Ann Thorac Surg 2007;84:988-994
© 2007 The Society of Thoracic Surgeons
Original Articles: General Thoracic
Seven-Day Artificial Lung Testing in an In-Parallel Configuration
Hitoshi Sato, MD,
Grant W. Griffith, BS,
Candice M. Hall, BS, LVT,
John M. Toomasian, MS,
Ronald B. Hirschl, MD,
Robert H. Bartlett, MD,
Keith E. Cook, PhD*
Department of Surgery, University of Michigan, Ann Arbor, Michigan
Accepted for publication March 5, 2007.
* Address correspondence to Dr Cook, 7679 Kresge I, 200 Zina Pitcher Pl, Ann Arbor, MI 48109 (Email: keicook{at}umich.edu).
 |
Abstract
|
|---|
Background: A thoracic artificial lung, the MC3 Biolung, is being developed as a bridge to lung transplantation or as a treatment for acute respiratory insufficiency.
Methods: The thoracic artificial lung was tested in 10 sheep with the goal of 7 days of respiratory support. The sheep were recovered from surgery and monitored awake for 7 days. Hemodynamics, blood gases, blood cell counts, and organ function were recorded, and after 7 days, all sheep were euthanized and necropsied.
Results: Seven sheep survived the full duration. Cardiac output and mean arterial blood pressure were unchanged, averaging 4.7 ± 0.8 L/min and 98 ± 10 mm Hg, respectively. Arterial oxygen tension and device oxygen transfer rate were also unchanged, averaging 110 ± 26 mm Hg and 97.7 ± 35 mL/min, respectively. Arterial carbon dioxide tension was within normal ranges during the entire experiment, averaging 37.4 ± 3.8 mm Hg. Artificial lung blood flow decreased from 51% ± 14% of cardiac output on day 1 to 30% ± 16% by day 7 because of changes in natural and artificial lung resistance. White blood cell counts were significantly elevated on days 5 and 7, and lastly, kidney and liver function remained normal, although signs of kidney infarction or hemorrhage were noted.
Conclusions: The thoracic artificial lung is suitable for 7-day attachment, but improvements in blood biocompatibility are warranted.
|
Introduction
|
|---|
In the past decade, lung transplantation has matured significantly as a treatment for end-stage lung disease. However, the number of donors has not kept up with the demand for transplantable organs. The waiting list for a lung transplant has grown steadily during the last decade, reaching 3,836 patients at the end of 2003 [1]. In addition, acute respiratory distress syndrome and acute lung injury affect 13.5 to 75 people per 100,000 [2, 3], with a mortality rate between 34% and 65% [4]. The main causes of acute respiratory distress syndrome deaths are sepsis or multiple organ failure rather than respiratory deficit. However, studies indicate that ventilator-induced lung trauma may contribute to multiple organ failure [5], and a mortality rate as low as 31% has been reported with improved ventilator settings, including lower tidal volumes [6]. Thus, less aggressive ventilation may decrease acute respiratory distress syndrome mortality.
The purpose of a thoracic artificial lung (TAL) is to act as either a bridge to lung transplantation or as a treatment for acute respiratory insufficiency that is unresponsive to mechanical ventilation. By assuming the majority of the respiratory requirements, the TAL can support the patient until transplantation or allow the natural lungs to heal. The purpose of this study is to examine 7-day attachment of the MC3 Biolung (Ann Arbor, MI) using an in-parallel, pulmonary artery (PA) to left atrial (LA) configuration. Previous studies have examined 3- to 7-day attachment of similar devices using in-series, PA to PA attachment [7–9], making this the first long-term study of the PA to LA attachment mode. The MC3 Biolung was thus attached in 10 adult sheep with minimal pharmacologic intervention and no transfusion of blood products. Various hemodynamic, gas transfer, and hematologic variables were measured to determine whether there are significant changes in device function or animal physiology and to determine whether device changes are warranted before longer-term, preclinical testing.
