Original Paper
Accepted: June 13, 2001
Nephron 2002;91:399–405
Risk Factor Analysis for Long-Term
Tunneled Dialysis Catheter-Related
Bacteremias
G. Jean B. Charra C. Chazot T. Vanel J.C. Terrat J.M. Hurot G. Laurent
Centre du Rein Artificiel de Tassin, Tassin la Demi-Lune, France
Key Words
Dialysis catheter-related bacteremia W Tunnel infection W
Staphylococcus aureus W Nasal carriers W Intravenous
iron therapy
Abstract
Infection, mainly related to vascular access, is one of the
main causes of morbidity and a preventable cause of
death in hemodialysis patients. From January 1994 to
April 1998 we conducted a prospective study to assess
the incidence and risk factors of catheter-related bacter-
emia. One hundred and twenty-nine tunneled dual-
lumen hemodialysis catheters were inserted percuta-
neously into the internal jugular vein in 89 patients. Bac-
teremia (n = 56) occurred at least once with 37 (29%) of
the catheters (an incidence of 1.1/1,000 catheter-days);
local infection (n = 45, 1/1,000 catheter-days) was associ-
ated with bacteremia in 18 cases. Death in 1 case was
directly related to Staphylococcus aureus (SA) septic
shock, and septicemia contributed to deaths in 2 addi-
tional cases. Catheters were removed in 48% of the bac-
teremic episodes. Treatment comprised intravenous
double antimicrobial therapy for 15–20 days. Bacterio-
logical data of bacteremia showed 55% involvement of
SA. Nasal carriage of SA was observed in 35% of the
patients with catheters. Bacteremic catheters were more
frequently observed in patients with diabetes mellitus
© 2002 S. Karger AG, Basel
ABC 0028–2766/02/0913–0399$18.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/journals/nef
(p = 0.03), peripheral atherosclerosis (p = 0.001), a pre-
vious history of bacteremia (p = 0.05), nasal carriage of
SA (p = 0.0001), longer catheter survival time (p = 0.001),
higher total intravenous iron dose (p = 0.001), more fre-
quent urokinase catheter infusion (p ! 0.01), and local
infection (p ! 0.001) compared with non-bacteremic cath-
eters. Monovariate survival analysis showed that signifi-
cant initial risk factors for bacteremia were nasal carriage
of SA (p = 0.00001), previous bacteremia (p = 0.0001),
peripheral atherosclerosis (p = 0.005), and diabetes (p =
0.04). This study confirms the relatively high incidence of
bacteremia with tunneled double-lumen silicone cathe-
ters and its potential complications. Possible preventive
actions are discussed according to the risk factors.
Copyright © 2002 S. Karger AG, Basel
Introduction
Bacterial infection is one of the major causes of mortal-
ity and morbidity among dialysis patients. Due to arterio-
venous (A-V) fistula failure and cardiovascular diseases, a
tunneled dual-lumen central venous silicone catheter is a
necessary alternative for long-term access in an increasing
number of patients, comprising about 18% of the total
dialysis population in our center. The main complications
of these devices are local or general infection, clotting and
central vein thrombosis or stenosis [1]. Hemodialysis
Dr. Guillaume Jean
Centre du Rein Artificiel de Tassin
Siriraj Medical Library, Mahidol University
41, avenue du 8 Mai 1945
198.143.39.97 - 3/1/2016 4:13:56 AM
F–69160 Tassin la Demi-Lune (France)
Tel. +33 472626124, Fax +33 478345940, E-Mail Gjean14357@aol.com
Downloaded by:
catheters are known to be one of the major risk factors for
bacteremia, especially when compared to synthetic or
native A-V fistulas [2]. The reported percentage of cases
complicated by bacteremia due to catheters varies be-
tween 1 and 12% [3, 4] with a mortality rate ranging from
0 to 18% [5–7]. Nasal carriage of Staphylococcus aureus
(SA) is the main risk factor reported for catheter-related
bacteremia [8–12]. We conducted a prospective study in
hemodialysis patients to assess the incidence and risk fac-
tors of catheter-related bacteremia with long-term tun-
neled double-lumen silicone hemodialysis catheters.
Subjects and Methods
From January 1994 to April 1998, 129 tunneled dialysis catheters
were inserted in 89 patients. All subjects were prospectively followed
in our institution. The minimum observation period was 6 months.
Most catheters were inserted for prolonged vascular access mostly
when an A-V fistula had failed or was contraindicated.
