I've noticed that when lines go hypermetrical with an extra syllable, it usually (though not always) is when two vowels in the middle can be sortof-almost elided together -- but only in the middle, as opposed to a wo at the end. But then, many of these pivoted jo constructions were conventional (some to the point of ossified), and so I'm guessing readily understandable without the case-marker.

I don't think the azusayumi is engo with anything else in the poem, though. I'm not exactly expert at all the associations, though.

---L.