APPLIED CHRONIC KIDNEY DISEASE · VOLUME 6

Chapter 16

Timing the Start

When to Begin Dialysis: Symptom-Guided, Not a Number

●Orientation & Knowledge◆Visualise & Map■Clinical Reasoning▲Safety & Evidence★Patient Decisions✓Apply & Test

Chapter Preamble

Signals declared

Sig-D — Diagnostic (primary). Recognise the symptoms and indications that should trigger dialysis, and separate them from an eGFR number.
Sig-T — Therapeutic (strong). Symptom-guided initiation, the incremental start that preserves residual function, and the management of the start itself.
Sig-E — Equipoise (strong). The exact timing within the symptom-guided window, and whether to start at all, are preference-sensitive — so the chapter builds a decision map and shared-decision scripts.
Sig-V — Evidence-dense (strong). The IDEAL trial settled the early-versus-late question, so the chapter grades and reflects on it.

Levels populated and omitted

Populated (20): L1–L5, L7, L8, L10–L22. The equipoise signal fires the preference-sensitive decisions map (L15) and the shared-decision scripts (L16); the therapeutic and evidence signals fire the absolute-risk table (L14), templates (L17), and reflective prompts (L21).

L6 / L9 mechanism levels — omitted. No Sig-M; this is a decision and evidence chapter, not a mechanistic one.

● Phase A Orientation & Knowledge

Phase A · Level 1

Learning Objectives

By the end of this chapter you should be able to:

Explain why dialysis is started by symptoms and indications rather than a fixed eGFR threshold.
Summarise the IDEAL trial and its conclusion that early start confers no benefit.
List the clinical indications that should trigger dialysis initiation.
Avoid both too-early and too-late starts.
Distinguish the effective-care triggers from the preference-sensitive timing decision.
Describe incremental dialysis and the value of preserving residual kidney function.
Conduct a shared decision about when — and whether — to start dialysis.
Integrate timing with modality choice, access readiness, and the conservative-care option.

Phase A · Level 2

Executive Summary

When to start chronic dialysis is decided by symptoms and clinical indications, not by reaching a particular eGFR.
The IDEAL trial compared early with late planned starts and found no difference in survival or complications.
Most patients in the 'late' arm started earlier than planned because symptoms developed — reinforcing that symptoms, not the number, drive the decision.
The indications to start are uraemic symptoms, refractory fluid overload, refractory hyperkalaemia or acidosis, declining nutrition, and the failure of conservative measures.
Starting too early confers no benefit and adds burden and dependence; starting too late risks uraemic emergencies and a crash start.
The 'sweet spot' is to start when the patient is genuinely clinically indicated — commonly around an eGFR of 6 to 10 when symptomatic, but the number alone does not dictate.
Some decisions are effective care: an emergent indication mandates starting now, and the symptom triggers are clinical, not values, judgements.
Other decisions are preference-sensitive: the exact timing within the symptom-guided window, and — above all — whether to start dialysis at all rather than choose conservative care.
Within the watchful window, the timing reflects the patient's values — their readiness, quality of life, and tolerance of symptoms against the burden of dialysis.
Whether to start at all is the larger preference-sensitive decision, shared with the patient and developed in the conservative-management chapter.
Residual kidney function is valuable and worth preserving, which underpins the incremental start.
Incremental dialysis — beginning with a reduced dose, such as twice-weekly haemodialysis or a lower peritoneal dose — preserves residual function and reduces burden in suitable patients, with monitoring.
Timing is integrated with the modality choice and access readiness of the previous chapters, so a symptom-driven start finds the patient prepared.
Done well, dialysis begins when the patient needs it and wants it — not when a number is crossed.

Phase A · Level 3

Main Narrative

For years, the question 'when should dialysis start?' was answered with a number — an eGFR threshold below which a patient was deemed to need it. A landmark trial dismantled that answer. Dialysis is started not when a number is crossed but when the patient develops the symptoms and indications that dialysis exists to treat, and — for the patient with a poor prognosis — only if they choose it over conservative care. The timing question is part evidence and part values, and this chapter holds both.

