Malfunction of chronic hemodialysis catheters

INTRODUCTION — The use of tunneled catheters for hemodialysis vascular access is associated with a relatively high incidence of complications, the most frequent of which is catheter dysfunction or low flow, which can lead to thrombotic complications. Catheter dysfunction is a major problem, with between 17 and 33 percent of chronic hemodialysis catheters requiring removal due to blood flow that is inadequate for hemodialysis [1-4]. The exact incidence of catheter dysfunction (generally expressed as cases per 1000 catheter-days) is not known for certain but is estimated to be between 0.5 and 3.42 episodes/1000 catheter-days, depending on the definition used [5,6]. As a surrogate marker, the use of intraluminal thrombolysis to restore flow has been reported in the range of 1.8 to 8.0 administrations/1000 catheter-days [3,7-10]. The incidence of catheter dysfunction is increased for subclavian vein compared with the internal jugular vein access sites [11], the presence of catheter malposition [11,12], prior catheter-related thrombosis [11,12], and increased body mass [13].

The consequences of catheter dysfunction include inadequate dialysis (by definition) [1], increased risk of sepsis [14], and shortened use-life of the dialysis catheter. If catheter dysfunction progresses to thrombosis, there is a requirement for the extra procedure of catheter removal or exchange. This has a negative impact on the patient's quality of life, adversely affects the dialysis schedule, and adds additional cost to the patient's therapy.

Malfunction of chronic hemodialysis catheters, including catheter-related dysfunction and thrombosis, is reviewed. Preventing catheter malfunction relies on proper catheter care, which is discussed separately. (See "Central venous catheters for acute and chronic hemodialysis access and their management".)

CATHETER DYSFUNCTION — Catheter dysfunction leading to inadequate dialysis has been defined based solely upon blood flow measurement. However, there is concern that using the blood flow rate as the only criterion can be misleading and has the potential to result in a significant number of unnecessary interventions [15,16]. Other factors besides blood flow rate that are important for determining the adequacy of hemodialysis include the weight of the patient, recirculation, ultrafiltration variables, duration of hemodialysis, and frequency of hemodialysis. In a review of 259 patients in which the relationship between catheter blood flow rate and hemodialysis adequacy was assessed, mean blood flows of 250, 275, and 300 mL/min did not reliably predict inadequate dialysis [16]. As a result of these considerations, catheter dysfunction has been more accurately defined as failure to attain and maintain an extracorporeal blood flow sufficient to perform hemodialysis without significantly lengthening treatment [15]. Nevertheless, decreasing catheter blood flow rate is a common problem frequently necessitating intervention.

Measuring catheter blood flow — The value recorded for blood flow rate during hemodialysis is typically the blood pump speed (based on pump revolutions per minute). However, this reading is not a true indicator of the actual blood flow to the catheter.

Measurement of blood flow rate depends on having a standardized volumetric blood pump segment of the dialysis tubing. This is the basis for two problems that can occur [17]:

●First, the pump relies on the elasticity of the pump segment to expand and refill as the rollers turn. However, despite the manufacturers' efforts to use tubing materials with minimal hysteresis (ie, failure to return to original size at no load), there is some flattening of the pump segment during dialysis as the tubing warms, which slightly reduces blood flow later in the dialysis treatment [18].

●Second, the relationship between prepump (arterial) pressure and blood flow rate is curvilinear. As the prepump pressure increases, blood flow rate decreases. There is a potential for prepump pressure to become more negative with increasing flow demands. The pump segment can be partially collapsed if the prepump pressure becomes excessive (becomes more negative). Refilling of this segment is less complete as prepump pressure increases and its reflection of true volumetric flow becomes progressively more inaccurate [19]. To prevent errors caused by excessive negative pressure, it is recommended that the prepump pressure be monitored and that alarms are set to detect when it drops below -240 to -260 mmHg [19].

Nevertheless, a blood flow rate problem that is likely to result in catheter dysfunction is generally progressive. So, to be of maximum value in detecting catheter dysfunction, blood flow rate should be trended over time. Available catheter designs allow a blood flow rate of >400 mL/min to be achieved at an acceptable prepump pressure; waiting until the blood flow rate declines to 300 mL/min may be inappropriate, missing the opportunity to detect catheter dysfunction earlier [19].

For the purposes of trending, it has been recommended that blood flow rate (from the blood pump indicator) should be measured at a standard prepump pressure (eg, -250 mmHg) with each hemodialysis session five minutes after the start of the session. A baseline blood flow rate should be established for a newly placed catheter, and the blood flow rate should be recorded for each treatment to detect changes. A decline of <10 percent at the same negative prepump pressure may result from a variety of factors and should not cause concern, but a change of >10 percent, particularly if progressive (indicating a progressive increase in resistance), should be taken as an indication of impending catheter dysfunction [17].

Indicators of dysfunction — Routine monitoring of parameters suggesting the presence of dysfunction should be performed for all dialysis patients who are using a hemodialysis access catheter. The variables listed below are recommended as indications of catheter dysfunction [17]:

●A decline in blood flow rate of >10 percent, particularly if progressive (measured at a prepump pressure of 250 mmHg five minutes after the start of the session).

●A delivered Kt/V of <1.2.

●An arterial pressure (prepump) more negative than -250 mmHg or a venous pressure (postpump) of >250 mmHg.

CLASSIFICATION — Catheter dysfunction can be described according to symptoms (asymptomatic, symptomatic), and timing relative to catheter placement (early, late). Catheter thrombosis, the most common cause of late dysfunction, is also classified by the location of the thrombus (extrinsic, intrinsic). Symptoms usually relate to the presence of catheter-related deep vein thrombosis (DVT). (See 'Catheter-related central venous thrombosis' below.)

Timing of dysfunction — Catheter dysfunction can be classified as early or late, with differing etiologies predominating [12].

●Early catheter dysfunction – Early catheter dysfunction is defined as a catheter that never functioned adequately after placement due to no blood flow, intermittent flow, or a blood flow rate insufficient to provide dialysis without significantly lengthening the hemodialysis treatment.

In general, early catheter dysfunction is related to technical problems with placement or catheter positioning, and these ideally should have been recognized and corrected at the time of catheter placement. Technical problems include an improperly positioned catheter, kinked catheter, or a catheter that is constricted by the exit-site suture(s). (See "Central venous catheters for acute and chronic hemodialysis access and their management", section on 'Catheter positioning'.)

If the catheter has been recently exchanged for a previously malfunctioning catheter, placement of the new catheter into a preexisting fibrin sheath must also be considered as a possible cause of catheter malfunction. (See 'Fibrin sheath thrombus' below.)

Severe obesity frequently complicates catheter placement. When a patient with obesity is upright, the weight and depth of the chest wall can pull on a catheter that was apparently properly positioned when the patient was supine, leading to a catheter that has retracted into the more distal superior vena cava, making it less functional [20].

