Protocol

Protocol for: Myhrvold SB, Brouwer EF, Andresen TKM, et al. Nonoperative or surgical treatment of acute
Achilles’ tendon rupture. N Engl J Med 2022;386:1409-20. DOI: 10.1056/NEJMoa2108447

This trial protocol has been provided by the authors to give readers additional information about the work.
This supplement contains the original protocol, final protocol and summary of changes.
The original statistical analysis plan (Version 1.0) was not changed.
Original protocol of January 29th 2013:

Non-operative treatment of acute Achilles tendon ruptures versus

open and mini-invasive surgery: A prospective randomized trial.
Principal Investigator Ståle Myhrvold MD (1)

Supervisor
Sigurd Erik Hoelsbrekken MD, PhD (1)

Co-supervisors
Kjetil Hvaal MD, PhD (2)
Lars Engebretsen MD, PhD (2)
Øivind Paulsrud MD,(2)

(1) Akershus University Hospital

(2) Oslo University Hospital, Ullevål

1. Background

The gastrocnemius-soleus musculotendinous complex, triceps surae, is located within the

posterior compartment of the calf. The two heads of the gastrocnemius merge together with
soleus to form the Achilles tendon, which inserts into the posterior surface of the calcaneus.
Plantarflexion of the ankle and consequently normal gait therefore relies upon the integrity of
the Achilles tendon, and injuries may lead to severe impairments. The tendon receives a load
stress up to several times the body weight during walking, and the load stress is amplified by
running or jumping1. This leaves the Achilles tendon vulnerable to ruptures, especially when
exposed to sudden and severe loads, which often occurs during sprinting and racket sports.
Men are more prone to injuries compared to women, and changes in sport and leisure
activities is commonly perceived as one of the most important factors causing the increased
incidence observed during the past decades2.

Ruptures of the Achilles tendon can be treated by surgical repair or by immobilization using a
cast or a brace allowing the tendon to heal without surgery. Even though such ruptures
represent a common injury, there is no consensus regarding the best treatment protocol3. In
general, the Norwegian orthopedic community has recommended surgical repair based on
studies published the past thirty years showing that surgical repair conveys a significantly
lower risk of rerupture compared to non-operative treatment. Moreover, the comparatively
low number of reruptures has been used to justify the risk of infections and nerve injuries
inferred by surgical treatment4. To reduce the number of complications associated with
operative treatment, mini-invasive surgical techniques have been developed thereby
significantly reducing the number of wound problems while retaining good functional result5.
This has led to an increasing popularity of the mini-invasive approach and a growing number
of patients are treated by mini-invasive surgery. The use of different suture materials and
techniques has also been investigated, but none have proven to be consistently better than
other alternatives and there are no clear recommendations6. There is also uncertainty as to
what time point reruptures tend to occur, and if reruptures happens when the healing tendon
carries most of the tensile strength, regardless of suture techniques and material used.
Mechanical loading stimulates bone healing and bone formation. This correlation was
deduced by the German surgeon Julius Wolff during the late eighteen century and has later
been known as Wolff’s law7. Similar to the relationship dictated by Wolff’s law, mechanical
loading of tendons stimulates increased tensile strength over time8,9. Awareness centered on

Figure 1
this biomechanical trait has led to the development of new rehabilitation protocols10,
especially concerning non-operative treatment. In analogue to Woffs law, by allowing early
controlled loading of the ruptured Achilles tendon, the presumed healing process is
accelerated thereby increasing tensile strength. If true, this approach would reduce the risk of
reruptures and several recent papers have shown that early mobilization actually do improve
the prognosis10-12. Furthermore, if non-operative treatment and surgery yield similar treatment
results, it would be of considerable importance since surgery is inherently associated with the
risk of infections and paresthesia. However, the studies carried out this far have included a
relatively small number of patients and no firm conclusions have been made13. It is therefore
of great interest to conclusively determine treatment results by conducting a prospective
randomized study comprising a sufficient number of patients, potentially altering current
treatment recommendations.

2. Study design

2.1 Inclusion and randomization

To compare the end-results of three
different methods of treatment, it is
necessary to establish identical
rehabilitation protocols. Traditionally,
early mobilization has been reserved
for patients treated surgically and this
may have unintentionally skewed
treatment results. We have therefore
designed a prospective randomized
trial performed as collaboration
between Akershus University Hospital
(Ahus), Oslo University Hospital (Oslo
Legevakt), Østfold Hospital
(Fredrikstad) and Drammen Hospital
(Figure 1). The four institutions were
chosen because of their geographical
proximity and because they jointly
have a substantial catchment area
(Figure 2). Treatment is divided into
three arms, and patients between 18
and 60 years of age sustaining first
time achilles tendon ruptures will be
invited to participate.

2.2 Surgical treatment

Figure 2
Patients are randomized to either
surgery or non-operative treatment. Surgery will be performed by open technique or by using
a mini-invasive approach. Open surgery remains widespread and arguably the best
documented treatment, however mini-invasive surgery is also included because of its potential
to lessen the risk of wound complications. Michael Amlang and Hans Zwipp at the University
of Dresden represent some of the leading foot- and ankle surgeons in Europe. They and their
co-workers have developed a protocol for mini-invasive surgery using specially adapted
proprietary instruments14. Michael Amlang visited Ahus in 2012 teaching the general
principles of the Dresden mini-invasive treatment as well as performing two operations live-
streamed by internet to the participating hospitals. To ensure optimal and reproducible results,
surgeons participating in the study must master both techniques. This requires a break-in
period of training and standardization of open and mini-invasive surgery prior to initiation of
the project.

2.3 Surgical procedures

All surgery will be performed in general anesthesia without a tourniquet, and all surgical
procedures utilize identical suture materials and thread sizes. By the open procedure a ten cm
dorsomedial longitudinal incision is made and the fascia and peritenon divided in the midline.
The tendon is then repaired by a double layer Krakow whip suture using non-absorbable
Ticron (polydioxanone) 2 thread size (Figure 3a). With the patient laying in prone position the
Ticron sutures are tightened so that the ankle joint demonstrates an equinus position
comparable to the noninjured side. Finally, the peritenon, fascia and subcutaneous tissues are
sutured using absorbable Polysorb (copolymer composed of glycolide and lactide) 2-0
sutures, and the skin is closed using non-absorbable Ethilon (nylon) 3-0 sutures. By the mini-
invasive exposure, a short three cm dorsomedial incision is made at least two cm proximal to
the rupture site. The fascia is then opened, whereas the peritenon is left intact. The especially
adapted Dresden instruments are introduced into the layer between the fascia and peritenon,
and the instruments apertures are positioned one cm proximal to the calcaneal insertion of the
tendon. The needle is passed through the apartures, and the instruments are then carefully
removed pulling out the suture ends (Figure 3b and c). With the patient laying in prone
position the Ticron 2 sutures are tightened so that the ankle joint demonstrates an equinus
position comparable to the noninjured side. The fascia and subcutaneous tissue are closed
with Polysorb sutures and the skin closed with Ethilon.

Figure 3
2.4 Postoperative treatment regiment
All patients included in the study will be immobilized in equinus position for the first two
weeks using a below-the-knee cast. Following the first two weeks the cast is replaced with a
brace maintaining ∼12° plantar flexion (three heel lifts). We have chosen to use standardized
heel lifts because it eases weightbearing compared to a hinged brace. After two weeks (four
weeks from injury) the ankle position is adjusted to ∼8° plantar flexion (two heel lifts) and
following additional two weeks the ankle position is adjusted to ∼4° plantar flexion (one heel
lift. The last week of the orthosis treatment (the 8th week after treatment initiation) the heel
lifts are completely removed. Hence, the brace is retained for a total of six weeks, and patients
are allowed full weight bearing as tolerated during the entire time period. The brace is worn
day and night for the first two weeks, but is removed during nighttime for the last four weeks.
Importantly, the immobilization regimen is identical for the three treatment groups. To ensure
identical rehabilitation following immobilization, physical therapy has to adhere to a
standardized protocol supervised by an experienced physiotherapist at the Norwegian Sport
Medicine Clinic (NIMI) as outlined in table 1.

Figure 4 Lateral radiographic evaluations of the ankle brace depicting the angle
between the longitudinal axis of the first metatarsal and the base of the brace using
none (A), one (B) or three (C) standardized heel lifts.

2.6 Calculating the size of the treatment groups

Although the minimum clinically important difference in ATRS score has yet to be
determined, Carmont and coworkers have shown that 7 points represent the smallest
detectable change20. Moreover, a study by Metz et al. found that patients with reruptures of
the achilles tendon presented on average 18 points lower scores whereas patients with injuries
of the sural nerve or superficial wound infections presented scores 10 and 9 points lower than
average, respectively21. In order to detect a difference of 7 points in ATRS score with 80%
power there must be 160 patients in each group. Power calculations are based on one-way
ANOVA analyses assuming a common standard deviation of 20. This concomitantly allows
us to detect differences as low as 8 % with respect to complications. This is of pivotal
importance since assessing the risk of reruptures is essential when establishing treatment
recommendations. We therefore plan to include a total of 530 patients allowing for expected
loss during follow-up. Ahus, The Emergency Department of Oslo University Hospital,
Drammen and Fredrikstad Hospitals have a catchment area of nearly 1.3 million people and
treats in excess of 300 achilles tendon ruptures annually (Figure 2). This allows for
completion of the inclusion period within three years. In our view, this is the first study to
include a sufficient number of patients to detect lesser, but yet important differences, and the
results will be published in a suitable international peer-reviewed journal.

