early initiation of prostacyclin in portopulmonary hypertension: 10 years of a transplant center’s...

PULMONARY HYPERTENSION

Early Initiation of Prostacyclin in Portopulmonary Hypertension:10 Years of a Transplant Center’s Experience

Rana L. A. Awdish • Hector R. Cajigas

Received: 14 February 2013 / Accepted: 5 August 2013 / Published online: 25 August 2013

� Springer Science+Business Media New York 2013

Abstract

Background Portopulmonary hypertension (PoPH) is a

subgroup of Group 1 pulmonary arterial hypertension

(PAH) with particularly poor prognosis. Delay in initiation

of parenteral therapy may be the reason for poor outcome.

Methods We conducted a prospective observational study

of all patients with Group 1 PoPH evaluated at Henry Ford

Hospital between January 2002 and July 2012. The cohort

of the REVEAL Registry patients with PoPH was used as

the comparator group. The patient survival rates at 5 years

after diagnosis and 2 years after enrollment, treatment

trends, and the freedom from all-cause hospitalization rates

at 2 years after enrollment were compared using v2

analysis.

Results Twenty-one patients were enrolled in the PH

Clinic with PoPH from January 2002 through July 2012.

Our patients were significantly more likely to be on pros-

tacyclin IV at 90 days as compared to REVEAL PoPH

patients (67 vs. 31 %; p = 0.002). Despite this, early out-

comes were not significantly different between the groups:

2-year survival from enrollment (70 and 67 %, respec-

tively; p = 0.77) and 2-year freedom from hospitalization

(35 vs. 49 %, respectively; p = 0.29). However, 5-year

survival from time of diagnosis was significantly higher in

our cohort of PoPH patients (71 and 40 %, respectively; p =

0.02).

Conclusions Early initiation of parenteral prostacyclin

therapy in PoPH patients at a single institution was asso-

ciated with improved 5-year patient survival from diag-

nosis as compared to the REVEAL Registry of PoPH

patients and allowed for clearance for transplant in 52 % of

patients within 1 year.

Keywords Model for end stage liver disease �Orthotopic liver transplant � Pulmonary arterial

hypertension � Portopulmonary hypertension �Transplant clearance � Prostacyclin

Abbreviations

ALT Alanine aminotransferase

AST Aspartate aminotransferase

6MWD Six minute walk distance

CO Cardiac output

ETRA Endothelin receptor antagonist

IPAH Idiopathic pulmonary arterial hypertension

mPAP Mean pulmonary arterial pressure

MELD Model for end stage liver disease

OLT Orthotopic liver transplant

PAH Pulmonary arterial hypertension

PASP Pulmonary artery systolic pressure

PCWP Pulmonary artery occlusion pressure

PDEI Phosphodiesterase inhibitor

PoPH Portopulmonary hypertension

PVR Pulmonary vascular resistance

NYHA New York Heart Association

REVEAL Registry to Evaluate Early and Long-term

PAH Disease Management

TPG Transpulmonary gradient

R. L. A. Awdish (&) � H. R. Cajigas

Pulmonary and Critical Care Medicine Division, Henry Ford

Hospital, 2799 West Grand Boulevard, K-17, Detroit, MI 48202,

USA

e-mail: [email protected]

123

Lung (2013) 191:593–600

DOI 10.1007/s00408-013-9501-5

Introduction

Portopulmonary hypertension (PoPH) is defined as pul-

monary arterial hypertension (PAH) associated with portal

hypertension, with or without liver disease [1]. PoPH is a

life-threatening disease characterized by elevation in pul-

monary vascular resistance (PVR), which leads to

increased pulmonary artery pressure, right ventricular

failure, and ultimately death [2]. Although the natural

history of the disease has been difficult to characterize,

multiple studies have demonstrated that despite what

would be considered ‘‘more favorable’’ hemodynamics

(higher cardiac index and lower PVR), mortality of these

patients is significantly higher than their idiopathic pul-

monary arterial hypertension (IPAH) counterparts [3, 4].

