Shunting

Shunting of the syrinx cavity can be performed using syringoperitoneal, syringopleural, syringosubarachnoid, or ventriculoperitoneal shunts. Most shunting procedures also involve laminectomies and duraplasties prior to the insertion of shunts (Figure 3).

Figure 3. Drainage of syrinx followed by shunt insertion. (Elliot, NSJ 2008b)

Table 1. Shunting

Author Year; Country Research Design Score Total Sample Size	Methods	Outcome
Schaan & Jaksche 2001; Germany Cohort N=30	Population: Gender: males=21, females=9; Level of severity: complete=24, incomplete=6. Intervention: Patients with syringomyelia were divided into 3 groups: Group 1 (n=18) received single or multiple shunting procedures; Group 2 (n=5) received shunting procedures before surgical creation of a pseudomeningocele; Group 3 (n=7) was treated only with the surgical pseudomeningocele. Outcome Measures: Sensory and motor deficit, Pain, Syringobulbia.	There were no significant differences in outcomes between groups. Prior to shunting (n=18) 15 had sensory deficits, 13 had motor deficits, 14 had pain and two had syringobulbia. Post-surgery: Sensory deficits improved in five, deteriorated in two, and were unchanged in eight patients. Motor deficits improved in five, deteriorated in two and were unchanged in four patients. Pain improved in five, deteriorated in four, and was unchanged in five patients. Syringobulbia improved in one and deteriorated in the second patient. Prior to shunting and pseudomeningocele (n=5) five had sensory and motor deficits, one had pain and one had syringobulbia. Post-surgery: Sensory deficits improved in four and deteriorated in one patient. Motor deficits improved in three, deteriorated in one and was unchanged in one patient. Pain and syringobulbia improved in both patients. Prior to pseudomeningocele only (n=7) seven patients had sensory deficits, six had motor deficits, three had pain, and one had syringobulbia. Post-surgery: Sensory deficits improved in six and were unchanged in one patient. Motor deficits improved in two, deteriorated in one, and were unchanged in three patients. Pain and syringobulbia improved in all patients.
Davidson et al. 2018; Australia Pre-Post N=41	Population: Mean Age: 42 yr; Gender: males=19, females=22; Injury etiology: SCI Trauma=13, Arachnoiditis=4, Chiari=5, Other=19; Localization of syrinx: cervical=11, cervicothoracic=5, cervicolumbar=3, thoracic=18, thoracolumbar=2. Intervention: Syrinx to subarachnoid shunt. Outcome Measure: Reduction, stabilization, or recurrence of syringomyelia, self-reported quality of life (SF-12), clinical symptoms.	Ninety percent (n=32) of patients experienced a reduction in syringomyelia size at 3 mo follow-up. Post-operatively, eight patients had clinically stable syringomyelias, while one patient’s condition was deemed clinically worse. There were no significant differences in SF-12 post-surgery for physical (p=0.64) or mental (p=0.74) health. The majority of patients had improved clinical symptoms (n=32) on pain, posterior column sensory loss, motor weakness, myelopathy, spasticity, and bladder dysfunction. Eight patient’s symptoms remained stable, while one patient had an increased incidence of spasticity.
Tassigny et al. 2017; Belgium Pre-Post N=17	Population: Mean age: 43.3 yr; Gender: males=10, females=7; Injury etiology: SCI-trauma=5, arachnoid cyst=2, tumor=1, Chiari malformation=4, unknown=7; Localization of syrinx: cervical=1, cervicothoracic=5, thoracic=1, thoracolumbar=2, pan-medullaris=4, multiple=4. Intervention: Placement of myringotomy tube with drain. Outcome Measure: Syringomyelia clinical status, radiological status (shrinkage), reoperation.	Three patients experienced syrinx disappearance, the clinical status of eight improved, five stabilized, and one patient experienced clinically worse progression of syringomyelia. Between one and three days post-operatively, an MRI was conducted to determine shrinkage. Eight patients experienced complete shrinkage, three almost complete, four a reduction is size, and two achieved stable size. A total of three patients required recurrent surgery, one ‘surgical exploration’ and two for a syringoperitoneal shunt.
