Fax: (313) 876-2554

Email: abesarab@pol.net

 The preparation for and construction and maintenance of vascular access for hemodialysis is radically different in the States compared to other parts of the world, particularly Europe. The autologous wrist arteriovenous fistula (AVF) has been acknowledged to be the preferred type of first hemodialysis access since 1966 (1). If a wrist fistula is not possible, options at the elbow (brachiocephalic, transposed brachiobasilic, autogenous vein transfers) can be pursued (2-4). When the elbow or antecubital veins cannot be used to create an AVF, the end of the cephalic vein can be mobilized and tunneled for an anastomosis with the proximal, above-elbow brachial artery. Alternatively, a short 6 mm-jump graft (ePTFE) can be used connecting the above-elbow brachial artery with the cephalic vein when the brachial artery and cephalic vein are separated by a distance too large for direct anastomosis. Polo and his co-workers (6) reported one- and five-year primary patency of 85% and 48% respectively for this latter type hybrid fistula, results much better than the corresponding primary one- and five-year patency rates of 40% and 10% reported for long jump grafts constructed in the forearm or arm (6).

Why is it then that, despite all these options for AVF and the clear advantage of AVF over grafts in terms of longevity, preference is given in the United States to prosthetic bridge grafts (7-9)? This preference is so great that use of the autogenous arteriovenous fistulae (AVF) in the United States has actually been rapidly declining during the last decade (7-10). This decline has been attributed to the increasing numbers of older and diabetic patients (9,10) in the United States. Comparisons of access types among nations, however, reveals that Europe, Japan, Canada, and other countries are experiencing the same increases in age, diabetes, and comorbidity yet are maintaining their AVF prevalence rates in excess of 50% (11). Autologous fistula prevalence rates are reported to be 84% in Japan, 66% in Australia, and 55% in Sweden, rates all higher than the target set by the NKF-DOQI for incident patients. Since even under the best conditions early loss of native AVF reduces the prevalence rate by 3-30 percentage points (12), construction of AVF in these countries obviously is attempted in more than two-thirds of patients. This was confirmed at the most recent Symposium on AngioAccess for Haemodialysis in Tours, France (2nd International Multidisciplinary Symposium for Angioaccess for Hemodialysis, May 31-June 2). Every country (including so-called third world Central European countries) repeatedly reported fistula construction rates above 80% in incident patients and use of the AVF in 60-85% of prevalent patients. Recently a study from the Catalan Spain showed a AVF to graft ratio of over 10:1 among prevalent hemodialysis patients (13)

So why can't we do it in the United States? Some have said that the NKF- DOQI guideline target of constructing AVF in 50% of incident patients is unrealistic (14). The European experience in which the proportion of AV fistulas to grafts is 4:1 (11), the exact opposite of the situation in the United States (9), says otherwise. Many suspect that financial considerations may be the primary driving forces since until recently payment for grafts exceeded that for native fistulae. Furthermore, studies by Sands and Miranda (15) and by Allon and co-workers (16) show that an integrated co-operative effort by nephrologist, surgeon, radiologist, and nurse can increase the construction rate of primary AVF even if it is necessary to bypass the wrist and go directly to one of the many elbow level fistula options (see below).

The frequent need to use elbow level fistulas probably arises from two sources: a serious problem in the logistics of access construction (due to late referral) or the limitations of anatomy in the elderly. Also autologous vein sites tare often thoughtlessly injured by intravenous lines and by venipuncture in pre-ESRD patients likely to need dialysis in the future. Intrinsic problems in access construction in older patients due to diseased vessels are compounded by late referral for dialysis. Inevitably a "hurry up" graft access is then created while a "temporary" catheter is used (17). As stated previously, we have rationalized our inability to construct autologous AV fistulae on the late referral and the aging of our population stating that these factors differentiates us form the rest of the world. However, late referral to dialysis is not a unique American experience nor is an aging population. In Catalan, 48% of over 2600 patients presented for hemodialysis without an access being available (13) and over half of the patients started on hemodialysis since 1993 were over 65 years of age. Again, virtually every European nephrologist I spoke to at the AngioAccess meeting in Tours admitted that late referral is a "global" phenomenon yet the focus remains creation of AVF. Catheters are used to bridge the gap of native fistula maturation but their use decreases with time on dialysis. In the Catalan registry, the catheter use rate fell to an impressive < 4% of prevalent patients after the 4^th year of dialysis. Impressive indeed.

