Incidence of secondary deep venous thrombosis after catheter implant for hemodialysis: evaluation by Doppler ultrasonography
(Portuguese
PDF version)
Felipe
José de Moura Vianna, Aldemar Araújo Castro, André Falcão
Pedrosa Costa, Guilherme Benjamin Brandão Pitta, Fausto
Miranda Júnior *
*
Hospital da Agroindústria do Álcool de Alagoas, Maceió,
AL, Brazil.
Correspondence:
Felipe José de Moura Vianna
Rua Dr. Abelardo Pontes Lima, 78
CEP 57052-695 - Maceió, AL, Brazil
Phone: + 55 82 241.0676
E-mail: felipejos@uol.com.br
ABSTRACT
Objective:
This study aimed at determining, by means of Doppler ultrasonography,
the incidence of deep venous thrombosis after temporary catheter
implant for hemodialysis in 60 chronic renal patients.
Method: These patients, at the onset of the replacement renal
therapy, were evaluated by Doppler ultrasonography before and after
the venous puncture. The sample was formed by consecutive individuals
with terminal chronic renal failure who needed temporary catheter
implant for hemodialysis by deep venous puncture. The presence of
internal subclavian and/or jugular vein thrombosis was the primary
variable. The secondary variables were puncture site, socioeconomic
condition, urea levels, presence of infection, time in which the
catheter remained implanted, and hematocrit.
Results: Sixty patients were studied, 31 males and 29 females,
with ages varying from 18 to 73 years. Eight presented thrombosis,
which indicates an incidence of 13% within a confidence interval
of 5% to 20%. Four cases (6.6%) were located in the right subclavian
vein, three cases (5%) in the left subclavian vein, and one case
(1.4%) in the right internal jugular vein.
Conclusion: The incidence of secondary deep venous thrombosis
after catheter implant for hemodialysis by Doppler ultrasonography
was 13%.
Key-words:
deep venous thrombosis, hemodialysis, Doppler ultrasonography.
J
Vasc Br 2005;4(2):176-82
The complications
caused by vascular accesses are not only the most frequent cause of
morbidity in hemodialysis patients, but also the major responsible for
the high cost of the treatment for the end-stage renal disease. It is
estimated that expenses with this type of patient, who presents any
complications, is approximately U$ 8,000 per patient-year.1
Several health organizations confirm the importance of the concern about
vascular access morbidity, which represents a quarter of all costs for
the patient with end-stage renal disease.2
However, deep venous thrombosis is the main complication inherent to
the prolonged use of the catheter, whose clinical symptomatology may
not appear in most cases, but can be dramatic when it is present.2
Repeated
cannulations, prolonged use, recurrent infection and trauma during the
catheter insertion are the factors that make the catheter cause thrombosis.3
The presence
of secondary deep venous thrombosis after catheter implant is reported
in the literature as predominant in the subclavian vein, with an incidence
varying from 40%4 to 50%,5
diagnosed by phlebography.
On the
other hand, this incidence is low in the internal jugular vein, ranging
from 0%4 to 10%,5
also verified by phlebography. In spite of phlebography being the gold
standard for the venous thrombosis diagnosis, a study performed in 19946
showed that it is possible to see the extension and frequency of the
thrombosis caused by the temporary catheter for hemodialysis.
Therefore,
the preventive evaluation of central vessels would lead to an early
detection of this complication, improving survival rates, access functionality,
and thus influencing in the morbidity and mortality of chronic renal
patients.7,8
Aiming
at determining the incidence of secondary deep venous thrombosis after
catheter implant for hemodialysis, the Doppler ultrasonography was performed
and evaluated according to the following methodology.
METHODS
A prospective,
descriptive, cohort study was performed. The research was carried out
at a tertiary care hospital (Hospital da Agroindústria do Açúcar e do
Álcool de Alagoas, Maceió, AL, Brazil).
Patients
with end-stage chronic renal insufficiency, submitted to a renal substitutive
program, or patients who lost permanent access and needed temporary
catheter implant (shilley double-lumen catheter, 15 cm) for hemodialysis
by deep venous puncture were included. Patients under 18 years old,
with end-stage hepatic disease and pregnant women were excluded. Patients
unable to be submitted to venous puncture of upper limb, in which it
is necessary to use the venous systems of lower limbs were not included
either, as well as patients who did not perform all the examination
times.
