- Email: [email protected]
Dynamic magnetic resonance imaging
Clin Sports Med 21 (2002) 403 – 415
Dynamic magnetic resonance imaging Dariusz Witonski, MD, PhD, DSc Department of Orthopedics, University of £o´dz´ School of Medicine, Drewnowska 75, 91-002 £o´dz´, Poland
Abnormalities of the patellofemoral joint function related to incongruency between the femoral trochlear groove and the patella frequently are a cause of patellofemoral problems. The clear presentation of patellofemoral maltracking and a type of incongruity are important to understanding a patient’s symptoms in light of observed patellar motion.
Anatomic conditions During active motion of the knee joint, the extensor mechanism is responsible for the stability of the patellofemoral articulation. Dynamic and static forces influence the patella, controlling its appropriate motion. The studies of Hirokawa [1] reveal that the patella undergoes rotation by approximately 70° in the sagittal plane, turns by 15° in the horizontal plane, and slopes 10° in the frontal plane when the knee is flexed within the range of 20° to 160°. Cadaver experiments by Buff et al [2] determined the ratio describing the relationship between the tension of the aponeurosis of the quadriceps femoris muscles and the patellar ligament. The value of ratio differs from one and increases with the degree of knee flexion. Similar observations, based on a mathematic model, were made by Hehne [3] and Van Eijden et al [4]. Reider et al [5] considered the width of the anatomic constituents of the quadriceps femoris in relation to the shape of patella and the associated forces in the assessment of extensor mechanism of the knee. Their study did not reveal any statistically significant difference between the width of the vastus lateralis and medialis. It suggested proportionality of measurements and had an effect on the muscle system as a whole. In studies on amputated limbs, Lieb and Perry [6] observed subluxation of the patella only when the vastus lateralis was loaded; this was especially apparent during knee extension from 90° to 60°. The subluxation is minimal
E-mail address: [email protected] (D. Witonski). 0278-5919/02/$ – see front matter D 2002, Elsevier Science (USA). All rights reserved. PII: S 0 2 7 8 - 5 9 1 9 ( 0 2 ) 0 0 0 3 0 - 3
404
D. Witonski / Clin Sports Med 21 (2002) 403–415
when the applied forces affect the long or oblique part of the vastus medialis. The activity of the oblique part of the vastus medialis balances the effect of the the vastus lateralis, resulting in subluxation. The disturbance in the patellar tracking is minor or does not occur with the isolated load of the remaining constituents of the quadriceps femoris muscle. At full extension, the patella is completely above the femoral articular surface for a full range of loads. At 30° of flexion, the inferior aspect of the patella contacts the upper part of the femoral condyles. The contact between the lateral femoral condyle and the lateral patellar facet is established, but by 30° is evenly distributed on both sides. At this part of motion, patellar alignment strongly depends on the quadriceps mechanical mechanism. In experimental studies on cadaver knees, Hungerford and Barry [7] found that patellofemoral contact was made between 10° to 20° of flexion along the inferior margin of the patella in a continuous band across medial and lateral facets. The contact area increases with increasing flexion, starting either at 20° or 30° of flexion, and increasing to 90° of flexion, patellar articular facets are brought into contact with the femoral condyles. Tension on the quadriceps produces a lateral displacement of the patella on the frontal plane. This lateral displacement is resisted by the oblique fibers of the vastus medialis obliquus, by the medial retinaculum, and by the lateral facet of the trochlea. During the first 20° of flexion, the tibia derotates, and this decreases the Q angle (the angle formed by a line from the anterior superior iliac spine of the pelvis to the center of the patella and from the center of the patella to the center of the tibial tubercle). Because of the Q angle, the patella enters the trochlea from the lateral side. The patellar tracking alignment, during the first 30° of knee flexion, when the patella lies over the trochlear groove, depends only on the balance of quadriceps muscle components [8– 13]. Extensor mechanism dysfunction does not compensate for the static force of the patellar ligament, which peripheral insertion moves laterally along with tuberositas tibiae as a result of external tibial rotation at the phase of extreme extension of the knee. A study of cadaver knee specimens by Nagamine et al [14] revealed that the patella may be unstable in the first phase of joint flexion, when the tibia is rotated and the degree of articular contrast is low.
Patellofemoral joint imaging Although there is great variability in measurement techniques of patellar tracking, patellofemoral articulation remains a great enigma. Gross malalignment and luxated patellae are easily diagnosed, but more subtle malalignment requires more precise expertise. The diagnosis may be difficult, as the correlation between clinical signs of patellar tracking and traditional axial radiographic images of the patellofemoral joint is poor. Axial radiographic evaluation requires flexion of at least 20° to 30°, which is difficult in practice [15]; therefore, the routine radiograms are not helpful in the evaluation of patients with patellofemoral problems [16,17]. Since 1921, when Settegast [18] introduced the radiographic
D. Witonski / Clin Sports Med 21 (2002) 403–415
405
method of visualizing the patellofemoral joint, many modifications and techniques have been introduced [19 – 22]. Patellofemoral instability is a dynamic abnormality of many origins: femurtibia alignment, bony structures, articular surface geometry (patella and femoral trochlear groove), and static (the medial and lateral retinaculum, patellar tendon) and dynamic stabilizers (quadriceps muscle). All efforts in diagnostic modalities are concentrated on developing a technique for evaluating the patellofemoral joint motion in its critical range from 30° of knee flexion to full extension. The introduction of computed tomography (CT) and MRI for the diagnosis of the knee joint abnormalities enables assessment of the patellar position. The first MRI study to estimate the normal position of the patella was reported by Shellock et al [22]. That study showed that the patella, usually centralized in the trochlear groove from full extension to 32° of flexion, in one healthy subject lateralized during full extension, but became centralized at 5° of flexion. Several other studies, using MRI or CT, report the same finding [23 –25]. DelgadoMartins [16] states that the patella lies laterally and is incongruent in contact with the lateral femoral condyle. Schutzer et al [21] reported that asymptomatic subjects may present with minor lateralization of the patella during extension (average congruence angle, 3°) and become negative in 5° of flexion of the knee, which is compatible with control group of patellofemoral mechanics. Pinar et al [25] stated that the mean congruence angle was positive in extension and in 10° of flexion and became negative at 20°. Witon´ski and Go´raj [26] showed that, in healthy subjects, the patella moves medially from the extension as the knee is flexed. The mean congruence angle is negative in extension and remains increasingly negative as the knee is flexed to 30°, according to the radiographic data of Merchant et al [20] and Aglietti et al [27]. At 45° of knee flexion, the mean congruence angle is 6° and 8°, respectively. This contrasts with Pinar et al’s [25] study, which revealed that a congruence angle as high as + 39° should be considered normal in full extension. Grelsamer et al [28], in the largest MRI study of patients with normal extensor mechanism, revealed that the normal patella is centered over the trochlea. The measurements are made with the quadriceps muscle relaxed. Schutzer et al [21,29] found that the normal patellar tilt angles are in the high-positive range and constant between 0° and 30° flexion, but are not less than 8° at any position. Martinez et al [30] found similar results between 0° and 45° flexion. Kujala et al’s [24] studies revealed, however, that normal knees are not congruent at less than 30° of flexion. The effect of thigh muscle contraction on the position of normal patella is still a matter of controversy. Some investigators find no difference in the mean value of lateral patellar displacement with or without muscle contraction in extension [24,26]. Also, in normal individuals, contraction of the thigh muscles does not influence patellar position at 30° flexion. Emphasis has been placed on dynamic visualization of patellar motion to detect an abnormal tracking pattern using dynamic CT and MRI. MRI offers the additional advantage of visualization of the surrounding structures of the patellofemoral joint, such as medial and lateral retinaculum, patellar tendon,
406
D. Witonski / Clin Sports Med 21 (2002) 403–415
quadriceps muscle, and cartilage—not only bone, which plays an important role in patellofemoral tracking. Sta¨ubli et al [31] evaluated axial MR arthrotomograms of the knee joint from 30 patients, revealing differences in geometry of the surface of the articular cartilage and corresponding subchondral osseous contours of the patellofemoral articulation. The surface geometry of the cartilage matched the osseous contour of the patella in only four of the 30 knees and of the femoral groove in seven knees. Furthermore, MRI is useful in internal derangement assessment. Conventional spin-echo proton density and T2-weighted images demonstrate the cruciate ligaments, menisci, and collateral ligaments. Fat-suppression techniques facilitate detection and the extent of abnormal signals from intramedullary processes, such as hemorrhage, edema or infiltrative diseases. Other pulse sequences, such as gradient-echo imaging, are used for detection of cartilage abnormalities. An ultrafast spoiled gradientrecalled acquisition in the steady state (GRASS) pulse sequence is used for examining the patellofemoral articulation during active movement. A 1.5-T/64MHz MR system is used for this purpose, with the following imaging parameters, according to Shellock et al [32]: imaging plane, axial; repetition time (TR), 6.5 milliseconds; echo time (TE), 2.1 milliseconds; number of excitations, two; matrix, 160 128; field of view, 38 cm; section thickness, 7 mm; and acquisition time, 8 seconds for six images at a single section location. The hardware and software upgrade of the 1.5-T/64-MHz MR system is required for ultrafast GRASS imaging. A load of 30 ft-lb/second is usually applied for dynamic examination of the patellofemoral joint [33]. A special nonferromagnetic positioning device is used for this purpose. With dynamic scanning, axial images are obtained through active flexion, and computer software is used to construct a cine loop.
