© 1991 S. Karger AG, Basel 0250-8095/91/0115-036952.75/0
Am J Nephrol 1991;11:369-373
Early Cannulation of Plasma TFE and Gore-Tex Grafts for Hemodialysis: A Prospective Randomized Study Gregory Jaffers, John D. Angstadl, James S. Bowman III Department of Surgical Transplantation, Wilford Hall USAF Medical Center, Lackland Air Force Base, Tex., USA
Key Words. Access • Dialysis • Grafts • Cannulation • Gore-Tex • Plasma TFE Abstract. Fifty-one consecutive vascular access procedures were randomized to either the Medtronic plasma TFE or Gore-Tex polytetrafluoroethylene (PTFE) conduits in patients requiring immediate dialysis from December 1989 to April 1990. There were 49 forearm loop fistulas and 2 upper arm grafts. Fifty of these fistulas were cannulated within 48 h of placement to avoid use of subclavian venous catheters for hemodialysis. Complications related to the early cannulation of these fistulas included 2 hematomas in the plasma TFE group, and 3 hematomas in the GoreTex group (p = 1.00). Two patients with Gore-Tex grafts were systemically heparinized prior to hematoma formation after thrombectomy of their accesses. There were no adverse sequelae in these 5 patients, and none of the hematomas interfered with further dialytic therapy. One patient in the plasma TFE group and 3 patients in the Gore-Tex group developed cellulitis within the first month of placement (p = 0.65). All were treated with intravenous vancomycin with resolution of the erythema. None of the plasma TFE and 3 of the Gore-Tex fistulas thrombosed within 30 days of placement (p = 0.22). All were salvaged by thrombectomy. Both the plasma TFE and Gore-Tex vascular conduits may be used after surgical placement for early dialytic therapy and are associated with minimal early complications. The early use of these fistulas may eliminate the need for subclavian venous cannulation in most patients with renal failure, thus diminishing the incidence of subclavian venous stenosis and thrombosis. Further observation of these grafts will be necessary to determine the effect of immediate cannulation on their long-term performance for hemo dialysis.
Subclavian venous cannulation provides a ready means of vascular access for patients requiring imme diate dialytic therapy. However, it is apparent that the use of this route for hemoaccess is associated with signif icant long-term morbidity due to subclavian vascular injuries [1-6]. These lesions may be progressive and result in severe stenosis or thrombosis of the subclavian or innominate veins, or the vena cava, significantly reducing the successful outcome of a vascular access sub sequently placed in the ipsilateral upper extremity [712], Due to an increasing number of central vein injuries encountered at our center with the use of subclavian
dialysis catheters, an attempt was made to eliminate sub clavian venous cannulation by using a surgically placed prosthetic upper extremity graft within 48 h of place ment. This report describes the results of a randomized prospective study comparing the early use of 25 Med tronic plasma TFE conduits with 25 Gore-Tex polytetra fluoroethylene (PTFE) hemodialysis conduits which were cannulated within 48 h for dialytic therapy. Patients and Methods All accesses were created under local, monitored anesthesia using 1% xylocaine. The 6-mm Medtronic plasma TFE or the 6-mm Gore-Tex graft was used. Two Gore-Tex grafts were placed in the
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
Introduction
JafTers/Angstadt/Bowman
370
Table 1. Group characteristics of fistulas cannulated within 24 h of surgical placement Gore-Tex
9/16 51 10
10/15 58 11
Results
Table 2. Timing of initial cannulation of prosthetic fistulas after surgical placement Hours to cannulation 24 2 0
