Hi, my name is Andrew Chan, and I'm one of the PGY6s at UCSF. And today, I'd like to present to you the culmination of the QOD Spondylolisthesis Study Group's investigation into the effectiveness of fusion for lumbar spondylolisthesis, titled Decompression and Fusion vs. Decompression Alone for Grade 1 Degenerative Lumbar Spondylolisthesis. Here are my disclosures. I'd like to thank the NREF for their contribution to our study. We know that lumbar spondylolisthesis is associated with disability and affects many people worldwide, including up to 12% of the United States population. For a subset of patients with symptoms who fail conservative management, surgical intervention has been shown to be effective in improving function and quality of life. To this point, in a 2007 New England Journal of Medicine as-treated analysis of participants with spondylolisthesis in the SPORT trial, the study showed a significant benefit of surgery compared to conservative management alone at two-year follow-up. However, what the study did not answer was what specific procedure patients should undergo, especially because greater than 90% of those undergoing surgery underwent fusion procedures. For the patients we see in clinic with spondylolisthesis, we are faced with the decision of whether to decompress the patient without fusion or decompress that patient with the addition of fusion, and the ideal surgical approach remains controversial despite some randomized controlled trials on the topic. In 2016, two RCTs investigating the utility of fusion for spondylolisthesis were published in the New England Journal of Medicine. In one study by Dr. Gogolwal and colleagues, in a study of stable, single-level spondylolisthesis, they found that those who underwent fusion in addition to decompression had a greater improvement in health-related quality of life using SF36, their primary outcome metric. Additionally, they found that those receiving decompression alone had a nearly 20% higher rate of re-operation within the study period. On the other hand, in the study by Dr. Forst and colleagues, in a study of one- or two-level lumbar stenosis with and without spondylolisthesis, they found no additional benefit of fusion for patients with regards to ODI, their primary outcome metric. Additionally, their rates of re-operation between the two cohorts were similar. So why were these results different? Well, for one, there were different populations studied, with the former studying a population of stable, single-level spondylolisthesis, and the other studying a population with one or two levels of lumbar stenosis with and without spondylolisthesis. Also, the studies studied different outcome measures, with one studying SF36 and the other studying ODI. And some authors have suggested that the fourth study did not have enough power to detect a difference for their primary outcome, ODI. Additionally, an overarching issue in limitation of RCTs stems from the narrow inclusion criteria employed by RCTs, which may not be applicable to the patients we actually see in clinic. To investigate the real-world practice experience of those managing spondylolisthesis, we leveraged the Prospective Quality Outcomes Database Registry. This registry has over 100 enrolled sites and includes over 80 demographic, surgical, clinical, and outcome variables. Twelve of the highest-enrolling sites, who had full-time data coordinators to ensure the acquisition of high-quality data, formed a focus study group to investigate the effectiveness of fusion for grade 1 degenerative lumbar spondylolisthesis. This represents a multi-center and multi-disciplinary effort. We prospectively enrolled patients from July 2014 through June 2016 and followed them for two to three years. These were all patients that underwent single-segment or single-disc level surgery for grade 1 degenerative lumbar spondylolisthesis. Overall, we enrolled 608 patients, of which 140 underwent decompression alone, and 468 underwent decompression and fusion. We compared outcomes at 24 months, including patient-reported outcomes for our primary outcome measure of ODI. We also investigated secondary outcome measures, including EQ5D, NRS back pain, NRS leg pain, and NAS satisfaction. We also compared readmission rates, cumulative re-operation rates, and also 30-day complication rates. Now, if we compare the two cohorts, you'll notice that there are multiple variables that are significantly different between them, and I've highlighted those differences in yellow here. For baseline demographics, the fusion cohort was younger, had a higher proportion of females, had a higher BMI, and were more often depressed, but less often had diabetes mellitus. For baseline patient-reported outcomes, fusions had a higher disability, worse back pain, and poor quality of life. If we look at their clinical presentation, fusions less often had motor deficits at presentation, and thus