This fact, in addition to the increased perioperative mor-

bidity and mortality of this group during liver transplant

makes PoPH especially worthy of study [5].

Given the severity and poor prognosis of PoPH, early

and appropriate therapy is considered critical; however,

studies in this area are lacking. Treatment of PoPH has

focused on vasodilator therapy as continuous IV adminis-

tration of epoprostenol reduces PVR, mPAP, and improves

RV function [6]. Independent of transplant status, IV

prostacyclin impacts long-term survival in PoPH [5].

Recently, a report from the Registry to Evaluate Early and

Long-term PAH Disease Management (REVEAL) descri-

bed a large cohort of patients with PoPH, and reaffirmed

that survival from the time of enrollment was worse than

that observed for IPAH [3]. Because only 57 % of the

newly diagnosed patients were receiving PAH treatment,

the authors postulated that delayed initiation of PAH

treatment may be implicated in their lower survival [3]. As

a high-volume liver transplant center, with a structured

protocol for screening for PoPH that is in accordance with

the international guidelines, and early initiation of paren-

teral therapy strategy, our cohort of patients offer oppor-

tunity to expand the knowledge in this area. We describe

the characteristics and outcomes of a cohort of patients

with PoPH.

Methods and Materials

This prospective, observational study was approved by the

Institutional Review Board of Henry Ford Hospital. The

medical records of patients diagnosed with PoPH and

evaluated in the Pulmonary Hypertension Clinic at Henry

Ford Hospital between January 2002 and July 2012 were

reviewed. All patients met the UNOS criteria for liver

transplantation and were referred to our clinic when

screening 2D transthoracic echocardiography reported a

systolic pulmonary artery pressure (PASP) greater than

40 mmHg. Date of diagnosis was taken as the date of

cardiac catheterization confirming the diagnosis, regardless

of whether it was within our system. Date of enrollment

was documented as the date the patient was first seen by the

Pulmonary Hypertension Clinic at Henry Ford Hospital.

All patients underwent cardiac catheterization and were

diagnosed with PoPH when the following specific criteria

were met: portal hypertension and/or evidence of advanced

liver disease, mPAP C25 mmHg, PVR C240 dynes/s/cm5,

PCWP B15 mmHg, and no evidence of alternative car-

diopulmonary diseases associated with PAH [3]. As

advocated by expert opinion, and given the high prevalence

of volume overload and high-output heart failure in

advanced liver disease, patients with PCWP [15 mmHg

were not excluded if their transpulmonary gradient (TPG)

was [12 mmHg [5]. This is modeled as well on REVEAL

registry enrollment criteria [7].

The following demographic data were collected: patient

age at diagnosis, gender, and self-identified race. Hemo-

dynamic data were collected at the time of diagnosis and

following therapy. Surrogate markers, including New York

Heart Association Functional Class (NYHA), Model for

End-Stage Liver Disease (MELD) and Child-Turcotte-

Pugh scores, REVEAL Risk Scores, treatment, and six-

minute walk distance (6MWD), also were collected before

and after therapy. Modality of therapy and timing of ini-

tiation of therapy were documented.

The cohort of the REVEAL registry patients with PoPH

as described in Chest by Krowka et al. [3] was used as the

comparator group. Group comparisons for continuous

variables and categorical variables were made using t tests

and Chi square tests, respectively. For categorical data

involving expected frequency counts \5, the Fisher exact

test was used instead of the Chi square test. Group com-

parisons for ordered data were made using the Cochran–

Armitage trend test. Patient survival (during the first

5 years after diagnosis and the first 2 years after enroll-

ment) and freedom from all-cause hospitalization (during

the first 2 years after enrollment) were assessed using

Kaplan–Meier estimation. A true Kaplan–Meier analysis

was performed on the Henry Ford cohort. Because we did

not have the original raw patient data from the REVEAL

registry cohort, we were not able to produce a direct

Kaplan–Meier comparison of the two cohorts. However,

when the Chi square comparisons of the survival rates at

specific time points were performed, the patient censoring

information from the Henry Ford cohort was directly

accounted for using the raw patient data, and the patient

censoring information from the REVEAL registry cohort

was indirectly accounted for using number of patients still

at risk at various time intervals. Data analysis was per-

formed using version 9.2 of the SAS statistical package.

