In-Depth Reviews

Catheter Management in Patients: Delivering Adequate Flow

Anatole Besarab and Rahul Pandey

Summary Over 330,000 individuals in the United States depend on hemodialysis (HD), the majority as a result of end- stage renal disease. Sustainable vascular access can be achieved through arteriovenous fistulas, arteriovenous Division of Nephrology grafts, or tunneled . Tunneled dialysis catheters (TDCs) often remain in use for months or even and Hypertension, years, long beyond their initial intended use as a bridging device. Research efforts are focused on identifying Department of strategies to prevent/minimize the risk of the most common -related complications: thrombotic oc- Medicine, Henry Ford clusion and infection. Thrombotic occlusion of TDCs prevents adequate dialysis but can be managed success- Hospital, Detroit, Michigan fully through thrombolytic agents to restore/improve flow in the majority of patients, allowing immedi- ate HD delivery and prolonging usability of the TDC. Occasionally, catheter exchange with fibrin sheath Correspondence: disruption is needed to preserve the site. Surface-treated catheters could improve the morbidity and mortality Dr. Anatole Besarab, associated with HD delivery via an indwelling catheter, but results from studies have been disappointing to Division of Nephrology date. We review the etiology of catheter-based access failure and the monitoring and interventional steps that and Hypertension, should be taken to maintain the patency and safety of catheters for HD. Wherever possible we note the areas Department of Medicine, Henry Ford Hospital, in which there is scant data where further randomized clinical trials are needed. 2799 West Grand Clin J Am Soc Nephrol 6: 227–234, 2011. doi: 10.2215/CJN.04840610 Boulevard, Detroit, MI 48202. Phone: 313-916- 2713; Fax: 313-916- 2554; E-mail: abesara1@ Introduction den (13,14). A retrospective analysis of resource use hfhs.org Hemodialysis (HD) is a life-saving and life-sustaining among 88 HD patients found that 51% experienced at procedure. In 2006, approximately 330,000 individu- least one access-related complication during fol- als in the United States were receiving HD (1). Sus- low-up (mean 487 days) (15). A follow-up of 674 tainable vascular access providing high-volume blood patients from 22 chronic HD units showed a signifi- flow rates (Qb) Ͼ300 ml/min is essential, either cant association between decreased dialysis adequacy through permanent arteriovenous access or tunneled and increased hospitalization because of complica- dialysis catheters (TDCs) (2). The majority of patients tions (11%), number of hospital days (12%), and in- begin HD with a TDC (3). In the United States, 60 to patient expenditure (US$940) (16). Of all accesses, 82% of incident HD patients start with a catheter (4,5). TDCs produce the majority of problems related to Up to 33% of patients in Canada are dialyzing with a delivering adequate blood flow for dialysis and re- long-term TDC (6). quire the most interventions (2,5). TDCs allow immediate vascular access, are initially almost always functional, require no venipunctures, Defining Access Dysfunction rarely produce hemodynamic consequences, and do The 2006 National Kidney Foundation Kidney Di- not require surgical placement (7). Despite efforts to alysis Outcomes Quality Initiative (KDOQI) guide- limit their use, TDCs often remain in use for months lines define access dysfunction as the inability to or even years (8). Studies have shown that approxi- achieve Qb of Ն300 ml/min (2) during the first 60 mately 40% of patients had TDCs 90 days after com- minutes of HD despite at least one attempt to improve mencing HD (5,9,10). flow. Since then, larger bore catheter design allows Proper catheter management to preserve patency much higher Qb (Ͼ400 ml/min) to be achieved at the and to prevent/minimize the risk of infection is vital same prepump pressure. Waiting until Qb declines to in improving patient outcomes (11). This article fo- 300 ml/min in these catheters may be inappropriate, cuses on the first of these, reviewing the etiology of missing the opportunity to detect catheter dysfunc- inadequate blood flow delivery and discussing the tion earlier. However, studies to support this hypoth- monitoring and interventional steps needed to main- esis are needed. tain the patency and usability of HD catheters.

Causes of Occlusive Access Dysfunction Access Dysfunction Early identification of catheter dysfunction enables The consequences of blood flow access dysfunction prompt intervention and salvage. Catheter occlusion, in HD can vary from inadequate dialysis dose deliv- a main cause of poor Qb, may be caused by kinking or ery to increased mortality (1,12). Access dysfunction malpositioning (17). In these cases, catheter dysfunc- increases resource utilization and healthcare-cost bur- tion generally emerges during the first HD session www.cjasn.org Vol 6 January, 2011 Copyright © 2011 by the American Society of Nephrology 227 228 Clinical Journal of the American Society of Nephrology

