The study, published in Frontiers in Veterinary Science, reviewed medical records from August 2015 to January 2025 and identified 16 dogs with traumatic cervical fractures confirmed by computed tomography that caused signs of cervical myelopathy. Every dog underwent segmental vertebral stabilization with the aid of custom drill guides, followed by an immediate postoperative CT scan to verify implant position. The dogs ranged from a 1.5-kilogram Chihuahua puppy to a 50-kilogram American Bulldog cross, with a median body weight of 11.9 kilograms and a median age of just 2 years. Neurological status was graded retrospectively using the Modified Frankel Score, a five-point scale in which grade 1 denotes tetraplegia without superficial nociception and grade 5 denotes cervical pain alone. At presentation, seven dogs were grade 4, five were grade 3, three were grade 2, and one was grade 5.

The technology behind the guides is conceptually elegant. Surgeons first acquired thin-slice CT images of each dog’s cervical spine at 0.6-millimeter thickness and exported the DICOM data into medical image processing software. A three-dimensional virtual model of each affected vertebra was then built in computer-aided design software, allowing the surgical team to manipulate the bone digitally in every plane and plan each screw trajectory with a virtual cylinder matching the intended implant diameter. Trajectories were chosen to maximize bone purchase—including bicortical engagement through the far cortex—while keeping a safe distance from the vertebral canal, spinal cord, nerve roots, and transverse foramina. Screw diameter was selected so that the screw-to-pedicle-width ratio exceeded 75 percent, a threshold associated with stronger fixation in bone that is often barely wider than the implant itself.

Once the trajectories were finalized, a virtual drill guide was generated for each vertebra. The contact surface of the guide was designed as an inverted replica of the ventral vertebral cortex, so the finished device could seat in only one unique position on the bone—essentially a patient-specific jig that locks the drill angle and depth before the first millimeter of bone is cut. Both the vertebral models and the guides were printed in a biocompatible, autoclavable methacrylate photopolymer resin certified under ISO 10993-1 and USP Class VI standards. The entire workflow, from CT acquisition to guide delivery, took between 2 and 7 days in most cases, and each guide was trial-fitted on its printed vertebral model and steam-sterilized at 134 degrees Celsius for 20 minutes before surgery.

At surgery, each dog was positioned in dorsal recumbency for a ventral approach to the cervical vertebrae. After the ventral bone surface was cleared of soft tissue, the guide was seated and bicortical pilot holes were drilled sequentially; the guide was then removed and self-tapping stainless steel or titanium screws, ranging from 1.5 to 3.5 millimeters in diameter, were placed through the pedicles into the vertebral bodies. The protruding screw heads, typically left 5 to 10 millimeters proud of the cortex, were bonded together with polymethylmethacrylate bone cement to create a rigid construct, with copious irrigation to limit thermal injury during the exothermic curing process. Each cervical vertebra received between two and four pedicle screws depending on its size.

The accuracy results were striking. Of 90 pedicle screws placed across the cohort, 85—94.4 percent—were graded as fully contained within the pedicle and vertebral body on postoperative CT, using a modified Zdichavsky classification. Only five screws, or 5.6 percent, partially penetrated the medial pedicle wall, and no screw completely breached it. This compares favorably with historical freehand techniques, and the authors note that freehand cervical pedicle screw placement in dogs has been associated with vertebral canal violation rates of 57 to 100 percent in prior biomechanical work. Given that a significant violation of the canal can cause severe neurological deficits or death, the near-absence of medial breaches in this series underscores the value of preoperative digital planning, particularly in toy breeds whose pedicles may be narrower than 3 millimeters.

Neurological outcomes were also encouraging. Twenty-four hours after surgery, 10 of 16 dogs had an unchanged Modified Frankel Score, 5 had improved, and 1 showed a transient deterioration that resolved by discharge. At the time of hospital discharge, half the dogs had improved neurologically and four were neurologically normal and free of cervical pain. At the 4-to-6-week follow-up examination, available for 14 dogs, 8 had improved further and 6 remained stable since discharge. Only two dogs reported intermittent, positionally elicited neck pain at follow-up.

The long-term picture, gathered through a standardized telephone questionnaire completed by 12 of 16 owners a median of 4.3 years after surgery, was equally favorable. All 12 dogs were reported to be ambulatory, with 9 walking with a normal gait and 3 showing a weak gait with scuffing of the toenails. Every owner rated their dog’s neurological signs as improved since the last hospital visit, none reported current pain, and quality of life was scored 5 out of 5 by 11 owners and 4 out of 5 by one. Nine of the 12 dogs were still alive at the time of data collection; the other three had died of unrelated or age-related causes.

The study is not without caveats, and the authors are candid about them. Postoperative complications occurred in 10 of 16 dogs, most commonly regurgitation, coughing or gagging, dysphonia, and laryngeal edema—adverse events consistent with the ventral cervical approach and prolonged anesthesia. Nine of these resolved with medical management, but one dog developed persistent regurgitation requiring ongoing treatment, and one dog died approximately nine days after surgery, with persistent laryngeal edema and suspected hypoventilation considered the most likely cause. The retrospective design, small sample size, incomplete follow-up, and reliance on owner-reported long-term outcomes all limit the strength of the conclusions, and neurological grading was reconstructed from medical records rather than assigned prospectively.

Nevertheless, the findings carry practical weight for veterinary surgeons and, by extension, for the growing field of patient-specific implant navigation in both animals and humans. The median interval from presentation to surgery was 5.87 days, largely reflecting the time needed for guide design, fabrication, and stabilization of concurrent injuries, yet only one dog deteriorated neurologically while awaiting surgery—suggesting that, with appropriate immobilization and monitoring, a controlled delay for surgical planning is clinically acceptable in selected cases. The authors conclude that 3D-printed patient-specific drill guides represent a feasible option for stabilizing cervical fractures in dogs, but they emphasize that larger prospective comparative studies are needed before the technique’s advantages over freehand placement can be definitively established.

Subject of Research: Use of 3D-printed patient-specific drill guides for pedicle screw placement in cervical vertebral fracture stabilization in dogs

Article Title: Clinical outcome and implant accuracy in cervical fractures stabilization using 3D-printed patient-specific drill guides for placement of bicortical screws in 16 dogs

Article References: Clinical outcome and implant accuracy in cervical fractures stabilization using 3D-printed patient-specific drill guides for placement of bicortical screws in 16 dogs. (n.d.). https://doi.org/10.3389/fvets.2026.1850379

Image Credits: AI Generated

DOI: 10.3389/fvets.2026.1850379

Keywords: 3D printing, patient-specific drill guides, cervical fracture, dogs, veterinary neurosurgery, pedicle screws, spinal stabilization, Modified Frankel Score, polymethylmethacrylate, computed tomography, surgical accuracy, veterinary orthopedics