Access to circulation by permanent arteriovenous fistula in regular dialysis treatment.

Serum immunoglobulin levels and meningococcal antibodies were measured in nine patients suffering from meningococcal infection. The bactericidal activity of normal serum to meningococci could be attributed largely to yM globulin. Severe yM deficiency, probably inherited, was found in two boys dying from fulminating meningococcal septicaemia. Other evidence supported the view that yM antibodies are an important humoral defence mechanism and that the normal presence of these antibodies in the blood stream restrains systemic meningococcal infection to the C.S.F. We thank Professor J. P. M. Tizard and Dr. J. A. Davis for allowing us to study patients under their care ; Dr. S. Woriledge, who estimated the blood groups and isoantibody titres ; and Professor L. P. Garrod for his advice. Requests for reprints should be addressed to Dr. J. R. Hobbs, Department of Chemical Pathology, the Royal Postgraduate Medical School, Ducane Road, London W.12. REFERENCES Adinolfi, M., Glynn, A. A., Lindsay, M., and Milne, C. M. (1966). Immunology, 10, 517 Brown, J. W., and Condit, P. K. (1965). Calif. Med., 102, 171. Burtin, P., Buffe, D., and Grabar, P. (1964). Ann. Inst. Pasteur, 106, 519. Cohen, I. R., and Norins, L. C. (1966). Science, 152, 1257. Evans, R. T., Spaeth, S., and Mcrgenhagen, S. E. (1966). 7. Immunol., 97, 112. Ewald, R. A., Williams, J. H., and Bowden, D. H. (1961). Vox Sang. (Basel), 6, 312. Goldman, J. M., and Hobbs, J. R. (1967). Immunology, 13, 421. Heist, G. D., Solis-Cohen, S., and Solis-Cohen, M. (1922). 7. Immunol., 7, 1. Hobbs, J. R., Russell, A., and Worlledge, S. M. (1967a). Clin. exp. Immunol., 2, 589. Sommerville, R. G., and McSwiggan, D. A. (1967b). Lancet, 1, 1108. Martern, P., Masseyeff, R., Michel, R., and Pereti, P. (1961). Ann. Inst. Pasteur, 101, 382. Michael, J. G., and Rosen, F. S. (1963). 7. exp. Med., 118, 619. Muschel, L. H., and Jackson, J. E. (1966). 7. Immunol., 97, 46. Ross, C. A. C. (1961). Lancet, 2, 527. - and Stevenson, J. (1962). 7. Hyg. (Lond.), 60, 501. Sample, D W., Gauld, J. R., Warram, J. H Winter, P. E., Runcik, K., and Gauld, R L. (1966). Amer. 7. Epidem., 84, 156. Schultze, H. E., and Heremans, J. F. (1966). The Molecular Biology of Human Proteins, 1, 737. London. Skiorten. F. (1966). Acta path microbiol. scand., 68, 517. South, M. A., Cooper, M. D., Wolheim, F. A., Hong, R., and Good, R. A. (1966). 7. exp. Med., 123, 615. ### Access to Circulation by Permanent Arteriovenous Fistula in Regular ### Dialysis Treatment* J. S. HANSON, M.B., B.SC.; M. CARMODY, M.B.; B. KEOGH, M.B.; W. F. O'DWYER, M.D. Brit. med. J3., 1967, 4, 586-589 Regular dialysis therapy has now become the accepted treat ment of selected cases with end-stage renal disease. Heretofore, repeated access to blood vessels for dialysis was achieved by means of the conventional Teflon or Teflon-Silastic prosthesis, establishing an external semipermanent arterio venous shunt (Quinton et al., 1960, 1962 ; Hegstrom, 1961). In spite of its considerable success this technique is far from perfect, and problems of infections and clotting remain as a major cause of trouble in the management of these patients. The average survival rate for external shunts varies considerably from one centre to another. In a series published from the Swedish Hospital in Seattle the mean shunt life for both arteries and veins varied from 11 days to 11.4 months (Murray et al., 1964). In a more' recent review from the same centre (Pendras and Smith, 1966) an average cannula survival of 12 to 14 months was reported. This they attributed to the prevention of infection in the patient before and shortly after operation, and also by avoiding cannulation in the severely uraemic state. We have been unable to achieve a similar standard of shunt survival since we began regular dialysis therapy in November 1964. Our survival rates in arteries and veins were 3.8 and 3.6 months respectively, the majority of the failures occurring i Department of Surgery and Renal Unit, Jervis Street Hospital, Dublin. in the first six weeks after surgery. The longest surviving shunt was 16 months. Infection and clotting, the major causes of our failures, were responsible for a high morbidity and often long periods of immobilization