Cauda equina syndrome (CES) is a rare condition arising from severe compression of the cauda equina to cause such symptoms as saddle and/or genital sensory disturbance and/or sexual, bladder and bowel dysfunctions ^[1]. Around 50%-70% of patients have urinary retention (CES-R), and 30%-50% have an incomplete syndrome ^[2]. In complete CES is characterized by partial urinary difficulties including weak urinary sensation, loss of desire to void and poor urinary stream, and the patients often have unilateral or partial saddle and genital sensory deficits with retained trigone sensation. CES-R (or complete CES) is characterized as painless urinary retention and overflow incontinence with extensive saddle and genital sensory deficits and deficiency of trigone sensation^[2].

Lumbar disc herniation (LDH) is a common cause of compression of the lumbar nerve root or cauda equine, resulting in such symptoms as low back pain, sciatica, leg weakness and sensory alteration. For patients with CES caused by LDH, prompt decompression is required to avoid permanent function loss, and 48 h or even 24 h after the onset of CES is thought to be the golden time for surgical intervention ^{[3, 4]}. But according to some reports, delayed decompression (mean 3.7 days) after the onset of CES also achieved good outcomes ^[5-7]. The consequences of further delayed decompression (by weeks or even months) remain undefined. In this study, we analyzed the long-term outcomes of 35 patients with CES secondary to LDH who underwent delayed surgical decompression and evaluated the long-term consequences of delayed surgical intervention.

We reviewed the data of consecutive patients with CES secondary to LDH (CES-LDH) treated in Nanfang Hospital from 2001 to 2015. The diagnosis of CES-LDH was established based on the following criteria: (a) presence of symptoms of the bladder with or without sexual or bowel dysfunctions; (b) abnormal muscle strength and sensation related to the compressed nerve, alterations of saddle sensation and anal squeeze pressure, and decreased penilo-cavernosus reflexes (in men); (c) sexual dysfunction in men in terms of morning erection revealed by questionnaire; and (d) severe disc herniation or spinal canal stenosis confirmed by magnetic resonance imaging (MRI) or computed tomography (CT).

In this study, we made the diagnosis of complete CES when the patients had any of the following symptoms: urine retention or incontinence, uncontrolled defecation, and penile erectile dysfunction. Incomplete CES was defined as partial loss of functions of the bladder, bowel or sexual activities.

The follow-up interviews were conducted in a cross-sectional manner, in which the patients were surveyed for their current general life, motor and sensory functions of the lower limbs, and urinary, bowel and sexual functions. Six patients with complete CES before operation received urodynamic analysis both before and after operation.

Out of 42 interviewed patients with the diagnosis of LDH-CES, 35 responded to the survey, including 18 patients who completed the survey in the physician's office and 17 patients receiving the interview on the phone. The 35 patients had a mean age of 42.8 years (ranging from 18 to 60 years) at the time of operation and the mean follow-up time was 43.0 ± 28.9 months (3-110 months).

A diagnosis of complete CES was established in 12 of the patients, including 4 patients requiring catheterization for urine drainage and 8 patients with loss of control of urination or defecation, or both; 5 patients did not have morning erection and could not finish sexual intercourse. Physical examination of the patients revealed multiple positive signs (Tab. 1). Eleven of the patients had the history of low back pain. In 5 patients, CES symptoms occurred immediately after chiropractic manipulations (type Ⅱ CES according to Tandon and Sankaran)^[8], and in 1 patient, uncontrollable urination and defecation occurred after a long motorcycle ride on rough roads. Eleven patients had LDH at the level of L_4/5, including 1 patient with also L₅/S₁ LDH; LDH at L₅/S₁ was found in 1 patient.

