Image-Guided Cryoablation of the Spine in a Swine Model: Clinical, Radiological, and Pathological Findings with Light and Electron Microscopy

Abstract

This study was designed to present the feasibility of an in vivo image-guided percutaneous cryoablation of the porcine vertebral body. Methods The institutional animal care committee approved this study. Cone-beam computed tomography (CBCT)-guided vertebral cryoablations (n = 22) were performed in eight pigs with short, 2-min, single or double-freezing protocols. Protective measures to nerves included dioxide carbon (CO2) epidural injections and spinal canal temperature monitoring. Clinical, radiological, and pathological data with light (n = 20) or transmission electron (n = 2) microscopic analyses were evaluated after 6 days of clinical follow-up and euthanasia. Results CBCT/fluoroscopic-guided transpedicular vertebral body cryoprobe positioning and CO2 epidural injection were successful in all procedures. No major complications were observed in seven animals (87.5 %, n = 8). A minor complication was observed in one pig (12.5 %, n = 1). Logistic regression model analysis showed the cryoprobe-spinal canal (Cp-Sc) distance as the most efficient parameter to categorize spinal canal temperatures lower than 19 °C (p<0.004), with a significant Pearson’s correlation test (p < 0.041) between the Cp-Sc distance and the lowest spinal canal temperatures. Ablation zones encompassed pedicles and the posterior wall of the vertebral bodies with an inflammatory rim, although no inflammatory infiltrate was depicted in the surrounding neural structures at light microscopy. Ultrastructural analyses evidenced myelin sheath disruption in some large nerve fibers, although neurological deficits were not observed. Conclusions CBCT-guided vertebral cryoablation of the porcine spine is feasible under a combination of a short freezing protocol and protective measures to the surrounding nerves. Ultrastructural analyses may be helpful assess the early modifications of the nerve fibers.