Table 1

3. Secondary studies

The prospective randomized trial embodies many different disciplines and will spur off
follow-up studies. As part of the main project we will conduct a treatment-cost analysis in
relation to individual results. We will also perform an investigation based on ultrasound
grading of Achilles tendon ruptures according to a classification system established by
Michael Amlang and colleagues. Previous studies have provided evidence suggesting that
treatment results may depend on the extension of the injury, and that treatment
recommendations should be based on type of rupture sustained22,23. This is of particular
interest as it may facilitate individualized treatment. Together with the prospective
randomized trial and the treatment-cost analysis, the ultrasound investigation will constitute
the core of the PhD project. Additionally, The Norwegian School of Sport Sciences has
established a method to assess tendon elasticity by ultrasound examinations, which would
enable us to evaluate the healing process qualitatively24,25. A biomechanical analysis
including three-dimensional gait analysis and pedographic measurements of patients treated
for Achilles tendon ruptures matched with a healthy control group will be carried out in
collaboration with the Norwegian School of Sport Sciences (Pedography registers the
distribution of the plantar pressure load during walking). These examinations may provide a
biomechanical understanding of the pathomechanical changes induced following treatment.
Reduced function in patients treated for Achilles tendon ruptures, regardless of choice of
treatment, is well documented19, but the orthopedic community has largely ignored the casual
relationships.

Figure 5 Sonographic classifications of achilles tendon ruptures. Type 1 ruptures

have complete adaptation of the tendon ends whereas type 2 only displays partial
adaptation with (2a) or without (2b) hematoma. In type 3 ruptures there is a diastasis
between the tendon ends with (3a) or without (3b) an organized hematoma. Type 4 is
a proximal rupture at the muscle/tendon intersection while type 5 represents distal
ruptures near the insertion.

4. Structure of the research group

The study will be carried out as part of a PhD project under supervision of Sigurd Erik
Hoelsbrekken employed at the orthopedic department, Akershus University Hospital. Lars
Engebretsen, Kjetil Hvaal and Øyvind Paulsrud at Oslo University Hospital, Ullevål, will
serve as co-supervisors. Knut Melhuus and Agathe Rønning will be responsible for the
implementation and follow-up at the Emergency Department of Oslo University Hospital.
Asbjørn Sorteberg and Lars Fredrik Høifødt will organize the implementation at Fredrikstad
hospital whereas Finnur Snorrasson and Faisal Butt are responsible for carrying out the study
at Drammen hospital. Karin Rydevik at the Norwegian Sport Medicine Clinic (NIMI)
supervises the physical therapy program as well as overseeing functional testing and
examinations at follow-up. Two master students in physical therapy under the supervision of
Jens Bojsen-Møller will execute the physical examinations at the Norwegian School of Sport
Sciences. The Dresden group under the leadership of Hans Zwipp has generously allowed us
to share some of their vast knowledge, and has also invited us to visit the university clinic at
Dresden. We have also been in contact with Professor Jon Karlsson at the University Hospital
in Gothenburg and Katarina Nilsson Helander visited Ahus earlier this year to deliver a
lecture on Achilles tendon ruptures. Hopefully the study will facilitate cooperation across
different institutions, and an axis of collaboration has already been established along the Oslo
Fiord. Furthermore, we are in the process of founding an international research network
(Figure 5).
5. Study progress and planning

During the final work on the study protocol, we organized an achilles day at Ahus in
conjunction with a visit by Dr. Amlang from Dresden in May 2012. He held a lecture and
performed two surgical procedures that were streamed by Internet to the participating
hospitals. We were also fortunate enough to welcome Dr. Nilsson Helander to Ahus 2012-06-
14 for another presentation on ruptures of the achilles tendon. By sharing their own
experiences in conducting similar projects, our international collaborators have proven
invaluable in the process of developing the study protocol. The start date of the prospective
randomized trial is set to 2013-02-15, and the study was approved by the Regional Ethics
Committee for Medical Research 2012-05-11 (REK Helseregion Øst, approval number
2012/530 D). For the time being, we are in the process of finishing a trial period aimed at
assessing the planned implementation and to ease cooperation across the different institutions.
Furthermore, the trial period has allowed us to identify and rectify potential problems and the
start date has been pushed forward to 2013-02-15. The inclusion period is planned to last until
31.12.15, and collection, analysis and preparation of data will be completed by the end of
2016. Publication of results and writing of the PhD thesis is set to 2017.

Figure 5
Literature

1. Giddings, V. L.; Beaupre, G. S.; Whalen, R. T.; and Carter, D. R.: Calcaneal loading during walking and
running. Med Sci Sports Exerc, 32(3): 627-34, 2000.
2. Jarvinen, T. A.; Kannus, P.; Maffulli, N.; and Khan, K. M.: Achilles tendon disorders: etiology and
epidemiology. Foot and ankle clinics, 10(2): 255-66, 2005.
3. Movin, T.; Ryberg, A.; McBride, D. J.; and Maffulli, N.: Acute rupture of the Achilles tendon. Foot and ankle
clinics, 10(2): 331-56, 2005.
4. Khan, R. J.; Fick, D.; Brammar, T. J.; Crawford, J.; and Parker, M. J.: Interventions for treating acute
Achilles tendon ruptures. Cochrane Database Syst Rev, (3): CD003674, 2004.
5. McMahon, S. E.; Smith, T. O.; and Hing, C. B.: A meta-analysis of randomised controlled trials comparing
conventional to minimally invasive approaches for repair of an Achilles tendon rupture. Foot and ankle surgery :
official journal of the European Society of Foot and Ankle Surgeons, 17(4): 211-7, 2011.
6. Maquirriain, J.: Achilles tendon rupture: avoiding tendon lengthening during surgical repair and rehabilitation.
Yale J Biol Med, 84(3): 289-300, 2011.
7. Wolff, J.: Das Gesetz der Transformation der Knochen. Edited, Hirschwald, 1892.
8. Palmes, D.; Spiegel, H. U.; Schneider, T. O.; Langer, M.; Stratmann, U.; Budny, T.; and Probst, A.: Achilles
tendon healing: long-term biomechanical effects of postoperative mobilization and immobilization in a new mouse
model. J Orthop Res, 20(5): 939-46, 2002.
9. Pneumaticos, S. G.; Phd, P. C. N.; McGarvey, W. C.; Mody, D. R.; and Trevino, S. G.: The effects of early
mobilization in the healing of achilles tendon repair. Foot Ankle Int, 21(7): 551-7, 2000.
10. Twaddle, B. C., and Poon, P.: Early motion for Achilles tendon ruptures: is surgery important? A randomized,
prospective study. Am J Sports Med, 35(12): 2033-8, 2007.
11. Gwynne-Jones, D. P.; Sims, M.; and Handcock, D.: Epidemiology and outcomes of acute Achilles tendon
rupture with operative or nonoperative treatment using an identical functional bracing protocol. Foot Ankle Int,
32(4): 337-43, 2011.
12. Nilsson-Helander, K.; Silbernagel, K. G.; Thomee, R.; Faxen, E.; Olsson, N.; Eriksson, B. I.; and Karlsson,
J.: Acute achilles tendon rupture: a randomized, controlled study comparing surgical and nonsurgical treatments
using validated outcome measures. Am J Sports Med, 38(11): 2186-93, 2010.
13. Tengberg, P. T.; Barfod, K.; Krasheninnikoff, M.; Ebskov, L.; and Troelsen, A.: Promising conservative
treatment using dynamic mobilisation after Achilles tendon rupture. Ugeskr Laeger, 173(44): 2778-84, 2011.
14. Amlang, M. H.; Christiani, P.; Heinz, P.; and Zwipp, H.: Percutaneous technique for Achilles tendon repair
with the Dresden Instruments. Unfallchirurg, 108(7): 529-36, 2005.
15. Nilsson-Helander, K.; Thomee, R.; Silbernagel, K. G.; Thomee, P.; Faxen, E.; Eriksson, B. I.; and Karlsson,
J.: The Achilles tendon Total Rupture Score (ATRS): development and validation. Am J Sports Med, 35(3): 421-6,
2007.
16. Garratt, A. M.; Ruta, D. A.; Abdalla, M. I.; Buckingham, J. K.; and Russell, I. T.: The SF36 health survey
questionnaire: an outcome measure suitable for routine use within the NHS? BMJ, 306(6890): 1440-4, 1993.
17. Melzack, R.: The short-form McGill Pain Questionnaire. Pain, 30(2): 191-7, 1987.
18. Silbernagel, K. G.; Gustavsson, A.; Thomee, R.; and Karlsson, J.: Evaluation of lower leg function in patients
with Achilles tendinopathy. Knee surgery, sports traumatology, arthroscopy : official journal of the ESSKA,
14(11): 1207-17, 2006.
19. Olsson, N.; Nilsson-Helander, K.; Karlsson, J.; Eriksson, B. I.; Thomee, R.; Faxen, E.; and Silbernagel, K.
G.: Major functional deficits persist 2 years after acute Achilles tendon rupture. Knee surgery, sports traumatology,
arthroscopy : official journal of the ESSKA, 19(8): 1385-93, 2011.
20. Carmont, M. R.; Silbernagel, K. G.; Nilsson-Helander, K.; Mei-Dan, O.; Karlsson, J.; and Maffulli, N.:
Cross cultural adaptation of the Achilles tendon Total Rupture Score with reliability, validity and responsiveness
evaluation. Knee Surg Sports Traumatol Arthrosc, 2012.
21. Metz, R.; van der Heijden, G. J.; Verleisdonk, E. J.; Kolfschoten, N.; Verhofstad, M. H.; and van der
Werken, C.: Effect of complications after minimally invasive surgical repair of acute achilles tendon ruptures:
report on 211 cases. Am J Sports Med, 39(4): 820-4, 2011.
22. Kotnis, R.; David, S.; Handley, R.; Willett, K.; and Ostlere, S.: Dynamic ultrasound as a selection tool for
reducing achilles tendon reruptures. Am J Sports Med, 34(9): 1395-400, 2006.
23. Thermann, H.; Zwipp, H.; Milbradt, H.; and Reimer, P.: [Ultrasound sonography in the diagnosis and follow-
up of Achilles tendon rupture]. Unfallchirurg, 92(6): 266-73, 1989.
24. Bojsen-Moller, J.; Hansen, P.; Aagaard, P.; Svantesson, U.; Kjaer, M.; and Magnusson, S. P.: Differential
displacement of the human soleus and medial gastrocnemius aponeuroses during isometric plantar flexor
contractions in vivo. J Appl Physiol, 97(5): 1908-14, 2004.
25. Fukashiro, S.; Itoh, M.; Ichinose, Y.; Kawakami, Y.; and Fukunaga, T.: Ultrasonography gives directly but
noninvasively elastic characteristic of human tendon in vivo. Eur J Appl Physiol Occup Physiol, 71(6): 555-7,
1995.
The final protocol of February 5th 2014:

Non-operative treatment of acute Achilles tendon ruptures versus

open and mini-invasive surgery: A prospective randomized trial.

