Preterm infants are recommended for prophylactic indomethacin (PIND) to promote closure of patent ductus arteriosus (PDA) and reduce morbidity and mortality. This study investigated the predictive factors of a non-response to PIND for PDA in preterm-birth infants. Consecutive preterm-birth infants (gestational age: < 28 weeks) who received PIND between 2009 and 2019 were retrospectively enrolled. Seventy-six eligible participants were classified as PIND responders (N=42) or non-responders (N=34). Information on potential confounders in maternal obstetric and perinatal data were collected from medical records. Multiple logistic regression analysis was carried out to identify the prognostic factors of a PIND response in preterm-birth infants. The prevalence of intrauterine infection and multiple births was significantly different between responders and non-responders to PIND (intrauterine infection: 2 [4.8%] vs. 8 [23.5%], P=0.036; twins: 3 [7.1%] vs. 9 [ 26.5%], P=0.029, respectively). In multivariate logistic regression analysis after adjustment for multiple births, intrauterine infection was a significant and independent predictive factor of a non-response to PIND (odds ratio [OR] 5.54, 95% confidence interval [CI] 1.05-29.2, P=0.044). A remarkable association was also noted for multiple births with a non-response to PIND (OR 4.22, 95% CI 0.99-17.8, P=0.050). Intrauterine infection and multiple births were identified as potential risk factors of a non-response to PIND for PDA in preterm infants.