Twin pregnancy has risks of adverse outcomes for mother and baby. Data synthesis is required to gain evidence to aid recommendations but may be hampered by variations in outcome reporting. Study design: Systematically review outcomes reported in twin pregnancy trials (PROSPERO -CRD42019133805). Searches were performed in MEDLINE, EMBASE, CINHAL, Cochrane library (inception-January 2019) for randomised control trials or their follow-up studies reporting prediction, prognosis, intervention or management outcomes in twin pregnancy. The study characteristics, outcomes definitions and measurements were extracted and descriptively analysed. Results: 49 RCTs and 8 follow-up studies evaluated 21 interventions, 1257 outcomes, categorised into 170 unique outcomes. 65 % of trials included all twin pregnancies, 12 % DCDA and 11 % MCDA only or MCMA and MCDA. Five (9 %) papers were prediction/ prognosis RCT's and 52 (91 %) related to an intervention. Of interventions, 40 (77 %) were medical, 34 (85 %) for preterm birth; 12 (23 %) surgical, 6 (50 %) related to TTTS interventions (83 % for monochrorionic studies). Commonest domains were: 'Neonatal' 77 %, 'Delivery' 70 % and 'Survival' 67 %. Least reported were longer term outcomes for 'Infant' or 'Parental'. Conclusions: Twin pregnancy outcomes are diverse and complex. This is related to the need to address maternal, single and double fetal outcomes and different types of chorionicity. The lack of outcome standardisation in selection, definition and reporting hinders evidence synthesis and the selection of outcomes important to women and health care professionals thus limiting the effectiveness of research.