Hemodialysis requires access to blood vessels capa-
ble of providing rapid blood flow, mainly achieved
by native fistulas and synthetic polytetrafluoroethyl-
ene grafts. However, complications associated with
vascular access often occur. Vascular access compli-
cations are not only a major cause of morbidity in
hemodialysis patients but also a major cost, ac-
counting for approximately 20% of the total spend-
ing for hemodialysis patients in the United States
(1). In addition, many reports confirm that access-
related morbidity comprises almost 25% of all hos-
pital admissions for end-stage renal disease patients
and may contribute to as much as 50% of all hospi-
Thrombosis is considered the most frequent com-
plication of vascular grafts and, when not correct-
ed, accounts for more than 80% of arteriovenous
access loss. The major predisposing factor is
anatomic venous stenosis, which is responsible for
85% of thromboses (1, 5).
In addition to correcting stenotic lesions, several
pharmacological strategies have been attempted to
lower the risk of thrombosis associated with chron-
ic vascular access. These attempts included treat-
ment with dipyridamole, low-dose aspirin with or
without sulfinpyrazone, and aspirin plus clopido-
grel (6-8). However, due to conflicting results, the
role of antiplatelet agents in preventing access
thrombosis remains unclear.
We decided to assess the risk of thrombosis in syn-
The Journal of Vascular Access 2005; 6: 29-33ORIGINAL ARTICLE
Clopidogrel diminishes hemodialysis access graft
For personal use only
H. TRIMARCHI1, P. YOUNG2, M. FORRESTER1, J. SCHROPP1, H. PEREYRA1, E. FREIXAS1
1Nephrology Section, and 2Internal Medicine Section, Department of Medicine, Hospital Británico de Buenos Aires -
Abstract: Background: The most common complication of hemodialysis access graft is thrombosis. Clopidogrel,
an inhibitor of platelet aggregation, was assessed to prevent this serious complication.
Methods: Nineteen patients on chronic hemodialysis whose vascular accesses were grafts were divided into two
groups: Group A (n=11, 58%) consisted of patients who did not receive anti-thrombotic therapy after graft place-
ment; Group B (n=8, 42%) received clopidogrel 75 mg/day from two days after surgery onwards. Both groups
were well matched with respect to age, gender, cause of renal failure, hematocrit, platelet count and Kt/V. All
patients’ thrombotic episodes were followed up from the day of graft surgery until thrombosis was diagnosed.
Finally, the survival difference between both groups was determined.
Results: Ten thrombotic episodes were diagnosed in Group A while no events were reported in Group B
(p<0.001). Graft access days of patency were significantly more in Group B than in Group A (350.8±166 vs
86.8±69, p<0.001). The time elapsed from dialysis initiation to graft placement was not different (Group A:
18±12 days; Group B: 20±10 days). Days in hemodialysis were different between both groups (Group A:
195.9±96; Group B: 545.5±291, p<0.001) and all patients of Group A (n=11, 57.9%) and two patients of Group
B (25%) died (p=0.001). No major bleeding events were reported.
Conclusions: Clopidogrel significantly decreased thrombotic graft episodes. Patients on clopidogrel had a pro-
longed vascular access patency, longer time on hemodialysis and longer survival. (The Journal of
Vascular Access 2005; 6: 29-33)
Key words: Clopidogrel, Hemodialysis, Graft, Vascular access, Thrombosis, Fistula, Atherosclerosis
1129-7298/ 029-05$15.00/ 0
© Wichtig Editore, 2005
thetic polytetrafluoroethylene grafts in patients re-
ceiving 75 mg of clopidogrel daily.
MATERIALS AND METHODS
For personal use only
Nineteen chronic hemodialysis patients with syn-
thetic polytetrafluoroethylene grafts were divided
into two groups according to the prescription of
clopidogrel. Group A consisted of eleven patients
(58%) who did not receive anti-platelet medication
after surgery, while Group B consisted of eight pa-
tients (42%) who were started on clopidogrel two
days after graft placement. Graft patency was moni-
tored until thrombosis was diagnosed and patient
survival from graft surgery until the writing of this
manuscript was determined.
Nineteen chronic hemodialysis patients with syn-
thetic grafts were included. Exclusion criteria were
malignancy, end-stage chronic heart disease, active
liver or thyroid disease, uncontrolled diabetes mel-
litus, moderate or severe malnourishment, hemat-
ocrit <32% or >36%, or acute infections. The pa-
tients were divided into two groups according to
the use of clopidogrel. Group A consisted of eleven
patients (58%) who did not receive clopidogrel,
while group B comprised eight patients (42%) who
were prescribed 75 mg of clopidogrel/ day; both
groups were well matched with respect to age, gen-
der, time on hemodialysis from the initiation of re-
placement therapy to graft placement, cause of re-
nal failure, baseline hematocrit, baseline homocys-
teine, and Kt/ V (Tab. I). High-flux hemodialysis
was performed with a bicarbonate bath, mean Qd
500 ml/ minute and mean Qb 350±50 ml/ minute.
