566378
research-article2015
JETXXX10.1177/1526602814566378Journal of Endovascular TherapyMaleux et al
Clinical Investigation
Percutaneous Rheolytic Thrombectomy of Thrombosed Autogenous Dialysis Fistulas: Technical Results, Clinical Outcome, and Factors Influencing Patency
Journal of Endovascular Therapy 2015, Vol. 22(1) 80–86 © The Author(s) 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1526602814566378 www.jevt.org
Geert Maleux, MD, PhD1, Bruno De Coster, MD1, Annouschka Laenen, PhD2, Johan Vaninbroukx, MD1, Björn Meijers, MD, PhD3, Kathleen Claes, MD, PhD3, Inge Fourneau, MD, PhD4, and Sam Heye, MD, PhD1
Abstract Purpose: To retrospectively analyze the technical and clinical outcome of percutaneous rheolytic thrombectomy with the use of the AngioJet device in thrombosed autogenous arteriovenous dialysis fistulas. Methods: A cohort of 38 consecutive patients (24 men; mean age 70.7±13.8 years) presenting with 59 thrombotic events in 39 autogenous dialysis fistulas were retrospectively analyzed. The AngioJet rheolytic thrombectomy device was used in all cases. Adjunctive therapies, as well as procedure-related complications, were noted. Primary, assisted primary, and secondary patency of the autogenous fistulas was assessed along with factors potentially influencing patency. Results: Initial technical success to recanalize the efferent vein was 100% (n=59), and a successful postprocedure dialysis session was possible in 97% (n=57) of cases. Adjunctive procedures included percutaneous thromboaspiration (n=4, 7%), balloon angioplasty (n=59, 100%), and stent placement (n=16, 27%). Complications related to thrombectomy occurred in 3 (5%) procedures (distal arterial embolus). Primary, assisted primary, and secondary patency rates at 12 months were 56.1% (95% CI 42.8% to 75.2%), 61.6% (95% CI 48.6% to 74.7%), and 86.2% (95% CI 74.9% to 97.5%), respectively. Risk factors for early fistula occlusion were greater patient age (p=0.045), the age of the fistula (p=0.045), previous stent insertion (p=0.019), and an upper arm fistula (p=0.047). Conclusion: Percutaneous rheolytic thrombectomy of autogenous dialysis fistulas is effective in restoring patency and allowing subsequent hemodialysis. The complication rate is acceptably low, and the large majority of the fistulas are still used for hemodialysis at 1-year follow-up. Older fistulas and upper arm fistulas are at higher risk for early rethrombosis. Keywords dialysis fistula, autogenous arteriovenous fistula, thrombosis, occlusion, rheolytic thrombectomy device, percutaneous thrombectomy, hemodialysis
Introduction Autogenous arteriovenous fistulas are considered the vascular access of choice as defined by the National Kidney Foundation’s Dialysis Outcomes Quality Initiative guidelines1; these recommendations are mainly based on superior patency, lower prevalence of vascular steal syndrome, and lower incidence of access-related infectious complications compared to synthetic grafts. Despite these advantages, early or late thrombosis of the efferent vein still occurs, and immediate revascularization is of utmost importance to provide continued hemodialysis without temporary use of dialysis catheters in this patient population.2 For more than 2 decades, percutaneous recanalization of hemodialysis access grafts has been described as a valuable
alternative to surgical thrombectomy,3 either with the use of various thrombolytic drugs alone4 or in combination with dedicated percutaneous catheter-based devices, including rheolytic5–8 and mechanical thrombectomy devices.9–14 1
Department of Radiology, University Hospitals Leuven / Department of Imaging & Pathology, KU Leuven, Belgium. 2 Interuniversity Institute for Biostatistics and Statistical Bioinformatics, KU Leuven and Universiteit Hasselt, Belgium. 3 Department of Nephrology, University Hospitals, Leuven, Belgium. 4 Department of Vascular Surgery, University Hospitals, Leuven, Belgium. Corresponding Author: Geert Maleux, Department of Radiology, University Hospitals Leuven, Herestraat 49, Leuven, B-3000, Belgium. Email: [email protected]
81
Maleux et al However, data on technical and clinical outcomes of the use of percutaneous thrombectomy devices for thrombosed autogenous dialysis fistulas mostly include the use in a mix of thrombosed autogenous fistulas and synthetic grafts. Additionally, the clinical outcome focuses only on shortterm (6 months) primary patency without reporting assisted primary or secondary patency rates. In this report, the technical and clinical outcomes of percutaneous rheolytic thrombectomy in thrombosed autogenous dialysis fistulas was assessed, including adjunctive thrombectomy procedures, the potential to perform subsequent hemodialysis through the recanalized fistula, complications during and immediately after the procedure, as well as primary, assisted primary, and secondary patency rates of the autogenous fistula. Finally, several factors potentially influencing prolonged patency of the fistula were analyzed.