Fifty-five percent of the patients were male, and the mean age of
the sample population was 64.5 B 13 (27–90) years. All patients were
on hemodialysis for a mean of 63.2 B 89 (1–320) months. Dialysis
was performed for 8 h three time each week in 75% of patients and
for 3 ! 5 h in 25%. All dialysis membranes were cellulose. Daugir-
das-2 Kt/V was 1.9 B 0.4, nPCR 1.15 B 0.3. Nephropathies were
nephrosclerosis in 22%, glomerulonephritis in 15%, interstitial ne-
phropathy in 13%, diabetes mellitus in 22%, polycystic kidney dis-
ease in 13%, kidney cancer in 3%, and undetermined in 12% of the
study group.
Sixty-four patients had 1 catheter, 16 patients had 2, 7 had 3, and
3 had 4 catheters during the study period. None of the patients were
treated with immunosuppressive therapy, although 5 received oral
methylprednisolone (4 mg daily) for chronic amyloidosis-related
joint pain. None had AIDS. Eighty-five percent of the patients had
received at least one intravenous iron therapy during the study peri-
od (i.e. intravenous iron polymaltose, 30–100 mg weekly, using per-
dialytic continuous infusion sessions for 10 weeks).
Catheters
One hundred and twenty-nine catheters were inserted into the
internal jugular vein, 64% of which were on the right side. Sixty-six
catheters were tunneled double-lumen cuffed silicone catheters, of
which 62 were Permacath and 4 were Mahurkar (Quinton® Instru-
ment Co., Seattle, Wash., USA). Sixty-three were tunneled non-cuf-
fed double TwinCath (Medcomp®, Harleysville, Pa., USA) catheters.
All were inserted percutaneously by a senior nephrologist using the
Seldinger technique and a J guide wire in an operating room under
local anesthesia. Chest X-ray was systematically performed to verify
placement of the catheter tip in the right atrium when fluoroscopy
was not available. No antimicrobial prophylaxis was given. A hepa-
rin lock of 5,000 U/ml was instilled in each catheter port to fill the 2
lumens after each hemodialysis session. The exit site was swabbed
with povidone iodine, rinsed after 3 min with sterile water and cov-
ered with gauze and a semi-occlusive dressing. The dressing was
changed before each dialysis session.
400 Nephron 2002;91:399–405
Bacteriological Samples
All patients were checked for the presence of nasal carriage of SA
starting at the first dialysis session with 2 successive cultures at 2-
week intervals, and thereafter 1 culture every 6 months. Patients
were considered as permanently colonized by SA if at least 2 nasal
cultures were positive before the first bacteremia episode as defined
previously [13]. Nasal cultures were not done systematically at the
time of bacteremia for comparison with the micro-organism respon-
sible for bacteremia. Patients were not treated with local mupirocin
or other topical antibacterial agents nor with long-term general anti-
biotic therapy during the study period.
All tunnel infections and exit-site infections were checked with
local cultures. Local infection was defined as a positive exit-site cul-
ture with local signs of inflammation or suppuration. Bacteremia was
defined as at least 2 positive blood cultures for coagulase-negative
Staphylococcus, Corynebacteria or Bacillus, or one positive blood
culture for other micro-organisms whatever the associated symptoms
(fever above 38 ° C, chills). Blood cultures were obtained from the
arterial dialysis line during the session or from peripheral veins or
dialysis catheters outside the dialysis sessions. Bacteremia from an
obvious source other than the catheter was excluded. Quantitative or
semi-quantitative bacterial analysis of blood cultures or catheter tip
cultures was not available. Recurrent bacteremia was defined as 2 or
more bacteremic episodes from the same catheter, or in the same
patient with a new catheter, regardless of the delay and/or the micro-
organism involved. Relapse of bacteremia was defined as recurrent
bacteremia within 90 days after the first episode due to the same
micro-organism.
Septicemia was defined as bacteremia with severe and prolonged
symptoms.
Treatment
Mild local infection was treated with oral antibiotics and local
irrigation with rifamycin. Catheter-related bacteremia was always
treated with double parenteral antibiotic therapy for 15–20 days. The
catheter was removed and changed to the other jugular side if apyrex-
ia was not obtained after 48 h of treatment or sooner in cases with
severe symptoms. The infected catheters were not removed from
patients in whom maintenance of the vascular access was critical and
when symptoms of infections rapidly improved (! 48 h). In cases
with a clotted catheter, urokinase (10,000 U/ml) was instilled into the
lumen to clear it.
