Microglia are known as the most motile cells in the central nervous system (CNS). It was shown in vivo that they permanently scan their direct microenvironment and react to pathological conditions within minutes. Many studies of brain pathologies use fixed brain tissue to investigate cellular changes. Unfortunately, due to technical reasons, the time span between the induction of the fixation procedure (start of the perfusion) and the finally-fixed tissue lasts several minutes, giving time to microglia to start reacting to the ischemic conditions due to perfusion start. Here, we investigated the microglial changes generated by the fixation itself in TgH(CX3CR1-EGFP) mice with fluorescent labelled microglia using confocal laser scanning microscopy (CLSM) of fixed brain tissue as well as two-photon laser scanning microscopy (2P-LSM) during the perfusion of a living animal. We revealed the impact of fixation and buffer parameters on cell morphology. The largest morphological differences compared to physiological in vivo branch arborization were observed when the directly dissected brain was immersed in paraformaldehyde fixation solution overnight, without prior fixative perfusion of the animal. But even perfusion with a fixative, followed by post-fixation leads to small changes in microglial process length and number and could not be prevented when compared to physiological in vivo microglia morphology acquired using in vivo 2P-LSM. Interestingly, perfusion with different buffers either oxygenated artificial cerebrospinal fluid or phosphate buffered saline prior to perfusion-fixation showed minor microglia changes in arborization and/or number of processes. Fixation methods influence microglia morphology. Therefore, to define microglia activation states immunohistochemical stainings or genetic labelling of the cells have to be included in addition to morphological analysis.