Karam et al. 2014; Canada Pre-Post N=27	Population: Mean age: 40 yr; Gender: males=24, females=3; Injury etiology: MVA=16, Bicycle/Tractor/AirplaneAccidents=3, Falls & All-terrain Vehicles=3, Motorcycle Accidents=2; Severity of Injury: AIS A=14, C=3. D=10. Intervention: Sixteen patients underwent a shunting procedure, 14 of which received a syringo-subarachnoid procedure and two received a syringo-pleural procedure. 11 patients underwent duraplasty including seven patients who received shunting and lysis of adhesions. Outcome Measures: Odom Score, Level of syrinx.	Amongst the 16 patients who received a shunt only, 10 required revision surgery whereas only three of the 11 patients who received a shunt plus duraplasty required revisions; the difference between the groups was non-significant (p=0.0718). No improvements in ASIA score from admission were reported. Sequential MRI scans were available for 11 of the patients with a significant correlation being found between clinical improvement assessed by the Odom Score and a reduction in the size of the syrinx (p<0.001). Size of syrinx reduced from an initial 12±6 levels to 6±7 levels at follow-up.
Ushewokunze et al. 2010; UK Pre-Post N=40	Population: Mean age: 32 yr; Gender: males=38, females=2. Intervention: A laminectomy with the creation of a cerebrospinal fluid conduit and shunting. Outcome Measures: Adverse events, Stabilization rate.	Overall, 27 patients reported a stabilization of symptoms while 13 patients experienced a deterioration of symptoms. At 6mo follow-up, a reduction in the size of syrinx was reported in 21 out of 33 patients. Twenty-three patients required no further surgery while 17 patients underwent further surgery for deteriorating symptoms. The most common early complications included pain (n=5), neurological deficit (n=4), wound infection (n=4), and CSF leak (n=2).
Li et al. 2021; USA Case Series N=34 (pre- and post- operative MRI for 24 patients)	Population: Mean age: 44 yr; Gender: males=15, females=9; T1=116; Severity of injury: AIS A=12, B=0, C=2, D=9, E=1. Intervention: Participants underwent surgical treatment consisting of syrinx fenestration, lysis of adhesions, and duraplasty. 21 of the patients underwent laminectomies, 21 patients underwent intradural lysis of adhesions, 17 underwent cyst fenestration, and 19 underwent duraplasty. In four patients, pre-existing syrinx shunts were removed; in two others, syrinx shunts (syringo- pleural and -subarachnoid) were placed. Outcome Measures: Syrinx length (measured in spinal segments), syrinx axial dimension (defined as the product of maximum syrinx transverse and antero-posterior dimensions), motor, sensory, bladder/bowel, and overall functional status.	The majority (15 of 24) of participants had cervical SCI. The average syrinx length, measured in spinal segments, was similar when comparing pre- and post-operative MRIs (5.5 and 5.4 spinal segments, respectively). Syrinx axial dimension was decreased in 16 of the patients postoperatively and stable or increased in the other eight. Of the 24 patients, 23 remained clinically stable in regards to motor, sensory, and bowel/bladder function. One patient, whose surgical treatment involved syringo- subarachnoid shunt placement, experienced neurologic deterioration; in this patient, motor function decline was noted postoperatively. The change in syrinx size did not correlate with clinical outcomes.
Hess & Foo 2001; USA Case Series N=8	Population: Level of injury: T=2, C=5, L=1; Level of severity: AIS A/B=7, C=1; Mean time since injury: 10 yr. Intervention: Charts of patients who received shunts were assessed. Outcome Measures: Outcome of shunting, Complications.	A significant reduction in pain was reported by >80% of patients post-surgery; improvement in strength (n=6) and sensory (n=2) was also reported. At follow-up, four patients had shunt failure resulting in neurologic decline, while two developed a new syrinx. In one patient a new cavity was found with MRI, while the original remained decompressed.