A catheter may have been placed into the wrong vessel. As an example, placement of a catheter into the azygous vein is not unusual due to the frequency of central vein stenosis/thrombosis, which can lead to enlargement of the azygous vein. Malposition of the catheter into the azygous vein should be suspected if the catheter runs obliquely toward the midline in the anteroposterior view (image 1).

●Late catheter dysfunction – Late catheter dysfunction is defined as the inability to attain and maintain a blood flow rate sufficient to perform hemodialysis without significantly lengthening the hemodialysis treatment in a catheter that was previously functioning adequately.

The main cause of late catheter dysfunction is thrombotic occlusion, either partial or complete, which can also provide a substrate for bacterial growth and catheter-related infection [21]. Thrombosis frequently results in catheter loss [22,23]. Late dysfunction can occur very early after initially successful use of the catheter or after a prolonged period of continuous use. In a review of 721 hemodialysis patients, thrombus formation was one of four parameters independently related to inadequate dialysis dose delivery [24]. The specific factors leading to thrombosis in an individual case are seldom obvious. A definable hypercoagulable state is a possibility but is rarely observed.

Thrombus location — The location of thrombus is described as extrinsic or intrinsic; although these are described as distinct entities, there is overlap between these categories.

●Extrinsic thrombus – Extrinsic thrombus refers to a thrombus that extends outside of the catheter into the vascular structure in which it resides. Extrinsic thrombus forms outside of the catheter and is not necessarily attached to it, although the catheter or its tip may be embedded within the thrombus. In addition to causing catheter blood flow problems during hemodialysis, each form of extrinsic thrombus presents with relatively unique clinical features.

•Central vein thrombus – The presence of a catheter within a central vein can precipitate complete thrombosis of the vessel (ie, DVT). (See 'Catheter-related central venous thrombosis' below.)

•Atrial (mural) thrombus – Thrombus within the right atrium (ie, catheter-related atrial thrombus) may be attached to the wall of the atrium or present as a mass within the atrium. (See 'Catheter-related atrial thrombus' below.)

●Intrinsic thrombus – Intrinsic thrombus forms within the catheter or surrounds it as a sleeve or sheath. Intrinsic thrombus is the main complication and cause of loss associated with hemodialysis catheters. The three types of intrinsic thrombi are as follows:

•Intraluminal thrombus – Refers to a thrombus that occurs within the catheter lumen. (See 'Intraluminal thrombus' below.)

•Catheter tip thrombus – Refers to a thrombus that develops at the tip of the catheter. (See 'Catheter tip thrombus' below.)

•Fibrin sheath thrombus – Refers to a thrombus that develops as part of the sleeve immediately surrounding the catheter. (See 'Fibrin sheath thrombus' below.)

Thrombus diameter — For the purposes of our discussion, we classify thrombus (catheter tip, atrial/mural) associated with hemodialysis catheters as small, intermediate, or large in diameter, regardless of type of thrombus (mural thrombus, catheter tip thrombus), as follows:

●Small thrombus is defined as <2 cm

●Intermediate is defined as >2 cm but <6 cm

●Large is defined as >6 cm

EXTRINSIC THROMBUS

Catheter-related central venous thrombosis — Catheter-related venous thrombosis is classified as a type of late catheter dysfunction. Catheter-related central venous thrombosis can occur in any vein through which the catheter traverses, including the subclavian, jugular, or brachiocephalic vein, or the vena cava. More than one venous structure can be involved as thrombus propagates (eg, axillary vein extension from a subclavian venous thrombosis). Catheter-related central venous thrombosis is common among patients with hemodialysis catheters.

Most thrombi associated with hemodialysis catheters are small and asymptomatic. However, catheter-related venous thrombosis can incite an acute inflammatory reaction, causing symptoms with clinical features of upper extremity deep vein thrombosis, which should prompt a diagnostic evaluation. Symptoms of venous thromboembolism include pain and tenderness overlying the location of the catheter, such as at the base of the neck for a jugular catheter, in the supraclavicular fossa or shoulder area for a subclavian catheter, or in the groin for a femoral catheter. (See "Clinical features, diagnosis, and classification of thoracic central venous obstruction" and "Clinical presentation and diagnosis of the nonpregnant adult with suspected deep vein thrombosis of the lower extremity".)

Some patients present with an acute onset of upper extremity swelling that can be striking in appearance. Alternatively, edema of the arm and hand may only be present only when the arm is being used vigorously. For some patients, extremity edema is only a subjective symptom, presenting as a feeling of fullness in the fingers or difficulty with finger rings feeling "too tight." In other patients, signs and symptoms of embolization, either to the pulmonary circulation or paradoxically to the systemic arterial bed, may be the first clue; however, embolization in association with hemodialysis catheters is rare [25].

Diagnosis — In the presence of a central venous catheter, signs and symptoms of acute thrombophlebitis are diagnostic. When there is catheter dysfunction, imaging evaluation (venography performed through the catheter site) yields the diagnosis [25]. While venography may be more sensitive for demonstrating the presence of a thrombus, duplex ultrasound is noninvasive, inexpensive, generally more readily available, and relatively easily performed, though can be operator dependent. (See "Clinical features, diagnosis, and classification of thoracic central venous obstruction", section on 'Diagnosis'.)

Treatment of venous thrombosis — Based on recommendations for treatment of venous thromboembolism, anticoagulation is the mainstay of treatment for patients with acute catheter-related deep venous thrombosis, provided there are no contraindications [26]. Anticoagulation should be continued for a period of at least three months after catheter removal to prevent extension of the thrombus and to allow the thrombus to organize. For those who cannot be anticoagulated, catheter removal alone may be sufficient to resolve symptoms. Specific therapeutic regimens are discussed in detail separately. (See "Overview of thoracic central venous obstruction", section on 'Anticoagulation'.)

Catheter management — When symptomatic venous thrombosis is present (extremity swelling, thromboembolism), the catheter should be removed, if possible. For cases in which potential sites for vascular access are depleted or extremely limited, it may be possible to preserve the catheter. Available clinical practice guidelines (non-hemodialysis catheters) suggest that a catheter that is functioning and for which there is ongoing need does not need to be removed [26]. The patient should remain systemically anticoagulated for as long as the catheter is in use [26]. However, it should be kept in mind that the presence of the catheter can promote the development of infection, which can lead to additional complications [21]. (See "Catheter-related upper extremity venous thrombosis in adults", section on 'Functioning'.)

Catheter-related atrial thrombus — Catheter-related atrial thrombus (CRAT) can lead to late hemodialysis catheter dysfunction. Although a distinction is not made in some reports, two types of CRAT are described: mural thrombus and catheter tip thrombus.