were more often independently ambulatory at the time of presentation. Additionally, there was a higher proportion of a symptom duration of over three months in the fusion cohort. If we look at socioeconomic variables, fusions had lower levels of education, but were more often employed. Now if we compare the perioperative outcomes between the two groups, you'll note that the fusion cohorts had more blood loss, longer operative times, and longer hospitalizations. However, the discharge disposition was not significantly different between the two cohorts, with both groups achieving around a 90% rate of discharge disposition to home or home health. If we compare patient-reported outcomes in a univariate fashion at 24 months, one, you'll notice that both groups improved significantly from baseline, but that fusions had a greater improvement in disability and back pain. Additionally, fusions had a higher rate of reaching ODI MCID, with 73% of patients in the fusion cohort reaching MCID, but only 56% in the decompression alone cohort reaching MCID. Fusions also had a 96% rate of fusion. There were no significant differences in univariate analysis, however, for NRS leg pain, EQ5D, or mass satisfaction. If we look at cumulative reoperation rate, 90-day readmission rate, and 30-day complication rates, there were no significant differences between the cohorts. However, if you zero in on the cumulative reoperation rate, you'll notice a trend towards higher reoperation in the decompression alone group at 9.3% versus 6.2%. Now zeroing in on the reasons for reoperations by cohort, we found that there were differences. If we look at the 13 reoperations in the decompression-only cohort, we found that 7 were for revision decompressions, whereas 6 were for transitions to fusions. In the fusion cohort, there were 31 reoperations in 29 patients. These were for 1 revision decompression for adjacent segment disease. There were 13 revision fusions, of which 8 were for adjacent segment disease, and 5 were for pseudoarthrosis. And there were 17 miscellaneous reasons, including 8 surgical site infections, 6 implant revisions and removals, 1 hematoma evacuation, 1 revision for suture granuloma, and 1 spinal cord stimulator placement. Additionally, if you look at the timing of reoperations, they differed also. If you bin the reoperations into windows less than 30 days, 30 days to 1 year, 1 to 2 years, and 2 to 3 years postoperatively, as was collected in the study, you'll see that in the first 30-day period, decompression alone had no reoperations, but decompression and fusions had 11 of their reoperations in this period, which was significantly different. On the other hand, from 30 days to 1 year, the decompression alone cohort had a significantly higher proportion of reoperations compared to the decompression and fusion cohort. There were no significant differences in the proportion of reoperations from 1 to 2 years and 2 to 3 years between the cohorts. Now, if you remember, there were significant differences amongst many variables at baseline between the two cohorts, so it's important when we're investigating effectiveness of fusion to adjust for those baseline differences, and we did just that in multivariable adjusted analysis. And despite adjusting for these different factors, we found that fusion was still independently associated with greater ODI improvement, greater NRS back pain improvement, and nearly two times odds of reaching MCID for ODI compared to decompression alone. Additionally, there was a higher odds of satisfaction in the cohort undergoing fusion. There were no significant differences for leg pain or EQ5D, however. So in conclusion, in multivariable risk-adjusted analyses, decompression and fusion was superior for disability, reaching a clinically meaningful improvement in disability, back pain, and patient satisfaction. The readmission and reoperation rates do not significantly differ between the procedures, though based on our experience with previous work, reoperation rates may further diverge with time. The bottom line is that when a spine surgeon selects the procedure they think is best for a given patient with lumbar spondylolisthesis, the addition of fusion is effective for the treatment of grade one degenerative lumbar spondylolisthesis. I'd like to thank everyone that makes the spondylolisthesis study group possible, including those that spearheaded it, including Dr. Praveen Moumaneni, Dr. Erika Bisson, Dr. Mohammed Biden, and Dr. Tony Asher. I'd like to thank the remainder of the QOD lumbar spondylolisthesis study site PIs, the research coordinators, the NPA, and the NREF, without which this study would not have been possible. Thank you so much for allowing me to contribute today. If you have any remaining questions, please do reach out by email here. Thank you so much.

fusion

lumbar spondylolisthesis

disability

patient satisfaction

prospective outcomes