The dif­ferences in clinical, radiological and treatment modalities of cervical intramedul­lary arachnoid cysts and cervical syringomyelia –  report of 12 cases

Overview

Background:
Intramedul­lary arachnoid cysts are rare cystic mas­ses that are usual­ly mixed up with syringomyelia. It is not neces­sarily a complication of any pathologic condition in the spine and/or brain but the etiology and management of this entity remains unclear. Syringomyelia is a disorder that is related to an abnormality of the brain and in most cases, occurs as a complication of other pathologies. The aim of this study was to establish the clinical and radiological characteristics of this intramedul­lary arachnoid cyst and discuss the dif­ferences between intramedul­lary arachnoid cyst and the syringomyelia.

Patients and methods:
A total of 77 patients with intramedul­lary cystic mas­ses were retrospectively as­ses­sed between 2002 and 2012 and 12 of them with intramedul­lary arachnoid cyst were analysed statistical­ly. All patients had initial scans and fol­lowed up with clinical examinations and MRI images and neuro-electrophysiological tests.

Results:
The average fol­low-up duration was 50 months (range 18– 120 months). The mean axial intramedul­lary arachnoid cyst size was 13.26 ± 4.76mm and the mean sagittal size was 23.29 ± 7.95m­m. Neurological examinations were normal in all except one patient who had minor sensory loss due to recent cervical trauma. There were no motor neurological deficits, and all patients remained asymp­tomatic or stable. The intramedul­lary arachnoid cyst size and shape remained unchanged radiological­ly, motor evoked potentials and somatosensory evoked potentials studies revealed normal median and tibial responses initial­ly and at the end of the fol­low-up in all patients.

Conclusion:
It is probable that intramedul­lary arachnoid cyst has dif­ferent radiologic findings and clinical signs/symp­toms from syringomyelia that are as­sociated with dif­ferent primary causes. And also, the treatment and fol­low-up modalities are dif­ferent in intramedul­lary arachnoid cyst.

Key words:
intramedullary arachnoid cyst – syringomyelia – magnetic resonance imaging

Introduction

Spinal arachnoid cysts are very rare nonspecific and usual­ly asymp­tomatic benign lesions. The incidence of spinal arachnoid cysts is low, with most of the cases hav­­ing an incidental dia­gnosis by MRI [1– 2]. Several clas­sification systems have been designed for description of the spinal arachnoid cysts based on anatomical locations and characteristics of the lesions [3– 4]. But the last clas­sification system was reported by Qi et al, as intramedul­lary, subdural extramedul­lary, subdural/ epidural, intraspinal epidural, or intraspinal/ extraspinal, based on the anatomical location and abnormalities observed by MRI, in order to evaluate a large case series and long fol­low-up [5].

Many authors put forward several theories, such as congenital, idiopathic, and secondary to bleeding, inflam­mation, infections, or puncture-related traumas [6– 7]. Especial­ly the main theory for formation of intramedul­lary arachnoid cyst is the misplaced cel­lular remnants that compose the cyst [8]. But the certain etiology of spinal arachnoid cysts is unclear yet.

MRI is the excel­lent imag­­ing modality for evaluat­­ing and delineat­­ing these le­sions [5,9– 11]. On conventional T1-weighted and T2-weighted images, intramedul­lary arachnid cysts can have general­ly similar characteristics as syringomyelia and rarely epidermoid tumours [12– 14]. The intramedul­lary arachnoid cysts are mostly reported as syringomyelia.

In many cases, fol­low-up with MRI and examination is enough. Spinal arachnoid cysts are only treated when symp­toms emerge. Intramedul­lary arachnoid cysts are very rare and only 31 cases have been reported and 10 of them were localized in cervical region [5,8,15– 17]. Literature review shows that these lesions are com­monly seen in the paediatric population more com­monly in the first decade. They are more com­mon in females as compared to males and are located more com­monly in the thoracic region than in the cervical region.

Causes of formation, shapes and radio­logic, clinical findings and treatment stra­tegy of the intramedul­lary arachnoid cysts and the syringomyelia are dif­ferent. Our experience with intramedul­lary arachnoid cysts defined as such in a retrospective case series is presented in as­sociation with a comprehensive review. Dif­ferences between these two conditions are discus­sed with the cur­rent base of evidence.