Principal Investigator Ståle Myhrvold MD (1)

Supervisor
Sigurd Erik Hoelsbrekken MD, PhD (1)

Co-supervisors
Kjetil Hvaal MD, PhD (2)
Lars Engebretsen MD, PhD (2)
Øivind Paulsrud MD,(2)

(1) Akershus University Hospital

(2) Oslo University Hospital, Ullevål

Protocol written 29.01.13

Last update 05.02.14

1. Background

The gastrocnemius-soleus musculotendinous complex, triceps surae, is located within the

posterior compartment of the calf. The two heads of the gastrocnemius merge together
with soleus to form the Achilles tendon, which inserts into the posterior surface of the
calcaneus. Plantarflexion of the ankle and consequently normal gait therefore relies upon
the integrity of the Achilles tendon, and injuries may lead to severe impairments. The
tendon receives a load stress up to several times the body weight during walking, and the
load stress is amplified by running or jumping1. This leaves the Achilles tendon vulnerable to
ruptures, especially when exposed to sudden and severe loads, which often occurs during
sprinting and racket sports. Men are more prone to injuries compared to women, and
changes in sport and leisure activities is commonly perceived as one of the most important
factors causing the increased incidence observed during the past decades2.

Ruptures of the Achilles tendon can be treated by surgical repair or by immobilization using
a cast or a brace allowing the tendon to heal without surgery. Even though such ruptures
represent a common injury, there is no consensus regarding the best treatment protocol3.
In general, the Norwegian orthopedic community has recommended surgical repair based
on studies published the past thirty years showing that surgical repair conveys a
significantly lower risk of rerupture compared to non-operative treatment. Moreover, the
comparatively low number of reruptures has been used to justify the risk of infections and
nerve injuries inferred by surgical treatment4. To reduce the number of complications
associated with operative treatment, mini-invasive surgical techniques have been
developed thereby significantly reducing the number of wound problems while retaining
good functional result5. This has led to an increasing popularity of the mini-invasive
approach and a growing number of patients are treated by mini-invasive surgery. The use of
different suture materials and techniques has also been investigated, but none have proven
to be consistently better than other alternatives and there are no clear recommendations6.
There is also uncertainty as to what time point reruptures tend to occur, and if reruptures
happens when the healing tendon carries most of the tensile strength, regardless of suture
techniques and material used. Mechanical loading stimulates bone healing and bone
formation. This correlation was deduced by the German surgeon Julius Wolff during the
late eighteen century and has later been known as Wolff’s law7. Similar to the relationship
dictated by Wolff’s law, mechanical loading of tendons stimulates increased tensile strength
over time8,9. Awareness centered on
Figure 1
this biomechanical trait has led to the development of new rehabilitation protocols10,
especially concerning non-operative treatment. In analogue to Woffs law, by allowing early
controlled loading of the ruptured Achilles tendon, the presumed healing process is
accelerated thereby increasing tensile strength. If true, this approach would reduce the risk
of reruptures and several recent papers have shown that early mobilization actually do
improve the prognosis10-12. Furthermore, if non-operative treatment and surgery yield
similar treatment results, it would be of considerable importance since surgery is inherently
associated with the risk of infections and paresthesia. However, the studies carried out this
far have included a relatively small number of patients and no firm conclusions have been
made13. It is therefore of great interest to conclusively determine treatment results by
conducting a prospective randomized study comprising a sufficient number of patients,
potentially altering current treatment recommendations.

2. Study design

2.1 Inclusion and randomization

To compare the end-results of three different methods of treatment, it is necessary to
establish identical rehabilitation protocols. Traditionally, early mobilization has been
reserved for patients treated surgically and this may have unintentionally skewed treatment
results. We have therefore designed a prospective randomized trial performed as
collaboration between Akershus University Hospital (Ahus), Oslo University Hospital (Oslo
Legevakt), Østfold Hospital (Fredrikstad) and Drammen Hospital (Figure 1). The four
institutions were chosen because of their geographical proximity and because they jointly
have a substantial catchment area (Figure 2). Treatment is divided into three arms, and
patients between 18 and 60 years of age sustaining first time achilles tendon ruptures will
be invited to participate.
2.2 Surgical treatment
Patients are randomized to either
surgery or non-operative treatment.
Surgery will be performed by open
technique or by using a mini-invasive
approach. Open surgery remains
widespread and arguably the best
documented treatment, however
mini-invasive surgery is also included
because of its potential to lessen the
risk of wound complications. Michael
Amlang and Hans Zwipp at the
University of Dresden represent
some of the leading foot- and ankle
Figure 2
surgeons in Europe. They and their
co-workers have developed a
protocol for mini-invasive surgery
using specially adapted proprietary
instruments14. Michael Amlang
visited Ahus in 2012 teaching the
general principles of the Dresden
mini-invasive treatment as well as
performing two operations
Figure 2
livestreamed by internet to the
participating hospitals. To ensure optimal and reproducible results, surgeons participating in
the study must master both techniques. This requires a break-in period of training and
standardization of open and mini-invasive surgery prior to initiation of the project.

2.3 Surgical procedures

All surgery will be performed in local anesthesia without a tourniquet, and all surgical
procedures utilize identical suture materials and thread sizes. By the open procedure a 10
cm dorsomedial longitudinal incision is made and the fascia and peritenon divided in the
midline. The tendon is then repaired by a three knot, double layer Krakow whip suture using
non-absorbable Ticron (polydioxanone) 2 thread size (Figure 3a). With the patient laying in
prone position the Ticron sutures are tightened so that the ankle joint demonstrates an
equinus position comparable or slightly overcorrected compared to the noninjured side.
Finally, the peritenon, fascia and subcutaneous tissues are sutured using absorbable
Polysorb (copolymer composed of glycolide and lactide) 2-0 sutures, and the skin is closed
using non-absorbable Ethilon (nylon) 3-0 sutures. By the miniinvasive exposure, a short
three cm dorsomedial incision is made at least two cm proximal to the rupture site. The
fascia is then opened, whereas the peritenon is left intact. The especially adapted Dresden
instruments are introduced into the layer between the fascia and peritenon, and the
instruments apertures are positioned 1-2 cm proximal to the calcaneal insertion of the
tendon. A straight needle with Ethibond 0 thread size is passed through the apartures so
that the thread ends will be pulled out together with the instruments. The procedure is
repeated three times, and the instruments are then carefully removed pulling out the
suture ends (Figure 3b and c). With the patient laying in prone position the Ethibon-0
sutures are tightened so that the ankle joint demonstrates an equinus position comparable
or slightly overcorrected compared to the noninjured side. The fascia and subcutaneous
tissue are closed with Polysorb 2-0 sutures and the skin closed with Ethilon 3-0.

Figure 3

2.4 Postoperative treatment regiment

All patients included in the study will be immobilized in equinus position for the first two
weeks using a below-the-knee cast. Following the first two weeks the cast is replaced with a
brace maintaining ~12° plantar flexion (three heel lifts). We have chosen to use
standardized heel lifts because it eases weightbearing compared to a hinged brace. After
two weeks (four weeks from injury) the ankle position is adjusted to ~8° plantar flexion
(two heel lifts) and following additional two weeks the ankle position is adjusted to ~4°
plantar flexion (one heel lift). The last week of the orthosis treatment the heel lift is
completely removed. Hence, the brace is retained for a total of six weeks, and patients are
allowed full weight bearing as tolerated during the entire time period. The brace is worn
day and night for the first two weeks, but is removed during nighttime for the last four
weeks. Importantly, the immobilization regimen is identical for all three-treatment groups.
To ensure identical rehabilitation following immobilization, physical therapy has to adhere
to a standardized protocol supervised by an experienced physiotherapist at the Norwegian
Sport Medicine Clinic (NIMI) as outlined in table 1.
Figure 4 Lateral radiographic evaluations of the ankle brace depicting the angle between the longitudinal axis of
the first metatarsal and the base of the brace using none (A), one (B) or three (C) standardized heel lifts.

