NICE vs SIGN: Management of Diabetic Ketoacidosis (2025)
This guideline provides a comparative summary of the National Institute for Health and Care Excellence (NICE) NG29 (updated 2021, with 2025 amendments under review) and the Scottish Intercollegiate Guidelines Network (SIGN) Guideline 168 (2024) for the management of Diabetic Ketoacidosis (DKA) in adults. It is intended for clinicians in the UK to understand the nuances and key differences between these two authoritative sources, facilitating informed, guideline-concordant care across different healthcare settings.
See how this translates to practice: Explore our Clinical governance features, visit the Patient Safety Hub, or review Clinical Safety & Assurance for enterprise rollout.
Diagnosis and Assessment
Both guidelines agree on the core diagnostic triad for DKA but differ in their specific biochemical thresholds and initial assessment priorities.
NICE NG29
- Diagnostic Criteria: Hyperglycaemia (blood glucose >11 mmol/L) or known diabetes mellitus, with:
- Ketonaemia ≥3.0 mmol/L or significant ketonuria (≥2+ on standard urine sticks).
- Acidaemia (venous pH <7.3) and/or bicarbonate (HCO₃⁻) <15.0 mmol/L.
- Assessment Emphasis: Rapid assessment of Airway, Breathing, Circulation, Disability (AVPU). Stresses the use of bedside blood ketone testing (beta-hydroxybutyrate) as the primary method for diagnosing and monitoring DKA, as it is more specific and accurate than urine ketone testing.
SIGN 168
- Diagnostic Criteria: A more traditional, stringent triad requiring all of the following:
- Hyperglycaemia (blood glucose >11 mmol/L).
- Ketonaemia ≥3.0 mmol/L or ketonuria ≥2+.
- Venous bicarbonate (HCO₃⁻) <15 mmol/L and/or venous pH <7.3.
- Assessment Emphasis: Also prioritises ABCDE approach. Strongly recommends venous blood gas (VBG) analysis as the initial investigation for all suspected cases, as it provides immediate data on pH, bicarbonate, potassium, and anion gap. Agrees on the superiority of blood ketone measurement but provides more detail on using the anion gap in scenarios where ketone testing is unavailable.
Key Difference: NICE allows for diagnosis in a known diabetic with significant ketosis and acidaemia, even if blood glucose is not markedly elevated (>11 mmol/L), acknowledging the phenomenon of "euglycaemic DKA". SIGN’s criteria are more classical, requiring marked hyperglycaemia. Both prioritise blood ketones over urine testing.
Treatment Protocol
The overall goals are identical: correct dehydration, hyperglycaemia, ketosis, and electrolyte imbalances. The differences lie in fluid choice, insulin dosing, and potassium replacement thresholds.
Fluid Resuscitation
- NICE: Recommends using 0.9% sodium chloride solution exclusively for the initial resuscitation. A fixed volume of 1 litre is given over the first hour.
- SIGN: Also recommends 0.9% sodium chloride but provides a weight-based fluid bolus of 10-15 mL/kg (typically 500-1000 mL) over the first hour. This may allow for more individualised dosing, particularly in patients at risk of fluid overload.
Insulin Therapy
- NICE: Recomments a fixed-rate intravenous insulin infusion (FRIII) of 0.1 units/kg/hour. Emphasises that the FRIII should not be diluted in the infusion bag but administered separately via a syringe pump.
- SIGN: Identical recommendation for a FRIII of 0.1 units/kg/hour. Provides additional guidance to consider a higher rate (e.g., 0.15 units/kg/hour) in severely insulin-resistant patients if response is inadequate, and to continue the FRIII until ketosis has cleared (blood ketones <0.6 mmol/L), not just until the anion gap is closed.
Potassium Replacement
- NICE: States that if the serum potassium is below 5.5 mmol/L, potassium chloride should be added to each litre of fluid, with a recommended concentration of 40 mmol/L.
- SIGN: Uses a more nuanced, tiered approach based on VBG results:
- K⁺ <3.5 mmol/L - Hold insulin, give potassium aggressively (e.g., 20-40 mmol/hour), and seek senior advice.
- K⁺ 3.5–5.5 mmol/L - Add 40 mmol/L to each bag.
- K⁺ >5.5 mmol/L - Do not add potassium, but recheck every 2 hours.
Key Difference: The most significant practical difference is in potassium management. SIGN's tiered protocol provides clearer, more actionable guidance for hypokalaemia, a critical risk during treatment. SIGN also offers more flexibility in initial fluid volume and insulin dosing in special situations.
