Neonatal Partial Liquid Ventilation for the Treatment and Prevention of Bronchopulmonary Dysplasia

Survival rates in extremely premature infants are increasing as are rates of bronchopulmonary dysplasia. Current therapeutic options are not sufficient to prevent or treat bronchopulmonary dysplasia in many infants. This presents a large cost, both human and economic. Perfluorocarbon liquids have been studied since the 1960s for various biomedical applications. After some promising studies in the 1990s, work in this field was halted. Despite considerable attention, including significant amounts of published data, the temporal gap has seen a generation of physicians who may not know about this potentially valuable therapy. This article will reintroduce partial liquid ventilation, a promising therapeutic option in acute and chronically ill infants, to clinicians by examining the history and presenting new data.After completing this article, readers should be able to:In their 1962 article entitled, “Of mice as fish,” Kylstra and colleagues wrote with excitement about the implications of a newfound mode of ventilation. In liquid ventilation using oxygenated saline solutions, these investigators saw a sort of reverse engineering of evolution—mammals regressing to their supposed piscine origins. Kylstra et al envisioned a world in which pool water, titrated to have an adequate saline concentration, would allow for more efficacious resuscitation from drowning. (1) In its most essential form, liquid ventilation involves filling the lungs of an animal with some kind of liquid. That liquid, at varying doses, may be cycled in and out of the lungs with the assistance of a ventilator. In 1966, Clark and Gollan first described the use of perfluorocarbons (PFCs) in liquid ventilation toward a similar end as did Kylstra and colleagues—undersea survival and submarine escape. (2) Since those early days, the unique qualities of PFCs have driven research in diverse fields—from creating better nonstick frying pans (the now-ubiquitous Teflon [Chemours, Wilmington, DE]) to improving lung mechanics. More recent research has sought to harness the potential of PFCs as a medium in liquid ventilation in an attempt to reduce the large burden of respiratory disease in infants, both acute and chronic.Perfluorooctyl bromide (PFOB) has emerged as the PFC most compatible with liquid ventilation. PFOB is inert and can carry large amounts of dissolved oxygen and carbon dioxide. (3) Over the past 40 years, PFOB has been used in animals and humans in a number of different ways, including imaging and blood replacement, as well as experimental liquid ventilation. Numerous studies in various populations have shown that PFOB can decrease inflammation, improve pulmonary mechanics, and reduce damage associated with conventional mechanical ventilation. (3)(4)(5)(6) This review will summarize the history of PFOB and its use in partial liquid ventilation (PLV), and discuss new avenues for research to prevent or mitigate lung injury in infants.PFOB is one of only a handful of PFC liquids approved for use in humans. PFOB has a simple backbone of 8 carbon molecules bonded together with single bonds. Instead of carbon-hydrogen bonds, this backbone is populated with carbon-fluorine bonds. The addition of a bromine molecule to one end completes the picture. This structure yields a unique combination of properties. The carbon-fluorine bonds are extraordinarily strong, once described as, “the strongest single bond found in molecular compounds.” (7) Fluorine has the highest electron affinity of any element, extremely low polarizability, and a dense electron cloud. As a result, fluorine atoms shield the carbon backbone both physically (sterically) and with regard to bonding availability. These strong intramolecular forces are complemented by very weak intermolecular forces. The low polarizability results in minimal van der Waals forces; there are no permanent dipoles and transient (induced) dipoles are unlikely. High intramolecular forces and low intermolecular forces result in a molecule that is kinetically and thermodynamically inert, and therefore able to dissolve large amounts of gases. The solubility of gases dissolved in PFC liquid is directly related to their partial pressure. (7)PFOB has additional qualities that separate it from other members of its chemical class. The bromine atom at the end of the molecule grants PFOB radiopacity. As such, it has been investigated as a contrast agent and, recently, its filling patterns have been investigated as a novel radiologic tool. (8) Many studies have sought to harness the chemical properties of PFOB for medical applications, using some combination of its radiopacity, inertness, and capacity to dissolve respiratory gases. PFOB has been used as an injectable contrast agent, a blood substitute, and a liquid ventilation medium, amongst others. (9)(10)When considering using PFOB in respiratory therapy, its chemical properties must be placed within a physiologic framework. The low surface tension and high density of PFOB may help to directly improve pulmonary mechanics and oxygenation. The ability of PFOB to dissolve large amounts of respiratory gases is an additional characteristic that gives liquid ventilation its lung therapeutic potential. Liquid ventilation is generally classified into 2 strategies: PLV and total liquid ventilation (TLV). In TLV, liquid completely fills the lungs and preoxygenated liquid tidal volumes are cycled in and out with gravity or with the use