HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2232010857v1 223/2/397    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hayashi, N. Articles by Tajiri, Y. Search for Related Content PubMed PubMed Citation Articles by Hayashi, N. Articles by Tajiri, Y.

Accuracy of Abnormal Paraspinal Muscle Findings on Contrast-enhanced MR Images as Indirect Signs of Unilateral Cervical Root-Avulsion Injury¹

¹ From the Departments of Radiology (N.H., T.M., O.A., S.A., K.O.) and Orthopedics (Y.T.), Graduate School of Medicine, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. Received April 30, 2001; revision requested June 7; revision received September 24; accepted October 22. Address correspondence to N.H. (e-mail: naoto-tky@umin.ac.jp).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: To evaluate the accuracy of abnormal magnetic resonance (MR) findings in the paraspinal muscles as indirect signs of nerve-root avulsion injury.

MATERIALS AND METHODS: Forty-three consecutive patients suspected of having unilateral root-avulsion injury underwent MR imaging and were evaluated. Paraspinal muscles were evaluated for abnormal signal intensity on T1- and T2-weighted images, abnormal enhancement on images obtained after contrast material enhancement, and muscle volume loss. MR images were interpreted independently by two observers for interobserver variability. MR findings were compared with findings of root continuity, determined with a combination of surgery and clinical evaluation. Sensitivities, specificities, and values of the findings were calculated.

RESULTS: Sensitivities of MR findings in the paraspinal muscles indicating root-avulsion injury were 88% (36 of 41 patients) for abnormal enhancement, 83% (34 of 41 patients) for high signal intensity on T2-weighted images, 37% (15 of 41 patients) for high signal intensity on T1-weighted images, and 71% (29 of 41 patients) for muscle volume loss. Specificities for all findings were 100% (two of two patients). Of the paraspinal muscles, findings in the multifidus muscle were the most accurate and provided the highest interobserver agreement ( = 0.81).

CONCLUSION: Contrast material–enhanced abnormal MR findings in the paraspinal muscles are accurate in indicating root-avulsion injuries, and abnormal enhancement in the multifidus muscle is the most accurate among paraspinal muscle findings.

© RSNA, 2002

Index terms: Muscles, denervation, 30.492 • Muscles, MR, 30.121411, 30.121415, 30.12143 • Nerves, injuries, 30.492 • Nerves, roots, 30.492

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cervical root-avulsion injury occurs in relation to severe traction force on the upper limb, mostly as a result of a motorcycle accident. It is critical to differentiate preganglionic lesions (root-avulsion injury) from postganglionic lesions (brachial plexus injury and peripheral nerve injury) to determine management and prognosis (1,2). Preganglionic lesions are not considered amenable to repair, so the functions of some denervated muscles are restored with nerve transfers from accessory nerves, the cervical plexus, or intercostal nerves, whereas postganglionic lesions are repaired with excision of the damaged segment and nerve autograft between two nerve ends (3). Conventional myelography and myelographic computed tomography (CT) are usually used to depict intradural portions of nerve roots. Although the noninvasive procedure of magnetic resonance (MR) imaging is preferred to myelography in most situations, MR imaging of the cervical spine cannot reliably enable delineation of the intradural nerve roots, probably because of various motion artifacts in the cervical region (4–7). Contrast material–enhanced MR imaging of the intradural roots has been reported to have high specificity but low sensitivity because of poor delineation of the intradural roots (6).

Recently, MR signal intensity (SI) changes in the paraspinal muscles have been shown to be suggestive of root-avulsion injury (8). Denervation of the posterior paraspinal muscles is known to be a possible indicator of root avulsion because these muscles are innervated with dorsal branches that arise most proximally from the spinal nerves. In theory, the posterior paraspinal muscles become denervated because of injuries involving the ventral root or spinal nerve proximal to the origin of the dorsal branch, while denervation of the posterior paraspinal muscles does not occur in nerve-root injuries distal to the origin of the dorsal branch (8). However, since the distance between the dorsal root ganglion and the dorsal branch origin is very short, the probability of posterior paraspinal muscle injury in postganglionic injury should be small.

