ossification in the region posterior as a myelopathy · ossification in the region oftheposterior...

J. Neurol. Neurosurg. Psychiat., 1970, 33, 263-268

Ossification in the region of the posterior longitudinalligament as a cause of cervical myelopathy

L. BAKAY, H. L. CARES, AND R. J. SMITH

From the Division of Neurosurgery and Department of Radiology, State University ofNew Yorkat Buffalo, U.S.A.

In 1967, we operated on a patient for cervical cordcompression caused by an unusual type of ectopicbone formation in the spinal canal which we ten-tatively identified as ossification within the posteriorlongitudinal ligament. Until that time we hadbelieved that calcification of the vertebral ligamentsis a phenomenon secondary to spondylosis (Oppen-heimer, 1942) and does not cause symptoms byitself. However, about the same time, a descriptionof no less than 18 similar cases was published byOnji, Akiyama, Shimomura, Ono, Hukuda, andMizuno (1967). Additional cases were also found inJapan (Yokoi, 1963). It seems, therefore, that thesecases, while probably rare, constitute a distinctpathological entity. Although calcification andectopic bone formation have been described beforein various ligaments of the spine (Oppenheimer,1942; Cloward, 1967), they were found accidentallywithout giving rise to clinical symptoms. A searchfor similar cases was undertaken and six additionalpatients with such abnormalities were found.

CASE REPORTS

1 ECTOPIC BONE FORMATION

CASE 1 M.N., a 60-year-old white woman, had a oneyear history of difficulty in walking. She also had chronicurinary incontinence which she attributed to a previoushysterectomy. Six weeks before admission, her gaitbecame much worse and she began having severe neckpain, paraesthesia and weakness in both arms, particu-larly in the right arm. One week before admission, shebecame unable to walk.

Examination on admission showed an obese whitefemale who was unable to stand or sit. Motion of theneck was limited, particularly in rotation and flexion.Both triceps muscles were weak, particularly the rightone, but biceps function was good. Both flexion andextension at the wrist were weak; the right grip wasespecially poor. There was no voluntary motion in eitherlower extremity. Sensation to pin prick and light touchwas decreased below the T4 dermatome. Position sensewas intact. Deep reflexes were brisk throughout, patho-logically active on the right side. The radial periosteal

reflex was inverted bilaterally and she had positiveBabinski reflexes on both sides.

Radiographs revealed a mild spondylosis of the cervicalspine with narrowing of the intervertebral spaces atC5_, and C., and minimal uncinate hypertrophy inthe lower cervical spine. The intervertebral foraminawere normal. A strip of bony density was located inthe spinal canal anteriorly (Fig. 1). The superior endof this structure was pointed and the lower end was notclearly defined. The sagittal diameter of the osseousdensity at C, vertebra was 5 mm, the transverse diam-eter 6 mm.A better resolution of the bony density was obtained

by laminography. It extended from the level of C1 toC5 vertebra and occupied the anterior half of the spinal

FIG. 1. Case 1. Lateral view of the cervical spine showingboneformation in the anterior spinal canal.

263

Protected by copyright.

on April 16, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.33.2.263 on 1 April 1970. D

ownloaded from

http://jnnp.bmj.com/

L. Bakay, H. L. Cares, and R. J. Smith

canal. It was continuous with the posterior corticalsurfaces of the bodies of C2 and C3 and possibly C4 andC, vertebrae.Myelography showed that the spinal cord was com-

pressed in its anteroposterior diameter and widenedin its transverse diameter at the level of C2 to C4 vertebraebecause of extradural encroachment on the neural canalby the bony bar (Fig. 2). The column of Pantopaquewas markedly delayed in its passage past this areaand the high cervical subarachnoid space filled onlywith difficulty.A decompressive laminectomy was performed from

C1 to Cf. After the laminae of C3, C4, and C5 vertebraewere removed, the dura appeared to be under tensionand without pulsation. Removal of the C2 lamina wasfollowed by dural pulsation. After carrying the decom-pression on both sides as far laterally as possible, thedura was gently retracted medially. The epidural veinswere engorged, but the nerve root sheaths appearednormal, although somewhat stretched due to the posteriordisplacement of the cord. The abnormal structurefilled out the entire anterior half of the spinal canal.It had the appearance of solid, marble-hard bone.Its surface was smooth, but, apart from its lateralbuttresses, it was tightly adherent to the dura. De-compression resulted in slackening and some posteriorprotrusion of the dura; it was not opened for fear ofinjury to the strongly compromised and kinked cord.The bony mass was not biopsied for the same reason.

