July 2017 - Donna Magid, M.D., M.Ed.

Scoliosis: Basic Reporting Template

The normally aligned spine in upright stance has a plumb line (center of gravity) which on lateral centers the foramen magnum over the thoracolumbar junction, proximal sacrum, knee, and plantar midfoot. On frontal views the midpoints of these landmarks should form an axis, without displacement to either side (ie head directly over spinous processes, mid sacrum, symphysis pubis). Side-to-side curvatures are scoliosis, anterior-posterior curves (normal or changed) are either kyphosis (concave anteriorly, as the normal thorax) or lordosis (concave posteriorly, as normal lumbar spine). If image demonstrates plumb line deviation, describe as “head carried approximately 7 cm. left lateral to sacrum” or “standing, pt. leans significantly left’ (Clinicans’ notes and older reports may help).

Less than 10 degrees scoliosis is considered clinically insignificant. Describe as ‘minimal, less than 10 degrees…’, or ‘approximately 7 degrees…’ and if possible compared to priors.

# of IMAGES: most are composites, ‘stitched’ or ‘fused’; so state ‘3 frontal views stitched’ or ‘3 frontal, 3 lateral, fused’,or “Scoliosis, 6 views”, etc. Scoli films are low-resolution, you will likely not be seeing trabecular detail, nuanced findings, etc. - if concerned about something one would have to request ‘dedicated spine films for better detail’. Describe suspected findings requiring better images as ‘possible’ or ‘cannot r/o, inadequately seen…” and if possible check older studies for more information.

ADDITIONAL VIEWS: Pre-operative lateral bending views (PA bending R, PA bending L) assess flexibility and potential for surgical improvement (a very rigid curve usually is fused as is, a more flexible curve can be reduced and then fused). Re-measure curves measured on neutral from SAME LANDMARKS and report as “37 degree right scoliosis from superior end plate T5 through inferior endplate L2 in neutral, which on bending to the R measures xy degrees and on bending to the L yz degrees using same landmarks”. Lateral bending views are far less common but one would again measure significant changes in lordosis/kyphosis, and if observed, vertebral body subluxations.

POSITION: Frontals presented as taken (and as pt. examined, from behind), PA standing (so R is yr. R; confirm with markers).

DOSE: true coned spine exams would be taken AP, so area of interest is closest to cassette and most geometrically accurate. By taking scoli films PA, one sacrifices some detail to magnification BUT the innate shielding of spine, ribs, pelvis drops dose to radiosensitive pediatric tissues— thyroid, gonads, female prepubertal breast—up to 95%. Further dose reduction is achieved by using screens, filters, modified beam dosage compared to true spinal exams, precise collimation, and shields. Most scoliosis patients are children; most idiopathic scoliosis patients are peripubertal females; and one anticipates scoliosis studies need to be repeated as often as every 6 months for those at risk of progression (scoliosis can progress a degree a month during growth spurts), making dose reduction essential. Scoliosis films are inherently low-resolution and sacrifice detail for the ‘big picture”: angle, curvature, and change from last exam; concerned parents can be reassured that the doses are more akin to chest radiography, the lowest-dose exam, than true spine radiography.

COUNT RIBS: ’12 pairs thoracic ribs’ (amounts to counting vertebral bodies; alerts you to anomalies, confirms numbering, good medical-legal habit  

  • CERVICAL RIBS have transverse processes that point DOWN, T1 transverse processes point UP.
  • COUNT LUMBAR VERT BODIES: ‘ 5 lumbar vertebral bodies’ (again confirms numbering, alerts you to anomalies, reminds you to look at L5, S1 for segmentation anomalies, pseudarthoses—surprisingly common, OFTEN missed, occasionally leads to litigation.)


  • Fusion/segmentation anomalies: try counting pedicles; may end up with something like ’11 R and 12 L thoracic ribs with apparent hemivertebra at T8’. If too confusing or too blurry—scoli films have very poor detail—just say ‘with apparent fusion anomalies at approximately T7-8-9’... or “ mid thoracic spine”, or 'Multiple vertebral/rib anomalies preclude numbering".

COBB MEASUREMENTS: While as of 6/2017 we are not specifying measurements in Ortho, in Peds or in the real world measurements will be mandated. Use end plates as landmarks for measuring tool lines; if can’t see endplates put superior line across cephalad pedicles, inferior line across caudal margin pedicles.

