The impact of kangaroo mother care on work of breathing and oxygen saturation in very low birth weight infants with respiratory insufficiency

BACKGROUND: Kangaroo mother care (KMC) is defined as prolonged skin to skin care between a mother and infant with the infant lying in prone position on mom’s chest. KMC decreases morbidity and mortality and promotes physiologic stability. The aim of this study is to measure work of breathing (WOB) during KMC in very low birth weight (VLBW) infants on non-invasive respiratory support. METHODS: A prospective observational pilot study was conducted comparing WOB indices during standard care (SC) and KMC. Respiratory inductive plethysmography (RIP) measured WOB indices non-invasively: phase angle and labored breathing index. VLBW infants who were stable on non-invasive respiratory support were randomized to receive RIP measurements during KMC or during SC first. Summary statistics and mixed linear models were used to compare WOB and vital signs. RESULTS: A total of 32 infants were consented for the study, data collection and analysis was completed on 28 infants. There were no significant differences in mean phase angle during KMC or SC (73.5±4.6 SE deg vs 66.8±3.9 SE deg, p = 0.25). No differences in WOB and vital signs were detected. Controlling for respiratory support or randomization/first location did not change the results. CONCLUSION: In this pilot cohort, infants demonstrated no differences in work of breathing indices or oxygen saturation during KMC or SC while receiving non-invasive respiratory support. KMC appears to be safe and well tolerated with no worsened WOB. Larger studies should be performed to confirm our findings.

Effects of xenon gas on human airway epithelial cells during hyperoxia and hypothermia

BACKGROUND

Hypothermia with xenon gas has been used to reduce brain injury and disability rate after perinatal hypoxia-ischemia. We evaluated xenon gas therapy effects in an in vitro model with or without hypothermia on cultured human airway epithelial cells (Calu-3).

METHODS

Calu-3 monolayers were grown at an air-liquid interface and exposed to one of the following conditions: 1) 21% FiO2 at 37°C (control); 2) 45% FiO2 and 50% xenon at 37°C; 3) 21% FiO2 and 50% xenon at 32°C; 4) 45% FiO2 and 50% xenon at 32°C for 24 hours. Transepithelial resistance (TER) measurements were performed and apical surface fluids were collected and assayed for total protein, IL-6, and IL-8. Three monolayers were used for immunofluorescence localization of zonula occludens-1 (ZO-1). The data were analyzed by one-way ANOVA.

RESULTS

TER decreased at 24 hours in all treatment groups. Xenon with hyperoxia and hypothermia resulted in greatest decrease in TER compared with other groups. Immunofluorescence localization of ZO-1 (XY) showed reduced density of ZO-1 rings and incomplete ring-like staining in the 45% FiO2- 50% xenon group at 32°C compared with other groups. Secretion of total protein was not different among groups. Secretion of IL-6 in 21% FiO2 with xenon group at 32°C was less than that of the control group. The secretion of IL-8 in 45% FiO2 with xenon at 32°C was greater than that of other groups.

CONCLUSION

Hyperoxia and hypothermia result in detrimental epithelial cell function and inflammation over 24-hour exposure. Xenon gas did not affect cell function or reduce inflammation.

High flow nasal heliox improves work of breathing and attenuates lung injury in a newborn porcine lung injury model

BACKGROUND

High flow nasal cannula (HFNC) has been shown to improve ventilation and oxygenation and reduce work of breathing in newborns with respiratory distress. Heliox, decreases resistance to airflow, reduces the work of breathing, facilitates the distribution of inspired gas, and has been shown to attenuate lung inflammation during the treatment of acute lung injury.

HYPOTHESIS

Heliox delivered by HFNC will decrease resistive load, decrease work of breathing, improve ventilation and attenuate lung inflammation during spontaneous breathing following acute lung injury in the newborn pig.

METHODS

Spontaneously breathing neonatal pigs received Nitrox or Heliox by HFNC and studied over 4 hrs following oleic acid injury. Gas exchange, pulmonary mechanics and systemic inflammation were measured serially. Lung inflammation biomarkers were assessed at termination.

RESULTS

Heliox breathing animals demonstrated lower work of breathing reflected by lower tracheal pressure, phase angle and phase relationship. Ventilation efficiency index was greater compared to Nitrox. Heliox group showed less lung inflammation reflected by lower tissue interleukin-6 and 8.

CONCLUSION

High flow nasal Heliox decreased respiratory load, reduced resistive work of breathing indices and attenuated lung inflammatory profile while ventilation was supported at less pressure effort in the presence of acute lung injury.

Feasibility of lung cancer hyperthermia using breathable perfluorochemical (PFC) liquids. Part I: Convective hyperthermia

Clinical studies have shown that hyperthermia in combination with radiotherapy and/or chemotherapy may be effective in the treatment of advanced cancer. No method of lung hyperthermia, however, has been accepted as standard or superior. This investigation sought to demonstrate in animals the thermal and physiologic feasibility of lung hyperthermia induced using heated breathable perfluorochemical (PFC) liquids, a method termed liquid-filled lung convective hyperthermia (LCHT). The ability to use LCHT is rooted in the development of both PFC liquid ventilation, now in clinical development with the PFC perflubron (LiquiVent), and a PFC blood substitute also in late Phase III trials (Oxygent). As LCHT background, the PFC technologies and biology are first reviewed. The physical properties of a variety of PFCs were evaluated for LCHT and it was concluded that more than one liquid is suitable based on such properties. Using total liquid ventilation type devices, LCHT was shown to deliver successfully localized (lobar) lung heating in sheep, and bilateral whole lung heating and whole-body hyperthermia in rabbits, cats and lambs. During LCHT, lung parenchymal temperatures were uniform (<1 degree C) across heated regions. In addition, based on patterns relating lung tissue temperatures to inspiratory and expiratory PFC liquid temperatures in the endotracheal tube, LCHT may minimize invasive thermometry requirements in the lung. Based on acute experiments, it was concluded that LCHT appears feasible and may simplify lung hyperthermia. It was recommended that potentially synergistic combinations of LCHT with other whole-body hyperthermia or local heating modalities, and with chemotherapeutic lung drug delivery, also be explored in the future.

Feasibility of lung cancer hyperthermia using breathable perfluorochemical (PFC) liquids. Part II: Ultrasound hyperthermia

Enhanced local control of disease in lung cancer has been shown to improve survival, and controlled clinical trials of hyperthermia adjunctive to radiotherapy in other cancers have shown improved disease control and survival over radiotherapy alone. The challenge of lung hyperthermia, however, persists. This investigation sought to demonstrate the feasibility of localized lung hyperthermia at depth via therapeutic ultrasound. The method is based on using breathable perfluorochemical liquids as acoustic coupling media in the lung, liquids that have also been shown to enable controlled liquid-filled lung convective hyperthermia (LCHT). The ability to use both lung convective hyperthermia and liquid-filled lung ultrasound hyperthermia (LUHT) provides potential flexibility in heating patterns for the hyperthermic treatment of lung cancer with concurrent radiotherapy and/or chemotherapy. Using custom ultrasound transducers designed and built for these studies, the acoustic properties of three candidate perfluorochemicals were characterized over a range of temperatures, gas contents and ultrasound frequencies and acoustic intensities. Both sound speed and attenuation were measured in the neat liquids and in isolated lungs filled with the perfluorochemicals. Successful ultrasound hyperthermia at depth was demonstrated in vivo in sheep lung lobes in intraoperative conditions. In addition, the use of ultrasound diagnostic imaging was explored as a tool for use in conjunction with lung ultrasound hyperthermia.

Cold perfluorochemical-induced hypothermia protects lung integrity in normal rabbits

To test the hypothesis that intrapulmonary perfluorochemical (PFC) liquid may induce hypothermia, and to compare the effects of internal (IC), external (EC), and combined cooling techniques (EC + IC), 14 juvenile rabbits were randomized to EC by a cold blanket (4 degrees C, n = 5), IC by intrapulmonary cold PFC liquid lavage (4 degrees C, n = 5), or combined IC with PFC and EC (n = 4). Arterial blood gas, blood pressure, and lung mechanics were monitored, and lung histology was examined by light microscopy. The results showed that cooling rates and the time needed to be cooled down to 30 degrees C were significantly faster in EC and EC + IC than IC (p < 0.05). Blood gas analysis and cardiopulmonary function were within the normal range in all groups. Histological assessment revealed varied atelectasis in all lung regions in EC, whereas PFC-filled lungs (IC and EC + IC) demonstrated more homogenous expansion and no evidence of atelectasis. The results indicate that intrapulmonary PFC may be an effective technique to induce and/or augment hypothermia while supporting gas exchange, lung volume and pulmonary architecture.

