The Effect of Systemic Anticoagulation in Patients Undergoing Angioaccess Surgery
Article Outline
The benefit of intravenous heparin as an anticoagulant to avoid thrombotic complications during angioaccess surgery for hemodialysis is unknown. We prospectively randomized 115 consecutive patients referred to our institution for permanent hemodialysis access to receive systemic anticoagulation or no anticoagulation during angioaccess surgery. Patient demographics, comorbid conditions, procedure time, complications, and patency were recorded in accordance with standards recommended by the Society for Vascular Surgery. Of the 115 patients randomized, 58 received no anticoagulation and 57 received systemic anticoagulation with intravenous heparin. Arteriovenous fistulas were created in 84 patients and 31 arteriovenous grafts were inserted. Operative times were longer for grafts compared to fistulas, but there were no significant differences in operative times between patients receiving anticoagulation and those not (p = 0.31). Perioperative bleeding complications were more common in patients receiving heparin (p = 0.008). The primary 30-day patency was 84% for patients receiving heparin and 86% for those not (p = 0.79). The 3-month functional patency was 68% for both groups (p = 0.99). Age, gender, operative time, and incidence of bleeding complications had no impact on patency. In our experience, systemic anticoagulation for angioaccess surgery is associated with an increased incidence of bleeding complications and offers no advantage in terms of early patency.
Introduction
Anticoagulation with unfractionated heparin administered intravenously is a common practice in patients undergoing vascular surgery. To our knowledge, there is currently no clinical evidence to support the use of systemic anticoagulation to prevent early thrombotic complications in patients undergoing angioaccess surgery. Furthermore, guidelines published by the National Kidney Foundation Dialysis Outcome Quality Initiative (NKF-DOQI) do not address the need for systemic anticoagulation during vascular access surgery for hemodialysis.1 As a result, current practice patterns vary with regard to the use of systemic heparin in patients undergoing angioaccess. While intraoperative anticoagulation could be justified if a beneficial effect on the outcome of angioaccess procedures were demonstrated, the use of systemic heparin is not without risk. Systemic anticoagulation with heparin during angioaccess could result in longer operative times, which would be necessary to achieve hemostasis, or an increase in the incidence of perioperative bleeding complications. We undertook this study to determine the value of systemic anticoagulation in patients undergoing vascular access surgery and the effect of this practice on procedural outcomes.
Methods
Participants included adult patients with end-stage renal disease in need of permanent vascular access for hemodialysis. These patients presented to the vascular surgical service at a single center from July 2004 to March 2006. Patients were considered suitable candidates for this study if they required maintenance hemodialysis and lacked permanent vascular access. Patients were excluded if they were undergoing revision of a previously placed fistula or graft. All procedures were performed by a single surgeon (M. D.), and written informed consent prior to enrollment was obtained from all patients.
Preoperative assessment with noninvasive upper extremity arterial and venous studies was routinely performed in the vascular laboratory to determine the best site for creation of an arteriovenous fistula or the need for arteriovenous graft insertion. Patients who lacked autogenous vein larger than 2.5 mm in diameter underwent insertion of an arteriovenous polytetrafluoroethylene (PTFE) graft. Published criteria were used to determine the optimal location and configuration for vascular access.1, 2 When choosing an access site, preference was given to the nondominant extremity, and when necessary, temporary vascular access was maintained by use of tunneled hemodialysis catheters until permanent functional access was obtained.
All operative procedures were performed under local anesthesia with intravenous sedation. After determining the most suitable vascular access procedure, patients were randomized to receive systemic anticoagulation in the form of unfractionated heparin at a standard dose of 5,000 units intravenously or no systemic heparin. Heparin was administered intraoperatively, just prior to vessel occlusion. The dose of heparin administered systemically was not adjusted according to weight. Local irrigation with heparinized saline was used in patients not receiving systemic heparin. All anastomoses were performed with 7-0 suture in running fashion. Protamine was not used perioperatively to reverse the anticoagulant effect of heparin. Patient demographics, including age, gender, race, and comorbid medical conditions such as diabetes mellitus, hypertension, dyslipidemia, and presence of coronary artery disease, were noted. Outcome variables included operative time, incidence of complications, and patency. Outcome data were recorded in accordance with recommended standards on reporting for arteriovenous hemodialysis access.3 Patients were followed in an outpatient setting and examined at monthly intervals until failure or a functional access was obtained. Follow-up was also obtained at 3-month intervals postprocedure.