表 1(Table 1)

Table 1 Results of functional examinations in 12 patients with complete CES before operation Patient No. Age (year) Perineal sense Bladder empty Bowel control Penile erection Spinal level Motor (M) and sense (S) in legs* Time of operation 1 35 Lost Catheterized Incontinence Null L4/5 Complete Loss 2 months 2 35 Lost Catheterized Constipation Null L4/5 Intact 9 days 3 37 ↓↓↓ Incontinence Incontinence Null L4/5 ↓↓ 3 months 4 40 Normal Incontinence Incontinence Null L4/5 ↓↓↓ 2 weeks 5 35 ↓↓↓ Catheterized Constipation Null L4/5 ↓↓ 3 weeks 6 41 Lost Incontinence Incontinence NA L4/5 M: Ⅳ; S: Lost 6 weeks 3 days 7 49 ↓↓ Incontinence Incontinence NR L4/5 M: Ⅰ; S: ↓↓ 7 days 8 50 ↓↓ Incontinence Constipation NR L4/5 M: Normal; S:↓↓ 11 days 9 60 ↓↓↓ Incontinence Incontinence NA L4/5 M: Ⅳ; S:↓↓↓ 1 months 5 days 10 49 Lost Catheterized Constipation NR L4/5, L5/S1 M: Ⅲ; S: ↓↓↓ 3 days 11 45 lost Incontinence Incontinence NR L4/5, L5/S1 M: Normal; S:↓↓↓ 2 days 12 38 ↓↓↓ Incontinence Constipation NA L5/S1 Normal 2 months 6 days ↓, ↓↓, and ↓↓↓ indicate slight, moderate and severe reduction, respectively. *The roman numbers indicate the grade of muscle strength. NR: Not recorded; NA: Not applicable.

Table 1 Results of functional examinations in 12 patients with complete CES before operation

The 12 patients with complete CES received delayed surgical decompression from 2 days to 3 months (mean 4.1 weeks) after the onset of symptoms. Full surgical decompression was performed in 11 cases in our department by senior surgeons specialized in spinal surgery; 1 patient was transferred to our department after operation in a local hospital with poor recovery, but the decompression was satisfactory as confirmed by MRI scan. The patients reported good recovery or significant improvement from 3 days to 20 months after operation (Tab. 2). In this group, 6 (50%) patients reported good recovery (2 with full recovery and 4 only had sensory alterations in the saddle area or lower limbs), and 6 patients still had sense deficits in the saddle area and difficulties in bladder or bowl empty or both, but none of them required catheterization in spite of occasional incontinence of defecation or urination.

表 2(Table 2)

Table 2 Results of functional examinations in 12 patients with complete CES after decompression Patient No. Time after operation (month) Perineal sense Bladder empty Bowel empty Penile erection Motor (M) and sense (S) in legs* 1 96 ↓↓↓ Often incontinence occ. incontinence Normal M:Ⅳ; S:↓↓ 2 28 ↓↓ Normal Normal Normal S: ↓ 3 14 Normal Normal Normal Normal Normal 4 12 Tingling Normal Normal Normal Normal 5 50 Normal Normal Normal Normal Normal 6 33 ↓↓ occ. incontinence occ. incontinence NA M: Ⅲ; S:Tingling 7 33 Normal occ. incontinence Constipation Normal M: Ⅲ; S: Normal 8 23 Tingling Normal Constipation Normal M:Normal; S: ↓ 9 3 ↓ Incontinence Incontinence NA M:Ⅳ; S:↓↓ 10 25 ↓↓ occ. incontinence occ. incontinence Normal M:Ⅲ; S: ↓ 11 30 ↓↓ occ. incontinence constipation Normal M:Normal; S:↓↓↓ 12 96 ↓ Normal occ. constipation NA Normal ↓, ↓↓, and ↓↓↓ indicate slight, moderate and severe reduction, respectively. Occ: Occasional; NA: Not applicable.

