CVCs play an important role in haemodialysis but may malfunction due to obstruction, stenosis or thrombosis, catheter plicature or tip malpositioning. This article considers solutions to obstruction
Josette Pengloan MD
University of Tours
Ever more patients are affected by chronic renal insufficiency. At least 1.6 million of them need treatment with haemodialysis, requiring vascular access (VA) which draws blood from the patient to a dialyser then returns it to the patient. In chronic treatment, the VA needs to provide a flow rate of 300-400 ml/min over at least four hours, three times a week.
The gold standard for VA is the native distal arteriovenous fistula (AVF). Nevertheless, ever more patients are treated using a central venous catheter (CVC) either in emergency, or as a bridge before creating or developing an arteriovenous VA. In some cases the CVC becomes a definitive VA, but the CVC must always remain viable for as long as the patient requires it.
The percentage of patients dialysed with a CVC is rising. In the USA in 2004, 57% of patients older than 75 years had a CVC as first access and 15-25% of prevalent patients according to age were dialysed with a CVC. In Europe, the percentage of prevalent patients dialysed with a CVC is also rising.,
Despite the importance of CVC in treating patients it has severe drawbacks which is why international recommendations (the US National Kidney Foundation Kidney Dialysis Outcomes Quality Intiative, and the European Renal Association/European Dialysis and Transplantation European Best Practice Guidelines; see Table 1) are for a reduction in CVC use to less than 10% of prevalent patients. The main complications are CVC-related infection which has high morbidity and mortality and, more commonly, CVC dysfunction. The main causes of dysfunction are obstruction of the CVC or stenosis or thrombosis of the host vessel. Other causes are mechanical, such as catheter plicature (folding) or tip malpositioning. Dysfunction can have severe consequences, including reduction in dose delivered and even, in cases of complete obstruction, preventing completion of a dialysis session. It could be responsible for up to 43% of cases of CVC removal accounting for more cases than those caused by infection (32%) in a prospective study of 812 tunnelled CVs.
Thrombosis development may be intrinsic (intraluminal, catheter-tip thrombus or fibrin sheath) or extrinsic (mural or atrial thrombus or central vein thrombosis).  A thrombus may be more or less extended, resulting in an obstruction which can either be complete, making dialysis impossible, or incomplete, making it difficult to obtain an adequate blood flow.
Thrombus formation and growth may be promoted by malpositioning the catheter tip so it rubs against the blood-vessel wall, or by CVC design. Leakage of anticoagulant through the lateral holes could explain the frequency of clotting at the CVC tip; this has led some investigators to recommend overfilling the CVC with locking solution by 20%. Leakage also promotes unwanted systemic effects due to the locking solution.
There is a close relationship between thrombosis and infection with every device. Experimental data suggest bacteria may preferentially adhere to plateletrich thrombi, and a concomitant increase in such adhesion has been observed as thrombus size increases. Venous thrombosis rate appears to be relatively higher in CVC-associated Staphylococcus aureus bacteraemia,  and prevention of thrombosis using heparin reduces the CVC-related infection rate. This relationship can be easily explained: fibrin straying into CVC lumens or collecting at the tips forms a nutrient for bacteria. In dysfunction, interventions via CVC connectors by nurses trying to resolve the problem raise the risk of introducing a pathogen which may colonise lumens.
Treatment of catheter occlusion
Whatever the degree of obstruction, several manoeuvres have to be undertaken: to try to aspirate the CVC content and to push and pull using saline. If these fail, fibrinolytics must be used. In cases of failure, imaging is recommended to locate the obstruction precisely and allow adequate treatment. This may include CVC repositioning in case of tip malpositioning, treatment of fibrin sheath via pharmacological methods or radiological intervention, or treatment of central venous stenosis. CVC exchange or removal remains the final option.
Lytics are used to convert plasminogen to plasmin, which dissolves fibrin and breaks down clots. The main thrombolytics available are streptokinase, urokinase, rtPA, tenecteplase and reteplase. Streptokinase is highly antigenic with low fibrin affinity, but cannot be used repeatedly due to allergenicity. Urokinase (UK) is indicated for occluded CVC in some European countries. It is obtained from human urine or human neonatal kidney cells grown in tissue culture. Tissue plasminogen activator (tPA) is secreted from normal vascular endothelium. Compared to UK, the recombinant-derived human tPA analogue rtPA has higher fibrin specificity and a shorter plasma half-life. It is FDA-approved for occluded CVC, but not indicated in France. Tenecteplase and reteplase are recombinant forms of tPA with longer half-lives than alteplase.
