Hemodialysis International 2010; 14:398402

The long-term effects of arteriovenous

fistula creation on the development
of pulmonary hypertension in
hemodialysis patients

Aydin UNAL,1 Kutay TASDEMIR,2 Sema OYMAK,3 Mustafa DURAN,4 Ismail KOCYIGIT,1
Fatih OGUZ,4 Bulent TOKGOZ,1 Murat Hayri SIPAHIOGLU,1 Cengiz UTAS,1 Oktay OYMAK1
Departments of 1Nephrology, 2Cardiovascular Surgery, 3Pulmonology, 4Cardiology, Erciyes University
Medical School, Kayseri, Turkey

Abstract
The aim of this prospective study was to evaluate long-term effects of arteriovenous stula (AVF) on
the development of pulmonary arterial hypertension (PAH) and the relationship between blood ow
rate of AVF and pulmonary artery pressure (PAP) in the patients with end-stage renal disease (ESRD).
This prospective study was performed in 20 patients with ESRD. Before an AVF was surgically created
for hemodialysis, the patients were evaluated by echocardiography. Then, an AVF was surgically
created in all patients. After mean 23.50  2.25 months, the second evaluation was performed by
echocardiography. Also, the blood ow rate of AVF was measured at the second echocardiographic
evaluation. Pulmonary arterial hypertension was dened as a systolic PAP above 35 mmHg at rest.
Mean age of 20 patients with ESRD was 55.05  13.64 years; 11 of 20 patients were males. Pulmo-
nary arterial hypertension was detected in 6 (30%) patients before AVF creation and in 4 (20%)
patients after AVF creation. Systolic PAP value was meaningfully lower after AVF creation than before
AVF creation (29.95  10.26 mmHg vs. 35.35  7.86 mmHg, respectively, P: 0.047). However, there
was no signicant difference between 2 time periods in terms of presence of PAH (P40.05). Pulmo-
nary artery pressure did not correlate with blood ow rate of AVF and duration after AVF creation
(P40.05). In hemodialysis patients, a surgically created AVF has no signicant effect on the devel-
opment of PAH within a long-term period. Similarly, blood ow rate of AVF also did not affect
remarkably systolic PAP within the long-term period.

Key words: Arteriovenous stula, stula blood ow rate, hemodialysis, pulmonary hypertension

INTRODUCTION of the disorder have been evaluated in chronic kidney

Pulmonary arterial hypertension (PAH) is a newly recog-
nized disorder in patients with renal disease. There are
a few studies in which the prevalence and pathogenesis
Correspondence to: A. Unal, MD, Erciyes University Medical
School, Department of Internal Medicine, Nephrology
Division, Erciyes Universitesi T{p Fakultesi, Organ Nakli ve
Diyaliz Hastanesi, 38039, Kayseri, Turkey.
E-mail: aydinunal2003@gmail.com
disease. In many studies, the prevalence of pulmonary
hypertension was detected in a high ratio of approxi-
mately 12% to 45%.15 There are several explanations
including high cardiac output resulting from arteriove-
nous fistula (AVF), anemia, hypervolemia, and pulmo-
nary artery calcifications for pulmonary hypertension
in patients with renal disease.13,6,7 The aim of this
prospective study was to evaluate the long-term effects
of AVF, which was created surgically for vascular access,
on the development of pulmonary hypertension, and

r 2010 The Authors

Hemodialysis International r 2010 International Society for Hemodialysis
398 DOI:10.1111/j.1542-4758.2010.00478.x

