Similar overall success rate seen for single- and multi-level cervical disc replacement

Huppert J. Eur Spine J. 2011. doi:10.1007/s00586-011-1722-9.

Issue: April 2011

There was no major significant clinical difference between the results of patients who underwent single- and multi-level cervical disc replacement, according to prospective study results.

Jean Huppert, MD, and colleagues at Clinique du Parc prospectively analyzed 231 patients with cervical degenerative disc disease (DDD) who were treated with cervical disc replacement. All patients in the trial had completed 24 months follow-up. In all, 175 patients were treated at one level, and 56 were treated at two or more levels.

The investigators compared the groups using the Neck Disability Index (NDI), the Visual Analog Scale (VAS), range of motion and satisfaction scores. The investigators also documented and compared rates of complications and subsequent surgeries.

For both groups, the investigators found that mean NDI and VAS scores for neck and arm pain improved similarly. They also saw a similar improvement in mobility at the treated segments.

In the multilevel group, analgesic use was significantly higher while heterotopic ossification was significantly lower than in the single-level group. Patient satisfaction was nearly equal; 94.2% of single-level patients and 94.5% of multilevel patients reported that they would undergo the surgery again. The overall success rates were not significantly different.

“Multilevel DDD is a challenging indication in the cervical spine,” the authors wrote. “We need further studies to know more about the impact of multilevel arthroplasty, especially on the adjacent segments, but these results demonstrate initial safety and effectiveness in this patient sample.”

I read this article with some interest. A bit of background: Anterior cervical discectomy and fusion (ACDF) surgery was devised in the 1950s. Though there have been many improvements in the techniques and technology over the decades, the basic operative concept has stood the test of time. In properly selected patients, patient satisfaction rates should exceed 90%.

If asked, most spine surgeons would list the procedure as being among the most reliable and effective procedures that we have to offer spine patients. In fact, if one considers the data from the control groups of the recent prospective randomized clinical trials for single-level disc herniations in the United States (one of which I was the primary author: Bryan Disc [Medtronic]), the data attest to the reliability and effectiveness of the operation in a more rigorous way than we have ever been able to do in 60 years.

There are some theoretical and practical disadvantages to fusing a cervical spine segment. The most obvious one being the need to trade motion for symptom relief. There is also some concern about whether such fusions accelerate the wear on adjacent segments, increasing the likelihood of further surgery in the years ahead. This theory drove the initial efforts to develop artificial cervical discs. The completed and published randomized investigational device exemption (IDE) trials to date (ProDisc-C [Synthes Spine], Bryan and Prestige ST [Medtronic]) have shown that the disc replacement cohorts have done statistically significantly better than controls in two (Bryan and Prestige) and equivalent to control (ProDisc-C) through 4 years of follow-up. We (Sasso et al) just submitted for publication the minimum 5-year data for the Bryan study, with no deterioration in the results.

But there are conflicting findings as to whether there are lower rates of surgery at the adjacent levels with time. The Bryan data show equal reoperation rates (~5%) in control and disc replacement groups. While the other two trials show less reoperation for adjacent segment disease among the artificial disc cohorts. In short, the jury is still out.

So what does this have to do with the paper under review?

First, the authors’ findings, though lacking a concurrent fusion control group, yield clinical outcomes from a one-level cervical disc replacement (CDR) that are consistent with those of the United States clinical trials. It is reassuring that such results translate across oceans and borders, and they suggest that the observations in the U.S. trials are generalizable. It is particularly reassuring, as some have sought to discredit the findings of the FDA sanction IDE trials, alleging that they are tainted because they were sponsored by industry.

Second, it is widely assumed that if adjacent segment degeneration is an issue for one-level ACDFs, then it must be more so for multi-level procedures. The data remain unclear on this point. But, the authors have prospectively collected data on multi-level CDRs to see if there is any difference in the outcomes for their patients. It is well to note that 51/56 patients were two-level procedures. Thus, this is really a comparison of one-level versus two-level CDRs (even though their figures illustrate two three-level CDRs). The only statistically significant difference observed among their primary and secondary outcome measures was the greater use of analgesics among the multi-level cohort. But, the patient satisfaction rates and clinical improvement reflected in the neck disability index (NDI), SF-36 and VAS scores were every bit as favorable as with one-level surgeries. This supports the notion of efficacy of CDR for patients with need for surgery at two levels. I would personally reserve judgment about three or more levels.

As a final bit of perspective, one needs to be aware of the differing reasons for operating on three or more levels of cervical disease and the available treatment options. If the issue is one of spinal cord compression and myelopathy, there is much debate as to the ideal choice. There are proponents for multi-level anterior fusions, multi-level posterior laminectomies with segmental instrumentation and fusion, combined anterior and posterior decompression/fusion procedures, or laminoplasty.

For myelopathic patients, laminoplasty effectively decompresses the spinal cord while preserving motion (a fusion is not inherent to the procedure). However, laminoplasty is not well suited to certain multilevel pathologies. But, for those situations where laminoplasty is ill-suited to the situation, multi-level CDRs would likely be contraindicated as well.

Thus, I would suggest that in situations that thought to be properly indicated for CDR at three or more levels, laminoplasty would be a more time-tested and cost-effective surgical option.

But CDRs will never substantially replace fusions, as they are indicated for only a subset of patients who would otherwise require an anterior discectomy and fusion.

– John Heller, MD
Baur Professor of Orthopaedic Surgery
Spine Fellowship Director
The Emory Spine Center
Atlanta

Disclosure: Heller receives royalties from Medtronic; is on the speakers bureau for Abbott and Medtronic; receives institutional support from Abbott, Medtronic and Synthes; and owns stock in Medtronic.