This is a retrospective study of a total of 37 consecutive patients with an average age of 43 years (32–55 years) who were treated with a modified posterior arthroscopic technique to repair irreducible or unstable locked BHMMTs from 2011 to 2016. Approval was acquired from the local ethics committee, and all participants filled out a written informed consent form before their operations. Inclusion criteria were patients aged 55 years or younger with a chronic post-traumatic locked BHMMT in the red-red or red-white zone and a concomitant ACL tear. Chronicity was defined as an interval of more than 12 weeks between the history of the last episode of locking and the surgery [10]. BHMMTs were limited to reducible but unstable locked menisci (Fig. 1A) and irreducible menisci with a complete rotation (Lim et al.’s [10] classification type 3) (Fig. 4C). Exclusion criteria were any other intra-articular surgery or revision surgery, which did not include any of the participants. Furthermore, all stable and reducible BHMMTs were excluded. All patients had a concomitant ACL tear and were operated on by the senior author, and all arthroscopic procedures were recorded. Patient information was extracted from medical records and the hospital PACS system.

A Locked unstable medial meniscus in an anterolateral portal view of a right knee. Anterolateral portal (B) and posterolateral trans-septal (C) views show the torn medial meniscus fixed using a needle (arrow) percutaneously. Posterolateral trans-septal (D), posteromedial (E), and anterolateral portal (F) views after meniscal repair. MM medial meniscus, TP tibial plateau, MFC medial femoral condyle, * arthroscopic sheath, PM.portal posteromedial portal

Magnetic resonance imaging (MRI) was performed preoperatively and 6 months postoperatively for all patients. Lysholm, International Knee Documentation Committee (IKDC), and Tegner activity scale scores were obtained from all patients before surgery. Patients were asked to score themselves on the Tegner activity scale based on their condition before suffering their injury. At 5–10 years (an average of 7 years) after the operation, these patient-reported outcomes were assessed again, and patients were clinically evaluated for meniscal repair failure. Failure was defined as the presence of at least one of Barrett et al.’s [12] criteria, including joint effusion, pain along the medial joint line, a positive provocation test, and the need for meniscectomy. Complete healing was assumed if none of the criteria were met [13].

Patients were placed in the supine position, and the affected limb was positioned over the edge of the operating table to allow access to either side of the knee, while the contralateral leg was placed in a lithotomy position. A high-thigh tourniquet was used. Standard anterior portals were used for routine diagnostic arthroscopic visualization and trans-portal ACL reconstruction, and a 30° lens was used throughout the surgery. First, the semitendinosus autograft was acquired, and the femoral closed socket tunnel was reamed. The meniscal repair process was then carried out.

Using the anterior portals, the adhesion bands of neighboring structures inside the notch were removed to mobilize the meniscus, and tear borders were refreshed in situ with a shaver before the reduction. With the patient’s knee positioned in 30° of flexion with valgus force, the meniscus was reduced by gentle manipulation using an arthroscopic trocar through an anteromedial portal. In the case of complete meniscal rotation (Lim et al.’s [10] classification type 3), a grasper was used through an accessory high anterolateral portal to derotate the fragment. The technique of reducing the torn meniscus depends on the skill of the surgeon and can easily be done using a trocar or grasper. For a better understanding, refer to the technique shown in Additional file 1: Video S1 of this article. Then, a 16-gauge needle was used as a percutaneous provisional fixation; this needle was inserted from anterior to posterior, perpendicular to the meniscal tear, to keep the reduced meniscus in the anatomical position (Fig. 1B and 2). This needle acted like a reduction in–out suture to fix the posteromedial part of the meniscus between the posterior and the middle-third parts at the beginning of the meniscal repair process (Fig. 1C).

Percutaneous provisional fixation in a right knee using a prebent needle inserted from anterior to posterior, perpendicular to the meniscal tear

To maintain better stability, the needle was slightly bent and then passed over the anterior horn of the meniscus and inserted into the posterior horn (Fig. 2). Usually, one needle was sufficient to perform this task, but more needles were sometimes required in cases of gross instability. The needle was removed after completion of the posterior meniscal repair.

After temporary meniscus fixation, posterior arthroscopy was performed following Keyhani et al.’s method [14], and the posteromedial and posterolateral trans-septal portals were created for repair and assessment, respectively. Medial collateral ligament (MCL) pie-crusting was not necessary to improve the visualization and reduction of the meniscus. Healing was enhanced by performing complete abrasion of the posterior meniscal tear borders with a shaver or meniscal rasp through the posteromedial portal. The posteromedial non-cartilaginous rim of the tibial plateau below the torn meniscus was abraded using a bur to reach the cancellous bleeding bone (Fig. 3). Once the tourniquet was released, these in-situ blood clots were confined to the undersurface of the meniscus to promote healing.

The posteromedial non-cartilaginous rim of the tibial plateau below the torn meniscus was abraded using a bur to reach the cancellous bleeding bone. MM medial meniscus, TP tibial plateau, MFC medial femoral condyle, PM.portal posteromedial portal

Under direct vision through the posterolateral trans-septal portal, a suture hook (Lasso ConMed-Linvatec, Utica, NY, USA) loaded with No. 1 PDS (polydioxanone suture; Ethicon, Somerville, NJ, USA) was passed through the peripheral and central segment of the medial meniscus. The posteromedial portal was used as a working portal, and a shuttle relay system was used to lift the peripheral fragment to the level of the central fragment to increase the contact surface for the repair process. The PDS was replaced with No. 2 fiber wire (Arthrex) and, finally, an SMC sliding knot was applied with a knot pusher. Suturing was repeated every 5 mm and terminated at the posterior corner point [15] (Fig. 1D–E).

After the posterior meniscal repair using an all-inside technique was completed, the classic outside-in technique was performed for the anterior third of the meniscus using anterior portals and No. 1 PDS (Fig. 1F). Sutures were placed anterior to the MCL to prevent MCL irritation, and abrasion was only done with a shaver for the mid-third of the meniscus.

The stability of the meniscus was re-evaluated with a probe at the end of the operation. Finally, the tibial tunnel was reamed, and a quadrupled semitendinosus autograft was used for a single-bundle anatomical ACL reconstruction. An endobutton or screw was used for fixation on the femoral side, and double fixation with a screw and ENDOTACK (KARL, STORZ) was used on the tibial side.

Toe-touch weight bearing was allowed for the first 2 weeks after the operation, while the knee was immobilized by an extension brace with an allowed passive range of motion (ROM) of up to 45°. During the third and fourth weeks, partial-weight-bearing ambulation and flexion with a ROM of 90° were allowed. Full-weight-bearing ambulation and full ROM were allowed only after 8–12 weeks after the operation.

SPSS software was used to analyze the data. Lysholm and IKDC scores (parametric variables) before and after the operation were compared using the paired t-test. The Tegner activity scale scores before the injury and after the operation were compared using the Wilcoxon test (non-parametric variable). The significance threshold was P < 0.001.