These international comparisons aside, It is clear that the factors of age and late referral cannot explain the large regional variation (from less than 10% to more than 50%) in the percentage of autologous AVF used in prevalent United States hemodialysis patients (9). Why is the rate of AVF in prevalent patients less than 10% in the South West yet is over 50% in New England region? Are the surgeons better in New England? The ongoing DOPPS (Dialysis Outcomes and Practice Patterns Study) clearly indicates that everyone must accept responsibility for this sad state of affairs. We can not simply point the finger at the American surgeon for the accesses our patients receive. Up to 25% of nephrologists and 50% of dialysis staff prefer and therefore accept graft accesses.

The reasons for preferential construction of grafts are simple. Grafts require less nurturing to develop, require less care (until they thrombose, that is), and unfortunately many of our staff do not have the skills needed to cannulate native AVF, particularly when the access is still "young". We are in a rush. "Get the patient on, get the patient off" seems to be the mantra of the 90's. When properly cannulated and cared for, the AVF access can serve the patient for prolonged trouble-free periods measured in decades. The need to improve construction of and maturation of AVF has never been more important. Alternatives to thrice weekly hemodialysis exist. These include nocturnal and short daily hemodialysis. Recent studies in Canada, USA, and Italy clearly indicate that the AVF is the ideal access for these therapies that require more frequent cannulation. The critical determinants for creating AVFs are early patient referral, vascular anatomy, and surgical skill. The solution lies in the integrated cooperation of nephrologist, radiologist, and surgeon. We should make the time to create native accesses.

With these considerations in mind what can we learn from the Multidisciplinary Symposium for Angioaccess for Hemodialysis in Tours, France (18). First, the success of AVFs depends upon adequate arterial inflow and on the anatomy of venous outflow; these need to be defined pre-operatively. SecondIy, the skill of the surgeon does matter. Finally the access must be carefully observed during the initial maturation period and then long-term. Most important of all, no turf battles between surgeons, nephrologists, and radiologists.

The prerequisites for a well functioning AV fistula demand adequate arterial inflow, no resistance to outflow, and a sufficient length of adequate vein for punctures. Following my presentation on "Native fistula vs. grafts in the DOQI: a nephrologist's perspective," the first major theme began. Several presentations clearly showed that pre-operative assessment beyond the physical examination can increase the ability to construct and obtain mature usable AV fistulas, particularly in the elderly and obese patients. If the patient's pre-operative pulse is poor, noninvasive Doppler, digital plethysmography and/or arteriography pre-operatively may identify the nature of the inflow problem and angioplasty or surgical correction may allow subsequent AVF construction. Such studies confirm the observations from almost 2 decades ago. An adequate AVF can develop in diabetics, elderly patients, even those with calcified arteries and with vasospastic vessels if intra-operatively there is good arterial flow (12).

JHM Tourdier (Maastricht, Holland) summarized the experience of his group in using pre-operative duplex as part of the assessment program. Using a feeding arterial diameter of 2 mm as a minimal criterion for construction, early failure rates decreased from 32% to 8%. This is a very important observation. G Franco (Paris, France) emphasized that visual inspection and physical examination have major limitations in the elderly, the obese, and in those with previous failed accesses. Vascular mapping with appropriate selection of vessels, both arterial (> 2 mm luminal diameter) and venous (> 2.5 mm luminal diameter) reduces unacceptable failure rates in these populations. R Alain (Paris) reviewed the indications for venography (obesity, limb edema, past history of central line placements) and a failure to create a first access. He summarized his experience with CO₂ phlebography to avoid contrast nephropathy in patients not yet on dialysis. Visualization was successful in 144 of 153 forearm veins, 90 of 91 cephalic and 50/51 basilic veins, and all 51 proximal central veins. We do not have to be afraid of precipitating the need for hemodialysis of venography is needed to determine the best access for the patient. M Maiovrh (Slovenia) found that changes in blood flow, resistance, and pulsatility following reactive hyperemia provided additional information that were predictive of successful construction and maturation of AVF, particularly in high risk elderly and/or diabetic patients. All of this exciting work indicates that with proper evaluation we should be able to do as well in the USA. The issue in the USA is who will pay. Medicare does not routinely pay for pre-operative evaluations. We must change this policy.

