Ebook: Research into Spinal Deformities 2

In June 1998, the International Research Society (IRSSD) held its second biannual scientific meeting in Burlington, Vermont on the shores of Lake Champlain. This book contains the Proceedings of that meeting in the form of short papers summarizing the papers and posters that were presented then. This year, the meeting returned to its original venue, in that the Society and these meetings are a direct result of a series of biannual meetings starting in 1980, when we met at Basin Harbor, Vermont. From 1980 to 1992 the meetings steadily evolved from a primary interest in three-dimensional and morphological aspects of spine and trunk asymmetry, towards improved understanding of mechanisms underlying the etiology, pathomechanism, and treatment of spinal deformity. The formation of the IRSSD in Pescara in 1994 reflected a desire to consolidate the momentum of these meetings, as well as to broaden the scope. The scope of the 1998 meeting, and of this book, includes work by researchers seeking answers to fundamental questions ranging from the causes of scoliosis, its mode of progression, to the conservative and surgical management of patients, especially children, with spinal deformities. Furthermore, several papers address emerging possibilities for more rational and predictable planning of treatment using deterministic models and knowledge bases.

These scientific proceedings provide the opportunity for readers to learn about the latest information in this field. Some areas of research have been further consolidated, while some new and exciting new areas are emerging. The meeting allowed us to discuss and speculate on the latest scientific ideas in a collegial, international atmosphere. There were 96 delegates from 22 countries. The meeting format intentionally encouraged in-depth communication, discussions, and debate of ‘hot’ issues. Colleagues in Montr´al organized pre-meeting workshops on Clinical Applications of 3-D Evaluation of Scoliosis. The IRSSD meeting included eight sessions for oral presentation of papers, two poster sessions, two invited keynote addresses, and three debates of controversial topics, all summarized in this book.

Ian Stokes

Burlington, August 1998

Although the primary cause may not lie in these structures, the deformity in scoliosis arises mostly from the vertebral bodies and intervertebral discs. Changes in mechanical stress can affect both vertebral body growth and disc remodelling, thus leading to wedging and the deformity.

Currently, assessments of scoliosis are done by means of clinical examination and full spinal x-rays. Multiple exposure to ionizing radiation, however, can be hazardous to the child and costly. Here we explain the use of a noninvasive imaging technique for quantifying the three-dimensional (3D) trunk surface topography that can be used to estimate parameters of 3D deformity of the spine. A laser optical scanning system produces a topographical mapping of the entire torso. In conjunction an accurate 3D reconstruction of the spine and rib cage can be developed from the digitized x-ray images of the scoliotic patient. Data from the surface laser scans and bone geometry from x-rays have been collected for 26 scoliotic patients. The technique provides the foundation for future work using neural networks to assess relations between internal bone geometry and external surface geometry of the torso.

The detection of anatomical landmarks on the body surface measured by rasterstereography is based on the analysis of surface curvatures. Landmarks such as the vertebra prominens or the lumbar dimples are localised at points of maximum convex or concave curvature. It is shown that the accuracy (reproducibility) of landmark localisation correlates with the magnitude of local curvature. This finding makes it possible to estimate the accuracy of a landmark a priori from a single record, instead of determining the statistical scatter from a large number of records taken in the same posture.

The reliability of repeated measurements of thoracic kyphosis from rasterstereographic back shape imaging was examined in 10 volunteers across different age cohorts and in a static back phantom. Coefficients of variation (CV) for five repeated measurements of the 10 subjects ranged from 2.4% to 3.0% for the kyphosis parameters ‘Maximum’, ‘Vertebra prominens to thoracolumbar inflexion point’, and ‘Vertebra prominens to T12’. Mean standard error of measurement values (SEM) as a percentage of mean kyphotic angles were 5.3%, 4.4% and 4.8% for the same three kyphosis parameters, ‘Maximum’, ‘Vertebra prominens to thoracolumbar inflexion point’, and ‘Vertebra prominens to T12’ respectively. The inter-trial CV for repeated imaging of the back phantom ranged from 0.4% to 1.3%. Intraclass correlation coefficients (ICC) ranged from 0.98 to 0.99 for all measurements. Consistent results were demonstrated across the varying age cohorts. The reliability of rasterstereographic evaluation of thoracic kyphosis is largely influenced by postural variations, while minimal errors are attributed to internal system inaccuracies. These results suggest the clinical utility of rasterstereographic imaging in the detection of relatively small curve changes and hence in monitoring of kyphosis progression.

This study examined the association between thoracic kyphosis derived from rasterstereographic measurement of back shape and lumbar bone mineral density (BMD) in 31 female subjects. Back shape imaging was performed following routine lumbar densitometry utilising dual energy X-ray absorptiometry. Thoracic kyphosis parameters were moderately correlated with mean BMD when second order polynomial regression models were fitted to the dataset. The association between accentuation of kyphosis and reductions in BMD was further confirmed following conversion of mean BMD values to age-matched Z-scores. These data conform to previous studies investigating radiographic indices of thoracic curvature and clinical lumbar BMD. Non-invasive examination of thoracic kyphosis may hence complement other existing methods for detecting early indicators of spinal osteoporosis.