Positive end-expiratory pressure modulates perfluorochemical evaporation from the lungs

To study the effects of positive end-expiratory pressure (PEEP) level on perfluorochemical (PFC) elimination profiles (E(L)), 6 ml/kg of perflubron were instilled into healthy anesthetized rabbits. The ventilation strategy was to maintain constant minute ventilation (300 ml/kg/min) and mean airway pressure (7-8 cm H(2)O) while randomly changing the PEEP levels from 5 to 0, 1, 3, and 10 cm H(2)O, each for a period of 15 min. The PFC content in the expired gas was measured and the E(L) was calculated. There was a significant reduction in the E(L) when decreasing the PEEP levels from 5 to 0 cm H(2)O, but no differences were seen when the PEEP was increased from 5 to 10 cm H(2)O. The results indicate that PEEP levels influence PFC elimination profiles; therefore, the measurement of the E(L) and PEEP levels should be considered when optimizing supplemental PFCs during partial liquid ventilation.

Temperature and heater responses during transition between radiant and incubator thermal environment in newborn preterm lambs

OBJECTIVE

Although both incubators and radiant warmer beds can provide thermal support to infants in the neonatal intensive care unit, the transition between devices can be a stressful event. The goal of this study was to evaluate a new device that combines these methods of warming and converts between them without requiring physical movement of the infant.

STUDY DESIGN

Twin preterm lambs received thermal support from a radiant warmer bed and an incubator (control), or from the Versalet 7700 Care Center (treatment) in the warmer and incubator configurations. Temperature of each lamb, as well as device heater power, were monitored every 2 min before, during, and after the transition. Physiological parameters were monitored every 15 min.

RESULTS

There was a significant difference in response time between closed-open and open-closed conditions for both groups. More adverse events occurred in the control group during transfers. There were no differences in temperatures or physiological variables during transitions in either group.

CONCLUSIONS

These data suggest the Versalet provides similar thermal stability to traditional devices, with fewer adverse events associated with the lack of physical movement between warming configurations. The impact of this device on the care of the preterm neonate will be evaluated in a clinical trial.

Perfluorochemical liquids do not stimulate endothelin-1 or nitric oxide production in human blood leukocytes

The purpose of these studies was to examine if perfluorochemical (PFC) liquids stimulate blood leukocytes to secrete nitric oxide (NO) and/or endothelin-1 (ET-1). As such, NO and ET-1 may modulate broncho- and vascular dilatation and constriction, respectively, and thereby influence the clinical condition of a patient in respiratory distress with persistent pulmonary hypertension. Blood leukocytes in their natural habitat (whole blood) were incubated in the presence of two different perfluorochemicals (perflubron and perfluorodecalin). The overall response in ET-1 or NO (indirectly measured as nitrite/nitrate) production was examined at increasing PFC percentages (wt/vol) of PFC/whole blood. The lowest proportion used, 0.001% (wt/vol), was relevant to serum concentrations of PFC observed in liquid-ventilated individuals, whereas the highest proportion PFC, 50% (wt/vol), would mimic a situation where leukocytes are presented to PFC-filled airways. Plasma levels of freshly drawn blood, similar to levels of incubated (6 h) non-PFC-supplemented cultures, were ET-1 0.59 +/- 0.07 pg/ml (6 h, mean +/- SEM) and NO(-2)/NO(-3) 50 +/- 9 microM (6 h). Perflubron or perfluorodecalin did not induce significant differences in ET-1 or NO(-2)/NO(-3) levels as function of PFC type or dose. In conclusion, PFC liquids do not stimulate production in leukocytes in vitro of substances that may modulate constriction or dilatation in the vascular and respiratory tract systems.

Perfluorochemical liquids modulate cell-mediated inflammatory responses

OBJECTIVE

To examine whether chemically different perfluorochemical liquids (PFC) (perfluorodecalin [PFD]; perflubron [PFB]) induce inflammatory responses in blood leukocytes.

SETTING

University research laboratory.

DESIGN

Whole blood from 12 healthy adults was incubated with increasing PFC concentrations and/or bacterial lipopolysaccharide.

MEASUREMENTS AND MAIN RESULTS

Adhesion molecules (CD62L, CD11b), reactive oxygen species, and cytokine responses in resting and activated leukocyte subtypes were studied. Scanning and transmission electron microscopies were performed. At the highest concentrations, PFB stimulated a significant increase in resting monocytic reactive oxygen species production; all types of blood leukocytes were unresponsive to PFD. Neither PFB nor PFD changed CD62L expression; PFB increased CD11b expression in monocytes and granulocytes. PFD induced a small though significant increase in interleukin-8 secretion. When simulating a condition in which patients with severe lung disease or sepsis would be ventilated with PFC, neither PFB nor PFD plus lipopolysaccharide stimulated tumor necrosis-alpha or interleukin-8 production above levels induced by lipopolysaccharide alone, but rather demonstrated a trend for decreased tumor necrosis factor-alpha production. Expression of CD11b and CD62L and the production of reactive oxygen species were not changed beyond the levels induced by lipopolysaccharide alone. As a morphologic correlate to the above proinflammatory changes, surface-bound blebs and intracellular vacuoles were seen by electron microscopy.

CONCLUSIONS

At PFC concentrations comparable with those in blood during liquid ventilation, PFC liquids did not induce variables associated with inflammation. In the presence of high PFC concentrations, simulating the condition in which bronchoalveolar cells are exposed to PFC, monocytes may be induced by PFB to produce reactive oxygen species, and blood leukocytes induced by PFB to express CD11b and by PFD to secrete interleukin-8; the presence of either PFC attenuated tumor necrosis factor-alpha production after lipopolysaccharide stimulation.

Intratracheal administration of perfluorochemical-gentamicin suspension: a comparison to intravenous administration in normal and injured lungs

Respiratory infections can lead to acute lung injury and perfusion abnormalities. We hypothesized that intratracheal (IT) administration of a perfluorochemical (PFC) gentamicin (G) suspension as compared to intravenous (IV) administration of gentamicin will result in higher lung tissue levels of gentamicin, while maintaining safe serum levels. To test this hypothesis, 21 lambs with normal and acid injured lungs were studied for 4 hr, using 2 different drug delivery methods, IT and IV. Lungs were injured with warm HCl acid in saline lavage, followed by partial liquid ventilation with perflubron (bolus FRC = 20 mL/kg). G at a dose of 5 mg/kg was delivered either IT (G-PFC; 20 mL/kg) or IV (aqueous injection with IT 20 mL/kg PFC alone). Throughout the study, serum G levels, arterial blood gases, respiratory system compliance, and mean arterial blood pressure were measured. Lung tissue G levels were measured at 4 hr and averaged across lobes. Physiologic gas exchange and pulmonary function were maintained throughout the protocol for both the normal and injured lungs. Intravenously administered G resulted in an initial 5-min serum concentration of 43 +/- 2.5 mcg/mL, followed by an exponential decline over the 4-hr protocol to a level of 2.1 +/- 0.23 mcg/mL at hr 4. The intratracheally administered G suspension resulted in a 5-min serum concentration of 1.8 +/- 0.98 mcg/mL and remained relatively constant throughout the protocol, with a 4-hr level of 1.6 +/- 0.29 mcg/mL. With respect to lung tissue G levels, IT administration was significantly more effective in delivering the drug to the normal lungs than IV (31.4 +/- 3.3 mcg/g vs. 4.0 +/- 0.7 mcg/g) 4 hr after administration. In the lung injury group, there was a small but significant difference in lung tissue G levels, with the IT-administered perfluorochemical-G suspension achieving greater levels than the IV-administered G (11.9 +/- 0.52 mcg/g vs. 10.1 +/- 0.8 mcg/g). Additionally, the drug delivered IV and IT in both the normal and injured lung models was homogeneously distributed throughout the lung. These data show that G lung tissue levels in both normal and injured lungs were higher in the IT group when compared to IV administration. The results of this study demonstrate that in normal and injured lungs, homogeneous G lung tissue levels can be more effectively achieved at lower serum levels when delivered IT in a G-PFC suspension as compared to IV administration.

A comparison of surfactant delivery with conventional mechanical ventilation and partial liquid ventilation in meconium aspiration injury

The objective of this study was to compare surfactant (SF) distribution and physiological effects after standard SF delivery during conventional mechanical ventilation (CMV) with that using partial liquid ventilation (PLV). A model of meconium aspiration syndrome (MAS) was developed using two groups of adult rats (n = 14). After meconium instillation of 2.5 ml kg(-1) (20% v/w), SF/CMV: (n = 7) CMV and SF/PLV: (n = 7) PLV, received 14C-labeled surfactant (4 ml kg(-1)) delivered intratracheally in four aliquots over 20 min in both groups. Sequential measurements of arterial blood chemistry and lung mechanics were performed in all animals. At the conclusion of experiments, lungs were inflated (30 cmH2O), dried, sectioned and evaluated for radioactivity in disintegrations per minute (DPM). Surfactant distribution was improved (P< 0.01) with PLV as compared to CMV with 48.8% of the pieces vs. 30.9% of the pieces receiving within 25% of the mean amount of surfactant, respectively. Further, regional distribution was also significantly more uniform with PLV than CMV: left vs right (P<0.01) lung and ventral vs. dorsal (P<0.01) regions. Finally, arterial PO2 and ventilation efficiency index were significantly (P<0.01) greater post-treatment in SF/PLV than SF/CMV. These data demonstrate surfactant delivery with PLV, as compared to CMV alone, to be an improved method of delivering surfactant in MAS and suggest the possible utility of SF/PLV combination therapy for its treatment of other etiologies of neonatal respiratory distress.

