Daily variations and levels of human chorionic gonadotropin before methotrexate treatment as predictors of treatment success.

Abstract

The aim of this study is to investigate the role of human chorionic gonadotropin (hCG) daily variations and levels prior to methotrexate treatment as predictors for treatment outcome. This retrospective study included patients who had a sonographically confirmed ectopic pregnancy at the International Peace Maternity and Child Health Hospital between November 2015 and June 2020. The associations of hCG levels and daily variations with the treatment success were evaluated by multivariable logistic regression and receiver operator characteristic (ROC) curve. Establish a nomogram that predicts how methotrexate (MTX) therapy will turn out. The performance of the model was assessed utilizing concordance index, receiver operating characteristic curves, and calibration plots. The median serum hCG levels before treatment and hCG daily variation in the failure group were higher than those in the success group (487.8 vs. 270.7 IU/L, -1.86% vs. 7.29%, both p < 0.01). According to the ROC curve analysis, the cutoff values of serum hCG level before treatment and daily variations were 617.35 IU/L and 1.76%/day. By multivariable logistic regression analysis, serum hCG levels before treatment (odds ratio [OR]: 1.001, 95% confidence interval [CI]: 1.000 ~ 1.001) and hCG daily variations were independently associated with the treatment success (OR: 1.033, 95% CI: 1.015 ~ 1.052). The nomogram was effective at predicting the outcome of MTX treatment with a receiver operating characteristic area under the curve of 0.717 (p < 0.001). The nomogram's calibration curve was almost parallel to the ideal diagonal line. We successfully created a nomogram based on serum hCG levels before treatment and hCG daily changes to anticipate the result of MTX therapy, which could assist medical professionals in selecting therapeutic schedule for patients with tubal pregnancies.