🩺 Reasoning GP — Clinical Algorithms

AKI is a medical emergency superimposed on CKD in up to 20% of CKD patients annually. Missing AKI causes irreversible nephron loss and accelerates progression to end-stage renal disease (ESRD). Hyperkalaemia is the most immediately life-threatening CKD complication — potassium ≥6.5 mmol/L or any ECG change requires emergency treatment within minutes. NICE NG203 (2021) mandates that eGFR <15 triggers immediate nephrology involvement for renal replacement therapy planning. Hypertensive emergency with renal involvement can cause irreversible target organ damage within hours. The AKI e-alert system (NHS England) is triggered by KDIGO criteria; GPs should be familiar with these thresholds. Stopping nephrotoxins promptly in AKI can allow near-complete recovery if done within hours.

The 90-day rule prevents over-diagnosis of CKD from transient eGFR dips (e.g. dehydration, acute illness, NSAID use). Labelling a patient with CKD incorrectly leads to unnecessary investigations, anxiety, insurance implications, and overmedicalisation. Conversely, missing CKD delays cardiovascular risk reduction — CKD independently doubles cardiovascular mortality. ACR is more sensitive than PCR for early diabetic nephropathy and is now the preferred marker. Isolated haematuria without proteinuria or eGFR reduction more commonly indicates urological pathology (bladder cancer, renal cell carcinoma) — hence the separate urology referral pathway.

The KDIGO heat map (G × A grid) identifies "high risk," "very high risk," and "highest risk" combinations that predict both CKD progression and cardiovascular events. For example, G3bA3 has a 40-fold higher risk of kidney failure than G1A1. Classification directly determines: monitoring intervals (NICE Table 2), when to refer to nephrology, treatment targets (BP, proteinuria), and when to start specific therapies (finerenone in diabetic CKD, dapagliflozin, SGLT2i). This KDIGO system replaced the old "Stage 1–5" system and better risk-stratifies patients — a patient with G2A3 (mild eGFR, severe proteinuria) needs MORE intensive management than G3aA1.

Blood pressure measurement in both arms is essential — a difference >15 mmHg suggests subclavian stenosis or aortic coarctation, both common in CKD patients with atherosclerosis. The target BP of <130/80 in proteinuric CKD is supported by the SPRINT trial (reduces CV events by 25%) and NICE NG203. Fluid overload is the most common cause of "resistant hypertension" in CKD — optimising loop diuretics can reduce systolic BP by 20–30 mmHg without adding antihypertensives. Palpable kidneys in a patient with bilateral renal enlargement should prompt urgent renal ultrasound for ADPKD, which has specific genetic counselling implications. Uraemic pericarditis is a dialysis indication regardless of eGFR.

Renal ultrasound at diagnosis is mandated by NICE NG203 because obstructive uropathy (e.g., prostatic hypertrophy, pelviureteric junction obstruction) is potentially fully reversible — catheterisation or urology intervention can restore kidney function. Bicarbonate monitoring from G3a is important because metabolic acidosis directly activates catabolic pathways that accelerate muscle loss and CKD progression — sodium bicarbonate supplementation (1–2g TDS) slows eGFR decline in the BICAR-RCT by 4 ml/min/yr. CKD-mineral bone disease (MBD) causes vascular calcification, which explains why CV mortality in CKD is 10–30× higher than the general population. Gadolinium MRI contrast is contraindicated in eGFR <30 (risk of nephrogenic systemic fibrosis) — check eGFR before ordering any contrast investigation.

The most important referral is eGFR <30 for RRT planning — this should happen 1–2 years before anticipated need. Arteriovenous fistula formation for haemodialysis takes 3–6 months to mature; late referral forces temporary vascular access catheters, which have significantly higher infection and mortality rates. The "rapid decline" threshold (>25% in 12 months) captures patients with accelerated CKD from renovascular disease, myeloma, or uncontrolled diabetes — these are reversible if caught early. ACR >300 mg/mmol is the nephrotic range threshold that often indicates primary glomerular disease requiring biopsy (done by nephrology). Approximately 60% of UK CKD is managed solely in primary care — the system depends on GPs understanding when NOT to refer, as much as when to refer.

The permitted creatinine/eGFR rise on ACEi/ARB initiation (up to 30% creatinine, up to 25% eGFR) is one of the most important practical points in CKD management. Stopping ACEi/ARB because of an expected haemodynamic eGFR dip deprives patients of renoprotection — the REIN trial showed ramipril halved the rate of progression to ESRD in proteinuric CKD. SGLT2 inhibitors represent the biggest advance in CKD management in 20 years — dapagliflozin's 39% reduction in CKD progression (DAPA-CKD) is superior to any previous agent. Statins in CKD: NNT to prevent one major CV event is approximately 15 over 5 years (SHARP trial), comparable to statins in post-MI patients. Avoiding combination ACEi + ARB is critical — this was definitively shown to cause harm (increased AKI, hyperkalaemia, death) without additional renal benefit in ONTARGET.

Salt restriction is one of the most under-utilised interventions in CKD hypertension. A 1g/day reduction in sodium reduces systolic BP by approximately 2–3 mmHg, and dietary restriction of 2g sodium/day reduces SBP by 10 mmHg — equivalent to adding a second antihypertensive. Sick day rules prevent an estimated 30% of community-acquired AKI episodes — every CKD patient should have a written plan. Hepatitis B vaccination efficiency drops dramatically once on dialysis (response rate falls from 70% to 50–60%) — pre-emptive vaccination is critical. Exercise in CKD: the STEPPING study showed structured exercise slows eGFR decline by 2 ml/min/yr compared to sedentary controls — the mechanism involves improved glycaemic control, BP reduction, and anti-inflammatory effects.

Plotting eGFR as a trend line over time is more informative than any single value — a patient with eGFR 38 but stable for 5 years is very different from one with eGFR 38 that has fallen from 60 in 2 years. The latter requires urgent nephrology referral and investigation. NICE NG203 introduced QOF indicator CKD006 (annual review with ACR, eGFR, and BP) — completing this for all CKD patients is both a quality requirement and the mechanism by which deterioration is detected early. The 1–2 week U&E check after ACEi/ARB initiation prevents the rare but serious complication of undetected severe hyperkalaemia — a potassium ≥6.5 mmol/L within the first week of ACEi treatment requires immediate same-day management. K+ home monitoring is not routinely available but should be discussed for patients on finerenone + ACEi. Advance care planning documentation from G4 onwards respects patient autonomy and prevents unwanted interventions near end-of-life — conservative kidney management (without dialysis) is a valid, NICE-endorsed option for frail elderly patients.