Patient selection for lumbar arthroplasty and arthrodesis: the effect of revision surgery in a controlled, multicenter, randomized study

Object. Patient selection is perhaps the most important factor in successful lumbar surgery. In this study, the authors analyzed the clinical outcomes of patients enrolled in the CHARITE investigational device exemption (IDE) trial who underwent revision surgery after primary total disc replacement with CHARITE or an anterior lumbar interbody fusion (ALIF) with placement of a BAK cage and iliac crest autograft. This revision surgery was either a supplemental posterior lumbar fixation or a 360 degrees fusion. Statistical analysis was conducted to compare clinical success in patients who underwent revision surgery with those who did not. Methods. The patients enrolled in the CHARITE IDE study were divided into 6 groups according to treatment and repeated operation status, and their Oswestry Disability Index (ODI) and visual analog scale (VAS) scores at the 2-year follow-up and at baseline were compared. The patients had received the following treatments by group: A) ALIF without reoperation; B) ALIF with conversion to 360 degrees fusion; Q arthroplasty (randomized) without repeated operation; D) arthroplasty with supplemental posterior lumbar fixation; E) arthroplasty (nonrandomized) without repeated operation; and F) arthroplasty (nonrandomized) with supplemental posterior lumbar fixation. Outcome scores in the groups of patients who required revision surgeries (Groups B, D, and F; 23 patients) were compared with the groups that did not require revision surgery (Groups A, C, and E; 299 patients). Results. Patients who required revision surgery had a significantly lower level of clinical improvement than those who did not. The mean change in ODI score was -53.0% in Groups A, C, and E, but just -12.7% in Groups B, D, and F (p < 0.0001). The mean change in VAS score was -59.1% in Groups A, C, and E, compared to -23.4% in Groups B, D, and F (p < 0.0001). No significant differences were identified in analyzing absolute change in scores and the percentage change. A comparison of outcomes in patients who had undergone arthroplasty without reoperation (Groups C and E) with all patients who had undergone revision surgery (Groups B, D, and F) demonstrated similar results (p < 0.0001). Conclusions. The 7.1% of patients who underwent a secondary stabilization procedure had poor clinical improvement. This finding may indicate that if the alternative treatment had been the initial treatment, these patients would not have benefited, and further implies a 7.1 % rate of imprecision in preoperative evaluation.