Cardiopulmonary Support and Physiology
`
`De Wet et al
`
`Inhaled prostacyclin is safe, effective, and affordable in
`patients with pulmonary hypertension, right heart dysfunction,
`and refractory hypoxemia after cardiothoracic surgery
`
`Charl J. De Wet, MBChBa,b
`David G. Affleck, MDb
`Eric Jacobsohn, MBChB, MHPE, FRCPCa,b
`Michael S. Avidan, MBBCha,b
`Heidi Tymkew, MHSa,b
`Laureen L. Hill, MDa,b
`Paul B. Zanaboni, MD, PhDa,b
`Nader Moazami, MDb
`Jennifer R. Smith, PharmDc
`
`Background: The purpose of this study was to describe our institutional experience
`in using inhaled prostacyclin as a selective pulmonary vasodilator in patients with
`pulmonary hypertension, refractory hypoxemia, and right heart dysfunction after
`cardiothoracic surgery.
`
`Methods: Between February 2001 and March 2003, cardiothoracic surgical patients
`with pulmonary hypertension (mean pulmonary artery pressure ⬎30 mm Hg or
`systolic pulmonary artery pressure ⬎40 mm Hg), hypoxemia (PaO2/fraction of
`inspired oxygen ⬍150 mm Hg), or right heart dysfunction (central venous pressure
`⬎16 mm Hg and cardiac index ⬍2.2 L · min⫺1 · m⫺2) were prospectively
`administered inhaled prostacyclin at an initial concentration of 20,000 ng/mL and
`then weaned per protocol. Hemodynamic variables were measured before the
`initiation of inhaled prostacyclin, 30 to 60 minutes after initiation, and again 4 to 6
`hours later.
`
`Results: One hundred twenty-six patients were enrolled during the study period. At
`both time points, inhaled prostacyclin significantly decreased the mean pulmonary
`artery pressure without altering the mean arterial pressure. The average length of
`time on inhaled prostacyclin was 45.6 hours. There were no adverse events attrib-
`utable to inhaled prostacyclin. The average cost for inhaled prostacyclin was $150
`per day. Compared with nitric oxide, which costs $3000 per day, the potential cost
`savings over this period were $681,686.
`
`Conclusions: Inhaled prostacyclin seems to be a safe and effective pulmonary
`vasodilator for cardiothoracic surgical patients with pulmonary hypertension, re-
`fractory hypoxemia, or right heart dysfunction. Overall, inhaled prostacyclin sig-
`nificantly decreases mean pulmonary artery pressures without altering the mean
`arterial pressure. Compared with nitric oxide, there is no special equipment required
`for administration or toxicity monitoring, and the cost savings are substantial.
`
`Because of a potential conflict of interest related to this manuscript on the part
`of our editors, Dr David A. Fullerton served as guest section editor, assigned
`reviewers, and made editorial decisions or recommendations leading to its
`acceptance for publication.
`
`Jacobsohn, Affleck, Moazami, Smith, and
`Tymkew (standing, left to right). De Wet, Hill,
`and Avidan (sitting, left to right).
`
`From the Department of Anesthesiologya
`and Division of Cardiothoracic Surgery,b
`Washington University School of Medicine
`and Barnes Jewish Hospital, St Louis, Mo,
`and Department of Pharmacy,c Barnes Jew-
`ish Hospital, St Louis, Mo.
`Funding was provided by a grant from the
`Barnes Jewish Hospital Foundation.
`Presented at the Society of Cardiovascular
`Anesthesiology Meeting, Miami, Fla, May
`2003, and the Twenty-ninth Annual Meeting
`of The Western Thoracic Surgical Associa-
`tion, Carlsbad, Calif, June 18-21, 2003.
`
`Received for publication June 17, 2003;
`revisions requested Sept 11, 2003; revisions
`received Nov 5, 2003; accepted for publi-
`cation Dec 1, 2003.
`Jacobsohn,
`Address
`for
`reprints: Eric
`MBChB, MHPE, FRCPC, Washington
`University School of Medicine, Department
`of Anesthesiology, 660 Euclid Ave, Cam-
`pus Box 8054, St Louis, MO 63110 (E-
`mail: jacobsoe@msnotes.wustl.edu).