594 Lung (2013) 191:593–600

123

The patient survival rates at 5 years after diagnosis, the

patient survival rates at 2 years after enrollment, and the

freedom from all-cause hospitalization rates at 2 years after

enrollment were compared between our group and the

REVEAL registry group using Chi square analysis. Sta-

tistical significance was attained for p \ 0.05.

Results

The analysis included 21 patients with PoPH compared

with the REVEAL registry’s 174 patients (Table 1). At the

time of diagnosis and enrollment in the PH Clinic (our

surrogate for enrollment in the REVEAL registry), our

subgroup was similar in age, gender, race, and NYHA

functional class with the majority of patients in each group

in functional class II (24 and 35 % of patients respectively)

or III (52 and 49 % respectively). Our patients were pre-

dominately Child-Turcotte-Pugh B at diagnosis (n = 11;

52 %); five patients (24 %) were Class A, and five patients

(24 %) were Class C. Our patients had a pretreatment

MELD score mean of 12.5 (SD ± 5.1). REVEAL registry

data does not capture severity of liver disease.

Neither the Child-Turcotte-Pugh Class nor the MELD

score had statistically significant change when pretreatment

and posttreatment variables were compared. This also was

true of the REVEAL Risk Score. The pre-REVEAL score

had a mean of 8.6 with a standard deviation of 1.8, whereas

the post-REVEAL score had a mean of 8.7 with a standard

deviation of 2.4. The pre to post change in the REVEAL

score was not statistically significant (p = 0.71). Where

available, hemodynamic variables and 6MWD were com-

pared before and after therapy as represented in Table 2.

Although the timeframe was variable, posttherapy values

are generally within 6–12 months after initiation of ther-

apy. 6MWD increased an average of 30 meters

(p = 0.009). We observed a decrease in mPAP from a

mean of 45 to 33 mmHg (p \ 0.001) as well as a decrease

in PVR from 6.71 to 3.43 WU (p \ 0.001). There also was

an increase in CO from a pretreatment mean of 5.48 to 7.82

L/min (p = 0.005); CI increased from a mean of 2.73 to

4.01 L/min/m2 (p \ 0.001).

Compared with the REVEAL registry data, our PoPH

patients were significantly more likely to be on prostacy-

clin IV at 90 days (p = 0.002) with 67 % of patients begun

on prostacyclin by that time compared with 31 % of

REVEAL registry patients (Table 3). Our patients were

significantly less likely to be on phosphodiesterase inhibi-

tors (PDEI) at 90 days (p \ 0.001) with only 24 % of our

patients on PDEI versus 65 % of REVEAL registry

patients. In addition, 5 % were initiated on inhaled pros-

tacyclin therapy within 90 days of diagnosis and 5 % on an

endothelin receptor antagonist (ETRA). No patients were

untreated at 90 days, nor were any patients were on com-

bination therapy by 90 days. Two patients were on com-

bination therapy at 1 year.

Although our patients did have significantly higher

levels of PCWP at diagnosis, the REVEAL registry authors

do explain that they conducted a separate analysis,

including the 26 patients they had excluded with

PCWP [15 mmHg, which did not affect the results or

conclusions of their study. A separate analysis of our seven

patients with PCWP [15 mmHg demonstrated that they

benefited from treatment with a statistically significant

decrease in mPAP posttreatment (pre-mPAP 49 mmHg;

post-mPAP 31 mmHg; p = 0.03) and improvement in

PVR (pre-PVR 5.29; post-PVR 2.29; p = 0.01). Taken

together, these improvements in hemodynamics would

allow for transplant listing for these patients, having been

brought into the lower-risk category through treatment,

suggesting that it is appropriate to include them in our

analysis.

