British Journal of Radiology (2005) 78, 596-600
© 2005 British Institute of Radiology
doi: 10.1259/bjr/32639616
Incidence of upper limb venous thrombosis associated with peripherally inserted central catheters (PICC)
B J J Abdullah, FRCR
1
N Mohammad, MBBS
1
J V Sangkar, MBBS, MRCP (UK)
1
Y F Abd Aziz, MRad
1
G G Gan, MRCP (UK)
1
K Y Goh, MRCP
1 and
I Benedict, MBBS
2
1 Departments of Radiology and Medicine, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur Malaysia, and 2 Sanofi-Synthelabo (Malaysia)
 |
Abstract
|
|---|
The objective of this study was to prospectively determine the incidence of venous thrombosis (VT) in the upper limbs in patients with peripherally inserted central catheters (PICC). We prospectively investigated the incidence of VT in the upper limbs of 26 patients who had PICC inserted. The inclusion criteria were all patients who had a PICC inserted, whilst the exclusion criterion was the inability to perform a venogram (allergies, previous contrast medium reaction and inability of gaining venous access). Both valved and non-valved catheters were evaluated. Prior to removal of the PICC, an upper limb venogram was performed. The number of segments involved with VT were determined. The duration of central venous catheterization was classified as; less than 6 days, between 6 days and 14 days and more than 14 days. VT was confirmed in 38.5% (10/26) of the patients. The majority 85.7% (12/14) were complete occlusive thrombi and the majority of VT only involved one segment. There was no statistical correlation between the site of insertion of the PICC and the location of VT. Neither was there any observed correlation between the occurrence of VT with the patient's history of hypertension, hypercholesterolaemia, coronary artery disease, diabetes mellitus, cardiac insufficiency, smoking or cancer. There was also no statistical correlation with the size of the catheter. In conclusion, PICCs are associated with a significant risk of upper extremity deep vein thrombosis (UEVT).
|
Introduction
|
|---|
The establishment of reliable, temporary central venous access is essential for an increasing number of patients who require vesicant drugs or solutions, long-term antibiotics, or chemotherapeutic or other agents. In 1992, it was estimated that 3 million central venous catheters were placed [1] and by 1996, that number was estimated at 5 million [2]. Peripherally inserted central catheters (PICCs) are becoming popular in providing such an access. PICCs have also been proved a safe and cost-effective in providing central venous access in both the inpatient and outpatient settings [3–9]. The incidence of pulmonary embolism secondary to upper extremity deep vein thrombosis (UEVT) in patients with central venous catheters has been reported to be greater than 35% [10, 11] and as high as that of lower extremity vein thrombosis (VT). In addition the risk of recurrent venous thrombosis and of late sequelae is again comparable with that of lower extremity deep vein thrombosis (DVT) [10, 12].
|
Patients and methods
|
|---|
A prospective study on the incidence of DVT in the upper limbs of 26 patients who had PICC inserted at University of Malaya Medical Centre was conducted. Ethical clearance was obtained from the hospital ethics committee prior to conducting the study. Informed consent was obtained from the patients prior to involvement in the study. The male to female ratio was 1.9:1 with the age ranging from 15 years to 70 years with an even distribution across the age groups. The majority of patients were Chinese, 46%, while Malays and Indians each made up 27%. The majority of PICCs were inserted for long term antibiotic therapy (65%) with the rest for chemotherapy. In the subgroups for chemotherapy, there were nine patients where acute myeloid leukaemia (AML) constituted the largest single group with 67%, acute lymphoblastic leukaemia (ALL) making up 22% and the remainder ALL combined with AML. None of these patients were on prophylactic anticoagulation with low dose heparin.
Inclusion criteria were all patients who had a PICC line inserted and the exclusion criteria were inability to perform a venogram (allergies, previous contrast medium reaction and inability of gaining venous access), previous history of lower limb DVT, patients who refused consent and those who were not haemodynamically stable enough to have venograms performed. Both valved and non-valved catheters were evaluated.