Methods Study Design This was a retrospective study approved by our institutional ethics committee (S 55351). A search of the interventional radiology database was performed between July 2009 and January 2013 to identify patients undergoing treatment to recanalize a thrombosed autogenous arteriovenous fistula with the use of the AngioJet Rheolytic Thrombectomy System (Boston Scientific, Natick, MA, USA). Patients presenting with graft occlusion or initially treated with other thrombectomy techniques were excluded from the study. Medical data, including dialysis and radiological reports, were reviewed by 2 radiologists (GM and BDC). Data collection included demographics, procedure details, and clinical outcomes related to the potential to perform hemodialysis through the target fistula.
Patient Demographics From July 2009 to January 2013, 38 consecutive patients (24 men; mean age 70.7±13.8 years) presented with 59 events of acute occlusion in 39 autogenous hemodialysis fistulas. All occluded autogenous fistulas were arm fistulas (Table 1). The mean age of the fistula was 3.4 years (0.1– 11.7), and 5 fistulas had a stent previously inserted. No patient had a thrombotic event prior to the first AngioJet thrombectomy procedure. No patients with thrombosed aneurysmal fistulas were identified.
Percutaneous Rheolytic Thrombectomy After clinical evaluation by the attending nephrologist, all recanalization procedures were performed within 24 hours of initial diagnosis of the fistula thrombosis. All procedures were performed by 3 interventional radiologists (GM, JV, and SH). After clinical evaluation of the afferent artery and
Table 1. Types of Occluded Fistulas Treated With the AngioJet Rheolytic Device.a Type of fistula (n=39) Elbow Brachiobasilic Brachiocephalic Brachial artery–median vein Radiobasilic Radiocephalic Radial artery–median vein Wrist Radiobasilic Radiocephalic Side of the fistula Right Left Previous interventions for stenosis
7 (18) 10 (26) 1 (3) 4 (10) 3 (8) 1 (3) 1 (3) 12 (31) 16 (41) 23 (59) 9 (23)
a
Data are presented as counts (percentage).
efferent vein and ultrasonographic confirmation of the thrombosis of the efferent vein, a retro- and/or antegrade puncture was performed under local anesthesia via the Seldinger technique. Through a 6-F high flow sheath (Merit Medical, South Jordan, UT, USA), the occluded efferent vein was recanalized with the use of a hydrophilic 0.035-inch guidewire (Glidewire; Terumo Europe, Leuven, Belgium) and a 4-F Cobra catheter (Slip-cath; Cook Medical, Bloomington, IN, USA). Contrast injection during retraction of the catheter was done to evaluate the extent of thrombosis. After intravenous injection of 5000 units of unfractioned heparin, the AngioJet AVX catheter, primed as described in the Instructions for Use, was passed over a 0.035-inch, 145cm polytetrafluoroethylene (PTFE) Movable Core Wire Guide (Cook Medical) through the thrombosed segment at a speed of 5 mm/s. High-pressure saline jets created a low pressure zone near the tip of the catheter, generating an effective vacuum for at least 12 mm. The thrombus was drawn into the catheter, where it was fragmented and evacuated. The pressure generated by the device unit and a pump set achieved isovolumetric balance between fluid delivery and removal.7 No thrombolytic agents were administered before, during, or after the rheolytic thrombectomy procedure. The number of passages performed was at the discretion of the attending interventional radiologist. Once the thrombus burden was removed on the venous side, the AngioJet catheter was navigated into the arteriovenous anastomosis after another retrograde puncture into the reopened segment of the efferent vein, distal to the first inserted high-flow sheath. Recanalization of this proximal, residually occluded segment was performed with the AngioJet AVX catheter as described above. If persistent thrombotic material was adherent to the vein wall, percutaneous thromboaspiration was performed through a 6-F Viking guiding catheter (Abbott Vascular, Santa Clara, CA, USA).
82
Journal of Endovascular Therapy 22(1)
Figure 1. Antegrade puncture of an acutely occluded brachiocephalic autologous fistula in the left elbow. (A) Contrast injection shows complete occlusion of the efferent vein (arrows) at 4 cm proximal to the fistula. (B) After 2 runs with the AngioJet rheolytic device over the occluded segment, the efferent vein is reopened, but several residual thrombotic deposits (arrows) persist. (C) After another 2 runs with the AngioJet and additional balloon angioplasty, the efferent vein is completely patent without any residual thrombotic deposit or residual stenosis.
