MRI of the Distal Biceps Femoris Muscle: Normal Anatomy, Variants, and Association with Common Peroneal Entrapment Neuropathy

Transcription

1 Vieira et al. MRI of the iceps Femoris Muscle Musculoskeletal Imaging Original Research 09_07_2308_Vieira.fm 7/27/07 Renata La Rocca Vieira 1 Zehava Sadka Rosenberg Kiril Kiprovski Vieira RLR, Rosenberg ZS, Kiprovski K Keywords: anatomy, biceps femoris muscle, common peroneal nerve, MRI, neuropathy DOI: /JR Received July 27, 2006; accepted after revision pril 19, ll authors: Department of Radiology, New York University (NYU) School of Medicine and NYU Hospital for Joint Diseases, 301 E 17th St., New York, NY ddress correspondence to R. La Rocca Vieira. JR 2007; 189: X/07/ merican Roentgen Ray Society MRI of the Distal iceps Femoris Muscle: Normal natomy, Variants, and ssociation with Common Peroneal Entrapment Neuropathy OJECTIVE. The objectives of our study were to describe the previously unreported normal MR anatomy of the distal biceps femoris muscle and its relationship with the common peroneal nerve and to present a case in which previously unreported MR evidence of an anatomic variation in the distal biceps femoris muscle was associated with common peroneal entrapment neuropathy. MTERILS ND METHODS. One hundred consecutive 1.5-T knee MR studies of 97 asymptomatic patients were retrospectively reviewed by two observers in consensus for, first, normal anatomy of the distal biceps femoris muscle; second, anatomic variations of the muscle; and, third, the relationship of the muscle to the common peroneal nerve. Measurements of the distal and posterior extents of the short and long heads of the biceps femoris were performed. n MR study of a symptomatic patient with clinical evidence of common peroneal neuropathy associated with a surgically proven anatomic variation of the distal biceps femoris was reviewed. RESULTS. Two MR anatomic patterns were seen in the asymptomatic patient group: First, in 77 knees (77%), the common peroneal nerve was located within abundant fat posterolateral to the biceps femoris; and, second, in 23 cases (23%), the common peroneal nerve traversed within a narrow fatty tunnel between the biceps femoris and lateral head of the gastrocnemius muscles. There was a positive correlation between the distal and posterior extents of the short head of the biceps femoris muscle and the presence of the tunnel. CONCLUSION. Variations in the posterior and distal extents of the biceps femoris muscle can produce a tunnel in which the common peroneal nerve travels. We also described a case of common peroneal neuropathy secondary to tunnel formation. he most frequent mononeuropathy T in the lower extremity involves the common peroneal nerve around the fibular head and neck area [1]. The superficial location of the common peroneal nerve in the knee and the fact that it is fixed at the fibular neck make it susceptible to injury. There are many causes of common peroneal neuropathy including trauma; idiopathic palsy; postural habits; rapid weight loss; intrinsic or extrinsic nerve tumors; or extraneural compression by a synovial cyst, soft-tissue tumor, osseous mass, or large fabella. Traumatic injury of the nerve may occur secondary to a fibular head fracture, knee dislocation, ankle sprains with stretching of the nerve, surgical procedures, and application of skeletal traction or a tight cast [1 5]. Recently we encountered a surgically proven case of common peroneal entrapment neuropathy secondary to an anatomic variation of the biceps femoris and an atypical relationship between the biceps femoris muscle and the nerve. The anatomic variation contributed to the formation of a muscular tunnel between the lateral head of the gastrocnemius muscle and the distal biceps femoris. In this case, the common peroneal nerve traversed within this narrow fatty tunnel. To our knowledge, this anatomic relationship has not been previously reported as a cause of common peroneal neuropathy. Furthermore, there is no detailed anatomic description in the literature of the distal biceps femoris muscle and its exact relationship with the common peroneal nerve in the posterior aspect of the knee (Fig. 1). ll of these issues motivated us to study the normal MRI anatomy of the distal biceps femoris muscle and the relationship of the muscle with the common peroneal nerve. Materials and Methods Institutional review board approval for this retrospective study was obtained, with waiver of informed consent. JR:189, September