From a number to symptoms

The old eGFR-threshold approach — start everyone at, say, an eGFR of 10 — was overturned by the IDEAL trial, which randomised patients to an early planned start, at an eGFR of 10 to 14, or a late one, at 5 to 7 or when symptoms demanded, and found no difference in survival or complications between them. Two findings drove the conclusion. First, early initiation bought nothing — no survival, no fewer complications — while committing patients to dialysis sooner. Second, most patients assigned to the late arm actually started before reaching the low threshold, because they developed symptoms that required it. The lesson is that symptoms, not the number, identify the right moment, and that the eGFR is a guide to when symptoms are likely, not a trigger in itself. Dialysis is started when the patient clinically needs it.

The indications that trigger dialysis

If symptoms drive the decision, the clinician must know which ones. The indications to start are the manifestations of kidney failure that dialysis treats: uraemic symptoms — nausea, anorexia, vomiting, fatigue, pruritus, restless legs, and, more severely, encephalopathy, pericarditis, or bleeding; refractory fluid overload that medical therapy can no longer control; refractory hyperkalaemia or metabolic acidosis; a declining nutritional state attributable to uraemia; and the general failure of conservative measures to keep the patient well. These are clinical triggers, and when they appear and cannot be managed otherwise, they indicate dialysis regardless of the exact eGFR. In practice most patients reach this point around an eGFR of 6 to 10, but the eGFR is the company the indications keep, not the indication itself — an asymptomatic patient at eGFR 9 does not need dialysis, while a symptomatic one at eGFR 11 may.

Neither too early nor too late

The timing has two failure modes. Starting too early — on the number, before symptoms — is the error IDEAL exposed: it confers no benefit, brings forward the burden and dependence of dialysis, and exposes the patient to its complications sooner, for nothing gained. This mirrors the acute lesson from the AKI volume, where accelerated initiation also failed to help. Starting too late — letting symptoms build into a uraemic emergency — is the opposite error, risking encephalopathy, pericarditis, severe overload, and the crash start the access and pathway chapters worked to prevent. The target is the middle: watchful monitoring as the patient approaches kidney failure, with a prompt, planned start when the indications appear — the same watchful-waiting-with-prompt-initiation logic as acute dialysis, applied over months rather than days.

What is effective care, and what is the patient's

As in the other equipoise chapters, the timing question divides into what is clinically determined and what depends on values. Some elements are effective care: an emergent indication — severe refractory hyperkalaemia, uraemic pericarditis — mandates starting now, not later, and the symptom triggers themselves are clinical judgements about when dialysis is needed. But two elements are genuinely preference-sensitive. The first is the exact moment of starting within the symptom-guided window: as symptoms gather, how much the patient is willing to tolerate before beginning, weighed against their readiness and quality of life, is theirs to decide. The second, and larger, is whether to start dialysis at all — the choice between dialysis and conservative kidney management, which for a frail or poor-prognosis patient is a values decision belonging to them, developed in the conservative-care and frailty chapters. Keeping these apart — the clinical triggers from the values-laden timing and the whether — is the discipline of the chapter.

Preserving residual function and the incremental start

Residual kidney function — the urine output and clearance a patient retains as they start dialysis — is valuable: it contributes to solute and fluid clearance, eases the dialysis prescription, and is associated with better outcomes, so it is worth preserving. This underpins the incremental start: rather than beginning every patient on a full, thrice-weekly haemodialysis or full peritoneal dose, a suitable patient with meaningful residual function can begin with a reduced dose — twice-weekly haemodialysis, or a lower peritoneal dialysis dose — that supplements rather than replaces their own kidneys, reducing treatment burden while residual function lasts. Incremental dialysis requires careful monitoring of that residual function so the dose is increased as it declines, and it is not for everyone, but it reflects the principle that dialysis should do what the patient's kidneys cannot, no more, and that the transition onto it can be gradual. It also fits the symptom-guided philosophy: start with what is needed, when it is needed.