Post-operatively, the patient became transiently worse

FIG. 2. Case 1. Lateral view ofcervical myelogram.

neurologically; herquadriparesis becamemore pronouncedwith complete paralysis of her right arm and leg. Twoweeks later, the left side improved to the pre-operativelevel. However, two months later improvement wasalso noted in the right side. Five months after the opera-tion there was considerable improvement in all fourlimbs. Although the right grip remained weak, thestrength in the left arm returned to normal and she wasable to feed herself. Muscle strength in the left leg re-turned to about 50% of normal and she was also ableto lift her right leg against gravity.CASE 2 A.A., a 55-year-old white woman, was admittedwith a three-month history of pain and inability tomove her right shoulder. Six years earlier, she had athyroidectomy followed by a period of hypocalcaemiaand 'seizures'. She was diagnosed as hypoparathyroidand treated with vitamin D.

There was limitation of neck motion in every direction.There was also restriction in mobility in the right shoulder,particularly in abduction, anterior flexion, and internalrotation. Neurological examination was essentiallynegative; there was no weakness, sensory deficit, orreflex change in the right upper extremity and therewas no evidence of long tract involvement of the spinalcord.

Radiographs revealed considerable spondylosis ofvarious segments of the spinal column. The cervicalspine revealed marked anterior spondylotic spurring,but posterior spur formation was minimal and therewas no encroachment upon the intervertebral foramina.The most striking finding was the presence of a bonydensity in the anterior spinal canal between C2 and C4vertebrae. Its characteristics in the lateral view werevery similar to that of case 1 (Fig. 3). On the frontalview, the mass appeared quite wide, reaching fromalmost side to side across the width of the spinal canal.

Radiographs of the shoulder were negative. Althoughthe pathological changes in the cervical spine suggestedthe possibility that the right shoulder pain was causedby radiculopathy rather than by a local articular dis-order, the lack of characteristic neurological findingsmade this diagnosis uncertain. The patient refusedmyelography.

She was treated by cervical traction, infra-red heatto the shoulder and gradual mobilization. This resultedin improvement of her pain and slight increase in therange of motion in the right shoulder. A follow-upexamination five years later revealed no significantchange in the shoulder; the neurological status remainednormal. The patient refused further radiographs.2 CALCIFICATION IN THE POSTERIOR LONGITUDINALLIGAMENTCASE 3 W.B., a 72-year-old Negro male, died at thehospital from disseminated carcinoma of the lung.Radiographs showed marked spondylotic changes inthe cervical spine from C2 through C7 vertebrae withnarrowing of all the corresponding intervertebral discsand slight narrowing of all intervertebral foraminabilaterally. There was a thin sliver of calcification inthe posterior longitudinal ligament, extending from C2to C5 vertebrae. Neurological examination was negative.

264



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


Ossification in the region of the posterior longitudinal ligament as a cause of cervical myelopathy 265

thicker in the higher segments and tapered toward itslower end.

CASE 6 J.W., an 85-year-old white male, had no historyof neurological symptoms until the day of admissionwhen he was found in his home lying on the floor un-conscious. He was quadriplegic with a sensory levelat C5 dermatome. The neck was painful. He had flaccidparalysis of all four limbs and an absence of tendonreflexes over the upper as well as lower extremities andabdominal reflexes. Babinski reflex was present on theleft side; there was no plantar reflex on the right.