The theme of arteriovenous creation included an exquisite video by P Bourlequot (France) on the use of microsurgery (in children) in creating native fistulae. In fact several presentations focused on the creation of AV fistulae at the extremes of life. During the discussion that followed I was truly amazed by the number of nephrologists in Europe (Germany, Italy, Poland, Greece, Macedonia, other central Europe countries) who assumed the responsibility of creating native AV fistulas in their patients. Not only did they create them but they became proficient in detecting stenosis in them (K. Konner: Köln, Germany). I was asked directly why DOQI did not even mention the construction of wrist ulnar-basilic fistulas. I had no good answer. The ulnar artery is larger and carries more flow than the radial artery, factors predicting successful maturation. The back of the forearm has to be used which is a little inconvenient for patient and staff. But in truth I can recall 2 patients whom I cared for and who had such a fistula constructed because the wrist radiocephalic fistulae had failed to mature. Both fistulas provided over a decade of useful access. P Bottet (France) concluded that in the elderly (age > 65, mean age of 74) protection of venous capital did not appear to be a sufficient argument to justify the use of more distal (wrist) fistulae because of a higher failure and re-operation rate. Because proximal elbow-crease fistulas in these individuals could achieve a patency rate of more than 50% at 4 years, they advocated direct construction of such elbow level accesses (this seems to be the direction we are following in the States). The session concluded with a number of presentations on elbow crease fistulae and on the resurrection of previously failed forearm cephalic vein fistulas in patients returning to dialysis after failed allografts (T Grochowiecki, Poland). Such arterialized veins could be used almost immediately. Although there were several papers on prosthetic devices (bioprosthetic, polyurethane, early cannulation, hooded grafts, influence of graft geometry) it appears that it will be some time before advances in this area will permit patency equal to the AV fistula. Clearly, in Europe the focus is on not using grafts and catheters if at all possible.

Once a fistula was created, a major issue discussed at the meeting was what to do with the native fistula with delayed maturation. Clearly we need better methods of maturing AVF (19). Beathard et al (20) have proposed a valuable algorithm for dealing with venous stenosis, accessory veins and inadequate vein diameter but we need nephrologists and staff who can promptly recognize the patients whose AVF is not developing within the first 40 days and not at a mean of 151 days. Prompt referral for intervention (nephrologist, surgeon, or radiologist) is mandatory. Certainly a venogram with angioplasty and/or accessory vein ligation is indicated at 4-8 weeks if the AVF remains immature but F Escobar and I believe that many potential problems can also be solved during the creation of the AVF (19). Pouchez and Turmel-Rodrigues (Tours, France) provided some maneuvers during physical examination to detect areas of stenosis or thickness and discussed the role of duplex-scanning in assessing the cause of delayed maturation. However, angiography of the brachial artery is the definitive diagnostic procedure. If the artery is diseased, a more healthy proximal anastomosis may be defined. If the vein is stenotic particularly in a wrist radiocephalic fistula, conversion to an ulnarbasilic fistula may be the most reasonable approach if the stenosis cannot be resolved by angioplasty without damaging healthy veins at the elbow. They emphasized that ligation of veins draining into the basilic system in the presence of a cephalic stenosis will increase the risk of access thrombosis since it does not increase flow.

In addition to the concerns about low flow and delayed maturation, I found my European colleagues quite concerned about high flow fistulas and the need to band such fistulas. Clear indications for access flow reduction do exist if there is steal. Dahan and Bourquelot (France) showed that the decrease in cardiac output 2 months after banding was proportional to the decrease in access flow achieved by banding but the magnitude of the former was twice as great as that of the latter. The presentation by Huu and co-workers (France) clearly indicated that high access flow (> 1.5 liters min) is not the sole causal factor for high-output heart failure in HD patients. I just did not get a sense of the indications for a flow-reduction procedure in patients with heart disease and autologous AV fistulae. Clearly, the balance is to minimize the hemodynamic effects of a peripheral AVF on the heart (increased stroke volume, increased end diastolic diameter, LVH) without decreasing the flow to the point where the access becomes at risk for thrombosis or is unable to deliver adequate flow for hemodialysis. Since we now have methods to measure both cardiac output and access flow on line, we need to do some prospective outcome studies in which baseline cardiac function and other factors are used to define those at risk from "high flow".

Other important topics were covered such as stenosis and thrombosis. In these areas there were less practice differences between the States and Europe with one important distinction. Because of the preponderance of AVF in Europe, the physical exam is more important and measuring flow is more useful than measuring pressures in detecting stenosis. Luc Turmel-Rodriguez (Tours, France) reported that AVF have an advantage over grafts in terms of patency after dilatation or stent placement. With stent placement, the radiologist must think of the future; stents must not obviate further access construction in the same extremity. When an AVF thromboses, there is real urgency in getting the thrombus out. "Within 48 hours" simply will not do. Two exciting paper on locking antibiotics into central vein catheters ( H Leray-Moaguez, France) or hemodialysis ports (K Sodemann, Germany) to reduce Catheter related bacteremia were also presented.

