Large-scale pharmacokinetic reconstruction of propofol effect-site concentrations during anaesthetic induction

This large-scale retrospective study reveals that despite age-based dose reductions in real-world practice, older adults still experience disproportionately high propofol brain exposure because weight-based dosing heuristics fail to adequately account for age-related increases in hypnotic sensitivity, supporting a shift toward exposure-informed and neurophysiologically guided induction titration.

The Gist

The Big Picture: The "Too Much Coffee" Problem for Older Adults

Imagine you are making coffee for a group of friends. You know that a young, energetic 20-year-old can handle a huge, strong cup of coffee without falling asleep. But if you give that same giant cup to your 80-year-old grandmother, she might pass out immediately because her system is much more sensitive to caffeine.

To be safe, you decide to give her a smaller cup. You think, "Okay, I'll give her half the amount I gave the young guy."

This study is about what happens when doctors give anesthesia (a drug that puts you to sleep) to older patients.

The researchers looked at over 250,000 surgeries to see if doctors were actually giving older adults the right amount of "sleep medicine" (propofol). They found a surprising problem: Doctors are reducing the dose for older people, but not enough.

Because older brains are like "super-sensitive coffee drinkers," the standard dose reductions doctors use are still too high. This means older patients are often getting a "brain buzz" that is way stronger than necessary, which can lead to dangerous drops in blood pressure or confusion after surgery.

---

The Analogy: The "Volume Knob" vs. The "Ear Sensitivity"

To understand why this happens, let's use a Volume Knob analogy.

1. The Drug (Propofol): This is the music playing in a room.
2. The Dose (mg/kg): This is how loud the doctor turns the volume knob.
3. The Patient's Brain: This is the listener's ear.

The Old Way of Thinking:
Doctors have been taught: "As people get older, their ears get more sensitive. So, if we turn the volume knob down by 30%, the older person will hear the same amount of music as the young person."

What This Study Found:
The researchers used a super-accurate computer simulation (like a high-tech sound engineer) to measure exactly how loud the music actually was inside the brain.

They discovered that while doctors did turn the volume knob down (reduced the dose) for older patients, the actual volume inside the brain didn't drop nearly as much as it should have.

• The Young Adult (18–24): Gets a dose that results in a "brain volume" of 3.70.
• The Older Adult (85–89): Gets a smaller dose, but the "brain volume" is still 3.06.

The Problem:
The older person's brain is so sensitive that they only need a volume of 2.22 to fall asleep.

• The Gap: The older patient is receiving a volume of 3.06, which is 38% louder than what they actually need.
• The Result: They are being "over-sedated." It's like turning the volume down just a little bit, but the listener's ear is so sensitive that they are still getting blasted by noise.

---

Why Didn't the Doctors Know This?

The study explains that doctors have been relying on a simple rule of thumb: "Give less drug per pound of body weight as people age."

Think of it like this:

• The Rule: "If the person is older, give them 30% less coffee."
• The Reality: The older person's body processes the coffee differently. Their liver clears it slower, and their brain absorbs it faster.

So, even though the doctor gave a smaller cup (less drug), the older person's body held onto the drug longer and sent more of it to the brain. The simple "less coffee" rule didn't account for the fact that the older person's "cup" (brain exposure) was still filling up way too fast.

The "Anchoring" Effect

The study found something really interesting: Doctors are stuck in the past.

Even for an 85-year-old, the amount of drug they receive is still very close to what a healthy 20-year-old gets.

• The Stat: Over 54% of 75-year-olds received a dose that would have been considered "too much" for a 20-year-old.
• The Metaphor: It's like driving a car. The doctor sees an 80-year-old and thinks, "I need to slow down." So they take their foot off the gas a little bit. But they don't realize the car has new, super-sensitive brakes. They are still driving way too fast for the conditions, even though they think they are driving slowly.

Why Does This Matter?

If you put an older person to sleep with too much "volume," bad things can happen:

1. Blood Pressure Crashes: Their heart might stop pumping hard enough because the "sleep signal" is too strong.
2. Confusion: They might wake up groggy, confused, or have trouble remembering things (delirium) because their brain was overwhelmed.
3. Brain Stress: It puts unnecessary stress on a brain that is already aging.

The Solution: Listen to the Brain, Not the Scale

The paper suggests that doctors need to stop just looking at the scale (how much drug per pound) and start thinking about the brain exposure (how much drug is actually hitting the brain).

Instead of guessing based on age and weight, the authors suggest using:

• Better Math: Using complex computer models (like the one they used in this study) to predict exactly how much drug the brain will see.
• Brain Monitors: Using EEG machines (like a stethoscope for brain waves) to see in real-time if the patient is actually asleep, rather than just guessing based on how much drug was injected.

Summary

The Takeaway: We are treating older patients like they are just "smaller versions" of young patients. In reality, their brains are "super-sensitive receivers." By only slightly lowering the dose, doctors are accidentally blasting the volume too high, leading to unnecessary risks. We need to tune the "radio" much more carefully for our older patients.