the relationship between blood flow rate of AVF and
pulmonary artery pressure (PAP) in patients with end-
stage renal disease (ESRD).
PATIENTS AND METHODS
The patients
This prospective study was performed in 50 patients with
ESRD. The study protocol was approved by the local
ethics committee. Before an AVF was surgically created
for hemodialysis, the patients, who signed an informed
consent form before the first echocardiographic evalua-
tion, were evaluated by echocardiography. Then, an AVF
was surgically created in all patients. After mean
23.50  2.25 months, the second evaluation was per-
formed by echocardiography. Eight of the 50 patients died
during follow-up. Two patients underwent renal trans-
plantation. Six patients refused control echocardiography.
Fourteen patients did not respond our call (loss to follow-
up). Finally, 20 patients were included in this study.
Before AVF creation, some patients were not on dialysis.
On the other hand, hemodialysis was performed via a
temporary hemodialysis catheter in the other patients, in
whom there was an acute indication for dialysis such as
severe hyperkalemia, severe metabolic acidosis unrespon-
sive to sodium bicarbonate infusions, and severe over-
volemia unresponsive to diuretics. Furthermore, before
AVF creation, the patients, who had hypervolemia symp-
toms and findings such as pretibial edema, hepatomegaly,
tachycardia, dyspnea, orthopnea, and paroxysmal noctur-
nal dyspnea, were dialyzed via a temporary hemodialysis
catheter until normovolemia was obtained. Thereafter,
an initial echocardiographic evaluation was performed.
Median duration between an AVF creation and the initial
echocardiographic evaluation was 1 (09) day. At evalua-
tion after AVF creation, no patient had hypervolemia symp-
toms and findings. The blood flow rate of AVF was
also measured at second echocardiographic evaluation.
All echocardiographic evaluations were performed within
24 hours after hemodialysis session. Patients with chronic
obstructive pulmonary disease (COPD), severe mitral
or aortic valve disease, connective tissue disease, left
ventricular ejection fraction o50%, history of pulmonary
embolism, or chest wall or parenchymal lung disease
were excluded.
Blood samples
Blood samples were taken from all patients for laboratory
examinations including levels of albumin, hemoglobin,
Hemodialysis International 2010; 14:398402
AVF and pulmonary hypertension
intact parathyroid hormone (iPTH), and high-sensitivity
C-reactive protein (hsCRP), and white blood cell (WBC)
count on the same day on which the echocardiographic
evaluation was performed. The iPTH level was measured
by radioimmunoassay (Immunotech, Marseille, France).
Serum hsCRP was measured by CardioPhase hsCRP on a
Dade Behring analyzer (both by Dade Behring, Marburg,
Germany).
Echocardiographic evaluation
Echocardiographic evaluations were performed by using
the Vivid 7 Dimension (GE Medical Systems, Horten, Nor-
way) echocardiography machine. Two-dimensional and
M-mode Doppler echocardiographic images were obtained
from apical or parasternal windows in the left lateral
recumbent position for each patient and control. In the
presence of tricuspid valve regurgitation, systolic right
ventricular (or systolic pulmonary artery) pressure was
calculated using the modified Bernoulli equation:
PAP= 4  (tricuspid systolic jet)2110 mmHg.8 Pulmonary
arterial hypertension is defined as an elevation of mean
PAP above 25 mmHg at rest in the setting of normal or
reduced cardiac output and normal pulmonary capillary
pressure.9 If echocardiographic criteria are used, PAH
is defined as systolic PAP435 mmHg at rest.9 Therefore,
in the present study, PAH was defined as systolic
PAP435 mmHg at rest. End diastolic left ventricular septal
and posterior wall thickness and internal dimensions were
used to calculate left ventricular mass (LVM) using the
following equation: LVM= 0.8f1.04[(LVIDD1PWTD1
IVSTD)3  IVSTD3]g10.6 g, where LVIDD is the left
ventricular internal diameter in diastole, PWTD is the pos-
terior wall thickness in diastole, and IVSTD is the inter-
ventricular septum thickness in diastole.10 Left ventricular
hypertrophy was defined as the left ventricular mass index
(LVMI, calculated as LVM in grams divided by body surface
area in square meters) higher than 116.0 g/m2 for men and
104.0 g/m2 for women.10 Body surface area was calculated
using Mostellers formula.11
Statistical analysis
SPSS 11.0 software (SPSSFW; SPSS Inc., Chicago, IL, USA)
was used for the statistical analysis. Kolmogorov-Smirnov
test was used for normality analysis of quantitative vari-
ables. Continuous variables with normal distribution were
presented as mean  standard deviation. Statistical analysis
for the parametric variables including systolic PAP, LVMI,
ejection fraction, body mass index (BMI), systolic and di-
astolic blood pressures, levels of hemoglobin and albumin,
399
Unal et al.

