There has been considerable debate regarding the best treatment of acute
ruptures of the Achilles tendon. Authors supporting operative intervention
have suggested that the normal tension and length of the tendon can be
restored only under direct visualization during an open
repair1-4.
Open repair is also thought to be associated with a lower rate of rerupture
and to allow an earlier return of ankle movement following removal of the
cast. The disadvantages of surgical intervention include morbidity from wound
problems and anesthesia as well as higher costs.
Studies of treatment of Achilles tendon ruptures have generally dealt with
small numbers of patients, and the results of surgical management have been
reported more frequently than have those of nonoperative treatment. To our
knowledge, only three prospective, randomized trials have been published to
date5-7,
and in both of those studies as well as in others that we have
reviewed5,6,8-10,
classic immobilization in a plaster cast was employed.
McComis et al.11
developed the concept of functional bracing as an alternative conservative
treatment for ruptured Achilles tendons. The brace allowed immediate
weight-bearing and active plantar flexion but limited dorsiflexion of the
ankle. McComis et al. found encouraging results after using this regimen for a
small number of patients (fifteen).
Results have been difficult to compare directly as there has been no widely
accepted scoring system with which to assess both subjective and objective
outcomes. Leppilahti et
al.12 addressed
this issue by modifying the criteria of Boyden et
al.13 to create a
point system to record pain, stiffness, muscle weakness, footwear
restrictions, range of motion, subjective outcome, and isokinetic muscle
strength. They used this system to rate the outcomes in a surgically treated
group, and reported that 79% had an excellent or good result.
We developed a combined conservative and orthotic treatment protocol for
acute ruptures of the Achilles tendon. This regimen combines the advantages of
a removable orthosis with those of traditional nonoperative therapy. The early
results, in the first thirty-two patients to be treated with the regimen, were
encouraging14.
This paper will outline our treatment protocol; describe how we assessed
the subjective, objective, and functional results of the regimen in 140
patients, and compare those results with published surgical outcomes.
Acute Achilles tendon ruptures were diagnosed clinically by one orthopaedic
consultant. The patients had been referred to the central orthopaedic service
for the province and included all patients referred to our institution because
of an acutely ruptured Achilles tendon during the period of this study. In all
cases, a short leg cast was applied with the ankle in the equinus position
within twenty-four hours after the diagnosis. At about two weeks, a cast was
made for the fabrication of a rigid polypropylene double-shell
patellar-tendon-bearing orthosis and the equinus cast was changed to a
lightweight equivalent. After a total of four weeks of cast immobilization
(during which time the patients remained non-weight-bearing), the patients
were fitted with the removable orthosis, which was worn for another four
weeks. The patients were asked to remove the orthosis when they went to bed at
night, for bathing, and to allow active exercise of the ankle and subtalar
joints while seated. At this stage, the patients received gait-training with
advice to progress to full weight-bearing using the orthosis.
After removal of the orthosis at the eight-week point, the patients
received additional physiotherapy and were encouraged to return gradually to
normal activities as appropriate.
All subjects who had been treated with this regimen were invited to
participate in this detailed evaluation. Each completed a questionnaire that
requested information on preinjury and postinjury work, preinjury and
postinjury activity levels, time to return to work and other activity, medical
history, drug history, history of Achilles tendon injury and treatment,
treatment complications, and details of physiotherapy. Questions were also
asked about pain, stiffness, subjective calf-muscle weakness, footwear
restrictions, and satisfaction with the result of the treatment.
The patients were invited to attend a physiotherapist-led review clinic
where various physiological measurements were made. The bilateral active and
passive ranges of ankle and foot plantar flexion, dorsiflexion, inversion, and
eversion were recorded while the patient was lying supine with the knee
extended. The calf circumference was measured bilaterally 10 cm distal to the
apex of the tibial tubercle with the patient standing.
The patients then warmed up with a ten-minute ergometric cycle followed by
three sets of thirty-second stretches of the gastrocnemius and soleus muscles
while they were standing. Peak torques (in newton-meters) were recorded for
plantar flexion and dorsiflexion strengths at speeds of 30°, 90°, and
240°/sec, on both the injured and the uninjured side, with use of a
Kinetic Communicator (KinCom) dynamometer (Chattecx, Chattanooga, Tennessee).
The testing position was supine with the knee extended. Before testing, the
patients performed some minimal, submaximal, and maximal repetitions, at a
velocity of 30°/sec, with each leg. They were given a five-minute rest
period before testing, a one-minute rest period between each contraction, and
a five-minute rest period between each leg. Three maximum voluntary
contractions were required at each speed.
