🔍 Key Findings Summary

• 62 fractures in 61 dogs (44 lateral condylar, 18 intracondylar); mostly Spaniels
• LEAP plate used in all cases; minimal intraoperative contouring needed (1 French Bulldog)
• Overall complication rate: ~33%, mostly minor; 1 amputation due to catastrophic infection
• Radiographic healing:
  • Lateral epicondylar part healed in 100%
  • Condylar part healed in ~61.5% LCF and ~57.1% ICF
• Functional outcomes:
  • 87% returned to full limb use
  • Median LOAD score: 2 for LCF, 6.5 for ICF
• Design adjustments made post-study to strengthen weak zones around 3rd–4th screw holes

🔍 Key Findings

• Combined ICTL and NIRF detected the same SLN in 80% of cases; each method alone missed sentinel lymph nodes in some dogs.
• ICLT failed in 5% of dogs, while NIRF failed in 20%; combined use resulted in only 5% failure in SLN detection.
• Metastatic lymph nodes were detected in 95% of dogs, despite most tumors being low or intermediate grade (95%).
• 27 of 41 nodes (65.8%) were histologically metastatic (HN2 or HN3), many of which would have been missed using ALN alone.
• ALN matched the SLN in only 45% of cases with ICTL and 30% with NIRF, supporting the need for SLN mapping.
• ICLT and NIRF were complementary, often identifying different SLN, with some only fluorescent or enhanced in one modality.
• Subcutaneous MCT had higher metastatic rates (7/8 dogs) than previously reported, challenging older assumptions about benign behavior.
• Removing both ALN and SLN (from both methods) improved detection of metastasis to 85–95%, reducing risk of understaging.

🔍 Key Findings

Study type: Cadaveric stifle arthroscopy in large breed dogs (n=12 joints from 6 dogs)
Goal: Describe and assess the proximal lateral portal for insertion of a Ventura stifle thrust lever (VSTL)
Main results:

• No damage to the long digital extensor tendon (LDE) with this portal
• VSTL could be placed without removing the arthroscope
• Portal creation time ~37 seconds (faster than previously reported)

Cartilage impact:

• Superficial iatrogenic articular cartilage injury (IACI) present in all specimens
• No difference in IACI between 5- and 10-minute lever durations
• Zone 4 (lateral femoral condyle) had significantly more damage than zones 1–3 (p < .05)

Conclusion: Portal was safe, repeatable, minimally invasive, and did not increase cartilage damage with up to 10-minute lever use

🔍 Key Findings

• 3D-printed cannulas (3DPCs) reduced mean surgical time significantly in cadaveric procedures (125.6 vs 95.2 min, p = 0.03).
• Cannula pullout events decreased from a mean of 10 to 2.2 per procedure when using only 3DPCs (p = 0.03).
• Instrument collisions were significantly fewer with 3DPCs (6.8 vs 2.6 collisions, p = 0.03).
• Live patients experienced no postoperative complications, including no incision site infections or discomfort.
• Initial versions of 3DPCs had minor issues, including valve leakage and looser trocar fit, requiring surgical workarounds.
• Customization of cannula shaft length (3 cm vs standard 5–8.3 cm) improved working space and reduced instrument interference.
• Production cost was under $5 per cannula, suggesting 3DPCs may be a cost-effective and reusable alternative for small patients.
• Study supports broader use of 3DPCs in laparoscopic procedures requiring long-jawed instruments or intricate tissue handling.

🔍 Key Findings

• Needle arthroscopy (NA) had 95% sensitivity and 100% specificity for diagnosing medial meniscal tears in dogs with CCLR.
• NA missed 1 medial and 1 lateral meniscal tear, both nondisplaced and stable.
• Visibility of all meniscal horns was significantly lower with NA compared to standard arthroscopy (SA) (P < .005 for all horns).
• Probing the caudal horn of the lateral meniscus was significantly more difficult with NA (P = .0017), though medial horn probing was similar.
• Mean NA procedure time was 8 ± 3 minutes, significantly shorter than SA (15 ± 9 min, P = .0041).
• No increase in lameness observed after NA, indicating minimal procedural morbidity.
• NA was performed under sedation in most dogs, though 10/26 required additional chemical restraint or short anesthesia.
• NA was most accurate for displaced vertical longitudinal tears, with all 14 correctly identified; one stable tear and a lateral tear were missed.