2 Vieira et al. Fig. 1 Drawing shows normal anatomy of common peroneal nerve (5). Note close relationship between distal biceps femoris insertion, lateral head of gastrocnemius muscle (7), and common peroneal nerve (circle). 1 = sciatic nerve, 2 = tibial nerve, 3 = long head of biceps femoris muscle, 4 = short head of biceps femoris muscle, and 6 = common tendon of biceps femoris. Reprinted with permission from HJD Graphics & Photo Department. Fig. 2 Measurement of posterior extent of short head of biceps femoris muscle performed on axial proton density (PD) weighted (TR/TE, 2,000/32) image of knee in 30-year-old asymptomatic man. Line is drawn posterior to femoral condyles (solid line). Second line (dotted line) measures posterior extent of short head of biceps femoris muscle (asterisk) relative to solid line. One hundred consecutive 1.5-T knee MR studies of 97 patients (41 males and 56 females; age range, years; mean age, 52.8 years) imaged for internal knee derangements were retrospectively reviewed by two musculoskeletal radiologists in consensus. The radiologists had 20 and 5 years experience in musculoskeletal radiology, respectively. Exclusion criteria included lesion of the posterolateral corner of the knee defined by signal abnormality or deformity of the joint capsule and ligaments of the posterolateral knee, and history or imaging findings compatible with common peroneal neuropathy. None of these criteria were met, so no examinations were excluded. The MR studies were obtained from July 2005 to January ll asymptomatic patients underwent MRI on a 1.5-T magnet. The MR protocol consisted of axial T1 (TR/TE, 466/9) or proton density (PD) weighted (2,000/32) and axial T2 fat-saturated (3,800/119), coronal PD (2,300/12) and T2 fat-saturated (2,400/17), and sagittal PD (2,900/21) and PD fat-saturated (2,750/56) images of the knee. ll MR sections were obtained with a slice thickness of 4 mm, a field of view of 16 cm, and a matrix of 256. The following parameters were recorded regarding the anatomic relationship of the common pero- Fig. 3 Comparison of two major patterns of downward course of common peroneal nerve in posterior knee on axial proton density (PD) weighted images (TR/TE, 2,000/32)., bsence of tunnel in 45-year-old asymptomatic woman. Common peroneal nerve (straight arrow) sits within large amount of fat. Short head of biceps femoris muscle (curved arrow) is quite small. sterisk = lateral head of gastrocnemius muscle., Presence of tunnel in 43-year-old asymptomatic woman. Common peroneal nerve (straight arrow) travels in narrow fatty passage (circle) between lateral head of gastrocnemius muscle (asterisk) and short head of biceps femoris muscle (curved arrow). Note that short head is much larger here in than in. 550 JR:189, September 2007

3 MRI of the iceps Femoris Muscle C Fig. 4 Presence of tunnel shown on sequential axial proton density (PD) weighted images (TR/TE, 2,000/32) of 44- year-old asymptomatic woman. D, Common peroneal nerve (straight arrow) is situated within narrow tunnel (circle) between muscle of lateral gastrocnemius muscle (asterisk) and short head of biceps femoris muscle (curved arrow). D femoris muscle. The variations that were noted included the posterior extent of the muscle belly of the short head of the biceps femoris muscle, the distal extent of the muscle belly (distal extent of the distal myotendinous junction) of the short head of the biceps femoris from the joint space, and the presence and distal extent of the muscle belly (distal extent of the distal myotendinous junction) of the long head of the biceps femoris. ll these parameters were performed in all 100 knees except the assessment of the posterior extent of the short head of the biceps femoris muscle, which was performed in only 74 knees because of technical difficulties, and the assessment of the distal extent of the muscle belly of the short head of the biceps femoris from the joint space, which was performed in only 98 knees again because of technical difficulties. We had technical difficulties in the retrieval of the examinations in our workstation to perform the muscle measurements in some cases. The posterior extent of the short head of the biceps femoris was measured in the following manner using an axial T1 or PD image (Fig. 2): line was drawn posterior to the femoral condyles, and a second line was drawn to measure the posterior extent of the short head of the biceps femoris muscle relative to the first line. The axial image of the knee that showed the deepest femoral notch (resembling a roman arch) was arbitrarily chosen to perform this measurement. We defined a short head of the biceps femoris as hypertrophied if its posterior extent measurement was greater than the mean value of the posterior extent in the asymptomatic population plus 2 SDs. We defined a tunnel as an instance in which the common peroneal