Radiographs of the cervical spine showed markedspondylosis and a rotated fracture-dislocation betweenC5 and C6 vertebrae. There was also a considerablecalcification of the posterior longitudinal ligamentwhich extended from C2 to C. or C6 vertebrae. On thelateral view, the diameter of this calcified structurewas quite impressive; it encroached upon the spinalcanal from the anterior direction, narrowing it to aconsiderable extent (Fig. 6). Upon considering thecompleteness of the motor and sensory deficit as wellas the patient's age, conservative measures of treatmentwere applied.

FIG. 3. Case 2. Abnormal bone formation in the anterior CASE 7 M.L., a 65-year-old white male with no localizingspinal canal between C2 and C4 vertebrae. Lateral view. neurological findings, but with advanced ankylosing

spondylitis, showed atlanto-axial fusion with C1 to C2calcification in the posterior longitudinal ligament.

CASE 4 C.T., a 65-year-old white male, was diagnosedduring his hospitalization as having amyotrophiclateral sclerosis. He was paraparetic with spastic lowerextremities, bilateral Babinski reflexes, and bilateralfacial palsy with pseudobulbar symptoms. Cervicalspine films showed what appeai ed to be a calcified pos-terior longitudinal ligament extending from C2 to C4vertebrae (Fig. 4).

CASE 5 H.W., a 51-year-old white male, was admittedwith the main complaint of postural dizziness whichparticularly occurred when looking upward. It wouldsubside upon flexing the neck. Neurological examinationshowed an essentially normal status, apart from thedizziness that was brought on by extending the neck.Radiographs of the spine showed severe changes com-patible with rheumatoid arthritis. Although the bambooappearance involved almost the entire spine, the changeswere most pronounced in the cervical region. Herethe bamboo-like appearance of the cervical spine withossification of the spinal ligaments was quite striking.Lateral masses appeared to be partially fused, althoughapophyseal joints could be seen. Fusion occurred betweenthe bodies of C4, C5, and C6 vertebrae, and very markedanterior osteophyte formation was seen between C6and C7 vertebrae. The cervical lordosis was fairly wellpreserved, but flexion and extension could not becarried out to any significant degree, the only motionhaving occurred at the C, and C2 levels and betweenthe occiput and Cl vertebra. The abnormal posterior FIG. 4. Case 4. Lateral radiograph of the cervical spinelongitudinal ligamentous calcification appeared between reveals thin calcification of the posterior longitudinalC2 and C5 vertebrae (Fig. 5). The calcification was ligament between C2 and C4 vertebrae.



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from



FIG. 5. Case 5. Lateral radiograph of the cervical spineshows extensive calcification of the posterior longitudinalligament between C2 and C6 vertebrae as well as severerheumatoid arthritis of the spine and osteoporosis ofthe vertebral bodies.

DISCUSSION

From a morphological point of view, there is agreat range of variation in these cases from a thinlayer of calcium deposition in the posterior longi-tudinal ligament to a thick, completely ossified,bony structure in the same anatomical region.The degree of associated spondylotic vertebralchanges varies, but there seems to be a distinctentity of cases with ectopic bone formation withoutany or only minimal vertebral pathology. Ourcases tended to be of two types. In the first group,exemplified by two cases (cases 1, 2), a hefty slabof bone is present in the anterior spinal canal andthere are either no spondylotic changes presentin the cervical spine or only to such a degree asis not unusual among the randomly selected popula-tion of the patient's age.The second group of patients consisted of five

cases (cases 3 to 7) representing ordinary degenerativecalcification within the confines of the posteriorlongitudinal ligament concomitant with otherchanges of marked cervical spondylosis. Althoughit is impossible to distinguish calcification fromossification radiographically a definite difference

FIG. 6. Case 6. Lateral radiograph of the cervicalspine. Marked calcification of the posterior longitudinalligament in thepresence ofmarked spondylosis.

