Lumbar spinal stenosis - surgical outcome and the odds of revision-surgery: Is it all due to the surgeon?

BACKGROUND Surgical decompression is the intervention of choice for lumbar spinal stenosis (LSS) when non-operative treatment has failed. Apart from acute complications such as hematoma and infections, same-level recurrent lumbar stenosis and adjacent-segment disease (ASD) are factors that can occur after index lumbar spine surgery. OBJECTIVE The aim of this retrospective case series was to evaluate the outcome of surgery and the odds of necessary revisions. METHODS Patients who had undergone either decompressive lumbar laminotomy or laminotomy and spinal fusion due to lumbar spinal stenosis (LSS) between 2000 and 2011 were included in this analysis. Demographic, perioperative and radiographic data were collected. Clinical outcome was evaluated using numeric rating scale (NRS), the symptom subscale of the adapted version of the german Spinal Stenosis Measure (SSM) and patient-sreported ability to walk. RESULTS Within the LSS- cohort of 438 patients, 338 patients underwent decompression surgery only, while instrumentation in addition to decompression was performed in 100 cases (22.3%). 38 patients had prior spinal operations (decompression, disc herniation, fusion) either at our hospital or elsewhere. Thirty-five intraoperative complications were documented with dural tear with CSF leak being the most common (33/35; 94.3%). Postoperative complications were defined as complications that needed surgery and differentiated between immediate postoperative complications (⩽ 3 weeks post operation) and complications that needed revisions surgery at a later date. Within all patients 51 revisions were classified as immediate complications of the index operation with infections, neurological deficits and hematoma being the most common. Within this group only 22 patients had fusion surgery in the first place, while 29 were treated by decompression. Revision surgery was indicated by 53 patients at a later date. While 4 patients decided against surgery, 49 revision surgeries were planned. 28 were performed at the same level, 10 at the same level plus an adjacent level, and 10 were executed at index level with indications of adjacent level spinal stenosis, adjacent level spinal stenosis plus instability and stand-alone instability. Pre- operative VAS score and ability to walk improved significantly in all patients. CONCLUSIONS While looking for predictors of revision surgery due to re-stenosis, instability or same/adjacent segment disease none of these were found. Within our cohort no significant differences concerning demographic, peri-operative and radiographic data of patients with or without revision wer noted. Patients, who needed revision surgery were older but slightly healthier while more likely to be male and smoking. Surprisingly, significant differences were noted regarding the distribution of intraoperative and early postoperative complications among the 6 main surgeons while these weren't obious within the intial index group of late revisions.