RENAL NURSING

Central venous access for haemodialysis: prospective evaluation of

possible complications
Denise de Andrade PhD, RN
University of São Paulo at Ribeirão Preto College of Nursing, WHO Collaborating Centre for Nursing Research Development,
Brazil

Viviane Ferreira MSc, RN

University of São Paulo at Ribeirão Preto College of Nursing, WHO Collaborating Centre for Nursing Research Development,
Brazil

Submitted for publication: 6 October 2005

Accepted for publication: 2 March 2006

Correspondence: D E A N D R A D E D & F E R R E I R A V ( 2 0 0 7 ) Journal of Clinical Nursing 16, 414–418

Dr Denise de Andrade
Escola de Enfermagem de Ribeirão Preto
Campus Universitário
14040-902 Ribeirão Preto-SP
Brazil
E-mail: dandrade@eerp.usp.br
Central venous access for haemodialysis: prospective evaluation of possible com-
plications
Aims and objectives. The combination of chronic renal insufficiency and haemo-
dialysis represents a challenge for health professionals. Chronic renal insufficiency
patients undergoing haemodialysis treatment through a temporary double-lumen
catheter were prospectively studied in order to identify the type and frequency of
local and systemic complications.
Methods. A six-month period was established with a view to the inclusion of new
cases. Data were acquired through interviews, clinical assessment and patient re-
cords, and entered into a Microsoft Excel database through a double entry system
and exported to the Statistical Package Social Sciences software. Sixty-four patients
were evaluated prospectively, of which thirty-eight (59.4%) were men and 35
(54.7%) required catheter insertion for immediate treatment. During the study
period, 145 catheters were inserted, ranging from 1 to 7 implants per patient, 29
(45.3%) were single insertions and 127 (87.6%) catheters were inserted into the
jugular vein. The catheters were left in place for an average of 30 days.
Results. Forty-one (64%) presented inadequate functioning, after about 26 days. A
febrile state occurred in 24 (37.5%) patients after 34 days, secretion at the catheter
entry site in 27 (42.2% after 26 days and bloodstream infection was encountered in
34(53%) after 34 days. Of the 61 blood culture samples, thirty (49%) were positive
for Staphylococcus aureus that was the microorganism most frequently isolated.
Conclusion. The findings indicate worrying aspects such as the catheters permanence
time, exposing patients to different complications, including infection. Furthermore,
inadequate catheter functioning leads to inefficient haemodialysis treatment.
Relevance to clinical practice. Knowledge about complications allows for systematic
care planning, prevention and control actions.

Key words: ambulatory care, chronic renal insufficiency, haemodialysis, implantable