●Mural thrombus — Mural thrombus (image 2) is the most common type of CRAT. Mural thrombus is thought to be caused by mechanical trauma to the atrial wall from catheter tip movement relative to the beating heart [27]. Endothelial damage results in activation of the coagulation cascade, platelet aggregation, and thrombus formation at the point of contact. The catheter tip is generally embedded within the thrombus, and when recognized, the diagnosis is generally made during the evaluation of catheter dysfunction.

●Catheter-tip thrombus — Catheter tip thrombus is thought to be due to the elongation of intraluminal thrombus or thrombus associated with the side holes of the catheter as a result of the fluid dynamics of the right atrium [28]. This proposed etiology is strengthened by the fact that positioning the catheter tip in the right atrium is highly associated with CRAT.

Although well-recognized as a complication of central venous catheters, the true incidence of CRAT is not known. The literature concerning CRAT consists of anecdotal single-case reports, small series reports, and compilations of cases derived from the literature. Reliable epidemiological data are not available [26,29]. In a postmortem study, the incidence of atrial mural thrombus associated with 42 central venous catheters was 29 percent, all of which had been asymptomatic [30]. In a study of 50 hemodialysis patients undergoing transesophageal echocardiography, a catheter tip thrombus was seen in nine (18 percent) cases, all of which are asymptomatic [31].

Clinical features and diagnosis — The majority of CRAT are small, cause no symptoms, and are recognized only incidentally by either transthoracic or transesophageal ultrasonography. Since it is unusual to perform an imaging study when a dialysis catheter is removed or exchanged, the incidence of CRAT is probably much higher than has been reported. Unless a patient becomes symptomatic, CRAT is generally not suspected. In a report of 213,236 chronic hemodialysis catheter procedures, among the 80 percent that were either catheter exchange or removal, symptoms leading to a diagnosis of pulmonary embolus were documented in only two cases [32]. This study points to the rarity of symptomatic CRAT. Nevertheless, when CRAT does occur, the consequences can be serious, including the possibility of death.

A high index of suspicion for CRAT is needed to make the diagnosis. CRAT should be suspected when a hemodialysis patient with a central venous catheter presents with one or more of the following clinical presentations:

●Mechanical problems

•Tricuspid regurgitation

•Right heart failure

•Cardiac electromechanical dissociation

•Cardiogenic shock

•Cardiac arrest

●Embolic problems

•Pulmonary embolism

•Systemic embolism (associated with patent foramen ovale)

●Infection (eg, sepsis, endocarditis)

While CRAT does not represent the most common cause of many of these clinical manifestations, it should be included in the differential diagnosis, and the presence of CRAT should be sought during the diagnostic evaluation using echocardiography, which should be considered for early diagnosis to reduce morbidity and mortality. The feature of CRAT that most closely corresponds to a fatal outcome is the presence of more serious symptoms [33].

Determination of thrombus size requires imaging. Many of these cases are first detected when echocardiography is performed. For accurate determination of CRAT size, transesophageal echocardiography has better sensitivity and specificity than transthoracic echocardiography. In addition, three-dimensional echocardiography can provide higher resolution of the size, point of attachment, and mobility of the thrombus [34].

Initial management — Patients with bacteremia should receive empiric antibiotic therapy regardless of the type of thrombus and tailored to culture and sensitivity results.

Treatment options — Management of hemodialysis patients with CRAT involves managing the thrombus and handling the catheter. Several treatment modalities have been used alone or in combination for the management of thrombus and are based on the size of the thrombus (algorithm 1). The catheter should generally be removed to prevent recurrence. Options for managing the thrombus include:

●Catheter-directed therapy (CDT)

•Catheter-directed thrombolysis

•Percutaneous catheter-directed suction thrombectomy (CDST)

•Ultrasound-assisted thrombolysis (USAT)

●Surgical thrombectomy

●Anticoagulation

●Systemic thrombolysis

The treatment selected largely depends on clinical judgment, availability of resources, and clinician experience, given the lack of evidence-based principles for management. Guidelines for management of hemodialysis patients with CRAT are nonexistent [26,29]. Treatment recommendations presented in the literature are opinions based heavily on strategies used in the treatment of pulmonary embolism, which may not be applicable. Older publications frequently to do not take into account newer developments in the management of this problem. In addition, the ongoing need for dialysis treatments and the associated comorbidities commonly represented in the dialysis patient exert a unique influence on management. Individualized care is important.

Catheter-directed therapy — CDT involves local-regional administration of a thrombolytic agent or physical aspiration of the thrombus, or both. CDT is increasingly used for the treatment of all types of atrial thrombus and for pulmonary embolism. The outcomes achieved by treating these problems suggest that CDT can be used as the initial therapy for the management of intermediate and large diameter CRAT (algorithm 1) when surgery is not otherwise indicated. However, CDT requires institutional resources, proper equipment, and appropriate expertise.

●Catheter-directed pharmacologic thrombolysis — CDT has been widely used in the treatment of pulmonary embolism, and there are some reports of its use in the treatment of right atrial thrombus not associated with a central venous catheter. Only sporadic cases of CDT to treat CRAT have been reported [35-37]. However, judging from the success reported for pulmonary embolism and an increased level of safety associated with this modality, CDT appears to be the preferred approach for the management of CRAT. In one case report, the dimensions of a CRAT decreased from 30 x 16 x 22 mm to 10 x 8 x 5 mm (96 percent reduction) after an 18-hour infusion of low-dose tissue plasminogen activator [35]. In another case report of right atrial thrombus not associated with a catheter, a thrombus measuring 6 x 1.5 cm was successfully resolved with catheter-directed thrombolysis [37].

In most instances when thrombolytic therapy is used, the agent is administered by continuous infusion over a period of hours via a catheter placed fluoroscopically near the thrombus. This approach has several advantages compared with systemic administration. Local intra-thrombus delivery of the thrombolytic agent results in a more rapid permeation of the thrombus, allowing for shorter infusion times and lower incidence of complications [38,39]. An additional advantage is that the dialysis catheter can be used during the infusion. In a meta-analysis, CDT was safe and effective and associated with a lower rate of major extracranial and intracranial hemorrhage compared with systemic thrombolysis [40].

●Ultrasound-assisted thrombolysis — USAT has been used primarily in the treatment of pulmonary embolism, with some reports of use in the treatment of right atrial thrombus [41-44]. USAT provides the attributes of CDT with addition of an acoustic field generated by an ultrasonic catheter core. The theory behind the device is that ultrasound serves to accelerate the lytic process. By shortening the time required for dissolution of the thrombus, the required dose of thrombolytic agent will also be reduced, lessening the risk of bleeding complications. However, in a study that compared USAT with CDT in the treatment of pulmonary embolism (36 and 27 cases, respectively), no significant differences were reported for clinical or hemodynamic outcomes, procedural complication rates, or total dose of thrombolytic agent, even though the duration of the treatment in the CDT group was significantly longer compared with the USAT group [45].

