Special Article
Complications of central venous catheters: Internal jugular versus
subclavian access—A systematic review
Sibylle Ruesch, MD; Bernhard Walder, MD; Martin R. Tramèr, MD, DPhil
C
entral venous catheters (CVCs)
are essential for the clinical
management of many patients.
Indications for a CVC are the
intravenous administration of specific
drugs (e.g., catecholamines), parenteral
nu-trition, hemodialysis, and hemodynamic
monitoring. In many institutions, patients
undergoing major surgery and patients
with critical illness or cancer routinely
re-ceive a CVC. Thus, percutaneous placement
of a catheter into a central vein is a
fre-quent procedure in many clinical settings.
At the Geneva University Hospitals, 35,087
patients were hospitalized in 1998, and
2,848 CVCs were used in that year. If we
assume that none of the patients received
more than one CVC, then 8.1% of all
hos-pitalized patients had a CVC.
The most frequently used anatomical sites
for CVC insertion are the internal jugular and
the subclavian vein (1). However, in an
indi-vidual patient, criteria for choosing one
ap-proach over the other often remain unclear.
This choice could depend on the
complica-tion rate with each approach. Indeed, there is
a substantial risk of mechanical lesions (e.g.,
arterial puncture, pneumothorax, cardiac
tamponade, or nerve lesions) and thrombotic
or septic complications with each CVC. These
complications are related to the procedure of
the insertion or to the catheter itself. An
im-proved understanding of CVC-related risks
might help clinicians to choose one approach
over the other in specific clinical settings. The
aim of this quantitative systematic review was
to gather the best available evidence on the
risks related to internal jugular and
subcla-vian central venous catheterization, to
criti-cally appraise and synthesize the data, and to
quantify the different risks.
METHODS
Search Strategy. An extensive search of the relevant literature without restriction to the English language was carried out using
MEDLINE (PubMed and KnowledgeFinder 4.19), EMBASE, and the Cochrane Library. The date of the last electronic search was June 30, 2000. Key words used for the final search were “central venous catheter,” “cathete-rization,” “catheterisation,” “subclavian,” “jugular,” “complication,” “infection,” “thrombosis,” “success rate,” “stenosis,” “pneumothorax,” “hematothorax,” “clinical trial,” and “prospective.” Reference lists of re-trieved reports and relevant review articles (2–5) were also checked. We did not contact manufacturers. Authors were contacted if there was ambiguity about original data.
Inclusion and Exclusion Criteria, End Points, and Data Extraction.Relevant studies had to be full reports of comparisons of inter-nal jugular vs. subclavian central venous cath-eterization, in adults or children, reporting dichotomous data on any complications that were possibly related to the CVC (insertion or catheter related), and published in a peer-reviewed journal. The type of CVC (e.g., dial-ysis) had no influence on these inclusion cri-teria. Data from abstracts, letters, review articles, animal studies, and postmortem stud-ies were not considered. Studstud-ies on tunneled catheters, implantable devices, radiologically assisted catheter insertion, or catheter place-ment by cut-down technique were not
ana-From the Divisions of Anaesthesiology (SR, MRT) and Surgical Intensive Care (BW), Department Anaes-thesiology, Pharmacology, and Surgical Intensive Care, University Hospitals of Geneva, Geneva, Switzerland.
MRT received a PROSPER Grant (32–51939.97) from the Swiss National Science Foundation.
Copyright © 2002 by Lippincott Williams & Wilkins
Objective:
To test whether complications happen more often
with the internal jugular or the subclavian central venous
ap-proach.
Data Source:
Systematic search (MEDLINE, Cochrane Library,
EMBASE, bibliographies) up to June 30, 2000, with no language
restriction.
Study Selection:
Reports on prospective comparisons of
inter-nal jugular vs. subclavian catheter insertion, with dichotomous
data on complications.
Data Extraction:
No valid randomized trials were found.
Sev-enteen prospective comparative trials with data on 2,085 jugular
and 2,428 subclavian catheters were analyzed. Meta-analyses
were performed with relative risk (RR) and 95% confidence
in-terval (CI), using fixed and random effects models.
