Mass spectrometry depends on separation of the analyte based on mass to charge (m/z). ICP-MS is used to detect metals, but under the plasma conditions many of these metals could be ionised multiple times, giving them a variety of (m/z) ratios making a difficult to identify peak for any particular metal. Additionally some metals would have m/z ratios that overlap with others. Then it is conceivable that depending on your solution the quantitation of some elements is impracticable. For example if you want to quanitate aluminium (mass ~27) then a solution high in manganese (mass ~54) would make this impracticable, if manganese were twice ionised to give the same mass to charge ratio as a single ionised aluminium ion. How does the ICP-MS ensure atoms are ionised only once, or that only singly ionised species are detected?
This is further complicated if you consider that the elements rarely exist in solution at oxidation states of 0, but rather aqueous ions with varying oxidation states.