Assessing the Approach to a Thrombosed AV Graft Pedro Ponce, Telmo Carvalho, Humberto Messias, Fernando Neves, and on behalf of the Lisbon VAC Interventional Nephrology Team Vascular Access Centre, NephroCare Portugal, Lisboa, Portugal
ABSTRACT The patency of the vascular access (VA) is a fight for the attending nephrologist. A retrospective observational study was conducted to compare the success rate of surgical versus endovascular technique percutaneous transluminal angioplasty (PTA) for graft thrombosis treatment. Of 3008 patients, 22.1% patients were dialyzed through grafts. Forty-five percent of all prevalent patients referred due to VA malfunction had a graft. For 18 months, 336 thrombosed grafts were submitted to surgery in 228 cases and to PTA in 126. PTA for thrombolysis included the Pharmaco-Mechanical Technique and the Arrow-Trerot-
ola Device. Procedures were performed as outpatient, with an average delay of 1 day. Immediate success was 100% for surgery and 87.3% for PTA. The unassisted patency for thrombosed grafts for surgery/PTAwas 265.12 15.30/230.59 19.83 days respectively, favoring surgery. The primary patency for thrombosed grafts treated by surgery/PTA at 30, 90, and 180 days was, respectively, 74.1%/81%, 63.2%/67.5%, and 53.9%/55.6% all in favor of PTA. AV grafts have a much higher rate of thrombosis than fistulas. Graft thrombosis can be dealt either by surgery or PTA, with identical success.
Vascular access (VA) malfunction and failure is a leading cause of morbidity in chronic hemodialysis patients, responsible for over 20% of all dialysis patients’ annual hospital admissions. In 2006, the United States spent $71,616 on access per-person, per-year with AV grafts and $59,374 with fistulas (1). Polytetrafluoroethylene (PTFE) grafts are particularly prone to VA failure; about half of the PTFE grafts experience their first episode of thrombosis within 6 months of construction (2). Our Portuguese Nephrocare network treated a universe of 4264 patients in 2012 and 17.4% were dialyzed through a PTFE graft. Our Lisbon VA centre (VAC) covers around 3008 patients, 22.1% being dialyzed through PTFE grafts. Only 7.2% of the incident VAs implanted are grafts, but 45% of all prevalent patients referred for VA malfunction have a graft, illustrating a much higher incidence of VA complications and need for intervention when compared with native AV fistulas. Available data do not indicate a clear-cut preference between surgical thrombectomy and revision versus percutaneous endovascular thrombolysis for the treatment and recovery of graft thrombosis. Comparative studies show conflicting results, with
similar technical success rates and long-term patencies between these two methodologies (3–5). In this prospective observational study, it was our intent to compare the success rate of surgical versus endovascular technique for PTFE graft thrombosis treatment. Patients and Methods A retrospective analysis of our Vascular Access Registry (AV online©) for a 18-month period, from July 1, 2011 to Dec 31, 2012, was conducted to compare the success rate of surgical thrombectomy/ revision versus endovascular thrombolysis in the treatment of 336 thrombosed PTFE grafts. Thrombosed grafts corresponds to 15.9% of all referrals to our VAC and the choice for referral to surgical or interventional nephrology was at the discretion of the referring nephrologist from the 19 dialysis units serviced by this VAC (a global population of 3008 patients, 22.1% dialyzed through PTFE grafts). Both techniques were equally available to all attending nephrologists in the subsequent 24 hours after referral, from 9:00 am to 7:00 pm, Monday through Friday. All procedures were performed as outpatient, with an average delay of 1 day, by five different surgeons and four distinct interventional nephrology teams. Only surgical thrombectomy was included and new graft interpositions (17 procedures overall— jump grafts) were excluded as they correspond to new accesses. Endovascular procedures for
