Large medial meniscus extrusion and varus are poor prognostic factors of arthroscopic partial meniscectomy for degenerative medial meniscus lesions

The DMMLs had a high incidence in the middle old-age. About 25% were aged 50–59, 35% aged 60–69, and 45% aged 70–79. The 2016 European Society for Sports Traumatology, Knee Surgery and Arthroscopy (ESSKA) Meniscus Consensus considered that nonsurgical treatment was the primary choice for DMMLs [1]. However, not all DMMLs respond to conservative treatment. For patients with normal or near-normal X-rays but abnormal MRI, surgical treatment was recommended after the failure of nonsurgical treatment for 3 months, or those who are undergoing persistent pain, and/or presenting mechanical symptoms (e.g., compression, locking and clicking) [2].

In past decades, a partial meniscectomy was considered a “traditional” treatment and was used widely, with more than 500,000 procedures performed annually in the USA. Between 1997 and 2017, 1.1 million APM were carried out in the United Kingdom [3, 4]. Although a partial meniscectomy can provide symptomatic relief in short-term follow-up, it can lead to progression of degenerative arthritis in mid- to long-term follow-up [5, 6]. A large number of relevant studies have shown that APM does not provide any additional benefit for DMMLs compared with non-operative or sham surgical treatment, and does not solve the underlying biomechanical abnormalities, which may increase the risk of osteoarthritis [7‐10]. The determination of preoperative prognostic factors was crucial to improve the outcome of surgery and delay the progress of osteoarthritis. Investigators have noted that before the surgical procedure for DMMLs, the role of varus alignment in clinical outcomes and the entire progress of osteoarthritis should be considered [11, 12]. Seil et al. [2] believed that the surgical decision-making process must include the negative imaging factors such as knee osteoarthrosis (K–L ≥ grade 2), age, terminal chondropathies, lateral meniscectomy and lower limb malalignment, for patients with persistent pain and no mechanical symptoms. However, the preoperative factors that affect the prognosis of DMMLs with APM are rarely used in quantitative studies, and previous studies mainly used reviews to describe their influencing factors.

Thus, the purpose of this study was to identify predictors of unfavorable clinical and radiologic outcomes after APM for DMMLs. We hypothesized that malalignment, poor cartilage status, medial meniscus extrusion may result in unsatisfactory clinical and radiographic outcomes in follow-up examinations for DMMLs.

Preoperative demographic data, preoperative and final results are shown in Table 1. 6 patients who had iatrogenic chondral injury were excluded due to tight medial compartment. Two patients had infrapatellar fat pad injury due to obscuration of the vision. And the pain has been eliminated by rehabilitation and physical therapy 2 weeks after the operation. Three patients still had medial joint line pain. It has shown a significantly improved postoperative Lysholm score (84.5 ± 9.7) compared with the preoperative score (63.8 ± 9.3) (P < 0.001), but a progression of K–L grade (20/36/30/0/0 ~ 15/27/25/19/0) (P < 0.001) in the image. Thirty-eight of 86 patients (44.2%) showed K–L grade progression. Nonetheless, there was no K–L grade progression to grade 4, and no total knee arthroplasty conversion patients (Table 1).

In Spearman correlation analysis, advancing age, smaller HKA and larger MME were negatively correlated with the final Lysholm score (Table 2). In the U test, Outerbridge grade ≥ 3 and K–L grade progression were associated with unfavorable final Lysholm score (Table 3). In multiple logistic regression analysis, advancing age (OR 1.109; P = 0.05) and smaller HKA (OR 0.255; P < 0.001) were an important factor causing unfavorable Lysholm score. Larger MME (OR 10.327; P < 0.001) is an adverse factor leading to the progression of K–L grading (Table 4).

The cutoff point for the relative value of preoperative medial meniscal extrusion associated with relatively unfavorable Lysholm scores was 2.05 mm (Area = 0.8668, P value < 0.0001, Sensitivity = 62.16%, Specificity% = 93.88%) (Fig. 4).

The principal findings of this study were that advancing age and smaller HKA were the primary reasons of poor postoperative clinical outcomes. Larger preoperative MME was a risk factor of postoperative K–L grading progression. Therefore, patients should be overall evaluated at the preoperative period before APM. Meanwhile, a better treatment strategy should be generated to maximize the knee function performance of patients and delay the progression of osteoarthritis.

The purpose of APM was to excise the irreparable meniscus margin of tear and retain the peripheric stability hoop so as to relieve symptoms caused by the meniscus lesions, compression and locking. The treatment of DMMLs by APM has attracted more and more attention from researchers but is still controversial. A multicenter, randomized, controlled trial showed that APM was not significantly superior to sham surgery in patients with DMMLs without knee osteoarthritis [15]. Currently, a large number of literature have reported that APM was not supported to be applied in first-line treatment of patients with DMMLs [16‐18]. Sihvonen et al. [18] performed APM on 70 patients with DMMLs and found that Lysholm score showed statistically significant difference in improvement compared with the preoperative condition, but APM did not improve knee symptoms or function better than the arthroscopic surgery group and showed no statistically significant difference. Furthermore, APM was associated with a slightly increased risk of knee osteoarthritis within 5 years after surgery. Therefore, for patients with conservative treatment and second-line APM treatment, the factors of poor prognosis should be analyzed, and a more reasonable treatment plan should be developed to improve the prognosis of patients.