Prior to removal of the PICC an upper limb venogram was performed. The venogram was performed by two experienced radiologists (BJJA, NM) using digital subtraction angiography (DSA) (Advantx DLX; GE, Milwaukee, WI) with no magnification (unless it was the area of interest). A tourniquet placed over the proximal arm. The patients were examined in the supine position with the involved upper limb in extension and immobilized through supination and abduction of 45°. Venous access was gained by inserting either a butterfly needle or a 20 gauge cannula. Three successive sequences were performed to include the forearm, arm and thorax. Patients asked to perform a Valsalva manoeuvre to decrease the venous runoff in the central veins. Approximately 20–40 ml of 300 mg ml–1 non-ionic iodinated contrast (Ultravist 300, Schering, Berlin) was injected via the cannula placed in the dorsum of the hand.
The studies were in the first instance assessed for completeness. Diagnostically adequate studies were classified as complete opacification of the entire venous system including the cephalic, basilic, median cubital, axillary, subclavian and brachiocephalic veins as well as the superior vena cava on multiple views with adequate penetration. The presence of non-opacification/non-filling and thrombus and number of segments involved was determined. The presence of risk factors was also noted. The images were viewed by two radiologists (NM/BJJA) and results agreed by consensus.
The type of catheter inserted was determined. The duration of central venous catheterization was classified as: less than 6 days, between 6 days and 14 days and more than 14 days.
|
Data analysis
|
|---|
A Pearson Chi-squared test (SPSS Ver. 10; SPSS Inc., Chicago, IL) was used to determine if there was any statistical correlation between the incidence of UEVT and location of VT, hypertension, hypercholesterolaemia, coronary artery disease (CAD), diabetes mellitus (DM), cardiac insufficiency, smoking, cancer or the size of the catheter inserted.
|
Results
|
|---|
The PICC lines were inserted on the right 65.3% (17/26) and on the left in 34.7% (9/26). With regards to the sites of placement, the right basilic vein 42.3% (11/26) was the most common site followed by the right cephalic and left median cubital veins being used in 15.3% (Table 1
). All the tips were placed in the proximal to mid superior vena cava (SVC). Majority of the PICC were in situ for more than 14 days (96.2%). Only one was valved (Groshong; Bard Access Systems, Salt Lake City, UT) while 25 were non-valved (Vaxcel; Boston Scientific, Natick MA and PICC Cook, Bloomington, IN). 14 of these were single lumen in nature. As regards the diameter, there were equal numbers of 4 F and 5 F diameter catheters used.
View this table:
[in this window]
[in a new window]
|
Table 1. Site of insertion of peripherally inserted central catheters (PICC). The majority of PICC were inserted in the right basilic vein
|
|
All 26 patients had diagnostically adequate studies. VT was confirmed in 38.5% (10/26) of the patients involving 14 segments. The majority 85.7% (12/14) were complete occlusive thrombi (Figures 1 and 2) and involving one segment (with (169/182) segments of the upper limb veins being normal). The sites of VT are as shown in Table 2. There was no statistical correlation between site of insertion of PICC and the location of VT (p=0.12). Neither were there any observed relationships between the occurrence of VT with patient's history of hypertension (p=0.50), hypercholesterolaemia (p=0.55), coronary artery disease (CAD) (p=0.26), diabetes mellitus (DM) (p=0.21), cardiac insufficiency (p=0.21), smoking (p=0.47) or cancer (p=0.90) (Table 3). There was also no statistical correlation with the diameter (p=0.72), or manufacturer (p=0.64) of the catheter inserted though 5 F catheters had a higher odds ratio (OR=1.33) of developing UEVT. Only one patient was symptomatic.
View larger version (78K):
[in this window]
[in a new window]
|
Figure 1. Right upper limb venogram. Completely occluding thrombus seen in the origin of the axillary vein (arrow) with no opacification of the brachial vein along the peripherally inserted central catheters.
|
|
View larger version (95K):
[in this window]
[in a new window]
|
Figure 2. Right upper limb venogram. There is opacification of the brachial (big arrows) and axillary veins while there is no opacification of the cephalic vein (small arrows).
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Sites of upper limb venous thrombosis. Even though there were only 10 patients with deep venous thrombosis, there were three patients with two or more sites of involvement, i.e. 14 segments in 10 patients
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Demographic data, underlying disease, type and location of peripherally inserted central catheters (PICC) as well as venographic findings
|
|
|
Safety
|
|---|
The patients were followed up 3 months after completion of the venography. There were no related adverse reactions or complications. There were two deaths within 3 months of the venography, these were not related to pulmonary embolism but secondary to sepsis.
|
Discussion
|
|---|
UEVT is estimated to constitute 1–4% of all cases of VT [13]. The reported rates of catheter-related DVT range from 2% to 42% [14–18] and are higher than the reported incidence rates of mechanical or septic complications [19, 20]. Despite this wide range of incidence, the actual incidence of catheter-related DVT is probably underestimated partly due to the indolent and frequently non-occlusive nature of the thrombi [10]. A 9% rate of asymptomatic catheter-related VT has been reported in cancer patients [21], and in our study only one of the patients was symptomatic, the lack of symptom probably being due to the low thrombus load.
