Upper motor neuron signs in acute stroke: Prevalence, interobserver reliability, and timing of initial examination.

There is little documentation in the neurologic literature of the prevalence or reliability of upper motor neuron (UMN) signs in acute stroke. This knowledge is important for those conducting neurologic surveys and clinical trials among stroke patients because it may aid in the design of data collection instruments. One-hundred stroke patients admitted to the Columbia-Presbyterian Medical Center Neurovascular Unit were independently examined by two neurologists who were blinded to the patient's clinical history and neuroimaging studies. Kappa values ranges from -0.09 to 0.57, and concordance rates ranged from 0.55 to 1.0. The prevalence of 13 UMN signs varied considerably (3-94%). More prevalent UMN signs such as pronator drift, diminished fine coordination, hyperreflexia, Babinski and Chaddock signs were also reasonably reliable. There were differences between the early-acute and late-acute motor stroke groups; however, these were not significant. Prevalence, reliability, and, possibly, the time of initial examination were inter-related. The clinical utility of each UMN sign is a function of three inter-related factors: prevalence, reliability, and time of examination. Hence, the time of initial neurologic assessment may have some impact on the prevalence and reliability of UMN signs.