Patients and methods

A total of 77 patients with cervical intrame­d­­-
ul­lary cysts were retrospectively analysed between 2002 and 2012. Clinical­ly; patient’s symp­toms, neurologic signs, neuro-electro­physiological tests, underly­­ing etiologies were as­ses­sed. Radiological­ly; the shape, location, size, signal intensities on T1-weighte­d and T2-weighted images were evaluated.

Intramedul­lary arachnoid cyst was de­fined as fol­lows:

• a) A wel­l-defined, ovoid, cystic lesion, which is isointense to CSF without contrast enhancement and without any as­sociated septae, nodularity or signal intensity changes in adjacent spinal cord.
• b) No additional pathology, for example, Chiari malformation, spinal cord tumours, any compres­sive lesions, vascular malformations, a tethered cord or any other congenital spinal disraphysm, any history of central nervous system infection, any history of spinal cord trauma (vertebral fracture, penetrat­­ing injury), hydrocephalus, or previous spinal surgery.
• c) Normal neuro-electrophysiological tests (tibial and median nerve soma­tosensory evoked potentials [SSEP] and motor evoked potentials [MEP]) at the time of the first dia­gnosis and in their fol­low-up studies.
• d) No clinical or radiologic progres­sion in at least their three fol­low-up examinations in a frequency of six months.

All medical records of these patients were reviewed and confirmed by two neurosurgeons who have 20 years and 19 years of experience regard­­ing the spine. Changes in clinical symp­toms, physical exams and neuro-electrophysiological tests were identified and documented.

MRI was performed using a 1.5 T MRI system (Gyroscan Intera, Philips Medical Systems, Best, The Netherlands). MRI sequences and parameters were sagittal T1--weighted TSE (FOV: 180, slice thicknes­s [ST]: 3 m­m, TR: 400 ms, TE: 7.4 ms, ACQ matrix: 196 × 145, FA: 90), sagittal T2-weighted TSE (FOV: 180, ST: 3 m­m, TR: 3,500 ms, TE: 120 ms, ACQ matrix: 196 × 140, FA: 90), axial T2-weighted TSE (FOV: 150, ST: 4 m­m, TR: 5,200, TE: 120, ACQ matrix: 176 × 120, FA: 90), sagittal T2-weighted TSE SPAIR (FOV: 180, ST: 3 m­m, TR: 3,700 ms, TE: 120 ms, ACQ matrix: 184 × 136, FA: 90, fat suppres­sion +) and axial T1-weighted TSE SPIR (FOV: 180, ST; 3 m­m, TR: 2,154 ms, TE: 7.8 ms, ACQ matrix: 176 × 121, FA: 90, fat suppres­sion +), sagittal T1-weighted TSE SPIR (FOV: 180, ST: 3 m­m, TR: 400 ms, TE: 7.4 ms, ACQ matrix: 196 × 145, FA: 90, fat suppres­sion +) after gadolinium injection. Two radiologists in consensus who have 17 years and 6 years of experience in dia­g­nostic imag­­ing as­ses­sed all the MRI images includ­­ing the fol­low-up images. Synapse (PACS) Workstation, Version 4.4 (Fujifilm Medical Systems, Stamford, USA) was used in the as­ses­sment of MRI images.

Intramedul­lary arachnoid cysts were defined as a wel­l-demarcated, non-enhanc­­ing cavity within the spinal cord that had the same signal intensity with CSF on T1-weighted and T2-weighted images without any septa, nodule and/ or abnormal signal intensity in the adjacent cord.

Changes in the shape of the intramedul­lary arachnoid cyst, size of the intramedul­lary arachnoid cyst in mil­limetres in the axial and sagittal planes and the number of cervical vertebral levels involved in the sagittal plane were as­ses­sed and recorded from the MRI studies at presentation and fol­low-up. Axial and sagittal T2-weighted images were used to determine the maximum dimensions of intramedul­lary arachnoid cyst.

