Study population
This was a prospective, randomized, comparative trial. All patients with acute Achilles tendon rupture (<7 days) were eligible for the study from May 2016 to January 2018 (n = 72 in 19 months). Achilles rupture was diagnosed by clinical examination (the Thompson test) and magnetic resonance imaging (MRI) in each patient [15].
The inclusion criteria were (1) an age of 18 to 45 years, (2) body mass index (BMI) of <34 kg/m2, (3) closed injury, and (4) acute Achilles tendon rupture (<7 days). The exclusion criteria were (1) partial Achilles tendon rupture (diagnosed by means of MRI or intraoperative probe), (2) rupture of the insertion point of the Achilles tendon (<2 cm), and (3) diseases that may affect the results of the functional study tests, such as autoimmune disease, deep vein thrombosis, or neuropathy.
The CONSORT flow chart of patient selection is shown in Figure 1. All patients provided written consent to participate in the study and return for standardized follow-up examinations at 1, 3, 6, and 12 months after surgery. The operative technique was selected by assigning each patient an odd or even number according to their order of hospital admission.
Open repair (bundle-to-bundle suture) was performed in Group A (n = 30), and percutaneous repair (modified Bunnell suture) was conducted in Group B (n = 31). All patients’ physical examinations and operations were performed by the same surgeon (FQ.Z).
Surgical procedures
Bundle-to-bundle suture
The posteromedial Achilles tendon approach was used with the site of Achilles tendon rupture as the center. After the Achilles tendon rupture was exposed, the congested tissue was carefully removed and the ruptured Achilles tendon was shifted to expose the deep soft tissue. The deep soft tissue was sutured first, and the soft tissue bed of the Achilles tendon was then repaired, which was beneficial for the recovery and healing of the Achilles tendon blood supply. The horsetail-like ends of the ruptured tendon were aligned according to the anatomical characteristics of the Achilles tendon. The tendon band of the superficial proximal gastrocnemius was turned laterally and sutured to the band of the distal lateral end of the tendon, while the tendon band of the deep proximal soleus was turned inward and sutured to the band of the distal medial end of the tendon. The Achilles tendon was typically divided into several anatomical bundles, and the tendon bundles were then repaired in an end-to-end fashion from the deep aspect to the lamina using absorbable 4-0 sutures to connect the ends and 5-0 Prolene to reinforce the repair. The long tendon band was sutured to the short tendon band using 4-0 Prolene to avoid a short, retracted Achilles tendon and excessive plantar flexion of the ankle. Nonabsorbable 4-0 suture was used to suture the outer membrane of the Achilles tendon, deep fascia, subcutaneous tissue, and skin. This bundle-to-bundle suture technique was first introduced by Zhao et al. [14] (Fig. 2).
Modified Bunnell suture
One mark was made in the middle of the depression of the ruptured Achilles tendon, and two or three marks were made on both sides of the distal and proximal Achilles tendon. Approximately 11 to 13 small 1-cm incisions were made at each mark to reach the deep fascia. PDS-II suture was threaded through the distal incision, passed through the tendinous tissue, and then threaded through the central incision. Next, a “Z” suture was performed by threading the PDS-II through the proximal incision and then through the central incision. The distal and proximal sutures were tightened to tie the ruptured end of the tendon, and the knot was then examined by the Thompson test. The transverse incision was sutured, and the stab incisions were closed. The sural nerve was protected during the performance of all incisions. This procedure was first introduced by Ma and Griffith [16] (Fig. 3).
Postoperative rehabilitation
The postoperative rehabilitation was performed by an experienced physical therapist. The affected limb was lifted up, and the quadriceps femoris and triceps crus were exercised by isometric contraction and relaxation on the second postoperative day. The stitches were removed at 2 weeks postoperatively. Partial protected weight bearing with walking boots was allowed until the patients had gradually achieved walking with a partial load at 4 weeks; the partial load was then changed to a full load within 4 to 6 weeks. Four layers of wedge insoles were placed inside the walking boots, and one layer was removed every l to 2 weeks. Flexion and extension of the ankle joint was allowed in the neutral position 6 weeks after surgery. After 2 weeks of full weight-bearing exercise with walking boots, the walking boots were removed. Walking with sneakers and functional ankle range-of-motion exercise were allowed at 2 months postoperatively, and heel-lifting functional exercise was allowed at 10 weeks postoperatively.
Functional evaluation
Follow-up was performed 1, 3, 6, 12, and 24 months after surgery. The functional evaluation included both a functional examination and patient-reported outcome measures. The functional examination findings and clinical score were evaluated at 12 months.
The functional examination consisted of three main items: measurement of the calf circumference, measurement of the Achilles tendon width and anteroposterior dimension, and performance of the single-leg heel-rise test. All postoperative evaluations were performed by nonsurgical personnel who were unaware of the patients’ surgical procedures. The calf circumference on the injured side was compared with that of the contralateral limb. The diameter and length of the Achilles tendon on the injured side were compared with those of the contralateral limb by MRI. The number of single-leg heel rises within 15 s (height of >5 cm) was compared with that of the uninjured side [17]. The height of the heel-rise test was monitored by one person. The ankle joint range of motion was recorded.
The patient-reported outcome measures were the Achilles tendon total rupture score (ATRS), American Orthopaedic Foot and Ankle Society (AOFAS) ankle-hindfoot scale score, and visual analog scale (VAS) pain score [18].
Statistical analysis
Statistical analysis was performed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA). The demographic data and all outcome parameters were tested for deviation from the normal distribution. Differences between the injured and uninjured sides and differences between the two treatment groups were tested with a two-tailed, unpaired t-test. Differences in age, sex, BMI, and follow-up duration were examined using a paired-samples t test and Fisher’s test or the χ2 test for continuous variables and categorical data, respectively. Pearson’s test differences in the ATRS, AOFAS score, and VAS score between the two treatment groups were determined with an unpaired-samples t test or the Mann–Whitney rank sum test. Pearson correlation analysis was performed during the functional examination of the patients. A P value of <0.05 was considered statistically significant.