About all

Image of acl tear: Imaging of the anterior cruciate ligament

Imaging of the anterior cruciate ligament

1. Yasuda K, van Eck CF, Hoshino Y, Fu FH, Tashman S. Anatomic single- and double-bundle anterior cruciate ligament reconstruction, part 1: basic science. Am J Sports Med. 2011;39:1789–1799. [PubMed] [Google Scholar]

2. Norwood LA, Cross MJ. Anterior cruciate ligament: functional anatomy of its bundles in rotatory instabilities. Am J Sports Med. 1979;7:23–26. [PubMed] [Google Scholar]

3. Amis AA, Dawkins GP. Functional anatomy of the anterior cruciate ligament. Fibre bundle actions related to ligament replacements and injuries. J Bone Joint Surg Br. 1991;73:260–267. [PubMed] [Google Scholar]

4. Girgis FG, Marshall JL, Monajem A. The cruciate ligaments of the knee joint. Anatomical, functional and experimental analysis. Clin Orthop Relat Res. 1975:216–231. [PubMed] [Google Scholar]

5. Cohen SB, VanBeek C, Starman JS, Armfield D, Irrgang JJ, Fu FH. MRI measurement of the 2 bundles of the normal anterior cruciate ligament. Orthopedics. 2009;32:687. [PubMed] [Google Scholar]

6. Takai S, Woo SL, Livesay GA, Adams DJ, Fu FH. Determination of the in situ loads on the human anterior cruciate ligament. J Orthop Res. 1993;11:686–695. [PubMed] [Google Scholar]

7. Sakane M, Fox RJ, Woo SL, Livesay GA, Li G, Fu FH. In situ forces in the anterior cruciate ligament and its bundles in response to anterior tibial loads. J Orthop Res. 1997;15:285–293. [PubMed] [Google Scholar]

8. Gabriel MT, Wong EK, Woo SL, Yagi M, Debski RE. Distribution of in situ forces in the anterior cruciate ligament in response to rotatory loads. J Orthop Res. 2004;22:85–89. [PubMed] [Google Scholar]

9. Zantop T, Herbort M, Raschke MJ, Fu FH, Petersen W. The role of the anteromedial and posterolateral bundles of the anterior cruciate ligament in anterior tibial translation and internal rotation. Am J Sports Med. 2007;35:223–227. [PubMed] [Google Scholar]

10. Larson RL. Prosthetic replacement of knee ligaments: Overview. In: Feagin JA, editor. The crucial ligaments: Diagnosis and treatment of ligamentous injuries about the knee. New York: Churchill Livingstone; 1988. pp. 495–501. [Google Scholar]

11. Weber WN, Neumann CH, Barakos JA, Petersen SA, Steinbach LS, Genant HK. Lateral tibial rim (Segond) fractures: MR imaging characteristics. Radiology. 1991;180:731–734. [PubMed] [Google Scholar]

12. Lubowitz JH, Bernardini BJ, Reid JB. Current concepts review: comprehensive physical examination for instability of the knee. Am J Sports Med. 2008;36:577–594. [PubMed] [Google Scholar]

13. Vahey TN, Broome DR, Kayes KJ, Shelbourne KD. Acute and chronic tears of the anterior cruciate ligament: differential features at MR imaging. Radiology. 1991;181:251–253. [PubMed] [Google Scholar]

14. Griffith JF, Antonio GE, Tong CW, Ming CK. Cruciate ligament avulsion fractures. Arthroscopy. 2004;20:803–812. [PubMed] [Google Scholar]

15. Bales CP, Guettler JH, Moorman CT. Anterior cruciate ligament injuries in children with open physes: evolving strategies of treatment. Am J Sports Med. 2004;32:1978–1985. [PubMed] [Google Scholar]

16. Goldman AB, Pavlov H, Rubenstein D. The Segond fracture of the proximal tibia: a small avulsion that reflects major ligamentous damage. AJR Am J Roentgenol. 1988;151:1163–1167. [PubMed] [Google Scholar]

17. Stallenberg B, Gevenois PA, Sintzoff SA, Matos C, Andrianne Y, Struyven J. Fracture of the posterior aspect of the lateral tibial plateau: radiographic sign of anterior cruciate ligament tear. Radiology. 1993;187:821–825. [PubMed] [Google Scholar]

18. Chan KK, Resnick D, Goodwin D, Seeger LL. Posteromedial tibial plateau injury including avulsion fracture of the semimembranous tendon insertion site: ancillary sign of anterior cruciate ligament tear at MR imaging. Radiology. 1999;211:754–758. [PubMed] [Google Scholar]

19. Donald L, Resnick MD, Heung Sik Kang MD, Michael L. Knee. 2th ed. Philadelphia: Saunders; 2007. Internal Derangements of Joints. p. 1823. [Google Scholar]

20. Pao DG. The lateral femoral notch sign. Radiology. 2001;219:800–801. [PubMed] [Google Scholar]

21. Vande Berg BC, Lecouvet FE, Poilvache P, Dubuc JE, Maldague B, Malghem J. Anterior cruciate ligament tears and associated meniscal lesions: assessment at dual-detector spiral CT arthrography. Radiology. 2002;223:403–409. [PubMed] [Google Scholar]

22. Antonio GE, Griffith JF, Yeung DK. Small-field-of-view MRI of the knee and ankle. AJR Am J Roentgenol. 2004;183:24–28. [PubMed] [Google Scholar]

