The validity of computed tomographic scan findings for the localization of cerebral functions. The relationship between computed tomography and hemiparesis.

We studied recovery from hemiparesis in 52 patients who participated in the Hennepin County (Minnesota) Recovery From Aphasia Study. Our intent was to determine the anatomic correlates of recovery from hemiparesis and then to contrast computed tomography (CT)-hemiparesis relationships with CT-aphasia relationships in this same population. Hemiparesis was assessed at one month after onset and again at six months after onset. Computed tomographic scans were obtained five months after onset and analyzed quantitatively for lesion location. The presence or absence of arm paresis was strongly predicted by CT scan findings in the supratentorial corticospinal pathway. However, three of 25 nonparetic patients had lesions in the corticospinal pathway, and one of 22 severely paretic patients lacked an appropriate lesion. This study demonstrated that a small but inherent lack of specificity and sensitivity existed in the predictive power of CT localization of lesions potentially affecting the motor system, suggesting that some caution is necessary in interpreting the lesion-deficit relationships in less well-localized functions, eg, language. Besides considering the lack of precision in localizing with CT, the existence of individual differences both in anatomy and localization of function must be taken into account.