upper arm. The remainder were all placed in a looped configuration in the forearm. Plasma TFE grafts were not preclotted. The vessels were isolated through transverse extremity incisions using vascular tapes. The venous anastomosis was completed first using a 5-0 or 6-0 prolene suture. The conduit was tunnelled through the subcuta neous tissue using a 6-mm tunnelling device. The arterial anastomo sis was then completed using a similar suture. Heparin was not rou tinely utilized. The incisions were closed using 3-0 dexon in the subcutaneous tissue and 4-0 Vicryl or 4-0 PDS as a running subcu ticular stitch. All patients without a functional hemodialysis access requiring dialysis within 48 h were considered for entry into the study. Con sent was obtained from all patients prior to their participation in the study. Randomization was accomplished by sealing 50 3 X 5 cards in 50 envelopes, half of which had ‘Plasma TFE’ printed on them, the other half ‘Gore-Tex’. A sealed envelope was selected and opened in the operating room when it was certain that a vascular conduit was to be used. A single dose of vancomycin was given prcopcratively for prophylaxis against infection in all patients. The patients were followed prospectively in the clinic oi by phone inter views with the patients or their dialysis center nurses. Data were updated continuously in a computer database. Initial dialysis treatments were accomplished utilizing 17-gauge needles. All dialysis nurses and technicians participated in the ini tial cannulation of these fistulas. Activated clotting time (ACT) was utilized to determine the amount of heparinization required for dialysis. If the initial ACT was less than 110s, 1,000 units of heparin were given upon initiation of dialysis and then hourly until comple tion of the treatment. With an ACT of greater than 110 s, no initial heparin was utilized, and the kidney was flushed with normal saline hourly. If evidence of clot formation became apparent in patients with an ACT greater than 110s, 1,000 units of heparin were admin istered and continued hourly until completion of the dialysis treat
Twenty-five plasma TFE and 25 Gore-Tex vascular conduits were cannulated within 48 h of placement for early hemodialysis. All 50 patients were prospectively followed after placement of their Fistulas. One patient who had a forearm Gore-Tex fistula created, thrombosed her access within 12 h. It was deemed that the access was not salvageable and she was excluded from the study since her access was never cannulated. Her 3 X 5 card was reentered for another patient. Nine of the 25 patients who received the Medtronic graft were male and 16 were female. The average age of patients in this group was 51 years. Ten of these recipi ents were diabetic. Of the 25 patients with Gore-Tex grafts 10 were male and 15 were female. The average age in this group was 58 years, and 11 patients were diabetic (table 1). One diabetic patient with a left forearm Gore-Tex fis tula developed severe extremity ischemia. Bilateral up per extremity angiography revealed severe segmental atherosclerotic disease. The access was ligated and a Permacath was subsequently placed into the internal jugular vein. A second diabetic patient with a forearm Gore-Tex fistula developed ipsilateral extremity ischemia requir ing eventual ligation of her access and initiation of peri toneal dialysis. One plasma TFE graft was cannulated within 12 h of surgery. No hematoma, cellulitis or infection developed within the first 30 days of initial cannulation in this patient. None of the Gore-Tex grafts were cannulated within 12 h of surgical placement. Forty-seven patients had their first dialysis treatment from 12 to 24 h after surgical placement of which 22 were plasma and 25 Gore-Tex grafts. One patient with a plasma TFE conduit and 3 with Gore-Tex grafts developed cellulitis within the first month of placement (p = 0.60). All of the patients responded to antibiotic therapy with complete resolution of the cellulitis. Five hematomas occurred within the first 30 days of initial cannulation, 2 in the plasma TFE group and 3 in the Gore-Tex group (p = 1.0). Of the 2 hematomas which occurred in patients with plasma TFE grafts, 1 developed immediately after surgi cal placement before cannulation for dialysis and the other after cannulation for dialysis. There were no fur-
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
Male/female Average age, years Diabetics
Plasma TFE
ment. Upon decannulation of the access, gentle pressure was held over the puncture sites for 15 min. Observed proportions were compared with the two-tailed Fish er’s exact test.
Early Access Cannulation
371
Table 3. Morbidity with early cannulation (> 48 h) after access placement Hours to cannulation
p
Plasma TFE < 12 Hematomas Infections Thromboses-'
0 0 0
Total
12-24 1 1 0
>24 1 0 0
> 12 0 0 0
3
12-24
>24 0 0 0
3 3 3
1.0 0.60 0.22
9
0.10
< 1 month after initial cannulation.