23. Pereira ER, Ryu KN, Ahn JM, Kayser F, Bielecki D, Resnick D. Evaluation of the anterior cruciate ligament of the knee: comparison between partial flexion true sagittal and extension sagittal oblique positions during MR imaging. Clin Radiol. 1998;53:574–578. [PubMed] [Google Scholar]

24. Niitsu M, Ikeda K, Fukubayashi T, Anno I, Itai Y. Knee extension and flexion: MR delineation of normal and torn anterior cruciate ligaments. J Comput Assist Tomogr. 1996;20:322–327. [PubMed] [Google Scholar]

25. Gene Saragnese. 3D imaging of the knee takes a step forward. Field strength. 2010;42:16–19. [Google Scholar]

26. Lee JK, Yao L, Phelps CT, Wirth CR, Czajka J, Lozman J. Anterior cruciate ligament tears: MR imaging compared with arthroscopy and clinical tests. Radiology. 1988;166:861–864. [PubMed] [Google Scholar]

27. McCauley TR, Moses M, Kier R, Lynch JK, Barton JW, Jokl P. MR diagnosis of tears of anterior cruciate ligament of the knee: importance of ancillary findings. AJR Am J Roentgenol. 1994;162:115–119. [PubMed] [Google Scholar]

28. Mink JH, Levy T, Crues JV. Tears of the anterior cruciate ligament and menisci of the knee: MR imaging evaluation. Radiology. 1988;167:769–774. [PubMed] [Google Scholar]

29. Robertson PL, Schweitzer ME, Bartolozzi AR, Ugoni A. Anterior cruciate ligament tears: evaluation of multiple signs with MR imaging. Radiology. 1994;193:829–834. [PubMed] [Google Scholar]

30. Tung GA, Davis LM, Wiggins ME, Fadale PD. Tears of the anterior cruciate ligament: primary and secondary signs at MR imaging. Radiology. 1993;188:661–667. [PubMed] [Google Scholar]

31. Lo IK, de Maat GH, Valk JW, Frank CB. The gross morphology of torn human anterior cruciate ligaments in unstable knees. Arthroscopy. 1999;15:301–306. [PubMed] [Google Scholar]

32. Duc SR, Zanetti M, Kramer J, Käch KP, Zollikofer CL, Wentz KU. Magnetic resonance imaging of anterior cruciate ligament tears: evaluation of standard orthogonal and tailored paracoronal images. Acta Radiol. 2005;46:729–733. [PubMed] [Google Scholar]

33. Hong SH, Choi JY, Lee GK, Choi JA, Chung HW, Kang HS. Grading of anterior cruciate ligament injury. Diagnostic efficacy of oblique coronal magnetic resonance imaging of the knee. J Comput Assist Tomogr. 2003;27:814–819. [PubMed] [Google Scholar]

34. Ng AW, Griffith JF, Law KY, Ting JW, Tipoe GL, Ahuja AT, Chan KM. Oblique axial MR imaging of the normal anterior cruciate ligament bundles. Skeletal Radiol. 2011 [PubMed] [Google Scholar]

35. Umans H, Wimpfheimer O, Haramati N, Applbaum YH, Adler M, Bosco J. Diagnosis of partial tears of the anterior cruciate ligament of the knee: value of MR imaging. AJR Am J Roentgenol. 1995;165:893–897. [PubMed] [Google Scholar]

36. Yao L, Gentili A, Petrus L, Lee JK. Partial ACL rupture: an MR diagnosis? Skeletal Radiol. 1995;24:247–251. [PubMed] [Google Scholar]

37. Van Dyck P, Vanhoenacker FM, Gielen JL, Dossche L, Van Gestel J, Wouters K, Parizel PM. Three tesla magnetic resonance imaging of the anterior cruciate ligament of the knee: can we differentiate complete from partial tears? Skeletal Radiol. 2011;40:701–707. [PubMed] [Google Scholar]

38. Brandser EA, Riley MA, Berbaum KS, el-Khoury GY, Bennett DL. MR imaging of anterior cruciate ligament injury: independent value of primary and secondary signs. AJR Am J Roentgenol. 1996;167:121–126. [PubMed] [Google Scholar]

39. Gentili A, Seeger LL, Yao L, Do HM. Anterior cruciate ligament tear: indirect signs at MR imaging. Radiology. 1994;193:835–840. [PubMed] [Google Scholar]

40. Kaplan PA, Walker CW, Kilcoyne RF, Brown DE, Tusek D, Dussault RG. Occult fracture patterns of the knee associated with anterior cruciate ligament tears: assessment with MR imaging. Radiology. 1992;183:835–838. [PubMed] [Google Scholar]

41. Schweitzer ME, Cervilla V, Kursunoglu-Brahme S, Resnick D. The PCL line: an indirect sign of anterior cruciate ligament injury. Clin Imaging. 1992;16:43–48. [PubMed] [Google Scholar]

42. Vahey TN, Hunt JE, Shelbourne KD. Anterior translocation of the tibia at MR imaging: a secondary sign of anterior cruciate ligament tear. Radiology. 1993;187:817–819. [PubMed] [Google Scholar]

43. Chiu SS. The anterior tibial translocation sign. Radiology. 2006;239:914–915. [PubMed] [Google Scholar]

44. Chan WP, Peterfy C, Fritz RC, Genant HK. MR diagnosis of complete tears of the anterior cruciate ligament of the knee: importance of anterior subluxation of the tibia. AJR Am J Roentgenol. 1994;162:355–360. [PubMed] [Google Scholar]

