Longitudinal split of the posterior cruciate ligament: description of a new MR finding and evaluation of its potential clinical significance

Results Seven of 12 patients (58.3%) had morphological or functional evidence of PCL injury or insufficiency according to the change of posterior instability on FU stress testing ( n = 3), insufficiency during arthroscopy ( n = 2), or decreased extent and altered shape of the PCL split on the FU MRI ( n = 3). One patient revealed both change of posterior instability on FU stress testing and insufficiency during arthroscopy. Combined injuries were revealed in seven patients. Five patients had isolated LS-PCL: two patients underwent arthroscopic PCL reconstructions; and another three patients revealed knee instability on stress testing. Conclusion Although LS-PCL has not been described before, it can be a type of partial tear of the PCL, which causes PCL insufficiency. Introduction Posterior cruciate ligament (PCL) injuries include a partial tear or intra-substance injury, complete ligamentous rupture, and avulsion of the PCL from its femoral or tibial attachment. 1,2 In the acute phase of a traumatic injury, physical examination can usually determine an acute tear of the PCL, and patients rarely describe hearing the “pop” at the time of injury, which is common with anterior cruciate ligament (ACL) tears. As a result, many PCL tears remain undiagnosed during the initial clinical evaluation. 3 However, magnetic resonance imaging (MRI) has revealed that the PCL is more often subject to injuries than was previously believed. Although the significance and presence of such an injury may not be immediately recognized, the late onset of instability and arthritis resulting from such an injury may be the start of irreversible limitations in activity, leading to eventual debilitation. Correct diagnosis of PCL tears can be challenging but rewarding for both physicians and patients. 1 An increased signal intensity (SI) was noticed in PCLs in a longitudinal direction along the course of the ligaments shown on knee MRI and was denoted intra-substance longitudinal split of the PCL. Unlike common types of PCL tear, longitudinal split of the PCL has not been described before. To the authors' knowledge, there have been no previous reports regarding this type of injury. The present study was conducted to describe the new MRI finding of the longitudinal split of the PCL and to evaluate its potential clinical significance regarding posterior instability of the PCLs and associated injuries on MRI. Materials and methods The present study was performed with institutional review board approval at two medical centres. Informed consent was waived because the study involved retrospective evaluation of patients’ MRIs and records in such a manner that the patients could not be identified. The databases of knee MRI obtained at two centres were searched for longitudinal split of the PCL: 5516 MRIs in centre A during 61 months (December 2003to January 2008) and 1401 MRIs in centre B during 11 months (March 2007 to January 2008). The knee MRI reports were searched for the words of “longitudinal” and “PCL” to obtain a list of 14 patients. The intra-substance longitudinal split of the PCL was defined as an increased signal intensity in a PCL in the longitudinal direction, but with an intact ligament outer surface on sagittal proton density (PD)- and T2-weighted images and coronal or axial fat-suppressed PD- or T2-weighted MRIs ( Fig 1 ). All MRIs were reviewed in consensus by two musculoskeletal radiologists. They excluded one patient during this case review because the image of the patient’s PCL was inadequate regarding the definition of intra-substance longitudinal split of the PCL. Thirteen patients exhibited the inclusion criteria, although one of these patients was excluded due to his history of ACL reconstruction. Finally, 12 patients were enrolled and comprised seven male and five female patients with an average age of 36 years (range 13–64 years). Available arthroscopic results, the degree of posterior knee instability, and changes of MRI findings or the degree of instability during follow-up (FU), were reviewed in their medical records and on their MRI images. The degree of instability was evaluated on physical examination by one of the two experienced orthopaedic surgeons with specialization in the knee joint, stress radiographs, or instrumented measurement using a KT-2000 arthrometer. The interval between initial MRI and stress radiographs or instrumented measurement was 16–75 days. The time intervals between the initial and FU MRI examinations were 70–374 days. Each patient’s medical records