Survival was assessed both from the time of diagnosis as

well as at the time of enrollment in the Henry Ford

Table 1 Baseline demographic and clinical comparison results for

HFHS patients versus the REVEAL registry

Comparison variable HFHS

study

(N = 21)

REVEAL

registry

(N = 174)

Comparison

p value

Patient age, year

(mean ± SD)

55 ± 10 53 ± 10 0.39 (T)

Gender (female) 48 % 52 % 0.72 (C)

Race (white) 67 % 79 % 0.19 (C)

NYHA at enrollment 0.23 (C–A)

I 24 % 10 %

II 24 % 35 %

III 52 % 49 %

IV 0 % 6 %

AST at enrollment 0.15 (C–A)

Normal 43 % 45 %

B3 times ULN 29 % 42 %

[3 and B5 times ULN 14 % 12 %

[5 times ULN 14 % 2 %

ALT at enrollment 0.04 (C–A)*

Normal 57 % 71 %

B3 times ULN 19 % 20 %

[3 and B5 times ULN 14 % 8 %

[5 times ULN 10 % 1 %

Data are mean ± standard deviation

T two-sample t test, C Chi square test, C–A Cochran-Armitage trend

test, F Fisher exact test, K–M Kaplan–Meier estimation

* Statistically significant, p \ 0.05

Lung (2013) 191:593–600 595

123

Pulmonary Hypertension Clinic. As some patients were

diagnosed outside of the PH clinic and in some cases,

outside of Henry Ford Hospital, the estimate from the time

of enrollment is representative of a mixed population of

newly and previously diagnosed patients. Despite a sig-

nificantly higher proportion of our patients on IV prosta-

cyclin therapy, there were no statistically significant

differences in 2-year freedom from death (patient survival)

or freedom from all-cause hospitalization detected between

the two groups. However, there was a statistically signifi-

cant difference in 5-year patient survival from diagnosis in

our cohort of patients, who benefited from early therapy

with IV prostacyclin. The 2-year survival from enrollment

in Henry Ford PH clinic compared with patients from

REVEAL registry with PoPH is represented in Fig. 1 (2-

year survival estimates 70 and 67 %, respectively;

p = 0.86). The 5-year survival from diagnosis compared

with patients from REVEAL registry with PoPH is repre-

sented in Fig. 2 (5-year survival estimates 71 and 40 %,

respectively; p = 0.02). In subgroup analysis, patients who

did not receive a liver transplant had a probability of both

2-year survival from enrollment and 5-year survival from

diagnosis of 57 % (Figs. 1, 2).

As demonstrated in the REVEAL registry data, patients

with PoPH are prone to frequent hospitalization. Despite

what would be agreed was a more aggressive treatment

strategy, our patients demonstrated similar 2-year freedom

Table 2 Pre- to posttreatment changes

Variable N Mean SD p value

Pre-6MWD 20 330.75 82.77

Post-6MWD 20 360.65 92.56

Post- minus pre-6MWD 20 29.9 91.93 0.009 (W)*

Pre-mPAP 16 45.56 7.73

Post-mPAP 16 33.31 9.65

Post- minus pre-mPAP 16 -12.25 10.72 \0.001 (T)*

Pre-mRAP 16 9.94 8.9

Post-mRAP 16 5.44 3.16

Post- minus pre-mRAP 16 -4.5 9.4 0.108 (W)

Pre-PVR 16 6.71 2.59

Post-PVR 16 3.43 2.21

Post- minus pre-PVR 16 -3.28 2.25 \0.001 (T)*

Pre-PCWP 16 12.88 3.54

Post-PCWP 16 11.19 6.02

Post- minus pre-PCWP 16 -1.69 6.45 0.312 (T)