The MR-based diagnosis of root-avulsion injury may be complemented by findings in the paraspinal muscles, since the primary findings in the intradural roots have low sensitivity (6). However, as far as we are aware, the accuracy of MR findings in cervical root-avulsion injury in paraspinal muscles has never been evaluated. Also, as far as we have searched, there are no studies of contrast-enhanced MR of the paraspinal muscles in the literature.

This prospective study was undertaken to evaluate the accuracy of abnormal MR findings in the paraspinal muscles as indirect signs of nerve-root avulsion injury.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
This clinical study was approved by the ethics committee of our institution’s faculty. Written informed consent was obtained from the patients at their first meeting with the attending orthopedist (Y.T.).

Forty-four consecutive patients, 42 male and two female patients (16–70 years of age; average age, 24.5 years) clinically suspected of having unilateral cervical root-avulsion injury were examined with MR imaging. MR imaging of the cervical spine was ordered by the attending orthopedist at the first meeting as part of the detailed presurgical evaluation process. All patients underwent serial neurologic examinations, myelography, and myelographic CT. One patient was excluded from our study because contrast-enhanced MR images were not obtained.

MR examinations were performed with a 1.5-T unit (Magnetom Vision; Siemens Medical Systems, Erlangen, Germany) by using spine array coils. Liquid fluorocarbon pads (Sat Pad N; Alliance Pharmaceutical, San Diego, Calif) were applied as a circumferential neck collar to reduce magnetic susceptibility differences. The interval between injury and MR imaging was 10–217 days, with an average of 70.6 days. In 32 of the 43 patients, transverse contiguous 4-mm-thick T1-weighted spin-echo (SE) (repetition time msec/echo time msec, 756/12) and transverse T2-weighted fast SE (4,700/112 [effective]; echo train length, 15) images were obtained from the vertebral levels of C3 to T2. Sagittal T2-weighted fast SE (4,500/112 [effective]) images were acquired with contiguous 4-mm-thick sections. Transverse postcontrast SE images (756/12) were obtained after intravenous administration through the cubital vein of 0.2 mmol per kilogram of body weight gadopentetate dimeglumine (Magnevist; Schering Japan, Tokyo). In the remaining 11 patients, only the levels of the clinically suspected nerve-root injury were imaged due to patient intolerance of extended examination times. Total MR imaging time was 30–60 minutes. A chemical fat-saturation technique was added in five of the 11 patients.

Hard-copy images of the transverse precontrast and postcontrast SE images and transverse fast SE images were evaluated by two radiologists (T.M., O.A.) who were not aware of the following clinical information: side of the body on which the injury had occurred, clinical history, results of neurologic examinations, and results of surgical or other diagnostic imaging procedures. The radiologists were instructed to evaluate abnormal findings (see next paragraph) in only the paraspinal muscles and to ignore other findings, if any, such as intradural roots or meningocele. First, images were interpreted independently by each radiologist and findings recorded on an assessment form. Then a consensus regarding the MR image interpretations was reached between the two readers by means of discussion; any initial inconsistencies were carefully reexamined and resolved. Interobserver agreement was evaluated by using statistics for the initial observation of the two radiologists.

The paraspinal muscles were evaluated in the transverse plane. Some muscles that could not be reliably separated in the transverse plane were grouped and evaluated as a single unit: splenius capitis and cervicis; iliocostalis cervicis, longissimus capitis, and cervicis; medial and posterior scalenus; and longus capitis and anterior scalenus. The SI of the paraspinal muscles on the T1- and T2-weighted images was defined as abnormal when there was a difference when compared with that of the trapezius muscle, which is not innervated by cervical root nerves. Enhancement of the paraspinal muscles was defined as abnormal if the SI of the muscle was higher than on precontrast images and higher than that of other normal muscles, especially the trapezius, on postcontrast images. Subtle differences in intensity were evaluated by comparison with the contralateral side. Muscle volume loss was defined as positive when the cross-sectional area of the muscle was smaller than the corresponding contralateral muscle.

