Résumé
Very massive stars (VMS) are defined as stars with an initial mass in excess of 100 Msun. Because of their short lifetime and the shape of the stellar mass function, they are rare objects. Only about twenty of them are known in the Galaxy and the Large Magellanic Cloud. However VMS are important in several ways. They efficiently spread nucleosynthesis products through their boosted stellar winds, they are predicted to explode as pair-instability supernovae or to form heavy black-holes from direct collapse, and they outshine all other types of stars in the ultraviolet light, thus dominating the integrated light of starbursts. Their presence is indirectly suspected across all redshifts, all the way to cosmic dawn where they may have played a key role in the formation of the first galaxies. Their search and identification is currently hampered by instrumental limitation, especially spatial resolution. An integral field spectrograph working at the diffraction limit of HWO (5mas) and with a spectral resolution of about 2000 would revolutionize the understanding of VMS. We make the case for such an instrument in this contribution.