catheters, institution, nurses, nursing, renal dialysis, renal nursing

414  2007 Blackwell Publishing Ltd

doi: 10.1111/j.1365-2702.2006.01654.x
Renal nursing
Introduction
Renal insufficiency represents a serious health problem due to
its high incidence, technological complexity and operational
cost of treatment and the need for specialized professionals. In
general, treatments have offered effective results in terms of
expectation and quality of life, as well as the reduction
of comorbidities for chronic renal insufficiency (CRI) patients.
Over the last few decades, nephrology has increasingly
developed new biomaterial and technologies with repercus-
sions on the outcome of renal insufficiency patients’ treatment.
In Brazil, dialysis is carried out in specialized centres, located
in hospitals or clinics supported by the national health services
of the country called the ‘Sistema Única de Saúde’ (SUS). It is
one of the largest public health systems in the world, which
guarantees integral and totally free care to the population,
including HIV patients, whether symptomatic or not, as well
as patients with chronic renal diseases or cancer (Sistema
Único de Saúde 2005). Renal insufficiency patients have the
following treatment options at their disposal: conservative (use
of medication and/or diet) or dialysis treatment, covering
haemodialysis (HD), Intermittent Peritoneal Dialysis (IPD),
Continuous Outpatient Peritoneal Dialysis (CAPD), Continu-
ous Cyclical Peritoneal Dialysis (CCPD) and Renal Transplant
(RT). Dialysis treatments do not fully substitute kidney
function, but represent a possibility of survival for these
patients and even allow them to resume social activities.
It is estimated that 13% of chronic renal insufficiency
patients undergoing haemodialysis treatment are regularly
treated by means of temporary or permanent catheters
(Canaud 1999). The use of the temporary double-lumen
catheter (TDLC), also called a non-tunneled venous catheter,
offers various benefits: practicality, rapid insertion, immedi-
ate use, painless during haemodialysis session, produces low
venous resistance and fast and easy withdrawal. Low blood
flow and inefficiency in haemodialysis can be associated with
inadequate location of the catheter tip or reduced central
bloodstream (Besarab & Raja 2003).
Literature reviews present a large number of research
results revealing high levels of complications associated
with temporary double-lumen catheters. Therefore, renal
therapy care practice should be based on a series of
carefully established activities, including epidemiological
surveillance of infection episodes and other complications
and, at the same time, prevention and control measure
proposals.
Recognizing the seriousness of complications in chronic
renal insufficiency patients under haemodialysis treatment
through a temporary double-lumen catheter, professionals
have been planning prevention and control measures,
 2007 Blackwell Publishing Ltd, Journal of Clinical Nursing, 16, 414–418
Complications central venous access
especially to avoid infections. Thus, this study aimed to
evaluate, prospectively, chronic renal insufficiency patients
submitted to haemodialysis treatment by means of a
temporary double-lumen catheter, to identify the type and
frequency of local and systemic complications.
Materials and methods
Study design
This prospective segment study was developed at a nephrology
clinic in the interior of São Paulo State-Brazil. Sixty-four CRI
patients under haemodialysis treatment through TDLC were
studied. This group included all patients who had a catheter
implanted over a six-month period (from 1 July 2003 to 31
December 2003). They were evaluated during 12 months,
until the final removal of the catheter. Inclusion criteria were:
accept study participation, be a CRI patient, undergoing HD
treatment at the clinic, the TDLC implanted between 1 July
2003 and 31 December 2003. Exclusion criteria were: catheter
insertion outside the period, patients who gave up HD
treatment or changed dialysis mode to CCPD or RT, patients
whose arteriovenous fistula was viable for puncture, cases of
hospitalization, transference to another clinic and death.
Data were collected each time patients attended the clinic
for haemodialysis through interviews, clinical exam and
patient record evaluation. Patients were assessed until final
withdrawal of the TDLC.
Ethical approval was obtained from the Ethics Committee
for Research involving human beings. Patients were informed
about the study aims and received guarantees about ano-
nymity and the right to withdraw from the investigation at
any time.
Definitions
The following concepts will be used in this study:
1 Local complication: adverse reactions or traumas that oc-
cur on the TDLC entry site for haemodialysis (haematoma,
pain, oedema, heat, hyperemia, purulent secretion, pruritis
and catheter malfunctioning). We considered catheter
malfunctioning or dysfunction to be the case when the
catheter did not provide a minimally adequate blood flow
needed for effective dialysis. This complication can be
caused by incorrect placement, fibrin formation in the
catheter tip and intraluminal thrombosis, among others.
2 Systemic complication: signs and symptoms that may or
may not affect the organism as a whole and, therefore, may
or may not put the patient at risk. These include fever,
pyrogenic reaction and bacteremia.
415
D de Andrade and V Ferreira

