High-dose glucocorticoids, pain, and adverse events. Protocol for an analysis of a natural experiment of patients undergoing hip and knee arthroplasty

This protocol outlines a natural experiment study using target trial emulation to evaluate the real-world impact of a single high-dose glucocorticoid administration on postoperative opioid consumption and adverse events in patients undergoing primary hip and knee arthroplasty in Denmark.

The Gist

Imagine you are trying to figure out if a new, powerful spice makes a soup taste better and keeps people from getting sick.

In the medical world, this "soup" is surgery (specifically hip and knee replacements), and the "spice" is a high dose of glucocorticoids (a type of steroid). Doctors have been using this spice to reduce pain and swelling after surgery. While small, controlled taste-tests (Randomized Controlled Trials) say the spice works great, those tests often leave out people who are older, sicker, or have other complications.

This paper is a plan to see how the spice works in the real world, with all the messy, complicated, real-life patients included.

Here is the breakdown of their plan, using some everyday analogies:

1. The "Natural Experiment" (The Accidental Lab)

Usually, to test a new drug, scientists randomly pick people to get the drug and others to get a placebo (sugar pill). But in Denmark, something interesting happened naturally: different hospitals started using this high-dose steroid at different times.

• The Analogy: Imagine a neighborhood where House A starts using a new brand of coffee maker in January, House B in March, and House C in June. You didn't force them to do it; they just decided to upgrade at different times.
• The Study: The researchers are going to look at the "coffee drinkers" (patients) at these houses. They will compare the people who got surgery before the new coffee maker arrived (the "Old Way" group) with the people who got surgery after it arrived (the "New Way" group). Because the change happened at different times in different places, it acts like a giant, real-life experiment.

2. The Goal: Less Pain, Fewer Side Effects

The researchers want to answer two main questions:

• Does it work? Do patients need less strong painkillers (opioids) after surgery?
• Is it safe? Does the new spice cause bad side effects, like infections or nausea?

They are looking at three specific things:

1. Pain Relief: How much "pain medicine" does the patient need in the first 24 hours?
2. The "Hangover": Did the patient get sick, dizzy, or have trouble breathing (side effects)?
3. The Recovery Time: How long did they stay in the hospital?

3. The "Target Trial" (Playing Pretend)

Since they can't go back in time and randomly assign patients, they are going to use a method called "Target Trial Emulation."

• The Analogy: Think of it like a detective trying to solve a crime. They can't recreate the crime scene, so they gather all the evidence (medical records) and try to build a "virtual courtroom" in their minds. They ask: "If we had run a perfect, fair experiment right now, what would the results look like based on the data we have?"
• They are using a massive digital database (called TRIPLE-A) that contains the electronic health records of over a million surgeries. It's like having a giant library of patient stories to read through.

4. The Rules of the Game

• Who is in the study? Adults having their first hip or knee replacement in specific regions of Denmark.
• The "Before" vs. "After": For each hospital, they look at the year before the steroid was introduced (Control Group) and the year after (Treatment Group).
• The "Real World" Factor: In a real hospital, not everyone follows the rules perfectly. Some patients with diabetes might be skipped for the steroid because doctors worry about blood sugar spikes. The researchers plan to count these "rule-breakers" to see if they skew the results.

5. How They Will Analyze the Data

They aren't just counting numbers; they are using a sophisticated statistical model (a "smart calculator") that accounts for:

• The Hospital: Some hospitals are just better at surgery than others. They will adjust for this.
• The Patient: Older patients or those with more health problems might react differently. They will adjust for this too.
• The Timing: They will make sure that changes in the results weren't just because of the time of year or other new medical trends.

6. Why This Matters

If this study shows that the high-dose steroid works well in the "real world" (with all the messy, complex patients), it could change how hospitals treat millions of people. It could mean:

• Patients go home sooner.
• Patients need fewer addictive painkillers.
• Doctors feel safer prescribing it to people they previously avoided.

In a nutshell: This paper is a roadmap for a giant detective story. The detectives are using a mountain of real-life data to see if a new pain-management "super-spice" actually works for everyone, or just the lucky few in the controlled lab tests. They are trying to prove that what works in a perfect world also works in our messy, real world.

Technical Summary

Based on the provided protocol document, here is a detailed technical summary. Note that as this is a study protocol (Version 2, dated February 2026) rather than a completed research article, it outlines the planned methodology and objectives but does not contain final results.

1. Problem Statement

• Clinical Context: High-dose glucocorticoids are increasingly used as part of multimodal pain management for patients undergoing primary total hip arthroplasty (THA), unicompartmental knee arthroplasty (UKA), and total knee arthroplasty (TKA).
• The Gap: Current evidence supporting their efficacy in reducing acute postoperative pain is primarily derived from Randomised Controlled Trials (RCTs). RCTs often suffer from strict exclusion criteria, limiting their external clinical validity (generalizability to real-world populations).
• The Concern: There are ongoing concerns regarding adverse events, particularly the risk of surgical site infections and prosthetic joint infections, which has led to a cautious, step-wise implementation of the treatment in Denmark rather than immediate universal adoption.
• Objective: To examine the real-world effect of a single high-dose glucocorticoid administration on postoperative opioid consumption and adverse events in a broad patient population undergoing THA, UKA, and TKA.