of a special liquid ventilator. PLV is the partial filling of the lungs to a resting liquid lung volume and uses conventional mechanical ventilation for oxygenation of, and carbon-dioxide removal from, the liquid medium.This review provides a capsule summary of the physiologic properties of the PFC-filled lung. A review by Kaisers et al provides an excellent description of the physiologic changes associated with PLV with the use of PFOB. (11) A review by Shaffer et al can provide further technical expertise on the subject. (12)PFC liquids are heavier than water (PFOB is twice as dense, 1.93 g/mL at 25°C) and thus, when infused into the airway, may quickly create structural and mechanical changes in the lung. Because of gravity, PFC liquids tend to settle in the more dependent lung regions once instilled. In the diseased lung, these dependent regions are often atelectatic and the most poorly ventilated. It is posited that the dense column of noncompressible PFC liquid can act as liquid positive end-expiratory pressure (PEEP), gently forcing open collapsed alveoli. (11) PLV may help recruit alveoli by providing additional distending pressure through its weight and improve ventilation-perfusion matching.A second crucial determinant of its behavior in the lung is the low surface tension of PFC liquids. In the surfactant-depleted lung, more ventilatory pressure is required to overcome high surface tension and maintain sufficient pulmonary volume. With the addition of a low surface tension liquid, especially PFOB, these forces can be mitigated, alveolar volume stabilized, and dynamic pulmonary compliance improved. Further, additional regions of the lung can take part in gas exchange in a more homogeneous manner. (13)By recruiting alveoli, decreasing surface tension, and improving distribution of ventilation, PLV may mitigate ongoing ventilator-induced injury including regional hyperoxia, barotrauma, and volutrauma.Finally, as previously discussed, PFC liquids can dissolve large amounts of respiratory gases. The synergistic effects of these 3 characteristics make PFOB a suitable molecule for liquid ventilation.Inflammation is an important contributor to ventilator-induced lung injury. (14)(15)(16)(17) Studies suggest that PFOB may have an anti-inflammatory effect at the cellular level. (11) The mechanism of this proposed effect is poorly understood. Thomassen and colleagues studied the response of alveolar macrophages to in vitro exposure to PFOB. They found that PFOB significantly reduced release of the inflammatory cytokines tumor necrosis factor α, interleukin 1, and interleukin 6. (15) Merz et al expanded on this cellular work, studying the reaction of surfactant-depleted newborn pigs to conventional mechanical ventilation, high-frequency oscillatory ventilation (HFOV), and PLV. They found that release of leukotriene B4 and interleukin 6 was reduced in PLV with PFOB, compared with conventional mechanical ventilation and HFOV. Furthermore, PLV with PFOB had a lavage action, removing inflammatory debris because of its immiscibility with other materials present in the lungs. (14) PFOB may alleviate the effects of local and systemic inflammatory factors on the course of respiratory disease. (12)PLV introduces a dense, incompressible liquid into the pulmonary space. Although these characteristics are important to the mechanical advantages of PLV, the addition of a dense liquid into the lung theoretically could decrease pulmonary and systemic blood flow. In 1979, Lowe et al found increased pulmonary vascular resistance with systemic redistribution of blood flow and metabolic acidosis during TLV in a normal lung model. (18) Further evaluation suggested that these effects were largely secondary to cardiovascular depression associated with anesthesia and were reversed with low-level volume resuscitation. In 1986, using an in situ isolated healthy lung model, they showed that pulmonary blood flow was not only preserved to dependent regions but was improved to less dependent areas when filled with PFC liquid. (19) More recently, several clinical studies have found no deleterious effects of PLV on the circulation. (11)(20)(21) One possible reason is that improved oxygen delivery with PLV may decrease vasoconstriction from regional hypoxia, and thus offset the effects of filling the lung with a dense liquid. (11)The unique ability of PFCs to carry dissolved respiratory gases has generated considerable interest in their ability to serve as a substrate for gas exchange in the lung. Over the past 50 years, multiple animal studies have explored the safety and potential toxicity of PFCs in biological systems, and whether liquid ventilation (in any form) may be an effective treatment for preventing severe lung injury from conventional mechanical ventilation. Holaday and colleagues administered TLV to adult dogs at a rate of 3 to 5 breaths/min with tidal volumes (approximately 400 mL) using the PFC liquid FX-80. (22) The dogs were supported with TLV for 1 to 8 hours. All survived the period of TLV, and a subgroup transitioned to air breathing. In postmortem analysis, they concluded that FX-80 was not metabolized and only deposited residually in small amounts, mainly in adipose tissue. (22) In 1972, Modell et al provided ventilation to 3 adult monkeys using PFC liquid caroxin-D for 10 to 60 minutes on 2 occasions, separated by 2 to 5 months. The transition from liquid back to air breathing was uneventful and 3 years later, there were no clinical indications that liquid ventilation had