Study Design
At initiation of the study, the mean values of the last 2 months
were collected for the following baseline analyses: serum albumin,
serum ferritin, C-reactive protein, and hemoglobin. The catheter-
related problems observed by the nursing staff or the patient were
recorded for all catheters on a special dialysis catheter chart after
each session. All treatments including antibiotic therapy and intrave-
nous iron doses were recorded. Cardiovascular antecedents and vas-
cular access history were recorded as well as antecedents of catheter-
or A-V fistula-related bacteremias. A subjective semi-quantitative
hygienic score (0–5) was used for each patient, a score of 5 being the
best hygienic state.
A monovariate analysis was performed comparing bacteremic
and non-bacteremic catheters and patients with and without nasal
carriage of SA.
We compared the bacteremia-free survival curve of catheters,
bacteremia being the censured event, according to potential risk fac-
Jean/Charra/Chazot/Vanel/Terrat/Hurot/
Laurent
Siriraj Medical Library, Mahidol University
198.143.39.97 - 3/1/2016 4:13:56 AM
Downloaded by:
tors. In the catheter survival study, the overall catheter survival rate
was defined based on survival that took into account all causes of
failure including patient-related causes (i.e. death, transplantation).
The technical survival rate was defined as catheter survival taking
into account only the catheter-related causes of failure (i.e. infection,
clotting, extrusion, thrombosis).
Statistical Analysis
The K² test, Mann-Whitney and Wilcoxon nonparametric U test
and Kaplan-Meier analysis with log rank test were applied when
appropriate. A significant result was considered for p ! 0.05. (Solo
6.0.4, BMDP®).
Results
Catheter Survival
Major complications did not occur during the insertion
of the catheters except for 3 moderate cervical hematomas
and 6 tunnel bleedings. The total time with functional
catheters was 52,030 catheter-days. The overall catheter
survival rate was 43% at 1 year, 30% at 2 years, and 10%
at 3 years. The 1-year technical survival rate was 65%. We
observed no difference between right and left jugular out-
comes.
One hundred catheters had stopped to be functional
for different reasons: death in 34%, infection in 21%, dys-
function in 19%, functional A-V fistulas in 17%, renal
transplantation in 5%, and accidental total extrusion in
4% of the cases.
Bacteremia Rate
Episodes of bacteremia (n = 56) were observed at least
once in 37 catheters (29%) with an incidence of 1.1/1,000
catheter-days. Twenty-five catheters had 1 episode of bac-
teremia, 6 had 2, 5 had 3, and 1 had 4 episodes. Local
infection (n = 45, 25%, incidence 1/1,000 catheter-days)
was present in 18/56 (32%) of the bacteremia episodes.
No significant differences were seen between the cuffed
and non-cuffed catheters according on the local or general
infection rate.
Death occurred in 1 case after catheter-related septic
shock due to methicillin-resistant SA. In 2 other cases,
catheter-related septicemias contributed to the fatal out-
come in multideficient diabetic patients. Prolonged bac-
teremia, exceeding 72 h, occurred in 5 cases including the
3 deaths. In 3 cases, secondary sepsis occurred which
responded to antibiotic therapy: 1 bacterial spondylitis
and 2 pneumonias. Endocarditis was not detected but
endoesophageal echocardiography was not done system-
atically. Catheters were changed or removed within 72 h
in the 5 cases with septicemia. When an episode of bacter-
Dialysis Catheter Bacteremia
emia occurred, the catheter was removed in 27/56 cases
(48%). Catheters were not changed over a guide wire. Of
the 13 catheters changed for recurrent infection, 8 were
complicated by recurrent bacteremia within 3 months
due to the new device.
The median time following catheter placement before
the first bacteremia was 94 (5–371) days for SA, 209 (1–
770) days for coagulase-negative Staphylococcus and 165
(35–685) days for gram-negative rods.
Bacteriological cultures showed 56 bacteremias: 31 SA
(55%, including 4 methicillin-resistant); 11 coagulase-
negative Staphylococcus (17%; including 4 methicillin-
resistant); 1 gram-positive Bacillus; 6 Serratia marcesens;
1 Beta-streptococcus; 2 Escherichia coli, and 4 Pseudo-
monas aeruginosa.
Local infection (n = 45, 1/1,000 catheter-days) showed
SA in most cases (29/45, 65%). A large majority of these
were observed in nasal carriers of SA (75%).
