IntroductionPrevious studies demonstrated that the implementation of the Kidney Disease Improving Global Outcomes (KDIGO) guideline-based bundle, consisting of different supportive measures in patients at high risk for acute kidney injury (AKI), might reduce rate and severity of AKI after surgery. However, the effects of the care bundle in broader population of patients undergoing surgery require confirmation.Methods and analysisThe BigpAK-2 trial is an international, randomised, controlled, multicentre trial. The trial aims to enrol 1302 patients undergoing major surgery who are subsequently admitted to the intensive care or high dependency unit and are at high-risk for postoperative AKI as identified by urinary biomarkers (tissue inhibitor of metalloproteinases 2*insulin like growth factor binding protein 7 (TIMP-2)*IGFBP7)). Eligible patients will be randomised to receive either standard of care (control) or a KDIGO-based AKI care bundle (intervention). The primary endpoint is the incidence of moderate or severe AKI (stage 2 or 3) within 72 hours after surgery, according to the KDIGO 2012 criteria. Secondary endpoints include adherence to the KDIGO care bundle, occurrence and severity of any stage of AKI, change in biomarker values during 12 hours after initial measurement of (TIMP-2)*(IGFBP7), number of free days of mechanical ventilation and vasopressors, need for renal replacement therapy (RRT), duration of RRT, renal recovery, 30-day and 60-day mortality, intensive care unit length-of-stay and hospital length-of-stay and major adverse kidney events. An add-on study will investigate blood and urine samples from recruited patients for immunological functions and kidney damage.Ethics and disseminationThe BigpAK-2 trial was approved by the Ethics Committee of the Medical Faculty of the University of Münster and subsequently by the corresponding Ethics Committee of the participating sites. A study amendment was approved subsequently. In the UK, the trial was adopted as an NIHR portfolio study. Results will be disseminated widely and published in peer-reviewed journals, presented at conferences and will guide patient care and further research.Trial registration numberNCT04647396.

Intraoperative hypotension is common and has been associated with adverse events, including acute kidney failure, myocardial infarction, and stroke. Since blood pressure is a multidimensional and measurable variable, artificial intelligence and machine learning have been used to predict it. To date, studies have shown that the prediction and prevention of hypotension can reduce the incidence of hypotension. This review describes the development and evaluation of an artificial intelligence predictive algorithm called Hypotension Prediction (HPI), which can predict hypotension up to 15 min before it occurs.