and WBC count was performed by the paired t test. Median

value was used where normal distribution is absent. The
Wilcoxon signed-rank test was used to compare nonpara-
metric variables including hsCRP and iPTH. The qualitative
data were defined as percentages. The chi-square test and
Fisher exact test were used to compare qualitative data
including the presence of pulmonary hypertension and left
ventricular hypertrophy. The correlation analysis was
evaluated by the Pearsons correlation test for parametric
variables and by the Spearmans correlation test for non-
parametric variables. A P value o0.05 was considered Figure 1 Changes of pulmonary artery pressure (PAP)
statistically significant. values in individual patients over time.

RESULTS Figure 1 shows changes of PAP values in individual

The mean age of 20 patients was 55.05  13.64 years; 11
of 20 patients were males. The comparison of clinical,
biochemical, and echocardiographic parameters in
patients before and after an AVF creation are summarized
in Table 1. There was no significant difference between
the 2 time periods with regard to the presence of pulmo-
nary hypertension although systolic PAP was meaning-
fully lower after an AVF creation than before an AVF
creation (P: 0.047). Levels of hemoglobin and serum
albumin were significantly higher after an AVF creation
than before an AVF creation (P: 0.001 and 0.001, respec-
tively). There was no significant difference between
2 time periods in terms of BMI, systolic and diastolic
blood pressures, WBC count, levels of iPTH and hsCRP,
ejection fraction, LVMI, and the presence of left ventric-
ular hypertrophy (P40.05).
patients over time.
Pulmonary artery pressure, which was evaluated after
an AVF creation, did not correlate with blood flow rate of
AVF, duration after an AVF creation, and iPTH level
(P40.05).
DISCUSSION
Pulmonary arterial hypertension is a common disorder
among patient receiving hemodialysis via arteriovenous
access. Its prevalence is higher of approximately 25%
to 45%.14 Similarly, in the present study, pulmonary
hypertension was observed in 6 (30%) of 20 patients
before an AVF creation while 4 (20%) of them had it after
an AVF creation.
In a study performed in 58 patients with ESRD, Yigla
et al. re-evaluated 6 predialysis patients, who had normal

Table 1 Comparison of clinical, biochemical, and echocardiographic parameters in patients before and after an AVF creation

Before AVF creation After AVF creation P value

2
Body mass index (kg/m ) 24.99  4.51 25.16  4.30 0.418
Systolic blood pressure (mmHg) 130.75  11.27 129.25  9.77 0.594
Diastolic blood pressure (mmHg) 77.75  8.02 80.50  8.87 0.231
White blood cell count (mm3) 8153  3585 8018  2685 0.814
Hemoglobin (g/dL) 9.30  1.55 11.40  1.40 0.001
Serum albumin level (g/dL) 3.00  0.85 4.12  0.49 0.001
Intact parathormone level (pg/mL) 197.95 (6.701599.00) 193.00 (11.30757.00) 0.550
hsCRP (mg/dL) 9.44 (3.17148.00) 4.29 (3.0890.00) 0.454
Ejection fraction (%) 61.60  5.84 63.15  7.05 0.430
Systolic PAP (mmHg) 35.35  7.86 29.95  10.26 0.047
Presence of pulmonary hypertension 6 (30%) 4 (20%) 0.657
LVMI (g/m2) 116.44  29.35 117.53  24.36 0.904
Presence of left ventricular hypertrophy 9 (45%) 14 (70%) 0.217
AVF= arteriovenous fistula; hsCRP = high-sensitive C-reactive protein; LVMI =left ventricular mass index; PAP = pulmonary arterial pressure.
*Signifies Po0.05.

