure for setting PEEP and tidal volume. (A) Slice compliance (C rs,dyn ) starting from low level and the descending shape of the plot indicate overdistension at the onset of inspiration. (B) A reduction of PEEP and V T result in an increase in the level of C rs,dyn, overdistension is still prominent. (C) A further reduction of both PEEP and V T results in decreased C rs,dyn, recruitment during the early phase of the delivery of V T and overdistension continuing at end-inspiration. (D) An increase in PEEP results in a more horizontal shape of the compliance plot for the major part of the tidal volume, although there is still overdistension at endinspiration. Grey area: Tidal volume. Values are mean (SD) for 20 consecutive breaths). Settings and resulting gas exchange: (A) PEEP 20 cm H 2 O, V T 330 ml, Pa O2 17 kpa, Pa CO2 5.8 kpa, stroke index (SI) 30 ml; (B) PEEP 15 cm H 2 O, V T 300 ml, Pa O2 18 kpa, Pa CO2 6.0 kpa, SI 45 ml; (C) PEEP 13 cm H 2 O, V T 280 ml, Pa O2 15 kpa, Pa CO2 6.2 kpa, SI 48 ml; (D) PEEP 15 cm H 2 O, V T 250 ml, Pa O2 18 kpa, Pa CO2 6.2 kpa, SI 46 ml. Please note: there is a 90-min interval between situation A and the nal situation D during which the condition of the lung probably changes, affecting the shape of the plot in addition to changes in PEEP and V T settings. screened off-line, using a recently developed method. 20 This method analyses the time-constant of passive expiration as a function of the expired volume to detect tube obstruction. Ventilator settings Ventilatory frequency was 25 min ±1, FI O2 0.3, inspirationto-expiration ratio 1:1 and inspiratory ow constant during all study modes. Both mechanics as well as Pa CO2 were taken into account for setting PEEP and V T : For the STAT settings, PEEP was set to the level of LIP, and V T initially at 10 ml kg ±1. During the subsequent 40 min, blood gases were checked every 10 min and V T was adjusted with respect to Pa CO2, while PEEP was adjusted with respect to the two point compliance of the respiratory system (C rs, 2P ). If Pa CO2 was above 6.5 kpa, V T was increased in steps of 1 ml kg ±1 until Pa CO2 5.5±6.5 kpa. With Pa CO2 5.5 kpa, V T was reduced in steps of 1 ml kg ±1. Airway pressures were also measured every 10 min and C rs, 2P was determined. If C rs, 2P decreased to 90% or less of its initial level where PEEP=LIP, reduction of PEEP in steps of 2 cm H 2 O was considered. For the DYN settings, PEEP was initially set at 12 cm H 2 O and V T at 10 ml kg ±1, which we have found gives adequate gas exchange in most animals. During the subsequent 40 min, blood gases were checked and dynamic mechanics were analysed every 10 min, and PEEP and V T were adjusted for an approximately horizontal shape of the slice-compliance curve. If an ascending shape of the slicecompliance appeared, PEEP was increased in steps of 2 cm H 2 O. If a descending shape appeared, PEEP was reduced in steps of 2 cm H 2 O. V T was adjusted to keep the Pa CO2 within 5.5±6.5 kpa and increased or decreased in steps of 1 ml kg ±1. (A representative example of the approach is given in Fig. 3.) Respiratory mechanics determined at PEEP and VT after ventilator settings made using static/dynamic measurements The end-inspiratory plateau pressure (P plat ) for both the STAT and the DYN settings was determined by performing an end-inspiratory hold for 5 s using the inspiratory hold function of the ventilator. The hold was performed with the V T and the PEEP level applied that had been set according to static or dynamic mechanics. Two point compliance of the respiratory system (C rs, 2P ) was calculated according to the formula: Tidal volume/ (end-inspiratory pressure ± end-expiratory pressure). To measure the end-expiratory pressure, the expiratory hold function of the ventilator was used for 5 s. C rs, 2P was determined with the V T and the PEEP level applied that had been set according to static or dynamic mechanics, respectively. Intrinsic PEEP was considered when expiratory ow had not decreased to zero at end-expiration. Re-expansion The lungs were re-expanded immediately after lavage, as well as before STAT and DYN by a 5-min period of pressure-controlled ventilation with a frequency of 20 min ±1, I:E 1:1, FI O2 0.3, PEEP 25 and a peak inspiratory airway pressure of 50 cm H 2 O. The re-expansion effect immediately after lavage was assessed in terms of C rs, 2P setting the ventilator for 2 min to the pre-lavage PEEP 4 settings. 580 Dynamic mechanics for setting the ventilator Monitoring Intravascular catheters were surgically placed to measure central venous, pulmonary artery (via the external jugular vein), and aortic pressures (via the carotid artery). The position of the catheters was con rmed by pressure tracing. Cardiac output was determined from arterial thermodilution curves 21 (Pulsion Medical Systems, Munich, Germany). Unlike measurements of right heart ow with thermodilution in the pulmonary artery, arterial thermodilution is in uenced minimally by ventilation-induced intrathoracic pressure changes. The anaesthetised-paralysed animals were studied in the physiological prone position. At the end of the experiment, the animals were killed with potassium chloride. Anesthesia and uid management Anaesthesia was induced with an injection of tiletamine 3mgkg ±1 ; zolazepam 3 mg kg ±1 ; xylazine 2.2 mg kg ±1 ; atropine 0.04 mg kg ±1 intramuscularly and deepened with ketamine 100 mg, and morphine 1 mg kg ±1 i.v. Anaesthesia was maintained with infusions of ketamine (20 mg kg ±1 h ±1 ) and morphine (0.5 mg kg ±1 h ±1 ), and muscle relaxation obtained by continuous infusion of pancuronium bromide (0.25 mg kg ±1 h ±1 ). The animals were given a solution of 4.5 g litre ±1 NaCl with 25 g litre ±1 glucose (Rehydrex, Pharmacia Infusion AB, Uppsala, Sweden) at 10 ml kg ±1 h ±1 and a bolus of dextran ml kg ±1 (Macrodex 70, Pharmacia Infusion AB) to ensure normovolaemia. Results Effects of lavage With lavage, EVLW increased from 7 (95% con dence interval 6±8) to 19 (15±21) ml kg ±1. Mean pulmonary artery pressure increased from 19 (13±22) to 34 (26±38) mm Hg, calculated venous admixture (Q va /Q t ) increased from 4 (3± 4.5) to 47 (38±55)%; C rs,2p decreased from 35 (29±41) to 14 ml cm H 2 O ±1, and Pa O2 /FI O2 decreased from 72 (60±(8± 16)88) to 12 (6±18) kpa, all determined at zero PEEP (P 0.01 for all differences). The lower in ection point was 20 (17±22) cm H 2 O (see Fig. 4 for the individual PVcurves). Upon re-expansion immediately after lavage, C rs,2p was 27 (20±36) ml cm H 2 O ±1 (P 0.05 to PEEP 4). Data presentation and statistics Data are presented as mean (SD) or mean (95% con dence interval). Differences were evaluated with a non-parametric analysis of variance (Friedman test). Signi cant differences were evaluated using the paired sign test with correction for multiple comparisons, 22 and signi cance was accepted with P 0.05. Where appropriate, the exact P value is also indicated. Fig 4 Pressure volume (PV) curves with lavage-induced surfactant de ciency. T
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks