Copyright © 2008 by The Journal of Bone and Joint Surgery, Inc.
Commentary & Perspective
Commentary & Perspective by
Christopher M. Bono, MD*,
Orthopaedic Spine Service, Brigham and Women's Hospital, Boston, Massachusetts
Posted September 2008
Lumbar arthrodesis for degenerative conditions of the spine
has come under exacting scrutiny over the past few years. Reviews of previously
published studies have suggested only modest improvements in results despite
increased use of technologies such as pedicle screws and interbody devices1,2,
and these results have raised doubts about the benefits of fusion for patients
affected by the wide spectrum of disorders categorized as "degenerative disc
disease."
It has been difficult, at best, to evaluate the results of
lumbar arthrodesis because of the lack of evidence-based studies. Despite the
use of a variety of accepted, validated, and disease-specific instruments, the
clinical relevance of statistically significant improvements in outcome and pain
scores has been increasingly contested. Consider the following illustrative
example. Study X found that Oswestry Disability Index scores were improved, on
average, from fifty to forty. With a large enough sample size, this ten-point
improvement can be statistically significant with a p value of <0.05. However,
it is unknown if a ten-point improvement in the Owestry Index represents a
clinically significant improvement for the patient.
The recognition of this disconnect between statistical and
clinically meaningful differences has had a justifiably pervasive impact in the
spinal literature. Thus, the concept of a minimum clinically important difference has been applied3. Notwithstanding
its importance, a minimum clinically important difference refers to the
proverbial "bare minimum" change that a patient would perceive as an improvement
in pain, function, or other assessed parameter. This method of assessment does
not take into consideration what a patient would consider to be an improvement sufficient
to elicit satisfaction with the outcome of the operation.
These outcome deficiencies are precisely what Glassman et al.
have attempted to answer in their study. Moreover, the investigators confined
themselves to a group of patients who had undergone treatment for degenerative
lumbar disorders, the diagnostic group most susceptible to interpretative
challenges. Their efforts should be applauded.
Glassman et al. have translated a patient's report of
"mostly satisfied" or "much better" into a numerical change in the Oswestry
Disability Index, Short-Form-36 (SF-36) questionnaire, and visual analog scale
pain scores. By doing so, one would hope that the reverse process might work
equally as well—that is, that the substantial clinical benefit parameters could
be applied post hoc to previously published and future outcome studies.
The current proposal of seemingly exact numerical
representations of substantial clinical benefit should be cautiously
interpreted, however. While Glassman et al. have suggested, on the basis of
their data, that a net change of 2.5 in leg or back pain (according to Table IV
in their paper) can distinguish between feeling "much better" or "about the
same," the threshold between these two patient-reported conditions is likely
much more complex. It is my impression that a reduction of pain by only 2.5 is
ostensibly a small number. In counseling a patient prior to surgery, it would
seem an overestimation to think that most patients would consider such a change
to be clinically substantial. Beyond the magnitude of change, one might also
question the equivalence of a change from disparate starting levels of pain. For
example, could (or should) one conclude that a 2.5-point change from 10 to 7.5
is equivalent to a change from 6 to 3.5?
Another observation is that the range of pain, Oswestry Disability Index, and SF-36 scores in the compared groups had considerable overlap. In their Table III, Glassman et al. documented that the average SF-36 scores and standard deviation for the "much better" and "about the same" groups were 13.8 ± 9.3 and 2.4 ± 6.1, respectively. Thus the calculated ranges, according to the standard deviations, would be 4.5 to 23.1 for the "much better" group and −3.7 to 8.5 for the "about the same" group. The upper and lower limits of these groups have substantial overlap. As the authors state in the Materials and Methods section, it is perhaps the value for the area under the curve (in this case 0.846) that is more meaningful, in that it implies that there was, on the average, an 84.6% chance that patients with an SF-36 change of at least 6.2 were truly "much better" than the "about the same" group.
As a final observation, the current study suggests that each
of the three outcome thresholds (SF-36, Oswestry Disability Index, and pain
score) is an independent criterion for substantial clinical benefit. However,
the three are likely to be interdependent. Future work might be directed at
formulating a combination of criteria that would more accurately represent the
full spectrum of what a patient interprets as a real or substantial clinical
improvement as a result of surgery.
In summary, Glassman et al. have produced a worthwhile contribution to the literature regarding an issue that is becoming increasingly crucial in the assessment of use of lumbar fusion for degenerative conditions. As they are the first group to suggest a criterion threshold for substantial clinical benefit, I would hope that their group continues their work, perhaps defining influential covariables, such as individual degenerative subgroups.
*The author did not receive any outside funding or grants in support of his research for or preparation of this work. Neither he nor a member of his immediate family received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which the author, or a member of his immediate family, is affiliated or associated.
References
1. Bono CM, Lee CK. Critical analysis of trends in fusion for degenerative disc disease over the past 20 years: influence of technique on fusion rate and clinical outcome. Spine. 2004;29:455-63; discussion Z5.
2. Gibson JN, Waddell G. Surgery for degenerative lumbar spondylosis: updated Cochrane Review. Spine. 2005;30:2312-20.
3. Copay AG, Glassman SD, Subach BR, Berven S, Schuler TC, Carreon LY. The minimum clinically important difference in lumbar spine surgery patients: a choice of methods using the Oswestry Disability Index, Medical Outcomes Study questionnaire Short Form 36, and Pain Scales. Spine J. 2008 Jan 15. [Epub ahead of print].
|