|
Material and Methods
|
|---|
Thoracic Artificial Lung Preparation
The TAL was the Biolung without a compliance chamber [10]1. The device has a 1.7-m2 surface area fiber bundle using 380-µm outer diameter, polymethylpentene Oxyplus fibers (Membrana Gmbh, Wuppertal, Germany) and an acrylonitrile butadiene styrene housing with a priming volume of 300 mL. Sweep gas through the device was attached to a vacuum to maintain a –10 to –20 mm Hg negative pressure at the TAL gas inlet to avoid gas embolism. Custom blood inflow and outflow conduits consisted of approximately 6 inches of 5/8 inner diameter Tygon tubing that is bonded to 18-mm vascular grafts (Boston Scientific, Natick, MA).
Surgery
All sheep received humane care in compliance with the Guide for the Care and Use of Laboratory Animals. Anesthesia induction consisted of 12 mg/kg intravenous sodium thiopental and 0.3 mg of intramuscular buprenorphine hydrochloride. Surgical ventilation followed previously published methods [10]. Penicillin (500 mg, intravenous) was administered, and arterial and venous monitoring catheters were placed. Baseline (BL) arterial blood samples were obtained to measure platelet and white blood cell counts, a blood chemistry panel, and plasma free hemoglobin. The blood chemistry panel includes aspartate aminotransferase, alanine aminotransferase, blood urea nitrogen, and creatinine.
Heparin (100 U/kg, intravenous) and ketorolac (60 mg, intravenous) were administered. The outlet and inlet graft conduits were attached to the LA and PA, respectively, by means of end-to-side anastomoses and then tunneled through the fifth intercostal space. A flow probe (Transonic 24AX, Ithaca, NY) was placed around the main trunk of the PA, distal to the conduit, and its cable was tunneled to the sheep’s back. The TAL was primed with saline solution and attached to the conduits and gas lines. Sweep flow was started at 1 L/min of oxygen. Blood flow to the TAL was then initiated by removing tubing clamps on the conduits. Chest tubes were placed, and the chest was closed. A nylon jacket was placed around the device and sutured to the skin. The sheep was moved to a custom-built cage that allowed all activity except turning around. The sheep were awake and ate and drank freely over the 7 days of TAL support.
A heparin drip (80 U/h, Multi-Phaser NE-1000, New Era Pump System Inc, Wantagh, NY) was supplied to the TAL blood inlet once the activated clotting time fell below 240 seconds, and the delivery was adjusted to maintain the activated clotting time between 180 and 220 seconds. The TAL sweep gas was replaced by 5 L/min of oxygen blended with 0% to 2% of carbon dioxide. For recovery, the sheep was switched to a Nellcor 800 series ventilator (Puritan-Bennet, Carlsbad, CA), weaned, and extubated.
Seven sheep died during or shortly after anesthesia as the preparation was being developed. Results for these sheep are not presented. In brief, the device uses polymethylpentene fibers, which do not allow passage of isoflurane [11]. This led to poor postoperative recovery in sheep. Three postoperative analgesic protocols were thus used in the 10 animals presented here as we sought to improve recovery: (1) ketorolac (30 mg, intravenous) and buprenorphine (0.3 mg, intravenous) every 6 hours (n = 2); (2) ketorolac, buprenorphine, and midazolam hydrochloride (0.05 mg/kg, intravenous, n = 3); and (3) fentanyl (75 µg/h, transdermal, n = 5). All sheep were given penicillin (500 mg, intravenous) and gentamicin (2.5 mg/kg, intravenous) every 6 and 8 hours, respectively. All pressures and flows were recorded hourly, and blood gases were taken every 1 to 4 hours.
The sheep were euthanized using Fatal-Plus (Vortech Pharmaceuticals, Ltd, Dearborn, MI) and necropsied. The heart was dissected to separate the right ventricular free wall from the left ventricle and septum. Each component was weighed to determine the right ventricular free wall to left ventricle and septum weight ratio, rw.