Thermal stability and transition studies with a hybrid warming device for neonates

OBJECTIVE

The use of both warmer beds and incubators is common in neonatal intensive care units (NICU), and transferring between these two warming devices is a routine and necessary event. This study was designed to evaluate the efficacy of a new hybrid-warming device, the Versalet, in transitioning a preterm animal from a warmer bed to an incubator mode and back.

STUDY DESIGN

Nine premature lambs were randomized, following delivery, to receive thermal support from a conventional warming bed and an incubator (control group), or from the Versalet (study group) in the warmer bed and incubator modes. Core and various surface temperatures, as well as physiological parameters were measured first during warming in the radiant warmer bed mode, Versalet or Resuscitaire and then during transition to the incubator mode, Versalet or Isolette, and then back to the warmer bed mode.

RESULTS

The animals remained stable during all the transitions. Despite careful planning, adverse events occurred in the control group during transfers. There were no significant differences in the temperature or physiologic profiles during any of the transitions in either group.

CONCLUSION

Compared with the standard warming technique used in NICUs (separate warmer bed and incubator), the Versalet provides similar thermal and cardiovascular stability without adverse events during transition to different modes of warming. The degree to which this device would contribute to ease of management and improved outcomes in humans needs to be evaluated in a clinical trial.

Role of ventilation strategy on perfluorochemical evaporation from the lungs

To study the effect of ventilation strategy on perfluorochemical (PFC) elimination profile (evaporative loss profile; E(L)), 6 ml/kg of perflubron were instilled into anesthetized normal rabbits. The strategy was to maintain minute ventilation (VE, in ml/min) in three groups: VE(L) (low-range VE, 208 +/- 2), VE(M) (midrange VE, 250 +/- 9), and VE(H) (high-range VE, 293 +/- 1) over 4 h. In three other groups, respiratory rate (RR, breaths/min) was controlled at 20, 30, or 50 with a constant VE and adjusted tidal volume. PFC content in the expired gas was measured, and E(L) was calculated. There was a significant VE- and time-dependent effect on E(L.) Initially, percent PFC saturation and loss rate decreased in the VE(H) > VE(M) > VE(L) groups, but by 3 h the lower percent PFC saturation resulted in a loss rate such that VE(H) < VE(M) < VE(L) at 4 h. For the groups at constant VE, there was a significant time effect on E(L) but no RR effect. In conclusion, E(L) profile is dependent on VE with little effect of the RR-tidal volume combination. Thus measurement of E(L) and VE should be considered for the replacement dosing schemes during partial liquid ventilation.

Effects of perfluorochemical distribution and elimination dynamics on cardiopulmonary function

Based on a physicochemical property profile, we tested the hypothesis that different perfluorochemical (PFC) liquids may have distinct effects on intrapulmonary PFC distribution, lung function, and PFC elimination kinetics during partial liquid ventilation (PLV). Young rabbits were studied in five groups [healthy, PLV with perflubron (PFB) or with perfluorodecalin (DEC); saline lavage injury and conventional mechanical ventilation (CMV); saline lavage injury PLV with PFB or with DEC]. Arterial blood chemistry, respiratory compliance (Cr), quantitative computed tomography of PFC distribution, and PFC loss rate were assessed for 4 h. Initial distribution of PFB was more homogenous than that of DEC; over time, PFB redistributed to dependent regions whereas DEC distribution was relatively constant. PFC loss rate decreased over time in all groups, was higher with DEC than PFB, and was lower with injury. In healthy animals, arterial PO(2) (Pa(O(2))) and Cr decreased with either PFC; the decrease was greater and sustained with DEC. Lavaged animals treated with either PFC demonstrated increased Pa(O(2)), which was sustained with PFB but deteriorated with DEC. Lavaged animals treated with PFB demonstrated increased Cr, higher Pa(O(2)), and lower arterial PCO(2) than with CMV or PLV with DEC. The results indicate that 1) initial distribution and subsequent intrapulmonary redistribution of PFC are related to PFC properties; 2) PFC distribution influences PFC elimination, gas exchange, and Cr; and 3) PFC elimination, gas exchange, and Cr are influenced by PFC properties and lung condition.

Meconium aspiration injury: Uncoupling between the in vivo physiologic and in vitro inflammatory responses

OBJECTIVE: To correlate the in vivo physiologic changes that occur with meconium aspiration injury to an associated in vitro cellular response to meconium. DESIGN: Experimental, prospective, randomized, controlled study. SETTING: University research laboratory. SUBJECTS: Eighteen adult Sprague-Dawley rats with meconium aspiration injury. INTERVENTIONS: Rats were given 3 mL/kg of a 25% meconium solution and were treated with conventional gas ventilation; nine rats were given exogenous surfactant therapy (Survanta, 4 mL/kg), and nine rats were not treated (control). Bronchoalveolar lavages were collected for total cell counts. Histologic samples also were taken for analysis. In addition, the in vitro effect of meconium on granulocytic elastase release from human neutrophils was determined. MEASUREMENTS AND MAIN RESULTS: Meconium caused significant morbidity in vivo, including poor oxygenation, elevated Paco(2), diminished compliance, and elevated white cell count in the bronchial lavages. Lung white cell count was significantly less in the surfactant group (p <.01). Meconium did not cause elastase release from human neutrophils in vitro. CONCLUSIONS: This study demonstrated uncoupling between in vivo physiologic responses to meconium injury in rats and the in vitro effect of meconium on human neutrophils. Surfactant therapy alleviated some of the perturbations associated with meconium injury, including a reduction in the inflammatory cell count in lung lavages. The absence of direct neutrophil activation by meconium suggests the requirement of an intermediary in the pathogenesis of meconium aspiration injury.

Liquid ventilation

Liquid breathing has been proposed as a means of improving gas exchange in infants with acute respiratory failure since the 1970s. In addition, there are potential clinical applications of perfluorochemical (PFC) liquids that span many specialties in medicine. The ability to lower surface tension directed the initial clinical focus on neonatal therapy in the treatment of premature lung disease. The first clinical trial of PFC ventilation was performed in neonates in 1989. Additional trials using LiquiVent (Alliance Pharmaceutical San Diego, CA), a medical grade PFC liquid, were initiated in 1993 in infants, children, and adults. These studies have concluded that liquid ventilation appeared to be safe, improve lung function, and recruit lung volume in patients from these populations. The results of such trials are encouraging, but randomized trials have yet to be completed. We await these pivotal trials, which will probably be completed in adult patients first, before this promising technique can be clinically available.

Perfluorochemical elimination from the lungs: effect of initial dose

Liquid-assisted ventilation with perfluorochemical (PFC) has been beneficial in a variety of respiratory diseases in animals and humans. Although PFC evaporation from the lungs is in part dependent on ventilation strategy and positioning, guidelines for initial and replacement dosing are unclear. We hypothesized that PFC evaporative loss over time is dependent on the size of the initial dose. Juvenile rabbits (n = 18) were ventilated using constant animal position and ventilator strategy. PFC (perflubron: LiquiVent ) was instilled endotracheally, using four groups with initial doses of 2, 6, 12, and 17 mL/kg. A previously described thermal detector that measures PFC in expired gas was used to calculate loss rate, residual perflubron in the lung, and volume loss as a % of initial fill volume. There was a significant dose, time, and dose-time interaction such that evaporative loss was dependent on initial PFC volume and time after fill (P < 0.05). Evaporative loss rate decreased earlier at lower doses. The percentage of initial volume lost to evaporation over time was inversely related to dose and could not be predicted by decreasing % PFC saturations, independent of dose. Evaporative loss should be considered to optimize both the application of PFC to the lung and replacement dosing during partial liquid ventilation.