UEVT complicating central venous catheterization is becoming more prevalent as catheters extending into the subclavian vein are increasingly used to facilitate the delivery of anti-neoplastic chemotherapy, total parenteral nutrition and prolonged parenteral antibiotic therapy as well as for haemodynamic monitoring and resuscitation. These catheters are usually placed via a percutaneous Seldinger method using the subclavian or internal jugular vein or a cut-down over the cephalic, external or internal jugular veins. Alternatively a PICC inserted into the basilic or cephalic vein can be used as a central venous catheter. Due to the easier insertion procedure and lesser morbidity, PICCs are replacing other types of central venous catheters for most indications.
The likelihood of developing catheter-related UEVT is affected by mechanical factors including the number of punctures during catheter insertion, the number of inserted and changed catheters, the location of catheter tip, the duration of catheterization, the material and diameter of catheter used, the type of fluid administered, catheter-related infection, hypercoagulable states and the presence of congestive heart failure. There are numerous other possible risk factors for catheter-related VT, and the cumulative effect makes it difficult to assess the individual contribution of each of these factors. Moreover, most of the previously reported studies that focused on catheter-related VT were based either on clinically symptomatic thromboses or on phlebography.
Catheter tips in the axillo-subclavian and innominate veins are associated with a higher incidence of UEVT (60%) compared with patients with tips in the SVC (21%) [22]. In another study [23], abnormalities were seen in 58.6% of patients despite being on prophylaxis; fibrin sleeves in 47% of patients; non-obstructive clots adherent to the vessel wall and/or the catheter in 8.3% and complete axillary vein thrombosis in 3.3%. No thrombosis was observed in patients in whom catheterization was limited to 6 days or less, two cases occurred among patients catheterized for 7–14 days and 5 cases were identified when catheterization was extended beyond 14 days. In our study since almost all patients had PICC lines for more than 14 days we are unable to ascertain if the duration of insertion is important. Similar to the findings of our study only one of these patient was symptomatic.
With regards to catheter material, clinical and in vitro studies have demonstrated that both polyurethane and silicone catheters are associated with a lower rate of catheter-related DVT [22, 25], as compared with polyethylene or Teflon-coated catheters. It has been reported [26] that an external catheter diameter of less than 2.8 mm is associated with a lower rate of catheter-related VT, while incorrect placement of a CVC in the superior vena cava resulted in a higher incidence of catheter-related VT. In our study we found that none of these factors showed any statistical correlation with UEVT. We are uncertain as to the exact reasons for this difference.
Three different types of thrombi can occur in association with central venous catheterization. The most common is the "fibrin sleeve", which starts at the point of entry of the catheter into the vein and extends toward the catheter tip, with the length proportional to the duration of catheterization. Catheter adherent fibrin sleeves are asymptomatic, but fragments can detach when the catheter is removed and therefore embolise to the lungs. Additional types of thrombi include the non-occlusive mural thrombus and a thrombus causing complete occlusion of the vein. Our study showed that the majority (85.7%) were completely occlusive in nature.
Contrast venography remains the gold standard for the diagnosis of UEVT though the invasiveness, cost and contrast medium load are disadvantages. Even though it has been suggested that venography itself could be responsible for some VT [27], we feel that with the newer non-ionic contrast medium this is much less likely, would take some time to develop and would not occur at the time of the study itself. None of the patients reported any complications after the venogram on the follow-up visits. Ultrasound has been extensively evaluated in lower limb VT, its role in the upper limb has been much less with only a few investigators using prospective Doppler ultrasound (US) screening for catheter-related VT. The reported sensitivities and specificity of colour flow Doppler are 100% and 93%, respectively [10].