ther hematomas and no resultant infections after these single bleeding events. Three patients with Gore-Tex fis tulas had hematomas develop along the tunnel. Two were systemically heparinized after successful thrombec tomy of their access within the first day of placement. One of these hematomas developed upon cannulation of the access with a 21-gauge needle for an intraoperative fistulagram during the thrombectomy procedure 12 h after initial placement of the conduit. The second patient similarly received intravenous heparin after thrombec tomy of the access immediately after surgery. Both fistu las were salvaged without sequelae. Neither developed an infection related to the hematoma, and both were cannulated for hemodialysis the following day. One other patient with a Gore-Tex graft developed a hema toma after his second dialysis treatment despite having no difficulties with the first attempt. This was felt to be related to a through and through puncture of the con duit. Two plasma grafts and none of the Gore-Tex grafts were cannulated between 24 and 48 h after placement. There was no morbidity after cannulation of these 2 fis tulas (table 2). Complications related to the early use of these fistulas are summarized below (table 3). There were no statistically significant differences between these groups (p = 0.45). Problems encountered with both types of grafts, unre lated to their early use for dialysis, included a forearm steal syndrome in 3 patients who received plasma TFE fistulas and 1 patient who received a Gore-Tex fistula. All 4 of these grafts required banding. The Gore-Tex fis tulas subsequently thrombosed after banding, and fur ther access attempts were abandoned due to severe pe ripheral vascular disease in this diabetic individual. Five patients developed significant arm edema after access
Table 4. Early morbidity unrelated to early cannulation: group characteristics
Forearm steal Edema Thrombosis before use Unrelated death < 1 month
Plasma TFE
Gore-Tex
3 2 3 3
1 3 3 0
placement, 2 in the plasma group and 3 in the Gore-Tex group, however, the edema did not preclude use of the access for immediate cannulation. Three patients with plasma TFE fistulas died within 1 month of their access surgery, of causes unrelated to their fistulas. Three pa tients with Gore-Tex grafts thrombosed their accesses before any attempts were made to cannulate these fistu las for hemodialysis. All were successfully thrombectomized and the fistulas were utilized within the subsequent 24 h. These unrelated complications after early cannula tion of the grafts are summarized below (table 4).
Discussion Since the introduction of the subclavian route for hemodialysis in 1969 [13], this method of central access to the vascular system for hemodialysis has received widespread use, particularly over the last decade, sup planting the various external shunts for immediate dialytic therapy. The apparent low morbidity initially en countered with the use of the subclavian catheter for dialysis appeared attractive when compared to the fre
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
;l
Gore-Tex
Jaffers/Angstadt/Bowman
372
to have no initial morbidity when cannulated within 48 h of placement. Since Gore-Tex has been the pre ferred conduit at our center, we elected to determine whether early cannulation of Gore-Tex was also feasible. There were no statistically significant differences in mor bidity between these two groups when the access was cannulated within 48 h of placement for hemodialysis. The success with early use of these grafts after their surgical placement is attributed to the use of a 6-mm tunneling device which ensures that the subcutaneous tissues are tightly apposed to the graft material. Addi tionally, the use of 17-gauge cannulas during the first dialysis treatment as well as maintaining pressure over the puncture site for at least 15 min after decannulation are also felt to be important in reducing the incidence of bleeding complications. Longer follow-up is obviously necessary to determine whether a difference in longevity will become important in attempts to maintain a patent vascular access in these patients with end-stage renal disease.
Acknowledgement The authors gratefully acknowledge the generous contributions of the dialysis staff.
References 1 Clark DD, Albina JE, Chazan JA: Subclavian vein stenosis and thrombosis: A potential serious complication in chronic hemodi alysis patients. Am J Kidney Dis 1990;15:265-268. 2 Selby JB, Tegtmeyer CJ, Amodeo C. Bittner L. Atuk NO: Inser tion of subclavian hemodialysis catheters in difficult cases: value of fluoroscopy and angiographic techniques. Am J Roentg 1989; 152/3:641-643. 3 Davis D, Petersen J, Feldman R, Cho C, Stevick CA: Subclavian venous stenosis: A complication of subclavian dialysis. JAMA 1984;252/24:3404-3406. 4 Barrett N, Spencer S, Mclvor J, Brown EA: Subclavian stenosis: A major complication of subclavian dialysis catheters. Nephrol Dial Transplant 1989;3:423-425. 5 Spinowitz BS. Galler MG, Golden RA. Rascoff JH, Schechter L, Held B, Charytan C: Subclavian vein stenosis as a complication of subclavian catheterization for hemodialysis. Arch Intern Med 1987;147:305-307. 6 Stalter KA, Stevens GF, Sterling WA: Late stenosis of the sub clavian vein after hemodialysis catheter injury. Surgery 1986; 100:924-927. 7 Cheung AK, Brownley R: Brachiocephalic vein occlusion in a hemodialysis patient. Am J Nephrol 1989;9:261-262. 8 Piotrowski JJ, Rutherford RB: Proximal vein thrombosis sec