Data Analysis
Device resistance and total and TAL oxygen transfer were calculated according to previously published methods [12]. Data from the 10 sheep that recovered from surgery and regained full consciousness were included in statistical analyses examining the effect of time on various indices. Data on postsurgical days 1 through 7 (D1 through D7) were averaged across each 24-hour period. Hemodynamic and gas transfer indices used only data off anesthesia and breathing room air (D1 through D7). Hematologic and blood chemistry data used data from BL and D1 through D7. All comparisons were performed with SPSS (SPSS, Chicago, IL) using a mixed model with the sheep number as the subject variable and time as the fixed, repeated-measure variable. Post hoc analysis using a Bonferroni-corrected confidence interval was then used to compare all variables to D1 in the case of hemodynamic and gas transfer data and BL in the case of hematologic and blood chemistry data. In addition, the natural log of platelet count was used for analysis to achieve a normal distribution. Lastly, a Student’s t test was used to compare the right ventricular free wall to left ventricle and septum weight ratio in the 7-day survivors to historical values from 7 healthy sheep weighing 35 ± 3.1 kg.
|
Results
|
|---|
Ten sheep regained full consciousness after surgery, and 7 sheep survived 7 days with the TAL (Table 1). One sheep died of bleeding at 24 hours, and 2 died at 49 and 95 hours of sepsis as a result of pneumonia.
Sheep Physiology
Both mean arterial blood pressure and central venous pressure remained within normal, healthy ranges with neither varying statistically (p = 0.67 and 0.18, respectively; Fig 1,
Table 2). The mean arterial blood pressure averaged 98 ± 10 mm Hg from D1 to D7, and the central venous pressure increased steadily but insignificantly from 6.2 ± 4.1 to 10.8 ± 2.7 mm Hg, respectively. Cardiac output was within a normal range during the entire experiment and remained statistically unchanged (p = 0.11), averaging 4.7 ± 0.8 L/min from D1 to D7 (Fig 1). Heart rate was significantly lower on D2 than D1 (p < 0.01) but not significantly different on D3 through D7. Overall, heart rate was elevated from normal, healthy ranges for sheep (100 to 120 bpm), averaging 137 ± 17 bpm from D1 to D7. The hemoglobin, arterial oxygen tension, and venous oxygen tension remained within normal ranges (Fig 2,
Table 3), averaging 8.1 ± 1.2 g/dL, 110 ± 26 mm Hg, and 35.5 ± 7.7 mm Hg, respectively, and did not vary significantly with time (p = 0.44, 0.20, and 0.33, respectively). Arterial carbon dioxide tension approached significance (p = 0.06) but remained within normal ranges during the entire experiment, averaging 37.4 ± 3.8 mm Hg from D1 to D7.
View larger version (18K):
[in this window]
[in a new window]
|
Fig 1. Mean arterial pressure (mAP; open symbols) and cardiac output (CO; closed symbols) at postoperative days 1 through 7 (D1–D7).
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Fig 2. Arterial oxygen tension (PaO
2; open symbols) and carbon dioxide tension (PaCO
2; closed symbols) and venous oxygen tension (PvO
2; x symbols) for postoperative days 1 through 7 (D1–D7).
|
|
Device Function
Device blood flow resistance did not vary significantly with time (p = 0.26), despite a seemingly steep increase in device resistance at D6 and D7 (Fig 3,
Table 2). Resistance averaged 2.5 ± 1.1 mm Hg · min · L–1 from D1 to D5, whereas on D6 and D7, the average increased to 4.3 ± 4.2 and 14.1 ± 28.7 mm Hg · min · L–1, respectively. Much of the large deviation results from one device used on D5 through D7 in sheep 17. This device’s resistance rose markedly to 13.7 mm Hg · min · L–1 on D6 and became nearly fully occluded with a resistance of 79.0 mm Hg · min · L–1 on D7. The remaining six devices have D6 and D7 resistances of 2.7 ± 0.7 and 3.2 ± 2.3 mm Hg · min · L–1, respectively.