Patients
who presented chronic renal insufficiency when admitted to the nephrology
service of the Hospital do Açúcar to start the dialytic treatment were
consecutively recruited. After agreeing with the consent form, in accordance
to the Helsinki declaration and the Regulation 196/96 of the Ministry
of Health, patients were selected.
The presence
of internal subclavian and/or jugular vein thrombosis is the primary
variable. Venous thrombosis is defined as the total occlusion of the
vessel at a certain site, due to blood elements.
The primary
variable was evaluated using the Doppler ultrasonography (GE Diasonics,
Santa Clara, CA, 10 MHz linear transducer), by the same examiner, in
the following times: zero (before puncture), 15 days after puncture
(time 1), and immediately after the catheter removal (time 2). The side
in which the catheter would be implanted was examined, as well as the
contralateral side.
The puncture
technique was performed according to the nephrologist in subclavian
or internal jugular veins, through the Seldinger technique, always guided
by the anatomical reference points used in this type of procedure.
Secondary
variables were:
a) Puncture site of hemodialysis catheter: it was performed by
the nephrologist of the service at the Hospital do Açúcar and registered
in the data collection manual by the researcher.
b) Hematocrit: it was collected at the moment of hospitalization
for the diagnosis of renal insufficiency and registered in the data
collection manual by the researcher.
c) Urea levels: they were collected at the moment of hospitalization
for the diagnosis of renal insufficiency and registered in the data
collection manual by the researcher.
d) Socioeconomic status: it was registered in the data collection
manual, through an interview between the researcher and the patient
during hospitalization. Three groups were established: up to a current
minimum wage, from one to three minimum wages, and above three minimum
wages.
e) Mean catheter permanence: it was the time registered in the
data collection manual by the researcher, comprehending the passage
of the catheter by the nephrologist until its removal, due to the presence
of a permanent access able to be used, detection of thrombosis using
Doppler ultrasonography and/or due to infection.
f) Presence of infection: it was detected in the patient who
presented inflammatory and infectious signs and symptoms (heat, flare,
pain, edema, secretion and fever) in the presence of thrombi, assessed
by the clinical examination.
Concerning
the statistical analysis, general data were expressed in mean and standard
deviation, with a 95% confidence interval. Thrombosis was the dependent
variable, and hematocrit, mean catheter permanence, urea, socioeconomic
status, infection, and puncture site were the independent variables.
Mean analyses
were performed using ANOVA for both groups (present or absent thrombosis),
tested by Bartlett's when there was variance homogeneity, or, on the
contrary, by Kruskal-Wallis.
Categorical
variables were analyzed through the odds ratio. In case there were cells
with values lower than 5, Fisher's test would be applied.
RESULTS
A total
of 60 patients with chronic renal insufficiency who implanted a temporary
catheter for hemodialysis were studied, including 31 male patients and
29 female, with ages varying from 18-73 yeas (mean of 46 years).
The etiologic
characteristic of the sample was 29% (18 patients) with systemic arterial
hypertension, 28% (17 patients) with chronic glomerulonephritis, 22%
(12 patients) with diabetes mellitus, 7% (four patients) with systemic
lupus erythematosus, 5% (three patients) with renal cyst, 3% (two patients)
with Alport syndrome, 3% (two patients) with undetermined cause, and
3% (two patients) with other causes (idiopathic thrombocytopenic purpura,
tuberculosis).
Thrombosis
was found in eight patients, presenting a 13% incidence (CI 95% = 4
to 21) (Figure 1).
 Figure
1 - Incidence of deep venous thrombosis using Doppler ultrasonography.
Four cases
(6.6%) were located in the right subclavian vein, three cases (5%) in
the left subclavian vein, and one case (1.4%) in the right internal
jugular vein (Figure 2).
Figure
2 - Puncture sites: with and without thrombosis.
In general,
patients without thrombosis presented the following distribution of
puncture sites for the temporary catheter: 61.8% in the right subclavian
vein (37 patients), 11.8% in the left subclavian vein (seven patients),
11.8% in the right internal jugular vein (seven patients), and 1.6%
in the left internal jugular vein (one patient) (Figure 2).
It is important
to highlight that the thrombosis in time 0 was detected in only one
patient. Since this thrombosis was located in the left internal jugular
vein and the site punctured with the temporary catheter was the right
subclavian vein, which did not develop thrombosis, this patient was
included in the group without thrombosis. In all other patients, in
which Doppler was performed previously to the catheter implant (time
0), the venous thrombosis was not detected in the Doppler ultrasonography.