Pre-CO 16 5.48 2.30

Post-CO 16 7.82 3.05

Post- minus pre-CO 16 2.34 2.87 0.005 (T)*

Pre-CI 16 2.73 1.02

Post-CI 16 4.01 1.44

Post- minus pre-CI 16 1.27 1.39 0.002 (T)*

Pre-TPG 16 31.81 7.72

Post-TPG 16 21.75 8.59

Post- minus pre-TPG 16 -10.06 9.57 \0.001 (T)*

Pre RV systolic pressure 18 71.89 22.70

Post RV systolic pressure 18 55.11 29.96

Post- minus pre-RV systolic

pressure

18 -16.78 23.63 0.008 (T)*

Each pre to post analysis involves only patients with both pre- and

post-data

W Wilcoxon signed-rank test, T paired t test, SD standard deviation


Table 3 Treatment and outcome comparison results for the HFHS

study versus the REVEAL registry

Comparison variable HFHS

study

(N = 21)

REVEAL

Registry

(N = 174)

Comparison

p value

Prostacyclin IV at 90 days 67 % 31 % 0.002 (C)*

Prostacyclin inhaled at

90 days

5 % 16 % 0.32 (F)

PDEI at 90 days 24 % 65 % \0.001 (C)*

ETRA at 90 days 5 % 8 % 1.00 (F)

Prostacyclin IV at 365 days 55 % 35 % 0.09 (C)

Prostacyclin inhaled at

365 days

10 % 20 % 0.37 (F)

PDEI at 365 days 50 % 71 % 0.06 (C)

ETRA at 365 days 10 % 14 % 1.00 (F)

mPAP at diagnosis

(mean ± SD)

46 ± 9 49 ± 11 0.23 (T)

mRAP at diagnosis

(mean ± SD)

9 ± 8 9 ± 6 1.00 (T)

PVR at diagnosis

(mean ± SD)

7 ± 2 8 ± 4 0.26 (T)

PCWP at diagnosis

(mean ± SD)

13 ± 4 9 ± 3 \0.001 (T)*

Cardiac output at diagnosis

(mean ± SD)

6 ± 2 5 ± 2 0.20 (T)

Cardiac index at diagnosis

(mean ± SD)

2.8 ± 1.0 2.7 ± 0.8 0.60 (T)

TPG at diagnosis

(mean ± SD)

32 ± 8 40 ± 11 0.001 (T)*

6MWD at diagnosis

(mean ± SD)

334 ± 82 354 ± 125 0.48 (T)

2-year patient survival from

enrollment (K–M)

70 % 67 % 0.77 (C)

5-year patient survival from

diagnosis (K–M)

71 % 40 % 0.02 (C)*

2-year all-cause non-hosp.

from enrollment (K–M)

35 % 49 % 0.29 (C)

Data are mean ± standard deviation

T two-sample t test, C Chi square test, C–A Cochran–Armitage trend

test, F Fisher exact test, K–M Kaplan–Meier estimation


596 Lung (2013) 191:593–600

123

from hospitalization estimates, 35 versus 49 %, respec-

tively; p = 0.29). Figure 3 also plots subgroup analysis of

2-year freedom from hospitalization in patients who were

not transplanted; the resulting probability was 50 %.

Discussion

The amassed data on PoPH is limited by the fact that it is a

rare presentation of a still underrecognized disease (PAH).

PoPH is characterized by the 5th World Symposium on

Pulmonary Hypertension as being within the same group as

idiopathic PAH, which is a reflection of similar pathology

[8]. Due to concomitant cirrhosis, many of these patients

require liver transplantation, and the cohort patients who

receive both PAH-specific therapy and transplant have the

best survival [5]. The prevalence of PoPH reported in the

patients undergoing evaluation for liver transplant has been

reported as 5.3 and 8.5 % in two large series [9, 10]. As

awareness of the disorder increases, this may rise. In fact,

Fig. 1 Kaplan–Meier curves

for patient survival with the

follow-up period starting at the

time of enrollment


for patient survival with the

follow-up period starting at the

time of diagnosis

Lung (2013) 191:593–600 597

123

PoPH represents 20 % of all PAH cases in the recent

French PAH registry, making it the most common form of

nonidiopathic PAH [7].

