Renal Replacement Therapy

Meet the patient

A 64-year-old man with diabetic kidney disease is in the pre-dialysis clinic at eGFR 12. He feels tired and nauseated but still works part-time; his son has offered to donate a kidney. Across the corridor in ED, a 55-year-old with septic shock and KDIGO Stage 3 AKI has a potassium of 6.9 with a widened QRS.[1]

Two examiners will ask you two different questions about these men, and the whole of renal replacement therapy hangs on telling them apart. For the first — planned maintenance dialysis in ESKD — you start for symptoms, plan access months ahead, and push for transplant. For the second — emergency dialysis in AKI — you start for AEIOU (Acidosis, Electrolytes, Ingestion, Overload, Uraemia), not for a creatinine number, and you expect the kidney to recover. Conflating the two is the single commonest error in a viva.[2][4]

What RRT actually is — and why it wears two faces

Renal replacement therapy is any treatment that takes over the kidney's excretory, fluid-balance, electrolyte and acid-base work — and, for transplant, also its endocrine work (erythropoietin, active vitamin D). The same word covers two entirely different jobs, and the job decides everything downstream.[1]

• Maintenance RRT in ESKD (also called kidney failure with replacement therapy, KFRT — CKD Stage 5, irreversible) is a planned pathway. The goal is to prepare early: vascular access, vaccination, transplant assessment, modality education, and preservation of residual function, nutrition and quality of life.
• Acute RRT in severe AKI is a temporary, supportive measure while you fix the underlying insult — sepsis, nephrotoxin, obstruction — and wait for the kidney to recover. The goal is physiological support, and timing is driven by complications (AEIOU), not by a creatinine number.[2][4]

The clinical skill in RRT is not the mechanics of the machine. The machine is a plumber's problem. The skill is deciding when to start, which modality for which patient, how to plan access and transplantation early, and how to recognise the modality-specific emergencies — access bacteraemia, peritonitis, rejection — that decide survival. And do not forget the fifth option: conservative (non-dialytic) care is a legitimate, evidence-supported choice for the frail elderly, focused on symptoms and dignity.[6]

AEIOU — the five reasons to dialyse tonight

When an examiner asks 'when do you start dialysis in AKI?', the answer is one word: AEIOU. Acidosis, Electrolytes, Ingestion, Overload, Uraemia. Memorise the letters and the thresholds verbatim — they are guaranteed marks.[3]

The classic trap: do not dialyse for a biochemical trigger alone. A creatinine of 700 with a normal ECG, a tolerable pH, no fluid overload and no uraemic symptoms is not an indication to dialyse tonight. Dialyse the patient, not the number. AEIOU is the whole of the acute indication; everything else is either planning or a watch-and-wait.[4]

AEIOU sits inside the KDIGO 2012 AKI framework — the international standard for AKI definition, staging, and the timing, modality and dose of RRT. If a viva asks you to 'name the guideline', KDIGO AKI 2012 is the answer.[3]

Start for AEIOU, not for a number — AKIKI, STARRT-AKI and IDEAL

Three landmark trials killed the reflex to dialyse early, and examiners love them. Learn the names, the finding, and the one-line conclusion.[1]

• AKIKI (Gaudry 2016, NEJM) — in critically ill adults with KDIGO Stage 3 AKI, an early (immediate) strategy gave no mortality benefit over a delayed strategy (start only for KDIGO complications or AEIOU, or after 72 hours of Stage 3). The punchline: about half the delayed group never needed dialysis at all.[5]
• STARRT-AKI (Bagshaw 2020, NEJM) — in severe AKI, an accelerated (early) strategy did not reduce 90-day mortality and actually increased adverse events — persistent RRT dependence and infections.[4]
• IDEAL (Cooper 2010, NEJM) — the same principle in ESKD: no survival or quality-of-life benefit of early initiation at eGFR 10 to 14 over late initiation at eGFR 5 to 7.[6]

The conclusion that ties them together: in AKI, start RRT for AEIOU, not for a creatinine number. In stable KDIGO Stage 3 AKI, a watch-and-wait strategy with close monitoring is correct and safer. In ESKD, start when symptoms or complications of uraemia appear — typically at eGFR 5 to 10 mL/min — and keep an asymptomatic patient under surveillance, with access created and transplant assessment underway, until symptoms or an AEIOU-type complication intervenes.[6]

The examiner's favourite follow-up: 'so would you have dialysed earlier to prevent this?'

