Abstract Surgical treatment of deep intracerebral hematomas remains a difficult problem to solve. One of its greatest difficulties is the lack of a clear differentiation into subtypes, that could benefit from the use of specific surgical approaches. There is also no agreement yet on which surgical techniques might be the most appropriate. This paper summarizes the current scientific knowledge on both limitations, and proposes a therapeutic algorithm based on the author’s experience, and on the possibility of appropriately treating associated problems.