The paired t test was used to calculate the significance of the differences
in the continuous variables between the injured and uninjured legs. Approval
was obtained from the Research Ethical Committee of the University of Ulster,
Jordanstown, and the Research and Development Committee of Greenpark
Healthcare Trust, Belfast.
One hundred and forty consecutive patients (101 men and thirty-nine women),
with a mean age of forty-five years (range, twenty-seven to seventy-nine
years), were evaluated. The mean duration of follow-up was 2.9 years (range,
0.4 to 8.2 years). The rupture was on the left in seventy-two patients and on
the right in sixty-eight. The mean time between the injury and treatment was
twenty-two hours (range, one hour to 17.5 days).
Before the injury, 28% of the patients had a desk job, 28% had a job
requiring physical labor, 31% had a professional occupation, and 13% were
unemployed. The mean time lost from work was seven days (range, zero to
fifty-two days), and 98% of the patients who had been working before the
injury returned to their full preinjury level of employment. No patient was
involved in a compensation claim.
Before the injury, 42% of the patients played sports regularly (two or more
times a week), 30% occasionally took part in sports activity (approximately
once a week), 25% were active but did not play a sport, and 3% were
housebound. Of the patients who had engaged in sports activity before the
injury, 4% returned to a better level of activity after treatment, 33%
returned to the same level, 54% returned to less sports activity, and 9% were
unable to return to sports activity. The mean time until the return to sports
activities following removal of the orthosis was eight weeks (range, two weeks
to six months).
The results of the subjective questionnaires and the patients' overall
satisfaction with the treatment are shown in Tables
I and
II.
The active ranges of dorsiflexion and plantar flexion, the calf
circumference, and the plantar flexion peak torque values are shown in
Table III. There were small but
significant differences between the injured and uninjured legs with regard to
all of those variables.
A comparison of the results in our patients with those in the surgically
treated patients in the study by Leppilahti et
al.12 is shown in
Table IV.
The major complications included three complete reruptures, which occurred
three months after the injury in two patients and nine months after the injury
in one. All of these reruptures were forced-dorsiflexion injuries: two of the
patients stumbled on steps, and one stumbled on cobblestones while wearing
high-heeled shoes. All of the patients were treated nonoperatively again, with
an excellent overall result (Leppilahti score).
The minor complications included a partial rerupture (as judged by the
senior author, R.G.H.W.) in five patients, deep venous thrombosis in two, and
a temporary dropfoot in one. Three of the partial reruptures occurred in the
first two months after the injury, in patients who did not comply with the
instructions on wearing the orthosis. The other two partial reruptures
occurred three months after the injury: one of them resulted when the patient
tripped, and the other occurred spontaneously for no apparent reason. These
reruptures were considered to be partial because the patients were found, on
assessment, to have moderate active plantar flexion of the ankle that was
greater than what would be expected from recruitment of the secondary plantar
flexors alone. The patients were treated for an additional four to six weeks
with the orthosis only. Three had an excellent overall result (Leppilahti
score); one, a good result; and one, a fair result. The dropfoot lasted less
than three months and was probably the result of pressure on the common
peroneal nerve resulting from a fall while the patient was wearing the initial
plaster cast. The patient made an excellent recovery.
The prevalence of Achilles tendon rupture is rising. Most authors have
concluded that surgery is the treatment of
choice5,15,
with nonoperative management reserved mainly for elderly or more sedentary
individuals.
Generally, the literature has shown that operative repair produces better
functional results, with a lower rerupture rate, than does nonoperative
management. Operative repair also has a higher rate of
complications16.
While the nonoperative approach may result in a poorer functional result, some
authors have noted that many of the postoperative complications can be
avoided17.
There have been few randomized, controlled trials comparing the two types
of
management5-7.
The most recent prospective study of which we are
aware7 showed a
significantly lower rate of rerupture (p < 0.001) in an operatively treated
group (1.7%) than in a nonoperatively treated group (20.8%). In that trial,
the nonoperative treatment was static, consisting of eight weeks of
immobilization in a cast, but a functional bracing regimen was used for the
surgically treated patients.
Other comparative studies have included heterogeneous groups of patients,
and consequently the results have been difficult to
evaluate8,9,18-20.
Also, the lack of a universally accepted and consistent scoring protocol for
the subjective and objective evaluation of Achilles tendon rupture has made
direct comparison of results between articles extremely difficult.
In the majority of previous studies of nonoperative
management5,6,8-10,
the patients were treated with rigid cast immobilization. Functional bracing
may improve the results of conservative treatment. A few small reports on
conservative functional management have been
published11,14,20-22,
but additional studies are needed to assess potential risks and benefits. The
current investigation is the largest study to date to review conservative
functional management, and it was designed to allow direct comparison of its
results with those in a surgically managed group—i.e., the patients in
the study by Leppilahti et
al.12, who used a
protocol for subjective and objective evaluation of Achilles tendon rupture.
