SPORTing Statistics: Slicing the Drive on Appropriate Treatment for Herniated Lumbar Discs
Abstract & Commentary
By Justin F. Fraser, MD, Roger Hartl, MD, and John Boockvar, MD, Department of Neurological Surgery Weill Medical College of Cornell University, New York, NY. Dr. Fraser, Dr. Hartl, and Dr. Boockvar report no financial relationship relevant to this field of study.
Synopsis: Both conservative and surgical therapies are effective in the treatment of herniated lumbar discs.
Sources: Weinstein JN, et al. Surgical vs nonoperative treatment for lumbar disk herniation: the Spine Patient Outcomes Research Trial (SPORT) observational cohort. JAMA. 2006;296(20):2451-2459 and Weinstein JN, et al. Surgical vs nonoperative treatment for lumbar disk herniation: the Spine Patient Outcomes Research Trial (SPORT): a randomized trial. JAMA. 2006;296(20):2441-2450.
Herniated lumbar discs are a debilitating neurological problem for many patients. While surgical microdiscectomy is a vital treatment of this disorder, 'conservative' management includes patient education, physical therapy, acupuncture, local steroid injections, oral anti-inflammatory and pain medication, and external orthotics. Given variations in treatment paradigms, Weinstein and colleagues undertook the Spine Patient Outcomes Research Trial (SPORT) to compare conservative treatment vs microdiscectomy for patients with herniated lumbar discs and persistent symptoms, despite conservative treatment for at least 6 weeks. In recognition of the problems of randomization for elective surgery and patient preference/crossover, SPORT was designed as a dual-arm study with a randomized branch and a prospective, cohort branch. Of 1991 eligible patients, 501 were randomized, 743 enrolled in the observational cohort, and 747 refused to participate. While intent-to-treat analysis demonstrated improvement in both surgical and conservative treatment groups with a trend toward greater improvement in the surgical group, there was no statistically significant difference between groups. However, given the frequent crossover, the study group also performed an as-treated analysis that demonstrated a strong and significant benefit in the surgical group at all follow-up intervals. Additionally, the observational cohort study demonstrated a greater improvement in the surgical group compared to the conservative group.
Commentary
Weinstein, et al, are to be congratulated for their work in attempting to answer a difficult question in treatment of spine disease. In conducting both intent-to-treat and as-treated analysis, and in publishing results from both prospective-randomized and observational-cohort arms, they make a valiant effort to provide the most appropriate statistical comparison of surgical and nonsurgical treatment. The essential question for any trial of this magnitude is: How will it affect practice patterns for spine surgeons and for physicians who refer patients for non-surgical and surgical therapy? The answer is that it should not; it reaffirms common practice for management of one-level lumbar disc herniations, and its limitations prevent it from providing level-one evidence in comparing surgical to non-surgical management.
Currently, patients referred to our institution for symptomatic nerve-root compression from a lumbar disc herniation are recommended for nonsurgical management prior to consideration for surgery. Operative indications typically include worsening symptoms despite conservative management, progressive neurologic deficit, or severe disability limiting quality of life. While SPORT reaffirms this practice, it also demonstrated that patients presenting with the above conditions are more likely to elect surgery.
Although demonstrating the success of a 'conservative-first' approach, SPORT falls far short of successfully declaring surgical and nonsurgical management to be at equipoise. The limitations of the study are particularly important because of the inappropriate interpretation of the results by the lay media. Despite the as-treated analysis and cohort data showing the superiority of surgical treatment, the intent-to-treat results, that showed no difference, garnered unjustified media exposure. According to ABC News, "Patients suffering from lower back pain could get the same benefits in pain relief and function from nonsurgical treatment as from back surgery…".1 Medical News Today reported that, "[i]f you suffer from a ruptured disk in your lower back you will recover whether you have surgery or not."2 While such reports hold little academic interest to spine surgeons, they are important informational resources to our patients. With such sound bytes, patients may question whether surgery carries any benefit. Furthermore, internists and neurologists may hesitate to refer patients based upon such a superficial reading of the study results. As academic medical practitioners, we have a responsibility to understand study limitations, to dispel inappropriate interpretations, and to inform patients on how SPORT impacts our practice.
The most prominent limitation was the extremely high crossover rate: 40% of patients randomized to surgery did not undergo an operation, while 45% of patients randomized to nonsurgical management underwent an operation. While the as-treated analysis showed a clear benefit of surgery, the fact that no statistically significant difference was found in the intent-to-treat analysis may reflect the fact that both groups (as analyzed under intent-to-treat) had significant numbers of both surgically and non-surgically managed patients. With the intent-to-treat analysis essentially meaningless, the as-treated analysis and cohort study results show a clear benefit of surgical intervention. However, as noted in an editorial accompanying the study, Dr. David Flum recognized the important limitation of placebo effect in elective surgical procedures.3 In any study in which patients elect for a surgical intervention over nonsurgical management, there is an expectation of benefit. That expectation may cloud subjective patient-reported results, rendering the placebo effect an important variable. This is recognized by the lay media. Newsweek reports, "[i]n general, researchers put less stock in observational studies than in randomized ones, in part because of the possibility of a placebo effect."4 However, this problem affects non-blinded randomized studies as well. The question is: Did the placebo effect have equal impact upon patients in the randomized and observational arm? This question is not answered by SPORT in its current publication. It would be interesting to see the separate analysis of patients who remained within their randomized group, excluding patients who crossed over. Thus, it would be academically negligent to affirm the intent-to-treat analysis without strongly conceding the crossover limitation, and without acknowledging the as-treated and observational cohort results. Therefore, we view the Sport Patient Outcomes Research Trial as a successful affirmation of our conservative-first approach to lumbar disc herniation, but also view the results as supporting the benefits of surgical decompression.
References
1. Karalakulasingam C. Beating Lower Back Pain — Without Surgery [online]. Available at: http://abcnews.go.com/Health/story?id=2670812&page=1.
2. Nordqvist C. Patients With Back Pain Recover Without Surgery [online]. Available at: http://www.medicalnewstoday.com/healthnews.php?newsid=57328.
3. Flum DR. Interpreting surgical trials with subjective outcomes: avoiding UnSPORTsmanlike conduct. JAMA. 2006;296(20):2483-2485.
4. Springen K. To Cut Or Not To Cut. In: Newsweek. 2006; 148(23):58.
Both conservative and surgical therapies are effective in the treatment of herniated lumbar discs.You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content