2.6 Calculating the size of the treatment groups

Although the minimum clinically important difference in ATRS score has yet to be
determined, Carmont and coworkers have shown that 7 points represent the smallest
detectable change20. Moreover, a study by Metz et al. found that patients with reruptures
of the achilles tendon presented on average 18 points lower scores whereas patients with
injuries of the sural nerve or superficial wound infections presented scores 10 and 9 points
lower than average, respectively21. In order to detect a difference of 7 points in ATRS score
with 80% power there must be 160 patients in each group. Power calculations are based on
one-way ANOVA analyses assuming a common standard deviation of 20. This concomitantly
allows us to detect differences as low as 8 % with respect to complications. This is of pivotal
importance since assessing the risk of reruptures is essential when establishing treatment
recommendations. We therefore plan to include a total of 530 patients allowing for
expected loss during follow-up. Ahus, The Emergency Department of Oslo University
Hospital, Drammen and Fredrikstad Hospitals have a catchment area of nearly 1.3 million
people and treats in excess of 300 achilles tendon ruptures annually (Figure 2). This allows
for completion of the inclusion period within three years. In our view, this is the first study
to include a sufficient number of patients to detect lesser, but yet important differences,
and the results will be published in a suitable international peer-reviewed journal.
Table 1

3. Secondary studies

The prospective randomized trial embodies many different disciplines and will spur off
follow-up studies. As part of the main project we will conduct a treatment-cost analysis in
relation to individual results. We will also perform an investigation based on ultrasound
grading of Achilles tendon ruptures according to a classification system established by
Michael Amlang and colleagues. Previous studies have provided evidence suggesting that
treatment results may depend on the extension of the injury, and that treatment
recommendations should be based on type of rupture sustained22,23. This is of particular
interest as it may facilitate individualized treatment. Together with the prospective
randomized trial and the treatment-cost analysis, the ultrasound investigation will
constitute the core of the PhD project. Additionally, The Norwegian School of Sport Sciences
has established a method to assess tendon elasticity by ultrasound examinations, which
would enable us to evaluate the healing process qualitatively24,25. A biomechanical analysis
including three-dimensional gait analysis and pedographic measurements of patients
treated for Achilles tendon ruptures matched with a healthy control group will be carried
out in collaboration with the Norwegian School of Sport Sciences (Pedography registers the
distribution of the plantar pressure load during walking). These examinations may provide a
biomechanical understanding of the pathomechanical changes induced following
treatment. Reduced function in patients treated for Achilles tendon ruptures, regardless of
choice of treatment, is well documented19, but the orthopedic community has largely
ignored the casual relationships.

Figure 5 Sonographic classifications of achilles tendon ruptures. Type 1 ruptures have complete adaptation of the
tendon ends whereas type 2 only displays partial adaptation with (2a) or without (2b) hematoma. In type 3
ruptures there is a diastasis between the tendon ends with (3a) or without (3b) an organized hematoma. Type 4 is
a proximal rupture at the muscle/tendon intersection while type 5 represents distal ruptures near the insertion.

4. Structure of the research group

The study will be carried out as part of a PhD project under supervision of Sigurd Erik
Hoelsbrekken employed at the orthopedic department, Akershus University Hospital. Lars
Engebretsen, Kjetil Hvaal and Øyvind Paulsrud at Oslo University Hospital, Ullevål, will serve
as co-supervisors. Knut Melhuus and Agathe Rønning will be responsible for the
implementation and follow-up at the Emergency Department of Oslo University Hospital.
Asbjørn Sorteberg and Lars Fredrik Høifødt will organize the implementation at Fredrikstad
hospital whereas Finnur Snorrasson and Faisal Butt are responsible for carrying out the
study at Drammen hospital. Karin Rydevik at the Norwegian Sport Medicine Clinic (NIMI)
supervises the physical therapy program as well as overseeing functional testing and
examinations at follow-up. She has also organized a group of centers providing
physiotherapy (Figure 2), and participants of the group hold regular meetings. Its members
treat all patients included in the study, and centers are located in Drammen and Sande
(Vestfold); Fredrikstad, Sarpsborg and Moss (Østfold); Oslo, Frogner and Lillestrøm (Oslo
and Akershus). Two master students in physical therapy under the supervision of Jens
Bojsen-Møller will execute the physical examinations at the Norwegian School of Sport
Sciences. The Dresden group under the leadership of Hans Zwipp has generously allowed us
to share some of their vast knowledge, and has also invited us to visit the university clinic at
Dresden. We have also been in contact with Professor Jon Karlsson at the University
Hospital in Gothenburg and Katarina Nilsson Helander visited Ahus earlier this year to
deliver a lecture on Achilles tendon ruptures. Hopefully the study will facilitate cooperation
across different institutions, and an axis of collaboration has already been established along
the Oslo Fiord. Furthermore, we are in the process of founding an international research
network (Figure 5).

5. Study progress and planning

During the final work on the study protocol, we organized an achilles day at Ahus in
conjunction with a visit by Dr. Amlang from Dresden in May 2012. He held a lecture and
performed two surgical procedures that were streamed by Internet to the participating
hospitals. We were also fortunate enough to welcome Dr. Nilsson Helander to Ahus in June
2012 for another presentation on ruptures of the achilles tendon. By sharing their own
experiences in conducting similar projects, our international collaborators have proven
invaluable in the process of developing the study protocol. The start date of the prospective
randomized trial is set to 15.02.2013, and the study was approved by the Regional Ethics
Committee for Medical Research 11.05.2012 (REK Helseregion Øst, approval number
2012/530 D). For the time being, we are in the process of finishing a trial period aimed at
assessing the planned implementation and to ease cooperation across the different
institutions. Furthermore, the trial period has allowed us to identify and rectify potential
problems and challenges. The inclusion period is planned to last until 31.12.15, and
collection, analysis and preparation of data will be completed by the end of 2016.
Publication of results and writing of the PhD thesis is set to 2017.

Figure 5
Literature
1. Giddings, V. L.; Beaupre, G. S.; Whalen, R. T.; and Carter, D. R.: Calcaneal loading during walking and
running. Med Sci Sports Exerc, 32(3): 627-34, 2000.
2. Jarvinen, T. A.; Kannus, P.; Maffulli, N.; and Khan, K. M.: Achilles tendon disorders: etiology and
epidemiology. Foot and ankle clinics, 10(2): 255-66, 2005.
3. Movin, T.; Ryberg, A.; McBride, D. J.; and Maffulli, N.: Acute rupture of the Achilles tendon. Foot and ankle
clinics, 10(2): 331-56, 2005.
4. Khan, R. J.; Fick, D.; Brammar, T. J.; Crawford, J.; and Parker, M. J.: Interventions for treating acute
Achilles tendon ruptures. Cochrane Database Syst Rev, (3): CD003674, 2004.
5. McMahon, S. E.; Smith, T. O.; and Hing, C. B.: A meta-analysis of randomised controlled trials comparing
conventional to minimally invasive approaches for repair of an Achilles tendon rupture. Foot and ankle surgery :
official journal of the European Society of Foot and Ankle Surgeons, 17(4): 211-7, 2011.
6. Maquirriain, J.: Achilles tendon rupture: avoiding tendon lengthening during surgical repair and rehabilitation.
Yale J Biol Med, 84(3): 289-300, 2011.
7. Wolff, J.: Das Gesetz der Transformation der Knochen. Edited, Hirschwald, 1892.
8. Palmes, D.; Spiegel, H. U.; Schneider, T. O.; Langer, M.; Stratmann, U.; Budny, T.; and Probst, A.: Achilles
tendon healing: long-term biomechanical effects of postoperative mobilization and immobilization in a new
mouse model. J Orthop Res, 20(5): 939-46, 2002.
9. Pneumaticos, S. G.; Phd, P. C. N.; McGarvey, W. C.; Mody, D. R.; and Trevino, S. G.: The effects of early
mobilization in the healing of achilles tendon repair. Foot Ankle Int, 21(7): 551-7, 2000.
10. Twaddle, B. C., and Poon, P.: Early motion for Achilles tendon ruptures: is surgery important? A randomized,
prospective study. Am J Sports Med, 35(12): 2033-8, 2007.
11. Gwynne-Jones, D. P.; Sims, M.; and Handcock, D.: Epidemiology and outcomes of acute Achilles tendon
rupture with operative or nonoperative treatment using an identical functional bracing protocol. Foot Ankle Int,
32(4): 337-43, 2011.
12. Nilsson-Helander, K.; Silbernagel, K. G.; Thomee, R.; Faxen, E.; Olsson, N.; Eriksson, B. I.; and Karlsson,
J.: Acute achilles tendon rupture: a randomized, controlled study comparing surgical and nonsurgical treatments
using validated outcome measures. Am J Sports Med, 38(11): 2186-93, 2010.
13. Tengberg, P. T.; Barfod, K.; Krasheninnikoff, M.; Ebskov, L.; and Troelsen, A.: Promising conservative
treatment using dynamic mobilisation after Achilles tendon rupture. Ugeskr Laeger, 173(44): 2778-84, 2011.
14. Amlang, M. H.; Christiani, P.; Heinz, P.; and Zwipp, H.: Percutaneous technique for Achilles tendon repair
with the Dresden Instruments. Unfallchirurg, 108(7): 529-36, 2005.
15. Nilsson-Helander, K.; Thomee, R.; Silbernagel, K. G.; Thomee, P.; Faxen, E.; Eriksson, B. I.; and Karlsson,
J.: The Achilles tendon Total Rupture Score (ATRS): development and validation. Am J Sports Med, 35(3): 421-6,
2007.
16. Garratt, A. M.; Ruta, D. A.; Abdalla, M. I.; Buckingham, J. K.; and Russell, I. T.: The SF36 health survey
questionnaire: an outcome measure suitable for routine use within the NHS? BMJ, 306(6890): 1440-4, 1993.
17. Melzack, R.: The short-form McGill Pain Questionnaire. Pain, 30(2): 191-7, 1987.
18. Silbernagel, K. G.; Gustavsson, A.; Thomee, R.; and Karlsson, J.: Evaluation of lower leg function in patients
with Achilles tendinopathy. Knee surgery, sports traumatology, arthroscopy : official journal of the ESSKA,
14(11): 1207-17, 2006.
19. Olsson, N.; Nilsson-Helander, K.; Karlsson, J.; Eriksson, B. I.; Thomee, R.; Faxen, E.; and Silbernagel, K.
G.: Major functional deficits persist 2 years after acute Achilles tendon rupture. Knee surgery, sports
traumatology, arthroscopy : official journal of the ESSKA, 19(8): 1385-93, 2011.
20. Carmont, M. R.; Silbernagel, K. G.; Nilsson-Helander, K.; Mei-Dan, O.; Karlsson, J.; and Maffulli, N.:
Cross cultural adaptation of the Achilles tendon Total Rupture Score with reliability, validity and responsiveness
evaluation. Knee Surg Sports Traumatol Arthrosc, 2012.
21. Metz, R.; van der Heijden, G. J.; Verleisdonk, E. J.; Kolfschoten, N.; Verhofstad, M. H.; and van der
Werken, C.: Effect of complications after minimally invasive surgical repair of acute achilles tendon ruptures:
report on 211 cases. Am J Sports Med, 39(4): 820-4, 2011.
22. Kotnis, R.; David, S.; Handley, R.; Willett, K.; and Ostlere, S.: Dynamic ultrasound as a selection tool for
reducing achilles tendon reruptures. Am J Sports Med, 34(9): 1395-400, 2006.
23. Thermann, H.; Zwipp, H.; Milbradt, H.; and Reimer, P.: [Ultrasound sonography in the diagnosis and
followup of Achilles tendon rupture]. Unfallchirurg, 92(6): 266-73, 1989.
24. Bojsen-Moller, J.; Hansen, P.; Aagaard, P.; Svantesson, U.; Kjaer, M.; and Magnusson, S. P.: Differential
displacement of the human soleus and medial gastrocnemius aponeuroses during isometric plantar flexor
contractions in vivo. J Appl Physiol, 97(5): 1908-14, 2004.
25. Fukashiro, S.; Itoh, M.; Ichinose, Y.; Kawakami, Y.; and Fukunaga, T.: Ultrasonography gives directly but
noninvasively elastic characteristic of human tendon in vivo. Eur J Appl Physiol Occup Physiol, 71(6): 555-7,
1995.
Summary of changes to the trial protocol:

