CHARACTERIZATION OF THE MAGIC ANGLE EFFECT IN THE EQUINE DEEP DIGITAL FLEXOR TENDON USING A LOW-FIELD MAGNETIC RESONANCE SYSTEM

Authors


Address correspondence and reprint requests to Mathieu Spriet, at the above address. E-mail: mspriet@ucdavis.edu

Abstract

Three isolated equine limbs were imaged with a low-field magnetic resonance system with a vertical magnetic field. Each limb was scanned in multiple positions with mild variation of the angle between the magnetic field and the long axis of the limb. When the long axis of the limb was not perpendicular to the magnetic field, a linear hyperintense signal was present at the palmar aspect of one of the deep digital flexor tendon lobes, at the level of the navicular bone and collateral sesamoidean ligaments, in proton density and T1-weighted pulse sequences. With increased angulation of the limb, the palmar hyperintense signal extended farther distally and proximally and additional signal hyperintensity was present at the dorsal aspect of the distal part of the other lobe of the deep digital flexor tendon. Increased signal intensity was also present in the collateral ligament of the distal interphalangeal joint on the same side as the palmar hyperintense signal in the tendon. The changes in the deep digital flexor tendon are due to the specific orientation of fibers at the palmar and dorsal aspect of the tendon, which is responsible for focal manifestation of the magic angle effect. Careful positioning of the limb perpendicular to the magnetic field can prevent this phenomenon. The association of palmar increased signal intensity in the deep digital flexor tendon with increased signal in the collateral ligament of the distal interphalangeal joint on the same side should be recognized as manifestations of the magic angle effect.

Introduction

In several equine feet examined in a low-field magnetic resonance (MR) system, we observed a linear hyperintense signal at the palmar aspect of one of the lobes of the distal aspect of the deep digital flexor tendon. This signal was observed on T1-weighted (T1-W) images and was responsible for an asymmetric appearance between the two lobes of the deep digital flexor tendon. We considered this change artifactual, although we could not identify its cause at that time. Previously, as we were investigating the magic angle effect in the collateral ligaments of the distal interphalangeal joint with a low-field MR unit, we observed focal signal intensity changes in the deep digital flexor tendon.1 Thereafter, we hypothesized that the signal observed in the deep digital flexor tendon could be related to the magic angle effect, and we subsequently reevaluated data from the previous study for changes in signal intensity in the deep digital flexor tendon.

Material and Methods

The distal extremities of the thoracic limbs of three skeletally mature horses with no known lameness were imaged shortly after euthanasia using a human knee radiofrequency coil in a 0.25 T magnetic resonance unit.* The imaging and positioning protocols have been described previously.1 Briefly, the feet were initially scanned as if the horse were in lateral recumbency, with the sagittal plane of the limb horizontal and thus perpendicular to the magnetic field. Then, imaging was repeated with different angulations of the sagittal plane relative to horizontal, with a maximum angulation of 20°. Two of these angulated series were obtained for the first digit, four for the second digit, and six for the third digit, for a total of three horizontal series and 12 angulated series.

Each series included transverse images, acquired using a spin echo T1-W pulse sequence and a turbo spin echo (TSE) dual proton density (PD) and T2-weighted (T2-W) pulse sequence, and dorsal images acquired with a 3D gradient echo (3D GE) T1-W pulse sequence (Table 1). The 3D GE T1 data were reformatted in the transverse and sagittal planes. The transverse plane was defined parallel to the sole of the hoof and thus perpendicular to the collateral ligaments of the distal interphalangeal joint, and the dorsal plane was parallel to the long axis of the collateral ligaments of the distal interphalangeal joint.

Table 1.   Pulse Sequence Parameters
SequenceTR (ms)TE (ms)NEXFlip AngleSlice Thickness (mm)Slice Interval (mm)FOV (mm)Acquisition Matrix
  1. FOV, field of view; NEX, number of acquisitions; SE T1, spin echo T1; TE, time of echo; TR, time of repetition; TSE dual echo, turbo spin echo dual echo; 3D GE T1, 3D gradient echo T1.