Biocompatible membranes were used: polysul-
phone F80®(Fresenius Germany) and cellulose tri-
acetate FB210® (Nipro, Japan). Each dialysis ses-
sion averaged 3.5±0.5 hours thrice weekly.
Group A: Eight patients had radiocephalic grafts
(forearm) and three had brachiocephalic grafts (up-
per arm). Group B: Five patients had radiocephalic
and three had brachiocephalic grafts. In accordance
with the routine protocol of our center, if venous
outflow occlusion was diagnosed the patient would
be immediately referred to the vascular surgery ser-
vice for prompt resolution of the stenosis.
Only in cases where thrombotic events were report-
ed did we investigate other causes of thrombophil-
ia. This consisted of screening for factor V Leiden
mutations, prothrombin G20210A variants, and
protein C, protein S or antithrombin III deficiency.
Common medications prescribed
All patients received alpha-erythropoietin (2000-4000
U subcutaneously three times a week) post-dialysis
and intravenous iron saccharate to reach a transfer-
rin saturation between 20% and 50%. Three thou-
sand units of standard heparin were routinely given
to all patients as an infusion by means of an automat-
ic pump during the whole length of each dialysis ses-
sion. All patients received folic acid (10 mg/day oral-
ly) and IV complex B vitamins (postdialysis).
Results are expressed as mean ± standard deviation
of the mean (SD), unless specified otherwise. The
Mann-Whitney U test was used for differences in
quantitative variables between groups. The chi-
square or Fisher test was used for qualitative vari-
Clopidogrel and graft thrombosis
TABLE I - PATIENT CHARACTERISTICS
(36.4) 64.0±13.833.2±0.8 11.2±2.41.1±0.2 (18.2) (18.2)(27.3) (36.4)
(62.5) 77.0±15.132.9±0.7 11.9±2.51.2±0.3(25) (12.5)(12.5)(50)
HD, hemodialysis; Hct, hematocrit; Hcy, homocysteine; DM, diabetes mellitus; GN, glomerulonephritis; SLE, systemic lupus erythematosus; ISCH
NEP, ischemic nephropathy. Differences are not statistically different
The two patient groups were well matched with re-
spect to age, gender, cause of renal failure, time on
hemodialysis to graft placement, time to initiation
of graft use, baseline hematocrit, platelet count, ho-
mocysteine level and Kt/ V (Tab. I). Hemodynamic
instability during dialysis sessions was similar
among patients of both groups. No cases of factor V
Leiden mutations, prothrombin G20210A variants,
or protein C, protein S or antithrombin III defi-
ciency were found in these patients.
All grafts thrombosed in this group. Arteriogra-
phies confirmed the presence of venous outflow ob-
struction. No episode of graft thrombosis could be
resolved surgically and catheters were transiently
inserted. All patients were monitored subsequent to
the event to record the complete duration of their
dialysis treatment. All patients in Group A died.
The time elapsed from graft thrombosis to death
was 68.1±15 days. In none of the patients was graft
thrombosis attributable to acute predisposing fac-
tors (infection, cancer, chronic hypotension).
No thromboses were recorded in this group. Two
patients (25%) died with their grafts functioning.
The cause of death was advanced cardiac insuffi-
ciency in both cases.
Ten thrombotic events were diagnosed in Group A
while no thromboses occurred in Group B (10 vs1,
p<0.001) (Tab. II). The time elapsed from first dial-
ysis to graft placement was similar in both groups:
Group A, 18 ± 12 days vsGroup B, 20±10 days. Also
the time from surgery to graft use was similar:
Group A, 38 ± 12 days vs. Group B, 37±15 days.
Graft patency was significantly longer in Group B
compared to Group A: 350.8±166 vs. 86.8±69 days,
p<0.001. Days in hemodialysis from first session to
death in Group A or to the end of this study in
Group B were statistically different: Group A:
195.9±96 vs Group B: 545.5±291, p<0.001. Patient
survival was statistically longer in Group B (89.5%
vs0%, p = 0.001).
Side effects of clopidogrel
At the prescribed dose of 75 mg daily, no major side
effects occurred. Although no bleeding time tests
or standard coagulation profiles were asked, none
of the patients reported clinically evident major
bleeding episodes in the gastrointestinal tract, fe-
male genital tract, airways or skull. Moreover, skin
hematomas and post-dialysis graft bleeding were
not different between patients of either group.