Such a direct comparison was not possible before, as no scoring scale was
available.
The overall score was excellent in 56% of our patients, good in 30%, fair
in 12%, and poor in 2%. These results were better than those reported by
Leppilahti et al.12
(Table IV). Pain was not a
common problem; only one subject, who had generalized osteoarthritis, still
had severe pain at time of the latest follow up. Mild stiffness in the morning
and mild stiffness after recreational activities were common symptoms and were
also reported by Leppilahti et
al.12.
Mild-to-moderate weakness was experienced only with recreational or sports
activities; the one individual who had severe weakness previously had been
very active and reported that she had not returned to sports because she was
afraid of a rerupture.
We found very small differences in ankle range of motion between the
injured and uninjured limbs, with slightly increased dorsiflexion and reduced
plantar flexion on the injured side. Our patients had smaller changes in the
range of motion than did the patients treated operatively in the study by
Leppilahti et
al.12. Similar
changes in the range of motion have been reported by others, following both
surgical and nonsurgical management, and may be a standard consequence of the
injury, regardless of the treatment
method23. McComis
et al.11 stated
that, although increased dorsiflexion has been associated with increased
Achilles tendon length after rupture, its relationship with decreased
force-generating capabilities of the ankle-joint complex has not been
established. They found no decrease in plantar flexion strength of the ankle
following brace treatment of acute ruptures.
In our series, the mean plantar flexion strength (as compared with that on
the contralateral, uninjured side) at both low and high speeds was similar to
that reported by Leppilahti et
al.12 in their
surgically treated patients. However,
Kuwada24 stated:
"Plantarflexory strength can be tested against resistance, but this may
not reveal the entire clinical picture. The patient's own perception of
strength and function is most important." In our study, 89% of the
subjects had no or mild subjective symptoms of calf muscle weakness. Thus, the
concern expressed by advocates of surgical treatment that lengthening of the
tendon can lead to a clinically important decrease in strength has not been
substantiated and should not be considered a reason for surgical
treatment.
Early return to work is a well-known benefit of nonoperative management. In
previous reports, the mean time lost from work has ranged from 10.5 to
thirteen weeks after operative treatment and from 8.5 to nine weeks after
nonoperative
treatment5,6,25.
In the present study, the mean time lost from work was only seven days (range,
zero to fifty-two days). Differences of this magnitude can have an important
economic impact on both the patient and the health-care system.
The subjects in this study returned to sports activity in a mean of only
eight weeks following conclusion of the treatment. Four percent returned to a
higher level of activity, 33% returned to the same level, 54% returned to a
lower level, and 9% were unable to return to sports for various reasons. In
the study by Leppilahti et
al.12, 73% returned
to the same level of sports activity, 4% returned to a lower level, and 23%
retired. Most of their subjects were competitive and recreational athletes,
and no information was given on their rehabilitation or on the time required
for a return to sports. This injury typically occurs at an age when many
individuals are reducing their level of sports activity, so there is not
always a strong motivation to return to sports following recovery. However,
the higher rate at which the surgically treated patients returned to sports
activity requires further study.
The rate of reruptures following conservative management has been the
central point of debate in discussions of alternative forms of treatment.
Rerupture rates as high as
17.7%26 following
nonoperative management have been mentioned as evidence that surgical
management is the better treatment option. Lo et
al.17 found that,
while data regarding complications were missing from certain studies, the
frequency of rerupture was four times higher for nonoperatively managed
patients (p < 0.001) but the total rate of complications for those patients
was only one-seventh of the rate for those treated operatively.
In our study, three patients had a rerupture, so the rate of major
complications was 2%. Minor complications (those that delayed recovery but did
not influence the overall study
results10,27)
occurred at a rate of 6% and included five partial reruptures, two deep venous
thromboses, and one temporary dropfoot. Three of the five partial reruptures
were due to non-compliance with the prescribed treatment regimen. While
noncompliance would not have occurred with full cast immobilization, the
number of complications resulting from non-compliance was low. Our
complication rates were much lower than those reported by Leppilahti et
al.12 (7% and 20%,
respectively) in their study of surgically managed patients.
In conclusion, it is our strong view that our nonoperative treatment
protocol for rupture of the Achilles tendon should be the treatment of choice,
but only when supervised by senior, experienced staff. Some problems have come
to light when patients were managed in other hospitals by staff with
insufficient experience with this form of treatment, reinforcing the
importance of experienced input in the care of this injury.