Introduction: Lumbar intervertebral disc degeneration (IVDD) impacts up to 90% of the aged population, with > 25% of Americans reporting low back pain (1, 2). Early degenerative changes typically occur in the central nucleus pulposus (NP), where progressive dehydration from proteoglycan loss compromises mechanical function. Our team has developed a non-crosslinked injectable NP replacement hydrogel based upon a polyvinyl alcohol and polyvinyl pyrrolidine co-polymer, treated with polyethylene glycol to yield a thermally processible hydrogel with 59-66% water. The objective of this study was to use chondroitinase ABC (CABC) to induce IVDD through degradation of NP proteoglycans (4) and then treat with our injectable hydrogel.

Methods: Three adult goats (50-80kg) underwent intradiscal injection of 1.5U CABC to induce IVDD at levels L1-2, L2-3, and L4-5(4). Two weeks later, the degenerated NP was replaced with injectable hydrogel under fluoroscopic guidance (Fig. 1) in levels L2-3 and L4-5; level L1-2 served as an untreated degenerated control (CABC) and level L3-L4 served as a native negative control. All animals recovered uneventfully and were maintained in a 20x30ft pen for 2.5-3-years. Animals were evaluated daily for pain, lameness, and neurologic deficits. Monthly standing lateral radiographs were used to calculate disc height index (DHI) through MatLab (5). After 2.5-3 years, contrast-enhanced MRI at 3T was performed to obtain T1 maps before and 30 minutes post-administration of 0.1 mmol/kg gadodiamide to assess trans-endplate small molecule diffusion, which is shown to be reduced with progressive end plate ossification seen in IVDD (6,7).

Results: After a minimum of 2.5 years, 4 of 6 hydrogels injected remained in the IVD (Fig. 2). In one animal, delivery assembly malfunction during injection caused dorsal rupture of the annulus fibrosus (AF) and hydrogel leakage into the vertebral canal. One hydrogel migrated laterally into the surrounding soft tissue at 20 months post-operatively. No lameness or neurologic deficits were observed. Longitudinal DHI showed hydrogel-treated discs maintained the same height as non-degenerated discs (Fig. 3). Contrast-enhanced T1 MRI revealed no difference in end-plate diffusion between hydrogel-treated and native discs. However, there was a reduction in T1 relaxation time change from pre- to post- contrast injection when comparing native discs to CABC only discs, suggesting that small molecule trans-endplate diffusion was reduced in untreated degenerating discs (Fig. 4).

Discussion: Taken together, our clinical findings demonstrate that hydrogel NP replacement mitigates disc degeneration and end-plate changes in an animal model of IVDD. These results are promising in support of minimally invasive injectable therapies for hydrogel-augmentation in early-onset human IVDD. This study has a small sample size, making statistical differences difficult to detect given the heterogeneity of measurements. Ongoing work is focused on increasing sample size and performing correlations across quantitative measures of native, hydrogel treated, and untreated degenerating discs (CABC).