The results of this study showed that lower limb malalignment was the main influencing factor for poor prognosis in clinical outcomes of DMMLs patients undergoing APM. Willinger et al. [19] found that the lower limb alignment and the amount of medial meniscus resection had significant effects on tibiofemoral contact pressure through researching the biomechanics of eight fresh cadavers. MME with varus increased mean contact pressure and peak contact pressure in the medial compartment, while lateral displacement of the line prevented overloading in the medial compartment. Willinger et al. [20], in another fresh cadaveric study, showed that the common layer tearing of DMMLs and intact meniscus have similar contact area and peak contact pressure. When APM was applied to one or two layers of the horizontal tearing, the contact area was significantly reduced and peak contact pressure was significantly increased. Both APM and varus resulted in increasing peak contact pressure in the medial compartment. The degenerative meniscus still had a protective effect on subchondral bone, even in osteoarthritis [21]. Therefore, the partial meniscus resection should be avoided in clinical practice to maintain the original biomechanical behavior. If partial meniscus resection was necessary, the structure of the meniscus should be retained as much as possible. In addition, the lower limb alignment played a key role in the load distribution between the medial and lateral of knee compartments, and varus was a potential risk factor for medial osteoarthritis. Therefore, clinicians should avoid APM in patients with layer tearing of DMMLs and considered alternatives (e.g., high tibial osteotomy in patients with varus) to reduce excessive stress in the medial compartment and removed the risk factors caused by varus to relieve knee symptoms and delay the further development of osteoarthritis [11]. More and more literature supported that DMMLs may be an early signal of knee osteoarthritis rather than a clinical problem of meniscus intervention alone [12]. For example, one study showed that there was no significant association between meniscus lesions and frequent knee pain in the middle-aged and the older, the occurrence of associated osteoarthritis should be taken into account [22]. The varus was an independent risk factor for medial osteoarthritis [23]. For patients with symptomatic DMMLs combined with varus, the osteotomy for line correction could be conducted as a preventative early measure to completely remove the risk factors without waiting for the confirmation of radiated osteoarthritis [12]. In this regard, we also conducted a study on the relationship between the lower limb alignment and DMMLs. We confirmed that varus was a risk factor of DMMLs and also one of the influencing factors for the poor prognosis of patients with DMMLs after APM.

In addition, MME was a risk factor of K–L grade progression in DMMLs patients after APM. The cutoff point for the relative value of preoperative MME associated with relatively unfavorable Lysholm scores was 2.05 mm. Kim et al. [24] had similar results with this study, which showed that the degree of meniscus degeneration and the manner of meniscus lesions (including horizontal, flap fissure and compound tear) were related to the MME, which had attracted more and more attention in clinical meniscus dynamic function evaluation and meniscus lesions diagnosis [25]. Meniscus extrusion was considered to be an indirect sign of meniscal pathology. It may be accompanied by loss of meniscus function, followed by deformity and loss of articular cartilage in the medial compartment, and eventually developed into osteoarthritis. There was a closed relationship between lower limb alignment and MME. Willinger et al. [25], through researching the biomechanics of eight fresh cadaveric knee specimens, investigated the relationship between MME and joint contact stress when the lower limb alignment was poor. The varus resulted in a higher MME and a corresponding increase in mean contact pressure and peak contact pressure compared to neutral or valgus. Goto [26], through retrospective multiple logistic regression analysis, showed that the varus was significantly correlated with MME, and the greater varus, the greater MME, and the more severe the osteoarthritis. Meniscus extrusion suggested dysfunction of meniscus, which was one of the causes of spontaneous osteonecrosis and was related to the development of spontaneous osteonecrosis [27]. The increased stress of femoral condyle of MME and varus may lead to enlargement of spontaneous osteonecrosis and changes of the secondary osteoarthritis [28]. Preoperative MME was negatively correlated with the clinical outcomes of partial meniscectomy, which could be used as a predictive factor of osteoarthritis after partial meniscectomy. The larger MME increased the risk of knee osteoarthritis. If APM was performed, this could increase the progression of osteoarthritis [15, 29].

Achtnich et al. [30] showed that the increase in age, BMI and load was significantly and positively correlated with the increase in meniscus extrusion. Varus, MME, DMMLs, and knee osteoarthritis were closely related and mutually influenced [1].The patients with lower limb malalignment and MME should be comprehensively evaluated. Patients accompanied with varus could choose osteotomy to remove the risk factors of osteoarthritis caused by lower limb malalignment, to reduce the pressure of medial compartment, and to relieve the pain at medial compartment. Astur et al. [31] studied the relationship between MME width and high tibial osteotomy, suggested that open wedge high tibial osteotomy could reduce MME and improve clinical efficacy and reactivate function. In addition, patients with a postoperative MME less than 1.5 mm had better clinical outcomes and activity levels than those with a postoperative MME of 1.5 mm or greater. Jing et al. [32] showed that for patients with medial meniscus posterior root tear, after high tibial osteotomy combined with root repair, the degenerated medial femoral condyle cartilage regenerated during the second-look arthroscopic findings. Therefore, for patients with DMMLs, a more satisfactory prognosis can be achieved by identifying the adverse factors affecting the operation, a comprehensive preoperative evaluation, a strict indication control and a reasonable treatment method.

The study had several limitations. First, this was a non-randomized retrospective study, so there could be a selective bias. Second, due to the absence of second-look arthroscopic findings, there was a lack of more intuitive evaluation indicators. Finally, because DMMLs were a signal of early osteoarthritis, the follow-up time was short, and there was a lack of long-term follow-up datum of imaging and clinical results.

In conclusion, advancing age, varus, and increased MME were risk factors leading to poor postoperative prognosis of APM for DMMLs and to progression of osteoarthritis. Lower limb alignment and MME should be assessed in each case of medial meniscal lesions. The most appropriate treatment plan could be selected, and corresponding intervention could be given in the early stage of osteoarthritis to improve treatment outcome, so as to effectively delay the progression of osteoarthritis.