Biomechanical Evaluation Of Anchor Sites In Early Onset Scoliosis Treated With Growing Rod Technique:from Clinics To Basic Study

1. Research backgroundEOS (early onset scoliosis) is defined as scoliosis diagnosed at the age of five or less. Because of the young age and rapid progression, patients whose chests and abdomens at the early stage of development are likely to develop shorter trunk and cardiopulmonary function failure and even end up living by ventilators if not treated in a timely and effective manner. So the main task in treating EOS is not to correct deformity, but to curb progression and allow for spine growth and development of thorax and visceral organs.Traditional treatment options for early onset scoliosis consist of bracing and spinal fusion. Early studies indicated that brace or cast treatment can delay the need for spinal fusion to treat infantile and juvenile idiopathic curves. However, it is often ineffective for severe scoliosis or cases with congenital or neuromuscular etiology. Moreover, bracing for early onset scoliosis may have adverse effects on thoracic rib cage growth and pulmonary development. Spinal fusion is effective in correcting and preventing curve progression but could lead to spine growth loss and pulmonary development restriction, while have a trend to "adding on" and crankshaft phenomenon.Non-fusion operation is the dominant choice of EOS treatment. Now one of the most widely applied non-fusion techniques in treating EOS is growing rod technique (single rod or dual rod), which allows for spine/thoracic cage growth and cardiopulmonary development by fixing screw anchors at the ends of curve and then placing a unilateral or bilateral growing rod subcutaneously or submuscular so as to achieve spine distraction at regular intervals. Despite of its great advantages, the accompanying clinical complications are still worrisome for most spine surgeons. And of them implants complications account for most which mainly result from improper anchor configuration. Therefore, as there has been a lack of related clinical and basic research on details of such complications and their specific mechanism, this paper aims to investigate the problem from clinical and biomechanical perspectives in order to provide reliable biomechanical evidences for spine surgeons in selecting proper anchor sites. 2. Research Objectives(1) To evaluate surgical outcomes of growing rod implantation in infantile and juvenile scoliosis and analyze the causes of complications, especially characteristics of implant complications and bring up the scientific problems to be solved through biomechanics.(2) To carry out distraction load-to-failure tests in the rib, lamina and pedicle of the thoracic and lumbar spine specimen aged0to5and evaluate distraction load-to-failure forces of three different anchor sites and compare that of different segments in different anchor sites.(3) To evaluate the distraction load-to-failure forces of double anchored segments in the whole spine specimen with intact rib cage and compare it with that of single anchored segment.3. Research Contents1. Analyze and Summarize the Features of Complications of Growing-rod Treatment for EOS and Bring up the Problems to Be Solved in Clinical Practice.This is a retrospective study of16patients with EOS treated by growing rod technique (single growing rod and dual growing rod) admitted to the scoliosis center of the First Affiliated Hospital of Sun Yat-sen University from September2007to October2012. All patients have adequate clinical radiographic materials and records of complications. In this paper, characteristics of implant complications which is most common in growing-rod treatment and the surgical outcomes are analyzed in detail and a solution for selecting proper anchor site is proposed combined with literature review.2. Evaluate Distraction Load-to-failure Forces Supported by Rib, Lamina and Pedicle of Mono-segmental Thoracic.After radiological and bone age evaluation of20pediatric cadaveric spines with rib cages. Two specimens were used for preliminary experiment, another eighteen of them were randomly assigned into three groups:Group A (n=6) for lamina (A1) and pedicle (A2) load-to-failure test, Group B (n=6) for rib (B1) and pedicle (B2) load-to-failure test and Group C (n=6) for rib (C1) and lamina (C2) load-to-failure test. Each thoracic specimen was sectioned into six test units (T1-T2、T3-T4、T5-T6、 T7-T8、T9-T10、T11-T12) from T1-T2to T11-T12. After special embedment and fixation with polymethylmethacrylate, a longitudinal load-to-failure test in single segmental lamina and pedicle (Group A), single segmental rib and pedicle (Group B), single segmental rib and lamina (Group C) simulating growing rod distraction force was performed with an ElectroForce(?)3500machine, and yield forces of three different anchor sites and different segments were statistically analyzed with pair t test and ANOVA by using SPSS17.0statistical software.3. Evaluate Distraction Load-to-failure Forces Supported by Bi-segmental thoracic and lumbar anchor.6pediatric cadaveric spines with intact rib cages were used after radiological and bone age evaluation. Each specimen was sectioned into eight test units (T1-T2、 T3-T4、T5-T6、T7-T8. T9-T10、T12-L1、L2-L3、L4-L5) from T1-T2to L4-L5. After embedment and fixation with polymethylmethacrylate, a longitudinal load-to-failure test in double thoracic ribs (Group D1)/laminas (Group D2)/pedicles (Group D3) of the same thoracic test unit and double lumbar laminas (Group E1)/pedicles(Group E2) of the same lumbar test simulating growing rod distraction force was performed with an ElectroForce(?)3500machine, and yield forces of different anchor sites and different segments were statistically analyzed with AVONA or pair t test by using SPSS17.0statistical