Address correspondence to: Pedro Ponce, Nephrocare, da Costa Pedreira, n 8, 1750-233 Lisboa, PortuRua Jose gal, e-mail: [email protected]. Seminars in Dialysis—Vol 27, No 5 (September–October) 2014 pp. 518–521 DOI: 10.1111/sdi.12189 © 2014 Wiley Periodicals, Inc. 518
519
VASCULAR ACCESS THROMBOSIS
thrombolysis included the Pharmaco-Mechanical technique (PMT) (only with heparinized saline + Angioplasty + Fogarty thrombectomy) and the Arrow-Trerotola Percutaneous Thrombolytic Device (PTD + angioplasty). No bare stenting or graft stenting was used in all these procedures. Data were analyzed with the SPSS 17.0, primary patency was assessed through the Kaplan–Meir actuarial survival and compared through the Logrank test. Results A total of 354 interventions were performed over 336 thrombosed grafts, 228 surgical interventions, and 126 endovascular interventions. A Pharmacomechanical procedure was performed in 100 (79.4%) of all endovascular interventions and the Arrow-Trerotola Percutaneous Thrombolytic Device (PTDâ) in 26 (20.6%) cases. Of the surgical revised grafts, 191 (83.8%) were located in the upper arm, 29 (11.8%) in the forearm, and 8 (3.5%) in the tight. Of all these operated grafts, 62 (27.2%) were experiencing their first intervention, while others were submitted on average to 2.19 2.39 interventions. Average number of previous thrombosis were 1.83 1.57 and 24 (10.5%) failed due to clotting for the first time. Immediate success (meaning graft flow when leaving the operating room) was obtained in 100% of all PTFEs. Endovascular thrombolysis was performed in 99 (78.6%) grafts in the upper arm, 17 (13.5%) in the forearm, and 10 (7.9%) in the tight. Of all cases referred to percutaneous transluminal angioplasty (PTA), 44 (34.9%) were intervened for the first time, while the remaining were intervened on average 2.19 2.57 times. Average number of previous episodes of access thrombosis was 1.21 1.11 and 36 (28.6%) failed due to clotting for the first time. Angiography had immediate success in 110 grafts (87.3%). A new therapeutic intervention was required in 116 grafts (50.9%) previously submitted to surgery and in 64 grafts (50.8%) previously treated in the angiography suite. Of all recurrent thrombosis in surgical cases, 67 (57.8%) were treated again by surgery, and the endovascular field recurrent thrombosis was treated by surgery in 36 (56.3%) grafts, an identical crossover for rescue treatment. A new intervention was required in the first 30 days in 59 (25.9%) of all grafts treated by surgery and in 24 (19%) of those treated by endovascular thrombolysis. At 90 days, 84 (36.8%) of all grafts treated by surgery and 41 (32.5%) of all grafts treated originally by angiography needed a new intervention. At 180 days, 105 (46.1%) of all the grafts submitted to surgery and 56 (44.4%) of those treated by angiography needed a new intervention (Table 1).
TABLE 1. Primary graft patency rate at 30, 90, and 180 days Primary patency rate
30 days
90 days
180 days
Surgery Angiography
74.1% 81.0%
63.2% 67.5%
53.9% 55.6%
Average primary (unassisted) patency rate for thrombosed grafts treated by surgery, or PTA was 265.12 15.30 days (median 227 days) and 230.59 19.83 days (median 178 days) respectively, a significant difference favoring surgery (Log-Rank (Mantel–Cox) test, p < 0.005) (Fig. 1). The average Qa measured before surgery was 763.61 455.90 ml/minute (n = 172), after 1 month 968.07 48,710 ml/minute (n = 122) and at 30 days 997.03 511.02 ml/minute (n = 98). Before endovascular intervention, average Qa before the procedure was 726.02 421.58 ml/minute (n = 114), at 30 days 1146.69 527.95 ml/minute (n = 80), and at 90 days 1017 516.07 ml/minute (n = 54). Average spKt/V before thrombosis and at 30 days of patients treated by surgery and angiography was, respectively, 1.63 0.50 (n = 205) and 1.8 0.44 (n = 130) versus 1.53 0.47 (n = 124) and 1.84 0.42 (n = 88). None of these differences in performance reached statistical significance.