Lee et al. 2001; USA Case Series N=45	Population: Mean age: 45.6 yr; Gender: males=30, females=15. Intervention: Records of patients who underwent surgical treatment for posttraumatic syringomyelia were assessed. Patients were divided into three groups: Group 1 underwent untethering only, Group 2 underwent shunting only, and Group 3 underwent both untethering and shunting. Patients were followed up to assess treatment efficacy. Outcome Measures: Improvement in symptoms, Magnetic Resonance Imaging (MRI), Complications.	There were no significant differences in outcomes between groups. Patients in the surgical untethering group: Demonstrated improvement in motor and spasticity symptoms in the majority of patients (60% and 58%, respectively). Experienced one treatment failure and two complications. Revealed cyst re-accumulation at one yr follow-up. The shunt only group experienced one complication and three treatment failures; 60% of patients in this group experienced improvement in gait followed by sensory (57%) and motor (54%). Among those who underwent both untethering and shunting, 33% had clinical recurrence, one experienced CSF leak, and 50% showed improvement in motor symptoms.
Lee et al. 2000; USA Case Series N=34	Population: Mean age: 43.2 yr; Gender: males=23, females=11. Intervention: Records of patients who underwent surgical treatment for posttraumatic syringomyelia were assessed. Patients underwent laminectomies and a syringosubarachnoid shunt was inserted. Patients were divided into three groups: Group 1 underwent untethering only, Group 2 underwent shunting only, and Group 3 underwent both untethering and shunting. Patients were followed up to assess treatment efficacy. Outcome Measures: Improvement in symptoms, Complications.	At follow-up (>1 yr), 26 patients’ resolution of one or more of the presenting symptoms was achieved post operatively; two patients experienced deterioration of motor function. A decrease in spasticity was the most common improvement in patients who underwent untethering only (67%), followed by motor functioning (57%) and sensory loss (50%); this group experienced one treatment failure and two complications. Improvement in gait was seen in the highest number of patients from the shunt only procedure group (60%), followed by motor (50%) and sensory loss (50%); in this group, two treatment failures and two complications occurred. Patients who underwent untethering and shunt procedures did not experience clinical reoccurrence; motor (67%) and gait (50%) improved in patients in this group.
Falci et al. 1999; USA Case Series N=59	Population: Mean age: 26 yr; Gender: males=49, females=10; Level of severity: AIS A=53, B=1, C=4, D=1. Intervention: All patients underwent spinal untethering and if a spinal cyst was present a lumbo-peritoneal shunt tube was placed along the length of the cyst. Outcome Measures: Pinprick, Motor and light touch scores, MRI findings, Somatosensory evoked potentials.	Participants with no previous surgery showed a significant increase in light touch (+2.38), pinprick (+3.88) and motor scores (+1.47) post-surgery. Participants who had previous surgery had a decrease in touch, pinprick and motor score, although it was minimal (0.7, 0.8, and 0.5, respectively). At 2wk post-surgery, MRI showed decreased cyst size or complete collapse. Somatosensory evoked potentials were improved in amplitude compared to baseline; latency of 2 milliseconds or greater was observed in 27 patients.
Ronen et al. 1999; Israel Case Control N=10	Population: Mean age: 31.3 yr; Gender: males=10, females=0; Level of injury: C=5, L=4; Level of severity: incomplete=5, complete=5. Intervention: Charts of patients with syringomyelia were reviewed. Patients were divided into two groups: patients receiving rehabilitation only and patients receiving rehabilitation and shunting. Outcome Measures: Functional and neurological outcome.	Four out of five patients in the shunt surgery and rehabilitation group showed functional and neurological deterioration; the fifth patient remained unchanged. Patients in the rehabilitation only group remained unchanged except for one who showed significant functional improvement without any change in neurological status.
Hida et al. 1994; Japan Case Series N=14	Population: Mean age: 48 yr; Gender: males=10, females=4; Level of injury: C=5, T=5, L=4. Intervention: Charts of patients who underwent syringosubarachnoid (n=6), syringoperitoneal (n=4), and ventriculoperitoneal (n=1) shunts were assessed. Outcome Measures: Neurological, motor, Sensory functioning, Shunt malfunction.	Neurological amelioration was obtained in all patients. Of the nine patients with motor function difficulty, eight improved. Sensory disturbance and relief of local pain or numbness improved in all patients. Malfunction was reported in three of four syringoperitoneal shunts and in the one ventriculoperitoneal shunt.

