Objectives: In this retrospective study, we investigated how spasticity developed in patients diagnosed with a prolonged DOC over an almost two-year observation period (21 months), and how it related to the patients’ age, gender, time since injury, etiology, level of consciousness, and anti-spastic medications. Methods: In total, 19 patients with a severe brain injury and prolonged DOC admitted to a long-term care facility were included in this study (14 male, age: 45.8 ± 15.3 years, 10 traumatic brain injury, 1.01 ± 0.99 years after brain injury, 11 minimally conscious state vs. 8 vegetative state). Each patient was assessed at admission and then quarterly, totaling eight assessments over 21 months. Spasticity was measured with the Modified Ashworth Scale (MAS) for both upper and lower limbs. The Western Neuro Sensory Stimulation Profile (WNSSP) was administered to assess the level of consciousness. Any other medical and demographic information of interest was obtained through medical records. Linear mixed models were used to assess each variable’s impact on the change of spasticity over time. Results: Significant differences were observed in the evolution of spasticity in patients based on their etiology for the upper limbs [F (7, 107.29) = 2.226; p = 0.038], and on their level of consciousness for the lower limbs [F (7, 107.07) = 3.196; p = 0.004]. Conclusion: Our preliminary results suggest that spasticity evolves differently according to the type of brain lesion and the level of consciousness. Spasticity in DOCs might therefore be mediated by different mechanisms and might have to be treated differently among patients. Future longitudinal studies should be performed prospectively in a bigger cohort and with data collection beginning earlier after brain injury to confirm our results and better understand the evolution of spasticity in this population.