`J Thorac Cardiovasc Surg 2004;127:1058-67
`
`0022-5223/$30.00
`
`Copyright © 2004 by The American Asso-
`ciation for Thoracic Surgery
`
`doi:10.1016/j.jtcvs.2003.11.035
`
`1058 The Journal of Thoracic and Cardiovascular Surgery ● April 2004
`
`CSP
`
`Liquidia's Exhibit 1025
`Page 1
`
`

`

`CSP
`
`De Wet et al
`
`Cardiopulmonary Support and Physiology
`
`Pulmonary hypertension, right ventricular dys-
`
`function, and perioperative hypoxemia are
`common problems that necessitate treatment
`in a cardiothoracic intensive care unit. Inhaled
`nitric oxide (iNO) was the first agent shown to
`be a selective pulmonary artery vasodilator,
`and it has also been shown to improve oxygenation in
`patients with acute lung injury and adult respiratory distress
`syndrome (ARDS).1,2 Nitric oxide, however, has become
`prohibitively expensive and has several associated toxicities
`that necessitate monitoring, which further increases the
`cost.3
`In the search for a selective pulmonary artery vasodila-
`tor, both as an alternative and possibly as a complement to
`iNO, several drugs administered via the inhalational route
`have been described in animal models and humans. These
`include inhaled sodium nitroprusside, nitroglycerin, class 5
`phosphodiesterase inhibitors (such as zaprinast and silde-
`nafil), milrinone, prostaglandin E1 (PGE1; alprostadil), PGI2
`(prostacyclin), and iloprost (the stable analog of PGI2).4 The
`use of intravenous PGI2 was first described in 1978 in
`canine experiments.5 As opposed to PGE1, PGI2 was shown
`not to have any pulmonary inactivation and was therefore
`10 times more potent than a systemic vasodilator, even
`␣ had few
`though its metabolite 6-keto-prostaglandin F1
`systemic vasodilator properties. However,
`intravenous
`PGI2, as opposed to intravenous PGE1, became an impor-
`tant pulmonary vasodilator in the treatment of pulmonary
`hypertension.6 However, in some patients with pulmonary
`hypertension, the intravenous doses required to decrease
`pulmonary artery pressures (PAPs) are often so high that
`significant systemic hypotension occurs. In 1993, Welte and
`colleagues7 reported that inhaled PGI2 (iPGI2) resulted in
`selective pulmonary artery vasodilation in dogs. It has been
`shown to be as effective as iNO in reducing pulmonary
`vascular resistance in heart transplant candidates,8,9 in de-
`creasing PAPs in primary and secondary pulmonary hyper-
`tension,10 and in improving right ventricular function in
`animals with hypoxic pulmonary vasoconstriction11; it is
`also effective as a selective pulmonary artery vasodilator,
`with improvement
`in oxygenation in patients with
`ARDS.12-14 Compared with iNO, it is less expensive, is
`easier to administer, is relatively free of side effects, and
`requires no special toxicity monitoring.
`The manufacturer has not sought Food and Drug Admin-
`istration (FDA) approval for inhalational administration of
`PGI2. The objective of this case series was to report our
`institutional experience with iPGI2 in critically ill cardio-
`thoracic surgical patients. Our secondary aim was to deter-
`mine whether it is easy to use during surgery as well as in
`the intensive care setting and to perform a cost analysis
`comparing it with iNO.
`
`Methods
`The Human Studies Committee at Washington University Medical
`School approved the study, and all enrolled patients signed written,
`informed consent. Before initiation of the study, the principal
`investigator (E.J.) obtained an investigational new drug application
`from the FDA. This was done so we could investigate a new route
`of administration of a previously approved drug.
`One hundred twenty-six patients (67 men and 59 women with
`a mean age of 56 ⫾ 15 years) were enrolled in this prospective,
`interventional study of iPGI2. The inclusion criteria were cardio-
`thoracic surgery patients with pulmonary hypertension, right ven-
`tricular dysfunction, or refractory hypoxemia in the perioperative
`period. Pulmonary hypertension was defined as a mean PAP
`(MPAP) 30 mm Hg or more or systolic PAP 40 mm Hg or more.
`Right ventricular failure was defined as a central venous pressure
`(CVP) 16 mm Hg or more and cardiac index less than 2.2 L ·
`min⫺1 · m⫺2. Refractory hypoxemia was defined as a ratio of
`arterial partial oxygen pressure to fraction of inspired oxygen
`(PaO2/FIO2 ratio) less than 150 mm Hg. Clinicians first attempted
`to improve oxygenation by conventional methods. These included
`strategies such as: 1) increasing the inspired oxygen tension, 2)
`optimizing functional residual capacity with the best positive end-
`expiratory pressure (PEEP) and recruitment maneuvers, 3) opti-
`mizing the inspiration-expiration ratio, 4) selecting the most ap-
`propriate mode of mechanical ventilation,
`and 5) using
`neuromuscular blockade when indicated. The exclusion criteria
`were patients younger than 18 years of age, pregnant patients, and
`those with a known allergy or sensitivity to PGI2 or the diluent
`(glycine). Hemodynamic variables were measured before the ini-
`tiation of iPGI2, 30 to 60 minutes after the initiation, and every 6
`hours thereafter. The study was designed to optimize patient safety
`and ease of use while maintaining clinician autonomy. Clinicians
`were free to titrate inotropes, vasoconstrictors, or vasodilators to
`maintain mean arterial blood pressure (MAP), cardiac output, or
`both. Side effects attributable to the use of iPGI2 were recorded.
`All adverse events were reported to the Human Studies Committee
`and then reviewed by a data safety monitoring committee for a
`further in-depth review.