A perfluorochemical loss/restoration (L/R) system for tidal liquid ventilation

Tidal liquid ventilation is the transport of dissolved respiratory gases via volume exchange of perfluorochemical (PFC) liquid to and from the PFC-filled lung. All gas-liquid surface tension is eliminated, increasing compliance and providing lung protection due to lower inflation pressures. Tidal liquid ventilation is achieved by cycling fluid from a reservoir to and from the lung by a ventilator. Current approaches are microprocessor-based with feedback control. During inspiration, warmed oxygenated PFC liquid is pumped from a fluid reservoir/gas exchanger into the lung. PFC fluid is conserved by condensing (60-80% efficiency) vapor in the expired gas. A feedback-control system was developed to automatically replace PFC lost due to condenser inefficiency. This loss/restoration (L/R) system consists of a PFC-vapor thermal detector (+/- 2.5%), pneumatics, amplifiers, a gas flow detector (+/- 1%), a PFC pump (+/- 5%), and a controller. Gravimetric studies of perflubron loss from a flask due to evaporation were compared with experimental L/R results and found to be within +/- 1.4%. In addition, when L/R studies were conducted with a previously reported liquid ventilation system over a four-hour period, the L/R system maintained system perflubron volume to within +/- 1% of prime volume and 11.5% of replacement volume, and the difference between experimental PFC loss and that of the L/R system was 1.8 mL/hr. These studies suggest that the PFC L/R system may have significant economic (appropriate dosing for PFC loss) as well as physiologic (maintenance of PFC inventory in the lungs and liquid ventilator) impact on liquid ventilation procedures.

Effects of partial liquid ventilation with perfluorodecalin in the juvenile rabbit lung after saline injury

OBJECTIVE

To evaluate the feasibility of using the perfluorochemical, perfluorodecalin, for partial liquid ventilation (PLV) with respect to gas exchange and lung mechanics in normal and saline-injured lungs of juvenile rabbits.

DESIGN

Experimental, prospective, randomized, controlled study.

SETTING

Physiology laboratory at a university medical school.

SUBJECTS

Seventeen juvenile rabbits assigned to three groups.

INTERVENTIONS

The conventional mechanical ventilation (CMV)-injury group (n = 5) was treated with CMV after establishing a lung injury; the PLV-injury group (n = 6) was treated with PLV after lung injury; and the PLV-healthy group (n = 6) was supported with PLV without lung injury. Lung injury was created by repeated saline lung lavages. PLV-treated animals received a single dose of intratracheal perfluorodecalin at a volume equal to the measured preinjury gas functional residual capacity (functional residual capacity = 18.6+/-1.5 [SEM] mL/kg).

MEASUREMENTS AND MAIN RESULTS

Sequential measurements of total respiratory compliance and arterial blood chemistries were performed in all groups. Oxygenation index (OI) and ventilation efficiency index were calculated. After lung injury, there was a significant (p < .05) decrease in PaO2, total respiratory compliance, and ventilation efficiency index and an increase in OI and PaCO2. In the PLV-injury group, PLV significantly (p < .05) improved PaO2 (+60%) and OI (-33%) over time. Compliance was significantly (p < .05) higher (90%) than in the CMV-injury group over time.

CONCLUSIONS

These results demonstrate that PLV with perfluorodecalin improved oxygenation and increased respiratory compliance in the saline-injured rabbit lung. In addition, similar to the effects of several other perfluorochemical liquids on normal lungs, pulmonary administration of perfluorodecalin was associated with a small impairment in gas exchange and a significant decrease in lung compliance in the juvenile rabbit model.

Liquid assisted ventilation update

The purpose of this summary is to highlight several areas of recent research performed in liquid assisted ventilation techniques. These areas were chosen specifically based on their contribution to elucidate mechanisms of improved pulmonary support or provoke additional and novel applications of this modality to reduce pulmonary complications of immaturity or acute lung injury.

Intra-tracheal delivery strategy of gentamicin with partial liquid ventilation

Patients with pulmonary infection often present with ventilation and perfusion abnormalities, which can impair intravenous antibiotic therapy. Intra-tracheal (i.t.) administration has met with obstacles, such as inadequate delivery to affected lung regions and the disruption of gas exchange. We hypothesized that i.t. administration of a gentamicin (G)/perfluorochemical (PFC) suspension (G/PFC) would effectively deliver and distribute gentamicin to the lung, while maintaining gas exchange and non-toxic serum levels. In addition, we sought to compare serum G and lung levels and distribution of G when G/PFC is administered at the initiation of partial liquid ventilation (PLV) vs. during PLV. To test this hypothesis, 17 newborn lambs were ventilated by PLV with perflubron (LiquiVent) for 4 h using three different G (5 mg kg-1) administration techniques: i.t. slow-fill (SF) (n = 6; G/PFC over 15 min at start of PLV), i.t. top-fill (TF) (n = 6; G/PFC 10-65 min after start of PLV), intravenous (i.v.) (n = 5, aqueous injection at start of PLV). Serum levels of gentamicin were obtained 1, 15, 30 and 60 min after administration, and hourly there after for the remainder of the protocol (4 h). Arterial blood gas and pulmonary function measurements were obtained throughout the protocol. At the conclusion of the protocol, representative samples from each lung lobe, the brain and kidney were homogenized and assayed for gentamicin. All results are presented as the mean +/- SEM; P < 0.05. Over time, serum gentamicin levels were greatest (P < 0.05) in i.v. (11.0 +/- 2.3 micrograms ml-1), followed by TF (2.3 +/- 0.1 micrograms ml-1) and SF (0.8 +/- 0.1 microgram ml-1). The percentage of the administered dose remaining in the lungs after 4 h was greater (P < 0.05) following i.t. delivery (SF 23.8 +/- 4.3%, TF 13.7 +/- 2.5%) as compared to i.v. (3.7 +/- 0.5%). These findings suggest that for a given dose of G, both SF and TF delivery methods of G/PFC can enhance pulmonary, relative to systemic, antibiotic coverage.

On-line techniques for perfluorochemical vapor sampling and measurement

The authors developed a compact gas sampling and perfluorochemical (PFC) measuring system for use in total and partial liquid ventilation systems, based on a precision two-thermistor thermal detector (TD). They describe the sensitivity and linearity of their on-line method for PFC analysis of expired gases and show how it may be used in partial liquid ventilation studies for determining PFC saturation and loss. Gas is sampled for a short time from a breathing circuit through a heated tube at a selectable point in the breathing cycle. Inspiration is sensed by a pressure transducer. The sample of gas is pulled into the heated (48 degrees C) thermistor chamber by suction and held there while the cooling effect of the vapor changes the thermistor temperature. Dry air in another chamber affects a second thermistor, and the difference of these responses is amplified. The raw signal is corrected for the effects of varying O2 levels by a fuel cell. This signal is sampled and held and displayed on a front panel display. Calibration is performed in percentage saturation at 37 degrees C using the PFC in use at that temperature, or another standard such as O2. In-vitro testing showed a linear response in the thermal detector device (R2 = 0.99) over the range of vapor pressures tested (0-14) mmHg) and was reproducible to within 3%. When electronically corrected for changes in O2 concentration, there was less than a 2% change in PFC saturation. The TD responses to CO2 (R2 = 0.99) and water vapor (R2 = 1.0) were linear and approximately equal and opposite over the normal operating ranges of expired gases. In-vivo results in rabbits showed a significant (R2 = 0.73; p < 0.01) correlation between the auto-sampler and manual collection modes for determination of PFC in expired gas.

Effect of single versus multiple dosing on perfluorochemical distribution and elimination during partial liquid ventilation

The objective of this study was to quantitate perfluorochemical (PFC) elimination kinetics during partial liquid ventilation (PLV) following an initial fill with or without hourly dosing. Young New Zealand rabbits were studied in two groups: Gr I (n = 6), PLV with a single dose of PFC liquid (perflubron: LiquiVent, Alliance Pharmaceutical Corp.); and Gr II (n = 5), PLV with PFC liquid and multiple hourly dosing . All rabbits were studied for 4 h, following initial instillation of a volume of PFC liquid equal to the measured gas functional residual capacity. Animals were ventilated at a constant breathing frequency (30 br/min), tidal volume (9.3+/-0.3 SE mL/kg), positive end expiratory pressure (4 cm H2O), and inspiratory time (0.30 s). PFC saturation of mixed expired gas (PFC-Sat) was assessed with a thermal conductivity analyzer, and PFC elimination was calculated from PFC-Sat, minute ventilation, and temperature of the expired gas. In GR II, PFC was supplemented hourly at a volume determined by PFC elimination calculations. The results demonstrated a decrease in PFC-sat and PFC loss with time, independent of group (P< 0.05). In addition, with hourly supplementation (GR II), PFC-Sat and PFC elimination over time was significantly (P < 0.05) greater than in animals (GR I) which did not receive additional doses. These data demonstrate that the PFC elimination rate is not constant and is related to the amount of PFC in the respiratory system. This may have occurred due to distributional differences of ventilation and PFC liquid between the single and multiple dosing groups. These findings also suggest that evaluation of PFC concentrations in expired gas may be a clinically useful index of intrapulmonary PFC distribution during PLV, and that maintained elevation of expired gas PFC saturation may guide optimal PFC dosing intervals and distribution to maximize protection against barotrauma.

