- Email: [email protected]
Flow controlled expiration does not impair pedal power during physical exercise on a bicycle ergometer
Respiratory Physiology & Neurobiology 271 (2020) 103303
Contents lists available at ScienceDirect
Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol
Flow controlled expiration does not impair pedal power during physical exercise on a bicycle ergometer
T
⁎
Stefan Schumanna,b, , Nina Bergera,b, Sara Lozano-Zahoneroa,b, Steffen Wirtha,b a b
Department of Anesthesiology and Critical Care, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany Laboratory of origin: Workgroup Clinical Respiration Physiology, Hugstetter Straße 55, D-79106 Freiburg, Germany
A R T I C LE I N FO
A B S T R A C T
Keywords: Breathing discomfort Breathing support Expiration Lung stabilization Perception Ventilation therapy
Flow Controlled Expiration (FLEX) has been demonstrated to be lungprotective in models of ARDS during controlled mechanical ventilation. However, modern ventilation strategies in critical care include spontaneous breathing. Therefore, we investigated breathing discomfort and potential performance constraints of FLEX in 24 healthy test persons under increased ventilation demand. The subjects generated 20, 50 or 100 W pedal power on a bicycle ergometer while breathing with and without FLEX and rated breathing discomfort on a scale ranging from 0 (comfortable) to 10 (not tolerable). Then the subjects were asked to indicate the power they could maintain for 30 min with and without FLEX. With FLEX, tidal volume was higher and respiratory rate lower than without. Breathing discomfort was slightly increased by FLEX (on average from 2.2 to 3.2, p = 0.002). The estimated maintainable power was similar with and without FLEX (p = 0.986). We conclude that FLEX does not intolerably increase breathing discomfort and does not impair physical performance.
1. Introduction
generally be tolerable by the conscious patient underlying increased ventilation demand. Therefore, we investigated if spontaneously breathing persons with increased ventilation demand could tolerate the decelerated expiration of FLEX. We hypothesized that FLEX would increase the ventilation discomfort and decrease physical performance, both within a tolerable range. To test this hypotheses, we requested lung healthy test persons to rate the perceived breathing discomfort while generating pedal power of 20–100 W on a bicycle ergometer and requested which pedal power they could well maintain while breathing without and with FLEX, respectively.
Mechanical ventilation carries a significant risk of ventilator-induced lung injury (Slutsky, 1999; Slutsky and Ranieri, 2013). Various modes of ventilation have been developed with the aim of giving the patient the best possible ventilation support while protecting the lung. Earlier, our group presented the Flow Controlled Expiration [FLEX, (Schumann et al., 2014)] as a new therapy option for lung protective mechanical ventilation (Goebel et al., 2014; Wirth et al., 2017). During passive expiration FLEX limits the initial expiratory peak flow rate in favour of a linearized expiratory flow throughout expiration time by artificially increasing the expiratory resistance in the beginning and continuously reducing this resistance during ongoing expiration. Thus a higher flow rate is maintained during late expiration, compared to conventional passive ventilation. The expiratory volume decrease in the lungs is thus decelerated without prolongation of the total expiration time. However, modern strategies in critical care therapy favour spontaneous breathing support over controlled ventilation (Guldner et al., 2014) and less sedation (Devlin et al., 2018). Consequently, new options for mechanical ventilation support must
2. Material and methods The study was approved by the ethics committee of the University Medical Center Freiburg (EK 412/13). 24 healthy adult volunteers were included in the study after obtaining informed written consent. Exclusion criteria were known pulmonary diseases, age < 18 years, and active implants like pace makers or cardioverters.
⁎
Corresponding author at: Department of Anesthesiology and Intensive Care Medicine, University Medical Center Freiburg, Hugstetter Straße 55, D-79106, Freiburg, Germany. E-mail addresses: [email protected] (S. Schumann), [email protected] (N. Berger), [email protected] (S. Lozano-Zahonero), steff[email protected] (S. Wirth). https://doi.org/10.1016/j.resp.2019.103303 Received 1 August 2019; Received in revised form 6 September 2019; Accepted 20 September 2019 Available online 20 September 2019 1569-9048/ © 2019 Elsevier B.V. All rights reserved.
Respiratory Physiology & Neurobiology 271 (2020) 103303
S. Schumann, et al.
the fashion of a two-alternative forced choice response (Persaud and McLeod, 2008). In the second experiment the test persons were asked to identify the maximal power level which they estimated to tolerate for 30 min. Therefore, blinded against the set pedal power, they pedalled with an initial pedal power of 20 W (being the lowest power, supported by the ergometer) and pedal power was continuously increased until the test persons indicated the current power as the requested maximum. Of this procedure six repetitions were performed of which three proceeded with and three without FLEX, again randomly allocated on the base of computer generated randomization lists. In the third experiment the test persons were asked to generate the power they had determined as maximal maintenance power, respectively once with and once without FLEX (randomly allocated) for 10 min. After that, capillary blood was drawn from the earlobe. Free comments of the test persons related to the experiments were protocolled throughout the measurements.
2.1. Set-up The test persons, unaware of the function of the FLEX, took place on a bicycle ergometer (ergometrics ER900 EL, Ergoline GmbH, Bitz, Germany) in a 45° head-up supine position and were asked to breathe spontaneously via a mouthpiece which was connected to a pneumotachograph (Type Fleisch 2, Dr. Fenyves und Gut, Hechingen, Germany) to measure the flow rate. A Y-piece which contained one-way valves for separating the inspiratory airway from the expiratory airway generated a directed air flow and prevented from rebreathing expired air. The FLEX-device was connected to the expiratory airway to control the expiratory flow rate. If activated, the FLEX-device occluded the expiratory airway nearly completely during inspiration and as soon as the beginning of expiration was detected, this occlusion was gradually reduced within 2.5 s until the airway was completely open. Breathing via the nose was disabled by means of a nose clip. Airway pressure, flow rate, the electrocardiogram (ECG), and CO2 partial pressure measured via a main stream sensor system (SC7000, Siemens Medical Systems, Danvers, MA) were recorded using a selfwritten software package based on LabView (v 7.1, National Instruments, Austin, TX). Tidal volumes were calculated by breathwise integration of the flow rate. Electrical impedance tomography (EIT) of the thorax was recorded using PulmoVista 500 (Dräger medical, Lübeck, Germany) with the electrode interface positioned at the height of the 5th intercostal space. EIT sampling rate was 50 images/s. From the EIT images the mean impedance was calculated as the mean of all pixels. From the resulting time series the beginnings of inspiration were identified as the local minima and the end of inspiration as the local maxima of the impedance curve. Baseline impedance was defined as the mean impedance at beginning of inspiration and tidal impedance change was defined as the difference of mean impedance between inspiration and the following expiration. The percentage of ventral ventilation was determined by dividing the tidal impedance change within only the ventral 50% of the images by the total tidal impedance change. Additionally, we determined the relative time courses of ventral and dorsal ventilation for a qualitative estimation of the effects of FLEX. Therefore, for each measurement the impedance change in the dorsal and ventral region were determined from the breath with the tidal volume closest to the mean tidal volume of this measurement. The maximal intratidal impedance change of this breath was aligned to 1. Since spontaneous breathing is characterized by varying inspiration and expiration times we resampled each inspiration and expiration phase of the respective breath by projecting these on 100 time points, respectively via cubic spline interpolation. This way we were able to generate average courses of impedance changes relative to the duration of the respective breathing phase. All data were retrospectively analysed using Matlab (v2011, The MathWorks, Natick, MA).
