MRI assessment of graft maturation after arthroscopic anatomical lateral ankle ligament reconstruction: One-year comparison between autograft and allograft

Abstract

Background

Chronic lateral ankle instability (CLAI) is a common condition often requiring surgical reconstruction of the anterior talofibular ligament (ATFL) and calcaneofibular ligament (CFL). While autografts are traditionally used, allografts are gaining interest due to reduced donor site morbidity. This study aimed to compare graft maturation between autograft and allograft using a multimodal MRI evaluation at one year.

Methods

This prospective, monocentric cohort study included 36 patients who underwent anatomical reconstruction of the ATFL and CFL between December 2020 and December 2023, using either gracilis autograft (n = 18) or frozen, non-irradiated allograft (n = 18). MRI performed at 8–15 months postoperatively assessed graft integration using three parameters: Signal-to-Noise Quotient Ankle (SNQA), Howell maturity score, and graft-to-bone interface signal. Measurements were performed independently by two blinded observers. Statistical comparison between groups was conducted using Mann–Whitney tests, and interobserver reproducibility was assessed using intraclass correlation coefficients (ICC).

Results

Interobserver reliability was excellent for SNQA (ICC = 0.976 for ATFL, ICC = 0.895 for CFL). No statistically significant differences were observed between autografts and allografts for any parameter: SNQA (ATFL, p = 0.267; CFL, p = 0.494), Howell score (ATFL, p = 0.436; CFL, p = 0.083), or graft-to-bone interface (talus, p = 0.332; calcaneus, p = 0.752; fibula, p = 0.289).

Conclusion

At one year, MRI-based maturation of autografts and allografts in anatomical lateral ankle ligament reconstruction was equivalent across all evaluated criteria. These findings support the use of allograft as a reliable alternative to autograft, particularly in selected patients where autograft harvesting is undesirable. Further longitudinal and functional outcome studies are warranted.

1

Introduction

Chronic lateral ankle instability (CLAI), often secondary to poorly healed recurrent sprains, represents a significant functional challenge in active patients . It has a high recurrence rate, with 20–40 % of patients at risk of developing CLAI . CLAI leads to recurrent sprains and may cause associated lesions such as osteochondral talar dome injury (ODTI), chondropathy, osteoarthritis, or tendinopathies , . Therefore, managing CLAI is essential to prevent long-term functional and articular complications.

Anatomic repairs aim to restore the integrity of damaged tissues and are sometimes reinforced by synthetic or biological augmentation. Reconstructions, on the other hand, involve replacing torn ligaments, most often with a graft. Both approaches can yield good outcomes when performed anatomically . Although the Broström repair remains widely used, anatomic reconstruction using autografts or allografts is preferred in cases of hyperlaxity or insufficient residual ligaments . While autografts remain the reference technique, allografts are gaining interest due to their ability to reduce donor site morbidity while providing comparable clinical outcomes. However, their biological effectiveness, particularly in terms of midterm graft integration, remains debated , .

A key determinant of surgical success lies in tendon-to-bone healing, which is a progressive biological process occurring at the graft–bone interface. As described by Ge et al. , healing begins with the formation of fibrous scar tissue, which is subsequently replaced by Sharpey-like fibers and new bone, gradually anchoring the graft within the tunnel. While a pronounced fibrous layer may indicate suboptimal osteointegration, it does not necessarily impair short-term mechanical stability, though it may influence long-term outcomes.

Because histological evaluation is rarely feasible in clinical practice, MRI has become the preferred non-invasive tool to assess graft healing and integration. It enables the analysis of both the morphology of the bone–graft interface and the tissue characteristics of the graft itself. Among the quantitative parameters derived from MRI, the Signal-to-Noise Quotient (SNQA) reflects signal intensity related to graft vascularity and water content—a high SNQA suggests an immature graft, while a decline over time is indicative of progressing maturation ,, .

Additionally, the Howell score was used to qualitatively assess graft signal outside the bone tunnels, grading the degree of signal homogeneity and ligamentous organization.

This study aimed to evaluate, for the first time, ligamentous integration on MRI one year after lateral ankle ligament reconstruction, comparing autografts and allografts. We hypothesized that both graft types would demonstrate comparable ligamentization profiles at one year postoperatively.

2

Methods

2.1

Study design and population

This was a monocentric, prospective, observational, comparative cohort study. All patients underwent surgery between December 2020 and December 2023, performed by the same surgeon (DA). The study was approved by the institutional ethics committee and conducted in compliance with regulatory requirements (RnIPH registry No. 2023–88, CNIL MR-004 protocol No. 2206723 v0).

Inclusion criteria were: age > 15 years; anatomical reconstruction of the lateral ankle ligaments (ATFL and CFL) using either gracilis autograft or tendon allograft; persistent CLAI symptoms > 3 months despite appropriate conservative treatment, with clinical laxity and MRI-confirmed injury of ATFL and CFL; no prior ankle surgery; and consent for a control MRI approximately one year postoperatively.

Exclusion criteria were: graft rupture before one-year MRI; symptomatic osteochondral talar dome lesion; MRI outside the predefined postoperative time window (8–15 months).

Between December 2020 and September 2022, all reconstructions were performed with autografts, as allografts were not yet available in our institution. From October 2022 onwards, once allografts became accessible, the surgeon systematically used them for ligament reconstruction. This chronological availability determined graft allocation without impacting patient selection, thereby ensuring that both groups remained comparable.

Of the initial 62 patients evaluated for anatomical ATFL and CFL reconstruction, 8 did not meet inclusion criteria, 4 were lost to follow-up, 11 underwent MRI outside the acceptable window (1–7 months or >15 months), 2 had graft rupture (1 autograft, 1 allograft), and 1 had a symptomatic talar dome lesion treated concomitantly. Therefore, 36 patients were included in the final analysis: 18 with autografts and 18 with allografts, ensuring comparable and homogeneous groups ( Fig. 1 ).

