Your criticism of a randomized, double-blind, placebo-controlled study is essentially "but it doesn't fit my own (unblinded, uncontrolled) anecdote". You offer

Your criticism of a randomized, double-blind, placebo-controlled study is essentially "but it doesn't fit my own (unblinded, uncontrolled) anecdote". You offer an edge case that was not tested by the study. Your criticism is constructive/relevant, but there is always some area of our understanding that the experiment doesn't reach... basically saying "here is a different but similar hypothesis that was not tested by the trial". Ok; fair enough. But in the eyes of most readers, this data set generated by an insanely tight scientific design suggests that the optimal allocation might lie in a different direction of research for the individuals suffering these symptoms. Interestingly: you can perform your own blinded experiment. If I were in your position, I would want to know whether my wife and I were planning our lives around a superstition. The way to answer that: an n-of-1 blinded clinical trial. It's a fair amount of work, but not as much as gluten avoidance, and would answer the question of your own circumstances better than any generalized scientific result from the literature.