2.3 Surgical procedures

Local anesthesia was the preferred choice of anesthesia, however surgical procedures were
not limited to local anesthesia, and other types of anesthesia were used dependent upon
indication.

Open surgical repair of the Achilles tendon was performed with a three-knot double layer
Krakow whip suture. Sutures were tightened so that the ankle joint demonstrated an
equinus position comparable to the non-injured foot. A slight overcorrection was also
accepted.

Three Ethibond 0 sutures with straight cutting needles were used for minimal invasive
surgery. Sutures were tightened so that the ankle joint demonstrated an equinus position
comparable to the non-injured foot. A slight overcorrection was also accepted.

All changes were implemented before inclusion of the first patient.

The Statistical Analysis Plan of September 24th 2020 has not been changed and is final:

Administrative information:
Sponsor name Akershus University Hospital

Sponsor address Sykehusveien 25, 1478 Lørenskog, Norway

EudraCT number / REC no 2012/530

Treatment results after Achilles tendon rupture: A multi-

Trial title center, randomized, clinical trial comparing conservative
treatment, open repair and minimal invasive surgery.
Trial ID The Achilles tendon project, nr 2739005

Trial registration number NCT01785264

SAP and protocol version:

SAP version and date: This SAP is version 1, dated 24. September 2020

Protocol version This document has been written based on information

contained in the study protocol version 1.1, dated 29.
January 2013

SAP revision history:

Protocol SAP Section number Description and reason for change Date changed
version version changed

1.0 1.0 NA First edition of SAP. 24 Sep 2020

 ABBREVIATIONS
ASA Acetylsalicylic acid
ATRS Achilles tendon Total Rupture Score
BMI Body Mass Index
CCS Complete Case Set
CI Confidence Interval
CMJ Counter-Movement Jump
CT Conservative Treatment
FAS Full Analysis Set
LSI Limb Symmetry Index
MCS Mental Component Summary
MIS Minimal Invasive Surgery
OR Open Repair
PCS Physical Component Summary
PF Physical Function
PPS Per Protocol Set
PROM Patient Reported Outcome Measure
RCT Randomized Controlled Trial
SD Standard Deviation
SF-36 Short Form 36
TABLE OF CONTENTS

1 INTRODUCTION ...................................................................................................................... 27
1.1 Background and rationale .................................................................................................. 27
1.2 Trial Objectives .................................................................................................................. 27
2 TRIAL METHODS .................................................................................................................... 27
2.1 Trial Design ........................................................................................................................ 27
2.2 Randomisation ................................................................................................................... 27
2.3 Sample size........................................................................................................................ 27
2.4 Statistical Framework ......................................................................................................... 28
2.5 Statistical Interim Analyses and Stopping Guidance ......................................................... 29
2.6 Timing of Final Analysis ..................................................................................................... 29
2.7 Timing of Outcome Assessments ...................................................................................... 29
3 STATISTICAL PRINCIPLES .................................................................................................... 30
3.1 Confidence Intervals and p-values .................................................................................... 30
3.2 Adherence and Protocol Deviations .................................................................................. 30
3.3 Analysis Populations .......................................................................................................... 30
4 TRIAL POPULATION ............................................................................................................... 31
4.1 Screening Data, Eligibility and Recruitment....................................................................... 31
4.2 Withdrawal/Follow-up ......................................................................................................... 31
4.3 Baseline Patient Characteristics ........................................................................................ 31
5 ANALYSIS ................................................................................................................................ 32
5.1 Outcome Definitions ........................................................................................................... 32
5.2 Analysis Methods ............................................................................................................... 34
6 SAFETY ANALYSES................................................................................................................ 39
6.1 Adverse Events .................................................................................................................. 39
6.2 Clinical Laboratory Parameters.......................................................................................... 39
6.3 Vital Signs .......................................................................................................................... 39
7 STATISTICAL SOFTWARE ..................................................................................................... 39
8 REFERENCES ......................................................................................................................... 40
8.1 Literature References ........................................................................................................ 40
8.2 Reference to Data Management Plan ............................................................................... 19
1 Introduction

1.1 Background and rationale

The management of acute Achilles tendon ruptures has been the subject of a long-standing
debate. Contemporary rehabilitation protocols emphasizing early weight bearing and
mobilization have furthermore blurred differences in treatment results. Open repair (OR)
may reduce the risk of re-rupture, but at the expense of surgical complications such as delay
in wound-healing and infection. This has led to the development of percutaneous and
minimal invasive surgical techniques (MIS). The demonstration of significant differences
comparing OR, MIS and conservative treatment (CT), requires a large sample, and although
the previously reported differences have lacked statistical significance, they may be of
substantial clinical value. We have therefore conducted a three-arm RCT comparing CT, OR
and MIS that is sufficiently large to be able to demonstrate significant differences.

1.2 Trial Objectives

1.2.1 Primary Objective

The primary objective is to assess if one of the three treatments is superior to at least one of
the others measured by the Achilles tendon Total Rupture Score (ATRS) [2, 3] score at 1-2
years post rupture.

1.2.2 Secondary Objectives

The secondary objectives of this study are to assess if there are differences between the
three treatment groups with regards to
• The ATRS at 3 and 6 months.
• The Short form 36 health survey (SF-36) [4, 5] at 6 months and 1-2 years.
• The physical test results in the Musclelab Measurement system [6] at 6 months and 1-2
years.
• Re-rupture rate between the three treatment groups at 1 year.

2 Trial Methods

2.1 Trial Design

The Achilles study is designed as a randomized, controlled, multicentre, single-country,
comparative study.

2.2 Randomisation
Eligible patients were allocated in a 1:1:1 ratio between CT, OR and MIS. Treatment
allocation was done by block randomization, stratified by hospital, with 6, 9 or 12 patients in
each block. Details of allocation sequence generation was provided in a separate document
unavailable to those who enrolled patients or assigned treatment.

2.3 Sample size

The sample size was based on the primary endpoint of ATRS at 1-2 years. The calculation
was based on a one-way ANOVA analysis, assuming a standard deviation of 20 in each
treatment group. A total sample size (study completers) of 480 (160 in each group) was
deemed necessary to achieve 80% power to detect a difference of ATRS of 7 between the
groups in two-sided tests at 5% significance level. To compensate for withdrawals/loss to
follow-up, a total enrolment of 530 patients was targeted.

2.4 Statistical Framework

2.4.1 Hypothesis Test

This trial will employ a hierarchical testing procedure as follows:
1. First, a test of overall effect of treatment at 1-2 years is performed. The null
hypothesis is that there is no difference in the change in ATRS score from baseline to
1-2 years between the three study arms. The alternative hypothesis is that there is a
difference between at least two of the three treatment arms. The test will be
performed at the (two sided) 5% significance level.

2. If the overall test shows that there is a significant difference on the 5% level, then
tests of a difference in the change from baseline to 1-2 years ATRS score will be
performed pairwise between the treatment arms. The null hypotheses in these tests
will be that there is no difference between the treatment arms. The alternative
hypotheses are that there is a difference between the two treatment arms being
compared. These tests will be performed at the (two sided) 5% significance level.

