Nephrol Dial Transplant (2005) 20: 1686–1692
doi:10.1093/ndt/gfh840
Advance Access publication 19 April 2005
Original Article
The pathogenesis of pulmonary hypertension in haemodialysis
patients via arterio-venous access
Farid Nakhoul1,4, Mordechai Yigla2, Rima Gilman1, Shimon A. Reisner3 and Zaid Abassi4
1
Department of Nephrology, 2Division of Pulmonary Medicine, 3Department of Cardiology and 4Department of
Physiology and Biophysics, Rambam Medical Center and Rappaport Faculty of Medicine, Technion-Israel Institute
of Technology, Haifa, Israel
Abstract
Background. We recently have shown a high incidence
of unexplained pulmonary hypertension (PHT) in
end-stage renal disease (ESRD) patients on chronic
haemodialysis (HD) therapy via arterio-venous (A-V)
access. This study evaluated the possibility that PHT in
these patients is triggered or aggravated by chronic HD
via surgical A-V access, and the role of endothelin-1
(ET-1) and nitric oxide (NO) in this syndrome.
Methods. Forty-two HD patients underwent clinical
evaluation. Pulmonary artery pressure (PAP) was
evaluated using Doppler echocardiography. Levels of
ET-1 and NO metabolites in plasma were determined
before and after the HD procedure and were compared
between subgroups of patients with and without PHT.
Results. Out of 42 HD patients studied, 20 patients
(48%) had PHT (PAP ¼ 46±2; range 36–82 mmHg)
while the rest had a normal PAP (29±1 mmHg)
(P<0.0001). HD patients with PHT had higher
cardiac output compared with those with normal
PAP (6.0±1.2 vs 5.2±0.9 l/min, P<0.034). HD
patients, with or without PHT, had elevated plasma
ET-1 levels compared with controls (1.6±0.7 and
2.4±0.8 fmol/ml vs 1.0±0.2, P<0.05) that remained
unchanged after the HD procedure. HD patients
without PHT and control subjects showed similar
basal plasma levels of NO2 þ NO3 (24.2±5.2 vs
19.7±3.1 mM, P>0.05) that was significantly higher
compared with HD patients with PHT (14.3±2.3 mM,
P<0.05). HD therapy caused a significant increase in
plasma NO metabolites that was greater in patients
without PHT (from 24.2±5.2 to 77.1±9.6 mM,
P<0.0001, and from 14.3±2.3 to 39.9±11.4 mM,
P<0.0074, respectively). Significant declines in PAP
(from 49.8±2.8 to 38.6±2.2 mmHg, P<0.004) and
cardiac output (CO) (from 7.6±0.6 to 6.1±0.3 l/min,
Correspondence and offprint requests to: Farid Nakhoul, MD,
Department of Nephrology, Rambam Medical Center, Haifa 31096,
Israel. Email: f_nakhoul@rambam.health.gov.il
ß The Author [2005]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please email: journals.permissions@oupjournals.org
Downloaded from ndt.oxfordjournals.org by guest on February 23, 2011
P<0.03) were found in 11 HD patients with PHT
that underwent successful transplantation. Similarly,
temporary closure of the A-V access by a sphygmo-
manometer in eight patients with PHT resulted in
a transient decrease in CO (from 6.4±0.6 to 5.3±
0.5 l/min, P ¼ 0.18) and systolic PAP (from 47.2±3.8
to 34.6±2.8 mmHg, P<0.028).
Conclusions. This study demonstrates a high prev-
alence of PHT among patients with ESRD on
chronic HD via a surgical A-V fistula. In view of the
vasodilatory and antimitogenic properties of NO, it
is possible that the attenuated basal and HD-induced
NO production in patients with PHT contributes to
the increased pulmonary vascular tone. Furthermore,
the partial restoration of normal PAP and CO in HD
patients that underwent either temporal A-V shunt
closure or successful transplantation indicates that
excessive pulmonary blood flow is involved in the
pathogenesis of the disease.
Keywords: arterio-venous access; endothelin;
end-stage renal disease; haemodialysis;
nitric oxide; pulmonary hypertension
Introduction