🔍 Key Findings

• 2.4 mm LCP with a 1.6 mm IM pin had the highest axial stiffness and yield strength among the tested constructs.
• Axial stiffness was significantly higher in the 2.4 mm LCP + 1.6 mm IM pin group compared to 2.7 mm LCP alone (p = .013).
• No significant difference in torsional stiffness was found among groups.
• 2.4 mm LCP + 1.0 mm pin had the lowest stiffness and failure load, underperforming both other constructs.
• All constructs failed via valgus bending, consistent with clinical observations in feline tibial fractures.
• A 1.6 mm pin (~50% canal fill) resulted in superior construct performance vs. 1.0 mm (~30% fill).
• Group 2 (2.4 LCP + 1.6 mm pin) outperformed the 2.7 mm LCP alone in stiffness, despite using a smaller plate.
• Plate–rod constructs may better preserve periosteal blood supply and support minimally invasive stabilization strategies.

🔍 Key Findings

• Minimally invasive repair of acetabular fractures using precontoured plates on 3D-printed models is feasible and technically reproducible in dogs.
• All cadavers had fracture gaps <2 mm and step defects <1 mm, indicating accurate reduction.
• Sciatic nerve injury was minimal or absent in all cases, supporting potential neuroprotection from indirect approaches.
• Pelvic angulation was maintained <5°, confirming preservation of alignment post-reduction.
• Surgical time averaged ~46 minutes in cadavers for both approaches and repair.
• Clinical case showed good radiographic healing by 8 weeks and full union by 3 months, with early weight-bearing post-op.
• Use of locking screws improved reduction fidelity, particularly across a broad plate span.
• 3D printing accelerated surgical planning, though its necessity remains debated due to the availability and cost concerns.

🔍 Key Findings

• SVAPs placed in axillary (aSVAP) or femoral/external iliac veins (fSVAP) during limb amputation remained functional in 92.3% and 100% of cases, respectively.
• Complication rates were lower in aSVAP (23.1%) and fSVAP (0%) compared to jSVAP (47.4%), although not statistically significant (P = .12).
• No catastrophic complications occurred with aSVAP or fSVAP; 2 deaths occurred with jSVAP due to port-related issues.
• Infection-related port removal was needed in 1 aSVAP (7.7%) and 2 jSVAPs (10.5%) — no removals were required in fSVAPs.
• Tip placement of the catheter (e.g., right atrium vs. vena cava) did not correlate with complications (P = .66).
• Shorter surgical time likely with aSVAP/fSVAP as they use the same surgical field as the limb amputation.
• Survival time median was similar across groups (jSVAP: 177 days, aSVAP: 125 days, fSVAP: 122 days).
• SVAP implantation during limb amputation offers a practical and safe alternative to separate jugular placement.

🔍 Key Findings

The study included 162 deep-pain-negative dogs undergoing decompressive surgery (hemilaminectomy) for acute thoracolumbar intervertebral disc extrusion (IVDE).

Ambulatory recovery occurred in 53.1% of dogs (86/162).

The best performing machine-learning model was XGBoost, with an AUC of 0.9502 and accuracy of 89.1%, outperforming Ridge, AdaBoost, and Naive Bayes models.

Preoperative-only XGBoost models were less accurate, with AUC dropping to 0.8271 and accuracy to 71.9%.

Top predictive features (by SHAP analysis) included:

1. T2-weighted to L2 spinal cord signal ratio (lower values predicted better outcome)
2. Use of fenestration (presence associated with better recovery)
3. Hospitalization duration
4. Imaging modality used
5. Duration of nonambulatory status

Machine learning provided better insight into prognostic factors than traditional statistical methods.

🔍 Key Findings

• Craniolateral approach (CLA) exposed significantly more radial surface area than craniomedial approach (CMA) (19.4 cm² vs. 13.8 cm²; p = 0.01).
• Proximal width of exposure was greater in CLA, especially at 12.5% length (P2 level, p = 0.016), aiding plate placement.
• No significant difference in exposed bone length between approaches.
• CLA avoided major neurovascular structures, making dissection cleaner and safer proximally.
• CMA consistently encountered median nerve/artery/vein, complicating proximal exposure.
• CLA allowed better access to proximal radius for locking plate application, which may benefit MIPO techniques.
• CLA also enables ulna fixation via the same incision, whereas CMA requires a separate skin incision.
• Anatomical tilt of the proximal cranial surface favored CLA, requiring less plate contouring than CMA for proper fit.