2.3. Statistical analysis Data are given as mean ± SD if not indicated otherwise. For mutual comparisons, multifactorial repeated measures ANOVA were calculated. Simple comparisons were performed using paired Student’s ttests. ‘Worse/better’ ratings for transitions from FLEX ‘on to off’ and ‘off to on’ were compared using four-field Chi2 test. A p < 0.05 was considered statistically significant. 3. Results Demographics and characteristics of test persons included in the study are given in Table 1. No test person aborted any breathing phase or identified his or her perception as ‘not tolerable’. With FLEX, tidal volume was higher and respiratory rate was lower than without FLEX (Table 2). Thereby, tidal volume increased with increasing pedal power but respiratory rate was not influenced by pedal power. The resulting minute volume increased with increasing pedal power but not significantly with FLEX. The minimal inspiratory pressure and maximal inspiratory flow increased with increasing pedal power but were independent from FLEX. The absolute values of maximal expiratory pressure and minimal expiratory flow increased with increasing pedal power and with FLEX. Heart rate and end-tidal CO2 partial pressure were independent of FLEX but increased with increasing pedal power. With FLEX, tidal impedance change was increased compared to without FLEX breathing but EIT measurements did not show significant baseline shifts or differences in dorsal vs. ventral ventilation shares between breathing with or without FLEX (Fig. 1, Table 3).The analysis of intratidal impedance change showed that FLEX did not affect the relative course of inspiration but delayed the intratidal impedance change during expiration. The latter was more pronounced in the ventral region (Fig. 2). Breathing discomfort with FLEX was on average rated 1 point worse compared to without FLEX (p = 0.002) but pedal power was not a
2.2. Experimental procedures Three experiments were performed with pauses of 10 min between each to allow for recovery. In the first experiment the test persons were asked to generate 20, 50 or 100 W pedal power for 10 min in ascending order. We omitted randomising the pedal power with the intention to avoid effects of fatigue when switching from a higher to a lower power level. At each power level the test persons performed six repetitions of which three proceeded with active and three with inactive FLEX. The respective orders of FLEX conditions were allocated on base of computer generated randomization lists. After each breathing period, the test persons rated their perception of breathing on a discomfort scale ranging from 0 (‘comfortable’) to 10 (‘not tolerable’). For statistical reasons, 10 was also intended if a test person had aborted the particular breathing phase. Furthermore, the test persons were asked if the breathing comfort of the actual phase was ‘better’ or ‘worse’ than the breathing comfort of the preceding phase in
Table 1 Demographic characteristics of volunteers included in the study(n = 24). Variable
Unit
Value
Age Bodyweight Bodyheight Body Mass Index Training-condition [0/1/2/3] Gender Smokers / Non-smokers
[years, (mean, range)] [kg, (mean, range)] [cm, (mean, range)] [kg/m², (mean, range)] [n] (male/female) [n]
32 (20 – 58) 73 (55 – 109) 174 (158 – 198) 24 (20 – 29) 6/9/7/2 14/10 4/20
Training condition: 0 = no training, 1 = moderate training, 2 = regular training, 3 = serious sport. 2
Respiratory Physiology & Neurobiology 271 (2020) 103303
S. Schumann, et al.
Table 2 Physiological variables during breathing without/with FLEX while pedaling.
Respiratory rate [1/min] Tidal volume [mL] Minute Volume L/Min Heart rate [1/ min] etCO2 [mmHg]
Maximal expiratory pressure [cmH2O] Minimal inspiratory pressure [1/ min] Maximal inspiratory flow [mL/s] Minimal expiratory flow [mL/s]
Pedal power [W]
FLEX off
20 50 100 20 50 100 20 50 100 20 50 100 20 50 100 20 50 100
14.8 ± 3.8 14.7 ± 4.2 15.4 ± 3.4 1140 ± 251 1216 ± 423 1471 ± 389 16.4 ± 3.4 16.7 ± 4.0 22.1 ± 5.6 87 ± 10 95 ± 10 112 ± 12 42 ± 2.7 43.5 ± 2.7 45.3 ± 2.8 3.1 ± 1.2 4.3 ± 1.9 6.3 ± 1.8
13.3 ± 3.6 13.0 ± 3.5 13.6 ± 2.5 1299 ± 378 1536 ± 381 1920 ± 419 16.5 ± 3.9 19.3 ± 4.1 25.1 ± 4.0 87 ± 10 94 ± 9 112 ± 12 41.8 ± 3.4 43.5 ± 3.2 45.3 ± 3.1 11 ± 4.8 13.8 ± 5.9 20.3 ± 9.9
20 50 100
−2.9 ± 0.7 −3.3 ± 0.8 −4.5 ± 0.9
−3.1 ± 0.8 −3.5 ± 0.9 −4.7 ± 0.8
pFLEX = 0.326 pPower < 0.001
20 50 100 20 50 100
796 ± 199 912 ± 205 1184 ± 258 −854 ± 203 −1021 ± 223 −1303 ± 270
842 ± 201 946 ± 217 1216 ± 176 −944 ± 256 −1098 ± 246 −1405 ± 311
pFLEX = 0.270 ppower < 0.001
Table 3 Measures of electrical impedance tomography during breathing without/with flow controlled expiration (FLEX).
