According to BSH Guidelines (and work performed by such luminaries as the late Prof Patrick Mollison - who knew a bit about blood transfusion!), those who have already made a clinically significant atypical alloantibody are more likely than others to make other specificities. It is for this reason that a patient who has made such an antibody is no longer eligible for a transfusion of blood selected by electronic issue (or, as people STILL insist on wrongly calling it, computer crossmatch - GRRRRRRRRRRR!!!!!!!!!!!!).
The more antibodies in a person's plasma, the more difficult it is to sort out all the various specificities, and the more difficult it is to supply antigen negative blood. As a (former) Reference Laboratory Chief, I prefer either to work on really easy samples, or really, really complex samples, but not on samples with two or three antibodies present, particularly where one or more are avoidable Rh and/or Kell Blood Group System (BGS) antibodies.
Therefore, although not mandated by either the BSH Guidelines, or, indeed, the UK Transfusion Services as a whole, as a personal preference, I would advise that Rh and Kell BGS,antigens are matched (particularly as, if the individual has already produced an antibody - probably from a previous transfusion - although other stimuli are "available" - and needs another transfusion, the chances are that this individual will require further transfusions in the future, if not actually become transfusion dependent).
I fully admit that this is a purely personal point-of-view, however, I would be more than a little annoyed if I was to be sent a sample at 3 o'clock in the morning, to sort out an additional, but avoidable specificity. It is, however, down to you.