SE T17602639030.3190 × 190288 × 186
TSE dual echo245028 (PD)
90 (T2)
19030.3200 × 200256 × 192
3D GE T138161650.60190 × 190 × 80192 × 192 × 80

If increased signal intensity was observed in the deep digital flexor tendon, its location and extent were noted. Associated changes in signal intensity in the collateral ligaments were also recorded. The angulation of the limb relative to the magnetic field was measured from the dorsal images.

Results

A focal linear hyperintense signal was observed in the 3D GE T1-W images in the deep digital flexor tendon in all series where the foot was angled with respect to the main magnetic field (Figs. 1 and 2). The hyperintense signal always appeared at the palmar aspect of one of the lobes of the distal aspect of the deep digital flexor tendon, which created an asymmetric appearance of the lobes of the deep digital flexor tendon on transverse and dorsal images (Fig. 1). The signal appeared either in the lateral or medial lobe depending on the direction of angulation. The intensity of the signal was dependent on the degree of angulation of the limb.

Figure 1.

 Gradient echo 3D T1-weighted dorsal images of foot 2 with an angle of 9° between the long axis of the foot and the perpendicular to the magnetic field. Images are shown at the level of the distal sesamoidean impar ligament and the collateral ligaments of the distal interphalangeal joint (A), navicular bone (B), and collateral sesamoidean ligament (C). A hyperintense signal is present at the palmar aspect of the lateral lobe of the deep digital flexor tendon (short white arrow), at the dorsal aspect of the medial lobe of the deep digital flexor tendon (long white arrow), and in the lateral collateral ligament of the distal interphalangeal joint (gray arrow).

Figure 2.

 Gradient echo 3D T1-weighted sagittal reformatted images of foot 2 with an angle of 9° between the long axis of the foot and the perpendicular to the magnetic field. Images are shown through the lateral lobe (A) and through the medial lobe (B) of the deep digital flexor tendon. A hyperintense signal is present at the palmar aspect of the lateral lobe of the deep digital flexor tendon (short white arrow), extending from the level of the distal sesamoidean impar ligament to the level of the middle phalanx. On image B, a hyperintense signal is present at the dorsal aspect of the medial lobe of the deep digital flexor tendon (long white arrow) at the level of the distal sesamoidean impar ligament and navicular bone.

The increased signal intensity was also present on the SE T1 and the TSE PD-weighted images, but was less pronounced than on the 3D GE T1-W images. No increase in signal intensity was observed on the TSE T2-W images.

In eight of 12 angulated positions, an additional hyperintense signal was present at the dorsal aspect of the opposite lobe of the deep digital flexor tendon (Figs. 1 and 2); this signal occurred with angulation of the limb of 7° and greater. On some images, the dorsal hyperintense signal appeared continuous to the palmar hyperintense signal in the opposite lobe through the junction between the two lobes at the level of the navicular bone (Fig. 1B).

When present, the palmar hyperintense signal always occurred at the level of the navicular bone, collateral sesamoidean ligaments, and distal aspect of the middle phalanx. It extended farther distally with increased angulation and reached the level of the distal sesamoidean impar ligament in three of 12 positions (Fig. 2A). The dorsal hyperintense signal always occurred at the level of the distal sesamoidean impar ligament and navicular bone (Fig. 2B). It extended to the level of the collateral sesamoidean ligaments and distal aspect of the middle phalanx in six of eight positions.

The palmar hyperintense signal was associated with an increase in signal intensity in the collateral ligament of the distal interphalangeal joint on the same side in 10 out of 12 series (Fig. 1A). In all positions where increased signal intensity was present in one of the collateral ligaments, a hyperintense signal was present at the palmar aspect of the lobe of the deep digital flexor tendon on the same side.

Discussion

As the linear hyperintense signal we observed at the dorsal and palmar aspect of the deep digital flexor tendon was related to a change in position of the foot relative to B0 and occurred in sequences with a short time of echo, it is logical to assume that the signal is due to the magic angle effect.2–5 Tendons and ligaments usually have a signal of very low intensity on all MR sequences, due to their very short T2 relaxation time. If the collagen fibers are aligned at 55±10° with the main magnetic field, the T2 relaxation time increases due to decreased magnetic interactions between the protons of the water molecules in the tendon.2,3 This leads to increased signal intensity with pulse sequences using a short time of echo, such as T1-W and PD pulse sequences.4,5 The 3D GE T1 pulse sequence has a very short time of echo, which explains the pronounced magic angle effect observed in this sequence.6

Histopathology was not available for these tendons but their normal appearance when imaged perpendicular to the magnetic field suggested that no lesions were present.