FLEX on
pFLEX = 0.005 pPower = 0.677
Tidal impedance change [arb]
pFLEX < 0.001 pPower < 0.001
Percentage ventral ventilation [%]
pFLEX = 0.058 pPower < 0.001
Baseline-Impedance [arb]
pFLEX = 0.812 pPower < 0.001
Pedal power [W]
FLEX off
FLEX on
20 50 100 20 50 100 20 50 100
9.5 ± 2.2 11.7 ± 1.9 14.5 ± 2.3 37.5 ± 9.6 32.8 ± 8.0 34.7 ± 7.6 0.2 ± 0.1 0.1 ± 0.1 0.3 ± 0.2
10.6 ± 2.9 12.5 ± 2.1 15.3 ± 1.9 39.5 ± 9.7 32.8 ± 8.0 35.3 ± 7.7 0.2 ± 0.1 0.1 ± 0.1 0.3 ± 0.1
pFLEX = 0.012 pPower < 0.001 pFLEX = 0.551 pPower = 0.006 pFLEX = 0.731 pPower < 0.001
pFLEX, pPower = probability that variable is independent from FLEX or pedal power, respectively.
pFLEX = 0.895 pPower < 0.001
estimated to be well maintainable did not differ between breathing with and without FLEX (Table 5). Free comments disclosed a large variability in the perception of FLEX: most test persons perceived (correctly) an increased resistance during expiration. Three test persons reported to have not felt any difference between the trials. Two persons reported ventilation with FLEX as pleasant, of which one specified that he felt pleasant not to breathe into the void.
pFLEX < 0.001 pPower < 0.001
4. Discussion
pFLEX = 0.026 pPower < 0.001
The main finding of this study is that in agreement with our hypothesis healthy test persons with increased ventilation demand perceive a slightly reduced breathing comfort when breathing with FLEX during physical exercise, compared to no FLEX. However, in contrast to our hypothesis, the physical power estimated to be maintainable for 30 min was nearly identical for breathing with or without FLEX, and furthermore all variables of blood gas analyses were not modified. Decelerating the expiratory flow during mandatory ventilation was shown to diminish ventilator induced lung injury in a porcine model of the ARDS lung (Goebel et al., 2014), homogenize regional ventilation (Wirth et al., 2017), improve oxygenation and decarbonisation in a porcine model of the healthy ventilated lung (Schmidt et al., 2018) and in lung healthy patients (Schmidt et al., 2019). Recently, we investigated the mechanisms underlying these lung-protective effects of FLEX by means of electrical impedance tomography. We found that
pFLEX, pPower = probability that variable is independent from FLEX or pedal power, respectively.
significant factor (p = 0.190) for breathing discomfort (Fig. 3). In the forced choice comparison, feigned transitions were on average rated as ‘better’ with 62% (Table 4). Transitions from breathing without FLEX to with FLEX were more often rated as ‘worse’ and changes from breathing with FLEX to without FLEX were rated more often as ‘better’ than otherwise. The maximal pedal power estimated to be well maintainable for 30 min was similar with (86 ± 39 W) and without (87 ± 42 W, p = 0.986) FLEX. Capillary blood gases after pedalling with the power that was
Fig. 1. Regional tidal ventilation in an exemplary test person breathing without (top) and with (bottom) flow controlled expiration (FLEX) while generating 20 (left), 50 (middle) or 100 (right) W pedal power on a bicycle ergometer. Regional tidal ventilation was determined as the difference between the endinspiratory and the following end-expiratory impedance. Warmer colours represent higher impedance changes and thus higher regional ventilation. According to the higher tidal volume, regional ventilation was increased during breathing with FLEX compared to breathing without FLEX.
3
Respiratory Physiology & Neurobiology 271 (2020) 103303
S. Schumann, et al.
Fig. 2. Temporal courses of ventral (red) and dorsal (black) impedances during breathing without (straight lines) and with (dotted lines) flow controlled expiration (FLEX) while generating 20 (left), 50 (middle) or 100 (right) W pedal power on a bicycle ergometer. For each measurement the breath with tidal volume closest to mean tidal volume was used. Total impedance change during each breath was aligned to 1 and the mean curve was built from the three repetitions. Then the curves were averaged from all test persons. The vertical lines separate inspiration from expiration. It is important to note that inspiration and expiration times represent the relative duration of the respective breathing phase and do not represent absolute values for the total duration of the breath.
Fig. 3. Histograms of breathing discomfort ratings during breathing without (top) and with (bottom) flow controlled expiration (FLEX) while generating 20 (left), 50 (middle) or 100 (right) W pedal power on a bicycle ergometer. The discomfort scale ranged from 0 (‘comfortable’) to 10 (‘not tolerable’). The histograms were built from the mean of three repetitions of discomfort ratings for each condition. Vertical lines indicate the median of the respectively perceived discomfort. PercHi = percentage of high values, defined as percentage of discomfort ratings ≥6. Table 4 Perception of Transitions between breathing without/with flow controlled expiration (FLEX). Values show the numbers of transitions from one FLEX condition to the next that were perceived as ‘worse’ or ‘better’.
Table 5 Venous blood gases after 10 min pedalling at self-estimated well maintainable pedal power without/with flow controlled expiration (FLEX). FLEX off
FLEX on
p-value
109.3 ± 23.4 34.7 ± 3.1 7.43 ± 0.02 98.3 ± 0.6 24.3 ± 1.2 −0.2 ± 1.5 1.6 ± 0.7
106.1 ± 19 34.3 ± 2.6 7.44 ± 0.01 98.3 ± 0.6 24.4 ± 1 0 ± 1.1 1.6 ± 0.6
0.229 0.708 0.604 0.435 0.352 0.339 0.956
FLEX Transition from/to
Worse [n] Better [n] Percentage ‚better ‘[%] a
off/off
on/on
off/on
on/off
16 35 68.6
22 28 56
93 36 27.9a
20 105 84
pO2 [mmHg] pCO2 [mmHg] pH sO2 (%) HCO3- (mmol/l) Base excess (mmol/l) Lactate (mmol/l)
= p < 0.05 compared to ‘on/off’.
pO2 = O2 partial pressure, pCO2 = CO2 partial pressure, sO2 = O2 saturation, HCO3- = bicarbonate.
FLEX caused lung recruitment and an alveolar stabilizing effect comparable to PEEP (Borgmann et al., 2018). In other words, the effects of FLEX are closely related to those that one expects from setting PEEP, however without the necessity of higher airway pressures. This might be potentially useful for critical care therapy. Transferring the lung protective potential of FLEX to the ICU would however require applicability during spontaneous breathing of spontaneous breathing support since modern strategies in critical care therapy more and more abstain from controlled ventilation (Guldner et al., 2014) and deep sedation (Devlin et al., 2018). In an earlier study in test persons breathing during rest, a subliminal perception of FLEX-related
breathing discomfort could be identified when the test persons were asked to rate transitions from one to another breathing mode in a forced choice fashion (Wirth et al., 2014). By contrast, in the current study in test persons with increased ventilation demand, breathing discomfort was significantly increased, however as indicated by the shift from 2.2 to 3.2 on a scale ranging up to 10 we feel justified to interpret this increase in breathing discomfort as well tolerable. Surprisingly, some test persons reported not to be able to identify a difference between the breathing phases and some perceived the decelerated expiration even as pleasant. 4
Respiratory Physiology & Neurobiology 271 (2020) 103303
S. Schumann, et al.
resistance. Therefore, further studies focusing on ventilation comfort and long term effects of FLEX in patients with injured lungs are required.