• Fever or hyperthermia: body temperature above 37Æ6 C,
more common in immunodepressed patients undergoing
haemodialysis through TDLC, making them more sus-
ceptible to infection.
• Pyrogenic reaction: complication caused by endotoxins,
characterized by acute responses such as: fever, chills,
shivering, headache, hypotension.
• Bacteremia: presence of bacteria in the bloodstream.
Symptoms include shivering, chills, fever, weakness, nau-
sea, vomiting.
3 Days until occurrence of complications: represents the
number of days until signs and symptoms of local and/or
systemic complications appear.
Data analysis
Facts for each variable in the data collection instrument were
coded in a Microsoft Excel database. After validation through
double entry, results were exported to Statistical Package Social
Sciences (SPSS) software version 10.0 for statistical analysis,
using means, standard deviation, median and quartiles.
Results
The age of the patients was 56Æ7 ± 15 years (range 16–
86 years, median 57Æ5, first and third quartiles 48 and
67Æ2 years), 38 (59Æ4%) were males and 26 females. Twenty
(31Æ2%) presented hypertensive nephrosclerosis and 19
(29Æ7%) diabetic nephropathy as the probable cause of CRI,
among other likely etiologies. Thirty-five (54Æ7%) patients
required a TDLC for immediate treatment, 20 (31Æ2%) for loss
of the arteriovenous fistula and nine (14Æ1%) without condi-
tions of obtaining fistula for dialysis treatment. One hundred
and forty-five catheters were implanted in the 64 participants,
29 (45Æ3%) had one insertion, 17 (26Æ6%) two and 15 (28Æ1%)
between three and seven insertions. One hundred and twenty-
seven catheters (87Æ6%) were inserted into the right or left
jugular vein, followed by 13 (9%) into a right or left subclavian
vein and five (3Æ4%) in the femoral vein. Catheter implantation
problems occurred in 21 (14Æ5%) cases, eight (5Æ5%) had
difficulty to tap a vein, four (2Æ8%) with intense bleeding
during insertion and four (2Æ8%) had hematomas. The cath-
eters were left in place 30 (SD 39) days (range 1–366 days).
The main reasons for catheter change or permanent removal
were: hyperthermia, an adequate arteriovenous fistula punc-
ture with an inadequate catheter functioning. Local and
systemic complications ranged from 0 to 36 (8Æ8 SD 8Æ2) equal
to 2 and 13 complications per patient. Further details of the
types of local complications, days until occurrence of compli-
cations and number of complications/patients may be seen in
Tables 1 and 2.
The investigation showed that 30 (49%) of 61 blood
culture samples were positive, confirming bloodstream infec-
tion seen after an average of 34 days. Gram-positive bacteria
were the microorganisms most frequently isolated, 10
(33Æ4%) samples were colonized with Staphylococcus aureus,
followed by eight (26Æ7%) with coagulase-negative Staphy-
lococci, among others.
It was found that topical antimicrobial agents were used on
the catheter entry site, that is, 33 (51Æ5%) patients used
mupirocin cream (Bactroban) and 31 (48Æ5%) 10% PVP-I
cream. Secretion was present in 18 (54Æ5%) patients using
mupirocin cream and in nine (29%) using 10% PVP-1 cream.
Details of the types of systemic complications, days until
occurrence of complications and number of complications/
patients may be seen in Tables 3 and 4. The number of
systemic complications in this study was relatively small
(range 1–4 complications per patient). At home, the most
frequent complication in 18 (28Æ1%) patients was fever,
followed by insertion pain in 16 (25%) and dirty catheter
dressing in 11 (17Æ2%).
Discussion
The predominance of male CRI patients is mentioned in
national and international literature. A Brazilian epidemio-

Table 1 Local complications as a result of TDLC insertion in CRI patients under haemodialysis treatment and days until occurrence,* Brazil,
2005

Complication No. % Minimum/maximum range Average Standard deviation Median Q1 Q3

Catheter malfunctioning 41 64 1–96 26 26 16 8 34

Insertion pain 28 43 1–140 28 37 15 1 33
Insertion secretion 27 42 3–83 26 20 23 8 41
Insertion bleeding 24 37Æ5 1–221 39 50 21 8 49
Insertion hyperemia 18 28 3–113 32 31 24 10 41
Catheter obstruction 12 18 3–89 28 20 15 8 40
Insertion pruritis 3 4Æ6 23–78 43 30 28 23 78

*Multiple answers.

416  2007 Blackwell Publishing Ltd, Journal of Clinical Nursing, 16, 414–418
Renal nursing Complications central venous access

Table 2 Local complications as a result

Minimum/ Standard
of TDLC insertion in CRI patients
Complication Maximum range Average deviation Median Q1 Q3
under haemodialysis treatment and
number of complications/patients,* Catheter malfunctioning 1–14 3 1 2 1 4
Brazil, 2005 Insertion pain 1–4 2 2 2 1 3
Insertion secretion 1–9 2 1 2 1 3
Insertion bleeding 1–3 1 2 1 1 1
Insertion hyperemia 1–5 2 1 2 1 2
Catheter obstruction 1–2 1 2 1 1 1
Insertion pruritis 1 1 1 1 1 1

*Multiple answers.

Table 3 Systemic complications in CRI patients under haemodialysis treatment through TDLC and days until occurrence of complications,*
Brazil, 2005

Minimum/ Standard
Problems No. % maximum range Average deviation Median Q1 Q3

Fever 24 37Æ5 1–174 34 42 16 5 51

Bacteremia 15 23Æ4 1–115 46 10 40 40 59
Pyrogenic reactions 13 20Æ3 1–98 38 36 21 7 82

*Multiple answers.