Dynamic versus static MRI During motion, the patellar position depends on patellofemoral bone –articular surface congruency, passive ligaments restraints, and active muscular action. Static MRI allows excellent visualization of patellofemoral articulation, cartilage abnormalities, and soft-tissue structures in static condition. Static images through the patellofemoral joint in different degrees of flexion reveal only the degree of patellar tilt or subluxation, parameters that can be measured also on the axial view of conventional radiography. The accuracy of patellar position on static axial MRI is limited by the absence of muscle contraction, movement, and loading. From diagnostic and therapeutic points of view, kinematic assessment during dynamic movement of the patellofemoral joint is the most important. Dynamic axial images of patellofemoral articulation can demonstrate the degree of flexion where patellar malalignment is maximal and assess whether or not it reduces. They also offer a view of the condition of passive stabilizers (ligaments, retinaculum, and patellar tendon) during kinematic action of its dynamic counterparts (muscles).
D. Witonski / Clin Sports Med 21 (2002) 403–415
407
Dynamic MRI—review of the literature In 1988, Shellock et al [22] described the kinematic MRI of the patellofemoral joint for assessing tracking abnormalities. The term ‘‘kinematic’’ was used to describe the motion of the body without reference to force or mass. Software provided with the magnetic resonance imager was used to produce a cine loop format. Images with the best visualization of the trochlear groove were selected and displayed. The speed of display was controlled by computer. According to Shellock et al [34], the following criteria have been used to describe patella position: Normal: The apex of the patella aligns with the femoral trochlear groove; there is no transverse deviation of the patella during flexion of the knee joint from 5° to 30°. Lateral subluxation: The apex of the patella is laterally displaced relative to the femoral trochlear groove or the centermost part of the trochlea, and the lateral facet of the patella overlaps the lateral aspect of the femoral trochlea. Excessive lateral pressure syndrome: The lateral facet of the patella tilts toward the lateral aspect of the femoral trochlea with little or no lateral subluxation of the patella. Medial subluxation: The apex of the patella is medially displaced relative to the femoral trochlear groove or the centermost part of the femoral trochlea.
Fig. 1. Axial fast spin-echo proton density (T1-weighted image) dynamic MRI at 0°, 10°, 20°, and 30° of knee flexion. Left patella tilts in all knee positions.
408
D. Witonski / Clin Sports Med 21 (2002) 403–415
Lateral-to-medial subluxation: A combination of the descriptions of the previous criteria, with the patella starting in a laterally-subluxed position and moving to a medially-subluxed position during increasing degrees of flexion. Witonski and Go´raj [26] observed patellar malalignment of this type in patients with anterior knee pain syndrome, analyzed with kinematic and dynamic axial MRI of the patellofemoral joint (Figs. 1– 3). In 1991, Shellock et al [32] introduced ultrafast spoiled GRASS MRI for patellofemoral joint evaluation during active flexion. The study was of five healthy subjects and seven patients with a clinical diagnosis of patellofemoral joint abnormality. Examination of the patellofemoral joint during active movement is important from a clinical point of view, because the contribution of activated muscles and soft-tissues structures around the knee is the main point of interest. In 1993, Brossmann et al [35] described the patellar tracking evaluation during active and passive extension with motion-triggered cine MRI. Cine MRI was compared with static MR images obtained during incremental extension of the knee. The investigators evaluated the same imaging parameters, such as patellar tilt angle, bisect offset, and lateral patellar displacement. The comparison between patients and healthy subjects yielded statistically significant differences for all parameters in actively-extended knees, but not in passively-extended knees. These
Fig. 2. Axial kinematic and dynamic MRI at 0°, 10°, 20°, and 30° of joint flexion. Left patella tilts and lateralizes; the congruence angle increases and patellar tilt angle decreases with progressive flexion.
D. Witonski / Clin Sports Med 21 (2002) 403–415
409
Fig. 3. T1-weighted axial kinematic and dynamic MRI obtained at the same section location at 0°, 10°, 20°, and 30° of knee flexion. Right patella lateralizes in extension to 10° of flexion and medializes at 20° to 30° of flexion.
studies reveal that kinematic examination of the patellofemoral joint during active movement, using fast gradient-echo or Hyperscan MRI, offers some important diagnostic advantages over the passive positioning methods. Images were obtained as the patellofemoral joint moved from flexion to extension, however, without the application of resistance. In 1996, Muhle et al [36] and Niitsu et al [37] demonstrated that dynamic MRI can be useful for studying cruciate ligaments injuries, some types of meniscal injuries, all patellofemoral conditions, and, most of all, cruciate ligament reconstruction. Vedi et al [38] revealed the usefulness of dynamic MRI in vivo study on meniscal movement in the normal knee under load. The most significant differences between weight-bearing and nonweight-bearing are the movement and vertical height of the anterior horn of the lateral meniscus. Bellelli and Nardis [39] used an open magnet MR unit for functional dynamic knee examination and acquired images at 0° to 120° degrees of flexion that permitted precise studies of knee flexion– extension. In 1993, Shellock et al [33] described the identification of patellofemoral joint abnormalities in active movement, comparing unloaded with loaded kinematic MRI technique. The active movement, loaded kinematic MRI examination, was performed with a special device that incorporates a mechanism that applies the resistance to patellofemoral joint when the joint moves through the range of motion from
410
D. Witonski / Clin Sports Med 21 (2002) 403–415
approximately 45° to extension. A force of 30 ft-lb/second was applied, which sufficiently stresses the patellofemopral joint of adult patients. Six of 23 symptomatic patellofemoral joints with normal patellar alignment in unloaded technique revealed an abnormality with the loaded technique. Powers et al [40] use dynamic MRI for quantification of patellar tracking. A nonferromagnetic device enables active, bilateral knee extension against the resistance (15% body weight) from 45° knee flexion to full extension. Patients were instructed to extend the knees at the rate of 9° per second, and the images at 45°, 36°, 27°, 18°, 9° and 0° were obtained. All images were assessed for medial – lateral patellar displacement, patellar tilt, and sulcus angle. Normal patellar movement was characterized by medial movement from 45° to 18°, followed by reversal toward lateral displacement from 18° to full extension. The results of patellar tilt revealed some tendency for lateral tilt decrease in the course of knee extension. Sulcus angle measurements indicated that patella was moving to a more shallow portion of the trochlear groove during extension. In 1999, Sheehan et al [41,42] published the results of a three-dimensional (3D) study of patellar-femoral-tibial joint kinematics using cine phase-contrast MRI (cine PC MRI) during leg extension, resisted by a 34 N weight. This technique was used primarily to measure heart motion and blood flow and has been adapted to study musculoskeletal system. The investigators found that the patella displaced laterally, superiorly, and anteriorly with the knee extended. The patellar flexion lagged knee flexion, patellar tilt was variable, and patellar rotation was constant throughout extension.
Contraindications to MRI MRI, as with other sophisticated diagnostic modalities, has several contraindications, including cardiac pacemakers, some types of cardiac prosthetic valves, intraorbital metallic foreign bodies, cochlear implants, metallic cerebral aneurysm clips, dorsal column stimulators, and bone-growth stimulators. Limitations to conventional MRI include metallic orthopaedic hardware, such as joint prosthesis and metallic plates and screws. Claustrophobic or uncooperative patients are not good candidates for MRI examination. The safety of this modality in pregnant women has not been established definitively. The safety committee of the Society of Magnetic Resonance Imaging announced special precautions for using this nonionizing form of diagnostic modality. Although the safety of MRI in pregnancy has not been proved [43], there are no reports on teratogenic effects associated with intense magnetic field.
Arthroscopy—clinical value of patellar tracking abnormalities MRI is a noninvasive modality that has made a significant contribution to the understanding of knee injury and pathologic changes of the knee, outside and
D. Witonski / Clin Sports Med 21 (2002) 403–415
411
inside the synovial cavity. Arthroscopy of the knee, however, still is standard as the technique of choice for detecting of intra-articular changes in the knee joint. Lindberg et al [44], in 1986, described the technique for arthroscopic examination of the patellofemoral joint. The alignment of the median ridge of the patella to the deepest part of the femoral groove was studied with the knee flexed from 0° to 90°. Subluxation, which was reduced during first 10° of flexion, was regarded as normal; if it was noted between 10° and 30°, but became normal on further degrees of flexion, it was classified as first degree. Subluxation that persisted at above 30° of flexion was classified as second degree. During examination, the probe was used to palpate the cartilage for softness, fragmentation, and ulceration. Witon´ski [45] used arthroscopic dynamic assessment of the patellofemoral joint in anterior knee pain syndrome patients. Results were compared with previously performed MRI study in this group. B} ull et al [23] compared motion-triggered MRI of patellar tracking with arthroscopy and concluded that, for patellar lateralization, both modalities show significant correlation, but comparisons of the patellar tilt are not as consistent. For the determination of intraarticular lesions, motion-triggered cine-MRI was not useful. Struhl [46] used limited volume gas arthroscopy for dynamic assessment of patellofemoral pathology. Also, Sojbjerg et al [47] and Delaunay and Kapandji [48] proved the usefulness of arthroscopy in determination of the patellofemoral alignment.
Cost-effectiveness assessment of MRI Boden et al [49], based on average regional medical cost and using a mathematic formula, calculate that diagnostic arthroscopy is more cost-effective than MRI of the acutely-injured knee if no less than 78% of patients examined with MRI undergo arthroscopy. Lundberg et al [50] found low diagnostic validity of MRI for intra-articular pathology with hemarthrosis. MRI is highly sensitive and specific for diagnosis of anterior cruciate ligament lesions, with values of 86% and 92%, respectively. Diagnosis of medial and lateral meniscal tears show specificity of 66% and 84%. The sensitivity for partial or total medial collateral ligament tears is 56%, the specificity 93%. Rupture of the medial retinaculum associated with patellar dislocation is detected using MRI with 27% sensitivity. Gelb et al [51] compared clinical examination and MRI findings with arthroscopy of the knee joint. Their study showed that clinical evaluation has sensitivity and specificity of 100% for diagnosis of anterior cruciate ligament injures, whereas MRI is 95% sensitive and 88% specific. For meniscal lesions, the clinical assessment has specificity and sensitivity of 91% compared, respectively, with 87% and 82% for MRI. The authors concluded that MRI was overused in the evaluation of knee disorders and was not a cost-effective method for evaluating injuries, if compared with a skilled examiner. But the cost-effectiveness analysis of medical diagnostic devices in cases of the knee disorders or injuries is not as explicit in nature. Bui-Mansfield et al [52], on the basis of a cost of $1000 for each MRI examination and $4000 for each diagnostic arthroscopy, concluded that
412
D. Witonski / Clin Sports Med 21 (2002) 403–415
cost-effective MRI requires 25% true negativity. With MRI examinations obtained before surgery, the authors find 21 of 50 arthroscopies unnecessary, thereby producing a savings of as much as $680 per MRI examination of the knee joint. Suarez-Almazor et al [53] performed an economic evaluation based on cost-effectiveness ratios (per averted arthroscopy), including direct and indirect costs. They revealed that, in general, MRI appears to be a cost-effective diagnostic procedure for patients with internal derangement of the knee requiring arthroscopy. Similar observations are described by Ruwe et al [54,55] and Chissell et al [56]. Le Vot et al [57] state that emergency diagnostic arthroscopy of the knee can be efficiently replaced by MRI examination.
Summary Among the devices helping with an accurate diagnosis, neither MRI nor arthroscopy is perfect; both delineate pathology in the knee joint with reasonable sensitivity and specificity. MRI, as a noninvasive and nonionizing modality, has made a significant contribution to the understanding of musculoskeletal disturbances. Static images through the patellofemoral joint in different degrees of flexion reveal only the degree of patellar tilt or subluxation, parameters that can be measured also on the axial view of conventional radiography. The accuracy of patellar position on static axial MRI is limited by the absence of muscle contraction, movement, and loading. Dynamic axial images of patellofemoral articulation can demonstrate the degree of flexion where patellar malalignment is maximal and assess whether or not it reduces. Arthroscopy, aside from its diagnostic values, provides the opportunity for treatment of intra-articular changes contributing to knee joint disturbances, but it is an invasive technique with potential risks of complications. The performed cost-effectiveness analysis of MRI is based mainly on estimation of intra-articular pathology of the acutelyinjured knee [49,52,56]. There are scarce data on the cost-effectiveness of MRI of patellofemoral alignment in patellofemoral pain knees. Total examination time for active movement dynamic MRI procedure is approximately 8 to 10 minutes, thus it can be performed during routine MRI examination of the knee. In cases of suspected patellofemoral malalignment with symptoms that mimic other types of internal derangement of the knee joint, dynamic MRI can be a procedure of choice for detection of transient patellar dislocation, whereas a single clinical examination cannot differentiate from other internal knee pathologies. Dynamic MRI, although in an experimental phase, gives us a new perspective for dynamic study of the patellofemoral joint.
References [1] Hirokawa S. Three-dimensional mathematical model analysis of the patellofemoral joint. J Biomech 1991;24:659.