400 Hemodialysis International 2010; 14:398402


PAP values before the onset of hemodialysis therapy, after
3 to 54 months after hemodialysis therapy and showed a
significant elevation of PAP values in 4 of 6 patients. Also,
PAP values decreased into the normal range after renal
transplantation in 4 of 5 patients with pulmonary hyper-
tension. Similarly, they observed that PAP values signifi-
cantly decreased after arteriovenous access compression
in 4 patients with pulmonary hypertension and hypoth-
esized that both uremia itself and arteriovenous access
contribute to the development of unexplained pulmonary
hypertension in patients with ESRD.1 In the present
study, however, we observed that the mean PAP value
was significantly lower after an AVF creation than before
the onset of hemodialysis via AVF. Also there was no sig-
nificant difference between 2 time periods in terms of the
presence of pulmonary hypertension; even it decreased
from 30% to 20% although there was no statistical sig-
nificance. Our observations disprove their hypothesis.
Havlucu et al. observed that AVF flow rate was posi-
tively correlated with PAP value in hemodialysis patients.6
On the other hand, Acarturk et al. reported that there was
no correlation between PAP value and AVF flow rate.2
Similarly, we observed no correlation between blood flow
rate of AVF and systolic PAP value.
Anemia and fluid overload, which are complications
that frequently occurred in patients with kidney disease,
deteriorate pulmonary hypertension.1 We observed that
levels of hemoglobin and serum albumin were signifi-
cantly higher after an AVF creation than before an AVF
creation. This condition possibly resulted from the use of
erythropoietin (EPO) and correction of malnutrition due
to uremia-induced loss of appetite. Before AVF creation,
the patients were not on routine dialysis and not receiving
EPO and post-AVF creation they were receiving EPO and
were better nourished and cared for. Correction of ane-
mia and hypoalbuminemia might contribute to the re-
duction of PAP observed after an AVF creation, although
no patient had hypervolemia symptoms and findings
such as pretibial edema, hepatomegaly, tachycardia, dys-
pnea, orthopnea, and paroxysmal nocturnal dyspnea.
It is well known that the relationship between inflam-
mation and cardiovascular disease in patients with
chronic kidney disease.12 In a study performed in
patients with COPD, it was reported that levels of
inflammation markers including CRP and tumor necro-
sis factor-a were higher in patients with pulmonary
hypertension compared with in those without pulmonary
hypertension and there was a significant relation between
CRP levels and systolic PAP values.13 In an other study
performed in patients with Gaucher disease, Elstein et al.
found that high CRP level is an important predictor for
Hemodialysis International 2010; 14:398402
AVF and pulmonary hypertension
pulmonary hypertension.14 On the other hand, we de-
tected no significant difference between peritoneal dialy-
sis (PD) patients with and PD patients without pulmonary
hypertension in terms of hsCRP.5 Similarly, in the present
study, there was no significant difference between before
an AVF creation and after an AVF creation with respect to
inflammation markers including WBC count and hsCRP.
Vascular calcifications are a very common finding in
patients with ESRD and an important risk factor for car-
diovascular death. Impairment of calcium-phosphorus
balance and secondary hyperparathyroidism play an im-
portant role in the pathogenesis of vascular calcifications
in patients with ESRD.15 It was reported that elevated
PTH levels induce pulmonary calcifications and pulmonary
hypertension in chronic renal failure.7 However, in the
present study, iPTH levels did not differ significantly be-
tween before an AVF creation and after an AVF creation.
Also the PAP value did not correlate with the iPTH
level.
In conclusion, in hemodialysis patients a surgically
created AVF has no significant effect on the development
of PAH within a long-term period. Similarly, blood flow
rate of AVF also did not affect remarkably systolic PAP
within the long-term period.
Limitations to the study
1. The number of patients in the study was relatively
low. In the literature, however, the number of