nerve, rather than traveling within abundant fat lateral to the biceps femoris and posterior to the lateral head of the gastrocnemius muscle (Figs. 3 and 4), was noted in a narrow passage between the two muscles. The fat quantification was subjective. passage shorter than 1 cm in its superior-to-inferior extent was not included in the definition of a tunnel. The longitudinal extent of the tunnel was measured. Statistical nalysis Patients with and those without a tunnel were compared with respect to, first, the distal extent of the muscle s bellies of the long head (when present) and short head of the biceps femoris from the joint space; and, second, the posterior extent of the short head of the biceps femoris using the Mann-Whitney test. ll statistical computations were performed using SS software (SS version 9.0, SS Institute) for Windows (Microsoft). neal nerve to the distal biceps femoris muscle and the lateral head of the gastrocnemius muscle: the presence of denervation edema of the muscles of the anterior and lateral compartments of the upper leg and signal abnormality of the common peroneal nerve in the popliteal fossa; and variations in the distal biceps Case Presentation One MRI study, obtained from our teaching file, of a 14-year-old boy with clinical evidence of pero- JR:189, September

4 Vieira et al. TLE 1: Frequencies of Tunnel Length in symptomatic Population Tunnel Length (cm) No. (%) of Patients 2 11 (45.8) > 2 to < 3 6 (25) 3 to < 4 6 (25) 4 1 (4.2) Total 24 (100) TLE 2: Posterior Extent of Short Head of iceps Femoris in Posterior Extent (cm) Group Median Mean SD Range With tunnel Without tunnel ll subjects neal entrapment neuropathy and MRI evidence of aberration of the insertion of the distal biceps femoris muscle was retrospectively reviewed in consensus by the same radiologists. The MRI protocol consisted of axial T1 (466/9) and T2 fat-saturated (3,800/119), coronal PD (2,300/12) and T2 fat-saturated (2,400/17), and sagittal PD (2,900/ 21) and PD fat-saturated (2,750/56) images of the knee and upper leg. ll MRI sections were obtained with a slice thickness of 4 mm, a field of view of 16 cm, and a matrix of 256 in a 1.5-T magnet (vanto, Siemens Medical Solutions). The same analysis performed in the asymptomatic population was also performed in this patient. The patient s medical records and electromyography (EMG) studies were reviewed. Results natomic Relationship of the Common Peroneal Nerve to the iceps Femoris Muscle and the Gastrocnemius Muscle In 77 of the 100 knees (77%), the common peroneal nerve was situated within an abundant amount of fat superficial to the lateral head of the gastrocnemius muscle and posterior to the short head of the biceps femoris muscle (Fig. 3). In the remaining 23 cases (23%), the common peroneal nerve was situated within a narrow tunnel between the lateral head of the gastrocnemius muscle and the short head of the biceps femoris muscle (Fig. 3). The tunnel averaged 2.4 cm in length (range, cm). Note in Table 1 that almost 50% of the asymptomatic population had a tunnel length of 2 cm or less and only 4.2% had a tunnel length of 4 cm or greater. TLE 3: Frequencies of Posterior Extent of Short Head of iceps Femoris in Distance (cm) No. (%) of Patients 1 37 (50) (34) (15) 3 1 (1) Total 74 (100) TLE 4: Distal Extent of Muscle elly of Short Head of iceps Femoris from Joint Space in symptomatic Population Distal Extent (cm) Group Median Mean SD Range With tunnel to 0.7 Without tunnel to 1.5 ll subjects to 1.5 Variations in the Distal iceps Femoris Muscle In the entire series of 100 knees, there was no accessory tendon of the biceps femoris. Posterior extent of the short head of the biceps femoris muscle The posterior extent of the short head of the biceps femoris muscle at the level of the femoral condyles in all cases averaged 1.19 cm (range, cm) (Table 2). In cases without tunnel, the posterior extent of the short head of the biceps femoris muscle averaged 1.02 cm (range, cm). In cases with tunnel, the posterior extent of the short head of the biceps femoris averaged 1.50 cm (range, cm). The short head of the biceps femoris met the criteria for muscle hypertrophy in two cases that is, the posterior extent of the muscle was greater than 2.59 cm, which corresponds to the mean value of the posterior extent of the short head of the biceps femoris in the asymptomatic population plus 2 SDs. In 50% of the asymptomatic population, the posterior extent of the short head of the biceps femoris was 1 cm or less (Table 3). Only 16% of the cases had posterior extension of the muscle 2 cm or greater. Distal extent of the muscle belly of the short head of the biceps femoris from the joint space