Pulmonary hypertension (PHT) is an elevation of
pulmonary arterial pressure (PAP) that can be the
result of heart, lung or systemic disorders [1,2].
Regardless of the aetiology, the morbidity and the
mortality from long-standing PHT exceed that
expected from the causative condition. PHT involves
vasoconstriction and obliteration of the lumen of
small vessels in the lungs by plexiform lesions resulting
in increased resistance to flow [3]. Proposed mecha-
nisms for the formation of the plexiform lesion include
dysregulation of endothelial growth and angiogenic
response to local triggers [4].
Pulmonary hypertension in haemodialysis patients
We have recently shown a 40% incidence of PHT
as detected by Doppler echocardiography in patients
with end-stage renal disease (ESRD) on chronic
haemodialysis (HD) therapy via arterial–venous
(A-V) access [5]. Affected patients had significantly
higher cardiac output (CO). Temporary A-V access
closure and successful kidney transplantation caused a
significant fall in PAP values to a normal range. Based
on these data, we presumed that the pulmonary
circulation of some uraemic patients on chronic HD
therapy is in a vasoconstriction state. Most probably,
the A-V access-induced high CO contributes to the
development of provoked PHT.
Local vascular tone and function are regulated by
the balance between vasodilators, such as prostacyclin
and nitric oxide (NO), and vasoconstrictors, such
as thromboxane A2 and endothelin-1 (ET-1) [1,4].
The current study was carried out to investigate the
possibility that PHT in these patients is triggered/
aggravated by chronic HD via a surgical A-V access,
and to explore the involvement of the ET-1
and NO systems in the pathogenesis of PHT in HD
patients.
Patients and methods
Patients
Our institutional clinical research ethics review board
approved the research protocol and informed consent
was obtained from all participants. Twenty-eight
patients from the initial study population (of 58
ESRD patients on chronic HD therapy via A-V
access) were excluded from this study, due to known
cardiac, pulmonary and systemic diseases. Five patients
underwent kidney transplantation, nine patients died
during the follow-up and 14 patients refused to
participate in the study. Twelve additional patients,
not included in the original study, increased the
study population to 42 eligible patients without
known cause of PHT. The control group consisted
of 20 normal volunteers, all without known cause
for PHT.
Patient evaluation
Systolic PAP was estimated in the 42 HD patients
by Doppler echocardiography as described by
Gladwin et al. [6]. To avoid overestimation of PAP
due to volume overload, the echo studies in HD
patients were performed within 1 h after completion
of HD, when the patients were at optimal dry weight
according to hydration status, blood pressure and
weight.
An experienced operator (S.A.R.) performed all the
echocardiographic studies, using an Acuson Sequoia,
Aspen or 128 XP (Mountain View, CA) ultrasound
machine. A complete two-dimensional, M-mode and
Doppler echocardiographic study was obtained from
each patient. A tricuspid regurgitation systolic jet
1687
was recorded from the parasternal or apical window
with the continuous-wave Doppler probe. Systolic
right ventricular (or pulmonary artery) pressure was
calculated using the modified Bernoulli equation:
PAP ¼ 4
(tricuspid systolic jet)2 þ 10 mmHg (esti-
mated right atrial pressure). The accuracy of systolic
PAP estimation by Doppler in our laboratory was
previously published [7]. PHT was defined as a
systolic PAP 35 mmHg. Stroke volume was esti-
mated from the left ventricular outflow tract velocity
time integral