The preganglionic continuity of individual nerve roots to the spinal cord was determined by using the results of neurologic examinations, exploratory surgery, CT myelography, and spinal MR imaging (eg, findings in the paraspinal muscle were not considered). If neurologic examination findings were definitively normal for a particular level, the corresponding nerve root was considered intact. When neurologic examination findings were considered to be abnormal, the corresponding nerve root was surgically explored, and intraoperative somatosensory evoked potentials of the spinal root nerves were recorded. Negative somatosensory evoked potentials were considered positive signs of preganglionic root injury. Nerve roots that were inaccessible for somatosensory evoked potentials because of surgical findings such as absent or severely scarred roots or large extravertebral meningoceles were also considered positive indicators of preganglionic root injury (1,9,10). In cases without surgical exploration of the brachial plexus, root continuity was comprehensively determined by using the results of neurologic examination and imaging (excluding paraspinal muscles) and clinical follow-up examinations. These judgments of preganglionic root continuity were used as a standard of reference to evaluate the sensitivity and specificity of the MR findings in the paraspinal muscles.

First, each MR finding in the paraspinal muscles (abnormal high signal intensities, abnormal enhancements, and muscle volume loss) was compared with preganglionic nerve-root continuity to examine the sensitivity and the specificity for muscle denervation. The sensitivity and specificity of the MR finding for detection of preganglionic injury were calculated on a per-patient basis. The MR finding was considered positive for preganglionic injury if any of the paraspinal muscles on the examined side of the body demonstrated the abnormality. A 95% CI was applied for sensitivity and specificity calculation (11).

Second, sensitivity, specificity, and interobserver agreement of the MR findings in various paraspinal muscles were compared to find the most accurate sign of root avulsion injury.

Finally, to find out whether there was a one-to-one correlation between each avulsed root and the MR-demonstrated abnormalities of the paraspinal muscles, the levels of roots with preganglionic injury were compared with the area of MR abnormalities of the multifidus muscle, a monosegmentally innervated paraspinal muscle.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Clinical Findings
Exploratory surgery of the brachial plexus was performed in 39 of the 43 patients, with recording of intraoperative somatosensory evoked potentials of the spinal root nerves. Thirty-eight of the 39 patients who underwent exploratory surgery had preganglionic injury, with a total of 133 avulsed roots. Four patients did not undergo exploratory surgery: Two did not initially have a preganglionic injury diagnosed at detailed neurologic examination (however, one patient had a preganglionic injury, confirmed with imaging of the intradural roots and with follow-up neurologic examination), one patient underwent therapy at another institution, and one patient was too elderly to undergo exploratory surgery. Three of the four patients had preganglionic root injury diagnosed on the basis of findings of neurologic examinations and imaging studies (excluding MR imaging of the paraspinal muscles), whereas one did not have preganglionic root injury diagnosed on the basis of findings of follow-up neurologic examinations. Thus, a total of 41 patients were confirmed to have preganglionic injuries, with 138 avulsed roots, whereas two patients were confirmed as not having preganglionic injuries. All patients had injury on only one side. None of the patients had initial injury directly to the cervical region, which might have caused abnormal SI in the paraspinal muscles.

MR Findings
Of the 41 patients with preganglionic root injury, 36 (88%) had abnormal enhancement in the ipsilateral paraspinal muscles (Figs 1, 2), 34 (83%) had high signal intensity on T2-weighted images, 15 (37%) had high signal intensity on T1-weighted images, 29 (71%) had muscle volume loss, and five (12%) had no abnormal MR findings in any paraspinal muscles. None of the patients had abnormal MR findings in the contralateral paraspinal muscles. Two patients without preganglionic root injury also had no abnormal MR findings in the paraspinal muscles.