Integrating timing with the rest of the pathway

Timing does not stand alone; it is the moment the preparation of the previous chapters pays off. A symptom-driven start works well only if the patient has already chosen a modality and has working access ready, so that when the indications appear, dialysis begins electively on the planned modality rather than urgently through a catheter. This is why the predialysis pathway times preparation by risk and creates access months ahead: not to start dialysis early, but to be ready to start it promptly and well when symptoms call for it. And the timing conversation flows into, or away from, the conservative-care decision — some patients, asked when they want to start, will be asking whether they want to at all. The chapter's scripts handle both the 'when' and the 'whether,' and the answer to each belongs, ultimately, to the patient.

Where the evidence is firm, and where values govern

The firm part is IDEAL's conclusion: early initiation confers no benefit, so dialysis is started on symptoms and indications, not an eGFR threshold — grade-A evidence that reshaped practice. The value of residual function and the rationale for incremental dialysis are well supported, if still being refined. What evidence cannot settle is the exact moment within the symptom-guided window, or whether a given patient should start at all rather than choose conservative care — because those turn on the patient's values, readiness, and prognosis. So the clinician's role is to apply the IDEAL lesson firmly (no early start on the number), to recognise the clinical triggers, to preserve residual function, and then to share the timing and the whether with the patient. The proper close is a start that is symptom-driven, prepared, and chosen.

Phase A · Level 4

Reference Tables

Table 16.1 — The timing question: number versus symptoms

Approach	Verdict
eGFR-threshold start	Outdated — a number does not identify who needs dialysis
Symptom/indication-guided start	Current standard — start when clinically indicated
Typical eGFR at symptomatic start	Often ~6–10, but the number does not dictate
Key principle	Start when the patient needs it, not when a number is crossed

Table 16.2 — The IDEAL trial and the evidence

Element	Detail
Design	Early start (eGFR 10–14) vs late start (5–7 or symptoms)
Result	No difference in survival or complications
Key observation	Most 'late' patients started earlier due to symptoms
Conclusion	Start on symptoms/indications, not an eGFR threshold

Table 16.3 — Indications to start dialysis

Category	Examples
Uraemic symptoms	Nausea, anorexia, fatigue, pruritus, encephalopathy, pericarditis, bleeding
Fluid overload	Refractory to medical therapy
Electrolytes / acid-base	Refractory hyperkalaemia or metabolic acidosis
Nutrition	Declining nutritional state attributable to uraemia
Conservative failure	General failure of medical management to maintain wellbeing

Table 16.4 — Too early versus too late

	Too early	Too late
Trigger	An eGFR number, no symptoms	Symptoms built into an emergency
Harm	No benefit; burden, dependence, complications sooner	Uraemic emergency; crash start
Correct course	Watchful monitoring	Prompt, planned start on indication

Table 16.5 — Effective care versus preference-sensitive

Effective care (clinical)	Preference-sensitive (the patient's values)
Emergent indication → start now	Exact timing within the symptom-guided window
Recognising the symptom triggers	Whether to start at all vs conservative care
Access readiness / planned start	Incremental vs full-dose start
Preserving residual function	How much symptom burden to tolerate first

Table 16.6 — Residual function and incremental dialysis

Item	Detail
Residual function	Valuable — aids clearance/fluid; better outcomes; preserve it
Incremental start	Reduced initial dose (e.g. twice-weekly HD, lower PD dose)
Suits	Patients with meaningful residual function
Requires	Monitoring residual function; increase dose as it declines
Principle	Dialysis supplements the kidneys; start with what is needed

Visualise & Map

◆ Phase B Visualise & Map

Phase B · Level 5

Imaging & Flowchart Specifications

Clinical Reasoning

■ Phase C Clinical Reasoning

Phase C · Level 8

Clinical Cases

CASE 1STARTING ON THE NUMBER

An eGFR is not an indicationSymptom-guided initiation

Presentation

An asymptomatic patient with an eGFR of 9, eating well, with controlled volume and electrolytes, is scheduled to start dialysis 'because the eGFR is below 10.'

❖ Pause and reflect

Does an eGFR of 9 in an asymptomatic patient indicate dialysis?