The incidence of pulmonary embolism secondary to UEVT in patients with central venous catheters as previously stated can be greater than 35% [10, 11] with PE being asymptomatic in a substantial proportion of patients [10]. High mortalities of up to 25% have been reported [28] and this may occur despite adequate anticoagulation [29]. We did not look for PE in our study and there were no clinically symptomatic cases. The post-thrombotic syndrome in the UEVT [13], as in the lower limb is characterized by pain, swelling and limitation of activity in the affected arm secondary to residual venous obstruction may be as high as 36–50% [10, 30]. Other complications from catheter-UEVT include a thrombophlebitis which can then evolve into systemic sepsis. It is estimated that thrombosis is related with a 2.6 fold increase in the risk of catheter related sepsis [31] with even the presence of a fibrin sleeve raising this risk. Venous gangrene and superior vena cava syndrome secondary to UEVT is extremely rare. The risk of recurrent venous thrombosis and of late sequelae is again comparable with that of lower extremity VT [10, 12].
With regard to the role of anti-coagulation in reducing the risk of UEVT, a randomized prospective study in patients with indwelling central venous catheters with 1 mg of warfarin sodium administered daily for 90 days beginning 3 days prior to catheter insertion in the treatment group, reported an incidence of UEVT of 9.5% in those patients receiving warfarin compared with 35.7% in the control group [32]. In another study in cancer patients with subclavian venous catheters using fixed doses of low molecular weight heparin, the incidence of UEVT was again significantly lower in those given prophylaxis, 6%, compared with 62% without prophylaxis [33]. These two studies suggest that prophylaxis with very low doses of anti-coagulants does protect against thrombosis in patients with central venous catheters. In our study even though we found the prevalence of VT to be 38.5%, all except one patient were asymptomatic, and there was no evidence of PE. There have been no long term studies looking at the delayed complications of venous thrombosis, e.g. a post-thrombotic syndrome. It would therefore be important to explore the role of prophylaxis in patients with PICCs and determine if it does reduce the risk of UEVT and long term complications.
In conclusion the insertion of PICCs is associated with a significant risk of UEVT and from the evidence it should be considered a serious thrombotic disorder with an incidence and morbidity similar to that of lower limb VT. It is therefore imperative that the role of anticoagulants in preventing theses complications be explored.
|
Footnotes
|
|---|
This study was funded by a research grant from Sanofi-sintalabo. 
Received for publication October 25, 2004.
Revision received January 2, 2005.
Accepted for publication January 26, 2005.
|
References
|
|---|
- Raad I, Bodey G. Infectious complications of indwelling vascular catheters. Clin Infect Dis 1992;15:197–210.[Medline]
- Herbst S. Options for venous access in ambulatory care: issues in selection and management. J Infusion Chemother 1996;6:186–94.
- Baranowski L. Central venous access devices: current technologies, uses and management strategies. J Intravenous Nurs 1993;16:167–94.[Medline]
- Goodwin ML, Carlson I. The peripherally inserted central catheter: a retrospective look at three years of insertions. J Intravenous Nurs 1993;16:92–103.[Medline]
- Richardson D, Bruso P. Vascular access devices: management of common complications. J Intravenous Nurs 1993;6:44–9.