View larger version (149K): [in this window] [in a new window] [Download PPT slide]   Figure 1a. Left C5 and C6 root-avulsion injury in a 25-year-old man 46 days after injury. (a) Transverse fat-suppressed postcontrast SE (756/12) MR image at the C4 vertebral level shows multifidus muscles (upper straight arrows) and semispinalis cervicis muscle (lower straight arrows) as having increased enhancement. Also, the group of iliocostalis cervicis, longissimus capitis, and cervicis muscles shows increased enhancement and atrophy (curved arrows), as compared with the contralateral side (arrowheads). (b) Schematic figure of paraspinal muscles at same C4 vertebral level muscles. Black areas = abnormal enhancement.

View larger version (49K): [in this window] [in a new window] [Download PPT slide]   Figure 1b. Left C5 and C6 root-avulsion injury in a 25-year-old man 46 days after injury. (a) Transverse fat-suppressed postcontrast SE (756/12) MR image at the C4 vertebral level shows multifidus muscles (upper straight arrows) and semispinalis cervicis muscle (lower straight arrows) as having increased enhancement. Also, the group of iliocostalis cervicis, longissimus capitis, and cervicis muscles shows increased enhancement and atrophy (curved arrows), as compared with the contralateral side (arrowheads). (b) Schematic figure of paraspinal muscles at same C4 vertebral level muscles. Black areas = abnormal enhancement.

View larger version (114K): [in this window] [in a new window] [Download PPT slide]   Figure 2a. Right C5 through C8 root-avulsion injury in a 17-year-old male patient 81 days after injury. The multifidus, semispinalis cervicis, and longus colli muscles, the group of splenius cervicis and capitis muscles, and the group of iliocostalis cervicis, longissimus cervicis, and longissimus capitis muscles show high SI (arrowheads) on the (a) transverse precontrast SE (756/12) and (b) transverse fast SE (4,700/112 [effective]; echo train length, 15) MR images and show increased enhancement on (c) the transverse postcontrast SE (756/12) MR image. Note that the semispinalis capitis muscle is preserved between the semispinalis cervicis muscle and the group of splenius cervicis and capitis muscles. Only the lateral portion (short arrows) of the group of splenius cervicis and capitis muscles shows abnormalities, presumably attributed to denervation of the splenius cervicis muscle, while the splenius capitis muscle is intact. The area of abnormal SI and abnormal contrast enhancement between the scalenus anterior and medius muscles is the injured brachial plexus (long arrow).

View larger version (118K): [in this window] [in a new window] [Download PPT slide]   Figure 2b. Right C5 through C8 root-avulsion injury in a 17-year-old male patient 81 days after injury. The multifidus, semispinalis cervicis, and longus colli muscles, the group of splenius cervicis and capitis muscles, and the group of iliocostalis cervicis, longissimus cervicis, and longissimus capitis muscles show high SI (arrowheads) on the (a) transverse precontrast SE (756/12) and (b) transverse fast SE (4,700/112 [effective]; echo train length, 15) MR images and show increased enhancement on (c) the transverse postcontrast SE (756/12) MR image. Note that the semispinalis capitis muscle is preserved between the semispinalis cervicis muscle and the group of splenius cervicis and capitis muscles. Only the lateral portion (short arrows) of the group of splenius cervicis and capitis muscles shows abnormalities, presumably attributed to denervation of the splenius cervicis muscle, while the splenius capitis muscle is intact. The area of abnormal SI and abnormal contrast enhancement between the scalenus anterior and medius muscles is the injured brachial plexus (long arrow).

