where Re[k] = Ve[k]/(1 – exp(–Te[k]/τ)), τ is the time constant, t the time (equal to zero at the beginning of an expiration), and k the index of the cycle.
The Supervisor Module
The supervisor module computes the inspiratory volume, Vi[k], and the expiratory volume, Ve[k], according to the clinician’s requests with regard to tidal volume, Vt[k], and the FRC correction, ΔV. In addition, the supervisor module compensates for errors in piston movement, when present. The following algorithm is used:
The desired inspiratory volume is computed with Vt[k] and ΔV (if it needs to be decreased):
At the end of inspiration, the effective inspiratory volume is measured, yi[k] = yi(t) for t = Ti[k], to compute the volume error, Ei[k], produced by the inspiratory pump:
The desired expiratory volume, Ve[k], is computed based on the Vt[k], ΔV (if it needs to be increased), Ei[k], and the previous expiratory error, Ee[k−1]:
At the end of expiration, the effective expiratory volume is measured, ye[k] = ye(t) for t = Te[k], to compute the volume error produced by the expiratory pump, Ee[k], which is memorized for the next respiratory cycle:
This error management algorithm (Equations 2 to 5) is stable and converges on a number of cycles, depending on the form of the error.
Finally, lower and upper pressure limits are set to protect the patient’s lung. If the upper pressure, Pmax, is reached during the inspiration phase, P(t) > Pmax, the inspiratory pump is stopped, and the cycle switches to the end-inspiratory pause. In this case, due to Equation 3, only the volume instilled in the lungs will be expired. Also, if the lower pressure limit, Pmin, is reached during the expiration phase with the risk of an airway closure, P(t) < Pmin, the expiratory pump is stopped, and the cycle switches to the end-expiratory pause. In such a case, the desired inspiratory volume for the next inspiration is equal to the volume expired from the lungs during the stopped expiratory cycle, Vi[k] = ye[k – 1], instead of the volume computed according to Equation 2. Hence, every time a pressure limit is reached, the volume inserted or expired from the lungs is adjusted to maintain the same FRC. Hence, each pump is time-limited, volume-controlled, and pressure-limited.
Testing of the Condenser
Before this test, the offsets on each thermocouple probe were previously removed. Throughout their calibration range, they maintained a standard deviation of ±0.3°C. Condenser efficiency was determined by using PFOB (perfluorooctyl-bromide) liquid. The gas and PFC vapor mixture coming from both oxygenators at 39°C was redirected to an external reservoir, where the condenser sits. In this reservoir, the amount of retrieved PFC liquid was measured and compared with the level variation in the buffer reservoir. During this test, the PFC liquid in the ventilator moves in a closed loop, using the preparation cycle.34,35 The bottom of the condenser was maintained at 5°C, and the gas flow was set to 3 l/min in each oxygenator. The measured condenser efficiency was 75% ± 5%, yielding a net PFOB loss of about 21 ml/h.
Testing of the Pumps and Their Control System
First, the supervisor module was tested to determine whether the algorithm converges and manages measurable pumping errors. A dSpace acquisition station (DS1003 processor board, DS2201 I/O board, dSpace Inc., Novi, MI) was used to precisely record the analog signals provided by the potentiometers of the expiratory and inspiratory pumps. Four Vt were chosen, from 75 to 150 ml, to reflect the weight range of a term newborn lamb. Over a 2-hour period, the differences between the inspired and expired volumes (ΔFRC) were summed. The results are presented as mean ± SD range in Table 1 Of note, the standard deviation of the inspired volume was close to the quantization error of the analog signal by the PLC to its discrete state, which is 0.14 ml. Thus, the supervisor module, presented with Equations 2 to 5, thereby proved its effectiveness in managing measurable errors.
Second, to test the operating capabilities of the expiratory and inspiratory piston pumps and their overall system control, in vivo TLV was mimicked by replacing the lungs with a fixed reservoir. A pressure sensor (Bentley Trantec 800, Irvine, CA) placed at the bottom of the reservoir was used to determine the volume variations within the reservoir, as a surrogate for Vt, over a 2-hour period. As performed earlier, four Vt were chosen, from 75 to 150 ml, to reflect the weight range of a term newborn infant during the first weeks of life. The results presented in Table 2 confirm that the measured Vt is within ±2% of the reference value.