and can be resolved by repositioning or, occasionally, excess of the endogenous fibrinolytic system’s capac- replacing the catheter. In a previously well function- ity develops, catheter thrombosis occurs. ing long-term catheter, inadequate flow may result from a change in position of the catheter tip during Monitoring for Access Dysfunction that session and respond to simply placing the patient Prompt identification of access dysfunction, a key in a recumbent position or adjusting the patient’s goal of the KDOQI guidelines (24), enables appropri- neck position. ate intervention (pharmacologic or mechanical) be- However, thrombotic occlusions, of which there are fore the emergence of access- and life-threatening four major types (Table 1) are a more serious cause of complications. Prospective monitoring for catheter access dysfunction. They occur in 30 to 40% of pa- Qb dysfunction should be a routine part of the med- tients, can occur within 24 hours after insertion or ical management of patients undergoing HD. Poten- after prolonged continuous successful usage (18), and tially dysfunctional catheters and reduction of Qb to can provide a substrate for bacterial growth (19). A “critical” levels can be detected before “emergency” study of 721 HD patients revealed that clot formation problems arise through systematic monitoring of Qb was one of four parameters, significantly (P Ͻ 0.001) and prepump negative arterial pressure (Pa) during and independently related to inadequate dialysis HD. The catheter channel is a long tube whose diam- dose delivery (20). eter and length determine resistance to flow (25). There is a curvilinear relationship between pressure Pathophysiology of Catheter Dysfunction and flow; the prepump pressure is virtually a nega- Injury to the vessel endothelium begins with inser- tive mirror of the venous pressure at Qb from 50 to tion of the catheter and is augmented by turbulent 500 ml/min (Figure 1a). Most large-gauge catheters flow around the catheter. “Line” reversal or catheter have a conductance (Qb/Pa) of 2 ml/min/mmHg. ”manipulation” as attempts to improve blood flow When examined serially over time at a prescribed Qb promote even more disruption in the fibrinolytic sys- (e.g. 350 to 450 ml/min), increasing negative prepump tem, initiating the coagulation and inflammatory cas- pressure to achieve the prescribed flow reflects alter- cade (21). Minute irregularities on the catheter poly- ations in inlet orifice TDC function. mer surface permit platelet adhesion and activation of We recommend the measurement of Qb at a preset the intrinsic coagulation pathway (17). Silicone may prepump pressure (e.g. Ϫ250 Ϯ 10 mmHg at each HD have less throbogenic potential than other materials (22). session 5 minutes after the start of each HD session It is the development of a fibrin sheath that deter- with trending over time. A Qb decline of Ͻ10% at the mines the long term patency of a catheter. This sheath, same negative prepump pressure may result from initially composed of fibrinogen, albumin, lipopro- many factors, but a change Ͼ10%, particularly if pro- teins, and coagulation factors, begins to form within gressive, i.e. lower conductance, suggests impending 24 hours of insertion (23). The fibrin sheath attracts access dysfunction, which may warrant intervention platelets and coagulation factors and promotes leuko- (Figure 1b). cyte adherence (21). Over weeks and months, collagen Qb decline to Ͻ300 ml/min during the first and/or is deposited as smooth muscle cells from the venous last 30 minutes of an HD session, delivered Kt/V of vessel wall migrate toward the tip. The rate of these Ͻ1.2, Pa more negative than 250 mmHg, and venous processes varies among patients because of inherited pressure of Ͼ250 mmHg are other signs of dysfunc- or acquired characteristics. Ultimately, if clotting in tion. Routine monitoring of these parameters with

Table 1. Types of thrombotic occlusions

Type Features Symptoms

Fibrin tail or flap Fibrin extends from the end of the Ability to infuse but not catheter causing partial occlusion withdraw blood (fibrin tail acts as one-way valve) Fibrin sheath Fibrin adheres to the external surface Inability to infuse and/or encasing the catheter, possibly withdraw blood extending the length of the catheter; thrombi trapped between sheath and catheter tip Mural thrombus Fibrin from vessel wall injury binds to Leakage of infusate from the fibrin-covered catheter; increased insertion site, swelling, risk of venous thrombosis pain, tenderness, engorged vessels Intraluminal Fibrin forms inside catheter lumen Inability to infuse and/or thrombus causing partial or complete withdraw blood occlusion Clin J Am Soc Nephrol 6: 227–234, January, 2011 Hemodialysis Catheter Flow, Besarab and Rahul 229