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
quent episodes of thrombosis and infection encountered with the external shunts. Ease of placement of the sub clavian catheter as well as the need to sacrifice a periph eral vessel when placing the external shunts were further important factors in the transition to the use of these central veins. Unfortunately, a significant incidence of late compli cations related to subclavian venous stenosis or throm bosis has become apparent. These lesions may not be initially obvious unless venous collaterals or edema de velop in the ipsilateral upper arm or chest. Often, mas sive edema of the arm after access placement is the first evidence of a subclavian injury. It is clear that cannulation of the subclavian vein may result in severe stenosis or thrombosis of the vein in up to 19% of patients [10]. The loss of the central vein for outflow in this manner may be responsible for recurrent thrombosis of many fis tulas subsequently constructed in the upper extremity. The resultant arm edema may preclude extended use of the arms for hemoaccess [14]. Early use of subcutaneously placed dialysis grafts for hemodialysis has been described previously [15]. Var ious graft types including bovine, PTFE and Impra were utilized and cannulated within 72 h of placement with an overall incidence of postcannulation bleeding of 1.6% and infectious complications of 2.7% within 8 weeks. The success of early cannulation was attributed to the use of a tunneling device to ensure a compressive tunnel around the dialysis graft, thus reducing dead space for potential hematoma formation. A more recent study evaluated 48 PTFE grafts with an average time to cannu lation of 5.4 days with no graft infections [ 16]. No bleed ing complications were mentioned. These studies as well as our recent experience with immediate cannulation of the Medtronic plasma TFE and the Gore-Tex PTFE grafts demonstrate the ability to have a prosthetic access cannulated as soon as 12 h after placement for urgent hemodialysis. The Medtronic plasma TFE graft is made from woven polyethylene terephthalate which is lined with covalently bonded tetrafluoroethylene to provide surface properties similar to PTFE. This unique construction provides a conduit which can safely be cannulated within 72 h of placement for hemodialysis. This feature was particu larly attractive due to the recurrent difficulties encoun tered at our center with subclavian venous injuries from catheterization for dialysis. Although historically, the use of woven conduits (dacron) for hemodialysis fistulas has proven to be suboptimal, the plasma TFE grafts were evaluated in 10 patients prior to this study and appeared
373
Early Access Cannulalion
10
11
12
13
14
15 Taucher LA: Immediate, safe hemodialysis into arteriovenous fistulas created with new tunneler. Am J Surg 1985; 150:212— 215. 16 Haag BW, Paramesh V, Roberts T: Early use of polytetrafluoroethylene grafts for hemodialysis access: in Sommer BG, Henry ML (eds): Vascular Access for Hemodialysis II. Gore and Asso ciates and Precept Press, 1990, vol 2, pp 173-178.
Received: February 20, 1991 Accepted: July 25, 1991 Dr. Gregory Jaffers Department of Surgical Transplantation Wilford Hall USAF Medical Center Lackland Air Force Base TX 78236 (USA)
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
9
ondary to hemodialysis catheterization complicated by arterio venous fistula. J Vase Surg 1987;5:876-878. Aujla N, McCauley J, Sorkin M: Superior vena cava syndrome due to subclavian hemodialysis catheters. Milit Med 1990; 155/6:274-277. Vanherweghem JL, Yassine T, Goldman M, Vandenbosch G, Delcour C, Struyven J, Kinnaert P: Subclavian vein thrombosis: A frequent complication of subclavian vein cannulation for hemodialysis. Clin Nephrol 1986;26/5:235—238. Fant GF, Dennis VW, Quarles LD: Late Vascular complications of the subclavian dialysis catheter. Am J Kidney Dis 1986;3: 225-228. Schwab SJ. Quarles LD, Middleton JP, Cohan RH, Saecd M, Dennis VW: Hemodialysis-associated subclavian vein stenosis. Kidney Int 1988;33:1 156-1159. Urben J, Kvasnicka J, Bastecky J, Groh J: Experience with rou tine use of subclavian vein cannulation in hemodialysis. Proc Eur Dial Transplant Assoc 1969;6:59. Coates GR. Clements JB, Yium JB, Watlington J: Painful edema of the arm after insertion of single-needle subclavian vein dialy sis catheters: Pathogenesis and treatment. South Med J 1988:81: 303-305.