The first three MRI scans and the last one (a total of four MRI scans for each patients) were included in the statistical analyses. An increase of 20% in the axial and/ or sagittal maximum diameter of intramedul­lary arachnoid cyst in any time was accepted as a progres­sive dis­ease. Statistical analysis was performed us­­ing SPSS statistics 21 for Windows (SPSS, Chicago, IL, USA). Descriptive analyses were presented as proportions for count data and as means with standard deviations (SD) for continuous data and the one-way Anova test was used to compare the means of three or more independent samples. A p value less than 0.05 was considered statistical­ly significant.

Neurologic examinations, imag­­ing studies and neuro-electrophysiological tests were performed at regular intervals (rang­­ing from 6 months to 1 year for MRI and neurological examinations, and 1 year for neuro-electro­physiological tests).

Results

A total of 77 patients were included in the study. Thirty-four (44%) of them were Chiari malformations type 1 or 2, 6 (7.7%) of them were posttraumatic, 7 (9%) of them were tumours (1 of them subependymoma, 3 of them ependymomas and 5 of them were astrocytomas), 8 (10%) of them were secondary to scoliosis and kyphosis, 3 (3.8%) were due to severe intervertebral disc herniation, 3 (3.8%) were due to severe cervical spinal stenosis, 2 (2.5%) were due to post-meningitis arachnoiditis, 1 (1.2%) were due to arachnoid cyst in posterior fos­sa, 1 (1.2%) were due to rheumatoid pan­nus.

Twelve (15.5%) of them were defined as intra­medul­lary arachnoid cyst accord­­ing to the inclusion criterion and they were analysed statistical­ly. The patients’ ages ranged between 21 and 45 years with an average of 33.5 years. Four were males, and 8 were females. The most com­mon present­­ing symp­tom was neck pain. Neurological examinations were normal but one patient who described sensory loss at the right C6 dermatome after recent minor cervical trauma. At the end of the fol­low-up, there were no neurological abnormalities in any patient but the pain without any significant changes continued even though a variety of conservative interventions includ­­ing medications, counsel­ling, physio­ther­apy and exercises. MEP and SSEP studies revealed normal median and tibial responses initial­ly and at the end of the fol­low-up in all patients (Tab. 1). The fol­low-up period ranged from 18 to 120 months (with a mean ± SD of 50 ± 28.06 months).

The first MRI revealed a mean ± SD axial diameter of 13.46 ± 4.94 (range 5.32– 21.0) mm and a mean ± SD sagittal diameter of 23.49 ± 8.30 (range 13.30– 39.20) mm. The second MRI revealed a mean ± SD axial diameter of 13.15 ± 4.75 (range 5.32– 22.4) mm and a mean ± SD sagittal diameter of 23.41 ± 8.41 (range 13.40– 39.80) mm. The third MRI revealed a mean ± SD axial diameter of 13.30 ± 4.95 (range 5.32– 23.0) mm and a mean ± SD sagittal diameter of 23.21 ± 8.13 (range 13.1– 38.7) mm. The last MRI revealed that a mean ± SD axial diame-ter of 13.13 ± 5.02 (range 5.32–23.0) mm and a mean ± SD sagittal diameter of 23.06 ± 8.09 (range 13– 38.0) mm (Tab. 2). There were not any significant dif­ferences of the axial and sagittal diameters between MRI studies (p = 0.99, > 0.05).

Discus­sion

The pathophysiology of the syringomyelia remains controversial. In general, all theories focused on three mechanisms:

1. CSF entrance from the 4^th ventricle;
2. CSF en­trance from the subarachnoid space;
3. extra­cel­lular fluid accumulation [18,19].

The exact mechanism however, is still debated. Syringomyelia may herald the presence of several primary conditions such as Chiari malformations, spinal cord compres­sion and tether­­ing due to previous spinal trauma, infections, spinal cord tumours and vascular malformations. If the syringo-myelia is secondary to a cause, treatment of the underly­­ing disorder is principal. Suc­­ces­s­ful treatment of the primary problem will often reduce or resolve the syringomyelia. The embryogenetic mo­d­­el of syringomyelia postulates pathologic alterations in neurogenesis as the basis of the structural abnormalities that lead to syringomyelia.

