Minimally invasive transforaminal lumbar interbody fusion (TLIF): technical feasibility and initial results.

Forty-nine patients underwent minimally invasive transforaminal lumbar interbody fusion (TLIF) from October 2001 to August 2002 (minimum 18-month follow-up). The diagnosis was degenerative disc disease with herniated nucleus pulposus (HNP) in 26, spondylolisthesis in 22, and a Chance-type seatbelt fracture in 1. The majority of cases (n = 45) were at L4-L5 or L5-S1. A paramedian, muscle-sparing approach was performed through a tubular retractor docked unilaterally on the facet joint. A total facetectomy was then conducted, exposing the disc space. Discectomy and endplate preparation were completed through the tube using customized surgical instruments. Structural support was achieved with allograft bone or interbody cages. Bone grafting was done with local autologous or allograft bone, augmented with recombinant human bone morphogenetic protein-2 in some cases. Bilateral percutaneous pedicle screw-rod placement was accomplished with the Sextant system. There were no conversions to open surgery. Operative time averaged 240 minutes. Estimated blood loss averaged 140 mL. Mean length of hospital stay was 1.9 days. All patients presenting with preoperative radiculopathy (n = 45) had resolution of symptoms postoperatively. Complications included two instances of screw malposition requiring screw repositioning and two cases of new radiculopathy postoperatively (one from graft dislodgement, the other from contralateral neuroforaminal stenosis). Narcotic use was discontinued 2-4 weeks postoperatively. Improvements in average Visual Analogue Pain Scale and Oswestry Disability Index (preoperative to last follow-up) scores were 7.2-2.1 and 46-14, respectively. At last follow-up, all patients had solid fusions by radiographic criteria. Results of this study indicate that minimally invasive TLIF is feasible and offers several potential advantages over traditional open techniques.