Am J Nephrol 1991;11:369-373
Early Cannulation of Plasma TFE and Gore-Tex Grafts for Hemodialysis: A Prospective Randomized Study Gregory Jaffers, John D. Angstadl, James S. Bowman III Department of Surgical Transplantation, Wilford Hall USAF Medical Center, Lackland Air Force Base, Tex., USA
Key Words. Access • Dialysis • Grafts • Cannulation • Gore-Tex • Plasma TFE Abstract. Fifty-one consecutive vascular access procedures were randomized to either the Medtronic plasma TFE or Gore-Tex polytetrafluoroethylene (PTFE) conduits in patients requiring immediate dialysis from December 1989 to April 1990. There were 49 forearm loop fistulas and 2 upper arm grafts. Fifty of these fistulas were cannulated within 48 h of placement to avoid use of subclavian venous catheters for hemodialysis. Complications related to the early cannulation of these fistulas included 2 hematomas in the plasma TFE group, and 3 hematomas in the GoreTex group (p = 1.00). Two patients with Gore-Tex grafts were systemically heparinized prior to hematoma formation after thrombectomy of their accesses. There were no adverse sequelae in these 5 patients, and none of the hematomas interfered with further dialytic therapy. One patient in the plasma TFE group and 3 patients in the Gore-Tex group developed cellulitis within the first month of placement (p = 0.65). All were treated with intravenous vancomycin with resolution of the erythema. None of the plasma TFE and 3 of the Gore-Tex fistulas thrombosed within 30 days of placement (p = 0.22). All were salvaged by thrombectomy. Both the plasma TFE and Gore-Tex vascular conduits may be used after surgical placement for early dialytic therapy and are associated with minimal early complications. The early use of these fistulas may eliminate the need for subclavian venous cannulation in most patients with renal failure, thus diminishing the incidence of subclavian venous stenosis and thrombosis. Further observation of these grafts will be necessary to determine the effect of immediate cannulation on their long-term performance for hemo dialysis.
Subclavian venous cannulation provides a ready means of vascular access for patients requiring imme diate dialytic therapy. However, it is apparent that the use of this route for hemoaccess is associated with signif icant long-term morbidity due to subclavian vascular injuries [1-6]. These lesions may be progressive and result in severe stenosis or thrombosis of the subclavian or innominate veins, or the vena cava, significantly reducing the successful outcome of a vascular access sub sequently placed in the ipsilateral upper extremity [712], Due to an increasing number of central vein injuries encountered at our center with the use of subclavian
dialysis catheters, an attempt was made to eliminate sub clavian venous cannulation by using a surgically placed prosthetic upper extremity graft within 48 h of place ment. This report describes the results of a randomized prospective study comparing the early use of 25 Med tronic plasma TFE conduits with 25 Gore-Tex polytetra fluoroethylene (PTFE) hemodialysis conduits which were cannulated within 48 h for dialytic therapy. Patients and Methods All accesses were created under local, monitored anesthesia using 1% xylocaine. The 6-mm Medtronic plasma TFE or the 6-mm Gore-Tex graft was used. Two Gore-Tex grafts were placed in the
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
Introduction
JafTers/Angstadt/Bowman
370
Table 1. Group characteristics of fistulas cannulated within 24 h of surgical placement Gore-Tex
9/16 51 10
10/15 58 11
Results
Table 2. Timing of initial cannulation of prosthetic fistulas after surgical placement Hours to cannulation 24 2 0
