Minimal Access Versus Open Transforaminal Lumbar Interbody Fusion Meta-Analysis of Fusion Rates

Study Design. A quantitative meta-analysis was conducted on published studies reporting fusion rates after open or minimally invasive/mini-open transforaminal lumbar interbody fusion (TLIF) procedures for single or multilevel degenerative disease including stenosis with spondylolisthesis and degenerative disc disease. Objectives. The primary aim of this study was to establish benchmark fusion rates for open TLIF and minimally invasive TLIF (mTLIF) based on published studies. A secondary goal was to review complication rates for both approaches. Summary of Background Data. Lumbar fusion for the treatment of degenerative disease has evolved from a purely posterior noninstrumented approach to a combination of anterior and/or posterior surgery with instrumentation. The increasingly popular transforaminal approach has advanced to incorporate minimally invasive spinal techniques. There currently exist no controlled comparisons between open TLIF and mTLIF. Methods. A Medline search was performed to identify studies reporting fusion rate on open TLIF or mTLIF with instrumentation. A database including patient demographic information, fusion rate, and complication rate was created. Fusion and complication rates were pooled according to whether TLIF was performed with open or minimally invasive technique. Publication bias was assessed with Egger's test, and adjustments were performed using Duval and Tweedie's Trim and Fill algorithm. Results. Twenty-three articles were identified that fit inclusion criteria. In each of the 23 studies, TLIF was performed with pedicle fixation and fusion was evaluated using radiograph or computed tomography scan at minimum 6-month follow-up. Overall, the studies included 1028 patients, 46.8% of which were female. The mean age of all patients was 49.7 (range, 38-64.9), and mean follow-up interval for assessment of fusion was 26.6 months (range, 6-46 months). The usage of recombinant bone morphologic protein was higher in the mTLIF group (50% vs. 12%). Mean fusion rate from 16 studies (716 patients) of open TLIF was 90.9%, whereas mean fusion rate from 8 studies (312 patients) of mTLIF was 94.8%. Complication rate was 12.6% and 7.5% for open and mTLIF, respectively. Conclusion. Fusion rates for both open and mTLIF are relatively high and in similar ranges. Complication rates are also similar, with a trend toward mTLIF having a lower rate. This analysis provides clear benchmarks for fusion rates in open and mTLIF procedures for spine surgeons.