45. Boeree NR, Ackroyd CE. Magnetic resonance imaging of anterior cruciate ligament rupture. A new diagnostic sign. J Bone Joint Surg Br. 1992;74:614–616. [PubMed] [Google Scholar]

46. Remer EM, Fitzgerald SW, Friedman H, Rogers LF, Hendrix RW, Schafer MF. Anterior cruciate ligament injury: MR imaging diagnosis and patterns of injury. Radiographics. 1992;12:901–915. [PubMed] [Google Scholar]

47. Smith JP, Barrett GR. Medial and lateral meniscal tear patterns in anterior cruciate ligament-deficient knees. A prospective analysis of 575 tears. Am J Sports Med. 2001;29:415–419. [PubMed] [Google Scholar]

48. Vinson EN, Gage JA, Lacy JN. Association of peripheral vertical meniscal tears with anterior cruciate ligament tears. Skeletal Radiol. 2008;37:645–651. [PubMed] [Google Scholar]

49. Laundre BJ, Collins MS, Bond JR, Dahm DL, Stuart MJ, Mandrekar JN. MRI accuracy for tears of the posterior horn of the lateral meniscus in patients with acute anterior cruciate ligament injury and the clinical relevance of missed tears. AJR Am J Roentgenol. 2009;193:515–523. [PubMed] [Google Scholar]

50. Allen CR, Wong EK, Livesay GA, Sakane M, Fu FH, Woo SL. Importance of the medial meniscus in the anterior cruciate ligament-deficient knee. J Orthop Res. 2000;18:109–115. [PubMed] [Google Scholar]

51. Thompson WO, Fu FH. The meniscus in the cruciate-deficient knee. Clin Sports Med. 1993;12:771–796. [PubMed] [Google Scholar]

52. Vinson EN, Major NM, Helms CA. The posterolateral corner of the knee. AJR Am J Roentgenol. 2008;190:449–458. [PubMed] [Google Scholar]

53. Meister BR, Michael SP, Moyer RA, Kelly JD, Schneck CD. Anatomy and kinematics of the lateral collateral ligament of the knee. Am J Sports Med. 2000;28:869–878. [PubMed] [Google Scholar]

54. Fineberg MS, Duquin TR, Axelrod JR. Arthroscopic visualization of the popliteus tendon. Arthroscopy. 2008;24:174–177. [PubMed] [Google Scholar]

55. Peduto AJ, Nguyen A, Trudell DJ, Resnick DL. Popliteomeniscal fascicles: anatomic considerations using MR arthrography in cadavers. AJR Am J Roentgenol. 2008;190:442–448. [PubMed] [Google Scholar]

56. Yu JS, Salonen DC, Hodler J, Haghighi P, Trudell D, Resnick D. Posterolateral aspect of the knee: improved MR imaging with a coronal oblique technique. Radiology. 1996;198:199–204. [PubMed] [Google Scholar]

57. Recht MP, Applegate G, Kaplan P, Dussault R, Schweitzer M, Dalinka MK, Resnick D. The MR appearance of cruciate ganglion cysts: a report of 16 cases. Skeletal Radiol. 1994;23:597–600. [PubMed] [Google Scholar]

58. McIntyre J, Moelleken S, Tirman P. Mucoid degeneration of the anterior cruciate ligament mistaken for ligamentous tears. Skeletal Radiol. 2001;30:312–315. [PubMed] [Google Scholar]

59. Bergin D, Morrison WB, Carrino JA, Nallamshetty SN, Bartolozzi AR. Anterior cruciate ligament ganglia and mucoid degeneration: coexistence and clinical correlation. AJR Am J Roentgenol. 2004;182:1283–1287. [PubMed] [Google Scholar]

60. Papadopoulou P. The celery stalk sign. Radiology. 2007;245:916–917. [PubMed] [Google Scholar]

61. Tzurbakis M, Mouzopoulos G, Morakis E, Nikolaras G, Georgilas I. Intra-articular knee haemangioma originating from the anterior cruciate ligament: a case report. J Med Case Reports. 2008;2:254. [PMC free article] [PubMed] [Google Scholar]

62. Majima T, Kamishima T, Susuda K. Synovial chondromatosis originating from the synovium of the anterior cruciate ligament: a case report. Sports Med Arthrosc Rehabil Ther Technol. 2009;1:6. [PMC free article] [PubMed] [Google Scholar]

Partial ACL Tear – Radsource

Clinical History: A 40 year-old female presents after a skiing injury. Proton-density weighted fat-suppressed sagittal (1a) and axial (1b) images are provided. What are the findings” What is your diagnosis?



Figure 1:

Proton-density weighted fat-suppressed sagittal (1a) and axial (1b) images



Figure 2:

Subtle focal hyperintensity in the substance of the proximal anterior cruciate ligament (arrow) and a slightly wavy contour of its anterior margin (arrowhead) are demonstrated on (2a), a fat-suppressed proton-density sagittal image.


Figure 3:

The fat-suppressed proton-density axial image (3a) demonstrates focal fluid signal along the anterior margin of the anterior cruciate ligament (arrow) in the intercondylar notch. Avulsive marrow edema (asterisk) is present in the lateral femoral condyle deep to the femoral origin of the anterior cruciate ligament. Posterior fibers of the anterior cruciate ligament remain attached to the femur (arrowhead). Edema and hemorrhage obscure the medial collateral ligament (short arrows).


Partial tear of the proximal anteromedial band of the anterior cruciate ligament and tear of the medial collateral ligament.