D. Witonski / Clin Sports Med 21 (2002) 403–415
413
[2] Buff HU, Jones LC, Hungerford DS. Experimental determination of forces transmitted through the patellofemoral joint. J Biomech 1988;21:17. [3] Hehne HF. Biomechanics of the patellofemoral joint and its clinical relevance. Clin Orthop 1990;258:73. [4] Van Eijden TMGJ, Kouwenhoven E, Verburg J, et al. A mathematical model of the patellofemoral joint. J Biomech 1986;19:219. [5] Reider B, Maeshall JL, Koslin B, et al. The anterior aspects of the knee joint. J Bone Joint Surg Am 1981;63:351. [6] Lieb FJ, Perry J. Quadriceps function. An anatomical and mechanical study using amputed limbs. J Bone Joint Surg Am 1968;50:1535. [7] Hungerford DS, Barry N. Biomechanics of the patellofemoral joint. Clin Orthop 1979;144:9. [8] Duchenne GB. Physiology of motion demonstrated by means of electrical stimulation and clinical observation and applied to study of paralysis and deformities. Philadelphia: J.B. Lippincott Co; 1949. [9] Last RJ. Some anatomical details of the knee joint. J Bone Joint Surg Br 1948;30:683. [10] Madigan R, Wissinger HA, Donaldson WF. Preliminary experience with a method of quadricepsplasty in recurrent subluxation of the patella. J Bone Joint Surg Am 1975;57:600. [11] Roberts JM. The surgical knee. Surg Clin North Am 1974;54:1313. [12] Smillie IS. Injures of the knee joint. Baltimore: Williams & Wilkins Co; 1962. [13] Steindler A. Kinesiology of the human body under normal and pathological conditions. Springfield (IL): Charles C Thomas; 1955. [14] Nagamine R, Otani T, White SE, et al. Patellar tracking measurement in the normal knee. J Orthop Res 1995;13:115. [15] Bentley G, Dowd G. Current concepts of etiology and treatment of chondromalacia patellae. Clin Orthop 1984;189:209. [16] Delgado-Martins H. A study of the position of the patella using computerized tomography. J Bone Joint Surg 1979;61B:443. [17] Ficat RP, Hungerford DS. Disorders of the patellofemoral joint. Baltimore: Williams & Wilkins; 1975. [18] Settegast L. Lower extremity and pelvis. In: Mescham I, editor. Radiographic positioning and related anatomy. Philadelphia: WB Saunders Company; 1978. [19] Laurin CA, Dussault R, Levesque HP. The tangential X-ray investigation of the patellofemoral joint: X-ray technique, diagnostic criteria and their interpretation. Clin Orthop 1979;144:16. [20] Merchant AC, Mercer RL, Jacobsen RH, et al. Roentgenographic analysis of patellofemoral congruence. J Bone Joint Surg Am 1974;56:1391. [21] Schutzer SF, Ramsby GR, Fulkerson JP. The evaluation of the patellofemoral pain using computerized tomography: a preliminary study. Clin Orthop 1986;204:286. [22] Shellock FG, Mink JH, Fox JM. Patellofemoral joint: kinematic MR imaging to assess tracking abnormalities. Radiology 1988;168:551. [23] B} ull CC, Brossmann J, Muhle C, et al. Die Untersuchung des Femoropatellargelenkes mit bewegungsgetriggerter MRT im Vergleich zur arthroskopichen Diagnostik. Arthroskopie 1993; 6:249. [24] Kujala UM, Osterman K, Kormano M, et al. Patellar motion analyzed by magnetic resonance imaging. Acta Orthop Scand 1989;60:13. [25] Pinar H, Akseki D, Genc I, et al. Kinematic and dynamic axial computerized tomography of the normal patellofemoral joint. Knee Surg Sports Traumatol Arthrosc 1994;2:27. [26] Witon´ski D, Go´raj B. Patellar motion analyzed by kinematic and dynamic axial magnetic resonance imaging in patients with anterior knee pain syndrome. Arch Orthop Trauma Surg 1999;119:46. [27] Aglietti P, Insall JN, Cerulli G. Patellar pain and incongruity. Measurements of incongruence. Clin Orthop 1983;176:217. [28] Grelsamer RP, Newton PM, Staron RB. The medial-lateral position of the patella on routine magnetic resonance imaging: when is normal not normal? Arthroscopy 1998;14:23.
414
D. Witonski / Clin Sports Med 21 (2002) 403–415
[29] Schutzer SF, Ramsby GR, Fulkerson JP. Computed tomographic classification of patellofemoral pain patients. Orthop Clin North Am 1986;17:235. [30] Martinez S, Korobkin N, Fondren FB, et al. Computed tomography of the normal patellofemoral joint. Invest Radiol 1983;18:249. [31] Sta¨ubli HU, D} urrenmatt U, Porcellini B, et al. Anatomy and surface geometry of the patellofemoral joint in the axial plane. J Bone Joint Surg Br 1999;81:452. [32] Shellock FG, Foo TKF, Deutsch AL, et al. Patellofemoral joint: evaluation during active flexion with ultrafast spoiled GRASS MR imaging. Radiology 1991;180:581. [33] Shellock FG, Mink JH, Deutsch AL, et al. Patellofemoral joint: identification of abnormalities with active-movement, ‘‘unloaded’’ versus ‘‘loaded’’ kinematic MR imaging techniques. Radiology 1993;188:575. [34] Shellock FG, Mink JH, Deutsch A, et al. Evaluation of patients with persistent symptoms after lateral retinacular release by kinematic resonance imaging of the patellofemoral joint. Arthroscopy 1990;6:226. [35] Brossmann J, Muhle C, Schroder C, et al. Patellar tracking patterns during active and passive knee extension: evaluation with motion-triggered cine MR imaging. Radiology 1993;187:205. [36] Muhle C, Brossmann J, Heller M. Kinematic MRI of the knee using a specially designed positioning device. J Comput Assist Tomogr 1996;20:522. [37] Niitsu M, Ikeda K, Fukubayashi T, et al. Knee extension and flexion: MR delineation of normal and torn anterior cruciate ligaments. J Comput Assist Tomogr 1996;20:322. [38] Vedi V, Williams A, Tennant SJ, et al. Meniscal movement. An in-vivo study using dynamic MRI. J Bone Joint Surg Br 1999;81:37. [39] Bellelli A, Nardis P. Dynamic magnetic resonance of the knee. Considerations on techniques an anatomy with magnetic resonance system with open magnet. Radiol Med (Torino) 1997;93:199. [40] Powers CM, Shellock FG, Pfaff M. Quantification of patellar tracking using kinematic MRI. J Magn Reson Imaging 1998;8:724. [41] Sheehan FT, Drace JE. Quantitative MR measures of three-dimensional patellar kinematics as a research and diagnostic tool. Med Sci Sports Exerc 1999;31:1399. [42] Sheehan FT, Zajac FE, Drace JE. In vivo tracking of the human patella using cine phase contrast magnetic resonance imaging. J Biomech Eng 1999;121:650. [43] Shellock FG, Kanal E. Policies, guidelines, and recommendations for MR imaging safety and patient management. J Magn Reson Imaging 1991;1:97. [44] Lindberg U, Hamberg P, Lysholm J, et al. Arthroscopic examination of the patellofemoral joint using a central, one-portal technique. Orthop Clin North Am 1986;17:263. [45] Witon´ski D. Arthroscopic dynamic assessment of the patello-femoral joint in patients with anterior knee pain syndrome. Chir Narz Ruchu Ortop Pol 1999;64:505. [46] Struhl S. Dynamic assessment of shoulder and patellofemoral pathology using limited volume gas arthroscopy. Arthroscopy 1998;14:546. [47] Sojbjerg JO, Lauritzen J, Hvid I, et al. Arthroscopic determination of patellofemoral malalignment. Clin Orthop 1987;215:243. [48] Delaunay C, Kapandji T. Effect of arthroscopic fluid inflow on the femoro-patellar tracking. Comparative study of the dry and fluid filled knee. Rev Chir Reparatrice Appar Mot 1998; 84:346. [49] Boden SD, Labropoulos PA, Vailas JC. MR scanning of the acutely injured knee: sensitive, but is it cost effective? Arthroscopy 1990;6:306. [50] Lundberg M, Odensten M, Thuomas KA, et al. The diagnostic validity of magnetic resonance imaging in acute knee injuries with hemarthrosis. A single-blinded evaluation in 69 patients using high-field MRI before arthroscopy. Int J Sports Med 1996;17:218. [51] Gelb HJ, Glasgow SG, Sapega AA, et al. Magnetic resonance imaging of the knee disorders. Clinical value and cost-effectiveness in sports medicine practice. Am J Sports Med 1996;24:99. [52] Bui-Mansfield LT, Youngberg RA, Warme W, et al. Potential cost savings of MR imaging obtained before arthroscopy of the knee: evaluation of 50 consecutive patients. Am J Roentgenol 1997;168:913.
D. Witonski / Clin Sports Med 21 (2002) 403–415
415
[53] Suarez-Almazor ME, Kaul P, Kendall CJ, et al. The cost-effectiveness of magnetic resonance imaging for patients with internal derangement of the knee. Int J Technol Assess Health Care 1999;15:392. [54] Ruwe PA, Wright J, Randall RL, et al. Can MR imaging effectively replace diagnostic arthroscopy? Radiology 1992;183:335. [55] Ruwe PA, McCarthy SM. Cost effectiveness of magnetic resonance imaging of the knee. Magn Reson Imaging Clin N Am 1994;2:475. [56] Chissell HR, Allum RL, Keightley A. MRI of the knee: its cost-effective use in a district general hospital. Ann R Coll Surg Engl 1994;76:26. [57] Le Vot J, Solarcroup JC, Leonetti P, et al. Correlations between the clinical test/MRI/arthroscopy in acute knee injures. J Radiol 1993;74:483.