After removal of all thrombotic material, the underlying stenoses were dilated using regular balloon catheters or, in case of resistant underlying stenosis, with a high-pressure balloon catheter (Conquest; Bard, Tempe, AZ, USA). If balloon angioplasty resulted in a flow-limiting intimal-wall dissection or in immediate recoil, a nitinol self-expanding stent was inserted [Maris (Medtronic Vascular, Santa Rosa, CA, USA); Epic (Boston Scientific Corp); or Zilver (Cook Medical)]. In case of vessel wall rupture, an ePTFE covered stent (Fluency; Bard) was placed. The nominal diameter of the stent was 1 mm larger than the efferent vein diameter proximal and distal to the stenosis. The procedure time from sheath insertion to completion fistulogram was recorded. The high flow sheaths remained in place, and the patient was sent to the dialysis department for subsequent hemodialysis. After dialysis, the sheaths were removed, and the puncture sites were manually compressed.
Definitions Primary patency is defined as the time interval between initial rheolytic thrombectomy and any new event necessitating a reintervention on the fistula to maintain successful dialysis, including efferent vein stenosis or thrombosis. Assisted primary patency is defined as the time interval between initial rheolytic thrombectomy and efferent vein thrombosis. Secondary patency is defined as the time interval between initial rheolytic thrombectomy and efferent vein thrombosis with failure to reopen the efferent vein with any endovascular or open surgical technique, necessitating the creation of a new fistula, dialysis graft or catheter insertion, or efferent vein thrombosis without attempts to recanalize the fistula.
Statistical Methodology Patency rates were estimated using the Kaplan-Meier method. Patients were censored in case of kidney transplantation, patient refusal to continue with hemodialysis, or death. Estimates are given with the 95% confidence intervals (CIs). Analysis of the effect of the prognostic factors on fistula patency was performed using univariate and multivariable Cox regression models. The multivariable model was obtained by a backward selection procedure: first, a model was fit including all predictors, and step-by-step the predictor with the highest p value was removed until all predictors had a p value
research-article2015
JETXXX10.1177/1526602814566378Journal of Endovascular TherapyMaleux et al
Clinical Investigation
Percutaneous Rheolytic Thrombectomy of Thrombosed Autogenous Dialysis Fistulas: Technical Results, Clinical Outcome, and Factors Influencing Patency
Journal of Endovascular Therapy 2015, Vol. 22(1) 80–86 © The Author(s) 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1526602814566378 www.jevt.org
Geert Maleux, MD, PhD1, Bruno De Coster, MD1, Annouschka Laenen, PhD2, Johan Vaninbroukx, MD1, Björn Meijers, MD, PhD3, Kathleen Claes, MD, PhD3, Inge Fourneau, MD, PhD4, and Sam Heye, MD, PhD1
Abstract Purpose: To retrospectively analyze the technical and clinical outcome of percutaneous rheolytic thrombectomy with the use of the AngioJet device in thrombosed autogenous arteriovenous dialysis fistulas. Methods: A cohort of 38 consecutive patients (24 men; mean age 70.7±13.8 years) presenting with 59 thrombotic events in 39 autogenous dialysis fistulas were retrospectively analyzed. The AngioJet rheolytic thrombectomy device was used in all cases. Adjunctive therapies, as well as procedure-related complications, were noted. Primary, assisted primary, and secondary patency of the autogenous fistulas was assessed along with factors potentially influencing patency. Results: Initial technical success to recanalize the efferent vein was 100% (n=59), and a successful postprocedure dialysis session was possible in 97% (n=57) of cases. Adjunctive procedures included percutaneous thromboaspiration (n=4, 7%), balloon angioplasty (n=59, 100%), and stent placement (n=16, 27%). Complications related to thrombectomy occurred in 3 (5%) procedures (distal arterial embolus). Primary, assisted primary, and secondary patency rates at 12 months were 56.1% (95% CI 42.8% to 75.2%), 61.6% (95% CI 48.6% to 74.7%), and 86.2% (95% CI 74.9% to 97.5%), respectively. Risk factors for early fistula occlusion were greater patient age (p=0.045), the age of the fistula (p=0.045), previous stent insertion (p=0.019), and an upper arm fistula (p=0.047). Conclusion: Percutaneous rheolytic thrombectomy of autogenous dialysis fistulas is effective in restoring patency and allowing subsequent hemodialysis. The complication rate is acceptably low, and the large majority of the fistulas are still used for hemodialysis at 1-year follow-up. Older fistulas and upper arm fistulas are at higher risk for early rethrombosis. Keywords dialysis fistula, autogenous arteriovenous fistula, thrombosis, occlusion, rheolytic thrombectomy device, percutaneous thrombectomy, hemodialysis
