Objective: This study was designated to quantitatively evaluate the effect of levodopa on spinal posture in patients with PD using a computer-assisted handheld SpinalMouse device. Methods: Prospective case-study involving 48 patients with definite PD. All patients were recruited between September 2011 and September 2013 and included 22 dopa-naïve, evaluated before and 3 months after initiation of treatment, and 26 patients with response fluctuations studied during the "off" and "on" states. The SpinalMouse instrument, a computer-assisted mechanical hand-held device, designed to noninvasively assess the curvature of the spine was guided along the midline of the vertebral column in upright, full flexion, and full extension positions to objectively assess spinal posture. Results: In the dopa-naïve patients, spinal incline in the upright position was 12.4±1.2° before and 7.6±1.3° after treatment; p=0.002. Corresponding area-under-the-curve (AUC) values were 131.7±8.0cm2 and 87.1±7.3cm2; p<0.0001. In the response fluctuations patients, spinal incline was 13.3±1.3° in the "off" and 9.3±1.2° in the "on" period; p=0.015. Corresponding AUC values were 144.6±9.2cm2 and 103.1±8.2cm2; p<0.0001. Conclusions: This is the first study that objectively measured and quantified abnormalities of spinal posture in patients with PD. Findings suggest that levodopa does have a beneficial effect on anterior flexion of the thoracolumbar spine, and thus indicate that the disorder of stooped posture in PD is mediated, at least in part, by dopamine deficiency.