Efficacy and safety of lytics in occluded CVC
Fibrinolytics have been used successfully for some 20 years to restore patency of occluded CVC, but issues remain relating to dose, concentration, regimen, injection modality (instillation or infusion) and duration of infusion.
In dialysis CVC, several regimens have been proposed: a bolus of 5,000 IU with 70% of patency rate after two infusions of 30 min,12 infusion of 20,000 IU/hour over six hours,13 infusion of 125,000 IU over two hours through one lumen followed by 125,000 IU through the other lumen, and intradialytic high dose of 250,000 IU of UK infused over three hours into the venous chamber during dialysis (with 81% success after the first infusion and 99% after the third infusion).13 We use a simpler modality: 100,000 IU of UK after reconstitution are diluted with 20â€”30 ml 0.9% serum chloride and infused via electric syringe over 30 min through both lumens before a dialysis session. This is a high-dose short-time infusion. In cases of incomplete success and/or suspected fibrin sheath formation, one or two more infusions are performed before the next sessions. The procedure is short enough that the schedule for the dialysis session need not be modified.
Safety, efficacy and dose regimen have been studied in nonhaemodialysis CVC via the COOL study, which confirmed the efficacy of up to two instilled 2 mg/2 ml doses of alteplase, with assessment at 30 and 90 min.
In CVC haemodialysis push protocols  were usually proposed for tPA instillation: 2 mg (1 ml) tPA were injected into the occluded catheter lumen and completed with saline; every 15 min 0.3 ml of saline were injected to move the active enzyme toward the catheter tip; and after three periods aspiration of the CVC content was attempted. Several protocols for alteplase infusion have also been proposed, including 2 mg/h over 4 h, gaining an 85% success rate, and 2.5 mg/h or 5 mg/h over 3 h through each port. Doses could be adapted according to severity of obstruction: 2 mg/h for blocked lines or 1mg/h for sluggish lines.
Lytics also seem to offer a valuable alternative
Treatment in cases of fibrin sheath buildup. Results similar to those for percutaneous fibrin sheath stripping have been reported with high doses of UK (250,000 IU)15 and with rtPA (2.5 mg) with a patency rate of 67% at 30 days.
The efficacy of alteplase compared to urokinase in curative strategies is hard to evaluate. One study found 1 mg of alteplase was likely equivalent in thrombolytic potency to 36,000 IU of urokinase.  Unsurprisingly, comparisons between low doses of UK 5,000 IU vs alteplase 1 mg yield better results for alteplase.
Lytics in preventing dysfunction and infection
At the end of the dialysis session the CVC is usually locked using heparin 5,000 IU/ml as an anticoagulant. Unfractionated and low-molecular-weight heparins stimulate S aureus biofilm formation, but alteplase may slightly inhibit this at 1-2 mg/ml. The ideal routine anticoagulant solution for locking CVC would also have antibacterial properties. rtPA 2 mg has been shown to be superior to heparin 2,000 IU for priming the Quinton CVC, and rtPA 1 mg/ml to be superior to heparin 5,000 IU/ml in paediatric central venous haemodialysis CVC. Some observers suggest lytic use could also reduce the infection rate. However, the cost-effectiveness of such a strategy needs further investigation.
Dysfunction in haemodialysis remains the most common cause of CVC removal. The safety and efficacy of lytics have been demonstrated in curative strategy, but more investigation is needed to evaluate them regarding prevention of CVC obstruction and infection.
1. USRDS. United States Renal Data System. Minneapolis: USRDS Coordinating Center; 2007.
2. Clin J Am Soc Nephrol 2006;1:246-55.
3. Nephrol Dial Transplant 2008:1-8.
4. Nephrol Dial Transplant 2008;23(1):275-81.
5. Semin Dial 2001;14(6):441-5.
6. Nephrol Dial Transplant 2007;22(12):3533-7.
7. ASAIO J 2000;46(6):S63-8.
8. Crit Care Med 2008;36(2):385-90.
9. BMJ 1998;316;969-75.
10. Nephrol Dial Transplant 2007;22;5 Suppl 2:ii88-ii117.
11. Am J Kidney Dis 2006;48;1;Suppl 1: S248-57.
12. J Vasc Interv Radiol 2004;15:575-9.
13. Nephrol Dial Transplant 1998;13:2203-6.
14. J Clin Pharm Ther 2004;29(6):517-20.
15. JVIR 2000;11(9):1121-9.
16. JVIR 2000;11:1131-6.
17. J Thromb Thrombolysis 2001;11(2):127-36.
18. Nephrol Dial Transplant 2006;21:2247-55.
19. Am J Kidney Dis 2000;35(1):130-6.
20. Arch Dis Child 2007;92(6):499-501.