Version May 2024
INTRODUCTION — Intravenous catheters cause endothelial trauma and inflammation and are often placed in patients who are hypercoagulable, leading to venous thrombosis. The majority (70 to 80 percent) of thrombotic events occurring in the superficial and deep veins of the upper extremity are due to intravenous catheters. The remainder are due to mechanical compression from anatomic abnormalities (eg, venous thoracic outlet syndrome) [1-3].
While superficial vein thrombosis and phlebitis related to peripheral intravenous catheters are generally self-limited once the catheter is removed, these patients can develop deep vein thrombosis (DVT) with its attendant risk for embolism. Thrombosis involving the deep veins of the upper extremity (axillary, brachial) or thoracic central veins (ie, subclavian, brachiocephalic, superior vena cava, intrathoracic internal jugular vein) can lead to pulmonary embolism and long-term sequelae in spite of timely diagnosis and treatment [4,5]. Pulmonary embolism from upper extremity sources accounts for approximately 6 percent of cases [6-8].
Upper extremity venous thrombosis (superficial and deep) as a complication of indwelling venous catheters is reviewed. Thrombosis involving the lower extremity veins is discussed elsewhere. (See "Superficial vein thrombosis and phlebitis of the lower extremity veins" and "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)
General issues regarding DVT, primary causes of upper extremity DVT, and thrombosis related to hemodialysis access are also discussed elsewhere. (See "Overview of the causes of venous thrombosis" and "Primary (spontaneous) upper extremity deep vein thrombosis" and "Central venous catheters for acute and chronic hemodialysis access and their management".)
Other complications of central venous catheters and the approach to catheter selection, including in those patients who have experienced a prior catheter thrombosis, are reviewed separately. (See "Central venous catheters: Overview of complications and prevention in adults" and "Central venous access: Device and site selection in adults".)
ANATOMIC CONSIDERATIONS — Venous catheters can be placed into the superficial or deep veins of the upper extremity, and for central catheters, centrally or peripherally. Detailed anatomy is described separately. (See "Placement of jugular venous catheters", section on 'Jugular vein anatomy' and "Placement of subclavian venous catheters", section on 'Subclavian vein anatomy'.)
●Superficial — The main superficial veins of the upper extremity include the cephalic, basilic, median antebrachial, median antecubital, and accessory cephalic veins (figure 1 and figure 2) [9]. There is considerable anatomic variation. Peripheral intravenous catheters are typically placed into the superficial veins of the forearm or hand, but the external jugular vein is a common access site during anesthesia. The basilic vein is one of the most common access sites used for placement of peripherally inserted central catheters
●Deep — The deep veins of the upper extremity include the paired ulnar, radial, and interosseous veins in the forearm; paired brachial veins of the upper arm; and axillary veins. The internal jugular vein is a deep vein in the neck and a common access site for catheter insertion. The deep veins of the chest (ie, thoracic central veins) include the subclavian veins (figure 3), which join the intrathoracic portion of the internal jugular veins to become the brachiocephalic veins, which in turn join together to become the superior vena cava (figure 4).
EPIDEMIOLOGY AND RISK FACTORS — For patients with central venous access, a wide variation in the incidence of venous thrombosis (1 to 66 percent) is reported, and the incidence depends upon the catheter type and location, criteria for diagnosis, type of test used, and population studied [10-15]. One retrospective study from a single institution estimated the frequency of thromboembolic disease of the brachiocephalic vein or superior vena cava due to central catheters to be 6 per 10,000 hospital admissions [11]. Although peripherally inserted central catheters (PICCs) reduce the incidence of pneumothorax by avoiding the deep veins of the neck and chest, they appear to have an incidence of upper extremity venous thrombosis that is similar to, if not greater than, that of centrally inserted catheters (CICCs), ranging from 3 to 58 percent [16-20]. In a study of postoperative general surgical patients, 58 percent of deep vein thromboses (DVTs) occurred in the upper extremity, all but one of which was catheter related [21].
The most common site of DVT for centrally placed catheters is the internal jugular vein [22]. For PICCs, the brachial, axillary, or subclavian veins are commonly involved. Thus, great caution should be used when considering intravenous access options in patients who may require hemodialysis access in the future due to the relatively high incidence of PICC-associated venous thrombosis. (See 'Peripheral versus central insertion' below and "Peripherally inserted central catheter (PICC)-related venous thrombosis in adults", section on 'Risk factors'.)
The highest reported incidences of upper extremity DVT are from studies that systematically screen for thrombus in patients with central catheters (asymptomatic and symptomatic), particularly those with malignancy. In one retrospective review, ultrasound confirmed the presence of upper extremity DVT in 60 percent of patients screened [12]. Another prospective study in patients with malignancy found that 75 percent of patients identified with upper extremity DVT were asymptomatic [13].
Risk factors — Any intravenous catheter has the potential to cause venous thrombosis [23]. These include peripheral intravenous catheters, PICCs [24], and tunneled and nontunneled central catheter port devices [25,26]. (See "Central venous access: Device and site selection in adults", section on 'Types of central venous catheters' and "Central venous access in adults: General principles", section on 'Central venous access'.)
An individual patient-level systematic review and meta-analysis identified 425 catheter-related thromboses among 5636 patients with malignancy (7.5 percent) from five trials and seven prospective studies [27]. The use of PICCs, previous history of DVT, subclavian venipuncture insertion site, and improper positioning of the catheter tip were associated with an increased risk for catheter-related thrombosis on multivariate logistic regression.
Catheter-related factors — Many studies have reported the association of specific catheter-related factors and upper extremity venous thrombosis.
Catheter type — The incidence of venous thrombosis is overall lower for port devices, and generally lower compared with PICCs, but increased for those with malignancy [28]. In the above systematic review, the risk of catheter-related thrombosis was lower for implanted ports compared with PICC catheters (odds ratio [OR] 0.43, 95% CI 0.23-0.80) [27].
●In a review of over 50,000 port devices, the overall risk for port-related venous DVT was at 1.8 percent. However, in a prospective multicenter study of patients with malignancy, ports were associated with a thrombosis rate of 3.8 percent; placement via the cephalic vein was associated with the greatest risk [28].
