I now have sufficient clinical information from the textbooks. Let me synthesize a comprehensive management review for this critically ill patient.
Clinical Review: DKA + AKI-on-CKD, Post-HDF, Ventilated
Current Status Summary
The patient has progressed significantly - ketosis cleared, mild residual acidosis, euglycaemia achieved, on HDF, ventilated, GCS 12/15, but critically anuric/severely oliguric (20 mL/hr x 2 days) which is the dominant ongoing problem.
1. DKA Resolution - Where Things Stand
Based on standard criteria, DKA is considered resolved when all three are met:
- Blood glucose < 11.1 mmol/L (200 mg/dL) ✓ (sugars ok)
- Serum bicarbonate ≥ 15 mmol/L (need to confirm - acidosis "mild")
- Venous pH > 7.3 (need to confirm)
- Anion gap normalised
- Ketonaemia resolved (beta-hydroxybutyrate < 1.0 mmol/L) ✓
The mild residual acidosis is likely a hyperchloraemic normal-gap metabolic acidosis - this is expected and normal. As the [Comprehensive Clinical Nephrology] explains: "During treatment, the anion gap metabolic acidosis transforms once again into a normal gap acidosis... The reversal of the hyperchloraemic acidosis takes several days as the HCO3- deficit is corrected by the kidney." Do not over-treat this with bicarbonate.
In this anuric patient on HDF, the kidneys cannot correct this acidosis independently - HDF is doing the work. Mild residual acidosis in this context is acceptable and expected.
2. Transition from IV Insulin (Critical Step)
Since DKA has essentially resolved, transition protocol:
- Continue IV insulin infusion until 1-2 hours after the first subcutaneous dose is given - do NOT stop IV insulin abruptly
- Start long-acting SC insulin (or patient's pre-admission regimen) before stopping IV infusion
- Target blood glucose 6-10 mmol/L (110-180 mg/dL) in ICU
- Risk of rebound ketosis is low now but monitoring continues - check beta-hydroxybutyrate or urine ketones 4-6 hourly for 24 hours post-transition
- Important in AKI/ESRD: Insulin half-life is prolonged - reduce doses, risk of hypoglycaemia is high. Monitor glucose hourly initially.
3. Fluid & Renal Status - Most Critical Issue
Urine output 20 mL/hr (480 mL/day) x 2 days = established oliguria, on background CKD
This patient is likely in AKI stage 3 on CKD (near-anuria/oliguria despite 2 days of HDF). Key questions:
Is this pre-renal, intrinsic, or mixed?
| Factor | Relevance |
|---|
| DKA-induced osmotic diuresis → severe dehydration | Initially pre-renal component |
| CKD background | Reduced renal reserve |
| Osmotic tubular injury from glucosuria | Intrinsic component |
| Ventilation (high PEEP → reduced renal perfusion) | Potentially contributory |
| Sepsis as DKA precipitant? | Must rule out |
Practical steps:
- Assess volume status - don't assume ongoing dehydration if on HDF x 2 days; patient may now be euvolaemic or fluid overloaded (especially ventilated)
- Target mean arterial pressure (MAP) ≥ 65 mmHg (higher if CKD baseline, aim ≥ 70-75)
- Avoid nephrotoxins: NSAIDs, aminoglycosides, contrast, ACE inhibitors/ARBs (hold until recovery)
- If not already checked: bladder scan to exclude obstruction; renal USS
4. HDF (Hemodiafiltration) - Ongoing Management
The patient has been on HDF (continuous/intermittent hemodiafiltration) - this combines diffusion (hemodialysis) and convection (hemofiltration), providing excellent clearance of both small and medium-weight solutes.
From [Fischer's Mastery of Surgery]: "CRRT is usually seen in patients with hemodynamic instability... CRRT allows for fewer hemodynamic changes because the rates of ultrafiltration are slower. There are also fewer abrupt changes in concentrations of solutes as well."