Discussion When essentially no monitoring or surveillance was being carried out, the incidence of graft thrombosis was 1 to 1.5 per patient per-year; nevertheless quite heterogeneous, as only 35–65% thrombosed in a 1-year period. Among those that clotted, 40% had it once, 40% had two to four episodes, and 20% had five or more (3). Of all cases of graft thrombosis, 85–90% are associated with an anatomical lesion, neointimal hyperplasia leading to venous stenosis. These lesions cause progressively increasing resistance and a concomitant decrease in blood flow (6). The graft survival following thrombectomy is quite poor, with primary patency of 30–63% at 3 months and 11–34% at 6 months (7–10). According to the NKF/KDOQI Practice Guidelines (11), Guideline 6 and 6.7, the immediate success rate should be at least 85% and primary unassisted patency goal at 3 months of 40%. The long-term patency rates for thrombolysis are not as good as for angioplasty, therefore our effort should be centered in thrombosis prevention by the prospective diagnosis and treatment of culprit lesions that induce graft clotting. Surgical unassisted patency is difficult to determine due to reporting of cumulative patencies; it is approximately 50% at 6 months and 20% at 1 year. The NKF/KDOQI Work Group recommends goals of 50% unassisted patency at 6 months and 40% unassisted patency at 1 year. Surgical correction of stenoses is held to a
520
Ponce et al.
Fig 1. Actuarial survival (Kaplan–Meier) of thrombosed grafts treated by both methods.
higher standard than PTA because the access is usually extended further up the extremity when surgical revision is performed In guideline 21 rationale, it is stated that in the work group opinion, current data suggest that surgical thrombectomy and endovascular thrombolysis are equally effective for resolving thrombosis, therefore technique choice should be based on the expertise of the center. They recommend based only on the opinion, an immediate patency rate of 85% for both techniques, 40% unassisted patency at 3 months (this one based on evidence) and for surgical revision 50% unassisted patency at 6 months. In a meta-analysis of randomized controlled trials, comparing surgical thrombectomy, consisting surgical thrombectomy followed by local revision of the access, and pharmaco-mechanical thrombolysis for thrombosed dialysis grafts, the authors (all of them surgeons) found a significant superiority of surgery over endovascular procedures at 30 days (relative risk of failure 1.32) and 90 days (RR 1.22) (12). On the other hand, some single-center studies were favorable to endovascular techniques (3,13). Data suggest that the benefit of surgical thrombectomy is seen mainly when the surgical procedure combines thrombectomy with some form of revision/interposition. It could be argued that this is similar to creating a new access. Although surgical thrombectomy and revision is at least as effective as endovascular thrombolysis, it has a number of disadvantages: It is invasive, causes higher blood loss, may require general anesthesia or hospitalization, has higher risk of infection, and last but not the least, if unsuccessful or suffering early failure, it does not allow rescue endovascular intervention in the following 3–4 weeks. In our study, immediate success rate was 100% for surgery and 87.3% for endovascular techniques. However, primary patency rates at 30, 90, and 180 days were for surgery 74.1%, 63.2%, and
53.9% respectively and for endovascular thrombolysis 81%, 67.5%, and 55.6% respectively, somehow favoring the percutaneous option. On the other hand, when counting the number of days of unassisted patency, we registered better survival for grafts intervened surgically. During this follow-up period, the crossover between techniques was identical at 43% of patients initially treated by one modality intervened subsequently by the other. Overall, the performances of both modalities were identical. Both interventions improved significantly Qa around 37% and spKt/V around 17%, although it is still not defined which one is the best way to measure functional success after VA intervention. Our results compare favorably with those published in the recent literature. We are well aware that the success rate of postthrombosis intervention is associated with poor outcomes of both the surgical and percutaneous techniques at much higher costs. Preemptive correction of significant lesions detected by prospective monitoring, however, is still an elusive goal awaiting better techniques with higher accuracy (14). The ideal protocol for effective VA surveillance is far from being consensual. The pursuit to find a noninvasive method to detect sub clinical stenosis is still not over (15). Disclaimer All authors work with NephroCare Portugal, having no other competing interests to declare. Acknowledgments The authors thank the outstanding performance of all the surgeons and interventional nephrologists who daily contributed to these results.