Table 4 Systemic complications in CRI patients under haemodialysis and 21 days in the jugular and subclavian veins (NKF-DOQI
treatment through TDLC and number of complications/patients,* 1997,Besarab & Raja 2003). Specialists mention risk factors
Brazil, 2005
for complications related to patient, catheter type, profes-
Minimum/ sional ability (difficult venous access), low immunity,
maximum Standard hospitalizations in intensive therapy centres, catheter perma-
Problems range Average deviation Median Q 1 Q 3 nence time and excessive and/or inadequate manipulation,
Fever 1–4 1 1 1 1 2 among others (NKF-DOQI 1997,Besarab & Raja 2003).
Bacteremia 1 1 0 1 1 1 The most frequent local complications resulted from
Pyrogenic 1–3 1 0Æ5 1 1 1 incorrect catheter positioning, causing insufficient blood-
reactions
stream flow for adequate dialysis. Another possible cause of
*Multiple answers. inadequate functioning is thrombosis, confirmed through
angiography (Dittmer et al. 1999, Moysés et al. 2000, Besarab
logical study showed that 52% of CRI patients were men & Raja 2003). According to Besarab and Raja (2003)
(Sesso 2000). As to etiologies of CRI, epidemiological catheters used over extended periods present varied average
research has indicated arterial hypertension, diabetes mellitus blood flow rates, ranging from 200 to 250 ml/minute. Using
and glomerulonephritis as the most frequent causes (Besarab access with low blood flow rates leads to inadequate dialysis
& Raja 2003,Sociedade Brasileira de Nefrologia 2005). and may be related to patient morbidity and mortality.
It was found that urgent situations and immediate need for Catheter problems such as insufficient blood flow, in-
haemodialysis as the most frequent causes of temporary creased venous pressure in the system and presence of blood
catheter implantation. Most patients arrive in an urgent recirculation may affect haemodialysis success. Recirculation
situation to start haemodialysis treatment, which is why may also be due to insufficient blood flow and may
nephrologists inserted the TDLC as a temporary access. The significantly affect urea release, thus decreasing dialysis
National Kidney Foundation-Dialysis Outcomes Quality efficacy. Dialysis treatment for patients with TDLC can only
Initiative (NFK-DOQI) guidelines recommend the internal be efficient when blood flow rates are high (350/400 ml/
jugular vein as the primary choice for vascular access (NKF- minute). Blood flows below this level may impair treatment
DOQI 1997). efficacy (Delmez & Windus 1996,Johnson 1998,Dittmer
It was observed that the average catheter permanence time et al. 1999,Baracetti 2001). However, vascular access infec-
in this study exceeded recommended levels. Scientific tions are common in haemodialysis patients. There is no
evidence establishes a five-day period for the femoral vein agreement about the source of TDLC colonization. Some