were also reviewed for possible trauma history, as well as their occupation. Six knee MRIs were performed using a 3 T MRI unit (Gyroscan Archieva; Philips Medical System International, Best, Netherlands) using a four- or eight-channel knee coil, a 14–16 cm field of view, and a 512 × 512 matrix. Imaging sequences included axial fast spin echo (FSE) proton density (PD) with fat suppression [repetition time/echo time (TR/TE) 3000 ms/29 ms; 8 echo-train length (ETL); 2 mm section thickness (ST)], sagittal PD (TR/TE 3743 ms/20 ms; 8 ETL; 1.5 mm ST), sagittal FSE T2 (TR/TE 6259 ms/100 ms; 16 ETL; 2.5 mm ST), sagittal T1 spin echo (TR/TE 458 ms/22 ms; 4 ETL; 2.5 mm ST), and coronal FSE PD images (TR/TE 2401 ms/20 ms; 8 ETL; 2.5 mm ST). Six other knee MRIs were performed on a 1.5 T MRI unit (Signa CV/I; General Electric Medical Systems, Milwaukee, WI, USA, n = 5; Magnetom Vision; Siemens Medical Systems, Erlangen, Germany, n = 1) using a knee coil, a 14–16 cm field of view, and a 3 mm ST. Imaging sequences included sagittal PD (TR/TE 2760 ms/15 ms), sagittal FSE T2 (TR/TE 3900 ms/105 ms), sagittal T1 spin echo (TR/TE 530 ms/12 ms), coronal PD (TR/TE 2760 ms/15 ms), and axial PD or T2 with fat-suppression images (TR/TE 3000 ms/29 ms or 2645 ms/50 ms). On MRI, the location of the PCL split into entire, mid-to-distal, and distal portions was recorded. Any associated injuries on knee MRI, e.g., menisci lesions, ligament injuries, and bone marrow changes, were evaluated. Results The causes of the patients’ injuries included sports injury ( n = 3), motor vehicle accident ( n = 1), and trauma during military training ( n = 1). For the other seven patients, their histories of knee injury were not specified in their medical records. Among the finally enrolled 12 patients, six underwent posterior instability test using a posterior drawer test, stress radiographs, or instrumented measurement using a KT-2000 arthrometer. Two patients underwent knee arthroscopy, three patients had FU MRI, and three underwent FU instability test. The history and available results of the patients’ MRIs and stress tests are listed in Tables 1 and 2 . Knee instability manifested on stress testing in six of the 12 patients (50%). Among these six patients, three demonstrated change of instability during the 1–2 month FU, i.e., aggravation in two patients and improvement in one patient. Two patients underwent arthroscopic PCL reconstructions due to posterior instability ( Fig 2 ). On arthroscopy, although the PCL preserved its continuity and appeared normal, the ligaments revealed insufficiency on probing, loss of normal strength and tension of the PCL and widening of intercondylar notch by pulling or pushing the ligament. Three patients, in whom FU MRIs were available, showed a decreased extent and altered shape of the PCL split on the FU MRI ( Fig 3 ). As a result, seven of 12 patients (58.3%) revealed morphological and functional evidence of PCL injury or insufficiency. In the other five patients, there was not sufficient evidence of PCL injury: one patient underwent total knee replacement arthroplasty (patient 7); and four patients were lost during FU (patients 3, 6, 8, and 11; Tables 1 and 2 ). Five of 12 patients had isolated longitudinal PCL splits: two patients underwent arthroscopic PCL reconstructions, as previously mentioned; and in other three patients knee instability was revealed on stress testing. Associated injuries were revealed in seven patients ( Fig 4 ): meniscus tear ( n = 7): medial meniscus ( n = 4); lateral meniscus ( n = 1); and both menisci ( n = 2); complete ACL rupture ( n = 2); bone marrow oedema ( n = 3); and ganglion cysts along the PCL ( n = 1). Locations of the longitudinal split of PCL within the ligaments were variable, i.e., mid-to-distal ( n = 5), distal ( n = 4), and the entire ligament ( n = 3). Discussion In chronic PCL injury, although the continuity of the ligament is preserved, it may be lax. 1 According to previous reports, 1,4 isolated PCL injuries have been treated conservatively with an excellent prognosis. However, complete interstitial tears of the PCL should be treated if other ligamentous injuries of the knee are present, such as ACL, medial or lateral collateral ligaments, or arcuate ligament complex injuries. Combined injuries involving the PCL have a more guarded prognosis and are therefore usually treated by surgical repair or reconstruction within 3 weeks of the injury. Surgical results are better than those results seen with conservative management; however, it is difficult to clinically distinguish between the two injuries. In the present study, all of the three patients who were treated