Dynamic magnetic resonance imaging Dariusz Witonski, MD, PhD, DSc Department of Orthopedics, University of £o´dz´ School of Medicine, Drewnowska 75, 91-002 £o´dz´, Poland
Abnormalities of the patellofemoral joint function related to incongruency between the femoral trochlear groove and the patella frequently are a cause of patellofemoral problems. The clear presentation of patellofemoral maltracking and a type of incongruity are important to understanding a patient’s symptoms in light of observed patellar motion.
Anatomic conditions During active motion of the knee joint, the extensor mechanism is responsible for the stability of the patellofemoral articulation. Dynamic and static forces influence the patella, controlling its appropriate motion. The studies of Hirokawa [1] reveal that the patella undergoes rotation by approximately 70° in the sagittal plane, turns by 15° in the horizontal plane, and slopes 10° in the frontal plane when the knee is flexed within the range of 20° to 160°. Cadaver experiments by Buff et al [2] determined the ratio describing the relationship between the tension of the aponeurosis of the quadriceps femoris muscles and the patellar ligament. The value of ratio differs from one and increases with the degree of knee flexion. Similar observations, based on a mathematic model, were made by Hehne [3] and Van Eijden et al [4]. Reider et al [5] considered the width of the anatomic constituents of the quadriceps femoris in relation to the shape of patella and the associated forces in the assessment of extensor mechanism of the knee. Their study did not reveal any statistically significant difference between the width of the vastus lateralis and medialis. It suggested proportionality of measurements and had an effect on the muscle system as a whole. In studies on amputated limbs, Lieb and Perry [6] observed subluxation of the patella only when the vastus lateralis was loaded; this was especially apparent during knee extension from 90° to 60°. The subluxation is minimal
E-mail address: [email protected] (D. Witonski). 0278-5919/02/$ – see front matter D 2002, Elsevier Science (USA). All rights reserved. PII: S 0 2 7 8 - 5 9 1 9 ( 0 2 ) 0 0 0 3 0 - 3
404
D. Witonski / Clin Sports Med 21 (2002) 403–415
when the applied forces affect the long or oblique part of the vastus medialis. The activity of the oblique part of the vastus medialis balances the effect of the the vastus lateralis, resulting in subluxation. The disturbance in the patellar tracking is minor or does not occur with the isolated load of the remaining constituents of the quadriceps femoris muscle. At full extension, the patella is completely above the femoral articular surface for a full range of loads. At 30° of flexion, the inferior aspect of the patella contacts the upper part of the femoral condyles. The contact between the lateral femoral condyle and the lateral patellar facet is established, but by 30° is evenly distributed on both sides. At this part of motion, patellar alignment strongly depends on the quadriceps mechanical mechanism. In experimental studies on cadaver knees, Hungerford and Barry [7] found that patellofemoral contact was made between 10° to 20° of flexion along the inferior margin of the patella in a continuous band across medial and lateral facets. The contact area increases with increasing flexion, starting either at 20° or 30° of flexion, and increasing to 90° of flexion, patellar articular facets are brought into contact with the femoral condyles. Tension on the quadriceps produces a lateral displacement of the patella on the frontal plane. This lateral displacement is resisted by the oblique fibers of the vastus medialis obliquus, by the medial retinaculum, and by the lateral facet of the trochlea. During the first 20° of flexion, the tibia derotates, and this decreases the Q angle (the angle formed by a line from the anterior superior iliac spine of the pelvis to the center of the patella and from the center of the patella to the center of the tibial tubercle). Because of the Q angle, the patella enters the trochlea from the lateral side. The patellar tracking alignment, during the first 30° of knee flexion, when the patella lies over the trochlear groove, depends only on the balance of quadriceps muscle components [8– 13]. Extensor mechanism dysfunction does not compensate for the static force of the patellar ligament, which peripheral insertion moves laterally along with tuberositas tibiae as a result of external tibial rotation at the phase of extreme extension of the knee. A study of cadaver knee specimens by Nagamine et al [14] revealed that the patella may be unstable in the first phase of joint flexion, when the tibia is rotated and the degree of articular contrast is low.
Patellofemoral joint imaging Although there is great variability in measurement techniques of patellar tracking, patellofemoral articulation remains a great enigma. Gross malalignment and luxated patellae are easily diagnosed, but more subtle malalignment requires more precise expertise. The diagnosis may be difficult, as the correlation between clinical signs of patellar tracking and traditional axial radiographic images of the patellofemoral joint is poor. Axial radiographic evaluation requires flexion of at least 20° to 30°, which is difficult in practice [15]; therefore, the routine radiograms are not helpful in the evaluation of patients with patellofemoral problems [16,17]. Since 1921, when Settegast [18] introduced the radiographic
D. Witonski / Clin Sports Med 21 (2002) 403–415
405
method of visualizing the patellofemoral joint, many modifications and techniques have been introduced [19 – 22]. Patellofemoral instability is a dynamic abnormality of many origins: femurtibia alignment, bony structures, articular surface geometry (patella and femoral trochlear groove), and static (the medial and lateral retinaculum, patellar tendon) and dynamic stabilizers (quadriceps muscle). All efforts in diagnostic modalities are concentrated on developing a technique for evaluating the patellofemoral joint motion in its critical range from 30° of knee flexion to full extension. The introduction of computed tomography (CT) and MRI for the diagnosis of the knee joint abnormalities enables assessment of the patellar position. The first MRI study to estimate the normal position of the patella was reported by Shellock et al [22]. That study showed that the patella, usually centralized in the trochlear groove from full extension to 32° of flexion, in one healthy subject lateralized during full extension, but became centralized at 5° of flexion. Several other studies, using MRI or CT, report the same finding [23 –25]. DelgadoMartins [16] states that the patella lies laterally and is incongruent in contact with the lateral femoral condyle. Schutzer et al [21] reported that asymptomatic subjects may present with minor lateralization of the patella during extension (average congruence angle, 3°) and become negative in 5° of flexion of the knee, which is compatible with control group of patellofemoral mechanics. Pinar et al [25] stated that the mean congruence angle was positive in extension and in 10° of flexion and became negative at 20°. Witon´ski and Go´raj [26] showed that, in healthy subjects, the patella moves medially from the extension as the knee is flexed. The mean congruence angle is negative in extension and remains increasingly negative as the knee is flexed to 30°, according to the radiographic data of Merchant et al [20] and Aglietti et al [27]. At 45° of knee flexion, the mean congruence angle is 6° and 8°, respectively. This contrasts with Pinar et al’s [25] study, which revealed that a congruence angle as high as + 39° should be considered normal in full extension. Grelsamer et al [28], in the largest MRI study of patients with normal extensor mechanism, revealed that the normal patella is centered over the trochlea. The measurements are made with the quadriceps muscle relaxed. Schutzer et al [21,29] found that the normal patellar tilt angles are in the high-positive range and constant between 0° and 30° flexion, but are not less than 8° at any position. Martinez et al [30] found similar results between 0° and 45° flexion. Kujala et al’s [24] studies revealed, however, that normal knees are not congruent at less than 30° of flexion. The effect of thigh muscle contraction on the position of normal patella is still a matter of controversy. Some investigators find no difference in the mean value of lateral patellar displacement with or without muscle contraction in extension [24,26]. Also, in normal individuals, contraction of the thigh muscles does not influence patellar position at 30° flexion. Emphasis has been placed on dynamic visualization of patellar motion to detect an abnormal tracking pattern using dynamic CT and MRI. MRI offers the additional advantage of visualization of the surrounding structures of the patellofemoral joint, such as medial and lateral retinaculum, patellar tendon,
406
D. Witonski / Clin Sports Med 21 (2002) 403–415
quadriceps muscle, and cartilage—not only bone, which plays an important role in patellofemoral tracking. Sta¨ubli et al [31] evaluated axial MR arthrotomograms of the knee joint from 30 patients, revealing differences in geometry of the surface of the articular cartilage and corresponding subchondral osseous contours of the patellofemoral articulation. The surface geometry of the cartilage matched the osseous contour of the patella in only four of the 30 knees and of the femoral groove in seven knees. Furthermore, MRI is useful in internal derangement assessment. Conventional spin-echo proton density and T2-weighted images demonstrate the cruciate ligaments, menisci, and collateral ligaments. Fat-suppression techniques facilitate detection and the extent of abnormal signals from intramedullary processes, such as hemorrhage, edema or infiltrative diseases. Other pulse sequences, such as gradient-echo imaging, are used for detection of cartilage abnormalities. An ultrafast spoiled gradientrecalled acquisition in the steady state (GRASS) pulse sequence is used for examining the patellofemoral articulation during active movement. A 1.5-T/64MHz MR system is used for this purpose, with the following imaging parameters, according to Shellock et al [32]: imaging plane, axial; repetition time (TR), 6.5 milliseconds; echo time (TE), 2.1 milliseconds; number of excitations, two; matrix, 160 128; field of view, 38 cm; section thickness, 7 mm; and acquisition time, 8 seconds for six images at a single section location. The hardware and software upgrade of the 1.5-T/64-MHz MR system is required for ultrafast GRASS imaging. A load of 30 ft-lb/second is usually applied for dynamic examination of the patellofemoral joint [33]. A special nonferromagnetic positioning device is used for this purpose. With dynamic scanning, axial images are obtained through active flexion, and computer software is used to construct a cine loop.
