Access Dysfunction & Thrombosis

The contract between this chapter and the reader.

1. 1. Recognise stenosis as the dominant cause of access failure.
2. 2. Detect access dysfunction by physical examination.
3. 3. Use dialysis-based and flow/pressure clues to flag stenosis.
4. 4. Interpret duplex and angiography to localise a stenosis.
5. 5. Define a “significant” stenosis — anatomy plus function.
6. 6. Treat stenosis with angioplasty, and know when to stent or revise.
7. 7. Salvage a thrombosed access urgently and treat the underlying lesion.
8. 8. Manage central venous stenosis and prevent it.
9. 9. Weigh the evidence on routine surveillance versus clinical monitoring.

A sixty-second reading. Each bullet stands alone.

• Stenosis, usually from neointimal hyperplasia, is the dominant cause of access dysfunction and failure.
• The aim is to find and treat stenosis before it thromboses, and to salvage a thrombosed access promptly.
• Physical examination is the cornerstone of monitoring: thrill, bruit, arm-elevation, and augmentation.
• A fistula that becomes pulsatile and stays distended on elevation has outflow stenosis; weak augmentation indicates inflow stenosis.
• Dialysis clues include rising venous pressures, recirculation, falling clearance, difficult cannulation, and prolonged bleeding.
• Access flow and static venous-pressure trends help, but the trend matters more than any single value.
• Angiography — a fistulogram — is the gold standard and allows treatment in the same sitting.
• Treat a stenosis only when it is significant: over about 50% narrowing plus a clinical or functional abnormality, not anatomy alone.
• Angioplasty is first-line; a stent is reserved for recoil, recurrence, or central and cephalic-arch lesions.
• A thrombosed access needs urgent salvage, and the underlying stenosis must be treated in the same procedure.
• Central venous stenosis causes arm, face, or breast swelling and is largely prevented by avoiding catheters and subclavian lines.
• Whether routine flow and pressure surveillance beyond clinical monitoring improves access survival is genuinely uncertain.

The medical core. An expert should agree access dysfunction is fully covered here.

Why it matters at the bedside

Access fails in a predictable way: a stenosis narrows, flow falls, and one day the access clots. Almost everything in this chapter is an attempt to interrupt that sequence — to catch the stenosis while the access still works, and, when it does clot, to open it fast and fix the lesion behind it.

Stenosis as the engine of failure

• The common pathway to access loss is stenosis from neointimal hyperplasia — at the juxta-anastomotic segment of a fistula, the venous anastomosis of a graft, the cephalic arch, or the central veins. As it tightens, flow falls, dialysis suffers, and thrombosis eventually follows. Behind almost every thrombosed access is a stenosis.

Physical examination — the cornerstone

• The most useful, immediate surveillance is the hand and the ear. A normal fistula has a soft continuous thrill that augments on occlusion and collapses on arm elevation. Outflow stenosis makes it hyperpulsatile and keeps it distended on elevation; inflow stenosis weakens the augmentation. A bruit that becomes high-pitched or discontinuous localises the lesion.

Dialysis-based and surveillance clues

• The dialysis unit reports the rest: rising venous pressures, recirculation, a falling delivered clearance, difficult cannulation, prolonged bleeding after needle removal, and aspirated clots. Access flow (Qa) and static venous-pressure measurements add objective trends — a falling Qa or a rising pressure ratio — but it is the trend, not a single reading, that signals stenosis.

Imaging and localisation

• Duplex ultrasound localises and grades a stenosis non-invasively. Angiography — the fistulogram — remains the gold standard and has the advantage that the same procedure can treat the lesion found.

What counts as significant stenosis

• Treat the patient, not the picture. A stenosis is significant — and worth treating — only when it combines an anatomic narrowing greater than about 50% with a clinical or functional abnormality (low flow, high venous pressure, recirculation, prolonged bleeding, or an abnormal exam). An incidental angiographic narrowing with a well-functioning access is left alone.

