It is important to remember that there is a sort of continuum between the various A types (including A1 and A2) in terms of the number of A antigen sites expressed per red cell, and that there is no such thing as an absolute A1, absolute A2, absolute A3 and so on and so forth. In addition, of course, the lectin Dolichos biflorus is NOT an anti-A1, but is a lectin that "recognises" the A antigen, the Tn antigen and Cad antigen, as well as the A1 antigen. It will not recognise the A antigen if it is diluted correctly, but every now and again, there will be an "overlap" between a "strong A2", and a stronger than normal Dol. b. reagent. In addition, there are people who are Aint, or A intermediate, who are somewhere between an A1 and an A2. SUch people usually hail from the south of the African continent, but not always.
As we are looking at ABO, I am not for one minute surprised that the reaction strength increases with cold incubation.
The next thing to remember is that, with gel, it is all but impossible to add the reagents to the reaction chamber so that they remain exactly at 37oC, which means that IgM antibodies, such as anti-A1, which it looks like this pregnant lady has in her circulation, will sensitise A1 red cells prior to true incubation at 37oC, but will not elute quickly enough to give negative reactions after 37oC incubation, particularly after centrifugation.
Anti-A1 is NOT clinically significant, unless it reacts STRICTLY at 37oC, and this can really only be shown by a pre-warming tube technique, in which case A2 blood (or other A subtypes, such as A3, Ax and Am) can safely be transfused, keeping your precious group O units for group O recipients. Anti-A1 has NEVER been implicated in clinically significant HDFN.
None of this is pregnancy related.