in hospital for surgical revision of the cannulas. This was particularly true for those in the lower limbs. The frequent manipulations involved in declotting were also a contributory factor to the psychological fixations which some of these patients developed about their cannulas. Our relative failure with the Quinton type shunt prompted us to adopt the surgically created arteriovenous fistula as described by Brescia et al. (1966). As a result of the arteriovenous fistula so created the arm veins are distended, allowing dialysis to be carried out by venepuncture. The veins are then in effect an extension of the arterial system. Surgical Technique Anaesthetic.-Brachial plexus block by the axillary route has been used routinely and has proved very satisfactory. Firstly, the accompanying sympathetic block dilates the blood vessels and prevents vascular spasm until the vein hag been denervated by the stripping of its fascia. The sympathetic blockage may lead to free bleeding from the line of incision, but #### 9 December 1967 Dialysis Therapy-Hanson et al. this can be controlled by circumvential infiltration around the wound with a local anaesthetic containing adrenaline. Secondly, the muscular paralysis enables the patient to lie in comfort with an outstretched arm for prolonged periods. Thirdly, the sensory effects of the brachial plexus block extends well into the postoperative period, and the analgesic effect is much appreciated by the patients. Fourthly, when performing this procedure under brachial block it is not necessary to admit the patient to hospital. Also it can be carried out at a time convenient to the patient and therefore does not interfere with his normal working hours. Choice of Vessels and Site of Fistula Formation In the early stages of this project we received patients who had had repeated shunt insertion. Frequently only ulnar arteries remained patent, the radial arteries having been cannulated. This does not make fistula formation impossible. Neither does it appear to impair seriously the blood supply to the hand if the sole remaining main vessel is used for fistula formation. In Tables I and 1I are given particulars of a series of 14 patients on regular dialysis treatment in whom an arteriovenous fistula was created surgically. TABLE I. Permanent Arteriovenous Fistula (A.V.F.) in Regular Dialysis Therapy Sex A.V.F. Formation I 28/12/66 2/ 1/67 16/ 1/67 17/ 1/67 18/ 1/67 21/ 1/67 21/ 1/67 16/ 2/67 23/ 1/67 24/ 1/67 31/ 1/67 20/ 3/67 1 9jq/ I10/ 15/ 14/ } } Artery Used Radial 3.. ,. Ulnar Radial Ulnar ,,1 ,.I 4/67 Radial ,, 4/67 ,, ,, 4/67 , , 6/67 3 Complications None Left A.V.F. non-function Right A.V.F. successful None 3, Right A.V.F. infection and haemor rhage Left A.V.F. successful None Transplanted 5/4/67 A.V. F. non-function (Quinton shunt replaced) None TABLE II No. of patients .. No. of arteriovenous fistulae .. No. of dialysis months .. No. of dialyses .. Failed arteriovenous fistula . * Two successful repeats, opposite limb. New patients introduced to the regular dialysis therapy pro gramme cause no problem in respect of suitability of artery and vein, as they are initially fitted with a temporary Carmody (1964) type shunt, leaving a virgin arm for fistula formation. When available the radial artery and cephalic vein are used routinely, but, as mentioned, the ulnar artery and any super ficial vein in close proximity may be used. Venae comitantes do not give the flow required in the superficial veins. When using an ulnar artery it is preferable to avoid use of the small superficial veins in the area of the skin creases on the ventral carpal surface. In this situation the skin is tethered to the deep fascia, making dissection difficult and increasing the possibility of vessel spasm due to excessive handling. Again, the veins in this area form a branching network ; therefore without ligating some branches it is unlikely that a segment of vessel adequate for fistula formation will be obtained. Finally, in association with brachial block there appears to be more superficial bleeding in the area. In choosing a suitable vein a venous tourniquet is applied and the arm inspected. The availability of a suitable vein really dictates the site of fistula formation, but it is desirable to create it as far distally as possible in the arm