Our study emulated an RCT using cohort data to evaluate the impact of medication on closing sPDA in ELBWI. Overall, our findings indicate that medication does not significantly influence prognosis favorably. However, subgroup analyses by administration routes revealed that OG administration had a more favorable impact on prognosis compared to IV therapy.
While there are no clearly defined strategies for appropriate treatment of PDA closure, recent trends in the United States, Canada, and Korea show a preference for conservative management, which is associated with improving outcomes (45–47). Park et al. reported a decrease in the proportion of infants with a birth weight under 1.5 kg receiving medication for PDA, from 62% in 2015 to 53% in 2018 in South Korea (22). In contrast, our study observed a medication rate of 53.5% and a surgical ligation rate of 25.6%, reflective of the different population characteristics.
Our results broadly align with previous studies. Sung et al. conducted a double-blind RCT targeting hsPDA in infants with a GA at 23–30 weeks, reported no significant differences in BPD stage and mortality between OG ibuprofen and placebo with PNA 6–14 days (16). Slaughter et al. reported no significant differences in severe BPD and all-cause mortality when medication was administered between PNA 2–28 days in infants with a GA of < 28 weeks (15).
The primary distinctions between our study and those mentioned are in the target population, medication period, and outcome criteria. In our study, we focused on ELBWI with sPDA, considering the medication period PNA 7 days to 28 days. Unlike Sung et al., our study focused exclusively on ELBWI with sPDA and extended the study period up to PNA 90 days. Also, unlike Slaughter et al., we categorized our study population based on the sPDA. The outcomes, notably the increase in mild BPD stage but no significant differences in treatment failure rates, and notably lower ORs for all-cause mortality when OG was administered, could be attributed to these differences in study design and population characteristics.
In our study, the IV group showed a higher BPD stage than the OG group. Given the delayed absorption by OG, IV was the preferred route for immediate PDA closure. Recent meta-analyses have indicated that the efficacy of OG administration is superior to that of IV administration, potentially due to the harmful effects of high peak concentrations observed with IV administration (18, 48, 49). Consequently, it was anticipated that the IV group would experience higher complications and mortalities compared to the OG group. The majority of the medication group receiving IV administration may explain the lack of significant differences in complications and all-cause mortality compared to the non-medication group.
Although not advocating treatment beyond conservative care for the closure of hsPDA, there are studies that explore cases where neither medication nor surgical ligation is performed, waiting for natural closure (50, 51). In these studies, outcomes such as BPD, IVH, and PVL did not show significant differences compared with the untreated group. However, long-term neurodevelopmental outcomes were not assessed. Similarly, our study did not evaluate neurodevelopmental outcomes, aligning with these findings
Considering that the existing body of research indicates the effectiveness of sPDA closure, conducting an RCT on this topic is ethically unfeasible (39). Therefore, through trial emulation, this study confirmed that medication for sPDA closing was no significant difference in efficacy, while OG medication was significant difference in mortality compared to IV.
However, the limitations of this study include the lack of data on the type and frequency of medications. While medication is typically administered for three consecutive days and usually approximately 2–3 times to achieve PDA closure, data on the exact frequency and duration were unavailable. Additionally, as a trial emulation was conducted, the study employed a non-random sample of participants. Finally, the study proceeded without a clear evaluation of the PDA size and hemodynamic findings, relying instead on data registered in the KNN.
Contrary to the significant correlation observed between PDA medication, and oxygen need at PNA 28 days, and BPD, the lack of a meaningful correlation with all-cause mortality suggests that BPD or the oxygen need resulting from it may not be significantly related to the survival of ELBWI. This consensus is generally acknowledged among Neonatal medical professionals (52). However, it should be thoughtfully considered from the quality of life, and longitudinal follow-up studies, possibly in collaboration with disciplines such as rehabilitation medicine, are necessary for a comprehensive understanding.