Introduction

Traumatic injuries of the thoracolumbar spine can result in conus medullaris syndrome (CMS) or cauda equina syndrome (CES). Fifteen percent of spinal cord injuries (SCI) occur in the thoracolumbar spine.1, 2 In ~10–38% of these injuries, the conus medullaris (CM) or cauda equina (CE) is involved; this percentage increases in case of a fracture dislocation.2 The clinical symptoms of patients with CMS or CES include the following: low back pain; lower limb weakness; perineum or saddle anesthesia; and bowel and/or bladder dysfunction. Muscle weakness is generally mild,3, 4 and lower limb weakness associated with CES is predominantly asymmetrical in patients with an incomplete injury.5, 6 The CMS and CES are assumed to be separate clinical entities, but there is no clear definition of neurological symptoms and level of injury. From a purely anatomical perspective, CMS and CES are separate structures, but clinical distinction between incomplete CMS and CES based upon neurological examination alone is difficult. Cadaveric and magnetic resonance imaging (MRI) studies from 1894 to the present show that the anatomical level of the CM ranges from the lower third of vertebrae Th11 to the upper third of L3, with mean on level L1 (Figure 1).7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17

Figure 1
figure 1

The position of the conus medullaris according to the literature. A full color version of this figure is available at the Spinal Cord journal online.

Full size image

Jefferson was the first author to report compression injury to the CM and the CE. He described that injury to vertebral level Th12 or L1 might affect the CM. In lower injuries, the CE is affected.18 Veraguth was the first author to describe the diagnosis and surgical treatment of a patient with a conus–cauda syndrome.19 In 1944, French and Payne described clinical symptoms that were caused by compression of the CE and named this as ‘cauda equina compression syndrome’.20 From 1982, CMS and CES were included in the International Standard for Neurological and functional Classification of Spinal Cord Injury (ISNCSCI) by the American Spinal Injury Association (ASIA), but were recently omitted because of the absence of a clear definition.21 The CM includes both upper and lower motor neurons, whereas the CE consists solely of lower motor neurons.22 It is assumed that the regenerative capacity in lower motor neuron lesions is superior to upper motor neuron lesions, and therefore the functional outcome of CES could be better than the functional outcome of CMS.23 For SCI trials, both CMS and CES lesions are often combined as ‘conus–cauda lesion’, although the outcome between these two syndromes might differ. According to Kirshblum et al.,24 the level of SCI can be determined by systematically examining the myotomes and dermatomes, whereby injury of the CES may involve only lower motor neurons, and injury to the CM may manifest as a mixed upper and lower motor neuron injury. A clear definition of both syndromes is required to determine functional outcomes, guide treatment decision-making and predict complications.25

The aim of this review was to search for the exact definitions of CMS and CES in trauma literature, including the vertebral level of injury, identify neurological symptoms and differentiate between the two syndromes.

Materials and methods

This study was designed and performed according to the Preferred Reporting Items for Systematic reviews and Meta-Analysis (PRISMA) statement.26

To answer the question what is the definition of CMS or CES in the trauma literature, a comprehensive electronic database search was performed in Medline (PubMed), Cochrane and Embase by an independent experienced librarian. All papers published up until June 2016 were included. Papers were identified by the following key words: CM, CE, spinal cord syndromes, thoracolumbar junction, thoracolumbar fracture, SCI and trauma. Papers written in English that involved patients with traumatic CMS, CES, traumatic thoracolumbar SCI and included 20 adult patients or more were included. The focus was to find the definition of the CMS and CES; therefore, it was not suitable to criticize the design of the articles. However, only original papers, no reviews, were accepted. Patients with other spinal deficits were excluded. Because of anatomical diversity, articles concerning children and animals were excluded.

Titles and abstracts were first screened by two investigators in June 2016 (EB and RB). Thereafter, the full texts of the eligible articles were screened. Reference lists were screened for missing articles (EB and BS). If the anatomical level or neurological description of the CMS or CES was reported, articles were included.

Because of the narrative design of this study, quality of the included studies and risk on bias were not assessed.