Only sporadic cases using USAT to treat CRAT have been reported [41-44]. In some of these reports, the right atrial thrombus was very large; however, the cases were conducted safely and were successful. In one case report, a hemodialysis-associated CRAT measuring 8.4 cm x 0.5 cm was successfully treated [41].

●Catheter-directed suction thrombectomy — CDST is an alternative catheter-directed approach to the treatment of CRAT [46-48]. To perform suction thrombectomy, percutaneous venous access via the common femoral vein is first obtained. Then, a specialized large bore catheter (eg, AngioVac with a 26 Fr sheath) is introduced into the atrium that uses mechanical suction to aspirate the thrombus. Although there are no reports using this modality specifically for the treatment of CRAT, treatment of other types of right atrial thrombus have reported a high level of success with relative freedom from complications, suggesting that it may be applicable in the treatment of CRAT [46-48]. CDST has been associated with mortality, mostly when used to treat large pulmonary emboli [49].

Surgical thrombectomy — Surgical thrombectomy has been recommended for the treatment of larger CRAT and in those with CRAT and other indications for cardiac surgery (algorithm 1) [33,50]. Surgery involves removal of the thrombus, removal of the catheter, and treatment of any contributing or other cardiac abnormalities that are surgically correctable. Indications for surgery include:

●Large CRAT >6 cm – In one report that compared surgical with nonsurgical therapy for thrombus (not all dialysis catheter related), surgical therapy was associated with significantly fewer complications regardless of thrombus size [51]. However, determining the need for surgery should not be based on the size of the thrombus alone in view of the results that have been obtained with CDT. Many dialysis patients are not ideal surgical candidates for the type of surgery that would be required.

●Patent foramen ovale – For patients with a patent foramen ovale, even a small embolus could create a serious complication from cerebral or peripheral arterial embolization.

●Infection – For those with infection (ie, endocarditis, septic emboli), removing the source of infection is an important aspect of management. (See "Overview of management of infective endocarditis in adults".)

Anticoagulation — In published reports, anticoagulation is used to manage CRAT larger than 2 cm in diameter in patients who will not undergo intervention to remove thrombus, or for those with residual thrombus after intervention (algorithm 1) [52,53]. Intravenous therapy is followed by oral anticoagulation for three to six months until the thrombus has resolved. In one report, anticoagulation for six months resulted in resolution of more than 50 percent of cases [54].

For some patients, anticoagulation may be contraindicated. Some patients have absolute contraindications to anticoagulation, but for others, the contraindication is relative [55-57]. Dialysis patients have other factors that increase the risk for bleeding, including older age [58], frailty [59,60], elevated fall risk [61], and an increased risk for fractures [62]. These issues have led to concern and controversy over the use of chronic anticoagulation in this population. Because of these issues, chronic anticoagulation must be individualized with careful attention to benefit versus risks. Strategies used to minimize the risk of bleeding associated with the use of anticoagulants in hemodialysis patients include minimizing heparin use with dialysis, use of citrate locks for catheters, and tight blood pressure control. If clinically reasonable, concurrent antithrombotic agents such as antiplatelet agents and nonsteroidal antiinflammatory agents should be discontinued, and prophylaxis for gastrointestinal bleeding should be prescribed [39].

Oral anticoagulants include vitamin K antagonists (eg, warfarin), direct oral anticoagulants (DOACs), factor Xa inhibitors, and direct thrombin inhibitors. Among these, the agent most widely used in hemodialysis patients with CRAT has been warfarin. Although warfarin has demonstrated benefits, it has a narrow therapeutic window and requires frequent monitoring in the dialysis patient. Maintaining an optimal international normalized ratio in patients with chronic kidney disease (CKD) is challenging [63,64]. This has been attributed to vitamin K deficiency from malnutrition, frequent antibiotic exposure, and chronic illness seen in these patients [64]. There are also numerous food and drug interactions that commonly occur in these patents with their dietary restrictions and burden of polypharmacy [64]. Chronic warfarin use has also been associated with calciphylaxis [65]. There are also specific risks associated with chronic warfarin administration in the dialysis patient [66,67]. These include an increased incidence of hemorrhage [64,68] and a higher risk of hemorrhagic stroke [69]. These issues make it necessary to carefully weigh the risk-benefit ratio associated with chronic warfarin administration before initiating therapy.

DOACs represent an alternative to warfarin and have been approved for use in the United States for patients with CKD stage 3 and 4. Compared with warfarin, DOACs have lower risk for bleeding [70,71], do not need regular monitoring, and do not interact with other drugs or food. Although there are no reports of the use of the factor Xa inhibitor, apixaban, in the treatment of CRAT, the reported safety profile of treating atrial fibrillation or venous thromboembolism in patients with creatinine clearance less than 25 mL/min suggests it may be a reasonable alternative to warfarin [71]. However, there is no consensus on the safety of apixaban in dialysis patients [69,72,73]. Apixaban has shown either no difference or lower complication rates compared with warfarin [74]. Apixaban is minimally affected by dialysis. In a four-hour dialysis session (use of Optiflux F180NR dialyzer, dialysate flow rate of 500 mL/min, blood flow rate 350 to 500 mL/min), only 6.7 percent of the drug was removed [75].

Systemic thrombolysis — Systemic thrombolysis involves peripheral administration of an initial loading dose of the thrombolytic agent followed by continuous infusion over a period of hours. The outcomes of systemic thrombolysis for CRAT have been mixed. In a systematic review that identified 71 cases, systemic thrombolysis was used in eight cases and was successful in only two [33]. However, among hemodialysis patients (44 of 66), the success rate for thrombolysis was 67 percent (8/12) compared with 74 percent for anticoagulation (24/33) [76]. In series of cases with right atrial thrombi (not CRAT) associated with a pulmonary embolism, systemic thrombolysis significantly improved survival rate when compared with either anticoagulation therapy or surgical embolectomy [77].

Because of complications (bleeding, possible embolism), systemic thrombolysis to treat CRAT is a second-line therapy relative to anticoagulation [78]. Even among patients without contraindications, the reported rate of major hemorrhage from systemic thrombolysis is approximately 20 percent, including a 3 to 5 percent risk of hemorrhagic stroke [79,80]. Mortality or hemorrhagic events among CKD patients who undergo systemic thrombolysis do not appear to differ compared with non-CKD patients [81]. Systemic thrombolysis also has the potential to precipitate pulmonary embolism by releasing a mobile thrombus [82].

Treatment comparisons — Comparing the various treatment options that have been used to treat CRAT is difficult. No systematic studies based upon a standardized protocol have been conducted. All reports have been affected by many variables that have influenced the management strategy used. Moreover, hemodialysis patients represent a heterogeneous group with variable comorbidities. In the presence of so many confounding variables, compiling cases from multiple published reports can give misleading results. Nevertheless, comparisons have been made. The most common approach for comparing modalities for the treatment of CRAT is to examine mortality rates in symptomatic cases.

