Conus-level combined dorsal and ventral lumbar rhizotomy for treatment of mixed hypertonia: technical note and complications

View More View Less
  • 1 Florida State University College of Medicine, Tallahassee, Florida;
  • | 2 Surgical Outcomes Center for Kids, Monroe Carell Jr. Children’s Hospital;
  • | 3 Department of Neurological Surgery and
  • | 5 Department of Physical Medicine and Rehabilitation, Vanderbilt University Medical Center;
  • | 4 Pediatric Rehabilitation, Monroe Carell Jr. Children’s Hospital, Vanderbilt University Medical Center;
  • | 6 Department of Pediatrics, Division of Pediatric Neurology, Vanderbilt University Medical Center; and
  • | 7 Department of Pediatrics, Division of Developmental Medicine, Vanderbilt University Medical Center, Nashville, Tennessee
Free access

Combined dorsal and ventral rhizotomy is an effective treatment for patients with concurrent spasticity and dystonia, with the preponderance of complaints relating to their lower extremities. This operative approach provides definitive relief of hypertonia and should be considered after less-invasive techniques have been exhausted. Previously, the surgery has been described through an L1–S1 laminoplasty. In this series, 7 patients underwent a conus-level laminectomy for performing a lumbar dorsal and ventral rhizotomy. Technical challenges included identifying the appropriate-level ventral roots and performing the procedure in children with significant scoliosis. Techniques are described to overcome these obstacles. The technique was found to be safe, with no infections, CSF leaks, or neurogenic bladders.

ABBREVIATIONS

BAD = Barry-Albright Dystonia; GMFCS = Gross Motor Function Classification System; GMFM = Gross Motor Function Measure; MAS = Modified Ashworth Scale; SDR = selective dorsal rhizotomy; TOF = train-of-four.

Combined dorsal and ventral rhizotomy is an effective treatment for patients with concurrent spasticity and dystonia, with the preponderance of complaints relating to their lower extremities. This operative approach provides definitive relief of hypertonia and should be considered after less-invasive techniques have been exhausted. Previously, the surgery has been described through an L1–S1 laminoplasty. In this series, 7 patients underwent a conus-level laminectomy for performing a lumbar dorsal and ventral rhizotomy. Technical challenges included identifying the appropriate-level ventral roots and performing the procedure in children with significant scoliosis. Techniques are described to overcome these obstacles. The technique was found to be safe, with no infections, CSF leaks, or neurogenic bladders.

ABBREVIATIONS

BAD = Barry-Albright Dystonia; GMFCS = Gross Motor Function Classification System; GMFM = Gross Motor Function Measure; MAS = Modified Ashworth Scale; SDR = selective dorsal rhizotomy; TOF = train-of-four.

In Brief

The objective of this study is to describe a combined dorsal and ventral rhizotomy for patients with concurrent spasticity and dystonia through a technical note. The key finding, and value added by this paper, is to perform the procedure through a conus-level laminectomy rather than an L1–S1 laminoplasty.

Children with cerebral palsy are often affected by spasticity, dystonia, or mixed hypertonia, which is a combination of both spasticity and dystonia.1 Although a cerebral insult is the underlying cause of each, treatment can vary due to differing pathophysiology, hypertonia subtype, severity, and anatomical distribution.1 Treatment of hypertonia is multifactorial, and can include physical therapy, bracing, enteral medications, botulinum toxin injections, intrathecal baclofen, rhizotomy, and/or orthopedic procedures.2 For children with mixed hypertonia in whom less-invasive treatments have failed, baclofen pumps are often used to manage tone.3 Unfortunately, baclofen pumps fail to provide adequate control of tone in approximately 15% of children with mixed hypertonia.4 Additionally, some children are poor candidates for baclofen pumps due to social circumstances or past complications from baclofen pumps.