diameter, and CO was calculated
by multiplying the stroke volume by the heart rate.
The general data concerning patients (age, sex,
co-morbidities and medications used) and kidney
disease (aetiology of renal failure, age at onset,
duration of HD therapy and access location, brachial
or radial) were recorded directly from the patients or
from their hospital files. Blood tests for haemoglobin,
haematocrit, calcium, phosphorus and parathyroid
hormone level were sampled at the end of the HD
Downloaded from ndt.oxfordjournals.org by guest on February 23, 2011
therapy within 1 week from the echocardiography
study. For measures sampled frequently (haemoglobin,
haematocrit, calcium and phosphorus), the mean
values from the 6 months preceding the echo study
were presented. Plasma levels of intact PTH were
determined by two site chemoluminescent enzyme-
labeled immunometric assay (DPC, LA, CA, USA).
Patients with PHT >35 mmHg were evaluated further
by an experienced pulmonologist (M.Y.) with the
intention to disclose other causes of PHT. They under-
went chest X-radiography and computed tomography,
complete pulmonary function tests, measurement
of blood gases and O2 saturation, and ventilation–
perfusion lung scan as previously described [1,5].
We also examined the effect of A-V access compres-
sion with a sphygmomanometer for 1 min (n ¼ 8) and
the effect of successful kidney transplantation (n ¼ 11)
on PAP and CO values in HD patients with PHT. The
corresponding PAP and CO values were compared
before and after the A-V closure for 1 min, or pre- and
post-kidney transplantation. Time intervals between
renal transplantation and measurement of PAP and CO
range between 11 months and 4 years.
Determination of ET-1
Plasma ET-1 levels were determined in 42 patients
on chronic HD, with either normal pulmonary blood
pressure (PAP <35 mmHg) or PHT (PAP >35 mmHg)
by enzyme-linked immunosorbent assay (ELISA;
Biomedica, Vienna, Austria). All the blood samples
were collected into pre-cooled tubes containing potas-
sium EDTA and the protease inhibitor aprotinin,
before and after the HD procedure. Plasma sepa-
rated immediately in a cold centrifuge at 3000 r.p.m.
and samples were stored at 70 C until analysis
was performed in a single assay. Twenty matched
healthy volunteer subjects served as controls. The
ET-1 antibody employed in this assay has a 100%
cross-reactivity with ET-2, and <5% and <1% cross-
reaction with ET-3 and Big ET-1, respectively.
1688 F. Nakhoul et al.
Determination of NO2 þ NO3 Table 1. Data on 42 patients with ESRD on HD via surgical
A-V access
NO metabolites (NO2 and NO3) in plasma of the
patients and volunteers described above were deter- Age (years)
mined before and after the dialysis procedure with Mean age 60.8±11.8
the Griess method using a commercial kit. This kit Range 31–81
is a colorimetric assay applying nitrate reductase, Sex
which converts nitrate to nitrite, and total nitrite Male/female 27/15
M/F ratio 1.8
was measured at 540 nm absorbance by reaction
Aetiology of renal failure
with Griess reagent (sulfanilamide and naphthalene- Diabetes mellitus 15
ethylene diamine dihydrochloride). Amounts of nitrite Hypertension 12
in the plasma were estimated by a standard curve Glomerulonephritis 6
obtained from enzymatic conversion of NaNO3 to Chronic pyelonephritis 4
nitrite. Twenty matched healthy volunteer subjects Nephrolithiasis 3
Unknown 2
served as controls for determining the basal circulatory
Duration of dialysis (months)
levels of ET-1 and NO metabolites. Mean 43±29
Range 6–120
Follow-up period from echo (months)
Data analysis and statistics Mean 34±35
Clinical variables were compared between HD patients Range 21–65