`PGI2 is supplied as a sodium salt of epoprostenol (Flolan;
`Glaxo Wellcome Inc, Research Triangle Park, NC). It is reconsti-
`tuted in 50 mL of glycine buffer diluent (sterile diluent for Flolan)
`to a final concentration of 20,000 ng/mL. It is then nebulized by
`using a continuous nebulizer system (MiniHEART nebulizer;
`Westmed, Tucson, Ariz). This is attached to the inspiratory limb of
`the ventilator circuit or via face mask with a Venturi attachment
`for aerosolization. For continuous administration, a 60-mL syringe
`of PGI2 (20,000 ng/mL) is attached to a standard intravenous pump
`and delivered at a constant rate of 8 mL/h to the nebulizer cham-
`ber. The nebulizer chamber is primed with 15 mL of the PGI2
`solution, and at a nebulizing oxygen flow rate of 2 to 3 L/min,
`approximately 8 mL/h is nebulized. The glycine buffer diluent is
`“sticky,” and we therefore decided to empirically change filters on
`the ventilator every 2 hours to prevent ventilator valve malfunc-
`tion. To support continuous nebulization, a heated wire circuit was
`added to the ventilator. The iPGI2 concentration was weaned by
`50% every 2 to 4 hours until a concentration of 2500 ng/mL was
`reached (20,000 to 10,000 to 5000 to 2500 ng/mL), as long as the
`patient did not have a negative response. A negative response was
`
`The Journal of Thoracic and Cardiovascular Surgery ● Volume 127, Number 4 1059
`
`Liquidia's Exhibit 1025
`Page 2
`
`

`

`Cardiopulmonary Support and Physiology
`
`De Wet et al
`
`Figure 1. Effect of iPGI2 on MPAP, MAP, the MPAP/MAP ratio, and cardiac index for all patients: P < .001. The
`smaller number of patients at the 4- to 6-hour time periods are compared with their own baseline. The numbers
`at this time period are smaller because some of the original patients were no longer receiving iPGI2 or had
`incomplete data.
`
`defined as an increase in PAP by 15% or a decline in cardiac index
`or the PaO2/FIO2 ratio by 10%. Systemic arterial pressure was
`closely monitored, and the PGI2 dose was adjusted accordingly if
`a clinical change in blood pressure was observed.
`The major aim of this study was to evaluate the immediate and
`sustained effects of iPGI2 on PAPs in patients with pulmonary
`hypertension, right ventricular dysfunction, and refractory hypox-
`emia. Previous studies have shown reductions in MPAP of at least
`8 mm Hg (SD, 7.5 mm Hg) with iPGI2. On the basis of these
`findings, 9 patients would be required to demonstrate a mean
`decrease in PAP of 8 mm Hg (SD, 8 mm Hg) with a power of 90%
`and a P value of .05. With the large number of patients we were
`planning to recruit (⬎100), the study was adequately powered to
`detect effects of iPGI2 on PAP in each subgroup.
`The Shapiro-Wilke test of normality was used to assess the
`distribution of the continuous data. Variables were normally dis-
`tributed. Paired Student t tests were used to evaluate the changes
`seen in hemodynamic and oxygenation parameters. The SPSS
`statistical package (SPSS, Chicago, Ill) was used for all analyses.
`
`A post hoc analysis of patients with refractory hypoxemia was
`performed.
`
`Results
`One hundred twenty-six patients were prospectively en-
`rolled during the study period. The demographics are shown
`in Table 1. The average time on iPGI2 was 45.6 hours
`(range, 0.1-390 hours). Tables 2 through 7 show the analysis
`for the 3 groups of patients that were enrolled. At both time
`points (after initiation and at 4-6 hours), iPGI2 significantly
`decreased the MPAP. The ratio of MPAP to MAP improved
`significantly at both time points. Further, no significant
`changes were observed in MAP at the same time points,
`demonstrating the selective pulmonary effects of iPGI2. The
`effect of iPGI2 on all hemodynamic parameters is reported
`in Tables 8 and 9. Although the entire group did not show
`an improvement in the PaO2/FIO2 ratio, post hoc analysis
`
`1060 The Journal of Thoracic and Cardiovascular Surgery ● April 2004
`
`CSP
`
`Liquidia's Exhibit 1025
`Page 3
`
`

`

`De Wet et al
`
`Cardiopulmonary Support and Physiology
`
`CSP
`
`Data
`
`56 ⫾ 15
`77 ⫾ 22
`
`67
`59
`78
`17
`11
`17
`12
`15
`6
`43
`5
`
`110
`14
`32
`
`TABLE 1. Demographics (n ⴝ 126)
`Variable
`
`Age (y) (mean ⫾ SD)
`Weight, (kg) (mean ⫾ SD)
`Sex
`Male
`Female
`Cardiac operation
`CABG
`CABG ⫹ valve
`Valve
`Heart transplantation
`LVAD
`Other*
`Lung transplantation
`Thoracic operation
`Enrollment criteria†
`Pulmonary hypertension
`RV dysfunction
`Refractory hypoxemia
`
`CABG, Coronary artery bypass graft; LVAD, left ventricular assist device;
`RV, right ventricular.
`*Other cardiac operations included atrial septal defect repair, maze-atrial
`septal defect, pulmonary thromboendarterectomy, post-extracorporeal
`membrane oxygenator removal after redo-aortic valve replacement, tho-
`racoabdominal aortic aneurysm repair, post–chest wound re-exploration.
`†Note that some patients met more than 1 of the 3 inclusion criteria.