Software for real-time control of a tidal liquid ventilator

The purpose of this project was to develop and test computer software and control algorithms designed to operate a tidal liquid ventilator. The tests were executed on a 90-MHz Pentium PC with 16 MB RAM and a prototype liquid ventilator. The software was designed using Microsoft Visual C++ (Ver. 5.0) and the Microsoft Foundation Classes. It uses a graphic user interface, is multithreaded, runs in real time, and has a built-in simulator that facilitates user education in liquid-ventilation principles. The operator can use the software to specify ventilation parameters such as the frequency of ventilation, the tidal volume, and the inspiratory-expiratory time ratio. Commands are implemented via control of the pump speed and by setting the position of two two-way solenoid-controlled valves. Data for use in monitoring and control are gathered by analog-to-digital conversion. Control strategies are implemented to maintain lung volumes and airway pressures within desired ranges, according to limits set by the operator. Also, the software allows the operator to define the shape of the flow pulse during inspiration and expiration, and to optimize perfluorochemical liquid transfer while minimizing airway pressures and maintaining the desired tidal volume. The operator can stop flow during inspiration and expiration to measure alveolar pressures. At the end of expiration, the software stores all user commands and 30 ventilation parameters into an Excel spreadsheet for later review and analysis. Use of these software and control algorithms affords user-friendly operation of a tidal liquid ventilator while providing precise control of ventilation parameters.

Liquid ventilation: clinical experiences

Liquid breathing has been in the medical literature for nearly 80 years and has been proposed as a means of improving gas exchange in patients with acute respiratory failure since the 1970s. There are many potential clinical applications of perfluorochemical (PFC) liquids that span many specialties in medicine. The ability to lower surface tension directed the initial clinical focus on neonatal therapy in the treatment of respiratory distress syndrome. The first clinical trial of PFC ventilation was performed in neonates in 1989. Additional trials using LiquiVent, a medical-grade PFC liquid, were initiated in 1993 in infants, children, and adults with severe respiratory distress. Based on the results of several studies during this time, it was concluded that the technique of liquid ventilation seemed to be safe, improve lung function, and recruit lung volume in patients from these various populations. More than 100 patients from preterm neonates to elderly patients with respiratory distress have been studied. In general, patients experience improvement in lung function and oxygenation, without adverse events. The results of such trials are encouraging and suggest the feasibility of this technique in the neonate with severe respiratory failure requiring respiratory support. Limitations of the technique include a limited understanding of how it works and its effectiveness in different patient populations. It has also become evident that there is a need for additional diagnostic and therapeutic instrumentation to expedite its implementation in the intensive care environment. This article discusses clinical experiences with liquid ventilation and highlights the technical needs to facilitate its implementation in respiratory care management.

Multifactorial analysis of exchanger efficiency and liquid conservation during perfluorochemical liquid-assisted ventilation

Liquid-assisted ventilation (LAV) of the lung with perfluorochemical (PFC) requires a method of oxygenating and removing CO2 from the liquid. Current PFC LAV techniques consist of total liquid ventilation, PFC lavage, and partial liquid ventilation. Because PFC liquid is volatile, it may be lost from the lung or ventilator circuit in the expired gas. This study evaluated the efficiencies of two types of exchangers (spray bubbler and membrane oxygenator) with respect to CO2 elimination from the PFC liquid and prevention of the loss of PFC liquid. A multifactorial analysis of exchanger efficiency was performed with respect to liquid conservation and CO2 removal. PFC losses and relative efficiencies of two types of exchangers to eliminate CO2 from expired PFC liquid were evaluated, along with two types of PFC liquids. Gas (100% O2 at 4 and 8 L/min) and PFC liquid were circulated countercurrently through the exchangers (oxygenator and bubbler) through a temperature-controlled (25 degrees or 37 degrees C) open circuit. To evaluate effectiveness of CO2 elimination, an exchanger efficiency index (EEI) for CO2 was calculated applying mass-transfer theory to characterize gas transport down a concentration gradient where EEI equals: [PPFCCO2 out--PPFCCO2 in]/PgasCO2 in--PPFCCO2 in]. Rate of PFC loss from the circuit was calculated from mixed expired gas samples using a thermal detector analyzer. EEI and PFC loss rate were analyzed with respect to gas: PFC liquid flow ratios (analogous to the V/Q ratio). The results showed that 1) PFC loss rate and exchanger efficiency to remove CO2 increased with increasing gas: PFC liquid flow rates independent of the type of exchanger or PFC liquid; 2) PFC loss rate at any gas or liquid flow rate was greater for the bubbler than for the oxygenator; 3) the oxygenator was more efficient than the bubbler with respect to CO2 elimination; 4) although PFC loss rate increased with temperature and vapor pressure, there was little difference in the EEIs for the temperatures studied. These results 1) identify exchanger requirements necessary to maintain effective CO2 elimination up to four times normal CO2 loading conditions during LAV; 2) suggest that using a membrane oxygenator as the gas exchanger, in concert with stringent fluid temperature control, improves PFC liquid conservation and CO2 elimination relative to bubbler exchanger configurations; 3) highlight the importance of matching the exchanger type to the physiocochemical properties of the specific PFC liquid to optimize CO2 elimination while reducing PFC liquid loss by minimizing gas relative to PFC liquid flow rates. Because PFC liquid loss occurs with all current means of oxygenating and removing CO2, this study raises the importance of developing alternative, bulk-gas-flow-independent, means to recondition PFC liquids.

Improved pulmonary function after surgical reduction of congenital diaphragmatic hernia in lambs

BACKGROUND/PURPOSE

Modern trends are toward delayed surgical reduction of congenital diaphragmatic hernia. This study was conducted to verify the hypothesis that the "ease" of ventilation found in the authors' postoperative experience in infants with congenital diaphragmatic hernia (CDH) is associated with postsurgical improvement in pulmonary mechanics.

METHODS

Very severe CDH was surgically induced in utero at 90 days' gestation in 31 lambs. At birth pulmonary mechanics (PeDS-Lab) was measured in these preterm lambs with (n = 24) and without (n = 7) early surgical reduction, and in eight non-CDH controls over the same period; functional residual capacity (FRC) also was obtained from 14 of the 31 CDH lambs (seven reduced animals and seven unreduced ones). Management excluded aspiration from the thorax and insertion of chest drains.

RESULTS

After 30 minutes of life CDH animals with early surgical reduction demonstrated significantly greater improvement in lung volume, pulmonary mechanics, and oxygenation than those without reduction: FRC, 235% versus 19%; compliance, 57% versus 14%; minute ventilation, 71% versus 30%; and PO2, 143% versus -15%. Over the same period, in preterm controls without CDH, only the compliance varied significantly, demonstrating a 32% increase.

CONCLUSION

Based on the mechanics of breathing in these lambs, the authors speculate that neonates with CDH could benefit from early surgical repair because of improvement in pulmonary function, provided extra care is taken to prevent pulmonary overdistension.

A method for isolating adult and neonatal airway smooth muscle cells and measuring shortening velocity

Methods are described for isolating smooth muscle cells from the tracheae of adult and neonatal sheep and measuring the single-cell shortening velocity. Isolated cells were elongated, Ca2+ tolerant, and contracted rapidly and substantially when exposed to cholinergic agonists, KCl, serotonin, or caffeine. Adult cells were longer and wider than preterm cells. Mean cell length in 1.6 mM CaCl2 was 194 +/- 57 (SD) microm (n = 66) for adult cells and 93 +/- 32 microm (n = 20) for preterm cells (P < 0.05). Mean cell width at the widest point of the adult cells was 8.2 +/- 1.8 microm (n = 66) and 5.2 +/- 1.5 microm (n = 20) for preterm cells (P < 0.05). Cells were loaded into a perfusion dish maintained at 35 degreesC and exposed to agonists, and contractions were videotaped. Cell lengths were measured from 30 video frames and plotted as a function of time. Nonlinear fitting of cell length to an exponential model gave shortening velocities faster than most of those reported for airway smooth muscle tissues. For a sample of 10 adult and 10 preterm cells stimulated with 100 microM carbachol, mean (+/- SD) shortening velocity of the preterm cells was not different from that of the adult cells (0.64 +/- 0.30 vs. 0.54 +/- 0.27 s-1, respectively), but preterm cells shortened more than adult cells (68 +/- 12 vs. 55 +/- 11% of starting length, respectively; P < 0.05). The preparative and analytic methods described here are widely applicable to other smooth muscles and will allow contraction to be studied quantitatively at the single-cell level.