Downloaded from ndt.oxfordjournals.org by guest on February 23, 2011

with and without PHT by using analysis of variance Outcome
Died 7 (16.6%)
(ANOVA), followed by t-test. The correlation between Alive 35
categorical variables, such as PHT and survival from
onset of HD therapy, was investigated by the 2 test.
Values are expressed as the mean±SEM. P-values of
<0.05 were considered as significant.
Seven (16.6%) patients died during follow-up, four
with and three without PHT, providing mortality rates
Results of 20 and 14%, respectively, P ¼ NS.

Data of the study patients (n ¼ 42), 27 men, 15 women,
mean age of 60.8±11.8 years, range 31–81, receiving
HD therapy are summarized in Table 1. Mean duration
of HD therapy was 43±29 months, range 6–120. The
common aetiologies of renal failure were diabetes
mellitus, arterial hypertension and glomerulopathy.
Twenty patients (48%) had PHT.
Data from the 20 HD patients with PHT were
compared with those of the 22 HD patients without
PHT (Table 2). The CO was significantly higher
among the PHT subgroup (5.95±1.2 vs 5.2±
0.9 l/min, P<0.034) (Table 2). The height and weight
were not significantly different between both sub-
groups. Similarly, the haemoglobin and haematocrit
levels did not differ significantly between patients with
and without PHT (10.6±1.6 vs 11.2±1.7 and 33.3±5.4
vs 35.7±5.1, respectively). The incidence of antihyper-
tensive medications and the distribution of these agents
did not differ significantly between the two subgroups.
The mean duration of HD therapy was not signifi-
cantly lower in the PHT subgroup (35±25 vs 50±30
months). Other variables, such as anatomic location
of the dialysis vascular access, parathyroid hormone
levels and calcium–phosphorus product, did not differ
significantly between the examined subgroups. Most
patients (70–73%) from both subgroups (with and
without PHT) tolerated the HD procedure and did
not develop severe hypotension. However, six patients
in each subgroup were classified as classic dialysis
intolerant. No correlation was found between the
incidence of HD intolerance and plasma NO levels.
Plasma ET-1 levels
As shown in Figure 1A, plasma concentrations of
ET-1 in HD patients (combined with and without
PHT) were significantly higher than those obtained
in normal subjects (2.13±0.37 vs 0.99±0.17 fmol/ml,
P<0.05). Plasma levels of ET-1 were not changed
by the dialysis procedure. The plasma levels of ET-1 did
not differ significantly between HD patients with and
without PHT (1.60±0.67 vs 2.35±0.77 fmol/ml before
HD and 1.58±0.68 vs 2.43±0.79 fmol/ml after HD)
(Figure 1B).
Plasma NO metabolite levels
As shown in Figure 2A, pre-HD plasma concentra-
tions of NO2 þ NO3 in the study population were
similar to those obtained in normal subjects
(19.73±3.13 vs 22.30±3.80 mM). However, follow-
ing the HD procedure, plasma levels of NO2 þ
NO3 increased from 19.73±3.13 to 60.29±8.00 mM
(P<0.001). Compared with HD without PHT, patients
with PHT had lower pre-HD circulatory levels of NO
metabolites (14.34±2.32 vs 24.17±5.21 mM, P<0.05).
Interestingly, following the HD procedure, plasma
NO2 þ NO3 levels of both subgroups of HD patients
increased significantly (Figure 2B). However, this
increase was more remarkable in patients without
PHT (from 24.17±5.21 to 77.10±9.60 mM, P<0.0001)
compared with PHT patients (from 14.34±2.32
to 39.87±11.41 mM, P<0.0074).
Pulmonary hypertension in haemodialysis patients 1689
Table 2. Clinical and laboratory data between patients with and A
without PHT 3.0
Control *
Elevated PAP Normal PAP P-value 2.5
HD Patients Before HD *
HD Patients After HD

Plasma ET-1 (fmol/ml)

2.0
No. of patients 20 (39.7%) 22 (60.3%)
Gender 1.5
Male 13 14
Female 7 8 1.0
Age (years)
Mean 62.8±11 59±12.2 NS 0.5
Range 39–81 31–78
Height (cm) 0.0
Mean 160.8±2.89 166.2±1.02 0.09 B
Range 154–168 164–170
Weight (kg) 3.2 Control
Mean 72.5±12.6 62.4±6.4 0.47 HD Patients Before HD
2.8

Plasma ET-1 (fmol/ml)

Range 54–109 46–85 HD Patients After HD
2.4
PAP (mmHg)
Mean 46.3±2.1 29.4±1.5 0.0001 2.0
Range 37–62 20–35 1.6