Selective dorsal rhizotomy (SDR)5–7 is an established neurosurgical treatment for children with spastic diplegia, resulting in significant improvements in spasticity and level of function. However, in children with significant dystonia as a component of their hypertonia, SDR can exacerbate the dystonia. As a result, SDR has not traditionally been recommended for children with mixed hypertonia. Albright and Tyler-Kabara8 first reported favorable results in performing combined dorsal and ventral rhizotomies in 6 children with refractory mixed hypertonia.8 Both family and caregivers reported increased ease in day-to-day patient care.8 In 2016, Abdel Ghany et al.9 reported combined dorsal and ventral lumbar rhizotomies in 50 children with meaningful improvement in their tone, joint range of motion, dystonia, and activities of daily living.9

Conus-level SDR has been well documented in the literature,10–13 but for combined dorsal and ventral rhizotomies only L1–S1 laminoplasty has been described.8,9 An advantage of their described techniques is direct identification as the nerve root is exiting the dura. However, many surgeons are more accustomed to performing conus-level rhizotomy procedures. Additionally, many of these patients have undergone scoliosis fusions, making the L1–S1 exposure more involved. Challenges of performing a conus-level dorsal and ventral rhizotomy include accessing the ventral roots, identifying the appropriate ventral root levels, and managing the rotated spinal cord with scoliosis. The objective of this paper is to describe a technique of combined dorsal and ventral rhizotomies for the treatment of patients with mixed hypertonia using a conus-level approach with short-term follow-up and complications.

Methods

Conus-level combined dorsal and ventral rhizotomy for mixed hypertonia was performed on 7 patients at Vanderbilt University Medical Center in 2019. Patients were selected based on the following criteria: presence of mixed hypertonia, predominance of lower-extremity complaints, failed conservative medical therapy, failed intrathecal baclofen pump or not a candidate for intrathecal baclofen, and understanding that therapy was palliative. Patients were excluded from this surgery if they had isolated spasticity or isolated dystonia as other treatments would be indicated. Patient with fixed contractures were not excluded, but decisions about their candidacy would be weighed against the benefits of orthopedic procedures versus no surgical procedures. The decision for combined rhizotomy in these patients hinged on the severity and impact of their overlying mixed hypertonia. All children were evaluated in a multidisciplinary spasticity clinic prior to recommendation for rhizotomy. Prior to presentation to the clinic, all patients had been evaluated by their primary cerebral palsy physician for the etiology of their hypertonia and been treated using other conservative measures. All patients had been evaluated by MRI of the brain. In these cases, because the etiologies of the hypertonia were clear, routine genetic testing or spine MRI was not performed. Spine MR images were not obtained for conus localization as has been reported by other surgeons who perform rhizotomies.10 Children were objectively assessed with a neurological examination, Gross Motor Function Measure (GMFM)-66, Barry-Albright Dystonia (BAD) scale, Modified Ashworth Scale (MAS), range of motion with a manual goniometer, and Care and Comfort Questionnaire. Patients were not referred for inpatient rehabilitation. All patients were already engaged in outpatient physical therapy prior to surgery, and they continued with this therapy postoperatively.

Surgical Technique

Preoperative Preparation, Pain Control, and Neuromonitoring

Patients were given preoperative antibiotics, underwent general anesthesia9 without muscle relaxation, and were positioned prone on gel rolls. Pain control was achieved using scheduled acetaminophen 15 mg/kg every 6 hours, scheduled ketorolac 0.5 mg/kg intravenously every 6 hours, scheduled valium 0.5 mg or 1 mg, depending on weight, every 6 hours, oxycodone (weight-based) every 6 hours, and morphine 0.025 mg/kg intravenously as needed every 6 hours. On postoperative day 2, the ketorolac is discontinued and scheduled ibuprofen 10 mg/kg is started, and the valium is converted to “as needed” administration. Bipolar electromyography was recorded bilaterally from the following muscle groups: adductor magnus, tibialis anterior, gastrocnemius, biceps femoris, and external anal sphincter muscles. Referential electromyography was recorded from vastus lateralis and vastus medialis bilaterally. Train-of-four (TOF) was set unilaterally on flexor digitorum brevis and abductor hallucis to verify the neuromuscular junctions are not chemically blocked. A Cadwell 16-channel IONM System with a Cadwell disposable right-angle double-nerve probe was used. Threshold stimulus parameters were 0–5 mA, repetition rate 2.79 Hz, pulse width 200–300 msec. TOF was the same as threshold activation intensity, repetition rate 50 Hz, and pulse width 200–300 msec. Radiography was used to localize the L1 spinal level.

