A 29-year-old schoolteacher from Brahmavar is 28 weeks into her first pregnancy. Her DIPSI test at 24 weeks was 162 mg/dL, and she has followed the dietitian’s advice for two weeks. Her glucometer diary now shows fasting readings of 100 to 104 and 2-hour post-breakfast readings of about 170. She weighs 64 kg. The obstetrician asks you to start insulin. This is an illustrative case, not a real patient.
“Which insulin, how much, and what are we aiming for?” By the end of this article you will be able to answer all three, and you will know what changes when she gets steroids, goes into labour, and delivers.
Use this at the bedside
Open the tool: Insulin in Pregnancy
Get the CDSS app: install the MEDiscuss CDSS on your phone
1. The targets, and why three documents give three sets
Start with the numbers we are aiming for. Three documents set them, and they agree more than they differ. Read across each row for the reading in front of you.
| Reading | ADA and ACOG | NICE NG3 | India, MoHFW 2018 |
|---|---|---|---|
| Fasting and pre-meal | 95 or below | Below 95 (5.3 mmol/L) | Below 95 |
| 1 hour after a meal | 140 or below | Below 140 (7.8) | Not used |
| 2 hours after a meal | 120 or below | Below 115 (6.4) | Below 120 |
| In labour | Below 110 (ACOG) | 72 to 126 (4 to 7) | 90 to 120 on the saline table |
| Overnight | Not below 60 (ACOG) | Above 72 (4.0) at all times | Not stated |
“Why does NICE say 115 at 2 hours, when ADA says 120?” NICE writes its targets in mmol/L, and its 2-hour figure of 6.4 mmol/L is 115 mg/dL, a little tighter than the ADA figure. The difference is small. What matters more is that the Indian national protocol (MoHFW 2018) uses the same fasting and 2-hour figures as ADA and ACOG, and does not use the 1-hour reading at all.
The tool titrates to the ADA and ACOG figures. In labour it prints the NICE and JBDS (Joint British Diabetes Societies) band of 72 to 126 mg/dL, because the infusion triggers are written against it, and names the ACOG band beside it.
Our schoolteacher’s fasting readings of 100 to 104 are above 95. Her post-breakfast reading of 170 is above 120. Both need treatment.
2. Two ways to start insulin in gestational diabetes
The first way is the national protocol. After two weeks of medical nutrition therapy, a 2-hour post-meal reading of 120 or above starts treatment. The protocol uses Human Premix 30/70, one dose 30 minutes before breakfast, sized by the post-meal reading: 4 units for 120 to 160, 6 units for 160 to 200, 8 units above 200. Then add 2 units every third day: before breakfast if the post-meal reading is high, before dinner if the fasting reading is high. Metformin 500 mg twice daily, up to 2 g, is permitted from 20 weeks. A woman needing more than 20 units a day, or more than 2 g of Metformin, is referred to a higher centre.
The second way is weight-based, from ACOG Practice Bulletin 190 and the ADA. Each high reading is matched to the insulin that controls it. A high fasting reading gets bedtime NPH at 0.1 to 0.2 U/kg. A high post-meal reading gets Regular insulin or a rapid analogue before that meal, at 1 unit per 10 to 15 g of carbohydrate, or 2 to 4 units to start.
“So which is right?” Both are. The premix start is simple and cheap, and it works well for the woman whose post-breakfast reading is the problem. It handles the fasting reading badly. Think of the intermediate part of the morning premix as a tank that runs dry by 3 a.m. Nothing is left to hold the glucose down before dawn. A fasting glucose that stays high on a morning premix needs the evening dose the protocol adds, or a bedtime NPH.
For our patient, the post-breakfast reading of 170 falls in the 160 to 200 band: 6 units of premix before breakfast. Her fasting reading is also high, so expect to add an evening dose. On the weight-based route, her bedtime NPH would start at 0.1 to 0.2 U/kg, which is 6.4 to 12.8 units for her 64 kg.
