This seems similar to asking what happens if you loose a number of parts whose loss does not produce immediately discernible detriment, or perhaps parts whose function may be approximated by something else in the event of a failure.
Of course, it is a perfectly valid question to ask what the effects of losing a part are.
Hence mention of a large population who lost those parts. The actual detriment may not be identifiable in a random individual, but it will show up at scale, as the noise of other confounders will average out.
Of course, it is a perfectly valid question to ask what the effects of losing a part are.