Introduction

Avulsion injury of the distal biceps femoris can occur as an isolated injury or as part of a multiligament injury pattern, with or without concomitant injury to additional lateral knee structures. ‖Through qualitative, quantitative, and biomechanical study, the anatomy and strength of the fibular collateral ligament (FCL), popliteofibular ligament, and popliteus tendon have been described thoroughly.^2,9,14,20 While the anatomy of the biceps femoris has also been qualitatively described,^{7,10,12,19,20,22} a quantitative analysis of its tendinous insertions on the fibula and proximal tibia is lacking.Previous qualitative description of the biceps femoris has elicited fascial and tendinous components of both the long head and the short head.^{7,10,12,18-22} The descriptions by Terry and LaPrade¹⁹ of the tendinous components of the long head included a direct arm, which inserts on the posterolateral aspect of the proximal fibula lateral to the styloid, and an anterior arm, in which insertion starts along the lateral edge of the proximal fibula lateral to FCL and continues distally to end in an anterior aponeurosis overlying the anterior compartment of the leg. Description of the tendinous components of the short head included a direct arm, which inserts just lateral to the fibular styloid and medial to the FCL, and an anterior arm, whose insertion begins medial to the FCL and continues anteriorly blending with the anterior tibiofibular ligament, with additional insertions continuing to the lateral tibia 1 cm posterior to the Gerdy tubercle. ¹⁹ Another qualitative description by Tubbs et al²² described a split of the biceps femoris insertion by the FCL into a medial slip and a lateral slip, with both components contiguous at their attachment site into the fibular head. The fibers anterior and posterior to the FCL convalesced to form a ring around the FCL, which tightened on traction of the tendon.The purpose of this study was to provide quantitative information regarding the tendinous insertions of the biceps femoris upon the proximal fibula and tibia, as repair of these structures often constitutes the goal of surgical treatment of avulsion injury. It was hypothesized that for the biceps femoris, there would be consistent relationships with pertinent osseous landmarks and the distal insertions of the FCL and anterolateral ligament (ALL). A comprehensive understanding of the anatomical attachment sites of the biceps femoris, FCL, and ALL will assist in guiding biceps femoris surgical repair and reconstruction techniques.

Methods

Specimen Preparation Twelve nonpaired, fresh-frozen cadaveric specimens (7 female, 5 male) with a mean age of 61.3 years (range, 53-66 years), mean height of 1.69 m (range, 1.55-1.78 m), mean weight of 93.58 kg (range, 45.36-127.0 kg), and mean body mass index (BMI) 33.2 kg/m² (range, 16-47 kg/m²) without history of injury were used in the study. We selected 12 specimens per group, as 10 and 12 specimens has proven a sufficient sample size in previous quantitative studies on ligamentous anatomy.^3,9 The knees were disarticulated, with obstructing soft tissue, ligaments, and tendons removed, including the femur and proximal musculature. Clear identification of the biceps femoris, FCL, and ALL, and their insertions upon the proximal tibia and fibula was performed. The tibia and fibula were fixed in cylindrical PVC pipe using automotive body filler (Bondo; 3M). The specimen was then clamped to a custom support structure, ensuring a constant and fixed spatial relationship to the measuring device throughout data capture. The ligamentous attachment centers of each structure were identified on the basis of their distinct insertions and were marked with a surgical marking pen (Figure 1).

Anatomic Measurements

Three-dimensional positional data were collected using a calibrated stylus utilizing standard system software (MicroScribe Utility Software; Revware Inc) that measured interspatial relationships between reference points (Figure 2). All data were captured by the senior author (A.W.A.). Data postprocessing was accomplished utilizing numerical computation software (MATLAB R2014b). For accurate measurement of the ligament and tendon footprints, immediately prior to data capture, the individual structures were tagged with sutures, removed from their proximal origin, and then sharply dissected off of their distal footprint to clearly identify their attachments. The periphery of each footprint was measured with points collected at 1-mm increments. Data points of pertinent osseous landmarks were also collected. From these points, the ligament attachment centers, distances to osseous prominences, and ligament footprint areas were calculated. Principal component analysis was used to reduce the dimensionality, allowing for plane fitting and 2-dimensional calculations of area. A point-inclusion algorithm was used to remove any points that were internal to the points defined by the outer polygon. Using the Heron formula, areas were calculated on the reduced 2-dimensional data set in an iterative process around the vertices set. Centroids were subsequently translated back to 3 dimensions and coordinated for linear comparison between landmarks. Data reduction and area calculation techniques were similar to those used by Brinkman et al² in a previous quantitative study on the posterolateral corner. All reported measurements were performed by the same individual, with guidance by the senior author (A.W.A.) to eliminate interobserver variability.

System Validation

To assess the accuracy of the measuring device, a single-point articulation performance test (SPAT) was performed based on the B89.4 22 American Society of Mechanical Engineers (ASME) standard and the manufacturer’s recommendations. The average SPAT result was 0.01 mm. The SPAT reflects both the repeatability and reproducibility of measurements using this device in our laboratory testing environment. Therefore, values <0.01 mm may be attributed to error within the measurement system, and values >0.01 mm may be considered significant.

Results

The tendinous portions of the long head and short head of the biceps were readily identifiable proximal to the insertion. Proximal to the fibula, the tendons of the long and short head coalesced, with some individual fibers continuing discretely from each muscle belly. The coalesced tendon began its insertion upon the fibula proximally and was split by the FCL as the insertion continued distally. Four easily identifiable and reproducible anatomic tendinous footprints were encountered: a proximal footprint of the biceps femoris upon the fibula (PBF) composed of fibers from both the short head and the long head, a distal lateral insertion of the biceps femoris upon the fibula (DBF) composed predominantly of fibers from the long head, a medial footprint of the biceps femoris upon the fibula (MBF) composed predominantly of fibers from the short head, and an insertion of the biceps femoris upon the tibia with a variable composition of fibers composed either predominantly from the long head or from the short head or a combination of the two. Footprint areas are presented in Figure 3, and distance from the center of the footprint areas to the surgical anatomic landmarks are presented in Table 1. The FCL and ALL were identifiable and distinct in all 12 specimens.