Discussion

In all of the studies, no shunting procedure was found to be superior to another. Schaan and Jaksche (2001) in a cohort study assessed the efficacy of syringomyelia treatment in three groups of patients: Group One received various shunts only, Group Two received shunting followed by surgical creation of a pseudomeningocele, and Group Three was treated with the pseudomeningocele only. The study found improvement in sensory and motor deficits, pain and syringobulbia in all three groups. However, more patients experienced greater pain post-surgery in the shunting only group than the other two groups. It should be noted that although groups did improve on some of the outcome measures, none of the groups were significantly different from each other.

Falci et al. (1999) demonstrated that untethering and shunt tube placement among individuals without prior surgery can significantly improve light touch, pinprick and motor scores. Two case series (Lee et al. 2001; Lee et al. 2000) further examined untethering and shunting treatment in individuals with syringomyelia. In the study by Lee et al. (2001), patients were divided into three groups: untethering only, shunting only, and untethering and shunting. Improvement in motor and sensory functioning was observed in all three groups, although the groups did not significantly differ from each other. In the first group, untethering only, improvement in spasticity was more common; while the shunting only group found gait improvement to be the most common. These results are also supported by Lee et al. (2000) earlier case series which found the same outcome measures were improved by the same corresponding intervention. Furthermore, shunting alone has been reported to significantly improve pain (Hess & Foo 2001; Hida et al. 1994; Davidson, Rogers & Stoodley 2018), strength (Hess & Foo 2001), motor function (Hida et al. 1994; Davidson et al., 2018) and sensation (Hess & Foo 2001; Hida et al. 1994) in patients with syringomyelia, although, a high rate of shunt failure (36-50%) has been reported (Hess & Foo 2001; Hida et al. 1994). Another case series by Li et al. (2021) investigated the radiographic and clinical outcomes following various surgical treatments, including syrinx fenestration, adhesion lysis, and duraplasty, in 24 individuals with syringomyelia following traumatic SCI. In four of the patients, pre-existing syrinx shunts were removed, and in two others, syrinx shunts were placed. It was found that all but one patient remained clinically stable regarding motor, sensory, and bowel/bladder function post surgery. The authors noted that the only patient who experienced deterioration in motor function post surgery had undergone syringo- subarachnoid shunt placement. Due to the lack of clarity in reporting post-surgical outcomes for each type of surgical intervention, as well as the small sample size, findings should be interpreted with caution (Li et al. 2021).

One case control study (Ronen et al. 1999) reviewed charts of patients receiving either rehabilitation only or rehabilitation and shunting for syringomyelia. The study found 80% of patients in the shunting and rehabilitation group experienced functional and neurological deterioration, while patients in the rehabilitation group remained either unchanged or improved. One must be careful when drawing conclusions from such a study because allocation to either group was dictated by receiving the treatment, which presumably was given to those patients already deteriorating or those were considered at a higher risk of deterioration.

Ushewokunze et al. (2010) studied the adverse events after laminectomy and shunting and reported a reduction in syrinx size among 21 of 40 patients and a stabilization of symptoms among 27 of 40. However, symptoms deteriorated for 13 individuals including pain, increased neurological deficits, infection and CSF leakage. In 17 individuals, a second surgical procedure was required to improve deteriorating symptoms.

Only one Pre-Post Study examined the use of a myringotomy tube (Tassigny et al. 2017). Of the 17 participants in the study, only one experienced clinically worse progression of the syrinx. Additionally, between days 1-3 post-surgery 15 patients experienced shrinkage of the syrinx, with eight of those being complete (Tassigny et al. 2017).

A Pre-Post study examining just shunting (Davidson et al. 2018) found that shunting alone was enough to reduce the size of the syrinx in up to 90% of patients. Although the majority of patients improved on clinical symptoms there were no significant differences between groups for quality of life. Finally, a study by Karam et al. (2014) revealed that patients who received both shunting and duraplasty were less likely to require reoperations (3/11 patients) compared with patients who received a shunt only with 10 of 16 requiring revisions and reoperations. However, this contrast was not statistically significant. Overall, 14 patients (52%) experienced an improvement in symptoms, 10 (37%) remained stable whilst three (11%) reported a progression of symptoms without improvement. In addition, a reduction in syrinx size was reported which correlated significantly with clinical improvement as measured by Odom Score. Only one patient experienced complications post-surgery with pain and dysphagia reported; however, it was later revealed the patient had developed another syrinx larger than before. The patients received another shunt and duraplasty which resulted in improvements reported at three-month follow-up.

Conclusions

There is level 2 evidence (from one cohort, four pre-post, and five case series studies: Karam et al. 2014; Ushewokenze et al. 2010; Schaan & Jaksche 2001; Falci et al. 1999; Lee et al. 2000; Lee et al. 2001; Hess & Foo 2001; Hida et al. 1994; Tassigny et al. 2017; Davidson et al. 2018) that shunting improves pain, motor function and sensory loss in some SCI patients with syringomyelia; however, a high rate of shunt failure has been observed; these findings are tempered by level 4 evidence (from one pre-post study: Karam et al. 2014) that reported minimal clinical improvement post shunting with or without duraplasty.

Shunting of the syrinx cavity improves pain, motor function and sensory loss in some SCI patients post syringomyelia.