Local insulin application has a dose-dependent effect on lumbar fusion in a rabbit model

The purpose of this study was to determine if locally applied insulin has a dose-responsive effect on posterolateral lumbar fusion. Adult male New Zealand White rabbits underwent posterolateral intertransverse spinal fusions (PLFs) at L5-L6 using suboptimal amounts of autograft. Fusion sites were treated with collagen sponge soaked in saline (control, n = 11), or with insulin at low (5 or 10 units, n = 13), mid (20 units, n = 11), and high (40 units, n = 11) doses. Rabbits were euthanized at 6 weeks. The L5-L6 spine segment underwent manual palpation and radiographic evaluation performed by two fellowship trained spine surgeons blinded to treatment. Differences between groups were evaluated by analysis of variance on ranks followed by post-hoc Dunn's tests. Forty-three rabbits were euthanized at the planned 6 weeks endpoint, while three died or were euthanized prior to the endpoint. Radiographic evaluation found bilateral solid fusion in 10%, 31%, 60%, and 60% of the rabbits from the control and low, mid, and high-dose insulin-treated groups, respectively (p < 0.05). As per manual palpation, 7 of 10 rabbits in the mid-dose insulin group were fused as compared to 1 of 10 rabbits in the control group (p < 0.05). This study demonstrates that insulin enhanced the effectiveness of autograft to increase fusion success in the rabbit PLF model. The study indicates that insulin or insulin-mimetic compounds can be used to promote bone regeneration.