Elsevier

Journal of Clinical Neuroscience

Volume 81, November 2020, Pages 290-294
Journal of Clinical Neuroscience

Clinical study
T2-SPACE imaging of the cauda equina for the assessment of leptomeningeal metastatic disease

https://doi.org/10.1016/j.jocn.2020.10.013Get rights and content

Highlights

  • T2-SPACE can accurately detect nodular lesions of the cauda equina.

  • Its greatest value is clarifying equivocal findings on standard sequences.

  • T2-SPACE improves temporal comparison when following-up known lesions.

  • It may negate the need for IV contrast when following suspected benign lesions.

Abstract

The diagnosis of leptomeningeal metastatic disease (LMD) is frequently challenging and MRI of the spine is an important part of the diagnostic paradigm. We sought to examine the value of adding 3-dimensional, heavily T2-weighted, Sampling Perfection with Application optimised Contrasts using different flip angle Evolution (T2-SPACE) imaging of the lumbar spine to the MRI protocol for patients with suspected LMD. MRI spine examinations including T2-SPACE imaging of the lumbar spine performed for suspected or known LMD were retrospectively reviewed by a neuroradiologist to determine the additional benefit of the T2-SPACE sequence. The accuracy of T2-SPACE was also compared to contrast-enhanced T1-weighted imaging (ceT1WI) and standard T2-weighted imaging (T2WI). 59 patients with T2-SPACE were identified over a 20-month period, 17 having abnormal appearances on ceT1WI, including 12 with appearances consistent with LMD. In eight of these 12 patients, nodules visible on T2-SPACE were visible on T2WI, though T2-SPACE improved the temporal comparison of slowly progressive cauda equina nodules in two cases. In three patients, T2-SPACE identified nodules which were not readily identifiable on T2WI, though were visible on ceT1WI. In one patient, LMD visible on ceT1WI was not appreciable on T2-SPACE or T2WI due to the lack of a nodular component. In six patients, T2WI showed equivocal nodularity, which could be confidently attributed to facet joint arthropathy or a tortuous vessel. In conclusion, T2-SPACE has high sensitivity and specificity for the detection of nodular lesions of the cauda equina and can confidently characterise equivocal findings on standard T2WI.

Introduction

Leptomeningeal metastatic disease (LMD) is an uncommon manifestation of a broad range of neoplasms, including solid organ malignancies, haematological malignancies and primary brain tumours [1]. Most commonly, it occurs late in the course of the disease and is associated with a poor prognosis [1]. The diagnosis of LMD can be challenging, due to the broad range of possible clinical presentations. Symptoms and signs may relate to cranial or spinal nerve dysfunction, be secondary to raised intracranial pressure, or be more general, such as headaches, nausea, vomiting and cognitive changes [1]. Management options include systemic chemotherapy (in particular if there is also uncontrolled systemic disease), intrathecal chemotherapy and radiotherapy (for larger discrete nodular and/or symptomatic lesions) [2].

Traditionally, the diagnosis of LMD has required the identification of malignant cells on cytological assessment of the cerebrospinal fluid (CSF) [3]. Cytology has a high specificity, of over 95%, but the sensitivity is relatively low, generally less than 50% [4]. This makes MRI of the brain and spine an important component of the diagnostic paradigm for suspected LMD [5], [1]. Positive MRI findings are adequate for a diagnosis of LMD in the appropriate clinical context, and do not require cytological confirmation [1]. Older studies have shown that the sensitivity of contrast-enhanced MRI for the diagnosis of LMD is approximately 70% [1], with a specificity of 77–100% [4], [3], [6], [1], though one would expect that the accuracy may have since increased due to technological improvements. The key finding of LMD in the spine is abnormal enhancement along the surface of the spinal cord or the nerve roots of the cauda equina, and this may be linear or nodular.

