Well, that seems to me to be totally different from what the current MEDDEV guidance expects.
I think that MEDDEV documents' role is not to replace the regulation or adjust it / keep it up to date. Their role is to clarify the regulation, and interpret it in line with their original authors' intents. In that sense I'm not sure rev. 4 is not a mistake. How does it accommodate the clause that started this thread? Maybe it never intended to, because it deals with Clinical Evaluation, and that clause completely bypasses the clinical evaluation path.
Keeping the regulations up to progress should be done through a proper legislative process, as has been done in issuing the MDR (whether a good job was done or not is a different question). The updated regulation - the MDR - will take effect in less than 3 years, and until that time the older regulation - the MDD, as amended in 2007 - should be the prevailing standard (upgrade in anticipation of the change may be recommended, but not yet mandated). If anyone thought that the MDD doesn't provide sufficient assurance, they should have pushed for another MDD amendment while the MDR was in the making (a very long time!). I know that it wasn't done in the name of efficiency and preserving resources, but it shouldn't have been used as a justification for hijacking regulatory power (as 2.7/1 in fact did, whether its authors meant it or not).
If there's an equivalent (which was made much more difficult in the last MEDDEV) and clinical data related to the equivalent which is deemed good enough to show compliance, you can use it, but if not, the only option is to create clinical data thru a clinical investigation.