Fig. 63.1
The four curvatures of the adult vertebral column: “cervical, thoracic, lumbar, and sacral.” Thoracic and sacral ones exist during fetal life and are called primary (1°) curvatures. The cervical and lumbar ones develop during infancy and childhood and are named secondary (2°) curvatures
When we talk about postoperative kyphosis that develops after posterior fossa tumor operations in children, we generally mean change of curve from cervical lordosis to cervical kyphosis.
Cervical lordosis is defined as a piece of a circle from C2 to T1 by Harrison and colleagues [16]. However, adult data for the cervical lordosis may not actually apply to children from birth to early adolescence.
Bagnall et al. found that up to 9.5 weeks, 83 % of fetuses have a cervical lordosis. This means that by 9.5 weeks, most of the fetuses are starting to use their extensor cervical muscles to pull the cervical curve away from the fetal “C” shape. The lordosis increases during postnatal life when the baby raises his/her head and begins to sit up [17].
These spinal curvatures develop under the influence of multivectorial forces, and this delicate construction is kept in balance by the help of various bony and soft tissue structures (Fig. 63.2).
Fig. 63.2
(a) Right lateral view of the ligaments in the cervical region. (b) Posterolateral view of the ligaments in the cervicothoracic region
With the normal cervical spine lordosis, the weight-bearing axis lies posterior to the vertebral bodies and approximately two thirds of the load is carried by the posterior columns [18].
The loss of the posterior ligamentous and bony elements after cervical laminectomy may shift the weight-bearing axis forward. This results in loss of lordosis and a change towards a straight or kyphotic alignment. That subsequently moves the weight-bearing axis in front of the vertebral bodies. As this kyphotic deformity progresses, the anterior column tends to be compressed and the posterior columns are placed under tension (Fig. 63.3). The surgical trauma weakens the posterior structures that stabilize the cervical vertebrae in their physiologic lordotic position. As a result, the kyphotic deformity progresses [19–21].
Fig. 63.3
Schematic drawings showing lordosis and kyphosis in the cervical spine. Kyphosis tends to be a self-progressive condition because the deformity causes the weight-bearing axis in the cervical spine to move to a more ventral position in front of the vertebral bodies
C2 and its muscular attachments are very important structures that protect physiologic cervical lordosis. Destruction of the second cervical vertebra leads to a highly unstable situation. The axis transfers the axial load of the two lateral masses of the atlas to three surfaces on the third cervical vertebra: the two articular facets and the vertebral body. Thus, pathological processes of this region are often treated less radically compared to other areas of the cervical spine [22].
The extensor cervical back muscles especially the semispinalis cervicis group play a critical role in maintaining cervical lordosis. Consequently, the involvement of C2 and its muscular attachments in laminectomy operations may increase the risk of postlaminectomy kyphosis [23]. Because of the destruction of the posterior column and posterior stabilizing ligament complexes, increasing the number of lamina involved in laminectomy also increases the risk of postlaminectomy kyphosis [24].
63.4 Pathology and Etiology
Deformity develops secondary to a combination of load imbalance and instability [25]. Instability is defined as the “loss of the ability of the spine under physiologic loads to maintain its pattern of displacement so that there is no initial or additional neurological deficit, no major deformity, and no incapacitating pain” by White and Panjabi [26]. Because of this, repetitive unbalanced physiological loads in combination with instability can lead to deformity [25]. The most common surgical procedure in the treatment of spinal and posterior fossa tumors has been laminectomy [27]. To limit postlaminectomy deformities, new techniques such as laminoplasty have been used. The prevalence of postlaminectomy deformities in the pediatric population is higher than in the adult population [27, 28].
There are significant differences between the pediatric and adult human spine. The pediatric spine is associated with greater viscoelasticity of its ligamentous structures. This allows the development of deformity when the supporting posterior structures are weakened. Posterior ligaments including the ligamentum flavum, interspinous and supraspinous ligaments, and the extensor spinal muscles normally counteract the flexion force of gravity on the cervical spine. After laminectomy, these stabilizing structures are damaged and make the spinal cord vulnerable to kyphosis [1]. Also in the pediatric patient, the cervical facets are in a more horizontal plane, the paraspinal muscles are not yet fully developed, ossification is ongoing, and the nucleus pulposus has a higher water content [1, 8, 10, 29]. These factors differ significantly from adults and these should be considered during surgery [25]. Therefore, the incidence of cervical postlaminectomy kyphosis is higher in pediatric patients than in adults [25].
The risk of spinal deformity in pediatric patients is thought to be due to C2 involvement, younger age, preoperative misalignment, irradiation, more elastic ligaments, more horizontal facets, ongoing ossification and remodeling process and weak paraspinal muscles [3, 8, 13, 29–31].