2. Methodology

The study is designed as a natural experiment reported using the Target Trial Emulation framework. It leverages a "step-wise" implementation of the treatment across different hospitals in Denmark as a quasi-experimental setup.

• Study Design:

  • Type: Natural experiment / Quasi-experimental study (Target Trial Emulation).
  • Setting: Public hospitals in the Capital Region and Region Zealand, Denmark (2017–2025).
  • Population: Adults (>18 years) undergoing primary, elective THA, UKA, or TKA.
  • Inclusion Window: Patients operated on within one year before and one year after the specific implementation date of high-dose glucocorticoids at each hospital.

• Intervention vs. Comparator:

  • Intervention (Treatment Arm): Patients operated on after the implementation of high-dose glucocorticoids. The dose is defined as $\ge$ 0.2 mg/kg or 16 mg dexamethasone (or equivalents like 80 mg methylprednisolone), administered after anesthesia induction.
  • Comparator (Control Arm): Patients operated on before implementation. These patients may have received low-dose glucocorticoids (4–8 mg dexamethasone) solely for anti-emetic purposes, but not high-dose analgesic protocols.
  • Assignment: Non-random, open-label. Allocation is based on the hospital's implementation timeline (Difference-in-Differences approach).

• Data Source:

  • TRIPLE-A Database: Validated electronic health record (EHR) data from the "Sundhedsplatformen" (Epic Systems) covering over one million surgical procedures.
  • Validation: Manual validation of treatment and co-intervention variables was performed on a random sample of 50 patients.

• Outcomes:

  • Primary Outcome: Opioid consumption in the first 24 hours post-surgery, expressed in Morphine Equivalent Doses (MEQs).
  • Secondary Outcomes:
    • Adverse Events: Incidence of opioid-related adverse events (ORADE) and serious adverse events.
    • Efficacy: Worst pain intensity (NRS 0-10) in 0-24h, length of stay in PACU, total hospital stay, and days alive and out of hospital at 30 days.

• Statistical Analysis Plan:

  • Model: Pre-defined Mixed-Effects Models (using R lme4 package).
    • Random Effect: Hospital (to account for clustering).
    • Fixed Effects: Implementation status (after_GCS), time period, and patient-level covariates (Age, Sex, Surgery Type, Anesthesia Type, Premedication, Chronic Opioid Use, Psychiatric diagnoses).
  • Handling Missing Data: Multiple imputation (mice package) if missingness is balanced; otherwise, analysis is halted.
  • Sensitivity Analysis: Per-protocol analysis excluding patients in the treatment arm who did not receive the intervention (e.g., due to diabetes contraindications).
  • Significance Thresholds:
    • Primary outcome: $P < 0.05$ (95% CI).
    • Secondary outcomes: Bonferroni corrected ($P < 0.01$) with 99% CIs to control for false positives.
  • Process Control: Statistical Process Control (SPC) charts (Xbar and P-charts) will be used to visualize trends and detect non-random patterns relative to implementation time.

3. Key Contributions

• Real-World Evidence (RWE): This study moves beyond the constraints of RCTs to provide high-quality evidence on the effectiveness and safety of high-dose glucocorticoids in a routine clinical setting with diverse patient comorbidities.
• Natural Experiment Design: By utilizing the staggered implementation of the treatment across Danish hospitals, the study mimics a randomized trial without the ethical or logistical constraints of randomization.
• Comprehensive Safety Profile: Unlike many RCTs that focus primarily on pain reduction, this protocol explicitly prioritizes the assessment of adverse events (specifically infection risks and opioid-related side effects) alongside efficacy.
• Methodological Rigor: The use of Target Trial Emulation, mixed-effects modeling to adjust for hospital-level clustering, and rigorous sensitivity analyses ensures robust causal inference.

4. Results

• Status: No results are available. This document is a protocol published in February 2026. It details the planned analysis and hypotheses but does not contain the study findings, statistical outputs, or conclusions regarding the efficacy or safety of the intervention.

5. Significance

• Clinical Impact: The findings will directly inform clinical guidelines for perioperative pain management in orthopedics. If high-dose glucocorticoids are proven effective and safe in real-world data, it could lead to widespread adoption, reducing opioid consumption and improving patient recovery. Conversely, if adverse event rates (e.g., infections) are elevated, it could prompt a re-evaluation of current protocols.
• Policy Implications: Given the step-wise implementation in Denmark, this study provides the evidence base needed to standardize treatment across the healthcare system, potentially influencing national and international guidelines.
• Research Paradigm: The study serves as a model for utilizing large-scale EHR databases and natural experiments to answer causal questions that are difficult or impossible to address with traditional RCTs.

Funding & Conflicts: The study received no external funding, and the authors declare no conflicts of interest. The protocol is publicly available via medRxiv.