No. AKIKI and STARRT-AKI together show that early, prophylactic RRT in severe AKI does not save lives and causes harm (persistent dialysis dependence, infection). You dialyse for AEIOU or for AKI that is not recovering — and a patient who meets AEIOU was, by definition, already overdue the moment they developed the indication, not before.[4][5]

Three modalities, one decision

There are three modalities — haemodialysis, peritoneal dialysis, kidney transplant — plus conservative care as a valid fourth. The choice is patient-centred, not nephrologist-centred. Lifestyle, comorbidity, home support, residual renal function and, above all, transplant suitability drive it.[1]

Haemodialysis is dominant worldwide — about 80 to 89 per cent of dialysis patients — because it is efficient, centre-based and scales. Peritoneal dialysis accounts for about 9 to 11 per cent and is over-represented in children, in resource-limited settings, and in countries with active home-dialysis programmes (Mexico, Hong Kong). Transplant rates vary widely by country. The leading causes of ESKD reaching dialysis are diabetic kidney disease (now the commonest in most regions), hypertensive nephrosclerosis, chronic glomerulonephritis (especially IgA nephropathy) and adult polycystic kidney disease.[7]

[1]

Haemodialysis — counter-current diffusion and the Kt/V number

Haemodialysis cleans blood by diffusion across a synthetic membrane, and the whole machine is built to keep that gradient maximal. Blood leaves the patient through the access, runs through a hollow-fibre dialyser (thousands of capillary-like membranes) and returns cleansed. Dialysate flows on the outside of the fibres in the opposite direction — counter-current — so the concentration gradient stays steep along the entire membrane length. Counter-current is the single design choice that makes HD efficient; remember it.[7]

Diffusion moves small molecules best; ultrafiltration removes the water. Small solutes (urea, creatinine, potassium) clear down their gradient from blood to dialysate. Middle molecules like beta-2 microglobulin (molecular weight 11,800) clear poorly on low-flux membranes and need high-flux membranes — the basis of dialysis-related amyloidosis. Water is pulled across by a trans-membrane hydrostatic pressure gradient the machine controls; the volume removed equals the interdialytic weight gain plus any prescribed negative balance. Clearance follows Fick's law (flux proportional to area, permeability, gradient).[1]

The HD prescription you must be able to recite:[1]

• Frequency and duration — conventional 3 sessions a week, 3 to 4 hours each; short-daily (5 to 6 times) or nocturnal (3 to 6 nights, 6 to 8 hours) in selected patients.
• Blood flow — 300 to 400 mL/min through a mature fistula.
• Dialysate flow — 500 to 800 mL/min, counter-current.
• Ultrafiltration — match the interdialytic weight gain; keep the UF rate under 10 to 13 mL/kg/h or you cause intradialytic hypotension and myocardial stunning.
• Dialysate composition — bicarbonate buffer; sodium around 140; potassium usually 2; calcium 1.25 to 1.5 mmol/L.
• Anticoagulation — unfractionated or low-molecular-weight heparin to keep the circuit patent; heparin-free dialysis with saline flushes in active bleeding.[7]

Kt/V is the dose number — know what the letters mean. K = dialyser urea clearance, t = treatment time, V = volume of urea distribution (roughly total body water). The target is single-pool Kt/V at least 1.2 per session (aim for 1.4), measured from paired pre- and post-dialysis urea. The simpler cousin is URR (urea reduction ratio) = (pre-urea minus post-urea) divided by pre-urea, times 100; target at least 65 per cent. Quote both numbers and you have the adequacy question answered.[7]