Bacteremia was probably related to a local infection in
18 cases since the same micro-organism was detected in
the local and blood culture at the same time. SA was
found in 11/18 cases. In cases of recurrent bacteremia, the
micro-organism was different from the previous one in
50% of cases.
Treatments
Treatment always comprised an initial intravenous dou-
ble antibiotic therapy. After blood culture, C-reactive pro-
tein measurement and blood cell count, immediate antimi-
crobial therapy was given. Initially this was on an empirical
basis or based on bacteriological history, and after the cul-
ture results were available, based on susceptibilities. The
protocol of first-line antibiotic therapy included: in case of
SA, a first-generation cephalosporin (cefazolin) with ami-
noglycoside (netilmicin); in cases of methicillin-resistant
Staphylococcus, vancomycin was used in combination with
initial netilmicin in 3 doses, and for gram-negative rods we
used a third-generation cephalosporin with an aminoglyco-
side or a systemic fluoroquinolone.
Risk Factors
Table 1 compares the 2 catheter groups. The first group
was free of bacteremia while the second was associated
with at least one bacteremia. The significant differences
between catheters with and without bacteremia included
male sex, younger age diabetes, more frequent nasal car-
riage of SA, a previous history of bacteremia, lower serum
ferritin levels, higher average urokinase infusion per cath-
eter, more frequent local infection, longer catheter use,
and higher total intravenous iron.
Nephron 2002;91:399–405 401
Siriraj Medical Library, Mahidol University
198.143.39.97 - 3/1/2016 4:13:56 AM
Downloaded by:
Table 1. Catheter- and patient-related characteristics recorded at Table 3. Characteristics of permanent nasal carriers of SA and non-
initiation and at the end of study, according to the occurrence of bac- carriers at study initiation and at study end
teremia (none vs. one or more; mean B SD)
No nasal SA Nasal SA
No bacteremia Bacteremia
Total n = 89 patients 58 (65%) 31 (35%)
Total = 129 catheters n = 92 (71%) n = 37 (29%)
Start of study
Start of study Male sex 29 (55%) 21 (68%)
Male sex 46 (50%) 25 (69%)* Age, years 68B12 58.6B15**
Age, years 64.6B13 59.8B13* Months on dialysis 62B92 59B15
Time on dialysis, months 58.9B88 62.7B76 Diabetes mellitus 12 (20 %) 8 (26%)
Nephrosclerosis 21 (23%) 4 (10%) Nephrosclerosis 12 (20%) 7 (22.5%)
Coronary disease 11 (12%) 10 (25.6%) Coronary disease 7 (12%) 8 (26%)
Peripheral atherosclerosis 12 (13%) 15 (38.4%)** Peripheral atherosclerosis 9 (15.5%) 12 (39%)*
Diabetes mellitus 15 (16.3%) 12 (33%)* History of bacteremia 23 (40%) 20 (65%)*
Nasal SA 27 (29%) 30 (82%)*** Right Jugular side 32 (65%) 17 (44%)
Hygienic score (/5) 3.16B0.9 2.5B1.2** Body mass index 24.8B5 23.9B5
Previous bacteremia 17 (18.4%) 18 (53%)*** Hygiene score /5 2.8B0.9 2.5B1.2
Cuffed catheter 40 (43.4%) 19 (54%) Serum ferritin, Ìg/l 263B294 225B284
Right jugular side catheter 57 (62%) 28 (70%) Serum albumin, g/l 35.6B6 37.1B5
Body mass index 24.5B5.2 24.6B5.4 Hb, g/l 86B20 86.7B16
Serum ferritin, Ìg/l 292B322 180B170* CRP, mg/l 30.4B38 27B35
Serum albumin, g/l 36.8B6 36.2B5
End of study
Hb, g/l 88.3B18 87.4B19
Catheter survival, days 482B350 720B550*
CRP, mg/l 45.5B63 53.9B62
Total i.v. iron, mg 1,850B1,500 2,229B1,600
End of study Local infection 7 (12%) 20 (65%)***
Urokinase infusion/catheter 1B2.3 3.5B5*** Number of bacteremia 0.2B0.6 1.6B1.6***
Local infection 14 (15.2%) 19 (51%)** Bacteremic patient 7 (12%) 25 (80%)***
Partial catheter extrusion 29 (31.5%) 17 (46%)
Catheter survival, days 296B322 648B542*** * p ! 0.05; ** p ! 0.001; *** p ! 0.0001.
