Ramp lesions are frequently missed in ACL-deficient knees and should be repaired in case of instability

Purpose The aim of the current study was (1) to provide an overview of common definitions and classification systems of ramp lesions (RL) and (2) to systematically review the available literature with regard to the diagnosis and treatment of RLs in anterior cruciate ligament (ACL)-deficient knees. Methods Following the PRISMA guidelines, MEDLINE and Scopus were searched for articles (1) reporting on acute or chronic ACL injuries, (2) with concomitant medial meniscus injury, (3) located at the posterior meniscocapsular attachment site (and red–red zone). Ex vivo studies, reviews and technical notes were excluded. Results Twenty-seven studies were included based on the criteria mentioned above. RLs are common in ACL-deficient knees with a prevalence ranging from 9 to 24%. RLs should especially be suspected in younger patients, patients with an increased meniscal slope and in patients with prolonged time from injury to surgery. The sensitivity of MRI for the detection of RLs ranges from 48 to 86% at a specificity of 79–99%. For arthroscopy, RLs are easily missed through standard anterior portals (sensitivity 0–38%). RL repair leads to a significant improvement of subjective knee scores, regardless of the specific fixation technique. For stable RLs, the literature suggests equivalent postoperative stability for trephination and abrasion compared to surgical RL repair. Conclusion Ramp lesions are frequently missed in ACL-deficient knees on standard arthroscopy with anterior portals only. If a RL is suspected, exploration via an additional posteromedial portal is indicated. In case of instability, RL repair should be performed. Level of evidence IV.


Introduction
The mechanisms of anterior cruciate ligament (ACL) rupture and concomitant knee injuries have been extensively investigated over the last decades, revealing that the minority of ACL ruptures occurs as an isolated injury [12,20]. Meniscal tears have been reported to be present in 55% up to nearly 80% of ACL injuries, with significantly higher rates in chronically ACL-deficient knees [12,26,46]. Accordingly, time to surgery correlates with the incidence of concomitant meniscus injuries [25,29]. In particular, the incidence of medial meniscus (MM) tears seems to significantly 1 3 increase with delayed surgery [26,50]. Conversely, patients with medial or lateral meniscus instabilities have a significantly increased risk of ACL failure after reconstruction [32]. Hence, early ACL reconstruction (ACLR) and meniscal repair are recommended to prevent secondary meniscus injuries and improve the long-term outcome [12,27].
Over the last years, more attention has been paid to tears located at the posteromedial meniscocapsular junction. Even though first characterizations date back at least 35 years [13], these injuries frequently remained unnoticed in knees with ACL injuries. Since their first description, a range of terms has been used synonymously to describe these lesions, including meniscocapsular separation (MCS), meniscosynovial tear, hidden lesion (HL) and ramp lesion (RL). The term ramp lesion was first used by Strobel in 1988, defining it as "a special type of meniscal injury involving the peripheral attachment of the posterior horn of the medial meniscus, typically associated with an ACL deficiency". Currently, there is no consensus regarding the definition of RLs [4] which has resulted in misleading descriptions. Some authors have also considered posteromedial meniscus tears in the red-red zone as ramp lesions [23,43].
Thaunat et al. classified five lesion subtypes by their exact tear pattern, location, degree of mobility and visibility during arthroscopy [43]. Type 1 is meniscocapsular junction tears located in the synovial sheath with very low mobility at probing. Type 2 includes partial superior meniscus tears which are stable and can only be diagnosed via the trans-notch approach. Type 3 is partial inferior meniscus tears (hidden lesions) associated with meniscotibial ligament disruptions resulting in high probing mobility. Type 4 involves complete longitudinal vertical meniscus tears in the red-red zone. Type 5 describes double longitudinal vertical tears (Fig. 1).

Fig. 1
Illustration of five ramp lesion subtypes depending on tear pattern, location, degree of mobility and visibility during arthroscopy according to Thaunat et al. [43] 1 3 An alternative classification has been proposed by Seil et al., paying more attention to the mediolateral extent of lesions and the properties of the capsule-ligament complex depending on the degree of flexion [34]. Furthermore, there seems to be confusion concerning the term "peripheral" in the context of RLs, since it has been referred both to the meniscocapsular and meniscotibial attachment sites of the PHMM [5,9], as well as to lesions in the RR zone [23,44]. Table 1 provides an overview of different RL definitions.
This review was conducted due to the growing number of publications regarding posteromedial meniscocapsular attachment tears, as well as the inconsistent nomenclature. The aim of this study was (1) to provide an overview of common ramp lesion definitions and classification systems and (2) to systematically review the available literature with regard to the diagnosis and treatment of ramp lesions (Fig. 2).

Materials and methods
A systematic database search of the literature concerning ramp lesions in ACL-deficient knees utilising pre-defined search terms was performed (see addendum). The study was conducted according to the preferred reporting items for systematic reviews and meta-analysis (PRISMA) guidelines [28]. The methodological index for non-randomised studies (MINORS) was used to assess the quality of all included non-randomised studies. Included databases were Medline and Scopus. English and German articles were considered because the authors speak both languages. Table 1 Definitions of ramp lesions as described in the literature Some authors consider also lesions in the red-red zone as ramp lesion

Authors
Year RL definition Definitions not including RR zone tears Di Vico et al. [9] 2017 Ramp lesions and hidden lesions represent subgroups of longitudinal lesions involving the peripheral attachment of the PHMM Seil et al. [34] 2017 Traumatic disruptions between the PHMM and its mensicoligamental junction Chen et al. [5] 2017 Peripheral attachment lesion of the PHMM Keyhani et al. [16] 2016 Longitudinal tear or detachment of the peripheral rim around the PHMM Li et al. [22] 2015 Peripheral attachment lesion of the PHMM Strobel et al. [42] 1988 Meniscal injury involving the peripheral attachment of the PHMM and is typically associated with an ACL deficiency Definitions including RR zone tears Sonnery-Cottet et al. [40] 2018 Disruption or tear of the peripheral meniscocapsular attachments of the posterior horn of the medial meniscus (exclusion criteria suggests that RR zone is included) Kim et al. [18] 2018 Peripheral longitudinal tear within 4 mm of the meniscocapsular junction of PHMM Thaunat et al. [43,44] 2016 Meniscosynovial or meniscocapsular tears + RR zone (see Fig. 1) Furumatsu et al. [11] 2013 Peripheral longitudinal tears of the MM included partial-or full-thickness, simple longitudinal tears ≥ 1 cm in length located in the outer one-third of the posterior segment Liu et al. [23,24] 2011 Tear of the peripheral attachment of the posterior horn of the medial meniscus (synovialmeniscus junction or red-red zone) Resnick et al. [31] 2007 Double vertical longitudinal tears of the PHMM  The search terms (see addendum) were assembled to cover a broad spectrum of meniscus and associated knee injuries, in order not to miss any relevant literature. As represented in our search terms, we paid special attention to keywords such as "ramp", "hidden", "meniscocapsular", "meniscosynovial" and "posteromedial" in the context of meniscus lesions. Next, relevant articles were preselected based on the abstract by two independent reviewers (AB and WW). A third reviewer (TT) was consulted to obtain consensus in case of discrepancy. Subsequently, the full text was carefully reviewed to check whether included lesions were (1) associated with acute or chronic ACL injury, (2) concerning the medial meniscus, and (3) located at the posterior meniscocapsular attachment site (and the periphery of the RR zone). Ex vivo studies, reviews and technical notes were excluded. Data regarding prevalence, diagnosis, surgical technique and outcome of RL repair were systematically extracted from the included studies.

Results
A total of 387 studies matched our search terms. After exclusion of duplicates, 37 studies were initially included based on their abstract. 27 studies were finally included. The average MINORS score of all non-randomised studies was 8.7/16 for non-comparative studies and 16.9/24 for comparative studies, respectively (Tables 2, 3).

Epidemiology
The reviewed literature reports a prevalence of RLs in ACLdeficient knees ranging from 9 to 24% [3,35]. RLs account for 17%-55% of all MM injuries [14,26]. There is evidence that RLs occur more frequently in younger and paediatric patients [23,25]. An increased medial meniscal slope [38] and prolonged time from injury to surgery [16,23,40] have also been associated with a higher incidence of RLs. Sonnery-Cottet et al. [40] reported that male sex, maximum age ≤ 30 years, revision ACL reconstruction (ACLR), side-toside laxity difference > 6 mm, and the presence of a lateral meniscal tear are all significant risk factors for RLs. Furthermore, RLs seem to be more prevalent in contact sports injuries [35] (Table 4).

Arthroscopy
Most authors have pointed out that RLs are easily missed through a standard anterolateral arthroscopic portal [1,3,23,39]. Thus, exploration via a trans-notch view or an additional posteromedial portal has been commonly suggested to specifically check for these lesions. The posteromedial portal has been reported to be safe concerning possible damage of popliteal neurovascular structures [5,9,16,30].
Kim et al. evaluated the accuracy of a sequential arthroscopic 4-step approach for the diagnosis of RLs [18]. The authors showed that 38% of all RLs were found using the initial standard exploration via an anterolateral portal. Fortyeight percent of RLs were identified through an intercondylar view using a 30° arthroscope. A trans-notch view using a 70° arthroscope and exploration through a posteromedial portal both resulted in a 100% detection rate. Additionally, the authors did not find a significant correlation between prolonged time from injury (TFI) and the overall prevalence of RLs; however, an association between TFI and the diagnostic step of detecting a RL was observed. RLs in chronic ACL tears (> 3 months) were more often detected through a standard anterolateral portal as compared to those in acute ACL injuries (< 3 months).
The difficulty of diagnosing a RL via an anterolateral portal has been confirmed by others [9,25,39] (Table 5).
Lesions affecting the meniscotibial ligament can be covered by an intact capsule and, therefore, remain undetected unless soft-tissue debridement is performed [9]. In case of meniscus instability without a corresponding lesion on standard arthroscopy, debridement via a posteromedial portal under trans-notch visualisation is recommended [9]. Zhang et al. [49] reported an association between a wavelike chondral injury of the medial femoral condyle and the presence of RL on arthroscopic exploration. The authors examined 1596 patients of which 4.9% (78/1596) presented with the so-called wave sign. A RL was confirmed in all of these patients.

Magnetic resonance imaging
Numbers regarding the sensitivity of MRI for the detection of RLs are very heterogeneous, ranging from 48 to 86% [6,41]. The specificity has been reported to range from 79 to 99% [2,48]. The criteria for the diagnosis of RLs on MRI in each individual study are displayed in Table 6. The diagnostic accuracy of hidden lesions on MRI remains unclear, since hidden lesions have not explicitly been considered in the aforementioned studies. Arthroscopy represented the gold standard for the diagnosis of RLs in all but one study; Kim et al. applied MRI as reference standard to demonstrate a correlation between the "uncovered medial meniscus sign" and an anterior tibial translation which could indicate instability of the PHMM in ACL-deficient knees [19]. The remainder of the studies applied similar methodologies, measuring the diagnostic accuracy of MRI in comparison to arthroscopy in an ACL-deficient patient collective (Table 6).    Outcome Eight studies were identified which report on the outcome of RL repair in a total of 855 ACL-deficient knees. The average age of patients at the time of surgery was 28.4 years and 74.9% (623/832) were males. Among the eight studies, one randomised controlled trial (RCT) [24], and two retrospective case-control studies were identified [40,47] (Table 7).
A range of different parameters regarding the outcome of RL repair in knees with concomitant ACL injuries has been reported in the literature. Among these are the Lysholm Knee Score (LNS), International Knee Documentary Committee (IKDC), Barrett criteria (presence of joint tenderness, RL repair leads to a significant improvement of subjective knee scores, regardless of the specific technique (Table 7).
In a recent study by Sonnery-Cottet et al. [40] RL repair was performed in 769 patients suffering from acute or chronic ACL injury, using the Arthrex Suture Lasso. The patients were divided into the following two groups: ACLR only + RL repair; and ACLR + anterolateral ligament repair (ALLR) + RL repair. Failure of RL repair was defined as the performance of secondary meniscectomy within the followup period of an average 45.6 months. The authors reported that patients who underwent ACLR + ALLR had a > twofold lower risk for reoperation due to failure of RL repair as compared to patients who underwent ACLR + RL repair. This study demonstrates that ALLR can have a protective effect on RL repair; yet, the effect of RL repair itself was not analysed.
Liu et al. conducted a randomised controlled trial, including 73 patients with ACL injury and a concomitant stable RL [24]. Criteria for stable RLs included a lack of excessive anterior translation of the PHMM on probing from the anteromedial portal and a maximum lesion length of 1.5 cm measured from the posteromedial portal. Patients underwent all-inside surgical repair (Suture hook, Linvatec) or abrasion and trephination only. Patients were followed up for at least 24 months evaluating both subjective and objective parameters. In terms of functional scores (LNS and IKDC), there was no significant difference between the two groups. Additionally, no difference in knee stability (pivot-shift test, Lachmann test, KT-1000 arthrometer) was observed. Finally, healing status was assessed on T2-weighted sagittal MRI scans, showing that there was no statistically significant difference between the study group and control group. In conclusion, this study indicates that all-inside repair of stable RL is non-superior to abrasion and trephination alone.
In a retrospective controlled analysis, Yang et al. compared arthroscopic abrasion and trephination and Fast-Fix repair of RLs measuring 1-2 cm with concomitant ACL injury in 68 patients [47]. At a minimum of 24 months after surgery, there was no significant difference between the groups, indicating that arthroscopic refreshment of stable RLs achieves similar results compared to Fast-Fix repair ( Table 7).

Discussion
The most important finding of this study was that ramp lesions are frequently missed in ACL-deficient knees and should be repaired in case of instability. Considering the high prevalence of RLs in ACL-deficient knees, exploration of the posteromedial meniscocapsular complex through a posteromedial (PM) portal seems indicated if a lesion is suspected [1,15,24,40]. Recently, Kim et al. demonstrated that a trans-notch view using a 70° arthroscope can achieve similar results in terms of RL detection rates [18]. However, an additional posteromedial portal should be considered the gold standard for the diagnosis of ramp lesions. Furthermore, a PM portal offers the benefit of a full exploration of the lesion extent and allows for a dynamic assessment of instability in flexion. If meniscus instability is observed at probing and no obvious lesion can be identified, soft-tissue debridement via a posteromedial portal under trans-notch visualisation could aid the identification of hidden lesions [9].
The current literature depicts limited and quite heterogeneous data on the diagnostic accuracy of MRI to detect RLs. Arner et al. demonstrated that MRI can be useful to rule out RLs [2]. MRI has a lower sensitivity and accuracy for RLs than for meniscus body tears [14]. On MRI, fluid filling between the posterior horn of the medial meniscus and the capsular margin can be indicative of a RL [14,48].
Based on the available case series, RL repair leads to a significant improvement of subjective knee scores,  [24] and 37.6 [5], most authors report significantly higher scores postoperatively as compared to preoperatively. These findings are in accordance with the results of a few case series reporting excellent healing results from 87 to 95% as evaluated during second-look arthroscopy [5,24]. Biomechanical studies have also demonstrated that RL repair can restore knee stability when simultaneous ACLR is performed [7,41]. A recent study by Sonnery-Cottet et al. demonstrated a significantly decreased risk of RL repair failure in patients undergoing concomitant ACLR and anterolateral ligament reconstruction, as compared to concomitant ACLR only [40]. For stable RLs, repair tends to be non-superior to trephination and abrasion only [24,47]. This has also been shown by the only available randomised controlled study by Liu et al. [24]. There is a considerable inconsistency regarding the nomenclature of ramp lesions in the literature [34,43]. The term ramp lesion has frequently been used as an umbrella term indicating posteromedial meniscus tears, rather than a specific type of tear (Table 1). Both tears of the meniscocapsular and meniscotibial attachment sites of the PHMM, as well as tears in the RR zone, have been termed ramp lesions [5,9,23,43]. To allow a comparison of studies, the authors of this work suggest to only consider meniscocapsular and meniscotibial lesions as ramp lesion. This is supported by a recent experimental study of DePhilippo et al. reporting that the meniscocapsular and meniscotibial attachments merge at a common attachment at the PHMM [8].
The following limitations have to be acknowledged: (1) generalised recommendations for the treatment of ramp lesions cannot be made based on the available literature. Ramp lesion repair leads to significant improvements in the clinical outcome; however, only two studies compared different surgical treatment options (trephination and abrasion vs RL repair). (2) Different ACLR techniques might lead to a different outcome. This review did not account for different fixation techniques. (3) Some authors considered lesions of the meniscocapsular attachment site and lesions at the periphery of the red-red zone as ramp lesions. Therefore, in a limited number of cases (Table 1), the reported postoperative improvements of RL repair may not exclusively be related to meniscocapsular lesions. (4) None of the MRI studies considered hidden lesions explicitly; therefore, it remains unclear whether these lesions can preoperatively be detected on MRI.
This work reports on the high prevalence of peripheral posteromedial meniscus lesions in ACL-deficient knees and provides an overview of common ramp lesion classification systems. A better understanding of the preoperative and intraoperative diagnostic accuracy of ramp lesions is of high clinical relevance. In daily practice, a thorough investigation of the posteromedial meniscus should always be performed during ACL reconstruction. Repair is indicated in case of instability.

Conclusion
Ramp lesions are common in ACL-deficient knees and are often missed through standard anterior portals. Exploration of the meniscocapsular complex via an additional posteromedial portal is recommended, if MRI suggests a lesion in this area or if instability is present at probing. If the posteromedial view still does not reveal any lesion despite significant instability, soft-tissue debridement might uncover hidden lesions (meniscotibial ligament disruptions).
Generalised recommendations for the treatment of ramp lesions cannot be made based on the available literature. Ramp lesion repair leads to a significant improvement of subjective knee scores, regardless of the specific fixation technique. For stable ramp lesions, trephination and abrasion might be equivalent to ramp lesion repair in terms of postoperative stability. In case of instability at probing, repair should be performed.