Anesthesia & Analgesia · 2013 · 34 citations · 16 references
MedicineOrthopaedic SurgerySurgeryOrthopedicsPain ManagementSpine SurgeryEpidural AnesthesiaFailed Back Surgery SyndromePain OutcomesSpinal DisordersEpidural ComplicationsLogistic RegressionMedical Records ReviewMultivariable Regression AnalysisRetrospective Case Study
BACKGROUND: Failed back surgery syndrome (FBSS) is a challenging problem. One treatment advocated to treat FBSS is epidural lysis of adhesions (LOA). The results of studies examining LOA for FBSS have been mixed, but are limited because no study has ever sought to identify factors associated with outcomes. METHODS: We performed this multicenter, retrospective study in 115 patients who underwent LOA for FBSS (n = 104) or spinal stenosis (n = 11) between 2004 and 2007. Twenty-seven demographic, clinical, and procedural variables were extracted from medical records and correlated with the outcome, defined as ≥50% pain relief lasting ≥1 month. Univariable analysis was performed, followed by multivariable logistic regression. RESULTS: Overall, 48.7% (95% confidence interval [CI], 39.3%-58.1%) of patients experienced a positive outcome. In univariable analysis, those who had a positive outcome were older (mean age 64.1 years; 95% CI, 59.7-68.6 vs 57.2; 95% CI, 53.0-61.4 years; P = 0.02), while higher baseline numerical rating scale pain scores were associated with a negative outcome (mean 6.7 years; 95% CI, 6.0-7.3 vs 7.5; 95% CI, 6.9-8.0; P = 0.07). Use of hyaluronidase did not correlate with outcomes in univariable analysis (odds ratio [OR], 1.2; 95% CI, 0.6-2.5; P = 0.65). In multivariable analysis, age ≥81 years (OR, 7.8; 95% CI, 1.4-53.7), baseline numerical rating scale score ≤9 (OR, 4.4; 95% CI, 1.4-16.3, P = 0.02), and patients on or seeking disability or worker's compensation (OR, 4.4; 95% CI, 1.1-19.5, P = 0.04) were significantly more likely to experience a positive outcome. CONCLUSIONS: Considering our modest success rate, selecting patients for epidural LOA based on demographic and clinical factors may help better select treatment candidates. Procedural factors such as the use of hyaluronidase that increase risks and costs did not improve outcomes, so further research is needed before these become standard practice.
16
Cross‐validation in survival analysis
Hans C. van Houwelingen · Statistics in Medicine · 1993 · 272 citations
Medicine, Biostatistics, Oncology +16
Revision Surgery Following Operations for Lumbar Stenosis
Richard A. Deyo, Brook I. Martin, William Kreuter et al. · Journal of Bone and Joint Surgery · 2011 · 171 citations