Irregular repetition slotted aloha (IRSA) is a massive random access protocol in which users transmit several replicas of their packet over a frame to a base station. Existing studies have analyzed IRSA in the single-cell (SC) setup, which does not extend to the more practically relevant multi-cell (MC) setup due to the inter-cell interference. In this work, we analyze MC IRSA, accounting for pilot contamination and multiuser interference. Via numerical simulations, we illustrate that, in practical settings, MC IRSA can have a drastic loss of throughput, up to $70\%$, compared to SC IRSA. Further, MC IRSA requires a significantly higher training length (about 4-5x compared to SC IRSA), in order to support the same user density and achieve the same throughput. We also provide insights into the impact of the pilot length, number of antennas, and signal to noise ratio on the performance of MC IRSA.