Dynamic versus static MRI During motion, the patellar position depends on patellofemoral bone –articular surface congruency, passive ligaments restraints, and active muscular action. Static MRI allows excellent visualization of patellofemoral articulation, cartilage abnormalities, and soft-tissue structures in static condition. Static images through the patellofemoral joint in different degrees of flexion reveal only the degree of patellar tilt or subluxation, parameters that can be measured also on the axial view of conventional radiography. The accuracy of patellar position on static axial MRI is limited by the absence of muscle contraction, movement, and loading. From diagnostic and therapeutic points of view, kinematic assessment during dynamic movement of the patellofemoral joint is the most important. Dynamic axial images of patellofemoral articulation can demonstrate the degree of flexion where patellar malalignment is maximal and assess whether or not it reduces. They also offer a view of the condition of passive stabilizers (ligaments, retinaculum, and patellar tendon) during kinematic action of its dynamic counterparts (muscles).
D. Witonski / Clin Sports Med 21 (2002) 403–415
407
Dynamic MRI—review of the literature In 1988, Shellock et al [22] described the kinematic MRI of the patellofemoral joint for assessing tracking abnormalities. The term ‘‘kinematic’’ was used to describe the motion of the body without reference to force or mass. Software provided with the magnetic resonance imager was used to produce a cine loop format. Images with the best visualization of the trochlear groove were selected and displayed. The speed of display was controlled by computer. According to Shellock et al [34], the following criteria have been used to describe patella position: Normal: The apex of the patella aligns with the femoral trochlear groove; there is no transverse deviation of the patella during flexion of the knee joint from 5° to 30°. Lateral subluxation: The apex of the patella is laterally displaced relative to the femoral trochlear groove or the centermost part of the trochlea, and the lateral facet of the patella overlaps the lateral aspect of the femoral trochlea. Excessive lateral pressure syndrome: The lateral facet of the patella tilts toward the lateral aspect of the femoral trochlea with little or no lateral subluxation of the patella. Medial subluxation: The apex of the patella is medially displaced relative to the femoral trochlear groove or the centermost part of the femoral trochlea.
Fig. 1. Axial fast spin-echo proton density (T1-weighted image) dynamic MRI at 0°, 10°, 20°, and 30° of knee flexion. Left patella tilts in all knee positions.
408
D. Witonski / Clin Sports Med 21 (2002) 403–415
Lateral-to-medial subluxation: A combination of the descriptions of the previous criteria, with the patella starting in a laterally-subluxed position and moving to a medially-subluxed position during increasing degrees of flexion. Witonski and Go´raj [26] observed patellar malalignment of this type in patients with anterior knee pain syndrome, analyzed with kinematic and dynamic axial MRI of the patellofemoral joint (Figs. 1– 3). In 1991, Shellock et al [32] introduced ultrafast spoiled GRASS MRI for patellofemoral joint evaluation during active flexion. The study was of five healthy subjects and seven patients with a clinical diagnosis of patellofemoral joint abnormality. Examination of the patellofemoral joint during active movement is important from a clinical point of view, because the contribution of activated muscles and soft-tissues structures around the knee is the main point of interest. In 1993, Brossmann et al [35] described the patellar tracking evaluation during active and passive extension with motion-triggered cine MRI. Cine MRI was compared with static MR images obtained during incremental extension of the knee. The investigators evaluated the same imaging parameters, such as patellar tilt angle, bisect offset, and lateral patellar displacement. The comparison between patients and healthy subjects yielded statistically significant differences for all parameters in actively-extended knees, but not in passively-extended knees. These
Fig. 2. Axial kinematic and dynamic MRI at 0°, 10°, 20°, and 30° of joint flexion. Left patella tilts and lateralizes; the congruence angle increases and patellar tilt angle decreases with progressive flexion.
D. Witonski / Clin Sports Med 21 (2002) 403–415
409
Fig. 3. T1-weighted axial kinematic and dynamic MRI obtained at the same section location at 0°, 10°, 20°, and 30° of knee flexion. Right patella lateralizes in extension to 10° of flexion and medializes at 20° to 30° of flexion.
studies reveal that kinematic examination of the patellofemoral joint during active movement, using fast gradient-echo or Hyperscan MRI, offers some important diagnostic advantages over the passive positioning methods. Images were obtained as the patellofemoral joint moved from flexion to extension, however, without the application of resistance. In 1996, Muhle et al [36] and Niitsu et al [37] demonstrated that dynamic MRI can be useful for studying cruciate ligaments injuries, some types of meniscal injuries, all patellofemoral conditions, and, most of all, cruciate ligament reconstruction. Vedi et al [38] revealed the usefulness of dynamic MRI in vivo study on meniscal movement in the normal knee under load. The most significant differences between weight-bearing and nonweight-bearing are the movement and vertical height of the anterior horn of the lateral meniscus. Bellelli and Nardis [39] used an open magnet MR unit for functional dynamic knee examination and acquired images at 0° to 120° degrees of flexion that permitted precise studies of knee flexion– extension. In 1993, Shellock et al [33] described the identification of patellofemoral joint abnormalities in active movement, comparing unloaded with loaded kinematic MRI technique. The active movement, loaded kinematic MRI examination, was performed with a special device that incorporates a mechanism that applies the resistance to patellofemoral joint when the joint moves through the range of motion from
410
D. Witonski / Clin Sports Med 21 (2002) 403–415
approximately 45° to extension. A force of 30 ft-lb/second was applied, which sufficiently stresses the patellofemopral joint of adult patients. Six of 23 symptomatic patellofemoral joints with normal patellar alignment in unloaded technique revealed an abnormality with the loaded technique. Powers et al [40] use dynamic MRI for quantification of patellar tracking. A nonferromagnetic device enables active, bilateral knee extension against the resistance (15% body weight) from 45° knee flexion to full extension. Patients were instructed to extend the knees at the rate of 9° per second, and the images at 45°, 36°, 27°, 18°, 9° and 0° were obtained. All images were assessed for medial – lateral patellar displacement, patellar tilt, and sulcus angle. Normal patellar movement was characterized by medial movement from 45° to 18°, followed by reversal toward lateral displacement from 18° to full extension. The results of patellar tilt revealed some tendency for lateral tilt decrease in the course of knee extension. Sulcus angle measurements indicated that patella was moving to a more shallow portion of the trochlear groove during extension. In 1999, Sheehan et al [41,42] published the results of a three-dimensional (3D) study of patellar-femoral-tibial joint kinematics using cine phase-contrast MRI (cine PC MRI) during leg extension, resisted by a 34 N weight. This technique was used primarily to measure heart motion and blood flow and has been adapted to study musculoskeletal system. The investigators found that the patella displaced laterally, superiorly, and anteriorly with the knee extended. The patellar flexion lagged knee flexion, patellar tilt was variable, and patellar rotation was constant throughout extension.