Partial tears of the anterior cruciate ligament (ACL) are common. The prognosis of a partial ACL tear is controversial and is dependent on the extent of the partial tear and associated meniscal, ligamentous, and osteochondral injuries. Small tears involving less than 25% of the ACL cross-section have a favorable prognosis of healing while maintaining stability of the knee. Tears involving 50-75% of the ACL demonstrate a significant probability of progressing to a complete tear.1

Biomechanics of the ACL

The anterior cruciate ligament is composed of densely organized fibrous collagenous connective tissue that attaches the femur to the tibia. The ACL is most commonly composed of two bands: the anteromedial and posterolateral. The anteromedial band is the stronger of the two and is taut in flexion while the posterolateral band is lax. In extension, the posterolateral band becomes taut and the anteromedial band is lax (4a). This reciprocal tension pattern in flexion and extension allows the ACL to remain under tension and contribute to knee stability throughout the normal range of motion. Both bands of the ACL restrain anterior tibial translation depending on the position of the knee. In flexion, the anteromedial bundle is the primary restraint to anterior tibial translation. Deficiency of the anteromedial bundle is tested for by the anterior drawer test, which is performed at 90 degrees of flexion. With the knee in or near extension, the posterolateral bundle serves as the primary restraint to anterior tibial translation, thus posterolateral band deficiencies are more commonly detected by the Lachman test and pivot shift test. In addition, the posterolateral bundle appears to be a crucial component in maintaining rotational stability with the knee near extension.2


Figure 4:

(4a) This 3-D graphic representation of the knee in extension with the medial femoral condyle removed depicts laxity of the anteromedial bundle (red) and a taut posterolateral band (blue).


Figure 5:

(5a) In flexion, the anteromedial band (red) becomes taut and the posterolateral band (blue) becomes lax.

Mechanism of Injury

The anteromedial band of the ACL is most commonly injured in partial ACL tears. Injury mechanisms are the same for partial tear and complete tear. The most common contact-related injury is the “clipping” type injury sustained in football. The knee is flexed, and with the foot planted, a valgus force is applied to the knee. Hyperextension injuries with or without varus can result in ACL disruption. Non-contact related injuries are increasingly recognized and are common in athletes involved in rapid deceleration with cutting, pivoting, or jumping.

Physical Examination

The accepted physical examination findings of a partial ACL tear include:

  • History of injury to the ACL
  • Positive Lachman or anterior drawer test with a firm endpoint
  • Negative pivot shift test
  • KT-1000 side to side difference of less than 5 mm.

However the accurate detection of partial ACL tears by physical exam is limited by several factors. The ACL may be difficult to evaluate in large patients. Patient guarding and soft tissue swelling in the acute setting, as well as strong secondary muscle restraints, may limit the sensitivity of the physical examination in detecting partial ACL tear.1 In addition cadaver studies have proven that a functionally significant partial tear involving the anteromedial band may be missed by physical exam and arthrometric testing. 3

MRI Technique



Figure 6:

(6a) A fat-suppressed proton-density weighted sagittal image demonstrates the normal straight anterior margin of the ACL correlating with the anteromedial band. Note the normal striated appearance of the ACL.



Figure 7:

(7a) The normal ACL is ovoid and of low signal on axial images at the mid-intercondylar level demonstrated on this axial T2-weighted FSE image. The regions of the anteromedial band (red) and the posterolateral band (blue) are indicated by the superimposed oval.

The ACL is normally seen as a longitudinally striated band, which parallels the intercondylar roof (6a) on slightly obliqued sagittal images. Close inspection of the axial (7a) and coronal views is also necessary for a thorough evaluation of the ACL. Thin T2-weighted sagittal images angled along the anticipated course of the ACL have become an accepted method for evaluating the ACL. This frequently enables a definitive diagnosis of a normal or torn ACL tear. However, we have found that coronal oblique T2-weighted images (8a, 9a, 10a) are more easily performed and more reliable for depicting the ACL, particularly when differentiating partial or complete tears from volume-averaged fluid in the intercondylar notch.


Figure 8:

(8a) A fat-suppressed proton-density sagittal image demonstrates a poorly defined ACL with intrasubstance increased signal (arrows). The yellow lines indicate the area covered and the angle used in obtaining coronal oblique images to better demonstrate the ACL.


Figure 9:

(9a) This T2-weighted coronal oblique view of the same knee as in illustration (8a) demonstrates a normal appearance of the ACL (arrows).


Figure 10:

(10a) In another patient, the T2-weighted coronal oblique image depicts a high-grade partial tear of the proximal ACL. The posterolateral bundle fibers are completely disrupted at the femoral origin (arrowhead) and the anteromedial band fibers are lax (arrow).

MRI Findings

Because of its ability to depict anatomy in multiple planes and its non-invasiveness, MRI offers distinct advantages over arthroscopy as a means of evaluating the ACL. MRI evaluation is effective in preventing unnecessary arthroscopy by assessing the severity of the ACL tear and coexisting injuries.4


Figure 11:

(11a) A fat-suppressed proton-density sagittal view of an acute partial ACL tear depicts focal increased signal of the ACL (arrowheads) and a wavy contour of posterior fibers, which remain in continuity (arrow).