NOTE: as of June 2017, rather than specifically measure angles, describe ('right thoracic and left lumar..."; "Slight decrease of right thoracic curve with traction compared to neutral view today...") and then say "See clinician's measurements".  There have been issues with insurance etc when our measurements are slightly different from the clinician of record-- and since SD of measurement is +/- 5 degrees, need to avoid 'contradicitons' in written record.

Use ‘top of top, bottom of bottom’ ie superior end plate most off-horizontal at cephalad portion curve, inferior endplate of most caudal level in curve.

DOESN’T MATTER that much usually if you can’t decide if you should use T7 or 8; DOES MATTER that you use same levels used in any older/comparison measurements.

State direction and extent of curves: ‘There is a 37 degree levo(left)/dextro (right) scoliosis extending from the superior endplate of T3 to the inferior endplate of T10”

DIRECTION: A right or dextroscoliosis ‘points’ to, or is convex to, the R; a left or levoscoliosis, convex to or ‘points’ to L.

OLD STUDIES: When available, state “….and unchanged from last exam, using same landmarks, 9/25/07’, or “ increasing from last exam, same landmarks….’ True increase or decrease, if one used same landmarks, means more than 5 degree change; if soft call (ie <5 degrees), say ‘possibly increasing…’ or ‘apparently increasing’. Go back to earlier studies—a slowly progressing scoliosis may be overlooked if only about 5 degree increase on each of several visits; yet by jumping back to earlier studies you may be able to document “increasing less than 5 degrees since last exam 9 months ago, but increasing 14 degrees since earlier exam 33 months ago”. If positional or technical differences between exams seem to limit comparison, say so.

HARDWARE: Describe hardware (look at older reports, need not be excruciatingly detailed); most are posterior rod and pedicle fusions, anterior disk space implants (bone graft, metal cages, implant material), anterior screws; the pediatric rods that look like portable radio antennae are telescoping rods to allow for growth.

Assess for CHANGE: most significant would be at most proximal/distal extents fusion, where screws create biggest ‘stress risers’ and are most likely to cut osteopenic bone once weight-bearing resumes.

Pedicle screws should go (PA) through pedicle to terminate anteriorly in vert. body with tip and threads within vert. body bone and not superimpose endplates or discs. Screw threads VERY sharp, most proximal or distal screw thread position superimposing endplate or disc space should be noted (may be cutting through soft bone). Also try to ascertain if this position has changed since old exam (“Anterior superior screw threads at L1 superimpose superior disc but unchanged from 5/12/12 and 9/23/11”; “Anterior superior screw thread now protruding 2 mm into disc space, minimally progressing since 5/12/12”, etc.)

HARDWARE FAILURE: both the vertical rods and transverse screw threads can fracture, creating a fine lucency that may be visible only on one of many views. Was it there previously? Known/unknown to clinician? Small washers (the ring often placed between screw head and rod) or nuts (the larger, often polygonal donut on proximal screw thread) may displace into tissue either during or after surgery. Confirm that all post-op foreign matter (drains, skin staples, sponge markers, etc) are eventually removed/gone. SCOLI TECH very limited for early metal-bone change.

HARDWARE INTERFACE: metal-bone interface should be <1mm and unchanged from priors. Wider than 1 mm may imply motion/loosening, even infection. Slight interface between pedicle screws and trabeculae common once weight bearing resumes; bone is both more compressible and more reactive than hardware but such interfaces should parallel surface of metal, not be bulbous (implying screw tip may be ‘windshield wiping’) or irregular (possible focal infection).

Corners, Dysplasias, Hiatal hernias, ...: as in any study, the non-scoli findings must also be mentioned:  "Again seen is hiatal hernia and 3-4 small 4 mm calcifications clustered in RUQ, presumed gallstones" "Characteristic changes of achondroplasia/osteogenesis imperfecta..."  "Inconpletely seen hip/shoulder arthroplasty..."

LAST LINE OF REPORT"Scoliosis technique which reduces detail to reduce dose". (And if specific concern-- hardware interface, disc space change-- "Coned and centered images of any area of interest would give better detail if clinically indicated")

July 2017 - Donna Magid, M.D., M.Ed.