Testing of the Oxygenator
To characterize the oxygenator, the equivalent gas exchange surface was defined by Seq = 6VO2/Db, with Db the bubble diameter and VO2 the oxygen volume in the PFC. The bubble diameter Db was obtained by taking a digital image of the bubbles in the oxygenator and by averaging a total of 50 images. The oxygen volume in the PFC VO2 was calculated by measuring the change in the level of PFC before and after initiating the flow of gas under the membrane. When using PFOB, the mean bubble diameter was 2.0 ± 0.6 mm and the global volume of gas in the PFC averaged 280 ± 10 ml at a total flow rate of 6 l/min. Thus, the Seq was 0.84 ± 0.1 m2. With two oxygenators in series, the equivalent gas exchange area is greater than an ECMO for children weighing less than 10 kg (based on a Medtronic ECMO model 0600, Minneapolis, MN).
In Vivo Testing
The experimental protocol was approved by our institutional Ethics Committee for Animal Care and Experimentation. Five healthy newborn lambs (<5 days) were placed in a supine position under a radiant heater to maintain a central temperature of 39 ± 0.5°C. Heart rate was recorded using a Hewlett-Packard cardiorespiratory monitor (model HP 78342A, Palo Alto, CA). Oxygen saturation (SpO2) was monitored with the use of a pulse oximeter probe placed on the base of the tail (8000R reflectance sensor, Nonin medical, Plymouth, MN). The analog signals generated from the TLV ventilator were recorded by using Signal Ranger I/O boards (SoftdB, Quebec, Canada). Lambs were sedated with atropine (0.1 mg/kg) subcutaneously, ketamine (10 mg/kg), and midazolam (0.1 mg/kg) intramuscularly and orally intubated with a cuffed endotracheal tube (ETT). Gas ventilation (Servo 300, Siemens-Elema AB, Solna, Sweden) was then initiated at a rate of 40 breaths/min, a peak inspiratory pressure of 15 cm H2O, and a positive end-expiratory pressure of 5 cm H2O. The fraction of inspired oxygen (FIO2) was adjusted to maintain an SpO2 of 100%. After cannulation of the right jugular vein, animals were anesthetized with an intravenous bolus of 20 mg/kg thiopental followed by a continuous intravenous infusion of 2 mg/kg per hour. Sedation was obtained with a continuous intravenous infusion of alfentanyl (2 μg/kg per hour) and intermittent intravenous administration of 1 μg/kg as needed. Curarization was obtained with intermittent intravenous administration of vecuronium bromide (0.1 mg/kg). Antibiotics (0.05 mg/kg duplocillin and 5 mg/kg gentamicin) were also injected intramuscularly. A 4F catheter was placed into the femoral artery for monitoring hemodynamic parameters and central temperature by the thermodilution method, using a PC 8000 PICCO monitor (Pulsion Medical System AG, München, Germany). Arterial blood gases were monitored with the use of a blood gas analyzer (model 1306, Coulter Electronic, Ltd, Hialeah, FL). A second catheter, 4F Swan-Ganz (Edwards LifeSciences, Irvine, CA), was placed into the right jugular vein for monitoring pulmonary artery and central venous pressures and for continuous infusion of 4 ml/kg per hour of 5% dextrose. A tracheotomy was then performed, and a 6.0-gauge, cuffed ETT was placed into the trachea.
