Abstract
BACKGROUND: Critical-care ventilators provide patient circuit compensation (CC) to counteract the loss of volume due to patient circuit compliance. No studies show the effect of inspiratory efforts (indicating maximal value of the muscle pressure waveforms [Pmax]) on CC function. The goal of this study was to determine how Pmax affects volume delivery with or without CC for both volume control continuous mandatory ventilation with set-point targeting scheme (VC-CMVs) and pressure control continuous mandatory ventilation with adaptive targeting scheme (PC-CMVa) modes on the Servo-u ventilator.
METHODS: A breathing simulator was programmed to represent an adult with moderate ARDS with different Pmax. It was connected to a ventilator set to VC-CMVs or PC-CMVa. The change in tidal volume (ΔVT) was defined as the difference between VT with CC on versus off. VT error was defined as the difference between the simulator displayed VT and the set VT with CC on versus off.
RESULTS: For both VC-CMVs and PC-CMVa modes, ΔVT decreased as Pmax increased. The VT error decreased as Pmax increas-ed for VC-CMVs. In contrast, VT error increased on PC-CMVa mode as Pmax increased and peaked 39.0% for Pmax = 15 cm H2O. For both modes, the difference in VT errors for CC on versus CC off decreased as Pmax increased.
CONCLUSIONS: CC corrected the delivered VT for volume lost due to compression in the patient circuit as expected. This compensation volume decreases as airway pressure drops due to patient Pmax.
Introduction
One of the key elements of lung-protective ventilation strategy for patients with ARDS is to control tidal volume (VT) within 4–8 mL/kg of predicted body weight (PBW).1,2 Previous studies showed that the compressible volume of the ventilator circuit can have important effects on the display of delivered VT.3,4 Therefore, the automatic compensation for compressible volume by the ventilator is a critical concern.
To date, many critical-care ventilators provide patient circuit compensation (CC) to counteract the loss of volume in ventilator circuits based on circuit compliance and airway pressure. In general, the volume required to compensate for circuit compliance is calculated by the ventilator as the change in airway pressure (during inspiration) times the circuit compliance (determined during pre-use check). But the specific algorithms used in such compensation are never described in ventilator manuals, making observed behavior difficult to interpret. Inspiratory efforts (indicating maximal value of the muscle pressure waveforms [Pmax]) decrease inspiratory pressure for volume-targeted modes, that is, volume control continuous mandatory ventilation with set-point targeting scheme (VC-CMVs) and pressure control continuous mandatory ventilation with adaptive targeting scheme (PC-CMVa).5 Therefore, Pmax may affect the way CC works.
To the best of our knowledge, previous studies did not investigate the effects of Pmax on the CC function. The goal of this study was to assess how Pmax affect volume delivery with or without CC for both VC-CMVs and PC-CMVa modes on the Servo-u ventilator.
QUICK LOOK
Current Knowledge
The circuit compensation (CC) function in critical-care ventilators is designed to compensate for the loss of volume due to circuit compliance. There is no study investigating the effects of inspiratory effort on CC function.
What This Paper Contributes to Our Knowledge
Increased simulated patient effort resulted in a change in the calculated circuit compressible volume. During VC-CMV the difference in CC were small. At excessive effort (−15 cm H2O) the error in delivered tidal volume during adaptive pressure modes peaked at 40%. Measurement of CC during patient effort alters accuracy if delivered VT in some modes.
Methods
We evaluated the performance of CC on delivered VT using different modes of ventilation using a critical-care ventilator and a breathing simulator.
Breathing Simulator
This study was conducted using the Active Servo Lung 5000 (ASL 5000) (sw3.6; IngMar Medical, Pittsburgh, Pennsylvania) programmed to simulate an adult patient of 64 kg of PBW with moderate ARDS and different Pmax. The ASL 5000 simulator was connected to a computer to analyze and record respiratory parameters.
Simulations using the ASL 5000 have 2 components: (1) a lung model comprising resistance (R) and compliance (C) and (2) an effort model that is a representation of the muscle pressure (Pmus) as a function of time. The lung model parameters are fairly easy to obtain from studies that record respiratory system mechanics. The effort model is more problematic because very few studies have measured Pmus. Furthermore, Pmus parameters can be derived in a variety of ways, such as a digitized signal (eg, esophageal pressure or electrical activity of the diaphragm signal from a neurally- adjusted ventilatory assist catheter). Usually, the ASL 5000 is programmed to represent Pmus as simply a modified sinusoidal function with parameters of frequency, Pmax (the maximal value of Pmus waveform), Increase percentage (time to reach Pmax, expressed as a percent of the total cycle time), Hold percentage (period of no-flow during the inspiratory phase, expressed as a percent of the total cycle time), and Release percentage (time to reach Pmus = 0, expressed as a percent of total cycle time). A passive patient is modeled as Pmax = 0 (making all other settings for Pmus inactive).
