• 
• 

Three Part Question

In [patients with low back pain], does [urinary dysfunction, compared to other symptoms and signs] accurately indicate [the presence of cauda equina syndrome]?

Clinical Scenario

A 40- year-old woman presents at the ED with complaint of severe lower back pain of 2 days duration after carrying a heavy load. She is able to ambulate with no sciatica but reports having urinary incontinence. Physical examination is unremarkable with no neurological deficits in limbs and intact perianal sensation with good anal tone. You wonder whether her urinary symptoms are indicative of cauda equina syndrome.

Search Strategy

Medline from 1975 to October 2013
("cauda equina"[All Fields] AND (("urinary"[All Fields]) OR ("urinary bladder"[MeSH Terms] OR ("urinary"[All Fields] AND "bladder"[All Fields]) OR "urinary bladder"[All Fields] OR "bladder"[All Fields])) AND (("diagnosis"[Subheading] OR "diagnosis"[All Fields] OR "diagnosis"[MeSH Terms]) OR ("physical examination"[MeSH Terms] OR ("physical"[All Fields] AND "examination"[All Fields]) OR "physical examination"[All Fields] OR "examination"[All Fields])) AND (("1975/01/01"[PDAT] : "2013/12/31"[PDAT]) AND "humans"[MeSH Terms] AND English[lang]))

Search Outcome

315 papers were found of which 4 were relevant. These 4 papers are summarised in the table below.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Bell et al. 2007 UK	23 patients with a suspected diagnosis of CES referred to a single tertiary neurosurgical referral centre. All underwent MRI lumbar spine.	Prospective cohort study	5 out of 23 (22%) urgent MRI scans were positive.	Only 22% of patients assessed by a trained clinician had positive MRI scan for CES	Small study though prospective. If neurosurgeon was confident patients did not have CES they were not imaged. Post-void bladder scan not used. Inter- and intra-observer error not studied. Not stated if interpretation of MRI was blinded to patient’s history or clinical examination.
			The diagnostic accuracy of urinary retention, urinary frequency, urinary incontinence, altered urinary sensation and altered perineal sensation were 0.57, 0.65, 0.61 ,0.65 and 0.60 respectively.	No individual sign/symptom is able to accurately predict CES. All patients with new onset urinary symptoms in context of back pain/sciatica should have urgent MRI.
Domen et al. 2009 Netherlands	58 patients with a suspected diagnosis of CES referred to the Neuro department at a single centre. All underwent MRI lumbar spine.	Retrospective study	8 out of 58 (13%) urgent MRI scans were positive. Of the 8 patients with CES, 7 (87.5%) had urinary retention. Urinary retention was present in 30 (60%) of patients without CES.	Agree with Bell et al to do MRI scan for every patient with new onset urinary symptoms in the context of back pain/sciatica.	Retrospective study -case records may be incomplete. Post-void bladder scan not done for all patients with CES. Inter- and intra-observer error not studied. Not stated if interpretation of MRI was blinded to patient’s history or clinical examination.
			Patients with 2 or more of the following: bilateral sciatica, subjective urinary retention, rectal incontinence were more likely to have CES on MRI. OR=48.00 (95% CI= 3.30–697.21), p=0.04.	PVRU is likely to be the most promising diagnostic tool for CES. Future prospective studies in CES should incorporate PVRU scans.
Rooney et al. 2009 UK	66 patients with suspected CES referred to the same neurosurgical unit as in the Bell study who underwent MRI lumbosacral spine.	Retrospective study	34 out of 66 (52%) urgent MRI scans showed structural pathology (not necessarily CES). 48% had a normal scan.	Patients with no structural cause of their symptoms are just as likely to present with symptoms suggesting CES.	Retrospective study – case records incomplete. 25% of potential subjects (32 out of 121) were excluded due to incomplete case records. If neurosurgeon was confident patients did not have CES they were not imaged. Use of discharge summary data may have inflated the numbers of patients with CES. No MRI definition of CES given.
			Of those with normal scans, 18 (59%) had weakness, 17 (57%) had saddle numbness, 24 (80%) had leg numbness, 13 (54%) had urinary incontinence and 9 (53%) had urinary retention. Percentages are derived from varying denominator due to incomplete case records.	This group of patients with “pseudo-CES” may have symptoms with a functional origin. More research is needed to define this group of patients, their prognosis and treatment.
Balasubramanian et al. 2010 UK	80 cases with suspected CES referred to on-call spine surgery team at a tertiary referral centre. All underwent MRI lumbar spine.	Retrospective cohort study	15 out of 80 (18.8%) urgent MRI scans were positive.	Clinical assessment alone cannot reliably predict the presence or absence of CES.	Retrospective study – case records incomplete. Post-void bladder scans not done routinely. Inter- and intra-observer error not studied. Not stated if interpretation of MRI was blinded to patient’s history or clinical examination.
			Urinary retention present in 14/79 patients (3/14 had CES), urinary incontinence present in 33/69 patients (6/33 had CES). Denominator changes from 80 as not all patients had symptoms documented.	No symptom or sign has absolute predictive value in excluding CES. All patients in whom CES is suspected clinically should have emergent MRI.

Comment(s)

There is lack of a universal definition of clinical cauda equina syndrome (CES) as well as what it means to have CES on MRI. Much of the existing literature on CES and urinary dysfunction comes from studies that were done to answer the question of how timing of surgery in CES affects long term prognosis of bladder or bowel function, not to diagnose CES based on clinical features. Post-void bladder scan as an objective test for bladder function is a promising diagnostic tool but no studies have been conducted on its accuracy nor has it been demonstrated to be useful in the absence of self-reported urinary symptoms.

Clinical Bottom Line

There is no one symptom or sign which can accurately diagnose cauda equina syndrome (CES) in patients with low back pain. Self-reported urinary dysfunction did not have a high sensitivity or specificity with regards to MRI evidence of CES. At the current time, the consensus is that whenever new onset urinary symptoms are reported In the context of back pain and sciatica, an emergent MRI scan should be done to exclude CES.

References

1. Bell DA, Collie D, Statham PF. Cauda equina syndrome: what is the correlation between clinical assessment and MRI scanning? Br J Neurosurg 2007;21(2):201–203.
2. Domen PM, Hofman PA, Santbrink H van. et al. Predictive value of clinical characteristics in patients with suspected cauda equina syndrome. Eur J Neurol. 2009;16(3):416–419.
3. Rooney A, Statham PF, Stone J. Cauda equina syndrome with normal MR imaging. J Neurol. 2009;256(5):721–725.
4. Balasubramanian K, Kalsi P, et al. Reliability of clinical assessment in diagnosing cauda equina syndrome. Br J Neurosurg. 2010;24(4):383–386.