In a comparable fashion to this preceding study, we observed that with and without FLEX similar peak flow rates were achieved by the test subjects despite the fact that FLEX is supposed to decelerate expiration. The test persons obviously adapted to the conditions under ventilation with FLEX, by increasing the expiratory peak pressure to compensate for the increases expiratory resistance during breathing with FLEX. This might have potentially increased the expiratory effort of breathing. However, this had neither severely increased the breathing discomfort nor shown a physiological sign in the blood gas analyses between breathing modes after pedalling. EIT images revealed a higher tidal impedance change, reflecting the higher tidal volume and a decelerated proceeding of the expiration during breathing with FLEX, compared to without FLEX. In contrast to studies in mechanically ventilated patients and in the porcine models of ARDS we found neither baseline shifts nor ventral-dorsal shifts of ventilation indicating global or regional recruitment. We attribute this to the fact that the lungs of our healthy and consciously spontaneously breathing test persons were well recruited. The test persons exercised in a 45° upright position and did not suffer from atelectasis, e.g. caused by anaesthesia induction and supine position and high inspiratory fraction of oxygen (Hedenstierna and Rothen, 2000) or reasoned by the lung injury. Thus, the lungs had only low potential for recruitment. A beneficial effect of FLEX might therefore only be expected for patients in the respective circumstances.
4.2. Conclusion FLEX does not impair the physical performance in lung healthy volunteers under increased ventilation demand and was throughout perceived as tolerable. We conclude that FLEX might be applicable in spontaneously breathing patients with increased ventilation demand. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References Borgmann, S., Schmidt, J., Goebel, U., Haberstroh, J., Guttmann, J., Schumann, S., 2018. Dorsal recruitment with flow-controlled expiration (FLEX): an experimental study in mechanically ventilated lung-healthy and lung-injured pigs. Crit Care 22, 245. Devlin, J.W., Skrobik, Y., Gelinas, C., Needham, D.M., Slooter, A.J.C., Pandharipande, P.P., Watson, P.L., Weinhouse, G.L., Nunnally, M.E., Rochwerg, B., Balas, M.C., van den Boogaard, M., Bosma, K.J., Brummel, N.E., Chanques, G., Denehy, L., Drouot, X., Fraser, G.L., Harris, J.E., Joffe, A.M., Kho, M.E., Kress, J.P., Lanphere, J.A., McKinley, S., Neufeld, K.J., Pisani, M.A., Payen, J.F., Pun, B.T., Puntillo, K.A., Riker, R.R., Robinson, B.R.H., Shehabi, Y., Szumita, P.M., Winkelman, C., Centofanti, J.E., Price, C., Nikayin, S., Misak, C.J., Flood, P.D., Kiedrowski, K., Alhazzani, W., 2018. Clinical practice guidelines for the prevention and management of pain, Agitation/Sedation, delirium, immobility, and sleep disruption in adult patients in the ICU. Crit. Care Med. 46, e825–e873. Goebel, U., Haberstroh, J., Foerster, K., Dassow, C., Priebe, H.J., Guttmann, J., Schumann, S., 2014. Flow-controlled expiration: a novel ventilation mode to attenuate experimental porcine lung injury. Br. J. Anaesth. 113, 474–483. Guldner, A., Pelosi, P., Gama de Abreu, M., 2014. Spontaneous breathing in mild and moderate versus severe acute respiratory distress syndrome. Curr. Opin. Crit. Care 20, 69–76. Hedenstierna, G., Rothen, H.U., 2000. Atelectasis formation during anesthesia: causes and measures to prevent it. J. Clin. Monit. Comput. 16, 329–335. Persaud, N., McLeod, P., 2008. Wagering demonstrates subconscious processing in a binary exclusion task. Conscious. Cogn. 17, 565–575. Schmidt, J., Gunther, F., Weber, J., Wirth, S., Brandes, I., Barnes, T., Zarbock, A., Schumann, S., Enk, D., 2019. Flow-controlled ventilation during ear, nose and throat surgery: a prospective observational study. Eur. J. Anaesthesiol. 36, 327–334. Schmidt, J., Wenzel, C., Mahn, M., Spassov, S., Cristina Schmitz, H., Borgmann, S., Lin, Z., Haberstroh, J., Meckel, S., Eiden, S., Wirth, S., Buerkle, H., Schumann, S., 2018. Improved lung recruitment and oxygenation during mandatory ventilation with a new expiratory ventilation assistance device: a controlled interventional trial in healthy pigs. Eur. J. Anaesthesiol. 35, 736–744. Schumann, S., Goebel, U., Haberstroh, J., Vimlati, L., Schneider, M., Lichtwarck-Aschoff, M., Guttmann, J., 2014. Determination of respiratory system mechanics during inspiration and expiration by FLow-controlled EXpiration (FLEX): a pilot study in anesthetized pigs. Minerva Anestesiol. 80, 19–28. Slutsky, A.S., 1999. Lung injury caused by mechanical ventilation. Chest 116, 9S–15S. Slutsky, A.S., Ranieri, V.M., 2013. Ventilator-induced lung injury. N. Engl. J. Med. 369, 2126–2136. Wirth, S., Best, C., Spaeth, J., Guttmann, J., Schumann, S., 2014. Flow Controlled Expiration is perceived as less uncomfortable than positive end expiratory pressure. Respir. Physiol. Neurobiol. 202, 59–63. Wirth, S., Springer, S., Spaeth, J., Borgmann, S., Goebel, U., Schumann, S., 2017. Application of the novel ventilation mode FLow-Controlled EXpiration (FLEX): a crossover proof-of-Principle study in Lung-Healthy patients. Anesth. Analg. 125, 1246–1252.