upper arm. The remainder were all placed in a looped configuration in the forearm. Plasma TFE grafts were not preclotted. The vessels were isolated through transverse extremity incisions using vascular tapes. The venous anastomosis was completed first using a 5-0 or 6-0 prolene suture. The conduit was tunnelled through the subcuta neous tissue using a 6-mm tunnelling device. The arterial anastomo sis was then completed using a similar suture. Heparin was not rou tinely utilized. The incisions were closed using 3-0 dexon in the subcutaneous tissue and 4-0 Vicryl or 4-0 PDS as a running subcu ticular stitch. All patients without a functional hemodialysis access requiring dialysis within 48 h were considered for entry into the study. Con sent was obtained from all patients prior to their participation in the study. Randomization was accomplished by sealing 50 3 X 5 cards in 50 envelopes, half of which had ‘Plasma TFE’ printed on them, the other half ‘Gore-Tex’. A sealed envelope was selected and opened in the operating room when it was certain that a vascular conduit was to be used. A single dose of vancomycin was given prcopcratively for prophylaxis against infection in all patients. The patients were followed prospectively in the clinic oi by phone inter views with the patients or their dialysis center nurses. Data were updated continuously in a computer database. Initial dialysis treatments were accomplished utilizing 17-gauge needles. All dialysis nurses and technicians participated in the ini tial cannulation of these fistulas. Activated clotting time (ACT) was utilized to determine the amount of heparinization required for dialysis. If the initial ACT was less than 110s, 1,000 units of heparin were given upon initiation of dialysis and then hourly until comple tion of the treatment. With an ACT of greater than 110 s, no initial heparin was utilized, and the kidney was flushed with normal saline hourly. If evidence of clot formation became apparent in patients with an ACT greater than 110s, 1,000 units of heparin were admin istered and continued hourly until completion of the dialysis treat
Twenty-five plasma TFE and 25 Gore-Tex vascular conduits were cannulated within 48 h of placement for early hemodialysis. All 50 patients were prospectively followed after placement of their Fistulas. One patient who had a forearm Gore-Tex fistula created, thrombosed her access within 12 h. It was deemed that the access was not salvageable and she was excluded from the study since her access was never cannulated. Her 3 X 5 card was reentered for another patient. Nine of the 25 patients who received the Medtronic graft were male and 16 were female. The average age of patients in this group was 51 years. Ten of these recipi ents were diabetic. Of the 25 patients with Gore-Tex grafts 10 were male and 15 were female. The average age in this group was 58 years, and 11 patients were diabetic (table 1). One diabetic patient with a left forearm Gore-Tex fis tula developed severe extremity ischemia. Bilateral up per extremity angiography revealed severe segmental atherosclerotic disease. The access was ligated and a Permacath was subsequently placed into the internal jugular vein. A second diabetic patient with a forearm Gore-Tex fistula developed ipsilateral extremity ischemia requir ing eventual ligation of her access and initiation of peri toneal dialysis. One plasma TFE graft was cannulated within 12 h of surgery. No hematoma, cellulitis or infection developed within the first 30 days of initial cannulation in this patient. None of the Gore-Tex grafts were cannulated within 12 h of surgical placement. Forty-seven patients had their first dialysis treatment from 12 to 24 h after surgical placement of which 22 were plasma and 25 Gore-Tex grafts. One patient with a plasma TFE conduit and 3 with Gore-Tex grafts developed cellulitis within the first month of placement (p = 0.60). All of the patients responded to antibiotic therapy with complete resolution of the cellulitis. Five hematomas occurred within the first 30 days of initial cannulation, 2 in the plasma TFE group and 3 in the Gore-Tex group (p = 1.0). Of the 2 hematomas which occurred in patients with plasma TFE grafts, 1 developed immediately after surgi cal placement before cannulation for dialysis and the other after cannulation for dialysis. There were no fur-
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
Male/female Average age, years Diabetics
Plasma TFE
ment. Upon decannulation of the access, gentle pressure was held over the puncture sites for 15 min. Observed proportions were compared with the two-tailed Fish er’s exact test.
Early Access Cannulation
371
Table 3. Morbidity with early cannulation (> 48 h) after access placement Hours to cannulation
p
Plasma TFE < 12 Hematomas Infections Thromboses-'
0 0 0
Total
12-24 1 1 0
>24 1 0 0
> 12 0 0 0
3
12-24
>24 0 0 0
3 3 3
1.0 0.60 0.22
9
0.10
< 1 month after initial cannulation.
