Deltoid ligament reconstruction: A novel technique with biomechanical analysis

Steven L. Haddad, Sunil Dedhia, Yupeng Ren, Jason Rotstein, Li Qun Zhang

Research output: Contribution to journalArticlepeer-review

64 Scopus citations


Background: Deltoid ligament insufficiency has been shown to decrease tibiotalar contact area and increase peak pressures within the lateral ankle mortise. This detrimental effect may create an arthritic ankle joint if left unresolved. Reconstructive efforts thus far have been less than satisfactory. We describe a novel technique that reconstructs both main limbs of the deltoid ligament in anatomic orientation while providing secure graft fixation. Materials and Methods: Six pairs of fresh frozen cadaveric lower extremities were utilized. Matched right and left lower limbs (one pair) were allocated either to a deltoid reconstruction group or an intact deltoid group. The anterior tibial tendon was chosen as the graft for ligament reconstruction, and was harvested from the ipsilateral specimen. Tunnels were created in the distal tibia at the deltoid origin, and at the talus (deep) and calcaneus (superficial) deltoid insertions. Following measurement, the graft was cut to the appropriate size and endobuttons weaved into both tendon ends. The graft ends were passed through the talus and calcaneus respectively. The residual graft loop was then routed through the tibial tunnel and secured proximally with a cancellous screw post and spiked washer. Following specimen mounting, a multiaxis testing apparatus with three separate motors allowed three planes (dorsiflexion/ plantarflexion; inversion/eversion; and internal/external rotation) of motion. Angular rotations and linear translations of the tibia in the X-Y-Z directions were measured for a given torque in external/internal rotation, dorsiflexion/plantarflexion, or eversion/inversion, under a constant velocity of 2 degrees per second. Testing consisted of a 2 Nm preload for 20 cycles in internal rotation/external rotation and inversion/eversion prior to data collection of 10 cycles at this level of torque. Similarly, a preload of 1 Nm for 20 cycles was used in dorsiflexion/plantarflexion prior to data collection of 10 cycles at this torque level. Data were collected in the control specimens (the matched contralateral extremity) with the deltoid ligament intact, and following complete sectioning of the ligament complex (both bundles). Results: Angular displacement at a 2 Nm level torque was significantly greater in the sectioned group compared to the deltoid reconstruction group in external rotation and eversion (p = 0.006 and p = 0.017 respectively). There was no statistical difference in angular displacement between the deltoid intact and reconstructed group in external rotation and eversion when tested at 2 Nm of torque (p = 0.865 and p = 0.470, respectively). The stiffness of the reconstruction was 136.4 ± 40.2% compared to the intact ligament. Stiffness data were statistically insignificant in both plantar flexion and dorsiflexion between the reconstructed and sectioned groups (p = 0.050 and p = 0.126). Conclusion: The described reconstruction technique under low torque was able to restore eversion and external rotation stability to the talus, which was statistically similar to the intact deltoid ligament. This novel technique developed its strength not only from the anatomic orientation of the reconstructed ligament, but the strength of the components chosen to fix the tendon graft to the bone. Clinical Relevance: This utilitarian reconstruction may be incorporated into total ankle arthroplasty, triple arthrodesis, and sports injuries to re-establish lost medial stability.

Original languageEnglish (US)
Pages (from-to)639-651
Number of pages13
JournalFoot and Ankle International
Issue number7
StatePublished - Jan 1 2010

ASJC Scopus subject areas

  • Surgery
  • Orthopedics and Sports Medicine


Dive into the research topics of 'Deltoid ligament reconstruction: A novel technique with biomechanical analysis'. Together they form a unique fingerprint.

Cite this