`
`put. Our study also demonstrates that there did not seem to
`be any appreciable tolerance to any of the beneficial effects
`after 4 to 6 hours of administration.
`Inhaled PGI2 binds to subunits of the prostaglandin
`G/protein receptor, which results in an increase in cyclic
`adenosine monophosphate,23 as opposed to iNO, which
`results in an increase in cyclic guanosine monophosphate.3
`Many of iNO’s associated toxicities are thought to be re-
`lated to the increase in cyclic guanosine monophosphate
`that interferes with normal cellular proliferation, including
`DNA strand breaks and potential mutagenic base alter-
`ations.3 PGI2 has a short half-life of 5 minutes. It is rapidly
`hydrolyzed at acid or physiologic pH to 6-keto-prostaglan-
`␣,24,25 which has been shown to have few vasodilator
`din F1
`properties.5 As opposed to PGE1 (alprostadil), which is
`commonly used to prevent reperfusion injury or as an in-
`travenous pulmonary artery vasodilator in the perioperative
`period, intravenous PGI2 is not inactivated in the pulmonary
`circulation. Because of this lack of pulmonary inactivation,
`PGI2 is 10 times more potent as a systemic vasodilator than
`PGE1.5
`As with iNO, there are theoretical concerns regarding the
`potential antiplatelet effects of PGI2 and its metabolite 6-ke-
`␣.24,26 Platelet/endothelial cell adhesion
`to-prostaglandin F1
`is regulated by endothelial cell– derived mediators, includ-
`ing PGI2 and endothelium-derived relaxing factor. PGI2
`
`showed that in patients with refractory hypoxemia, there
`was a significant improvement in the PaO2/FIO2 ratio (Tables
`2 and 3). The initial PaO2/FIO2 increased from 85 ⫾ 33 to
`158 ⫾ 114 (P ⫽ .001). For those patients with refractory
`hypoxemia who remained on the treatment at 4 to 6 hours,
`the PaO2/FIO2 ratio increased from 95 ⫾ 36 to 186 ⫾ 111
`(P ⫽ .001).
`There was a substantial cost savings compared with iNO.
`The total cost for administering iPGI2 is approximately
`$150 per day. The cost for iNO is $125 per hour, or $3000
`per day. Total patient hours of iPGI2 administration were
`5740.50 hours. The total iPGI2 cost was $35,878, versus a
`calculated nitric oxide cost of $717,564. This amounts to a
`calculated cost savings of $681,686.
`There were few adverse events attributable to iPGI2. No
`patients complained of facial flushing. There was no signif-
`icant systemic hypotension. There were no clinically signif-
`icant bleeding events. Within 30 days, there were 2 re-
`explorations for postoperative bleeding (2/126; 1.6%), 6
`cases of renal failure (6/126; 1.6%), and 1 new stroke
`(1/126; 0.8%). There were 2 perioperative cardiac arrests
`(2/126; 1.6%). There was 1 serious adverse event related to
`the use of the “sticky” glycine diluent. An exhalation valve
`on a ventilator became stuck, and this resulted in significant
`auto-PEEP and hypotension. The problem was diagnosed
`early, and there were no sequelae. On several patients, the
`mainstream end-tidal CO2 analyzers ceased to function be-
`cause of excess moisture accumulation. The overall 30-day
`mortality rate was 12.7%.
`Discussion
`Inhaled aerosolized PGI2 has been described in a number of
`small human studies for use in patients with ARDS,12-15
`primary and secondary pulmonary hypertension,10,16 and
`right ventricular failure or dysfunction9,17 and to improve
`oxygenation.18,19 It has also been shown to be as effective
`as iNO.8,12,13,20-22 However, centers that use iPGI2 use it via
`a non–FDA-approved route of administration. Similarly,
`iNO used for indications other than pulmonary hypertension
`is used for a non–FDA-approved indication. After obtaining
`a FDA investigational new drug application, we set out to
`document and demonstrate its efficacy, safety, ease of ad-
`ministration, and cost in a large cardiothoracic surgical
`operating room and intensive care unit. We found that iPGI2
`was effective as a selective pulmonary artery vasodilator in
`patients with pulmonary hypertension and right ventricular
`dysfunction, and improved oxygenation in patients with
`refractory hypoxemia. It acutely decreased PAPs while
`MAP was maintained. This resulted in an improvement in
`the MPAP/MAP ratio. Inhaled PGI2 was also effective in
`improving oxygenation regardless of the etiology of hypox-
`emia. In patients with poor oxygenation, iPGI2 not only
`improved oxygenation, but was also effective as a pulmo-
`nary vasodilator while maintaining MAP and cardiac out-
`
`The Journal of Thoracic and Cardiovascular Surgery ● Volume 127, Number 4 1061
`
`Liquidia's Exhibit 1025
`Page 4
`
`

`

`Cardiopulmonary Support and Physiology
`
`De Wet et al
`
`TABLE 2. Hemodynamic parameters for patients with refractory hypoxemia: baseline and after iPGI2
`Variable
`n
`Before iPGI2
`After iPGI2
`
`PaO2/FIO2 ratio
`MAP (mm Hg)
`PAP (mm Hg)
`MPAP/MAP ratio
`CI (L · min⫺1 · m⫺2)
`CVP (mm Hg)
`Wedge (mm Hg)
`PVR (dynes 䡠 s 䡠 cm⫺5)
`SVR (dynes 䡠 s 䡠 cm⫺5)
`
`85 ⫾ 33
`77 ⫾ 11
`32 ⫾ 9
`0.43 ⫾ 0.13
`2.7 ⫾ 1.1
`17 ⫾ 6
`21 ⫾ 4
`191 ⫾ 47
`1315 ⫾ 655
`
`158 ⫾ 114
`75 ⫾ 10
`28 ⫾ 9
`0.35 ⫾ 0.20
`2.6 ⫾ 0.7
`17 ⫾ 7
`17 ⫾ 2
`179 ⫾ 146
`1163 ⫾ 364
`
`27
`32
`26
`26
`13
`24
`2
`2
`8
`
`P value
`
`.001
`.331
`.014
`.039
`.504
`.358
`.5
`.892
`.297
`
`iPGI2, Inhaled prostacyclin; MAP, mean arterial blood pressure; PAP, pulmonary artery pressure; MPAP, mean pulmonary artery pressure; CI, cardiac index;
`CVP, central venous pressure; PVR, pulmonary vascular resistance; SVR, systemic vascular resistance.