An in vivo study investigated 2D and 3D relationships between global and local vertebral and discal descriptors of the scoliotic spine deformity. It was performed on a cohort of 40 patients representing a wide range of thoracic curve progression. Global descriptors referred to Cobb angle in PA view as well as corresponding Cobb angle in the plane of maximum deformity and its angular orientation. Local descriptors included vertebral axial rotation, wedging angles in the global and local vertebral frontal planes as well as a 3D maximum wedging angle and its angular orientation. Statistical analyses indicated that, at the thoracic level, vertebral wedging and axial rotation increase with curve severity. The 3D maximum wedging tends to shift towards the left posterolateral vertebral region with curve severity, with simultaneous displacement of the plane of maximum deformity toawrds the coronal plane. These results confirm the importance of 3D measurements in the understanding of biomechanics of idiopathic scoliosis.

This study evaluates a registration technique for intraoperative tracking of the spine which consists in matching intraoperative measurements with specific surfaces defined on a 3-D preoperative model. The validation study was undertaken on a cadaveric spine. The accuracy for four approaches combining different tracking systems and 3-D reconstruction techniques were compared. The accuracy resulting from the combination of a radiographic 3-D reconstruction and a magnetic digitizer was 5.9±2.7mm. The corresponding errors on vertebral rotations were 4.4±3.3°, 6.7±5.8° and 5.0±3.8° in frontal, sagittal and transverse planes, respectively. This approach is minimally invasive (only 2 X-rays) and may provide sufficient accuracy for certain clinical applications. With CT scan 3-D reconstruction, the accuracy was increased by about 2mm, but the high radiation exposure associated with CT scan imaging for long spinal segments makes it unfavorable to most clinical use. As for the mechanical arm, considering the small increase in accuracy and its awkwardness, its use during surgery is not suggested.

This paper presents a comparison of 4 different methods used to describe the “plane of maximum deformity” of scoliotic spines. It was conducted on 31 scoliotic subjects with a right thoracic curve. Most techniques are reproducible and are little influenced by reconstruction errors. The measurement of the plane of maximum deformity gives quite different orientations and there is not a “best technique”, even if most of them are correlated. The spinal curves presented different shape patterns, which may explain the discrepancy in the results between the methods. We recommend using the segment between the end vertebrae to calculate the plane of maximum deformity, and being careful when comparing published results using these techniques because they are not equivalent, even if correlated.

A mathematical model was created to predict the location and orientation of vertebrae relative to artificial markers. Because the stereovision system detects the position and orientation of artificial markers, the model was developed to transform the marker motion into vertebra motion. Geometry, vector analysis, and stereo-radiographs were used to relate die artificial markers to the vertebrae. Experiments were done using stereo-radiographs and an electromagnetic digitizer to determine die orientation and position of vertebrae in a simulated operating room (OR) environment. Results demonstrated Uiat the mathematical model performed as expected when supplied accurate data from an electromagnetic digitizer. Systematic errors in stereo-radiographs however resulted in 3D reconstruction inaccuracies of 6-7mm

Surface topography is used in spinal deformity to monitor cosmetic change. In order to be useful, it must not only yield values that correlate with radiological data, but must reliably change in tandem with radiological change. Using the average of 4 consecutive repeat scans to reduce error margins in a Quantec Imaging System, changes on radiograph and in several parameters of contemporary scans were analysed to estimate their agreement. Enough correlation was found to encourage further study.

CPT (complex phase tracing) profilometry was selected as a method of choice for clinical use in back surface topography recording. Two mutually out-of-phase patterns are projected on the back surface of the patient, recorded and demodulated according to the CPT procedure. Ten subjects have been investigated, 3 subjects being healthy controls. The system was reliable in measuring back shape with millimetric definition (+/- 1 mm). It proved to be fast in acquiring topographical data, requiring less than 0.3 sec and, then, overcoming problems related to breathing of the patient. Data analysis was enough fast to provide a visual output within few minutes, allowing the examiner to perform a further acquisition, if needed, and the patient to appreciate the output of the analysis. Presence of asymmetry was easily detected and clearly highlighted; improvements related to treatment was evidenced in those subjects where a follow-up exam was performed.