4.1. Critique of methods We could not apply invasive haemodynamic measurements. However, as the haemodynamics were not negatively influenced in any preceding investigation on FLEX (Goebel et al., 2014) and as the heart rates were independent from FLEX, we assume that haemodynamics were not significantly influenced by FLEX in our study. The pedal powers were generated in ascending and not in randomized order. Therefore, the measurements could be biased by the duration of the experiments. However, the test subjects were allowed to pause between each trial and the main focus of our study was on the effect of FLEX which was randomly active or inactive. Therefore, we would not expect this to have influenced our results to a relevant extent. When the test persons were asked to identify the pedal power they could maintain for 30 min, the trial was stopped after identification of the level and the test persons had never to generate this power for 30 min. We cannot exclude that the subjects under- or overestimated the power they actually could have generated. As our primary intention was to investigate the perception of FLEX, we omitted to produce potential fatigue in favour of the possibility of repeated measurements. However, we have to concede that our results are restricted to short term ventilation periods in non-invasively supported lung-healthy persons. By contrast, patients who require mechanical ventilation support suffer from lung injury and are often ventilated for hours or even days. These patients might be more sensitive to ventilation discomfort. The active compensation of the expiratory peak flow may have played a crucial role in this context. A lung injured patient may not be able to generate an increased pressure to compensate the FLEX related
5
Contents lists available at ScienceDirect
Respiratory Physiology & Neurobiology journal homepage: www.elsevier.com/locate/resphysiol
Flow controlled expiration does not impair pedal power during physical exercise on a bicycle ergometer
T
⁎
Stefan Schumanna,b, , Nina Bergera,b, Sara Lozano-Zahoneroa,b, Steffen Wirtha,b a b
Department of Anesthesiology and Critical Care, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany Laboratory of origin: Workgroup Clinical Respiration Physiology, Hugstetter Straße 55, D-79106 Freiburg, Germany
A R T I C LE I N FO
A B S T R A C T
Keywords: Breathing discomfort Breathing support Expiration Lung stabilization Perception Ventilation therapy
Flow Controlled Expiration (FLEX) has been demonstrated to be lungprotective in models of ARDS during controlled mechanical ventilation. However, modern ventilation strategies in critical care include spontaneous breathing. Therefore, we investigated breathing discomfort and potential performance constraints of FLEX in 24 healthy test persons under increased ventilation demand. The subjects generated 20, 50 or 100 W pedal power on a bicycle ergometer while breathing with and without FLEX and rated breathing discomfort on a scale ranging from 0 (comfortable) to 10 (not tolerable). Then the subjects were asked to indicate the power they could maintain for 30 min with and without FLEX. With FLEX, tidal volume was higher and respiratory rate lower than without. Breathing discomfort was slightly increased by FLEX (on average from 2.2 to 3.2, p = 0.002). The estimated maintainable power was similar with and without FLEX (p = 0.986). We conclude that FLEX does not intolerably increase breathing discomfort and does not impair physical performance.
1. Introduction
generally be tolerable by the conscious patient underlying increased ventilation demand. Therefore, we investigated if spontaneously breathing persons with increased ventilation demand could tolerate the decelerated expiration of FLEX. We hypothesized that FLEX would increase the ventilation discomfort and decrease physical performance, both within a tolerable range. To test this hypotheses, we requested lung healthy test persons to rate the perceived breathing discomfort while generating pedal power of 20–100 W on a bicycle ergometer and requested which pedal power they could well maintain while breathing without and with FLEX, respectively.
Mechanical ventilation carries a significant risk of ventilator-induced lung injury (Slutsky, 1999; Slutsky and Ranieri, 2013). Various modes of ventilation have been developed with the aim of giving the patient the best possible ventilation support while protecting the lung. Earlier, our group presented the Flow Controlled Expiration [FLEX, (Schumann et al., 2014)] as a new therapy option for lung protective mechanical ventilation (Goebel et al., 2014; Wirth et al., 2017). During passive expiration FLEX limits the initial expiratory peak flow rate in favour of a linearized expiratory flow throughout expiration time by artificially increasing the expiratory resistance in the beginning and continuously reducing this resistance during ongoing expiration. Thus a higher flow rate is maintained during late expiration, compared to conventional passive ventilation. The expiratory volume decrease in the lungs is thus decelerated without prolongation of the total expiration time. However, modern strategies in critical care therapy favour spontaneous breathing support over controlled ventilation (Guldner et al., 2014) and less sedation (Devlin et al., 2018). Consequently, new options for mechanical ventilation support must
2. Material and methods The study was approved by the ethics committee of the University Medical Center Freiburg (EK 412/13). 24 healthy adult volunteers were included in the study after obtaining informed written consent. Exclusion criteria were known pulmonary diseases, age < 18 years, and active implants like pace makers or cardioverters.
⁎
Corresponding author at: Department of Anesthesiology and Intensive Care Medicine, University Medical Center Freiburg, Hugstetter Straße 55, D-79106, Freiburg, Germany. E-mail addresses: [email protected] (S. Schumann), [email protected] (N. Berger), [email protected] (S. Lozano-Zahonero), steff[email protected] (S. Wirth). https://doi.org/10.1016/j.resp.2019.103303 Received 1 August 2019; Received in revised form 6 September 2019; Accepted 20 September 2019 Available online 20 September 2019 1569-9048/ © 2019 Elsevier B.V. All rights reserved.
Respiratory Physiology & Neurobiology 271 (2020) 103303
S. Schumann, et al.
the fashion of a two-alternative forced choice response (Persaud and McLeod, 2008). In the second experiment the test persons were asked to identify the maximal power level which they estimated to tolerate for 30 min. Therefore, blinded against the set pedal power, they pedalled with an initial pedal power of 20 W (being the lowest power, supported by the ergometer) and pedal power was continuously increased until the test persons indicated the current power as the requested maximum. Of this procedure six repetitions were performed of which three proceeded with and three without FLEX, again randomly allocated on the base of computer generated randomization lists. In the third experiment the test persons were asked to generate the power they had determined as maximal maintenance power, respectively once with and once without FLEX (randomly allocated) for 10 min. After that, capillary blood was drawn from the earlobe. Free comments of the test persons related to the experiments were protocolled throughout the measurements.
2.1. Set-up The test persons, unaware of the function of the FLEX, took place on a bicycle ergometer (ergometrics ER900 EL, Ergoline GmbH, Bitz, Germany) in a 45° head-up supine position and were asked to breathe spontaneously via a mouthpiece which was connected to a pneumotachograph (Type Fleisch 2, Dr. Fenyves und Gut, Hechingen, Germany) to measure the flow rate. A Y-piece which contained one-way valves for separating the inspiratory airway from the expiratory airway generated a directed air flow and prevented from rebreathing expired air. The FLEX-device was connected to the expiratory airway to control the expiratory flow rate. If activated, the FLEX-device occluded the expiratory airway nearly completely during inspiration and as soon as the beginning of expiration was detected, this occlusion was gradually reduced within 2.5 s until the airway was completely open. Breathing via the nose was disabled by means of a nose clip. Airway pressure, flow rate, the electrocardiogram (ECG), and CO2 partial pressure measured via a main stream sensor system (SC7000, Siemens Medical Systems, Danvers, MA) were recorded using a selfwritten software package based on LabView (v 7.1, National Instruments, Austin, TX). Tidal volumes were calculated by breathwise integration of the flow rate. Electrical impedance tomography (EIT) of the thorax was recorded using PulmoVista 500 (Dräger medical, Lübeck, Germany) with the electrode interface positioned at the height of the 5th intercostal space. EIT sampling rate was 50 images/s. From the EIT images the mean impedance was calculated as the mean of all pixels. From the resulting time series the beginnings of inspiration were identified as the local minima and the end of inspiration as the local maxima of the impedance curve. Baseline impedance was defined as the mean impedance at beginning of inspiration and tidal impedance change was defined as the difference of mean impedance between inspiration and the following expiration. The percentage of ventral ventilation was determined by dividing the tidal impedance change within only the ventral 50% of the images by the total tidal impedance change. Additionally, we determined the relative time courses of ventral and dorsal ventilation for a qualitative estimation of the effects of FLEX. Therefore, for each measurement the impedance change in the dorsal and ventral region were determined from the breath with the tidal volume closest to the mean tidal volume of this measurement. The maximal intratidal impedance change of this breath was aligned to 1. Since spontaneous breathing is characterized by varying inspiration and expiration times we resampled each inspiration and expiration phase of the respective breath by projecting these on 100 time points, respectively via cubic spline interpolation. This way we were able to generate average courses of impedance changes relative to the duration of the respective breathing phase. All data were retrospectively analysed using Matlab (v2011, The MathWorks, Natick, MA).
