Tarlov cysts (TC) are generally considered to be incidental radiological findings.  However recent studies suggest their possible association with neurological symptoms such as pain, numbness and urogenital complaints (1). The aim of this study was to explore the relationship between TC and sacral nerve root functions using pelvic neurophysiology tests, and to correlate changes with clinical symptoms and magnetic resonance imaging (MRI) findings. To our knowledge, this is the first study assessing objectively both sensory and motor sacral nerve functions in patients with TC and pelvic symptoms.

Consecutive patients with sacral TC, presenting with at least one symptom related to the pelvic area and undergoing neurophysiology testing, participated in a cross-sectional review of symptoms using validated questionnaires. Bladder, sexual and bowel symptoms were objectively assessed using the following validated questionnaires: the Urinary Symptom Profile (USP) to assess stress urinary incontinence (SUI), overactive bladder (OAB) and low stream symptoms, the Constipation Scoring System (CSS)13 and the Arizona Sexual Experience Scale (A-SEX) to identify sexual dysfunction. 
 Findings of pelvic neurophysiology (pudendal sensory evoked potentials (SEPs), S2 and S3 dermatomal SEPs, external anal sphincter electromyography (EAS EMG)) and urodynamics testing were collected retrospectively. MRI of the lumbosacral spine acquired closest to the date of the neurophysiology investigation was reviewed. Relationship between neurophysiology, MRI findings, and patients’ symptoms were assessed using Fischer and ANOVA tests.

Sixty-five females were included (mean age 51.2±12.1y). The commonest symptom was pain (92%). Urinary (91%), bowel (71%) and sexual (80%) symptoms were also frequently reported. Among the patients reporting lower urinary tract symptoms, 82% reported storage and 58% voiding symptoms. Thirty-seven patients (57%) had abnormal neurophysiology findings reflecting sacral root dysfunction. Twenty-five (42%) patients had abnormal dermatomal SEPs, 17 (26%) had abnormal pudendal SEPs, and 9 (16%) had abnormal EAS EMG. No association was seen between MRI findings (size, location of the cysts, severity of compression) and neurophysiology. 
A negative association was found between neurophysiology abnormalities and both a higher OAB USP sub-score (p<0.01) and reported urgency urinary incontinence (p = 0.03). Similarly, a negative association was found between neurophysiology abnormalities and a higher SUI USP sub-score (p=0.02). A low-stream USP sub-score was higher in patients with abnormal neurophysiology, but this did not reach significance (p=0.17).
No association was found between uroflowmetry findings – morphology of the curve, elevated post void residual or reduced flow rate – and neurophysiology findings, concomitant medication use, or the number of cysts. Only twenty-two patients underwent multichannel urodynamic study. A negative association was observed between detrusor overactivity and abnormal neurophysiology results (p<0.01).
No significant association was found between bowel and sexual symptoms and neurophysiology findings. All results related to urogenital and bowel symptoms and neurophysiology are reported in Table 1.

To our knowledge, this is the first study to objectively evaluate the impact of TC within the sacral canal on both sensory and motor nerve root functions. Nearly 60% of our cohort had abnormal neurophysiology findings. However, these did not correlate with the number, size and location of the cysts. Moreover, the negative association between abnormal neurophysiology and urinary incontinence suggests that these symptoms are not related to TC. 

The commonest abnormalities were seen on SEP studies, suggesting that the sensory pathways are more susceptible to injury. This is possibly related to the smaller size and lesser degree of myelination of sensory fibres compared to motor fibres and is consistent with the presumed pathophysiological mechanism underpinning the clinical course of lumbar or cervical nerve root compression in disc herniation where the initial symptom is pain (usually without neurophysiological abnormalities) followed by sensory disturbances and finally motor deficits. Furthermore, TC are preferentially related to the dorsal rather than ventral nerve roots and this may account for the greater prevalence of sensory symptoms and signs and neurophysiology abnormalities (2).

The absence of correlation between the neurophysiological findings and the number of cysts, their size, or their location, indicates that MRI is an unreliable predictor of sacral nerve dysfunction and that instead, pelvic neurophysiology tests provide more useful information as a diagnostic tool in the assessment of neurological damage in patients with symptomatic TC.

No association could be established between abnormal neurophysiology findings and voiding dysfunction expected in lesions affecting the sacral spinal nerve roots. Different explanations can be advanced for this. Firstly, the neurophysiology abnormalities encountered in our cohort mostly affected the sensory innervation suggesting dorsal root involvement, whereas detrusor contraction during voiding is mediated by the parasympathetic innervation derived from the sacral spinal cord segments S2,3,4 travelling through the ventral sacral roots. Furthermore, changes were more often seen unilaterally and damage to a single nerve root or even several nerve roots unilaterally, may not result in significant dysfunction of detrusor functions which receives its innervation bilaterally. Confounding factors such as medications, age and concomitant co-morbidities such as pelvic organ prolapse could have also contributed to voiding dysfunction.

A high prevalence of storage symptoms was reported in our cohort (82%). However, contrary to the literature, the results of this study demonstrate a negative correlation between findings of sacral nerve root injury and reports of OAB symptoms, findings of detrusor overactivity on urodynamics, and SUI (3). OAB symptoms and detrusor overactivity are therefore unlikely manifestations of TC, and this is consistent with the expected pattern of bladder dysfunction for lesions affecting the sacral nerve roots.
Bowel symptoms such as constipation were prevalent although no correlation was observed with findings of sacral root injury. Similar to the lower urinary tract, the lower bowel receives innervation from multiple sacral roots, explaining the absence of correlation, and the wide use of medications such as opiate painkillers in our cohort may have contributed to bowel symptoms.
A substantial number of women reported sexual dysfunction, and given the location of the cysts, it is plausible that sacral TC impact various sexual functions such as genital sensation or the ability to achieve orgasm through injury to the somatic or splanchnic innervation in the sacral roots. However no association with sacral root injury could be established and a future study including an age-matched control population is required to explore this association further and the role of confounding factors such as peri- and post-menopausal changes and pain on sexual performance.

This study demonstrates objective neurological impairment of sacral root functions in patients with symptomatic sacral TC, particularly of the sensory innervation. Although women with TC report a high rate of bladder, sexual and bowel symptoms, urinary incontinence is unlikely to reflect TC-related nerve damage.

Hulens M, Rasschaert R, Bruyninckx F, et al. Symptomatic Tarlov cysts are often overlooked: ten reasons why-a narrative review. Eur Spine J Off Publ Eur Spine Soc Eur Spinal Deform Soc Eur Sect Cerv Spine Res Soc. 2019;28(10):2237-2248. doi:10.1007/s00586-019-05996-1
Tarlov IM. Spinal perineurial and meningeal cysts. J Neurol Neurosurg Psychiatry. 1970;33(6):833-843. doi:10.1136/jnnp.33.6.833
Baker M, Wilson M, Wallach S. Urogenital symptoms in women with Tarlov cysts. J Obstet Gynaecol Res. 2018;44(9):1817-1823. doi:10.1111/jog.13711