The present study shows that clopidogrel at a dose
of 75 mg daily significantly reduced thrombotic
events in polytetrafluoroethylene hemodialysis
grafts, thereby increasing the useful life of this type
of vascular access. Moreover, duration of hemodial-
ysis treatment and patient survival were longer in
this population. Finally, clopidogrel was well toler-
ated by all patients and no major side effects were
Several aspects need to be mentioned about the
present study. The thrombotic episodes in Group A
could not be due to early vascular access, to acute
or clinical conditions that could predispose to graft
occlusion, or to patient characteristics, because in
Trimarchi et al
TABLE II - FINAL RESULTS
GroupTime from HD
Time from graft
Time on HD
Group A 18.0±1238.0±12 86.8±69195.9±96010 11 (100%)0
Group B 20.0±1037.0±15 350.8±166545.5±2910 2 (10.5%)89.5
p Ns Ns<0.001 <0.001<0.001 = 0.001 = 0.001
For personal use only
Group B these conditions were highly similar. Al-
though several reports assure that grafts can be
used almost immediately after surgery, we start us-
ing them only after the soft tissue edema has dis-
appeared and the prosthesis can be clearly felt by
the physician and dialysis technician. Doppler
sonograms are usually performed before applying
the grafts to confirm the clinical impression.
As to the longer survival of Group B (Tab. II), this
could be due to a lower rate of thrombotic com-
plications, which, if not surgically corrected,
would be followed by hospitalizations, catheter in-
sertions, catheter-associated complications and
lower hemodialysis efficacy rates.
Cardiovascular disease is the main cause of death
in the end-stage renal disease population (9, 10).
Goldsmith et al reported an angiographically con-
firmed prevalence of significant coronary stenosis
that varied from 24% in a young non-diabetic he-
modialysis population to 85% in long-standing di-
abetic dialysis patients over 45 years of age (11).
Moreover, most hemodialysis patients with angio-
graphically significant coronary artery disease are
symptomless (12). In this regard, the CAPRIE
study has shown that clopidogrel is effective in re-
ducing ischemic events in patients with myocar-
dial infarction in the general population (13).
Thus, another reason why clopidogrel may pro-
long patient survival in uremic patients could be
its cardioprotective effect. It should be noted that
the two patients of Group B who died had their
grafts functioning by the time of death.
Uremic patients have an acquired functional
platelet defect favoring bleeding. However, in-
creased platelet adhesiveness in atherosclerotic
uremic patients has also been reported (11). Al-
though coagulation studies were not performed
in the present study, patients on clopidogrel did
not show evidence of major bleeding events, and
hematocrit falls or clinical overt hemorrhagic
episodes were not reported. Moreover, minor
post-dialysis bleeding at the puncture site of the
graft was similar in both groups. Data on the use
of clopidogrel alone in this population are lack-
ing, but a recent study suggests that the dose of
clopidogrel need not be adjusted in patients with
moderate to severe renal failure, although it pro-
longs the bleeding time almost two-fold (14, 15).
Moreover, a pharmacodynamic study of clopido-
grel in chronic hemodialysis patients indicated
that the inhibition of platelet aggregation is safe
and unaltered by hemodialysis (16).
Clopidogrel is a thienopyridine derivative that
achieves its antiplatelet effect by inhibiting the
binding of adenosine diphosphate (ADP) to its
membrane receptor. Its antiplatelet aggregation
is concentration dependent and appears to be
highest after four to seven days of therapy at the
standard dose of 75 mg daily, at which time 30%
to 50% of platelet aggregation is inhibited. This
antiplatelet effect is irreversible. Clopidogrel is
extensively metabolized by the liver, and approxi-
mately 50% of the drug is excreted in the urine
The most common complication of chronic he-
modialysis vascular access is thrombosis, which,
when not corrected, accounts for approximately
85% of vascular loss (1, 5, 17). Grafts are 3.8 times
more likely to require a thrombectomy and 3.0
times more likely to need access intervention
than native fistula (18). One of the main reasons
for this increased predisposition to develop
thrombosis is their propensity for venous outflow
stenosis caused by endothelial and fibromuscular
hyperplasia. In addition, failure of permanent
vascular access in hemodialysis patients occurs in
80% of cases due to thrombosis (1, 5, 17). Finally,
only 50% of arteriovenous grafts survive longer
than three years (19). In this respect, clopidogrel
does not solve venous outflow stenosis, but it may
diminish platelet adhesion to turbulent partially
occluded flows, probably slowing the rate of
thrombosis and vessel occlusion.
Strategies to prevent thrombosis other than treat-
ment of venous outflow stenosis in hemodialysis
patients are scant and include dipyridamole and
low-dose aspirin and aspirin plus clopidogrel (6-
8). The first combination did not appear to be ef-
fective in patients with previous thrombosis, al-
though the thrombosis rate in patients with new
grafts was substantially reduced with dipyri-
damole (6). A recent study assessed the effective-
ness of aspirin plus clopidogrel in preventing
graft thrombosis. This approach would affect
platelet activation in two ways: cyclo-oxygenase ac-
tivation and thromboxane generation by aspirin
plus platelet ADP binding by clopidogrel. Howev-
er, the study was interrupted because of a marked-
ly increased risk of bleeding among those under
treatment (8). Our results show that clopidogrel
alone can be used safely to prolong graft survival
in chronic hemodialysis patients and may also
contribute to cardioprotection.