Special Situations
Euglycaemic DKA
- NICE: Explicitly includes this in its diagnostic criteria. Management is the same as for hyperglycaemic DKA, but highlights the importance of using ketone levels, not glucose, to guide therapy.
- SIGN: Discusses euglycaemic DKA as a specific entity, often associated with SGLT2 inhibitor use, prolonged fasting, or pregnancy. Management is similar, but cautions against using glucose-containing fluids too early.
SGLT2 Inhibitor-Associated DKA
- Both guidelines strongly advise immediate and permanent discontinuation of the SGLT2 inhibitor during the acute episode.
- SIGN provides more detailed advice on patient education upon discharge regarding sick-day rules and the risks of re-initiating this drug class without careful consideration.
Management of Cerebral Oedema (Rare in Adults)
- NICE: Briefly mentions it as a complication, advising immediate senior review and consideration of mannitol or hypertonic saline.
- SIGN: Offers a more structured approach, including immediate neuroimaging, raising the head of the bed, and specific protocols for hypertonic (3%) saline administration.
Practical Clinical Flow: A Synthesis
For a UK clinician, a practical synthesis of both guidelines would be:
- ABCDE Assessment: Immediate stabilisation. Attach monitoring.
- Diagnostic Tests: Take VBG (pH, HCO₃⁻, K⁺), bedside blood ketones, U&E, FBC, VBG. Use SIGN's criteria but be aware of NICE's euglycaemic DKA caveat.
- Hour 1: Give 0.9% NaCl (NICE: 1L; SIGN: 10-15 mL/kg). Start FRIII at 0.1 units/kg/hr via syringe pump. Commence potassium replacement as per SIGN's tiered protocol if K⁺ <5.5 mmol/L.
- Ongoing Management (Hours 1-6): Continue 0.9% NaCl with K⁺ added, adjusting rate based on clinical status. Monitor capillary glucose, VBG (pH, HCO₃⁻, K⁺), and blood ketones hourly.
- Switching Fluids & Monitoring Resolution: When blood glucose falls below 14 mmol/L, add 10% glucose 125 mL/hour to the IV fluids (or bag with glucose 10% and NaCl 0.45%/0.9%). Continue FRIII until blood ketones <0.6 mmol/L and VBG pH >7.3 and/or HCO₃⁻ >18 mmol/L.
- Transition to SC Insulin: Once DKA resolved and patient eating/drinking, overlap SC insulin with IV insulin for at least 30-60 minutes before stopping the infusion.
Frequently Asked Questions (FAQs)
1. Which guideline should I follow in England? In Scotland?
In England and Wales, NICE is the standard. In Scotland, SIGN is the standard. However, the guidelines are largely aligned. The most pragmatic approach is to adopt the strongest recommendations from both: use NICE's emphasis on blood ketone monitoring and incorporate SIGN's more detailed potassium replacement protocol.
2. What is the single most important difference in practical management?
The potassium replacement protocol. SIGN's tiered approach (especially the guidance for severe hypokalaemia <3.5 mmol/L) is more explicit and safety-focused, reducing the risk of dangerous hypokalaemia during insulin infusion.
3. How should we monitor for resolution of DKA?
Both guidelines agree that blood ketone levels are the superior marker. Resolution is indicated by a blood ketone level <0.6 mmol/L, alongside a normalising venous pH (>7.3) and/or bicarbonate (>18 mmol/L). Do not use blood glucose alone to judge resolution.
4. What about patients on SGLT2 inhibitors?
Both guidelines mandate stopping the drug. Be highly vigilant for euglycaemic DKA, as glucose levels may be only mildly elevated. The diagnosis rests on high ketones and metabolic acidosis.
5. When should a variable-rate insulin infusion (VRIII or "sliding scale") be used?
Never as initial therapy for DKA. A FRIII is essential to suppress ketogenesis effectively. A VRIII may be considered only after DKA has fully resolved if ongoing IV insulin is needed for glycaemic control before SC insulin can be safely re-established.
Source Links
- NICE Guideline NG29 (Diabetes (type 1 and type 2) in children and young people: diagnosis and management) - The DKA section is within this broader guideline. [Last updated: September 2021, with 2025 update in progress].
NICE NG29 - SIGN Guideline 168 (Management of Diabetic Ketoacidosis) - A dedicated DKA guideline for adults.
SIGN 168 DKA (PDF)
Disclaimer: This comparison is for informational purposes within a clinical context. Always refer to the full, original guidelines for comprehensive detail and the most up-to-date information. Local hospital trust policies may also provide specific operational protocols.