The hierarchical testing procedure abides by the closed testing procedure, allowing a 5%
significance level to be used in each test, while at the same time maintaining a family-wise
error rate at 5%.
All other tests than described in step 1 and 2 above, will be regarded as supportive or
exploratory.
Note that if the test in step 1 shows that there is no significant difference between
treatment groups, a pairwise comparison between treatment groups will still be reported,
but these are to be considered as secondary analyses.

2.4.2 Decision Rule

This protocol is designed to address a single primary endpoint, in a hierarchical fashion as
described above. A difference in the effect of the treatment arms will be claimed if null
hypothesis in step 1 of the hierarchical testing procedure outlined in Section 2.4.1 is
rejected. That is, the two-sided p-value is less than 5%.
If step one of the testing procedure in step 1 in Section 2.4.1 shows significance, then tests
outlined in step 2 of Section 2.4.1 will be performed.
1. Superiority of the OR group over the MIS group will be claimed if the two-sided p-
value in the test comparing the change from baseline to 1-2 years ATRS score is less
than 5%, and if the effect goes in favour of the OR group. Superiority of the MIS
group will be claimed if the effect goes in favour of this group.
 2. Superiority of the OR group over the CT group will be claimed if the two-sided p-
value in the test comparing the change from baseline to 1-2 years ATRS score is less
than 5%, and if the effect goes in favour of the OR group. Superiority of the CT group
will be claimed if the effect goes in favour of this group.

3. Superiority of the MIS group over the CT group will be claimed if the two-sided p-
value in the test comparing the change from baseline to 1-2 years ATRS score is less
than 5%, and if the effect goes in favour of the MIS group. Superiority of the CT group
will be claimed if the effect goes in favour of this group.

2.5 Statistical Interim Analyses and Stopping Guidance

There will be no interim analyses in this trial.

2.6 Timing of Final Analysis

The main analysis is planned when all patients have concluded a minimum of 334 days of
follow up, all data up to two years have been entered, verified and validated and the
primary database has been locked.

2.7 Timing of Outcome Assessments

For all clinically planned measures, visits should occur within a window of the scheduled
visit. Visits outside visit window is regarded a protocol deviation. The target day and visits
window are defined in the protocol as:
Visit Label Target Day Definition (Day window)
Diagnosis -1 Prior to Day 0
V1. Baseline Day 0 (Randomization) Day 0
V2. Surgery/Casting Day 1 < 8 days from injury
V2. 2 weeks 14 days from casting Target day ± 7 days
V3. 4 weeks 28 days from casting Target day ± 7 days
V4. 8 weeks 56 days from casting Target day ± 14 days
V5. 3 months 91 days from casting Target day ± 30 days
V6. 6 months 182 days from casting Target day ± 30 days
Last study visit* 364 days from casting 334 to 728
*The patients are physically tested by a physiotherapist at 1-2 years. Most often they are
asked to complete the PROMs prior to the testing, either by link send by e-mail, on site on a
tablet, or on paper when a tablet is not present or online. Some patients did not show up for
the 1-2 years testing but did complete the PROMs. Some patients showed up for testing at 1-
2 years but did not complete the questionnaires. The last study visit is defined as the last of
either the 1-2 years test day or completion of the 1-2 years PROMs.

For analysis and tabulation purposes, we define study time points as

Time Point Label Target Day Definition (Day window)
TP1. Baseline Day 0 (Randomisation) Information up to
randomisation + 14 days
TP2. Month 3 91 Days 61 to 121
TP3. Month 6 182 Days 151 to 212
TP4. Year 1-2 364 Days 334 to 728
3 Statistical Principles

3.1 Confidence Intervals and p-values

All calculated p-values will be two-sided and compared to a 5% significance level. If a p-value
is less than 0.05, the null hypothesis in the test will be discarded. Efficacy estimates for the
comparison of specific treatment arms will be presented with two-sided 95% confidence
intervals. This trial utilizes a hierarchical testing procedure for the primary endpoint (see
Section 2.4.1), which maintains the type I error rate at 5%.

3.2 Adherence and Protocol Deviations

3.2.1 Adherence to Allocated Treatment

The number and proportion of patients that received the intervention they were
randomized to will be presented.

3.2.2 Protocol Deviations

The following are pre-defined major protocol deviations regarded to affect the efficacy of
the intervention:
• Entering the trial when the eligibility criteria should have prevented trial entry
o Lack of sufficient skills in Norwegian language to understand the questionnaires.
o Outside the age criteria (18-60 years).
o Misdiagnosed.
o Earlier injury to one of the Achilles tendons.
• Received other intervention than allocated to.
• Lack of technical competence for randomized surgical method.

The number (and percentage) of patients with major protocol deviations will be summarised
by treatment group with details of type of deviation provided. All randomized patients will
be used as the denominator to calculate the percentages. No formal statistical testing will be
undertaken.

3.3 Analysis Populations

We define the following patient population in this trial.
• All randomized patients: All patients that have been randomized regardless if they actually
received treatment or not.
• Full analysis set (FAS): All patients that are randomized, received treatment, and where ATRS
was measured at least once post baseline (i.e. at 3 months, 6 months or 1-2 years). Patients
are allocated to the treatment they were randomized to.
• Per protocol set (PPS): Similar to the FAS, but patients are allocated to the treatment they
actually received.
• Complete Case set (CCS): The subset of patients in the FAS that has ATRS measurements at
all follow-up visits. Patients are allocated to the treatment they were randomized to.

The FAS will be used for the primary analysis, while the PPS and the CCS will be used for
sensitivity analyses.
4 Trial Population

4.1 Screening Data, Eligibility and Recruitment

The total number of screened patients and reasons for not entering the trial will be
summarised and tabulated.
A CONSORT flow diagram (appendix A) will be used to summarise the number of patients
who were:
• assessed for eligibility at injury
• eligible at injury
• eligible and randomised
• eligible but not randomised*
• received the randomised allocation
• did not receive the randomised allocation*
• lost to follow-up*
• randomised and included in the primary analysis
• randomised and excluded from the primary analysis*

*reasons will be provided.

4.2 Withdrawal/Follow-up
The status of eligible and randomised patients at trial end will be tabulated by treatment
group according to whether they
• completed intervention, but not assessments.
• completed assessments, but not intervention.
• withdrew consent.
• lost to follow-up.
• excluded due to delay from injury to surgery (>7 days).
• excluded due to protocol deviations after randomisation but occurring prior to surgery.
• Unable to measure the primary endpoint due to:
o comorbidity that compromise rehabilitation or testing.
o death during follow up.
o re-rupture on the contralateral side during follow up.
o other surgery on the lower extremities during follow up.

Time from randomisation to treatment discontinuation and time from randomisation to

withdrawal/lost to follow-up will be presented graphically using a CONSORT flow diagram.

4.3 Baseline Patient Characteristics

The patient demographics and baseline characteristics to be summarised include study
centre, age in years, gender, Body mass index (BMI), injured side (left/right), education,
smoking status, ASA1/ASA2 and baseline ATRS.
Patient demographics and baseline characteristics will be summarised by randomised
treatment arm and overall using descriptive statistics (N, mean, standard deviation, median)
for continuous variables, and number and percentages of patients for categorical variables.
Any clinical important imbalance between the treatment groups will be noted.
5 Analysis

5.1 Outcome Definitions

5.1.1 General Definitions and Derived Variables

5.1.1.1 Body Mass Index

Body Mass Index (BMI) = Body weight in kilograms divided by the square of the height in meters.

5.1.1.2 Achilles tendon Total Rupture Score

The acute Achilles tendon Total Rupture Score (ATRS) was developed in Sweden and
published in 2007 (2). The ATRS is a patient reported outcome measure (PROM) designed to
assess outcomes in patients with ruptures of the Achilles tendon. It contains ten questions,
and each question is answered on an 11-point Likert scale ranging from 0 to 10. The total
score is calculated by summing the individual Likert items. A score of 100 represents the
absence of symptoms, whereas a score of 0 represents severe symptoms. The Norwegian
version of the ATRS was validated, reliability tested and published in 2017/2018 (3). The
Norwegian adaption of the ATRS demonstrates acceptable validity and reliability for use in
the Norwegian population to assess clinical outcomes in patients with Achilles tendon
ruptures.
If any of the 10 questions lacks an answer, this trial will follow the ATRS manual (8). Most
importantly, if one or more of the 10 questions lacks an answer, the total ATRS score is
treated as a missing value.

5.1.1.3 Short Form 36

The Short Form 36 (SF-36) is a self-assessment health status questionnaire composed of 36
questions sorted into eight multi-item scales. The SF-36 also provides two summarized
measures represented by the physical component summary (PCS) and the mental
component summary (MCS). The validity and reliability of the Norwegian translation of SF-36
have been found to be satisfactory (4, 5).

5.1.2 Primary Outcome Definition

ATRS was assessed at baseline, 3 months, 6 months and 1-2 years post injury. The primary
outcome is the change in the ATRS score from baseline to 1-2 years post injury. The last
follow-up was planned 12 months post-injury, preferably at the 12 months test visit at one
of the study physiotherapists. Patients that did not conduct the test battery or of some other
reason did not complete the questionnaires at 12 months were contacted by phone and
asked to answer the PROMs electronically via an encrypted link to the questionnaires sent
by e-mail. In order to minimize loss to follow up the patients were allowed to answer the
questionnaires up to 24 months post injury. This time extension was not done for the last
included patients who did not answer the last set of PROM´s or did not meet for the 12
months testing as the registration of completed forms and testing of patients was stopped
12 months after the last patient was included. An earlier publication reported only minor
improvements occurring between the 1- and 2-year evaluations (7).