Downloaded from ndt.oxfordjournals.org by guest on February 23, 2011

CO (l/min)
1.2
Mean 5.95±1.2 5.2±0.9 0.034
Range 3.6–7.5 3.84–7 0.8
Time interval (months)a 0.4
Mean 34.9±25.2 50±30.1 NS
Range 6–106 12–120 0.0
Control HD patients HD patients
Medications W/O PHT With PHT
Anti-hypertensive therapy 14 14 NS
Beta blockers 6 5 NS Fig. 1. Plasma endothelin (ET-1) levels in normal volunteers
(n ¼ 29) and all HD patients included in the present study (n ¼ 42)
Vasodilators
pre- and post-HD (A), *P<0.05 compared with normal volunteers.
Ca channel blockers 4 5 NS
(B) Plasma levels of ET-1 in HD patients with (n ¼ 20) and without
ACE inhibitors 9 8 NS
(n ¼ 22) PHT pre- and post-HD.
Haemoglobin (mean±SD) 10.6±1.6 11.2±1.7 NS
Haematocrit (%) 33.3±5.4 35.7±5.1 NS
Intact PTH (ng/h)
Mean 260±174 274±197
Range 32–649 20–771 NS
was closed (PAP decreased from 47±2.4 to
Ca
P product (mg/dl) 37.1±1.7 mmHg, P<0.014, n ¼ 7) or opened (PAP
Mean 47±10.1 47.5±11.5 NS
Range 32.7–67 35.2–67.6 decreased from 54.8±6.1 to 40.8±5.1 mmHg, P ¼ 0.13,
Shunt location
n ¼ 4). Similarly, CO was reduced in transplanted
Brachial 17 18 NS patients with both closed A-V fistula (from 8.1±0.92
Radial 3 4 to 6.0±0.35 l/min, P<0.05) and opened fistula (from
Outcome 6.8±0.74 to 5.9±0.20 l/min, P ¼ 0.27).
Alive 16 19 NS
Dead 4 3
A-V access compression
PAP ¼ pulmonary artery pressure; PTH ¼ parathyroid hormone.
a
Time interval from onset of HD therapy to detection of PAP and CO were measured in eight HD patients
pulmonary hypertension. with significant PHT before and 1 min after A-V access
compression. During this maneouvre, the systolic
PAP decreased from 47.2±3.8 to 34.6±2.8 mmHg,
P<0.028 (Figure 3C), while the CO decreased from
Kidney transplantation 6.4±0.6 to 5.3±0.5 l/min, P ¼ 0.18 (Figure 3D).
Eleven HD patients with PHT underwent success-
ful kidney transplantation during the study
period. The pre-transplantation mean systolic PAP Discussion
of 49.8±2.8 mmHg (range, 37–65) decreased to
38.6±2.2 mmHg (range, 28–50), P<0.004 (Figure 3A). PHT is defined as a sustained elevation of PAP
The observed reduction in PAP following transplan- to >25 mmHg at rest or to >30 mmHg with exercise.
tation was associated with a significant decline The ability of the pulmonary circulation to dilate
in CO from 7.6±0.6 to 6.1±0.3 l/min, P<0.03, and recruit unused blood vessels prevents development
(Figure 3B). The reduction in PAP following kidney of PHT in conditions characterized by increased
transplantation was observed whether the A-V fistula pulmonary blood flow [3]. Pulmonary hypertensive
1690
A that affected patients had diminished NO production
70
Control
60 HD patients-Before HD
Plasma NO2+NO3 (µM)
HD Patients-After HD
50
40
30
20
10
0
B endothelial cells and is involved in the regulation
90
Control
80 HD Patients Before HD
Plasma NO2+NO3 (µM)
HD Patients After HD
70
60
50
40
30
20
10
0
Control HD patients
W/O PHT
Fig. 2. Plasma NO2 þ NO3 levels in normal volunteers (n ¼ 29) and
all HD patients included in the present study (n ¼ 42) pre- and
post-HD (A), *P<0.05 compared with normal volunteers (n ¼ 20),
#
P<0.05 compared with HD patients before HD. (B) Plasma levels
of NO2 þ NO3 in HD patients with (n ¼ 20) and without (n ¼ 22)
PHT pre- and post-HD. $P<0.05 compared with pre-dialysis NO
metabolite levels in the same subgroup of HD patients.
disease in HD patients is a relatively new entity in
which elevated CO is an important factor. Previously,
we reported that PAP increased significantly follow-
ing initiation of HD therapy via A-V access and
decreased significantly after successful kidney trans-
plantation, as well as after short A-V access compres-
sion, indicating that both ESRD and A-V access
contribute to the pathogenesis of PHT. However, the
fact that only about half of the uraemic patients
developed PHT suggests that mechanisms other than
uraemia may be involved in this disorder. These may
include underlying concomitant diseases such as