●In a subsequent systematic review that included 80 studies, the overall incidence of symptomatic venous thromboembolism was 2.76 percent (95% CI, 2.24-3.28) [29]. A subgroup analysis showed that the odds of venous thrombosis was significantly lower for ports compared with PICCs in patients with malignancy (OR 0.20, 95% CI 0.09-0.43).
The reported rates of thrombosis associated with midline catheters are more variable. Differences in midline catheter materials, insertion techniques, types of infusates, care and management of midlines, as well as strategies to diagnose DVT may explain differences in study findings [30].
●In one review that included over 2500 catheters, midline catheters were associated with an increased risk of thrombosis compared with PICCs (7 versus 4.7 percent) [31]. Whereas, in a multicenter review of 1161 midlines, DVT occurred in 1.4 percent [32].
●In a multicenter cohort study of hospitalized patients who received PICCs or midlines for the indication of difficult venous access, midlines were associated with similar odds of DVT compared with PICCs (OR 0.93, 95% CI 0.63-1.37). However, DVT events in patients with midlines occurred over a shorter dwell time when compared with PICCs. Thus, when expressed as a hazard ratio, the risk for DVT when comparing midlines with PICCs on a day-to-day basis was greater for midlines (hazard ratio 0.53, 95% CI 0.38-0.74) [33].
Catheter diameter/number of lumens — The diameter of the catheter relative to the size of the vein determines whether or not blood will flow freely around the catheter or stagnate [27,34,35]. For a vein of similar size, thrombosis is more likely with a large-diameter, centrally placed catheter (eg, plasmapheresis, dialysis, multilumen catheters) compared with a small-diameter catheter (eg, single lumen).
As such, for PICCs, insertion is recommended only when the catheter-to-vein ratio does not exceed 45 percent [19,36]. Some studies suggest this ratio should be lowered even further, reaching values of 33 percent [36,37]. In one study involving PICC catheters of varying diameters, 5 and 6 French PICCs were significantly more likely to develop DVT compared with 4 French PICCs [38,39]. In a quality improvement initiative, preferential use of single-lumen PICCs (as recommended by published appropriateness criteria [40]) was associated with a reduction in occlusion and thrombotic complications [41]. For central venous catheters placed into large vessels such as the internal jugular or subclavian veins, the ratio between the catheter to the vessel tends to be more favorable with lower reported rates of thrombosis [42].
Peripheral versus central insertion — PICCs appear to be associated with a greater risk for venous thrombosis overall (superficial and deep thrombosis) compared with CICCs, particularly among those who are critically ill or who have malignancy [16,18,19,27,43-53]. The incidence of DVT for PICCs is between 5 and 15 percent for hospitalized patients and 2 to 5 percent for ambulatory patients [44]. Other systematic reviews have suggested that the risk of DVT associated with PICCs placed using modern techniques (eg, ultrasound-guided insertion, micropuncture needle) may be less; however, these reviews may be subject to study selection bias [54,55]. In a systematic review of 62 studies, PICCs were associated with a 2.5-fold greater risk of thromboembolism compared with CICCs [16]. (See "Peripherally inserted central catheter (PICC)-related venous thrombosis in adults".)
Among CICCs, reported rates of thrombosis vary across sites of insertion, with higher rates reported for catheters placed in the femoral vessels compared with those placed in the internal jugular and subclavian veins [42]. (See "Central venous access: Device and site selection in adults", section on 'Access site'.)
Catheter malposition — Following placement of central catheters, the position of the catheter tip has traditionally been confirmed with plain chest radiography. However, alternative approaches are accurate and cost effective. Proper catheter positioning is discussed elsewhere. (See "Central venous access in adults: General principles", section on 'Confirming catheter tip position'.)
Malposition of the tip of a central catheter may be associated with an increased risk of venous thrombosis. In one study, upper extremity DVT developed in 46 percent of patients whose catheter tip was in the brachiocephalic vein or at the junction of the brachiocephalic vein with the superior vena cava, compared with 6 percent of patients with a properly positioned catheter with its tip in the right atrium or superior vena cava [13].
Catheter infection — Bacterial infection may originate from the patient's skin, with migration along the catheter to the vascular space from contaminated intravenous fluids or from hematogenous spread from a remote site [56]. Prior catheter infection may also be a risk factor for the development of catheter-related thrombosis [57]. In surgical patients, the bacteria isolated are often associated with the bacteria present at the site of surgery (eg, patients who undergo gastrointestinal surgery often have gastrointestinal flora isolated). (See "Catheter-related septic thrombophlebitis".)
The use of sterile technique during insertion and timely removal of indwelling catheters are important practices to prevent catheter-related infection [58]. (See "Routine care and maintenance of intravenous devices", section on 'Universal care strategies'.)
History of deep vein thrombosis — Prior DVT also increases the risk for catheter-related upper extremity venous thrombosis (OR 2.03, 95% CI 1.05-3.92) [27]. In a risk model examining prediction of PICC-related DVT, a history of DVT was a strong and independent risk factor for catheter-related thrombosis [59].
Prothrombotic states — The presence of a number of congenital or acquired systemic prothrombotic conditions may increase the risk for catheter-related upper extremity venous thrombosis [3,60-66].
It is well documented that factor V Leiden and prothrombin gene mutation 20210 are risk factors for non-catheter-related upper extremity DVT. Studies have also assessed the risk of catheter-related upper extremity DVT in carriers of these common mutations [67]. One prospective study of 252 consecutive patients found a nearly threefold increased risk of developing catheter-related upper extremity DVT in carriers of factor V Leiden or prothrombin gene mutations [68]. In a meta-analysis of 10 studies including 1000 patients with malignancy, an association was confirmed for both mutations [60]. The estimated attributable risk of catheter-related upper extremity DVT was 13.1 percent for factor V Leiden and 4.5 percent for prothrombin gene mutation. The indications for screening patients for such a prothrombotic condition are reviewed in detail elsewhere. (See "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)
In a prospective case-control study of 260 patients, a higher incidence of upper extremity DVT occurred in patients with antibody-positive heparin-related thrombocytopenia (HIT) with a central catheter compared with those without a central catheter (9 versus 0 percent, respectively) [69]. All thrombotic sites corresponded to the side with the central venous catheter, strongly implicating an interaction between the underlying hypercoagulable state of HIT and the local injury caused by the catheter.