a) the end of dialysis had deranged clotting 1 hour after 250 200  completion of dialysis. Heparinized , using a 150   lower total dose of heparin per catheter lumen, sig- Pv  100  nificantly reduced this rate. A lower-concentration   50   (1000 versus 10,000 U/ml) heparin lock was associated 0  with a higher use of tissue plasminogen activator -50   (tPA) (35). Citrate locks do not produce this apparent -100 

Pressure (mmHg) 

Pa -150  systemic effect (27,29,36).  -200  Currently there is great interest in trisodium citrate -250 0 50 100 150 200 250 300 350 400 450 500 (TSC), given its antithrombotic and potential antibac- Qb (mL/min) terial properties (37). A study in 19 patients (1370 b) sessions) showed that a low-concentration 4% citrate 2 lock was associated with a higher rate of catheter

1.5 thrombosis than standard heparin but had no effect on dialysis efficiency, effective blood flow, the use of Qb/Pa 1 thrombolytic therapy, or safety (38). A subsequent double-blind, crossover study in 28 patients showed 0.5 that a 5% citrate lock was just as efficacious as a 10% 0 solution (39). 01234 Months Two prospective observational studies conducted in Canada (40,41) showed either equivalence or a lower rate of TDC exchange and tPA use without a Figure 1. | (a) Relationship of prepump pressure (Pa) and ve- change in hospitalization for TSC 4% versus heparin. nous return pressure (Pv) to blood pump flow. Note that the The largest experience reported is that from Grudz- two curves are nearly mirror images of each other, although inski et al. (42) during a conversion study from hep- the negative Pa pressure has an absolute value slightly higher arin 10,000 U/ml to 4% citrate locks. The rate of than Pv. The dashed lines shows the effect of orifice obstruc- tion at either the inflow or outflow ports. The circles indicate flow-related catheter exchange, tPA use, and bactere- the effect of “deliberate” clamping of the tubing to the arterial mia did not differ in the year of heparin locks com- transducer to obviate machine alerts. Prepump pressure be- pared with the subsequent year of citrate locks with comes independent of flow. (b) Sequential measurements of over 30,000 catheter days at risk in each period. Un- pump flow (Qb) at the same prepump pressure early into expectedly, a higher concentration of TSC (47%) pro- dialysis. duced a significantly higher catheter thrombosis rate (measured by the use of a urokinase lock) versus stan- immediate alerts to physicians should be encouraged dard heparin lock (43). for all HD patients with a TDC. The antimicrobial activity of TSC is concentration dependent (37), and higher concentrations would be Maintaining Patency with Catheter Locking Solutions desirable. However, there is concern over using higher- between Dialysis Sessions concentration TSC, and currently the Food and Drug Ad- The standard procedure for maintaining patency ministration (FDA) restricts tricitrosol in the United States between dialysis treatments has been heparin instil- to Ͻ4% (44). Questions remain regarding the optimal lock- lation (1000 to 10,000 U/ml) into the lumens in a ing frequency and dose of any agent to optimize catheter volume sufficient to fill to the lumen tip (the lock). patency (45). The heparin concentration used for locking has been The American Society of Diagnostic and Interven- reduced because catheter lumens have increased in tional Nephrology Clinical Practice Committee (46) volume so as to reduce the possibility of unintentional recommends using a locking solution of 1000 U/ml systemic anticoagulation of patients (26–29). Locking heparin or 4% TSC to maintain TDC patency. The with lower heparin concentrations (2500 or 1000 need for tPA for maintaining catheter patency may be U/ml) prevents catheter thrombosis as efficiently as increased by using 1000 U/ml heparin lock compared 5000 U/ml (30,31). Heparin loss (“leak phenomenon”) with higher heparin concentrations, but safety consid- through diffusion from the lumen within the first erations mandate that higher concentrations of hepa- hour or after injecting the locking solution (slow ver- rin lock be reserved for patients with evidence of sus fast) may accidentally administer up to half the catheter occlusion or thrombosis after 1000 U/ml hep- anticoagulant systemically (27) and may explain re- arin. ports of bleeding episodes after heparin lock (32,33). No catheter seems to be free of this problem. An in vitro study of five different double lumen catheters Managing Thrombotic Access Dysfunction with showed that systemic leakage was greatest with 20% Anticoagulants and Lytic Agents overfill (34). Thrombus formation within or at the tip of an HD The degree of systemic anticoagulation appears to catheter can be resolved by catheter replacement be proportional to the dose of heparin instilled. In a alone, catheter replacement with balloon disruption small study in patients undergoing heparin-free dial- of the fibrin sheath, or stripping of the fibrin sheath ysis, all patients with catheters and heparin locks at from a femoral approach (47). In a randomized 230 Clinical Journal of the American Society of Nephrology