●In an early systematic review of CRAT, 71 published reports involving 119 subjects were summarized [53]. Among the cases, 93 percent presented with abnormal cardiac signs and symptoms, and 7 percent were asymptomatic but had risk factors for pulmonary embolism. Thrombolysis, surgery, anticoagulation, and no therapy were compared. Except for the no therapy group, there was no significant difference between these groups.

•Among those without an associated pulmonary embolus, the estimated probability of survival for patients receiving heparin, thrombolytic agents, surgical embolectomy, or none of the above, was 92, 89, 89, and 53 percent, respectively.

•Among those with an associated pulmonary embolus, the estimated probability of survival for patients receiving heparin, thrombolytic agents, surgical embolectomy, or none of the above, was 70, 62, 62, and 19 percent, respectively.

●More than 20 years later, a separate systematic review identified 71 cases of CRAT that also compared thrombolysis, surgery, anticoagulation, and no therapy [33]. Results in this report were very similar. Survival was 81, 86, 94, and 56 percent, respectively. Again, survival was similar between the groups, except for the no therapy group.

Approach to management — The two types of CRAT, mural thrombus and catheter tip thrombus, can generally be managed in a similar fashion with only a few differences. Detailed imaging is required to determine the type, size, and points of thrombus attachment. (See 'Clinical features and diagnosis' above.)

The treatment strategy selected for the treatment of CRAT should be individualized, taking into consideration the patient's comorbidities, the urgency of the situation, and the available clinical expertise (algorithm 1). In the absence of a body of high-quality clinical evidence to guide management, recommendations for treatment are necessarily opinion-based, much of which have been the product of experience dealing with pulmonary embolism and right atrial thrombus that are not catheter related. This experience, along with reported cases involving CRAT, suggests that a catheter-directed approach is effective and associated with a low risk of complications. The lower risk is related to the agent used and method of administration, which reduce the overall duration of time required for thrombus resolution. As such, for patients without indications for surgery, we suggest CDT rather than anticoagulation alone or systemic thrombolysis.

In addition, since the presence of the catheter is integral to the pathogenesis of CRAT, catheter removal is important in management. The timing of catheter removal is determined by thrombus diameter and type of thrombus. Failure to remove the catheter is associated with CRAT recurrence and mortality, due to either pulmonary embolism or septic complications [76,83,84]. If the patient continues to require a catheter, a new catheter can be inserted, but the tip should be positioned in a different location, either the superior vena cava or inferior vena cava. If the patient is a suitable candidate, the placement of an early cannulation graft [85-87] or initiating peritoneal dialysis should be considered to avoid the continued use of a catheter.

Small diameter thrombus — For small diameter thrombus <2 cm in diameter, we suggest catheter removal only, which should limit any further increase in thrombus size with time. The catheter can be removed regardless of the thrombus type (mural, catheter tip).

Intermediate diameter thrombus — For intermediate diameter thrombus >2 cm but <6 cm, we suggest CDT, if available (no other indication for surgery). If CDT is not available, thrombus should be treated with anticoagulation (using apixaban) or systemic thrombolysis, depending on the clinical situation. Patients in whom anticoagulation, systemic thrombolysis, and surgery are all contraindicated should be monitored; patients who fall into this category would be expected to have a limited longevity, potentially receiving palliative dialysis. Treatment failures following CDT can be managed with either anticoagulation or surgery according to the clinical situation and clinical judgment.

To manage the catheter:

●For mural thrombus, the catheter can be removed as CDT is initiated since the thrombus is attached to the atrial wall and the risk of embolization is low. After resolution of thrombus, follow-up imaging to rule out recurrence should be obtained 7 to 10 days after completion of treatment for mural thrombus. Since the cause of the thrombus (the catheter) is no longer present and in view of the associated risk, chronic anticoagulation is not necessary unless indicated for another condition.

●For catheter-tip thrombus, the catheter should remain in place to avoid the risk of embolization until the thrombus has resolved, after which time the catheter can be removed.

Large diameter thrombus — For large thrombus >6 cm, surgery is often recommended, but a decision for surgery should be based upon the clinical situation. For the patient who is clinically stable without signs and symptoms of cardiac dysfunction, CDT can be attempted first. For those who undergo surgery, the catheter is removed after thrombus removal.

INTRINSIC THROMBUS

Types

Intraluminal thrombus — When an intraluminal thrombus occurs, the catheter becomes totally occluded. Thrombosis of the lumen of the catheter results from one of the following:

●An inadequate volume of anticoagulant being placed within the catheter lumen

●Failure to close the clamp on the catheter before the heparin syringe is detached

●Anticoagulant being lost from the catheter between dialysis treatments

●The presence of blood within the catheter

Catheter tip thrombus — Since the portion of the dialysis catheter from the side holes to the tip does not retain heparin, a thrombus can form in this location (image 3). A catheter tip thrombus may be occlusive or act as a ball valve. If the thrombus is not addressed because it does not cause a mechanical problem, it can progress and give rise to what is classified as a catheter-related atrial thrombus. (See 'Catheter-related central venous thrombosis' above.)

Preventive measures that are used to avoid intraluminal thrombosis are generally adequate for the prevention of these thrombi. It is probable that forcible flushing before and after dialysis aids in clearing poorly attached catheter tip thrombi. (See "Central venous catheters for acute and chronic hemodialysis access and their management", section on 'Routine care and access for hemodialysis'.)

Fibrin sheath thrombus — Fibrin sheath thrombus (image 4) develops as a sleeve immediately surrounding the catheter and is the most common reason for central vein catheter dysfunction. As it develops, it eventually covers the inlet and outlet holes of a catheter, interfering with the blood flow, leading to ineffective dialysis. This problem generally occurs weeks or months after catheter placement, but it has been observed as early as 48 hours after catheter placement.

All central venous catheters likely become encased in a layer of fibrin within a few days of insertion. It starts as early as 24 hours after insertion of the catheter [88,89], which then becomes encased along its entire length within five to seven days [88,89]. In an autopsy study of 55 patients with central venous catheters, a fibrin sheath was found in 100 percent of cases [90]. This is not to say that all fibrin sheaths are problematic; they are not.

●When the vein is cannulated, the endothelium is injured, and a thrombus forms around the catheter at the insertion site. This becomes organized and is eventually converted to fibroepithelial tissue. The term "fibrin sheath" refers to a sleeve that surrounds the intravenous portion of the catheter. Animal studies have suggested that this sheath is actually not composed entirely of fibrin [89-91]. Fibrin continues to be deposited at the leading edge of this tissue and consequently becomes organized. In this manner, a sleeve migrates down the catheter to form a sheath composed of thrombus, fibroblasts, endothelial cells, and collagen, forming a layer about 1 mm thick around the outside of the catheter.

●The fibrin sheath is only loosely attached to the catheter. When a catheter is removed, angiography performed through the partially retracted catheter can demonstrate a "wind sock" of residual fibrin sleeve. The reported incidence for fibrin sheath has ranged from 13 to 76 percent [23,92-94].