2.3. Statistical analysis Data are given as mean ± SD if not indicated otherwise. For mutual comparisons, multifactorial repeated measures ANOVA were calculated. Simple comparisons were performed using paired Student’s ttests. ‘Worse/better’ ratings for transitions from FLEX ‘on to off’ and ‘off to on’ were compared using four-field Chi2 test. A p < 0.05 was considered statistically significant. 3. Results Demographics and characteristics of test persons included in the study are given in Table 1. No test person aborted any breathing phase or identified his or her perception as ‘not tolerable’. With FLEX, tidal volume was higher and respiratory rate was lower than without FLEX (Table 2). Thereby, tidal volume increased with increasing pedal power but respiratory rate was not influenced by pedal power. The resulting minute volume increased with increasing pedal power but not significantly with FLEX. The minimal inspiratory pressure and maximal inspiratory flow increased with increasing pedal power but were independent from FLEX. The absolute values of maximal expiratory pressure and minimal expiratory flow increased with increasing pedal power and with FLEX. Heart rate and end-tidal CO2 partial pressure were independent of FLEX but increased with increasing pedal power. With FLEX, tidal impedance change was increased compared to without FLEX breathing but EIT measurements did not show significant baseline shifts or differences in dorsal vs. ventral ventilation shares between breathing with or without FLEX (Fig. 1, Table 3).The analysis of intratidal impedance change showed that FLEX did not affect the relative course of inspiration but delayed the intratidal impedance change during expiration. The latter was more pronounced in the ventral region (Fig. 2). Breathing discomfort with FLEX was on average rated 1 point worse compared to without FLEX (p = 0.002) but pedal power was not a
2.2. Experimental procedures Three experiments were performed with pauses of 10 min between each to allow for recovery. In the first experiment the test persons were asked to generate 20, 50 or 100 W pedal power for 10 min in ascending order. We omitted randomising the pedal power with the intention to avoid effects of fatigue when switching from a higher to a lower power level. At each power level the test persons performed six repetitions of which three proceeded with active and three with inactive FLEX. The respective orders of FLEX conditions were allocated on base of computer generated randomization lists. After each breathing period, the test persons rated their perception of breathing on a discomfort scale ranging from 0 (‘comfortable’) to 10 (‘not tolerable’). For statistical reasons, 10 was also intended if a test person had aborted the particular breathing phase. Furthermore, the test persons were asked if the breathing comfort of the actual phase was ‘better’ or ‘worse’ than the breathing comfort of the preceding phase in
Table 1 Demographic characteristics of volunteers included in the study(n = 24). Variable
Unit
Value
Age Bodyweight Bodyheight Body Mass Index Training-condition [0/1/2/3] Gender Smokers / Non-smokers
[years, (mean, range)] [kg, (mean, range)] [cm, (mean, range)] [kg/m², (mean, range)] [n] (male/female) [n]
32 (20 – 58) 73 (55 – 109) 174 (158 – 198) 24 (20 – 29) 6/9/7/2 14/10 4/20
Training condition: 0 = no training, 1 = moderate training, 2 = regular training, 3 = serious sport. 2
Respiratory Physiology & Neurobiology 271 (2020) 103303
S. Schumann, et al.
Table 2 Physiological variables during breathing without/with FLEX while pedaling.
Respiratory rate [1/min] Tidal volume [mL] Minute Volume L/Min Heart rate [1/ min] etCO2 [mmHg]
Maximal expiratory pressure [cmH2O] Minimal inspiratory pressure [1/ min] Maximal inspiratory flow [mL/s] Minimal expiratory flow [mL/s]
Pedal power [W]
FLEX off
20 50 100 20 50 100 20 50 100 20 50 100 20 50 100 20 50 100
14.8 ± 3.8 14.7 ± 4.2 15.4 ± 3.4 1140 ± 251 1216 ± 423 1471 ± 389 16.4 ± 3.4 16.7 ± 4.0 22.1 ± 5.6 87 ± 10 95 ± 10 112 ± 12 42 ± 2.7 43.5 ± 2.7 45.3 ± 2.8 3.1 ± 1.2 4.3 ± 1.9 6.3 ± 1.8
13.3 ± 3.6 13.0 ± 3.5 13.6 ± 2.5 1299 ± 378 1536 ± 381 1920 ± 419 16.5 ± 3.9 19.3 ± 4.1 25.1 ± 4.0 87 ± 10 94 ± 9 112 ± 12 41.8 ± 3.4 43.5 ± 3.2 45.3 ± 3.1 11 ± 4.8 13.8 ± 5.9 20.3 ± 9.9
20 50 100
−2.9 ± 0.7 −3.3 ± 0.8 −4.5 ± 0.9
−3.1 ± 0.8 −3.5 ± 0.9 −4.7 ± 0.8
pFLEX = 0.326 pPower < 0.001
20 50 100 20 50 100
796 ± 199 912 ± 205 1184 ± 258 −854 ± 203 −1021 ± 223 −1303 ± 270
842 ± 201 946 ± 217 1216 ± 176 −944 ± 256 −1098 ± 246 −1405 ± 311
pFLEX = 0.270 ppower < 0.001
Table 3 Measures of electrical impedance tomography during breathing without/with flow controlled expiration (FLEX).