Tidal Liquid Ventilation Protocol
After a brief period of stabilization, a liquid FRC of 30 ml/kg of warmed, preoxygenated PFC (PFOB from F2 Chemicals, Lancashire, UK) was rapidly instilled into the trachea through the ETT, using a syringe with an 8F gastric feeding tube. Tidal liquid ventilation was then initiated at a rate of 3 to 4 breaths/min, a Vt of 25 ml/kg, and an inspiratory/expiratory (I/E) ratio of 1/3, with a FIO2 of 1.0. An exponential profile was used during expiration and a ramp profile during inspiration. Arterial blood specimens were drawn at 30-minute intervals for determination of PaO2, PaCO2, pH, BE (Base-Excess), and glucose. The ventilator settings were then adjusted to maintain PaO2 >100 mm Hg and PaCO2 at 30 to 50 mm Hg. Sodium bicarbonate or tromethamine (THAM) was used to maintain pH >7.25. The rate of dextrose intravenous infusion was adjusted to maintain glucose blood level at 40 to 100 mg/dl. Normal saline was used as needed to maintain a mean arterial pressure of 65 mm Hg. Hemodynamic parameters were continuously monitored and recorded at 30-minute intervals. Dynamic compliance Cdyn was also recorded at 30-minute intervals, using quasi-static end-inspiration and end-expiration pauses to monitor airway pressures measured at the distal end of the ETT, using a Mikro-Tip catheter 2.3F (SPR524, Millar Instruments Inc., Houston, TX).40,41 After 120 minutes, the animals were given a lethal dose of pentobarbital (100 mg/kg), and the lungs were carefully examined for the presence of perfluorothorax.
Table 3 presents the main results obtained in 5 healthy newborn lambs (<5 days old) with a mean weight of 4.40 ± 1.39 kg. All results are presented as mean ± SD. No perfluorothorax was observed after examination of the lungs. Figure 6 presents typical tracheal pressure and volume profiles during a TLV.
The Condenser-Heater-Oxygenator Unit
The majority of liquid ventilators developed to date use an ECMO with an integrated water heater.6,19–21,24,26,29,30,32 However, to design an optimal integrated CHO and avoid the use of an expensive ECMO, the bubble gas exchanger was selected. On the basis of the characterization results, we were able to demonstrate that the latter compares favorably with an ECMO designed for children weighing <10 kg (Medtronic ECMO model 0600, Minneapolis, MN). Moreover, it can easily be sterilized, and the membrane can be changed between patients.
The major novelty of the presented CHO is its modularity. To reach a specified gas exchange surface, it is possible to add several CHO in parallel or in series, whereas oxygenator sections can be modified to increase the gas exchange area. Hence, the presented prototype can be easily extended from an infant to an adult version. This feature could prove important for future clinical use of the TLV.
The condenser mounted on the top of the gas exchanger limits PFC losses at 21 ml/hr (using PFOB). Based on the data of Wolfson et al.,42 PFC losses from the CHO are lower than that with an Ultrox III (Avecor Cardiovascular Inc., Plymouth, MN) ECMO membrane. However, when comparing PFC losses per square meter of gas exchange area, the CHO and the Ultrox III ECMO membrane are equivalent.
A Volume-Controlled and Pressure-Limited Ventilator
Treatment of a critical respiratory patient requires that safety measures are in place to limit the pressure and to precisely control the volumes.32 In the event of airway closure43 during the expiratory phase, the expiratory pump is immediately stopped when the lower pressure limit is reached. Thereafter, the inspiratory volume to be inserted into the lungs is automatically adjusted to ensure a constant FRC level with the TLV maintaining these new parameters. Compared with other prototypes presented in the literature that halt the TLV and await new instructions, this new approach gives more time for health care providers to investigate and correct the situation while the patient still undergoes TLV. If an FRC or Vt correction needs to be made, the volume inspired or expired will be adjusted consequently. Hence, the proposed system is volume controlled by adjusting Vt and FRC and is pressure-limited.
Control of Functional Residual Capacity
In TLV, piston pumps have proven their efficiency27–30 when compared with peristaltic or gravimetric systems.31 Such a choice is necessary to implement an accurate volume control system. However, the main concern with pumping systems is the difference that can occur between the expiratory and inspiratory volumes. It can be calculated that a small volume difference repeated at each cycle leads to a dramatic FRC increase (or decrease) after a 2-hour experiment. Such problems may be related to a small mechanical and electronic bias between each pump components.
