Endocrine Feedback Loop EFL057: Weaning Glucocorticoids and HPA Recovery

Date: January 16, 2025
Host: Dr. Chase Hendrickson, Vanderbilt University
Guests: Dr. Celia Kura, Columbia University & Dr. Felix Beuschlein, University Clinic Zurich
Featured Article: Arshad MF, et al. (JCEM, Nov 2024) – Retrospective Study on Weaning Glucocorticoids and Recovery of the Hypothalamic-Pituitary-Adrenal (HPA) Axis

Episode Overview

This episode explores a recent study published in the Journal of Clinical Endocrinology and Metabolism on best practices for weaning patients off long-term glucocorticoid therapy, specifically examining the recovery of the HPA axis and comparing outcomes between patients maintained on prednisolone and those switched to hydrocortisone. The discussion considers the implications for clinical practice, the study's methodological strengths and limitations, and how these findings intersect with current Endocrine Society guidelines.

Key Discussion Points and Insights

1. Background and Rationale

• Prevalence and Risks:

  • 1–3% of the general population is on systemic glucocorticoids (02:38, Dr. Kura).
  • Up to 50% of patients are at risk for adrenal suppression; suppression risk with >5 mg prednisolone (or equivalent) for ≥4 weeks.

• Assessment and Guidelines:

  • Suppression can result from any administration route; risk increases with cumulative dose and treatment duration.
  • Guidelines suggest both clinical and biochemical assessment (morning cortisol, 250 µg ACTH stimulation test), with some variability due to resource differences (04:57, Dr. Beuschlein).

“The guideline… appreciates both [clinical and hormonal evaluation] as equal possibilities… when patients are referred to me there's already a higher pretest probability of an HPA axis problem…”
— Dr. Felix Beuschlein (04:57)

2. Weaning Regimens and Biochemical Considerations

• Prednisolone vs. Hydrocortisone:
  • Prednisolone: higher receptor occupancy, longer half-life than hydrocortisone.
  • Some believe hydrocortisone (split dosing) is more physiological, possibly aiding axis recovery, but evidence is lacking (05:53, Dr. Kura).
  • Study aims to assess if staying on prednisolone or switching to hydrocortisone affects HPA recovery.

3. Study Design and Population

• Methodology:

  • Retrospective cohort/case series at a dedicated UK steroid clinic (Sheffield, 2015–2022) (07:26, Dr. Hendrickson).
  • All adult patients with suspected tertiary adrenal insufficiency (AI) after glucocorticoid exposure, mostly on prednisolone.
  • Exclusions: those on glucocorticoids other than prednisolone, certain clinical confounders (e.g., protein-losing disorders, night shifts).

• Referral Bias:

  • Referral patterns to specialized clinics affect generalizability—more complex or higher-risk cases in this cohort.

“It’s at least possible that [this group is] more likely to be complicated cases or other challenges… not as well representative of the general population.”
— Dr. Chase Hendrickson (09:02)

4. Key Methodological Debates

• Exposure Groups:
  • Only prednisolone at baseline included; hydrocortisone conversions happened during follow-up due to symptoms.
  • Dr. Beuschlein notes usefulness of including patients started directly on hydrocortisone for comparison (12:05).
• Prednisolone vs. Prednisone:
  • Used interchangeably in guidelines; small pharmacokinetic differences exist, but largely a practical/distinct tablet availability issue worldwide (13:14).

“From the work with the guidelines, I learned that availability of preparations worldwide is quite different… That can pose some practical challenges in the lower dose of the tapering…”
— Dr. Felix Beuschlein (13:14)

• Definition of Adrenal Insufficiency:
  • Used 30-min ACTH-stimulated cortisol <430 nmol/L (15.6 mcg/dL).
  • Cutoff debated; clinical correlation complex, endpoint (adrenal crisis) is rare and hard to measure (15:16, Dr. Beuschlein).

5. Results and Findings

Population:

• 837 patients screened → 276 with suspected tertiary AI → 206 on prednisolone included

Primary Outcome: HPA Axis Recovery on Prednisolone

• Of 176 patients who remained on prednisolone:
  • 112 (63.6%) passed first ACTH stim and were weaned off glucocorticoids.
  • 18 more recovered on second stim test, 7 on subsequent tests.
  • Only 3 still suppressed after three or more tests.
  • Overall recovery: 137/176 (77.8%) on prednisolone reactivated their HPA axis.

Secondary Outcome: Prednisolone vs. Hydrocortisone (Dose-Matched)

• 10 on prednisolone (4–5 mg), 13 converted to hydrocortisone (20 mg):
  • Recovery rates: 70% (prednisolone) vs. 15% (hydrocortisone), p=0.008 (22:48)
• In all suppressed patients:
  • 61% (prednisolone ≤5mg) vs. 27% (hydrocortisone ≤25mg), p=0.004.
• Hospitalizations: No adrenal crises on prednisolone; all adrenal crisis admissions occurred in hydrocortisone group.