This paper reports a pilot study analysing axial rotation in patients with adolescent idiopathic scoliosis using three-dimensional MR. The objective was to define the proportion of segmental axial rotation that occurs due to intravertebral deformity in patients with adolescent idiopathic scoliosis. Ten patients with adolescent idiopathic scoliosis with right thoracic curve (Cobb angle 44° – 78°) selected sequentially from clinic were included in the study. Patients were imaged with a Siemens IT Impact scanner using dual echo steady state gradient echo T2-weighting (TR 30msec/ TE 9/45 msec/ 40°). Three-dimensional volume images of the apical ten vertebrae were obtained in the axial plane and were post-processed through multiplanar reconstruction, allowing axial reconstructions to be obtained in the plane of each endplate. Axial rotation was measured, and an absolute value determined by reference to a neutrally-rotated vertebra. The proportion of intravertebral and intervertebral deformity within each scoliotic curve was subsequently determined. Overall change in observed axial rotation from end vertebra to apical vertebra ranged from 20° to 34°, with a mean of 28°. The mean proportion of axial rotation within the overall scoliotic curve occurring due to intravertebral basis was 34%, with a range of 9% – 76%. It was concluded that a significant but variable amount of the overall scoliotic deformity in patients with adolescent idiopathic scoliosis occurs as a result of intravertebral rotation, contributing over 45% of the total scoliotic axial rotation in half of the patients imaged. This study implies that assessment of axial rotation in the plane of individual endplates with three-dimensional MRI may be a useful means for identifying a subgroup of patients in whom derotational surgery is likely to be of limited benefit.

Adolescent idiopathic scoliosis is a continuous process that produces a three-dimensional deformity. Observation of radiographs taken at irregular intervals fails to provide a continuous view of curve development. Morphing is a computer technique by which a ‘start’ image can be blended into an ‘end’ image. The application of this technique to sequential pairs of images collected from a patient during visits to the clinic produces movies in which die continuous development of scoliosis can be seen. Such movies can be created from radiographs, cosmesis images, or combined images where the radiographs have been superimposed on the cosmesis images. Real-time movies can also be created showing appropriate patient data to accompany the curve development. The movies have been found to be especially useful in patient education but are also expected to provide improvements in clinical management as well as information relating to the aetiology of AIS. It must always be remembered that the interpolated images are computer generated but there are few limitations of Uie technique in relation to validity.

This paper reports a new real-time ultrasound method for measuring vertebral rotation to detect lateral spinal curves of adolescent idiopathic scoliosis (AIS) in 50 adolescents referred consecutively to hospital from the scoliosis school screening service in Nottingham, England (girls 39, boys 11, median age 15 years). Radiographically the curve types were thoracic 21, thoracolumbar 14 and lumbar 15 with a mean Cobb angle of 17° (range 2-40°). Ultrasound laminal rotation was measured in the prone position from Tl-Sl by one of three observers. The ability of ultrasound laminal rotation to predict the scoliosis curve (Cobb) angle was compared with surface back shape angles calculated at 18 levels from angles of trunk inclinations (ATIs) measured by a Scoliometer in the standing forward bending position. The findings show that ultrasound laminal rotation (and Scoliometer ATI) each with a threshold of 8° predicted Cobb angles of 20° or more with a sensitivity of 95% (79%), specificity of 68% (35%) and positive predictive value of 64% (43%). Overall, the ultrasound method is significantly better than the Scoliometer method for discriminating curves with Cobb angles of 20° (p=0.003, McNemar test). The findings suggest that in school screening for scoliosis ultrasound laminal rotation measured in the community may reduce the number of false positives who are referred to hospital and x-rayed.

We studied 40 normal subjects and 136 patients with idiopathic scoliosis. Both the Quantec Spinal Image System (QSIS) and a AP x-ray view had evaluated all of them. Discriminate analysis showed functional scores can be used as classification of scoliotic groups with different Cobb angles. It supports that classification of scoliosis is able to be based upon the QSIS metrics rather than X-ray exclusively. Functional analysis is best to distinguish patients with less than 8° of Cobb angle from those with more than 20°.

A new indicator to assess trunk asymmetry in scoliosis patients is introduced. The medio-lateral length difference index at the axilla and the waist, and off balance index at C7 were defined as Frontal Asymmetry Index (FAI-C7, FAI-A, FAI-T). The height difference at the shoulder, the axilla and the waist were defined as Height Difference Index (HDI-S, HDI-A, HDI-T). The total sum of these six indices was defined as Posterior Trunk Symmetry Index (POTSI). The average intra-observer and inter-observer error of POTSI were 5.5 and 6.4 respectively. Slight trunk asymmetries were detected in normal children, whereas POTSI were significantly larger in scoliosis patients than those in normal children. This measurement of trunk asymmetry defined as POTSI is valid for the clinical assessment of scoliosis and to evaluate objectively the effect of surgery on trunk deformity.

Moiré pictures of fifty-five normal children and 195 cases of scoliosis patients were analyzed. Among them 40 cases were treated surgically by 3-D Harrington instrumentation. Cobb angle, and the coronal plane asymmetry indices (FAI, HDI), POTSI and Hump Sum were analyzed. A slight trunk asymmetry were detected in the normal children. The mean values of all the parameters in the scoliosis group were significantly larger than those in the normal cases (p< 0.01). A weak correlation was found between POTSI and Cobb angle (r=0.435, p< 0.0001). The preoperative POTSI of 46.9 decreased to 24.3 postoperatively. Lateral curvature and trunk deformity including hump and trunk asymmetry should be evaluated separately because these are three different major factors of scoliois. Therefore, Cobb angle, Hump Sum and POTSI must be measured in the evaluation of scoliosis treatment. The new parameter introduced here as POTSI is clinically useful to assess the trunk deformity quantitatively.