Statistical analysis was performed using commercially available software (R software; R Foundation, Vienna, Austria).4 Linear regression was used to compute mean operative times and to adjust for age, gender, and race. Logistic regression was used to compute 30-day and 3-month functional patency and complication rates and to adjust for age, gender, and race. All mean values are reported along with their standard error (SE). Analysis of patency data was performed using the intention to treat rule. Survival analysis was used with the log-rank test for differences in patency for the heparin and no heparin groups. All analyses followed that suggested by the Committee on Reporting Standards of the Society for Vascular Surgery.3
Results
A total of 115 patients were randomized in this study, with 58 receiving no systemic anticoagulation and 57 receiving 5,000 units of unfractionated heparin intravenously. Three patients were lost to follow-up and excluded from analysis. All three of these patients underwent arteriovenous fistulas, and one received heparin. No perioperative complications occurred in these three patients. The remaining 112 patients, 56 in each group, form the basis for this report. Arteriovenous grafts were inserted in 31 of the 115 patients. Of the 31 patients who underwent arteriovenous graft insertion, 18 received systemic heparin and 13 did not. Systemic heparin was given to 38 of the patients undergoing arteriovenous fistulas (Table I). The mean age of patients in the no heparin group was 60 years (SE = 1.8). This did not differ from the mean age of patients in the heparin group, which was 62 (SE = 1.8, p = 0.4). Both groups also had a comparable gender distribution, no racial differences, and similar comorbid conditions (Table II).
Table I. Distribution of arteriovenous (AV) fistulas and grafts in the systemic heparin and no heparin groups
| AV fistulas | AV grafts | |
|---|---|---|
| Systemic heparin | 38 | 18 |
| No systemic heparin | 43 | 13 |
Table II. Demographics and baseline characteristics by study group
| Measure | No heparin | Heparin | p |
|---|---|---|---|
| Age (years, mean) | 60 (SE = 1.8) | 62 (SE = 1.8) | 0.40 |
| Sex (% male) | 53% | 57% | 0.63 |
| Race (%) | 0.26∗ | ||
 Black | 40% | 54% | |
| Hispanic | 19% | 21% | |
| White | 34% | 23% | |
| Other | 7% | 2% | |
| Comorbidities (%) | |||
| DM | 0.66 | 0.47 | 0.05 |
| HTN | 0.90 | 0.88 | 0.97 |
| Dyslipidemia | 0.22 | 0.28 | 0.48 |
| CAD | 0.36 | 0.38 | 0.79 |
The mean operative time for the no heparin group was 79 min (SE = 4.7), not significantly different from the mean operative time for the heparin group, which was 86 min (SE = 4.7, p = 0.31). Arteriovenous fistulas had a mean operative time of 71 min (SE = 3.3), significantly lower than the 114 min for arteriovenous grafts (SE = 5.4, p = 3e−10). There were no differences in operative times for arteriovenous fistulas receiving no heparin compared to those receiving heparin, at 70 min (SE = 4.9) and 73 min (SE = 4.8, p = 0.64), respectively. Likewise, there were no differences in operative times for arteriovenous grafts whether or not they received heparin, with 116 min (SE = 12) for the no heparin group and 113 min (SE = 9.8, p = 0.60) for the heparin group.
No thrombotic events occurred intraoperatively. Perioperative bleeding complications were seen in 14 patients. Bleeding from the operative site was seen in the postanesthesia care unit in 10 patients, and two of these patients developed a wound hematoma. Wound hematoma was also seen in four other patients with no external bleeding. None of these patients required operative intervention. A total of four blood transfusions were necessary in three patients, and thrombosis of the vascular access was seen in three patients with bleeding complications. These three patients with thromboses of their access all underwent arteriovenous fistulas. All of the bleeding complications except for one hematoma were seen in the heparin group, and this was statistically significant (p = 0.008). The incidence of bleeding complications was 1.8% for the no heparin group and 23% for the heparin group. Additional complications included perioperative myocardial infarction in one patient and hand ischemia secondary to a steal syndrome in another, with both occurring in the heparin group. The patient with hand ischemia had a brachial-axillary arteriovenous PTFE graft inserted and required ligation of her graft 6 months postoperatively. There were no perioperative deaths. However, 14 deaths occurred in the follow-up period. One patient underwent successful renal transplantation.
The 30-day primary patency rate was 86% for the no heparin group and 84% for the heparin group (p = 0.79). The functional patency rate was 68% for both groups at 3 months (p = 0.99). Controlling for the procedure type, arteriovenous fistula versus arteriovenous graft, there was no significant difference in 30-day primary patency (p = 0.30) or 3-month functional patency (p = 0.18). Arteriovenous fistulas receiving no heparin had a 30-day primary patency rate of 82%. The 30-day primary patency for arteriovenous grafts receiving no heparin was 100%. There were also no differences in the 30-day primary patency for arteriovenous fistulas and arteriovenous grafts receiving heparin, at 84% for both groups. The functional patency rate at 3 months for the no heparin group was 61% for arteriovenous fistulas and 92% for arteriovenous grafts. For the heparin group, arteriovenous fistulas and arteriovenous grafts both had a 3-month functional patency of 68%. Logistic regression analysis showed that age, gender, operative times, and incidence of complications had no effect on either the 30-day or 3-month functional patency rate (Table III).
Table III. Primary patency rates at 1- and 3-month follow-up for the groups with heparin and without heparin
| Period | No heparin | Heparin | p |
|---|---|---|---|
| 30-day | 0.86 | 0.84 | 0.79 |
| 3-month | 0.81 | 0.80 | 0.92 |
| 3-month functional patency | 0.68 | 0.68 | 0.99 |
A Kaplan-Meier analysis was run for patency. Figure 1 shows the probability of patency versus days since surgery for the two groups: no heparin and heparin. A log-rank test on the difference between the two groups gave a chi-squared of 0.2 and a p value of 0.66, showing no significant difference. Controlling for procedure, arteriovenous fistula versus arteriovenous graft, did not make any difference (p = 0.82). Post-hoc power analysis suggested that a sample size of 112 patients would yield 80% power to detect a difference in outcome of 20%.
