Abstract Background Pancreatoduodenectomy (PD) is a complex surgical intervention comprising multiple components which, when delivered differently, can impact outcomes, introduce bias into randomised controlled trials (RCTs) and impair accurate interpretation of results. This study aimed to systematically explore and summarise variation in PD description and standardisation amongst contemporary RCTs in pancreas surgery. Methods Systematic literature searches for RCTs of PD were undertaken from 2020 to 2023. RCTs including robotic PD, or other types of pancreas resection (e.g. total, distal), were excluded. The technical aspects of PD were deconstructed into constituent components and steps (pre-operative considerations, before skin incision, incision, dissection, resection, haemostasis, reconstruction, insertion of adjuncts, closure and after closure) using a methodological framework. Trial reports and protocols were scrutinised to examine whether the components/steps were described, and whether (and how) adherence was assessed. Results Of 83 screened articles, 29 RCTs were included, reporting on 3440 patients across 16 countries (1 trial compared pre-operative interventions, 18 trials compared intra-operative interventions, and 10 trials compared post-operative interventions). Forty-one PD steps were identified. No studies described all steps and five did not describe any. Fifteen component parts were described in one RCT. Only three RCTs reported that delivery of PD was recorded and assessed, using intra-operative photographs in all cases. Conclusions In a contemporary series of RCTs evaluating PD, descriptions of the procedure were inadequate. Clear descriptions of PD using pre-existing methodological frameworks in future trials are warranted to improve outcomes, decrease bias and facilitate cross-RCT comparisons in systematic reviews.