Direct signs of an acute partial ACL tear include focal angulation of the ACL and focal intrasubstance high signal intensity on T2-weighted images, while still visualizing intact ACL fibers (10a, 11a). A chronic partial tear of the ACL may appear angulated and attenuated in caliber. Frequently there is overlap in the appearance of a partial tear and complete tear. In such cases, MRI, by evaluating the extent of ACL signal abnormality and the presence of coexisting injuries, is effective in stratifying ACL tears into low-risk partial ACL tears versus high-risk partial or complete ACL tears. High-grade partial tears are more commonly found with other major ligamentous (2a, 3a), meniscal, and osteochondral injuries.4 Partial tears that demonstrate a narrowed transverse dimension of the ACL on axial images with an otherwise normal anteroposterior dimension and a normally oriented ACL correlate to a stable partial tear.5 Absence of the anteromedial or posterolateral bundle (12a, 13a, 14a) correlates with an unstable high-grade partial tear. Partial tears associated with bone bruises are higher grade and at greater risk of progressing to complete tears.6 However, in the skeletally immature patient, typical pivot-shift bone bruises can occur without injury to the ACL.7


Figure 12:

(12a) Fat-suppressed proton-density sagittal view demonstrates subtle laxity of the anteromedial portion of the ACL (arrow).


Figure 13:

(13a) The next lateral slice for the patient in (9a) demonstrates abnormal laxity of the posterolateral bundle fibers (arrow).


Figure 14:

(14a) Fat-suppressed proton-density axial images of the patient in (12a and 13a) confirm the posterolateral band tear (arrowhead). The anteromedial band is thickened and edematous (arrow). This represents a single bundle ACL, which is at high risk for functional ACL instability.

The indirect signs of a partial ACL tear are the same as for a complete ACL tear, but are typically more subtle. Indirect signs of an ACL injury include pivot-shift bone bruises and fractures of the lateral femoral condyle and posterolateral tibia, “contrecoup” bone bruises in the posteromedial tibial plateau near the semimembranosus insertion, anterior displacement of the tibia, Segond fracture, avulsion fracture of the fibula, lateral meniscal tears at the junction of the posterior meniscofemoral ligament (Wrisberg Rip Web Clinic), and buckling of the PCL.


The ultimate goal of the treatment of a patient with a complete or partial ACL tear is achieving functional stability of the knee and prevention of osteoarthritis. Not all ACL tears warrant surgical repair. Treatment is tailored to the complexity of the coexisting injuries, the future athletic demands of the patient, objective signs of instability by physical exam, and subjective symptoms of instability by the patient. In order to more closely reconstruct the functional biomechanics of the ACL, double bundle ligament reconstructions are being performed with increasing frequency. The single bundle technique, which has been widely used, is effective at reducing or eliminating anterior translation of the tibia by reproducing the function of the anteromedial bundle. The goal of the double ligament ACL reconstruction technique is not only to reproduce the anteromedial bundle function, but to also reproduce the posterolateral bundle function, thus improving rotational stability.8 Relatively early data suggests an improved outcome using this method.9


Partial tears of the ACL are a common injury. A significant percentage of partial tears will progress to a functionally complete ACL tear. MRI helps guide the treatment decision process by demonstrating the extent of the ACL injury and by demonstrating co-existing injuries of the ligaments, menisci, articular cartilage and capsule.


1 Noyes FR, Mooar LA, Moorman CT 3rd, McGinniss GH: Partial tears of the anterior cruciate ligament. Progression to complete ligament deficiency. J Bone Joint Surg Br 1989 Nov; 71(5): 825-33.

2 Dienst M, Burks RT, Greis PE. Anatomy and biomechanics of the anterior cruciate ligament. Orthop Clin North Am. 2002;33(4):605-20.

3 Lintner DM, Kamaric E, Moseley JB, Noble PC: Partial tears of the anterior cruciate ligament. Are they clinically detectable” Am J Sports Med 1995 Jan-Feb; 23(1): 111-8.

4 Vincken PW, ter Braak BP, van Erkell AR, et al: Effectiveness of MR imaging in selection of patients for arthroscopy of the knee. Radiology 2002 Jun; 223(3): 739-46.

5 Roychowdhury S, Fitzgerald SW, Sonin AH, et al: Using MR imaging to diagnose partial tears of the anterior cruciate ligament: value of axial images. AJR Am J Roentgenol 1997 Jun; 168(6): 1487-91.

6 Zeiss J, Paley K, Murray K, Saddemi SR: Comparison of bone contusion seen by MRI in partial and complete tears of the anterior cruciate ligament. J Comput Assist Tomogr 1995 Sep-Oct; 19(5): 773-6.

7 Snearly WN, Kaplan PA, Dussault RG: Lateral-compartment bone contusions in adolescents with intact anterior cruciate ligaments. Radiology 1996 Jan; 198(1): 205-8.

8 Yagi M, Wong EK, Kanamori A, Debski RE, Fu FH, Woo SL. Biomechanical analysis of an anatomic anterior cruciate ligament reconstruction. Am J Sports Med. 2002;30(5):660-666.

9 ACL Study Group 2004.

Network core switches – what they are, what they are for and what they look like / Sudo Null IT News This is understandable, because a large number of tasks require a diverse fleet of equipment: access points, access level switches, firewalls, and so on.

In the case of a corporate IT infrastructure, all of these components operate on the “lower floors”, providing access to users and end devices to the network.

But little is said about the network core level. The reason is quite understandable – there are fewer large organizations than small ones, so there are also fewer large corporate networks. Let’s try to fill this gap. First, let’s talk about the general features and then move on to specific models (description and use cases). In addition to general principles, we will analyze specific models by cogs (including literally – with a screwdriver) to see what works and how.

