To The Editor:
We took great interest in the article "Comparison of
Arthroscopic and Open Anterior Shoulder Stabilization. A Two to
Six-Year Follow-up Study" (82-A: 1108-1114, August 2000),
by Cole et al. We agree with the authors about the importance of
comparing the outcomes of arthroscopic repair and open surgery for shoulder
stabilization. However, we would like to comment on some of the
issues that the authors have attempted to address.
The title of this article implies that the authors have compared the
effectiveness of open procedures with that of arthroscopic methods
in the treatment of recurrent shoulder instability. However, it
is clear that patients with isolated Bankart lesions were treated
arthroscopically while patients with evidence of anteroinferior
capsular laxity were treated with an open procedure alone. The study
groups were obviously different and were not randomized; therefore, direct
comparison of the two methods cannot be made without the risk of
significant systematic bias.
Furthermore, in their introduction the authors state that it
was their aim to determine the effectiveness of their method of patient
selection for open repair or arthroscopic stabilization surgery.
To do so, they tested the null hypothesis that there is no difference
between the two treatment groups with respect to the final clinical
outcome. We are unable to understand how the finding of no statistically
significant difference in the final outcome between two distinctly different
groups of patients with diverse pathology can possibly lead to validation
of a method of treatment selection. Conversely, would they have
rejected their null hypothesis (and thus have concluded that their
method of treatment selection was invalid) if one of the two methods
had resulted in significantly superior results compared with the
other?
Moreover, regarding the authors’ conclusion that the
two surgical methods "yield comparable results," it
appears to us that this trial had little chance of detecting major
differences between the two groups studied due to the small sample
size (thirty-seven arthroscopic repairs and twenty-two open reconstructions). Although
the article does not include the standard deviations for the American
Shoulder and Elbow Surgeons’ standardized assessment score,
if one assumes a standard deviation of 16 points1-3 and
a study power of 80%, detection with use of the t test
of a real difference between the two groups in the score (10 points) would
necessitate inclusion of at least forty-two patients in each group
(Minitab 12.1; Minitab Inc., State College, Pennsylvania). Also,
although the authors did not do a power study and they did not report
a beta value for their study, they nevertheless state that "there
was no significant difference between the two groups";
this clearly contradicts the statistical policy of The Journal4.
We would suggest that, in future studies, prospective, randomized methodology
be used to test the relative effectiveness of open versus arthroscopic
anterior stabilization for shoulder instability in comparable patient
groups.
B.J. Cole, J. L’Insalata, J. Irrgang, and J.J.P.
Warner reply:
We appreciate the interest of Mr. Karnezis and Mr. Sarangi in
our article. They appropriately highlight the fact that the title
implies that we presented a population of patients whose treatments
were comparable simply because the underlying clinical manifestation of
their pathoanatomy was that of a singular entity, recurrent anterior
shoulder instability. We agree that the title, read out of context,
could be misconstrued in this fashion. It was our intention to report
retrospectively the results of patients with similar clinical presentations
but differing pathology, ranging from isolated Bankart lesions to
frank capsular laxity, in an effort to demonstrate the efficacy
of contemporary decision-making in determining the appropriate surgical
procedure.
We were careful to elaborate on this issue of group comparability in
the Abstract, Materials and Methods, and Discussion sections. We
clearly stated that this study reports a series of consecutive patients
diagnosed with recurrent traumatic anterior shoulder instability
for whom the choice of surgical procedure was made according to
the pathology identified at the time of the index procedure. Furthermore,
we believe and we acknowledged that these study groups were comparable
in terms of etiology, age, and chronicity of the instability. We
admitted in the Discussion section, in which we described the limitations
of this study, that "we did not perform a truly randomized
prospective study with an absolutely pure patient population to
compare arthroscopic and open stabilization techniques, [but
rather] we sought to optimize the indications for each
technique in order to improve their respective outcomes."
Mr. Karnezis and Mr. Sarangi raise an interesting issue regarding
our stated null hypothesis that, using our method of treatment selection
(based upon defined perioperative decision-making), there would
be no difference in the final clinical outcomes between the two treatment
groups. In fact, implicitly our conclusion was that our results
supported the premise that the application of the described selection
criteria led to comparable outcomes in terms of the recurrence of
instability. We agree, however, that conceptually there is much
work to be done to statistically validate our method of decision-making
in order to truly define which groups are optimally treated by arthroscopic
repair and which are optimally treated by open procedures. Admittedly,
the variables that we currently believe to be critical for success
may, as of yet, be only poorly defined. Furthermore, generalizations
from the results of a single arthroscopic technique performed by
a single surgeon are of limited value to the general orthopaedic community.
Nevertheless, most of us who manage patients with shoulder instability
are frequently faced with the issue of which patients are optimally treated
by either an arthroscopic or an open stabilization technique. It
was our goal to help the readership understand this process more
clearly through an honest appraisal of a single surgeon’s
experience. It is important to note that the senior author (J.J.P.W.)
initiated his arthroscopic practice in 1991, a time when arthroscopic
techniques were heavily scrutinized.
Just as we would be reluctant to conclude that our selection
process was absolute and all-inclusive had our null hypothesis been rejected,
we most certainly would not have concluded that patients should
have all been treated by one method or the other. In other words,
we believe that refined selection criteria and an understanding
of the pathology associated with this clinical entity are critical
to the success of any surgical procedure selected. Appropriate decision-making
with proper respect for surgical indications will always be the
standard to which we will all be held.
Mr. Karnezis and Mr. Sarangi accurately recognize that no formal
power study was performed for this analysis. We were careful to
indicate in our Results that, "with the numbers available, there
was no significant difference between the groups." Thus, we
admit that, with larger numbers of patients in each group, statistically
significant differences between the two groups may have occurred.
Unfortunately, the lack of a power analysis is an inherent limitation
of any retrospective, nonrandomized study of consecutive patients
over a defined time-period.
All of the authors of this study agree that the arthroscopic
technique applied in this report has certain limitations and that
contemporary techniques that address both the capsule and the labrum
(e.g., suture anchors) may offer results comparable with even the
best results following open stabilization. We also agree, however,
that a truly prospective, randomized methodology comparing arthroscopic
with open stabilization techniques is critical to validate this
conclusion and to obviate the inherent limitations of our study.
Once again, we would like to thank Mr. Karnezis and Mr. Sarangi
for the time and energy spent generating their observations and
commentary; we sincerely hope that we have adequately addressed
their concerns and would look favorably upon any future correspondence.