Laminectomy

A 1.5- to 2-level laminectomy was performed to optimize visualization of the ventral roots. Most patients did not have a preoperative spine MRI, and therefore the conus level was unknown. An L1 laminectomy was always performed and then an ultrasound was used to determine the need for further laminectomy to optimize exposure. In all 7 patients the conus terminated at L1, and in order to assist with the ventral rhizotomy, a partial or complete T12 laminectomy was performed. The patients necessitating a full two-level laminectomy included patients who underwent fusion for scoliosis. The additional level allowed better visibility around the fusion mass.

Nerve Root Identification

A midline durotomy was performed, and the operative microscope was used for microdissection. Bed rotation was often necessary for patients with scoliosis for proper visualization of intradural anatomy. The dentate ligaments were sectioned for mobilization of the spinal cord. Some patients had a single level of dentate ligaments, while other patients would have two levels of dentate ligaments exposed.

The L2 nerve roots were anatomically identified exiting below the L2 pedicle. After splitting the dorsal and ventral L2 roots from one another, the L2 dorsal root was followed back to its entry zone and then used as a landmark to reflect the dorsal roots posteriorly and separate from the ventral roots. Next, the S3–5 dorsal nerve roots were identified at their entry sites and isolated for exclusion from the rhizotomy. The L2–S2 dorsal nerve roots were then bundled together using a small silastic ribbon similar to common techniques for SDR,5 but rather than suturing this silastic bundle to the ipsilateral dura, it was temporarily sutured to the contralateral dura, providing visualization of the ipsilateral ventral roots (Fig. 1).

FIG. 1.
FIG. 1.

Upper: Dorsal roots L2 through S2 have been bundled in silastic, and the silastic bundle has been sutured to the contralateral dura. Lower: By suturing the dorsal bundle to the contralateral dura, the ipsilateral ventral roots are viewed. Figure is available in color online only.

Next, work proceeded to bundle the ipsilateral L2–S1 ventral roots. The technique evolved into performing the ventral root identification, bundling, and rhizotomy between the T12 and L1 exiting nerve roots. Dissection began by identifying the most inferior sacral ventral roots so that they could be excluded from the forthcoming bundle (Fig. 2A and B). The L2 ventral nerve root was followed from its dural exit site to the ventral root exit zone. In every case, discrete ventral roots were easier to identify than dorsal roots. In comparison, the dorsal rootlets often do not converge into discrete root bundles at the level of the conus, making discrete nerve roots difficult to identify and count; instead, the ventral rootlets converge quickly into discrete roots and allow for counting, which contributes to the ability to reliably identify the desired ventral roots (Fig. 2B and C). Because the spinal cord obstructs the view of the ventral root exit zone, great care must be exercised to ensure that contralateral ventral roots are not included in this bundle. Once isolated, a silastic ribbon bundles the L2–S1 ventral roots and then is sutured to the ipsilateral dura (Fig. 2D). The ipsilateral dorsal root bundle, which had been sutured to the contralateral dura to optimize ventral exposure, is then released from the contralateral dura and sutured to the ipsilateral side. The same technique is then used to isolate the dorsal and ventral roots on the contralateral side so that the appropriate bilateral dorsal and ventral roots have been isolated (Fig. 3).

FIG. 2.
FIG. 2.

A: The ipsilateral ventral roots are isolated and prepared for identification. B: Sacral roots are separated. The ventral roots are discrete. C: Two ventral roots are isolated for demonstration of their discrete nature. This allows for counting of nerve roots for neurological level identification. D: The L2–S1 ventral roots have been bundled and sutured to the ipsilateral dura. Figure is available in color online only.

FIG. 3.
FIG. 3.

Prior to starting the rhizotomy, bilateral dorsal L2–S2 and ventral L2–S1 ventral roots have been bundled in silastic and the bundles have been secured to their ipsilateral dura. The silastic bundle from one side of the dorsal roots has been opened to begin the dorsal rhizotomy. Figure is available in color online only.