Data Synthesis:
In six trials (2,010 catheters), there were
significantly more arterial punctures with jugular catheters
com-pared with subclavian (3.0% vs. 0.5%, RR 4.70 [95% CI, 2.05–
10.77]). In six trials (1,299 catheters), there were significantly less
malpositions with the jugular access (5.3% vs. 9.3%, RR 0.66
[0.44 – 0.99]). In three trials (707 catheters), the incidence of
bloodstream infection was 8.6% with the jugular access and 4.0%
with the subclavian access (RR 2.24 [0.62– 8.09]). In ten trials
(3,420 catheters), the incidence of hemato- or pneumothorax was
1.3% vs. 1.5% (RR 0.76 [0.43–1.33]). In four trials (899), the
incidence of vessel occlusion was 0% vs. 1.2% (RR 0.29 [0.07–
1.33]).
Conclusions:
There are more arterial punctures but less
cath-eter malpositions with the internal jugular compared with the
subclavian access. There is no evidence of any difference in the
incidence of hemato- or pneumothorax and vessel occlusion. Data
on bloodstream infection are scarce. These data are from
non-randomized studies; selection bias cannot be ruled out. In terms
of risk, the data most likely represent a best case scenario. For
rational decision-making, randomized trials are needed. (Crit Care
Med 2002; 30:454 –460)
K
EYW
ORDS: meta-analysis; bloodstream infection;
pneumotho-rax; hematothopneumotho-rax; arterial puncture; malposition; catheter;
com-plications; risk; harm
lyzed. Abstracts of all potentially relevant re-ports were screened by two authors (SR and BW). Papers that did not clearly meet the inclusion criteria were excluded at this stage. From included studies, dichotomous data on complications per insertion site were ex-tracted as reported by the original authors. Data on patients, clinical settings, and opera-tors were extracted from each included study. Data abstraction was carried out by one inves-tigator (SR) and cross-checked by the two oth-ers.
Validity Assessment.There was an inten-tion to include data from randomized trials only. However, our search strategies did not retrieve one single randomized trial that was relevant for the purpose of our study (i.e., in which patients were randomly allocated to one of the two CVC accesses). In clinical studies that are not properly randomized, there is a risk of selection bias; selection bias may lead to overestimation of the effect of a treatment (6). Another frequent problem with nonran-domized studies is unequal group sizes. This may happen when trialists have a free choice of what intervention to use, and when for external reasons (e.g., institutional policies) one intervention is systematically used more often compared with another. Undue weight would then be given to the intervention that is used more often. We assume that the magni-tude of inequality between group sizes in a comparative but nonrandomized trial to some extent reflects the degree of selection bias in that trial. Thus, to minimize the risk of selec-tion bias, we excluded trials from further anal-yses when their group sizes differed more than twofold. Finally, we excluded trials with retro-spective data collection, inasmuch as those are prone to observational bias and because data selection cannot be ruled out.
Meta-analyses.Dichotomous data on com-plications per number of inserted CVCs were analyzed. We calculated relative risks (RR) with 95% confidence intervals (CI) (RevMan version 4.0, Cochrane Library, Oxford, UK). We used the fixed-effect model when the data were homogeneous (p⬎.1) and, otherwise, a random effects model. As an estimate of the clinical relevance of any difference between the two CVC accesses, we calculated the num-ber-needed-to-treat (7). The number-needed-to-treat indicated how many catheters had to be inserted by one route for one catheter to lead to a complication that would not have happened had the other approach been cho-sen. A positive number-needed-to-treat indi-cated that the end point happened more often with the jugular approach. A negative num-ber-needed-to-treat indicated the opposite.
RESULTS
Retrieved Reports.
We screened 747
reports (Fig. 1). Of the 35 potentially
rel-evant trials, 4 were excluded because data
on complications were lacking (8 –11),
and 7 because the insertion site was
un-clear (12–18). Seven trials were excluded
because the number of analyzed jugular
and subclavian catheters was highly
un-equal (i.e., ratio
⬎
2); ratios were 2.6 (19),
2.8 (20), 3.0 (21), 3.1 (22), 3.9 (23), 4.9
(24), and 8.1 (25). We contacted authors
of seven original studies for further
infor-mation; none replied to our enquiry.
We eventually analyzed 17 prospective,
comparative, nonrandomized reports,
pub-lished between 1982 and 1999, with data on
2,085 jugular and 2,428 subclavian
cathe-ters (Table 1) (26 – 42). One trial was in
children (33), all others were in adults.
Av-erage group size was 150 catheters for
jug-ular access (range, 10 –306), and 139 for
subclavian access (range 19 – 498).
Cathe-ters varied from single to quadruple-lumen,
and were most often made of polyurethane
or polyethylene. Most patients were from
surgical or medical intensive care units.