in the appearance of this group from that of thefirst was noticed.Symptomatic ossification in the spinal canal

was first reported in large numbers in patients ofOriental, and particularly of Japanese, extraction(Yokoi, 1963; Onji etal., 1967). However, alater reportis based on patients of Caucasian extraction (Minagiand Gronner, 1969). It still remains to be seenwhether this condition is more prevalent in Japanthan elsewhere. Onji et al. (1967) reported a frequencyof 1-7% among their patients with cervical symptomsand a similar Japanese report by Yokoi (1963)calculated the frequency as being 1 %. At the sametime, the survey of a large number of Cloward'spatients (Onji et al., 1967) of Japanese ancestryfailed to demonstrate such lesions.Our cases, as well as those reported in the liter-

ature, represent patients in later life, with nostriking sexual predilection. Grossly, the bar iscomposed of hard bone as described in case 1,located between the posterior longitudinal ligamentand the dura to which it is tightly adherent. Theseabnormalities have been identified histologicallyas varying from unorganized calcification to fullyformed bone with fatty marrow and Haversiancanals (Onji et al., 1967).

266



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


Ossification in the region of the posterior longitudinal ligament as a cause of cervical myelopathy 267

Although it may be merely another expressionof degenerative change, ectopic bone formationdiffers from the usual secondary spondyloticcalcification because: (1) it exceeds the anatomicallimits of the posterior longitudinal ligament;(2) it is continuous with the vertebral bodies incontradistinction to the posterior longitudinalligament which attaches only to the annulus; (3) itspresence is independent of the degree of spondylosis;and, (4) its predilection in the cervical spine is ata higher level than the usual spondylotic changes.The presence or severity ofneurological symptoms,

particularly those of cord compression, dependson the relative width of the upper cervical spinalcanal and the bulk of the cord. If the sagittal diam-eter of the canal is great and the cord relativelysmall, cord compression might not occur (case 2).On the other hand, formation of a wide layerof ectopic bone in a cervical canal, where the reservespace between the cord and its bony confines issmall, will inevitably lead to severe compression(case 1). Associated bony spur formation encroachingon the intervertebral foramina could be responsiblefor the radicular symptoms, but a posterior dis-placement of the cord with stretching of the nerveroots might be a sufficient explanation. In ourmost pronounced case (case 1), the sagittal diam-eter of the spinal canal at the affected level wasreduced to 11 mm, as compared with the normalof 17-9 mm (16&0 to 200) (Payne, 1959). Whetherthis bone formation is progressive is not known,but eventually the formation of the voluminous slabof bone far exceeds the anatomical limits of theposterior longitudinal ligament itself. This seemsto distinguish it from a mere calcium depositwithin a normal-sized posterior longitudinal ligamentwhich is seen as a manifestation of severe spondylosisof the cervical spine.The aetiology of this ectopic bone formation is

unknown. The factors that induce cdlls to form bonein unusual sites are poorly understood. Calcifiedmatter of some sort seems to be essential (Hamand Harris, 1956). Ectopic ossification may occurpractically anywhere in the body without any knownmetabolic disorder, although it is stated thatectopic bone cannot form in normal connectivetissue. Interestingly, lime salts that deposit in newbone originate from the adjacent bone which thusbecomes rarefied (Oppenheimer, 1942). In neitherof our ectopic bone cases was there any rarefactionof the vertebral bodies, but in most of the secondgroup the vertebral bodies were 'washed out'.Immobilization favours ossification, and this mayexplain why the bar occurs in the high cervicalarea where mobility is less. In case 2, the anteriorspurring, just below the inferior end of the bony

bar, contributed undoubtedly to the immobilityof the cervical spine at this level.An infectious aetiology has been suggested

