To The Editor:
Upon reading "Treatment of Ruptures of the
Lateral Ankle Ligaments: A Meta-Analysis" (82-A: 761-773,
June 2000), by Pijnenburg et al., I was extremely surprised to find
that their results demonstrated that operative treatment yielded
superior results compared with those of functional treatment. The
authors specifically cited four previous literature reviews, all
of which found the results of functional treatment to be superior
to those of operative treatment, but they faulted these previous
studies for being merely descriptive, with no statistical analysis. They
noted that only seven studies met the rigid inclusion criteria for
their meta-analysis of the results of operative treatment
versus those of functional treatment1-7.
In order to evaluate the findings of this meta-analysis
further, I reviewed the studies that were available to me of those
seven.
I did not review the paper by Van der Ent7,
as it is apparently a thesis that was not published in a peer-reviewed
journal, nor did I review the article by Eggert et al.2, which is written in German. I did review
the other five articles and was surprised by some of the findings
of Pijnenburg et al. in their literature review. The reference by
Prins, apparently a doctoral thesis that was published in a supplement
to a peer-reviewed journal6,
was cited in the comparison of results of operative treatment with
those of functional treatment. However, according to my reading
of the article, Prins described the results of primary surgical
repair of the ligaments followed by two weeks in a short leg nonwalking cast,
one week in a short leg walking cast, and then use of the Unna boot.
The nonoperative group in that study was treated with six weeks
in a short leg walking cast, which is not functional treatment and
therefore should not be included in the comparison. Also, Pijnenburg
et al. stated that there were ninety-four patients in the operative
group, yet, in my review of the article, there were 104 patients.
In the article by Evans et al.3,
the patients in the functional treatment group were immobilized
for three weeks in a short leg walking cast. This would stretch
the definition of functional treatment according to most treating
physicians today. Also, I was surprised by Figure 2-A, which showed the
relative risk for pain in patients in the operative group to be
lower than that in the nonoperative group. In my close reading of
this article, pain is only implied when the authors state that "persistent
symptoms" reduced or prevented a return to sporting activities
for ten patients treated operatively and for only five patients
treated conservatively. Similarly, in Figure 2-B, Pijnenburg et
al. showed that the relative risk for giving-way was greater following functional
treatment than it was after operative treatment. However, Evans
et al. stated that at the two-year follow-up thirteen patients had
symptoms of giving-way following surgery, while only four of the
patients in the functional treatment group had such symptoms.
Pijnenburg et al., in Fig. 2-A, showed that the study by Korkala
et al.5 demonstrated a lower risk
of pain after operative treatment than after functional treatment.
However, Korkala et al. did not evaluate pain. The only measure
of discomfort was tenderness, which was slightly higher in the functional
treatment group than it was in the surgical treatment group. Tenderness, an
objective physical finding, is not synonymous with pain, a subjective complaint
by the patient. The article by Kaikkonen et al.4 did
demonstrate less pain following functional treatment and similar
symptoms of giving-way in both treatment groups.
Broström1 did note
less risk of giving-way in his operatively treated group compared with
that in the functional treatment group, as noted by Pijnenburg et
al. However, in his discussion, he advocated that, in principle,
all injuries to the ankle ligaments should be treated functionally
due to the risks of surgery and that surgical treatment centers could
be inundated by patients with this condition if all were treated
operatively.
While I believe that the authors should be commended for the
great deal of work performed in this meta-analysis, I am
concerned that some of the above discrepancies may alter some of
their conclusions regarding the results of operative treatment versus
those of functional treatment. I was pleased to note that, in the
final paragraph of their discussion, they stated that one should not
necessarily make surgery the treatment of choice. They did mention
the increased risk of complications and the higher costs associated
with operative treatment. Additionally, none of the studies evaluating
functional treatment described any additional methods for treating
residual symptoms when functional treatment fails. Current standard treatment
for all patients with persistent instability or pain is to undergo
a course of rehabilitation. The majority of patients respond to
this treatment regimen and do not require surgery. I do not believe
that one should conclude that operative treatment is the treatment
of choice, since it is more expensive, is associated with a greater
risk of complications, and, as the authors pointed out in their
final sentence, "when conservative treatment fails, secondary operative
reconstruction of the ruptured ligaments can be performed, with
similar good results, even years after the initial injury."
A.C.M. Pijnenburg, C.N. van Dijk, P.M.M. Bossuyt, and
R.K. Marti reply:
We thank Dr. Thordarson for his comments. The study performed
by Prins6 that was published in
1978 was indeed a doctoral thesis. In his study, Prins included
five different treatment groups: Group 1 was treated with surgery
followed by three weeks of casting; Group 2, with cast immobilization for
six weeks; Groups 3 and 4, with elastic bandaging (the former without
a lesion and the latter with a lesion of the anterior talofibular
ligaments only); and Group 5, with cast immobilization for three
weeks (in patients with a lesion of the anterior talofibular ligament
only)6. One of the inclusion criteria
in our study was the presence of a ligament lesion. Since one of
the assumptions we made on the basis of the literature is that there
is no difference in outcome between single and multiple ligament lesions,
we included all patients with either single or multiple ligament lesions.
In Prins’s study, we thus compared the results of operative
treatment followed by three weeks of casting (Group 1) with those
of three weeks of casting only (Group 5). Forty-five patients
in the first group and forty-nine patients in the second group were
evaluated at the six-month follow-up.
In the article by Evans et al.3,
the relative risks for giving-way and pain were derived from Table
III. Since this is the only table that presents data on both giving-way
and pain at the same follow-up time point, we stated in our Materials
and Methods section that a short period of cast immobilization (up
to three weeks) was considered to be a form of functional treatment,
as the immobilization was for such a short period of time.
From the study by Korkala et al.5 we
used the results in Table 4 to evaluate the number of patients with
pain. Since we are aware of the fact that the word "tenderness" is
not synonymous with "pain," we contacted the first author
personally. Dr. Korkala stated that all of these patients had residual pain.
He used the word "tenderness" as a broader word
for pain since there was variety in the severity of pain experienced
by the patients. We included the number of patients with pain in
our analysis.
We agree with the final conclusions of Broström1. However, we tried to find the best treatment
for lateral ligament lesions on evidence-based grounds. After our
evaluation of the literature, it became evident that operative treatment
leads to better results than functional treatment.