PoPH is a known risk factor for perioperative morbidity

and mortality in liver transplantation; evidence suggests that

a mean pulmonary arterial pressure (mPAP) [35 mmHg

portends increased risk and mPAP [50 mmHg is prohibi-

tively dangerous, with a nonlinear increase between the two

values [10, 11]. This knowledge is at the root of current

practice guidelines that recommend that all patients under-

going liver transplant evaluation be screened for PoPH via

transthoracic echocardiography [12].

Confirmation of PoPH requires right heart catheterization

and interpretation of data by clinicians well versed in the

complexity of PoPH [13]. mPAP C25 mmHg that is asso-

ciated with PVR C240 dynes/s/cm5 and a pulmonary artery

occlusion pressure (PCWP) B15 mmHg is diagnostic [3].

However, because patients may present with high-volume

status in the setting of PoPH, the transpulmonary gradient

(TPG) also must be considered, as must the relative cardiac

output (CO) [5]. It is the complex relationship between all of

these variables that allows for decisions as to whether the

patient may benefit from PAH-specific therapy [14]. Treat-

ment with IV prostacyclin impacts long-term survival in

PoPH regardless of transplant status, but additionally, if

PoPH is controlled it may make a previously impossible

orthotopic liver transplant (OLT) possible [15]. Indeed,

results of a multicenter database of 66 patients cited degree

of pulmonary hypertension as the reason for denying

transplant in 30 of the patients [16].

Our paper describes a cohort of patients with PoPH at a

high-volume liver transplant center who were followed

prospectively, and subsequently compared to the largest

described cohort of patients with PoPH from published

REVEAL registry data. Hemodynamic parameters of our

patients were similarly favorable, with lower mPAP, PVR,

and mRAP and higher CO than in patients with IPAH.

Functionally, our patients were similarly distributed

between NHYA Functional Classes and had similar

6MWD. Although our patients did have significantly

higher levels of PCWP at diagnosis, the REVEAL registry

authors do explain that they conducted a separate analysis,

including the 26 patients they had excluded with

PCWP [15 mmHg; this did not affect the results or con-

clusions of their study. This information, taken in total,

allows for comparison of treatment patterns, given that the

two groups share very similar baseline characteristics. A

separate analysis of our seven patients with

PCWP [15 mmHg demonstrated that they benefited from

treatment with improvement in hemodynamics that placed

them solidly in the lower-risk category and would allow for

transplant listing. This suggests that they continue to rep-

resent the target group for PoPH that requires therapy

despite the PCWP [15 mmHg and further that it is

appropriate to include them in our analysis.

Significant findings include a seemingly more aggres-

sive treatment pattern than what has been defined previ-

ously. Unlike the REVEAL registry PoPH patients, our

patients were significantly more likely to be on prostacy-

clin IV at 90 days (p = 0.003) and significantly less likely


for freedom from all-cause

hospitalization with the follow-

up period starting at the time of

enrollment

598 Lung (2013) 191:593–600

123

to be on PDEI at 90 days (p \ 0.001). This treatment

pattern is influenced by the Mayo Clinic experience and

other published data clearly demonstrating the survival rate

for untreated PoPH patients is abysmally low and that

survival improves with vasodilator therapy regardless of

whether the patient receives liver transplant [5]. Although

the REVEAL registry data were not available to us at the

time of decision regarding initiation of therapy, it too

reinforces that patients with PoPH have significantly worse

survival compared with patients with IPAH and that the

simple diagnosis of PoPH compared with other PAH sub-

groups is an independent risk factor for increased mortality

[3]. Our treatment strategy is comparable to that used for

other types of PAH, with consideration of severity of dis-

ease, functional class, MELD score, and hemodynamics all

considered. Because of the poor prognosis of PoPH, how-

ever, all patients are considered for aggressive therapy with

a prostacyclin infusion. This is in contrast to PAH where

‘‘sicker’’ patients only are considered for more aggressive

therapy. In the setting of PoPH, what are traditionally

considered ‘‘favorable’’ hemodynamics for an IPAH

patient are relatively meaningless, as the prognosis even in

these patients is poor.