Rhizotomy

In all cases, the bilateral dorsal rhizotomy was performed prior to the ventral rhizotomy. This is necessary because the neurophysiology feedback from stimulation would lose value after a ventral rhizotomy had been performed. The dorsal rhizotomy was performed using the same well-described technique as in other studies.5,14 In summary, for each dorsal nerve root, a threshold was established. Each nerve root was then split into 3–5 rootlets and each rootlet would then be stimulated with a train of tetanic stimulation. Using previously described criteria,12 60%–70% of dorsal rootlets would be cut from L2 to S2. The L1 dorsal root would not be stimulated, but 60% of the nerve root would also be sectioned.

Once the dorsal rhizotomy had been performed, attention would then be turned to the ventral roots (Fig. 4). The ventral roots were stimulated in a similar manner to the dorsal roots. Working from L2 to S1, each nerve root was stimulated to establish a threshold. The minimum amplitude for stimulation was 0.05 mA, and most of the nerve roots were hyperactive at this level. Therefore, for most, a true threshold could not be established. Each nerve root was divided into 3 segments. A train of tetanus was then applied to the nerve rootlets. However, in most cases the ventral nerve roots were too reactive to be useful in identifying nerve root level and often too reactive to provide meaningful information about selecting which bundles to cut. Therefore, as the procedure evolved, the ventral portion of the rhizotomy became more anatomical and less selective. On average, two-thirds of the ventral roots were sectioned based on the variability in reported percentages in previous reports.8,9 There was a concern of cutting a higher percentage and leading to severe muscle atrophy, possibly secondarily contributing to worsening osteoporosis, and poor venous return from the lower extremities.

FIG. 4.
FIG. 4.

Bilateral dorsal rhizotomies have been completed. One ventral silastic bundle has been opened, preparing the ventral roots for rhizotomy. The contralateral ventral silastic bundle can also be seen on the contralateral side, still sutured to the dura. Figure is available in color online only.

Scoliosis

Over half of the children had scoliosis (Fig. 5). With scoliosis, both the bones of the spine and the spinal cord are rotated. Therefore, as the spinal cord rotates, on the concave side of the curve the ventral roots will rotate anteriorly, or away from the surgeon, and become more challenging to access and isolate. To counter this issue, the surgeon should use the same technique described above, but the surgeon should begin by bundling the nerve roots on the concave side first. By starting on the concave side, in the process of suturing the bundled dorsal roots to the contralateral dura, the conus slightly rotates, bringing the ventral roots into a more visible position. When performing the rhizotomy, the same technique described above should be followed, but it is important to save the concave ventral roots for the last bundle to cut. Again, by this point, all tethering will have been removed and the nerve roots will be gently rotated and elevated into view.

FIG. 5.
FIG. 5.

Standing radiographs showing 4 of the 7 patients had scoliosis, contributing to the complexity of the surgery.

Results

Of the 7 children, the median age of the cohort was 12 years (range 4–16 years) with a predominance of male (71%) and Caucasian (71%) patients. Anoxia (57%) was the most common etiology of the mixed hypertonia (Table 1). All patients had a Gross Motor Function Classification System (GMFCS) level of 5 and suffered spastic and dystonic quadriplegia with the predominance of complaints related to their lower extremities, with mean MAS scores of 3.1 and BAD scale scores of 3.8. A baclofen pump had been removed from 2 patients and remained in 1 patient, where it continued to be used for upper extremity and trunk tone management. Four patients were not good baclofen pump candidates as there were geographic or social barriers to obtaining refills. Scoliosis was present in 57% of patients and all but 1 of these patients had undergone fusion. In surgery, a mean of 64% of dorsal roots were cut and 65% of ventral roots cut at the selected levels. All patients had a Foley catheter removed on postoperative day 1. Two patients retained urine. One patient performed in-and-out catheterizations for 1 day and the other patient for 5 days. All patients were voiding at their baseline at last follow-up. Additionally, there were no infections, CSF leaks, or readmissions within 3 months. Average length of stay was 2.5 days. There has been an average of 6 months follow-up thus far on this cohort and all patients have had a significant reduction in their tone with an average MAS score of 0.4. No patients were ambulatory before surgery and no patients are ambulatory after surgery. All patients maintain movement in their lower extremities, but none with strength greater than 3/5. A complete tabulation of patient information can be found in Table 1.

