Impact of perioperative complications on clinical outcome scores in lumbar fusion surgery

Abstract

Although lumbar fusion is effective in well-selected patients, it is not without complications associated with short-term morbidity. There is a paucity of literature on the effect of these complications on long-term clinical outcomes. The purpose of this study was to determine whether perioperative complications-that is, those occurring within 30 days after surgery-alter the long-term clinical outcomes after lumbar fusion. The authors retrospectively reviewed surgical and clinical databases for the period from 2001 to 2008 to identify patients who had undergone instrumented lumbar spinal fusion and had complete preoperative and 2-year postoperative outcome measures data. Outcome measures included the 36-Item Short-Form Health Survey (SF-36) Physical Component Summary, SF-36 Mental Component Summary, Oswestry Disability Index (ODI), and Numeric Rating Scales (0-10) for back and leg pain. Three patient groups were created for comparison-one with major complications, one with only minor complications, and another with no complications-using propensity matching techniques based on demographics, baseline clinical outcome scores, and surgical characteristics. Preoperative and 2-year postoperative outcome scores in the groups were compared. One-way ANOVA was used to compare continuous variables, and the Fisher exact test was used to compare categorical variables between the groups. Significance was set at p < 0.001. In the database with 1144 patients, 81 had a major complication. Of these 81 patients, 78 were propensity matched to a similar group of patients with minor complications and another group with no complications. Comparison of the 3 groups revealed that 2-year postoperative outcomes were not statistically different for any of the measures. Overall ODI at 2 years was better in patients having no complications (39.6) or only minor complications (37.0) than in those having major complications (44.5), but this difference was not statistically significant (p = 0.074). The proportion of patients reaching a minimum clinically important difference (MCID) for ODI was statistically significantly smaller in the major complication group (31%) than in the minor complication (51%) and no complication groups (65%; p < 0.001). A smaller proportion of patients achieved MCID for ODI 2 years after a major perioperative complication following lumbar fusion than after a minor complication or no complications.