Nonsteroidal anti-inflammatory drugs are associated with a variety of adverse effects, some pharmacologic and most of the others idiosyncratic. Cutaneous adverse effects include bruising from antiplatelet effects, urticaria secondary to NSAIDs as a class or to specific NSAIDs, chronic idiopathic urticaria aggravated by NSAIDs, dermatologic conditions aggravated by NSAIDs (eg dermatitis herpetiformis [Griffiths, C. E. M., J. N. Leonard, and LIONEL FRY. "Dermatitis herpetiformis exacerbated by indomethacin." British Journal of Dermatology 112.4 (1985): 443-445]), photosensitivity reactions, and other dermatologic adverse effects. I believe your patient falls into the last category as the description was a desquamating rash without urticaria. I am not sure if the eye swelling was due to angioedema or nonspecific fluid retention and the dermatitis. An older, general review of indomethacin (Boardman, P. L., and F. Dudley Hart. "Side-effects of indomethacin." Annals of the rheumatic diseases 26.2 (1967): 127.) describes 2% of adverse effects as being dermatologic.
NSAIDs are divided into chemical classes and there may be a greater chance of a reaction with other members of the class if the adverse effect is due to specific response. I am skeptical that this is the case for your patient, but I would suggest you avoid members of the “acetic acid” group (diclofenac, ketorolac, tolmetin, sulindac) and if you must use an NSAID choose a member of the propionic acid family (eg ibuprofen, naproxen), oxicam family (eg meloxicam) or fenamate family (eg meclofenamate). I would suggest avoiding NSAIDs and use acetaminophen as an analgesic. However, a graded challenge with one of the other NSAIDs would be reasonable if NSAID is absolutely necessary. NSAIDs are known to cause photosensitive reactions, usually phototoxic, with piroxicam and ketoprofen, the most likely to cause.
In summary, I agree with you that the eye swelling and dermatitis are likely due to indomethacin and I would avoid for the foreseeable future. There is no testing that would likely help in defining a mechanism. I would avoid all NSAIDs if possible and if an NSAID was necessary would use a member of a family other than the ‘acetic acid group’. Ideally would perform a graded challenge or at least give the first dose in clinic, even though the reaction to the indomethacin was delayed. I do not think a COX II selective NSAID would necessarily be any safer from the standpoint of the rash. There is seldom a situation, other than the use of aspirin for antiplatelet effect, where a NSAID is essential.
I hope this information is of help to you and your patient.
All my best.
Dennis K. Ledford, MD, FAAAAI