Development and Validation of a Nomogram for Predicting Oral Ibuprofen Treatment Failure in Preterm Infants With Hemodynamically Significant Patent Ductus Arteriosus.

Abstract

Background: Hemodynamically significant patent ductus arteriosus (hsPDA) is common in extremely and very preterm infants and has been associated with pulmonary overcirculation, systemic hypoperfusion, and adverse neonatal outcomes. However, the optimal timing and selection of pharmacologic treatment remain controversial, highlighting the need for tools that can help identify infants who are unlikely to respond to oral ibuprofen.

Objective: Hemodynamically significant patent ductus arteriosus remains a frequent and critical complication in preterm infants. Due to the lack of clinical consensus, management strategies vary significantly, often leading to inconsistent treatment outcomes. This study aimed to develop and validate a risk prediction model for oral ibuprofen treatment failure in extremely and very preterm infants with hsPDA.

Methods: We conducted a retrospective study of extremely and very preterm infants (gestational age <32 weeks, birth weight <1500 g) diagnosed with hsPDA between January 2020 and August 2024. Patients were categorized into treatment-effective and treatment-ineffective groups based on their response to oral ibuprofen. Clinical and echocardiographic parameters were compared using univariate analysis, and clinically relevant candidate predictors were entered into a multivariable logistic regression model using the enter method to develop the prediction model. A predictive nomogram was constructed; its performance was evaluated using discrimination, calibration, and decision curve analysis to determine clinical utility.

Results: Among 165 eligible infants, 143 with complete candidate predictor data were included in model development. Six clinically relevant predictors were included in the final model: gestational age, hypertensive disorders of pregnancy, pretreatment hemoglobin level, lung ultrasound score, patent ductus arteriosus diameter, and ratio of left atrial diameter to aortic root diameter ratio. The model demonstrated good discrimination, with an area under the receiver operating characteristic curve of 0.889 (95% CI, 0.826-0.952), and acceptable calibration.

Conclusion and relevance: This nomogram may help identify infants who are less likely to respond to oral ibuprofen and support individualized decision-making when pharmacologic treatment is being considered. External validation is needed before routine clinical implementation.