Systemic Effects of Perineural Glucocorticoids on Fasting Serum Glucose, Potassium, and White Blood Cell Count in Total Hip Arthroplasty

Citation

Sharma, A., Dai, F., Tseng, L., Effraim, P. R., Zhou, B., Schonberger, R. B., & Li, J. (2023). Systemic effects of perineural glucocorticoids on fasting serum glucose, potassium, and white blood cell count in total hip arthroplasty. Journal of Pain Research, 16, 553–561. https://doi.org/10.2147/JPR.S395336 Free full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC9946012/

Chapter Summary

This is a nine-page open-access research article built in the usual order of abstract, introduction, materials and methods, results, discussion and conclusion. It is a retrospective cohort study from one academic hospital that asks whether glucocorticoids added to nerve blocks, which are meant to act locally, leave a measurable systemic footprint in the laboratory values that patients receive after hip replacement.

Abstract and Introduction

Glucocorticoids (dexamethasone and methylprednisolone acetate) are used as adjuvants to lengthen and improve nerve blocks, with dexamethasone described as quick in onset and methylprednisolone as acting over days. Little had been published on whether perineural doses change fasting serum glucose, serum potassium or white blood cell count (WBC). The practical concern is that abnormal values after surgery can trigger unnecessary work-up, treatment and delayed discharge. The stated primary hypothesis is that the combination in anterior quadratus lumborum and lateral femoral cutaneous nerve blocks would produce modest changes that carry little clinical weight.

Materials and Methods

With institutional review board approval and waived consent, the team reviewed electronic health records of patients who had elective, unilateral, primary total hip arthroplasty for osteoarthritis under spinal anesthesia between July 2017 and July 2019. Of 228 consecutive patients, 18 were excluded, leaving 210 operations: 132 with periarticular local anesthetic injection alone (PAI) and 78 with PAI plus peripheral nerve blocks (PAI+PNB). Each block contained ropivacaine with 5 mg dexamethasone and 40 mg methylprednisolone acetate. The primary outcome was the change in fasting serum glucose from the preoperative baseline on postoperative days (POD) 1, 2 and 3; secondary outcomes were changes in WBC and potassium. Analysis used a mixed-effects model for repeated measures adjusted for baseline value, time and group, and complications were tracked for 90 days.

Results

The groups were similar in age, sex, body mass index and ASA class, but type 2 diabetes was more common in the PAI-only group. After adjustment, the extra rise in glucose with the nerve blocks was 19.87 mg/dL on POD 1 and 17.5 mg/dL on POD 2, and not significant on POD 3. Potassium differed by 0.16 mEq/L on POD 1 only, and WBC by 3.18 × 1000/mm³ on POD 2 only. The authors describe these as statistically significant but not clinically significant.

Discussion and Conclusion

The discussion sets the glucose difference against a meta-analysis that found a similar roughly 20 mg/dL rise with systemic glucocorticoids and treated it as clinically unimportant, and against guidance that keeps perioperative glucose below 200 mg/dL; the peak fasting glucose here was about 149 mg/dL. It cautions that a WBC rise to roughly 13 × 1000/mm³ is non-specific and cannot rule out infection concerns, while noting that glucocorticoid-related leukocytosis usually lacks a left shift or fever. The limitations are the retrospective single-center design, moderate sample size, imbalance in diabetes, missing data and the lack of adjustment for multiple testing, so the results are treated as hypothesis-generating. A prospective randomized trial is proposed. The conclusion is that the transient changes are unlikely to matter clinically.

Key Concepts

Abstract and Introduction

• Perineural glucocorticoid adjuvants: steroids placed with local anesthetic around a nerve to prolong and strengthen the block.
• Dexamethasone and methylprednisolone acetate: the two agents studied, the first faster and shorter acting, the second thought to act for days.
• Systemic effect of a local drug: absorption from the block site that can alter glucose, electrolytes or blood counts.

Methods and Results

• PAI versus PAI+PNB: the two analgesic strategies compared, local periarticular injection alone versus injection plus fascial-plane nerve blocks.
• Change from baseline: each outcome is expressed as the difference from the preoperative value, so patients serve as their own reference.
• Mixed-effects model for repeated measures: a model that handles the same patient measured on several days.

Discussion

• Statistical versus clinical significance: a difference can be reliably detected yet too small to change care, which the authors tie to the idea of a minimal clinically important difference.
• Hyperglycemia and surgical site infection: the reason glucose thresholds matter in arthroplasty.
• Hypothesis-generating study: a finding that motivates a trial rather than settling the question.

Critical Analysis

Strengths

• A focused, clinically meaningful question. The outcomes (glucose, potassium, WBC) are routine tests that could lead to extra work-up or delayed discharge, so the study speaks to real postoperative decisions.
• Well-defined cohort and standardized care. The inclusion criteria, spinal technique, fasting practice, block doses and timing of morning blood draws are described, which limits some sources of noise.
• Appropriate repeated-measures analysis. Adjusting for baseline and for time-by-group effects, and reporting confidence intervals rather than only p-values, makes the size of each difference visible.
• Honest interpretation. The discussion puts the findings in the context of thresholds and previous studies, and the limitations paragraph says plainly that the results are exploratory.

Limitations

• Observational and unrandomized. Patients were not assigned to the nerve block; the paper itself notes fewer diabetic patients among those who received steroid-containing blocks, possibly reflecting clinician caution.
• Single center, 210 patients. The sample cannot detect rare events such as infection, and the findings may not transfer to other protocols or doses.
• Combined exposure. Every block contained both steroids, so the effect of each drug, or of dose, cannot be separated.
• Multiple testing and borderline results. No correction for multiple comparisons was specified, and the potassium result (P about 0.03) is marginal.

Where It Fits in the Course

Of the three clinical research papers in the list, this one is the smallest and most self-contained: a single-center retrospective cohort with a short methods section, one primary outcome and clear tables. It is therefore a good first paper to appraise before the multicenter variance study of inotrope use, which needs multilevel modeling, or the cholecystectomy cost analysis drawn from a large hospital database.

The books in the list supply the framework for judging it. How to Read a Paper is a brief guide to the questions to ask of a cohort study; Evidence-Based Medicine goes further into appraisal and applying results to individual patients; Testing Treatments explains, without technical language, why randomized comparisons are needed when observational data suggest an effect.

How to Use This Book

This article suits medical students, residents in anesthesiology, orthopedics or perioperative medicine, and anyone practicing critical appraisal. Basic familiarity with confidence intervals, p-values and the idea of confounding is enough; the mixed-effects model can be read for its purpose rather than its algebra.

A practical plan: read the abstract, then the methods before the results, and write down the exposure, comparison, outcomes and exclusions in your own words. Next read the results with the adjusted table in mind and ask whether a 20 mg/dL difference in glucose would change what you do for a patient. Then read the discussion to see how the authors defend the clinical judgment, and the limitations to compare with your own list. As an exercise, write the one-paragraph question a randomized trial would need to answer, including the groups, the primary outcome, the sample size consideration and what result would change practice.

Edition and Sources

Version covered: the open-access article as published in Journal of Pain Research 2023;16:553–561 (Dove Medical Press, Creative Commons Attribution–NonCommercial licence), freely available at https://pmc.ncbi.nlm.nih.gov/articles/PMC9946012/. This guide draws on the full text: abstract, introduction, methods, results, discussion and conclusion. The tables and figures were not consulted.