Concerning
the hematocrit mean, results were as follows: patients with thrombosis
had a mean hematocrit of 28.7% (SD = 8.3; CI 95% = 23.3 to 34.1), and
without thrombosis, of 26% (SD = 6; CI 95% = 24.3 to 27.6) (Figure 3).
Figure
3 - Mean levels of hematocrit in patients who presented and did
not present thrombosis.
Regarding
the urea levels presented by the patients at the onset of the dialytic
treatment, the group with thrombosis presented a mean of 172 mg/dl (SD
= 44.9; CI 95% = 141 to 203), and in the group without thrombosis, the
mean was 183 mg/dl (SD = 69.9; CI 95% = 164 to 202) (Figure 4).
Figure
4 - Mean levels of urea in patients with and without thrombosis.
As to the
socioeconomic status, there was the following distribution: up to a
minimum wage in the group with thrombosis (6.6%) and in the group without
thrombosis (43.4%), from one to three minimum wages in the group with
thrombosis (3.2%) and in the group without thrombosis (25%), above three
minimum wages in the group with thrombosis (3.2%) and in the group without
thrombosis (18.6%).
Mean catheter
permanence was 33 days (SD = 11), 37.5 days (SD = 14.1) for the group
with thrombosis (CI 95% = 27.7 to 47.3), and 32.5 days (SD = 10.6) for
the group without thrombosis (CI 95% = 29.6 to 35.3) (Figure 5).
Figure
5 - Mean catheter permanence in patients with and without thrombosis.
Regarding
cases of thrombosis associated to infection, six cases were found in
which there was the presence of signs and symptoms of bacterial contamination
by the temporary catheter, located on the same side of the venous obstruction
(Figure 6).
Figure
6 - Catheters that provoked thrombosis in the presence of infection.
After a
statistical analysis, there were no significant correlations between
the presence of thrombosis and the secondary variables.
Any predictive
factors of thrombosis were observed among these variables.
DISCUSSION
The Doppler
ultrasonography to verify the prevalence of secondary deep venous thrombosis
after the catheter implant for hemodialysis was used for being a non-invasive
method and, therefore, presenting no risks for patients and possible
to be routinely used. Moreover, the sensitivity of the technique has
proven to be efficient for detecting thrombotic phenomena.9
A study
performed by Fischer et al.6 reported a
good method capacity for detecting present catheters in the internal
jugular and subclavian veins. A critical analysis concerning the thrombosis
detection of deep venous catheters is that the phenomenon may occur
in sites where the subclavian vein cannot be seen by the method.
In a study
performed by Brady et al.,10 50 patients
who had been submitted to a temporary catheter implant for hemodialysis
were consecutively examined, using Doppler ultrasonography and signs
of deep venous thrombosis of upper limbs were described, confirmed by
phlebography and compatible with signs and clinical symptoms.10
Despite
the lack of studies in the literature showing the use of the Doppler
ultrasonography for diagnosing secondary venous thrombosis of upper
limbs after catheter implant for hemodialysis, a research published
in 199011 by Knudson et al. reported a sensitivity
of around 78% and specificity of 92% of the Doppler ultrasonography
in the deep venous thrombosis of upper limbs, without determining which
etiologic agent was involved.
In this
study by Knudson et al.,11 the assessment
using Doppler ultrasonography detected 30% of deep venous thrombosis.
The difference in the result of deep venous thrombosis incidence, in
relation to our study, can be attributed to several factors. Among them,
the different catheters and their therapeutical objectives and the fact
that the Doppler ultrasonography is dependent on the device and examiner.
Deep venous
thrombosis associated with catheter for hemodialysis is a complication
that does not, in most cases, present clinical signs and symptoms, whose
incidence is quite variable, being described as around 3.7%12
when the assessed method is exclusively the clinical evaluation. Such
data do not match the findings of phlebography,13
in a study performed in 1981, and necropsy,14
in a study performed in 1982, which indicate a percentage result higher
than the one described by clinical findings.
Thrombosis
of the subclavian and internal jugular veins is a major problem for
the patient with end-stage chronic renal insufficiency. Its presence
negatively alters survival, since a patent access with a good function
provides a good renal substitution therapy. It is also often needed
to close an arteriovenous fistula with good function, due to limitations
imposed by signs and symptoms of the venous obstruction not previously
detected.
Since the
venous thrombosis of subclavian and internal jugular veins is a major
limiting factor of the chronic renal patient's survival, specialists,
nephrologists and vascular surgeons show great concern about diagnostic
means that are able to early identify such alterations.