Total i.v. iron, mg 1,116B949 2,012B1,900***

* p ! 0.05; ** p ! 0.001; ***p ! 0.0001.

Table 2. Log rank test of catheter-related bacteremia survival Table 2 shows the results of catheter-related survival
analysis analysis, with bacteremia being considered as failed sur-
vival, according to some initial potential risk factors. Sig-
Factors log rank
nificant factors were peripheral atherosclerosis, diabetes
Sex 0.3 mellitus, previous bacteremia and, above all, nasal car-
Nephrosclerosis 0.3 riage of SA.
Peripheral atherosclerosis 0.005
Coronary disease 0.07 Nasal Carriage of SA
Diabetes mellitus 0.04
Nasal carriage of SA was found in 35% of the patients
Previous bacteremia 0.0001
Nasal SA 0.00001 (n = 31). Table 3 depicts the characteristics of nasal car-
Catheter type 0.3 riers of SA. This condition was more frequently observed
Jugular side 0.2 in younger patients with a previous history of bacteremia
Serum albumin ! 35 0.1 and with more frequent peripheral atherosclerosis. Eighty
CRP 1 40 0.21
percent of nasal carriers of SA had at least 1 bacteremia
Hemoglobin ! 80 g/l 0.8
Serum ferritin 1 500 ng/l 0.2 vs. 12% of non-carriers and for a longer exposure time. In
Dialysis time 1 2 years 0.4 nasal carriers of SA, the incidence of bacteremia per
patient was higher than that for non-carriers and local

402 Nephron 2002;91:399–405 Jean/Charra/Chazot/Vanel/Terrat/Hurot/

Laurent
Siriraj Medical Library, Mahidol University
198.143.39.97 - 3/1/2016 4:13:56 AM
Downloaded by:
infections were more frequent. Nasal SA was found in
90% of SA bacteremias, in 60% of coagulase-negative sta-
phylococcus and in 100% of gram-negative bacteremias.
Figure 1 displays a survival curve showing the risk of
bacteremia on the nasal carriage of SA. The median cathe-
ter survival before bacteremia occurred was 128.5 (1–
760) days for patients with positive nasal carriage of SA
vs. 209.5 (0–1,425) days for those free of nasal SA (p !
0.00001).
Discussion
Hemodialysis catheters are known to be a major risk
factor for bacteremia particularly when compared to syn-
thetic or native A-V fistulas [2, 14]. The wider use of tun-
neled dialysis catheters for long periods increases the risk
of bacteremia. Our results show that the risk of infection
increases with catheter survival time. The catheter surviv-
al rate was similar to that reported in the literature for
different catheter types, with results varying from 43 to
80% [4, 15–17].
The reported incidence of bacteremia varies greatly
according to the literature. It depends on the catheter
type, their survival rate, the cuff and tunnel used, the cen-
tral vein used and its side, and the percentage of diabetics
in the study. Most often the incidence is given as a per-
centage of catheters complicated with bacteremia. It var-
ies from 1 to 49% [3, 18–23]. Elsewhere it is more ade-
quately related to the exposure risk and expressed as 0.5–
5.5/1,000 catheter-days [4, 5, 21, 24]. The infection rate
reported here seems high when expressed as 29% of the
catheters with one or more related bacteremia, although
this frequency may not be relevant. When reported as
exposure time, given here as 1.1 bacteremia/1,000 cathe-
ter-days, it becomes comparable to the lower bacteremia
rate reported by others. Local infection was reported to
occur in 16–21% of catheters [4, 19], figures that are simi-
lar to our 25% incidences. Expressed as catheter-days, our
value of 1/1,000 is close to the 1.25/1,000 catheter-days
reported in a recent study [5]. The carriage of nasal SA is
an important risk factor for local SA infection. In 32% of
cases, local infection was probably the origin of the bacter-
emia. Tunnel and cuff catheters do not appear to be an
efficient bacteriological barrier. This led us to review our
exit-site care protocol and to treat the carriers of nasal SA
with local mupirocin.
In a recent prospective study it was suggested that
infected catheters should be changed over a guide wire in
cases of moderate bacteremia-associated symptoms [25].
Dialysis Catheter Bacteremia
Fig. 1. Survival curve of catheters with a first bacteremia in carriers
and non-carriers of nasal SA (log rank ! 0.00001).