This construct is at the root of our treatment patterns,

and in fact, none of our patients with PoPH were untreated

at 90 days. In addition to the 65 % who were initiated on

IV prostacyclin, an additional 5 % were initiated on

inhaled prostacyclin therapy, 25 % on a PDEI, and 5 % on

an ETRA. Seven of our patients were transplanted suc-

cessfully; an additional four patients were cleared for

transplant (Table 4). Few of our patients were on combi-

nation therapy, however, at 365 days. This is perhaps

reflective of our reliance on prostacyclin to achieve

favorable hemodynamics. A majority of our patients were

treated with IV epoprostenol, with a mean dose of 20.8 ng/

kg/min (SD ± 13.9 ng/kg/min). Hemodynamic improve-

ment was noted with a significant decrease in mPAP, PVR,

and improvement in CO as well as CI (Table 2).

In the French series by Le Pavec et al. [17], patients with

PoPH demonstrated a 68 % 5-year survival from diagnosis.

This is despite only 15 % of their patients receiving IV

prostacyclin and is likely related to severity of liver disease

in a cohort where Child-Pugh class A predominated at

51 % of patients. In contrast, REVEAL registry data noted

a 40 % survival. The REVEAL registry researchers had

postulated that the severity of liver disease may have been

the variable influencing the difference in 5-year survival

between the two groups, and this is possible. However, our

cohort had a 5-year survival that surpassed the French

series at 71 %, despite our patients being predominately

Child-Pugh class B (55 %) and C (20 %) at the time of

diagnosis. We believe that the improved 5-year survival in

our patients, despite the severity of liver disease is a direct

reflection of an aggressive treatment strategy (Fig. 2). In

the French series, only 15 % of patients were on IV pros-

tacyclin therapy [17]. In the Mayo Clinic series, the

patients who received PAH therapy only, the 5-year sur-

vival was 45 % [5]. This may have reflected uncontrolled

and variable treatment patterns.

This improved 5-year survival was not apparent in

change in MELD score, Child-Pugh class, or REVEAL risk

score as there was not a statistically significant change

observed pre- and posttreatment in any of these variables.

This suggests to us that the improvement garnered with

therapy is not captured in a measured variable that is

included in these scores.

Small, prospective, cohort studies such as these provide

valuable information that can help broaden the under-

standing of this rare disease; nevertheless, there are many

limitations. Some patient censoring occurred before the end

of the 2- and 5-year follow-up periods. However, the Chi

square comparisons of the 2- and 5-year outcome propor-

tions were based on the underlying Kaplan–Meier esti-

mates and the denominators of those proportions were

adjusted accordingly and did not involve censored patients.

Given the small sample size, the ability to adjust our

Table 4 Patient outcomes

Patient

no.

Date of

death

Narrative outcome

1 Jul-11 Patient elected hospice enrollment due to

HCC

2 Alive Transplanted



5 Alive Developed HCC


7 Alive Continues on medical therapy

8 May-08 Cleared for transplant; elected hospice

enrollment


10 Jun-12 Elected hospice enrollment

11 Aug-11 Patient declined transplant

12 Alive Patient elected hospice enrollment due to

HCC


14 Apr-08 Patient elected hospice enrollment

15 Aug-07 Patient elected hospice enrollment


17 Alive Patient cleared for transplant

18 Jul-07 Patient cleared for transplant

19 Alive Not an OLT candidate due to ongoing

drinking



Lung (2013) 191:593–600 599

123

analysis for possible confounders is limited. Nevertheless,

we believe there is value in a single-center study of this

nature. The consistency and relatively uniform treatment

pattern, which is not true or possible in large registries,

allows insight into this complex and deadly disease.

Conclusions

Early initiation of IV prostacyclin therapy in PoPH patients

at a single institution was associated with improved 5-year

patient survival from diagnosis compared with the

REVEAL registry of PoPH patients and allowed for

clearance for transplant in 52 % of patients within 1 year.

