Just some strayed thoughts:
Be cautious with reviewers who give overly dogmatic "rules of thumb" about writing. A few of those rules are reasonable and legitimate, the majority are either learned practices and unsupported or outdated conventions.
To me (hint: opinion), starting a paragraph with a question is cliche. 1) I can understand the use of it in other literature, but it's overly stylized in scientific writing. 2) People usually start reading the abstract because they are attracted by your title. There isn't really a strong need to use another bait. 3) The first part of the abstract is usually used to set the stage by presenting some background or key information to bring the readers up to speed. Starting with a question can throw some experienced readers off. 4) Most of the time, the follow up "answer" to that question tend to use very similar words, ending up wasting word count. 5) I work in biomedical field and the general style is to state the research question at the end of the background. Starting the same paragraph with a similar question would be, like the point above, wasteful.
For these reasons, I'd suggest try two versions with different starting style and all else equal, do another poll with some friends and colleagues and see how that goes.