Catheters were inserted by house officers,
residents, or fellows. In three reports, no
such information was available (32, 38, 39).
In some trials, additional central vein
ac-cesses (e.g., femoral) were tested; these data
were not analyzed.
Arterial Puncture.
Arterial puncture
(six trials, 2,010 catheters) (27–29, 31,
36, 41) was significantly more often
re-ported with the jugular approach
com-pared with the subclavian approach
(3.0% vs. 0.5%, RR 4.70 [95%CI, 2.05–
10.77]) (Fig. 2). The
number-needed-to-treat was 39 (95%CI, 27–73).
Catheter Malposition.
Catheter
mal-position (six trials, 1,299 catheters) (29,
31, 33, 34, 36, 41) was significantly less
often reported with the jugular access
(5.3% vs. 9.3%, RR 0.66 [95%CI, 0.44 –
0.99]) (Fig. 3). The
number-needed-to-treat was
⫺
25 (95%CI,
⫺
15 to
⫺
82).
Bloodstream Infection.
Three trials
(707 catheters) (32, 38, 40) reported on
bloodstream infection. The incidence was
8.6% with the jugular access and 4.0%
with the subclavian access. The data were
heterogenous; the RR (random effects
model) was 2.24 (95% CI, 0.62– 8.09)
(Fig. 4).
Hemato- and Pneumothorax.
In ten
trials (3,420 catheters) (26 –29, 31–33,
35, 36, 41), there was no difference in the
incidence of hemato- or pneumothorax
(1.3% vs. 1.5%, RR 0.76 [0.43–1.33]) (Fig.
5).
Vessel Occlusion.
Four trials (899
catheters) (28, 31, 33, 40) reported on
vessel stenosis or thrombosis. There was
no significant difference in the incidence
of vessel occlusion (0% vs. 1.2%, RR 0.29
[0.07–1.33]) (Fig. 6).
Further reported complications were
insertion site infection and local
hema-toma. These were reported in no more
than two trials each; no sensible
conclu-sions could be drawn.
DISCUSSION
In many institutions, the anatomical
site of CVC insertion is chosen on the
grounds of personal experience or local
policies rather than on evidence-based
guidelines. The aim of this quantitative
systematic review was to clarify some of
the controversies that exist on the
rela-tive risk of internal jugular compared
Figure 1.Flow chart of the literature search.with subclavian access. If there was
evi-dence for an increased risk of specific
complications with one approach, then
clinicians may take advantage of that
knowledge for insertion of a CVC in an
individual patient.
The main methodological problem
was that there were no relevant
random-ized trials. There is empirical evidence
that data from studies that lack proper
randomization may overestimate the
ef-fect of a treatment (6). Observational data
on treatment efficacy do not necessarily
need to be different from randomized
data (43). Also, our analysis is about harm
and not about efficacy, and there is
evi-dence from systematic review that
obser-vational studies may underestimate the
Figure 2.Relative risk (RR) of arterial puncture with jugular vs. subclavian access.n, number with arterial puncture;N, number of catheters;95% CI Fixed, 95% confidence interval, fixed effect model.Table 1.Analyzed trials
Reference
Analyzed Catheters
Catheter Type Setting Operator
Jug Subcl
Barrera 1996 (26) 52 63 CVC, 3-lumen (7-Fr), introducer sheaths (8-Fr)
ICU and ward ICU fellows
Bo-Linn 1982 (27) 201 179 CVC 10% cardiopulmonary arrest; 59% emergencies (catheterization within 1 hr); 31% elective (catheterization within 1 day)
Medical house officers in training
Eisenhauer 1982 (28) 248 286 CVC Surgical service (volume resuscitation, CVP monitoring), 39.9% as an emergency procedure
Surgical house officers in training Fisher 1999 (29) 306 352 CVC, 3-lumen (7-Fr) or 4-lumen
(8-Fr), PAC (8,5-Fr), dialysis catheter (12-Fr)
Gastroenterology unit, patients with liver disease (INR 1.5 or greater, platelet count 150 or less⫻109/L)
Attending clinician
Gil 1989 (30) 118 87 CVC, 1-lumen (polyurethane) and 3-lumen (polyvinylchlorid)
Medical ICU Treating physician
Kaiser 1981 (31) 52 62 CVC Oncology Main author
Lehr 1988 (32) 290 498 CVC Abdominal surgery Unknown
Luyt 1996 (33) 56 94 CVC, 1-, 2-, and 3-lumen Multidisciplinary ICU Residents
McGee 1996 (34) 65 37 CVC, 16 cm and 20 cm Community teaching hospital Surgical, medical, or anesthesia residents Miller 1999 (35) 230 176 PAC (7,5-Fr, polyvinylchlorid), CVC
(7-Fr, polyurethane)
Surgical and medical ICU and general medical ward
