Access is the lifeline of dialysis, and a good one is built before it is needed. The fistula that will carry a patient's dialysis for years has to be planned months ahead, on veins that have been carefully protected, in a patient for whom it is the right choice. Get the planning right — early referral, vein preservation, the right access matched to the patient — and dialysis starts on a working fistula; get it wrong and it starts on a catheter, with all the harm that follows. This chapter is about the planning; the surgical and insertion technique belongs to the modality books.
The vascular access hierarchy
For haemodialysis, three access types form a clear outcome hierarchy. The arteriovenous fistula — a direct surgical anastomosis of artery to vein — sits at the top: it has the lowest rates of infection and thrombosis, the best long-term patency, and is associated with the best patient survival. Its drawbacks are time and reliability — it takes weeks to months to mature into something usable, and a substantial fraction fail to mature at all (primary failure), needing revision or a new attempt. The arteriovenous graft, a synthetic conduit bridging artery and vein, is the option when the patient's own veins are inadequate; it can be used sooner than a fistula but thromboses and needs intervention more often, though it still outperforms a catheter. The central venous catheter is the worst on every count — bacteraemia, thrombosis, central venous stenosis, and excess mortality — and is reserved for urgent or bridging use or when all other options are exhausted. The planning task is to move patients up this hierarchy and off catheters.
From 'fistula first' to a life-plan
For years the guiding slogan was 'fistula first,' and it drove a welcome shift away from catheters. But the imperative has matured into something more individualised: the right access for the right patient at the right time, the so-called life-plan approach. The recognition is that a fistula is not the best choice for every patient. A frail elderly patient with poor vessels and limited life expectancy may be better served by a graft or even a catheter than by a fistula that is unlikely to mature before it is needed or before the patient dies; a patient heading for an imminent pre-emptive transplant may need no permanent access at all; a patient with excellent veins and years ahead is the ideal fistula candidate. So the plan weighs vessel quality, likely maturation, life expectancy, the time until dialysis is needed, and transplant prospects — aiming not for a fistula in everyone, but for the access that best serves this patient over their whole kidney journey, while still avoiding catheters wherever possible.
Timing: months, not weeks
Timing is where access planning most often fails. A fistula needs to be created months before it will be used, because maturation takes weeks to months and because the common event of primary failure must be detected and salvaged with time to spare. Create it too late and the patient starts dialysis on a catheter while the fistula matures — the very crash-start scenario the predialysis pathway exists to prevent. But creating it too early has costs too: a fistula made years ahead in a slowly progressing patient may never be needed, or may develop complications, or may be wasted if the patient is transplanted pre-emptively. The balance is struck using the same risk tools as the pathway — the kidney-failure risk equation and the trajectory — to create access when dialysis is genuinely likely within the maturation-plus-salvage window. Access planning and the predialysis pathway are therefore one coordinated effort, not two.
Vein preservation: the silent imperative
The most underappreciated act in access planning is protecting the veins before any access is created. The veins of the forearm and non-dominant arm are the raw material for a future fistula, and they are routinely destroyed by ordinary hospital care — repeated venepuncture, intravenous cannulae, and above all peripherally inserted central catheters, which cause central venous stenosis that can render the entire arm useless for access. Subclavian central lines do the same. So vein preservation is an active discipline: the access arm is flagged, staff and the patient are educated that no needles, cannulae, or lines go into it, blood is drawn from elsewhere or the back of the hand, and PICCs and subclavian lines are avoided in any patient who may need dialysis access. A single thoughtless PICC can foreclose years of future access, and this is one of the highest-yield, lowest-cost interventions in all of CKD care. Vessel mapping with ultrasound then guides which vein and site to use.
Peritoneal access
For peritoneal dialysis, the access is a Tenckhoff catheter placed into the peritoneal cavity, and it too is planned ahead of need. Ideally it is placed with enough lead time to heal before use — around two weeks — though it can be used sooner when necessary. Planning requires assessment of the abdomen: prior abdominal surgery and adhesions, hernias and abdominal-wall defects, obesity, stomas, and the patient's ability to perform or be supported in the technique. Hernias and abdominal-wall defects are repaired before or at the time of catheter placement, because they otherwise cause leaks and failure. The exit site is planned for hygiene and comfort. As with vascular access, the peritoneal plan is made alongside the modality choice of the previous chapter — a patient who has chosen PD needs their catheter planned in the same coordinated, timely way a haemodialysis patient needs their fistula.
Minimising catheters, and the transplant exception
Running through the whole chapter is the imperative to minimise central venous catheters, because their infection and mortality burden is the largest avoidable harm in dialysis access. Every element — early referral, the access hierarchy, timely fistula creation, vein preservation, planned PD access — serves this end. The one exception that can render dialysis access unnecessary altogether is the pre-emptive transplant: a patient who will be transplanted before needing dialysis may need no permanent access at all, so transplant prospects, established in the predialysis pathway, feed directly into the access plan — there is no point creating a fistula for a patient about to receive a kidney. The art is to plan access neither too early nor too late, matched to the patient and the chosen modality, on preserved veins, with the catheter as the avoided last resort and the transplant as the ideal that obviates it.
Where the evidence is firm, and where it is judgement
The firm parts are the outcome ones: fistulas outperform grafts, which outperform catheters, and catheters carry a heavy excess of infection and death — this hierarchy is robust. So too is the harm of PICCs and subclavian lines to future access. The judgement parts are the individualisation: exactly when to create access, and for whom a fistula versus a graft versus a planned catheter is right, depend on vessel quality, prognosis, and transplant timing in ways no rule fully captures — which is precisely why 'fistula first' became a life-plan. The disciplined approach is to drive patients off catheters and up the hierarchy as the default, to preserve veins relentlessly, to time creation with the risk tools, and to individualise the specific access to the patient's whole kidney journey. Insertion and salvage technique — the surgery, the radiology, the cannulation — are the province of the modality books this chapter points to.