(Cloward, 1967), implying that pharyngeal infectionproceeds by the paravertebral venous plexus tothe anterior spinal canal and there forms dystrophiccalcification and eventually ossification. Neitherour cases nor those reported elsewhere have shownevidence of infection. Interestingly, ossificationof the lateral paravertebral ligaments with nodegenerative changes in the spine occurs in psoriaticarthritis (Bywaters and Dixon, 1965) in which aninfectious agent (PPLO) has been implicated(Wright and Reed, 1964). Ossification within thespinal canal has not been reported in this disease(Kaplan, Plotz, Nathanson, and Frank, 1963),but the subject is mentioned in passing in theconnection of an infectious agent causing paraverte-bral ligamentous ossification.Trauma has also been suggested as a cause of

the ectopic ossification: small splinter fracturescalcify and join posteriorly. However, this charac-teristically results in the confluence of posteriorosteophytes and is not similar morphologicallyto the bony intraspinal bar. In Schneider's reportof chronic cervical spinal trauma, there was bridgingand calcification in the region of the anteriorlongitudinal ligament but no ossification withinthe spinal canal (Schneider, Papo, and Alvarez, 1962).About half of the cases reported by Onji et al.

(1967) required surgical therapy. In their series,decompr-ssive laminectomy had the best results.Our experience in case 1 attests to the fact thatposterior decompression can be followed by adefinite improvement. The experience of Onjiet al. (1967) with anterior interbody fusion wasdisastrous. This is not surprising because adequatedecompression cannot be carried out by this method.On the other hand, a piecemeal removal of theossified structure through several intervertebralburr holes separated by parts of the vertebral bodiescould hardly be carried out without injuring thealready severely compromised cord or interferingwith its blood supply through the anterior spinalartery.

SUMMARY

Seven cases of calcification and ossification withinthe spinal canal have been described. A distinctionwas attempted between ectopic ossification pro-ducing spinal cord compression as a clinical entityand degenerative calcification in the posterior longi-tudinal ligament as a secondary and usually asympto-matic manifestation of cervical spondylosis. Al-though theyboth may represent a different expression



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from



of the same underlying disease process, theformer should be included in the differential diag-nosis of cervical myelopathy because of its uniqueclinical behaviour. Decompressive laminectomyseems to be the treatment of choice.

REFERENCES

Bywaters, E. G. L., and Dixon, A. St. J. (1965). Paravertebral ossifi-cation in psoriatic arthritis. Ann. rheum. Dis., 24,313-331.

Cloward, R. (1967). Personal communication.Ham, A. W., and Harris, W. R. (1956). Repair and transplantation

of bone. In: Biochemistry and Physiology of Bone. P. 504.Edited by Geoffrey H. Bourne. Academic Press: New York.

Kaplan, D. A., Plotz, C. M., Nathanson, L., and Frank, L. (1963).Cervical spine in psoriasis and psoriatic arthritis. Arthr. andRheum., 6,281.

Minagi, H., and Gronner, A. T. (1969). Calcification of the posteriorlongitudinal ligament: A cause of cervical myelopathy. Amer.J. Roentgenol., 105, 365-369.

Onji, Y., Akiyama, H., Shimomura, Y., Ono, K., Hukuda, S., andMizuno, S. (1967). Posterior paravertebral ossification causingcervical myelopathy. A report of eighteen cases. J. Bone Jt.Surg., 49A, 1314-1328.

Oppenheimer, A. (1942). Calcification and ossification of vertebralligaments (spondylitis ossificans ligamentosa): Roentgen studyof pathogenesis and clinical significance. Radiology, 38,160-173.

Payne, E. E. (1959). The cervical spine and spondylosis. Neuro-chirurgia (Stattg.), 1, 178-196.

Schneider, R. C., Papo, M., and Alvarez, C. S. (1962). The effects ofchronic recurrent spinal trauma in high-diving. A study ofAcapulco's divers. J. Bone Jt. Surg., 44A, 648-656.

Yokoi, K. (1963). Ectopic calcification in the epidural space. Orthop.Surg. (Tokyo), 14,1262.

Wright, V., and Reed, W. B. (1964). The link between Reiter'ssyndrome and psoriatic arthritis. Ann. rheum. Dis., 23, 12-21.

268



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


ossification in the region posterior as a myelopathy · ossification in the region oftheposterior...

Documents