Residents or 4th-year medical students Peres 1990 (36) 58 107 CVC, 3-lumen Operating room and ICU Anesthesiologist Pinilla 1983 (37) 50 44 CVC (14 G and 17 G), PAC, 4-lumen
G188(7 Fr)
Surgical ICU and operating room Residents Poisson 1991 (38) 76 107 CVC (polyurethene and polyethylene) ICU Unknown
Richet 1990 (39) 114 194 CVC ICU Unknown
Schillinger 1994 (40) 50 50 Dialysis catheter Hemodialysis Nephrology medical staff
Sessler 1987 (41) 101 72 CVC, 1-lumen (16 G) or 3-lumen (7-Fr), Side port introducer (8.5-Fr), PAC
ICU and medical center House officers in training Singh 1982 (42) 18 19 PAC Critical care center, operating room and
emergency room
ICU fellows or residents Jug, jugular; Subcl, subclavian; CVC, central venous catheter; Fr, French; ICU, intensive care unit; PAC, pulmonary artery catheter; G, gauge.
harm of a treatment compared with
ran-domized controlled trials (44). Another
issue is related to the operators in these
studies. For instance, to use exclusively
one CVC approach in an institution may
reduce the CVC-related risk inasmuch as
all operators would then accumulate
ex-perience and improve dexterity. Finally,
there may have been the possibility of
underreporting of CVC-related
tions in studies where these
complica-tions were secondary end points. This
may explain, at least in part, some of the
observed variability in event rates in
these trials. However, there was no
evi-dence that this potential underreporting
would have favored one approach
com-pared with the other. Consequently, these
data are likely to represent a best case
scenario in terms of risk assessment for
both CVC accesses.
Figure 4.Relative risk (RR) of bloodstream infection with jugular vs. subclavian access.n, number with bloodstream infection;N, number of catheters;95% CI Random, 95% confidence interval, a random effects model.
Figure 5.Relative risk (RR) of hemato- or pneumothorax with jugular vs. subclavian access.n, number with hemato- or pneumothorax;N, number of catheters;95% CI Fixed, 95% confidence interval, fixed effect model.
Figure 3.Relative risk (RR) of catheter malposition with jugular vs. subclavian access.n, number with catheter malpositions;N, number of catheters;95% CI Fixed, 95% confidence interval, fixed effect model.
One of the most frequently reported
complications of CVC insertion is arterial
puncture (45). Based on these
systemat-ically searched data, of 39 attempts to
access the internal jugular vein, 1 will
lead to an arterial puncture that would
not have been the case had the subclavian
approach been chosen. We do not know if
this significantly increased risk with the
jugular access was the result of an
under-reporting of arterial punctures with the
subclavian approach inasmuch as
punc-ture of a carotid artery is usually easier to
detect than puncture of a subclavian
ar-tery. Although the puncture of a carotid
artery seems to happen more often,
effec-tive hemostasis is much easier (manual
compression). Also, Doppler ultrasound
guidance is an effective method in
de-creasing the incidence of this
complica-tion (46). It is unlikely that clinicians will
abandon the internal jugular access based
on these risk data.
The data on catheter malpositioning
may have more impact on clinical
deci-sion-making. Malpositioning has been
re-ported in 14% of CVCs even when they
were inserted by experienced clinicians
(47). In the analyzed trials,
malposition-ing happened significantly less often with
the internal jugular access. Of 25 CVCs
inserted via the internal jugular vein, one
will not be malpositioned that would have
been the case had the subclavian route
been chosen. Also, malposition of a
sub-clavian catheter may include entry into
the opposite subclavian vein or the neck
veins, whereas many jugular catheters
may simply be pulled back if the tip lies in
the right atrium. This is yet another
ar-gument in favor of the jugular approach.
The jugular access should be chosen if a
fast and correct catheter tip placement is
mandatory (e.g., hemodynamic
monitor-ing in a patient in shock). Malpositionmonitor-ing
of a CVC
per se
, independent of the access
may lead to serious complications. The
positioning of catheter tips within the
cardiac silhouette is associated with an
increased risk of cardiac tamponade (48).
