Kasumov R.D., Klimenko N.B., Dzhabbarova L.B., Nikulina T.A., Alekseev V.V.
(Polenov Research Neurosurgical Institute, Saint Petersburg, Russia)
Prognostication of early outcomes of treatment of severe craniocerebral trauma (SCCT) is one of the most prospective trends of neurotraumatology development. It is of great medical, socio-economic and juridical importance. Prognosis makes it possible to choose a necessary number of diagnostic and therapeutic measures in every specific case. Use of modern information technologies allowed to solve prognostication problems with the help of methods of mathematical modeling.
The study was based on data, obtained in 300 casualties with SCCT, who were operated for crush foci and intracranial hematomas. A clinical picture and dynamics of neurologic disorders, duration of their diagnosis, postoperative intracranial and somatic complications were analyzed in detail; their prognostic significance was estimated. Methods of discriminant analysis, logistic regression and analysis of functions of "survival period" were used for creating mathematical models of prognosis. A minimum list of symptoms and syndromes of cerebral lesions, sufficient for reliable staged prognostication, was compiled. Developed prognostic models possess high informative capacity (from 86.00% up to 93.45%).
Models are presented in the form of a table. Thus, neurologists and neurosurgeons can use them in their practice and calculate prognosis without any difficulty.