In all the asymptomatic knees, the distance between the short head of the biceps femoris muscle and the joint space averaged 0.05 cm below the level of the joint space TLE 5: Frequencies of Distal Extent of Muscle elly of Short Head of iceps Femoris from Joint Space in Distance (cm) No. (%) of Patients +2 2 (2) (19.5) 0 50 (51) 1 22 (22.4) 2 5 (5.1) Total 98 (100) TLE 6: Distal Extent of Muscle elly of Long Head of iceps Femoris from Joint Space in Distal Extent (cm) Group Median Mean SD Range With tunnel Without tunnel ll subjects (range, from 1.5 cm above the joint to 2.0 cm below the joint) (Table 4). In knees with a tunnel, the distance between the short head of the biceps femoris muscle and the joint space averaged 0.6 cm below the level of the joint space (range, from 0.7 cm above the joint to 2.0 cm below the joint space). In knees without a tunnel, the distance between the short head of the biceps femoris muscle and the joint space averaged 0.1 cm above the level of the joint space (range, from 1.5 cm above the joint to 1.0 cm below the joint). The myotendinous junctions were situated at the level of the joint space in 51% of the asymptomatic population, as shown in Table 5. The rest of the cases were almost equally divided into those in which the myotendinous extended above the joint (21.4%) and those in which the myotendinous extended below the joint (27.5%). Presence and distal extent of the muscle belly of the long head of the biceps femoris The muscle belly of the long head of the biceps femoris was identified in 40 of the 100 knees (40%). The distance between the long head myotendinous junction to the joint space averaged 5 cm (range, cm above the joint) (Table 6). In only one case (1%), the low-lying muscle belly of the long head of the biceps femoris contributed to the formation of a tunnel (Fig. 5). 552 JR:189, September 2007

5 MRI of the iceps Femoris Muscle There was no evidence of denervation edema of the muscles of the anterior and lateral compartments of the upper leg or signal abnormality of the common peroneal nerve in the popliteal fossa in any of the cases in the asymptomatic population. Statistical Results The presence of a tunnel was significantly predicted by the posterior extent of the short head of the biceps femoris (p =0.005) and the distal extent of the muscle belly of the short head of the biceps femoris from the joint space (p < 0.001). However, there was greater correlation between the posterior extent compared with the distal extent of the short head of the biceps femoris and the presence of a tunnel. Symptomatic Case The MR studies of a 14-year-old boy with clinical and EMG evidence of common peroneal neuropathy were studied. The patient presented with nonprogressive weakness of the right foot that had begun 3 months earlier. He had difficulty lifting the foot off the ground. One month after the onset of his Fig. 5 Distal extent of long head of biceps femoris muscle shown on sequential proton density (PD) weighted axial images (TR/TE, 2,000/30) of knee in 50-year-old asymptomatic man. and, Long head (asterisk) ends more distal than is typical in this patient. Straight arrow = common peroneal nerve, curved arrow = short head. symptoms, he twisted his right ankle. He was also tripping frequently, but he had not fallen. He denied previous trauma-related surgery or systemic disease. Physical examination revealed dorsiflexion, eversion weakness, and diminished touch sensation over the distal part of the lateral aspect of the right leg and dorsum of the right foot. The EMG and nerve conduction studies revealed severe right common peroneal nerve dysfunction at the fibular head. MRI was performed (Fig. 6), and the short head of the biceps femoris was subjectively interpreted as hypertrophied. In retrospect, we analyzed the posterior extent of the short head, which measured 1.8 cm and contributed to the formation of a tunnel. The short head of the biceps femoris extended 1.0 cm below the level of the joint space. The common peroneal nerve traveled within a tunnel between the lateral head of the gastrocnemius muscle and the short head of the biceps femoris. The longitudinal distance of the tunnel was 4.5 cm. Focal increased signal of the common peroneal nerve was noted in the posterior knee and around the fibular head. Denervation edema and minimal atrophy of the anterior and lateral compartments of the upper third of the leg were characterized by high signal intensity on T2-weighted images and iso-signal on T1-weighted sequence. The degree of muscle atrophy was subjectively estimated in comparison with the muscle of the posterior compartment of the calf. The involved muscles were the anterior