Fig. 1

Flowchart (ATFL, Anterior Talofibular Ligament; CFL, Calcaneofibular Ligament).

Of the initial 62 patients evaluated for anatomical ATFL and CFL reconstruction, 8 did not meet inclusion criteria, 4 were lost to follow-up, 11 underwent MRI outside the acceptable window (1–7 months or >15 months), 2 had graft rupture (1 autograft, 1 allograft), and 1 had a symptomatic talar dome lesion treated concomitantly. Therefore, 36 patients were included in the final analysis: 18 with autografts and 18 with allografts, ensuring comparable and homogeneous groups ( Fig. 2 ).

Fig. 2

Anatomic reconstruction technique of the ATFL and CFL.

2.2

Surgical technique ( Fig. 2 )

Both groups underwent anatomical reconstruction of the ATFL and CFL using an arthroscopically assisted technique similar to that described by Lopes et al. , The autograft group received gracilis tendons harvested from the ipsilateral knee, while the allograft group received frozen, non-irradiated tendons obtained from a certified tissue bank. Various allograft tendons were used depending on availability, including extensor hallucis longus (EHL), flexor hallucis longus (FHL), gracilis, flexor digitorum longus (FDL), peroneus longus (PL), and peroneus brevis (PB), all of which have biomechanical properties comparable to native ankle ligaments. Details of patient positioning, graft preparation, fixation, and rehabilitation are described below.

2.2.1

Patient positioning

Patients were positioned supine with the affected leg draped free, and a pneumatic thigh tourniquet was applied. The ankle was maintained in neutral dorsiflexion and eversion throughout the procedure.

2.2.2

Graft harvesting and preparation

In the autograft group, the gracilis tendon was harvested through a small incision at the pes anserinus using a tendon stripper. Both autografts and allografts were prepared by whipstitching both ends with FiberLoop 2.0 sutures (Arthrex, Naples, FL, USA) and soaked in a solution of 500 mg vancomycin diluted in 100 mL saline prior to implantation.

2.2.3

Arthroscopic evaluation and tunnel preparation

Standard anteromedial and anterolateral arthroscopic portals were used to inspect the joint for intra-articular pathologies, including synovitis, chondral lesions, or other ligament injuries. Bone tunnels were drilled at the anatomical insertion sites of the ATFL and CFL: the talus and fibula tunnels were created first, followed by a percutaneous calcaneal tunnel for the CFL.

2.2.4

Graft passage and fixation

The graft was passed sequentially through the talar, fibular, and calcaneal tunnels. Fixation was achieved using biocomposite interference screws (Bio-Tenodesis™, Arthrex) for the talar and calcaneal tunnels and an endobutton device (ACL TightRope® RT, Arthrex) for the fibular tunnel. Appropriate tension was applied to replicate the native ligament properties, and final arthroscopic inspection confirmed proper graft positioning.

2.3

Postoperative rehabilitation

Postoperative care followed a standardized protocol:

  • •

    Phase 1 (0–6 weeks): Weight-bearing as tolerated with an ankle brace and early range-of-motion exercises.

  • •

    Phase 2 (6–12 weeks): Gradual brace weaning, straight-line activity, and muscle strengthening focusing on peroneal muscles and proprioception.

  • •

    Phase 3 (3–6 months): Sport-specific training including balance and agility drills, with return to non-contact sports as tolerated.

  • •

    Phase 4 (>6 months): Full return to competitive sports after favorable clinical and functional evaluation.

2.4

MRI analysis

MRIs were performed on different 1.5-Tesla scanners, following a standardized protocol for all patients. Imaging was acquired with the ankle at 90° of flexion and in neutral pronation-supination. Axial, sagittal, and coronal slices were obtained using proton density fat-suppressed (PD-FS) sequences and volumetric proton density (PD) acquisition. MRIs were performed between 8 and 15 months postoperatively.

All measurements were performed independently and blinded to surgical technique by two orthopedic surgery residents using a PACS workstation (Horizon Rad Station; McKesson).

Three parameters were assessed to evaluate graft integration:

2.4.1

Signal-to-Noise Quotient Ankle (SNQA)

The SNQA was calculated as:

(signal intensity of graft − signal intensity of peroneal tendons)/ background noise (measured in air posterior to the Achilles tendon) .

This measurement was performed separately for the ATFL and CFL grafts ( Fig. 3 ).

Fig. 3

Axial DP FS MRI: Different measurements for SNQA calculation f the ATFL.

2.4.2

Howell Maturity Score

Graft signal intensity outside the bone tunnels was graded according to the Howell scale, which reflects water content and maturation of the graft:

  • •

    Stage 1: low, homogeneous signal indistinguishable from peroneal tendons (normal) ( Fig. 4 ).

    Fig. 4

    Axial DP FS MRI: reconstructed ATFL(*) assessed as grade 1 on the Howell scale (T: Talus, F: Fibula).

  • •

    Stage 2: normal ligament-like signal in > 50 % of the graft.

  • •

    Stage 3: increased signal in > 50 % of the graft with normal signal in the remainder.

  • •

    Stage 4: diffuse increased signal throughout the graft with no normal signal band.

Tags:
Sep 5, 2026 | Posted by in ORTHOPEDIC | Comments Off on MRI assessment of graft maturation after arthroscopic anatomical lateral ankle ligament reconstruction: One-year comparison between autograft and allograft

Full access? Get Clinical Tree
Get Clinical Tree app for offline access