conservatively and followed by MRI, showed decreased PCL split on MRI. Conversely two other patients with isolated longitudinal split of the PCL revealed ligament insufficiency during knee arthroscopy and subsequently underwent PCL reconstruction. Therefore, isolated intra-substance longitudinal split of the PCL can be one of the types of partial tear of the PCL, which causes instability of the knee. Reports of longitudinal tears of the ligaments in the joints could not be found. However, longitudinal tears are well-known in the tendons of biceps long head of the shoulder and peroneus brevis of the ankle. Several researchers have recently proposed that there is a correlation between a longitudinal tear in the peroneus brevis tendon and chronic lateral ankle-ligament instability. 5–9 Similar to the peroneus brevis tendon, even though ligaments and tendons are not similar histologically, longitudinal splitting of the PCL could be one of the causes of knee instability. The pathogenesis of this longitudinal splitting of the PCL can not be clearly explained. It is possible that a minor injury with a shearing or stretching force might cause the injury as per the patients’ injury history. The PCL possesses two functional bundles: anterolateral and posteromedial. The names reflect their relative positions on the femur. These are not strictly anatomic separations of the ligament but can be isolated by their distinct patterns of tension at differing angles of knee flexion. 2,10–13 However, these two bundles of normal PCL are not separately visualized in knee MR images. The longitudinal split in the present cases does not indicate the normal borders of two bundles of PCL. There may be confusion between the longitudinal split of the PCL and intra-ligamentous ganglion cyst of PCL. However, unlike ganglion cysts, absence of mass effect and resolution of the longitudinal split of the PCL during FU may be differential points between them. The present study has several limitations, the first being the small number of patient in the study population and the retrospective nature of the study. Another limitation is the small number of patients who underwent arthroscopy, FU stress testing, and FU MRI. In addition, four patients were lost during FU. Therefore, not all of the MRI findings were correlated with the arthroscopic findings and instability testing. In conclusion, although intra-substance longitudinal split of the PCL has not been described before, it can in fact be a type of partial tear of the PCL, which causes insufficiency of the ligament. It commonly occurs with meniscus or ACL injuries, but it can also occur as an isolated injury. Acknowledgements The authors thank Bonnie Hami, MA (USA), for assistance with language correction in the manuscript. References 1 C.T. Servant J.P. Ramos N.P. Thomas The accuracy of magnetic resonance imaging in diagnosing chronic posterior cruciate ligament injury Knee 11 2004 265 270 2 A.A. Amis C.M. Gupte A.M. Bull Anatomy of the posterior cruciate ligament and the meniscofemoral ligaments Knee Surg Sports Traumatol Arthrosc 14 2006 257 263 3 W. Rodriguez E.N. Vinson C.A. Helms MRI appearance of posterior cruciate ligament tears AJR Am J Roentgenol 191 2008 155 159 4 M.D. Miller D.L. Johnson C.D. Harner Posterior cruciate ligament injuries Orthop Rev 22 1993 1201 1210 5 M. Sobel W.H. Bohne J.A. Markisz Cadaver correlation of peroneal tendon changes with magnetic resonance imaging Foot Ankle 11 1991 384 388 6 J. Karlsson S. Brandsson P. Kälebo Surgical treatment of concomitant chronic ankle instability and longitudinal rupture of the peroneus brevis tendon Scand J Med Sci Sports 8 1998 42 49 7 M. Sobel M.J. Geppert R.F. Warren Chronic ankle instability as a cause of peroneal tendon injury Clin Orthop 296 1993 187 191 8 O. Minoyama E. Uchiyama H. Iwaso Two cases of peroneus brevis tendon tear Br J Sports Med 36 2002 65 66 9 J. Karlsson P. Wiger Longitudinal split of the peroneus brevis tendon and lateral ankle instability: treatment of concomitant lesions J Athl Train 37 2002 463 466 10 J.C. Hughston J.A. Bowden J.R. Andrews Acute tears of the posterior cruciate ligament: results of operative treatment J Bone Joint Surg Am 62 1980 438 450 11 F.G. Girgis J.L. Marshall A. Monajem The cruciate ligaments of the knee joint anatomical, functional and experimental analysis Clin Orthop Related Res 106 1975 216 231 12 A. Race A.A. Amis The mechanical properties of the two bundles of the human posterior cruciate ligament J Biomech 27 1994 13 24 13 A. Edwards A.M. Bull A.A. Amis The attachments of the fiber bundles of the posterior cruciate ligament: an anatomic study Arthroscopy 23 2007 284 290