Analysis

It does not. The IDEAL trial showed that starting dialysis early, on an eGFR threshold rather than symptoms, confers no benefit while bringing forward the burden and dependence of dialysis. This patient is asymptomatic with controlled volume, electrolytes, and nutrition, so there is no clinical indication to start — the number is a guide to when symptoms are likely, not a trigger in itself. Starting now would be the too-early error.

Plan

Do not start on the number. Continue watchful monitoring of symptoms, volume, electrolytes, and nutrition, with the modality chosen and access ready, and start promptly when clinical indications appear.

Teaching point

An eGFR is not an indication — start dialysis on symptoms and clinical indications, not on a number (IDEAL).

Cross-reference

Exercises rules R1 and R3; the IDEAL figure (16.2); Tables 16.1 and 16.2.

CASE 2THE SYMPTOMATIC PATIENT

Now it is indicatedRecognising the triggers

Presentation

A patient with an eGFR of 8 develops worsening nausea, anorexia, fatigue, and refractory fluid overload despite maximal medical therapy, with declining nutrition. The team hesitates because 'the eGFR isn't that low yet.'

❖ Pause and reflect

Should the eGFR delay starting in a symptomatic, overloaded patient?

Analysis

This patient is clinically indicated regardless of the eGFR. Uraemic symptoms, refractory fluid overload, and declining nutrition are exactly the triggers dialysis exists to treat, and the symptom-guided approach says to start when they appear and cannot be managed otherwise. Hesitating because the eGFR 'isn't low enough' inverts the lesson — symptoms, not the number, indicate dialysis, and waiting risks a uraemic emergency.

Plan

Start dialysis on the chosen modality with the prepared access, treating the clear clinical indications. The eGFR of 8 is the company the symptoms keep, not a reason to delay.

Teaching point

When uraemic symptoms, refractory overload, or declining nutrition appear, start — don't let a 'not-low-enough' eGFR delay an indicated start.

Cross-reference

Exercises rules R2 and R4; Table 16.3; the sweet-spot figure (16.1).

CASE 3WHEN, OR WHETHER?

The bigger questionTiming versus the conservative choice

Presentation

A frail elderly patient approaching kidney failure is asked by the team 'when' she wants to start dialysis. In the conversation it becomes clear she is unsure she wants dialysis at all, valuing time at home over its burdens.

❖ Pause and reflect

Is her question really 'when,' or is it 'whether'?

Analysis

Her question is the larger, preference-sensitive one: whether to start dialysis at all, rather than merely when. For a frail patient who values time at home over the burdens of dialysis, conservative kidney management is a legitimate choice, and pressing only the 'when' would miss the decision she is actually facing. The timing conversation must be ready to become a whether conversation, sharing the dialysis-versus-conservative choice grounded in her values.

Plan

Step back from 'when' to 'whether': share the dialysis-versus-conservative decision honestly, elicit her values, and support her choice, with the conservative pathway developed in the conservative-care chapter. Do not assume dialysis is the destination.

Teaching point

Asking 'when' can surface 'whether' — be ready to share the dialysis-versus-conservative decision, not just the timing.

Cross-reference

Exercises rule R6; the L15 map and L16 scripts; conservative management in Chapter 17; frailty in Chapter 12.

CASE 4PRESERVE WHAT'S LEFT

The incremental startResidual kidney function

Presentation

A patient starting haemodialysis still passes good volumes of urine and retains meaningful residual kidney function. The default plan is to begin full thrice-weekly dialysis immediately.

❖ Pause and reflect

Should a patient with good residual function start full-dose dialysis, or is there an alternative?

Analysis

His meaningful residual kidney function is valuable — it aids clearance and fluid handling and is associated with better outcomes — and full thrice-weekly dialysis from the outset ignores it. An incremental start, beginning with a reduced dose such as twice-weekly haemodialysis that supplements rather than replaces his own kidneys, preserves residual function and reduces burden while it lasts, with the dose stepped up as that function declines. Dialysis should do what his kidneys cannot, no more.