TABLE 1.

Patient demographic information, preoperative data, and outcome

Patient No.Age (yrs)EtiologyBaclofen PumpScoliosis/FusionGMFM-66 Score*MAS Mean Preop ScoreBAD Preop Score% Dorsal Roots Cut% Ventral Roots CutMAS Mean Postop Score
115AnoxiaYesYes/yes03.6462620
210AnoxiaNoYes/no03461671
316PrematurityRemovedYes/yes92.5464651
414AnoxiaRemovedYes/yes03469670
56MeningitisNoNo/no82.5465640
64HoloprosencephalyNoNo/no283362660
712Anoxia (placental abruption)NoNo/no02.75363671

Value of 0 given to patients who were unable to complete any tasks.

Average of all lower extremity scores given.

Average of lower extremity BAD scores.

Discussion

A conus-level dorsal and ventral rhizotomy is a feasible and safe technique for treating patients with mixed hypertonia who have a preponderance of lower-extremity complaints and in whom conservative management has failed. There are two unique technical challenges of conus-level combined dorsal and ventral rhizotomy: identifying the appropriate ventral roots, and managing the rotated spinal cord configuration in patients with scoliosis. Identifying the appropriate ventral roots for sectioning is challenging for multiple reasons. Cadaveric studies find a consistent pattern of nerve root orientation/identification around the conus medullaris15 and differences in the sizes of the roots depending on spinal level.16,17 However, this predictable pattern is not reliable in surgery for nerve root identification and can easily be disturbed by standard operative dissection. Identification of the ventral root neurological level is challenging because the dorsal roots obstruct the view of the ventral roots and because the spinal cord blocks the view of the ventral root exit zone. Second, neurophysiological feedback was not typically helpful because of the hypersensitivity of these nerve roots, even at the lowest possible stimulation thresholds. To assist with optimizing the visibility of the ventral roots, it was useful to section the dentate ligaments to allow mobilization of the conus medullaris. Typically, the lowest dentate ligament is at the T12–L1 level, but it can occasionally be seen at the L1–2 level.17 The ipsilateral dorsal root bundle is then transiently sutured to the contralateral dura, providing subtle, but important, rotation of the conus medullaris, allowing improved visibility of the ventral roots.

Initially, we performed the ventral rhizotomy immediately inferior to the conus medullaris (L1–2 level) because there would be more space within the spinal canal for manipulating the nerve roots. Unfortunately, this made identification of the appropriate ventral roots more challenging. At that inferior location, the ventral roots could not be easily tracked back to their ventral root exit zone and therefore ordering of the nerve roots from cranial to caudal was extremely difficult. Additionally, even differentiating left from right was challenging for some ventral roots at this level. Therefore, the technique evolved to working between the L1 and L2 exiting nerve roots. At this level, the ventral nerve roots could be properly ordered because they could be tracked back to their ventral root exit site. Lastly, the ventral roots at this level were consistently very distinct, especially compared to the dorsal roots. Whereas the dorsal roots can often be difficult to discriminate from one another at the level of the conus due to lack of an interradicular space and failure to coalesce into an individual nerve root bundle quickly,16 the ventral roots consistently formed single, discrete bundles. This is consistent with cadaveric findings of Arslan et al.,16 who found that although the dorsal roots may have multiple discrete bundles at their dural exit site, the ventral roots were consistently found to be a single bundle.16 Combining the ability to properly order the nerve roots based on their ventral root exit site and their discrete nature, they could be reliably counted to ensure that the operation included only the desired ventral nerve roots (L1–S1). Lastly, nerve root counting was always performed from superior to inferior because cadaveric studies have shown that although there is always a dorsal S5 nerve root, there is not reliably a ventral S5 nerve root.17 Without the knowledge of the presence/absence of these nerve roots, the nerve roots could not be reliably counted from the caudal direction.