 2007 Blackwell Publishing Ltd, Journal of Clinical Nursing, 16, 414–418 417
D de Andrade and V Ferreira
authors indicate the skin around the catheter as the most
important infection source, while others defend intraluminal
contamination, that is, the catheter connection as the most
important source (Marcondes et al. 2000).
It is possible that risk factors for infection are related to
catheter permanence time and number of insertions. More
specifically, excessive manipulation such as disconnecting and
connecting the catheter protectors during haemodialysis
should be taken into consideration. In this sense, basic asepsis
principles, hand washing with antiseptic products, use of
adequate clothing during catheter insertion and handling,
among others must be rigorously followed (Delmez & Windus
1996, Johnson 1998, Marcondes et al. 2000). Other aspects
related to infection risks include: catheter entry site, number of
hospitalizations, albumin level, HIV infection, age and diabe-
tes mellitus (Nassar & Ayus 2001, Stevenson et al. 2002).
As far as the topical use of antimicrobial agents on the
catheter entry site is concerned, professionals have a wide
range of options at their disposal, which often leads to doubts
about the best option. Furthermore, the literature reviews
present controversial results for these products’ antimicrobial
activities.
Vascular access infections are the main sources of morbid-
ity and mortality among haemodialysis patients. Studies
demonstrate that the TDLC is the main cause of between 48
and 73% of all bacteremias during haemodialysis treatment
(Castro 2001).
On the other hand, another preoccupying source of infection
is water quality. Patients under haemodialysis are more liable
to water contamination, which may lead to pyrogenic reac-
tions and/or bacteremia, in spite of modern water treatment
systems through reversed osmosis (Hoen et al. 1998).
Nassar and Ayus (2001) highlight that bacteremia in
haemodialysis patients is frequently related to vascular access
infection, which may turn into hematogenic pneumonia. The
most frequent microorganisms are Staphylococcus aure-
us,Escherichia coli, Staphylococcus epidermidis and other
gram-negative bacteria.
Conclusions
Our prospective analysis of CRI patients revealed various
worrying aspects, including catheter permanence time, which
greatly exposes them to different complications, especially
infection. It seems imperative that other research on the use
of temporary double-lumen catheters is needed to clarify
inquiries that remain unanswered, to be of help in decision-
making in controversial situations, to support the implemen-
tation of new technologies, produce knowledge and work
towards its practical applicability, with a view to increased
418  2007 Blackwell Publishing Ltd, Journal of Clinical Nursing, 16, 414–418
care efficiency and quality of life for chronic renal insuffi-
ciency patients under haemodialysis.
Contributions
Study design: DA, VF; data collection and analysis: VF; and
manuscript preparation: DA, VF.
References
Baracetti S (2001) Indication for the use of central venous catheters as
vascular access for haemodialysis. The Journal of Vascular Access
2, 20–27.
Besarab A & Raja RM (2003) Acesso vascular para hemodiálise.
In Manual de dálise (Daugirdas JT & Ing TS eds). Medsi, Rio de
Janeiro, pp. 68–102.
Canaud B (1999) Haemodialysis catheter-related infection: time for
action. Nephroloy Dialysis Transplantation 14, 2288–2290.
Castro MCM (2001) Atualização em diálise: complicações agudas
em hemodiálise. Jornal Brasileiro de Nefrologia 23, 108–113.
Delmez J & Windus D (1996) Impaired delivery of dialysis: diagnosis
and correction. American Journal of Nephrology 16, 29–34.
Dittmer ID, Sharp D, McNulty CAM, Williamsa AJ & Banks RA
(1999) A prospective study of central venous haemodialysis
catheter colonization and peripheral bacteremia. Clinical
Nephrology 51, 34–39.
Hoen B, Paul-Daupin A, Hestin D & Kessler M (1998) EPI-
BACDIAL: a multicenter prospective study of factors for bacter-
emia in chronic haemodialysis patients. Journal of the American
Society Nephrology 9, 869–876.
Johnson MS (1998) Catheter access for haemodialysis. Seminars in
Dialysis 11, 326–330.
Marcondes CR, Biojone CR, Cherri J, Moryia T & Piccinato CEB
(2000) Complicações precoces e tardias em acesso central: análise
de 66 implantes. Acta Cirúrgica Brasileira 15, 73–75.
Moysés-Neto M, Vieira-Neto OM & Costa JAC (2000) Complicações
infecciosas do acesso vascular em hemodiálise. In Moysés-Neto H
Atualidades de Nefrologia, Guanabara Koogan, São Paulo,
pp. 343–357.
Nassar GM & Ayus JC (2001) Infectious complications of the hae-
modialysis access. Kidney Internacional 60, 1–13.
NKF-DOQI Clinical Practice Guidelines For Vascular Access (1997)
Guidelines 13, 14, 15. National Kidney Foundation, New York,
pp. 44–48.
Sesso R (2000) Inquérito epidemiológico em unidades de diálise do
Brasil. Journal Brasileiro de Nefrologia 22, 23–26.
Sistema Único de Saúde (2005) Aspectos Gerais. Available at: http://
www.sespa.pa.gov.br. (accessed 5 april 2005).
Sociedade Brasileira de Nefrologia (2005) Para o público: orim e suas
funções. Available at: http://www.sbn.org.br/index14.html
(accessed 10 january 2005).
Stevenson KB, Hannah EL, Lowder CA, Adcox MJ, Davidson RL,
Mallae MC, Narasimhan N & Wagnild JP (2002) Epidemiology
of haemodialysis vascular access infections from longitudinal
infection surveillance data: predicting the impact of NKF-DOQI
Clinical Practice Guidelines for vascular access. American Journal
of Kidney Diseases 39, 549–555.