ther hematomas and no resultant infections after these single bleeding events. Three patients with Gore-Tex fis tulas had hematomas develop along the tunnel. Two were systemically heparinized after successful thrombec tomy of their access within the first day of placement. One of these hematomas developed upon cannulation of the access with a 21-gauge needle for an intraoperative fistulagram during the thrombectomy procedure 12 h after initial placement of the conduit. The second patient similarly received intravenous heparin after thrombec tomy of the access immediately after surgery. Both fistu las were salvaged without sequelae. Neither developed an infection related to the hematoma, and both were cannulated for hemodialysis the following day. One other patient with a Gore-Tex graft developed a hema toma after his second dialysis treatment despite having no difficulties with the first attempt. This was felt to be related to a through and through puncture of the con duit. Two plasma grafts and none of the Gore-Tex grafts were cannulated between 24 and 48 h after placement. There was no morbidity after cannulation of these 2 fis tulas (table 2). Complications related to the early use of these fistulas are summarized below (table 3). There were no statistically significant differences between these groups (p = 0.45). Problems encountered with both types of grafts, unre lated to their early use for dialysis, included a forearm steal syndrome in 3 patients who received plasma TFE fistulas and 1 patient who received a Gore-Tex fistula. All 4 of these grafts required banding. The Gore-Tex fis tulas subsequently thrombosed after banding, and fur ther access attempts were abandoned due to severe pe ripheral vascular disease in this diabetic individual. Five patients developed significant arm edema after access
Table 4. Early morbidity unrelated to early cannulation: group characteristics
Forearm steal Edema Thrombosis before use Unrelated death < 1 month
Plasma TFE
Gore-Tex
3 2 3 3
1 3 3 0
placement, 2 in the plasma group and 3 in the Gore-Tex group, however, the edema did not preclude use of the access for immediate cannulation. Three patients with plasma TFE fistulas died within 1 month of their access surgery, of causes unrelated to their fistulas. Three pa tients with Gore-Tex grafts thrombosed their accesses before any attempts were made to cannulate these fistu las for hemodialysis. All were successfully thrombectomized and the fistulas were utilized within the subsequent 24 h. These unrelated complications after early cannula tion of the grafts are summarized below (table 4).
Discussion Since the introduction of the subclavian route for hemodialysis in 1969 [13], this method of central access to the vascular system for hemodialysis has received widespread use, particularly over the last decade, sup planting the various external shunts for immediate dialytic therapy. The apparent low morbidity initially en countered with the use of the subclavian catheter for dialysis appeared attractive when compared to the fre
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
;l
Gore-Tex
Jaffers/Angstadt/Bowman
372
to have no initial morbidity when cannulated within 48 h of placement. Since Gore-Tex has been the pre ferred conduit at our center, we elected to determine whether early cannulation of Gore-Tex was also feasible. There were no statistically significant differences in mor bidity between these two groups when the access was cannulated within 48 h of placement for hemodialysis. The success with early use of these grafts after their surgical placement is attributed to the use of a 6-mm tunneling device which ensures that the subcutaneous tissues are tightly apposed to the graft material. Addi tionally, the use of 17-gauge cannulas during the first dialysis treatment as well as maintaining pressure over the puncture site for at least 15 min after decannulation are also felt to be important in reducing the incidence of bleeding complications. Longer follow-up is obviously necessary to determine whether a difference in longevity will become important in attempts to maintain a patent vascular access in these patients with end-stage renal disease.
Acknowledgement The authors gratefully acknowledge the generous contributions of the dialysis staff.
References 1 Clark DD, Albina JE, Chazan JA: Subclavian vein stenosis and thrombosis: A potential serious complication in chronic hemodi alysis patients. Am J Kidney Dis 1990;15:265-268. 2 Selby JB, Tegtmeyer CJ, Amodeo C. Bittner L. Atuk NO: Inser tion of subclavian hemodialysis catheters in difficult cases: value of fluoroscopy and angiographic techniques. Am J Roentg 1989; 152/3:641-643. 3 Davis D, Petersen J, Feldman R, Cho C, Stevick CA: Subclavian venous stenosis: A complication of subclavian dialysis. JAMA 1984;252/24:3404-3406. 4 Barrett N, Spencer S, Mclvor J, Brown EA: Subclavian stenosis: A major complication of subclavian dialysis catheters. Nephrol Dial Transplant 1989;3:423-425. 5 Spinowitz BS. Galler MG, Golden RA. Rascoff JH, Schechter L, Held B, Charytan C: Subclavian vein stenosis as a complication of subclavian catheterization for hemodialysis. Arch Intern Med 1987;147:305-307. 6 Stalter KA, Stevens GF, Sterling WA: Late stenosis of the sub clavian vein after hemodialysis catheter injury. Surgery 1986; 100:924-927. 7 Cheung AK, Brownley R: Brachiocephalic vein occlusion in a hemodialysis patient. Am J Nephrol 1989;9:261-262. 8 Piotrowski JJ, Rutherford RB: Proximal vein thrombosis sec