The second challenge is managing the rotated spinal cord in scoliotic patients. Over half of the patients in this series had scoliosis and had undergone fusion. With the scoliotic curve, the spinal cord itself also rotates. The surgeon should begin on the side of the concave side of the curve because the ventral roots will be rotated anteriorly. Additional challenges in managing the children with scoliosis and fusion included identifying the correct level and performing the laminectomy through the fusion mass. To identify the correct level, typically a combination of anteroposterior and lateral localizing radiographs were taken at the start of the case. We then referenced the preoperative spine radiographs to identify the appropriate level by counting instrumentation screws and identifying ribs. Additionally, once the presumed L1 laminectomy had been performed, ultrasound was used to visualize the conus before extending the laminectomy superior to T12. In these patients with fusion the laminectomy was performed using a combination of drilling and rongeurs. The fusion bone was often hypervascular and diligent waxing of the bone edges throughout the exposure was needed to maintain hemostasis.

Conclusions

A conus-level combined dorsal and ventral rhizotomy is a safe alternative technique for treating children with mixed hypertonia who have failed or maximized other therapeutic options. Technique refinement has helped overcome operative obstacles associated with identifying the correct ventral root levels and performing the procedure in scoliotic children. Although this technique has been shown to be safe, it will require longer follow-up and additional patients with functional measures to show durable functional improvements, as reported by Abdel Ghany et al.9 and Albright and Tyler-Kabara.8

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Ahluwalia, Naftel. Acquisition of data: Ahluwalia, Naftel. Analysis and interpretation of data: Ahluwalia, Lawrence, Naftel. Drafting the article: all authors. Critically revising the article: all authors. Reviewed submitted version of manuscript: Ahluwalia, Bass, Flynn, Martin, Riordan, Naftel. Approved the final version of the manuscript on behalf of all authors: Ahluwalia. Administrative/technical/material support: Naftel. Study supervision: Naftel.

References

  • 1

    Nahm NJ, Graham HK, Gormley ME Jr, Georgiadis AG. Management of hypertonia in cerebral palsy. Curr Opin Pediatr. 2018;30(1):5764.

  • 2

    Verrotti A, Greco R, Spalice A, et al. Pharmacotherapy of spasticity in children with cerebral palsy. Pediatr Neurol. 2006;34(1):16.

  • 3

    Bonouvrié LA, Becher JG, Vles JSH, et al. Intrathecal baclofen treatment in dystonic cerebral palsy: a randomized clinical trial: the IDYS trial. BMC Pediatr. 2013;13:175.

    • Search Google Scholar
    • Export Citation
  • 4

    Albright AL, Barry MJ, Shafton DH, Ferson SS. Intrathecal baclofen for generalized dystonia. Dev Med Child Neurol. 2001;43(10):652657.

    • Search Google Scholar
    • Export Citation
  • 5

    Warsi NM, Tailor J, Coulter IC, et al. Selective dorsal rhizotomy: an illustrated review of operative techniques. J Neurosurg Pediatr. 2020;25(5):540547.

    • Search Google Scholar
    • Export Citation
  • 6

    Park TS, Dobbs MB, Cho J. Evidence supporting selective dorsal rhizotomy for treatment of spastic cerebral palsy. Cureus. 2018;10(10):e3466.

    • Search Google Scholar
    • Export Citation
  • 7

    Wang KK, Munger ME, Chen BP, Novacheck TF. Selective dorsal rhizotomy in ambulant children with cerebral palsy. J Child Orthop. 2018;12(5):413427.

    • Search Google Scholar
    • Export Citation
  • 8

    Albright AL, Tyler-Kabara EC. Combined ventral and dorsal rhizotomies for dystonic and spastic extremities. Report of six cases. J Neurosurg. 2007;107(4)(suppl):324327.

    • Search Google Scholar
    • Export Citation
  • 9

    Abdel Ghany WA, Nada M, Mahran MA, et al. Combined anterior and posterior lumbar rhizotomy for treatment of mixed dystonia and spasticity in children with cerebral palsy. Neurosurgery. 2016;79(3):336344.