tibial, extensor digitorum, peroneus longus, and peroneus brevis. The MRI findings showed that a variation in the distal length of the short head of the biceps femoris muscle was producing entrapment neuropathy of the common peroneal nerve and secondary subacute denervation of the anterior and lateral compartments. The patient underwent decompressive surgery with removal of the short head of the biceps femoris. The patient described significant daily improvement of symptoms 2 weeks after surgery. The neurologic examination performed 1 month after surgery was normal except for a mild Tinel sign over the fibular head. Repeat EMG and nerve conduction studies depicted a significant functional improvement with only minimal residual neurogenic dysfunction of the right common peroneal nerve at the fibular head. Discussion The biceps femoris is the most lateral component of the hamstring muscle. It is a strong flexor of the knee joint and has two heads of origin. The long head arises together with the semitendinosus muscle from the ischial tuberosity, crosses laterally, and becomes tendinous 7 10 cm above the knee joint [6]. The short head arises from the lateral prolongation of the linea aspera of the femur [7]. The tendons of the short head and long head merge above the knee joint, and the combined tendon inserts into the head of the fibula, crural fascia, and proximal tibia [6]. The common peroneal nerve arises from the sciatic nerve at the upper level of the popliteal fossa. It descends obliquely along the lateral side of the popliteal fossa, posterior to the short head of the biceps femoris muscle, and then lateral and superficial to the lateral head of the gastrocnemius muscle. More inferiorly, it winds around the neck of the fibula, enters the anterior and lateral muscle compartments of the leg, and divides into the deep and superficial peroneal branches [4] (Fig. 1). The common peroneal nerve and its branches provide motor innervation to the anterior compartment of the leg (tibialis ante- JR:189, September

6 Vieira et al. C Fig. 6 Common peroneal nerve entrapment in 14-year-old boy with clinical and electromyography evidence of common peroneal neuropathy. D, xial T1 images (TR/TE, 466/9) (,, and D) and T2 fat-saturated image (3,800/119) (C) of knee show hypertrophied muscle belly of short head of distal biceps femoris muscle (curved arrow, and ) associated with tunnel formation (circle, and ). Common peroneal nerve (black arrow, C) is lying inside this tunnel, between short head of biceps femoris and lateral head of gastrocnemius muscle (asterisk, and ) and has increased signal on T2-weighted image. There is also denervation edema and minimal denervation atrophy (white arrow, C and D) of anterior compartment of leg. D rior, extensor hallucis longus, extensor digitorum longus, and peroneus tertius muscles) and lateral compartment of the leg (peroneus longus and brevis muscles) and sensory innervation to the distal two thirds of the leg [3]. The most frequent mononeuropathy in the lower extremity involves the common peroneal nerve around the fibular head and neck area [1]. There are many causes of common peroneal neuropathy [1 5], and no report in the literature, to our knowledge, describes common peroneal entrapment neuropathy secondary to an anomalous relationship between the biceps femoris muscle and the nerve. On the basis of the MRI findings in our asymptomatic population, we believe that the relationship between the common peroneal nerve and the distal biceps femoris muscle can be categorized into two major patterns. In the most common pattern, noted in 77% of knees in our study, the common peroneal nerve is situated within abundant fat superficial to the lateral head of the gastrocnemius muscle and posterior to the short head of the biceps femoris muscle. In the remaining 23% of knees, the common peroneal nerve is situated within a narrow tunnel between the lateral head of the gastrocnemius muscle and the short head of the biceps femoris muscle. third pattern, which was seen in one asymptomatic patient, was distal extension of the long head of the biceps femoris muscle that contributed to the formation of the tunnel. The long head was depicted in 40 knees, and in only one case did a low-lying long head of the biceps femoris muscle contribute to the formation of a tunnel. To our knowledge, the presence of the tunnel and the different patterns of anatomic relationships between the common peroneal nerve and the short head of the biceps femoris have not yet been described in the Englishlanguage literature. We found that there is no description in the literature of the distal extent of the short head of the biceps femoris relative to the joint level. In Gray s natomy [7], the only information available is the short head arises from the lateral lip of the linea aspera, between the adductor magnus and vastus lateralis, extending up almost as high as the insertion of the gluteus maximus; from the lateral prolongation of the linea aspera to within 5 cm of the lateral condyle. We found that the distal myotendinous junction of the short head of the biceps femoris ended at the level of the joint space in 51% of the cases and ranged from 1.5 cm above to 2.0 cm below the level of the joint. In 554 JR:189, September 2007