Peritoneal dialysis — the peritoneum as a living membrane

In PD the peritoneum is the membrane, glucose is the pump, and the patient's own peritoneal capillaries are the blood side. It is gentler, continuous and home-based — which is why it preserves residual renal function longer than HD and suits children, heart-failure patients and anyone who values independence.[1]

The three-pore model — the viva answer for how PD clears solute and water:[1]

• Small pores (about 4 to 5 nm) — the bulk pathway. Urea, creatinine and sodium diffuse down their gradient. The rate is the peritoneal solute transport rate, measured by the peritoneal equilibration test (PET).
• Ultrasmall pores (aquaporin-1) — water-only channels, responsible for free-water clearance in the early, hypotonic phase of a dwell.
• Large pores (over 15 nm) — let macromolecules like albumin leak, which explains the protein loss in PD (about 5 to 15 g a day).[1]

The osmotic gradient comes from glucose — and it dissipates with dwell time. Dialysate comes as 1.36, 2.27 or 3.86 per cent dextrose, or as icodextrin (a polyglucose, iso-osmotic but oncotic, used for the long overnight dwell to sustain ultrafiltration without glucose absorption). As glucose is absorbed the gradient fades, so ultrafiltration is maximal early and falls with dwell time. High transporters (rapid glucose absorption) suit short-dwell APD; low transporters suit long-dwell CAPD. The PET — a 4-hour dwell with 2.27 per cent dextrose — sorts patients into high, high-average, low-average and low transporters, and decides the prescription.[1]

Two PD prescriptions to know:[1]

• CAPD (continuous ambulatory PD) — 3 to 5 manual exchanges a day, dwell volumes 1.5 to 2.5 L. The original method; no machine.
• APD (automated PD) — a cycler runs 8 to 12 hours overnight with 3 to 5 cycles plus a daytime 'last-fill' dwell, often icodextrin for the long daytime dwell. Suits children, working adults and high transporters.[1]

PD adequacy is a weekly number, not a session number. Target total Kt/V at least 1.7 per week (renal plus peritoneal clearance combined), with creatinine clearance targets of 50 to 60 L per week per 1.73 m squared. PD preserves residual renal function — a major survival advantage — so dose is adjusted as residual function declines.[1]

Dialysis access — fistula first, catheter last

Vascular access is the single biggest determinant of survival on haemodialysis, and the order is non-negotiable: AV fistula, then AV graft, then tunnelled catheter. A catheter carries 2 to 3 times the mortality of a fistula. The KDOQI Fistula First Catheter Last initiative exists for this reason.[7]

Why a fistula needs time — and why late referral is a disaster. An AV fistula joins an artery to a vein (classically the radial artery to cephalic vein at the wrist — the Cimino-Brescia fistula). High-pressure arterial flow triggers flow-mediated dilatation and wall thickening (arterialisation) over 6 to 12 weeks, producing a vein you can cannulate repeatedly with large-bore needles at 300 to 400 mL/min. That is why fistulae are created 6 to 12 months before anticipated dialysis. A working fistula cannot be improvised at the moment of need — this single fact is the strongest argument for early nephrology referral at CKD Stage 4 (eGFR under 30).[7]

Three cardinal signs of a patent, mature fistula — examine every dialysis patient for them:[1]

• Thrill — palpable throughout the cardiac cycle (not just systole); confirms flow.
• Bruit — a continuous machinery bruit, systolic and diastolic.
• Pulse augmentation — compress the fistula proximally and the pulse augments.[1]

The access red flags — named traps that lose marks if missed:[1]

• Absent thrill or bruit — thrombosis. Urgent vascular review; a fistula can often be salvaged by thrombolysis or thrombectomy if caught within 24 to 48 hours.
• Erythema, swelling, pus — infection. Cultures and IV antibiotics; cover S. aureus including MRSA. Do not cannulate an infected fistula.
• Cold, painful, pale hand with weak pulses — steal syndrome. Urgent vascular review.
• Swollen arm with prominent chest-wall veins — central venous stenosis, especially after a previous subclavian catheter.[1]