Quantitative bronchoscopic assessment of airway collapsibility in newborn lamb tracheae

To date, quantitative studies of the inherent characteristics of the developing airway wall have required excision of an airway segment or surgical creation of an isolated segment. We hypothesized that airway wall characteristics, at various collapsing pressures, and attendant changes in stiffness after smooth muscle stimulation could be quantitated bronchoscopically from airway pressure-area relationships. Neonatal lamb tracheal segments (n = 12) were suspended over hollow mounts, in a buffer-filled chamber, and subjected to a range (0 to -4.0 kPa) of pressures to determine wall stiffness under collapsing forces before and after stimulation of the trachealis with methacholine. Luminal images were recorded through a 3.6-mm flexible bronchoscope under the same conditions, subsequently corrected for distortion, and a cross-sectional area was quantitated. Both pressure-volume and pressure-area relationships detected significant changes in airway wall stiffness after methacholine administration (p < 0.002), and the magnitude of change was similar between methods. These data suggest that quantitative flexible bronchoscopy can be used clinically in the intact airway to assess wall stiffness.

Antenatal triiodothyronine improves neonatal pulmonary function in preterm lambs

OBJECTIVE

To characterize 1) pulmonary gas exchange, 2) pulmonary function, 3) lung fluid and tissue phospholipid content, and 4) thyroid hormone in the premature lamb (0.85 of term) after intra-amniotic administration of 100 micrograms of triiodothyronine (T3) 2 weeks before delivery.

METHODS

Nine fetal lambs were given 100 micrograms of intra-amniotic T3 under ultrasound guidance at 112 +/- 1 days' gestation and delivered at 126 +/- 1 days (term = 149 days). Five saline-injected animals served as controls. Arterial blood gases, pulmonary mechanics, and lung volumes were compared between groups for 1 hour after delivery. At delivery, tracheal fluid and blood was taken for T3, and thyroxine (T4) levels. Tracheal fluid and lung tissues were assayed for total phosphorus and disaturated phosphatidylcholine.

RESULTS

Triiodothyronine-treated lambs had significantly higher mean arterial pH and lower PCO2 than controls (P < .05) with a trend toward higher mean PO2. The dynamic lung compliance was increased by 54% with a 40% proportional increase in tidal volume and minute ventilation in the T3-treated group (P < .05). Functional residual capacity increased 69% (P < .05) without a change in specific compliance. The tracheal fluid and pulmonary phospholipids and tracheal fluid and plasma T3 and T4 levels were not different between the two groups.

CONCLUSION

A single 100 micrograms dose of antenatal T3 significantly improves neonatal gas exchange and lung compliance. The improvement in lung function was not accompanied by an increase in pulmonary surfactant production. It is inferred that T3 improved lung function via accelerated structural development of the lung with an alternative possible effect on parenchymal connective tissue matrix.

Cardiopulmonary interaction during partial liquid ventilation in surfactant-treated preterm lambs

Gas ventilation following instillation of perfluorochemical (PFC) liquid, partial liquid ventilation (PLV), improves gas exchange and pulmonary mechanics in neonatal animals and humans with severe respiratory distress. The effect of PLV on cardiac contractility, performance, pulmonary blood flow and ductal shunt has not been fully described. To this end, we evaluated these indices of cardiopulmonary function in eight conventionally gas ventilated, surfactant-treated premature lambs (125 days gestation) before and during PLV. Animals were instrumented with central venous and aortic lines. Serial evaluation of arterial blood chemistry/pressure, and pulmonary mechanics was performed; cardiac contractility, performance, pulmonary blood flow and ductal shunts were serially assessed by echocardiography. As compared to conventional gas ventilation, during PLV there was a significant decrease in left ventricular meridian (22.5+/-6.6 SE vs 8.1+/-1.4 SE g/cm2, P < 0.02) and circumferential wall stress (54.1+/-16.5 vs 24.4+/-3.8 SE g/cm2, P < 0.04) at end systole. The fall in wall stress at end systole was associated with a significant decrease in left ventricular internal diameter (1.2+/-0.05 SE vs 1.04+/-0.045 SE cm; P < 0.01). There were no significant changes in heart rate, systemic arterial and central venous pressures, systemic vascular resistance, left ventricular shortening and ejection fractions during PLV. The decrease in wall stress was associated with a significant decrease in mean airway pressures (15.9+/-1.1 SE vs 9.9+/-0.2 SE cmH2O; P < 0.05) and ostensibly a change in intrathoracic pressures during PLV. There were no significant differences in blood flows (pre vs during PLV; ml/min/kg): pulmonary (226+/-62 SE vs 293+/-65 SE), aortic (237+/-36 SE vs 204+/-21 SE), and left to right ductal (119+/-25 SE vs 105.5+/-26 SE) measured before and during PLV.

CONCLUSION

Cardiac output and pulmonary blood flow do not change significantly during PLV and therefore do not appear to contribute to improved gas exchange. This stable cardiac performance occurs at lower wall stress and thereby more advantageous energetic conditions.

Perfluorochemical rescue after surfactant treatment: effect of perflubron dose and ventilatory frequency

To test the hypotheses that perfluorochemical (PFC) liquid rescue after natural surfactant (SF) treatment would improve pulmonary function and histology and that this profile would be influenced by PFC dose or ventilator strategy, anesthetized preterm lambs (n = 31) with respiratory distress were studied using nonpreoxygenated perflubron. All animals received SF at 1 h and were randomized at 2 h as follows and studied to 4 h postnatal age: 1) conventional mechanical gas ventilation (n = 8), 2) 30 ml/kg perflubron with gas ventilation [partial liquid ventilation (PLV)] at 60 breaths/min (n = 8), 3) 10 ml/kg perflubron with PLV at 60 breaths/min (n = 7), and 4) 10 ml/kg perflubron with PLV at 30 breaths/min (n = 8). All animals tolerated instillation without additional cardiopulmonary instability. All perflubron-rescued groups demonstrated sustained improvement in gas exchange, respiratory compliance, and reduction in pressure requirements relative to animals receiving SF alone. Improvement was directly related to perflubron dose and breathing frequency; peak inspiratory pressure required to achieve physiological gas exchange was lower in the higher-dose and -frequency groups, and mean airway pressure was lower in the lower-frequency group. Lung expansion was greater and evidence of barotrauma was less in the higher-dose and -frequency group; regional differences in expansion were not different as a function of dose but were greater in the lower-frequency group. Regional differences in lung perflubron content were reduced in the higher-dose and -frequency groups and greatest in the lower-dose and -frequency group. The results suggest that, whereas PLV of the SF-treated lung improves gas exchange and lung mechanics, the protective benefits of perflubron in the lung may depend on dose and ventilator strategy to optimize PFC distribution and minimize exposure of the alveolar-capillary membrane to a gas-liquid interface.

Physiologic, biochemical, and histologic correlates associated with tidal liquid ventilation

Tidal liquid ventilation (TLV) with perfluorochemical fluid (PFC) has been successfully used experimentally for up to 4 h. However, no studies of prolonged TLV have been reported. We hypothesized that full-term newborn lambs can safely and effectively be liquid-ventilated for up to 24 h. To test this hypothesis, 17 lambs were liquid-ventilated; 7 for 4 h, 5 for 12 h, and 5 for 24 h. Arterial blood samples were obtained for PFC uptake, lipid analysis, and blood gas measurements. Tissues were obtained for histologic and biochemical analysis. Arterial blood gas and mean arterial blood pressure were as follows (mean +/- SEM): pH 7.48 +/- 0.04; PaCO2 30.6 +/- 2.8; PaO2 424 +/- 17; mean arterial pressure 76 +/- 16 mm Hg. PFC blood levels increased rapidly to a mean of 5.2 +/- 3.9 microg/mL. PFC tissue levels increased significantly (p < 0.01) from 260 +/- 45 microg/g at 4 h to 400 +/- 140 microg/g at 12 h. There was no further increase in PFC tissue levels by 24 h (456 +/- 181 microg/g). There was a significant difference in PFC concentration as a function of tissue (p < 0.01). Furthermore, there was a significant correlation (r = 0.88; p < 0.01) between the amount of PFC and lipid in blood and tissue. Microscopic examination of the lungs demonstrated no evidence of barotrauma. These data demonstrate that prolonged TLV can be safe and efficacious for up to 24 h in full-term newborn lambs.

Liquid-assisted ventilation: physiology and clinical application

Liquid-assisted ventilation, as an alternative ventilation strategy for respiratory distress, is progressing from theory and basic science research to clinical application. Biochemically inert perfluorochemical liquids have low surface tension and high solubility for respiratory gases. From early immersion experiments, two primary techniques for liquid-assisted ventilation have emerged: total liquid ventilation and partial liquid ventilation. While computer-controlled, time-cycled, pressure/volume-limited total liquid ventilators can take maximum advantage of these liquids by completely eliminating the gas phase in the distressed lung, partial liquid ventilation takes advantage of having these liquids in the lung while maintaining gas ventilation. The benefits of both partial and total techniques have been demonstrated in animal models of neonatal and adult respiratory distress syndrome, aspiration syndromes and congenital diaphragmatic hernia and also in combination with other therapeutic modalities including extracorporeal membrane oxygenation, high-frequency ventilation and nitric oxide. Additionally, nonrespiratory applications have expanding potential including pulmonary drug delivery and radiographic imaging. Since its use in neonates in 1989, liquid-assisted ventilation in humans has progressed to a variety of clinical experiences with different aetiologies of respiratory distress. The future holds the opportunity to clarify and optimize the potential of multiple clinical applications for liquid-assisted ventilation.