View larger version (22K):
[in this window]
[in a new window]
|
Fig 3. Artificial lung resistance (closed symbols) and percentage of cardiac output (CO) flowing to the thoracic artificial lung (TAL; open symbols) for postoperative days 1 through 7 (D1–D7).
|
|
Flow through the TAL was significantly lower at D2 (p < 0.01) and D7 (p < 0.05) than at D1 (Fig 3, Table 2). Natural lung flow rate was recorded in only 8 of the 10 sheep that recovered owing to PA flow probe malfunctions. There was no significant difference for any times. The percentage of cardiac output going to the TAL decreased, albeit not significantly (p = 0.24). It was initially 51% ± 14%, decreased and then stabilized between 43% ± 14% and 46% ± 21% from D2 to D4, and then decreased thereafter to 30% ± 16% by D7. Much of this decrease at D7 was again attributable to the device in sheep 17, whose flow dropped from 1.5 L/min at D5 to 0.2 L/min by D7.
Total oxygen transfer was stable, with no significant difference among times (p = 0.96). The TAL oxygen transfer decreased slightly during the course of the experiment, but this change did not reach statistical significance (p = 0.13). The drop in TAL oxygen transfer was predominantly attributable to the drop in device flow, as TAL blood outlet oxygen saturations did not change significantly with time (p = 0.6), averaging 99.5% ± 1.5% for D1 to D7.
On explant, fiber bundles were predominantly free from thrombus. Most devices, however, possessed thrombus within areas of stasis, either at the distal end of the device or small areas where the bundle was in close proximity to the housing. Little to no thrombus was found within the inlet and outlet passages, with the exception of the second device used by sheep 17, which was nearly completely occluded.
Coagulation and Inflammation
Platelet counts decreased significantly between BL and D1 (p < 0.01; Fig 4,
Table 4) and remained reduced at D2 (p = 0.06), although this fell short of significance. After D2, platelet counts increased and were not significantly different from BL at D3 through D7 (p = 1.0). The increase in platelet count was reflected by the rate of heparin delivery, which increased significantly from D1 to D4 (p < 0.05; Table 4). Delivery rates continued to be significantly larger for D5 (p < 0.001) through D7 (p < 0.000001), peaking at 2,220 ± 992 U/h at D7. The activated clotting times for the experiment were relatively stable, averaging 228 ± 18 seconds on D1 before settling out to a nearly constant value of 195 ± 14 seconds for D2 through D7.
View larger version (21K):
[in this window]
[in a new window]
|
Fig 4. Platelet (closed symbols) and white blood cell counts (open symbols) at baseline (BL) and for postoperative days 1 through 7 (D1–D7).
|
|
The white blood cell counts (Fig 4, Table 4) did not vary significantly from BL at D1 through D4 (p = 1.0 for all) and D6 (p = 0.11), but were significantly increased from BL at D5 and D7 (p < 0.05 and 0.001, respectively). Plasma-free hemoglobin increased significantly between BL and D1 (p < 0.0001) and decreased thereafter. Free hemoglobin remained significantly elevated from BL at D2 (p < 0.05) but was not significantly different from BL at D3 through D7 (p = 0.54 to 1.0).
Organ Function
There was no significant effect of time on aspartate aminotransferase or alanine aminotransferase concentrations (p = 0.13 and 0.14, respectively; Table 4). Both aspartate aminotransferase and alanine aminotransferase concentrations increased between BL and D1 through D4, but much of the increase was attributable to sheep 8. This sheep had preexisting pneumonia and aspartate aminotransferase and alanine aminotransferase concentrations ranging from 3,000 to 4,200 (U/L) and 330 to 470 (U/L), respectively, and died on D4. This death returned overall concentrations of each to values that were not significantly different from BL (p = 0.1 to 1.0).