The classic access trap: never use a subclavian catheter if you can avoid it. A subclavian line scars the central vein and destroys future fistula veins in that arm. The internal jugular is always preferred — lowest infection and stenosis — and the femoral is acceptable in an emergency or coagulopathy. If dialysis is needed immediately and no mature fistula exists, place a non-tunnelled internal jugular dialysis catheter as a bridge, then get a tunnelled line or fistula as soon as possible.[1]

PD peritonitis — cloudy effluent is peritonitis until proven otherwise

Peritonitis is the cardinal complication of PD, and 'cloudy effluent' is the phrase that should trigger your whole pathway. Any PD patient with cloudy effluent and abdominal pain has peritonitis until proved otherwise — do not wait for culture.[9]

The diagnostic triad — reproduce it verbatim:[9]

• Cloudy effluent with a white cell count over 100 per mL and over 50 per cent neutrophils, after a dwell of at least 2 hours.
• Abdominal pain, often with fever.
• Gram stain is usually negative; culture is positive in about 70 to 90 per cent. Commonest organism is coagulase-negative staphylococci (Staph epidermidis), then S. aureus, then Gram-negatives; fungal in recurrent disease.[9]

The empirical treatment — ISPD 2016, give it within 6 hours:[9]

1. Send effluent for cell count and culture immediately.
2. Start empirical intraperitoneal antibiotics within 6 hours — gram-positive cover (vancomycin or a first-generation cephalosporin) plus gram-negative cover (ceftazidime or an aminoglycoside), both intraperitoneally.
3. Add heparin 500 to 1000 units per litre intraperitoneally to reduce fibrin and catheter blockage.
4. Most cases respond within 48 to 72 hours. Catheter removal is indicated for refractory, fungal, or relapsing peritonitis.[9]

Kidney transplant — the goal, not the last resort

Transplant is the best survival and the best quality of life ESKD medicine can offer, and it halves long-term mortality against staying on the waiting list (Wolfe 1999). It is the goal, not the last resort. Every suitable patient should be assessed and listed; age alone is not a contraindication, and pre-emptive transplant before dialysis (from a living donor) gives the best outcomes of all.[8]

The donor ladder — best to worst:[1]

• Living donor (related or unrelated, including spousal) — best graft and patient survival, can be planned pre-emptively.
• Deceased donor after brain death (DBD) — heart-beating; the standard deceased source.
• Deceased donor after circulatory death (DCD) — non-heart-beating; expanding the donor pool.
• Expanded-criteria donor (ECD) — older donors or donors with comorbidity, used selectively to expand the pool.[1]

The surgery, in one line: the donor kidney sits extraperitoneally in the right or left iliac fossa, with the renal artery anastomosed end-to-side to the external iliac artery, the renal vein to the external iliac vein, and the ureter implanted into the bladder (ureteroneocystostomy). The native kidneys are usually left in situ — except in polycystic disease with massive enlargement, or uncontrolled hypertension or infection.[1]

The immunosuppression regimen — TIM, the triple therapy

The maintenance regimen is a calcineurin inhibitor plus an antiproliferative plus a steroid — tacrolimus, mycophenolate, prednisolone. Learn it as TIM. It targets the three signals of T-cell activation: signal 1 (TCR binds donor peptide on MHC), signal 2 (CD28 co-stimulation), signal 3 (IL-2 driven proliferation).[1]

[1]

Induction is given peri-operatively to most recipients. Basiliximab (an interleukin-2 receptor antagonist) for low-immunological-risk recipients; anti-thymocyte globulin (ATG) for high-risk recipients (sensitised, delayed graft function, re-transplant). Alternatives to tacrolimus exist when you need to spare the kidney: ciclosporin (more hirsutism, gingival hypertrophy and hypertension, less diabetes) and sirolimus or everolimus (mTOR inhibitors — used to spare CNI nephrotoxicity; side-effects hyperlipidaemia, impaired wound healing, proteinuria, mouth ulcers). Steroid withdrawal or minimisation in low-risk recipients reduces diabetes and osteoporosis.[1]

