Three patients were identified by either physical exam or image study to have a thrombosed or partially thrombosed arteriovenous fistula estimated to have happened in the past 48 hours. Patients who had a recent bleeding event (eg. gastrointestinal, intracranial) were excluded. Written consent was obtained from all participants prior to infusion of tPA in the dialysis unit.

Alteplase (tPA) was prepared to a final concentration of 1 mg/ml by withdrawing 2.2 ml of sterile water and injecting the water into a 2 mg vial containing alteplase in a powder form. The reconstituted solution should be used within 8 hours and stored at 3 to 30°.

An ultrasound or fistulogram were reviewed and the location of the clot was identified. If the clot was too close to the anastamosis site (Figure 1), the arterial needle was placed retrograde into the fistula (pointing towards the anastamosis), and the dialysis circuit was put into reverse flow (blood is infused back through the arterial needle and withdrawn from the venous needle).

Similarly, if the clot was between the arterial needle insertion site and the venous needle insertion site (Figure 2), the flow will be reversed again but this time both needles will be placed antegrade, away from the anastomosis.

However, if the clot was beyond the venous needle site in the venous outflow towards to the heart, needles were placed antegrade and the patient was dialyzed in the conventional circuit where blood is withdrawn from the arterial needle and returned via the venous needle. Once the needles were in place, tPA was infused via a Y-connector attached to the dialysis needle. A dose of 2 mg of tPA in 50 ml of normal saline was infused over 30 minutes if the arterial needle was used as the venous return, to reduce recirculation time. If the venous needle was used as the venous return, tPA was infused over 60 minutes. Once the tPA is infused, the circuit was returned back to conventional mode. The needle size was not altered from the patient’s usual needle size used for cannulation of their fistula. The extra-corporeal heparin dose used during dialysis was not changed. We defined the tPA infusion procedure as being successful based on changes in the physical assessment of the AVF (softness, presence of bruit and thrill) and on resolution of clot burden assessed by either ultrasound, fistulogram or absence of clot aspiration on needling the fistula.

Case 1

A 66 year old man presented to our outpatient hemodialysis unit with a thrombosed left radio-cephalic AVF. A month prior he had an access flow of 1190 ml/min. He underwent an urgent fistulogram (Figure 3a) that demonstrated a completely occluded fistula. An angioplasty was performed. Partial recovery of flow was achieved but heavy residual clot burden was seen (Figure 3b). Once back in the hemodialysis unit, the fistula was cannulated and tPA was infused as per protocol. Following the tPA infusion the patient was dialyzed successfully. A fistulogram one week later demonstrated no clot but there was evidence of residual stenosis still present (Figure 3c). This was successfully treated with repeat angioplasty. Access flow evaluation a month after angioplasty was 1420 ml/min. He was on ASA 81 mg once daily during this time period. He has since passed away from a malignancy related complication.

Case 2

A 66 year female due for her regular dialysis, was found to have no thrill on careful physical examination of her left radio-cephalic AVF and dark clotted blood was aspirated on cannulation. Her access flow was adequate a month prior to presentation at 1020 ml/min. She underwent an urgent fistulogram and angioplasty but the procedure was deemed unsuccessful since adequate flow was not established and there was evidence of large residual clot burden (Figure 4a).

She returned the following day for insertion of a dialysis catheter. On clinical assessment of the fistula, a faint thrill was present. It was felt she could have a salvageable fistula and likely had vessel spasm post angioplasty, which can make it difficult to judge the viability of fistulae immediately post procedure. An ultrasound (Figure 4b) showed a large clot in the venous aneurysm. She had tPA infused as per our protocol on three consecutive days. Significant reduction of clot burden was demonstrated a week later with a repeat fistulogram (Figure 4c). The fistula was needled and successful dialysis followed. No clot burden was demonstrated on repeat ultrasound in vascular access clinic. A follow up access flow measurement a month later was 940 ml/min. She has not required any further interventions to date. She was not on any oral anticoagulants.

Case 3

A 73 year old female from our Kidney Care Clinic, (pre-dialysis, CKD stage 5), with a mature brachiocephalic AVF was assessed at 6 months post fistula creation. On physical examination of her AVF, she was found to have changes on palpation and auscultation of her AVF suggestive of a venous outflow stenosis. A fistulogram confirmed a venous outflow stenosis. Successful angioplasty of the stenosis followed. A few months later she complained of persistent pain at the fistula site. Physical exam demonstrated a hard marble like structure within the AVF. Dialysis was attempted unsuccessfully. A repeat fistulogram indicated persistent stenosis mid-humerus (Figure 5a) and was treated with angioplasty. A follow up ultrasound a week later showed residual clot (Figure 5b). She underwent four tPA infusions in the dialysis unit, on four separate days. Decrease of clot burden was evident on a repeat ultrasound (Figure 5c). On physical assessment, the marble like structure was absent and persistent pain resolved. Successful dialysis treatments were initiated. First access flow was 970 ml/min after initiation of dialysis. She underwent a repeat angioplasty for a stenosis 2 months later.

1. Kloss K, Shusterman N. Bolus administration of streptokinase to declot thrombosed arteriovenous accesses. ANNA J. 1985;12 suppl 5:313–7.PubMedGoogle Scholar
2. Schon D, Mishler R. Pharmacomechanical trombolysis of natural vein fistulas: reduced dose of TPA and long term follow up. Sem Dialysis. 2003;16 suppl 3:272–5.View ArticleGoogle Scholar
3. Sofocleous CT, Cooper SG. Retrospective comparison of the amplatz thrombectomy device with modified pulse-spray pharmacomechanical thrombolysis in the treatment of thrombosed hemodialysis access grafts. Radiology. 1999;213:561–7.View ArticlePubMedGoogle Scholar
4. Cynamon J, Lakritz PS. Hemodialysis graft declotting; description of the ‘lyse and wait’ technique. J Vasc Interv Radiol. 1997;8:825–9.View ArticlePubMedGoogle Scholar
5. Patrick M, Vogel M, Vipin B. Thrombosed hemodialysis grafts: Lyse and wait with tissue plasminogen activator urokinase compared to mechanical thrombolysis with the Arrow-Trerotola percutaneous thrombolytic device. J Vasc Interv Radiol. 2001;12:1157–65.View ArticleGoogle Scholar
6. United State Renal Data System. URDS annual data report. The National Institutes of Health, National Institute of Digestive and Kidney Disease. Am J Kidney Dis. 1999;34:S9-S19.Google Scholar
7. Valji K, Bookstein JJ. Pharmacomechanical thrombolysis and angioplasty in the management of clotted hemodialysis grafts: early and late clinical results. Radiology. 1991;178:243–7.View ArticlePubMedGoogle Scholar
8. Bent C, Sahni V. The radiological management of the thrombosed arteriovenous dialysis fistula. Clin Radiol. 2011;66:1–12.View ArticlePubMedGoogle Scholar
9. Falk A, Mitty H, Guller J. Thrombolysis of clotted hemodialysis grafts with tissue-type plasminogen activator. J Vasc Interv Radiol. 2001;12:305–11.View ArticlePubMedGoogle Scholar
10. Constantinos T, Sofocleous M, Clay R. Alteplase for hemodialysis access graft thrombolysis. J Vasc Interv Radiol. 2002;13:775–83.View ArticleGoogle Scholar