Plan

Consider an incremental start with a reduced initial dose, monitoring residual kidney function closely and increasing the dose as it falls. Reserve full dosing for when residual function no longer supplements adequately.

Teaching point

Residual kidney function is valuable — an incremental start preserves it and reduces burden in suitable patients, with monitoring.

Cross-reference

Exercises rule R5; the incremental-start figure (16.3); Table 16.6.

Phase C · Level 10

Clinical Pearls

Start dialysis on symptoms and indications, not an eGFR number.

IDEAL: early start (eGFR 10–14) vs late — no survival/complication difference.

Most 'late' patients started early due to symptoms — symptoms drive it.

Indications: uraemic symptoms, refractory overload, refractory K/acidosis, declining nutrition, conservative failure.

Too early = no benefit + burden/dependence; too late = emergency/crash start.

Aim for the symptom-guided sweet spot.

Most start ~eGFR 6–10 when symptomatic — but the number doesn't dictate.

An asymptomatic patient at eGFR 9 does not need dialysis.

A symptomatic patient at eGFR 11 may.

Emergent indication → start now (effective care).

Exact timing within the window is preference-sensitive.

Whether to start at all (vs conservative) is the bigger preference-sensitive decision.

Residual kidney function is valuable — preserve it.

Incremental dialysis (e.g. twice-weekly HD) preserves residual function, reduces burden.

Monitor residual function and step up the dose as it declines.

Integrate timing with modality choice and access readiness — a prepared, planned start.

Safety & Evidence

▲ Phase D Safety & Evidence

Phase D · Level 11

Red Flags & Never-Do

Panel A — Red flags

▲

An asymptomatic patient being started on dialysis because of an eGFR number — no indication; do not start (IDEAL).

▲

Uraemic pericarditis, encephalopathy, severe overload, or refractory hyperkalaemia — emergent indication; start now.

▲

Symptoms building unaddressed toward a uraemic emergency — too-late start looms; act on the indications.

▲

A frail patient pressed on 'when' who is really unsure 'whether' — share the conservative-care decision.

▲

Full-dose dialysis started despite good residual function — consider an incremental start.

Panel B — Never do

✖ NEVER — start dialysis on an eGFR threshold in an asymptomatic patient.

✖ NEVER — delay an indicated start because the eGFR 'isn't low enough.'

✖ NEVER — treat 'when to start' as settled when the patient is questioning 'whether.'

✖ NEVER — ignore residual kidney function when prescribing the start.

Phase D · Level 12

Common Pitfalls

Pitfall 1 — Starting on the number

✖

WRONG Starting dialysis because the eGFR fell below a threshold.

✓

RIGHT Starting on symptoms and clinical indications.

✉

WHY IDEAL showed early, number-driven starts confer no benefit.

Pitfall 2 — Delaying the indicated start

✖

WRONG Withholding dialysis from a symptomatic, overloaded patient because the eGFR 'isn't low enough.'

✓

RIGHT Starting when the clinical indications appear, regardless of the eGFR.

✉

WHY Symptoms, not the number, indicate dialysis; delay risks an emergency.

Pitfall 3 — Missing the 'whether'

✖

WRONG Pressing only 'when' with a patient unsure about dialysis at all.

✓

RIGHT Sharing the dialysis-versus-conservative decision.

✉

WHY Whether to start is the larger preference-sensitive choice.

Pitfall 4 — Ignoring residual function

✖

WRONG Starting full thrice-weekly dialysis despite good residual function.

✓

RIGHT Considering an incremental start with monitoring.

✉

WHY Residual function is valuable and worth preserving.

Pitfall 5 — An unprepared start

✖

WRONG Waiting for symptoms with no modality chosen or access ready.

✓

RIGHT Preparing in advance so a symptom-driven start is planned and elective.

✉

WHY Symptom-guided timing only works if the patient is prepared (Chapters 14–15).

Phase D · Level 13

Evidence Grading

GRADE

HIGH CONFIDENCE

The effect is real and the estimate is stable.

RCTs at low risk of bias; multiple concordant prospective cohorts; meta-analyses.