A higher incidence (up to 32 percent) of symptomatic catheter-related upper extremity DVT is found in patients with underlying malignancies [1,28,70]. In one prospective cohort study of over 3000 patients with a variety of solid malignancies and implanted ports, 3.8 percent developed a catheter-related thrombosis over 12 months [28]. In a study of ambulatory patients with malignancy in the National Health System, in the presence of a PICC, DVT occurred in 6.2 percent of patients [71]. Cephalic vein insertion, as opposed to subclavian or jugular vein, was identified as a risk factor (highlighting the importance of the catheter-to-vein ratio in thrombosis). Patients with malignant processes involving the right lung, lymph nodes, or other mediastinal structures may present with catheter thrombosis as a consequence of thoracic central vein compression and thrombosis. The management of patients with central venous obstruction and superior vena cava syndrome is discussed elsewhere. (See "Risk and prevention of venous thromboembolism in adults with cancer" and "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)" and "Malignancy-related superior vena cava syndrome".)
Hormonal therapy — An association between hormonal therapy and primary (ie, non-catheter-related) upper extremity DVT has been established. Observational studies of pregnant women who receive PICCs for a variety of indications suggest that these patients experience a high rate of thrombosis [72,73]. (See "Deep vein thrombosis in pregnancy: Epidemiology, pathogenesis, and diagnosis".)
Whether hormonal therapy increases the risk of catheter-related thrombosis is unclear. Oral contraceptives may increase the risk for catheter-related upper extremity DVT when used in a patient with prothrombotic mutations, such as prothrombin 20210 or factor V Leiden [67,74]. Case reports of catheter-related upper extremity DVT (with or without coagulation abnormalities) are also reported in women undergoing ovulation induction and in vitro fertilization [75-77]. (See "Combined estrogen-progestin contraception: Side effects and health concerns", section on 'Venous thromboembolism'.)
Chemical irritation — Chemical phlebitis can occur as a reaction to the catheter material or to the infused drugs contacting the venous intima. Common culprits include potassium chloride, diazepam, certain antibiotics (eg, vancomycin and oxacillin), chemotherapy agents, and hypotonic (<250 mOsm/kg) or hypertonic (>350 mOsm/kg) electrolyte solutions [78].
The majority of these agents generally cause minor-to-moderate reactions that cease when the infusion is discontinued; however, chemotherapeutic agents can cause severe, persistent venous irritation. Some such agents also have irritant or vesicant properties and can cause necrosis of the subcutaneous tissues if leakage outside the vein occurs (table 1). It is important to note that not all vesicant agents cause venous irritation during the injection. (See "Extravasation injury from cytotoxic and other noncytotoxic vesicants in adults".)
CLINICAL PRESENTATION — The symptoms and signs of venous thrombosis are related to local effects and to embolization, either to the pulmonary circulation or, paradoxically, to the systemic arterial bed.
Local effects are due to the inflammatory reaction incited by the thrombus causing primarily pain and tenderness along the course of the vein. These findings are more obvious when a superficial vein is affected. Obstruction of upper extremity flow may cause congestion of the collateral veins of the shoulder and chest wall on the affected side. The patient may or may not take notice of the venous patterning that results.
Thrombosis of the deep veins is often asymptomatic or can present with mild symptoms. A high index of suspicion is necessary to make the diagnosis, and some cases go unrecognized, especially in patients with a variety of other major medical problems. This may partially account for the wide variation in reported incidence of thrombosis, and it underscores the need for careful ongoing clinical assessment of patients with central venous catheters. (See "Clinical features, diagnosis, and classification of thoracic central venous obstruction", section on 'Diagnosis' and 'Diagnostic evaluation' below.)
No symptoms — The reported incidence of asymptomatic catheter-related upper extremity deep vein thrombosis (DVT) in prospective series of patients with subclavian catheters has ranged from 5 to 13 percent [13,79,80]. However, thrombi are detected in over 50 percent of patients with malignancy and those undergoing bone marrow transplantation [1,81]. These thrombi range from nonocclusive mural thrombi to thrombi large enough to totally occlude the subclavian veins bilaterally. Why some individuals are asymptomatic, while others with thrombi in similar anatomic locations have debilitating symptoms, is largely unknown.
Inability to withdraw blood from or infuse into an indwelling catheter in an otherwise asymptomatic patient is a common occurrence, noted in 14 to 36 percent of patients within one to two years of central catheter placement [82]. The mechanism of the obstruction may be either thrombotic or mechanical (eg, kink in the catheter tubing, tight suture, catheter tip positioned against vessel wall, precipitation of infused drugs or lipid residues). Mechanical problems account for 40 percent of catheter occlusions, while thrombi account for the remainder.
Occluding thrombus usually does not originate in the lumen of the catheter but rather begins as a "fibrin sheath" encasing the catheter tip. The fibrin sheath represents a natural response of the vein wall to the catheter tip, with as many as 100 percent of catheters developing this phenomenon over time. Formation of the fibrin sheath can block the withdrawal of blood from the catheter by creating a one-way valve over the catheter tip, impairing the ability to withdraw blood from the catheter but allowing infusion through it, the so-called "ball-valve effect" or "withdrawal occlusion." Propagation of thrombus around the catheter can ultimately prevent both infusion and withdrawal from the catheter.
Symptomatic — The clinical manifestations of catheter-related venous thrombosis are variable, and patients can be mildly symptomatic or have debilitating symptoms. Clinicians caring for patients with indwelling catheters must be vigilant in their attempts to identify and treat thrombotic catheter complications.
Phlebitis — Phlebitis refers to an inflammatory reaction within the vein that gives rise to clinical findings of pain, tenderness, induration, and/or erythema along the course of that vessel. While phlebitis does not represent infection, it is usually difficult to differentiate from infection, and phlebitis predisposes to and often leads to infected lines/veins.
Phlebitis of a superficial veins with or without thrombosis is frequently associated with the use of peripheral intravenous catheters and is more common in patients who are immunocompromised (eg, burns, transplant) [11].