View larger version (114K): [in this window] [in a new window] [Download PPT slide]   Figure 2c. Right C5 through C8 root-avulsion injury in a 17-year-old male patient 81 days after injury. The multifidus, semispinalis cervicis, and longus colli muscles, the group of splenius cervicis and capitis muscles, and the group of iliocostalis cervicis, longissimus cervicis, and longissimus capitis muscles show high SI (arrowheads) on the (a) transverse precontrast SE (756/12) and (b) transverse fast SE (4,700/112 [effective]; echo train length, 15) MR images and show increased enhancement on (c) the transverse postcontrast SE (756/12) MR image. Note that the semispinalis capitis muscle is preserved between the semispinalis cervicis muscle and the group of splenius cervicis and capitis muscles. Only the lateral portion (short arrows) of the group of splenius cervicis and capitis muscles shows abnormalities, presumably attributed to denervation of the splenius cervicis muscle, while the splenius capitis muscle is intact. The area of abnormal SI and abnormal contrast enhancement between the scalenus anterior and medius muscles is the injured brachial plexus (long arrow).

View this table: [in this window] [in a new window]   TABLE 3. Values for Interobserver Agreement of MR Imaging Findings in Each Paraspinal Muscle

View larger version (156K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Right C5 through C8 root-avulsion injury in a 21-year-old man 50 days after injury. Coronal postcontrast three-dimensional gradient-echo (37/10; flip angle, 45°) MR image shows abnormal enhancement in the multifidus muscle (arrowheads) and a portion of the splenius cervicis muscle (arrows). Fibers of enhanced multifidus muscle are attached to the C4 spine and run downward laterally to insert at articular processes of vertebrae below.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The feasibility of MR signal abnormalities in the paraspinal muscles as an indication of root-avulsion injury has been demonstrated by Uetani et al (8). The current study added the additional observation of abnormal enhancement in the paraspinal muscles and included evaluation of the accuracy of MR findings in the paraspinal muscles to indicate root-avulsion injury. Although SI changes and abnormal enhancement of denervated muscles have been studied extensively in the past few years (13–21), the accuracy of these findings have not, to our knowledge, been evaluated. Our results show that MR findings in the muscles can be accurate indirect signs of nerve injury in certain clinical settings.

An interesting aspect of our results is that abnormal enhancement of the multifidus muscle was always present when there was any abnormal enhancement in the paraspinal muscles and was never present without root avulsion. This indicates that the multifidus muscle may be the sole paraspinal muscle that needs to be evaluated for abnormal enhancement in cervical root avulsion. The multifidus muscles are easy to identify on transverse images. Since the fascicles of the multifidus muscles run nearly perpendicular to the transverse plane, the subtle differences in SI between the fascicles can be evaluated. In fact, SI differences between fascicles on a single image were actually observed in some cases in our study. The high values in the evaluation of multifidus muscles also indicate the high interobserver consistency of these findings. We believe that abnormal enhancement in the multifidus muscles is a useful sign of cervical root avulsion. The clinical usefulness of the MR findings in the lumbar multifidus muscles has also been demonstrated in lumbar radiculopathy (22,23).

Anatomic studies (24,25) have indicated that the multifidus muscles are innervated by a single root, in contrast with the rest of the paraspinal muscles, which receive multisegmental innervation. This anatomy of innervation led us to expect a one-to-one correlation between each avulsed root and the abnormal enhancement of the multifidus muscle. Contrary to our expectations, our results showed that a one-to-one correlation is not always present. A substantial number of avulsed roots did not have corresponding abnormal enhancement in the multifidus muscle. One possible explanation for this discrepancy is that the multifidus muscle is not actually innervated by a single root. The concept of monosegmental innervation of the multifidus muscle has recently been challenged with an electrophysiologic demonstration of polysegmental innervation of lumbar multifidus muscles (26,27) and with a report of polysegmental innervation variation in the lumbar multifidus muscle in a cadaver (28). Our result also supports the notion that the multifidus muscle may not always be monosegmentally innervated.