diabetes mellitus or systemic hypertension, although
the percentages of diabetic or hypertensive patients
in both subgroups of patients with and without PHT
were identical.
This study investigated the role of two endothelial-
derived molecules, ET and NO, in the pathogenesis
of this syndrome. Diminished vasodilatation response
to the A-V access-induced elevated CO is a possible
explanation for the elevated PAP in some uraemic
patients. Endothelial dysfunction that has been
described in PHT [8,9] and in uraemia [10] is a proposed
mechanism.
In line with our previous findings that were published
2 years ago [5], 48% of the patients had PHT, with
significantly elevated CO. The present findings indicate
F. Nakhoul et al.
#
*
while their ET system remained intact.
Based on the data presented, we may conclude
that the pulmonary hypertensive disease of HD patients
is a unique form of PHT in which elevated CO
and uraemic-induced endothelial dysfunction exist.
This endothelial dysfunction manifests as reduced
NO production, augments the tonus of the pulmonary
vascular, reduces the capacity of the pulmonary
circulation to maintain the A-V access-mediated
elevated CO and subsequently causes PHT.
NO is a gaseous molecule that is produced in the
$ of tonus in both the pulmonary and systemic circula-
tions [11]. Normally, the pulmonary endothelial cells
increase the synthesis of NO in response to increased
$ pulmonary blood flow and pressure in an attempt to
restore normal vascular tone [12]. Reduced expression
of endothelial NO synthase in the lungs of patients
with PHT suggests that reduced production of this
Downloaded from ndt.oxfordjournals.org by guest on February 23, 2011
molecule may be involved in the pathogenesis of PHT
[13]. Impaired NO production and reduced sensitivity
to NO have been described in patients with chronic
HD patients renal failure [14], but the impact of this fault on the
With PHT pulmonary circulation in uraemia has not been studied
previously. So far, endothelial dysfunction in uraemia
has been related to the pathogenesis of hypertension
and to accelerated atherosclerosis. This study demon-
strates, for the first time, a link between impaired NO
production and PHT in uraemic patients receiving
chronic HD therapy via A-V access. Previous studies
demonstrated that NO production is enhanced in HD
patients following the HD procedure. This effect is
attributed to the biocompatibility of the dialyser. The
mechanisms that interfere with NO activity in uraemia
are still not clear. Occurrence of endothelial dysfunc-
tion in all spectra of chronic renal failure suggests
that uraemia may directly promote this fault. Reduced
NO bio-availability of the NO substrate L-arginine,
reduced expression of NO synthase in the relevant
organs, interaction of NO with reactive oxygen species
and accumulation of endogenous inhibitors of NO
synthase, such as asymmetric dimethylarginine and
homocysteine, are all proposed mechanisms [10,15].
Changes in shear stress can induce patterns of gene
expression that can alter endothelial function or the
release of mediators that influence the behaviour of
endothelial and smooth muscle cells [16]. For example,
increased shear stress is considered a major event in
the pathogenesis of PHT and vascular remodelling in
congenital left-to-right shunt. Similarly, creation of
A-V access for HD therapy may increase both CO and
pulmonary blood flow, alter shear stress and contribute
indirectly to the development of PHT.
ET-1 is a potent vasoconstrictor and a powerful
mitogen that has been related to PHT in many ways
[8,9,17]. ET-1 levels are increased in humans with PHT
[11]. High arterial levels compared with venous levels
suggest that it is produced in the lungs [17]. Enhanced
expression of the ET-1 gene and its receptor subtypes
(ETA and ETB) has been demonstrated in lungs of
Pulmonary hypertension in haemodialysis patients 1691
A C
Before Transplantation Before A-V closure
50 50
After Transplantation After A-V closure