The novel solution proposed with the presented prototype is to monitor and control FRC by means of the buffer reservoir and to use inherent pump independence for FRC readjustments, when necessary. To maintain a constant FRC, the supervisor module automatically adjusts the volumes to compensate measurable errors made by the pumps. However, it is important to note that only measurable errors are controlled. In other words, mechanical, electrical, and quantization errors cannot be treated by the pump supervisor module but could influence FRC over a long period of time. Hence, FRC monitoring during TLV is essential for future clinical use.
With the proposed prototype, PFC can be added in the ventilator to compensate for PFC losses.44 Thus, the ventilator volume is kept constant and the buffer reservoir can be used directly to monitor FRC changes. In the present case, any variation in volume level measured in the reservoir at the end-expiratory pause implies an opposite variation in FRC volume. Ongoing work is currently aimed at developing an online algorithm able to process data obtained during TLV (based on measured volumes and pressures) and deliver an accurate estimation of FRC.
In Vivo Results
The results obtained in 5 healthy term newborn lambs demonstrate the efficiency and safety of our TLV ventilator in maintaining adequate gas exchange and normal acido-basis equilibrium during a short, 2-hour TLV trial. The relatively low PaO2 obtained herein as compared with gas ventilation in newborn lambs with normal lungs has been previously described as the so-called Kylstra effect.45 In their theoretical model of gas exchange, Kylstra et al. demonstrated that a gradient (from the alveolus center up to its wall) exists between the inspired PiO2 and alveolar PaO2. Because oxygen diffusion in gases is very fast, this gradient is small. However, oxygen diffusion is much slower in liquids (including PFC liquids), creating a larger gradient, which explains the lower values of PaO2 in TLV compared with gas ventilation.
Cardiovascular stability was also maintained and was not different from results obtained in a similar gas-ventilated animal preparation (data not shown). Furthermore, airway pressure, lung volumes, and ventilation scheme were maintained in the targeted range, including the absence of any observed perfluorothorax. Overall, the present results are in accordance with previous studies on TLV with healthy term newborn lambs and support further testing of our TLV ventilator in animal models of lung injury.46
The distinct advantage of our TLV prototype is its ability to control FRC by using a system of independent pumps and the ability to estimate FRC by means of the buffer reservoir. Current developments will enable better control of FRC and reduction in airway collapse. These developments include a more accurate model of the airway system during TLV, thus improving pump control.
The TLV ventilator presented is also shown to be efficient and safe for volume-controlled, pressure-limited tidal liquid ventilation in 5 term newborn lambs with healthy lungs. Further TLV studies will be performed on experimental models of lung injury, especially acute respiratory distress syndrome.
This work was supported in part by the Quebec Foundation for Research into Children’s Diseases and the Fonds de Recherche sur la Nature et les Technologies.
The authors thank Valerie Cardinal, Johann Lebon, Valerie Provost, and Caroline Ouellet for their technical assistance supporting the in vivo experiments. The authors also thank Rémy Oddo, Guillaume Drolet, and Oliver Berbuer for their technical assistance and Pulsion Medical Systems AG, which provided the acquisition software for the PiCCO.
1.Clark LC, Gollan F: Survival of mammal breathing organic liquids equilibrated with oxygen at atmospheric pressure. Science
152: 1755–1756, 1966.
2.Tooley R, Hirschl RB, Parent A, Bartlett RH: Total liquid ventilation with perfluorocarbons increases pulmonary end-expiratory volume and compliance in the setting of lung atelectasis. Crit Care Med
24: 268–273, 1996.
3.Reickert CA, Rich PB, Crotti S, et al
: Partial liquid ventilation and positive end-expiratory pressure reduce ventilator induced lung injury in an ovine model of acute respiratory failure. Crit Care Med
30: 182–189, 2002.
4.Cox CA, Wolfson MR, Shaffer TH: Liquid ventilation: A comprehensive overview. Neonatal Network
15: 31–43, 1996.
5.Hirschl RB, Tooley R, Parent A, et al
: Evaluation of gas exchange, pulmonary compliance, and lung injury during total and partial liquid ventilation in the acute respiratory distress syndrome. Crit Care Med
24: 1001–1008, 1996.
6.Dani C, Costantino ML, Martelli E, et al
: Perfluorocarbons attenuate oxidative lung damage. Pediatr Pulmonol
36: 322–329, 2003.
