#373 Strategies for initiating insulin in type 2 diabetes
Reading Tools for Practice Article can earn you MainPro+ Credits
Join NowAlready a CFPCLearn Member? Log in
- Results statistically significant unless indicated.
- 3 systematic reviews in the past 15 years (14-22 RCTs, 4,379-9,548 participants), comparing initiating basal insulin once daily (mostly glargine) versus biphasic (premix short and long acting) insulin twice-daily.1-3 Largest (4 months–3 years):1
- HbA1c reduction from baseline (7 RCTs, 3,472 participants): Biphasic 0.19% greater reduction versus basal.1
- If small RCT removed: 0.11% difference.1
- Other systematic reviews showed larger difference;2,3 metagraphs may have incorrect numbers; 2 largest RCT not included.3
- Weight gain (4 RCTs, 2,600 patients): Biphasic 1.25kg greater increase versus basal.1
- Total hypoglycemia (6 RCTs, 3,548 patients): Biphasic more than basal, absolute events not provided (basal Odds Ratio 0.72).1
- Severe/nocturnal hypoglycemia: Not different.
- HbA1c reduction from baseline (7 RCTs, 3,472 participants): Biphasic 0.19% greater reduction versus basal.1
- Largest RCT: 2,091 participants, baseline HbA1C ~9%, glargine once-daily (basal) versus premix (25% lispro, 75% lispro protamine suspension) twice-daily (biphasic). At 24 weeks:4
- HbA1c reduction from baseline: -1.7% basal versus -1.8% biphasic.
- % patients reaching HbA1C< 7%: 40% basal versus 48% biphasic. Number Needed to Treat=13.
- Insulin dose: Average ~36 Units basal versus ~44 Units biphasic [PEER calculation].
- Weight gain: 2.5kg basal versus 3.6kg biphasic.
- Symptomatic hypoglycemia (blood glucose ≤3.9): 34% basal versus 44% biphasic. Number Needed to Harm=10.
- Severe hypoglycemia rare.
- Limitations: RCTs open label; different insulin combinations/doses, sulfonylurea common co-intervention, now rarely used with insulin.
- Effects may be driven by total insulin dose, rather than type: one systematic review found when adjusted for total insulin dose, basal and biphasic HbA1c difference no longer significant.2
- RCTs (with limitations) show that insulin does not reduce mortality/cardiovascular outcomes.5
- Sample patient instructions online.6 Example: Start 10 units basal daily, titrate 1 unit/day until fasting glucose 4-7.