Let’s try to crack this nut of knowledge to get to the core.


– the corporate network can be conditionally divided into three levels:

  • Access level – designed to connect client devices.
  • The aggregation/distribution layer, which, as the name implies, is intermediate and serves for preliminary traffic control.
  • Network core level.

Figure 1. Corporate network layers

The core switches are located at the very center of the corporate network and provide general switching (and if necessary, routing) that connects all other segments.

Of course, each level cannot be considered separately from the previous one.

The general increase in traffic at the access layer leads to an additional load on the switches of the distribution layer, which ultimately affects the core load. Of course, situations are possible when a burst of traffic occurs locally within one segment (within the limits of one switch of the aggregation level or even the access level). But if there is a general trend towards an increase in traffic and transmitted volumes, this still leads to an increase in the load on the network core.

Therefore, it is important to consider not only current needs, but also what awaits in the future.

Features of the load in the kernel are very different from the load at the access layer. If the access level switch is tied to the work of users (who may simply not be in the office), then the core switch will have data exchange traffic between servers, storage systems, cloud systems for backup, etc. Therefore, core switches are not necessarily the fastest, but certainly the most reliable, designed for long-term loading

An important nuance – the kernel level is the most critical to downtime when performing technical work. Turning off and replacing one core level switches leads to inactivity of a large number of participants in the network exchange. Therefore, the desire to reduce the number and duration of such “stops” is quite understandable. To do this, it is necessary: ​​firstly, to choose the optimal architecture of the future network, and secondly, to select the most suitable core switches.

Note . Given the massive nature of purchases, especially when deploying a network from scratch, an error in choosing switches for the distribution / aggregation level and even the access level can also result in significant financial losses. And although the scale of the “catastrophe” is usually estimated by the number of idle nodes for a selected period of time, the choice of equipment for the “junior” levels should be approached no less responsibly.

Features of core switches

As mentioned above, in the IT infrastructure of a corporate network, core level switches are the central link that unites other segments (usually the aggregation/distribution level, less often the access level). Most of all traffic between clients, servers, the Internet, and so on passes through the core.

Therefore, the main “skill” of the network core is not to crash at maximum load. This layer always consists of high-speed switches and routers, efficient and fault-tolerant. An important role is played by hardware, including the characteristics of the switching matrix, the performance of the processor or controller.

Note . “ Generalists vs Narrow Pros” There is an opinion that for high-speed traffic transfer, core switches should not perform any packet manipulation, such as routing between VLANs, ACL (Access Control List) and so on – in such an architecture, all these functions are assigned to the aggregation/access level switches. However, it is far from always possible to build an ideal infrastructure and meet the allocated budget. Often, in practice, a certain mixed version is used, in which the kernel level and the aggregation / access level are some kind of common level.0005 core + distribution “. Of course, from the point of view of classical architecture, this looks like a flagrant deviation from the rules, but from the financial side, it is quite reasonable.

And now, in short, simple and understandable words

Simply put, core level switches are very reliable high-performance L3 or L2 + switches that can perform certain tasks, but the main thing is stable traffic transmission. Below we will elaborate on some of the nuances.


As mentioned above, packet forwarding speed and switching capacity are important characteristics for a core switch in enterprise networks. The kernel must provide the required speed and throughput.

The good news is that traffic is not taken from nowhere. That is, knowing who, what and how much you are going to connect to the network and what “thickness” the external channels will be, you can predict the upper and lower figures for the network core load. And then it’s up to the choice of equipment.

Of course, corporate life sometimes throws up surprises, like the birth of new business units from the ground up or the building of new segments like private clouds. Therefore, reserving from 20 to 35% of the performance margin “for growth” and the same reserve in terms of the number of ports for the network core is not a bad idea at all. As mentioned above, justifying a stop or temporary slowdown in the operation of almost the entire corporate network in order to replace the switch in the core is another task.

Hardware reliability

Kernel design places more emphasis on redundancy than other layers. Everything seems to be clear: why and why, but let’s look in more detail.

As mentioned above, the load on the core level switches has a different character than the access level. Accordingly, the temperature effect is also higher, and most importantly, it stays at the same mark. And this should be taken into account when designing the cooling system.

Another important nuance is the power supply. Having two power supplies is not a luxury, but a necessity. Of course, you can use additional “cunning” external modules ATS (Automatic Transfer of Reserve) or SmartPDU, which allow you to switch the power supply to the backup line, even if the device itself has one power supply. But what happens to the network core if the only power supply inside the switch fails? Does it need to be checked?

With a second power supply, when one of them fails, the other immediately takes over all power supply functions. That is, the standard scheme: Active-Passive is quite useful.

Much depends on the manufacturer of the power supply and the element base. If everything inside is made out of nowhere and it’s not clear how, it’s probably not worth installing such equipment at all, and even more so in the core of the network.

Attack and peak resilience

Since the core switches are the hub of the network, they must not only be able to quickly forward Ethernet frames, but also have advanced DDoS protection using layer 2 and 3 protocols. And this is not only “ evil hackers. A crookedly working network application can “make a rustle” no less than the “dark knights of the keyboard.”

In addition to protection against attacks, the ability to handle peak loads is an important feature in itself. It is generally advised to avoid such configurations as meticulous access lists and packet filtering, especially against the backdrop of performance degradation. But in any case, the power reserve does not hurt.

Stack and zoom. Channel aggregation.