Contraindications to MRI MRI, as with other sophisticated diagnostic modalities, has several contraindications, including cardiac pacemakers, some types of cardiac prosthetic valves, intraorbital metallic foreign bodies, cochlear implants, metallic cerebral aneurysm clips, dorsal column stimulators, and bone-growth stimulators. Limitations to conventional MRI include metallic orthopaedic hardware, such as joint prosthesis and metallic plates and screws. Claustrophobic or uncooperative patients are not good candidates for MRI examination. The safety of this modality in pregnant women has not been established definitively. The safety committee of the Society of Magnetic Resonance Imaging announced special precautions for using this nonionizing form of diagnostic modality. Although the safety of MRI in pregnancy has not been proved [43], there are no reports on teratogenic effects associated with intense magnetic field.
Arthroscopy—clinical value of patellar tracking abnormalities MRI is a noninvasive modality that has made a significant contribution to the understanding of knee injury and pathologic changes of the knee, outside and
D. Witonski / Clin Sports Med 21 (2002) 403–415
411
inside the synovial cavity. Arthroscopy of the knee, however, still is standard as the technique of choice for detecting of intra-articular changes in the knee joint. Lindberg et al [44], in 1986, described the technique for arthroscopic examination of the patellofemoral joint. The alignment of the median ridge of the patella to the deepest part of the femoral groove was studied with the knee flexed from 0° to 90°. Subluxation, which was reduced during first 10° of flexion, was regarded as normal; if it was noted between 10° and 30°, but became normal on further degrees of flexion, it was classified as first degree. Subluxation that persisted at above 30° of flexion was classified as second degree. During examination, the probe was used to palpate the cartilage for softness, fragmentation, and ulceration. Witon´ski [45] used arthroscopic dynamic assessment of the patellofemoral joint in anterior knee pain syndrome patients. Results were compared with previously performed MRI study in this group. B} ull et al [23] compared motion-triggered MRI of patellar tracking with arthroscopy and concluded that, for patellar lateralization, both modalities show significant correlation, but comparisons of the patellar tilt are not as consistent. For the determination of intraarticular lesions, motion-triggered cine-MRI was not useful. Struhl [46] used limited volume gas arthroscopy for dynamic assessment of patellofemoral pathology. Also, Sojbjerg et al [47] and Delaunay and Kapandji [48] proved the usefulness of arthroscopy in determination of the patellofemoral alignment.
Cost-effectiveness assessment of MRI Boden et al [49], based on average regional medical cost and using a mathematic formula, calculate that diagnostic arthroscopy is more cost-effective than MRI of the acutely-injured knee if no less than 78% of patients examined with MRI undergo arthroscopy. Lundberg et al [50] found low diagnostic validity of MRI for intra-articular pathology with hemarthrosis. MRI is highly sensitive and specific for diagnosis of anterior cruciate ligament lesions, with values of 86% and 92%, respectively. Diagnosis of medial and lateral meniscal tears show specificity of 66% and 84%. The sensitivity for partial or total medial collateral ligament tears is 56%, the specificity 93%. Rupture of the medial retinaculum associated with patellar dislocation is detected using MRI with 27% sensitivity. Gelb et al [51] compared clinical examination and MRI findings with arthroscopy of the knee joint. Their study showed that clinical evaluation has sensitivity and specificity of 100% for diagnosis of anterior cruciate ligament injures, whereas MRI is 95% sensitive and 88% specific. For meniscal lesions, the clinical assessment has specificity and sensitivity of 91% compared, respectively, with 87% and 82% for MRI. The authors concluded that MRI was overused in the evaluation of knee disorders and was not a cost-effective method for evaluating injuries, if compared with a skilled examiner. But the cost-effectiveness analysis of medical diagnostic devices in cases of the knee disorders or injuries is not as explicit in nature. Bui-Mansfield et al [52], on the basis of a cost of $1000 for each MRI examination and $4000 for each diagnostic arthroscopy, concluded that
412
D. Witonski / Clin Sports Med 21 (2002) 403–415
cost-effective MRI requires 25% true negativity. With MRI examinations obtained before surgery, the authors find 21 of 50 arthroscopies unnecessary, thereby producing a savings of as much as $680 per MRI examination of the knee joint. Suarez-Almazor et al [53] performed an economic evaluation based on cost-effectiveness ratios (per averted arthroscopy), including direct and indirect costs. They revealed that, in general, MRI appears to be a cost-effective diagnostic procedure for patients with internal derangement of the knee requiring arthroscopy. Similar observations are described by Ruwe et al [54,55] and Chissell et al [56]. Le Vot et al [57] state that emergency diagnostic arthroscopy of the knee can be efficiently replaced by MRI examination.
Summary Among the devices helping with an accurate diagnosis, neither MRI nor arthroscopy is perfect; both delineate pathology in the knee joint with reasonable sensitivity and specificity. MRI, as a noninvasive and nonionizing modality, has made a significant contribution to the understanding of musculoskeletal disturbances. Static images through the patellofemoral joint in different degrees of flexion reveal only the degree of patellar tilt or subluxation, parameters that can be measured also on the axial view of conventional radiography. The accuracy of patellar position on static axial MRI is limited by the absence of muscle contraction, movement, and loading. Dynamic axial images of patellofemoral articulation can demonstrate the degree of flexion where patellar malalignment is maximal and assess whether or not it reduces. Arthroscopy, aside from its diagnostic values, provides the opportunity for treatment of intra-articular changes contributing to knee joint disturbances, but it is an invasive technique with potential risks of complications. The performed cost-effectiveness analysis of MRI is based mainly on estimation of intra-articular pathology of the acutelyinjured knee [49,52,56]. There are scarce data on the cost-effectiveness of MRI of patellofemoral alignment in patellofemoral pain knees. Total examination time for active movement dynamic MRI procedure is approximately 8 to 10 minutes, thus it can be performed during routine MRI examination of the knee. In cases of suspected patellofemoral malalignment with symptoms that mimic other types of internal derangement of the knee joint, dynamic MRI can be a procedure of choice for detection of transient patellar dislocation, whereas a single clinical examination cannot differentiate from other internal knee pathologies. Dynamic MRI, although in an experimental phase, gives us a new perspective for dynamic study of the patellofemoral joint.
References [1] Hirokawa S. Three-dimensional mathematical model analysis of the patellofemoral joint. J Biomech 1991;24:659.