For this study, the lung model was programmed as a single constant airway resistance (including endotracheal tube and normal airway resistance) and single linear respiratory-system compliance using evidence-based values for an adult patient with moderate ARDS.6 Kallet et al7 suggested a model for Pmus that we believe is unrealistic because it included a Hold percentage > 0, which we believe is not present in normal breathing. Therefore, the effort models for our study were set in terms of the parameters of the simulated Pmus waveform based on data from a previous study.8 Briefly, actual Pmus waveforms from that study were analyzed to determine representative values for Increase percentage and Release percentage (Fig. 1). A “ruler” was constructed from a group of lines in an Excel spreadsheet, and the timing of Increase percentage and Release percentage was measured by hand. Pmax values were arbitrarily set based on our estimate about how much effort would be needed to generate a normal VT given the lung model parameters. The resultant simulation parameters are shown in Table 1.
Method to estimate effort model parameters for simulation with the ASL 5000 breathing simulator. Muscle pressure (Pmus) waveforms are modified from Kondili et al.8
Simulation Parameters
Ventilator
The Servo-u ventilator (Getinge, Gothenburg, Sweden) was used for all experiments. The ventilator precheck was performed with a standard-length heated wire circuit. The experiment was performed with a filled heated humidifier that was turned off (ie, the experiment was conducted at room temperature). In addition, because the ventilator corrects delivered volume for body temperature and pressure saturated (BTPS) with water conditions, the simulator was set to correct measured values to BTPS.
Two modes were used: volume control, classified as VC-CMVs; and pressure regulated volume control, classified as PC-CMVa.5 Ventilator settings for each mode are shown in Table 2. For all experiments, FIO2 was set at 0.21 and PEEP at 12 cm H2O.
Ventilator Settings
Outcome Variables
VT measured by the ASL 5000 was designated as patient inspiratory VT, corrected for BTPS. The effect of Pmax on CC was defined in 2 ways. First, the change in VT due to CC was calculated as the difference between VT with CC on versus off (ΔVT expressed as a percent) for different levels of Pmax as shown in Equation 1:
Second, we calculated the effect on VT delivery error, where the target VT was that set on the ventilator (VT set) that was compared to that actually delivered as measured by the simulator (VT meas) expressed as a percentage as shown in Equation 2:
Defined this way, error represents the deviation from the desired VT such that positive numbers for error indicate that the actual delivered volume is in excess of what is expected, a situation that could increase risk of VT overdosage. VT error was calculated for both CC on and CC off.
Procedure
The ASL 5000 simulator was calibrated according the manufacturer's instructions. The pre-use check including internal leakage test, patient circuit test with patient circuit compliance, and flow and pressure transducer calibration on the Servo-u ventilator was performed.
Each combination of mode and Pmax value was considered an individual experiment (ie, 8 experiments).
Data Analysis
We recorded the mean value of patient inspired VT as recorded by the ASL 5000 for 10 breaths after stabilization. All calculated values were rounded to the nearest 0.1%.
Results
The effects of simulated Pmax on pressure, volume, and flow waveforms are shown in Figure 2. The first waveform (Pmax = 0) shows the effects for passive inflation. As Pmax increased, inspiratory pressure decreased as indicated by the distorted Pvent waveforms.
Representative pressure, volume, and flow waveforms showing effects of different levels of inspiratory effort, simulated as maximal value of the muscle pressure (Pmax) in cm H2O. The top row is volume control continuous mandatory ventilation with set-point targeting scheme (VC-CMVs) and the bottom row is pressure control continuous mandatory ventilation with adaptive targeting scheme (PC-CMVa). See Table 2 for ventilator settings.
Figure 3 shows the effects of Pmax on ΔVT. During VC-CMVs, the ΔVT was 8.0%, 7.0%, 4.0%, and 1.6% on Pmax of 0, 5, 10, and 15 cm H2O, respectively. During PC-CMVa, the ΔVT was 6.8%, 4.3%, 2.0%, and 0% on Pmax of 0, 5, 10, and 15 cm H2O, respectively. When Pmax increased, ΔVT during both VC-CMVs and PC-CMVa decreased (Fig. 1).
The effect of muscle pressure on change in tidal volume. ΔVT = change in tidal volume; VC-CMVs = volume control continuous mandatory ventilation with set-point targeting scheme; PC-CMVa = pressure control continuous mandatory ventilation with adaptive targeting scheme; Pmax = maximal value of the muscle pressure.