The blood urea nitrogen concentrations varied significantly with time (Table 4), but remained within normal ranges during the course of the experiment. Concentrations for D1 through D3 and D5 through D7 were not significantly different from BL, averaging 16.0 ± 4.7 mg/dL. The D4 concentration, 25.5 ± 11.7 mg/dL, was significantly greater than that of BL (p < 0.01). Creatinine concentrations did not vary significantly with time (p = 0.08) and averaged 0.91 ± 0.26 mg/dL from D1 to D7.
Lastly, all sheep were awake and alert during their course with no signs of stroke or brain damage.
Sheep Necropsy
All sheep demonstrated pathologic changes consistent with cardiac surgery. Atelectasis was seen in most sheep (n = 5), but this was localized to regions where grafts compressed lung tissue. Additional cardiovascular lesions included a 2-cm subvalvular interventricular septal defect (n =1) and chronic pericarditis (n = 2), endocarditis (n = 2), or myocarditis (n = 1). Lastly, the ratio of right ventricular free wall to left ventricle and septum was significantly greater after 7 days of artificial lung attachment than in normal sheep (p < 0.0001), increasing from 0.33 ± 0.03 to 0.42 ± 0.03. With one exception, the sheep were free of gross signs of infection at either the graft entry sites or neck incisions. The exception was sheep 1, which possessed swelling and an abscess near the jugular vein access site. Wound care was improved thereafter, eliminating wound infection in subsequent sheep. One sheep showed signs of preexisting chronic systemic infection, as indicated by chronic pneumonia, pleuritis, pericarditis, myocarditis, nephritis, and hepatitis.
Most important for PA to LA attachment, infarction and hemorrhage were not found in any of the lungs, livers, or spleens. An acute focal infarct was observed in the kidney of 1 sheep, and acute microinfarcts were noted in the kidneys of 2 other sheep. Lastly, 1 sheep possessed a mild periarterial hemorrhage, and 1 had nephritis with slight petechia on one kidney and slight hemorrhage on the renal pelvis of the other.
|
Comment
|
|---|
Sheep Physiology and Device Function
Seven of the 10 sheep that recovered from surgery survived for 1 week with increasing stability during the course. All of the measured hemodynamic variables were within normal ranges and did not change significantly with time. Thus, in most cases, if the sheep recovered from surgery and regained full consciousness, they became progressively more stable and survived the week. The only exception was sheep with pneumonia. Sheep can frequently possess low-grade, subclinical pneumonia that becomes more severe only after operative stress [13]. These sheep may have benefited from purposeful treatment of the underlying infection or the use of PA banding to divert more blood flow from the dysfunctional native lung to the TAL. Even including these sheep, device function was sufficient for respiratory support and was maintained for the course of the experiment. There were no significant changes in device outlet saturations or resistance, and thrombus formation, although present, was not sufficient to affect device function in nearly all cases.
Device resistance was unchanged in all sheep but 1 during the first 6 postoperative days. On D7, however, 4 of the 7 surviving sheep had devices with increased resistance. Resistances for these sheep had jumped from initial values of 2.2, 1.4, 2.2, and 3.9 mm Hg · min · L–1 to D7 values of 4.1, 2.8, 7.5, and 79 mm Hg · min · L–1. The late, inconsistent increase led to the lack of statistical significance, but this may signal an approaching acceleration in thrombus growth that would have appeared more consistently in subsequent days. The most severe case was sheep 17, the only sheep to receive a replacement device. The resistance of this replacement device grew from a value of 4.8 mm Hg · min · L–1 on D5, its first day of use, to 79 mm Hg · min · L–1 on D7. This starting resistance was more than twice the normal starting value, 2.2 ± 0.8 mm Hg · min · L–1. As the housings are quite uniform, this indicates an overly tight fiber bundle wind during device construction. In addition, necropsy indicated that, unlike the other 7-day survivors, this sheep was chronically septic and may have had a more procoagulant blood supply.
The average fraction of cardiac output through the Biolung decreased from 52% at D1 to 30% by D7 (Table 2). An initial drop in the fraction at D2 was likely related to a postoperative decrease in pulmonary vascular resistance, as TAL resistance did not change significantly during that period. Flow rate through the TAL was stable from D2 to D4, but the fraction through the TAL decreased again thereafter. This decrease is caused by increased device resistance and the death of sheep 8, who had a larger cardiac output fraction to the TAL owing to natural lung hypoxia.