`
`P value
`
`.001
`.465
`⬍.001
`.001
`.635
`.015
`.41
`.307
`.708
`
`TABLE 3. Hemodynamic parameters for patients with refractory hypoxemia: baseline and after 4 to 6 h oniPGI 2
`Variable
`4 to 6 h after
`n
`Before iPGI2
`
`CSP
`
`PaO2/FIO2 ratio
`MAP (mm Hg)
`PAP (mm Hg)
`MPAP/MAP ratio
`CI (L · ⫺1 · m⫺2)
`CVP (mm Hg)
`Wedge (mm Hg)
`PVR (dynes 䡠 s 䡠 cm⫺5)
`SVR (dynes 䡠 s 䡠 cm⫺5)
`
`95 ⫾ 36
`76 ⫾ 11
`32 ⫾ 7
`0.44 ⫾ 0.11
`2.4 ⫾ 0.5
`18 ⫾ 5
`20 ⫾ 4
`212 ⫾ 78
`1147 ⫾ 414
`
`186 ⫾ 111
`78 ⫾ 12
`26 ⫾ 9
`0.34 ⫾ 0.15
`2.6 ⫾ 0.6
`15 ⫾ 6
`16 ⫾ 0
`139 ⫾ 23
`1069 ⫾ 218
`
`17
`25
`22
`22
`10
`22
`2
`2
`7
`
`iPGI2, Inhaled prostacyclin; MAP, mean arterial blood pressure; PAP, pulmonary artery pressure; MPAP, mean pulmonary artery pressure; CI, cardiac index;
`CVP, central venous pressure; PVR, pulmonary vascular resistance; SVR, systemic vascular resistance.
`
`TABLE 4. Hemodynamic parameters for patients with pulmonary hypertension: baseline and after the initiation of iPGI2
`Variable
`n
`P value
`Before PGI2
`After PGI2
`
`PaO2/FIO2 ratio
`MAP (mm Hg)
`PAP (mm Hg)
`PAP/MAP ratio
`CI (L · min⫺1 · m⫺2)
`CVP (mm Hg)
`Wedge (mm Hg)
`PVR (dynes 䡠 s 䡠 cm⫺5)
`SVR (dynes · s 䡠 cm⫺5)
`
`300 ⫾ 184
`76 ⫾ 12
`36 ⫾ 9
`0.49 ⫾ 0.14
`2.6 ⫾ 0.9
`18 ⫾ 7
`18 ⫾ 2
`358 ⫾ 164
`1199 ⫾ 483
`
`319 ⫾ 168
`78 ⫾ 11
`30 ⫾ 8
`0.40 ⫾ 0.12
`2.8 ⫾ 0.9
`18 ⫾ 8
`18 ⫾ 3
`246 ⫾ 122
`1093 ⫾ 390
`
`55
`109
`107
`107
`30
`78
`11
`11
`19
`
`.322
`.148
`⬍.001
`⬍.001
`.054
`.22
`.933
`.005
`.291
`
`iPGI2, Inhaled prostacyclin; MAP, mean arterial blood pressure; PAP, pulmonary artery pressure; CI, cardiac index; CVP, central venous pressure; PVR,
`pulmonary vascular resistance; SVR, systemic vascular resistance.
