RESUMO
Introduction: Scores are available to predict the probability of contrast-induced nephropathy (CIN) after cardiac interventions, but not many scores are available for non-cardiac interventions and there are none for intravenous exposure to contrast. We designed this study to develop a simplified score to determine the probability of developing CIN in patients exposed to the parenteral contrast medium. Methods: This was a prospective study of patients who received parenteral contrast. Of 1300 patients, the first 1000 comprised the derivation cohort and the next 300 comprised the validation cohort. The patient variables in the development cohort were studied using univariate analysis. Statistically significant individual variables were used as independent variables, and CIN was used as the dependent variable in the final multivariate logistic regression model. Then, the risk score was obtained and validated. Results: The incidence of CIN was 3.8%. The risk factors, namely the presence of diabetes mellitus, e-GFR, and route and volume of contrast material were significantly associated with the risk of CIN (P < 0.05). The developed risk score had a sensitivity of 90.4% and specificity of 98.78%. The overall accuracy was 97.8%. The values of AUC of ROC in the development and validation datasets were high. This indicated that the predicted CIN risk score correlated well with the calibration and discriminative characteristics. Conclusions: The route and volume of contrast administered, low e-GFR, and diabetes mellitus were the significant risk factors. The developed risk score exhibited very good sensitivity and specificity and excellent accuracy in predicting the probability of CIN.
RESUMO
INTRODUCTION: In laparoscopic surgery, looking in the target areas is an indicator of proficiency. However, gaze behaviors revealing feedforward control (i.e., looking ahead) and their importance have been under-investigated in surgery. This study aims to establish the sensitivity and relative importance of different scene-dependent gaze and motion metrics for estimating trainee proficiency levels in surgical skills. METHODS: Medical students performed the Fundamentals of Laparoscopic Surgery peg transfer task while recording their gaze on the monitor and tool activities inside the trainer box. Using computer vision and fixation algorithms, five scene-dependent gaze metrics and one tool speed metric were computed for 499 practice trials. Cluster analysis on the six metrics was used to group the trials into different clusters/proficiency levels, and ANOVAs were conducted to test differences between proficiency levels. A Random Forest model was trained to study metric importance at predicting proficiency levels. RESULTS: Three clusters were identified, corresponding to three proficiency levels. The correspondence between the clusters and proficiency levels was confirmed by differences between completion times (F2,488 = 38.94, p < .001). Further, ANOVAs revealed significant differences between the three levels for all six metrics. The Random Forest model predicted proficiency level with 99% out-of-bag accuracy and revealed that scene-dependent gaze metrics reflecting feedforward behaviors were more important for prediction than the ones reflecting feedback behaviors. CONCLUSION: Scene-dependent gaze metrics revealed skill levels of trainees more precisely than between experts and novices as suggested in the literature. Further, feedforward gaze metrics appeared to be more important than feedback ones at predicting proficiency.