Conflict of interest Dr. Cajigas has served as a consultant for

United Therapeutics Inc., Actelion and Gilead. He has served in the

speaker’s bureau for Actelion and United Therapeutics Inc, received

honoraria for service on Advisory boards from United Therapeutics,

Actelion and Gilead. Dr. Cajigas has received Grants and research

support from Pfizer, Actelion and United Therapeutics. Dr. Awdish

has received Grants and research support from Pfizer, Actelion and

United Therapeutics.

References

1. Rodriguez-Roisin R, Krowka MJ, Herve P et al (2004) Pulmon-

ary-hepatic vascular disorders (PHD). Eur Respir J 24:861–880

2. Krowka MJ, Edwards WD (2000) A spectrum of pulmonary

vascular pathology in portopulmonary hypertension. Liver

Transpl 6:241–242

3. Krowka MJ, Miller DP, Barst RJ et al (2012) Portopulmonary

hypertension: a report from the US-based REVEAL registry.

Chest 141:906–915

4. Kawut SM, Taichman DB, Ahya VN et al (2005) Hemodynamics

and survival of patients with portopulmonary hypertension. Liver

Transpl 11:1107–1111

5. Swanson KL, Wiesner RH, Nyberg SL et al (2008) Survival in

portopulmonary hypertension: Mayo Clinic experience catego-

rized by treatment subgroups. Am J Transplant 8:2445–2453

6. Fix OK, Bass NM, De Marco T et al (2007) Long-term follow-up

of portopulmonary hypertension: effect of treatment with epo-

prostenol. Liver Transpl 13:875–885

7. Frost AE, Badesch DB, Barst RJ et al (2011) The changing

picture of patients with pulmonary arterial hypertension in the

United States: how REVEAL differs from historic and non-US

Contemporary Registries. Chest 139:128–137

8. Simonneau G, Robbins IM, Beghetti M et al (2009) Updated

clinical classification of pulmonary hypertension. J Am Coll

Cardiol 54:S43–S54

9. Krowka MJ, Swanson KL, Frantz RP et al (2006) Portopulmo-

nary hypertension: results from a 10-year screening algorithm.

Hepatology 44:1502–1510

10. Ramsay MA, Simpson BR, Nguyen AT et al (1997) Severe

pulmonary hypertension in liver transplant candidates. Liver

Transpl Surg 3:494–500

11. Krowka MJ, Plevak DJ, Findlay JY et al (2000) Pulmonary

hemodynamics and perioperative cardiopulmonary-related mor-

tality in patients with portopulmonary hypertension undergoing

liver transplantation. Liver Transpl 6:443–450

12. Murray KF, Carithers RL Jr (2005) AASLD practice guidelines:

evaluation of the patient for liver transplantation. Hepatology

41:1407–1432

13. Sussman NL, Kochar R, Fallon MB (2011) Pulmonary compli-

cations in cirrhosis. Curr Opin Organ Transplant 16:281–288

14. McLaughlin VV, Archer SL, Badesch DB et al (2009) ACCF/

AHA 2009 expert consensus document on pulmonary hyperten-

sion a report of the American College of Cardiology Foundation

Task Force on Expert Consensus Documents and the American

Heart Association developed in collaboration with the American

College of Chest Physicians; American Thoracic Society, Inc.;

and the Pulmonary Hypertension Association. J Am Coll Cardiol

53:1573–1619

15. Hollatz TJ, Musat A, Westphal S et al (2012) Treatment with

sildenafil and treprostinil allows successful liver transplantation

of patients with moderate to severe portopulmonary hypertension.

Liver Transpl 18:686–695

16. Krowka MJ, Mandell MS, Ramsay MA et al (2004) Hepatopulmo-

nary syndrome and portopulmonary hypertension: a report of the

multicenter liver transplant database. Liver Transpl 10:174–182

17. Le Pavec J, Souza R, Herve P et al (2008) Portopulmonary

hypertension: survival and prognostic factors. Am J Respir Crit

Care Med 178:637–643

600 Lung (2013) 191:593–600

123