Hematology and Organ Function
Progressive hemostatic complications, hemolysis, and resultant organ dysfunction do not appear to be major impediments to 7-day, PA to LA use of the Biolung. Platelet counts were significantly lower at D1 than at BL but returned to normal levels thereafter. There is no way to separate the roles of surgical trauma and blood–biomaterial contact on platelet count within this study. However, platelet counts returned to normal levels by D3, suggesting that surgical trauma is the predominant factor decreasing platelet counts in these experiments. Similarly, plasma-free hemoglobin concentrations were significantly elevated at D1 but returned to BL levels by D2. Thus, TAL itself did not cause significant platelet consumption or hemolysis, and transfusion of blood products should not be necessary during clinical use of the Biolung unless there are surgical bleeding complications.
Clot was formed in the devices, particularly at the end opposite the entrance where blood flow velocities are the lowest. Thus, some microthromboembolic or macrothromboembolic material may have been delivered to the systemic circulation. However, the amount delivered does not appear to be significant, as liver damage and renal dysfunction did not contribute to morbidity or mortality. There were no significant changes in aspartate aminotransferase, alanine aminotransferase, or creatinine, and although the blood urea nitrogen concentration was significantly increased at D4, it returned to normal thereafter. Mild infarcts or hemorrhage were observed in the kidneys of 3 and 2 sheep, respectively, but this was not sufficient to create significant changes in renal function. Infarction and hemorrhage were not observed in the other major organs.
Unlike platelet count or hemolysis, white blood cell counts did increase progressively during the course of the experiment. Counts were significantly higher than BL at D5 and D7, but no evidence of infection was found at wound sites in any but the first sheep. Blood bacterial culture was not performed, so it is not known whether the increase in white blood cell counts was caused by infection or a chronic inflammatory response caused by blood contact with the foreign surface of the artificial lung. In either case, the relatively low level of inflammation did not result in typical inflammatory complications such as hypotension, edema, or respiratory dysfunction in most sheep. On the other hand, sheep with preexisting pneumonia or sepsis experienced a further decrease in natural lung respiratory function while on the TAL. It may be that their respiratory dysfunction was the result of the combined force of their initial infection, coupled with an inflammatory response from surgical trauma and the blood–biomaterial interaction.
Comparison With Previous Long-Term Studies
This study is the first to examine long-term TAL attachment in an in-parallel, PA to LA configuration. However, three studies have examined in-series, PA to PA attachment of the Biolung for periods between 3 and 7 days [7–9]. The first study examined attachment in 8 sheep for periods of up to 168 hours [7]. That study, using an earlier MC3 TAL prototype, resulted in 2 sheep surviving for 168 hours, 2 deaths attributable to right ventricular failure, and 4 deaths attributable to blood loss either from device damage (3 sheep) or anastomosis bleeding (1 sheep). The device was redesigned to improve hemodynamics and tested again for a period of 72 hours in 7 sheep [8]. The devices were nearly identical to those tested in the current study but used Celgard porous polypropylene fibers. In that study, right ventricular function was improved, but 4 deaths still occurred because of bleeding complications and 1 death occurred because of cardiac failure of unknown origin. The third study tested the device up to 105 hours using a severe acute respiratory distress syndrome model. Impressively, 6 sheep survived, with 1 death attributable to pulmonary hypertension and 1 death attributable to sepsis from pneumonia [9].
In all of the PA to PA attachment studies, gas exchange was excellent, and in the third study, redesign of the TAL to reduce its impedance led to animals with stable hemodynamics for the experimental course. No clots were found in the devices or in the pulmonary vasculature. This difference from the current study is likely attributable to a slightly higher activated clotting time range (200 to 300 seconds) and the higher flow rate through the devices. Device flows were approximately 4 L/min in the PA to PA studies versus 2 L/min in the PA to LA studies. Blood flow velocity within the Biolung fiber bundle is relatively slow, with shear stresses averaging between 1 and 10 dynes/cm2 [14], and thus stasis is the predominant cause of coagulation.