3. Pre-existing diabetes: the requirement climbs
A woman who had diabetes before pregnancy is a different problem. Insulin resistance rises as the placenta grows. ACOG Practice Bulletin 201 sizes the total daily dose by trimester: 0.7 to 0.8 U/kg/day in the first, 0.8 to 1.0 in the second, 0.9 to 1.2 in the third, about half basal and half prandial. The dose that was right at 20 weeks is short at 30.
A common mistake, and it is easy to see why: a fall in insulin requirement after 36 weeks looks like good news. It is not. It is one of the signs of placental insufficiency, and it needs fetal surveillance, not just a smaller dose.
Which insulins? NPH, Regular, Aspart, Lispro and Detemir have pregnancy data. Glargine U-100 has observational data in several hundred pregnancies with no signal of harm, and the ADA does not require a woman stable on it to change. Oral agents other than Metformin stop; Glibenclamide crosses the placenta and gives more neonatal hypoglycaemia and macrosomia than insulin. In Type 1 diabetes, hypoglycaemia awareness is often lost in the first trimester, so teach the partner to give glucagon at the booking visit.
4. When the steroids go in
At 32 weeks our patient has a threatened preterm labour, and the obstetrician gives antenatal corticosteroids for the baby’s lungs. Her glucose will rise from the evening of the first dose, peak on days 2 and 3, and return to baseline by day 5 or 6. ACOG says to anticipate the rise over five days, but gives no numbers.
Mathiesen and colleagues (2002) tested a schedule on the subcutaneous doses:
- Day 1: raise the evening insulin by 25%.
- Days 2 and 3: raise all doses by 40%.
- Day 4: by 20%.
- Day 5: by 10 to 20%.
- Days 6 and 7: back to the pre-steroid dose.
The JBDS guidance (2017, updated 2022) expects a 40 to 50% increase over the course, or runs a variable rate intravenous insulin infusion from the first steroid dose, targeting 72 to 140 mg/dL, where hourly monitoring is possible.
“And the woman with gestational diabetes on diet alone?” Check her glucose four to six times a day for five days. No guideline gives a threshold for starting insulin during the course. A reading persistently above 140 is a reasonable one, and the tool says so.
5. Labour and the operation
In established labour, JBDS and NICE check capillary glucose hourly and aim for 72 to 126 mg/dL. In Type 1 diabetes the variable rate infusion starts at established labour. In Type 2 and gestational diabetes it starts after two consecutive readings above 126. The infusion is 50 units of soluble insulin in 50 mL of 0.9% sodium chloride, 1 unit per mL, with 5% dextrose alongside.
Many of our labour rooms have no syringe pump. For them the national protocol adds Regular insulin to 500 mL of normal saline, run at 100 mL/h (16 drops a minute). The glucose sets the number of units: none at 90 to 120, 4 units at 120 to 140, 6 at 140 to 180, 8 above 180. Glucose is checked two-hourly, and the morning insulin and Metformin are withheld on the day of induction.
For an emergency caesarean, the question is what long-acting insulin is already in her. A dose given within 12 hours is still working through the operation and recovery, with no meal to meet it. And whatever the route of delivery, tell the paediatric team before the delivery: the baby feeds within 30 minutes and has a glucose check 2 to 4 hours after birth.
6. After the placenta
When the placenta is delivered, the insulin resistance of pregnancy goes with it. In gestational diabetes, stop the infusion and all insulin. Check pre-meal readings for 24 hours. A fasting glucose of 126 or above, or a random of 200 or above, after 48 hours is diabetes that the pregnancy found. In Type 1 diabetes, restart insulin at once, eating or not, at a much smaller dose than late pregnancy needed; the tool prints the JBDS and ACOG figures.
“Then she is cured?” Her glucose is normal. Her risk is not. About half of women with gestational diabetes develop Type 2 diabetes within a decade. The postpartum test is the action item of the whole admission: a 75 g oral glucose tolerance test at 6 weeks (MoHFW), or at 4 to 12 weeks and then every 1 to 3 years for life (ADA). A line in the discharge summary asking someone else to arrange it is not an appointment.
