Predictive value of uterocervical angle combined with phosphorylated insulin-like growth factor binding protein-1 testing for preterm delivery and adverse neonatal outcomes.

BACKGROUND: Preterm birth is a leading cause of perinatal morbidity. While transvaginal cervical length (CL) is the standard screening tool for symptomatic women, its low specificity frequently leads to overtreatment and unnecessary hospital admissions. Although the uterocervical angle (UCA) and phosphorylated insulin-like growth factor binding protein-1 (phIGFBP-1) are recognized as strong independent predictors, the clinical value of integrating these biomechanical and biochemical markers into a single triage model remains unclear. OBJECTIVE: To investigate the predictive accuracy of a combined model incorporating UCA, CL, and the phIGFBP-1 test for spontaneous preterm delivery, and to evaluate its clinical utility and association with neonatal outcomes. STUDY DESIGN: A cohort study was conducted on 130 pregnant women presenting with symptoms of threatened preterm labor. The latency period (time-to-delivery) was analyzed using Kaplan-Meier survival curves and a multivariable Cox proportional hazards regression model. Diagnostic accuracy was compared utilizing receiver operating characteristic curves. Clinical utility was evaluated by the number of preventable hospitalization days, and neonatal outcomes were compared based on the phIGFBP-1 test results. RESULTS: The combined predictive model significantly improved diagnostic accuracy, achieving an excellent area under the curve of 0.776 compared to 0.731 for CL alone (P<.001). Multivariable Cox regression identified a positive phIGFBP-1 test as the strongest independent factor accelerating preterm delivery (hazard ratio=1.675, P=.006), whereas CL (< 25 lt; 25 mm) and cervical funneling lost statistical significance. Kaplan-Meier analysis demonstrated that the longest latency period belonged to the lowest-risk group (narrow UCA and negative phIGFBP-1). Utilizing this combined model as a triage tool could have helped low-risk patients avoid 34 unnecessary hospital bed days and unwarranted exposure to tocolytics and corticosteroids. Furthermore, a positive phIGFBP-1 test was closely associated with adverse neonatal outcomes, including a significantly lower birth weight (2582.98±518.52 g vs 2899.26±400.57 g, P<.001) and reduced 1- and 5-minute Apgar scores (P<.001 and P=.002, respectively). CONCLUSION: Integrating the UCA and the biochemical marker (phIGFBP-1) provides superior predictive accuracy for preterm delivery compared to the traditional CL measurement. This protocol not only optimizes healthcare resource allocation by safely reducing unnecessary hospitalizations but also effectively identifies pregnancies at risk for adverse neonatal outcomes. More importantly, it empowers physicians to confidently triage low-risk groups, thereby restricting medication overuse, minimizing unnecessary admission rates, and mitigating the economic burden on both families and the healthcare system.