GRADE

MODERATE CONFIDENCE

The effect is likely real but may shift with new data.

Observational studies, registries, mechanistic human studies.

GRADE

LOW CONFIDENCE

Rests on physiology, reasoning, or consensus rather than outcomes.

Pathophysiological reasoning; extrapolation; consensus without outcomes.

Graded statements (by evidence type)

Statement	Grade	Basis (evidence type)
Early dialysis initiation confers no survival benefit over a later, symptom-guided start.	A	RCT (IDEAL)
Dialysis should be started on symptoms/indications, not an eGFR threshold.	A	RCT and guideline consensus
Starting too late risks uraemic emergencies and crash starts.	B	Observational data
Residual kidney function is associated with better outcomes.	B	Observational data
Incremental dialysis preserves residual function in suitable patients.	B	Observational and emerging trial data
Whether to start dialysis at all is preference-sensitive.	C	Ethical reasoning and consensus
Emergent indications mandate immediate dialysis.	A	Physiology and consensus

Patient Decisions

★ Phase E Patient Decisions

Phase E · Level 14

Absolute Risk in Natural Frequency

Natural-frequency estimates for orientation, from the timing evidence; they vary with population. They convey the size of the timing decisions, expressed per 100 comparable patients.

Per 100 patients…	Outcome	Roughly how many	See
Started early vs symptom-guided	Survive / avoid complications	About the same — no benefit to early	L13 row 1
Assigned to a 'late' (symptom-guided) strategy	Actually start early due to symptoms	Most of them	L13 row 1
Started symptom-guided with good preparation	Have a planned (not crash) start	More than the unprepared	L13 row 3
Started incrementally with residual function	Preserve residual function longer	More than full-dose starts	L13 row 5

★ How to read these

Read these as orientation, not promises; outcomes vary with population and preparation. The stable signals: early initiation adds no benefit, symptoms drive most starts anyway, and residual function is worth preserving. Communicate them as people out of 100, not as a hazard ratio — and alongside what the patient wants.

Phase E · Level 15

Preference-Sensitive Decisions

First, what is NOT on this map. Some elements are effective care, not choices: an emergent indication mandates starting now, the symptom triggers are clinical judgements, and access readiness and residual-function preservation are good care. Starting on an eGFR number alone is simply wrong, not a preference.

What IS preference-sensitive. The decisions below depend on the patient's values, readiness, and prognosis, and belong in a shared conversation.

Decision	The options	What it trades / whose values decide
Timing within the symptom window	Start nearer the first symptoms vs nearer the threshold	Symptom burden vs readiness and quality of life — the patient's tolerance
Whether to start at all	Dialysis vs conservative kidney management	Length vs burden and quality — the patient's values (Chapters 12, 17)
Incremental vs full-dose start	Reduced initial dose vs full dose	Preserving residual function vs simplicity — clinician and patient together
Place / pace of starting	Planned elective start vs delay a little longer	Readiness vs risk of waiting — the patient's wishes

Phase E · Level 16

Shared-Decision Scripts

Language for the 'when' and the 'whether' conversations. Adapt to the patient; the teach-back confirms understanding.

Script 1 — We start when you need it, not at a number

CLINICIAN “Your kidney function is low, and I know a number on the report can feel alarming. But we don't start dialysis just because a number crosses a line.”

PATIENT “So how will we know when to start?”

CLINICIAN “We watch for how you feel and for specific problems — nausea, poor appetite, fatigue, fluid you can't shift, or blood tests that medicines can no longer control. A big trial showed starting early, just on the number, doesn't help people live longer — so we start when your body actually needs it.”

PATIENT “And until then?”

CLINICIAN “We keep your access ready and watch closely, so that when the time comes we can start smoothly and on your terms — not in an emergency. You won't be caught out.”

TEACH-BACK “So I've explained it clearly — how would you describe what we're watching for, rather than just the number?”

Script 2 — When becomes whether

CLINICIAN “As we talk about when to start dialysis, I want to make sure we're answering the right question — because for some people, the real question is whether dialysis is the path they want at all.”