Prophylaxis post-transplant — three drugs you will see on every drug chart:[1]

• Co-trimoxazole (trimethoprim-sulfamethoxazole 480 mg daily, or 960 mg three times a week) for 6 to 12 months — prevents PJP, and also toxoplasmosis, listeria and nocardia.
• Valganciclovir for a CMV donor-positive, recipient-negative mismatch — typically 900 mg daily for 3 to 6 months.
• Nystatin or fluconazole for oral candidiasis prophylaxis for 1 month; entecavir or tenofovir for HBsAg-positive recipients to prevent hepatitis B reactivation under immunosuppression.[1]

The transplant timeline — what threatens a graft and when

A transplant patient faces two threats for life — rejection and opportunistic infection — and the timeline tells you which is more likely at any moment. A rising creatinine is not always rejection; biopsy is definitive, and the differential includes CNI toxicity, BK nephropathy, obstruction and recurrence of native disease.[1]

Rejection — four types by speed

Sort rejection by how fast it happens — that one fact points to mechanism and treatment.[1]

Treating acute rejection — the doses to quote:[1]

• Acute cellular rejection — IV methylprednisolone 500 mg to 1 g daily for 3 days; if steroid-resistant, anti-thymocyte globulin (ATG).
• Antibody-mediated rejection — plasmapheresis (3 to 6 sessions), IV immunoglobulin (about 2 g/kg, often divided), rituximab (anti-CD20), and in refractory cases bortezomib (a proteasome inhibitor targeting plasma cells).
• Then optimise baseline immunosuppression — confirm the tacrolimus is in range and the patient is adherent.[1]

Chronic allograft nephropathy is the slow killer — interstitial fibrosis and tubular atrophy driven by chronic antibody-mediated rejection, CNI toxicity and recurrence of native disease, with transplant glomerulopathy and transplant arteriopathy on biopsy. It is irreversible; graft loss is gradual over years.[10]

Opportunistic infection — the timeline after transplant

The month after transplant tells you the likely organism. Memorise the three windows.[1]

Rising creatinine after transplant — the differential trap. Do not assume rejection. A biopsy settles it. The competitors:[1]

• CNI nephrotoxicity — high tacrolimus or ciclosporin trough; biopsy shows isometric vacuolisation of tubules and arteriolar hyalinosis; improves on dose-lowering.
• BK polyomavirus nephropathy — rising BK PCR, decoy cells in urine; biopsy shows intranuclear viral inclusions in tubular cells; treat by reducing immunosuppression.
• Ureteric obstruction — hydronephrosis on ultrasound; ureteric stenosis or clot.
• Recurrence of native disease — FSGS (early, nephrotic-range proteinuria), IgA nephropathy, MPGN, atypical HUS.
• Volume depletion, ATN (delayed graft function), drug nephrotoxicity (NSAIDs, contrast, ACE inhibitors).[1]

Malignancy after transplant is part of the same coin. Non-melanoma skin cancer is roughly 100-fold more common — annual dermatology review and sun protection are mandatory. Post-transplant lymphoproliferative disorder (PTLD) is an EBV-driven B-cell lymphoma, especially in an EBV donor-positive, recipient-negative mismatch.[1]

The dialysis emergency at 3am — dialysis disequilibrium

The first dialysis of a heavily uraemic patient can cause cerebral oedema, seizures and coma — dialysis disequilibrium — and you prevent it by starting gentle and short. Rapid solute shift drops plasma osmolality faster than brain osmolality, water moves into the brain, and intracranial pressure climbs. The risk rises with a urea over 35 to 40 mmol/L at the first session.[1]

The prevention, verbatim:[1]

• A short, gentle first session — about 2 hours, low blood flow, reduced dialysate flow.
• Avoid over-rapid ultrafiltration.
• Symptoms — headache, nausea, restlessness, seizures, coma — appear during or just after the session.[1]

This is a safety-critical trap: humour is off, and the dose is 'gentle and short' for the first run of any heavily uraemic patient.[1]