5.1.3 Secondary Outcomes Definitions

 5.1.3.1 SF-36
The 36-Item Short Form Health Survey (SF-36) was assessed at baseline, 6 months and 1-2
years post injury. The PCS and MCS is composed of five and three subscales, respectively.
The two SF-36 component summaries PCS and MCS in addition to the subscale physical
function (PF) will be compared between the three treatment arms. As the ATRS and SF-36
are presented simultaneously, the time extension of up to 24 months post injury also applies
to 12-month assessment of the SF-36 questionnaire. The difference in scores from baseline
will be assessed at the 6 months and 1-2 years visits.

5.1.3.2 Re-rupture
Re-rupture is here defined as a total rupture of the same Achilles tendon as treated in the
trial within the 12-month visit. Re-rupture will be treated as a dichotomous outcome. The
diagnosis is done clinically without any need of imaging techniques. Re-ruptures were
continuously reported to the principal investigator of the study. Patients suffering from
either a re-rupture of the same Achilles tendon or a new rupture of the contralateral Achilles
tendon were excluded from further testing and questionnaires.

5.1.3.3 ATRS score at intermittent visits

The change in ATRS scores from baseline to the intermittent visits (3 and 6 months) are
regarded as continuous secondary endpoints.

5.1.3.4 Muscle lab measurements

The following 6 muscle lab measurement endpoint will be considered.
1. Heel rise height (cm): For description see 3.
2. Hopping: A continuously rhythmical jump similar to skipping. The patients performed 25
jumps. The average air flight and floor contact times were documented, and the plyometric
quotient (flight time/contact time) was used for data analysis.
3. Drop counter-movement jump (drop CMJ): The patients started by standing on one leg on a
20-cm-high wooden box. They were instructed to “fall” down onto the floor and, directly on
landing, perform a maximum vertical one-legged jump. The maximum jumping height in
centimeters was used for data analysis (see 1).
4. Concentric heel rise: For description see 5.
5. Eccentric-concentric heel rise: For the strength tests (4 and 5), a linear encoder was used. A
spring-loaded string was connected to a sensor inside the linear encoder unit. When the
string was pulled, the sensor gave a series of digital pulses proportional to the distance
travelled. The resolution is approximately 1 pulse every 0.07 mm. By counting the number of
pulses per time, the displacement as a function of time can be recorded and thus allow
calculation of time, length, velocity, force, and power (force x velocity). In this experiment,
the spring-loaded string of the linear encoder was attached to the heel of the participant´s
shoe and thus the height (in centimeters) and time (in seconds) of the heel displacement of
the heel-rise could be measured. The weight of the participant and the extra external weight
were entered into the MuscleLab software and peak power in watts was calculated. The best
trial (ie, with the highest power in watts) for each weight was used for data analysis.
6. Muscular endurance test: A standing heel-rise test. The total amount of work performed (in
joules) and the maximum heel-rise height were used for data analysis.
For each of these endpoints, a similar measurement for the healthy Achilles is performed.
The ratio of the test result in the treated foot versus the healty foot is calculated and
multiplied with 100. This measurement is referred to as the Limb symmetry index (LSI).
These are also considered as secondary endpoints. Furthermore, the physical tests are
performed both at the 6-month visit and at the 1-2 years visit, the results of which are both
considered as secondary endpoints. Furthermore, the change in the test results, from the 6-
month to the 1-2 years visit, are also considered secondary endpoints. Thus, there are 36
secondary endpoints relating to the physical tests.

5.1.4 Overview of Outcomes

Level Outcome Timeframe Type

Primary ATRS 1-2 years Continuous
Secondary SF-36 1-2 years Continuous
Re-rupture rate 12 months Dichotomous
Muscle-lab measurement 6 and 1-2 years Continuous
system (see 5.1.3.4 for
description)

5.2 Analysis Methods

5.2.1 Primary Outcome

5.2.1.1 Primary Analysis

The primary outcome in this trial is the change in the ATRS score from baseline to the 1-2
years visit.
The primary analysis will be performed by applying a linear mixed effects model to model
the longitudinal change from baseline ATRS measurements at the 3-month, 6-month and 1-
2-year visits. A fixed interaction term between the follow-up visit and treatment group will
be included in the model. The baseline ATRS measurement will be included as a fixed
covariate in the model. Study centre, which was used as a stratification variable in the
randomization, will also be included as a fixed covariate. An individual-specific random
intercept term will be included in the model to account for the dependencies within
individuals. An unstructured correlation structure will be assumed for the repeated
measurements.
From the fitted model, the estimated marginal means will be computed for each time point
and each treatment. Based on the estimated marginal means, an omnibus test, where the
null hypothesis is that there is no difference between the three treatment groups, will be
performed for the 1-2-year visit. This corresponds to step 1 in Section 2.4.1. If the test is
statistically significant on the 5% level, pairwise testing between the three treatment groups
will be performed for the 1-2-year visit (step 2 in Section 2.4.1). A 5% significance level will
be used in each of these tests. As described in Sections 2.4.1-2.4.2, the hierarchical testing
procedure maintains the overall type I error rate at 5%. The estimated marginal means with
95% confidence intervals of the change in ATRS from baseline will be presented for each of
the treatment groups for the 1-2 years visit.
If the omnibus test at the 1-2 years visit does not show statistical significance, then no
further hypothesis testing will be performed. However, the estimated marginal means and
the accompanying 95% confidence intervals will still be presented, but should be considered
as secondary analyses.
The analysis will be conducted in the FAS.

5.2.1.2 Summary Measures

The primary effect estimate will be the change in the ATRS from baseline to the 1-2 years
visit. If the omnibus test of a treatment difference at the 1-2 years visit is significant, the
estimated marginal mean change in ATRS from baseline will be displayed for the three
treatment together with the p-values of the pairwise tests of the three treatment groups
(See Sections 2.4.1, 2.4.2 and 5.2.1.1). If the omnibus test does not identify a treatment
difference, then no pairwise testing will be performed, and no p-values will be presented.
However, the 1-2 years estimated marginal mean change in ATRS from baseline will be
reported for each treatment group, together with 95% confidence intervals, but these
should be regarded as secondary analyses.

5.2.1.3 Assumption Checks and Alternative Analyses

Assumption checks of the fitted model will be done by visual inspection of the residuals.
Marginal residuals will be plotted against the predicted marginal responses, and a check for
trend will be performed. To check if the dependencies in the data has been sufficiently
accounted for by the fitted model, inspection of scatter plots of conditional residuals will be
done for all combinations of visits (i.e. visit 3-month vs. 6-month, 3-month vs 1-2 years and
6-month vs. 1-2 years). Normality of conditional residuals will be assessed by inspecting qq-
plots for each visit.
If there are substantial deviation for normality, or there are indication of substantial residual
dependence, then further random effects will be introduced in the mixed model.

5.2.1.4 Missing Data

At baseline and at the three follow-up visits (3-month, 6-month and 1-2-year), patients
answer the ATRS questionnaire. Each questionnaire contains ten questions, and each
question is answered on an 11-point Likert scale ranging from 0 to 10. The total score is
calculated by summing the individual Likert items. For each question, the value is set
according to the guidelines in the ATRS manual (8). Importantly, if at least one of the 10
questions lack an answer, the ATRS is regarded as missing for the patient for that particular
visit.
Missing baseline ATRS data will be imputed using mean imputation (9). If baseline ATRS are
skewed, median imputation will be used instead.
Missing values in follow-up ATRS measurements (at the 3-month, 6-month and 1-2-year
visits) will be handled by the mixed modelling approach. However, individuals with no
follow-up measurements will then be excluded from the analysis. If the proportion of
patients that has no follow-up measurements is less than 5%, no action will be taken to
include these in the analysis. If this proportion is larger than 5%, multiple imputation will be
used to impute the 1-2 years ATRS for these individuals.

5.2.1.5 Sensitivity Analyses

The following sensitivity analyses will be performed:
1. If there are patients with missing baseline ATRS, a sensitivity analysis will be conducted in the
patients that has baseline a measurement (i.e. excluding the patients with missing baseline
measurement).
 2. If the missing baseline ATRS measurements were imputed using median imputation (see
Section 5.2.1.4), a sensitivity analysis will be conducted using mean imputation instead.
3. The ATRS questionnaire is known to sometimes be misinterpreted, and inverted, by patients
(10, 11). If the study group suspects that this has happened for one or more patients at any
time point, a sensitivity analysis will be conducted where these possibly inverted ATRS
measurements are inverted back. I.e. if a patient e.g. has answered the value 2 on an item in
the questionnaire, this is converted to 10-2=8. This is only done for patients where the study
group suspects that the scale has been inverted.
4. A sensitivity analysis will be conducted in the PPS, the per protocol set. That is, if any patients
received a different treatment than randomized to, they will in this sensitivity analysis be
placed in the treatment group corresponding to the treatment they actually received.
5. A sensitivity analysis will be conducted in patients that has complete follow-up ATRS
measurements, the complete case set (CCS).

5.2.1.6 Subgroup Analyses

No subgroup analyses will be performed.

5.2.2 Dichotomous Secondary Outcome: Re-ruptures

5.2.2.1 Main Analysis

The probability of re-rupture will be compared pairwise between the three treatment
groups.
A logistic regression model will be fitted to the data, where the treatment variable will be
adjusted by study centre (the stratification factor used in the randomization). From the
model, the estimated marginal means of the risk difference between the groups will be
reported pairwise.
We expect that the number of re-ruptures will be low, which may lead to the normality
approximation to the binomial distribution becoming questionable. If any of the expected
numbers in one of the cells in the 3-by-2 contingency table, summarizing the number re-
ruptures in the treatment groups, is less than 5, the following analysis will be performed in
place of fitting the logistic regression model: The confidence intervals of the risk difference
between the groups will pairwise be calculated using the Newcombe Hybrid Score[12]. Thus,
the risk difference will not be adjusted for study centre.
The analysis will be conducted in the FAS.