PATIENT “I'm honestly not sure I want to be tied to a machine.”

CLINICIAN “That's an important thing to say, and it's a legitimate choice. There's an alternative — conservative kidney management — where we treat your symptoms and look after you actively without dialysis, focusing on how you feel and your time at home.”

PATIENT “Would that mean giving up?”

CLINICIAN “Not at all — it's an active plan of care, just a different one, and whichever you choose we'll look after you. What matters is what fits your life and what you value. Let's take the time to think it through together.”

TEACH-BACK “Just to check I've been clear and not pushed you either way — how would you describe the two paths in front of you?”

Apply & Test

✓ Phase F Apply & Test

Phase F · Level 17

Documentation Templates

Paste-ready notes. Tick the boxes that apply and delete the rest; make the symptom-based rationale and the when/whether distinction explicit.

Template 1 — Dialysis-start decision

eGFR ___ (a guide, not a trigger); symptom/indication review: ☐ uraemic symptoms ☐ refractory overload ☐ refractory K/acidosis ☐ declining nutrition ☐ none.
Emergent indication present → start now: ☐ yes ☐ no.
If no indication → watchful monitoring; NOT started on the number alone: ☐ confirmed.
Whether-to-start considered (vs conservative care, if poor prognosis): ☐ yes → shared decision (Chapter 17).
Preparation in place: ☐ modality chosen ☐ access ready (a planned, not crash, start).
Exact timing shared with patient (values/readiness): ☐ yes.

Template 2 — Symptom-guided monitoring and incremental start

Monitoring: symptoms, volume, electrolytes/acid-base, nutrition — frequency ___ .
Start trigger defined (the indications that will prompt initiation): ___ .
Residual kidney function: ___ (urine output/clearance) — valuable, to be preserved.
Incremental start considered (reduced initial dose): ☐ yes (e.g. twice-weekly HD / lower PD dose) ☐ not suitable.
Residual-function monitoring plan; dose to step up as it declines: ☐ yes.
Integrated with modality/access (Chapters 14–15): ☐ yes.

Phase F · Level 18

Cheat Sheet

Start dialysis on SYMPTOMS/indications, not an eGFR number.

IDEAL: early (10–14) vs late — no survival/complication difference.

Most 'late' patients started early due to symptoms.

Indications: uraemic symptoms, refractory overload, refractory K/acidosis, declining nutrition.

Too early = no benefit + burden; too late = emergency/crash start.

Symptom-guided sweet spot.

Most start ~eGFR 6–10 when symptomatic — number doesn't dictate.

Asymptomatic at eGFR 9 → don't start.

Symptomatic at eGFR 11 → may start.

Emergent indication → start now (effective care).

Exact timing within window = preference-sensitive.

Whether to start at all (vs conservative) = the bigger preference-sensitive decision.

Residual kidney function is valuable — preserve it.

Incremental dialysis (e.g. twice-weekly HD) preserves it; monitor and step up.

Integrate with modality + access (planned, not crash, start).

'When' can become 'whether' — be ready to share the conservative option.

Phase F · Level 19

Flashcards

CARD 1

Q. When should chronic dialysis be started?

Show answer

A. On the basis of uraemic symptoms and clinical indications, not a fixed eGFR threshold — the eGFR is a guide to when symptoms are likely, not a trigger.

DETAILED. The IDEAL trial showed early, number-driven starts confer no benefit.

CLINICAL. Start when the patient clinically needs it.

CARD 2

Q. What did the IDEAL trial show?

Show answer

A. Early planned start (eGFR 10–14) versus late (5–7 or symptoms) gave no difference in survival or complications, and most 'late' patients started early because of symptoms.

DETAILED. Symptoms, not the number, set the moment.

CLINICAL. Start on indications, not an eGFR threshold.

CARD 3

Q. What are the indications to start dialysis?

Show answer

A. Uraemic symptoms (nausea, anorexia, fatigue, pruritus, encephalopathy, pericarditis), refractory fluid overload, refractory hyperkalaemia or acidosis, declining nutrition, and failure of conservative measures.

