What is the preferred target‑controlled infusion pharmacokinetic model for propofol in adult patients?

Target-Controlled Infusion Propofol Models

The Eleveld model is the preferred pharmacokinetic model for target-controlled infusion (TCI) of propofol in adult patients, as it demonstrates superior predictive accuracy across diverse populations including adults, elderly, and obese patients, though the Schnider model may offer better bias control specifically in elderly patients. 1

Understanding TCI Models

Target-controlled infusion systems use population-based pharmacokinetic models programmed into computerized pumps to calculate and deliver propofol infusion rates that achieve predetermined target concentrations—either in plasma or at the effect site 2. The system continuously adjusts infusion rates based on patient characteristics (age, weight, sex) to maintain stable drug levels, avoiding the accumulation that occurs with fixed-rate infusions 2.

Primary Model Options

Eleveld Model (Recommended)

• Broad applicability: Validated across ages from 27 weeks post-menstrual age to 88 years and weights from 0.68-160 kg 3
• Comprehensive covariates: Incorporates age, post-menstrual age, weight, height, sex, and presence of concomitant anesthetic drugs 3
• Superior clearance estimation: Particularly advantageous in obese patients 4
• Effect-site targeting: Recommended dosing at 0.5-1 μg/mL effect-site concentration 1

Critical caveat: The Eleveld model shows higher incidence of deep anesthesia events (69.2% vs 30.8%) and burst suppression events (28.2% vs 5.1%) compared to Schnider, especially in elderly patients 5. This necessitates closer BIS monitoring when using Eleveld in patients ≥65 years.

Schnider Model (Alternative for Elderly)

• Better bias control in elderly: Demonstrates superior performance specifically in patients ≥65 years 4
• Lower deep anesthesia risk: Fewer burst suppression events in elderly (4.5% vs 31.6% with Eleveld) 5
• Anesthetic hysteresis: Shows predictable concentration-effect relationships 5

Marsh/Diprifusor Model (Historical)

• Limited accuracy: Higher bias (16%) and inaccuracy (26%) compared to newer models 6
• Not recommended: Superseded by models with better covariate incorporation 4

Clinical Implementation Algorithm

For patients <65 years:

1. Use Eleveld model with effect-site targeting
2. Target 0.5-1 μg/mL for sedation during awake procedures 1
3. Avoid bolus dosing—use gradual titration 1
4. Monitor with BIS targeting 40-60 for general anesthesia 5, 7

For patients ≥65 years:

1. Consider Schnider model to minimize deep anesthesia events 5
2. If using Eleveld, increase BIS monitoring frequency
3. Expect 31.6% incidence of burst suppression with Eleveld vs 4.5% with Schnider 5
4. Reduce initial targets by approximately 30% compared to younger adults 8

For obese patients (BMI ≥35):

1. Prefer Eleveld model—uses lean body weight (Janmahasatian formula) for clearance parameters 7
2. Allometric scaling improves accuracy over total body weight 7
3. Target BIS 40-60 with propofol infusion guided by effect-site concentration 7

Safety Considerations

Avoid exceeding 1.5 μg/mL effect-site concentration when using TCI for sedation, particularly with concomitant opioids, due to risk of over-sedation and hypoventilation 1. The guidelines explicitly warn against bolus dosing in awake procedures 1.

Titration intervals: Allow 3-5 minutes between dose adjustments to assess clinical effects, as propofol's effect-site equilibration half-time (ke0) is approximately 0.146 min⁻¹ 8, 3.

Model Limitations

No single model achieves universal applicability 4. The Eleveld model, despite superior overall performance, shows inconsistencies in predicting brain propofol concentrations across all populations 4. Ongoing research explores machine learning and adaptive dosing strategies, but these require external validation before clinical implementation 4.

Key pitfall: Failure to reduce infusion rates during prolonged administration (>10 days) leads to tissue accumulation and delayed awakening, as propofol's context-sensitive half-time increases dramatically with infusion duration 8. After extended infusions, reduce rates by up to 50% to maintain constant plasma levels 8.