Also, positioning of a catheter tip in a
subclavian vein is associated with a high
risk of thrombus formation and vessel
occlusion (49). In cancer patients, for
in-stance, rates of catheter-related thrombi
of
⬎
50% have been reported (50). When
a CVC is leading to a thrombosis, the risk
of catheter-related sepsis may be
in-creased (19). In this meta-analysis, only
four trials reported on vessel occlusion
(vessel stenosis or thrombosis), and there
was no evidence of any difference
be-tween the two CVC approaches.
Catheter-related bloodstream
infec-tion remains an important cause of
nos-ocomial infection. In patients with a CVC,
the risk of acquiring a bloodstream
infec-tion was reported to range between 1%
and 10% (51). The economic burden of
bloodstream infection is considerable; it
may prolong the hospital stay and
in-crease mortality in critically ill patients
(52). Mortality of catheter-related
blood-stream infection increases with older age,
longer length of hospital stay, cancer,
dis-ease of the digestive tract, and
Candida
species detection (53). Data on
blood-stream infection could be analyzed from
three trials (707 catheters) only.
Blood-stream infection happened more often
with jugular catheters. The absolute risk
reduction compared with the subclavian
access was almost 5%; this would
trans-late into a number-needed-to-treat of
about 20. This potentially clinically
rele-vant result was, however, not statistical
significant (i.e., the 95% CI of RR
in-cluded 1).
There was no evidence of any
differ-ence in the inciddiffer-ence of hemato- and
pneumothorax with the two approaches.
This result may be somewhat unexpected
because many clinicians believe that the
subclavian access is more prone to these
complications. This equality may reflect
the lack of randomization in the original
trials, leading to selection bias. Thus, the
conclusion must be that, in experienced
hands, both accesses have the same low
risk of hemato- and pneumothorax.
Pa-tients at increased risk of pulmonary
complications (e.g., patients with chronic
obstructive pulmonary disease or acute
respiratory distress syndrome) have not
been included in these studies.
CONCLUSIONS
These data from nonrandomized
clin-ical trials are likely to reflect a best case
scenario in terms of risk assessment.
There is some evidence that there are
more arterial punctures but less catheter
malpositions with the internal jugular
compared with the subclavian access.
There is no evidence of any difference in
the incidence of hemato- or
pneumotho-rax and vessel occlusion. Bloodstream
fection happened more often with the
in-ternal jugular access, but the data were
heterogeneous and there was no evidence
of statistical significance. The lack of
ran-domized comparisons of jugular and
sub-Figure 6.Relative risk (RR) of vessel occlusion with jugular vs. subclavian access.n, number with vessel occlusion;N, number of catheters;95% CI Fixed, 95% confidence interval, fixed effect model.
T
here is some
evi-dence that there
are more arterial
punctures but less catheter
malpositions with the
inter-nal jugular compared with
the subclavian access.
clavian catheters is highly unsatisfactory.
Randomized comparisons are needed to
verify these findings. In the meantime,
clinical decision-making will have to be
based on the best available evidence.
ACKNOWLEDGMENTS
We thank Daniel Haake from the
Doc-umentation Service of the Swiss Academy
of Medical Sciences for his help in
search-ing electronic databases.
REFERENCES
1. Felleiter P, Gustorff B, Lierz P, et al: Use of electrocardiographic placement control of central venous catheters in Austria.Acta Med Austriaca1999; 26:109 –113
2. Agee KR, Balk RA: Central venous catheter-ization in the critically ill patient.Crit Care Clin1992; 8:677– 686
3. Hagley MT, Martin B, Gast P, et al: Infectious and mechanical complications of central ve-nous catheters placed by percutaneous veni-puncture and over guidewires.Crit Care Med
1992; 20:1426 –1430
4. Paternoster G, Molino A, Alloatti S: Emer-gency and temporary vascular access. Min-erva Urol Nefrol1998; 50:1– 8
5. Skolnick ML: The role of sonography in the placement and management of jugular and subclavian central venous catheters. Am J Roentgenol1994; 163:291–295
6. Schulz KF, Chalmers I, Hayes RJ, et al: Em-pirical evidence of bias. Dimensions of meth-odological quality associated with estimates of treatment effects in controlled trials.