What you find at the bedside

The bedside exam in RRT is modality-specific — the fistula, the PD exit-site, the transplant graft — and each has its own three-sign script.[1]

Examining the AV fistula — thrill (throughout the cycle), bruit (continuous machinery), pulse augmentation; then screen for the red flags: absent thrill (thrombosis), erythema or pus (infection), cold painful hand (steal), aneurysm, and a swollen arm with chest-wall collaterals (central venous stenosis). Do an Allen test before radio-cephalic fistula creation to confirm ulnar supply to the hand.[1]

Examining the PD patient — inspect the exit-site and score it with the Twardowski criteria (0 excellent to 4 tunnel infection); feel for hernias (inguinal, incisional, umbilical), leaks around the catheter, and signs of peritonitis (guarding, rebound). PD patients can be volume-overloaded or depleted, so assess clinically.[1]

Examining the transplant recipient — palpate the graft in the iliac fossa (tenderness suggests rejection or infection), auscultate for a bruit (anastomotic stenosis), check the blood pressure (CNI and steroids raise it), run a fever work-up (the immunosuppressed patient has subtle signs — examine skin, oropharynx, perianal area), and screen for malignancy (full skin exam, lymphadenopathy, splenomegaly for PTLD). The trajectory of creatinine matters more than any single value — a 25 per cent rise warrants investigation.[1]

[1]

Long-term dialysis — the complications that decide survival

Annual mortality on dialysis is about 10 to 20 per cent, and the leading cause of death is cardiovascular disease (40 to 50 per cent of deaths), followed by infection and withdrawal. The long-term complications you manage decide whether your patient beats that average.[1]

The HD-specific complications:[1]

• Intradialytic hypotension — the commonest symptom, from rapid ultrafiltration exceeding vascular refill; recurrent episodes cause myocardial stunning.
• Dialysis-related amyloidosis — beta-2 microglobulin deposition (poorly cleared by low-flux membranes) over 5 to 10 years or more, causing carpal tunnel syndrome, arthropathy and bone cysts; reduced by high-flux membranes.
• Allergic and anaphylactoid reactions — to the dialyser membrane (AN69), the sterilant ethylene oxide, or heparin.
• Hepatitis B and C — the historical scourge of dialysis units, now controlled by isolation, hepatitis B vaccination and universal precautions; HCV is curable with direct-acting antivirals.[1]

The shared complications of any long-term dialysis — manage them actively:[1]

• Anaemia — IV iron (target TSAT over 20 per cent, ferritin over 200 micrograms per litre) then an ESA to haemoglobin 100 to 115 g/L (avoid over 130 — cardiovascular risk).
• Mineral bone disease (CKD-MBD) — phosphate binders with meals; activated vitamin D and cinacalcet for secondary hyperparathyroidism (target PTH 2 to 9 times the upper limit of normal); parathyroidectomy for refractory disease.
• Acidosis — oral sodium bicarbonate to keep bicarbonate over 22.
• Hypertension — the single most powerful intervention is dry-weight reduction; then ACE inhibitor or ARB (cautious), calcium channel blocker, beta-blocker.
• Cardiovascular risk — statin (note simvastatin is dose-limited with amlodipine), aspirin if indicated, smoking cessation.[1]

Conservative care is the legitimate fourth option. In the over-80s with heavy comorbidity, mean survival on conservative care is often comparable to dialysis (under 2 years either way) but with fewer hospital days and better quality of life. Offer it openly; it is not a failure.[1]

Continuous or intermittent RRT in the ICU

For the haemodynamically unstable patient, continuous renal replacement therapy (CRRT) or sustained low-efficiency dialysis (SLED) is gentler than intermittent HD, and the choice is driven by stability, anticoagulation and centre expertise — not by a survival difference.[1]