`
`activates platelet adenylate cyclase and augments cyclic
`adenosine monophosphate formation by way of specific
`membrane receptors. van Heerden and colleagues14 demon-
`strated that even though significant levels of 6-keto-prosta-
`␣ could be demonstrated in patients receiving
`glandin F1
`iPGI2 for ARDS, there was no effect on platelet aggregation
`as measured by the response to adenosine diphosphate. In a
`double-blind, placebo-controlled randomized study over 4
`to 6 hours of administration of iPGI2, Haraldsson and as-
`sociates27 reported that even though antiplatelet effects
`could be detected by in vitro measurements such as platelet
`
`aggregation and thromboelastography, these effects were
`not clinically detected as measured by chest tube output and
`bleeding times after on-pump cardiac surgery. On the basis
`of these findings, we therefore did not monitor blood levels
`of either PGI2 or its metabolite. Our results corroborate this
`because the re-exploration rate for bleeding was low. In
`contrast to the increasing bleeding times, there is also a
`theoretical concern that when platelets are exposed to ex-
`tended periods of PGI2 or its chemical analogs it may result
`in a time- and dose-dependent desensitization of the pros-
`tacyclin receptor.28 Darius and associates28 showed that
`
`1062 The Journal of Thoracic and Cardiovascular Surgery ● April 2004
`
`Liquidia's Exhibit 1025
`Page 5
`
`

`

`De Wet et al
`
`Cardiopulmonary Support and Physiology
`
`TABLE 5. Hemodynamic parameters for patients with pulmonary hypertension: baseline and after 4 to 6 h oniPGI 2
`Variable
`4 to 6 h after
`n
`P value
`Before PGI2
`
`PaO2/FIO2 ratio
`MAP (mm Hg)
`PAP (mm Hg)
`PAP/MAP ratio
`CI (L · min⫺1 · m⫺2)
`CVP (mm Hg)
`Wedge (mm Hg)
`PVR (dynes 䡠 s 䡠 cm⫺5)
`SVR (dynes 䡠 s 䡠 cm⫺5)
`
`282 ⫾ 164
`76 ⫾ 11
`36 ⫾ 9
`0.48 ⫾ 0.13
`2.6 ⫾ 0.8
`19 ⫾ 7
`17 ⫾ 4
`212 ⫾ 40
`1133 ⫾ 572
`
`302 ⫾ 166
`78 ⫾ 11
`25 ⫾ 8
`0.33 ⫾ 0.13
`2.9 ⫾ 0.8
`13 ⫾ 6
`16 ⫾ 5
`157 ⫾ 38
`931 ⫾ 249
`
`37
`89
`86
`86
`33
`67
`6
`5
`22
`
`.442
`.16
`⬍.001
`⬍.001
`.042
`⬍.001
`.615
`.078
`.131
`
`iPGI2, Inhaled prostacyclin; MAP, mean arterial blood pressure; PAP, pulmonary artery pressure; CI, cardiac index; CVP, central venous pressure; PVR,
`pulmonary vascular resistance; SVR, systemic vascular resistance.
`
`CSP
`
`TABLE 6. Hemodynamic parameters for patients with right ventricular dysfunction: baseline and after the initiation of iPGI2
`Variable
`n
`P value
`Before iPGI2
`After iPGI2
`
`PaO2/FIO2 ratio
`MAP (mm Hg)
`PAP (mm Hg)
`MPAP/MAP ratio
`CI (L · min⫺1 · m⫺2)
`CVP (mm Hg)
`Wedge (mm Hg)
`PVR (dynes 䡠 s 䡠 cm⫺5)
`SVR (dynes · s 䡠 cm⫺5)
`
`287 ⫾ 157
`78 ⫾ 11
`34 ⫾ 10
`0.43 ⫾ 0.12
`1.9 ⫾ 0.3
`21 ⫾ 5
`18 ⫾ 2
`321 ⫾ 130
`1338 ⫾ 549
`
`357 ⫾ 134
`81 ⫾ 11
`29 ⫾ 10
`0.35 ⫾ 0.11
`2.2 ⫾ 0.6
`19 ⫾ 6
`17 ⫾ 3
`243 ⫾ 130
`1104 ⫾ 407
`
`7
`16
`15
`15
`12
`16
`6
`6
`10
`
`.11
`.356
`.007
`.002
`.036
`.058
`.135
`.05
`.014
`
`iPGI2, Inhaled prostacyclin; MAP, mean arterial blood pressure; PAP, pulmonary artery pressure; MPAP, mean pulmonary artery pressure; CI, cardiac index;
`CVP, central venous pressure; PVR, pulmonary vascular resistance; SVR, systemic vascular resistance.
`
`TABLE 7. Hemodynamic parameters for patients with right ventricular dysfunction: baseline and after 4 to 6 h oniPGI 2
`Variable
`4 to 6 h after
`n
`P value
`Before PGI2
`
`PaO2/FIO2 ratio
`MAP (mm Hg)
`PAP (mm Hg)
`MPAP/MAP ratio
`CI (L · min⫺1 · m⫺2)
`CVP (mm Hg)
`Wedge (mm Hg)
`PVR (dynes 䡠 s 䡠 cm⫺5)
`SVR (dynes 䡠 s 䡠 cm⫺5)
`
`250 ⫾ 86
`80 ⫾ 11
`34 ⫾ 11
`0.42 ⫾ 0.12
`1.9 ⫾ 0.3
`20 ⫾ 5
`17 ⫾ 1.4
`202 ⫾ 18
`1231 ⫾ 518
`
`284 ⫾ 105
`72 ⫾ 7
`24 ⫾ 9
`0.34 ⫾ 0.14
`2.6 ⫾ 0.6
`14 ⫾ 6
`9.5 ⫾ 5
`147 ⫾ 8
`999 ⫾ 281
`
`5
`13
`13
`13
`7
`13
`2
`2
`9
`
`.552
`.056
`.002
`.032
`.038
`.006
`.205
`.212
`.216
`
`iPGI2, Inhaled prostacyclin; MAP, mean arterial blood pressure; PAP, pulmonary artery pressure; MPAP, mean pulmonary artery pressure; CI, cardiac index;
`CVP, central venous pressure; PVR, pulmonary vascular resistance; SVR, systemic vascular resistance.