Patients with central venous catheters, particularly peripherally inserted central catheters (PICCs), can also present with phlebitis of the superficial veins where those devices are inserted (eg, basilic vein, cephalic vein). PICCs can cause extensive superficial vein thrombosis and phlebitis due to the length of vein traversed by the catheter [17]. Phlebitis in superficial veins due to peripheral venous catheters is generally a benign, self-limited disorder once the catheter is removed. (See 'Treatment' below and "Overview of thoracic central venous obstruction", section on 'Management'.)
The phlebitic reaction in deeper veins leads to pain and tenderness overlying the insertion site of the catheter. Induration, erythema, or congestion of tributary veins may be appreciated at the base of the neck, infraclavicular fossa, shoulder, or arm depending upon the location and type of catheter.
The differential diagnosis of these symptoms in patients with underlying malignancy includes local tumor invasion, malignant lymphadenopathy, and bone metastasis, which may be complicated by pathologic fracture (especially of the clavicle).
Suppurative (infected) thrombophlebitis causes significant local symptoms, and the patient may be toxic with bacteremia or fungemia. Suppurative thrombophlebitis can be initiated by infection of catheter-associated thrombus. Purulent material may emanate from the catheter exit site in patients with peripheral intravenous catheters and tunneled catheters. The management of suppurative thrombophlebitis is discussed in detail elsewhere. (See "Catheter-related septic thrombophlebitis" and "Intravascular non-hemodialysis catheter-related infection: Clinical manifestations and diagnosis" and "Intravascular non-hemodialysis catheter-related infection: Treatment".)
Extremity edema — Obstruction of the major thoracic veins can cause edema (usually unilateral) of the arm and hand, depending upon the extent of collateral venous flow from the upper extremity. Swelling is often exercise dependent or may appear if the arm is kept dependent for prolonged periods or when the arm is used vigorously. In some patients, the edema is a subjective symptom, described as a feeling of fullness in the fingers with rings feeling "too tight." The differential diagnosis of arm swelling is broad and includes lymphedema, muscle injury, intracompartmental hematomas, or thoracic outlet obstruction. Therefore, signs and symptoms of edema are not reliable for the diagnosis of thrombosis because of the poor specificity of clinical findings. (See "Overview of thoracic central venous obstruction", section on 'Clinical presentations'.)
If bilateral upper extremity DVT is suspected in a patient with bilateral upper extremity edema, it can be distinguished from more generalized fluid retention by the absence of lower extremity edema. (See "Clinical manifestations and evaluation of edema in adults".)
Embolization — A high index of suspicion for occult subclavian vein thrombosis is necessary when evaluating a susceptible patient with symptoms suggestive of pulmonary emboli or acute neurologic insult. Symptoms consistent with embolization in an otherwise asymptomatic patient with a central catheter may provide the first clues to the presence of upper extremity DVT. This possibility is often overlooked clinically. Although low, the risk of embolization is not negligible and more likely associated with thrombosis in the deep compared with the superficial upper extremity veins [4]. Fatal pulmonary embolism has been reported, even in patients treated with anticoagulants [83]. (See "Clinical presentation, evaluation, and diagnosis of the nonpregnant adult with suspected acute pulmonary embolism".)
The incidence of pulmonary embolus is relatively low for both catheter-related and spontaneous upper extremity DVT [5]. In patients with upper extremity DVT, one United States registry (324 patients) and one European Registry (11,564 patients) reported rates of symptomatic pulmonary embolus of 1 and 8 percent, respectively [5,6]. The incidence is higher in patients with malignancy. A higher incidence of 36 percent was reported in one study but may reflect the higher incidence of patients with thrombotic disorders and malignancy [84]. In the GARFIELD-VTE registry, rates of recurrence of thromboembolism did not differ in patients who developed upper versus lower extremity DVT (4 versus 5.5 per 100 person-years, respectively) [14].
Clinical prediction score — A clinical score for predicting the presence of upper extremity DVT was developed in a derivation cohort of 140 patients and confirmed in two separate validation cohorts with a total of 217 additional patients. All episodes were objectively confirmed by ultrasound examination. A risk score was generated from the following four parameters [85]:
●Presence of a catheter or access device in a subclavian or jugular vein or a pacemaker (plus 1 point)
●Unilateral pitting edema (plus 1 point)
●Presence of localized pain in that extremity (plus 1 point)
●Another diagnosis at least as plausible (minus 1 point)
Total scores were then rated as low probability (zero points or less, prevalence of upper extremity DVT 9 to 13 percent), intermediate probability (one point, prevalence 20 to 38 percent), or high probability (2 to 3 points, prevalence 64 to 70 percent).
A PICC-specific score is also used and presented separately. (See "Peripherally inserted central catheter (PICC)-related venous thrombosis in adults", section on 'Risk prediction'.)
DIAGNOSTIC EVALUATION — The methods to diagnose catheter-related upper extremity deep vein thrombosis (DVT) are similar to those of the primary upper extremity DVT (ie, Paget-Schroetter syndrome). However, a greater emphasis should be placed on noninvasive methods, usually beginning with duplex ultrasound and, possibly, D-dimer testing (a fibrin degradation product) [86]. (See "Primary (spontaneous) upper extremity deep vein thrombosis", section on 'Diagnosis' and "Primary (spontaneous) upper extremity deep vein thrombosis", section on 'D-dimer'.)
At least two studies have evaluated the role of D-dimer in patients with suspected upper extremity DVT [87,88]. Both studies found that in patients with image-confirmed thrombosis, D-dimer had high sensitivity but suffered from poor specificity. Therefore, a strategy using D-dimer to rule in or rule out upper extremity DVT cannot be recommended.