VASCULAR ACCESS THROMBOSIS
References 1. USRDS: Annual Data Report: Atlas of End-stage Renal Disease in the United States. Bethesda, MD: National Institute of Health, 2008 2. Huber T, Carter J, Carter R, et al.: Patency of autologous and PTFE upper extremity arteriovenous hemodialysis accesses. J Vasc Surg 38:1005–1011, 2003 3. Bethard G: Thrombolysis versus surgery for the treatment of thrombosed dialysis access grafts. J Am Soc Nephrol 6:1619–1624, 1995 4. Sands J, Miranda C: Prolongation of hemodialysis access survival with elective revision. Clin Nephrol 44:334–337, 1995 5. Schwartz C, McBrayer C, Sloan J, et al.: Thrombosed dialysis grafts: Comparison of treatment with transluminal angioplasty and surgical revision. Radiology 194:337–341, 1995 6. Beathard G: Arteriovenous graft stenosis and thrombosis. In: Arif A, Agarwal A, Yevzlin A eds. Interventional Nephrology. New York: McGraw-Hill Inc, 2012:179 7. Cohen M, Kumpe D, Durham J, et al.: Improved treatment of thrombosed hemodialysis access sites with thrombolysis and angioplasty. Kidney Int 46:1375–1380, 1994 8. Lilly R, Carlton D, Barker J, et al.: Predictors of arteriovenous graft patency after radiologic intervention in hemodialysys patient. Am J Kidney Dis 37:945–953, 2001
521
9. Turmel-Rodrigues L, Pengloan J, Baudin S, et al.: Treatment of stenosis and thrombosis in hemodialysis fistulas and grafts by interventional radiology. Nephrol Dial Transplant 15:2029–2036, 2000 10. Valji K, Bookstein J, Roberts A, et al.: Pulse-spray pharmacomechanical thrombolysis of thrombosed hemodialysis access grafts. Am J Roentgenol 164:1495–1500, 1995 11. National Kidney Foundation: KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for 2006 Updates. Am J Kidney Dis 48:S1–S322, 2006 12. Green L, Lee D, Kucey D, et al.: A metaanalysis comparing surgical thrombectomy and pharmacomechanical thrombolysis for thrombosed dialysis grafts. J Vasc Surg 36:939–945, 2002 13. Dougherty M, Calligaro K, Shindler N, et al.: Endovascular vs surgical treatment for thrombosed hemodialysis grafts. A prospective randomized study. J Vasc Surg 30:1016–1023, 1999 14. Ponce P, Mateus A, Santos L: Anatomical correlation of a well-functioning access graft for haemodialysis. Nephrol Dial Transplant 24:535–538, 2009 15. Paulson WD: Vascular access surveillance: an ongoing controversy. Kidney Int 81:132–142, 2012
ABSTRACT The patency of the vascular access (VA) is a fight for the attending nephrologist. A retrospective observational study was conducted to compare the success rate of surgical versus endovascular technique percutaneous transluminal angioplasty (PTA) for graft thrombosis treatment. Of 3008 patients, 22.1% patients were dialyzed through grafts. Forty-five percent of all prevalent patients referred due to VA malfunction had a graft. For 18 months, 336 thrombosed grafts were submitted to surgery in 228 cases and to PTA in 126. PTA for thrombolysis included the Pharmaco-Mechanical Technique and the Arrow-Trerot-
ola Device. Procedures were performed as outpatient, with an average delay of 1 day. Immediate success was 100% for surgery and 87.3% for PTA. The unassisted patency for thrombosed grafts for surgery/PTAwas 265.12 15.30/230.59 19.83 days respectively, favoring surgery. The primary patency for thrombosed grafts treated by surgery/PTA at 30, 90, and 180 days was, respectively, 74.1%/81%, 63.2%/67.5%, and 53.9%/55.6% all in favor of PTA. AV grafts have a much higher rate of thrombosis than fistulas. Graft thrombosis can be dealt either by surgery or PTA, with identical success.