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
quent episodes of thrombosis and infection encountered with the external shunts. Ease of placement of the sub clavian catheter as well as the need to sacrifice a periph eral vessel when placing the external shunts were further important factors in the transition to the use of these central veins. Unfortunately, a significant incidence of late compli cations related to subclavian venous stenosis or throm bosis has become apparent. These lesions may not be initially obvious unless venous collaterals or edema de velop in the ipsilateral upper arm or chest. Often, mas sive edema of the arm after access placement is the first evidence of a subclavian injury. It is clear that cannulation of the subclavian vein may result in severe stenosis or thrombosis of the vein in up to 19% of patients [10]. The loss of the central vein for outflow in this manner may be responsible for recurrent thrombosis of many fis tulas subsequently constructed in the upper extremity. The resultant arm edema may preclude extended use of the arms for hemoaccess [14]. Early use of subcutaneously placed dialysis grafts for hemodialysis has been described previously [15]. Var ious graft types including bovine, PTFE and Impra were utilized and cannulated within 72 h of placement with an overall incidence of postcannulation bleeding of 1.6% and infectious complications of 2.7% within 8 weeks. The success of early cannulation was attributed to the use of a tunneling device to ensure a compressive tunnel around the dialysis graft, thus reducing dead space for potential hematoma formation. A more recent study evaluated 48 PTFE grafts with an average time to cannu lation of 5.4 days with no graft infections [ 16]. No bleed ing complications were mentioned. These studies as well as our recent experience with immediate cannulation of the Medtronic plasma TFE and the Gore-Tex PTFE grafts demonstrate the ability to have a prosthetic access cannulated as soon as 12 h after placement for urgent hemodialysis. The Medtronic plasma TFE graft is made from woven polyethylene terephthalate which is lined with covalently bonded tetrafluoroethylene to provide surface properties similar to PTFE. This unique construction provides a conduit which can safely be cannulated within 72 h of placement for hemodialysis. This feature was particu larly attractive due to the recurrent difficulties encoun tered at our center with subclavian venous injuries from catheterization for dialysis. Although historically, the use of woven conduits (dacron) for hemodialysis fistulas has proven to be suboptimal, the plasma TFE grafts were evaluated in 10 patients prior to this study and appeared
373
Early Access Cannulalion
10
11
12
13
14
15 Taucher LA: Immediate, safe hemodialysis into arteriovenous fistulas created with new tunneler. Am J Surg 1985; 150:212— 215. 16 Haag BW, Paramesh V, Roberts T: Early use of polytetrafluoroethylene grafts for hemodialysis access: in Sommer BG, Henry ML (eds): Vascular Access for Hemodialysis II. Gore and Asso ciates and Precept Press, 1990, vol 2, pp 173-178.
Received: February 20, 1991 Accepted: July 25, 1991 Dr. Gregory Jaffers Department of Surgical Transplantation Wilford Hall USAF Medical Center Lackland Air Force Base TX 78236 (USA)
Downloaded by: UCL 144.82.238.225 - 5/1/2018 5:10:34 PM
9
ondary to hemodialysis catheterization complicated by arterio venous fistula. J Vase Surg 1987;5:876-878. Aujla N, McCauley J, Sorkin M: Superior vena cava syndrome due to subclavian hemodialysis catheters. Milit Med 1990; 155/6:274-277. Vanherweghem JL, Yassine T, Goldman M, Vandenbosch G, Delcour C, Struyven J, Kinnaert P: Subclavian vein thrombosis: A frequent complication of subclavian vein cannulation for hemodialysis. Clin Nephrol 1986;26/5:235—238. Fant GF, Dennis VW, Quarles LD: Late Vascular complications of the subclavian dialysis catheter. Am J Kidney Dis 1986;3: 225-228. Schwab SJ. Quarles LD, Middleton JP, Cohan RH, Saecd M, Dennis VW: Hemodialysis-associated subclavian vein stenosis. Kidney Int 1988;33:1 156-1159. Urben J, Kvasnicka J, Bastecky J, Groh J: Experience with rou tine use of subclavian vein cannulation in hemodialysis. Proc Eur Dial Transplant Assoc 1969;6:59. Coates GR. Clements JB, Yium JB, Watlington J: Painful edema of the arm after insertion of single-needle subclavian vein dialy sis catheters: Pathogenesis and treatment. South Med J 1988:81: 303-305.