FLEX on
pFLEX = 0.005 pPower = 0.677
Tidal impedance change [arb]
pFLEX < 0.001 pPower < 0.001
Percentage ventral ventilation [%]
pFLEX = 0.058 pPower < 0.001
Baseline-Impedance [arb]
pFLEX = 0.812 pPower < 0.001
Pedal power [W]
FLEX off
FLEX on
20 50 100 20 50 100 20 50 100
9.5 ± 2.2 11.7 ± 1.9 14.5 ± 2.3 37.5 ± 9.6 32.8 ± 8.0 34.7 ± 7.6 0.2 ± 0.1 0.1 ± 0.1 0.3 ± 0.2
10.6 ± 2.9 12.5 ± 2.1 15.3 ± 1.9 39.5 ± 9.7 32.8 ± 8.0 35.3 ± 7.7 0.2 ± 0.1 0.1 ± 0.1 0.3 ± 0.1
pFLEX = 0.012 pPower < 0.001 pFLEX = 0.551 pPower = 0.006 pFLEX = 0.731 pPower < 0.001
pFLEX, pPower = probability that variable is independent from FLEX or pedal power, respectively.
pFLEX = 0.895 pPower < 0.001
estimated to be well maintainable did not differ between breathing with and without FLEX (Table 5). Free comments disclosed a large variability in the perception of FLEX: most test persons perceived (correctly) an increased resistance during expiration. Three test persons reported to have not felt any difference between the trials. Two persons reported ventilation with FLEX as pleasant, of which one specified that he felt pleasant not to breathe into the void.
pFLEX < 0.001 pPower < 0.001
4. Discussion
pFLEX = 0.026 pPower < 0.001
The main finding of this study is that in agreement with our hypothesis healthy test persons with increased ventilation demand perceive a slightly reduced breathing comfort when breathing with FLEX during physical exercise, compared to no FLEX. However, in contrast to our hypothesis, the physical power estimated to be maintainable for 30 min was nearly identical for breathing with or without FLEX, and furthermore all variables of blood gas analyses were not modified. Decelerating the expiratory flow during mandatory ventilation was shown to diminish ventilator induced lung injury in a porcine model of the ARDS lung (Goebel et al., 2014), homogenize regional ventilation (Wirth et al., 2017), improve oxygenation and decarbonisation in a porcine model of the healthy ventilated lung (Schmidt et al., 2018) and in lung healthy patients (Schmidt et al., 2019). Recently, we investigated the mechanisms underlying these lung-protective effects of FLEX by means of electrical impedance tomography. We found that
pFLEX, pPower = probability that variable is independent from FLEX or pedal power, respectively.
significant factor (p = 0.190) for breathing discomfort (Fig. 3). In the forced choice comparison, feigned transitions were on average rated as ‘better’ with 62% (Table 4). Transitions from breathing without FLEX to with FLEX were more often rated as ‘worse’ and changes from breathing with FLEX to without FLEX were rated more often as ‘better’ than otherwise. The maximal pedal power estimated to be well maintainable for 30 min was similar with (86 ± 39 W) and without (87 ± 42 W, p = 0.986) FLEX. Capillary blood gases after pedalling with the power that was
Fig. 1. Regional tidal ventilation in an exemplary test person breathing without (top) and with (bottom) flow controlled expiration (FLEX) while generating 20 (left), 50 (middle) or 100 (right) W pedal power on a bicycle ergometer. Regional tidal ventilation was determined as the difference between the endinspiratory and the following end-expiratory impedance. Warmer colours represent higher impedance changes and thus higher regional ventilation. According to the higher tidal volume, regional ventilation was increased during breathing with FLEX compared to breathing without FLEX.
3
Respiratory Physiology & Neurobiology 271 (2020) 103303
S. Schumann, et al.
Fig. 2. Temporal courses of ventral (red) and dorsal (black) impedances during breathing without (straight lines) and with (dotted lines) flow controlled expiration (FLEX) while generating 20 (left), 50 (middle) or 100 (right) W pedal power on a bicycle ergometer. For each measurement the breath with tidal volume closest to mean tidal volume was used. Total impedance change during each breath was aligned to 1 and the mean curve was built from the three repetitions. Then the curves were averaged from all test persons. The vertical lines separate inspiration from expiration. It is important to note that inspiration and expiration times represent the relative duration of the respective breathing phase and do not represent absolute values for the total duration of the breath.
Fig. 3. Histograms of breathing discomfort ratings during breathing without (top) and with (bottom) flow controlled expiration (FLEX) while generating 20 (left), 50 (middle) or 100 (right) W pedal power on a bicycle ergometer. The discomfort scale ranged from 0 (‘comfortable’) to 10 (‘not tolerable’). The histograms were built from the mean of three repetitions of discomfort ratings for each condition. Vertical lines indicate the median of the respectively perceived discomfort. PercHi = percentage of high values, defined as percentage of discomfort ratings ≥6. Table 4 Perception of Transitions between breathing without/with flow controlled expiration (FLEX). Values show the numbers of transitions from one FLEX condition to the next that were perceived as ‘worse’ or ‘better’.
Table 5 Venous blood gases after 10 min pedalling at self-estimated well maintainable pedal power without/with flow controlled expiration (FLEX). FLEX off
FLEX on
p-value
109.3 ± 23.4 34.7 ± 3.1 7.43 ± 0.02 98.3 ± 0.6 24.3 ± 1.2 −0.2 ± 1.5 1.6 ± 0.7
106.1 ± 19 34.3 ± 2.6 7.44 ± 0.01 98.3 ± 0.6 24.4 ± 1 0 ± 1.1 1.6 ± 0.6
0.229 0.708 0.604 0.435 0.352 0.339 0.956
FLEX Transition from/to
Worse [n] Better [n] Percentage ‚better ‘[%] a
off/off
on/on
off/on
on/off
16 35 68.6
22 28 56
93 36 27.9a
20 105 84
pO2 [mmHg] pCO2 [mmHg] pH sO2 (%) HCO3- (mmol/l) Base excess (mmol/l) Lactate (mmol/l)
= p < 0.05 compared to ‘on/off’.
pO2 = O2 partial pressure, pCO2 = CO2 partial pressure, sO2 = O2 saturation, HCO3- = bicarbonate.
FLEX caused lung recruitment and an alveolar stabilizing effect comparable to PEEP (Borgmann et al., 2018). In other words, the effects of FLEX are closely related to those that one expects from setting PEEP, however without the necessity of higher airway pressures. This might be potentially useful for critical care therapy. Transferring the lung protective potential of FLEX to the ICU would however require applicability during spontaneous breathing of spontaneous breathing support since modern strategies in critical care therapy more and more abstain from controlled ventilation (Guldner et al., 2014) and deep sedation (Devlin et al., 2018). In an earlier study in test persons breathing during rest, a subliminal perception of FLEX-related
breathing discomfort could be identified when the test persons were asked to rate transitions from one to another breathing mode in a forced choice fashion (Wirth et al., 2014). By contrast, in the current study in test persons with increased ventilation demand, breathing discomfort was significantly increased, however as indicated by the shift from 2.2 to 3.2 on a scale ranging up to 10 we feel justified to interpret this increase in breathing discomfort as well tolerable. Surprisingly, some test persons reported not to be able to identify a difference between the breathing phases and some perceived the decelerated expiration even as pleasant. 4
Respiratory Physiology & Neurobiology 271 (2020) 103303
S. Schumann, et al.
resistance. Therefore, further studies focusing on ventilation comfort and long term effects of FLEX in patients with injured lungs are required.