Table 2 Results of functional examinations in 12 patients with complete CES after decompression

Twenty-three of the 35 patients were diagnosed to have incomplete CES before surgery. Full decompression was performed at 2 days to 7 months (mean 5.5 weeks) after the onset of symptoms (including numbness or tingling in the saddle area, reduced erectile time or sexual desire, slow start and prolonged void time of urination, constipation, and reduced pressure for defecation). Thirteen of the patients had L_4/5 LDH, 8 patients had L₅ or S1 disc herniation, 1 patient had herniation at both levels, and 1 patient had herniation at L_3/4 and L_4/5. In all the 23 patients, the condition did not evolve into complete CES before operation (Tab. 3). Nine of the patients received chiropractic manipulation before the onset of CES. In follow-up, 19 (83%) patients had full recovery, and 4 (17%) patients still had mild sensory abnormalities in the saddle area or the lower limbs (Tab. 4).

表 3(Table 3)

Table 3 Results of functional examinations in 23 patients with incomplete CES before operation Patient No. Age (year) Perineal sense Bladder empty Bowel empty Penile erection Spinal level Motor (M) and sense (S) in legs* Timing of operation 1 41 Tingling ↓ ↓↓ ↓ L5/S1 M: Normal; S: ↓↓ 5 days 2 56 Normal ↓↓ ↓ ↓ L4/5 M: LⅠ; S: ↓ 1 month 3 47 Tingling ↓ Normal NR L4/5/S1 M:LⅢ; S: ↓↓ 10 days 4 48 ↓ ↓ ↓ Normal L4/5 M: Normal; S: ↓↓ 2 days 5 53 ↓↓ NR NR Normal L5/S1 M: Normal; S: ↓↓↓ 1 month 6 32 ↓ Normal Normal NR L5/S1 M: Normal; S: ↓ 2 weeks 7 41 ↓ ↓ ↓ ↓ L4/5 M: Normal; S: ↓↓ 7 months 8 45 ↓ ↓↓ Normal NR L4/5 M: Normal; S: ↓ 3 months 9 59 ↓↓ ↓ Constipation ↓↓ L5/S1 M: Normal; S: ↓ 3 months 10 33 ↓ ↓ ↓ ↓ L5/S1 M: Normal; S: ↓↓↓ 3 days 11 54 ↓ ↓↓ Constipation ↓ L4/5 M: Normal; S: ↓↓ 2 months 12 42 Normal Normal Constipation NR L4/5 M: Normal; S: ↓↓ 15 days 13 56 Tingling ↓ NR Normal L4/5 M: Normal; S: ↓ 6 months 14 45 Tingling ↓↓ NR NR L4/5 M: Normal; S: ↓ 4 days 15 29 ↓↓ ↓ Constipation NR L5/S1 M: Normal; S: ↓↓ 12 days 16 51 ↓ ↓↓ ↓ ↓ L4/5 M: Normal; S: ↓ 1 month 17 50 ↓↓ ↓↓ Constipation NR L4/5 M:Ⅲ; S:↓↓ 1 month 3 days 18 26 ↓ ↓ ↓ ↓↓ L4/5 M:Ⅳ; S: ↓↓ 8 days 19 35 ↓↓ ↓ ↓ NA L5/S1 Normal 4 days 20 41 ↓↓ ↓↓ Constipation NA L4/5 M:Normal; S: Tingling 2 days 21 40 Normal ↓ Normal NR L5/S1 M:Normal; S: ↓ 6 days 22 43 ↓↓ Normal Normal ↓↓↓ L3/4/5 M:Ⅱ; S: ↓↓ 5 days 23 18 Normal Normal Normal ↓↓↓ L4/5 M:Ⅳ; S: Normal 3 months 3 days

Table 3 Results of functional examinations in 23 patients with incomplete CES before operation

表 4(Table 4)