●Although not considered part of the fibrin sheath, the fibroepithelial sleeve surrounding the catheter begins peripherally at the catheter exit site, continues through the subcutaneous tunnel, and is continuous with the intravenous portion of the catheter.

It remains unknown whether a fibrin sheath can lead to or contribute to the formation of a stenotic lesion within the vein. Histology studies of human autopsy material of long-term catheters have shown the development of vein wall thickening and bridges from the vein wall to the fibrin sheath covering the catheter [95]. These changes have led some interventionalists to question whether a fibrin sheath should be ablated even if the catheter is being permanently removed to minimize the risk of developing a later venous stenosis.

Treatment approach for intrinsic thrombus — Catheter dysfunction is common and can have serious consequences if not addressed in a timely fashion. Once recognized, intrinsic catheter dysfunction should be addressed immediately; delayed treatment results in inadequate dialysis and excessive catheter manipulation, which can lead to catheter-related bacteremia.

Since intrinsic catheter dysfunction is usually first recognized in the dialysis clinic, successful management allows the patient to continue receiving their scheduled dialysis treatment, which has great value. Management should be initiated using "bedside maneuvers," which includes repositioning the patient (placing in a Trendelenburg position) or using a "forceful saline flush" [15]. (See 'Forceful saline flush' below.)

If bedside maneuvers are unsuccessful, reversing the catheter lumens may allow for completion of the hemodialysis treatment. Intraluminal lytic enzyme has the potential for providing a more prolonged benefit if the problem is related to an intrinsic catheter thrombosis. If the problem cannot be successfully resolved within the dialysis facility or if the treatment applied does not have an effect of sufficient duration, then the case should be referred to a vascular specialist for diagnostic imaging and possible intervention [96]. (See 'Referral for intervention' below.)

Forceful saline flush — A forceful saline flush is frequently successful in restoring catheter function [96,97]. To minimize the chance of catheter rupture, the syringe should not be smaller than 5 mL; a 10 mL syringe is optimal. The force that can be generated with a syringe is inversely proportional to the diameter of the syringe (or more precisely related to the ratio of the diameter of the plunger to the syringe outlet opening). Thus, a small (but not too small) syringe should be used, and it should include a Luer lock to prevent detachment during the procedure.

To provide a forceful saline flush, fill the syringe with saline, firmly attaching it to the catheter, and flush into the catheter with as much force as can be generated with the hand. Once done, attempt to aspirate blood. If no blood can be aspirated, repeat the saline flush, which can be done several times. If blood can be aspirated, the flush procedure should be repeated several times with aspirated blood until flow seems to be free and easy.

The saline flush technique has the advantages of being easily performed, economical, safe, and frequently effective. Although a successful saline flush means that the small amount of thrombus that occluded the catheter has embolized, this does not appear cause any clinically significant problem.

The disadvantages of this procedure are the following:

●It may not be successful.

●It is not a permanent solution.

●There is a small risk of rupturing the catheter. However, if the catheter does rupture, it is not a major problem. An intravascular rupture will not be obvious, although the catheter will continue to be dysfunctional. An extravascular rupture should be clamped central to the rupture point until the catheter can be replaced.

Intraluminal lytic enzyme — Instillation of intraluminal lytic enzyme (eg, tissue plasminogen activator [tPA]: alteplase, tenecteplase reteplase; urokinase) has been extremely useful for managing catheter dysfunction in the dialysis clinic when flushing is ineffective. The lytic agent is typically maintained within the catheter for a prescribed period of time, but continuous infusion has also been tried.

General points and guidelines — In summary, with respect to the available information on instillation of intraluminal lytic agents, which are described in the sections below:

●Lytic therapy is effective in the treatment of catheter dysfunction.

●Tissue plasminogen activator may offer an advantage over urokinase. (See 'Administration' below.)

●The duration of success with treatment is relatively short, and the duration decreases progressively with subsequent treatments.

●Using a "push" protocol may be more effective than the standard dwell protocol. (See 'Dwell versus push technique' below.)

Each dialysis clinic should have a protocol in place for the use of a lytic agent to manage intrinsic thrombus. Such therapy can be accomplished with minimal disruption to the patient's dialysis schedule. The question arises as to how many times intraluminal lytic therapy should be used before the patient is referred for a catheter exchange. The critical points are the patient's need for dialysis on a timely basis and the risks of excessive catheter manipulation. A catheter that is not able to provide the blood flow necessary for adequate dialysis according to the patient's prescription must not be tolerated. All catheter flow problems must be resolved in a timely fashion, and the emphasis for managing catheter dysfunction should shift to intervention at a relatively early stage of dysfunction.

Reasonable guidelines are as follows:

●If lytic therapy fails to restore blood flow to a level that will provide for adequate dialysis according to the patient's prescription, the catheter should be exchanged.

●If the duration of the effectiveness of lytic therapy is less than two weeks, serious consideration should be given to exchanging the catheter.

Effectiveness of intraluminal lytic therapy — Several studies have demonstrated the effectiveness of intraluminal thrombolytic therapy or have provided comparisons between agents, doses, or techniques for administration. Most are observational studies, though a few trials have been reported.

●In one trial, 149 patients with a dysfunctional hemodialysis catheter were randomly assigned to tenecteplase (2 mg for one hour) or placebo [98]. Dysfunctional was defined as a blood flow rate <300 mL/min at -250 mmHg pressure in the arterial line. After a one-hour dwell, the percentage of patients with a functional catheter was significantly higher for the tenecteplase group compared with placebo (22 versus 5 percent). At the end of dialysis, the mean change in blood flow rate was 47 mL/min in the tenecteplase group compared with 12 mL/min in the placebo group.

●A prospective study analyzed patients with dysfunctional hemodialysis catheters over a 2.5-year period [3]. Instillation of a lytic agent (alteplase) was necessary in 2.8 percent of dialysis sessions. The median time from the first to second treatment or catheter removal for nonfunction or thrombosis was 27 days (95% CI 15.7-32.3). The additional median catheter survival advantage from each subsequent treatment ranged from 10 to 18 days. In other words, treatment of recurrent catheter malfunction provided a median of only five to seven additional dialysis sessions before the treatment had to be repeated or the catheter exchanged.

It is unknown for certain whether one lytic agent is superior to another, or for a specific agent, the optimal dosing and dwell time. Regarding agents and dosing:

●While tPA (alteplase) is the more commonly used agent in the US, urokinase is also effective. In one trial, hemodialysis patients were randomly assigned alteplase (1 mg/mL) or urokinase (5000 international unit [IU]/mL) [99]. The dwell time for both groups was 40 minutes. The percentage of patients achieving post-thrombolytic blood flow of ≥250 mL/min was similar between the groups (alteplase: 42/44 [95 percent]; urokinase: 46/56 [82 percent]). Failure to achieve blood flow after a second dose occurred in more patients for urokinase compared with alteplase (7 versus 1 patient), but the difference was not significant.

