November 2006

Lumbar Spine Surgery At A Crossroads
By Alan Scarrow, MD, JD, Chairman, St. John's Neurosurgery

Low back pain has become a significant public health problem in the U.S. Approximately 80 percent of all people will have back pain at some point in their life and back pain is second to upper respiratory problems as a reason to see a physician.

Fortunately, 80-90 percent of all patients with back pain will have their symptoms resolve within six weeks of onset. Nonetheless there remains a large number of Americans who despite good conservative therapy continue to have significant low back problems. As a result, demand for a surgical ‘solution’ to low back problems has been significant. 

Lumbar spine surgery has become the third most common surgical procedure performed in American hospitals. However, wide differences in the rate of lumbar spine surgery exist within the U.S.  Additionally there is almost no consensus on the indications for lumbar spine surgery. This became particularly evident when the American Association of Neurological Surgeons and Congress of Neurological Surgeons Joint Section on Spinal Disorders published an exhaustive set of guidelines for the performance of all types of spine surgery and found alarmingly little Class I evidence from which to draw their conclusions. 

This has led many spine surgeons, patients, and insurers to question the current state of affairs. Over the next several issues of this bulletin, I will review the literature on lumbar spine surgery with special attention to what has been proven and what remains unanswered. This first article will look at the data on lumbar fusion surgery for all-comers regardless of diagnosis. Subsequent articles will review surgical outcome data on specific lumbar spine pathologies. 

Does fusion matter in lumbar spine surgery?

Spine surgeons spend a lot of time talking and worrying about whether or not we can achieve a proper fusion (the surgically induced bony union of adjacent spine levels) following spine surgery. Much of this worry seems to rise from the orthopedic experience where fusion of the long bones of the extremities is clearly important. Yet in spine surgery, while the importance of decompressing the neural structures (spinal cord and nerve roots) is unequivocal, the utility of fusion has become less clear. 

There are 3 Class I studies (prospective, randomized controlled trials) that have sought to answer the question of whether or not fusion matters in lumbar spine surgery. The first looked at a group of 114 patients undergoing lumbar fusion for degenerative disc disease, stenosis, scoliosis or spondylolisthesis. 

All patients were treated with decompression (laminectomies of the lumbar spine) and then randomized to one of three groups: autologous bone graft (ABG), ABG plus semirigid pedicle screw fixation, or ABG plus rigid pedicle screw fixation. At the end of 16 months of follow up, the group with ABG plus rigid pedicle screw fixation had the highest fusion rate (95 percent) but there was no significant difference in outcome (pain and functionality) between the groups.

The second study looked at a similar group of 69 patients with a variety of low back pathologies. This study compared patients treated with decompression and pedicle screw fixation and ABG versus a group with just ABG.  At the last follow up 40 months there was no difference between the two groups in pain and overall patient satisfaction. Fusion rates for the group with pedicle screw fixation and ABG were higher (76 pecent versus 64 percent). However, there was no statistical difference in outcome between patients who had radiographic evidence of a solid fusion and those that didn’t.

The third study reviewed a group of 48 patients with structural spine problems potentially amenable to fusion who had failed conservative therapy. This study looked at the difference in outcome for patients undergoing lumbar decompression with pedicle screw fixation and either anterior lumbar interbody fusion (ALIF) or posterior lumbar interbody fusion (PLIF). Three years of follow up showed high rates of fusion for both groups but no difference in outcome between the two groups.

In summary, when looking at the little Class I evidence that exists for all-comers undergoing lumbar spine surgery, radiographic fusion rates are higher for patients having instrumented fusion (pedicle screws with or without interbody devices). However, there is not any statistically detectable difference in outcome (pain, functionality or overall satisfaction) between these groups and no apparent attributable difference to the effect of a solid fusion in the spine. 

To many of us this seems counterintuitive and against what we have been taught. Nonetheless, the data speaks for itself and may beg us to think more critically about what we thought to be the universal truths of lumbar spine surgery.

In subsequent articles we will look into this issue of lumbar spine surgery efficacy in more detail by reviewing data on lumbar spine surgery for specific diagnosis such as lumbar stenosis, chronic low back pain and spondylolisthesis.

To refer a spine patient to Dr. Scarrow, please call his office at 417-820-5150 or e-mail him at ascarrow@sprg.mercy.net.

Read more about St. John's Spine Center