• CRRT runs 24 hours a day. The modalities are CVVH (continuous veno-venous haemofiltration — convective clearance), CVVHD (diffusive) and CVVHDF (combined). Gentle fluid removal suits the shocked patient; the costs are anticoagulation (often regional citrate), immobility, expense, and ongoing drug and nutrient losses.
• SLED is the hybrid — 6 to 8 hour sessions, daily or alternate — gentler than intermittent HD, less resource-intensive than CRRT.
• There is no clear mortality difference between CRRT and intermittent HD in AKI.[1]

Special populations in one line each:[1]

• Children and infants — PD is the preferred initial modality (APD overnight preserves school and family routine); transplant is the goal; growth hormone for short stature; weight-based dosing throughout.
• Pregnancy on dialysis — intensified HD (daily, or 5 to 6 sessions a week, targeting 20 or more hours per week) improves fetal survival to over 70 per cent; PD is rarely used in pregnancy.
• Conception after transplant — defer 1 to 2 years with stable graft function; switch mycophenolate to azathioprine pre-conception (mycophenolate is teratogenic); the safe regimen is tacrolimus plus azathioprine plus prednisolone.
• Hepatitis B, C or HIV — isolate the HBV-positive patient on a dedicated machine; HCV is curable with DAAs before transplant; HIV is no longer a contraindication to transplant.[1]

Exam pearls

• Start dialysis for symptoms (eGFR around 5 to 10) in ESKD, or for AEIOU in AKI — never for a number alone. IDEAL, AKIKI and STARRT-AKI all say so.[4][5][6]
• AEIOU = Acidosis (pH under 7.1 to 7.15), Electrolytes (K over 6.5 or ECG changes), Ingestion (lithium, salicylate, methanol, ethylene glycol, metformin), Overload (refractory pulmonary oedema), Uraemia (pericarditis, encephalopathy).[3]
• HD: extracorporeal diffusion, AV fistula best access, 3 times a week, Kt/V at least 1.2; risks access infection, hypotension, beta-2 microglobulin amyloidosis.[7]
• PD: peritoneal membrane, three-pore model, glucose osmotic gradient, Kt/V at least 1.7 per week; peritonitis is cloudy effluent; Staph epidermidis commonest.[9]
• Transplant halves mortality (Wolfe 1999); living-donor, pre-emptive preferred.[8]
• Immunosuppression is TIM: Tacrolimus (CNI) plus Mycophenolate plus Prednisolone; induction basiliximab or ATG.[1]
• Transplant twin threats: rejection (hyperacute, acute cellular, antibody-mediated, chronic) and opportunistic infection (CMV 1 to 6 months, BK, PJP); plus malignancy (skin 100-fold, PTLD).[1]
• AV fistula matures 6 to 12 weeks and is created 6 to 12 months ahead; thrill plus continuous bruit equals patent; absent equals thrombosis.[7]
• Subclavian catheters destroy future fistula veins — internal jugular always. Mycophenolate is teratogenic — switch to azathioprine pre-pregnancy. Dialysis disequilibrium is the first dialysis of a uraemic patient — start gentle and short.[1]

[3]

Ward-round test

Stem 1 — the number that does not justify dialysis. A 70-year-old with CKD Stage 5 is admitted with a creatinine of 750 micromol/L but is asymptomatic, euvolaemic, potassium 4.8, pH 7.36, and declines dialysis. Two questions: do you dialyse tonight, and what is the trial that backs your decision?[6]

Stem 1 — start for the patient, not the number (answer)

Not tonight. This patient meets no AEIOU criterion and is asymptomatic, so a creatinine of 750 is not an indication to start. IDEAL (Cooper 2010) showed no survival or quality-of-life benefit of early initiation at eGFR 10 to 14 over late initiation at eGFR 5 to 7 in ESKD. Keep him under surveillance, ensure access is planned and transplant assessment is underway, and start when symptoms or an AEIOU-type complication appears.[6]

Stem 2 — the ICU timing question. A 60-year-old in septic shock develops KDIGO Stage 3 AKI but is not yet acidotic, not hyperkalaemic, not overloaded and not uraemic. The registrar wants to 'start dialysis now to get ahead of it'. What do the trials say, and what do you do?[4]

Stem 2 — AKIKI and STARRT-AKI say wait (answer)