Phlebography
is considered the gold standard for diagnosing the deep venous thrombosis
and is recommended by some authors7 for
exclusion of thrombosis before creating an arteriovenous fistula, mainly
on the same side in which a temporary catheter for hemodialysis had
been previously implanted. Its routine use, however, is arguable, considering
the risks inherent to this method.
Our study
recruited 60 patients consecutively, similar to the study performed
by Hernandez et al. in 1998,15 in which
42 patients were assessed using phlebography in order to verify the
presence of secondary deep venous thrombosis after catheter implant
for hemodialysis.
In a study
performed by Wanscher et al.,7 53 patients
were examined using phlebography after the removal of the catheter for
hemodialysis, aiming at verifying the presence of deep venous thrombosis
of the subclavian vein and superior vena cava. The authors did not perform
the examination before the catheter was implanted and, therefore, could
not avoid possible thrombotic phenomena not directly related to the
presence of the venous catheterization.
In this
study, we verified the presence of deep venous thrombosis in 13%: four
cases (6.6%) were located in the right subclavian vein, three cases
(5%) in the left subclavian vein, and one case (1.4%) in the right internal
jugular vein. In 1986,12 Vanherweghem et
al., in a similar study, but using phlebography, studied 42 asymptomatic
patients who had been submitted to a subclavian vein cannulation, and
found a percentage of 19% with thrombosis and 14% with minimal defect
in the venous lumen. Once there were no patients with internal jugular
vein puncture, it is not possible to compare the predominance of thrombosis
of one vessel in relation to another. Nevertheless, the incidence found
in our study, using a different diagnostic method, is very similar to
the study performed in 1986.12
In a study
performed by Schillinger et al.,4 using
phlebography, 100 patients were assessed in a 4-year period, observing
that the prevalence of stenosis was around 26% (5% in the internal jugular
vein and 21% in the subclavian vein). Concerning the severity of the
stenosis, 15% presented severe stenosis, and the total obstruction was
detected in 2% of cases, all present in the subclavian vein. Methodologically,
we cannot compare findings in the angiography and in the ultrasonography.
In the angiography, the objectivity of the analysis is a result of the
undisputable photographic data; the ultrasonography is an examiner-dependent
technique that is often unable to accurately detect stenosis when they
are located in the deep veins of upper limbs. Moreover, the detection
of thrombosis in the ultrasonography can be mistaken by the presence
of subjacent severe stenosis.
In one
of the patients of the study, a deep venous thrombosis of the right
jugular vein in time 0 was detected, despite being a case in which the
patient had been undergoing hemodialysis and there had been loss of
the permanent access. The patient was treated as if belonging to the
group without thrombosis, since the puncture for the study was performed
in the right subclavian artery, which did not present venous thrombosis
in all the times of the examination.
Of the
eight cases of deep thrombosis found in our study, six patients presented
an association with catheter infection, in 1993 and 1998. Hernandez
et al. found 75%15 and 66%16,
respectively, of catheter infection in association with venous thrombosis,
data which are very similar to ours.
The predominant
site for venous thrombosis was, in our study, in the subclavian vein,
similar to what is reported in the literature, in the studies performed
by Cimochowski et al.5 and Schillinger et
al..4 For these authors, the close anatomical
relation of the subclavian vein to bone structures and the proximity
of the endothelial lesion, caused by the catheter, to the heart were
the probable causes for the prevalence of thrombosis in the jugular
vein.
In general,
this study shows that the possibility of using the Doppler ultrasonography
before creating an arteriovenous fistula is one more diagnostic method
to guide the specialist in the treatment of the patient with chronic
renal insufficiency. Patients who will benefit the most are those who
present a suggestive history of venous thrombosis, that is, those who
had the puncture of the temporary catheter in the subclavian or jugular
vein on the side where it is intended to create a permanent access.
Thus, the
Doppler ultrasonography would reduce the financial cost for these patients,
and would be mainly contributing to a lower morbity and mortality rate
of the chronic renal patient.
In conclusion,
we can affirm that the incidence of secondary deep venous thrombosis
after catheter implant for hemodialysis using Doppler ultrasonography
was 13%.