5.2.2.2 Summary Measures

The 3-by-2 Contingency table (see Section 5.2.2.1) will be presented.

5.2.2.3 Assumption Checks

A check of the expected numbers in the 3-by-2 Contingency table (see Section 5.2.2.1) will
be done. If any of these numbers are less than 5, the non-parametric Newcombe Hybrid
Score will be used to calculate the risk differences between the groups pairwise

5.2.2.4 Missing Data

All patients are followed-up for re-ruptures until 12 months. The participating hospitals are
instructed to report any re-rupture (total or partial) detected during the follow up period
directly to the principal investigator. This will secure that the chance of missing the count of
any re-rupture is regarded as insignificant.

5.2.2.5 Sensitivity Analyses

No sensitivity analyses will be performed.

5.2.2.6 Subgroup Analyses

No subgroup analyses will be performed.

5.2.3 Continuous Secondary Outcome: SF-36v2

5.2.3.1 Main Analysis

The outcome is the change in the SF-36 score from baseline to the 6-month and the 1-2-year
visit. The two summary scores, PCS and MCS, as well as the sub-score PF will be considered.
Thus, there are three SF-36-based secondary outcome variables. Note that the SF-36 scores
will be standardized to the 1998 US population, as provided by the ProCore software by
Qualitymetric (https://www.optum.com/business/solutions/life-sciences/answer-
research/patient-insights/sf-health-surveys.html).
The analyses will be conducted in the FAS.

5.2.3.2 Summary Measures

The 6-month and 1-2-year visit estimated marginal mean change in SF-36 score from
baseline will be reported for each treatment group, together with 95% confidence intervals.

5.2.3.3 Assumption Checks

Similar as for the primary endpoint.

5.2.3.4 Missing Data

Partially answered SF-36 questionnaires will be treated according to the SF-36v2 user´s
manual [13]. Note that the manual allows for a certain degree of missingness in a given
questionnaire. If the manual deemed a questionnaire as not sufficiently answered, then the
score value for the individual at the visit in question will be treated as a missing value, in the
same manner as completely unanswered questionnaires.
Missing baseline SF-36 score data will be imputed using mean imputation (9). Missing values
in follow-up SF-36 scores (at the 6-month and the 1-2 years visits) will be handled by the
mixed modelling approach. However, individuals with no follow-up measurements will then
be excluded from the analysis. If the proportion of patients that has no follow-up
measurements is less than 5%, no action will be taken to include these in the analysis. If this
proportion is larger than 5%, multiple imputation will be used to impute SF-36 scores for
these individuals.

5.2.3.5 Sensitivity Analyses

No sensitivity analyses will be performed.

5.2.3.6 Subgroup Analyses

No subgroup analyses will be performed.
 5.2.4 Continuous Secondary Outcome: Muscle lab measurements
The six muscle lab measurement variables will be analysed in the same way.

5.2.4.1 Main Analysis

There are three outcomes for each physical test:
• The difference between the 6-month and the 1-2 years test performance.
• The 6-month test performance.
• The 1-2-year test performance.

Linear regression models will be fitted to each of these outcomes, for each physical test. The
treatment variable will be adjusted by study centre (the stratification variable used in the
randomization). Note that for each of the 6 physical tests at each visit, a test is also
performed in the healthy foot. The ratio between the performance in the treated foot and
the healthy foot, multiplied with 100, (the LSI) is also considered an outcome variable for
each physical test.
The analyses will be conducted in the FAS.

5.2.4.2 Summary Measures

The effect estimates will be the pairwise differences in the estimated marginal means (that
is, mean difference adjusted for study centre) between the treatment groups, reported
together with 95% confidence intervals.

5.2.4.3 Assumption Checks

Visual inspection of residual plots will be performed (qq-plots, histograms and residuals vs.
fitted values). Substantial deviations from normality will be addressed by suitable
transformations of the data.

5.2.4.4 Missing Data

We expect three types of missing data for these outcome variables:
1. Individuals is present in the gym, tries to perform the test, but do not manage to fulfil the
minimum requirements to have the test approved. These values are in reality censored
rather than missing.
2. Value may be missing due to e.g. non-working equipment in the gym where the test takes
place, even if the patient were physically there.
3. Values may be missing because the patient did not show up for the physical tests.

The following handling of the three types of missingness will be performed:

1. Threshold imputation: These values will be imputed in the following way:
I. The 1% highest and 1% lowest observed values for the variable will be
removed, and a normal distribution will be fitted to the ‘trimmed’
observations.
II. Values “klarte ikke” (did not manage) will be replaced by drawing values from
the distribution in point I, truncated by 0 and the 10th percentile of the
distribution.
2. These values are assumed to be missing at random, and multiple imputation by chained
equation will be performed.
 3. Multiple imputation using chained equations will be used to impute these values.

The multiple imputation in point 2 and 3 will be done simultaneously, after point 1 has been
performed. The following variables will be used to impute the test scores at 6 months and 1-
2 years: Sex, Age, Weight at baseline, weight at the time of the test, study centre, results in
the other 5 physical tests, results from all 6 physical test in the healthy foot, and ATRS scores
at 6 and 1-2 years.
Note that the LSI variables, which is the ratio between the injured and the healthy foot
multiplied by 100, will not be imputed as described above. For the threshold imputation, the
value in the injured and healthy foot will be imputed separately, and the LSI value is
obtained from these. In the multiple imputation procedure, the LSI variables for the
different tests will be imputed by passive imputation.

5.2.4.5 Sensitivity Analyses

Performing the analysis described in Section 5.2.4.1 on the population of patients that
showed up for the physical test. I. e. after the threshold imputation described in Section
5.2.4.4 has been performed.

5.2.4.6 Subgroup Analyses

No subgroup analyses will be performed.

5.2.5 Time to event secondary outcomes

Not applicable.

5.2.6 Additional Analyses

Not applicable.

6 Safety Analyses
Complications and adverse events were continuously registered. For instance, thromboembolic
events, wound healing problems, infections and nerve-damage.

6.1 Adverse Events

Any complications and adverse events were continuously documented and any new rupture
on the injured side (re-rupture) or the healthy side were immediately reported to the study
coordinator. Re-ruptures are considered as a secondary outcome (see Section 5.1.3.2).
Adverse events other than re-ruptures will be tabulated.

6.2 Clinical Laboratory Parameters

Not applicable.

6.3 Vital Signs

Not applicable.

7 Statistical Software
All statistical analyses will be done using R version 3.6.3 (R Core Team (2020). R: A language
and environment for statistical computing. R Foundation for Statistical Computing, Vienna,
Austria. URL https://www.R-project.org/).
8 References

8.1 Literature References

1. Guidelines for the Content of Statistical Analysis Plans in Clinical Trials.
2017;318(23):2337-2343. doi:10.1001/jama.2017.18556.
2. Nilsson-Helander K, Thomee R, Silbernagel KG, Thomee P, Faxen E, Eriksson BI et al
(2007) The Achilles tendon Total Rupture Score (ATRS): development and validation. Am
J Sports Med 35:421-426
3. Myhrvold SB, Sandnes Ø, Hoelsbrekken SE (2018) Validity and reliability of the
Norwegian translation of the Achilles tendon Total Rupture Score. Knee Surg Sports
Traumatol Arthrosc 26:2045-2050
4. Tests of Data Quality, Scaling Assumptions, and Reliability of the SF-36 in Eleven
Countries: Results From the IQOLA Project, International Quality of Life Assessment. J
Clin Epidemiol. 1998 Nov;51(11):1149-58. doi: 10.1016/s0895-4356(98)00106-1
5. Measurement properties and normative data for the Norwegian SF-36: results from a
general population survey. Health and Quality of Life Outcomes 15, Article number: 51
(2017)
6. Evaluation of Lower Leg Function in Patients With Achilles Tendinopathy. Knee Surg
Sports Traumatol Arthrosc. 2006 Nov;14(11):1207-17.doi: 10.1007/s00167-006-0150-6
7. Major Functional Deficits Persist 2 Years After Acute Achilles Tendon Rupture. Kne Surg
Sports Traumatol Arthrosc. 2011 Aug;19(8):1385-93. doi: 10.1007/s00167-011-1511-3
8. Performance of the Achilles Tendon Total Rupture Score Over Time in a Large National
Database: Develepment of an Instruction Manual for Accurate Use. Am J Sports Med.
2020 May;48(6):1423-1429. doi: 10.1177/0363546520912222.
9. Sullivan, T. R., White, I. R., Salter, A. B., Ryan, P., & Lee, K. J. (2018). Should multiple
imputation be the method of choice for handling missing data in randomized
trials?. Statistical methods in medical research, 27(9), 2610-2626).
10. Cross cultural adaptation of the Achilles tendon total rupture score with reliability,
validity and responsiveness evaluation. Knee Surg Sports Traumatol Arthrosc
2013;21(6):1356-1360.
11. Reliability and validation of the Dutch Achilles tendon Total Rupture Score. Knee Surgery,
Sports Traumatology, Arthroscopy. July 2016
12. Fagerland, Morten, Stian Lydersen, and Petter Laake. Statistical analysis of contingency
tables. CRC press, 2017.
13. User´s manual for the SF36v2 Health Survey. ME Maruish, M Maruish, M Kosinski, JB
Bjorner, B Gandek, DM Turner-Bowker, JE Ware. 2011. www.scienceopen.com.

8.2 Reference to Data Handling Plan

Not applicable.