Vascular access (VA) malfunction and failure is a leading cause of morbidity in chronic hemodialysis patients, responsible for over 20% of all dialysis patients’ annual hospital admissions. In 2006, the United States spent $71,616 on access per-person, per-year with AV grafts and $59,374 with fistulas (1). Polytetrafluoroethylene (PTFE) grafts are particularly prone to VA failure; about half of the PTFE grafts experience their first episode of thrombosis within 6 months of construction (2). Our Portuguese Nephrocare network treated a universe of 4264 patients in 2012 and 17.4% were dialyzed through a PTFE graft. Our Lisbon VA centre (VAC) covers around 3008 patients, 22.1% being dialyzed through PTFE grafts. Only 7.2% of the incident VAs implanted are grafts, but 45% of all prevalent patients referred for VA malfunction have a graft, illustrating a much higher incidence of VA complications and need for intervention when compared with native AV fistulas. Available data do not indicate a clear-cut preference between surgical thrombectomy and revision versus percutaneous endovascular thrombolysis for the treatment and recovery of graft thrombosis. Comparative studies show conflicting results, with
similar technical success rates and long-term patencies between these two methodologies (3–5). In this prospective observational study, it was our intent to compare the success rate of surgical versus endovascular technique for PTFE graft thrombosis treatment. Patients and Methods A retrospective analysis of our Vascular Access Registry (AV online©) for a 18-month period, from July 1, 2011 to Dec 31, 2012, was conducted to compare the success rate of surgical thrombectomy/ revision versus endovascular thrombolysis in the treatment of 336 thrombosed PTFE grafts. Thrombosed grafts corresponds to 15.9% of all referrals to our VAC and the choice for referral to surgical or interventional nephrology was at the discretion of the referring nephrologist from the 19 dialysis units serviced by this VAC (a global population of 3008 patients, 22.1% dialyzed through PTFE grafts). Both techniques were equally available to all attending nephrologists in the subsequent 24 hours after referral, from 9:00 am to 7:00 pm, Monday through Friday. All procedures were performed as outpatient, with an average delay of 1 day, by five different surgeons and four distinct interventional nephrology teams. Only surgical thrombectomy was included and new graft interpositions (17 procedures overall— jump grafts) were excluded as they correspond to new accesses. Endovascular procedures for
Address correspondence to: Pedro Ponce, Nephrocare, da Costa Pedreira, n 8, 1750-233 Lisboa, PortuRua Jose gal, e-mail: [email protected]. Seminars in Dialysis—Vol 27, No 5 (September–October) 2014 pp. 518–521 DOI: 10.1111/sdi.12189 © 2014 Wiley Periodicals, Inc. 518
519
VASCULAR ACCESS THROMBOSIS
thrombolysis included the Pharmaco-Mechanical technique (PMT) (only with heparinized saline + Angioplasty + Fogarty thrombectomy) and the Arrow-Trerotola Percutaneous Thrombolytic Device (PTD + angioplasty). No bare stenting or graft stenting was used in all these procedures. Data were analyzed with the SPSS 17.0, primary patency was assessed through the Kaplan–Meir actuarial survival and compared through the Logrank test. Results A total of 354 interventions were performed over 336 thrombosed grafts, 228 surgical interventions, and 126 endovascular interventions. A Pharmacomechanical procedure was performed in 100 (79.4%) of all endovascular interventions and the Arrow-Trerotola Percutaneous Thrombolytic Device (PTDâ) in 26 (20.6%) cases. Of the surgical revised grafts, 191 (83.8%) were located in the upper arm, 29 (11.8%) in the forearm, and 8 (3.5%) in the tight. Of all these operated grafts, 62 (27.2%) were experiencing their first intervention, while others were submitted on average to 2.19 2.39 interventions. Average number of previous thrombosis were 1.83 1.57 and 24 (10.5%) failed due to clotting for the first time. Immediate success (meaning graft flow when leaving the operating room) was obtained in 100% of all PTFEs. Endovascular thrombolysis was performed in 99 (78.6%) grafts in the upper arm, 17 (13.5%) in the forearm, and 10 (7.9%) in the tight. Of all cases referred to percutaneous transluminal angioplasty (PTA), 44 (34.9%) were intervened for the first time, while the remaining were intervened on average 2.19 2.57 times. Average number of previous episodes of access thrombosis was 1.21 1.11 and 36 (28.6%) failed due to clotting for the first time. Angiography had immediate success in 110 grafts (87.3%). A new therapeutic intervention was required in 116 grafts (50.9%) previously submitted to surgery and in 64 grafts (50.8%) previously treated in the angiography suite. Of all recurrent thrombosis in surgical cases, 67 (57.8%) were treated again by surgery, and the endovascular field recurrent thrombosis was treated by surgery in 36 (56.3%) grafts, an identical crossover for rescue treatment. A new intervention was required in the first 30 days in 59 (25.9%) of all grafts treated by surgery and in 24 (19%) of those treated by endovascular thrombolysis. At 90 days, 84 (36.8%) of all grafts treated by surgery and 41 (32.5%) of all grafts treated originally by angiography needed a new intervention. At 180 days, 105 (46.1%) of all the grafts submitted to surgery and 56 (44.4%) of those treated by angiography needed a new intervention (Table 1).