●When using tPA (alteplase), 1 or 2 mg (1 mg dose prepared in a concentration of 1 mg/2 mL normal saline or 2 mg dose prepared in a concentration of 2 mg/2 mL normal saline) may be equally effective, and a one hour dwell time, rather than longer, appears to be sufficient. In a trial that included 252 catheter events, the rate of clot resolution was similar for 2 mg versus 1 mg (85.7 percent and 84.9 percent, respectively) [100]. In a trial that included 60 hemodialysis patients, catheter patency (blood flow >250 mL/min) was similar for short (one hour) compared with long (>48 hours) dwell time at the subsequent treatment (76.9 and 79.4 percent, respectively) or at two weeks (42.3 and 52.9 percent, respectively). Overall, after tPA installation, catheter patency was 78 percent at the subsequent hemodialysis session, with patency falling to 48 percent. There was no statistically significant difference between the short and long tPA dwell groups for catheter patency.

●When using urokinase, a higher dose (100,000 IU/mL, rather than 25,000 IU/mL) appears to be more effective. In a trial of 65 patients, successful restoration of blood flow >250 mL/min was recorded in all patients (100 percent) in the high-dose group and only four (13.7 percent) in the low-dose group [5].

Administration — The protocol for administering a thrombolytic agent involves instillation of the agent into the occluded catheter lumen, allowing it to dwell for a period, followed by an assessment of the catheter blood flow. In most cases, it is only the arterial lumen that is occluded and requires treatment. If both lumens are occluded, both will require treatment. The manufacturer's instructions for the use of alteplase (with modifications for clarification) are as follows:

●Step 1: Instill the 2 mL dose into the occluded catheter lumen.

●Step 2: After 30 minutes of dwell time, assess catheter function by attempting to aspirate blood. If the catheter is functional, proceed with step 5 below. If it is not functional, proceed with step 3.

●Step 3: Extend the dwell time 120 minutes and assess catheter function by attempting to aspirate blood. If the catheter is functional, proceed with step 5 below. If it is not functional, proceed with step 4.

●Step 4: If the catheter function is not restored after one dose of alteplase, a second dose of equal amount may be instilled. Follow steps 1 through 3 above. If function is not restored, the catheter should be replaced. If function restored, proceed with step 5 below.

●Step 5: If catheter function has been restored, aspirate 4 to 5 mL of blood to remove the alteplase and residual clot, and gently irrigate the catheter with an 0.9% sodium chloride.

Two alternative administration protocols have been tested.

●A technique referred to as the "push" technique is based upon the fact that the lytic agent within the catheter lumen and not in contact with the circulation is still pharmacologically active. Advancing (pushing) the agent toward the catheter tip by the intermittent injection of saline into the catheter lumen could potentially prolong its pharmacologic action.

●An infusion technique has also been studied in which the lytic agent is administered by continuous infusion over an extended period of time either during or off of dialysis.

Dwell versus push technique — A multicenter trial compared the efficacy of an alteplase "dwell" protocol (n = 43) with a "push" protocol (n = 40) for restoring function to occluded hemodialysis catheters [4]. Catheter dysfunction was defined as either an inability to aspirate the catheter to initiate dialysis or maintain a blood flow (pump speed) of 200 mL/min on dialysis, with arterial and venous pressures not exceeding ± 250 mmHg, respectively. In this study:

●With the dwell protocol, 2 mg of alteplase in 2 mL of saline was instilled to fill both catheter lumens. After 30 minutes of dwell time, catheter function was assessed by attempting to aspirate blood. If the catheter was functional, 4 to 5 mL of blood was aspirated to remove alteplase. The catheter was irrigated with saline, and hemodialysis was initiated. If the catheter was not functional after 30 minutes of dwell time, the alteplase was allowed to dwell in the catheter for an additional 90 minutes, and aspiration as described was again attempted.

●With the push protocol, 2 mg of alteplase in 2 mL of saline plus 0.1 mL of saline was instilled to fill both catheter lumens. After 10 minutes dwell time, 0.3 mL of saline was instilled into each lumen. This was repeated again after an additional 10 minutes had elapsed. Ten minutes later (a total of 30 minutes from initiation), catheter function was assessed by attempting to aspirate blood.

Based upon intention-to-treat analysis, the percentage of patients achieving a post-thrombolytic blood flow rate ≥300 mL/min was higher in the push group (82 percent versus 65 percent). The highest stable blood flow rate post-treatment was a mean of 318 ± 56 mL/min (increase of 194 ± 88 ml/min) in the push group and 302 ± 81 mL/min in the dwell group (increase of 157 ± 110 mL/min). Despite the noted differences, neither the success rate nor post-thrombolytic blood flow were found to be statistically significant, likely due to the study being underpowered.

Continuous infusion — The results obtained with continuous infusion of lytic therapy are no better than those obtained with intermittent intraluminal enzyme instillation. In addition, it is more difficult and more expensive to accomplish. Thus, its overall value is questionable, and it is not generally recommended.

In one series, 55 hemodialysis catheters, which were felt to be related to fibrin sheath development, were treated with an infusion of tPA (2.5 mg of tPA in 50 mL of normal saline was given as a three-hour infusion) through each hemodialysis catheter lumen. The technical success rate, defined as at least one successful dialysis session, was 91 percent. A Kaplan-Meier survival analysis yielded patency rates at 30, 60, 90, and 120 days of 55, 36, 25, and 15 percent, respectively. The dose of enzyme used was not large enough to result in systemic effects, and no complications were encountered.

Referral for intervention — If intrinsic thrombus cannot be successfully resolved within the dialysis facility or if the treatment applied does not have an effect of sufficient duration, the patient should be referred to a vascular specialist for diagnostic evaluation. Appropriate treatment should then be based upon the findings of this evaluation. Treatment of extrinsic thrombus is discussed above.

●In one report, 17 percent of the cases with early catheter failure in a large series of tunneled catheters were attributed to catheter replacement into a preexisting fibrin sheath [101].

●A randomized controlled pilot trial was conducted to investigate the impact of angioplasty sheath disruption on catheter patency and function [102]. Forty-seven long-term hemodialysis patients with secondary, refractory catheter dysfunction underwent guidewire exchange to replace their catheters. Sheaths were present in 33 (70 percent) of the 47 patients. In 18 patients who were randomly assigned to disruption, the median time to repeat dysfunction was 373 days, compared with 97.5 days in patients who did not undergo disruption (p = 0.22), and the median time to repeat catheter exchange was 411 and 198 days, respectively (p = 0.17). Fourteen patients had no sheaths, and their median times to repeat dysfunction and repeat exchange were 849 and 879 days, respectively [102]. Although striking numerical differences were seen between these two groups, the differences did not reach statistical significance because the study was underpowered.