In general, abnormal enhancement and abnormal SI of muscles are not specific for denervation. They may also be seen in inflammation, neoplasm, or direct injury. Inflammation and neoplasm, however, are unlikely to be concomitant in the clinical setting of suspected root-avulsion injury. Direct injury to the paraspinal muscles would most likely occur in the superficial location and not in the deepest multifidus muscle, in which root avulsion is best demonstrated. It should also be noted that no abnormal findings were found in the paraspinal muscles contralateral to the injury. Since the existence of causes, other than denervation, of abnormal enhancement or abnormal SI in the paraspinal muscles in the clinical setting of suspected unilateral root avulsion injury is unlikely, we believe that the MR abnormalities in the paraspinal muscles are adequately specific in this setting.

One limitation of this study was that four patients did not undergo exploratory surgery to determine whether they had root avulsion. Exploratory surgery, however, cannot be justified unless there is high expectation of patient benefit, such as surgical repair of the injured roots. These four patients did not have adequate reasons to undergo exploratory surgery.

In conclusion, this study shows that contrast-enhanced MR findings in the paraspinal muscles are accurate for diagnosis of cervical root avulsion; in particular, abnormal enhancement of the multifidus muscle demonstrates the abnormality more accurately than do other paraspinal muscle findings. Recognition of this finding can direct the radiologist to further examine the continuity of the intradural nerve roots.

