Urine Tx-M and PGI-M concentrations in ADAT participants categorized by treatment adherence, group assignment, and aspirin use.

Two-way ANOVAs were performed for each eicosanoid in compliant and non-compliant subjects for the three treatment groups. Group assignment was significantly (+P<0.0001) related to urine Tx-M and PGI-M in the compliant group, but not for the non-compliant group (∧P>0.05). ASA use was associated with reduced urine Tx-M concentrations (#P<0.05) in the compliant and non-compliant groups, and reduced urine PGI-M in only the compliant group (#P<0.05). A significant interaction between group assignment and ASA use (P<0.05) was detected only for urine Tx-M concentrations in the compliant group. Bonferroni-corrected post-hoc comparisons evaluated the effects of ASA use in each of the twelve groups; only urine Tx-M in compliant Placebo and compliant celecoxib participants was significant (*P<0.0001) while all others had P>0.05. In addition, values for ASA compliant and noncompliant individuals were calculated regardless of treatment group assignment, presented in rows labeled subtotal, and compared using t-tests (++P<0.0001, **P<0.05).