Results

A total of 1046 articles were initially identified as potentially interesting. Thirty duplicate articles were excluded. After title and abstract screening of 1016 articles, 75 articles were available for full-text analysis. Seven articles reported an anatomical or neurological definition of traumatic CMS or CES.27, 28, 29, 30, 31, 32, 33 In addition, 7 articles were found in reference lists,4, 34, 35, 36, 37, 38, 39 which provided a total of 14 articles for qualitative analysis (Figure 2). Most articles were retrospective evaluations of single-center studies. The period of published articles spanned from 1982 to 2010.

Figure 2
figure 2

Flow chart of eligible articles.

Full size image

Definition of traumatic conus medullaris syndrome

A compilation of neurological symptoms or vertebral level of the CMS was defined in 14 articles (Table 1). Generally, the vertebral level was used to determine the level of the CM.4, 27, 29, 30, 31, 32, 34, 35, 37, 39 A fracture (burst fractures and/or distraction dislocation fractures) at level Th12 and L1 was repeatedly associated with CMS. In one article, an affected CMS was observed distally from injury to vertebral level Th10,35 and two articles described a CMS from an injured level Th11.32, 38 The termination level of CM was associated with vertebral levels L1 and L2. Neurological symptoms in patients with CMS were described by McAfee et al.,28 Clohisy et al.33 and Willén et al.36 The diagnosis of CMS in the article of McAfee et al.28 was based on clinical findings of a neurogenic bladder and loss of third, fourth and fifth sacral nerve root sensation. In the article of Clohisy et al.,33 the CMS was assessed by clinical evaluation of rectal tone, perianal sensation and the ability to spontaneously void. None of their patients had a complete lesion. Willén et al.36 described a CM lesion when there was saddle anesthesia, bladder and/or bowel disturbances or anal problems but no motor impairment in the lower limbs. So, all refer to an acute loss of sensation in the saddle region, bladder and/or bowel dysfunction and sexual dysfunction after trauma, without muscle weakness in lower extremities.

Table 1 The definitions of CMS found in literature
Full size table

Definition of traumatic cauda equina syndrome

Seven articles were identified that described traumatic CES (Table 2). Most studies identify vertebral level (L2) as the cranial border of the CE.29, 31, 32, 38, 39 Podnar described pathology caudal from level L1 to L2 to result in CES.4 Only Willén et al.36 described the neurological symptoms of a traumatic CES—signs of nerve root lesion without saddle anesthesia, bladder and/or bowel dysfunction or sexual dysfunction.

Table 2 The definitions of CES found in literature
Full size table

According to Podnar, the definition of a traumatic CMS and CES is based on the following: (1) a history of a causal event followed by lower limb weakness, bladder and/or bowel or sexual dysfunction; (2) clinical examination—reduced sphincter contraction, sacral sensory loss or, in men, a diminished penilocavernosus reflex; (3) signs of denervation on electromyographic; (4) decreased sacral reflexes; and (5) radiological findings of lumbosacral pathology. In his article, Podnar could not differentiate the CMS and CES based on neurological outcomes because in 20 patients with a CMS, additional root lesion could not be expelled. Patients with a lesion at Th12–L1 were assumed to have a CMS; pathology at L1–L2 intervertebral level and below was associated with a CES.4

Mixed CMS/CES cases were described with vertebral lesions between Th10 and L1, whereby no differentiation was made based on neurological function (Table 3).4, 27, 29, 32, 36, 38

Table 3 Neurological symptoms of CMS and CES
Full size table

Discussion

The articles included in this review indicate that the definition of both traumatic CMS and CES in the acute phase is based on radiological findings. A distinction can be derived by level of injury whereby trauma to vertebral level Th12–L2 will likely cause a CMS, whereas damage to vertebral level L3 and below will cause a CES. However, in the traumatology literature a distinction is not made on neurological examination in the acute setting. Patients with CMS tend to have symmetric sensory–motor deficits, whereas patients with CES tend to have more asymmetric sensory–motor deficits.4, 22, 28, 33, 36 These findings correspond with previous reviews by Wagner and Jagoda,3 Orendacova et al.5 and Fraser et al.6 Our review was established from trauma literature whereby the diagnosis was based on motor and sensor examination and radiology in the acute phase. Bladder function is an important clinical sign to distinguish between CMS and CES. However, we did not find any literature dealing with this topic in the acute setting. Most of these patients already have an indwelling urinary catheter, and further investigations on this topic are not warranted in the emergent situation. We did not find data of additional physical or neurophysical examination directly after trauma to specify the diagnosis.