Sequential prediction rule — A sequential approach incorporating risk stratification and biomarkers may prove useful when approaching upper extremity thrombosis. Extrapolating from the approach in lower extremity DVT, investigators in the ARMOUR study examined whether an algorithm beginning with risk stratification using the rule by Constans, followed by D-dimer testing and compression ultrasonography, could better rule in/rule out patients with upper extremity DVT [86]. Evaluation of the rule in a cohort of patients at varying risk of upper extremity thrombosis revealed that the tool performed well in identifying patients at low risk of events. Approximately one-fifth of patients in the cohort of 406 patients avoided ultrasound imaging through use of this approach. However, application of the approach in another study consisting of a higher-risk cohort (patients older than 75, hospitalized patients, those with malignancy, or those with upper extremity catheters/devices) showed that the ability to distinguish a subset of patients where imaging could be withheld in this setting was low [89]. Thus, this approach might be more useful in patients deemed to be at low risk of venous thromboembolism.
Vascular imaging
Duplex ultrasonography — Ultrasound examination is the screening tool of choice to evaluate for upper extremity venous thrombosis because it is noninvasive, has no ionizing radiation or contrast-dye exposure, and is relatively easy to perform [90,91]. Similar to the lower extremities, a completely normal finding on ultrasonography of the upper extremity can safely exclude DVT [90]. Noncompressibility of the vein, abnormal flow (eg, flow reversal), and visible intraluminal thrombus are the major criteria for the diagnosis of thrombosis. Sensitivity and specificity range from 80 to 100 percent [84,92]. The chance of a false-positive study is very low; however, nonocclusive mural thrombus and thrombus in the proximal subclavian or brachiocephalic veins (generally shadowed by the clavicle and sternum) may not be adequately visualized. Additional imaging using computed tomographic venography or magnetic resonance (MR) imaging venography may be needed if the clinical suspicion for thrombosis is high.
When asymptomatic upper extremity DVT is suspected, the same techniques effective in the diagnosis of symptomatic thrombi are used. However, duplex ultrasound has a lower sensitivity in this setting, making a negative study noncontributory to patient care.
Patients with asymptomatic or nonocclusive deep vein thrombus or superficial vein thrombus of veins approaching the deep venous system (eg, basilic vein, cephalic vein) who do not respond to removal of the catheter and other conservative measures, or have progression of their symptoms, should undergo repeat duplex examination to evaluate for progression or extension of clot. Advanced imaging techniques may also be considered in these situations.
MR venography — Because MR imaging avoids radiation exposure and can provide detailed imaging of the veins, some studies have evaluated the role of MR venography to detect upper extremity DVT. In one small study of 18 patients, this technique had a sensitivity of 71 percent and a specificity of 89 percent for imaging without contrast. A pilot study of three patients with DVT identified on ultrasound similarly confirmed thrombosis in all vein segments, suggesting that the approach may be feasible [93]. However, because of the cost of this technique and limited evidence of effectiveness, its use cannot be widely recommended at this time [94].
Catheter-based contrast venography — Venography may be indicated if clinical suspicion for upper extremity DVT is high and the ultrasound is negative or nondiagnostic. For mechanical catheter problems, instillation of intravenous contrast into the lumen of the catheter under fluoroscopy may readily demonstrate the presence of thrombus at the tip of the catheter.
Digital subtraction venography can be performed if an aggressive diagnostic approach is required. Venography requires cannulation of either a peripheral vein or a central vein (eg, subclavian, femoral). Because poor venous patency is one of the reasons that indwelling catheters are needed in the first place, access for venography can be challenging.
TREATMENT — The goals in managing catheter-related upper extremity venous thrombosis include alleviation of symptoms, minimizing the risk for embolization, and providing for continued intravenous access, if it is needed.
Superficial vein thrombosis and phlebitis — There are limited data to guide management of upper extremity superficial vein thrombosis and phlebitis of upper extremity veins. Fortunately, it appears that pulmonary embolus from superficial vein thrombosis and phlebitis is very rare.
Initial symptomatic care — The initial management of superficial vein thrombosis and phlebitis related to peripheral intravenous catheters consists of discontinuing the intravenous infusion and removing the peripheral catheter.
Symptomatic care includes extremity elevation, warm or cool compresses, and oral nonsteroidal anti-inflammatory agents (NSAIDs). Topical NSAIDs may also be effective in relieving local burning associated with phlebitis.
When managing phlebitis, the catheter insertion site should be closely monitored. Excessive pain may indicate suppurative phlebitis, chemical phlebitis, or extravasation of infused fluid into the subcutaneous tissues. Additional treatment, including surgery, may be required to manage extravasation injuries. Antibiotics should be initiated if there is any concern for suppurative thrombophlebitis. (See "Catheter-related septic thrombophlebitis" and "Extravasation injury from cytotoxic and other noncytotoxic vesicants in adults".)
Role of anticoagulation — Extrapolating data regarding the role of anticoagulation from the lower extremity, it would be reasonable to also consider anticoagulation for patients with upper extremity superficial vein thrombosis who are at risk for DVT (eg, superficial vein thrombosis in proximity to the deep veins, thrombophilia) [4,95-97]. For spontaneous lower extremity superficial vein thrombosis, patients who are not anticoagulated have a higher incidence of persistent pain and thrombus recurrence/extension [98]. Based upon reviews of randomized trials, anticoagulation is suggested in patients with lower extremity superficial vein thrombosis who are at risk for deep vein thrombosis (DVT). As an example, in one systematic review and meta-analysis involving 6862 patients, anticoagulation with fondaparinux (a synthetic anticoagulant, factor Xa inhibitor) in patients with lower extremity superficial vein thrombosis was associated with reduced rates of subsequent thromboembolism [99]. Providing anticoagulation may be more important for patients with underlying malignancy in whom a predisposition to thrombosis is known to exist. (See "Superficial vein thrombosis and phlebitis of the lower extremity veins", section on 'Treatment of SVT'.)
Patients with basilic or cephalic vein thrombosis who remain symptomatic (eg, edema, pain) despite catheter removal may also be considered for anticoagulation treatment to manage symptoms. As with superficial vein thrombosis and phlebitis of the lower extremity, anticoagulation in these patients may be effective in alleviating symptoms, especially in the setting of malignancy. Whether anticoagulation improves the future patency of the superficial veins has not been adequately investigated. The duration of therapy under these circumstances needs to be guided by clinical judgment. Ultrasound can be repeated if symptoms progress to rule out extension into the deep venous system. (See 'Duplex ultrasonography' above.)
