Patent ductus arteriosus (PDA) is a common cardiovascular problem in preterms that places additional morbidity burden on them particularly when it is large enough to be haemodynamically significant (i.e disrupts the physiologic milieu of the individual). Studies have shown that 52.0% of preterms will have PDA. Routine echocardiography for the determination of haemodynamically significant PDA (HsPDA) in all preterms is not feasible in most centres. Hence researchers have tried to identify biomarkers for its detection. Some studies have demonstrated a relationship between elevated serum brain natriuretic peptides (BNP) and HsPDA. However, these studies were predominantly conducted among the Caucasian populations. Considering the universally acknowledged differences in morbidity and outcome between Caucasian and Black preterms, these findings may be inappropriate to extrapolate to African preterms. Thus, this study aimed to determine the burden of HsPDA, and as well determine the utility of day 3 BNP in predicting HsPDA in preterms in Ilorin. This was a cross sectional analytical study conducted between May to November 2018 at the neonatal intensive care unit of University of Ilorin Teaching Hospital. Relevant biodata was obtained and physical examination findings documented. Blood sample was taken at recruitment on the third day of life to determine the plasma BNP levels using BNP Enzyme Immunoassay (EIA) Kit (Elabscience ELISA kit®, USA). Transthoracic echocardiography was done with SSI8000 SonoScape echocardiography machine (China) with a 7.5MHz probe on the seventh day of life to check for the presence of PDA and when present determine if haemodynamically significant. Data were analyzed using SPSS 20.0 software package (SPSS Inc., USA). The variables were presented and subjected to appropriate statistical tests to determine their significance of relationships and differences obtained. A p-value of 0.05 was considered significant. A total of 110 preterms with gestational ages 28 to 34 weeks were initially recruited on the third day of life. Fourteen preterms were unable to go through echocardiography (due to death or too ill to be taken for echocardiography) and subsequently excluded. Ninety-six preterms had both BNP and echocardiography data and was analysed. The median (IQR) gestational age was 32 (30-33) weeks while the median (IQR) age at recruitment was 60 (52-70) hours and the birth weight was 1505.0 (1285.0- 1722.0) g. Males were 54 constituting 56.3% of the study population. The median (IQR) plasma BNP on the third day of life was 738.0pg/ml (648.3- 903.8). There was no significant difference of plasma BNP levels across gender, gestational age groups, birth weight categories, p > 0.05 in each. Eighteen (18.8%) of the 96 preterms had PDA with ductal diameter ranging between 1.06mm and 3.20mm and a mean ± SD of 1.88 ± 0.51mm. Of the eighteen subjects with PDA, thirteen subjects had an haemodynamically significant PDA (HsPDA) with a prevalence of 13.5% amongst all subjects and represents 72.2% of those with PDA. There was no significant difference in the prevalence of HsPDA across gender, gestational age groups and birth weight categories (all p> 0.05). The left atria diameter; left atrium/aortic root (LA/Ao) and the ductal diameter were significantly higher in the subjects with HsPDA as compared to subjects without HsPDA (p= <0.001; <0.001 and 0.003 respectively). The left ventricular internal diameter in diastole, left ventricular internal diameter in systole, ejection fraction and fractional shortening showed no significant difference in subjects with HsPDA as well as those without HsPDA (all p>0.05). The median (IQR) plasma BNP in subjects with HsPDA was 754.0 pg/ml (565.0-977.5) and 734.0pg/ml IQR (666.0-864.0) in subjects without HsPDA. There was no significant difference in the plasma BNP levels of subjects with and without HsPDA (p = 0.957). There was no significant correlation between the day 3 plasma BNP and presence of HsPDA on day 7(r = - 0.005; p = 0.958). Similarly, there was no significant correlation between plasma BNP and LA/Ao (r = -0.048; p = 0.640) as well as with ductal diameter (r = -0.191; p = 0.447). The optimal cut-off of day 3 plasma BNP that maximize the predictiveness for day 7 HsPDA was >710pg/ml with an area under the curve (AUC) of 0.523. At this cut-off, the sensitivity, specificity, positive predictive value and negative predictive value were 61.5%, 49.3%, 17.4% and 88.1% respectively. In conclusion, day 3 plasma BNP is a poor predictive tool for the occurrence of HsPDA in our environment.