In a comparable fashion to this preceding study, we observed that with and without FLEX similar peak flow rates were achieved by the test subjects despite the fact that FLEX is supposed to decelerate expiration. The test persons obviously adapted to the conditions under ventilation with FLEX, by increasing the expiratory peak pressure to compensate for the increases expiratory resistance during breathing with FLEX. This might have potentially increased the expiratory effort of breathing. However, this had neither severely increased the breathing discomfort nor shown a physiological sign in the blood gas analyses between breathing modes after pedalling. EIT images revealed a higher tidal impedance change, reflecting the higher tidal volume and a decelerated proceeding of the expiration during breathing with FLEX, compared to without FLEX. In contrast to studies in mechanically ventilated patients and in the porcine models of ARDS we found neither baseline shifts nor ventral-dorsal shifts of ventilation indicating global or regional recruitment. We attribute this to the fact that the lungs of our healthy and consciously spontaneously breathing test persons were well recruited. The test persons exercised in a 45° upright position and did not suffer from atelectasis, e.g. caused by anaesthesia induction and supine position and high inspiratory fraction of oxygen (Hedenstierna and Rothen, 2000) or reasoned by the lung injury. Thus, the lungs had only low potential for recruitment. A beneficial effect of FLEX might therefore only be expected for patients in the respective circumstances.
4.2. Conclusion FLEX does not impair the physical performance in lung healthy volunteers under increased ventilation demand and was throughout perceived as tolerable. We conclude that FLEX might be applicable in spontaneously breathing patients with increased ventilation demand. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References Borgmann, S., Schmidt, J., Goebel, U., Haberstroh, J., Guttmann, J., Schumann, S., 2018. Dorsal recruitment with flow-controlled expiration (FLEX): an experimental study in mechanically ventilated lung-healthy and lung-injured pigs. Crit Care 22, 245. Devlin, J.W., Skrobik, Y., Gelinas, C., Needham, D.M., Slooter, A.J.C., Pandharipande, P.P., Watson, P.L., Weinhouse, G.L., Nunnally, M.E., Rochwerg, B., Balas, M.C., van den Boogaard, M., Bosma, K.J., Brummel, N.E., Chanques, G., Denehy, L., Drouot, X., Fraser, G.L., Harris, J.E., Joffe, A.M., Kho, M.E., Kress, J.P., Lanphere, J.A., McKinley, S., Neufeld, K.J., Pisani, M.A., Payen, J.F., Pun, B.T., Puntillo, K.A., Riker, R.R., Robinson, B.R.H., Shehabi, Y., Szumita, P.M., Winkelman, C., Centofanti, J.E., Price, C., Nikayin, S., Misak, C.J., Flood, P.D., Kiedrowski, K., Alhazzani, W., 2018. Clinical practice guidelines for the prevention and management of pain, Agitation/Sedation, delirium, immobility, and sleep disruption in adult patients in the ICU. Crit. Care Med. 46, e825–e873. Goebel, U., Haberstroh, J., Foerster, K., Dassow, C., Priebe, H.J., Guttmann, J., Schumann, S., 2014. Flow-controlled expiration: a novel ventilation mode to attenuate experimental porcine lung injury. Br. J. Anaesth. 113, 474–483. Guldner, A., Pelosi, P., Gama de Abreu, M., 2014. Spontaneous breathing in mild and moderate versus severe acute respiratory distress syndrome. Curr. Opin. Crit. Care 20, 69–76. Hedenstierna, G., Rothen, H.U., 2000. Atelectasis formation during anesthesia: causes and measures to prevent it. J. Clin. Monit. Comput. 16, 329–335. Persaud, N., McLeod, P., 2008. Wagering demonstrates subconscious processing in a binary exclusion task. Conscious. Cogn. 17, 565–575. Schmidt, J., Gunther, F., Weber, J., Wirth, S., Brandes, I., Barnes, T., Zarbock, A., Schumann, S., Enk, D., 2019. Flow-controlled ventilation during ear, nose and throat surgery: a prospective observational study. Eur. J. Anaesthesiol. 36, 327–334. Schmidt, J., Wenzel, C., Mahn, M., Spassov, S., Cristina Schmitz, H., Borgmann, S., Lin, Z., Haberstroh, J., Meckel, S., Eiden, S., Wirth, S., Buerkle, H., Schumann, S., 2018. Improved lung recruitment and oxygenation during mandatory ventilation with a new expiratory ventilation assistance device: a controlled interventional trial in healthy pigs. Eur. J. Anaesthesiol. 35, 736–744. Schumann, S., Goebel, U., Haberstroh, J., Vimlati, L., Schneider, M., Lichtwarck-Aschoff, M., Guttmann, J., 2014. Determination of respiratory system mechanics during inspiration and expiration by FLow-controlled EXpiration (FLEX): a pilot study in anesthetized pigs. Minerva Anestesiol. 80, 19–28. Slutsky, A.S., 1999. Lung injury caused by mechanical ventilation. Chest 116, 9S–15S. Slutsky, A.S., Ranieri, V.M., 2013. Ventilator-induced lung injury. N. Engl. J. Med. 369, 2126–2136. Wirth, S., Best, C., Spaeth, J., Guttmann, J., Schumann, S., 2014. Flow Controlled Expiration is perceived as less uncomfortable than positive end expiratory pressure. Respir. Physiol. Neurobiol. 202, 59–63. Wirth, S., Springer, S., Spaeth, J., Borgmann, S., Goebel, U., Schumann, S., 2017. Application of the novel ventilation mode FLow-Controlled EXpiration (FLEX): a crossover proof-of-Principle study in Lung-Healthy patients. Anesth. Analg. 125, 1246–1252.
4.1. Critique of methods We could not apply invasive haemodynamic measurements. However, as the haemodynamics were not negatively influenced in any preceding investigation on FLEX (Goebel et al., 2014) and as the heart rates were independent from FLEX, we assume that haemodynamics were not significantly influenced by FLEX in our study. The pedal powers were generated in ascending and not in randomized order. Therefore, the measurements could be biased by the duration of the experiments. However, the test subjects were allowed to pause between each trial and the main focus of our study was on the effect of FLEX which was randomly active or inactive. Therefore, we would not expect this to have influenced our results to a relevant extent. When the test persons were asked to identify the pedal power they could maintain for 30 min, the trial was stopped after identification of the level and the test persons had never to generate this power for 30 min. We cannot exclude that the subjects under- or overestimated the power they actually could have generated. As our primary intention was to investigate the perception of FLEX, we omitted to produce potential fatigue in favour of the possibility of repeated measurements. However, we have to concede that our results are restricted to short term ventilation periods in non-invasively supported lung-healthy persons. By contrast, patients who require mechanical ventilation support suffer from lung injury and are often ventilated for hours or even days. These patients might be more sensitive to ventilation discomfort. The active compensation of the expiratory peak flow may have played a crucial role in this context. A lung injured patient may not be able to generate an increased pressure to compensate the FLEX related
5