Growing Spine Case: 5-Year-old with Increasing Spinal Deformity
A 5-year-old male, with an uncharacterized congenital myopathy, presented for increasing spinal deformity. He was ambulatory, but was having some recent "balance" issues, which the parents reported as frequent falling due to spinal and trunk imbalance, although the patient was also afflicted with "unstable" knees, for which he wore braces to aid stability.
The patient's family history was notable for an older sibling with a similar myopathy, who was more severely involved and currently wheelchair bound. His review of systems was otherwise unremarkable, and there was no history of respiratory impairment, frequent upper respiratory infections, or other infections. A brace had been prescribed by another physician, but was not tolerated due to pain from pressure over the rib prominence.
Clinical evaluation showed a tall, thin, 5-year-old male with significant left convex kyphoscoliosis, lumbar hyperlordosis, and generalized weakness, especially around the pelvic and shoulder girdles (Figure 1A-1E).
The spinal deformity was noted to be quite flexible with underarm suspension of the trunk. He had a +ve Gower's sign, hyperextensible joints, his knees showed 3+ laxity in all planes, and his patellae were easily dislocatable, although there was little discomfort associated with provoked dislocation.
When wearing braces (modified knee "cages" with outside hinges designed to prevent excessive varus-valgus and anterior-posterior movement), the patient was able to ambulate, but had become cautious due to the unbalanced forward position of the trunk combined with his hip girdle weakness.
Spine x-rays (Figures 2A-2C) showed a 102-degree thoraco-lumbar curve with significant kyphosis. However, a supine stretch film demonstrated significant correctability (Figure 2C).
Suggest TreatmentIndicate how you would treat this patient by completing the following brief survey. Your response will be added to our survey results below.
Fusionless instrumentation was recommended based on the patient's age, curve flexibility, and intolerance of external bracing. Preoperative pulmonary function tests showed a forced vital capacity of 1.02 L. (57% pred.), although it was noted that the patient was unable to fully exhale due to "weakness". At age 5+9, the patient underwent implantation of fusionless instrumentation from T1-L4, producing local fusion of the cephalad hook claw anchors at T1-T2 and the distal L3-L4 pedicle screw claws (Figure 3A, 3B).
An additional C7 sublaminar wire was used to augment the upper claw construct against posterior pullout. Using long subfascial rods from the lumbar anchors, contoured for appropriate sagittal plane alignment, the thoraco-lumbar kyphosis was corrected by "convex" technique pushing the kyphotic apex ventrally. The caudal rods were dominoed to shorter thoracic rods, again contoured for kyphosis, in an area where subsequent distraction would not produce undesired flattening or rod prominence, because the overlapping rods had the same radius of kyphotic contour appropriate for the mid thoracic spine (Figure 3C).
Spinal elongation and deformity correction was quite satisfactory (Figures 3A, 3B, 3C). The Cobb angle was reduced to 45o, while overall T1-S1 length was increased from 25.2 to 34.1 cm as an immediate result of the implantation and distraction. T1-T12 length was increased from 17.4 to 19.2 cm immediately, while T6 thoracic width decreased from 17.5 to 16.9 cm. In follow-up, the patient's imbalance problems improved, most likely due to the improvement in sagittal balance (Figure 4A-4C).
Over the next 3-½ years, 4 additional lengthenings were performed. At age 9+6 (Figure 5A, 5B), the T1-S1 length was 39 cm, T1-12 = 25 cm, and T6 width = 19.8 cm.
To place these values in perspective, Dimeglio reports normal T1-S1 length to be 35 cm at age 10 (45 cm at maturity) in males. Emans' data suggests this T1-T12 length is about the fiftieth percentile based on pelvic width (Figure 6).
On the other hand, the other thoracic parameters, T6 coronal width and T6-sternum sagittal width, were fifth percentile or less (Figures 7, 8).
Pulmonary function tests performed at this time (the patient was again noted to have difficulty with the performance of the test) showed minimal change in absolute forced vital capacity = 1.07 L. (39% pred.), perhaps representing lack of pulmonary development due to failure of the thorax to grow in the coronal width and sagittal depth components, even though the thoracic length seemed to have been sufficiently lengthened. Obviously, the congenital myopathy also played an important role in the forced vital capacity impairment.
Currently and Future Options
The patient has just recently had rod revision to continue lengthening. Discussion now centers on how much further to attempt to lengthen based on length parameters, which appear to show adequate spinal length. Options in the future include a period of observation until puberty, and if progression occurs, conversion to a final correction and fusion. Due to the underlying myopathy and pulmonary status, avoiding an anterior procedure to control crankshaft would be attractive, and thus management by posterior techniques only is the goal.