40
* 40

PAP (mmHG)
*
PAP (mmHg)

30 30

20 20

10 10

0 0

B D Before A-V Closure

Before Transplantation
8 After Transplantaion 7 After A-V Closure
Cardiac Output (L/min)

Cardiac Output (L/min)

7
6
*
6
5
5

Downloaded from ndt.oxfordjournals.org by guest on February 23, 2011

4
4
3
3
2
2

1 1

0 0

Fig. 3. Pulmonary arterial pressure (PAP) and cardiac output (CO) in HD patients with pulmonary hypertension (PHT) before and after
kidney transplantation (A and B, n ¼ 11), or A-V access compression (C and D, n ¼ 8). *P<0.05 compared with the relevant value before
either transplantation or A-V closure.

humans with PHT [9,17]. Increased ET activity has
also been reported in uraemia [8,9,17]. The current
study does not indicate a role for ET-1 in the
pathogenesis of this novel type of PHT. Nevertheless,
taking into account that ET-1 is known to be secreted
by the endothelial cells toward the vascular smooth
muscle cells in a polar way, we do not exclude the
possibility that local concentrations of ET-1 are
enhanced in the pulmonary tissue of HD patients
with PHT. This notion is supported by the findings that
bosentan, an ET-1 antagonist, significantly reduced
PHT in a 19-year-old woman with ESRD [18].
As we have suggested previously, an additional
mechanism that may underlie the elevated PAP in HD
patients may be haemodynamic changes associated
with the placement of A-V fistula. The increase in CO
due to enhanced venous return to the heart and
subsequently exaggerated pulmonary blood flow fol-
lowing the creation of the A-V shunt indicate that these
haemodynamic alterations may also play a crucial
role in the development of PHT in these patients. This
notion is supported by our findings that PAP and
CO decreased significantly after successful kidney
transplantation, as well as after short A-V access
compression. Since the PAP was reduced comparably
in kidney recipients whether the fistula was opened
or closed, despite a more remarkable decline in CO in
the latter, these data suggest that non-haemodynamic
factors may be involved in the beneficial effects of
kidney transplantation. These factors may include
improvement in their inflammatory status following
the immunosuppressive therapy, cytokines, toxins,
vascular remodelling, endothelial function and others.
Therefore, in the present study, reduction in PAP
following reduction in CO seems to be due not only
to the flow dependency of PAP but also to a decrease
in pulmonary resistance [19]. Clarkson et al. [20]
recently reported PHT of 60 mmHg in an ESRD
patient receiving chronic HD therapy via an aneu-
rysmal brachicephalic A-V fistula. In line with our find-
ing, surgical ligation of the A-V fistula and insertion
of a tunnelled semi-permanent internal jugular dialysis
catheter reduced the PAP to 30 mmHg. We confirmed
this observation, with significant reduction of PAP
values (sometimes even to normal range) following
temporary A-V compression by sphygmomanometer
among eight HD patients, and noted normalization
of PAP values following successful kidney transplanta-
tion in nine additional patients. From these data, we
may conclude that the presence of both uraemia and
A-V access-mediated elevated CO is mandatory for
the development of PHT. The data presented by
Clarkson et al. [20] indicate an unusual pattern of
pulmonary hypertensive disease in HD patients, almost
complete reversibility following CO reduction or
amelioration of the uraemia. This reversibility of the
PHT is surprising considering the data concerning the
remodelling of pulmonary vessels in PHT. Although
1692
pathological studies are absent, this reversibility of the
PHT suggests that the pulmonary vasculature of the
study population only underwent minor remodelling
changes, if any.
Development of PHT in ESRD patients due
to endothelial dysfunction and A-V access-mediated
haemodynamic changes involves a therapeutic
dilemma. PHT is associated with increased morbidity
