Results of Prompt Surgical Intervention in Hemodialysis Radiocephalic Fistula Secondary Dysfunction

End-stage renal disease (ESRD) incidence and the number of patients undergoing dialysis are on the increase throughout the world.¹ Vascular access is the prerequisite for hemodialysis treatment and is directly related to morbidity and mortality in hemodialysis patients.² Autogenous arteriovenous fistulas (AVFs) provide longevity for hemodialysis patients with high dialysis efficiency and quality of life with low morbidity. The National Kidney Foundation's Kidney Disease Outcome Quality Initiative (NKF-K DOQI) recommends that the first choice of vascular access in hemodialysis patients should be autogenous AVF, and radiocephalic fistulas (RCFs) should be the first among them.³ RCFs are AVFs that are easy to create and use. They cause less complications, protect the proximal vessels, and provide the ideal blood flow for hemodialysis.

Besides their known advantages, RCFs have such disadvantages as high primary failure rate, maturation problems, or low patency rates.^4–9 It is believed that these disadvantages will become more important as the population of elderly people increases. For the quality of life in patients with RCF dysfunction, it is crucial to provide RCF refunction and hemodialysis continuation and thus to prevent catheter placement, and when not possible, to shorten the use of a catheter, to protect the limited vascular structure and to perform an operation with high success and patency.¹⁰ To achieve all these, surgical or endovascular procedures are being used at present. In surgical procedures, the routine practice is revisions performed in the forearm. Early AVF-saving operations in the upper arm can also yield the same benefits.

Materials and Methods

Inclusion criteria for the study were: (1) a RCF created at least 3 months previously, (2) sufficient blood flow (>300 mL/min) in the hemodialysis for at least 1 month, and (3) a surgical intervention for RCF dysfunction at the level of snuff box or wrist without catheter placement. Between June 2005 and February 2011, 1303 patients underwent AVF or AV graft operations in Acibadem University Hospital and Mustafa Kemal University Hospital. Of these, 295 patients who met the criteria were evaluated retrospectively.

An AVF use procedure was created in our clinics and dialysis nurses (nurses who have certification for hemodialysis cannulation and treatment follow-up), dialysis physicians, and hemodialysis patients were educated by vascular surgeons about AVF use procedures. In this context, the protocols to be followed in AVF dysfunctions were determined. With the coordinated work of the patients, the dialysis center and our surgical clinics, after patients, physicians, or nurses experienced AVF problems (such as thrill loss), patients were enabled to reach our surgical clinics as soon as possible. Patients who developed RCF dysfunction were routinely evaluated by a surgeon by physical examination (PE) and color Doppler ultrasonography (USG). With these procedures, arterial and venous anatomy and stenosis or thrombosis areas were evaluated, and thus the site and method of the operation were determined.

According to the surgical protocol used in our clinics for RCF dysfunction, the patients who were detected to have perianastomotic stenosis or thrombosis underwent new proximal anastomosis (NEO) if they had a suitable venous structure in the forearm. In the proximal NEO procedure, the radial artery and cephalic vein were reached with an incision created proximal to the previous anastomosis. Thus, the patients could enter dialysis through this newly created NEO 24–48 hours after the procedure. When the patients did not have enough or suitable venous structure, brachiocephalic AVF was performed if their upper arm cephalic vein diameter was over 3 mm in the evaluation with a tourniquet, and if the vein structure was suitable. These patients were evaluated for the thrill, and if thrill was established, they began to have hemodialysis through the newly created AVF 24–48 hours after the operation. In patients whose fistula was considered immature, a temporary catheter was placed surgically in internal jugular veins and the hemodialysis was performed through this catheter. During the follow-up, if the AVF was believed to be mature, the hemodialysis was performed through the AVF, and the catheter was removed after 3 trouble-free dialyses. In patients who were not suitable for brachiocephalic AVF but had a median basilic vein diameter of 3.5 mm or larger, side-to-side brachiobasilic AVF + basilic superficialization was performed and the patients were able to have hemodialysis within 24–48 hours (Fig. 1).

View Figure

• Download as Powerpoint
• Download Figure

Surgical Techniques

Operations were performed routinely under local anesthesia; five-twenty mL Citanest 2% Flakon (agent prilocaine) was used as local anesthetic. Uncooperative patients underwent intravenous sedoanalgesia with Dormicum (midazolam) and or fentanyl.