The hyperintense signal we observed has an unusual focal linear appearance for the magic angle effect, being different from that previously reported in the deep digital flexor tendon and collateral ligaments of the distal interphalangeal joint,1,7 where diffuse increased signal intensity was observed throughout. This focal linear appearance can be explained by the local fiber orientation and the specificity of the magic angle effect to fiber orientation (Fig. 3).2,3 At the palmar aspect of the deep digital flexor tendon, the most superficial fibers have a diverging orientation from proximal to distal, from the level of the distal aspect of the middle phalanx to the navicular bone, whereas at the dorsal aspect of the deep digital flexor tendon, the superficial fibers converge from proximal to distal, at the level of the navicular bone and distal sesamoidean impar ligament. This explains the association of hyperintense signal at the palmar aspect of one lobe with hyperintense signal at the dorsal aspect of the other lobe; the fibers of these two areas are parallel to each other. The level of divergence or convergence of fibers explains the level of occurrence of the hyperintense signal, with the palmar hyperintense signal occurring more proximal than the dorsal hyperintense signal.

Figure 3.

 Palmar (A) and dorsal (B) macroscopic views of the distal aspect of an isolated deep digital flexor tendon. The superficial fibers at the palmar aspect of the tendon are diverging from the level of the distal aspect of the middle phalanx to the navicular bone. At the dorsal aspect of the tendon, the superficial fibers are converging at the level of the navicular bone and distal sesamoidean impar ligament.

The manifestation of the magic angle effect in this study is different from that previously reported with high-field systems.7 This is due to a different orientation of the main magnetic field. With most high-field MR systems, the magnetic field is parallel to the long axis of the limb being imaged, which leads to a diffuse increased signal intensity in the distal part of the deep digital flexor tendon.7 With MR systems similar to the one used in this study where the magnetic field is perpendicular to the long axis of the limb being imaged, the magic angle effect results in an asymmetric appearance of the two lobes of the deep digital flexor tendon.

It is important to be aware of this manifestation of the magic angle effect to avoid misinterpreting it as a superficial lesion of the deep digital flexor tendon. If a change in signal intensity is observed in the deep digital flexor tendon on PD or T1-W images, the tendon should always be compared with T2-W images, which are not susceptible to the magic angle effect. Superficial lesions of the deep digital flexor tendon should be conspicuous on T2-W images, at least in the acute phase, and they should have irregular margins due to fibrillation. Additional changes, such as modified size and shape, and abnormal signal in surrounding soft tissue, should also be expected in a real lesion. Signal change alone, without morphologic changes or associated findings in surrounding structures, is more likely to be artifactual.

It is interesting to note the association of the magic angle effect in the deep digital flexor tendon and in the collateral ligaments of the distal interphalangeal joint with the same change in position. The association between the two can be useful in recognizing the artifact. Changes in the appearance of the collateral ligament of the distal interphalangeal joint should be interpreted in concert with the appearance of the deep digital flexor tendon and vice versa. If increased signal intensity is observed in a collateral ligament of the distal interphalangeal joint, the palmar aspect of the lobe of the deep digital flexor tendon on the same side should be carefully examined for increased signal. If there is increased signal in both the collateral ligament and the deep digital flexor tendon, the change observed in the collateral ligament is most likely due to the magic angle effect. If the deep digital flexor tendon has a normal signal, the abnormal signal in the collateral ligament is more likely to represent a true lesion.

Ideally, the occurrence of the magic angle effect should be prevented by careful positioning of the limb. The limb should be perfectly horizontal, so that the long axis of the digit is perpendicular to the magnetic field.1

Changes in the deep digital flexor tendon, similar to those described in this study, are also encountered using standing MR systems. However, angulations, such as the ones used in this study, are unlikely to occur with a standing horse, except if major mediolateral foot imbalance is present. The characteristics of the magic angle effect in the deep digital flexor tendon with standing MR systems are still to be determined.

Footnotes

  1. *Esaote Vet MR Grande, Genoa, Italy.

Ancillary