Of course, few people are happy with the situation when a large segment, or even the entire corporate network, does not work due to a problem with the central switch. To avoid situations when a single device combines a large number of connections and in the event of a failure nothing can take over its functions, they use redundancy and combine network equipment into a stack.

A stack is a connection of several physical switches into one “super switch” when the fault-tolerant circuit continues to work when one of the physical devices goes out.

However, on fault tolerance alone, the light did not converge like a wedge. Sooner or later, the network will grow and there will be a shortage of computing resources and free ports. Even if at the beginning switches were purchased with a good supply of ports and power, sooner or later you will still have to upgrade. The switch stack gives us the ability to add new devices to the kernel without decommissioning the old ones.

For example, the XGS4600 series supports a stack of up to 4 switches, and the XGS3700 supports a stack of up to 8. Simply put, if you have, say, two XGS4600-52F switches in the core, you can double their number, bringing their number to 4, without interrupting work networks.

It also seems to be useful to use failover protocols such as VRRP to build a failover routing scheme.

It is extremely important that other participants in the network exchange do not lose communication with the core. To do this, link aggregation is used, when several physical ports on the aggregation / distribution level switch are combined into a common UPLink and connected to two ports on the core level switches. Thus, if the connection is interrupted on one of the ports, the connection is still not lost.


“Quality of Service” (QoS) is an important feature that allows you to ensure the stable passage of certain types of traffic. For example, modern enterprises require video conferencing. Such traffic requires continuous transmission of voice and video data, unlike, for example, browsing text pages in html format. Another example is backup, when the data is in a dense stream and it is necessary to have time to transfer everything in a short “backup window”. In such cases, the use of a priority system and bandwidth limitation helps. That is, QoS.

QoS makes it possible for core switches to provide different bandwidth to different applications according to performance. Compared to traffic that is less demanding on bandwidth and time delays (such as email), critical traffic will receive a higher priority, and will be transmitted at a high speed and guaranteed low packet loss.


To describe the basic principles of working with network core switches, the well-known proverb is very suitable: “Does it work? Do not touch!”.

But there are situations when you need to touch, for example, when upgrading the entire network, connecting additional segments, and so on.

And, of course, it is necessary to receive data on the operation of network equipment in time.

Therefore, the core switches support a variety of monitoring and management methods ranging from SNMP to console connection.

It is also useful to have a dedicated management port (not shared with data) that can be connected to a separate VLAN or even a switch. In addition to improving security, this allows you to streamline the network architecture and maintain control even when traffic increases dramatically through the core.

Below are descriptions and performance characteristics of specific models from Zyxel. If you don’t like it when the manufacturer describes the specifications and capabilities of their own devices in their own blog and consider this “solid advertising”, you can immediately go to the next section: “Summarizing and recommendations”.

Let’s look at specific models

As an example, we have chosen a line of switches designed for core and aggregation/distribution levels. Where does this dual purpose come from? It all depends on the goals and objectives, primarily on the architecture of the corporate network. There are situations when the aggregation/distribution level switches are under a load comparable to the level of the network core. For example, if routing between VLANs, access lists (ACLs), traffic filtering, and so on are actively used.

Power reserve and a wide range of options in any case will not hurt.

Which models are we talking about?

Today, the XGS4600 line includes 3 switches: XGS4600-32, XGS4600-32F, XGS4600-52F. The main difference between them is in the number and design of the ports. Below is a table that highlights the main differences and common points.

Characteristic XGS4600–32 XGS4600-32F XGS4600-52F
Total ports 32 32 52
Gigabit SFP 24 48
100/1000 Mbps 24
Gigabit combo (SFP/RJ‑45) 4 4
10-Gigabit SFP+ 4 4 4
Switching capacity (Gbps) 136 136 176
Packet forwarding rate (Mpps) 101. 1 101.1 130.9
Packet buffer (bytes) 4 MB 4 MB 4 MB
MAC address table 32 KB 32 KB 32 KB
Transfer table L3 Max. 8K IPv4 entries; Max. 4K IPv6 entries Max. 8K IPv4 entries; Max. 4K IPv6 entries Max. 8K IPv4 entries; Max. 4K IPv6 entries
Routing table 12 thousand 12 thousand 12 thousand
Number of IP interfaces 256 256 256
Flash/RAM 64 MB / 1 GB 64 MB / 1 GB 64 MB / 1 GB

Below we briefly describe why these switches are suitable for use as the core of a network.

Stack and High Availability

Up to 4 switches can be physically stacked using one or two 10-Gigabit SFP+ slots. Dynamic routing is also supported to facilitate communication between subnets. This feature is very useful for large hotels, universities and other companies that use complex network infrastructure. For XGS4600 series switches, you can purchase an additional license that supports OSPFv3 and RIPng protocols for IPv6 dynamic routing.


XGS4600 Series is equipped with gigabit ports and four integrated 10-Gigabit SFP+ slots.

Other Security Measures

In addition to stacking, the series switches store two configuration files and two firmware images. This is a kind of protection against accidental failures. Imagine that the uploaded firmware file was corrupted while being transmitted over the network. The presence of the second file allows you to solve this problem “without unnecessary blood” by simply rebooting the device with a working firmware.

Approximately the same recovery algorithm if the configuration changes turned out to be “incompatible with life”. We just load another file – and the trick is in the bag.

Power Scheme – Two Independent Units


XGS4600 Series supports Active-Standby power redundancy. If the main power supply fails, the switch will operate from the backup power supply.

The power supplies themselves are from the renowned manufacturer DELTA Electronics.