Confounders Noted:

• More opioid use in prednisolone group.
• No significant difference in baseline demographics, disease, or initial cortisol/ACTH.
• Fewer stim tests and shorter follow-up in prednisolone group.

“During the study period… there were no hospital admissions due to adrenal crisis in the prednisolone group, while there were two patients in the hydrocortisone group who had at least one hospital admission due to adrenal crisis…”
— Dr. Celia Kura (24:39)

6. Discussion and Interpretation

• Primary Takeaway:

  • Most long-term prednisolone patients recover normal HPA function once referred for testing.
  • Conversion to hydrocortisone does not appear to accelerate HPA recovery.

• Reservations & Selection Bias:

  • Likelihood of selection bias due to non-randomized switch to hydrocortisone group based on symptoms (18:33, Dr. Beuschlein).
  • Possible that hydrocortisone group comprised more complex cases (hospitalizations for adrenal crisis might reflect this).

• Guideline Context:

  • Current guidelines do not mandate conversion to hydrocortisone as clearly superior for HPA recovery.
  • “Personal practice” (hydrocortisone vs. prednisolone taper) remains reasonable.

“For the time being, both theories are possible with no really convincing data supporting one or the other. The guideline holds true…”
— Dr. Felix Beuschlein (26:51)

• Limitations (Acknowledged by Authors and Panelists):
  • Retrospective design—variables and biases hard to control.
  • Adrenal insufficiency rate only applies to those tested with ACTH stim.
  • Lack of data on cumulative steroid dosing.
  • Conversion to hydrocortisone not randomized.

7. Application to Clinical Practice

• The panel is cautious about changing practice based on this single, retrospective report:
  • Dr. Kura: “If anything, [the study shows] it’s not inferior to wean on prednisolone… probably wouldn’t change my practice based on these results…” (30:20).
  • Dr. Beuschlein: Remains a ‘hydrocortisone person,’ not convinced to change; clinical and psychological considerations for patient need to be individualized (30:53).

“…sometimes it’s important that the patients understand that we are talking about weaning. We are not talking about treating their disease maybe any longer… From a psychological point of view, it sometimes helps the patient.”
— Dr. Felix Beuschlein (30:53)

• Both panelists and the host agree that a definitive answer requires a prospective, randomized, ideally blinded trial, but such studies may not be forthcoming soon (32:26).

Notable Quotes & Memorable Moments

• On guideline interpretation and practice patterns:

  “I think it’s also about your personal practice, whether you’re rather a hydrocortisone person or a… prednisolone person. If you feel more comfortable with that and treating your patients with that, that should also come into play…”
  — Dr. Felix Beuschlein (27:16)

• On the nature of evidence so far:

  “For the time being both theories are possible with no… really convincing data supporting one or the other.”
  — Dr. Felix Beuschlein (26:51)

• On study design and application:

  “I think the fact that it’s a retrospective design makes it difficult to know exactly what to do in your own practice based on these results.”
  — Dr. Celia Kura (29:54)

• On patient communication:

  “We are not talking about treating their disease maybe any longer. So we are in a situation where we try to get rid of any kind of steroid and sometimes the change… helps the patient.”
  — Dr. Felix Beuschlein (30:53)

Timestamps for Major Segments

| Segment | Timestamp | |-------------------------------------------------|------------| | Introduction, Importance of HPA Recovery | 00:00 | | Background & Prevalence of Adrenal Suppression | 02:38 | | Panel Expert Comments on Guidelines | 03:29 | | Clinical Practice Variation | 04:57 | | Drug Characteristics: Prednisolone vs. Hydrocortisone | 05:53 | | Study Design and Referral Bias | 07:26–09:30| | Inclusion/Exclusion Decisions | 12:05 | | Definition Debates: Prednisolone vs. Prednisone | 13:14 | | Diagnostic Thresholds for Adrenal Insufficiency | 14:06–15:16| | Symptom-driven Conversion and Biases | 16:34–18:33| | Primary and Secondary Outcomes Walkthrough | 20:24–24:50| | Results and Initial Interpretation | 24:50–27:44| | Discussion: Guideline fit, limitations | 27:44–29:54| | Practice Implications & Panel Opinions | 29:54–33:15|

Summary: Practical Takeaways

• Both prednisolone and hydrocortisone can be used as part of a glucocorticoid weaning regimen, with neither shown conclusively superior for HPA axis recovery in this retrospective study.
• Individual practice and patient context remain important, pending further evidence.
• Educating patients about the rationale for weaning and the goals of therapy (weaning rather than ongoing immunosuppression) can support successful transitions.
• A need remains for high-quality, prospective research comparing weaning strategies and evaluating patient outcomes including real-world risk of adrenal crisis.

For clinicians:
This study supports the current flexible approach to glucocorticoid weaning, with no strong mandate to convert to hydrocortisone. Continue to base tapering strategies on patient-specific factors, your clinical experience, and guideline recommendations until further evidence emerges.