Watch and wait, do not start. AKIKI (Gaudry 2016) showed no mortality benefit of early over delayed RRT in Stage 3 AKI — and about half the delayed group never needed dialysis at all. STARRT-AKI (Bagshaw 2020) went further: an accelerated early strategy did not reduce 90-day mortality and increased adverse events (persistent RRT dependence, infection). Monitor closely and start the moment an AEIOU indication appears.[4][5]

Stem 3 — the access trap. A patient needs emergency dialysis tonight and the only available central access is a subclavian line, or an internal jugular line. Which do you choose, and why does it matter for the future?[1]

Stem 3 — internal jugular always (answer)

Internal jugular, every time. A subclavian catheter causes central venous stenosis that destroys the future fistula veins in that arm — the named access trap. The internal jugular has the lowest infection and stenosis rate; the femoral is acceptable in an emergency or coagulopathy. If dialysis is needed immediately with no mature fistula, place a non-tunnelled internal jugular dialysis catheter as a bridge, then plan a fistula or tunnelled line.[1]

Stem 4 — the transplant complication. Six weeks after a deceased-donor transplant, a patient has a fever, a leukopenia and a rising creatinine. What is the differential, what is the classic opportunistic infection at this window, and what is the single investigation that settles rejection?[1]

Stem 4 — CMV window, biopsy to settle rejection (answer)

At 1 to 6 months post-transplant the classic opportunist is CMV — fever, leukopenia, pneumonitis, colitis — especially in a CMV donor-positive, recipient-negative mismatch. The rising creatinine puts acute cellular or antibody-mediated rejection on the differential alongside BK nephropathy (decoy cells, rising BK PCR) and CNI toxicity (high trough). The single investigation that settles rejection is a renal allograft biopsy graded by the Banff schema — never treat rejection empirically when BK, obstruction or CNI toxicity could be the cause. Send a CMV PCR and a BK PCR, check the tacrolimus trough, and biopsy.[1]

Coverage self-check

If you cannot answer any ward-round stem from this page alone, re-read the matching section. The page is built to be self-sufficient for final-prof, NEET-PG, INICET, USMLE and PLAB questions on Renal Replacement Therapy.[1]

References

1. [1]Villa G, Ricci Z, Ronco C. Renal Replacement Therapy Crit Care Clin, 2015.PMID 26410148
2. [2]Neri M, Villa G, Garzotto F, et al. Nomenclature for renal replacement therapy in acute kidney injury: basic principles Crit Care, 2016.PMID 27719682
3. [3]Kellum JA, Lameire N Diagnosis, evaluation, and management of acute kidney injury: a KDIGO summary (Part 1) Crit Care, 2013.PMID 23394211
4. [4]Bagshaw SM, Wald R, Adhikari NKJ, et al. Timing of Initiation of Renal-Replacement Therapy in Acute Kidney Injury N Engl J Med, 2020.PMID 32668114
5. [5]Gaudry S, Hajage D, Schortgen F, et al. Initiation Strategies for Renal-Replacement Therapy in the Intensive Care Unit N Engl J Med, 2016.PMID 27181456
6. [6]Cooper BA, Branley P, Bulfone L, et al. A randomized, controlled trial of early versus late initiation of dialysis N Engl J Med, 2010.PMID 20581422
7. [7]Himmelfarb J, Ikizler TA. Hemodialysis N Engl J Med, 2010.PMID 21047227
8. [8]Wolfe RA, Ashby VB, Milford EL, et al. Comparison of mortality in all patients on dialysis, patients on dialysis awaiting transplantation, and recipients of a first cadaveric transplant N Engl J Med, 1999.PMID 10580071
9. [9]Li PK, Szeto CC, Piraino B, et al. ISPD Peritonitis Recommendations: 2016 Update on Prevention and Treatment Perit Dial Int, 2016.PMID 27282851
10. [10]Nankivell BJ, Borrows RJ, Fung CL, et al. The natural history of chronic allograft nephropathy N Engl J Med, 2003.PMID 14668458