	FOOTNOTES

 
Abbreviations: SE = spin echo, SI = signal intensity

Author contributions: Guarantor of integrity of entire study, N.H.; study concepts, N.H., Y.T.; study design, N.H., T.M.; literature research, N.H.; clinical studies, N.H., T.M., O.A.; data acquisition and analysis/interpretation, N.H., T.M., O.A.; statistical analysis, N.H.; manuscript preparation, N.H., S.A.; manuscript definition of intellectual content, N.H., Y.T.; manuscript editing and revision/review, N.H., S.A., K.O.; manuscript final version approval, all authors.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Millesi H. Brachial plexus injuries. In: Jupiter JB, eds. Flynn’s hand surgery. 4th ed. Baltimore, Md: Williams & Wilkins, 1991; 457-463.
2. Kline DG, Hudson AR. Acute injuries of the peripheral nerves. In: Youmans JR, eds. Neurological surgery. Philadelphia, Pa: Saunders, 1996; 2103-2181.
3. Glasby MA, Hems TE. Repairing spinal roots after brachial plexus injuries. Paraplegia 1995; 33:359-361.
4. Ochi M, Ikuta Y, Watanabe M, Kimori K, Itoh K. The diagnostic value of MRI in traumatic brachial plexus injury. J Hand Surg [Br] 1994; 19:55-59.
5. Vielvoye GJ, Hoffmann CF. Neuroradiological investigations in cervical root avulsion. Clin Neurol Neurosurg 1993; 95:S36-S38.
6. Hayashi N, Yamamoto S, Okubo T, et al. Avulsion injury of cervical nerve roots: enhanced intradural nerve roots at MR imaging. Radiology 1998; 206:817-822.
7. Volle E, Assheuer J, Hedde JP, Gustorf-Aeckerle R. Radicular avulsion resulting from spinal injury: assessment of diagnostic modalities. Neuroradiology 1992; 34:235-240.
8. Uetani M, Hayashi K, Hashmi R, Nakahara N, Aso N, Ito N. Traction injuries of the brachial plexus: signal intensity changes of the posterior cervical paraspinal muscles on MRI. J Comput Assist Tomogr 1997; 21:790-795.
9. Sunderland S. Mechanisms of cervical nerve root avulsion in injuries of the neck and shoulder. J Neurosurg 1974; 41:705-714.
10. Nagano A, Ochiai N, Sugioka H, Hara T, Tsuyama N. Usefulness of myelography in brachial plexus injuries. J Hand Surg [Br] 1989; 14:59-64.
11. Newcombe RG, Altman DG. Proportion and their differences. In: Altman DG, Machin D, Bryant TV, Gardner MJ, eds. Statistics with confidence. 2nd ed. Bristol, Conn: BMJ, 2000; 45-56.
12. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977; 33:159-174.
13. Fleckenstein JL, Watumull D, Conner KE, et al. Denervated human skeletal muscle: MR imaging evaluation. Radiology 1993; 187:213-218.
14. Polak JF, Jolesz FA, Adams DF. Magnetic resonance imaging of skeletal muscle: prolongation of T1 and T2 subsequent to denervation. Invest Radiol 1988; 23:365-369.
15. Polak JF, Jolesz FA, Adams DF. NMR of skeletal muscle: differences in relaxation parameters related to extracellular/intracellular fluid spaces. Invest Radiol 1988; 23:107-112.
16. Shabas D, Gerard G, Rossi D. Magnetic resonance imaging examination of denervated muscle. Comput Radiol 1987; 11:9-13.
17. Uetani M, Hayashi K, Matsunaga N, Imamura K, Ito N. Denervated skeletal muscle: MR imaging—work in progress. Radiology 1993; 189:511-515.
18. Davis SB, Mathews VP, Williams DW, III. Masticator muscle enhancement in subacute denervation atrophy. AJNR Am J Neuroradiol 1995; 16:1292-1294.
19. Murakami R, Baba Y, Nishimura R, et al. CT and MR findings of denervated tongue after radical neck dissection. AJNR Am J Neuroradiol 1997; 18:747-750.
20. Murakami R, Baba Y, Nishimura R, et al. MR of denervated tongue: temporal changes after radical neck dissection. AJNR Am J Neuroradiol 1998; 19:515-518.
21. Russo CP, Smoker WR, Weissman JL. MR appearance of trigeminal and hypoglossal motor denervation. AJNR Am J Neuroradiol 1997; 18:1375-1383.
22. Campbell WW, Vasconcelos O, Laine FJ. Focal atrophy of the multifidus muscle in lumbosacral radiculopathy. Muscle Nerve 1998; 21:1350-1353.
23. Kader DF, Wardlaw D, Smith FW. Correlation between the MRI changes in the lumbar multifidus muscles and leg pain. Clin Radiol 2000; 55:145-149.
24. Bogduk N, Wilson AS, Tynan W. The human lumbar dorsal rami. J Anat 1982; 134:383-397.
25. Macintosh JE, Bogduk N, Munro RR. The morphology of the human lumbar multifidus. Clin Biomechanics 1986; 1:196-204.
26. Wu PB, Kingery WS, Frazier ML, Date ES. An electrophysiological demonstration of polysegmental innervation in the lumbar medial paraspinal muscles. Muscle Nerve 1997; 20:113-115.
27. Wu PB, Date ES, Kingery WS. The lumbar multifidus muscle is polysegmentally innervated. Electromyogr Clin Neurophysiol 2000; 40:483-485.
28. Shindo H. Anatomical study of the lumbar multifidus muscle and its innervation in human adults and fetuses. Nippon Ika Daigaku Zasshi 1995; 62:439-446[Japanese].

This article has been cited by other articles:

				T. Yoshikawa, N. Hayashi, S. Yamamoto, Y. Tajiri, N. Yoshioka, T. Masumoto, H. Mori, O. Abe, S. Aoki, and K. Ohtomo Brachial Plexus Injury: Clinical Manifestations, Conventional Imaging Findings, and the Latest Imaging Techniques. RadioGraphics, October 1, 2006; 26(suppl_1): S133 - S143. [Abstract] [Full Text] [PDF]

				A J P Goddard and A Gholkar Diagnostic and therapeutic radiology of the spine: an overview Imaging, October 1, 2002; 14(5): 355 - 373. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2232010857v1 223/2/397    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hayashi, N. Articles by Tajiri, Y. Search for Related Content PubMed PubMed Citation Articles by Hayashi, N. Articles by Tajiri, Y.