If they stick to Roman numerals, the development of maths in the West is severely limited. (Interestingly, I can not think of any surviving Classical Latin text about mathematics - as opposed to both Classical Greek and Medieval Latin - not to mention Arabic or Indian stuff. Ready to be corrected on the point though). It's hard to imagine the Scientific Revolution happening in Europe here, for example.
Honestly, I do not see it lasting forever. Even if they do not adopt the very effective Indian/Arab system (and I'd ask why; it's so handy, and Medieval cultural exchange with Islamicate cultures so sustained, that they'll come across it and see it's useful), they'd likely switch to something akin the letters-for-numbers systems used in Greek. Unwieldy, but still considerably better than Roman numerals (it spread to Cyrillic, and I think also Runic, for instance). Perhaps, if a letter system is adopted and accepted before Gerbert promoted introduction of Arab numerals, the Arab system may be seen as both alien enough and not necessary enough not to take hold (I am under the impression that this is what happened in the Eastern Roman Empire). Then, admitting that a *Columbian Exchange happens roughly on schedule and the greater currents of Western intellectual history are only marginally affected by this (huge assumptions, I know) at some point someone would note that the Mayan notation system for numbers is too practical not to adopt. Or else, maths stagnates and European learning at large with it. Endgame, probably, no global Euro dominance unless they are smart enough to pick a better system (more likely option: there's at least a good one available right next door) or invent one from scratch.