In the proximal NEO procedure, the radial artery and cephalic vein were reached with an incision created proximal to the previous anastomosis. The branches of the cephalic vein were first ligated and incised, then the vein was ligated at the distal end and incised. If the patient was thought to have a thrombus in the vein, the thrombus was removed by pressing out at the proximal end of the venotomy. When the thrombus could not be removed completely, the complete removal was achieved with thrombectomy using a 4 F Fogarty catheter. After these procedures, a 14 F cannula was placed on the open end of the vein. The vein was then washed with a solution of 100 mL physiological saline (PS) and 5000 IU heparin and was dilated. If any resistance was encountered during PS administration, vein passage was controlled with vascular dilators. In patients with vein stenosis, venous dilatation was performed with vascular dilators. All the accessible branches of the radial artery were ligated and incised. To provide arterial dilatation during the operation, Papaverine and Citanest were administered locally. In all cases, anastomosis was performed with a side-to-side method using 7/0 Prolene. In patients with an adequate thrill, detected by palpating the venous outflow from the anastomosis, the distal end of the vein is ligated and the anastomosis was converted into an end-to-side type. In patients with weak or no thrill, the artery was entered through the distal end of the open vein with vascular dilators (2.5–3.0–3.5 mm), and the artery was dilated. But if sufficient thrill still could not be provided, the arterial system was dilated with a 4 F Fogarty thrombectomy catheter starting from the elbow area. When it was detected that there was enough thrill, the distal end of the vein was ligated at the end of the anastomosis and converted into an end-to-side type.

In the operation of conversion into brachiocephalic AVF, the cephalic vein was prepared as was done in the proximal NEO operation. Side-to-side anastomosis was performed in all patients and the anastomosis was converted into an end-to-side type. In brachiocephalic AVF, if the cephalic vein could not be freed sufficiently, brachiocephalic AVF was created using a 6 mm short interposition polytetrafluoroethylene graft between the brachial artery and the cephalic vein.

In the brachiobasilic AVF + basilic vein superficialization operation, after the preparation of the brachial artery and median-basilic vein, side-to-side anastomosis was performed using a 7/0 Prolene continuous suture. After anastomosis, basilic vein superficialization was performed through a second incision 8–10 cm in length.

Definitions

Statistics

Data were entered into a Microsoft Excel (Microsoft Corporation, Redmond, Washington) spreadsheet for analysis. Statistical analysis was performed using the SPSS software (version 15, SPSS Inc, Chicago, Illinois). Kaplan–Meier analysis was used to determine the primary and secondary patency rates.

Results

Of the patients, 68.5% (n = 202) were male and 31.5% (n = 93) were female. The mean age was 58.4 (range: 22–80). Of these patients, 74.2% (n = 219) had hypertension (HT), 65.4% (n = 193) had diabetes mellitus (DM), 24.7% (n = 73) had peripheral arterial disease (PAD), 16.9% (n = 50) had coronary artery disease (CAD), and 30.2% (n = 89) had smoking history (Table 1).

Table 1 Patient characteristics

View Fullscreen

Of the 295 patients who underwent surgical intervention for RCF dysfunction between June 2005 and February 2011, 77.28% (n = 228) underwent proximal NEO, 14.23% (n = 42) underwent brachiocephalic AVF, and 8.47% (n = 25) underwent brachiobasilic AVF + basilic vein superficialization.

A total of 95.6% (n = 218) of the 228 patients who underwent NEO started to have trouble-free dialysis within 24 to 48 hours. In 4.4% (n = 10) of the patients, a temporary catheter was placed and dialysis was performed through this catheter. Five of these catheters were placed due to complications (vascular injury, hematoma, etc.) that developed during the operation or service follow-up to allow healing in the surgical area. The other 5 NEOs were placed in patients whose NEO operations were unsuccessful, or whose NEOs could not be used due to complications (thrombosis, low flow, etc.) that developed during follow-up (Table 2). In these patients, an AVF was created in the elbow area 5–15 days later. When the AVF matured, temporary catheters were removed. In NEO performed patients, the rate of the temporary catheter was 4.4%, and the mean duration of temporary catheter use was 18.8 ± 13.8 days (range: 5–40 days).

Table 2 Postoperative early and late complications

View Fullscreen

Within the follow-up period, 51 patients were withdrawn from the study. Some were excluded due to death or transplantation, and some were excluded because they could not be reached. The surgical success in patients who underwent NEO was 98.7%. The primary patency rates for 1, 2, and 3 years were 91.2%, 85.3%, and 77.9% and the secondary patency rates were 92.7%, 88.8%, and 82.8%, respectively.

