3D photography could improve assessment of chest wall deformity in scoliosis surgery


Plastic surgery’s 3D imaging techniques could simplify and improve the assessment of deformity and the measurement of post-operative improvements in scoliosis surgery, according to a talk by Bronek Boszczyk, Nottingham University Hospitals, Nottingham, UK at the Brussels International Spine Symposium (BISS; 20-21 November, Brussels, Belgium).

The most common methods of assessment for chest wall deformity are complicated, broadly cumbersome and produce varied results, according to Boszczyk, who began the presentation with a review of current methods for measuring deformity. Some, like the scoliometer, are relatively simple to use, but subject to high rates of user error. More sophisticated methods, like the posterior trunk symmetry index, fail to reveal much indication about the depth of deformities. Methods like the Suzuki Hump Sum and rasterstereography may offer this information, but are too complex to be used routinely in everyday practice.

Boszczyk suggested that borrowing 3D imaging techniques practised by plastic surgeons could more accurately measure deformity, surgical outcomes and the effects that scoliosis and its treatment can have on patients. This technology is simple and can effectively simulate the patient’s deformity, allowing surgeons to accurately determine asymmetry in terms of volume discrepancy. Boszczyk told Spinal News International, “It will help determine how effective our derotation techniques and costoplasty (thoracoplasty) techniques are.” Comprising a cage mounted with multiple cameras and either Vultus or Canfield software—which is commonly used to create facial reconstructions—surgeons can combine simultaneous photographs from all angles of a patient’s body to produce very realistic torso images.

Boszczyk demonstrated the benefits of the technology—which can be used for all patients but those with severe rib humps—with the example of an 18-year-old female patient with adolescent idiopathic scoliosis (AIS) and a modest rib hump. Standard costoplasty and deformity correction had corrected the scoliosis from 57 to 34 degrees. By super-imposing these images on to pre- and post-operative 3D images of the patient’s torso, surgeons can observe the changes in soft tissue as well as the spine itself. This exposes the way that the patient’s waist actually changes its contour. Boszczyk emphasised that the 3D images could also be combined with EOS and CT images to more comprehensively simulate corrections.

The merged images can be used to evaluate the changes in the chest wall shape after surgery, by superimposing the pre- and postoperative torso images. Alternatively, symmetry can be determined by mirroring and superimposing each side of the back, divided along the midpoint. The software can calculate the volume difference between each side, and by subtracting the pre-operative result from the postoperative result, we can objectively measure asymmetry as an outcome of surgery. Boszczyk demonstrated this method with images of the 18-year-old AIS patient. An initial volume mismatch of 1070cc on the curve concavity was reduced to 411.03cc; a difference of 600cc after surgery. An improvement of 362cc on the curve convexity suggests that the costoplasty was less effective.

As well as measuring treatment outcomes, Boszczyk told Spinal News International that the technology could also be useful for assessing treatment options, “It will be useful once we can merge imaging and 3D photography, as is already the case for dental and maxillofacial surgery, for example.”

Boszczyk confirmed that inter-observer variability, for example still needed to be calculated, but that the comparison of the mirrored volume could potentially give a standardisable measure for symmetry. The use of software to calculate these values should make the system easy to use, and the availability of an EOS system would remove one of the biggest limits to this simplicity—the ability to pick the perfect midline.

A further application of this technology comes in the form of contour measurement. A band can be selected from the torso reconstruction, and used in a similar way to the torso images themselves to calculate the difference in surface area at any part of the torso. By boxing thin bands of torso into quadrants and mirroring them, the inside or outside differences in surface area, and the differences in distances between the midlines, can be calculated. This could allow surgeons to measure outcomes in terms of body contours.

Work still needs to be done to determine which of these measurements is most reliable, but Boszczyk asserted that this technology improves upon its predecessors in a number of ways. Technology has advanced enough to make these measurements simple and accurate, which could offer surgeons a standardisable, automated method to calculate body contours and asymmetry. Boszczyk, who has been using this technology for about three years, told Spinal News International, “I believe that this technology will have a substantial impact on our ability to predict and measure our effectiveness in surgically correcting scoliosis. In the future it may also be a good counselling tool for patients, allowing surgeons to predict what their appearance might be after corrective surgery.”