What about the hardware?

  • The central node is the processor (CPU) – 1GHz ARM cortex-A9.
  • Switch controller – BCM56340.
  • RAM – 1GB.
  • Flash 64MB.

Of course, it is better to see once than to hear a hundred times (and even better to touch with your own hands). And right in the office we opened two models to see what was inside.

Below are some photos taken right at the Zyxel Russia office.

Interesting information. Zyxel does not try to “catch” its customers on trifles. “Cunning” seals, mild steel bolts (in which the slots are damaged even with the most careful unscrewing), holographic stickers and other “tricks” in order to deprive the consumer of warranty service – all this is not necessary.

Figure 2 XGS4600 Series Switches Front View: Top – XGS4600-32F, Bottom XGS4600-32

Figure 3 XGS4600 Series Switches Rear View: Top – XGS4600 -32F, bottom XGS4600-32.

All core models have two power supplies.

Figure 4. Inside the XGS4600-32 switch.

Proper layout and careful mounting of boards and connectors is very important. The manufacturer should not have a desire to “shove the unpushable” into a small case.

There are powerful heatsinks and a block of three fans. It is important for network core switches to have good cooling.

Figure 5 XGS4600-32 Switch – Power Supplies.

Figure 6 XGS4600-32 switch. Fragment of the motherboard with memory chips.

Figure 7. Close up.

Figure 8. Inside the XGS4600-32F.

Figure 9. XGS4600-32F switch power supply.

Figure 10. On the right side are the UPLINK, the MGMT port for managing the switch, and the console port.

Notice the dedicated management port (OOB) – it is shown as MGMT on the panel. Unlike the console RS-232 (which is also available), this port is intended for remote control of the device over the network.

There is also a switch number indicator in the stack – Stack ID.

Various functions

As mentioned above, despite the fact that the main task of the kernel is stable operation under load, from time to time it becomes necessary to manage traffic, and this requires certain tools.

For example, VLAN support as well as QoS and access lists are pretty useful features.

A complete list of features can be found here.

Summing up and recommendations

It is impossible to grasp the immensity, so our story about core switches is coming to an end.

There are many factors that determine which core switches are best suited for the network core in any given case. However, there are some general recommendations that should be followed to avoid long network infrastructure outages.

In addition to the “bare theory”, we have shown how these features look like on the example of a specific implementation. The described principles are suitable for evaluating any other network core level switches. We hope this will help in the development of new projects and the modernization of existing ones.

Useful links

  1. Telegram chat Zyxel
  2. Zyxel Hardware Forum
  3. A lot of useful videos on the Youtube channel
  4. L2, L2+ and L3 switches – what, when, where, where, how, why and why?
  5. Zyxel L3 XGS4600 Series Switches
  6. Building a network infrastructure based on Nebula. Part 1 – Challenges and Solutions
  7. Building a network infrastructure based on Nebula. Part 2 – Network Example
  8. Features of the use of managed and unmanaged switches
  9. How SFP, SFP+ and XFP make our lives easier

Hatred and PDF — djvnsk


Posted by


No comments yet

Disclaimer: the post is inspired by Lurkov’s articles and is entertaining, it is not obliged to please anyone, it expresses only the imaginary point of view of the author.

The PDF looks like an empty-headed high school girl and is almost as fresh and rosy, only 1993. Further text seems to hint at parallels between patients.


Attractiveness and aesthetics

PDF is originally intended for the presentation of printing in electronic form.

Useless in everyday life

For some reason it is used to store 95% of documentation. At the same time, it is absolutely not adapted for different screens (sizes and aspect ratios), which is passed off as “displayed exactly in the form in which it was conceived and created by the author” (but we know how it is done). If this is not enough for you, add bloody tears from fonts, especially for owners of is not the same as , the world’s best paid operating system for housewives and designers.

Imaginary security and internal complexity

The words “vulnerability” and “another one” have long been stuck to the news about PDF.

Wants your money

For editing, for the width of the Internet channel, because PDF files weigh more than documents in other formats, for the power of your PeC, because the official reader takes so long to start up and consumes so many resources that it seems like it suffers from bulimia and constipation at the same time.

Well, you understand

PDF became an open format in 2008, and this, for a minute, at 15 years old.

Like big guys Thermorectally promoted in business (including at the expense of manufacturers of printing and copying equipment) in order to suppress competitors. After slipping all sorts of whistles and fakes into it, it has a functionality comparable to the skills of a goldfish. The main feature is the ability to prohibit users from doing anything.

Helps sell

It helps to sell something else, for example, paid editors and iPads (now also Pro versions!), because it is possible to read comfortably with a cup of coffee and a pie a little less than anywhere else, except for the mentioned lukhuri devices.

In the world

In the English Internet you will find 9000+ pictures with memes about PDF. And it’s not from a good life!

The origin of the disease

Kmk, thanks to consumers, magazines mutated from periodicals into premium luxury advertising catalogs, ceased to be carriers of useful information, even graphic content did not save them. Then, along with documentation and booklets, all this moved to electronic form … and then merchants joined.

Again, let’s go back to the first paragraph of “characteristics”: PDF was originally designed to represent printing in electronic form. And the presentation of information in electronic form has long evolved and moved away from paper progenitors. After scratching our empirical data, we can conclude (regarding most of the existing information) that in terms of the structure of information presentation, we have hypertext instead of linear text (Wikipedia reports that the term “hypertext” in relation to information technology was introduced by Theodore Nelson in the distant 1965).