been provided at all. (23) These early studies suggested that PFC liquids could be safely used as a ventilatory medium and that prolonged exposure was not necessarily dangerous.Shaffer et al further developed the technique of mechanical TLV for the support of mammals. They designed and built a liquid ventilator that cycled PFC liquid tidal volumes through a membrane oxygenator (24) and carbon dioxide scrubber and delivered these into a PFC liquid–filled lung. They demonstrated that TLV enabled adequate oxygenation and ventilation in a normal lung. The group went on to investigate physiologic adjustments in a normal lung during TLV, and noted the development of metabolic acidosis (18) and increased pulmonary vascular resistance. (25) In 1983, Shaffer and colleagues demonstrated improved lung function in a preterm lamb model of respiratory distress syndrome (RDS) treated with TLV. Their most remarkable finding was that the very preterm lambs (ranging from 106 days of gestation to 70% of term gestation) treated with TLV had lung compliance and gas exchange that resembled those of much more mature (132–138 days of gestation) lambs. (24)The 1990s brought an abundance of preclinical liquid ventilation work on evaporation rates, positioning, liquid selection, and filling patterns. In a 1992 study, Wolfson and colleagues demonstrated that preterm lambs exposed to TLV incurred less barotrauma than those receiving conventional mechanical ventilation. TLV allowed for better distribution of ventilation in the lung compared with conventional mechanical ventilation, likely because of the high gas solubility and low surface tension of the liquid. (26) A prospective, randomized, controlled trial in adult cats by Richman and colleagues confirmed some of these physiologic effects using a PFC lavage in which they noted improvements in pulmonary mechanics and gas exchange. (27)Perhaps the most significant shift in the 1990s was the switch from TLV to a new liquid ventilation strategy, PLV, developed by Fuhrman and colleagues. Whereas TLV required specific and complex machinery to dissolve gas into the liquid and cycle liquid tidal volumes through the lungs, Fuhrman discovered that many of the benefits of TLV could be achieved with a standard gas-driven ventilator. In PLV, a volume approximating the functional residual capacity (FRC) is instilled and oxygenated in vivo with a conventional ventilator. PLV was first described as a standalone strategy by Fuhrman et al in a piglet model. (28) This new approach demonstrated for the first time that liquid ventilation could be accomplished in a less and et al went on to the technique of PLV in a preterm lamb model, and in demonstrated that PLV improved lung function and gas exchange in compared with conventional mechanical ventilation. studies in various animal to the development of PLV were published in the years these These found that PLV improved lung compliance and gas exchange compared with conventional mechanical ventilation in multiple lung injury was compatible with and could be using and reduced lung injury in lung function likely from a more homogeneous distribution of ventilation and changes in mechanical properties of the lung normal pulmonary 1990s brought the first of liquid ventilation with PFC liquid in humans. to the preclinical several different have been used the of liquid ventilation. human published by in and used TLV. The first human trial of PLV was in by et al and published in This was a by the and in preterm infants with severe who had therapy. The liquid was a was seen for liquid and conventional mechanical ventilation were to maintain a liquid in the infants PLV for an of with a of hours. infants who required for had during PLV with the use of a conventional some in lung They were from the trial less than to high-frequency ventilation by the of these infants to have improved lung the The 10 infants had improvements in several clinical oxygen tension improved by and dynamic compliance by and colleagues administered PFOB to infants as well as 10 and receiving support for respiratory had their filled to a and low-level conventional mechanical ventilation was pulmonary compliance increased from to Although these were they provided valuable to support the of PLV in ill infants, and studies in were built on the of those small of PLV. et al administered PFOB to 6 infants who had conventional respiratory support and were receiving with no 2 pulmonary compliance increased in infants receiving PFOB and were to (3) a separate and colleagues filled the lungs of several ill infants with After for a of days, showed in oxygenation and ventilation, with of lung PFOB was in 1 because of multiple and of PLV was to the in and ill infants and contrast to early studies in infants, a of large studies in found benefits of PLV with PFOB. specific In et al studied adult years, with severe respiratory who were receiving This found improved gas exchange in receiving PLV. the previously in animal and Despite these the results were positive to support a In et al found positive results in a trial of adult receiving PLV had a in their to acute respiratory distress syndrome and than years of were from mechanical ventilation significantly was not improved with PLV and there was an in transient respiratory and in those receiving PLV. In a controlled and colleagues studied adult with This group compared PLV to PLV and conventional mechanical ventilation. They found that PLV did not improve to conventional mechanical were and there were significantly