Combined Cardiac Ultrasound Indicators to Predict Increased Extravascular Lung Water in Bipolar Hysteroscopic Surgery

Abstract

Background: Hysteroscopic surgical dilatation fluid overload is a serious complication that can lead to circulatory overload, increased extravascular lung water (EVLW) and even acute heart failure pulmonary edema. Because bipolar hysteroscopy uses electrolyte fluid as dilatation fluid, we cannot determine dilatation fluid overload by the traditional method of blood sodium ion concentration, and currently we urgently need a method that can quickly diagnose distended fluid overload during hysteroscopy early to prevent operative hysteroscopy intravascular absorption (OHIA). Objective: To construct a predictive model of fluid overload caused by bipolar hysteroscopic procedures monitored by bedside cardiac ultrasound and evaluate its predictive ability. Screening bedside cardiac ultrasound indicators to advance the diagnosis of fluid overload-induced increased EVLW. Methods: 110 patients undergoing bipolar electric knife hysteroscopic release of uterine adhesions were selected. Cardiac ultrasound indices included inferior vena cava collapse rate (IVC-CI), cardiac output (CO), left ventricular ejection fraction (LVEF), E-peak to A-peak ratio (E/A), the ratio of peak mitral valve early diastolic flow velocity (E) on pulsed Doppler ultrasound to peak mitral annular early diastolic motion velocity (e’) on tissue Doppler ultrasound (E/e’), mitral annular systolic displacement (MAPSE), and tricuspid annular systolic displacement (TAPSE). Mann-Whitney U non-parametric test was used to screen the significant single-factor indicators mentioned above for multi-factor logistic regression analysis, identify the independent influencing factors indicators, establish a prediction model, draw the ROC (receiver operating characteristic) curves, and evaluate predictive ability of the prediction model. Results: Multifactor logistic regression analysis showed that IVC-CI and E/A were independent risk factors for increased lung water in bipolar hysteroscopic surgery. The expression of the joint detection factor model was obtained as Logit (p) (y = lung water increase) = 3.893 – 0.081 × ICV-CI – 5.839 × E/A. The area under the ROC curve of the constructed joint prediction model was 0.938 (95% CI (confidence interval): 0.896–0.980), with a sensitivity of 84.6% and a specificity of 94.8%. Conclusions: The combined predictive model of cardiac ultrasound indicators constructed based on logistic regression analysis is of high diagnostic value for early detection of complications of fluid overload leading to increased EVLW during bipolar hysteroscopy.