DETAILED. These are clinical triggers regardless of the exact eGFR.

CLINICAL. Start when they appear and cannot be managed otherwise.

CARD 4

Q. What are the two failure modes of timing?

Show answer

A. Too early — starting on the number, no benefit, with added burden and dependence; and too late — letting symptoms build into a uraemic emergency and a crash start.

DETAILED. The target is the symptom-guided middle.

CLINICAL. Watchful monitoring with a prompt, planned start on indication.

CARD 5

Q. Which timing decisions are preference-sensitive?

Show answer

A. The exact moment of starting within the symptom-guided window, whether to start at all rather than choose conservative care, and incremental versus full-dose starting.

DETAILED. Emergent indications and the symptom triggers are effective care.

CLINICAL. Share the values-laden decisions; provide the clinical ones.

CARD 6

Q. Why preserve residual kidney function, and how?

Show answer

A. It aids clearance and fluid handling and is associated with better outcomes; an incremental start — a reduced initial dose such as twice-weekly haemodialysis — preserves it while it lasts.

DETAILED. Dialysis should supplement the kidneys, not replace them prematurely.

CLINICAL. Consider incremental dialysis with monitoring in suitable patients.

CARD 7

Q. Why must timing be integrated with the predialysis pathway?

Show answer

A. A symptom-driven start works well only if the patient has already chosen a modality and has working access, so dialysis begins electively rather than through an emergency catheter.

DETAILED. Preparation is what makes a prompt start a planned one.

CLINICAL. Time preparation by risk so a symptom-driven start is ready.

CARD 8

Q. How can the 'when' question become a 'whether' question?

Show answer

A. A patient asked when to start may be unsure they want dialysis at all; conservative kidney management is a legitimate alternative, so the timing conversation must be ready to become a dialysis-versus-conservative one.

DETAILED. Whether to start is the larger preference-sensitive decision.

CLINICAL. Be ready to share the conservative option, not just the timing.

Phase F · Level 20

One-Minute Preceptor

SCENE 1

The intern starting on the number

GET A COMMITMENT“You've booked this asymptomatic patient to start dialysis — why?”

PROBE FOR EVIDENCE“Her eGFR is below 10” — ask: “Does she have any symptoms or indications, and what did IDEAL show about starting on the number?”

TEACH A GENERAL RULEDialysis is started on symptoms and indications, not an eGFR threshold — early, number-driven starts confer no benefit.

REINFORCE WHAT WAS RIGHTWatching her function closely was correct.

CORRECT A MISTAKEHold dialysis, monitor symptoms, and start when she is clinically indicated.

SCENE 2

The resident answering the wrong question

GET A COMMITMENT“You're arranging when this frail patient will start — has she said she wants dialysis?”

PROBE FOR EVIDENCE“She's approaching failure” — ask: “Is her question really 'when,' or might it be 'whether'?”

TEACH A GENERAL RULEFor a frail patient, whether to start at all is a preference-sensitive choice — conservative care is a legitimate alternative that the timing conversation must surface.

REINFORCE WHAT WAS RIGHTEngaging her about timing was a good start.

CORRECT A MISTAKEStep back and share the dialysis-versus-conservative decision against her values.

Phase F · Level 21

Reflective Prompts

Genuine tensions this evidence and these values leave open; sit with them rather than resolving them too quickly.

IDEAL showed early start gains nothing, yet eGFR thresholds persist in practice and patients' minds. Why is a discredited number so hard to let go of?
Symptom-guided timing requires tolerating a falling eGFR and a worsening patient without acting. How do you hold that nerve, and how do you know when watching has become waiting too long?
The 'when' and the 'whether' blur at the bedside. How do you make space for a patient to say 'not at all' when the whole pathway has been pointing toward dialysis?
Incremental dialysis preserves residual function but demands monitoring and individualisation. Where is the line between elegant, kidney-sparing care and under-dialysis?
Preparation makes a symptom-driven start possible, but over-preparing can pressure a patient toward a start they have not chosen. How do you prepare without nudging?

Phase F · Level 22