software.Results1. Clinical Treatment Results of Growing Rods in Infantile and Juvenile Patients with Early Onset ScoliosisSub-muscular growing rods in16children with EOS are successfully distracted for79times, of which40are single rod distraction and39are dual rod distraction. The mean distraction times for each patient are4.9. The mean age at the first surgery is52.6±36.3months, with a mean follow-up period of31.8±14.4months. At the last follow-up, the main curves are41.6%corrected; the mean length increase in anchored segments is2.7±2.3cm, with a lengthening speed of1.1±0.6cm per year; the height of thoracic cage in both concave and convex sides increases by2.9±1.6cm and2.3±2.2cm respectively. Among these16cases,9(56%) suffer from23times of complication, with a mean complication rate of27.8%at each surgery (40%in single rod distraction compared to17.9%in dual rod distraction, P=0.139). Complications related to implants are most common in growing rods therapy, accounting for73.9%, of which anchors failure take up70.6%.2. Distraction Load-to-Failure Test Results in Mono-segmental Pedicle, Vertebral Body, and Lamina of Thoracic SpineExperimental results in Group A show that the distraction load-to-failure forces of lamina and pedicle are close to each other, being222.4±107.3N and231.6±163.2N respectively without a statistical difference (P=0.159). Results in Group B and Group C show that distraction load-to-failure forces of lamina and pedicle are approximately twice as much as those in the ribs. In Group B, the load-to-failure distraction force of rib is129.3±35.1N, while that of the pedicle is212.1±71.2N. In Group C, the load-to-failure distraction force of rib is87.5±42.9N, while that of the lamina is183.6±77.4N. Data in Group B and Group C both indicate statistical difference (P<0.01). Moreover, all experimental results indicate that the distraction load-to-failure forces of T3, T5and T7pedicles and laminas are lower than those of proximal/distal thoracic segments. Pedicle breach mainly occur at T5and T7levels, which provides75.6%of distraction force as compared to the same segments with intact pedicle insertion.3. Distraction Load-to-Failure Test Results in Different Bi-segmental Thoracic/Lumbar Anchor sitesThe mean distraction load-to-failure force in Subgroup D1(bi-segmental ribs), Subgroup D2(bi-segmental thoracic laminas) and Subgroup D3(bi-segmental thoracic pedicles) are373.3±92.6N,444.6±120.3N and473.9±118.4N, respectively. Results show that the distraction load-to-failure forces of bi-segmental pedicles and bi-segmental laminas have no statistical difference (P=0.336), while those of bi-segmental thoracic laminas/bi-segmental thoracic pedicles and bi-segmental ribs have significant differences (Both P value were0.021and0.001); distraction load-to-failure forces of bi-segmental pedicles, laminas and ribs are2.0,2.1and2.6times those of mono-segments respectively (All P value<0.01). It indicates that the three types of distraction load-to-failure forces in thoracic spine increase as anchored segments increase. Moreover, the results showed no significant differences existed among different segments of the same anchor site,(All P value>0.05).The distraction load-to-failure forces of Subgroup E1(lumbar bi-segmental laminas) and Subgroup E2(lumbar bi-segmental pedicles) are822.1±216.7N and951.4±143.1N respectively, with statistical significant difference (P=0.010), which were1.8times and2.0times as much as those of bi-segmental thoracic laminas and pedicles (Both P value<0.01). We also find no significant differences existed among different segments when the data of the same anchor site were used to analyze, P value all>0.05.Conclusions1. Analysis of clinical outcomes and complication in16EOS case indicates that growing rod technique can successfully control the progress of EOS and allow development of spine and thoracic cage; Implant-related complications are the main complications in growing rod therapy (accounting for73.9%), of which70.6%are related to anchor sites.2. The potential causes of anchor complications including single growing rod, rib hook, proximal single hook, hooks with the same direction, main concave side and multipledistraction, moreover, flexiable anchoring segments and rigid anchoring segments are the risk factors for hook dislodgement and hook anchor site failure, respectively, so it has important clinical value for the biomechanical characteristics study of different anchor constructs;3. Infantile thoracic mono-segmental pedicles and lamina can provide distraction load-to-failure force similar to each other, which is twice as much as that of mono-segmental rib.4. Based on the results of this study, it is not recommended to use proximal mono-segmental rib hooks in growing rod treatment. 5. Mono-segmental thoracic distraction load-to-failure test results indicate that the distraction load-to-failure forces at T5and T7are lower than those of other segments. It may be due to their anatomical structures, but further biomechanical study is needed to discover the specific reasons.6. Pedicle breach may potentially reduce its capacity against distraction. If pedicles are breached intra-operatively, it is suggested to control correction or change anchor strategy so as to prevent from intra-operative or post-operative anchor construct failure.7. Distraction load-to-failure force in thoraccic bi-segmental anchor constructs may twice compared to that in mono-segmental constructs.8. Distraction load-to-failure forces in lumbar bi-segmental laminas and pedicles are obviously greater than those in thoracic bi-segments. It may indicates that the upper edge of lamina and the lower edge of pedicle may provide a greater distraction load-to-failure force compared to the lower edge of lamina and the upper edge of pedicle.