Two bags of fluid run at the same time into the same line. They are identical except for dextrose — same saline, same potassium. One has none, one has dextrose.

Because only the dextrose differs, sliding volume from one bag to the other changes the sugar the child receives and nothing else. The total fluid, the sodium and the potassium all stay exactly the same. That is the whole point of the method: you can chase the glucose without disturbing the rehydration.

A third pump carries the insulin. It runs at a steady dose and is not used to control the glucose — the dextrose slider does that.

Work left to right through the tabs: confirm the diagnosis, set up your fluids, run the bags, then log each set of bloods and let the tool tell you what to change.

Enter the weight in the header — nothing calculates without it — then the age below. Everything else on this page is optional, but the more you enter the more the tool can check.

These change every number further down. Set them to what your unit actually stocks before you read a rate.

Protocols differ here. Agree it locally before the tool is used on a patient.

The potassium must be identical in both bags. That is what lets the ratio move dextrose without moving anything else.

Start 1 to 2 hours after fluids begin, once potassium is at least 3.5. The infusion corrects the acidosis, not the glucose — do not reduce it to manage a falling glucose while the patient is still acidotic.

All fluid rates, deficit replacement and insulin infusion rates require the patient weight. Enter it above, or pick a preset to start.

The shared rate never changes when you move this — only the proportion of it that carries dextrose. Sodium and potassium delivery are unaffected.

Use Add New Lab Values for each round of bloods. It asks only for what changes, updates the patient, logs the set, and works out the rate of fall and what to do about it. The timestamp is editable afterwards if the sample was drawn earlier than it was entered.

Calculate the anion gap each time, and follow β-hydroxybutyrate where available. Urine ketones stay positive for days after the acidosis clears and must not be used to judge resolution.

• Criteria: glucose > 33.3 mmol/L, effective osmolality > 320 mOsm/kg, HCO₃ > 15 mmol/L, pH > 7.25
• Picture: more dehydrated, severe electrolyte disturbance, minimal acidosis or ketosis, negative to trace ketones
• Management: these patients need more aggressive fluid resuscitation than DKA, not less
• Action: discuss with the paediatric referral centre promptly. A mixed picture is possible.

1. Set your centre's fluids once. Open Settings and choose the dextrose bag you stock, the saline base, the potassium, the bag volumes and the rate ceiling. It is remembered on this device and you will not be asked again.
2. Type the weight. Every rate comes from it. Nothing else is needed to get the two bag rates.
3. Check the phase. The tool opens on Pre-Insulin — the first one to two hours, bag A carrying the whole rate and bag B at a to-keep-open trickle. Switch to Insulin Active the moment the insulin starts.
4. Read the two pump numbers. Pump A and Pump B are the mL/h you program. They always add up to the shared rate above them.
5. Move the slider to change dextrose. It shifts volume between the bags and changes nothing else — same total, same sodium, same potassium. Move about 20% per hour and watch the response.
6. Read the red boxes. They are the limits that should not be crossed. Amber is one you are near.
7. Copy the handover before you hand the patient over.

The large figure is the dose, in units per hour. The figure beside it is the pump setting, in millilitres per hour of fluid.

They differ because bag C is dilute. At 0.1 units/mL there are ten millilitres of fluid for every unit, so a 140 kg patient on 0.1 units/kg/h receives 14 units per hour while the pump reads 140 mL per hour. At 1 unit/mL the same 14 units per hour runs at 14 mL per hour.

In a large adolescent the dilute preparation can consume a quarter or more of the entire fluid order. The tool says so when it does.

• Start it when glucose reaches 15–17 mmol/L, or is falling faster than 5 mmol/L/h once insulin has begun.
• Aim for glucose 8–12 mmol/L and a fall of 2–5 mmol/L/h.
• Falling too fast? Raise dextrose first. Hold the insulin rate steady until the pH is close to normal.
• Still low with bag B at 100%? In order: cut insulin by about 25% if the acidosis is correcting; then a higher dextrose bag; then, with intensive care and central access, D20.
• Never stop the insulin infusion to fix a low glucose. Its half-life is minutes and the ketosis returns.

The thresholds here follow the Canadian Paediatric Society practice point on paediatric diabetic ketoacidosis, and through it the CPEG-endorsed TREKK algorithm and the ISPAD 2022 consensus guidelines beneath it. They are not specific to any one hospital — the same figures appear in provincial and institutional protocols across Canada.