- Cardella JF, Fox PS, Lawler JB. Interventional radiologic placement of peripherally inserted central catheters. J Vasc Intervent Radiol 1993;4:653–60.[Medline]
- Gabriel J. Care and management of peripherally inserted central catheters. Br J Nurs 1996;5:594–9.[Medline]
- Andrews JC, Marx MV, Williams DM. The upper arm approach for placement of peripherally inserted central catheters for protracted venous access. AJR Am J Roentgenol 1992;158:427–9.[Free Full Text]
- Miller KD, Dietrick CL. Experience with PICC at a university medical center. J Intravenous Nurs 1997;20:141–7.[Medline]
- Pradoni P, Polistena P, Benardi E, et al. Upper-extremity deep vein thrombosis: risk factors, diagnosis, and complications. Arch Intern Med 1997;157:57–62.[Abstract/Free Full Text]
- Moser KM, Fedullo PF, Little John JK, et al. Frequent asymptomatic pulmonary embolism in patients with deep vein thrombosis. JAMA 1994;271:223–5.[Abstract/Free Full Text]
- Khamashta MA, Cuadrado MJ, Mujic F, et al. The management of thrombosis in the anti-phospholipid –antibody syndrome. N Engl J Med 1995;332:993–7.[Abstract/Free Full Text]
- Hill SL, Berry RE. Subclavian vein thrombosis: a continuing challenge. Surgery 1990;108:1–9.[Medline]
- Brismar B, Hardstedt C, Jacobson S. Diagnosis of thrombosis by catheter phlebography after prolonged central venous catheterization. Ann Surg 1981;194:779–83.[Medline]
- Brismar B, Nystrom B. Thrombophlebitis and septicemia: complications related to intravascular devices and their prophylaxis—a review. Acta Chir Scand Suppl 1986;530:73–7.[Medline]
- Padberg FT Jr, Ruggiero J, Blackburn GL, Bistrian BR. Central venous catheterization for parenteral nutrition. Ann Surg 1981;193:264–70.[Medline]
- Beheshti MV, Protzer WR, Tomlinson TL, Martinek E, Baatz LA, Collins MS. Long-term results of radiologic placement of a central vein access device. AJR Am J Roentgenol 1998;170:731–4.[Abstract/Free Full Text]
- Chemaly RF, et al. Venous thrombosis associated with peripherally inserted central catheters: a retrospective analysis of the Cleveland clinic experience. Clin Infect Dis 2002;34:1179–83.[CrossRef][Medline]
- Paoli C, Bassot V, el-Bez M, et al. Cervicofacial cancerology: complications of implantable systems of venous access—prospective study. Presse Med 1994;23:649–52 [In French].
- Douard MC, Ardoin C, Payri L, Tarot JP. Infectious complications of long term intravenous devices: incidence, risk factors, diagnostic tools. Pathol Biol 1999;47:288–91.[Medline]
- Bonfils P, Luciani JA, Potard G, Bassot V. Prospective study of thromboses of the subclavian vein after setting implantable venous access systems in cervico-facial cancerology. Ann Otolaryngol Chir Cervicofac 1996;113:425–9.[Medline]
- Kearns PJ, Coleman S, Wehner JH. Complications of long arm catheters: a randomized trial of central versus peripheral tip location. J Parenter Entral Nutr 1996;20:20–4.
- Martin C, Viviand X, Saux P, et al. Upper extremity deep vein thrombosis after central venous catheterisation via axillary vein. Crit Care Med 1999;27:2626–9.[CrossRef][Medline]
- Borow M, Crowley JG. Evaluation of central venous catheter thrombogenicity. Acta Anaesthesiol Scand Suppl 1985;81:59–64.[Medline]
- Boswald M, Lugauer S, Bechert T, Greil J, Regenfus A, Guggenbichler JP. Thrombogenicity testing of central venous catheters in vitro. Infection 1999;27(Suppl 1):S30–S33.
- Lokich JJ, Becker B. Subclavian vein thrombosis in patients treated with infusion chemotherapy for advanced malignancy. Cancer 1983;52:1586–9.[CrossRef][Medline]
- Cronan JJ. Contemporary venous imaging. Cardiovasc Intervent Radiol 1991;14:87–97.[Medline]
- Kerr TM, Lutter KS, Moeller DM, et al. Upper extremity venous thrombosis diagnosed by duplex scanning. Am J Surg 1990;160:202–6.[CrossRef][Medline]
- Monreal M, Raventos A, Lerma R, et al. Pulmonary embolism in patients with upper extemity DVT associated to venous catheter lines – a propsective study. Thromb Haemost 1994;72:548–50.[Medline]
- Leebeek FW, Kappers-Klunne MC, Gomez-Garcia EB. Deep vein thrombosis of the arm: etiology, diagnosis and treatment. Ned Tijdschr Geneeskd 2000;144:361–4.[Medline]
- Timsit JF, Farkas JC, Boyer JM, et al. central vein catheter related thrombosis in intensive care patients. Incidence, risk factors, and relationship with catheter related sepsis. Chest 1998;114:207–13.[Abstract/Free Full Text]
- Bern MM, Lokich JJ, Wallach SR, et al. Very low doses of warfarin can prevent thrombosis in central venous catheters. A randomized prospective trial. Ann Intern Med 1990;112:423–8.
- Monreal M. Upper extremity deep venous thrombosis in cancer patients with venous access devices – prophylaxis with a low molecular weight heparin (Fragmin). Thromb Haemost 1996;75:251–3.[Medline]
Top
Abstract
Introduction
