73 patients had complete records available for image reading data analysis of PA images (table 1).
No significant differences in biometric parameters were found between both study groups populations:
Table 1
Group
|
n
|
Females
|
Age
|
Weight
|
Height
|
Cobb
|
Risser
|
Surgery.
|
A
|
40
|
30
|
13.7
|
49.4 kg
|
157 cm
|
28º
|
2.7
|
4
|
B
|
33
|
28
|
13.9
|
50.7
|
158
|
24º
|
2.6
|
2
|
Lateral views,
were obtained in 25 of 40 patients of group A,
and in 26 of 33 patients of group B. Lateral exposure data were analysed but were not used in the Cobb's angle or Risser's grade assessments.
No studies were repeated or rejected for quality problems in both groups.
Average scoliosis was 27º (SD 16º,
range 4º-79º).
No differences in exposed area between both techniques were found as exposed field length Fig. 3 and exposed square meter Fig. 4 that may justify dose differences other than technique.
The radiographic parameters and the recorded DAP were significantly lower in the optimized technique (group B DAP 1.68 dGy·cm2,
SD 1.08 dGy·cm2) than for standard technique (group A DAP 3.5 dGy·cm2,
SD 2.17 dGy·cm2),
and 168,25 μGy and 54,21 μGy for techniques A and B respectively, Fig. 5, Fig. 6, and Fig. 7.
Statically significant differences were present for all exposure related parameters favouring the optimized technique as a 50-70% exposure reduction,
both in PA projection and in lateral projection, Fig. 8,
and Fig. 9. Lateral projections convey the same exposure derived parameters than PA projections.
The lateral projection in low-dose group displays,
just on the box and whisker plot,
a not significant exposure increment, Fig. 10.
Small,
but not significant differences were found among the different rounds of lecture and among the readers, Fig. 11.
Cobb's angle and Risser grading assessment were feasible in all patients with both techniques,
with overall better agreement for angle (R 0.83) than for Risser (R 0.73),
while group B had slightly better agreement (R 0.77) for Risser grading (table 2).
Intraobserver (R 0.86) and interobserver (R 0.83) agreement were good or excellent in both techniques,
either in senior or junior readers.
Table 2
|
Interobserver
|
Intraobserver
|
Senior
Observer
|
Junior
Observer
|
Cobb Angle
|
0.83
|
0.83
|
0.86
|
0.83
|
Group A
|
0.83
|
0.80-0.90
|
|
0.80
|
Group B
|
0.82
|
0.82-0.87
|
|
0.87
|
Risser Grading
|
0.69
|
0.65-0.83
|
0.62
|
0.73
|
Group A
|
0.66
|
0.71-0.83
|
|
|
Group B
|
0.72
|
0.72-0.82
|
|
|
The magnitudes of agreement for Universal Agreement R are the same as for Kappa: <0 indicate that there is no agreement; 0-0.20 as poor; 0,21-0,40 as regular; 0.41-0.60 as moderate; 0.61-0.80 as good; and 0.81-1 as very good.