patients was also low in studies in which PAH was
investigated in patients with ESRD. In the future, the
matter will be evaluated in multicenter studies.
2. The patients with double-lumen tunneled cuffed
catheters could be included in the study as a control
group. However, we never prefer the catheters as the
first choice for vascular access. The patients, who
were enrolled in the study, had newly diagnosed
ESRD and started HD for renal replacement therapy.
Therefore, the patients with tunneled cuffed cathe-
ters were not included in the present study because
of the above-mentioned reason.
3. Right heart catheterization is the best method to es-
timate PAP. However, in the literature, to the best of
our knowledge, in all studies in which pulmonary
hypertension was investigated in patients with ESRD,
PAP was estimated by the echocardiographic method.
Also, Marangoni et al. reported that there was an ex-
cellent correlation between the measurements of
PAP by Doppler echocardiography and by invasive
method.16 In addition, invasive method in estimating
PAP is a difficult and a traumatic procedure for the
401
Unal et al.
patients. Also, it was difficult to obtain the approval
of patients, who were included in the study, for in-
vasive procedure because of the above-mentioned
reasons. Therefore, we chose the echocardiographic
method to estimate systolic PAP because of the above
mentioned reasons.
4. Volume status of the patients was subjectively eval-
uated according to clinical findings and symptoms.
Unfortunately, this evaluation was not performed
with relatively objective methods such as bioimped-
ance analysis.
Manuscript received January 2010; revised July 2010.
REFERENCES
1 Yigla M, Nakhoul F, Sabag A, et al. Pulmonary hyperten-
sion in patients with end-stage renal disease. Chest. 2003;
123:15771582.
2 Acarturk G, Albayrak R, Melek M, et al. The relationship
between arteriovenous fistula blood flow rate and pul-
monary artery pressure in hemodialysis patients. Int Urol
Nephrol. 2008; 40:509513.
3 Amin M, Fawzy A, Hamid MA, et al. Pulmonary hyper-
tension in patients with chronic renal failure: Role of
parathyroid hormone and pulmonary artery calcifica-
tions. Chest. 2003; 124:20932097.
4 Tarrass F, Benjelloun M, Medkouri G, et al. Doppler
echocardiograph evaluation of pulmonary hypertension
in patients undergoing hemodialysis. Hemodial Int. 2006;
10:356359.
5 Unal A, Sipahioglu M, Oguz F, et al. Pulmonary hyper-
tension in peritoneal dialysis patients: Prevalence and
risk factors. Perit Dial Int. 2009; 29:191198.
402
6 Havlucu Y, Kursat S, Ekmekci C, et al. Pulmonary
hypertension in patients with chronic renal failure.
Respiration. 2007; 74:503510.
7 Akmal M, Barndt RR, Ansari AN, et al. Excess PTH in
CRF induces pulmonary calcification, pulmonary hyper-
tension and right ventricular hypertrophy. Kidney Int.
1995; 47:158163.
8 Dabestani A, Mahan G, Gardin JM, et al. Evaluation of
pulmonary artery pressure and resistance by pulsed
Doppler echocardiography. Am J Cardiol. 1987; 59:
662668.
9 Rosenkranz S. Pulmonary hypertension: Current diag-
nosis and treatment. Clin Res Cardiol. 2007; 96:527541.
10 Devereux RB, Wachtell K, Gerdts E, et al. Prognostic
significance of left ventricular mass change during treat-
ment of hypertension. JAMA. 2004; 292:23502356.
11 Mosteller RD. Simplified calculation of body surface area.
N Engl J Med. 1987; 317:1098.
12 Menon V, Wang X, Greene T, et al. Relationship between
C-reactive protein, albumin, and cardiovascular disease
in patients with chronic kidney disease. Am J Kidney Dis.
2003; 42:4452.
13 Joppa P, Petrasova D, Stancak B, et al. Systemic inflam-
mation in patients with COPD and pulmonary hyper-
tension. Chest. 2006; 130:326333.
14 Elstein D, Nir A, Klutstein M, et al. C-reactive protein
and NT-proBNP as surrogate markers for pulmonary hy-
pertension in Gaucher disease. Blood Cells Mol Dis. 2005;
34:201205.
15 Davies MR, Hruska KA. Pathophysiological mechanisms
of vascular calcification in end-stage renal disease. Kidney
Int. 2001; 60:472479.
16 Marangoni S, Quadri A, Dotti A, et al. Noninvasive as-
sessment of pulmonary hypertension: A simultaneous
echo-Doppler hemodynamic study. Cardiology. 1988;
75:401408.
Hemodialysis International 2010; 14:398402