clinical trial pilot study in 47 patients, 70% of whom dwell (repeated if the first instillation is unsuccessful) had sheaths, disruption of the sheath compared with has been the most common protocol studied. In some no disruption prolonged the median time till repeat protocols the lytic is slowly pushed in by injection of dysfunction from 97 to 373 days, modestly increased saline 0.2 ml/lumen every 10 to 15 minutes to ad- mean blood flow, and produced a higher URR, 72% vance the lytic to the catheter tip during the 1-hour versus 66% (48). The choice of technique should be dwell. This strategy decreased the need for repeat guided by factors including cost and patient and phy- lytic dwells by 81% while maintaining the proportion sician preference. Catheter replacement with sheath of successful declottings (56). disruption is invasive, inconvenient, costly, and time In all of these instances, lytic use is intermittent and consuming and increases patients’ risk for additional conditioned by catheter performance at any moment complications, but in our opinion it produces more in time. The use of lytics as locking solutions between durable results. all dialysis sessions has been recommended by Prophylactic warfarin has shown some effect in Gabutti et al. (57). Postdialysis lock with urokinase reducing thrombus formation rates in patients with a versus conventional heparin (5000 U/lumen) was as- TDC (49–51). The importance of adequate systemic sociated with increased Qb and reduced occurrence of anticoagulation (international normalized ratio, 1.5 to dysfunction. 2.0) is crucial (49). Malfunction-free catheter survival If maximizing flow is the key issue, then the study at 9 months was 47.1% in patients with adequate of Twardowski (58) has particular relevance; Twar- anticoagulation compared with 8.1% in patients with- dowski’s study sought to minimize the occurrence of out (P ϭ 0.01). However, the risk of bleeding and the inadequate Qb (Ͻ400 ml/min during any dialysis need for regular monitoring and possibly dose adjust- session). Locking the catheter with 5000 to 9000 U of ment makes this a less than ideal therapy for routine urokinase between treatments was successful in re- use in HD (33). Moreover, use of warfarin in HD storing flow in only three of 21 instances, whereas patients has become controversial, because warfarin infusion of 20,000 to 40,000 U of urokinase into the may promote vascular calcification (52). catheter restored function in 10 of 25 instances. Even Of the commonly available antiplatelet agents, nei- when there was an inability to aspirate from the cath- ther acetylsalicylic acid or dipyrimadole have shown eter at all, infusion rather than locking was successful consistent efficacy in preventing TDC thrombosis. in restoring the ability to dialyze with adequate blood Clodiprogel has not been systematically studied. flow in eight of nine patients. Twardowski suggested Noninvasive pharmacotherapy with lytics has a “prophylactic” strategy for lytics. Whenever a non- proved to be effective in restoring catheter patency, positional progressive deterioration of blood flow and a number of lytic agents are currently available was noted, 250,000 U of urokinase was infused during (53). However, there are no thrombolytic agents ap- dialysis over 3 hours, if there were no contraindica- proved for the clearance of occluded TDCs . Use of tions. Successful full restoration of Qb (Ͼ400 ml/min) lytics is on the basis of data of thrombus clearance was achieved in 132 of 162 (81%) instances. In those found in other smaller central venous catheters. failing to achieve the desired Qb, repeat infusions Safety issues with urokinase for this purpose led to its increased the success rate. Whether the increased cost withdrawal in the United States in 1999 (54). Both the of intradialytic high-dose urokinase could be offset by manufacturers and subsequently the FDA warned the high probability of positive results, saving on that streptokinase should not be used for thrombo- transportation expenses and nursing and patient lytic clearance of indwelling catheters because of the time, and avoidance of catheter stripping or catheter risk of allergic reaction and the frequency of serious exchange was not formally studied. adverse events (55). By default tPA (alteplace) became The remaining discussion will focus on the recom- the only approved agent for thrombus clearance of binant tPA alteplase, reteplase, and tenecteplase (an central venous catheters. emerging thrombolytic agent).