The anatomical level of the CM was described by Thomson in 1894.17 In his report of collective investigations of 198 cadavers, the average termination of the CM was situated at vertebral level L1. More recent cadaveric and MRI studies have shown the same results, indicating that the tip of the CM is situated at the (middle/lower third of) vertebral level L1 (Table 1). There is a sex difference in the mean termination level of the CM, whereas the spinal cord in females tends to be longer compared to that in males—L1–L2 vs Th12–L1 intervertebral space.7, 11, 13, 14 However, only in the study of Demiryürek et al.13 is this difference significant. In newborns and infants up to the age of 12, the mean termination level of the tip of the CM is at L2–L3. The mean termination level of the CM in adolescents and early adults (age 13–29) is at L1–L2.10 Anatomy studies support the assumption, derived from the trauma literature in this review, that the tip of the CM is situated between Th12 and L2. It should be noted that the individuals in our included articles, as well as those in the anatomy studies, did not have any spinal deformities.

The minimal number of patients who were reported in a study to be included might be considered a limitation. However, while designing the study, we assumed that the quality of the results of studies reporting on very few patients can be questioned. Furthermore, we are confident that inclusion of a small case series will not alter the conclusions of our study. Therefore, we did not introduce them after the results were obtained. Another flaw could be the focus on traumatic lesions. We were interested in traumatic CMS and CES because we assumed that there is no differentiation between these syndromes in the acute phase. Therefore, we have limited our search to traumatic CMS and CES, despite the fact that more causes are known for these syndromes—for example, tumor and hernia nucleus pulposus. Moreover, a major difference between a traumatic SCI and non-traumatic cases is the acute moment of injury to the spine. Because of acute compression, secondary ischemia, hemorrhage and/or edema may occur. Therefore, its course can be different from non-traumatic CMS/CES. The articles were retrospective single-center studies, presenting neurological outcomes after surgery. We used these articles to investigate which definition of the CMS and the CES was used. The low quality of the studies is a matter of concern. However, as CMS and CES were only descriptions of the subject of research and not of an outcome itself, we are convinced that these were reliable and did not affect our results and conclusions.

The strength of this study arises from our search for combined neurological, radiological and anatomical studies to find differences between the CMS and the CES. Until now, the diagnosis of CMS and the CES was based on non-specific neurological symptoms (low back pain, lower limb weakness, perineum or saddle anesthesia and bowel and/or bladder dysfunction) in cases with injury to the lower thoracic or lumbar spinal cord. X-ray, computerized axial tomography and MRI scans seem to be the deciding factor for the diagnosis. However, besides the initial level of injury, it is important to recognize the syndrome at an early stage to predict neurological recovery,40 especially as the prognosis of the CES might be superior compared to CMS.23 From the standpoint of decision-making and informing the patient and family, it is important to minimize ambiguity about these syndromes. To add the ISNCSCI gradation to our results from the literature (Table 4), a clear cutoff point for the border between CMS and CES could be derived. This addition is necessary in future research to examine the hypothesis that CES patients tend to have a better functional outcome.

Table 4 Proposal for the definition of the CMS and the CES
Full size table

Conclusion

A CMS may result from a trauma to vertebra T12–L2 in combination with neurological impairment in dermatomyotomes Th12–S5. The most cranial ISNCSCI level for CMS is Th11. A trauma to vertebra L3–L5 with neurological impairment of the nerve roots is always a CES in patients without pre-existing spinal deformities. The most cranial ISNCSCI level for CES is L2. MRI is the gold standard for the discrimination between CMS and CES. These findings are necessary to distinguish the CMS from the CES, whereby the differences in functional outcome between these patient groups can be investigated Figure 3.

Figure 3
figure 3

Schematic presentation of the conus medullaris and cauda equina. A full color version of this figure is available at the Spinal Cord journal online.

Full size image

Data archiving

There were no data to deposit.