Risk factors for coexistence of cervical elongation in uterine prolapse

To identify factors predicting cervical elongation in women with uterine prolapse.

The medical records of women with uterine prolapse who underwent vaginal hysterectomy were reviewed. Multivariable logistic regression analysis was performed to identify predictors of cervical elongation.

Of 295 women with uterine prolapse, 136 (46.1%) patients had cervical elongation, according to Berger et al. Classification (i.e., cervical length >3.38 cm and/or cervix-to-corpus lengths ratio >0.79). Multivariable analysis revealed that lower parity (odds ratio = 0.85, 95% confidence interval [CI] = 0.73 to 0.99, P = 0.04) and advanced stage of uterine prolapse (odds ratio = 1.97, 95% CI = 1.35-2.88, P < 0.001) were predictors for cervical elongation. Based on a receiver operating characteristic curve (ROC) analysis, the following optimum cut-off values were determined for cervical elongation: (1) parity ≤3, ROC area = 0.60 (95% CI = 0.53 to 0.66); (2) stage of uterine prolapse ≥3, ROC area = 0.63 (95% CI = 0.56 to 0.69). Thus, the predicted logit(p) for a given parity (a) and stage of uterine prolapse (b) can be denoted by logit(p) = -1.26 - 0.16 x a + 0.68 x b. The optimum cut-off values of logit(p) ≥-0.18 to predict cervical elongation were determined using ROC analysis (area = 0.66, 95% CI = 0.59 to 0.73). For women with parity ≤6, we can use either (1) stage 2 uterine prolapse and parity ≤1, or (2) ≥ stage 3 uterine prolapse as criteria to predict cervical elongation.

Lower parity and advanced stage of uterine prolapse are predictors of cervical elongation in women with uterine prolapse. Thus, stage of uterine prolapse ≥3 or logit(p) ≥-0.18 may be useful for predicting cervical elongation.