Introduction Autogenous arteriovenous fistulas are considered the vascular access of choice as defined by the National Kidney Foundation’s Dialysis Outcomes Quality Initiative guidelines1; these recommendations are mainly based on superior patency, lower prevalence of vascular steal syndrome, and lower incidence of access-related infectious complications compared to synthetic grafts. Despite these advantages, early or late thrombosis of the efferent vein still occurs, and immediate revascularization is of utmost importance to provide continued hemodialysis without temporary use of dialysis catheters in this patient population.2 For more than 2 decades, percutaneous recanalization of hemodialysis access grafts has been described as a valuable
alternative to surgical thrombectomy,3 either with the use of various thrombolytic drugs alone4 or in combination with dedicated percutaneous catheter-based devices, including rheolytic5–8 and mechanical thrombectomy devices.9–14 1
Department of Radiology, University Hospitals Leuven / Department of Imaging & Pathology, KU Leuven, Belgium. 2 Interuniversity Institute for Biostatistics and Statistical Bioinformatics, KU Leuven and Universiteit Hasselt, Belgium. 3 Department of Nephrology, University Hospitals, Leuven, Belgium. 4 Department of Vascular Surgery, University Hospitals, Leuven, Belgium. Corresponding Author: Geert Maleux, Department of Radiology, University Hospitals Leuven, Herestraat 49, Leuven, B-3000, Belgium. Email: [email protected]
81
Maleux et al However, data on technical and clinical outcomes of the use of percutaneous thrombectomy devices for thrombosed autogenous dialysis fistulas mostly include the use in a mix of thrombosed autogenous fistulas and synthetic grafts. Additionally, the clinical outcome focuses only on shortterm (6 months) primary patency without reporting assisted primary or secondary patency rates. In this report, the technical and clinical outcomes of percutaneous rheolytic thrombectomy in thrombosed autogenous dialysis fistulas was assessed, including adjunctive thrombectomy procedures, the potential to perform subsequent hemodialysis through the recanalized fistula, complications during and immediately after the procedure, as well as primary, assisted primary, and secondary patency rates of the autogenous fistula. Finally, several factors potentially influencing prolonged patency of the fistula were analyzed.
Methods Study Design This was a retrospective study approved by our institutional ethics committee (S 55351). A search of the interventional radiology database was performed between July 2009 and January 2013 to identify patients undergoing treatment to recanalize a thrombosed autogenous arteriovenous fistula with the use of the AngioJet Rheolytic Thrombectomy System (Boston Scientific, Natick, MA, USA). Patients presenting with graft occlusion or initially treated with other thrombectomy techniques were excluded from the study. Medical data, including dialysis and radiological reports, were reviewed by 2 radiologists (GM and BDC). Data collection included demographics, procedure details, and clinical outcomes related to the potential to perform hemodialysis through the target fistula.
Patient Demographics From July 2009 to January 2013, 38 consecutive patients (24 men; mean age 70.7±13.8 years) presented with 59 events of acute occlusion in 39 autogenous hemodialysis fistulas. All occluded autogenous fistulas were arm fistulas (Table 1). The mean age of the fistula was 3.4 years (0.1– 11.7), and 5 fistulas had a stent previously inserted. No patient had a thrombotic event prior to the first AngioJet thrombectomy procedure. No patients with thrombosed aneurysmal fistulas were identified.
Percutaneous Rheolytic Thrombectomy After clinical evaluation by the attending nephrologist, all recanalization procedures were performed within 24 hours of initial diagnosis of the fistula thrombosis. All procedures were performed by 3 interventional radiologists (GM, JV, and SH). After clinical evaluation of the afferent artery and
Table 1. Types of Occluded Fistulas Treated With the AngioJet Rheolytic Device.a Type of fistula (n=39) Elbow Brachiobasilic Brachiocephalic Brachial artery–median vein Radiobasilic Radiocephalic Radial artery–median vein Wrist Radiobasilic Radiocephalic Side of the fistula Right Left Previous interventions for stenosis
7 (18) 10 (26) 1 (3) 4 (10) 3 (8) 1 (3) 1 (3) 12 (31) 16 (41) 23 (59) 9 (23)
a
Data are presented as counts (percentage).