Pulmonary administration of gentamicin during liquid ventilation in a newborn lamb lung injury model

OBJECTIVES

Newborns with pulmonary infection frequently present with acute lung injury leading to ventilation/perfusion abnormalities in which intravenous delivery of antibiotics to the lung can be suboptimal. Tidal liquid ventilation (TLV) has been shown to be an effective means for delivering drugs directly to the pulmonary system. The objective of this study was to compare, with lung injury, antibiotic delivery achieved by conventional techniques (gas ventilation and intravenous gentamicin) with that using pulmonary administration of drug (PAD) during TLV.

METHODS

Twelve newborn lambs with an acid lung injury were randomized to receive gentamicin either intravenously during gas ventilation or via PAD during TLV using LiquiVent (Alliance Pharmaceutical Corporation, San Diego, CA, and Hoechst-Marion Roussel, Bridgewater, NJ) perfluorochemical. Gentamicin (5 mg/kg) was administered over 1 minute, and serum levels were obtained at 15-minute intervals. Arterial blood gases and pulmonary mechanics were measured. Ventilation efficiency index and arterial/alveolar oxygen ratio were calculated. Lung-tissue gentamicin levels were measured 4 hours after administration and corrected to dry weight.

RESULTS

Serum gentamicin levels were similar in both groups. Lung gentamicin levels (micrograms/g) were significantly higher for TLV. Also, TLV resulted in significantly more of the total delivered dose in the lung after 4 hours. Ventilation efficiency index and arterial/alveolar oxygen ratios were significantly higher for TLV.

CONCLUSIONS

In this lung injury model, both methods achieved equivalent serum gentamicin levels with higher lung levels using PAD during TLV. This study suggests that TLV may provide an effective vehicle for gentamicin delivery in infants with severe pulmonary infection and ventilation/perfusion abnormalities.

Analysis of perfluorochemical elimination from the respiratory system

We describe a simple apparatus for analysis of perfluorochemicals (PFC) in expired gas and thus a means for determining PFC vapor and liquid elimination from the respiratory system. The apparatus and data analysis are based on thermal conduction and mass transfer principles of gases. In vitro studies were conducted with the PFC vapor analyzer to determine calibration curves for output voltage as a function of individual respiratory gases, respiratory gases saturated with PFC vapor, and volume percent standards for percent PFC saturation (%PFC-Sat) in air. Voltage-concentration data for %PFC-Sat of the vapor from the in vitro tests were accurate to within 2.0% from 0 to 100% PFC-Sat, linear (r = 0.99, P < 0.001), and highly reproducible. Calculated volume loss of PFC liquid over time correlated well with actual loss by weight (r = 0.99, P < 0. 001). In vivo studies with neonatal lambs demonstrated that PFC volume loss and evaporation rates decreased nonlinearly as a function of time. These relationships were modulated by changes in PFC physical properties, minute ventilation, and postural repositioning. The results of this study demonstrate the sensitivity and accuracy of an on-line method for PFC analysis of expired gas and describe how it may be useful in liquid-assisted ventilation procedures for determining PFC volume loss, evaporation rate, and optimum dosing and ventilation strategy.

Virtual bronchoscopy with perfluoronated hydrocarbon enhancement

RATIONALE AND OBJECTIVES

Bronchoscopic computed tomography (CT) is limited by machine resolution and air-soft-tissue contrast. The objective of this study was to determine whether improving the contrast by using the contrast agent perflubron (PFOB) in the lung would improve the bronchoscopic CT technique and permit visualization of small airways.

MATERIALS AND METHODS

Bronchoscopic CT was performed in an anesthetized 8-week-old New Zealand white rabbit before and after the endotracheal administration of PFOB.

RESULTS

Bronchoscopic CT performed with PFOB permitted navigation of bronchi as small as 0.8 mm in diameter, which are much smaller than those that can be navigated without PFOB.

CONCLUSION

In this example, the use of perfluorochemicals with bronchoscopic CT enhanced the capabilities of virtual bronchoscopy.

Comparison of perfluorochemical fluids used for liquid ventilation: effect of endotracheal tube flow resistance

Neonatal endotracheal tubes with small inner diameters are associated with increased resistance regardless of the medium used for assisted ventilation. During liquid ventilation (LV) reduced interfacial tension and pressure drop along the airways result in lower alveolar inflation pressure compared with gas ventilation (GV). This is possible by optimizing liquid ventilation strategies to overcome the resistive forces associated with liquid density (rho) and viscosity (mu) of these fluids. Knowledge of the effect of rho, mu, and endotracheal tube (ETT) size on resistance is essential to optimize LV strategies. To evaluate these physical properties, three perfluorochemical (PFC) fluids with a range of kinematic viscosities (FC-75 = 0.82, LiquiVent = 1.10, APF-140 = 2.90) and four different neonatal ETT tubes (Mallincrokdt Hi-Lo Jet ID 2.5, 3.0, 3.5, and 4.0 mm) were studied. Under steady-state flow, flow and pressure drop across the ETTs were measured simultaneously. Resistance was calculated by dividing pressure drop by flow, and both pressure-flow and resistance-flow relationships were plotted. Also, pressure drop and resistance were each plotted as a function of kinematic viscosity at flows of 0.01 L.s-1 for all four ETT sizes. Data demonstrated a quadratic relationship with respect to pressure drop versus flow, and a linear relationship with resistance versus flow: both were significantly correlated (R = 0.92; P < 0.01) and were inversely related to ETT size. Additionally, there was a significant correlation between pressure drop or resistance and kinematic viscosity (R = 0.99; P < 0.01). For LV in neonates these data can be used to select the optimum ETT size and PFC liquid depending OR the chosen ventilation strategy.

Enhanced distribution of adenovirus-mediated gene transfer to lung parenchyma by perfluorochemical liquid

Although gene therapy holds great promise for the treatment of inherited and acquired diseases of the lung, a number of issues including efficient delivery and distribution of genes to pulmonary target cells must still be addressed. In this study we evaluated the use of perfluorochemical (PFC) liquid as a vehicle for delivery of recombinant adenovirus (AdCBlacZ) to lungs of juvenile rabbits. Virus was instilled into trachea of rabbits, and 4 days later the lungs were removed, cut into multiple pieces, and assayed for beta-galactosidase (beta-Gal) activity. Total lung expression of the beta-Gal reporter gene was increased two- to three-fold by instillation of the virus (10(11) particles/kg body weight) in saline (1.5 ml/kg) simultaneously with perflubron liquid (15 ml/kg) compared to virus+saline alone (control). Uniformity of beta-Gal activity between lobes was significantly improved by the PFC liquid. In perflubron-treated lungs approximately 45% of the lung pieces had beta-Gal-specific activity values within 50-150% of the mean specific activity for the total lung, compared to only approximately 15% of the pieces in control lungs. More of total lobar beta-Gal activity was recovered in the distal lung tissue (approximately two-fold greater than controls, p < 0.05). Morphological assessment of X-Gal-stained, fresh-frozen lung sections showed increased levels and more complete staining of alveolar wall cells in the PFC group. These data indicate that the PFC liquid perflubron enhances distribution of virus-mediated gene expression to the lung parenchyma in healthy rabbits. PFC liquid may be a useful treatment vehicle for accessing distal spaces of the damaged or diseased lung.

A comparison of intratracheal and intravenous administration of gentamicin during liquid ventilation

Pulmonary absorption of aminoglycosides is poor with intravenous administration, but may be enhanced by direct intratracheal administration of these drugs using perfluorochemical liquid ventilation (LV). To test this hypothesis, gentamicin sulfate was administered to two groups of newborn lambs during LV. Serum and lung tissue levels of gentamicin were compared after either pulmonary intratracheal (IT) or intravenous (IV) routes of administration. Serial serum levels of gentamicin were obtained every 15 min for the 1st h, every 30 min for the 2nd h, and then hourly until sacrifice (maximum 6 h). At sacrifice, representative samples of each lung lobe were homogenized and analyzed for tissue gentamicin content. At 1 h, serum gentamicin levels were similar in both groups: IT administration levels were 3.7 +/- 0.55 SE micrograms/ml and IV levels were 3.5 +/- 0.85 SE micrograms/ml. The peak serum gentamicin level of 4.8 +/- 0.8 SE micrograms/ml for the pulmonary administration group occurred 1.5 h after administration. Lung tissue levels of gentamicin for IT administration (4.04 +/- 0.62 SE micrograms/g) were significantly greater than for IV administration (1.75 +/- 0.33 SE micrograms/g; P < 0.05). There were no significant differences in interlobar gentamicin distribution for either mode of administration.

