Standard intracranial in vivo animal models of delayed cerebral vasospasm.

Object: Animal models provide a basis for clarifying the complex pathogenesis of delayed cerebral vasospasm (DCVS) and for screening of potential therapeutic approaches. Arbitrary use of experimental parameters in current models can lead to results of uncertain relevance. The aim of this work was to identify and analyze the most consistent and feasible models and their parameters for each animal.

Methods: An online search of the MEDLINE PubMed and EMBASE medical databases (1969 to week 21 of 2007) was performed using the key words "canine", "mice", "rabbit", "pig", "rat", "cat", and "primate" in combination with "subarachnoid hemorrhage", "model", and "vasospasm". Cross references of each model were checked. Analysis of identified publications was considered in accordance with predetermined eligibility criteria.

Results: 1254 abstracts were reviewed and 516 studies were included in the analysis. Then, 66 models in 7 animals were identified. Most often used blood amounts (ml) lead to degree (% vessel narrowing) and peak onset (day) of DCVS within animal models as follows: mice endovascular puncture (various, day 3, 20-62%); rat single injection (0.3 ml, day 2, 19-29%); rat double injection (2 x 0.3 ml, day 7, 28-47%); rabbit single injection (3 ml, day 3, 19-55%); rabbit double injection (not established, day 5, not established); dog double injection (2 x 4-5 ml, day 7, 45-66%); primate clot placement (5 ml, day 7, 32-52%).

Conclusions: Among the great number of experimental SAH methods and associated parameters only a fistful reliable and consistent models can be identified and recommended. Implementation of more standardized experimental techniques could increase the relevance of future experimental studies.