Fig. 1
The probability of patency versus days since surgery for the two groups: no heparin (solid line) and heparin (dashed line). There were no significant differences between the two groups.
Discussion
Vascular access surgery for hemodialysis is associated with a high early failure rate.1, 2, 5, 6 While early failures are more common with autogenous arteriovenous fistulas than with prosthetic arteriovenous grafts, NKF-DOQI guidelines promote the use of primary arteriovenous fistulas whenever possible.1 These guidelines recommend placement of native fistulas in 50% of all new patients undergoing permanent hemodialysis access because of the higher long-term patency rate and lower incidence of infectious complications associated with such fistulas. The overall stated goal should be to achieve a 40% prevalence of arteriovenous fistulas as an access route in patients undergoing hemodialysis. So as to not discourage the creation of arteriovenous fistulas in patients with difficult anatomy, an acceptable early failure rate is not stated in these guidelines. However, early failure rates of 36-56% have been reported.2, 5
In an attempt to maximize the use of autogenous arteriovenous fistulas and improve the early patency rate of angioaccess procedures, Silva et al.2 proposed a protocol for the use of arterial and venous imaging in the noninvasive vascular laboratory. Using this approach, they reported a decrease in the early failure rate for arteriovenous fistulas from 36% to 8% and an increased utilization rate for arteriovenous fistulas from 14% to 63%.2 Such imaging protocols, which facilitate identification of suitable arterial and venous anatomy for angioaccess surgery, are now widely accepted and commonly used in an attempt to minimize early failures and improve long-term outcomes.7, 8 Other authors have reported improved outcomes with the use of interrupted vascular clips when compared with sutured anastomoses.9
To our knowledge, the effect of systemic anticoagulation on the early patency rate and outcomes of angioaccess procedures has not been previously studied. In our review of the literature, we found variable practice patterns regarding the intraoperative use of anticoagulants for angioaccess surgery. While some authors routinely heparinize all patients undergoing vascular access for hemodialysis, others do not or make use of heparinized saline solution administered locally.7, 8, 10 We conducted this study to clarify the role of systemic anticoagulation in patients undergoing angioaccess surgery. We chose to administer a standard dose of heparin and did not adjust this dose according to weight as this is a common practice in vascular surgery. Our results do not support the use of routine systemic anticoagulation for angioaccess surgery as no difference in patency was found. Contrary to our expectations, operative times were not prolonged in patients receiving systemic anticoagulation. However, minor perioperative bleeding complications were substantially higher in those patients who received intravenous heparin. Although three of these patients lost their access after developing wound hematomas, the incidence of bleeding complications did not statistically affect the patency rate. The patency rate was also unaffected by age, gender, or operative times, a finding consistent with that reported by other authors.5
Conclusion
The high early failure rate associated with angioaccess surgery is not affected by the intraoperative administration of systemic anticoagulation. Systemic anticoagulation is associated with a higher bleeding risk perioperatively, although these bleeding complications are minor and do not alter the long-term patency of angioaccess procedures. Our results do not support the routine use of systemic anticoagulation for angioaccess surgery.
References
- National Kidney Foundation. K/DOQI clinical practice guidelines for vascular access. Am J Kidney Dis. 2001;37(Suppl. 1):S137–S181
- A strategy for increasing use of autogenous hemodialysis access procedures: impact of preoperative noninvasive evaluation. J Vasc Surg. 1998;27:302–307
- Recommended standards for reports dealing with arteriovenous hemodialysis accesses. J Vasc Surg. 2002;35:603–610
- R Development Core Team. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, 2006, URL http://www.R-project.org.
- Longitudinal comparison of dialysis access methods: risk factors for failure. J Vasc Surg. 1997;26:1009–1019
- . Impact of secondary procedures in autogenous arteriovenous fistula maturation and maintanence. J Vasc Surg. 2001;34:866–871
- Prospective validation of an algorithm to maximize native arteriovenous fistulae for chronic hemodialysis access. J Vasc Surg. 2002;36:452–459
- Basilic vein transposition fistula: a good option for maintaining hemodialysis access site options?. J Vasc Surg. 2004;39:1043–1047
- A multicenter study of permanent hemodialysis access patency: beneficial effect of clipped vascular anastomotic technique. J Vasc Surg. 2003;38:229–235
- Changes in the practice of angioaccess surgery: impact of dialysis outcome and quality initiative recommendations. J Vasc Surg. 2000;31:84–92
Presented at the Annual Meeting of the Society of Clinical Vascular Surgery, Orlando, FL, March 21-24, 2007.
PII: S0890-5096(07)00306-8
doi:10.1016/j.avsg.2007.09.002
© 2008 Annals of Vascular Surgery Inc. Published by Elsevier Inc All rights reserved.