Deep vein thrombosis — For patients with upper extremity DVT, embolism is less common compared with lower extremity DVT due to differences in epidemiology and risk factors [14]. (See 'Epidemiology and risk factors' above.)
Nevertheless, venous thromboembolism is still a serious problem. Thus, therapy for upper extremity catheter-related DVT is also directed toward preventing this complication. Anticoagulant therapy is generally effective for preventing pulmonary embolism in patients with lower extremity DVT. By inference, anticoagulation should also be effective for preventing embolization from thrombosed thoracic veins, although treatment failures occur at both locations [83,95-97]. However, no treatment scheme has been rigorously evaluated for its efficacy in preventing embolization from upper extremity sources. As a result, recommendations for the treatment of upper extremity DVT are based largely upon indirect evidence from the experience with DVT of the lower extremities. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)".)
Patients with catheter-related thrombosis become rapidly asymptomatic with various combinations of observation, anticoagulation, and removal of the thrombogenic stimulus (the catheter). Individual case reports suggest that some patients with catheter-related thrombosis have symptoms that do not resolve; however, these patients appear to be in the minority.
Catheter-related upper extremity DVT is generally managed more conservatively compared with primary (spontaneous) upper extremity DVT. The more aggressive approach used with spontaneous thrombosis is due to a higher incidence of postphlebitic sequelae. In a systematic review, the incidence of post-thrombotic syndrome (PTS) ranged from 7 to 46 percent following upper extremity DVT in adults [100]. Residual thrombosis and axillosubclavian vein thrombosis were associated with PTS, but catheter-associated upper extremity DVT was associated with reduced risk. (See "Primary (spontaneous) upper extremity deep vein thrombosis".)
Anticoagulation — Despite a lack of direct evidence proving safety and efficacy of anticoagulation for upper extremity DVT, anticoagulant therapy remains the cornerstone of therapy with a goal of relieving acute symptoms and preventing embolization [96,101-103]. In patients with acute upper extremity DVT involving the axillary or more proximal veins, we recommend anticoagulation, as described for lower extremity DVT, provided there are no contraindications, with or without catheter removal. For isolated brachial vein thrombosis, the intensity and duration of anticoagulation is uncertain, and decision-making should be individualized. This recommendation is consistent with guidelines from the American College of Chest Physicians (ACCP) and other international guidelines [96,97,104]. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)", section on 'Patients at low risk of bleeding'.)
The type and intensity of anticoagulant therapy with catheter-related DVT should be similar to that given to prevent embolization of thrombi from the deep veins of the legs. We suggest initial therapy with parenteral anticoagulants (low-molecular-weight heparin [LMWH], fondaparinux, unfractionated heparin) followed by LMWH or a vitamin K antagonist (eg, warfarin). Sufficient data are lacking to recommend the use of a direct oral anticoagulant (DOAC) for the management of the acute phase of catheter-related upper extremity DVT. However, observational and small controlled studies examining the utility of DOACs in upper extremity thrombosis are emerging [105,106]. Therapeutic options for DVT are discussed in detail elsewhere. (See "Overview of the treatment of proximal and distal lower extremity deep vein thrombosis (DVT)".)
In general, we manage catheter-related upper extremity deep vein thrombosis in the following manner:
●For patients who have an ongoing need for the catheter, it is reasonable to administer anticoagulant therapy without catheter removal, provided the device remains functional and its tip is well positioned (see 'Catheter management' below) [7,97]. This approach has been associated with good clinical outcomes in small series of patients, including patients with malignancy [107-109]. The optimal duration of anticoagulation when the catheter stays in place has not been defined. The ACCP guidelines recommend continuing anticoagulation provided the central venous catheter remains in place, especially in patients with malignancy. If the catheter needs to be removed, no data are available to indicate whether central venous catheter removal should be preceded by a brief period of anticoagulation to minimize risk of embolization.
●There is uncertainty about the need to anticoagulate patients with thrombosis confined to the brachial vein [97]. The risk of long-term chronic venous sequelae from venous obstruction at this site generally appears to be quite small. The ACCP guidelines favor anticoagulation for up to three months if the thrombosis is symptomatic, is associated with malignancy, or the catheter remains in place [97].
●The optimal strategy for patients with upper extremity DVT who have contraindications to systemic anticoagulation is unknown. In most cases, removal of the catheter is recommended to prevent clot propagation and speed resolution of thrombosis. Localized procedures such as thrombectomy or catheter-directed therapies may be performed in instances of massive thrombosis but must be considered on a case-by-case basis. (See 'Deep vein thrombolysis' below and "Overview of thoracic central venous obstruction", section on 'Endovenous intervention'.)
●Whether asymptomatic upper extremity DVT should be treated is more controversial and less well defined than treatment of symptomatic thrombosis. While asymptomatic thrombosis in the lower extremity is well recognized as a source of embolization, the rate of this or other complications with asymptomatic thoracic central vein thrombosis is not known [84]. Consequently, the risk/benefit analysis of any potential treatment cannot be determined. However, there is no theoretical reason to believe that the risk of embolization is any different compared with symptomatic thoracic central vein thrombi, unless chronic in nature. In addition, asymptomatic upper extremity DVT can cause permanent obstruction in the subclavian vein and may interfere with subsequent catheter placement, resulting in loss of central venous access on the affected side. As an example of this last point, one series noted that 14 percent of patients undergoing a second or third catheter placement failed the procedure because of obstruction.
Deep vein thrombolysis — Most patients with catheter-related upper extremity DVT do not need catheter-directed thrombolytic therapy. Although some authors have advocated its use, there is no hard evidence that better outcomes are achieved compared with conservative therapy with anticoagulation. These interventions are time-consuming, expensive, and not without morbidity. Since thrombolytic therapy is indicated to minimize long-term symptoms, the impact of this potential problem on the patient's quality of life should first be determined.
Catheter-directed thrombolytic therapy would seem logical for individuals with significant symptoms in the acute stages of thrombosis (symptoms less than 14 days) who have met all of the following criteria (see "Endovenous intervention for thoracic central venous obstruction"):
●Have not adequately responded to systematic anticoagulation
●Have a low risk for bleeding
●Have a good long-term prognosis relative to their underlying disease
●Have a lifestyle that requires vigorous use of the affected arm
Patients who do not meet these criteria (eg, a patient with limited life expectancy from their coexistent medical problems who does not require extensive use of their arm) are likely to do just as well with conservative therapy.