Syringomyelia is widely used as a term describ­­ing the presence of a fluid-fil­led cavity oriented in a rostral-caudal axis within the tissue of the spinal cord. This cyst expands and elongates over time, destroy­­ing a portion of the spinal cord at its centre and it expands outwardly. The in­ner floor of the syringomyelia is covered with ependymal or glial cel­ls and is presumed to be fil­led with a derivative of CSF [20]. Syringomyelia is located at the centre of the spinal cord probably in con­nection with the central canal that has an ependymal wal­l [21].

The origin of intramedul­lary arachnoid cysts is not yet well defined. Aithala et al described the first intramedul­lary arachnoid cyst [22]. The relations with dysraphic anomalies of spinal cord with arachnoid cysts were defined later [23]. Gelabert et al described the occur­rence of intramedul­lary cyst without neural tube defects [24]. Accord­­ing to the hypothesis of Fortua and Mercuri, the intramedul­lary arachnoid cysts arise as secondary cystic development of the atypical intramedul­lary arachnoid granulations [25]. The other alternative hypothesis postulated describes the anato­m­­-
ical com­munication between the cyst and subarachnoid space as a one-way valve al­low­­ing the CSF to seep into the cyst and cause the consequent expansion of the cyst [24]. There was no obvious com­munication between the cyst and the spinal canal that was evident due to no CSF leak on Valsalva manoeuvre in our cases.

Syringomyelia is as­sociated with symp­toms and findings such as cervical pain, stif­f­­ness in the back, shoulders, arms, or legs; arm and/ or leg weaknes­s, headaches, and a loss of the ability to feel extremes of hot or cold [26]. Postural changes or Valsalva-like manoeuvres may exacerbate symp­toms, a phenomenon that is consistent with the hydrodynamic theories of the genesis of syringomyelia. These symp­toms may vary with the extent and the location of the syringomyelia within the spinal cord and are progres­sive.

On the other hand, all cases of intramedul­lary arachnoid cysts are recognized incidental­ly. They are usual­ly asymp­tomatic, but rarely become symp­tomatic once the cyst starts compres­s­­ing the cord or nerve roots. The most com­mon present­­ing symp­tom is slowly progres­sive weakness in the limbs because of gradual and continuous enlargement of the cyst. If it becomes symp­tomatic, several surgical techniques were described such as decompres­sive laminectomy, percutaneous drainage, shunt­­ing procedures, com­municat­­ing the cyst to the subarachnoid space, myelotomy and partial or total cyst wall excision for intramedul­lary arachnoid cyst [5,9– 11,27]. Neck pain was the most com­mon and usual­ly the only present­­ing symp­tom in our cases. In almost all cases, the cause of this neck and shoulder pain is myofascial pain syndrome, which is amenable to treatment with physiother­apy and exercises. We did not perform any surgery for intramedul­lary arachnoid cyst in our cases.

The syringomyelia is frequently found in the cervical region at the vertebral column, but sometimes it can extend to the thoracic spine, however, the intramedul­lary arachnoid cysts are frequently limited to only the cervical spine.

MRI is an excel­lent modality for identi­f­ication of the syringomyelia and accom­pany­­ing pathologies [28]. MRI al­lows the demonstration of any cystic lesion in a normal or atrophic spinal cord. Additional­ly, it permits determination of the lower and upper borders of the cyst, determination of any septa in the cysts, detection of as­sociated disorders such as cysts in the posterior fos­sa, tethered cord, Chiari malformation etc., detection the complications of treatments and trauma, study­­ing the fluid circulation within the subarachnoid space and in the cyst by us­­ing flow imag­­ing techniques [29]. T1 and T2 relaxation characteristics of the syringomyelia suggest a non-enhanc­­ing CSF-like fluid [30]. Syringomyelia usual­ly extends to more than two or three levels, whereas intramedul­lary arachnoid cysts are almost usual­ly limited in three or less cervical vertebrae levels on the sagittal images. Syringomyelia was recognized as a lopsided round or fusiform shape [30] in contrast to the intramedul­lary arachnoid cyst, which has an ovoid shape like an American football ball (Fig. 1). The shape of the syringomyelia may be complex with septations (unlike the syrinx) and general­ly involves a portion of the central canal at some level. It has been reported recently that unenhanced MRI includ­­ing sagittal and axial T2-weighted imag­­ing has been demonstrated to dif­ferentiate syringomyelia from intramedul­lary arachnoid cyst as­sociated mass with a 100% sensitivity and negative predictive value by us­­ing imag­­ing features such as nodularity, septations, a spinal cord signal intensity abnormality or a mass adjacent to the intramedul­lary arachnoid cyst [31]. None of these MRI features were recognized in our patients with intramedul­lary arachnoid cyst and this also supports our consideration that the intramedul­lary arachnoid cyst and syringomyelia are dif­ferent entities (Fig. 2, 3).

