Basal Insulin Should Replace Sliding Scale Insulin in Hospitalized Diabetic Patients
Basal insulin regimens (either basal-bolus or basal-plus) should be used instead of sliding scale insulin (SSI) alone for hospitalized adults with type 1 or type 2 diabetes, as SSI is associated with worse glycemic control and higher rates of complications. 1
The Evidence Against Sliding Scale Insulin
Sliding scale insulin is widely condemned in clinical guidelines despite its continued use in many hospitals 1. The fundamental problem is that SSI is reactive—it treats hyperglycemia only after it has already occurred, leading to:
- Clinically significant hyperglycemia in most patients 1
- Higher complication rates including postoperative wound infections, pneumonia, bacteremia, acute renal failure, and respiratory failure 1
- Poor glycemic control with mean blood glucose levels 14.8 mg/dL (0.8 mmol/L) higher than basal-bolus regimens 2
Critical exception: SSI alone should NEVER be used in type 1 diabetes patients 1. It may only be appropriate for patients without diabetes who have mild stress hyperglycemia 1.
Recommended Insulin Strategies
For Type 2 Diabetes: Basal-Bolus or Basal-Plus
Basal-Bolus Regimen (preferred for most patients):
- Basal insulin once or twice daily PLUS rapid-acting insulin before meals PLUS correctional doses
- Dosing for insulin-naive patients: 0.3-0.5 U/kg total daily dose
- Split 50/50: half as basal (1-2 times daily), half as rapid-acting (divided before 3 meals)
- Dosing for high-risk patients (age >65, renal failure, poor oral intake): use lower end of range
- For patients on home insulin ≥0.6 U/kg/day: reduce total daily dose by 20% during hospitalization to prevent hypoglycemia 1
Basal-Plus Regimen (preferred for specific situations):
- Single dose of basal insulin (0.1-0.25 U/kg/day) PLUS correctional doses only
- Use when:
- Mild hyperglycemia (blood glucose <200 mg/dL)
- Decreased oral intake
- Perioperative period 1
For Type 1 Diabetes
Multiple daily injections or continuous subcutaneous insulin infusion with both basal and prandial components are required 3. SSI alone is contraindicated 1.
Clinical Outcomes: The Numbers
Randomized trials consistently demonstrate superiority of basal-based regimens over SSI:
- Better glycemic control: Basal-bolus achieved lower fasting glucose (10.8 vs 11.6 mmol/L) and mean glucose throughout hospitalization 4
- Reduced complications: Composite outcome of infections and organ failures significantly lower with basal-bolus 1
- Treatment failure rates: 0-2% with basal regimens vs 19% with SSI 5
The Hypoglycemia Trade-off
Important caveat: Basal-bolus regimens carry a 12-30% incidence of mild hypoglycemia in controlled settings 1. However, severe hypoglycemia (<40 mg/dL) rates are similar between approaches 2, 5. The basal-plus approach may offer a middle ground with lower hypoglycemia risk while maintaining superior glycemic control compared to SSI 5.
Practical Implementation Algorithm
Assess diabetes type:
- Type 1 → basal-bolus mandatory (never SSI alone)
- Type 2 → proceed to step 2
Evaluate hyperglycemia severity and oral intake:
- Severe hyperglycemia (>200 mg/dL) + eating well → basal-bolus
- Mild hyperglycemia (<200 mg/dL) OR poor oral intake OR perioperative → basal-plus
- Non-diabetic with stress hyperglycemia → SSI acceptable
Calculate starting dose:
- Insulin-naive: 0.3-0.5 U/kg (lower for elderly/renal disease)
- On home insulin <0.6 U/kg: continue same dose
- On home insulin ≥0.6 U/kg: reduce by 20%
Prescribe glucagon for all patients on insulin 3
Common Pitfalls to Avoid
- Do not use premixed insulin (70/30) in the hospital—unacceptably high hypoglycemia rates 1
- Do not continue SSI out of habit—it has been condemned in guidelines for nearly 20 years 6
- Do not forget to reduce home insulin doses by 20% for patients on high doses to prevent hypoglycemia with reduced oral intake 1
- Do not use SSI as monotherapy in type 1 diabetes—this is dangerous 1
The evidence is clear and consistent: basal insulin strategies provide superior glycemic control and reduce complications compared to the outdated sliding scale approach, which should be abandoned except in the narrow circumstance of stress hyperglycemia in non-diabetic patients.