Catheter management
Functioning — Maintenance of the catheter in a thrombosed vein is justified if it is mandatory, functional, in the correct position, and not infected. Anticoagulation is instituted and clinical symptoms monitored closely for signs of improvement for as long as the catheter is present [101]. Worsening of symptoms while anticoagulated indicates a need to remove the line and repeat the duplex examination.
Occluded — Occluded catheters often need to be removed and replaced. However, many patients have limited access sites. Replacement of the catheter risks recurrent endothelial injury and recurrence of thrombosis at the new catheter site. Thus, salvage of a catheter that has a thrombotic (ie, not mechanical) occlusion can be attempted with the instillation of fibrinolytic agents [110].
The decision to use thrombolytic therapy (eg, tissue plasminogen activator [tPA]) for upper extremity catheter thrombosis that is refractory to usual initial measures (ie, normal saline flush, repositioning) is mainly clinical, considering the available resources (drug availability, interventional radiology expertise) and the degree of need for rapid restoration of catheter function. For thrombotic occlusions unresponsive to two or more attempts using the thrombolytic agent, objective evaluation with radiologic techniques is advisable since some instances may be due to mechanical problems (eg, fibrin sheath or tip malposition) amenable only to mechanical solutions. (See 'Catheter-based contrast venography' above and "Clinical features, diagnosis, and classification of thoracic central venous obstruction", section on 'Diagnosis'.)
Thrombolytic agents that have predominantly been studied for this indication include t-PA or alteplase, and urokinase [110-120]. In the United States and elsewhere, alteplase (Cathflo, Activase) is more commonly used to manage catheter occlusion. Alteplase is available in 2 mg vials for use in this setting. For adults weighing ≥30 kg, one or two doses of 2 mg (1 mg/mL) of alteplase (t-PA) with a recommended dwell time of 30 to 120 minutes per dose are generally successful in restoring catheter function [121,122]. Lower-weight patients (<30 kg) should receive a reduced dose adjusted to catheter lumen volume. The efficacy and safety of local instillation of alteplase was confirmed with catheter function restored in 74 and 90 percent after one or two doses, respectively [119,121]. The approach using thrombolytics to restore hemodialysis catheter function and its effectiveness are reviewed separately. (See "Malfunction of chronic hemodialysis catheters", section on 'Treatment approach for intrinsic thrombus'.)
Urokinase is not available in the United States and has been found to be less effective compared with alteplase in small randomized trials [113,118,123]. Outside of the United States, urokinase may still be in use [124]. Guidance on urokinase dosage for the management of dysfunctional central venous catheters has been published for use in the United Kingdom with suggested doses of urokinase (Syner-Kinase) 10,000 international units for each lumen using a PushLock technique or dwell time of 30 to 60 minutes [124]. For persistent withdrawal occlusion or for hemodialysis catheters, a higher dose of urokinase (25,000 international units per lumen) is recommended. The value of intraluminal lytic enzyme infusion is unknown.
THROMBOSIS PREVENTION — Routine prophylactic systemic anticoagulation is not recommended for patients with indwelling central venous catheters [97,101,104,125-127]. Although randomized trials have not definitively demonstrated a clinically significant benefit for prophylactic anticoagulation in patients with malignancy, a decision for prophylactic anticoagulation may be reasonable if the clinical evaluation suggests the risks associated with thrombosis outweigh the risks of bleeding, taking into account tumor location and any prior history of thrombosis or bleeding [127-129]. It may be possible to identify high-risk patients who will benefit from prophylactic anticoagulation based upon specific factors (eg, chemotherapy) [27]. (See 'Prophylactic anticoagulation' below.)
Site and catheter management — Catheter-related thrombosis may be reduced using antithrombotic catheter lock solutions (eg, heparin, citrate). When thrombosis occurs, thrombolytic therapy may restore lumen patency, but thrombosis related to mechanical problems usually requires replacement of the catheter. (See 'Catheter management' above.)
Prophylactic anticoagulation — Most randomized trials in adults evaluating prophylactic anticoagulation with low-dose warfarin, unfractionated heparin (UFH), or low-molecular-weight heparin (LMWH) have been performed in adults with malignancy [79,101,126,130-144]. These subsets of patients have a high risk for venous thrombosis, yet a definitive clinical benefit for prophylactic anticoagulation in preventing catheter-related venous thrombosis has yet to be proven.
Early systematic reviews did not show any definitive benefit [136,137,139-142,145]; however, later meta-analyses that included a subsequent large trial found a decreased risk of symptomatic deep vein thrombosis (DVT) with prophylactic heparin and a decreased risk of asymptomatic DVT with low-dose warfarin, each compared with no anticoagulation [143-146]. The effects regarding vitamin K antagonists (VKAs) were uncertain, and comparison of prophylactic anticoagulation on mortality and adverse outcomes remains unclear in this study and in other reviews. The data have been inconsistent, likely due to differences in chosen endpoints, malignancy type, screening modality, and improvements in catheter-related factors over time [137,139-142,145,147]. Given this uncertainty, we continue to agree with various society guidelines that do not recommend routine prophylactic anticoagulation for the sole purpose of preventing catheter-related thrombosis [104,126,127]. However, in patients with malignancy, guidelines have endorsed pharmacologic prophylaxis of hospitalized patients at high risk of venous thromboembolism (eg, Khorana score of 2 or higher). The direct oral anticoagulants rivaroxaban and edoxaban have been added to recommendations as appropriate agents for thromboprophylaxis in hospitalized patients with malignancy, including those with central venous catheters [129]. While routine thromboprophylaxis is not recommended for outpatients with malignancy, a placebo-controlled trial of ambulatory patients with malignancy who were about to begin chemotherapy reported that prophylactic treatment with apixaban compared with placebo was associated with a significant reduction in DVT but with increased rates of bleeding [148]. (See "Risk and prevention of venous thromboembolism in adults with cancer".)