more days in the conventional mechanical ventilation The increased of and in the PLV This trial was the to PLV in A review on PLV in with acute lung injury and concluded there was no to support the use of PLV in those with or acute lung injury. effects associated with in these adult may be to in the and and colleagues suggest that the and by some PLV were the result of in the of residual PFOB in the lungs at the time of The their ventilatory strategy as well as as potential of their A rate in the group than that used in the and a in for with at the of may have PLV safety for are by in and for the with and which may have of PLV These diverse of a and colleagues have noted that survival of treated with PLV increased for than Furthermore, in the large trial had and which could to PFOB distribution in the dependent areas of the is in dependent lung regions in with This may have multiple of PLV, to increasing alveolar pressure and seen in the can be from the of clinical with PLV in humans. PLV can be a and in infants. there is about the to clinical of PLV. on of PFOB, of therapy, and After the results of the adult further research into PLV in infants The studies in are an especially important to in the of PLV. Their to the of the of PFOB, a of PFOB and the of interest in at of an new from the for in infants with bronchopulmonary dysplasia to clinical research into this 2 its have the safety of, and clinical response PLV with PFOB in preterm infants than and with severe of the research or clinical had PLV for with severe the of has in the years since the of PLV in infants. A significant has been in survival to for extremely premature infants at to gestation and for infants in the there has been a significant in for infants at through With the in who would have early in years are who are at increased for as severe at a as there is a for new and to these in clinical there are no that can significantly the course of severe with severe are than many of those studied previously and, as a result of their are the of PLV on a physiologic with a and present for new therapeutic options to treat severe in part as a at the in is a single randomized, clinical In the first and were treated with PFOB instilled for PLV 5 All of these had severe and were at a of 5 months. The safety and of oxygen and carbon dioxide for incurred a safety 2 had mainly and to be related to These but did a of The which 2 infants did not to to an lung volume. After the was to administered once a the course of the This was had been an through several and part of the of must on finding the in and disease The of is significantly than that used in which from 10 to These to the of in the are not be more may to the of the the results from the of the trial are Although not to has shown a significant and in in the PLV group 3 additional of this recent trial has been the of the development of novel radiologic to patterns of PFOB in the lung (8) this small group of preterm infants with severe PLV with PFOB was and patterns were but there were no effects related to residual PFOB to days of of have been on PLV and many and clinical later, use of PLV in infants has not been liquid ventilation, in its is to and PLV may a second The potential of PFOB as a therapeutic agent for use in PLV in its A unique ability to carry respiratory a low surface tension, and anti-inflammatory effects PLV with PFOB as a promising therapy. part of this second is from the This article is to the history in a that will the generation of to and about are several important to studies that can be from that has been studied as a As such, the have been extremely ill at the of therapy. This it to and the effects of the In a who other disease could have been the In infants, the lungs can The of lung injury more severe and more infants may of PFOB and may have more because their lungs have not had the time to with changes and vascular the to study, is in a of can the behavior of the lungs. This was an in the adult than years had different to PLV than did a a novel of a with the who have standard of therapeutic the those are some of the most from to adult the of PLV research in One of the by years of is the of to in a with complex disease In is not a sufficient of safety or A at and various will more and more published work on PLV in infants has been with it to the of In a recent the found only 1 controlled trial in with and The that the of could be because of They noted that further the analysis, who were treated with PLV were ill and was it is to any to PLV the of liquid has been an especially in adult for infants, it has been to the of PFC liquid in the lung at any have been but clinical has not been of PFC liquid has been a in the development of effective therapeutic may be for more research in this but there is not liquid to present studies on treatment of severe the liquid and anti-inflammatory effects of PLV with PFOB in infants with severe severe and other Many studies the is administered in the disease that the of PLV in premature infants in of therapy. this could lung volumes and, prevent the inflammatory that is to have a on the development of on the et al in very premature infants and 2 additional premature infants treated at in the 1990s, that not only the liquid improve the of the but it can have a on mechanical ventilator and oxygen This is to the and of PLV in premature addition to more studies should be may to novel radiologic to filling patterns and lung research should not only clinical but attempt to of lung and inflammatory