Level of evidence and citation index in current neurosurgical publications.

Systematic clinical reviews or meta-analyses offer scientifically valid sources of clinical information. They provide information in a concise form and can contribute to clinical quality management. Such studies, however, are only able to reflect the quality of the articles reviewed. The aim of this study was to evaluate the quality of the neurosurgical literature according to evidence-based medicine (EBM) standards. We reviewed all articles published in 1999 in three major neurosurgical journals. These articles were subdivided according to the level of evidence (LOE) scale (from 0 to V), article type, and citation index. Nine hundred eighty-two articles were published in these journals in 1999. Of these, 346 (35%) were clinical studies, 287 (29%) case reports, 153 (16%) experimental studies, 122 (13%) technical reports, and 74 ( 8%) other types. Subdivision according to LOE was: Ia 0.3%, Ib 2.5%, IIa 0.2%, IIb 4.3%, IIc 9.5%, IIIa 0.1%, IIIb 3.9%, IV 22.4%, and V 1.6%. Fifty-five percent of all published studies were case reports, experimental studies, technical reports, or others and thus could not be subdivided according to the EBM standards. The number of articles published with high LOE seems to be rather low in 1999. If these data reflect overall publication practice, it seems unclear whether enough articles with high LOE are published to propose scientifically sound clinical treatment suggestions according to EBM standards.