Alteplace Efficacy of Lytic Therapy in Catheter Clearance A variety of protocols are used by dialysis centers Two studies have demonstrated efficacy for alte- to restore TDC blood flow. Frequently a lytic is used plase 1 mg/ml instillation in restoring the patency of as an intradialytic dwell of 1 hour when the Qb occluded HD catheters (59,60). A single instillation of achieved (Ͻ250 ml/min) would be unable to provide low-dose alteplase with a 30-minute dwell restored or adequate dialysis even after moderate extension of maintained Qb Ͼ300 without line reversals in 36 of 50 treatment time. Two strategies attempt to improve Qb (72%) patients, with a second instillation restoring before development of an “emergency” TDC flow patency for a further four patients (40 of 50 patients; problem: so-called pre-emptive postdialysis lytic lock 80%) (59). The majority of patients (70%) required or intradialysis lytic infusions. There have been no additional intervention within the 4-month follow-up comparative outcome studies of the various strate- of further alteplase instillation (62%; median time to gies. next exposure 14 days) and/or radiologic interven- Short-term dwells should produce suboptimal re- tion (59). The financial savings over replacement of sults because diffusion of lytics to the site of occlusion dysfunctional catheters in these 50 patients was esti- (thrombus or fibrin sheath) is limited, yet a 1-hour mated at approximately 22,000 Canadian dollars. In Clin J Am Soc Nephrol 6: 227–234, January, 2011 Hemodialysis Catheter Flow, Besarab and Rahul 231

the second study, Qb of Ͼ300 ml/min was completely catheters; catheters remained patent to 30 days in 54% restored in 23 of 62 (37%) episodes of low Qb (Ͻ250 of patients and to 45 days in 33% of patients (67). ml/min; mean pretreatment Qb, 130 ml/min; and Reteplase has demonstrated similar efficacy in re- mean post-treatment Qb, 320 ml/min) (60). Partial storing the flow of occluded HD catheters in two improvement (mean pre- to post-treatment Qb from retrospective analyses (69,70). Reteplase (0.4 U/lu- 69 to 233 ml/min) was restored for 20 of 62 episodes. men; dwell time, 30 minutes) restored/improved Qb Clearly the degree of flow restoration varied inversely in 44 of 50 patients (69); 50 of 59 patients were able to with the pretreatment blood flow. Nineteen episodes complete HD in another study (70). failing to respond to alteplase installation therapy These published studies of thrombolytics for the were shown to have malposition or sheath formation treatment of HD catheter flow dysfunction have been (60). Neither low-dose study systematically collected limited by small numbers of patients, differing dosing safety data. In two other studies, a single instillation regimens, lack of control data, and inconsistent defi- of a higher dose (2 mg/ml) restored Qb to Ն200 nitions of treatment success (71), with success often ml/min in 15 of 18 episodes (83%; mean post-treat- defined as flow rates that would be considered as ment Qb, 260 ml/min) (61) and in 49 of 56 episodes dysfunctional according to the current KDOQI guide- (88%; mean post-treatment Qb, 248 ml/min) among lines (2,24). More recent studies with tenecteplase, 22 patients (62). The mean flow rate in both of these which has increased fibrin specificity, greater resis- studies would be regarded as dysfunctional using tance to plasminogen activator inhibitor-1, and a current guidelines (2), reflecting the previous ten- longer half-life, have adopted more rigorous defini- dency to delay thrombolytic therapy until the dys- tions of treatment success (72,73). functional process was very advanced. Tenecteplase was evaluated in two large pivotal Studies have shown alteplase more effective than studies: TROPICS (Tenecteplase for the Restoration of urokinase in restoring flow to occluded HD catheters Function in Dysfunctional HD Catheters) 3 (72) and 4 Ͻ (62,63). Single dwell instillation (up to 60 minutes) of (73). In both trials, the KDOQI Qb criteria of 300 Ն low-dose (1 mg/ml) alteplase led to significantly ml/min and a flow of 25 ml/min less than pre- more patients achieving Qb Ͼ300 ml/min than with scribed were used for inclusion. Acute therapy as well urokinase (5000 U/ml) (70% versus 35%; P ϭ 0.013) as extended interdialytic dwell were evaluated on and completing an HD session (93% versus 70%; P ϭ TDC delivered blood flow in both studies. The pri- Ն 0.023) (63). The mean post-treatment Qb was 291 ml/ mary efficacy end point was defined as Qb 300 Ն min in the alteplase group and 203 ml/min in the ml/min and an absolute increase of 25 ml/min urokinase group. Using a 30-minute push protocol, from baseline after the 1 hour lytic dwell that per- alteplase restored Qb (Ͼ200 ml/min) in 92% of par- sisted at 30 minutes before and at the end of HD. If tially occluded catheters (mean post-treatment Qb 260 both of the above criteria were not achieved, open label tenecteplase, given as an extended intradialytic ml/min) and 85% of completely occluded catheters dwell period (up to 72 hours), was evaluated. (mean post-treatment Qb 246 ml/min) (64). An anal- An overnight extended dwell was more effective in ysis of 14 patients who had previously received uroki- restoring adequate flow (Ͼ300 ml/min) than the short nase for catheter occlusion revealed that alteplase was dwells. Overall treatment success (Qb of Ͼ300 ml/ significantly more likely to partially restore the flow min and increase by 25 ml/min) was achieved in 22% (Qb Ͼ200 ml/min) of completely occluded catheters, of patients after single 1-hour dwell (no extended 88.2% versus 42.8% (64). dwell; TROPICS 3), 40% when the initial 1-hour dwell Optimal dwell times for lytics have yet to be for- was followed by extended dwell in TROPICS 3, and mally defined. Indeed, both short (1 hour) and long 49% when the initial dwell was followed by an inter- (Ͼ48 hours) dwell times preserved catheter patency dialytic dwell in TROPICS 4. for the subsequent HD session; patency was pre- served for a mean of 14 days with both protocols (65). Delivery of Lytic Therapy as an Interdialytic The long-term patency rate after alteplase instillation “Locking” Solution remains to be determined. In a 36-month follow-up of Five studies suggest that use of tPA as a catheter 570 HD catheters, the overall catheter half-life was 10.2 lock solution may effectively reduce the risk of vessel months, the median time from first to second alteplase occlusion between HD sessions with no increased risk instillation was 27 days, and the period from second to for bleeding events (74–76). In a prospective, double- subsequent instillations was 10 to 18 days (66). blind study of alteplase versus heparin among nine Alteplase infusion protocols also successfully re- children, alteplase 1 mg/ml between HD sessions store HD catheter flow, even when associated with significantly reduced the risk for clot formation (P ϭ fibrin sheaths (67,68). Administration of alteplase 2.5 0.001) (76). In the following year, using interdialytic mg per over 3 hours resulted in a 91% success tPA locks, a significant reduction in both the inci- rate (effortless aspiration and infusion capability dence of blocked lines requiring tPA infusion (P Ͻ through both lumens) for the subsequent HD session, 0.03) and the need for surgical replacement (P Ͻ 0.02) but primary patency rates were relatively short (68). was observed. In a separate study, a 2-mg tPA locking Infusion of alteplase 2.5 mg/h/port over 2 hours solution preserved catheter performance between HD (total dose 10 mg) restored flow to Ͼ250 ml/min in all sessions in 10 patients undergoing HD (75). Further 25 patients presenting with poorly functioning HD evaluation of lytic locking solutions is warranted . 232 Clinical Journal of the American Society of Nephrology