Technical Summary

1. Problem Statement

Current clinical guidelines for propofol induction in geriatric patients recommend dose reductions based on age, acknowledging that hypnotic sensitivity increases with age (i.e., older adults require lower effect-site concentrations to achieve loss of consciousness). However, real-world practice is often evaluated using static metrics like weight-normalized dose (mg kg⁻¹).

The study identifies a critical gap:

• Static vs. Dynamic: Weight-based dosing ignores the temporal dynamics of drug delivery (bolus vs. infusion) and the complex pharmacokinetic (PK) changes associated with aging (e.g., altered volume of distribution and clearance).
• The Hypothesis: Clinicians may reduce the mass of the drug administered to older adults, but due to PK changes, this reduction may not translate proportionally into a reduction in actual brain exposure (effect-site concentration, $C_e$). Consequently, older patients may be systematically over-exposed to propofol despite receiving "reduced" doses.

2. Methodology

The study employed a retrospective, large-scale pharmacokinetic reconstruction approach using high-fidelity electronic health record (EHR) data.

• Data Source: A perioperative data warehouse from UCLA containing anesthesia documentation from March 2013 to February 2026.
• Cohort: 250,640 adult anesthetic inductions (ages 18–89). Cases with missing data, physiologic outliers, or implausible dosing were excluded.
• Pharmacokinetic Reconstruction:
  • Input: High-resolution medication timestamps (bolus doses in mg and infusion rates in µg kg⁻¹ min⁻¹) extracted within a 10-minute window of the induction event.
  • Model: The Eleveld model was used as the primary PK/PD framework due to its broad applicability across ages and body compositions. The Schnider model was used for sensitivity analysis.
  • Simulation: Using the PyTCI library, effect-site concentration trajectories ($C_e(t)$) were simulated at 1-second resolution for 900 seconds post-induction.
  • Primary Metric: Peak modeled effect-site concentration ($C_{e,max}$).
• Benchmarking:
  • $C_{e,max}$ was compared against the age-adjusted $C_{e50}$ (the concentration required for loss of consciousness in 50% of the population), derived from the Eleveld model: $C_{e50}(Age) = 3.08 \cdot e^{-0.00635(Age-35)}$.
  • Overexposure was defined as $C_{e,max} > C_{e50}(Age)$.
• Statistical Analysis:
  • Covariate-adjusted natural cubic splines (5 degrees of freedom) were used to model the relationship between age and outcomes, controlling for BMI, sex, and ASA physical status.
  • Logistic regression was used to estimate the probability of overexposure.

3. Key Contributions

• Scale and Resolution: This is one of the largest studies (N > 250,000) to reconstruct patient-specific effect-site concentrations from real-world manual bolus dosing, moving beyond static dose audits.
• Exposure-Based Framework: The study shifts the paradigm from analyzing "dose administered" to "exposure achieved," demonstrating that dose metrics are poor surrogates for brain exposure in the elderly.
• Quantification of the "Titration Offset": The research quantifies the divergence between the decline in administered dose and the decline in physiological requirement, revealing a systematic mismatch.

4. Key Results

• Dose vs. Exposure Divergence:
  • From young adulthood (18–24) to the oldest cohort (85–89), the weight-normalized induction dose decreased by 32% (from 3.16 to 2.16 mg kg⁻¹).
  • However, the modeled peak effect-site concentration ($C_{e,max}$) declined by only 17% (from 3.70 to 3.06 µg ml⁻¹).
  • In contrast, the physiological requirement ($C_{e50}$) declined by 34% (from 3.37 to 2.21 µg ml⁻¹).
• Systematic Overexposure:
  • Because the achieved concentration ($C_{e,max}$) declined much slower than the required concentration ($C_{e50}$), the "surplus" exposure widened with age.
  • At age 75, the adjusted probability of a patient exceeding their individual hypnotic requirement ($C_{e,max} > C_{e50}$) was 89.6%.
  • Notably, 54.7% of 75-year-old patients achieved peak exposures exceeding the average requirement of a healthy 20-year-old (3.37 µg ml⁻¹), indicating that exposure levels remain anchored to youthful standards despite dose reductions.
• Robustness: These findings were consistent across sensitivity analyses using the Schnider model, unadjusted analyses, inclusion of patients ≥90 years, and adjustments for co-administered drugs (fentanyl, ketamine, midazolam, etomidate).

5. Significance and Clinical Implications

• Inadequacy of Current Heuristics: The study demonstrates that simple weight-based dose reductions are insufficient to account for the exponential increase in hypnotic sensitivity and the complex PK changes in the elderly.
• Patient Safety Risks: Systematic overexposure in geriatric patients may contribute to:
  • Post-induction hypotension.
  • Excessive anesthetic depth (e.g., EEG burst suppression), which is linked to adverse postoperative outcomes such as delirium and cognitive dysfunction.
• Future Practice: The authors advocate for a transition from weight-based dosing heuristics to exposure-informed titration. This suggests a greater reliance on:
  • Pharmacokinetic modeling at the bedside.
  • Neurophysiologic monitoring (e.g., BIS, entropy) to guide induction depth directly, rather than relying on calculated doses.

In conclusion, the paper provides robust evidence that current clinical practices for propofol induction in the elderly result in a significant and widening gap between the intended and actual brain exposure, necessitating a shift toward more precise, exposure-guided anesthetic management.