CONCLUSION

Perfluorochemical can be used as a vehicle for intratracheal delivery of antimicrobials. This route provides equivalent serum levels at 1 h, higher lung tissue levels, and uniform interlobar distribution relative to intravenous administration of gentamicin. We speculate that pulmonary administered gentamicin during LV may provide an effective alternative treatment modality in the management of severe neonatal pneumonia.

Halothane administration during liquid ventilation

The objective of this study was to test the hypothesis that perfluorochemical (PFC) liquid ventilation (LV) can be used as a vehicle to deliver halothane and induce and maintain analgesia. Seven hamsters were paralysed and stabilized with mechanical gas ventilation, ventilated in alternating cycles with gas and either neat oxygenated PFC liquid or oxygenated PFC liquid mixed with liquid halothane (PFC:hal) 1:50% (volume/vapour); arterial pressure and blood gases were monitored throughout the protocol. After each cycle, the animal was stimulated with a foot clamp for 2 s. Mean arterial pressure (MAP:mmHg) response to this stimulation (percent change from the resting MAP) was used as an index of analgesia. Mean arterial pressure was significantly lower during ventilation with PFC:hal (73 +/- 7 SE) as compared with MAP during neat PFC (113 +/- 5 SE) or gas ventilation (107 +/- SE). Mean arterial pressure response (% change in MAP from baseline) to foot-clamp stimulation was significantly lower with PFC:hal ventilation (+ 12 +/- 5% SE) as compared with neat PFC (+ 28 +/- 8% SE) and gas ventilation (+ 29 +/- 9% SE). There was no statistically significant difference in resting MAP or MAP response to foot-clamp stimulation between cycles of ventilation with neat PFC alone or gas ventilation; arterial blood gases were not significantly different between modes of ventilation or levels of analgesia. The data indicate that halothane can be administered during LV while supporting gas exchange, and demonstrate the feasibility of inducing analgesia while using PFC LV techniques.

Morphometrics of normal and hypoplastic lungs in preterm lambs with gas and partial liquid ventilation

To test the hypothesis that perfluorochemical (PFC) instillation may reduce the pulmonary trauma commonly associated with conventional gas ventilation, we studied 12 lambs with normal lungs and 10 with pulmonary hypoplasia secondary to congenital diaphragmatic hernia (CDH). We used mechanical ventilation for up to 3.5 h, with and without tracheal instillation of LiquiVent PFC liquid. At the end of experimentation lungs were fixed for morphometrical analysis of their components and pulmonary trauma was evaluated by measurement of the perivascular compression index (PCI = % perivascular emphysema/% vessels). In normal lungs good gas exchange and respiratory mechanics were obtained with all modes of ventilation, with no statistical difference in the index of pulmonary trauma with or without instillation of LiquiVent (P >0.05). In the hypoplastic lungs, tracheal instillation of PFC liquid after 30 min of conventional gas ventilation significantly improved PaCO2 (from 107+/-8 to 55+/-6 mmHg, P <0.05), pH (from 7.00+/-0.03 to 7.29+/-0. 04, P <0.05), compliance (from 0.08+/-0.01 to 0.25+/-0.03 ml/cmH2O . kg, P <0.05), and ventilatory index (from 1,445+/-148 to 794+/-139, P <0.05). Survival was 6/6 animals with PFC ventilation compared to 1/4 with conventional gas ventilation with no more pulmonary trauma (mean PCI 12.6+/-1.8 vs. 11.4+/-4.0%, P >0.05) for a longer mean ventilatory period in the PFC group. We conclude that the PFC liquid technique of ventilation can improve respiratory physiology when conventional gas ventilation alone is proving inefficient. There was no significant difference in pulmonary trauma at morphometrics between gas and partial liquid ventilation.

Partial liquid ventilation in critically ill infants receiving extracorporeal life support. Philadelphia Liquid Ventilation Consortium

OBJECTIVES

To demonstrate that a period of partial liquid ventilation (PLV) with perflubron improves pulmonary function, without adverse events, in a select group of critically ill infants receiving extracorporeal life support (ECLS) with a high likelihood of mortality.

METHODS

This was an open-label, noncontrolled, phase I and II trial of PLV in two infants with congenital diaphragmatic hernia and four infants with acute respiratory distress syndrome (ARDS) who were failing to improve while receiving ECLS. PLV was performed by instilling and maintaining a functional residual capacity of sterile perflubron for 4 to 96 hours.

RESULTS

Four infants were successfully weaned off ECLS for at least 3 days, and two infants (both with ARDS) are long-term survivors after PLV. All infants demonstrated lung recruitment and improved lung compliance, and there were no adverse events related to PLV.

CONCLUSIONS

The study suggests that perflubron PLV is safe, improves lung function, and recruits lung volume in critically ill infants receiving ECLS. PLV therapy for infants with ARDS seems to have a great deal of promise. Based on this and other phase I and II trials, studies of PLV on selected full-term infants before ECLS have been initiated.

Quantitative structure-activity relationships of perfluorinated hetero-hydrocarbons as potential respiratory media. Application to oxygen solubility, partition coefficient, viscosity, vapor pressure, and density

It has been extensively reported that liquid-assisted ventilation, using inert perfluorocarbon liquids (PFCs), can reduce interfacial surface tension and allow for improved ventilation at decreased alveolar pressures. PFCs are bioinert, minimally absorbed, and have no deleterious histologic, cellular, or biochemical effects when used as respiratory media. Although several types of PFCs have been characterized, a select few are considered to be compatible with life. Compatibility is often related to the physicochemical profile inherent to the PFC liquids. It is essential that certain physical properties such as respiratory gas solubility, vapor pressure, density, viscosity, and tissue permeability be within a narrow, acceptable range for a PFC to be considered as a possible candidate for respiratory media. The current study sought to characterize the physicochemical profile of commercially available PFCs. This was accomplished by creating a method for accurate, rapid prediction of a host of unknown physical characteristics of PFCs. The physicochemical properties of 16 perfluorinated hetero-hydrocarbons were catalogued from the literature. The input data were categorized into three major groups: empiric properties, geometric indices, and quantum mechanical descriptors, to generate a database. Algorithms were then developed, one for each dependent variable (FUNCTION), including oxygen solubility, partition coefficient (logP), vapor pressure, viscosity, and density, that related the values of these physical properties of potential breathable PFC liquids to the parameters listed in the database. The general form of the algorithm can be written as follows: FUNCTION = sigma (CiPi/magnitude of Pi) + constant; where the FUNCTIONS are oxygen solubility, logP, vapor pressure, viscosity, and density. Ci is a coefficient that weights the relative contribution of each parameter. Each independent parameter, Pi, was normalized by the average value of the parameter used in the analysis, magnitude of Pi. Residual analysis demonstrated validity with all five equations. This method is expected to assist in the prediction of physical properties of PFC liquids with acceptable accuracy, such that PFC production and selection from currently available liquids can be optimized for each liquid ventilation application.

Partial liquid ventilation with perflubron in premature infants with severe respiratory distress syndrome. The LiquiVent Study Group

BACKGROUND

The intratracheal administration of a perfluorocarbon liquid during continuous positive-pressure ventilation (partial liquid ventilation) improves lung function in animals with surfactant deficiency. Whether partial liquid ventilation is effective in the treatment of infants with severe respiratory distress syndrome is not known.

METHODS

We studied the efficacy of partial liquid ventilation with perflubron in 13 premature infants with severe respiratory distress syndrome in whom conventional treatment, including surfactant therapy, had failed. Partial liquid ventilation was initiated by instilling perflubron during conventional mechanical ventilation to a volume approximating the functional residual capacity. Infants were considered to have completed the study if they received partial liquid ventilation for at least 24 hours.

RESULTS

Ten infants received partial liquid ventilation for 24 to 76 hours. In the other three infants, partial liquid ventilation was discontinued within four hours in favor of high-frequency ventilation, which was not permitted by the protocol, and the data from these infants were excluded from the analysis. Within one hour after the instillation of perflubron, the arterial oxygen tension increased by 138 percent and the dynamic compliance increased by 61 percent; the mean (+/- SD) oxygenation index was reduced from 49 +/- 60 to 17 +/- 16. Chest radiographs showed symmetric filling, with patchy clearing during the return from partial liquid to gas ventilation. There were no adverse events clearly attributable to partial liquid ventilation. Infants were weaned from partial liquid to gas ventilation without complications. Eight infants survived to 36 weeks' corrected gestational age.

CONCLUSIONS

Partial liquid ventilation leads to clinical improvement and survival in some infants with severe respiratory distress syndrome who are not predicted to survive.