Does whole-spine lateral radiograph with clavicle positioning reflect the correct cervical sagittal alignment?

To evaluate the differences of cervical alignment between standing cervical lateral radiograph and whole-spine lateral radiograph with clavicle position. We prospectively evaluated 101 asymptomatic adults from whom standing cervical lateral radiograph with hands on both side followed by whole-spine lateral radiographs with clavicle position were obtained from April 2012 to December 2013. On two radiographs, cervical sagittal alignment from C2 to C7 was analyzed by Gore angle (GA) and Cobb angle (CA); head position was evaluated using the translation distance (TD, distance of the anterior tubercle of C1 compared with the vertical line through the posterior-inferior body of C7) and McGregor angle (MA, angle between the McGregor and horizontal lines). T1-slope was also evaluated. Cervical alignment on the cervical radiograph (GA -13.59A degrees [-15.58 to -11.60], CA -9.76A degrees [-11.65 to -7.86]) was significantly more lordotic than that on whole-spine radiographs (GA -6.28A degrees [-8.65 to -3.91] and -4.14A degrees [-6.40 to -1.89]). TD and MA on cervical radiographs (TD 34.98 mm [33.22-36.75]; MA 7.20A degrees [6.35-8.35]) were meaningfully higher than those on whole-spine radiographs (TD 31.31 mm [29.47-33.16]; MA 6.32A degrees [5.25-7.39]), but the MA values were not significant (p = 0.064). T1-slope was significantly lower in whole-spine radiographs (20.11A degrees [18.88-21.35]) than in cervical radiographs (24.37A degrees [23.14-25.6]). Values are expressed as mean (95 % confidence interval). Clavicle position during whole-spine radiograph caused a substantial decrease in the T1-slope; head position posteriorly translated followed by the cervical sagittal alignment became more hypo-lordotic, with slight downward gazing in comparison with the cervical radiograph.