Elsewhere, it was reported that antibiotic therapy, com-
bined with catheter exchange, produced a better success
rate than antibiotic therapy alone [5]. Conservative man-
agement of catheter-related sepsis, with the catheter re-
maining in place for at least the first 48 h of the antibiotic
treatment, has been reported as in our study [26] This
conservative attitude is questionable because, theoretical-
ly, it increases the risk of a bacteremia relapse. However,
the risk of a new catheter being colonized with the same
micro-organism present on the patient’s skin is high,
therefore increasing the risk of relapse especially when
changing a catheter over a guide wire or using the same
tunnel. In our experience, 61% of the patients whose cath-
eters were changed due to recurrent bacteremia displayed
another bacteremia within 4 months. Some patients with
long-term catheters had central venous stenosis, thrombo-
sis or central venous wall stent, and changing the catheters
could prove difficult or impossible. In addition, changing
the catheter side increases the risk of central venous steno-
sis. Moreover, some morbidity due to catheter changes is
always possible. The management of each case must be
considered by the nephrologist based on the patient’s
medical conditions.
Studies have indicated that the prevalence of a causa-
tive micro-organism is largely variable, but, as in our
experience, SA is the most frequently reported bacteria
causing infections [1, 18, 27, 28]. In a recent study, Ditt-
mer et al. [29] reported that 68% of catheters became
colonized after a mean time of 27 (5–115) days and 35%
Nephron 2002;91:399–405 403
Siriraj Medical Library, Mahidol University
198.143.39.97 - 3/1/2016 4:13:56 AM
Downloaded by:
developed bacteremia with the same organism. Regular
intralumen culture is advisable. From our results, as for
most data reported in the literature, nasal carriage of SA is
the main risk factor for catheter-related SA bacteremia
[8–10, 12]. As Levin et al. [9] observed, we found the same
main risk factors for both SA and coagulase-negative Sta-
phylococcus infections. Being a SA carrier may reflect
favorable conditions for skin carriage of other micro-
organisms. Nasal SA was frequent in our patients with
catheters (35%). Compared to some other studies, the fre-
quency displays significant variability depending on local
conditions, ranging from 20 to 58% [30].
Our studies indicate that a history of previous bacter-
emia is a risk factor [2]. This seems to be related to the
nasal carriage of SA, which is also a major risk factor for
repeated bacteremia [9].
Nasal mupirocin has been reported to significantly
decrease the hemodialysis SA-related infection rate [8, 31,
32] and nasal SA elimination is frequently observed in
more than 90% of treated patients [10, 30, 31]. However,
long-term results in hemodialysis patients are not avail-
able and the risk of SA resistance is unknown.
Diabetes mellitus has been shown to be a risk factor for
catheter-related sepsis [11]. We also observed that the risk
of bacteremia is increased in diabetic patients, with the
sensitivity of diabetics to infection being well known.
Peripheral atherosclerosis, another significant risk factor,
is mainly observed in case of diabetes.
A high serum ferritin level is also known to be a signifi-
cant risk factor for infection [2, 33, 34]. When serum ferri-
tin is above 1,000 Ìg/l, rHuEPO reverses polymorphonu-
clear granulocytes dysfunction in iron-overloaded hemo-
dialysis patients following transfusion [35]. Desferriox-
amine has beneficial effects on infection susceptibility in
iron-overloaded patients (serum ferritin 500–1,000 Ìg/l)
[34]. These facts are in contrast with our data where the
initial serum ferritin levels were lower in patients who
subsequently developed bacteremia. One could argue that
none of our patients had a history of blood transfusions or
had a serum ferritin of 1700 ng/ml. On the other hand,
bacteremic patients received more intravenous iron treat-
ment during the observation period, which was related to
their lower initial serum ferritin and to longer catheter
survival rates. Collins et al. [36] showed that frequent
intravenous iron dosing may be related to more frequent
infection-related deaths. The deleterious role of this treat-
ment was not evident here even if theoretical arguments
do exist. However, stopping intravenous iron therapy dur-
ing infection is advisable.
Conclusion
Patients using a long-term tunneled central venous
catheter are at high risk of developing bacteremia, mostly
due to SA. The main significant risk factors are nasal car-
riage of SA, a previous history of bacteremia, and long
catheter survival. The deleterious role of intravenous iron
treatment needs further evaluation. The infection-related
mortality and morbidity lead us to take preventive actions
with the use of such catheters. The usefulness of systemat-
ic and long-term treatment of nasal carriers of SA remains
to be studied in depth.