Table 4 Results of functional examinations in 23 patients with incomplete CES after decompression. Patient No. Time after operation (month) Age (year) Perineal sense Bladder empty Bowel empty Penile erection Motor (M) and sense (S) in legs 1 28 41 Normal Normal Normal Normal M: ↓; S: ↓ 2 80 56 Normal Normal Normal Normal Normal 3 45 47 Normal Normal Normal Normal Normal 4 71 48 tingling Normal Normal Normal ND 5 93 53 Normal Normal Normal Normal ND 6 42 32 Normal Normal Normal Normal ND 7 35 41 Normal Normal Normal Normal Normal 8 91 45 Normal Normal Normal Normal ND 9 42 59 Normal Normal Normal Normal Normal 10 3 33 Normal Normal Normal Normal Normal 11 55 54 ↓ Normal Normal Normal ND 12 41 42 Normal Normal Normal Normal Normal 13 47 56 Normal Normal Normal Normal ND 14 45 45 Normal Normal Normal Normal ND 15 110 29 Normal Normal Normal Normal Normal 16 78 51 Normal Normal Normal Normal Normal 17 27 50 Normal Normal Normal Normal Normal 18 29 26 Normal Normal Normal Normal Normal 19 22 35 Normal Normal Normal NA M: Normal; S: Numbness 20 24 41 Normal Normal Normal NA Normal 21 19 40 Normal Normal Normal Normal Normal 22 13 43 Normal Normal Normal Normal Normal 23 21 18 Normal Normal Normal Normal Normal ND: Not detected (because the patients answered the phone call only without showing up personally).

Table 4 Results of functional examinations in 23 patients with incomplete CES after decompression.

Urodynamic analysis was performed in 6 patients with complete CES (Tab. 5). At 3 to 96 (36.7 ±31.18) months after the operation, residual volume reduced significantly in all the 6 patients and to the normal level in 5 patients. The peak flow rate increased in 4 patients. Contractions of intra-urethral sphincter demonstrated by urethral pressure were normal or reduced before operation, and became normal in the follow-up in all the 6 cases. The most significant finding was the sensory function of the bladder, as we found that the first-desire volume was in the normal range in all the patients after the operation. But the detrusor contraction still remained paralyzed except in one patient who showed weak contraction during voiding. The non-inhibitory contraction collaborated by the detrusor and autonomic nerves were absent in all the patients. Strong intra-abdominal pressure was detected in all the 6 patients during voiding.

表 5(Table 5)

Table 5 Comparison of urodynamic findings in 6 patients with complete CES before (Pre) and after (Post) the operation Findings No.1 No.6 No.7 No.9 No.10 No.11 Gender M F M F M M Age of diagnosis (year) 35 41 49 60 49 45 Peak flow rate (mL/s) Pre 4 5 10 3 0 3 Post 13 10 25 5 14 4 Residual volume (ml) Pre 350 30 120 220 400 150 Post 240 10 35 50 65 20 Urethral pressure Pre ↓ Normal ↓ Normal ↓ Normal Post Normal Normal Normal Normal Normal Normal Bladder filling first sensation (mL) Pre 202 500 58 478 495 530 Post 196 137 135 411 161 329 First desire (mL) Pre 330 508 136 490 510 680 Post 315 218 184 411 216 377 Urination urgency (mL) Pre 839 509 169 501 512 689 Post 442 267 223 412 369 448 Maximum bladder volume (mL) Pre 1202 511 198 505 525 690 Post 489 300 255 412 456 588 Intra-abdominal pressure (cm H2O) Pre 12 2 15 10 7 6 Post 36 31 44 23 35 20 Urethremphraxis Pre None None None None None None Post None None None None None None Detrusor contraction Pre Null Null Null Null Null Null During voiding Post 444 Null Null Null Null Null Non-inhibitory contraction Pre Null Null Null Null Null Null Post Null Null Null Null Null Null The case No. assigned to the individual patients in Tab. 1 and 2 is used consistently this table. ↓, ↓↓, and ↓↓↓ indicate slight, moderate and severe reduction, respectively.