JAMA1995; 273:408 – 412
7. McQuay HJ, Moore RA: Using numerical re-sults from systematic reviews in clinical practice.Ann Intern Med1997; 126:712–720 8. Eyer S, Brummitt C, Crossley K, et al: Cath-eter-related sepsis: Prospective, randomized study of three methods of long-term catheter maintenance. Crit Care Med 1990; 18: 1073–1079
9. Mermel LA, McCormick RD, Springman SR, et al: The pathogenesis and epidemiology of catheter-related infection with. Am J Med
1991; 91:197S–205S
10. Senagore A, Waller JD, Bonnell BW, et al: Pulmonary artery catheterization: A prospec-tive study of internal jugular and subclavian approaches.Crit Care Med1987; 15:35–37 11. Sznajder JI, Zveibil FR, Bitterman H, et al:
Central vein catheterization. Failure and complication rates by three percutaneous ap-proaches. Arch Intern Med 1986; 146: 259 –261
12. Ahmed N: Thrombosis after central venous cannulation.Med J Aust1976; 1:217–220 13. Cairoli R, De Gasperi A, Gimmelli A, et al:
Untunnelled central venous catheters in the treatment of acute leukemia.Haematologica
1993; 78:418 – 419
14. Farkas JC, Liu N, Bleriot JP, et al: Single-versus triple-lumen central catheter-related sepsis: A prospective randomized study in a critically ill population.Am J Med1992; 93: 277–282
15. Lindblad B, Wolff T: Infectious complications of percutaneously inserted central venous catheters.Acta Anaesthesiol Scand1985; 29: 587–589
16. McGee WT, Ackerman BL, Rouben LR, et al: Accurate placement of central venous catheters: A prospective, randomized, mul-ticenter trial. Crit Care Med 1993; 21: 1118 –1123
17. Pettigrew RA, Lang SD, Haydock DA, et al: Catheter-related sepsis in patients on intra-venous nutrition: A prospective study of quantitative catheter cultures and guidewire changes for suspected sepsis.Br J Surg1985; 72:52–55
18. Queiros J, Cabrita A, Maximino J, et al: Cen-tral catheter for hemodialysis: Six months experience of 103 catheters. Nephrologie
1994; 15:113–115
19. Timsit JF, Farkas JC, Boyer JM, et al: Central vein catheter-related thrombosis in intensive care patients: Incidence, risks factors, and relationship with catheter-related sepsis.
Chest1998; 114:207–213
20. Almirall J, Gonzalez J, Rello J, et al: Infection of hemodialysis catheters: Incidence and mechanisms.Am J Nephrol1989; 9:454 – 459 21. Horowitz HW, Dworkin BM, Savino JA, et al: Central catheter-related infections: Compar-ison of pulmonary artery catheters and triple lumen catheters for the delivery of hyperali-mentation in a critical care setting.
J Parenter Enteral Nutr1990; 14:588 –592 22. Gray P, Sullivan G, Ostryzniuk P, et al: Value
of postprocedural chest radiographs in the adult intensive care unit. Crit Care Med
1992; 20:1513–1518
23. Heberer M, Moser J, Durig M, et al: Prospec-tive study of complications of central venous catheters. Infusionsther Klin Ernahr 1984; 11:254 –261
24. Jantsch H, Lechner G, Mauritz W, et al: Dis-placements and complications of superior vena cava catheter. Radiologe 1981; 21: 463– 469
25. Scalea TM, Sinert R, Duncan AO, et al: Per-cutaneous central venous access for resusci-tation in trauma. Acad Emerg Med 1994; 1:525–531
26. Barrera R, Mina B, Huang Y, et al: Acute complications of central line placement in profoundly thrombocytopenic cancer pa-tients.Cancer1996; 78:2025–2030 27. Bo-Linn GW, Anderson DJ, Anderson KC, et
al: Percutaneous central venous catheteriza-tion performed by medical house officers: A prospective study.Cathet Cardiovasc Diagn
1982; 8:23–29
28. Eisenhauer ED, Derveloy RJ, Hastings PR: Pro-spective evaluation of central venous pressure
(CVP) catheters in a large city-county hospital.
