Effectiveness of hysteroscopic resection of a uterine caesarean niche can be predicted: a prospective cohort study

Abstract

This study aimed to develop and validate a model for the preoperative prediction of the effectiveness of hysteroscopic resection of a uterine cesarean niche in patients with postmenstrual spotting. The predictive model was developed in a primary prospective cohort consisting of 208 patients with niche treated by hysteroscopic resection. Multivariable logistic regression analysis was performed to develop the predictive model, which incorporated preoperative menstrual characteristics and magnetic resonance imaging (MRI) findings. Surgical efficacy was defined as a decrease in postmenstrual spotting duration of at least 3 days at the 3-month follow-up compared with baseline. The predictive model was presented with a nomogram, and the performance was assessed with respect to its calibration, discrimination, and clinical use. Internal validation was performed using tenfold cross-validation. The predictive factors in the final model were as follows: preoperative menstrual duration, thickness of the residual myometrium (TRM), length, TRM/thickness of the adjacent myometrium ratio, angle γ, area, and presence of a lateral branch of the niche. The model showed good performance in predicting the effectiveness of hysteroscopic niche resection. Incorporating the preoperative duration of the menstrual period and MRI findings of the niche into an easy-to-use nomogram facilitates the individualized prediction of the effectiveness of a hysteroscopic niche resection by 26 Fr resectoscope, but multicenter prospective studies are needed to validate it.

Figure 1

Flow chart.

Figure 2

Nomogram of the predictive model. TRM (1: ≥ 2.2 mm; 2: < 2.2 mm); Length (1: < 9 mm; 2: ≥ 9 mm); Angle γ (1: < 90°; 2: ≥ 90°); Area (1: < 50mm²;2: ≥ 50 mm²); Lateral branch (0: No; 1:Yes); The degree of severity of the niche (1: Severe (TRM/TAM < 50%) ; 2: Mild (TRM/TAM ≥ 50%)).

Figure 3

Receiver operating characteristic (ROC) curves of the predictive model. Area under the ROC curve to determine the predictive ability of the model, representing the sensitivity on the ordinate axis and specificity in the abscissa. The green line represented the model 1. The red line represented the model 2.

Figure 4

Calibration plot of the predictive model. Calibration plot depicted the agreement between the predicted effective rate of hysteroscopic surgery and observed outcomes of effective rate. The y-axis represented the actual effective rate. The x-axis represented the predicted effective rate. The diagonal solid line represents a perfect prediction by an ideal model. The dotted line represented the performance of the model, of which a closer fit to the diagonal solid line represented a better prediction. (a) represented the model 1. (b) represented the model 2.

Figure 5

Decision curve analysis (DCA) for the predictive model. The y-axis measures the net benefit. The blue line represented the assumption that all patients can benefit from hysteroscopic surgery. The gray line represented the assumption that no patients can benefit from hysteroscopic surgery. The green line represented the model 1. The red line represented the model 2. The net benefit was calculated by subtracting the proportion of all patients who are false positive from the proportion who are true positive, weighting by the relative harm of forgoing treatment compared with the negative consequences of an unnecessary treatment. Threshold probability is where the expected benefit of treatment is equal to the expected benefit of avoiding treatment.