Ten patients with severe spasticity were evaluated according to a standardized protocol in order to be treated by intraspinal baclofen. Entry criteria in the protocol were the following: 1) Stable central nervous system lesion, 2) Severe spasticity and/or flexo-extensor spasms not controllable by oral treatment, 3) Normal CSF circulation and 4) Informed consent. All patients received a test dose of twenty-five micrograms of baclofen injected intrathecally. At intervals of at least one day, doses were increased in 10-25 microgram steps until total abolition of spontaneous spasms was achieved in complete spinal cord lesions. In patients with residual motor function, doses were titrated until the optimal dose was found that reduced spasms and enabled performance of maximum daily life activities according to the patient's neurological level. In nine patients a multidose reservoir was implanted to deliver intrathecal baclofen. Effective dosage was 60 +/- 31 micrograms in the entire group. Ashworth score was reduced from 4.6 +/- 0.7 to 1.2 +/- 0.4 (mean +/- SD) (p less than 0.0001) and spasms from 3.2 +/- 0.8 to 0.2 +/- 0.4 (p less than 0.0001). Follow-up of the nine patients in whom a reservoir was implanted has been 18 +/- 9 months. Initial dosage requirements and tolerance were significantly different in complete (Frankel's A grade) or incomplete lesions (Frankel's B, C and D grades). Complete spinal cord lesions required a greater initial dose (156 +/- 43) than incomplete lesions (44 +/- 24), these differences being statistically significant (Student's t-test, p less than 0.05). Tolerance was observed only in patients with complete motor and complete sensory lesions. In incomplete lesions, dose increase was insignificant.