Treating stenosis: angioplasty, stent, revision

• Percutaneous transluminal angioplasty is first-line for a significant stenosis. A stent or stent-graft is reserved for elastic recoil, rapid recurrence, or the difficult cephalic-arch and central lesions; surgical revision suits some juxta-anastomotic lesions and aneurysms. Restenosis is common, so many accesses need repeated intervention.

The thrombosed access

• A thrombosed access — no thrill, no bruit, a hard cord — is an emergency for the access. Salvage is by percutaneous pharmacomechanical thrombectomy or surgical thrombectomy, and it is time-sensitive (grafts tolerate a longer window than fistulae). Crucially, the underlying stenosis is treated in the same sitting, or the access simply re-clots.

Central venous stenosis

• Central stenosis presents as arm, facial, or breast swelling and dilated chest-wall collaterals, usually from previous catheters or subclavian lines. It is treated by angioplasty, sometimes with a stent, but it recurs and is hard to cure — which is why prevention, by avoiding the catheters that cause it, matters most.

The surveillance controversy

• It is intuitive that measuring flow and pre-emptively angioplastying every stenosis should prevent thrombosis — yet the trials do not clearly bear this out. Routine surveillance detects more stenoses and triggers more interventions, but its benefit over careful clinical monitoring for thrombosis and access survival is uncertain, and pre-emptive angioplasty can start a restenosis treadmill. Guidance therefore leans on clinical monitoring, with surveillance added in the knowledge that its added value is unproven.

Evidence base

• Angioplasty as first-line and the need to treat the lesion behind a thrombosis rest on consistent procedural and observational data. The surveillance question rests on conflicting randomised trials and meta-analyses — the reason this chapter treats it as a genuine uncertainty rather than a settled rule.

Five fully-built tables.

Table A — Where stenosis forms

Table B — Detecting dysfunction

Table C — Localising by examination

Table D — Treatment options

Table E — What makes a stenosis “significant”

Visualise & Map

Numbered rules. These numbers are the cross-reference handle for the cases and flowcharts.

Clinical Reasoning

Five cases. Each stops you at a decision before it answers it.

Exhaustive. Every rule in the chapter is here.

Safety & Evidence

Panel A — Red flags

Panel B — NEVER DO

Anti-patterns clinicians fall into. Each becomes a Level 22 distractor.

The grade reflects strength of evidence, not importance.

Patient Decisions

Outcomes as natural frequencies. Figures are representative; the direction of effect is given where precise numbers are uncertain.

Apply & Test

Copy-paste chart notes that map to the real decisions in this chapter.

Template 1 — Access assessment note

• Exam: thrill ___; bruit ___; arm-elevation ___; augmentation ___.
• Dialysis clues: venous pressure trend ___; recirculation ___; bleeding ___.
• Flow (Qa) / static pressure trend (if used): ___.
• Suspected lesion: inflow / outflow / central; functional abnormality present? yes/no.
• Plan: continue monitoring / image (duplex or fistulogram).

Template 2 — Intervention note (angioplasty / declot)

• Indication: significant stenosis (> 50% + functional sign) / thrombosis.
• Procedure: angioplasty / stent / thrombectomy + thrombolysis.
• Underlying stenosis treated (if declot): yes/no; site ___.
• Result: patency restored ___; residual stenosis ___.
• Follow-up and monitoring plan: ___.

Pre-rounds compression. Rules only.

Active recall. At least one card per objective.

Micro-teaching for rounds. Two scenarios, five steps each.

Metacognition anchored to this chapter's tensions. No answers provided.

1. 1. It feels negligent to leave a stenosis untreated; how do you sit with watchful waiting when the evidence says treating the picture does not help?
2. 2. Surveillance promises control over an access that keeps failing; how do you tell a real benefit from the comfort of ‘doing something’?
3. 3. Each angioplasty buys time but may shorten the interval to the next; when does the restenosis treadmill argue for a different access altogether?
4. 4. The cheapest, most sensitive monitor is your own hands; what stops a busy unit from using the examination well, and how would you fix that?
5. 5. Central stenosis is mostly self-inflicted by past lines; how does that change how you weigh every catheter you are tempted to place today?

Nine items, each anchored in this chapter. At least one per objective.