`
`there was an in vitro platelet PGI2 receptor desensitization
`that caused a marked augmentation of platelet/endothelial
`cell adhesion. This could theoretically cause thrombus for-
`mation. Our data and safety monitoring committee reviewed
`all adverse events and found no hematologic adverse events
`attributable to iPGI2.
`Inhaled PGI2 is a potent vasodilator.5 However, as in
`other studies, we did not detect significant systemic hypo-
`tension. It has subsequently been confirmed that iPGI2 has
`no systemic hypotensive effects in doses up to 50 ng · kg⫺1
`· min⫺1, even though significant levels of its metabolite
`
`␣ were detected.14 At
`the dose
`6-keto-prostaglandin F1
`ranges used in our study and in others, not enough PGI2
`reaches the systemic circulation to cause significant vaso-
`dilation. To maintain ease of administration and considering
`the principles of inhalational drug delivery, we elected to
`use a concentration-based rather than a weight-based pro-
`tocol. Even at the highest concentration (20,000 ng/mL),
`this amounted to an average of 37 ng · kg⫺1 · min⫺1. On the
`basis of previously published dose-response studies, this
`dose is well below the recommended safe inhalational dose
`of 50 ng · kg⫺1 · min⫺1.14,29 The beneficial effects of the
`
`The Journal of Thoracic and Cardiovascular Surgery ● Volume 127, Number 4 1063
`
`Liquidia's Exhibit 1025
`Page 6
`
`

`

`Cardiopulmonary Support and Physiology
`
`De Wet et al
`
`TABLE 8. Effect of inhaled prostacyclin (iPGI2) on hemodynamic parameters: baseline and after the initiation of iPGI2*
`Variable
`n
`P value
`Before PGI2
`After PGI2
`
`260 ⫾ 186
`77 ⫾ 12
`35 ⫾ 9
`0.47 ⫾ 0.14
`2.6 ⫾ 0.9
`18 ⫾ 7
`18 ⫾ 2
`335 ⫾ 160
`1255 ⫾ 534
`70 ⫾ 14
`
`290 ⫾ 173
`78 ⫾ 11
`30 ⫾ 8
`0.39 ⫾ 0.13
`2.7 ⫾ 0.9
`17 ⫾ 8
`18 ⫾ 3
`246 ⫾ 112
`1121 ⫾ 405
`72 ⫾ 10
`
`PaO2/FIO2 ratio
`MAP (mm Hg)
`MPAP (mm Hg)
`MPAP/MAP ratio
`CI (L · min⫺1 · m⫺2)
`CVP (mm Hg)
`Wedge (mm Hg)
`PVR (dynes 䡠 s 䡠 cm⫺5)
`SVR (dynes 䡠 s 䡠 cm⫺5)
`SvO2 (%)
`FIO2, Fraction of inspired oxygen; MAP, mean arterial blood pressure; MPAP, mean pulmonary artery pressure; CI, cardiac index; CVP, central venous
`pressure; PVR, pulmonary vascular resistance; SVR, systemic vascular resistance; SvO2, mixed venous oxygen saturation.
`*The small n in the number of wedge pressures and cardiac output measurements (and hence derived parameters, such as PVR and SVR) reflects the
`inability to obtain these parameters for a variety of reasons, including tricuspid regurgitation (inaccurate, meaningless cardiac output) or the inability to
`wedge the catheter. See Discussion section.
`
`68
`125
`116
`116
`37
`91
`13
`13
`24
`24
`
`.075
`.371
`⬍.001
`⬍.001
`.113
`.113
`.802
`.015
`.116
`.332
`
`CSP
`
`TABLE 9. Effect of inhaled prostacyclin (iPGI2) on hemodynamic parameters: baseline and after 4 to 6 h oniPGI 2*
`Variable
`4 to 6 h after
`n
`P value
`Before PGI2
`
`256 ⫾ 167
`76 ⫾ 11
`35 ⫾ 9
`0.47 ⫾ 0.14
`4.6 ⫾ 1.5
`2.5 ⫾ 0.8
`18 ⫾ 6
`17 ⫾ 4
`209 ⫾ 37
`1143 ⫾ 553
`69 ⫾ 12
`
`281 ⫾ 164
`78 ⫾ 11
`24 ⫾ 8
`0.32 ⫾ 0.12
`5.3 ⫾ 1.6
`2.9 ⫾ 0.8
`13 ⫾ 6
`15 ⫾ 5
`156 ⫾ 34
`961 ⫾ 265
`69 ⫾ 11
`
`PaO2/FIO2 ratio
`MAP (mm Hg)
`MPAP (mm Hg)
`MPAP/MAP ratio
`Cardiac output (L/min)
`CI (L · min⫺1 · m⫺2)
`CVP (mm Hg)
`Wedge (mm Hg)
`PVR (dynes 䡠 s 䡠 cm⫺5)
`SVR (dynes 䡠 s 䡠 cm⫺5)
`SvO2 (%)
`FIO2, Fraction of inspired oxygen; MAP, mean arterial blood pressure; MPAP, mean pulmonary artery pressure; CI, cardiac index; CVP, central venous
`pressure; PVR, pulmonary vascular resistance; SVR, systemic vascular resistance; SvO2, mixed venous oxygen saturation.