Bell and colleagues reviewed 89 patients with a mean age of 5.7 years who had an average of 4.7 levels removed between C2 and C7. After a 5-year follow-up, they found that 46 (52 %) of 89 patients developed cervical spinal deformity; however, they did not find a correlation between the number of levels removed and deformity [8].
Yasuoka and colleagues reviewed 248 patients younger than 25 years of age and concluded that patients younger than 15 years of age had a higher incidence of postlaminectomy deformity. Yasuoka and colleagues [32] found no link between deformity and gender or the number of laminae resected in their 58 patients [28].
Yeh and colleagues compared laminectomy and laminoplasty techniques with a 45-month follow-up. They found that increasing the number of laminae operated increases the risk of postlaminectomy kyphosis in the cervical region, whereas laminoplasty reduces the risk of postoperative kyphosis [13].
There are many case-control studies in literature concerning the involvement of the cervical vertebra above or below C2 in the laminectomy procedure. Most of them state that in suboccipital craniotomy, the involvement of cervical laminectomies below C2 increases kyphotic deformity in the cervical region.
This is because more fibrous bands and muscles are scarified in axial and subaxial laminectomies. In suboccipital craniotomy with C1 laminectomy, muscular and ligamentous attachments to the C2 spinous process should not be detached. Even partial involvement of the C2 lamina in the laminectomy procedure may leave enough structural integrity to compensate for laminectomy-related instability [25].
Cervical intradural tumors which involve multiple cervical levels and posterior fossa tumors which also involve cervical levels below C2 may require multilevel cervical laminectomy. Multilevel laminectomies also increase the risk of postlaminectomy kyphosis. Sciubba et al. reported that there is an increased risk of kyphosis in the pediatric population after patients had surgeries which require greater than three levels of laminectomies for the intradural tumor resection [24].
63.5 Treatment
There have been many attempts to limit the development or progression of deformity. Many authors suggest postoperative bracing [8, 27, 28, 31]. On the other hand, bracing may actually reduce the role of the paraspinal muscles and lead to atrophy.
Bracing is thought to play a limited role in the long-term management of preexisting deformity. It may prove useful for shorter periods, for early intervention, or perhaps to postpone surgical intervention until growth spurts pass [25].
To limit the chance of postlaminectomy vertebral deformities, the surgeon should first protect the C2 lamina, muscles, and ligaments that are attached on it. Also the surgeon should try not to damage facet joints. It is because most of the strong ligaments responsible for vertebral integrity have attachments to the C2 lamina, spinous process, and facet joints [33, 34].
There are different surgical procedures to limit postoperative vertebral deformity. Iatrogenic deformity mainly develops after dorsal tension band disruption. A variety of surgical procedures, especially laminoplasty techniques, have been tried to protect this structure [35].
There are not so many studies that show the actual benefit of laminoplasty over laminectomy in pediatric patients. Besides this, the most commonly accepted procedure in current practice is laminoplasty.
There are studies that propose instrumentation with fusion during initial surgery [14]; others advise close follow-up with x-rays every 6 months for the first year after surgery and then yearly controls with x-rays until bone maturity is achieved [28].
The time period between initial surgery and diagnosis of spinal deformity ranges from right after surgery to 74 months [14, 28, 36, 37]. Close follow-up and early diagnosis and intervention may prevent the progression of deformity. Not all patients with iatrogenic spinal deformity need to be surgically corrected so early because surgical correction may lead to many unnecessary and expensive procedures in these children.
Before making a decision about surgery, patients should be evaluated for cervical instability with flexion and extension x-rays using the criteria described by Panjabi et al. [21]. According to these criteria, the presence of more than 3.5 mm or 20 % sagittal plane displacement and more than 20° of sagittal plane rotation on flexion-extension x-rays and also the presence of 3.5 mm sagittal plane translation or more than 11° sagittal plane angulation on neutral x-rays are suggestive findings of instability. It should be noted that the Panjabi instability criteria were based on an adult spine research investigation, and there are no parallel studies to determine the criteria for instability in pediatric patients.
Patients with signs and symptoms of instability such as excessive pain, gross deformity and neurological symptoms should be treated surgically [25].
In surgery, anterior fixation, anterior and posterior fixation, and posterior fixation alone are options.
Posterior fixation is generally indicated in the setting of posterior column instability [38, 39]. One of the earliest posterior cervical fusion techniques was sublaminar wiring which was first described in the early 1890s for the management of deformity caused by Pott’s disease [39]. Early strategies for craniovertebral junction or cervical spinal stabilization used simple onlay bone graft technique. Suboccipital and sublaminar wiring techniques were used for stabilizing the onlay graft. However, to achieve optimal stability, more cervical vertebral levels should be involved in this technique which limits the patient’s motion range [40, 41] (Figs. 63.4, 63.5, and 63.6).