TABLE 1. Primary graft patency rate at 30, 90, and 180 days Primary patency rate
30 days
90 days
180 days
Surgery Angiography
74.1% 81.0%
63.2% 67.5%
53.9% 55.6%
Average primary (unassisted) patency rate for thrombosed grafts treated by surgery, or PTA was 265.12 15.30 days (median 227 days) and 230.59 19.83 days (median 178 days) respectively, a significant difference favoring surgery (Log-Rank (Mantel–Cox) test, p < 0.005) (Fig. 1). The average Qa measured before surgery was 763.61 455.90 ml/minute (n = 172), after 1 month 968.07 48,710 ml/minute (n = 122) and at 30 days 997.03 511.02 ml/minute (n = 98). Before endovascular intervention, average Qa before the procedure was 726.02 421.58 ml/minute (n = 114), at 30 days 1146.69 527.95 ml/minute (n = 80), and at 90 days 1017 516.07 ml/minute (n = 54). Average spKt/V before thrombosis and at 30 days of patients treated by surgery and angiography was, respectively, 1.63 0.50 (n = 205) and 1.8 0.44 (n = 130) versus 1.53 0.47 (n = 124) and 1.84 0.42 (n = 88). None of these differences in performance reached statistical significance.
Discussion When essentially no monitoring or surveillance was being carried out, the incidence of graft thrombosis was 1 to 1.5 per patient per-year; nevertheless quite heterogeneous, as only 35–65% thrombosed in a 1-year period. Among those that clotted, 40% had it once, 40% had two to four episodes, and 20% had five or more (3). Of all cases of graft thrombosis, 85–90% are associated with an anatomical lesion, neointimal hyperplasia leading to venous stenosis. These lesions cause progressively increasing resistance and a concomitant decrease in blood flow (6). The graft survival following thrombectomy is quite poor, with primary patency of 30–63% at 3 months and 11–34% at 6 months (7–10). According to the NKF/KDOQI Practice Guidelines (11), Guideline 6 and 6.7, the immediate success rate should be at least 85% and primary unassisted patency goal at 3 months of 40%. The long-term patency rates for thrombolysis are not as good as for angioplasty, therefore our effort should be centered in thrombosis prevention by the prospective diagnosis and treatment of culprit lesions that induce graft clotting. Surgical unassisted patency is difficult to determine due to reporting of cumulative patencies; it is approximately 50% at 6 months and 20% at 1 year. The NKF/KDOQI Work Group recommends goals of 50% unassisted patency at 6 months and 40% unassisted patency at 1 year. Surgical correction of stenoses is held to a
520
Ponce et al.
Fig 1. Actuarial survival (Kaplan–Meier) of thrombosed grafts treated by both methods.
higher standard than PTA because the access is usually extended further up the extremity when surgical revision is performed In guideline 21 rationale, it is stated that in the work group opinion, current data suggest that surgical thrombectomy and endovascular thrombolysis are equally effective for resolving thrombosis, therefore technique choice should be based on the expertise of the center. They recommend based only on the opinion, an immediate patency rate of 85% for both techniques, 40% unassisted patency at 3 months (this one based on evidence) and for surgical revision 50% unassisted patency at 6 months. In a meta-analysis of randomized controlled trials, comparing surgical thrombectomy, consisting surgical thrombectomy followed by local revision of the access, and pharmaco-mechanical thrombolysis for thrombosed dialysis grafts, the authors (all of them surgeons) found a significant superiority of surgery over endovascular procedures at 30 days (relative risk of failure 1.32) and 90 days (RR 1.22) (12). On the other hand, some single-center studies were favorable to endovascular techniques (3,13). Data suggest that the benefit of surgical thrombectomy is seen mainly when the surgical procedure combines thrombectomy with some form of revision/interposition. It could be argued that this is similar to creating a new access. Although surgical thrombectomy and revision is at least as effective as endovascular thrombolysis, it has a number of disadvantages: It is invasive, causes higher blood loss, may require general anesthesia or hospitalization, has higher risk of infection, and last but not the least, if unsuccessful or suffering early failure, it does not allow rescue endovascular intervention in the following 3–4 weeks. In our study, immediate success rate was 100% for surgery and 87.3% for endovascular techniques. However, primary patency rates at 30, 90, and 180 days were for surgery 74.1%, 63.2%, and