ACKNOWLEDGEMENT
To the
professors of Vascular Surgery at the Department of Surgery at Escola
Paulista de Medicina da Universidade Federal de São Paulo (SP),
Drs. Newton de Barros Junior, Jorge Eduardo Amorim, Maria Del Carmem
Janeiro Perez, Wellington Gianotti Lustre, José Carlos Costa
Baptista Silva, Luis Francisco Poli de Figueiredo, João Francisco
Junior, for the knowledge transmitted on Vascular Surgery, and to Nildo
Batista, for the teachings on didactics.
To Professor
Laís Záu Serpa de Araújo, Head of Bioethics at
Fundação Universitária de Ciência da Saúde
de Alagoas Governador Lamenha Filho/Escola de Ciências Médicas
de Alagoas, Maceió (AL, Brazil), for the praiseworthy task of
sharing the ethical knowledge for the research.
To the
radiologist Sirlene Maria de Lima Oliveira Mota, specialist in imaging
diagnosis, for her great collaboration and availability to perform the
Doppler ultrasonography examinations, only aiming at the scientific
contribution and the sense of brotherhood.
To the
nephrology and radiology service assistants, Hilza, Maria José
and Maria de Jesus, for their collaboration and effort in order to achieve
the desired sample.
REFERENCES
1.
U. S. government sponsored work. Renal Data System: The cost effectiveness
of alternative types of vascular access and the economic cost of ESRD.
Am J Kidney Dis 1995;26:S140-56.
2. Schwab SJ. Vascular access for hemodialysis. Kidney
Int 1999;55:2078-90.
3. Jassal SV, Pierratos A, Roscoe JM. Venous stenosis
and thrombosis associated with the use of internal jugular vein catheters
for hemodialysis. ASAIO J 1999;45:356-9.
4. Schillinger F, Schillinger D, Montagnac R, Milcent
T. Post catheterisation vein stenosis in hemodialysis: comparative angiographic
study of 50 subclavian and 50 internal jugular accesses. Nephrol Dial
Transplant 1991;6:722-4.
5. Cimochowski GE, Worley E, Rutherford WE, Sartain
J, Blondin J, Harter H. Superiority of the internal jugular over the
subclavian access for temporary dialysis. Nephron 1990;54:154-61.
6.
Fischer M, Krünes U, Thieme T. Thrombosis of central veins induced
by catheters, Doppler-Duplex ultrasound examinations of frequency and
dimensions. Ultraschall in Med 1994;15:304-7.
7.
Wanscher M, Frifelt JJ, Smith-Sivertsen C, et al. Thrombosis caused
by polyurethane double-lumen subclavian superior vena cava catheter
and hemodialysis. Crit Care Med 1988;16:624-8.
8. Surrat RS, Picus D, Hicks ME, Darcy MD, Kleinhoffer
M, Jendrisak M. The importance of preoperative evaluation of the subclavian
vein in dialysis access planning. Am J Radiol 1991;156:623-5.
9.
Comerota AJ, Katz ML, Greenwald LL, Leefmans E, Czeredarcczuk M, White
JVJ. Venous duplex imaging: should it replace hemodynamic tests for
deep vein thrombosis? J Vasc Surg 1990;11:53-9.
10. Brady HR, Fitzcharles B, Goldberg H, et al. Diagnosis
and management of subclavian vein thrombosis occurring in association
with subclavian cannulation for hemodialysis. Blood Purif 1989;7:210-7.
11. Knudson GJ, Wiedmeyer DA, Erickson SJ, et al. Color
Doppler sonographic in the assessment of upper-extremity deep venous
thrombosis. AJR 1990;154:399-403.
12. Vanherweghem JL, Yassine T, Goldman M, et al. Subclavian
vein thrombosis: a frequent complication of subclavian vein cannulation
for hemodialysis. Clin Nephrol 1986;26:235-8.
13. Brismar BO, Hardstedt C, Jacobson S. Diagnosis
of thrombosis by catheter phlebography after prolonged central venous
catheterization. Ann Surg 1981;194:779-83.
14. Vanholder R, Lameire N, Verbanck J, Rattinghe R,
Kunnen M, Ringoir S. Complications of subclavian catheter hemodialysis:
a 5-year prospective study in 257 consecutive patients. Int J Artif
Organs 1982;5:297-303.
15.
Hernandez D, Diaz F, Suria S, et al. Subclavian catheter-related infection
is a major risk factor for the late development of subclavian vein stenosis.
Nephrol Dial Transplant 1993;8:227-30.
16. Hernandez D, Diaz F, Rufino M, et al. Subclavian
vascular access stenosis in dialysis patients: natural history and risk
factors. J Am Soc Nephron 1998;9:1507-10.
|
