Spinal Versus General Anesthesia for Acute Kidney Injury and Transfusion in One-Week-Staged Bilateral Total Knee Arthroplasty.

Background/

Objectives

Evidence on spinal versus general anesthesia in unilateral total knee arthroplasty (TKA) may not extend to one-week-staged bilateral surgery, where older patients receive two anesthetics in a short interval and intra-operative spinal-to-general conversion is common but rarely reported transparently. We compared peri-operative acute kidney injury (AKI) and transfusion between strategies in this setting.

Methods

We retrospectively analyzed 207 patients (414 surgeries) undergoing one-week-staged bilateral primary TKA at one center. Co-primary endpoints were creatinine-based AKI (patient level) and packed-red-blood-cell transfusion (surgery level). Because 42 general-anesthesia-classified surgeries had an attempted spinal injection, the primary analysis used the initial anesthetic plan (an intention-to-treat analogue), reclassifying these as spinal, with as-treated classification as a sensitivity analysis; AKI was modeled at the patient level (any general anesthesia versus spinal-spinal) and transfusion per surgery.

Results

Median age was 75 years and 82.6% were female; AKI affected 74 of 207 patients (35.7%) and transfusion 185 of 414 surgeries (44.7%). The adjusted any-general-anesthesia versus spinal-spinal estimate was not statistically significant and opposite the spinal-protective hypothesis (adjusted odds ratio 0.49, 95% confidence interval 0.23-1.01, p = 0.054), and no pre-specified sensitivity scenario survived Benjamini-Hochberg correction. Transfusion did not differ between strategies; among secondary endpoints, length of stay, hemoglobin drop, peak C-reactive protein, and intra-operative hypotension likewise showed no significant difference after multiplicity correction.

Conclusions

These hypothesis-generating findings do not support changing anesthetic practice; the choice should remain individualized. Approximately 12% of attempted spinal anesthetics converted intra-operatively to general anesthesia-a record-based observation, not a validated failure rate.