Kidney International, Vol. 52 (1997), pp. 1704—i 705

Hemodialysis vascular access: An ounce of prevention

EDITORIAL

Chronic maintenance hemodialysis requires repeated reliableaccess to the systemic circulation. The United States Renal DataService reports that hemodialysis access failure continues to bethe most frequent cause of hospitalization in ESRD patients inthe United States [1, 2]. In addition, in some centers it accountsfor the largest number of hospital days [3]. Thus, failed hemodi-alysis access is not only a major cause of patient morbidity, butalso a significant financial liability, estimated to cost the healthcare system in the United States almost 1 billion dollars per year[4]. It is not surprising that the National Kidney Foundationidentified hemodialysis access as one of the four topics to beaddressed by the dialysis outcomes quality initiative (DOQI) [1].

Ideally, access to the systemic circulation should accommodatea flow rate adequate for the hemodialysis prescription, remainusable for a long period of time, and have a low complication rate.No access meets all these criteria. Native arterio-venous (A-V)fistulas come closest, but even they are associated with somecomplications. Furthermore, in the United States, as the numberof patients undergoing dialysis for ESRD has grown, the use ofnative A-V fistulas has declined and the use of synthetic A-Vgrafts for permanent vascular access has increased [2]. Unfortu-nately, patency rates for A-V grafts are inferior to those of nativeA-V fistulas, and complication rates are at least threefold greater[1, 2, 4—6].

The primary complication, accounting for 90% of access dys-function of both A-V fistulas and grafts, is vascular accessthrombosis. In more than 90% of instances, vascular accessthrombosis is mediated by a venous outflow stenosis [1, 5—9].These venous stenoses, caused by endothelial and vascularsmooth muscle hyperplasia of the vein wall, occur in areas markedby turbulence in the venous system. Although they may develop atsites of vein bifurcation and vein injury associated with centralvein cannulation, they are most common at the vein graft anas-tomosis in A-V grafts [1, 8—10]. Innovative strategies to minimizeaccess thrombosis to date have borrowed a page from thecardiologists: with intensive monitoring using access flows andpressures, venous outflow stenosis can be prospectively identifiedand anatomically treated [1, 7—11]. While this significantly im-proves access patency, it is costly and may increase patientmorbidity. In addition, it does not address the issue of why thisphenomenon occurs and how it can be prevented.

The prevailing consensus is that venous stenosis is caused byendothelial injury and vascular smooth muscle hyperplasia inareas where flow turbulence is created. While many instances ofoutflow stenoses do occur at such locations, very little stenosisoccurs at the equally turbulent artery graft anastomosis or at thesite of the artery vein anastomosis in A-V ñstulas. Thus, additionalfactors must be present for hyperplasia to occur. Venipuncture inthe hemodialysis access forms a platelet plug with the exposure ofthe downstream vessel to a host of platelet-associated cytokinessuch as platelet-derived growth factor (PDGF). In this issue ofKidney International, Himmelfarh and Couper take the first steps

in elucidating the causes of this phenomenon [12]. Previously:,Sreedhara and associates reported that high-dose dipyridamoledecreased access thrombosis in new, but not previously estab-lished, A-V grafts [13]. The observations of Himmelfarb andCouper suggest that the mechanism for this effect may not berelated to the effect of dipyridamole on coagulation, but rather toa contributory effect of dipyridamole on the inhibition of plateletderived cytokines (PDGF and others) that act on the vessel wall.The theory that vascular smooth muscle hyperplasia reflects "aresponse to injury" in which platelet-derived growth factor isprominent suggests that this pathway to vein stenosis may bepreventable. It has been disappointing that while vast amounts ofresearch and development assets, by both industry and the NIH,have been directed at coronary stenosis post-percutaneous trans-luminal coronary angioplasty, almost no basic cellular or molec-ular research has been funded for the problem of hemodialysisvascular access stenosis.

As previously suggested by Sukhatme, it is time to apply thetools of cellular and molecular biology to the clinical problem ofvascular access stenosis [14]. The role of vitamin E and otheragents in addition to dipyridamole must be explored, and clinicaltrials using radiotherapy to prevent smooth muscle vascularhyperplasia are needed.

As outlined in the DOQI Practice Guidelines, a return toprimary A-V fistulas will reduce access problems, but even fistulasfail because of endothelial injury and vascular smooth musclehyperplasia. Active efforts to understand and prevent the basiccauses of venous stenoses are needed. As remarked long ago, anounce of prevention is worth many pounds of cure.

REFERENCES

1704

STEVE J. SCHWABDuke University

Durham, North Carolina, USA

1. NKF DOQI Clinical Practice Guidelines for hemodialysis vascularaccess. Work group: SCHWAB SJ, BESARAB A, BEATHARD G, BROUWERD, ETHERE[X:I: E, HARTIGAN M, L1vlNe M, MCCANN R, SHERMAN R,TRER0l'otA S. Am J Kidney Dis 30 (Suppl 3), 1997

2. U.S. RENAL DATA SYSTEM: X. The cost effectiveness of alternativetypes of vascular access and the economic cost of ESRD. Am JKidneyDis 26 (Suppl 2):S140—Sl56, 1995

3. MAYERS JD, MARKELI. MS, COHEN LS, HoNe; J, LUNDIN P, FRIEDMANEA: Vascular access surgery for maintenance hcmodialysis: Variablesin hospital stay. ASAJO J 38:113—115, 1992

4. FBI A)MAN HI, K0BRIN 5, WASSERMAN A: Hemodialysis vascular accessmorbidity. JAm Soc Nephrol 7:523—535, 1996

5. FAN PY, Sci IWAB Si: Vascular access: Concepts for the 1990s. J AmSoc Nephrol 3:1—Il, 1992

6. ALBERS F: Causes of hemodialysis access failure. Adv Ren ReplaceTher 1:107—118, 1994

7. BESARAB A, SULLIVAN KL, Ross RP, MoRrrz MJ: Utility of intra-access pressure monitoring in detecting and correcting venous outletstenoseS prior to thrombosis. Kidney mt 47:1364—1373, 1995

8. WINDUS DW: Permanent vascular access: A mephrologist'sview.AmJKidney Di,s 21:457—471, 1993

Schwab: Editorial 1705

9. Vui K, B00K5TEIN JJ, ROBERTS AC, DAvis GB: Pharmacomechani-cal thrombolysis and angioplasty in the management of clottedhemodialysis grafts: Early and late clinical results. Radiology 178:243—247, 1991

10. SCHWAB SJ, RAYMOND JR, SAEED M, NUWMAN GE, DeNNIS PA,B0L,LINGER RR: Prevention of hemodialysis fistula thrombosis. Earlydetection of venous stenoses. Kidney mt 36:707—71], 1989

Ii. TURMEL-RODRIGUES L, PENGLOAN J, BLANCHIER D, ABAZA M, BIR-MULE B, HAILLOT 0, BLANCHARD D: Insufficient dialysis shunts:Improved long-term patency rates with close hemodynamic monitor-

ing, repeated percutaneous balloon angioplasty, and stent placement.Radiology 187:273—278, 1993

12. HIMMELFARB J, COUPER L: Dipyridamole inhibits PDGF and bFGF-induced vascular smooth muscle proliferations. Kidney mt 52:1671—1677, 1997

13. SREEDHARA R, HIMMELFARB J, LAZARUS JM, HAKM RM: Antiplate-let therapy in graft thrombosis: Results of a prospective randomizeddouble blind study. Kidney tnt 45:1477—1438, 1994

14. SUKHATME V: Vascular access stenosis: Prospects for prevention andtherapy. Kidney mt 49:1161—1174, 1996