An observational study evaluated reteplase as an advantages in the acute setting, acting as a bridge to interdialytic lock solution (77). The mean Ϯ SD dura- more permanent vascular access, and are useful for tion of catheter patency was 45 Ϯ 39 days. Among 85 patients ineligible for arteriovenous fistula or arterio- episodes of catheter dysfunction, higher reteplase venous graft. Indeed, continued improvements in the doses (4 to 6 U versus 1 U) were associated with a design and performance of catheters along with poor higher but not statistically significant rate of restored access preparation before initiation of dialysis sug- catheter function (91% versus 84%, respectively). gests that they will continue to be the initial method of access for the majority of patients initiating long- Unmet Needs and Unanswered Questions with Lytic term HD. Therapy in Catheter Clearance Attention to monitoring for dysfunction and infec- For patients with Qb Ͻ250 ml/min, acute lytic ther- tion, the two major clinical complications of catheter apy offers a short-term solution to restore indwelling use, as well as prompt intervention to salvage the catheter patency and allows a window of opportunity functionality of the indwelling catheter, are essential to implement alternative HD delivery access (66), em- in preventing or minimizing potential morbidity and phasizing the need to monitor patients to detect early mortality. Lytic therapy to restore patency after signs of occlusive dysfunction. Studies are needed in thrombus formation has proven effective. Future this setting to determine the benefits, with regard to studies should focus on evaluating prophylactic use long-term catheter patency, of early lytic intervention of lytics. on an individual basis using, for example, percentage changes in Qb between HD sessions rather than a Acknowledgments defined Qb to trigger therapy. Support for some third-party writing assistance for this Clinical studies are also needed to determine the manuscript was provided by Genentech, Inc. The content of relative merits of the various approaches to lytic ther- the manuscript was entirely within the control of the au- apy, particularly the value of prophylactic therapy thors. using an interdialytic locking solution between HD sessions for prolonging catheter patency. The Pre- Disclosures CLOT (Prevention of Catheter Lumen Occlusion with Dr. Pandey has no disclosures. Dr. Besarab has been a tPA versus heparin) study may help answer this ques- consultant to Genentech, Inc. in the past and has received tion (78). Patients will be randomized to tPA 1 mg per honoraria and consulting fees. He has lectured on thrombo- lumen once per week, with 5000 U/ml heparin as a lytic use and received honoraria from Roche, Canada. catheter-locking solution for the remaining two ses- sions or to the control arm receiving 5000 U/ml hep- arin as a catheter-locking solution after each dialysis References session. The primary outcome is catheter malfunction 1. U.S. Renal Data System: USRDS 2008 Annual Data on the basis of mean blood flow parameters while on Report: Atlas of chronic kidney disease and end-stage HD, and a cost-effectiveness analysis is planned with renal disease in the United States, Bethesda, MD, Na- tional Institutes of Health, National Institute of Diabetes catheter maintenance using once weekly tPA versus and Digestive and Kidney Diseases, 2007 heparin locking solution. 2. National Kidney Foundation: Clinical practice guide- lines for vascular access: 2006 update. Am J Kidney Dis Current and Future Perspectives 48[Suppl 1]: S176–S285, 2006 Catheters continue to be used acutely and long 3. Bhola C, Lok CE: Central venous catheters: Optimizing the suboptimal. Nephrol News Issues 22: 38–44, 2008 term in a significant proportion of HD patients. HD 4. Pisoni RL, Young EW, Dykstra DM, Greenwood RN, catheter care should routinely involve pre-emptive Hecking E, Gillespie B, Wolfe RA, Goodkin DA, Held surveillance for emerging access dysfunction (in PJ: Vascular access use in Europe and the United States: terms of declining Qb and delivered dialysis dose). Results from the DOPPS. Kidney Int 61: 305–316, 2002 5. U.S. Renal Data System: USRDS 2008 Annual Data Early detection of a failing catheter should permit Report: Atlas of chronic kidney disease and end-stage prompt lytic therapy to salvage the indwelling cath- renal disease in the United States, Bethesda MD, Na- eter, thereby avoiding or at least delaying the need for tional Institutes of Health, National Institute of Diabetes catheter replacement. In patients with a need for fre- and Digestive and Kidney Diseases, 2008 quent lytic therapy, further thrombolysis may not be 6. Mendelssohn DC, Ethier J: Hemodialysis vascular ac- cess problems in Canada: Results from the Dialysis as effective as sheath disruption. The evaluation of Outcomes and Practice Patterns Study (DOPPS II). combined antibiotic and antithrombotic locking solu- Nephrol Dial Transplant 21: 721–728, 2006 tions in prolonging the life of TDCs is ongoing (79,80). 7. Ash SR: Advances in tunneled central venous catheters Catheters surface-treated with antithrombotic coat- for dialysis: Design and performance. Semin Dial 21: 504–515, 2008 ings (81) appear to reduce platelet adhesion and 8. Wang J, LaBerge JM, Chertow GM, Kerlan RK, Wilson thrombus formation on the catheter. However, there MW, Gordon RL: Tesio catheter access for long-term is an absence of clinical data confirming a reduction in maintenance hemodialysis. Radiology 241: 284–290, catheter-related complications (81). 2006 9. Besarab A, Brouwer D: Aligning hemodialysis treatment practices with the National Kidney Foundation’s Conclusions K/DOQI vascular access guidelines. Dial Transplant 33: Although catheters are not regarded as appropriate 694–702, 2004 for long-term HD delivery, they offer a number of 10. Moist LM, Trpeski L, Na Y, Lok CE: Increased hemodialy- Clin J Am Soc Nephrol 6: 227–234, January, 2011 Hemodialysis Catheter Flow, Besarab and Rahul 233

