Threaded cortical bone dowels for lumbar interbody fusion: technique and short-term results

Abstract

The authors retrospectively reviewed a series of cases involving mechanical low-back or disogenic pain; 35 patients underwent lumbar interbody fusion in which threaded cortical bone dowels (TCBDs) were placed to treat degenerative disc disease. The series was composed of 18 females, and 17 males whose mean age was 46 years (range 17-76 years). There were nine smokers in the group. All patients presented with symptoms consistent with mechanical low-back or discogenic pain, and magnetic resonance imaging–documented degenerative changes and disc collapse greater than 50%, as compared with the adjacent normal-appearing level, were confirmed. Twenty-three patients underwent a posterior lumbar interbody fusion (PLIF) procedure for placement of the TCBD, whereas 12 underwent an anterior lumbar interbody fusion (ALIF) procedure for placement of the TCBD. In all patients undergoing PLIF procedures pedicle screw and rod constructs were used without posterolateral fusion except one. In all cases of ALIF except one TCBDs were used as “stand-alone” devices without supplemental fixation. All TCBDs were packed with morselized cancellous autograft prior to implantation. The success of fusion was determined at follow-up intervals and was defined as: the absence of lucency around the TCBD; an increase in subchondral endplate sclerosis; and the presence of bridging bone incorporating the anterior bone graft as demonstrated on static lumbar radiographs and/or computerized tomography scans. Stability was also determined by an absence of movement on dynamic lumbar radiographs. The degree of lumbar lordosis at the diseased level was measured immediately postoperatively and compared with the change in lordosis at follow up. Outcomes were assessed using a modified Prolo outcome scale and rated as excellent, good, fair, or poor. Excellent and good outcomes were considered satisfactory; fair or poor outcomes were considered unsatisfactory. In 27 patients radiographic and clinical follow-up results were considered adequate (nine ALIF and 18 PLIF patients). The mean follow-up duration was 7.9 months. Overall satisfactory outcome was 70%: a 77% satisfactory outcome in PLIF patients and a 55% in ALIF patients. Osseous fusion was present in 94% of the patients in the PLIF group and in 33% of those in the ALIF group. Complications included one L-5 nerve root injury and two postoperative wound infections, all in patients who underwent PLIF; there was also a case of breakout of one implant at 8 months postoperatively. The degree of vertebral body angulation measured at last follow up compared with the measurement obtained immediately postoperative was 3.4° of kyphosis in the ALIF group and 3.1° of kyphosis in the PLIF group, which represented an 11% and 9% loss of lordosis, respectively. Preliminary results indicate that there is a dramatically higher fusion rate in PLIF compared with ALIF procedures in which TCBDs are used. There is a corresponding trend seen in patient outcomes, but no distinct difference seems apparent in terms of restoration of lordosis when performing either procedure. The results suggest that TCBDs may best be used in PLIF procedures in conjunction with pedicle screws and rod constructs. Moreover, in patients in whom TCBDs and supplemental tension band constructs are used fusion rates appear to be comparable with those reported in other series but at a faster rate (94% at 7.9 months mean follow up). Longer follow-up periods and a larger series of patients are needed to confirm these preliminary observations.