Insulin Dose Adjustments for Type 1 Diabetes
Discontinue metformin immediately—it is not indicated for type 1 diabetes and poses unnecessary risk of lactic acidosis, particularly given the low bicarbonate (17 mmol/L) and anion gap (13) suggesting early metabolic acidosis 1, 2, 3.
Critical Context
This 19-year-old has type 1 diabetes with suboptimal control (all glucose readings elevated except the 4 AM value of 97 mg/dL) and is inappropriately on metformin. The blood glucose pattern shows:
- Morning hyperglycemia (184 mg/dL at 8 AM)
- Progressive daytime rise (220→262 mg/dL)
- Evening elevation (237 mg/dL at 10 PM)
- Appropriate overnight nadir (97 mg/dL at 4 AM)
The low bicarbonate (17 mmol/L) with anion gap of 13 raises concern for chronic metabolic acidosis, potentially exacerbated by metformin in the setting of type 1 diabetes.
Basal Insulin (Lantus) Adjustments
AM Lantus Dose
Increase from 15 units to 20 units (approximately 33% increase). The 8 AM glucose of 184 mg/dL indicates insufficient basal coverage during the overnight-to-morning period. The 4 AM glucose of 97 mg/dL is appropriate, but the rise to 184 mg/dL by 8 AM suggests dawn phenomenon requiring more morning basal insulin 1.
PM Lantus Dose
Maintain 35 units or consider slight reduction to 32 units (approximately 10% decrease). The 4 AM glucose of 97 mg/dL indicates the evening dose is providing adequate overnight coverage. The progressive daytime hyperglycemia (220→262 mg/dL) is primarily a prandial insulin issue, not a basal problem. Given the patient's weight (86 kg) and the risk of nocturnal hypoglycemia, a modest reduction may be prudent 4, 5.
Total daily basal insulin would be approximately 52 units (0.6 units/kg/day), which is appropriate for this patient's weight and represents roughly 50% of total daily insulin dose 5, 6.
Insulin-to-Carbohydrate Ratios (ICR)
Breakfast ICR
Change from 1:6 to 1:5 (more aggressive). The rise from 184 mg/dL (8 AM) to 220 mg/dL (noon) indicates insufficient breakfast bolus coverage. This represents a 20% increase in insulin per gram of carbohydrate 6.
Lunch ICR
Change from 1:7 to 1:5 (more aggressive). The dramatic rise from 220 mg/dL (noon) to 262 mg/dL (5 PM) indicates grossly inadequate lunch coverage. This represents a 40% increase in insulin per gram of carbohydrate 6.
Dinner ICR
Change from 1:7 to 1:6 (moderately more aggressive). The modest decline from 262 mg/dL (5 PM) to 237 mg/dL (10 PM) suggests dinner coverage is closer to adequate but still insufficient. This represents a 17% increase 6.
Insulin Sensitivity Factor (ISF)
Change from 25 mg/dL per unit to 20 mg/dL per unit (more aggressive correction). Using the 1800 rule: 1800 ÷ estimated total daily dose of ~100 units = 18 mg/dL per unit. However, starting at 20 mg/dL per unit is safer to avoid overcorrection. The current ISF of 25 is too conservative given persistent hyperglycemia throughout the day 6, 5.
Monitoring and Safety Considerations
Critical pitfall: The low bicarbonate (17 mmol/L) requires immediate evaluation for diabetic ketoacidosis risk, especially after stopping metformin. Check beta-hydroxybutyrate and ensure adequate insulin coverage during the transition 7, 3, 8.
Hypoglycemia prevention: The 4 AM glucose of 97 mg/dL is appropriate, but with increased insulin doses, monitor for overnight hypoglycemia. Consider continuous glucose monitoring if not already in place 5, 9.
Titration approach: Implement these changes gradually over 3-5 days, adjusting one parameter at a time if possible. Start with basal insulin adjustments, then modify ICRs, then fine-tune ISF 4.
Expected total daily insulin: With these adjustments, total daily insulin should approximate 0.8-1.0 units/kg/day (69-86 units/day for 86 kg), which is typical for adolescents with type 1 diabetes 6, 10.