Figure 4 shows the effects of Pmax on VT error. During VC-CMVs with CC off, the VT error was −3.4%, −1.0%, 0.6%, and 2.4% for Pmax of 0, 5, 10, and 15 cm H2O, respectively. During VC-CMVs with CC on, the VT error was 4.3%, 5.9%, 4.8%, and 4.0% for Pmax of 0, 5, 10, and 15 cm H2O, respectively. During PC-CMVa with CC off, the VT error was −1.8%, 0.7%, 8.3%, and 39.0% for Pmax of 0, 5, 10, and 15 cm H2O, respectively. During PC-CMVa with CC on, the VT error was 4.9%, 5.0%, 10.7%, and 39.0% for Pmax of 0, 5, 10, and 15 cm H2O, respectively.
The effect of maximal value for muscle pressure on tidal volume error. VT = tidal volume; VC-CMVs = volume control continuous mandatory ventilation with set-point targeting scheme; CC = circuit compensation; PC-CMVa = pressure control continuous mandatory ventilation with adaptive targeting scheme; Pmax = maximal value of the muscle pressure.
The VT error during VC-CMVs trended in different directions for CC off versus CC on, but the absolute value remained within 5%. In contrast, VT error on PC-CMVa increased with increased Pmax and peaked at 39.0% for Pmax = 15 cm H2O. For both modes, the difference in VT error between CC on and CC off (ie, VT error with CC on minus VT error with CC off) decreased as Pmax increased.
Discussion
Change in Tidal Volume (ΔVT)
This is the first study to investigate the effect of CC on different modes with the presence of Pmax in a simulator setting. Both VC-CMVs and PC-CMVa modes are volume targeted (ie, allow VT set). The compliance of the breathing circuit (Ccircuit) was defined by the change in volume (V) divided by the change in pressure (P), as shown in Equation 3:
Therefore, as airway pressure increases during inspiration the compressible volume increases and the greater the difference between target and delivered VT. Rearranging Equation 3, we get the equation for the volume lost due to compression during inspiration, as shown in Equation 4:
The equation of motion for the respiratory system is shown in Equation 5, where Pvent is the pressure provided by the ventilator, Pmus is the pressure provided by the patient's respiratory muscle, C is the compliance of the respiratory system, V is the volume, R is the resistance of airway, and V̇ is the flow over time.9
In our experiments, C, R, and V were held constant. Pmax simulates the peak Pmus. Therefore, as Pmax increases, Pvent decreases to maintain the right-hand side of Equation 5 constant. As a result of the decrease in pressure, the lost volume decreases according to Equation 4. This explains the pressure waveform deformations in Figure 2 and the decrease in ΔVT shown in Figure 3. This is an example of work shifting as described elsewhere.10,11
VT Error
The operator's manual for the Servo-u ventilator specifies inspiratory VT error of ± (4 mL + 7% of actual volume) for VT range of 100–4,000 mL.12 The International Standards Organization has specified an error standard of ± (5 + 10% of the set volume).13
For VC-CMVs, VT error was within this range with CC on or off for all simulated Pmax. Hence, there are no clinical implications of our findings for this mode.
For PC-CMVa, VT error was within this range with CC on or off only for low simulated Pmax (ie, Pmax ≤ 5 cm H2O). The adaptive targeting scheme of this mode attem-pts to automatically decrease Pvent as Pmax increases to maintain the average VT at the set target value.14 However, the ventilator can decrease Pvent only to the PEEP level. Beyond that, increases in Pmax result in increased VT. The clinical implication is that patients with high Pmax (eg, patients with COVID-19) VT overdosage is possible and perhaps common.
The major limitation of this study was that it was based a simulation of a single kind of subject, one with moderate ARDS. The function we used to Pmax was evidence based, but real Pmax can take a wide variety of forms that varies randomly. Finally, we did not study situations of patient-ventilator asynchrony.
Conclusions
CC corrected the delivered VT for volume lost due to compression in the patient circuit as expected. This compensation volume decreases as airway pressure drops due to patient Pmax. The difference between set and delivered VT was minimal for VC-CMVs but increased to excessive amounts during PC-CMVa in the presence of large Pmax.
Footnotes
- Correspondence: Robert L Chatburn MHHS RRT RRT-NPS FAARC, Cleveland Clinic, Respiratory Therapy T03-35, 9500 Euclid Avenue, Cleveland, OH 44121. E-mail: chatbur{at}ccf.org
The study was performed at Cleveland Clinic, Cleveland, Ohio.
Mr Chatburn discloses relationships with IngMar Medical, Vyaire Medical, and ProMedic Consulting. Ms Liu has disclosed no conflicts of interest.
- Copyright © 2022 by Daedalus Enterprises