sis catheter use in Canada and associated mortality risk: predictors for its malfunction. Am J Nephrol 28: 298– Data from the Canadian Organ Replacement Registry 303, 2008 2001–-2004. Clin J Am Soc Nephrol 3: 1726–1732, 2008 31. Thomas CM, Zhang J, Lim TH, Scott-Douglas N, Hons 11. Sands JJ: Vascular access: The past, present and future. RB, Hemmelgarn BR: Alberta Kidney Disease Network: Blood Purif 27: 22–27, 2009 Concentration of heparin-locking solution and risk of 12. Owen WF Jr., Chertow GM, Lazarus JM, Lowrie EG: central venous hemodialysis catheter malfunction. Dose of hemodialysis and survival: Differences by race ASAIO J 53: 485–488, 2007 and sex. JAMA 280: 1764–1768, 1998 32. Yevzlin AS, Sanchez RJ, Hiatt JG, Washington MH, Wa- 13. Balwit A: Clinical and Economic Issues in Vascular Ac- keen M, Hofmann RM, Becker YT: Concentrated hepa- cess for Hemodialysis, Madison WI, Ahrens Balwit & rin lock is associated with major bleeding complica- Associates Inc., 2002 tions after tunneled hemodialysis catheter placement. 14. Hakim RM, Breyer J, Ismail N, Schulman G: Effects of Semin Dial 20: 351–354, 2007 dose of hemodialysis on morbidity and mortality. Am J 33. Moritz Ml, Vats A, Ellis D: Systemic anticoagulation Kidney Dis 23: 661–669, 1994 and bleeding in children with hemodialysis catheters. 15. Rocco MV, Bleyer AJ, Burkart JM: Utilization of inpa- Pediatr Nephrol 18: 68–70, 2003 tient and outpatient resources for the management of 34. Sungur M, Eryuksel E, Yavas S, Bihorac A, Layon AJ, hemodialysis access complications. Am J Kidney Dis Caruso L: Exit of catheter lock solutions from double 28: 250–256, 1996 lumen acute haemodialysis catheters: An in vitro study. 16. Sehgal AR, Dor A, Tsai AC: Morbidity and cost implica- Nephrol Dial Transplant 22: 3533–3537, 2007 tions of inadequate hemodialysis. Am J Kidney Dis 37: 35. Holley JL, Bailey S: Catheter lock heparin concentra- 1223–1231, 2001 tion: Effects on tissue plasminogen activator use in tun- 17. Leblanc M, Bosc JY, Paganini EP, Canaud B: Central neled cuffed catheters. Hemodial Int 11: 96–98, 2007 venous dialysis catheter dysfunction. Adv Ren Replace 36. Rioux JP, De Bortoli B, Troyanov S, Madore F: The ef- Ther 4: 377–389, 1997 fect of sodium citrate 4% locking solution for central 18. Whitman ED: Complications associated with the use of venous dialysis catheter on the international normalized central venous access devices. Curr Probl Surg 33: ratio (INR) value. Nephrol Dial Transplant 23: 1772– 319–378, 1996 1773, 2008 19. Fux CA, Uehlinger D, Bodmer T, Droz S, Zellweger C, 37. Weijmer MC, Debets-Ossenkopp YJ, Van De Vonder- Mu¨hlemann K: Dynamics of hemodialysis catheter col- voort FJ, ter Wee PM: Superior antimicrobial activity of onization by coagulase-negative staphylococci. Infect trisodium citrate over heparin for catheter locking. Control Hosp Epidemiol 26: 567–574, 2005 Nephrol Dial Transplant 17: 2189–2195, 2002 20. Sehgal AR, Snow RJ, Singer ME, Amini SB, DeOreo PB, 38. Hendrickx L, Kuypers D, Evenepoel P, Maes B, Messi- Silver MR, Cebul RD: Barriers to inadequate delivery of aen T, Vanrenterghem Y: A comparative prospective hemodialysis. Am J Kidney Dis 31: 593–601, 1998 study on the use of low concentrate citrate lock versus 21. Suajanen JN, Brophy DP, Nasser I: Thrombus on in- heparin lock in permanent dialysis catheters. Int J Artif dwelling cenral vein catheters: The histopathology of Organs 24: 208–211, 2001 fibrin sheaths. Cardiovasc Intervent Radiolog 23: 194– 39. Kuypers DR, Claes K, Evenepoel P, Maes B, Van- 197, 2000 renterghem Y: A prospective, randomized, double-blind 22. Curelaru I, Gustavsson B, hansson AH, Linder LE, Sten- crossover study on the use of 5% citrate lock versus qvist O, Wojciechowski J: Material thrombogenicity in 10% citrate lock in permanent hemodialysis catheters. central vein catheterization II: A comparison between Blood Purif 23: 101–105, 2005 plain silicon elastomer and plain polyethylene, long 40. Lok CE, Appleton D, Bhola C, Khoo B, Richardson RM: antebrachial catheters. Acta Anaesthesiol Scand 27: Trisodium citrate 4%: An alternative to heparin capping 158–164, 1983 of haemodialysis catheters. Nephrol Dial Transplant 22: 23. Mehall JR, Saltzman DA, Jackson RJ: Fibrin sheath en- 477–483, 2007 hances infection. Crit Care Med 41. Macrae JM, Dojcinovic I, Djurdjev O, Jung B, Shalan- 30: 908–912, 2002 sky S, Levin A, Kiaii M: Citrate 4% versus heparin and 24. National Kidney Foundation Dialysis Outcomes Quality the reduction of thrombosis study (CHARTS). Clin J Am Initiative (KDOQI): NFK-DOQI clinical practice guide- Soc Nephrol 3: 369–374, 2008 lines for vascular access, 2000. Am J Kidney Dis 42. Grudzinski L, Quinan P, Kwok S, Pierratos A: Sodium 37[Suppl 1]: S137–S181, 2001 citrate 4% locking solution for central venous dialysis 25. Naka T, Egi M, Bellomo R, Baldwin I, Fealy N, Wan L: catheters: An effective, more cost-efficient alternative to Resistance of vascular access catheters for continuous heparin. Nephrol Dial Transplant 22: 471–476, 2007 renal replacement therapy: An ex vivo evaluation. Int J 43. Power A, Duncan N, Singh SK, Brown W, Dalby E, Ed- Artif Organs 31: 905–909, 2008 wards C, Lynch K, Prout V, Cairns T, Griffith M, 26. Karaaslan H, Peyronnet P, Benevent D, Lagarde C, Rince McLean A, Palmer A, Taube D: Sodium citrate versus M, Leroux-Robert C: Risk of heparin lock-related bleeding heparin catheter locks for cuffed central venous cathe- when using indwelling venous catheter in haemodialysis. ters: A single-center randomized controlled trial. Am J Nephrol Dial Transplant 16: 2072–2074, 2001 Kidney Dis 53: 1034–1041, 2009 27. Vorweg M, Monaca E, Doehn M, Wappler F: The “hep- 44. U.S. Food and Drug Administration: 2000 Safety Alerts arin lock”: Cause for iatrogenic coagulopathy. Eur J An- for Human Medicinal Products. Available at: http://www. aesthesiol 23: 5053, 2006 fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlerts 28. Markota I, Markota D, Tomic M: Measuring of the hep- forHumanMedicalProducts/ucm173095.htm. Accessed arin leakage into the circulation from central venous December 2009 catheters: An in vivo study. Nephrol Dial Transplant 45. Oran NT, Eser I: Impact of heparin locking frequency 24: 1550–1553, 2009 on preventing temporary dialysis catheter dysfunction in 29. Pepper RJ, Gale DP, Wajed J, Bommayya G, Ashby D, hemodialysis patients. J Ren Care 34: 199–202, 2008 McLean A, Laffan M, Maxwell PH: Inadvertent postdi- 46. Moran JE, Ash SR: for the ASDIN Clinical Practice alysis anticoagulation due to heparin line locks. Hemo- Committee: Locking solutions for hemodialysis cathe- dial Int 11: 430–434, 2007 ters; heparin and citrate: A position paper by ASDIN. 30. Brzosko S, Hryszko T, Malyszko J, Malyszko JS, Mazer- Semin Dial 21: 490–492, 2008 ska M, Mys´liwiec M: Femoral localization and higher 47. Janne d’Othe´e B, Tham JC, Sjeiman RG: Restoration of ultrafiltration rate but not concentration of heparin used patency in failing tunneled hemodialysis catheters: A for canal locking of hemodialysis catheter are negative comparison of catheter exchange, exchange and bal- 234 Clinical Journal of the American Society of Nephrology

