Insulin Drip to Subcutaneous Transition Calculator
Medically reviewed by Dr. Chhaya Makhija, MD, DipABLM — September 2026
Estimate the starting subcutaneous TDD when transitioning a patient from IV insulin infusion to a basal-bolus regimen. Based on the infusion rate averaged over the prior 6–8 hours. Clinical supervision required.
🔁 IV → SubQ Transition Estimator
Subcutaneous Starting Regimen Estimate
(50% of TDD)
(50% of TDD)
(÷ 3 meals)
units/day
Transition Protocol Overview
Standard IV → SubQ Steps
- Calculate IV TDD: average rate (units/hr) × 24
- Apply reduction factor (typically 70–80% of IV TDD)
- Split into 50% basal (long-acting) + 50% bolus (rapid-acting)
- Give first long-acting dose 1–2 hours before stopping IV infusion
- Divide bolus pool across meals (3 equal doses as starting point)
- Monitor closely for 24–48 hours after transition
Overlap the injections. Long-acting insulin takes 1–4 hours to reach meaningful activity. If the IV is stopped simultaneously with the first basal injection, a dangerous hyperglycemic gap can occur. Always administer the first basal dose while the IV is still running, then stop the IV 1–2 hours later per institutional protocol.
Why the timing of the switch matters
Intravenous insulin clears the bloodstream within minutes of the drip being switched off. Subcutaneous insulin, by contrast, takes time to absorb and start working — long-acting insulin may need an hour or two to kick in. If you simply stop the infusion and give the first injection at the same moment, you create an uncovered gap where no effective insulin is present, and glucose can rebound sharply. In ketoacidosis, that gap can even let ketones return.
The fix is overlap. The first dose of long-acting subcutaneous insulin is given before the drip is stopped — commonly one to two hours ahead — so the new insulin is already active when the infusion ends. Getting this hand-off right is the single most important safety step in the transition.
How the subcutaneous dose is estimated from the drip
The patient's recent infusion rate is the best clue to their true daily insulin requirement, because it reflects what actually controlled their glucose. A common approach is to take the average hourly rate over the last stable several hours, project it across 24 hours to estimate the total daily dose, and then deliberately give only a portion of that — often around 75–80% — as the new subcutaneous regimen.
Why clinicians use only part of the calculated dose
Scaling back provides a safety margin. Stress, illness, and IV dextrose can all inflate the drip requirement, so the full projected dose may be more than the patient needs once they're recovering and eating. Starting slightly conservative and adjusting upward is far safer than overshooting into hypoglycemia.
How much of the IV dose carries over: what the trials found
Published protocols disagree on two things — how many hours of infusion you average, and what share of that projected dose you give subcutaneously. The disagreement is not sloppiness. Each protocol was built for a different patient.
| Protocol | Averaging window | Share of IV-projected TDD | First-day split |
|---|---|---|---|
| Schmeltz 2006 (randomised trial) | Final 6 h | 80% — beat 40% and 60% | Glargine only; rapid-acting added once eating |
| Avanzini 2011 (DDD protocol) | Final 12 h | 100% — but only when IV dextrose matched the oral carbohydrate load and glucose held 100–139 mg/dL for 24 h | 50% basal / 50% prandial |
| Bode & Braithwaite 2004 | Final 6–8 h | ~80% | Basal, plus prandial once eating |
Schmeltz randomised patients to 40%, 60% and 80% and found 80% controlled glucose best over the first 24 hours. That trial remains the strongest single piece of evidence for the number this calculator defaults to. Avanzini's group carried over the full 100% and still kept roughly half of all readings inside a tight target — but their patients ate a measured carbohydrate load that matched the IV dextrose they had been receiving, so the projection described a real steady state rather than an acute-stress requirement.
The practical reading: use 80% when the patient's intake after the switch is uncertain. A higher fraction fits only when carbohydrate intake is known and glucose has already been stable for a day.
Splitting the mealtime dose: equal thirds is a simplification
Most quick references divide the bolus pool into three equal meal doses. The DDD protocol splits it 20% at breakfast, 40% at lunch, 40% at dinner, matching how carbohydrate actually distributes across a day. Morning insulin resistance pushes the other way, so local protocol decides this. Equal thirds is a starting point, not a finding.