The 61.9% (n = 26) of 42 patients who underwent brachiocephalic AVF started to have trouble-free dialysis within 24 to 48 hours. A temporary catheter was placed in 38.1% (n = 16) of the patients. Thirteen of these catheters were placed due to complications (vascular injury, hematoma, etc.) that developed during the operation or follow-up to allow the surgical area to heal. In 1 patient, the catheter was placed due to the thrombosis that developed in the AVF during the follow-up. The catheter was kept in place until the AVF created in the other arm in another session became mature. Since the vein diameters were not suitable, catheters were used in 2 patients for longer times. In brachiocephalic AVF performed patients, the rate of the temporary catheter was 38.1%, and the mean temporary catheter use duration was 12.8 ± 9.7 days (range: 7–45 days).

Within the follow-up period, 15 patients were withdrawn due to death or transplantation, and some were withdrawn because they could not be reached. The surgical success in patients who underwent brachiocephalic AVF was 100%. The primary patency rates for 1, 2, and 3 years were 86.1%, 74.2%, and 59.3%, and the secondary patency rates were 91.6%, 83.8%, and 74.0%, respectively.

The 72% (n = 18) of 25 patients who underwent brachiobasilic AVF + basilic vein superficialization started to have trouble-free dialysis within 24 to 48 hours. To provide this, the arterial needle was inserted at the distal end of the anastomosis, and the venous needle was inserted into the other superficial veins in the arms or feet. The superficialized basilic vein segment became ready for use in about 3 weeks. Thus, both needle entries became suitable for AVF creation. The rate of temporary catheter placement in patients with brachiobasilic AVF + basilic vein superficialization was 29.2% and the mean duration of catheter use was 15.1 ± 6.8 days (range: 7–25; Table 3).

Table 3 Rate and duration of temporary catheter use in dysfunctional radiocephalic arteriovenous fistulas after intervention

View Fullscreen

Within the follow-up period, 13 patients were excluded due to death or transplantation, and some were excluded because they could not be reached. The surgical success in patients who underwent brachiobasilic AVF + basilic vein superficialization was 100%. The primary patency rates for 1, 2, and 3 years were 80.0%, 62.5%, and 41.7% and the secondary patency rates were 85.0%, 68.8%, and 50.0%, respectively (Table 4).

Table 4 Primary and secondary patency rates in dysfunctional radiocephalic arteriovenous fistulas after intervention

View Fullscreen

In total, 88.5% (n = 262) of the 295 patients who underwent the aforementioned operations due to RCF dysfunction were able to continue trouble-free dialysis without catheter placement. In 11.2% (n = 33) of the patients, catheter placement was performed and the mean period of catheter use was 15.1 ± 10.7 days (range: 5–45).

The surgical success of the interventions performed due to RCF dysfunction was 98.9%. The primary patency rates for 1, 2, and 3 years were 91.8%, 85.9%, and 77.9%, and the secondary patency rates were 94.1%, 90.6%, and 84.5%, respectively (Figs. 2, 3).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Discussion

In the meta-analysis of 8 prospective and 30 retrospective studies conducted on RCFs by Albee et al in 2013, the 12-month primary patency rate of the RCFs was reported to be 62.5% (95% CI, 54.0%–70.3%), and the secondary patency rate was reported to be 66.0% (95% CI, 58.2%–73.0%).¹¹ According to the results of this meta-analysis, RCFs had medium patency rates, and one third of the RCFs became irreversibly dysfunctional within a year.¹² Additionally, in the results of the meta-analysis, the primary and secondary patency rates were very close to each other, and this indicates that the revision procedures that make the RCFs functional again were rarely performed or not performed effectively.

RCFs often require a second intervention due to low maturation rates and medium patency rates.^13,14 Since they are time-saving, practical, and relatively successful, endovascular procedures have attracted technical interest and have been used more widely in recent years. The technical advances in endovascular procedures have increased the success rates and decreased the rates of complications, such as thrombus migration, vessel rupture, infection, hand ischemia, remote or local bleeding, and pseudoaneurysm.^15–17 Despite the high technical success, endovascular treatment has low patency rates. Furthermore, comparative studies revealed that in endovascular treatment, the restenosis rates and pulmonary embolism risk are higher than the surgical intervention.^15,18,19 There are very few studies that compare surgical and endovascular treatment of thrombosed AVFs. The most extensive study on this issue was made by Turmel-Rodrigues et al¹⁶ In this study, published results are given from 93 endovascular interventions applied to 73 patients. The thrombosis in these patients was reported as 56 in the forearm and17 in the upper arm. According to these results, the technical success rate was 93% for the forearm, and 76% for the upper arm. The 1-year primary and secondary patency rates were 49% and 81%, respectively.

Tordoir and colleagues compared endovascular treatment to surgical treatment of thrombosed AVF. In this meta-analysis study, the technical success rate of endovascular treatment varied between 73% and 93%, and the 1-year primary patency rate ranged from 18% to 70%, with the 1-year secondary patency rate ranging from 44% to 89%.