and mortality regardless of its aetiology. Similarly,
HD patients with PHT had significantly higher
mortality rates. Calcium channel blockers and
angiotensin-converting enzyme inhibitors alone or
in combination are likely to ameliorate endothelial
dysfunction. On the other hand, poor response of
PHT to medical therapy and reversibility of the PHT
in this population of patients following A-V access
closure or kidney transplantation are a call to refer
affected patients to other modes of dialysis without
A-V access, such as peritoneal dialysis, or to kidney
transplantation.
In summary, the present study demonstrates that
some uraemic patients on chronic HD therapy via
A-V access exhibit decreased NO production. This
endothelial dysfunction reduces the capacity of the
pulmonary circulation to maintain the A-V access-
mediated elevated CO and contribute to the develop-
ment of PHT.
Conflict of interest statement. None declared.
References
1. Archer S, Rich S. Primary pulmonary hypertension: a vascular
biology and translational research ‘work in progress’. Circulation
2000; 102: 2781–2791
2. Yigla M, Dabbah S, Azzam ZS, Rubin AH, Reisner SA.
Background diseases in 671 patients with moderate to
severe pulmonary hypertension. Isr Med Assoc J 2000; 2:
684–689
3. Rubin LJ. Pathology and pathophysiology of primary
pulmonary hypertension. Am J Cardiol 1995; 75: 51A–54A
F. Nakhoul et al.
4. Jeffery TK, Morrell NW. Molecular and cellular basis of
pulmonary vascular remodeling in pulmonary hypertension.
Prog Cardiovasc Dis 2002; 45: 173–202
5. Yigla M, Nakhoul F, Sabag A, Tov N, Gorevich B, Abassi Z,
Reisner SA. Pulmonary hypertension in patients with end-stage
renal disease. Chest 2003; 123: 1577–1582
6. Gladwin MT, Sachdev V, Jison ML et al. Pulmonary
hypertension as a risk factor for death in patients with sickle
cell disease. N Engl J Med 2004; 350: 886–895
7. Reisner SA, Azzam Z, Halmann M et al. Septal/free wall
curvature ratio: a noninvasive index of pulmonary arterial
pressure. J Am Soc Echocardiogr 1994; 7: 27–35
8. Chen YF, Oparil S. Endothelial dysfunction in the pulmonary
vascular bed. Am J Med Sci 2000; 320: 223–232
9. Giaid A, Yanagisawa M, Langleben D et al. Expression of
endothelin-1 in the lungs of patients with pulmonary
hypertension. N Engl J Med 1993; 328: 1732–1739
10. Morris ST, Jardine AG. The vascular endothelium in chronic
renal failure. J Nephrol 2000; 13: 96–105
11. Giaid A. Nitric oxide and endothelin in pulmonary hyper-
tension. Chest 1998; 114: 208S–212S
12. Stamler JS, Loh E, Roddy MA, Currie KE, Creager MA.
Nitric oxide regulates basal systemic and pulmonary vascular
Downloaded from ndt.oxfordjournals.org by guest on February 23, 2011
resistance in healthy humans. Circulation 1994; 89: 2035–2040
13. Wever R, Boer P, Hijmering M et al. Nitric oxide production
is reduced in patients with chronic renal failure. Arterioscler
Thromb Vasc Biol 1999; 19: 1168–1172
14. Vaziri ND. Effect of chronic renal failure on nitric oxide
metabolism. Am J Kidney Dis 2001; 38: S74–S79
15. Vapaatalo H, Mervaala E. Clinically important factors
influencing endothelial function. Med Sci Monit 2001; 7:
1075–1085
16. Dschietzig T, Richter C, Bartsch C et al. Flow-induced
pressure differentially regulates endothelin-1, urotensin II,
adrenomedullin, and relaxin in pulmonary vascular endo-
thelium. Biochem Biophys Res Commun 2001; 289: 245–251
17. Chen YF, Oparil S. Endothelin and pulmonary hypertension.
J Cardiovasc Pharmacol 2000; 35 [4 Suppl 2]: S49–S53
18. Rubin LJ, Badesch DB, Barst RJ et al. Bosentan therapy for
pulmonary arterial hypertension. N Engl J Med 2002; 346:
896–903
19. Naeije R. Pulmonary vascular resistance. A meaningless
variable? Intensive Care Med 2003; 29: 526–529
20. Clarkson MR, Giblin L, Brown A, Little D, Donohoe J.
Reversal of pulmonary hypertension after ligation of a
brachiocephalic arteriovenous fistula. Am J Kidney Dis 2002;
40: E8
Received for publication: 8.12.04
Accepted in revised form: 21.3.05