The goal of treatment of Early Onset Scoliosis (EOS) is to achieve and maintain a reasonable correction of scoliosis and to allow the growth of the spine to continue. Normalization of the thoracic growth is believed to be a sound strategy to counteract the pulmonary morbidity seen with EOS, regardless of its etiology. This also prevents worsening of pulmonary function. Although there is no conclusive evidence to indicate that the restoration of thoracic volume leads to better pulmonary function, it seems reasonable to assume that improving thoracic volume will provide a better environment for lung function.
In the child presented, the spinal deformity involves the thoraco-lumbar and lumbar spine rather than the thoracic spine and his space available for the lungs seems to be in the normal range, although the function may be abnormal because of the primary neuromuscular disease. Therefore restoration of the thoracic space is not the primary issue in this case, although prevention of deformity progression is absolutely mandatory if this patient is to avoid functional and pulmonary morbidity later. Therefore we should discuss the best treatment option to provide us with the best results.
The goal of treatment in this patient should be to relieve him from using a brace which has caused some pressure problems, interfering with his care and has failed to control his curve. The goals are not only to make him brace free, but to allow a better balance of the spine and hopefully prevent frequent falls. With his curve being flexible enough and not having any congenital spinal anomalies, he is an excellent candidate for growing instrumentation technique. I agree that it is best to avoid approaching the spine anteriorly thereby avoiding further pulmonary compromise.
There are two basic posterior growing techniques available to treat this type of deformity; surgery on the spine and that on the rib cage. Surgical procedures for the ribs are reserved for significant pulmonary problems and thoracic insufficiency syndrome and when there are congenital spinal anomalies present.
I feel that posterior growing rod is an excellent choice for surgical treatment of this case (1, 5) where there is a normal chest wall without anomalies, and any chest wall deformity would result from it being imposed by a progressive spinal deformity. One can debate whether rib attachment devices are appropriate, or in fact can even be harmful, in the setting of a normal chest wall where non-rib attachment devices can prevent the superimposed deformity.
It should be emphasized that no matter what type of treatment is used, managing children with EOS is challenging. Underlying problems such as myopathy in this case, multiple surgeries, and length of treatment period all contribute to possible complications. (2) Dual growing rod technique was used in this case and instrumentation spanned from the upper thoracic to lower lumbar spine, which is necessary in most cases of neuromuscular etiology to provide adequate correction and balance. This also reduces the possibility of adding more levels to the curve and the need for further extension of the spinal fusion.
Post-operative radiographs show good correction in both coronal and sagittal alignment. It is important to achieve the most stable anchors. The anchors are well placed and additional support is provided by sublaminar wires at the upper thoracic spine. The anchors may be hooks, screws, or a combination. A transverse connector can increase the foundation strength if only one set of hook claws is utilized. (3) The connection between two rods can be end-to-end versus side-to-side. If the end-to-end connector is used, the location of the connectors is usually at the thoraco-lumbar junction to have minimal affect on the sagittal alignment. The side-to-side connector used in this case allows more contouring of the rods and conforming to the sagittal alignment and kyphosis.
The frequency of lengthening is also important for achieving maximal correction and length. We are now recommending routine lengthening every 6 months. (4) The patient has had several lengthening and the initial correction and further growth over the treatment period is very close to the normal growth of the spine. Also the length of the thoracic spine is within normal range. So the question is raised regarding when one should stop the lengthening and do the final fusion. Timing of the final procedure varies in each case. The difficulty seems to be the fact that prediction of the growth rate may be difficult when there is an underlying disease. In a normal situation, one expects some crankshaft and rotational deformity to develop if the final fusion is performed at the age of 9+6. To avoid anterior fusion, our approach is to continue lengthening, maybe less frequently, until the patient has stopped growing. One may leave the instrumentation until it fails and then, either remove without fusion, or do the final fusion.
1. Akbarnia, B., et al., Dual growing rod technique for the treatment of progressive early-onset scoliosis: a multicenter study. Spine, 2005. 30(17 Suppl): p. S46-57.
2. Akbarnia, B., et al. Complications of Dual Growing Rod Technique in Early Onset Scoliosis: Can We Identify Risk Factors? in 41st Annual Meeting of Scoliosis Research Society. 2006. Monterey, California, USA.
3. Bagheri, R., et al. Biomechanical Comparisons of Different Anchors (Foundations) Used in the Growing Dual Rod Technique. in IMAST 11th Annual Meeting. 2004. Southhampton, Bermuda.
4. Akbarnia, B., et al. End results of dual growing rod technique followed until final fusion. The effect of frequency of lengthening. Spine (Accepted for publication ) April 2008.
5. Thompson, G., et al., Comparison of single and dual growing rod techniques followed through definitive surgery: a preliminary study [In Process Citation]. Spine, 2005. 30(18): p. 2039-44.