In the same study, surgical treatment in patients ranged from 82% to 100% technical success rate and the 1-year primary patency rate ranged from 51% to 93%, with the secondary patency rate ranging from 69% to 95%. In the study of Tordoya et al, in autologous AVFs both surgical treatment and endovascular treatment had technical achievement over 90%, but the primary and secondary patency rates as compared with surgery in endovascular treatment was very low.¹⁸ In our study, the success of the operation was 98.9%. The primary patency rates for 1, 2, and 3 years were 91.8%, 85.9%, and 77.9%, and the secondary patency rates were 94.1%, 90.6%, and 84.5%, respectively, consistent with the results of other studies.

Treatment of RCF dysfunction is controversial due to the variable treatment choices. According to the international guidelines, each center should establish its own treatment protocols on the basis of its own resource and the experience of its employees.²⁰ Although randomized studies on the ideal treatment method to be used in saving RCFs are quite few, the available evidence suggests that surgery is better.^21–23 The procedures used in saving dysfunctional AVFs can successfully be performed in immature fistulas as well as obstructed RCFs. In our clinic, our preference for RCF dysfunctions is to use surgical methods due to high surgical success and good clinical results in the long term.

NEO is the routinely performed surgical method for RCF dysfunctions. There are publications reporting that this method has been performed with high success and low restenosis rates.^24–26 The procedure to create a new anastomosis at the proximal end of the radial artery using the mature cephalic vein is called NEO. This surgical procedure provides the re-use of the invaluable forearm veins, and prevents switching to the elbow area.

The main cause of RCF dysfunctions is the problem in the perianastomotic area. The creation of a new anastomosis with NEO using mature radial artery and cephalic vein allows us to bypass the problem in the perianastomotic area. The ideal fistula flow can immediately be reached with this procedure, and the venous bed is ready for early use, and thus the patients can have dialysis without catheter placement or without setting back their sessions. Additionally, new fistula chances, which are crucial for hemodialysis patients, would not be wasted.

Similarly, RCFs provide favorable conditions for early surgical interventions in the vascular structure of the upper arm. In patients whose forearm anatomy is not suitable for NEO, upper arm-saving operations would provide a chance for early use, and thus would decrease the catheter requirement and the duration of catheterization.

A catheter is the only choice for hemodialysis in patients without permanent vascular access.^27–29 However, catheters lead to complications, such as infection, thrombosis, venous stenosis, or vascular injury.³⁰ Furthermore, catheter placement deteriorates the quality of life. Central venous stenosis that develops due to the catheter is associated with the duration of catheter use, and can eliminate the chances of AVF or AV grafts that can be performed in that extremity. Besides the known complications, a patient who has lost his/her AVF is in a serious psychologic trauma. The catheter placement procedure is the most feared and anxiety-creating procedure for dialysis patients. Hemodialysis with a catheter leads to inadequate dialysis, decreases the quality of life, and reduces the desire to live.

Avoiding catheter use in hemodialysis patients is crucial, but if not possible, shortening the length of use is also important.

One of the most important factors in dysfunctional AVFs is the intervention time. Intervention in a dysfunctional AVF within the first 24 to 48 hours would prevent catheter placement, and thus the quality of life would improve and survival would increase.^15,21,31,32 Delayed intervention can lead to thrombus, make the intervention more difficult and reduce the rate of success. Studies on this subject have shown that the success rates in interventions carried out within the first 48 hours reach up to 96%.¹⁵ In conclusion, interventions in the early period increase the chance to save the AVF and decrease the catheter requirement of the patient.³¹

The increase in the number of vascular surgeons is important for early intervention. It is also important to diagnose the AVF dysfunction as early as possible and direct the patient to a surgeon. Therefore, the patient, the dialysis team, and the vascular surgeon should work in close cooperation. In our clinic, patients with RCF dysfunction can be intervened within 24 to 48 hours due to the education given to the patients and dialysis staff.

To provide patients with all these advantages, patient and dialysis staff education, and patient, dialysis team, and surgeon cooperation are crucial in the early intervention. In patients with RCF, the vascular system of the entire arm should be evaluated with USG before catheter placement, and if there is no chance for forearm revision, the upper arm AVF choices, brachiocephalic or brachiobasilic AVF operations, should be performed in patients whose vascular structure is suitable.

Although the effectiveness and success rate of early intervention are known, early intervention is not a routine practice. This indicates that there is a need to develop a new point of view in evaluating AVF dysfunctions. In the near future, RCF dysfunctions must be considered as a clinical condition requiring immediate surgical intervention.