7 MRI of the iceps Femoris Muscle our study, subjects with a tunnel had significantly greater distal extent of the muscle belly of the short head of the biceps femoris relative to the joint space than subjects without a tunnel (p < 0.001). We also measured the distance between the distal myotendinous junction of the long head of the biceps femoris from the joint space. In our study, the muscle belly of the long head of the biceps femoris was identified in 40 knees (40%). The distance between the long head myotendinous junction to the joint space was 5 cm on average (range, cm above the joint space). Our findings are different from the previously reported data (7 10 cm above the knee joint) [6]. nother interesting issue, which motivated us to start this study, was the case of common peroneal neuropathy associated with the presence of a tunnel. Most interestingly, the length of the tunnel in the symptomatic patient (4.5 cm) was almost twice the mean length of the tunnel in the asymptomatic group (2.4 cm). Moreover, only 4.2% of the asymptomatic population had a tunnel length of 4 cm or greater. lthough we believe that the length of the tunnel in our symptomatic patient contributed to his neuropathy, we think that the presence of the tunnel is a normal variation that is not necessarily associated with but that can predispose to common peroneal nerve entrapment. In our 100 consecutive cases, we did not find any signal abnormality compatible with common peroneal neuropathy, such as denervation edema of the anterior and posterior compartments of the proximal leg or signal abnormality of the common peroneal nerve in the popliteal fossa. It is important to highlight that a passage shorter than 1 cm in its superior-to-inferior extent was not included in the definition of a tunnel because we believe that a tunnel shorter than 1 cm could not represent a real tunnel anatomically. Our study has several limitations, such as the retrospective design, lack of cadaveric correlation, and only one symptomatic case of common peroneal neuropathy associated with abnormal distal biceps femoris. lso, we did not study the contribution of the lateral head of the gastrocnemius muscle to the formation of the tunnel. On the other hand, we believe that in our population the variations in the gastrocnemius muscle did not contribute significantly to tunnel formation because we did not see many anatomic variations in the gastrocnemius muscle. In conclusion, we describe the normal MRI anatomy of the distal biceps femoris and the relationship of this muscle with the common peroneal nerve. The common peroneal nerve typically courses downward within abundant fat posterior to the short head of the biceps femoris muscle and superficial to the lateral head of the gastrocnemius muscle, but it can also descend in a tunnel between the two muscles (23%). The presence of this tunnel, previously undescribed in the literature, correlates with the posterior and distal extents of the short head of the biceps femoris. We also describe a case of common peroneal neuropathy secondary to an unusual relationship between the common peroneal nerve and the distal biceps femoris. References 1. prile I, Caliandro P, La Torre G, et al. Multicenter study of peroneal mononeuropathy: clinical, neurophysiologic, and quality of life assessment. J Peripher Nerv Syst 2005; 10: Iverson DJ. MRI detection of cysts of the knee causing common peroneal neuropathy. Neurology 2005; 65: LeGeyt MT, mbrose J. Nontraumatic compression of the common peroneal nerve: a case report and review of the literature. m J Orthop 1998; 27: Loredo R, Hodler J, Pedowitz R, Yeh LR, Trudell D, Resnick D. MRI of the common peroneal nerve: normal anatomy and evaluation of masses associated with nerve entrapment. J Comput ssist Tomogr 1998; 22: Dellon L. Postarthroplasty palsy and systemic neuropathy: a peripheral-nerve management algorithm. nn Plast Surg 2005; 55: Sneath RS. The insertion of the biceps femoris. J nat 1955; 89: Gray H. Muscle and fasciae. In: Gray H, ed. Gray s anatomy, 15th ed. New York, NY: ounty ooks, 1977:57 59 JR:189, September