Most clear to anyone attending the meeting was the spirit of cooperation. People seemed to talk to each other. As a result the patient got the best access and the best follow-up for that access. We can learn a lot from our European colleagues.

References

1. Brescio MJ, Cimino JE, Appel K, et al. Chronic hemodialysis using venipuncture and a surgically created arteriovenous fistula. N Engl J Med 1966: 275;1089- 1092
2. Bender MH, Bruyinckx CM, Gerlag PG. The brachiocephalic elbow fistula: A useful alternative angioaccess for permanent hemodialysis. J Vasc Surg 1994; 20:808-813.
3. Gade J, Aabech J, Hansen RI. The upper arm arteriovenous fistula&endash;an alternative for vascular access in haemodialysis. Scand J Urol Nephrol 1995; 29:121-124.
4. Lindner J. Transposition of the basilic vein in the arm for vascular access in hemodialysis. Rozhl Chir 1997; 76:126-128.
5. Polo JR, Vázquez R, Polo J, Sanabia J, Rueda J, Lopez-Baena JA. Brachiocephalic jump graft fistula. An alternative for dialysis use of elbow crease veins. Am K Kidney Dis 1999; 33:904-909
6. Mehta S. Statistical summary of clinical results of vascular access procedures for hemodialysis. In: Vascular access for Hemodialysis. Sommer BG, Henry ML (eds.), Chicago ILL, WL Gore & Associates Inc and Precept Press, Inc. 1991, pp. 145-157.
7. USRD 1995 Annual Data Report. X. The cost effectiveness of alternative types of vascular access and the economic cost of ESRD. Am J Kidney Dis 1995; 26(suppl 2): S140-S156
8. USRD 1997 Annual Data Report. IV. USRDS dialysis morbidity and mortality study: WAVE 2. Am J Kidney Dis 1997; 30(suppl 1): S67-S85.
9. Hirth RA, Turenne MN, Woods JD, Young EW, Port FK, Pauly MV, Held PJ. Predictors of type of vascular access in hemodialysis patients. JAMA 1996; 276:1303-8.
10. Kaufmann JL. The decline of the autogenous hemodialysis access site. Seminars Dial 1995; 8:59-61.
11. Bleyer AJ, Disney AP, Hylander B, Nomoto Y, Coates PTH, delaTorre E, Bjorklund A, Burkart JM. The multinational study of dialysis therapy: hemodialysis access (abstract). Am J Kidney Dis 1995; 26:483
12. Reilly DT, Wood RFM, Bell PR. Prospective study of dialysis fistulas: problem patients and their treatment. Br J Surg 1982; 69:549-555.
13. Rodriguez JA, Lopez J, Cleries M, Vela E, and the Renal Registry committee. Vascular access for haemodialysis - an epidemiological study of the Catalan renal registry. Nephrol Dialysis Transplant 1999; 14:1651-1657
14. Schwab S, Besarab A, Beathard G, et al. NKF-DOQI: Clinical practice guidelines for vascular access. Am J Kid Dis 1997; 30(Suppl 3): S150-S189.
15. Sands J, Miranda CL. Increasing numbers of AV Fistulas for hemodialysis access. Clin Nephrol 1997; 48:114- 117
16. Allon M, Bailey R, Ballard R, Deierhoi MH, Hamrick K, Oser R, Rhynes VK, Robbin ML, Saddekni S, Zeigler ST. A multidisciplinary approach to hemodialysis access: Prospective evaluation. Kidney Int 1998; 53:473-479.
17. Ifudu O, Dawood M, Homel P, Friedman EA. Excess morbidity in patients starting uremia therapy without prior care by a nephrologist. Am J Kidney Dis 1996; 28:841-845.
18. Communications of the 2^nd International Multidisciplinary Symposium on AngioAccess for Haemodialysis, Tours France, May 31-June 2, 1999.
19. Besarab A, Escobar FS III. A Glimmer of hope: Increasing the construction and maturation of autologous arteriovenous fistulas. [Editorial] Am J Kid Dis 1999; 33:977-979.

Beathard GA, Settle SM, Shields MW. Salvage of the nonfunctioning arteriovenous fistula. Am K Kidney Dis 1999;33:910-914.