7.Richman PS, Wolfson MR, Shaffer TH: Lung lavage with oxygenated perfluorochemical liquid in acute lung injury. Crit Care Med
21: 768–774, 1993.
8.Foust R III, Tran NN, Cox C, et al
: Liquid assisted ventilation: An alternative ventilatory strategy for acute meconium aspiration injury. Pediatr Pulmonol
21: 316–322, 1996.
9.Wolfson MR, Greenspan JS, Shaffer TH: Liquid assisted ventilation: An alternative respiratory modality. Pediatr Pulmonol
26: 42–63, 1998.
10.Rotta A, Steinhorn D: Partial liquid ventilation reduces pulmonary neutrophil accumulation in an experimental model of systemic endotoxemia and acute lung injury. Crit Care Med
26: 1707–1715, 1998.
11.Rüdiger M, Some M, Jarstrand C, et al
: Influence of partial liquid ventilation on bacterial growth and alveolar expansion in newborn rabbits with group B-streptococcal pneumonia. Pediatr Res
54: 808–813, 2003.
12.Hirschl RB, Parent A, Tooley R, et al
: Liquid ventilation improves pulmonary function, gas exchange, and lung injury in a model of respiratory failure. Ann Surg
221: 79–88, 1995.
13.Hirschl RB, Croce M, Gore D, et al
: Prospective, randomized, controlled pilot study of partial liquid ventilation in adult acute respiratory distress syndrome. Am J Respir Crit Care Med
165: 781–787, 2002.
14.Pedneault C, Renolleau S, Gosselin R, et al
: Total liquid ventilation using a modified extracorporeal gas exchange circuit: Preliminary results in lambs. Pediatr Pulmonol
18: A241, 1999.
15.Wolfson MR, Greenspan JS, Deoras KS et al
: Comparison of gas and liquid ventilation: Clinical, physiological and histological correlates. J Appl Physiol
72: 1024–1031, 1992.
16.Cox C, Fox WW, Wolfson MR, Shaffer TH: Liquid ventilation: Gas exchange, perfluorochemical uptake, and biodistribution in an acute lung injury. Pediatr Crit Care Med
3: 288–296, 2002.
17.Sekins KM, Shaffer TH, Wolfson MR: Apparatus for pulmonary delivery of drugs with simultaneous lavage and ventilation. US Patent
5: 562–608, 1996.
18.Sekins KM, Leeper DB, Horman JK, et al
: Feasibility of lung cancer hyperthermia using breathable perfluorochemical (PFC) liquids, part I: Convective hyperthermia. Int J Hyperthermia
20: 252–277, 2004.
19.Degraeuwe PLJ, Dohmen LRB, Geilen JM, Blanco CE: A feedback controller for the maintenance of FRC during tidal liquid ventilation: Theory, implementation, and testing. Int J Artif Organs
23: 680–688, 2000.
20.Hirschl RB, Merz SI, Montoya JP et al
: Development and application of a simplified liquid ventilator. Crit Care Med
23: 157–163, 1995.
21.Corno C, Fiore GB, Martelli E, et al
: Volume controlled apparatus for neonatal tidal liquid ventilation. ASAIO J
49: 250–258, 2003.
22.Matsuda K, Sawada S, Bartlett RH, Hirschl RB: Effect of ventilatory variables on gas exchange and hemodynamics during total liquid ventilation in a rat model. Crit Care Med
31: 2034–2040, 2003.
23.Meinhardt JP, Sawada S, Quintel M, Hirschl RB: Comparison of static airway pressures during total liquid ventilation while applying different expiratory modes and time patterns. ASAIO J
50: 68–75, 2004.
24.Curtis SE, Furhman BP, Howland DF: Airway and alveolar pressures during perfluorocarbon breathing in infant lambs. J Appl Physiol
68: 2322–2328, 1990.
25.Koen PA, Wolfson MR, Shaffer TH: Fluorocarbon ventilation: Maximal expiratory flows and CO2
elimination. Pediatr Res
24: 291–296, 1988.