`*The small n in the number of wedge pressures and cardiac output measurements (and hence derived parameters, such as PVR and SVR) reflects the
`inability to obtain these parameters for a variety of reasons, including tricuspid regurgitation (inaccurate, meaningless cardiac output) or the inability to
`wedge the catheter. See Discussion section.
`
`43
`98
`92
`92
`36
`36
`73
`7
`6
`24
`24
`
`.269
`.253
`⬍.001
`⬍.001
`.037
`.036
`⬍.001
`.424
`.044
`.137
`.708
`
`drug were therefore observed even at these dosages that
`were lower than previously published.
`Inhaled PGI2 has several advantages over iNO. It is easy
`to administer and requires no special monitoring equipment
`during administration. It can also safely be administered to
`extubated patients via a face mask. This route of adminis-
`tration was very well tolerated and did not cause any eye
`irritation or sore throat, unlike iNO, which can cause met-
`hemoglobinemia and buildup of oxygen free radicals and
`nitrogen dioxide.3 Inhaled PGI2 does not have toxic metab-
`olites that require monitoring. It is stored in a crystallized
`powder form at room temperature, its final preparation with
`the special glycine diluent is fast and easy, and it is readily
`available.
`There are certain precautions that have to be considered
`when administering iPGI2. The final preparation has to be
`shielded from light because of the potential for light deg-
`radation. Care has to be taken during transport to prevent
`
`accidental spilling of the drug down the inspiratory limb
`into the patient’s trachea. Another disadvantage of iPGI2 is
`the inability to adequately deliver the nebulized drug to
`patients in whom large-volume pulmonary edema develops.
`As previously mentioned, the glycine buffer makes the
`aerosol “sticky.” Filters on ventilators and BIPAP machines
`therefore need to be frequently changed to prevent the
`development of a stuck ventilator valve and auto-PEEP.
`We have demonstrated that there are potentially large
`cost savings when iPGI2 is used instead of iNO. These cost
`savings would be even more substantial if the costs associ-
`ated with monitoring for iNO toxicity were considered. The
`relatively low cost, together with the excellent safety pro-
`file, may allow for the use of iPGI2 in many patients in
`whom iNO might not have been considered because of the
`enormous costs currently associated with it. Inhaled NO
`should still be readily available in cases in which patients do
`not respond to iPGI2. Similarly, in patients with severe
`
`1064 The Journal of Thoracic and Cardiovascular Surgery ● April 2004
`
`Liquidia's Exhibit 1025
`Page 7
`
`

`

`De Wet et al
`
`Cardiopulmonary Support and Physiology
`
`CSP
`
`pulmonary hypertension or refractory hypoxemia, these 2
`agents combined, may provide additive effects.30
`There are several limitations to the study. There was no
`control group in our study, nor did we directly compare
`iPGI2 with iNO. We did not believe that a control or placebo
`group could be used in these critically ill patients, nor could
`we stop the agent to allow patients to serve as their own
`controls. However, this has previously been addressed in
`other studies, albeit in small numbers of patients.8,12,13,20-22
`Another limitation is that there are many confounding vari-
`ables that affect the hemodynamic parameters that we mea-
`sured. We did not standardize the intraoperative and anes-
`thetic management or
`the postoperative critical care,
`including vasoactive drug strategies and ventilation strate-
`gies. This would not have been possible, considering the
`heterogeneous group of patients. As a result, not all the
`changes observed in our study may be attributable to iPGI2.
`It is likely that the immediate changes (before and after) are
`linked to iPGI2 but that some of the changes seen 4 to 6
`hours after initiation of the drug were due to confounding
`variables and the natural history of the pathophysiological
`process. For example, a decrease in CVP may have been
`due to a decrease in volume status or an improvement in
`right ventricular function. Similarly, the decrease in MPAP
`may be due to an improvement in many of the factors that
`are known to affect pulmonary artery pressures, including
`oxygenation, CO2 levels, acid-base status, functional resid-
`ual capacity, vasopressor use, and shivering. Another limi-
`tation is that we do not have complete data for some
`variables, such as pulmonary vascular resistance, wedge
`pressure, and CVP, at the 4- to 6-hour time point. This is
`due to a number of factors that commonly occur in these
`patients. First, some of the patients were already weaned
`from iPGI2 by the time of the 4- to 6-hour measurements.
`Second, it is a safe and common intensi