The Use of a Novel Perfusion-Based Cadaveric Simulation Model with Cerebrospinal Fluid Reconstitution Comparing Dural Repair Techniques: A Pilot Study

Abstract Background Context Watertight dural repair is crucial for both incidental durotomy and closure after intradural surgery. Purpose To describe a perfusion-based cadaveric simulation model with cerebrospinal-fluid (CSF) reconstitution and to compare spine dural repair techniques. Study Design/Setting Fresh Tissue Dissection Laboratory. Sample Size 8 fresh human cadavers. Outcome Measures A watertight closure was achieved when pressurized saline up to 40 mm Hg did not cause further CSF leakage beyond the suture lines. Methods Fresh human cadaveric specimens underwent cannulation of the intradural cervical spine for intrathecal reconstitution of the CSF system. The cervicothoracic dura was then exposed from C7-T12 via laminectomy. The entire dura was then opened in 6 cadavers (ALLSPINE) and closed with 6-0 Prolene (n=3) or 4-0 Nurolon (n=3) and pressurized with saline via a perfusion system to 60 mm Hg to check for leakage. In 2 cadavers (INCISION), 6 separate 2 cm incisions were made and closed with either 6-0 Prolene or 4-0 Nurolon and then pressurized. A hydrogel sealant was then added and the closure was pressurized again to check for further leakage. Results Spinal laminectomy with repair of intentional durotomy was successfully performed in 8 cadavers. The operative microscope was used in all cases, and the model provided a realistic experience of spinal durotomy repair. For ALLSPINE cadavers (mean 240 mm dura/cadaver repaired), mean pressure thresholds for CSF leakage was observed at 66.7 (± 2.9) mm Hg in the 6-0 Prolene group and at 43.3 (± 14.4) mm Hg in the 4-0 Nurolon group (p >0.05). For INCISION cadavers, mean pressure thresholds for CSF leakage without hydrogel sealant were significantly higher in 6-0 Prolene group than the 4-0 Nurolon group (6-0 Prolene: 80.0 ± 4.5 mm Hg vs. 4-0 Nurolon: 32.5 ± 2.7 mm Hg; p < 0.01). Mean pressure thresholds for CSF leakage with the hydrogel sealants were not significantly different (6-0 Prolene: 100.0 ± 0.0 mm Hg vs. 4-0 Nurolon: 70.0 ± 33.1 mm Hg). The use of a hydrogel sealant significantly increased the pressure thresholds for possible CSF leakage in both the 6-0 Prolene group (p = 0.01) and the 4-0 Nurolon group (p < 0.01) when compared to mean pressures without the hydrogel sealant. Conclusions We described the feasibility of utilizing a novel cadaveric model for both the study and training of watertight dural closure techniques. 6-0 Prolene was observed to be superior to 4-0 Nurolon for watertight