Ann Surg1982; 196:560 –564
29. Fisher NC, Mutimer DJ: Central venous can-nulation in patients with liver disease and coagulopathy—a prospective audit.Intensive Care Med1999; 25:481– 485
30. Gil RT, Kruse JA, Thill-Baharozian MC, et al: Triple- vs single-lumen central venous cath-eters. A prospective study in a critically ill population. Arch Intern Med 1989; 149: 1139 –1143
31. Kaiser CW, Koornick AR, Smith N, et al: Choice of route for central venous cannula-tion: Subclavian or internal jugular vein? A prospective randomized study.J Surg Oncol
1981; 17:345–354
32. Lehr L, Brandmair W, Kovacs J, et al: Expe-riences with 3151 central venous catheters at a surgical clinic. Dtsch Med Wochenschr
1988; 113:1218 –1223
33. Luyt DK, Mathivha LR, Litmanovitch M, et al: Confirmation of the safety of central ve-nous catheterisation in critically ill infants and children—the Baragwanath experience.
S Afr Med J1996; 86:603– 606
34. McGee WT, Moriarty KP: Accurate placement of central venous catheters using a 16-cm catheter. J Intensive Care Med 1996; 11: 19 –22
35. Miller JA, Singireddy S, Maldjian P, et al: A reevaluation of the radiographically detect-able complications of percutaneous venous access lines inserted by four subcutaneous approaches.Am Surg1999; 65:125–130 36. Peres PW: Positioning central venous
cathe-ters—a prospective survey. Anaesth Intensive Care1990; 18:536 –539
37. Pinilla JC, Ross DF, Martin T, et al: Study of the incidence of intravascular catheter infec-tion and associated septicemia in critically ill patients.Crit Care Med1983; 11:21–25 38. Poisson DM, Touquet S, Bercault N, et al:
Polyurethane versus polyethylene: In vivo
randomized study of infectious complica-tions of central catheterization.Pathol Biol
1991; 39:668 – 673
39. Richet H, Hubert B, Nitemberg G, et al: Pro-spective multicenter study of vascular-catheter-related complications and risk fac-tors for positive central-catheter cultures in intensive care unit patients.J Clin Microbiol
1990; 28:2520 –2525
40. Schillinger F, Schillinger D, Montagnac R, et al: Central venous stenosis in hemodialysis: Comparative angiographic study of subcla-vian and internal jugular access.Nephrologie
1994; 15:129 –131
41. Sessler CN, Glauser FL: Central venous can-nulation done by house officers in the inten-sive care unit: A prospective study. South Med J1987; 80:1239 –1242, 1248
42. Singh S, Nelson N, Acosta I, et al: Catheter colonization and bacteremia with pulmonary and arterial catheters.Crit Care Med1982; 10:736 –739
controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med2000; 342:1887–1892
44. Trame`r MR, Moore RA, Reynolds DJ, et al: Quantitative estimation of rare adverse events which follow a biological progression: A new model applied to chronic NSAID use.
Pain2000; 85:169 –182
45. Yilmazlar A, Bilgin H, Korfali G, et al: Com-plications of 1303 central venous cannula-tions.J R Soc Med1997; 90:319 –321 46. Randolph AG, Cook DJ, Gonzales CA, et al:
Ultrasound guidance for placement of central venous catheters: A meta-analysis of the lit-erature.Crit Care Med1996; 24:2053–2058 47. Gladwin MT, Slonim A, Landucci DL, et al:
Cannulation of the internal jugular vein: Is postprocedural chest radiography always necessary? Crit Care Med 1999; 27: 1819 –1823
48. Collier PE, Blocker SH, Graff DM, et al: Car-diac tamponade from central venous cathe-ters.Am J Surg1998; 176:212–214 49. Bennegard K, Curelaru I, Gustavsson B, et al:
Material thrombogenicity in central venous catheterization. I. A comparison between un-coated and heparin-un-coated, long antebra-chial, polyethylene catheters.Acta Anaesthe-siol Scand1982; 26:112–120
50. Wu X, Studer W, Skarvan K, et al: High incidence of intravenous thrombi after short-term central venous catheterization of the
internal jugular vein. J Clin Anesth1999; 11:482– 485
51. Adal KA, Farr BM: Central venous catheter-related infections: a review.Nutrition1996; 12:208 –213
52. Pittet D, Tarara D, Wenzel RP: Nosocomial bloodstream infection in critically ill pa-tients. Excess length of stay, extra costs, and attributable mortality. JAMA 1994; 271: 1598 –1601
53. Pittet D, Li N, Woolson RF, et al: Micro-biological factors influencing the out-come of nosocomial bloodstream infec-tions: A 6-year validated, population-based model. Clin Infect Dis 1997; 24: 1068 –1078