53.9% respectively and for endovascular thrombolysis 81%, 67.5%, and 55.6% respectively, somehow favoring the percutaneous option. On the other hand, when counting the number of days of unassisted patency, we registered better survival for grafts intervened surgically. During this follow-up period, the crossover between techniques was identical at 43% of patients initially treated by one modality intervened subsequently by the other. Overall, the performances of both modalities were identical. Both interventions improved significantly Qa around 37% and spKt/V around 17%, although it is still not defined which one is the best way to measure functional success after VA intervention. Our results compare favorably with those published in the recent literature. We are well aware that the success rate of postthrombosis intervention is associated with poor outcomes of both the surgical and percutaneous techniques at much higher costs. Preemptive correction of significant lesions detected by prospective monitoring, however, is still an elusive goal awaiting better techniques with higher accuracy (14). The ideal protocol for effective VA surveillance is far from being consensual. The pursuit to find a noninvasive method to detect sub clinical stenosis is still not over (15). Disclaimer All authors work with NephroCare Portugal, having no other competing interests to declare. Acknowledgments The authors thank the outstanding performance of all the surgeons and interventional nephrologists who daily contributed to these results.
VASCULAR ACCESS THROMBOSIS
References 1. USRDS: Annual Data Report: Atlas of End-stage Renal Disease in the United States. Bethesda, MD: National Institute of Health, 2008 2. Huber T, Carter J, Carter R, et al.: Patency of autologous and PTFE upper extremity arteriovenous hemodialysis accesses. J Vasc Surg 38:1005–1011, 2003 3. Bethard G: Thrombolysis versus surgery for the treatment of thrombosed dialysis access grafts. J Am Soc Nephrol 6:1619–1624, 1995 4. Sands J, Miranda C: Prolongation of hemodialysis access survival with elective revision. Clin Nephrol 44:334–337, 1995 5. Schwartz C, McBrayer C, Sloan J, et al.: Thrombosed dialysis grafts: Comparison of treatment with transluminal angioplasty and surgical revision. Radiology 194:337–341, 1995 6. Beathard G: Arteriovenous graft stenosis and thrombosis. In: Arif A, Agarwal A, Yevzlin A eds. Interventional Nephrology. New York: McGraw-Hill Inc, 2012:179 7. Cohen M, Kumpe D, Durham J, et al.: Improved treatment of thrombosed hemodialysis access sites with thrombolysis and angioplasty. Kidney Int 46:1375–1380, 1994 8. Lilly R, Carlton D, Barker J, et al.: Predictors of arteriovenous graft patency after radiologic intervention in hemodialysys patient. Am J Kidney Dis 37:945–953, 2001
521
9. Turmel-Rodrigues L, Pengloan J, Baudin S, et al.: Treatment of stenosis and thrombosis in hemodialysis fistulas and grafts by interventional radiology. Nephrol Dial Transplant 15:2029–2036, 2000 10. Valji K, Bookstein J, Roberts A, et al.: Pulse-spray pharmacomechanical thrombolysis of thrombosed hemodialysis access grafts. Am J Roentgenol 164:1495–1500, 1995 11. National Kidney Foundation: KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for 2006 Updates. Am J Kidney Dis 48:S1–S322, 2006 12. Green L, Lee D, Kucey D, et al.: A metaanalysis comparing surgical thrombectomy and pharmacomechanical thrombolysis for thrombosed dialysis grafts. J Vasc Surg 36:939–945, 2002 13. Dougherty M, Calligaro K, Shindler N, et al.: Endovascular vs surgical treatment for thrombosed hemodialysis grafts. A prospective randomized study. J Vasc Surg 30:1016–1023, 1999 14. Ponce P, Mateus A, Santos L: Anatomical correlation of a well-functioning access graft for haemodialysis. Nephrol Dial Transplant 24:535–538, 2009 15. Paulson WD: Vascular access surveillance: an ongoing controversy. Kidney Int 81:132–142, 2012