Who has a rough transition, and who should not transition yet
Avanzini's team tracked which patients lost glucose control on day one. Three variables predicted a poor transition:
| Predictor | Threshold | Odds of a poor first day |
|---|---|---|
| Age | 72 years or older | 2.29× |
| Mean IV insulin rate over the prior 24 h | 1.6 units/h or higher | 2.20× |
| Glucose variability (coefficient of variation) | 11.9% or higher | 1.75× |
All three describe diabetes that was already hard to control on the drip. None of them blocks a transition. They tell the team to monitor harder and titrate sooner.
The same protocol also set a floor. Patients whose glucose sat in target on less than 0.5 units/hour — roughly a projected 12 units a day — were not transitioned to a scheduled subcutaneous regimen at all. Below that, the arithmetic produces doses too small to be meaningful.
What the first three days actually look like
Hypoglycaemia is common after the switch, and honest numbers help teams plan for it. In the Avanzini cohort, 7.7% of patients had a reading below 70 mg/dL on day one, rising to 26.8% across the first three days. Nearly three-quarters of those episodes sat between 60 and 69 mg/dL and none produced symptoms. Doses were then adjusted by 10–20% every one to two days against pre-meal and post-meal readings.
One more finding deserves attention: the protocol was mis-applied in 16% of patients, and the errors were mostly arithmetic. That is the argument for computing the dose rather than doing it in your head at handover.
Avoiding rebound high blood sugar after the switch
Even with good timing and dosing, glucose needs close monitoring for the first day after the transition. Frequent checks catch an early rebound while it's still easy to correct, and they confirm the new basal and mealtime doses are holding. This whole process is an inpatient, clinician-led procedure — the numbers here are an educational illustration of the principles, not a protocol to follow on your own.
Frequently Asked Questions
How do you convert an IV insulin drip to subcutaneous insulin?
Estimate the 24-hour IV requirement from a stable infusion period, multiply by a reduction factor of about 70–80%, then split the result roughly 50% basal (long-acting) and 50% bolus (rapid-acting) divided across meals.
What is the 2/3 – 1/3 rule for insulin?
The 2/3 – 1/3 rule splits a twice-daily NPH and regular regimen: two-thirds of the total daily dose goes before breakfast, one-third before dinner. Each of those doses splits again, two-thirds NPH and one-third regular. It belongs to split-mixed regimens, not to the basal-bolus regimen used after an IV transition, and it carries a higher hypoglycaemia risk because NPH peaks.
How is the dose adjusted after the first day?
Teams titrate the basal and mealtime doses by 10–20% every one to two days. Pre-meal readings drive the basal adjustment and post-meal readings drive the mealtime doses. Expect changes: in one published cohort the median total daily dose rose from 36 units on day one to 40 units by day three.
Why multiply by 70–80% instead of the full IV dose?
IV requirements often reflect acute stress, and subcutaneous insulin has different pharmacokinetics. Applying a reduction factor (commonly 80%) lowers the risk of hypoglycemia as the patient stabilizes. The exact factor is set by clinical judgment and protocol.
When should the first long-acting dose be given?
Give the first basal (long-acting) injection 1–2 hours before stopping the IV infusion. Long-acting insulin needs 1–4 hours to take effect, so overlapping prevents a hyperglycemic gap when the drip is discontinued.
Is this transition done at home?
No. The IV-to-subcutaneous transition happens in the hospital under a protocol with close glucose monitoring for 24–48 hours. This calculator is an educational reference for clinicians and students.
Sources
- ADA/AACE Inpatient Glycemic Control Consensus Statement. Diabetes Care. 2009.
- Avanzini F, Marelli G, Donzelli W, et al. "Transition From Intravenous to Subcutaneous Insulin: effectiveness and safety of a standardized protocol and predictors of outcome." Diabetes Care. 2011;34(7):1445–1450. Link
- Schmeltz LR, DeSantis AJ, Schmidt K, et al. "Conversion of intravenous insulin infusions to subcutaneously administered insulin glargine in patients with hyperglycemia." Endocr Pract. 2006;12(6):641–650. Link
- Bode BW, Braithwaite SS, Steed RD, Davidson PC. "Intravenous insulin infusion therapy: indications, methods, and transition to subcutaneous insulin therapy." Endocr Pract. 2004;10(Suppl 2):71–80.
- Umpierrez GE et al. "Management of Hyperglycemia in Hospitalized Patients." J Clin Endocrinol Metab. 2012. Link
Written by Ryan Mitchell · Last updated: September 2026