• Canadian Paediatric Society. Current recommendations for the management of paediatric diabetic ketoacidosis. Practice point, Acute Care Committee. Source of the cerebral injury risk factors, the cumulative bolus ceiling, the corrected sodium and anion gap definitions, and the disposition criteria.
• TREKK — Translating Emergency Knowledge for Kids, endorsed by the Canadian Pediatric Endocrine Group. Paediatric DKA algorithm, updated June 2023 to ISPAD 2022. Source of the severity table, the weight bands, the dextrose trigger and the 500 mL/h ongoing maximum.
• ISPAD. Glaser N, et al. Clinical Practice Consensus Guidelines 2022: Diabetic ketoacidosis and hyperglycemic hyperosmolar state. Pediatr Diabetes 2022;23(7):835–56.
• Grimberg A, Cerri RW, Satin-Smith M, Cohen P. The two-bag system for variable intravenous dextrose and fluid administration: benefits in diabetic ketoacidosis management. J Pediatr 1999;134(3):376–8. The method itself is theirs, and is in routine use internationally.
• Kuppermann N, et al; PECARN DKA FLUID Study Group. Clinical trial of fluid infusion rates for pediatric diabetic ketoacidosis. N Engl J Med 2018;378(24):2275–87.

Most of this is settled. These points are not, so the tool lets you set them rather than choosing silently.

• Maximum ongoing rate. TREKK states 500 mL/h and describes it as an increase on earlier figures. Some rehydration tables print 250 mL/h for the same band. The tool defaults to 500 and offers 250. At 100 kg the two differ by 150 mL/h in the bags, so agree it locally rather than at the bedside.
• Corrected sodium. Written as Na + 0.3 × (glucose − 5) and as Na + 0.36 × (glucose − 5.6). Both are shown, and the tool warns when they land on opposite sides of the 140 mmol/L switch to half-normal saline.
• Anion gap. Usually without potassium, where above 12 is diagnostic; some flowsheets include it, where 16 ± 2 is normal. Both are shown.
• Potassium. Added when measured potassium is below 5.0 to 5.5, and always only once urine output is documented.
• Phosphate. Replacement thresholds of 0.32, 0.4 and 0.5 mmol/L all appear. The tool prompts below 0.5 and flags below 0.4.
• Glucose target. Stated as 7–11, 8–12 and 8–15 mmol/L. The tool uses 8–12, which sits inside all of them.

This is a single file with nothing loaded from the network — no fonts, no scripts, no analytics, no tracking. Once you have the file it works with aircraft mode on.

• Phone: open the file, then Share → Add to Home Screen. It opens like an app.
• Computer: save it to the desktop and double-click.
• Ward: put it on a shared drive or email it around. It is one file.

• Nothing is transmitted. No network requests of any kind.
• No identifiers are collected. The tool asks for age and weight and nothing else — no name, no date of birth, no chart number.
• Data stays until you clear it. A DKA admission runs 24 to 48 hours, so the working set survives a reload, a flat battery and a shift change. Reset Patient removes it.
• Clear it before handing the device on. Nothing is deleted automatically.
• Device settings persist — bag concentrations, potassium, ceiling, local protocol. They carry no patient data and save you setup time.
• No cookies, no identifiers, no analytics.

Scope. Paediatric diabetic ketoacidosis in patients under 18 years. It is not validated for adults, for hyperosmolar hyperglycaemic state, or for any other condition.

Intended user. A regulated health professional already trained in the management of paediatric DKA, working within their own institution’s protocol.

Responsibility. This is a reference aid. It performs arithmetic on figures you enter and compares them with published thresholds. It does not examine the patient, does not diagnose, and does not decide anything. Clinical judgment, verification of every figure, and responsibility for the care given rest entirely with the treating clinician. Any error in entry, interpretation or application is the responsibility of the operator.

It has not been reviewed or endorsed by the Canadian Paediatric Society, the Canadian Pediatric Endocrine Group, TREKK, ISPAD or any other body.

Guidelines change. Confirm every figure against a current source before acting on it. Provided without warranty of any kind, express or implied; the author disclaims any liability, loss or damage arising from its use.