efferent vein and ultrasonographic confirmation of the thrombosis of the efferent vein, a retro- and/or antegrade puncture was performed under local anesthesia via the Seldinger technique. Through a 6-F high flow sheath (Merit Medical, South Jordan, UT, USA), the occluded efferent vein was recanalized with the use of a hydrophilic 0.035-inch guidewire (Glidewire; Terumo Europe, Leuven, Belgium) and a 4-F Cobra catheter (Slip-cath; Cook Medical, Bloomington, IN, USA). Contrast injection during retraction of the catheter was done to evaluate the extent of thrombosis. After intravenous injection of 5000 units of unfractioned heparin, the AngioJet AVX catheter, primed as described in the Instructions for Use, was passed over a 0.035-inch, 145cm polytetrafluoroethylene (PTFE) Movable Core Wire Guide (Cook Medical) through the thrombosed segment at a speed of 5 mm/s. High-pressure saline jets created a low pressure zone near the tip of the catheter, generating an effective vacuum for at least 12 mm. The thrombus was drawn into the catheter, where it was fragmented and evacuated. The pressure generated by the device unit and a pump set achieved isovolumetric balance between fluid delivery and removal.7 No thrombolytic agents were administered before, during, or after the rheolytic thrombectomy procedure. The number of passages performed was at the discretion of the attending interventional radiologist. Once the thrombus burden was removed on the venous side, the AngioJet catheter was navigated into the arteriovenous anastomosis after another retrograde puncture into the reopened segment of the efferent vein, distal to the first inserted high-flow sheath. Recanalization of this proximal, residually occluded segment was performed with the AngioJet AVX catheter as described above. If persistent thrombotic material was adherent to the vein wall, percutaneous thromboaspiration was performed through a 6-F Viking guiding catheter (Abbott Vascular, Santa Clara, CA, USA).
82
Journal of Endovascular Therapy 22(1)
Figure 1. Antegrade puncture of an acutely occluded brachiocephalic autologous fistula in the left elbow. (A) Contrast injection shows complete occlusion of the efferent vein (arrows) at 4 cm proximal to the fistula. (B) After 2 runs with the AngioJet rheolytic device over the occluded segment, the efferent vein is reopened, but several residual thrombotic deposits (arrows) persist. (C) After another 2 runs with the AngioJet and additional balloon angioplasty, the efferent vein is completely patent without any residual thrombotic deposit or residual stenosis.
After removal of all thrombotic material, the underlying stenoses were dilated using regular balloon catheters or, in case of resistant underlying stenosis, with a high-pressure balloon catheter (Conquest; Bard, Tempe, AZ, USA). If balloon angioplasty resulted in a flow-limiting intimal-wall dissection or in immediate recoil, a nitinol self-expanding stent was inserted [Maris (Medtronic Vascular, Santa Rosa, CA, USA); Epic (Boston Scientific Corp); or Zilver (Cook Medical)]. In case of vessel wall rupture, an ePTFE covered stent (Fluency; Bard) was placed. The nominal diameter of the stent was 1 mm larger than the efferent vein diameter proximal and distal to the stenosis. The procedure time from sheath insertion to completion fistulogram was recorded. The high flow sheaths remained in place, and the patient was sent to the dialysis department for subsequent hemodialysis. After dialysis, the sheaths were removed, and the puncture sites were manually compressed.
Definitions Primary patency is defined as the time interval between initial rheolytic thrombectomy and any new event necessitating a reintervention on the fistula to maintain successful dialysis, including efferent vein stenosis or thrombosis. Assisted primary patency is defined as the time interval between initial rheolytic thrombectomy and efferent vein thrombosis. Secondary patency is defined as the time interval between initial rheolytic thrombectomy and efferent vein thrombosis with failure to reopen the efferent vein with any endovascular or open surgical technique, necessitating the creation of a new fistula, dialysis graft or catheter insertion, or efferent vein thrombosis without attempts to recanalize the fistula.
Statistical Methodology Patency rates were estimated using the Kaplan-Meier method. Patients were censored in case of kidney transplantation, patient refusal to continue with hemodialysis, or death. Estimates are given with the 95% confidence intervals (CIs). Analysis of the effect of the prognostic factors on fistula patency was performed using univariate and multivariable Cox regression models. The multivariable model was obtained by a backward selection procedure: first, a model was fit including all predictors, and step-by-step the predictor with the highest p value was removed until all predictors had a p value