Prophylactic systemic anticoagulation (heparin) may have a role in some high-risk patients when the perceived risk of thrombosis outweighs the risk of bleeding and the burden of anticoagulation. High-risk factors include hospitalization for malignancy, previous venous thrombosis, bulky disease, or hereditary thrombophilia [27,128]. Research efforts should focus on identifying patients most likely to benefit from prophylactic anticoagulation.
●A large trial randomly assigned 420 high-risk ambulatory patients about to undergo chemotherapy for malignancy to no anticoagulation, LMWH, or warfarin (1 mg/day) for three months [146]. Compared with no anticoagulation, prophylactic anticoagulation significantly reduced the incidence of catheter-related DVT (LMWH: 8 versus 20 events; warfarin: 14 versus 20 events), and catheter-nonrelated DVT (LMWH: 1 versus 7 events; warfarin: 1 versus 7 events), with no difference between warfarin and LMWH. A later meta-analysis that included trials similar to those in an earlier review as well as the large trial found a significant benefit for prophylactic anticoagulation (heparin, warfarin) over no anticoagulation for symptomatic DVT (risk ratio [RR] 0.61, 95% CI 0.42-0.88) [144].
●An updated Cochrane review identified 13 randomized trials that evaluated the efficacy and safety of LMWH or VKAs for prevention of catheter-related venous thrombosis in patients with malignancy [143,147]. Among six adult trials, the incidence of symptomatic catheter-related thrombosis at three months was lower for LMWH compared to without (RR 0.43, 95% CI 0.22-0.81); VKA did not alter the risk for symptomatic catheter-related thrombosis. There were also no significant differences in bleeding, infection, or mortality for either LMWH or VKAs compared with no treatment.
Heparin-bonded catheters — A number of central venous catheters are available that have heparin-bonded coatings designed to reduce catheter-associated thrombosis. To date, technology has not been able to produce heparin bonding that is sufficiently sustainable for use in longer-term catheters. (See "Central venous access: Device and site selection in adults", section on 'Coated and impregnated catheters'.)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Superficial vein thrombosis, deep vein thrombosis, and pulmonary embolism" and "Society guideline links: Venous access".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Superficial vein phlebitis and thrombosis (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Catheter-related thrombosis – Intravenous catheters cause endothelial trauma and vein wall inflammation, which can result in thrombosis. Superficial vein thrombosis and phlebitis is most often related to peripheral intravenous catheters. Deep vein thrombosis can also occur and may lead to complications such as pulmonary embolism or post-thrombotic syndrome of the upper extremity. The majority of upper extremity deep vein thrombosis (DVT) events are due to peripherally inserted central catheters (PICCs). (See 'Introduction' above.)
●Risk factors – Risk factors for upper extremity catheter-related venous thrombosis include catheter-related factors (eg, catheter malposition, catheter size), the presence of prothrombotic states (congenital or acquired), hormonal therapy, and infusion of vesicant or irritant medications. (See 'Epidemiology and risk factors' above.)
●Clinical features – The majority of symptoms and signs of catheter-related thrombosis are related to local effects of the thrombus (ie, phlebitis, venous obstruction [edema, pain]). Although less common, symptoms and signs arising from embolization to the pulmonary circulation, or paradoxically to the systemic circulation via a patent foramen ovale (eg, brain, periphery), can also occur. A high index of suspicion for thrombosis is needed when evaluating susceptible patients with catheters who have symptoms suggestive of pulmonary emboli or acute neurologic insult. (See 'Clinical presentation' above.)
●Diagnosis – Duplex ultrasound is the initial imaging test of choice when upper extremity catheter-related venous thrombosis is suspected in asymptomatic and symptomatic patients. Patients who have had prior DVT should be screened for patency of the thoracic central veins prior to placement of a subsequent catheter. (See 'Diagnostic evaluation' above.)
●Superficial phlebitis – Superficial phlebitis related to peripheral intravenous catheters should be managed first by discontinuing the intravenous infusion and removing the peripheral catheter. Symptomatic care consists of extremity elevation, warm or cool compresses, and oral nonsteroidal anti-inflammatory drugs (NSAIDs). Excessive pain not controlled with NSAIDs may indicate the presence of a suppurative phlebitis, chemical phlebitis, or extravasation injury. (See 'Phlebitis' above.)
●Upper extremity deep vein thrombosis – For patients diagnosed with upper extremity catheter-related DVT, we suggest anticoagulation (Grade 2B). For uncomplicated cases, three months of anticoagulation therapy should be sufficient. A longer duration of anticoagulation may be warranted if the catheter remains in place, particularly for patients with malignancy. (See 'Treatment' above and "Overview of thoracic central venous obstruction", section on 'Management'.)
•For most patients with upper extremity catheter-related DVT, we suggest not instituting deep vein thrombolysis as a first-line therapy (Grade 2C). Compared with anticoagulation, there is insufficient evidence to suggest that thrombolysis leads to any better outcomes. (See 'Deep vein thrombolysis' above and "Overview of thoracic central venous obstruction", section on 'Endovenous intervention'.)
•The catheter can be left in place if it is functional, has a well-positioned tip, and there is an ongoing clinical indication for its use. For patients without malignancy with thrombosis confined to the brachial vein, there is uncertainty about the need for anticoagulation once the catheter has been removed. (See 'Catheter management' above.)
●Prophylaxis against catheter-related thrombosis – For most patients with indwelling upper extremity catheters, we recommend not administering prophylactic anticoagulation (vitamin K antagonists, low-molecular-weight heparin, unfractionated heparin) (Grade 1A). For patients with malignancy, we suggest prophylactic anticoagulation, especially among hospitalized patients (Grade 2A). It may be reasonable to administer prophylactic anticoagulation to high-risk patients in the ambulatory setting when the perceived risk of thrombosis outweighs the risk of bleeding. High-risk factors include previous venous thrombosis, bulky disease, hereditary thrombophilia, or suboptimal catheter tip location. (See 'Thrombosis prevention' above.)
ACKNOWLEDGMENTS — The editorial staff at UpToDate acknowledge Caroline Bérubé, MD, and James L Zehnder, MD, who contributed to an earlier version of this topic review.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
