Institutional members access full text with Ovid®

Does Symptom Duration Correlate Negatively With Outcome After Posterior Lumbar Interbody Fusion for Chronic Low Back Pain?

Marshman, Laurence A. G. MD, FRCSN; Kasis, Ata MRCS; Krishna, Manoj FRCS, MCh(Orth); Bhatia, Chandra K. MOrth

doi: 10.1097/BRS.0b013e3181bb8ebd
Health Services Research

Study Design. Original report.

Objective. To investigate the putative negative correlation between the duration of symptoms (DOS) and outcome after surgery for chronic low back pain (CLBP).

Summary of Background Data. Posterior lumbar interbody fusion (PLIF) is a well established treatment for CLBP. Anecdotally, a prolonged DOS is associated with a poor prognosis for recovery of CLBP. In one recent study, a DOS greater than 3 years predicted a poor prognosis for subsequent clinical improvement with CLBP.

Methods. Patients (n = 209) underwent PLIF for CLBP who had proven unresponsive to nonoperative management for at least 6 months. A wide variety of physical and mental outcome scores were simultaneously assessed pre- and after surgery: i.e., the Oswestry Disability Index (ODI), SF-36 body score, SF-36 mental score, Visual Analogue Score (VAS) for back pain, VAS for leg pain, Hospital Anxiety Score (HAS), and Hospital Depression Score.

Results. Despite a prolonged mean DOS of 84.3 ± 6.6 months, there was a significant postoperative improvement in all 7 outcome scores after PLIF. Significant improvement occurred in 181 patients (i.e., 86.6%) and was sustained at 51.6 ± 12.0 months follow-up. No significant correlation was found between DOS and any outcome score (ODI: rs = 0.013, P = 0.877; SF-36 bodily pain: rs = 0.013, P = 0.87; VAS for back pain: rs = 0.038, P = 0.656; VAS for leg pain: rs = 0.086, P = 0.310; HAS: rs = 0.511, P = 0.056; Hospital Depression Score: rs = 0.056, P = 0.509, or SF-36 mental score rs = 0.007, P = 0.935). No arbitrary DOS “cut-off” was found for which significantly different outcomes were recorded either side of the cut-off; or for which a significant correlation was revealed either side of the cut-off. Finally, no significant partial correlation was found between DOS and any outcome score after controlling for pain severity (VASback pain) before surgery. There were no significant differences in terms of age, sex, or DOS between those with improved ODI scores less than 10 compared with those with improved ODI scores greater than 10.

Conclusion. The putative negative correlation between DOS and outcome was not observed under any analysis in our study. PLIF procured a rapid and sustained improvement in CLBP, even where the DOS was excessively prolonged; and even after having allowed for pain severity. Symptom chronicity, therefore, does not represent a poor prognostic indicator for CLBP outcome after PLIF: PLIF should be considered irrespective of DOS. Because DOS and pain severity are likely mediators of “central sensitization,” the hypothesis that central sensitization may be prevalent in CLBP patients selected for PLIF is therefore questioned.

The putative negative correlation between duration of symptoms and outcome after posterior lumbar interbody fusion for chronic low back pain was not observed across 7 outcome scores in our study. Symptom chronicity does not represent a poor prognostic indicator for chronic low back pain outcome after PLIF. PLIF should be considered irrespective of duration of symptoms.

From the Department of Spinal Surgery, University Hospital of North Tees, Hardwick, Stockton on Tees, United Kingdom.

Acknowledgment date: June 25, 2008. First revision date: September 1, 2008. Second revision date: April 2, 2009. Third revision date: May 26, 2009. Acceptance date: July 1, 2009.

The device(s)/drug(s) is/are FDA-approved or approved by corresponding national agency for this indication.

No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.

Manoj Krishna is a consultant for De Puy and has shares in GPOS.

This article has been significantly contributed to by all named authors.

Address correspondence to reprint requests to Laurence A. G. Marshman, MD, FRCSN, Pollard Cottage, Church Rd, Lingfield, Surrey, RH7 6AH; E-mail: l.a.g.marshman@btinternet.com

© 2010 Lippincott Williams & Wilkins, Inc.