    • Search Google Scholar
    • Export Citation
  • 10

    Park TS, Johnston JM. Surgical techniques of selective dorsal rhizotomy for spastic cerebral palsy. Technical note. Neurosurg Focus. 2006;21(2):e7.

    • Search Google Scholar
    • Export Citation
  • 11

    Bales J, Apkon S, Osorio M, et al. Infra-conus single-level laminectomy for selective dorsal rhizotomy: technical advance. Pediatr Neurosurg. 2016;51(6):284291.

    • Search Google Scholar
    • Export Citation
  • 12

    Graham D, Aquilina K, Cawker S, et al. Single-level selective dorsal rhizotomy for spastic cerebral palsy. J Spine Surg. 2016;2(3):195201.

  • 13

    Ou C, Kent S, Miller S, Steinbok P. Selective dorsal rhizotomy in children: comparison of outcomes after single-level versus multi-level laminectomy technique. Can J Neurosci Nurs. 2010;32(3):1724.

    • Search Google Scholar
    • Export Citation
  • 14

    Garriz-Luis M, Sanchez-Carpintero R, Alegre M, Tejada S. Selective dorsal rhizotomy: a review of the literature on this technique for the treatment of spasticity in infantile cerebral palsy. Article in Spanish. Rev Neurol. 2018;66(11):387394.

    • Search Google Scholar
    • Export Citation
  • 15

    Wall EJ, Cohen MS, Abitbol JJ, Garfin SR. Organization of intrathecal nerve roots at the level of the conus medullaris. J Bone Joint Surg Am. 1990;72(10):14951499.

    • Search Google Scholar
    • Export Citation
  • 16

    Arslan M, Cömert A, Açar , et al. Lumbosacral intrathecal nerve roots: an anatomical study. Acta Neurochir (Wien). 2011;153(7):14351442 Published correction in Acta Neurochir (Wien). 2013;155(3):557.

    • Search Google Scholar
    • Export Citation
  • 17

    Hauck EF, Wittkowski W, Bothe HW. Intradural microanatomy of the nerve roots S1-S5 at their origin from the conus medullaris. J Neurosurg Spine. 2008;9(2):207212.

    • Search Google Scholar
    • Export Citation
Panels from figures in Aung et al. (pp 36–46).
  • View in gallery

    Upper: Dorsal roots L2 through S2 have been bundled in silastic, and the silastic bundle has been sutured to the contralateral dura. Lower: By suturing the dorsal bundle to the contralateral dura, the ipsilateral ventral roots are viewed. Figure is available in color online only.

  • View in gallery

    A: The ipsilateral ventral roots are isolated and prepared for identification. B: Sacral roots are separated. The ventral roots are discrete. C: Two ventral roots are isolated for demonstration of their discrete nature. This allows for counting of nerve roots for neurological level identification. D: The L2–S1 ventral roots have been bundled and sutured to the ipsilateral dura. Figure is available in color online only.

  • View in gallery

    Prior to starting the rhizotomy, bilateral dorsal L2–S2 and ventral L2–S1 ventral roots have been bundled in silastic and the bundles have been secured to their ipsilateral dura. The silastic bundle from one side of the dorsal roots has been opened to begin the dorsal rhizotomy. Figure is available in color online only.

  • View in gallery

    Bilateral dorsal rhizotomies have been completed. One ventral silastic bundle has been opened, preparing the ventral roots for rhizotomy. The contralateral ventral silastic bundle can also be seen on the contralateral side, still sutured to the dura. Figure is available in color online only.

  • View in gallery

    Standing radiographs showing 4 of the 7 patients had scoliosis, contributing to the complexity of the surgery.

  • 1

    Nahm NJ, Graham HK, Gormley ME Jr, Georgiadis AG. Management of hypertonia in cerebral palsy. Curr Opin Pediatr. 2018;30(1):5764.

  • 2

    Verrotti A, Greco R, Spalice A, et al. Pharmacotherapy of spasticity in children with cerebral palsy. Pediatr Neurol. 2006;34(1):16.