loon disruption of the fibrin sheath, and femoral strip- alysis catheter dysfunction. Am J Kidney Dis 39: 86–91, ping. J Vasc Interv Radiol 17: 1011–1015, 2006 2002 48. Oliver MJ, Mendelssohn DC, Quinn RR, Richardson EP, 67. Dowling K, Sansivero G, Stainken B, Siskin G, Dolen E, Rajan DK, Pugash RA, Hiller JA, Kiss AJ, Lok CE: Cathe- Ahn J, Mitchell N: The use of tissue plasminogen acti- ter patency and function after catheter sheath disrup- vator infusion to re-establish function of tunneled he- tion: A pilot study. Clin J Am Soc Nephrol 2: 1201– modialysis catheters. Nephrol Nurs J 31: 199–200, 1206, 2007 2004 49. Coli L, Donati G, Cianciolo G, Raimondi C, Comai G, 68. Savader SJ, Ehrman KO, Porter DJ, Haikal LC, Oteham Panicali L, Nastasi V, Cannarile DC, Gozzetti F, Piccari AC: Treatment of hemodialysis catheter-associated fibrin M, Stefoni S: Anticoagulation therapy for the prevention sheaths by rt-PA infusion: Critical analysis of 124 proce- of hemodialysis tunneled cuffed catheter (TCC) throm- dures. J Vasc Interv Radiol 12: 711–715, 2001 bosis. J Vasc Access 7: 118–122, 2006 69. Falk A, Samson W, Uribarri J, Vassalotti JA: Efficacy of 50. Willms L, Vercaigne LM: Does warfarin safely prevent reteplase in poorly functioning hemodialysis catheters. clotting of hemodialysis catheters? Semin Dial 21: 71– Clin Nephrol 61: 47–53, 2004 77, 2008 70. Hyman G, England M, Kibede S, Lee P, Willets G: The 51. Zellweger M, Bouchard J, Raymond-Carrier S, Laforest- efficacy and safety of reteplase for thrombolysis of he- Renald A, Que´rin S, Madore F: Systemic anticoagula- modialysis catheters at a community and academic re- tion and prevention of hemodialysis catheter malfunc- gional medical center. Nephron Clin Pract 96: c39– tion. ASAIO J 51: 360–365, 2005 c42, 2004 52. Danziger J: Vitamin K-dependent proteins, warfarin, 71. Clase CM Crowther MA, Ingram AJ, Cina`CS: Throm- and vascular calcification. Clin J Am Soc Nephrol 3: bolysis for restoration of patency to haemodialysis cen- 1504–1510, 2008 tral venous catheters: A systematic review. J Thromb 53. McFarland HF, Dinwiddie L, Ferrell J, Forloines-Lynn S: Thrombolysis 11: 127–136, 2001 Lytic therapy for central venous catheters for hemodial- 72. Tumlin JA, Goldman J, Spiegel D, Roer D, Adu Ntoso ysis. Nephrol Nurs J 29: 355–360, 2002 K, Blaney M, Jacobs J, Gillespie BS, Begelman SM: A 54. U.S. Food and Drug Administration: Important drug warn- phase III, randomized, double-blind, placebo-controlled ing, urokinase. Available at: http://www.fda.gov/Drugs/ study of tenecteplase for improvement of hemodialysis DevelopmentApprovalProcess/HowDrugsareDeveloped catheter function: TROPICS 3 [Abstract]. JAmSoc andApproved/ApprovalApplications/TherapeuticBiologic Nephrol 20: Abstract F-FC298, 2009 Applications/ucm113512.htm#99. Accessed December 73. Fishbane S, Milligan S, Lempert K, Hertel JE, Wetmore 2009 J, Oliver MJ, Blaney M, Gillespie B, Jacobs JR, Be- 55. U.S. Food and Drug Administration: Important drug warn- gelman SM: A phase III, open-label study of tenect- ing, streptokinase. Available at: http://www.fda.gov/Safety/ eplase for improvement of hemodialysis catheter func- MedWatch/SafetyInformation/SafetyAlertsfor tion: TROPICS 4 [Abstract]. J Am Soc Nephrol 20: HumanMedicalProducts/default.htm. Accessed December Abstract F-PO1555, 2009 2009 74. Gittins NS: Comparison of alteplase and heparin in 56. Seddon PA, Hrinya MK, Gaynord MA, Lion CM, maintaining the patency of paediatric central venous Mangold BM, Bruns FJ: Effectiveness of low dose uroki- haemodialysis lines: A randomised controlled trial. Arch nase on dialysis catheter thrombolysis. ASAIO J 44: Intern Med 92: 499–501, 2007 M559–M561, 1998 75. McGill RL, Blas A, Bialkin S, Sandroni SE, Marcus RJ: 57. Gabutti L, Mosconi E, Pellegrini L, Duchini F, Marone Clinical consequences of heparin-free hemodialysis. C: Chronic and acute consequences of a post-dialysis urokinase lock on permanent hemodialysis catheter Hemodial Int 9: 393–398, 2005 function. J Nephrol 19: 183–188, 2006 76. McGill RL, Spero JA, Sysak JC, Sandroni SE, Marcus RJ: 58. Twardowski ZJ: High-dose intradialytic urokinase to Tissue plasminogen activator as a hemodialysis catheter restore the patency of permanent central vein hemodi- locking solution. Haemodial Int 12: 348–351, 2008 alysis catheters. Am J Kidney Dis 31: 841–847, 1998 77. Dunlay RW, Packard KA, Hilleman DE: Reteplase for 59. Haymond J, Shalansky K, Jastrzebski J: Efficacy of low- dysfunctional hemodialysis catheter clearance. Pharma- dose alteplase for treatment of hemodialysis catheter cotherapy 23: 137–141, 2003 occlusions. J Vasc Access 6: 76–82, 2005 78. Hemmelgarn BR, Moist L, Pilkey RM, Lok C, Dorval M, 60. O’Mara NB, Ali S, Bivens K, Sherman RA, Kapoian T: Tam PY, Berall MJ, LeBlanc M, Toffelmire EB, Manns Efficacy of tissue plasminogen activator for thrombolysis BJ, Scott-Douglas N: Canadian Hemodialysis Catheter in central venous dialysis catheters. Hemodial Int 7: Working Group: Prevention of catheter lumen occlu- 130–134, 2003 sion with rT-PA versus heparin (Pre-CLOT): Study proto- 61. Paulsen D, Reisoether A, Aasen M, Fauchald P: Use of col of a randomized trial [ISRCTN35253449]. BMC tissue plasminogen activator for reopening of clotted Nephrol 7: 8, 2006 dialysis catheters. Nephron 64: 468–470, 1993 79. Steczko J, Ash SR, Nivens DE, Brewer L, Winger RK: 62. Daeihagh P, Jordan J, Chen J, Rocco M: Efficacy of tis- Microbial inactivation properties of a new antimicrobi- sue plasminogen activator administration on patency of al/antithrombotic catheter lock solution (citrate/methyl- hemodialysis access catheters. Am J Kidney Dis 36: 75– ene blue/parabens). Nephrol Dial Transplant 24: 1937– 79, 2000 1945, 2009 63. Eyrich H, Walton T, Macon EJ, Howe A: Alteplase ver- 80. Weijmer MC, van den Dorpel MA, van de Ven PJ, ter sus urokinase in restoring blood flow in hemodialysis- Wee PM, van Geelen JA, Groeneveld JO, van Jaarsveld catheter thrombosis. Am J Health Syst Pharm 59: 1437– BC, Koopmans MG, le Poole CY, Schrander-Van der 1440, 2002 Meer AM, Siegert CE, Stas KJ: CITRATE Study Group: 64. Zacharias JM, Weatherston CP, Spewak CR, Vercaigne Randomized, clinical trial comparison of trisodium ci- LM: Alteplase versus urokinase for occluded hemodialy- trate 30% and heparin as catheter-locking solution in sis catheters. Ann Pharmacother 37: 27–33, 2003 hemodialysis patients. J Am Soc Nephrol 16: 2769– 65. MacRae JM, Loh G, Djurdjev O, Shalansky S, Werb R, 2777, 2005 Levin A, Kiaii M: Short and long alteplase dwells in 81. Dwyer A: Surface-treated catheters: A review. Semin dysfunctional hemodialysis catheters. Hemodial Int 9: Dial 21: 542–546, 2008 189–195, 2005 66. Little MA, Walshe JJ: A longitudinal study of the re- Published online ahead of print. Publication date available peated use of alteplase as therapy for tunneled hemodi- at www.cjasn.org.