68. Spinal fusion with a resorbable mesh pouch in a preclinical posterolateral model

BACKGROUND CONTEXT Delivery and handling of graft material during surgery can be challenging often due to the particulate nature of materials, irrigation and risk of migration. Resorbable mesh pouches (RMP) offer a simple and effective way to handle and implant particulate grafts, maintain host bone contact and avoid migration. The current RMP was composed of a PLGA mesh. Two studies were performed using a well reported single level posterolateral rabbit spinal fusion model. Study 1 evaluated the RMP prefilled with demineralized bone matrix (DBM) and used as an extender with autograft while Study 2 evaluated the RMP filled with corticocancellous iliac crest autograft. An Autograft alone group was used as a positive control group in each study. PURPOSE To confirm that a porous resorbable mesh pouch, used to facilitate placement and containment of graft material does not impede bone formation. STUDY DESIGN/SETTING Preclinical in vivo study. PATIENT SAMPLE Not applicable. OUTCOME MEASURES Not applicable. METHODS Eighty skeletally mature female New Zealand White rabbits were enrolled following ethical approval in two studies. The transverse processes were exposed and decorticated with a pneumatic high-speed burr. The implant groups were 2cc of autograft, 2cc of autograft inside the RMP, or RMP pre-filled with 2cc DBM placed on top of 1cc of autograft bilaterally in the posterolateral gutters. Fusion assessment at 3 days (n=2 per group), 4 and 8 weeks (n=5 per group) and 12 weeks (n=8 per group) included manual palpation, X-rays, Faxitron radiographs, micro CT, histology and histomorphometry for bone formation as well as polymer resorption and reaction using polarized light. Statistical analyses were performed using a non-parametric Kruskal Wallis test for manual palpation, radiographs and micro CT grading while an ANOVA was used for histomorphometry and non-destructive range of motion testing. RESULTS All animals recovered uneventfully following surgery. No adverse reactions were noted macroscopically. The positive control groups performed consistently with the model and literature. Manual palpation at 12 weeks for RMP groups was 6/8 in both studies. Radiograph and micro computed tomographic grading and pmMA histomorphometry did not detect any differences between the groups at any time point. Paraffin histology revealed no adverse reactions associated with the RMP. The RMP pre-filled with DBM and combined with autograft or filled with autograft alone remodeled in the same manner as the positive control. The RMP progressively resorbed with time and was nearly completely resorbed by 8 weeks with no adverse cellular or tissue reactions. CONCLUSIONS The RMP facilitated graft placement and handling and did not impede the biology of healing and remodeling of autograft in the single-level rabbit spine model. FDA DEVICE/DRUG STATUS Resorbable Mesh Bag (Ballast) (Approved for this indication)