Conclusions
The MC3 Biolung is capable of 1 week of respiratory support in the in-parallel, PA to LA attachment mode.
|
Acknowledgments
|
|---|
This work was supported from a grant from the National Institutes of Health (R01 HL69420).
|
References
|
|---|
- OPTN/SRTR 2004 Annual Report. Available at: http://www.optn.org.
- Luhr OR, Antonsen K, Karlsson M, et al. Incidence and mortality after acute respiratory failure and acute respiratory distress syndrome in Sweden, Denmark, and Iceland Am J Respir Crit Care Med 1999;159:1849-1861.[Abstract/Free Full Text]
- Ware LB, Matthay MA. The acute respiratory distress syndrome N Engl J Med 2000;342:1334-1349.[Free Full Text]
- Vincent L, Sakr Y, Ranieri VM. Epidemiology and outcome of acute respiratory failure in intensive care unit patients Crit Care Med 2003;31(Suppl):S296-S299.[Medline]
- Plotz FB, Slutsky AS, van Vught AJ, Heijnen CJ. Ventilator-induced lung injury and multiple system organ failure: a critical review of facts and hypotheses Intensive Care Med 2004;30:1865-1872.[Medline]
- The Acute Respiratory Distress Syndrome Network Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome N Engl J Med 2000;342:1301-1308.[Abstract/Free Full Text]
- Zwischenberger JB, Wang D, Lick SD, Alpard SK, Deyo DJ, Chambers SD. The paracorporeal artificial lung improves 5-day outcomes from lethal smoke/burn induced acute respiratory distress syndrome in sheep Ann Thorac Surg 2002;74:1011-1018.[Abstract/Free Full Text]
- Lick SD, Zwischenberger JB, Wang D, Deyo DJ, Alpard SK, Chambers SD. Improved right heart function with a compliant inflow artificial lung in series with the pulmonary circulation Ann Thorac Surg 2001;72:899-904.[Abstract/Free Full Text]
- Lick SD, Zwischenberger JB, Alpard SK, Witt SA, Deyo DM, Merz SI. Development of an ambulatory artificial lung in an ovine survival model ASAIO J 2001;47:486-491.[Medline]
- Sato H, McGillicuddy JW, Griffith GW, et al. Effects of artificial lung compliance on in vivo pulmonary system hemodynamics ASAIO J 2006;52:248-256.[Medline]
- Wiesenack C, Wiesner G, Keyl C, et al. In vivo uptake and elimination of isoflurane by different membrane oxygenators during cardiopulmonary bypass Anesthesiology 2002;97:133-138.[Medline]
- Cook KE, Perlman CE, Seipelt R, Backer CL, Mavroudis C, Mockros LF. Hemodynamic and gas transfer properties of a compliant thoracic artificial lung ASAIO J 2005;51:404-411.[Medline]
- Brooks DL, Tillman PC, Niemi SM. Ungulates as laboratory animalsIn: Fox JG, Cohen BJ, Loew FM, editors. Laboratory animal medicine. Orlando, FL: Academic Press; 1984. pp. 273-295.
- Cook KE, Maxhimer J, Leonard DJ, Mavroudis C, Backer CL, Mockros LF. Platelet and leukocyte activation and design consequences for thoracic artificial lungs ASAIO J 2002;48:620-630.[Medline]
This article has been cited by other articles:
|
|
|
|
 
H. Sato, C. M. Hall, N. G. Lafayette, J. R. Pohlmann, N. Padiyar, J. M. Toomasian, J. W. Haft, and K. E. Cook
Thirty-Day In-Parallel Artificial Lung Testing in Sheep
Ann. Thorac. Surg.,
October 1, 2007;
84(4):
1136 - 1143.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