Fibrin sheath disruption can be accomplished by using several different devices; however, the most frequently used technique involves the use of an angioplasty balloon [103-105]. The size of the balloon used has varied from 8 to 12 mm in diameter. By using an angioplasty balloon, it is possible to not only disrupt the fibrin sheath but also dilate any associated venous stenosis.

When the fibrin sheath is left behind, even if disrupted, the question as to what becomes of this material once the catheter is removed has been raised. In an early study that involved catheters that have been in place for a relatively short period of time (mean of 27 days, range of 3 to 111 days), it was found that while the fibrin sheath was adherent to the vessel wall in most cases, in a few cases it was observed to embolize [106]. In 3 of 60 cases, the patients were symptomatic, and lung scans confirmed the occurrence of a pulmonary embolus. In another series of 25 cases [94] in which the temporary dialysis catheter had also been in place for only a short period (mean of 29 days, range of 12 to 48 days), fibrin sheath embolization did not occur in any patient. The material remained adherent to the vein at the catheter insertion site.

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: Dialysis" and "Society guideline links: Hemodialysis vascular access".)

SUMMARY AND RECOMMENDATIONS

●Definition and classification – Catheter dysfunction is a major problem for patients on hemodialysis and can be described according to symptoms (asymptomatic, symptomatic), timing relative to catheter placement (ie, early, late), and size (small: <2 cm; intermediate: 2 to 6 cm; large: >6 cm) and location (extrinsic, intrinsic) of thrombus, the most common cause of dysfunction. (See 'Introduction' above and 'Classification' above.)

•Early catheter failure occurs immediately after placement and is caused by catheter position or technical problems that should be corrected at the time of catheter placement.

•Late dysfunction occurs in a catheter that initially functioned well and is generally the result of extrinsic or intrinsic thrombus.

-Extrinsic thrombus extends outside the catheter into the vascular structure in which the catheter resides.

-Intrinsic thrombus forms within the catheter (intraluminal), at the catheter tip, or surrounds the catheter as a sleeve or sheath. Intrinsic thrombus is the main cause of loss associated with hemodialysis catheters.

●Monitoring and indications of dysfunction – Routine monitoring of parameters suggesting the presence of dysfunction should be performed for all patients who are dialyzing with a central venous catheter. Catheter dysfunction leading to inadequate dialysis has been defined based solely upon blood flow measurement, but trends and other factors are important for determining the adequacy of hemodialysis. Indicators of catheter dysfunction may include: (See 'Catheter dysfunction' above.)

•A decline in blood flow rate of >10 percent, particularly if progressive (measured at a prepump pressure of 250 mmHg five minutes after the start of the session).

•A delivered Kt/V of <1.2.

•An arterial pressure (prepump) more negative than -250 mmHg or a venous pressure (postpump) of >250 mmHg.

●Catheter-related venous thrombosis – Central vein thrombosis may or may not cause symptoms. Duplex ultrasonography is the initial imaging study, although venography may be needed to make the diagnosis. Central vein thrombosis is managed with systemic anticoagulation for three months and may include removal of the catheter. (See "Catheter-related upper extremity venous thrombosis in adults", section on 'Deep vein thrombosis' and "Catheter-related upper extremity venous thrombosis in adults", section on 'Catheter management' and "Overview of thoracic central venous obstruction", section on 'Thrombotic TCVO'.)

●Catheter-related atrial thrombus – Catheter-related atrial thrombus (CRAT) includes mural thrombus and catheter tip thrombus, both of which can lead to late hemodialysis catheter dysfunction. Most CRAT do not cause symptoms and are diagnosed incidentally on cardiac imaging studies. CRAT requires a high index of suspicion and should be considered in hemodialysis patients with cardiac problems (eg, tricuspid regurgitation, right heart failure, cardiac arrest, embolism [pulmonary, systemic], unknown source of infection). CRAT is generally managed according to thrombus size and location individualized according to patient's comorbidities, the urgency of the situation, and the available clinical expertise (algorithm 1).

•For small diameter thrombus (<2 cm) in diameter, we suggest catheter removal only, rather than anticoagulation or thrombolysis (Grade 2C). Catheter removal alone should limit any further increase in thrombus size with time. The catheter can be removed regardless of the thrombus type (mural, catheter tip).

•For intermediate diameter thrombus (>2 cm but <6 cm) in patients with no other indication for surgery, we suggest catheter-directed therapy (CDT), rather than anticoagulation or systemic thrombolysis (Grade 2C). If CDT is not available or is unsuccessful, anticoagulation (eg, apixaban) or systemic thrombolysis can be used depending on the clinical situation. Patients in whom anticoagulation, systemic thrombolysis, and surgery are all contraindicated are monitored for complications.

To manage the catheter associated with intermediate CRAT:

-For mural thrombus, the catheter can be removed as CDT is initiated since the risk of embolization is low. After resolution of thrombus, follow-up imaging should be obtained 7 to 10 days after completion of treatment to rule out recurrence.

-For catheter-tip thrombus, the catheter should remain in place to reduce the risk of embolization until the thrombus has resolved, after which time the catheter can be removed.

•For large thrombus >6 cm, surgery may be indicated for symptomatic patients or those with other contributing indications but is individualized based on the clinical situation. For the patient who is clinically stable without signs and symptoms of cardiac dysfunction, CDT can be attempted first. For those who undergo surgery, the catheter is removed along with the thrombus.

●Intrinsic catheter dysfunction – Intrinsic catheter dysfunction (intraluminal thrombus, catheter tip thrombus, fibrin sheath) is a major cause of catheter loss and should be addressed when identified. Initial management includes using "bedside maneuvers," such as repositioning the patient (placing in a Trendelenburg position), using a "forceful saline flush," or instillation or intraluminal instillation of lytic enzymes. (See 'Treatment approach for intrinsic thrombus' above.)

•The initial treatment consists of forceful saline flushes and intraluminal lytic enzyme instillation if flushing is not effective. Intraluminal tissue plasminogen activator (tPA; 1 to 2 mg) is effective in many patients but provides only a short-term benefit, and more definitive therapy should be sought when this measure is employed. Infusion of urokinase or tPA is more difficult and expensive. (See 'Treatment approach for intrinsic thrombus' above.)

•For patients in whom initial treatment in the dialysis facility is unsuccessful, catheter exchange over a guidewire should be performed. When performing a catheter exchange, we disrupt any fibrin sheath that is present. (See 'Referral for intervention' above.)

●Prevention – Measures to prevent thrombus formation include proper flushing and locking the catheter with an appropriate anticoagulant following dialysis. Routine use of systemic antithrombotic prophylaxis for hemodialysis catheters to prevent catheter dysfunction is not recommended. (See "Central venous catheters for acute and chronic hemodialysis access and their management", section on 'Routine care and access for hemodialysis'.)