Table 5 Comparison of urodynamic findings in 6 patients with complete CES before (Pre) and after (Post) the operation

The occurrence of CES caused by LDH can be either gradual or sudden. In slowly developing CES, the symptoms evolve progressively and sustain over a long period of time as in the case of incomplete CES, whereas in the acute type, complete CES can occur in seconds. In our study, 11 out of the 12 patients with complete CES reported a history of low back pain, 4 patients developed complete CES shortly after chiropractic manipulation, 1 patient developed complete CES in 2 days after chiropractic manipulation, and 1 developed complete CES after a long motorcycle ride. In the 23 patients with incomplete CES, progression of the condition was gradual, and 9 patients reported worsening of the condition after chiropractic manipulation. As a common therapy for back pain associated with muscle strain, chiropractic manipulation should be performed with caution after careful examination to exclude such contraindications as LDH, tuberculosis, and tumors.

In patients with CES, neurological deterioration occurs in a progressive manner ^{[10, 11]}. If the patients' condition allows, surgery should be performed as soon as possible to prevent neurological deterioration ^[12-19] and promote recovery of the neural functions. According to Ahn et al^[18], decompression within 48 h after the onset of CES could achieve favorable outcomes, but sooner (within 24 h) surgical intervention did not significantly improve the patients' outcomes; Kohles et al ^[3] and Delong et al ^[19], however, believed that surgical intervention of CES within 24 h could bring further benefit. In our case, all the patients received much delayed decompression (mean 4.1 weeks for complete CES and 5.5 weeks for incomplete CES). Six (50%) of the patients with complete CES achieved good recovery and the other 6 also reported significant improvement after surgery, and 23 patients with incomplete CES all had normal life after delayed decompression.

Several reasons might explain the functional recovery of the patients after delayed decompression for CES. Firstly, in slowly developing CES, the compressed nerve roots may slowly develop tolerance to hypoxic conditions and still retain the potential of functional recovery after delayed compression; secondly, as peripheral nerves, the cauda equina nerves possess regeneration ability and may recover from reversible injuries caused by compression, which also explains the better results of early decompression than delayed decompression, especially in terms of bladder and bowel functions. During the surgery, we found that the cauda equina roots were swollen with a reddish purple color, but it could be difficult to accurately assess the damages without histopathological evidence. The injury of the cauda equine roots is not necessarily irreversible when patients receive late decompression. In addition, progression of incomplete CES into complete CES occurred in none of our cases even after months' delay of decompression. Thirdly, this long-term follow-up study allowed assessment of neural functional recovery of the patients long after the operation. Chang et al^[20] followed 4 patients with CES for 6.4 years and found that the long-term outcome of the patients was not necessarily poor in cases of poor short-term recovery of bladder function ^[20], which was also supported by the results of similar studies^{[7, 21]}.

An important finding of this study is that in patients who reported recovery of urination, their recovery was far from good as indicated by the results of urodynamic analysis. This discrepancy had been noticed before, and the bladder function can be disturbed seriously even in the patients without symptoms ^[22]. Our results of urodynamic analysis indicate that although the detrusor remained paralyzed, the recovered sensation of the bladder and the compensation by increased intra-abdominal pressure allowed the patients to finish urination. This highlights the importance of training the patients to practice the skills of increasing the intraabdominal pressure (Valsalva maneuver), which serves as one of the rehabilitation strategies for these patients.

Although the patients with CES can have favorable motor function recovery of the lower limbs ^{[5, 23, 24]} (as we also found in our patients) and achieve significant regeneration of the neuromuscular junctions even after denervation for up to one year^[25], the recovery of detrusor function remained poor in these patients. The relations among the recovery of bladder sensation, motor function of the extremities and detrusor remain so far unclear, but this might point the direction of further study.

In conclusion, we believe that patients with LDH-CES who missed the chance of early decompression can still expect favorable functional recovery in the long term. The improvement of bladder function following decompression is probably a result of recovery of bladder sensation and the compensation by increased intra-abdominal pressure. The key strategy to promote bladder function recovery in these patients is to promote the detrusor recovery, and long-term follow-up of the patients is of critical importance. Nevertheless, we could not accurately evaluate the benefit of delayed decompression surgery for ethical reasons, and further study is warranted to address this issue.