Back to our patient
She starts 6 units of premix before breakfast, with an evening dose added for the fasting readings, titrated by 2 units every third day. Her steroid course is covered by a planned rise and fall in her doses. In labour she is on the saline table, glucose two-hourly. When the placenta is out, her insulin stops. She leaves with a date for her oral glucose tolerance test at 6 weeks, and advice on contraception, because the next pregnancy should begin with an HbA1c below 6.5% and folic acid already started.
Try this
A woman with gestational diabetes at 30 weeks is on 8 units of premix before breakfast. Her post-breakfast readings are now 110 to 118, but her fasting readings remain 104 to 110. Her colleague suggests increasing the morning premix to 10 units. Is that right? Illustrative case.
Answer. No. Her post-breakfast reading is already at target (120 or below at 2 hours). The problem is the fasting reading, and the morning premix has run out by 3 a.m. The protocol adds 2 units before dinner when the fasting reading is high; the weight-based alternative is a bedtime NPH. Raising the morning dose would push her daytime readings lower without touching the fasting glucose.
Take-home points
- Aim for 95 or below fasting and 120 or below at 2 hours; the Indian protocol and ADA agree on these.
- A morning premix controls the post-breakfast reading but not the fasting one; a high fasting needs an evening dose or bedtime NPH.
- In pre-existing diabetes the dose climbs each trimester; a fall after 36 weeks needs fetal surveillance.
- After antenatal steroids, expect the glucose to peak on days 2 and 3 and plan the doses for five days.
- The postpartum glucose tolerance test is the action item of the admission: give her a date, not a line in the summary.
For how gestational diabetes is diagnosed, see the queue’s article Diagnosing Diabetes and Prediabetes: The Evidence Behind the Thresholds. For when the DIPSI test falls in the antenatal schedule, see Pregnancy Dating and the Antenatal Schedule: The Evidence Behind the Tool. For insulin on the ward outside pregnancy, see Inpatient Insulin: The Evidence Behind the Protocol.
Use this at the bedside
Open the tool: Insulin in Pregnancy
Get the CDSS app: install the MEDiscuss CDSS on your phone
Declaration: the author built and maintains the tool. It is free to use. The figures in this article are those the tool carried in its Evidence tab on 5 November 2026, each from the source the tool names. The illustrative cases follow the tool’s rules; check the tool’s result for a real patient.
References
- American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes - 2026. Section 15: Management of Diabetes in Pregnancy. Diabetes Care. 2026;49(Suppl 1). [Current edition]
- American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 201: Pregestational Diabetes Mellitus. Obstet Gynecol 2018;132(6):e228-e248.
- American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 190: Gestational Diabetes Mellitus. Obstet Gynecol 2018;131(2):e49-e64.
- National Institute for Health and Care Excellence. Diabetes in pregnancy: management from preconception to the postnatal period (NG3). London: NICE; 2015, updated December 2020.
- Dashora U, Murphy HR, Temple RC, et al. Managing hyperglycaemia during antenatal steroid administration, labour and birth in pregnant women with diabetes (JBDS 12). Diabet Med 2018;35(8):1005-1010.
- Dashora U, Levy N, Dhatariya K, et al. Managing hyperglycaemia during antenatal steroid administration, labour and birth in pregnant women with diabetes: an updated guideline from the Joint British Diabetes Society for Inpatient Care. Diabet Med 2022;39(2):e14744.
- Ministry of Health and Family Welfare, Government of India. Diagnosis and Management of Gestational Diabetes Mellitus: Technical and Operational Guidelines. New Delhi: Maternal Health Division, MoHFW; 2018.
- Mathiesen ER, Christensen AB, Hellmuth E, Hornnes P, Stage E, Damm P. Insulin dose during glucocorticoid treatment for fetal lung maturation in diabetic pregnancy: test of an algorithm. Acta Obstet Gynecol Scand 2002;81(9):835-839.
- Research Society for the Study of Diabetes in India. RSSDI Clinical Practice Recommendations for the Management of Type 2 Diabetes Mellitus. Int J Diabetes Dev Ctries 2022;42(Suppl 1):1-143.
- Seshiah V, Balaji V, Balaji MS, et al. Diagnosis of gestational diabetes mellitus in the community. J Assoc Physicians India 2012;60:15-17.