MRI for the investigation of LMD is typically performed with IV contrast, but unenhanced MRI also frequently demonstrates positive findings [7]. An older study by Moulopoulos et al found that precontrast T1-weighted imaging (T1WI) demonstrated abnormal findings – such as loss of CSF clarity, poor definition of the conus medullaris, thickened and clumped nerve roots, and nodules – in 41 of the 48 patients with leptomeningeal enhancement on contrast-enhanced T1WI (ceT1WI) [7]. Of note, however, all patients in this cohort were known to have LMD [7], and the specificity of some of the more subtle findings – such as loss of CSF clarity or poor definition of the conus medullaris – is debatable in a typical clinical setting. Nevertheless, the presence of multiple nodules would be expected to be a more specific feature of LMD, and the addition of post-contrast imaging does not necessarily provide incremental value, as non-malignant differentials such as nerve sheath tumours would also be expected to enhance.

A recent report has suggested that a Magnetic Resonance myelography (MRM) technique utilising a 3-dimensional, heavily T2-weighted, Sampling Perfection with Application optimised Contrasts using different flip angle Evolution (T2-SPACE) sequence of the lumbar spine, performed in the coronal plane, may provide a useful adjunct to the MRI assessment of suspected LMD [8]. MRM is predominantly utilised for the investigation of intracranial hypotension, based on the detection of cerebrospinal fluid (CSF) leaks. A CSF leak is suspected if streaky T2-hyperintensity extends lateral to the nerve root sleeves, with the confidence increasing with greater transverse extension [9]. The heavy T2-weighting allows the distinction between fluid and normal epidural fat [9], though also provides excellent differentiation between CSF and other tissues, in this case the nerve roots [8], [9].

The potential utility of MRM for the investigation of LMD is based on the detection of nodular lesions of the cauda equina which are not readily visualised or confidently characterised on standard sagittal and axial sequences, which are typically performed with 3–4 mm slices and an inter-slice gap. Detection of nodules is facilitated by the high-resolution volumetric nature of the sequence and the high contrast between CSF and the nerve roots (and their associated pathology) [8]. The lack of significant effects of gadolinium administration means that an inability to detect abnormal linear enhancement of the cauda equina is an expected limitation of this technique. As such, post-contrast imaging remains an important component of the MRI protocol, especially if no abnormal nodularity is demonstrated.

Having since added T2-SPACE with an MRM technique to the MRI protocol for the investigation of LMD at our institution, we sought to review our experience.

Section snippets

Materials & methods

Institutional ethics committee approval was obtained. A list of MRI studies of the spine performed at our institution (a dedicated tertiary oncologic hospital) over a 20-month period (from February 2018 – the time at which this sequence was added – to September 2019) was obtained and examinations including T2-SPACE of the cauda equina were identified. T2-SPACE was routinely included in the MRI protocol for the investigation of known or suspected LMD over this period. The imaging protocol for

Results

T2-SPACE imaging of the cauda equina was performed in 59 patients over this period. This included 30 males (51%) and 29 females, with a median age of 58 years (range 18–85 years). The patients’ underlying cancer diagnoses included a variety of haematological malignancies (n = 14), breast cancer (n = 13), melanoma (n = 11, including one patient with primary meningeal melanoma), lung cancer (n = 9), primary central nervous system (CNS) tumours (two medulloblastomas, two pineal primaries and one

Discussion

T2-SPACE added little incremental sensitivity for the detection of LMD beyond standard ceT1WI, but provided other valuable benefits. Importantly, T2-SPACE was able to confidently characterise equivocal findings on T2WI in several patients. While the reporting radiologist will generally suspect, for example, that an apparent nodule at the lateral aspect of the central vertebral canal reflects volume averaging from degenerative changes, true nodular lesions can also occur in this location.

Conclusion

T2-SPACE imaging of the cauda equina is able to characterise equivocal findings on standard T2WI and improves the detection of nodularity when there is a contraindication to IV contrast administration. There is also value in following-up known nodular lesions of the cauda equina, improving temporal comparison and, in the case of suspected benign lesions, potentially negating the need for IV gadolinium administration at follow-up.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (11)

There are more references available in the full text version of this article.

Cited by (0)

View full text