An intramedul­lary arachnoid cyst is usual­ly a stagnant pathology, whereas syringomyelia is a dynamic pathology that progres­ses over time and can be as­sociated with T2-weighted hyperintensity within the adjacent cord tissue, which is probably due to myelin degeneration, edema or gliosis [29]. The size of the sy­ringomyelia usual­ly increases in the sagittal and axial planes unless surgical intervention is performed [20,21]. Increas­­ing widespread use of Cine-MR flow studies provide additional information about CSF flow velocity and patterns [32,33]. Application of other techniques, such as dif­fusion-weighted imag­­ing and dif­fusion tensor imaging, may eventual­ly provide useful additional data on the intrinsic properties of the neuroaxis.

Electrodia­gnostic findings as­sociated with syringomyelia may play an important role in the quantification of deficits induced by the spinal cord cavity and they are es­sential components of clinical dia­gnostics [34]. SSEPs and MEPs obtained after non-invasive electrical or magnetic stimulation of the motor cortex are sensitive tools for the detection of morphological and functional lesions of the spinal cord pathways [35,36]. SSEP sensitivity for centromedul­lary symp­toms such as hypalgesia or hypesthe-sia is reported to be 54% and 81%, resp. [37]. MEP recordings have been shown to have good cor­relation with motor deficits [38]. In spite of all these SSEP and MEP ab­normalities described in patients with syringomyelia [36,37,39], there were no abnormalities of the SSEP and MEP records at the time of presentation and dur­­ing the fol­low-up in our cases with intramedul­lary arachnoid cyst.

Also, the mean limitation that the small sample size which is limited to only twelve intramedul­lary arachnoid cyst cases, should be taken into account when interpret­­ing the results of this study. Despite the limitations mentioned, the strengths of this study included a relatively long duration of fol­low-up.

The etiological causes were dif­ferent between syringomyelia and idiopathic intramedul­lary arachnoid cysts. The sy­ringomyelia occurs secondary to a primary pathology, but the etiology of intramedul­lary arachnoid cyst is unknown. Also, the clinical signs and symp­toms are dif­ferent in these two conditions. The syringomyelia usual­ly has progres­sive clinical symp­toms and signs whereas the intramedul­lary arachnoid cyst is recognized incidental­ly without any related symp­tom. The intramedul­lary arachnoid cyst has wel­l-defined borders within homogeneous internal signal intensity on MRI images. The intramedul­lary arachnoid cyst shows no progres­sion in size or signal intensity changes over time in contrast with the dynamic mechanism of syringomyelia of which progres­sion is demonstrated in most cases due to myelin degeneration with accompany­­ing gliosis. We created a sum­mary table accord­­ing to the results of our study to help the understand­­ing of this topic by show­­ing the dif­ferences between intramedul­lary arachnoid cyst and syringomyelia (Tab. 3).

Conclusion

It is very important to be able to dif­ferentiate between the intramedul­lary arachnoid cyst and syringomyelia dur­­ing MRI report­­ing because cases of intramedul­lary arachnoid cyst can be managed succes­sful­ly without surgery by us­­ing conservative approaches.

The authors declare they have no potential conflicts of interest concerning drugs, products, or services used in the study.

The Editorial Board declares that the manu­ script met the ICMJE “uniform requirements” for biomedical papers.

Accepted for review: 21. 10. 2017

Accepted for print: 13. 3. 2018

Halil İbrahim Secer, MD

Department of Neurosurgery

Faculty of Medicine

University of Kyrenia Şehit

Yahya Bakır Sokak

Karakum Kyrenia

Cyprus

e-mail: hisecer@yahoo.com