In the future, possible clinical solutions, apart from the
reduced use of catheters, may come from improvements
in catheter technology (for example, the development of
thrombo- and infection-resistant materials) and from effi-
cient preventive antibiotic applications.

References

1 Dupont D, Morinière P, Pourchez T, el Esper
N, Fournier A: Long-term development of Per-
macath Quinton catheters used as a vascular
access route for extra-renal detoxification (in
French). Néphrologie 1994;15:105–110.
2 Hoen B, Kessler M, Hestin D, Mayeux D: Risk
factors for bacterial infections in chronic hae-
modialysis adult patients: A multicentre pro-
spective survey. Nephrol Dial Transplant
1995;10:377–381.
3 Schwab SJ, Beathard G: The hemodialysis
catheter conundrum: Hate living with them,
but can’t live without them. Kidney Int 1999;
56:1–17.
4 Moss AH, Vasilakis C, Holley JL, Foulks CJ,
Pillai K, McDowell DE: Use of a silicone dual-
lumen catheter with a Dacron cuff as a long-
term vascular access for hemodialysis patients.
Am J Kidney Dis 1990;16:211–215.
5 Saad TF: Bacteremia associated with tunneled,
cuffed hemodialysis catheters. Am J Kidney
Dis 1999;34:1114–1124.
6 Quarles LD, Rutsky EA, Rostand SG: Staphy-
lococcus aureus bacteremia in patients on
chronic hemodialysis. Am J Kidney Dis 1985;
6:412–419.
7 Nielsen J, Kolmos HJ, Espersen F: Staphylo-
coccus aureus bacteraemia among patients un-
dergoing dialysis – Focus on dialysis catheter-
related cases. Nephrol Dial Transplant 1998;
13:139–145.
8 Bloom BS, Fendrick AM, Chernew ME, Patel
P: Clinical and economic effects of mupirocin
calcium on preventing Staphylococcus aureus
infection in hemodialysis patients: A decision
analysis. Am J Kidney Dis 1996;27:687–694.
9 Levin A, Mason AJ, Jindal KK, Fong IW,
Goldstein MB: Prevention of hemodialysis
subclavian vein catheter infections by topical
povidone-iodine. Kidney Int 1991;40:934–
938.

404 Nephron 2002;91:399–405 Jean/Charra/Chazot/Vanel/Terrat/Hurot/

Laurent
Siriraj Medical Library, Mahidol University
198.143.39.97 - 3/1/2016 4:13:56 AM
Downloaded by:
10 Perl TM, Golub JE: New approaches to reduce
Staphylococcus aureus nosocomial infection
rates: Treating S. aureus nasal carriage. Ann
Pharmacother 1998;32:S7–S16.
11 Roubicek C, Brunet P, Mallet MN, et al: Nasal
carriage of Staphylococcus aureus: Prevalence
in a hemodialysis center and effect on bacter-
emia (in French). Néphrologie 1995;16:229–
232.
12 Yu VL, Goetz A, Wagener M, et al: Staphylo-
coccus aureus nasal carriage and infection in
patients on hemodialysis. Efficacy of antibiotic
prophylaxis. N Engl J Med 1986;315:91–96.
13 Mupirocin Study Group: Nasal mupirocin pre-
vents Staphylococcus aureus exit-site infection
during peritoneal dialysis. J Am Soc Nephrol
1996;7:2403–2408.
14 Kessler M, Hoen B, Mayeux D, Hestin D, Fon-
tenaille C: Bacteremia in patients on chronic
hemodialysis. A multicenter prospective sur-
vey. Nephron 1993;64:95–100.
15 McLaughlin K, Jones B, Mactier R, Porteus C:
Long-term vascular access for hemodialysis us-
ing silicon dual-lumen catheters with guidewire
replacement of catheters for technique salvage.
Am J Kidney Dis 1997;29:553–559.
16 De Meester J, Vanholder R, De Roose J, Rin-
goir S: Factors and complications affecting
catheter and technique survival with perma-
nent single-lumen dialysis catheters. Nephrol
Dial Transplant 1994;9:678–683.
17 Shaffer D, Madras PN, Williams ME, D’Elia
JA, Kaldany A, Monaco AP: Use of Dacron
cuffed silicone catheters as long-term hemodi-
alysis access. ASAIO J 1992;38:55–58.
18 Vanholder V, Hoenich N, Ringoir S: Morbidity
and mortality of central venous catheter hemo-
dialysis: A review of 10 years’ experience.