In our study, clopidogrel significantly decreased
thrombotic graft episodes. Patients on clopido-
grel had prolonged vascular access patency, a
longer time on hemodialysis and better survival.
We are aware that the study involved a very small
number of patients, but in view of the promising
results we believe that our preliminary findings
Clopidogrel and graft thrombosis
For personal use only
deserved prompt communication. However, the Download full-text
data must be interpreted with caution because
the pharmacological approach to prevent vascu-
lar access thrombosis in hemodialysis is still in
its infancy. A prospective randomized controlled
trial including a larger number of patients is
For personal use only
Address for correspondence:
Dr. Hernán Trimarchi
1280 Buenos Aires - Argentina
Trimarchi et al
1. Windus DW. Permanent vascular access: A nephrol-
ogist’s view. Am J Kidney Dis 1993; 21: 457-71.
2. Feldman HI, Kobrin S, Wasserstein A. Haemodialysis
vascular access morbidity. J Am Soc Nephrol 1996; 7:
3. Feldman HI, Held PJ, Hutchinson JT, et al. He-
modialysis vascular access morbidity in the United
States. Kidney Int 1993; 43: 1091-6.
4. Schwab SJ. Vascular access for hemodialysis. Kidney
Int 1999; 55: 2078-90.
5. Fan PY, Schwab SJ. Vascular access. Concepts for the
1990s. J Am Soc Nephrol 1992; 3: 1-11.
6. Domoto DT, Bauman JE, Joist JH. Combined aspirin
and sulpynpirazone in the prevention of recurrent
hemodialysis vascular access thrombosis. Throm Res
1991; 62: 737-43.
7. Sreedhara R, Himmelfarb J, Lazarus M, Hakim R.
Anti-platelet therapy in graft thrombosis: Results of a
prospective, randomized, double-blind study. Kidney
Int 1994; 45: 1477-83.
8. Kaufman JS, O’Connor TZ, Zhang JH, Cronin RE.
Randomized controlled trial of clopidogrel plus as-
pirin to prevent hemodialysis access graft thrombo-
sis. J Am Soc Nephrol 2003; 14: 2313-21.
9. Best JM, Lennon R, Ting HH, et al. The impact of re-
nal insufficiency on clinical outcomes in patients un-
dergoing percutaneous coronary interventions. Am
J Cardiol 2002; 39: 1113-9.
10. US Renal Data System: USRDS 1999 Annual Data Re-
port. Bethesda, MD, The National Institutes of
Health, National Institute of Diabetes and Digestive
and Kidney Diseases, 1999.
11. Goldsmith DJA, Covic A. Coronary artery disease in
uremia: Etiology, diagnosis, and therapy. Kidney Int
2001; 60: 2059-78.
12. Braun WE, Phillips DF, Vidt DG. Coronary artery dis-
ease in 100 diabetics with end-stage renal failure.
Transplant Proc 1984; 16: 603-7.
13. CAPRIE Steering Committee. A randomized, blind-
ed, trial of clopidogrel versus aspirin in patients at
risk of ischaemic events (CAPRIE). Lancet 1996; 348:
14. Deray G, Bagnis C, Brouard R, et al. Clopidogrel ac-
tivities in patients with renal function impairment.
Clin Drug Invest 1998; 4: 319-28.
15. Boneau B, Destelle G, on behalf of the Study Group.
Platelet anti-aggregating activity and tolerance of
clopidogrel in atherosclerotic patients. Thromb
Haemost 1996; 76: 939-43.
16. Kaufman JS, Fiore L, Hasbargen JA, O’Connor T,
Perdriset G. A pharmacodynamic study of clopido-
grel in chronic hemodialysis patients. J Thromb
Thrombolysis 2000; 10: 127-31.
17. Galbraith S, Fan P, Collins D, et al. Hemodialysis fis-
tula thromboses: A prospective evaluation of
anatomic versus non-anatomic causes (abstract). J
Am Soc Nephrol 1992; 3: 365.
18. Young EW, Dykstra DM, Goodkin DA, et al. He-
modialysis vascular access preferences and outcomes
in the Dialysis Outcomes and Practice Patterns Study
(DOPPS). Kidney Int 2002; 61: 2266-71.
19. Munda R, First MR, Alexander JW, et al. Polytetra-
fluoroethylene graft survival in hemodialysis. JAMA
1983; 249: 219-22.