  • 3

    Bonouvrié LA, Becher JG, Vles JSH, et al. Intrathecal baclofen treatment in dystonic cerebral palsy: a randomized clinical trial: the IDYS trial. BMC Pediatr. 2013;13:175.

    • Search Google Scholar
    • Export Citation
  • 4

    Albright AL, Barry MJ, Shafton DH, Ferson SS. Intrathecal baclofen for generalized dystonia. Dev Med Child Neurol. 2001;43(10):652657.

    • Search Google Scholar
    • Export Citation
  • 5

    Warsi NM, Tailor J, Coulter IC, et al. Selective dorsal rhizotomy: an illustrated review of operative techniques. J Neurosurg Pediatr. 2020;25(5):540547.

    • Search Google Scholar
    • Export Citation
  • 6

    Park TS, Dobbs MB, Cho J. Evidence supporting selective dorsal rhizotomy for treatment of spastic cerebral palsy. Cureus. 2018;10(10):e3466.

    • Search Google Scholar
    • Export Citation
  • 7

    Wang KK, Munger ME, Chen BP, Novacheck TF. Selective dorsal rhizotomy in ambulant children with cerebral palsy. J Child Orthop. 2018;12(5):413427.

    • Search Google Scholar
    • Export Citation
  • 8

    Albright AL, Tyler-Kabara EC. Combined ventral and dorsal rhizotomies for dystonic and spastic extremities. Report of six cases. J Neurosurg. 2007;107(4)(suppl):324327.

    • Search Google Scholar
    • Export Citation
  • 9

    Abdel Ghany WA, Nada M, Mahran MA, et al. Combined anterior and posterior lumbar rhizotomy for treatment of mixed dystonia and spasticity in children with cerebral palsy. Neurosurgery. 2016;79(3):336344.

    • Search Google Scholar
    • Export Citation
  • 10

    Park TS, Johnston JM. Surgical techniques of selective dorsal rhizotomy for spastic cerebral palsy. Technical note. Neurosurg Focus. 2006;21(2):e7.

    • Search Google Scholar
    • Export Citation
  • 11

    Bales J, Apkon S, Osorio M, et al. Infra-conus single-level laminectomy for selective dorsal rhizotomy: technical advance. Pediatr Neurosurg. 2016;51(6):284291.

    • Search Google Scholar
    • Export Citation
  • 12

    Graham D, Aquilina K, Cawker S, et al. Single-level selective dorsal rhizotomy for spastic cerebral palsy. J Spine Surg. 2016;2(3):195201.

  • 13

    Ou C, Kent S, Miller S, Steinbok P. Selective dorsal rhizotomy in children: comparison of outcomes after single-level versus multi-level laminectomy technique. Can J Neurosci Nurs. 2010;32(3):1724.

    • Search Google Scholar
    • Export Citation
  • 14

    Garriz-Luis M, Sanchez-Carpintero R, Alegre M, Tejada S. Selective dorsal rhizotomy: a review of the literature on this technique for the treatment of spasticity in infantile cerebral palsy. Article in Spanish. Rev Neurol. 2018;66(11):387394.

    • Search Google Scholar
    • Export Citation
  • 15

    Wall EJ, Cohen MS, Abitbol JJ, Garfin SR. Organization of intrathecal nerve roots at the level of the conus medullaris. J Bone Joint Surg Am. 1990;72(10):14951499.

    • Search Google Scholar
    • Export Citation
  • 16

    Arslan M, Cömert A, Açar , et al. Lumbosacral intrathecal nerve roots: an anatomical study. Acta Neurochir (Wien). 2011;153(7):14351442 Published correction in Acta Neurochir (Wien). 2013;155(3):557.

    • Search Google Scholar
    • Export Citation
  • 17

    Hauck EF, Wittkowski W, Bothe HW. Intradural microanatomy of the nerve roots S1-S5 at their origin from the conus medullaris. J Neurosurg Spine. 2008;9(2):207212.

    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 561 562 0
Full Text Views 184 184 46
PDF Downloads 168 168 36
EPUB Downloads 0 0 0