Search

In the case of a patient with sudden-onset focal neurological deficit, clinicians must answer three fundamental questions: is it a stroke? is it ischemia or hemorrhage? and what kind of ischemic stroke is it? Clinical information (i.e., history and examination) is available in any situation, and its role in answering these questions is extremely important, even though certainty can only be achieved from instrumental diagnostic tools. In fact, when diagnosis is based on properly designed clinical criteria, the percentage of mistakes is quite low. Clinical methods are still the best way to orient topographic and etiologic diagnosis, as well as estimate prognosis. In addition, time might be saved if randomization in clinical trials were performed using clinical methods before initiating complex investigations.

The great technological progress in the Neurosciences area, which has appeared in the last few years, can cause, beside obvious scientific and practical advantages, an important risk: in fact, efficiency might be preferred at the expense of efficacy. In this paper we try and outline a sort of "efficacy route in Neurology", based on well known general principles of Clinical Epidemiology, both for diagnosis and treatment. We stress the fact that clinical evaluation is still an essential instrument in the diagnostic process, and that the result of any therapeutic procedure must be evaluated by means of hard end points, strictly related to the real problems of the neurological patient.

The "N of 1" study is a randomized, placebo-controlled, cross over study of a single drug in a single patient, which can be particularly useful when physicians are faced with chronic, stable diseases and/or complex treatments. We applied this model in 2 patients with neurological disorders. In the first case (a movement disorder) the N of 1 trial allowed us (and the patient!) to conclude that only one drug was really needed to control the disease. In the second case (a painful neuropathy) we could avoid using a potentially useful, but toxic drug, which turned out to be no more effective than placebo. We suggest that the "N of 1" trial model should be frequently applied in general as in well as in neurological practice.