[Effects of postoperative urine protein levels on predicting acute kidney injury in critically ill patients undergoing non-cardiac surgery].

Objective: To investigate the predictive value of postoperative urine protein level in critically ill patients undergoing non-cardiac surgery with acute kidney injury (AKI). Methods: A total of 661 critically ill patients undergoing non-cardiac surgery, who visited the Department of Critical Care Medicine of Peking University First Hospital from May 20, 2019 to November 24, 2020, were enrolled in this prospective study. The clinical data of the patient's age, gender, body mass index, laboratory examination, surgical status, etc. were collected. AKI diagnostic criteria of the 2012 KDIGO guidelines were used to diagnose the occurrence of AKI after surgery. The independent predictors of AKI were determined by multivariate logistic regression. Results: The age of this patient cohort was (69±15) years. The prevalence of AKI was 45.4% (300/661). Multivariate logistic regression showed that urine protein semi-quantitative ≥2+(OR=2.62, 95%CI: 1.05-6.56, P=0.039) was independent factor for postoperative AKI in critically ill patients undergoing non-cardiac surgery, other independent factors include higher age (OR=1.04, 95%CI: 1.02-1.06, P=0.001), higher body mass index (BMI) (OR=1.12, 95%CI: 1.04-1.21, P=0.004), lower plasma hemoglobin level (OR=0.98, 95%CI: 0.97-1.00, P=0.019), lower central venous pressure (OR=0.89, 95%CI: 0.83-0.97, P=0.005) and lower total hypotension time (OR=1.01, 95%CI: 1.00-1.01, P=0.041). Conclusions: Urine protein semi-quantitative ≥2+after surgery is an independent predictive factor for the occurrence of postoperative AKI in critically ill patients undergoing non-cardiac surgery. It is important to check urine routine immediately after surgery to detect and deal with high-risk patients.

Colonic motility diagnosis based on colonic pressure.

Manometry of the alimentary tract is a valuable and widely used means to evaluate and diagnose the function of the alimentary tract. However, the measurement can be inconvenient due to the invasive method used, and the many factors affecting results. Research on colonic pressure data is even more insufficient. This paper deals with colonic pressure data via an improved method ensuring that pressure data of the whole colon is available. The data is analysed based on the learning vector quantization (LVQ) method. Testing results show that this method distinguishes the normal data and the abnormal data, consistently with the original diagnoses. This method can serve as an assistant diagnosis of colonic motility and contributes to further research on colonic motility based on pressure data.