For this ventilated, likely haemodynamically borderline patient:
- Continue HDF/CRRT - preferable to intermittent HD given ventilation and cardiovascular stress
- Adjust net fluid balance targets based on clinical examination (jugular venous pressure, lung auscultation, chest X-ray, lung ultrasound, echocardiography)
- Target neutral to mildly negative fluid balance if euvolaemic; negative balance if fluid overloaded
- Electrolyte monitoring during HDF - potassium, phosphate, magnesium, bicarbonate every 6-12 hours
- Bicarbonate-buffered replacement fluid (preferred over lactate-buffered) to assist acidosis correction
- Watch for HDF-related hypoglycaemia - glucose-containing or glucose-monitored replacement fluids
Phosphate:
Large DKA-associated phosphate deficits become apparent during treatment. Phosphate replacement may need to be added to HDF replacement fluid or given IV (hypophosphataemia can cause respiratory muscle weakness - relevant in ventilated patient).
5. Ventilation Considerations
GCS 12/15 - not fully obtunded. Possible causes:
- Residual metabolic encephalopathy from acidosis/uraemia
- Sedation for ventilation
- Cerebral oedema (rare in adults with DKA, more common if glucose corrected too rapidly or aggressive hypotonic fluid given)
Ventilator settings:
- Avoid aggressive hyperventilation to compensate for metabolic acidosis - once ketosis is cleared, this is no longer needed and auto-PEEP/air trapping risk increases
- Target pH ≥ 7.25-7.30 with HDF doing the primary work for acidosis correction
- Aim for lung-protective ventilation (Vt 6 mL/kg IBW, plateau pressure < 30 cmH2O)
- Daily spontaneous breathing trials when metabolic and haemodynamic stability allows
- Sedation hold daily - GCS 12 may allow progressive reduction of sedation
Weaning readiness checklist:
6. Electrolytes - Ongoing Monitoring
These are particularly dynamic in DKA recovery + HDF:
| Electrolyte | Issue | Action |
|---|
| Potassium | Risk of hypokalaemia during insulin therapy; HDF removes K+ | Check every 4-6 hrs; supplement via HDF fluid or IV |
| Phosphate | Total body deficit despite normal/high serum levels at presentation | Replace if < 1.0 mg/dL; watch for refeeding-type syndrome |
| Magnesium | Depleted in DKA | Target > 0.8 mmol/L |
| Sodium | Corrected slowly; osmotic demyelination risk if corrected too fast | Correct by < 10 mmol/L per 24 hours |
| Bicarbonate | Mild deficit expected; do NOT push bicarb aggressively | HDF will correct this over days |
7. Precipitant - Have You Found It?
DKA in a known CKD patient on a ventilator suggests a serious precipitant. Common ones per [Goldman-Cecil]:
- Infection (most common) - check blood/urine/sputum cultures, procalcitonin, CRP
- Acute coronary syndrome - ECG, troponin
- Inadequate insulin or SGLT2 inhibitor use (euglycaemic DKA risk)
If a precipitant (sepsis, pneumonia) is driving the clinical picture, treating the underlying cause is as important as the DKA/AKI management.
8. Prognosis Markers
A large 2026 retrospective study (PMID
40550726) covering 464,057 US hospitalisations found AKI complicating DKA is associated with substantially higher morbidity and healthcare costs. Persistent oliguria at 48 hours despite renal replacement therapy in a CKD patient is a poor prognostic sign for renal recovery - early nephrology involvement is essential.
Summary Priority Actions
- Confirm DKA resolution biochemically (pH, bicarbonate, anion gap, beta-hydroxybutyrate) and transition IV insulin to SC with overlap
- Do not over-treat the mild residual acidosis - it is expected hyperchloraemic, HDF will correct it
- Continue HDF (CRRT-mode preferred) with target neutral-to-negative fluid balance, bicarbonate-buffered replacement fluid
- Electrolyte monitoring every 6 hours - focus on K+, PO4, Mg2+
- Assess and manage underlying precipitant aggressively
- Ventilator weaning plan: daily sedation holds, SBTs when ready, avoid compensatory hyperventilation now that ketosis has cleared
- Renal prognosis discussion: sustained oliguria/near-anuria on background CKD on day 2 of RRT - prepare family; chance of renal recovery is uncertain
- Nutrition: initiate enteral nutrition (NG/NJ), account for renal and glucose targets; avoid parenteral nutrition unless enteral route unavailable
- Comprehensive Clinical Nephrology 7th Ed., p. 3496-3508
- Goldman-Cecil Medicine, p. 2484-2485 (DKA treatment protocol)
- Fischer's Mastery of Surgery 8th Ed., p. 228 (CRRT vs iHD)