I used to read an old book about openings written by Ludek Pachman. It was probably published in 1980 though I am not sure of that because I read a translation... anyway, I digress :)
The interesting part of his analysis started after those standard moves
[FEN "rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1"]
1. e4 e5 2. Nf3 Nc6 3. Bb5 Nf6 4. O-O Nxe4 5. d4
At this point, he mentions that 5... exd4 is bad and mentions two moves. First, 5... Be7 that he recommends (I think he puts a (!) evaluation to it):
[FEN "r1bqkb1r/pppp1ppp/2n5/1B2p3/3Pn3/5N2/PPP2PPP/RNBQ1RK1 b kq - 0 5"]
5... Be7
Funnily, this move is now only rarely seen.
And, to comment the famous line that Kramnik revived in 2000:
[FEN "r1bqkb1r/pppp1ppp/2n5/1B2p3/3Pn3/5N2/PPP2PPP/RNBQ1RK1 b kq - 0 5"]
5... Nd6 6. Bxc6 dxc6 7. dxe5 Nf5 8. Qxd8+ Kxd8
He mentions that this line cannot really be recommended as Black has some troubles finishing his development: the rooks are not connected, and also White's Kingside majority seems more efficient than Black's queenside majority. I guess this statement reflects the general feeling about this opening at that time: hard to play in practice.
I don't have the book next to me so I cannot really check my memories, but it should be close.
So, kudos to Kramnik for having gone beyond this impression and proved that the whole line was playable.