D. Witonski / Clin Sports Med 21 (2002) 403–415
413
[2] Buff HU, Jones LC, Hungerford DS. Experimental determination of forces transmitted through the patellofemoral joint. J Biomech 1988;21:17. [3] Hehne HF. Biomechanics of the patellofemoral joint and its clinical relevance. Clin Orthop 1990;258:73. [4] Van Eijden TMGJ, Kouwenhoven E, Verburg J, et al. A mathematical model of the patellofemoral joint. J Biomech 1986;19:219. [5] Reider B, Maeshall JL, Koslin B, et al. The anterior aspects of the knee joint. J Bone Joint Surg Am 1981;63:351. [6] Lieb FJ, Perry J. Quadriceps function. An anatomical and mechanical study using amputed limbs. J Bone Joint Surg Am 1968;50:1535. [7] Hungerford DS, Barry N. Biomechanics of the patellofemoral joint. Clin Orthop 1979;144:9. [8] Duchenne GB. Physiology of motion demonstrated by means of electrical stimulation and clinical observation and applied to study of paralysis and deformities. Philadelphia: J.B. Lippincott Co; 1949. [9] Last RJ. Some anatomical details of the knee joint. J Bone Joint Surg Br 1948;30:683. [10] Madigan R, Wissinger HA, Donaldson WF. Preliminary experience with a method of quadricepsplasty in recurrent subluxation of the patella. J Bone Joint Surg Am 1975;57:600. [11] Roberts JM. The surgical knee. Surg Clin North Am 1974;54:1313. [12] Smillie IS. Injures of the knee joint. Baltimore: Williams & Wilkins Co; 1962. [13] Steindler A. Kinesiology of the human body under normal and pathological conditions. Springfield (IL): Charles C Thomas; 1955. [14] Nagamine R, Otani T, White SE, et al. Patellar tracking measurement in the normal knee. J Orthop Res 1995;13:115. [15] Bentley G, Dowd G. Current concepts of etiology and treatment of chondromalacia patellae. Clin Orthop 1984;189:209. [16] Delgado-Martins H. A study of the position of the patella using computerized tomography. J Bone Joint Surg 1979;61B:443. [17] Ficat RP, Hungerford DS. Disorders of the patellofemoral joint. Baltimore: Williams & Wilkins; 1975. [18] Settegast L. Lower extremity and pelvis. In: Mescham I, editor. Radiographic positioning and related anatomy. Philadelphia: WB Saunders Company; 1978. [19] Laurin CA, Dussault R, Levesque HP. The tangential X-ray investigation of the patellofemoral joint: X-ray technique, diagnostic criteria and their interpretation. Clin Orthop 1979;144:16. [20] Merchant AC, Mercer RL, Jacobsen RH, et al. Roentgenographic analysis of patellofemoral congruence. J Bone Joint Surg Am 1974;56:1391. [21] Schutzer SF, Ramsby GR, Fulkerson JP. The evaluation of the patellofemoral pain using computerized tomography: a preliminary study. Clin Orthop 1986;204:286. [22] Shellock FG, Mink JH, Fox JM. Patellofemoral joint: kinematic MR imaging to assess tracking abnormalities. Radiology 1988;168:551. [23] B} ull CC, Brossmann J, Muhle C, et al. Die Untersuchung des Femoropatellargelenkes mit bewegungsgetriggerter MRT im Vergleich zur arthroskopichen Diagnostik. Arthroskopie 1993; 6:249. [24] Kujala UM, Osterman K, Kormano M, et al. Patellar motion analyzed by magnetic resonance imaging. Acta Orthop Scand 1989;60:13. [25] Pinar H, Akseki D, Genc I, et al. Kinematic and dynamic axial computerized tomography of the normal patellofemoral joint. Knee Surg Sports Traumatol Arthrosc 1994;2:27. [26] Witon´ski D, Go´raj B. Patellar motion analyzed by kinematic and dynamic axial magnetic resonance imaging in patients with anterior knee pain syndrome. Arch Orthop Trauma Surg 1999;119:46. [27] Aglietti P, Insall JN, Cerulli G. Patellar pain and incongruity. Measurements of incongruence. Clin Orthop 1983;176:217. [28] Grelsamer RP, Newton PM, Staron RB. The medial-lateral position of the patella on routine magnetic resonance imaging: when is normal not normal? Arthroscopy 1998;14:23.
414
D. Witonski / Clin Sports Med 21 (2002) 403–415
[29] Schutzer SF, Ramsby GR, Fulkerson JP. Computed tomographic classification of patellofemoral pain patients. Orthop Clin North Am 1986;17:235. [30] Martinez S, Korobkin N, Fondren FB, et al. Computed tomography of the normal patellofemoral joint. Invest Radiol 1983;18:249. [31] Sta¨ubli HU, D} urrenmatt U, Porcellini B, et al. Anatomy and surface geometry of the patellofemoral joint in the axial plane. J Bone Joint Surg Br 1999;81:452. [32] Shellock FG, Foo TKF, Deutsch AL, et al. Patellofemoral joint: evaluation during active flexion with ultrafast spoiled GRASS MR imaging. Radiology 1991;180:581. [33] Shellock FG, Mink JH, Deutsch AL, et al. Patellofemoral joint: identification of abnormalities with active-movement, ‘‘unloaded’’ versus ‘‘loaded’’ kinematic MR imaging techniques. Radiology 1993;188:575. [34] Shellock FG, Mink JH, Deutsch A, et al. Evaluation of patients with persistent symptoms after lateral retinacular release by kinematic resonance imaging of the patellofemoral joint. Arthroscopy 1990;6:226. [35] Brossmann J, Muhle C, Schroder C, et al. Patellar tracking patterns during active and passive knee extension: evaluation with motion-triggered cine MR imaging. Radiology 1993;187:205. [36] Muhle C, Brossmann J, Heller M. Kinematic MRI of the knee using a specially designed positioning device. J Comput Assist Tomogr 1996;20:522. [37] Niitsu M, Ikeda K, Fukubayashi T, et al. Knee extension and flexion: MR delineation of normal and torn anterior cruciate ligaments. J Comput Assist Tomogr 1996;20:322. [38] Vedi V, Williams A, Tennant SJ, et al. Meniscal movement. An in-vivo study using dynamic MRI. J Bone Joint Surg Br 1999;81:37. [39] Bellelli A, Nardis P. Dynamic magnetic resonance of the knee. Considerations on techniques an anatomy with magnetic resonance system with open magnet. Radiol Med (Torino) 1997;93:199. [40] Powers CM, Shellock FG, Pfaff M. Quantification of patellar tracking using kinematic MRI. J Magn Reson Imaging 1998;8:724. [41] Sheehan FT, Drace JE. Quantitative MR measures of three-dimensional patellar kinematics as a research and diagnostic tool. Med Sci Sports Exerc 1999;31:1399. [42] Sheehan FT, Zajac FE, Drace JE. In vivo tracking of the human patella using cine phase contrast magnetic resonance imaging. J Biomech Eng 1999;121:650. [43] Shellock FG, Kanal E. Policies, guidelines, and recommendations for MR imaging safety and patient management. J Magn Reson Imaging 1991;1:97. [44] Lindberg U, Hamberg P, Lysholm J, et al. Arthroscopic examination of the patellofemoral joint using a central, one-portal technique. Orthop Clin North Am 1986;17:263. [45] Witon´ski D. Arthroscopic dynamic assessment of the patello-femoral joint in patients with anterior knee pain syndrome. Chir Narz Ruchu Ortop Pol 1999;64:505. [46] Struhl S. Dynamic assessment of shoulder and patellofemoral pathology using limited volume gas arthroscopy. Arthroscopy 1998;14:546. [47] Sojbjerg JO, Lauritzen J, Hvid I, et al. Arthroscopic determination of patellofemoral malalignment. Clin Orthop 1987;215:243. [48] Delaunay C, Kapandji T. Effect of arthroscopic fluid inflow on the femoro-patellar tracking. Comparative study of the dry and fluid filled knee. Rev Chir Reparatrice Appar Mot 1998; 84:346. [49] Boden SD, Labropoulos PA, Vailas JC. MR scanning of the acutely injured knee: sensitive, but is it cost effective? Arthroscopy 1990;6:306. [50] Lundberg M, Odensten M, Thuomas KA, et al. The diagnostic validity of magnetic resonance imaging in acute knee injuries with hemarthrosis. A single-blinded evaluation in 69 patients using high-field MRI before arthroscopy. Int J Sports Med 1996;17:218. [51] Gelb HJ, Glasgow SG, Sapega AA, et al. Magnetic resonance imaging of the knee disorders. Clinical value and cost-effectiveness in sports medicine practice. Am J Sports Med 1996;24:99. [52] Bui-Mansfield LT, Youngberg RA, Warme W, et al. Potential cost savings of MR imaging obtained before arthroscopy of the knee: evaluation of 50 consecutive patients. Am J Roentgenol 1997;168:913.
D. Witonski / Clin Sports Med 21 (2002) 403–415
415
[53] Suarez-Almazor ME, Kaul P, Kendall CJ, et al. The cost-effectiveness of magnetic resonance imaging for patients with internal derangement of the knee. Int J Technol Assess Health Care 1999;15:392. [54] Ruwe PA, Wright J, Randall RL, et al. Can MR imaging effectively replace diagnostic arthroscopy? Radiology 1992;183:335. [55] Ruwe PA, McCarthy SM. Cost effectiveness of magnetic resonance imaging of the knee. Magn Reson Imaging Clin N Am 1994;2:475. [56] Chissell HR, Allum RL, Keightley A. MRI of the knee: its cost-effective use in a district general hospital. Ann R Coll Surg Engl 1994;76:26. [57] Le Vot J, Solarcroup JC, Leonetti P, et al. Correlations between the clinical test/MRI/arthroscopy in acute knee injures. J Radiol 1993;74:483.