In a series of 60 children from 3 to 18 years of age who met typical clinical criteria for TCS, Wehby et al. retrospectively analyzed the outcome over a mean follow-up period of 13.9 months after sectioning of the tight filum terminale [79]. All patients in this series had their conus in a normal position on MRI. After surgery, urinary deficiency completely resolved in 52 % and markedly improved in 35 %. Fecal incontinence resolved in 56 % and improved in 41 %. Weakness, sensory abnormalities, and pain improved or resolved in all patients. Surgery was indicated only, when there were signs of spina bifida occulta, progressive neurogenic bladder instability unresponsive to conservative measures, and two or more of the following: fecal incontinence or chronic constipation, lower extremity weakness, gait changes, reflex/tone abnormalities, sensory disturbances, back/leg pain, limb length discrepancy, scoliosis/lordosis, recurrent urinary tract infections, abnormal voiding cystourethrogram/ultrasound, syringomyelia, or neurocutaneous stigmata.

This comprehensive list illustrates the complex clinical work-up involved in decision making for the surgical indication in TCS patients. The study also demonstrates that significant improvement of neurogenic bladder disturbances may be achieved even in patients in whom the disorder has already caused some urinary deficiency as long as primary symptoms are not ignored or accepted and diagnosis of TCS is established early by employing the whole range of diagnostic criteria.

Syringomyelia is associated with spinal dysraphism in a significant number of patients and may successfully be treated by untethering the spinal cord alone. Lee et al. [42] reported that 33 in a series of 135 patients with closed spinal dysraphism presented with a syrinx. In 31 of the 32 patients whose spinal cord was surgically untethered, follow-up imaging showed long-term stability or a decrease in the syrinx index. New urinary symptoms had developed in one patient whose syrinx increased after surgery and retethering could be proved.

When surgery is delayed over a longer period in established disease, results are not as favorable [35, 48, 49, 55, 57, 61, 82]. Since the disease leads to progressive neurological deterioration in most patients with neuroradiologically proven TCS including a low conus and spina bifida occulta, prophylactic surgery has been suggested and performed in children and adults with good results for several years follow-up in asymptomatic patients [62, 74].

Previous surgery (e.g., for myelomeningocele or spinal tumors) does not appear to compromise the results of surgical untethering [33] and therefore should not be a criterion to delay an operation in clinically progressive disease with clear radiological signs of TCS.

Having observed histological changes that reflected decreased elasticity of the filum terminale in urinary incontinent patients with normal results in radiological studies, Selcuki et al. even suggested sectioning of the filum terminale in all of these patients [67].

The indication of surgery in adult patients should be based on clinical symptoms, i.e., the presence and progression of a neurological deficit and pain syndromes that are clearly related to a tethered spinal cord. While radiological evidence in some severely symptomatic cases may not readily be obtained, there are other cases with an unequivocal radiological finding that has no clinical correlate.

From a series of 85 cases, Klekamp [34] has concluded that surgery in adult patients with a tethered cord syndrome should be reserved for those with symptoms. In the 43 patients who had been operated on with or without additional resection of a lipoma or dysraphic cyst, satisfying long-term pain relief was achieved in the majority, and long-term neurological stabilization tended to persist more often than it did in 25 conservatively treated patients.

While there is no consensus about the timing of surgery in primary TCS today, there is clearly a tendency toward early operation in established disease, especially when there are progressive clinical signs. Conservative wait-and-see management in patients with a clinical manifestation possibly related to primary tethering of the spinal cord and no radiological markers would at least necessitate a search for associated anomalies [48], urodynamic investigation, and close follow-up. On the other hand, asymptomatic patients with clear radiological signs of TCS may not become symptomatic for years [10, 28] but should be referred to surgery on appearance of the first symptoms [48, 62].

Adherence to some general neurosurgical principles will help to improve operative results. With the patient in prone position, special efforts to ensure that the abdomen is free of pressure should be undertaken in order to reduce abdominal venous pressure and prevent unnecessary blood loss. In the presence of spinal curvature, this may be difficult. A midline incision will usually be carried out which avoids additional cutaneous manifestations and follows scoliotic deformities. In the presence of associated malformations and in recurrent surgery, as a rule the surgeon will work from normal to abnormal anatomy.

In primary TCS without associated lesions, hemilaminectomy will be sufficient to approach and transect the filum. Laminectomy and laminotomy will be necessary in cases of additional tumors or myelomeningocele; the latter should be carefully considered in the presence of multiple arch defects [35, 54, 61].

If primary dural closure cannot be achieved due to a small dural sac or a large surface after tumor resection that would promote retether, dural plasty with either thoracolumbar fascia or artificial material will be required [61], perhaps combined with retention sutures that may help to maintain a relatively normal position of the spinal cord within the thecal sac, thus decreasing the potential adherence of the dorsally scarred aspect of the dysmorphic cord to an overlying graft [70, 72].

Intraoperative monitoring with anal sphincter EMG has long been introduced in surgery for TCS [27]. A more elaborate setup with continuous EMG recording of leg muscles, continuous recording of tibial nerve SSEPs, recording of MEPs evoked by transcranial electrical stimulation, and recording of compound muscle action potentials or SSEP from the scalp upon electrical stimulation of the nerve roots (mapping) provides the surgeon with functional information on the state of the motor and sensory pathways and enables anatomical identification of nerve roots and their distinction from fibrous or neoplastic structures [38, 39, 52, 56].

Because electrical stimulation of the filum will also lead to motor activation due to activation of neighboring nerve roots, and because there is considerable inter-patient variability in electrical thresholds of nerve fibers, von Koch et al. have proposed a ratio, rather than an absolute number, for establishing the electrical criteria for the distinction of the filum and the roots [75]. In over 70 % of their 63 patients, muscle activation via the filum required 100 times the voltage needed to activate a motor root.

In the literature, the term primary tethered cord syndrome apparently stands for at least three different entities – tight filum terminale with and without low conus medullaris as well as TCS associated with myelomeningocele and tumors. These entities are diagnosed at various different stages. It is this variety that mainly accounts for the different outcome reports and the ambiguity with respect to the indication for surgery. Nevertheless, there is common ground to stand on: neurological deterioration in a majority of patients with untreated primary TCS is natural and obvious. It is thus important that the diagnosis is established as early as possible. Surgical untethering ultimately remains the method of choice and should be offered to all patients who experience worsening of their condition. In some with clear evidence of TCS, even prophylactic surgery may be indicated.