| Abstrakt: |
To estimate the risk of failed medical treatment in patients with tubo-ovarian abscess (TOA) and assess the external validity of a clinical risk score. External validation study. The probability of failed medical treatment predicted from the reference risk score was compared with the observed rates in a retrospective cohort of patients with TOA. Safety-net teaching hospital system in Houston, Texas. 169 consecutive patients admitted with TOA between 2011 and 2018 were included. Some were treated conservatively with IV antibiotics; others required a drainage procedure. Predictors of failed conservative treatment were captured. The previously published predictors by Fouks et al, included age>36, WBC ≥ 16000), abscess diameter (≥ 7 cm) and bilateral abscesses. A clinical risk score was calculated for each patient and correlated with the risk of failed medical treatment. A multivariate logistic regression using the patient characteristics was created, in addition to the prediction model using the risk score. Then the calibration, discrimination and accuracy of the model were evaluated to perform the external validation analysis. Among 169 eligible patients, 50.2% were successfully treated with IV antibiotics and 49.8% needed drainage via minimally invasive or traditional surgery. Patients undergoing the intervention were more likely to be older, diabetic, to present with elevated white blood cell count, have fever and a large abscess size on univariate analysis. Large abscess size was found to be the strongest independent predictor for intervention in our cohort. The AUC curve for the adjusted model was 0.77 (0.71-0.84), indicating good discrimination. The Brier score was favorable (0.19) and the observed and predicted rates were similar ranging across different risk scores. Our results provide external validation to a simple clinical risk score predicting failed medical treatment in patients with TOA. Large abscess size was found to be the strongest independent predictor for intervention. [ABSTRACT FROM AUTHOR] |
