Integrity constraints play a major role when the logical consistency of a data set has to be evaluated during quality assurance. The paper reviews established approaches of integrity constraint classification and proposes a profound categorisation which distinguishes semantic integrity constraints according to the semantic domains of the involved properties. The categorisation provides the preconditions for formalisation and the identification of conflicts and redundancies in sets of integrity constraints. Therewith the management and usability of integrity constraints will be improved which will also result an improvement of spatial data quality.