26.Heckman JL, Horman J, Shaffer TH, Wolfson MR: Software for real-time control of a tidal liquid ventilator. Biomed Eng Tech
33: 268–276, 1999.
27.Sekins KM, Nugent L, Mazzoni M et al
: Recent innovation in total liquid ventilation system and component design. Biomed Eng Tech
33: 277–284, 1999.
28.Meinhardt JP, Quintel M, Hirschl RB: Development and application of a double-piston configured, total-liquid ventilatory support device. Crit Care Med
28: 1483–1488, 2000.
29.Larrabe JL, Alvarez FJ, Cuesta EG et al
: Development of a time-cycled volume-controlled pressure limited respirator and lung mechanics system for total liquid ventilation. IEEE Trans Biomed Eng
48: 1134–1144, 2001.
30.Tredici S, Komori E, Funakubo A et al
: A prototype of a liquid ventilator using a novel hollow-fiber oxygenator in a rabbit model. Crit Care Med
32: 2104–2109, 2004.
31.Meinhardt JP, Ashton BA, Annich GM, et al
: The dependency of expiratory airway collapse on pump system and flow rate in liquid ventilated rabbits. Eur J Med Res
8: 212–220, 2003.
32.Baba Y, Taenaka Y, Akagi H et al
: A volume-controlled liquid ventilator with pressure-limit mode: Imperative expiratory control. Artif Organs
20: 1052–1056, 1996.
33.Cox C, Stavis RL, Wolfson MR, Shaffer TH: Long-term tidal liquid ventilation in premature lambs: Physiologic biochemical and histological correlates. Biol Neonate
84: 232–242, 2003.
34.Robert R, Micheau P, Cyr S, et al
: Method and apparatus for conducting liquid ventilation with control of residual volume and ventilation cycle profile. US Patent Application Publication 2004/0134486 A1, 2004.
35.Micheau P, Praud JP, Robert R, et al
: Method and apparatus for conducting total liquid ventilation with control of residual volume and ventilation cycle profile. Canadian Patent CA
2: 451–261, 2003.
36.Parker JC: Liquid ventilator system and use thereof. US Patent
5: 706–830, 1998.
37.Shaffer TH, Wolfson MR, Heckman JL, et al
: Liquid ventilator. US Patent
6: 105–572, 2000.
38.Diaz FJA, Barrena JLL, Valls I, Soler A, et al
: Dispositivo para respiración asistida de líquido con control de volumen. Spanish Patent ES
2: 154–596, 2001.
39.Shaffer TH, Wolfson MR: Process control for liquid ventilation and related procedures. US Patent
5: 335–650, 1994.
40.Iotti GA, Braschi A: Measurements of Respiratory Mechanics During Mechanical Ventilation. Rhäzüns, Switzerland, Hamilton Medical Scientific Library, 1999.
41.Egan DF, Scanlan CL, Wilkins RL, Stoller JK: Fundamentals of Respiratory Care. 7th edition. Russell J, ed. St Louis, MO, Mosby Inc, 1999.
42.Wolfson MR, Miller TF, Peck G, Shaffer TH: Multifactorial analysis of gas exchanger efficiency and liquid conservation during perfluorochemical liquid-assisted ventilation. Biomed Eng Tech
33: 260–267, 1999.
43.Baba Y, Brant D, Brah SS, et al
: Assessment of the development of choked flow during liquid ventilation. Crit Care Med
32: 201–208, 2004.
44.Libros R, Philips CM, Wolfson MR, Shaffer TH: A perfluorochemical loss/restoration (L/R) system for tidal liquid ventilation. Biomed Eng Tech
34: 351–360, 2000.
45.Kylstra JA, Paganelli CV, Lanphier EH: Pulmonary gas exchange in dog ventilated with hyperbarically oxygenated liquid. J Appl Physiol
21: 177–184, 1966.
Copyright © 2006 by the American Society for Artificial Internal Organs
46.Stavis R, Wolfson MR, Cox C, et al
: Physiologic, biochemical, and histologic correlates associated with tidal liquid ventilation. Pediatr Res
43: 132–138, 1998.