Nephron 1987;47:274–279.
Dialysis Catheter Bacteremia
19 McDowell DE, Moss AH, Vasilakis C, Bell R,
Pillai L: Percutaneously placed dual-lumen sili-
cone catheters for long-term hemodialysis. Am
Surg 1993;59:569–573.
20 Almirall J, Gonzalez J, Rello J, et al: Infection
of hemodialysis catheters: Incidence and mech-
anisms. Am J Nephrol 1989;9:454–459.
21 Dryden MS, Samson A, Ludlam HA, Wing AJ,
Phillips I: Infective complications associated
with the use of the Quinton ‘Permcath’ for
long-term central vascular access in haemodial-
ysis. J Hosp Infect 1991;19:257–262.
22 Dunea G, Domenico L, Gunnerson P, Win-
ston-Willis F: A survey of permanent double
lumen catheters in hemodialysis patients. AS-
AIO Trans 1991;37:M276–M277.
23 Cheesbrough JS, Finch RG, Burden RP: A pro-
spective study of the mechanisms of infection
associated with hemodialysis catheters. J Infect
Dis 1986;154:579–589.
24 Hirsch DJ, Bergen P, Jindal KK: Polyurethane
catheters for long-term hemodialysis access.
Artif Organs 1997;21:349–354.
25 Beathard GA : Management of bacteremia as-
sociated with tunneled-cuffed hemodialysis
catheters. J Am Soc Nephrol 1999;10:1045–
1049.
26 Capdevila JA, Segarra A, Planes AM, et al: Suc-
cessful treatment of haemodialysis catheter-
related sepsis without catheter removal. Neph-
rol Dial Transplant 1993;8:231–234.
27 Nielsen J, Ladefoged SD, Kolmos HJ: Dialysis
catheter-related septicaemia – Focus on Sta-
phylococcus aureus septicaemia. Nephrol Dial
Transplant 1998;13:2847–2852.
28 Hoen B, Paul-Dauphin A, Hestin D, Kessler
M: EPIBACDIAL: A multicenter prospective
study of risk factors for bacteremia in chronic
hemodialysis patients. J Am Soc Nephrol 1998;
9:869–876.
Nephron 2002;91:399–405
29 Dittmer ID, Sharp D, McNulty CA, Williams
AJ, Banks RA: A prospective study of central
venous hemodialysis catheter colonization and
peripheral bacteremia. Clin Nephrol 1999;51:
34–39.
30 Boelaert JR, Van Landuyt HW, Godard CA, et
al: Nasal mupirocin ointment decreases the
incidence of Staphylococcus aureus bacteraem-
ias in haemodialysis patients. Nephrol Dial
Transplant 1993;8:235–239.
31 Kluytmans JA, Manders MJ, van Bommel E,
Verbrugh H: Elimination of nasal carriage of
Staphylococcus aureus in hemodialysis pa-
tients. Infect Control Hosp Epidemiol 1996;17:
793–797.
32 Sesso R, Barbosa D, Leme IL, et al: Staphylo-
coccus aureus prophylaxis in hemodialysis pa-
tients using central venous catheter: Effect of
mupirocin ointment. J Am Soc Nephrol 1998;
9:1085–1092.
33 Boelaert JR, Daneels RF, Schurgers ML, Mat-
thys EG, Gordts BZ, Van Landuyt HW: Iron
overload in haemodialysis patients increases
the risk of bacteraemia: A prospective study.
Nephrol Dial Transplant 1990;5:130–134.
34 Tielemans CL, Lenclud CM, Wens R, Collart
FE, Dratwa M: Critical role of iron overload in
the increased susceptibility of haemodialysis
patients to bacterial infections. Beneficial ef-
fects of desferrioxamine. Nephrol Dial Trans-
plant 1989;4:883–887.
35 Boelaert JR, Cantinieaux BF, Hariga CF, Fon-
du PG: Recombinant erythropoietin reverses
polymorphonuclear granulocyte dysfunction in
iron-overloaded dialysis patients. Nephrol Dial
Transplant 1990;5:504–517.
36 Collins A, Ebben J, Ma J: Frequent i.v. dosing
is associated with higher infectious deaths (ab-
stract). J Am Soc Nephrol 1997;8:190.
405
Siriraj Medical Library, Mahidol University
198.143.39.97 - 3/1/2016 4:13:56 AM
Downloaded by: