Anterolateral ligament injuries on magnetic resonance imaging and pivot-shift testing for rotational laxity

Abstract

Maintaining anterolateral rotational stability of the knee requires a complex set of structures, most notably the anterior cruciate ligament. However, lateral knee structures such as the anterolateral ligament (ALL) also play an important role. There has been controversy over the role the ALL plays in an ACL deficient knee to maintain rotational stability. In this study, we examined ACL deficient knees with and without intact ALLs, for rotatory laxity using a pivot shift examination. This was graded using International Knee Document Committee (IKDC) criteria. MRI was used to view the ALL and its status. We found no statistically significant difference in rotational stability of ACL deficient knees, with or without intact ALLs. We did however find a statistically significant association between injury to the ALL and other concomitant lateral structures such as the lateral collateral ligament, biceps femoris tendon, and iliotibial band. This supports that the ALL works in concert with the other lateral structures in the knee and the ACL, to provide rotational stability. This suggests that as an isolated structure, the ALL’s contribution to clinical rotational stability is not significant, even in the presence of an ACL tear.

1. Introduction

Anterolateral rotatory instability of the knee is complex and remains among the least understood and most controversial concepts in orthopaedic sports medicine. Extensive research has been dedicated to further defining the lateral structures of the knee and to understanding their role in stability. Still, significant disagreement persists regarding rotatory knee laxity, particularly in the setting of anterior cruciate ligament (ACL) injury.

Injury to the anterior cruciate ligament is one of the most common sports injuries evaluated by orthopaedic surgeons, with an estimated 200,000 cases each year in the US, resulting in about 100,000 annual reconstructions.¹ The integrity of the ACL can be assessed clinically as well as with magnetic resonance imaging (MRI). The most sensitive and specific clinical examinations for detection of ACL injury are the Lachman test²⁸ and the pivot shift test.^{2, 3, 4} While the Lachman test is very reliable for evaluating anterior tibial translation, and diagnosing an ACL injury, it has not been shown clinically to correlate with subjective functional outcomes.^{5, 6, 7, 8, 9, 10, 30, 33} The pivot shift examination, which tests rotatory laxity, has been shown to strongly correlate with patients’ subjective perceptions of short and long-term knee function.^{27, 5, 31} The importance of the pivot shift and correcting it when reconstructing an unstable knee is emphasized in the literature. Leitze et al. were able to show that patients with a persistent pivot “shift” (IKDC 2 or 3) had significantly worse subjective functional outcomes compared to those with a pivot “glide” (IKDC 1).²⁷ Kaplan et al. showed that patients with a persistent shift were much less likely to return to high level sports participation than those with a glide.³¹ A positive test has also been linked to the development of osteoarthrosis.^{32, 35}

First described in the 1960′s, the pivot shift itself as well as our understanding of its components have evolved greatly over the years. It is known that there are many contributing factors to the presence and severity of the pivot shift, including the ACL, lateral meniscus, anterolateral capsule, iliotibial band, other lateral soft tissues, posteromedial joint capsule, and morphology of the tibial plateau as well as the femoral condyles.^{23, 24} The interplay between these factors is not yet fully understood. While its association with rotatory laxity is unquestioned, the performance of the pivot shift and its grading have been shown to be highly subjective and noted to demonstrate significant inter-examiner variability.^{29, 33, 37} Several clinical and biomechanical studies have been performed in an effort to produce a more standardized method of grading and performing the test.^{34, 36}

Recent attention has focused on one of the less studied lateral knee structures, the anterolateral ligament (ALL). The first noted reference to the ALL is credited to the French surgeon Paul Segond, who in 1879 described a “pearly, fibrous band” associated with the tibial avulsion fracture that would later bear his name.^{11, 12} The anterolateral ligament has since been alluded to numerous times in the literature by different names, including the “lateral capsular ligament”,¹³ the “midthird lateral capsular ligament”,¹⁴ and the “capsulo-osseous layers of the iliotibial band”.¹⁵ The term anterolateral ligament has been used for many years in reference to the entire iliopatellar band and iliotibial tract complex.²⁰ Vieira et al. were the first to refer to the specific structure in question by the name “anterolateral ligament” in their anatomic study of the iliotibial tract.¹⁶ In 10 cadaver knees, they were able to identify a “capsular-osseous” deep layer of the ITB that seemed to act as an independent ligamentous structure. More recently, several studies have attempted to further define the anatomic and histologic characteristics of the ALL, and have speculated on its functional contribution to rotatory stability.^{17, 18, 19} Controversy exists regarding the true significance of the ligament. Some authors suggest that the ALL is an isolated, independently functioning structure crucial in preventing rotatory laxity of the knee, while other authors believe it to be simply an extension of the iliotibial band or reflection of the anterolateral capsule.

The purpose of this study is to accurately identify the ALL on magnetic resonance imaging, as well as to evaluate its integrity in the setting of traumatic anterior cruciate ligament rupture. By correlating the degree of injury, if any, to the ALL with pre-operative pivot shift examinations under anesthesia, the contribution of the ALL to rotatory stability in the ACL deficient knee will be assessed. In this setting, based on previous anatomic studies, it is our hypothesis that injury to the ALL will have a significant effect on the degree of clinical rotatory instability, manifested by the pivot shift examination.

2. Methods

This study was conducted in accordance with our Institutional Review Board guidelines and appropriate waivers obtained. A retrospective chart review was performed including 76 patients from two surgeons’ practices with full thickness anterior cruciate ligament tears over a 2-year period. Four patients were excluded due to the absence of an available operative report. Twenty-six were excluded because a pivot shift examination was not completely or uniformly documented in the operative report. A total of 50 knees from 50 patients were selected for inclusion in the study. The subjects included 12 females with a mean age of 32 years, and 38 males with a mean age of 27 years. The mean age of all patients was 28.4 years with an age range of 17–54 years. Operative reports detailing the pivot shift examination under anesthesia were reviewed. The pivot shift examination was graded by one of two fellowship trained sports medicine orthopaedic surgeons according to the International Knee Document Committee (IKDC) criteria (0 = absent, 1 = glide, 2 = clunk, 3 = gross). Grade 2 and 3 pivot shift were grouped together as a “shift.” Pre-operative MRIs were reviewed by a fellowship trained musculoskeletal radiologist with over 10 years of experience blinded to the clinical details and surgical findings. MRI evaluations were performed utilizing standard clinical 1.5-T MRI units with standardized 8-channel knee coils. The MRI protocol included axial, coronal and sagittal pulse sequences using a moderate echo time (TE), fast spin-echo (FSE) technique, with repetition time (TR) 3200 to 6800 milliseconds; TE, 28 to 38 milliseconds (effective); and matrix 512 × 384 (sagittal), 512 × 256 (coronal), and 512 × 256 (axial) at 2 excitations. Additional axial, coronal, sagittal frequency selective fat-suppression pulse sequences were performed with a TR of 4000 to 6000 milliseconds; TE, 40 to 50 milliseconds (effective); and matrix 256 × 224, at 2 excitations. All images were obtained with a 3.5- 4 mm slice thickness, without gap. Studies were viewed on high resolution 5 megapixel monitors via a picture archiving and communication system (PACS). Axial, sagittal, and coronal cuts were used to identify the presence and degree of injury to the ALL (Grade 0–3 as per our classification). The popliteus tendon, lateral collateral ligament, biceps femoris tendon, and iliotibial band were analyzed and injuries were graded 0–3 (0 = normal, 1 = sprain/strain, 2 = partial-thickness tear, 3 = complete full-thickness tear). Grade 1 injuries were defined as increased “fluid sensitive” signal intensity within the structure of interest, without discontinuity. Grade 2 injuries were defined as partial discontinuity of the structure, without full-thickness interruption. Grade 3 injuries were defined as complete full-thickness discontinuity of the structure, with or without corrugation of the torn ends. The presence or absence of a Segond fracture was also documented. The ALL was identified on MRI using similar criteria to previously published results. Statistical analysis using Stata 11.0 (StataCorp, College Station, TX) was then utilized for interpretation.

3. Results

The ALL was identified in 100 percent of the anterior cruciate ligament deficient knees evaluated. In 17 knees, there was no MRI evidence of ALL injury (Grade 0). A grade 1 injury was noted in 26 knees. A grade 2 injury was noted in 5 knees. A grade 3 injury was observed in 2 knees. No Segond fractures were noted. Fifty six percent (28/50) of knees showed a positive pivot shift (IKDC 2 or 3) on examination under anesthesia. A positive pivot glide was noted in 44 percent (22/50) of patients. No patients had a negative pivot shift test. Of the patients with a positive pivot shift on examination, the mean time from injury to examination was 6.81 months. Of the patients with a pivot glide, the mean time from injury to examination was 5.45 months. This time difference was determined to be non-significant. MRIs were all obtained within two weeks of injury in all subjects.

Without controlling for injury to the other lateral knee structures, no significant association was found between injury to the anterolateral ligament and pivot shift test results, regardless of injury severity (P = 0.34) (Table 1).

Table 1.

Association between injury to the anterolateral ligament and pivot shift test results.

ALL Injury	Pivot Shift
	Glide (N = 22)	Shift (N = 28)
No injury.	5 (29%)	12 (71%)
Grade 1 injury	15 (58%)	11 (42%)
Grade 2 injury	0	5 (100%)
Grade 3 injury	2 (100%)	0

P = 0.34 from Cochran-Armitage Trend Test.

Injuries to the popliteus, lateral collateral ligament, biceps femoris, and iliotibial band were then reviewed. Statistically significant associations were found between injury to the ALL and the LCL (P < 0.01), biceps femoris (P = 0.01), and iliotibial band (P < 0.01). There was not a statistically significant association between injury to the ALL and popliteus (P = 0.11). (Table 2)

Table 2.

Association between injury to the ALL and injury to other posterolateral corner structures.

	ALL Status
	No Injury	Injury	P Value
Popliteus	3 (18%)	14 (82%)	0.11
LCL	0	13 (100%)	<0.01
Biceps femoris	0	10 (100%)	0.01
IT band	1 (7%)	14 (93%)	<0.01

Results obtained using Fisher’s Exact Test.

Finally, when controlling for all other injuries, there was not a significant association between injury to the anterolateral ligament and pivot shift test result. (Table 3, Table 4).

Table 3.

(No Injury to structures other than ACL/ALL).

ALL	Pivot Shift
	Glide (N = 8)	Shift (N = 17)
No injury	3 (23%)	10 (77%)
Injury	5 (42%)	7 (58%)

P = 0.41 using Fisher’s Exact Test.

Table 4.

(Injury to structure other than ACL/ALL).

ALL	Pivot Shift
	Glide (N = 14)	Shift (N = 11)
No injury	2 (50%)	2 (50%)
Injury	12 (57%)	9 (43%)

P > 0.99 using Fisher’s Exact Test.

4. Discussion

The ALL originates in the femur, with lack of consensus on its exact origin on the femur. It then travels distally in a posterior to anterior direction crossing the femorotibial joint line, and inserts into the anterolateral tibia just distal to the articular surface and posterior to Gerdy’s tubercle.⁴³ On MR imaging, the ALL can be identified in the axial and coronal planes as a structure closely associated but separate from the fibular collateral ligament at the femoral attachment and separate from the iliotibial band (Fig. 1). Because of its oblique course, the entire ALL cannot be captured in a single image, but instead requires scrolling through several consecutive MR cuts to completely evaluate its origin and insertion. The oblique course of the ALL and its close proximity to the LCL, IT band and popliteus tendon, can result in volume averaging artifact and make it difficult to identify on sagittal MR sequences.

Fig. 1.

Axial proton density MR image just below the level of the lateral femoral epicondyle demonstrates the location of the ALL (straight arrow) relative to its neighboring structures, including the popliteus tendon (curved arrow). Both the ALL and fibular collateral ligament (lateral collateral ligament complex) are closely associated at the femoral attachment, and together are part of the deepest layer (III), separate from the iliotibial band (layer I) and lateral patellar retinaculum (layer II).

Numerous cadaveric studies have attempted to define the precise anatomy of the anterolateral knee. Controversy still exists as to the presence and location of the ALL. Terry et al.²⁰ divided the iliotibial tract into 5 distinct layers: the aponeurotic layer, the superficial layer, the middle layer, the deep layer, and the capsulo-osseous layer. Their assertion was that the deep, capsule-osseous, and superficial layers work in concert to function as an “anterolateral ligament.” Sanchez et al.²¹ reported that the capsule-osseous layer functions like an anterolateral ligament, forming a sling over the lateral femoral condyle, but did not go so far as a ligament on its own. Similarly, Vieira¹⁶ identified the capsule-osseous layer as the anterolateral ligament, but continued to acknowledge it as a distinct portion of the iliotibial band. Only more recent studies have distinguished the ALL as a clearly defined, independent structure. Vincent,¹⁷ Claes,¹⁸ Helito,¹⁹ and Caterine³⁹ have all produced anatomic descriptions of the ligament over the last several years, although the specific anatomic characteristics they found were often variable.

Vincent et al. identified the ligament in 30 patients undergoing total knee arthroplasty as well as 10 cadaver specimens.¹⁷ They consistently found the origin to be just anterior to the popliteus, with some of its proximal fibers blending with those of the popliteus as it extends inferiorly. The insertion was found to be both into the lateral meniscus as well as the tibial plateau, on average 5 mm distal to the articular surface and posterior to Gerdy’s tubercle. After histologic analysis of the cadaver specimens, they concluded that the anterolateral ligament is clearly a structure that is distinct from the lateral capsule as well as the iliotibial band.

Claes et al. examined 41 cadaver knees and identified the ALL in 40.¹⁸ In contrast to Vincent, they noted the origin to be more closely associated with the lateral collateral ligament (LCL) than to the popliteus, with numerous connecting fibers between the LCL and ALL. The two insertion sites were consistent with Vincent, with the first at the middle 1/3 of the body of the lateral meniscus and the second noted to be proximal and posterior to Gerdy’s tubercle. Again, a clear distinction was found to exist between the ALL and both the anterolateral capsule and iliotibial band. A “lateral collateral ligament” complex was also proposed, consisting of the ALL and LCL and being functionally akin to the medial collateral ligament complex.^{21, 22} Helito et al. dissected the anterolateral knee in 6 cadaver specimens. Their findings included highly variable origin sites.¹⁹ Two cadaver ligaments originated proximal to the LCL, 3 originated distal, and one at the same level. The insertion site was noted to be consistent with the previous studies. The size of the ligament and histologic findings were consistent with previous studies as well^{17, 18}

Caterine et al. examined 19 knees, and found 3 separate variations of origin and insertion of the ligament. Their histologic findings also confirmed its consistencies with a ligamentous structure.³⁹ In addition, they noted a network of peripheral nerves throughout the ALL on immunohistochemistry, establishing the ALL as a possible proprioceptive role player in normal knee kinematics. (redundant)

We were able to identify the ALL on all patients using MR imaging. We found a statistically significant association between injury to the ALL and LCL, which supports the anatomic description by Claes et al.¹⁸ The real controversy regards the independent functional role it plays in effecting the pivot shift phenomenon, and as a result the rotatory stability of the knee. The relationship between the anatomy and function of the lateral knee structures was described extensively by Hughston et al. They observed 6 types of lateral compartment instability, including anterolateral rotatory instability. It was asserted that this type of instability is “caused by a tear of the middle one third of the lateral capsular ligament but it may be accentuated by other tears, principally a tear of the anterior cruciate.²⁶” Norwood et al. noted “injury to the lateral capsular ligament, the anterior cruciate ligament, or both” to be present in all knees with acute anterolateral rotatory instability. Dodds et al.³⁸ noted lengthening of the ligament in cadavers when imposing internal tibial rotation, although this finding was inconsistent with previous and subsequent studies,^{17, 18, 39} and it has been suggested that a different structure was in fact being measured.³⁹ Despite a lack of data, several authors have proposed that what we now know as the anterolateral ligament plays a key role in rotatory stability of the knee. Etienne et al. found that “There was an in vivo relation between a high-grade pivot shift and ALL injury in ACL-deficient knees.”, however due to the small number of patients in their study they did not find any statistically significant results when comparing all 4 pivot shift grades using MRI⁴⁰

The results of this study suggest otherwise. We found no significant correlation between injury to the anterolateral ligament and grading of the pivot shift exam, our best clinical detector of rotatory laxity. There was, however, a link between injury to the ALL and the other lateral knee structures. This suggests that the ALL does function as part of the “lateral ligament complex,” which as a unit contributes greatly to rotational stability. Hence, the magnitude of injury to the lateral ligament complex and other lateral structures plays a role to rotational stability, where an isolated ALL injury was insignificant, an injury to the ALL and the LCL was found to be significant (P < 0.01), biceps femoris (P = 0.01), and iliotibial band (P < 0.01). (Table 2) This is supported by Fu and Herbst, who discussed pivot-shift phenomenon as multifactorial involving “Bone morphology, inherent ligamentous hyperlaxity, and injury to other soft tissue structures such as lateral meniscus or posteromedial and posterolateral corner, also play a role in rotatory instability”.⁴¹ This supports our conclusion that the ALL must work in concert with the ACL and other lateral knee structures to maintain rotational instability, and it does not play a major role on its own. The recent commentary by Rossi also supports this finding, stating the ALL does not provide significant rotational stability in the setting of an ACL deficient knee.⁴²

How much bearing ALL disruption on an ACL deficient knee, will have on the management of patients with ACL injury is still uncertain. As stated, statistically significant results were found for clinical rotational instability by pivot shift examination, for ALL injuries with concomitant LCL, biceps femoris tendon, or iliotibial band injury. Pointing to the importance of the lateral knee structures as a whole, to provide rotational stability. During surgery patients who are undergoing ACL reconstruction that demonstrate excess instability on manipulation under anesthesia and have a disrupted ALL then repaired along with other damaged lateral knee structures.

Limitations of this study include the inherent weaknesses of a retrospective review, although every attempt was made to eliminate selection bias by initially including all ACL injured patients undergoing reconstruction within the selected time period. Recall bias was not a major concern as all operative reports were dictated at the time of surgery and the dictating physicians were not aware that they would be involved in the study. Also, any time a subjective clinical examination is used to provide data points, accuracy and precision is brought into question. Musahl, et al. demonstrated this point as it pertains to the pivot shift, showing a wide variation in clinical grading among 12 “expert” surgeons.²⁵ Further, other factors that can contribute to the pivot shift were not evaluated or controlled for, including bony morphology and injury to medial sided structures.

5. Conclusion

The anterolateral ligament is a structure that can be clearly identified on MRI, including pathology to the ligament. (Fig. 1, Fig. 2) Given the fact that there is statistically significant association between ALL injuries and injuries to other structures of the lateral knee, we suggest that the ALL works in concert with the other lateral structures in the knee as well as the ACL, to provide rotational stability. Our results suggest that as an isolated structure, its contribution to clinical rotational stability is not significant, in the presence of an ACL deficient patient.

Fig. 2.

A and B − Coronal PD FS images showing proximal tear of the ALL (arrows).

Funding

The authors have received no funding in the preparation of this manuscript.

Conflict of interest

The authors declare no conflict of interest.