Pharmacokinetics, dosing exposure and safety of oral ibuprofen in Chinese preterm neonates with patent ductus arteriosus: preliminary report

Abstract

Objective

The use of oral ibuprofen to treat haemodynamically significant patent ductus arteriosus (hsPDA) in premature neonates is common. However, among Chinese preterm neonates, the available pharmacokinetic (PK) findings for its enantiomers remain merely descriptive. This study aimed to characterise the PKs, dosing exposure and safety of oral ibuprofen in Chinese preterm infants treated with hsPDA.

Design

A pilot trial.

Setting

Three perinatal centres across Southwest China.

Patients

Preterm infants (≤35 weeks’ gestational age, ≥48 hours postnatal age) with suspected hsPDA (August 2022–May 2023).

Intervention

Oral ibuprofen (10, 5 and 5 mg/kg at 24-hour intervals) was used and sparse sampling was performed.

Main outcome measures

Plasma concentrations were analysed via ultrahigh-performance liquid chromatography–tandem mass spectrometry. The PK parameters were calculated via ADAPT 5 software. Adverse events and treatment efficacy were assessed.

Results

14 neonates provided 42 plasma samples. The R-ibuprofen (R-IBU) concentrations were lower than the S-ibuprofen (S-IBU) concentrations. The clearance values for rac-ibuprofen (rac-IBU), R-IBU and S-IBU were 24.22±37.04, 206.1±143.1 and 12.67±26.53 mL/kg/hour, respectively. The area under the curve (AUC_0-72h) values for rac-IBU, R-IBU and S-IBU were 799.8±392.8, 75.9±51.9 and 537±246.4 µg hour/mL, respectively. DA closure was achieved in 8 (57%) infants. The indirect bilirubin level increased significantly following the administration of oral ibuprofen (p<0.05).

Conclusions

The recommended dosage regimen for Chinese preterm neonates with hsPDA may be insufficient to achieve targeted therapeutic exposure. Importantly, our study indicated that even at a relatively high dose, drug exposure might remain within the therapeutic window. Moreover, we observed a higher conversion ratio of R-IBU to S-IBU in these neonates than in adults. The use of ibuprofen may be associated with an elevated risk of jaundice in preterm infants.

Trial registration number

Chinese Clinical Trial Registry (ChiCTR2300067714).

WHAT IS ALREADY KNOWN ON THIS TOPIC.

WHAT THIS STUDY ADDS

• The findings of this study indicate that the recommended dosage regimen for Chinese preterm neonates with hsPDA may not provide adequate target exposure. There was a higher conversion ratio of R-ibuprofen to S-ibuprofen in Chinese neonates than in adults. Ibuprofen may increase the risk of jaundice in preterm infants.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• This study fills the gap in knowledge concerning the PK of ibuprofen enantiomers for PDA treatment in Chinese neonates, providing updated information on its off-label use. High-dose ibuprofen may be requested for premature Chinese patients to achieve the target exposure.

Introduction

Closure of haemodynamically significant patent ductus arteriosus (hsPDA) is often delayed by more than 1 week, which may complicate the clinical course of neonates and even increase the incidence rates of poor neonatal outcomes.¹ The clinical approach to PDA closure has become quite controversial, particularly given the published research within the past few years.² Oral ibuprofen, which is used off-label for PDA closure, is considered the first-line treatment because of its higher efficacy and favourable safety profile, not merely for its low cost.³ Some studies have reported that the favourable concentration–time pharmacokinetic (PK) effect following oral administration might be attributable to delayed absorption and/or increased pre-systemic conversion to the active S-ibuprofen (S-IBU).^{3 4}

The recently reported success rate (21%–89%) of PDA closure with oral standard-dose ibuprofen varies widely among preterm neonates.5,7 The variation in the DA closure rate might be because most PK data on oral ibuprofen enantiomers are merely descriptive and scarce.8,10 Moreover, the exposure–response relationship of intravenous/oral ibuprofen is also unclear in preterm neonates with varying gestational and postnatal ages (PNAs).^{11 12} Hirt et al suggested dosing schemes based on PNAs to achieve a total R-ibuprofen (R-IBU)/S-IBU area under the curve (AUC_0-72h) of 900 mg hour/L, thereby increasing the treatment efficacy for PDA in neonates.¹³

Rac-ibuprofen (rac-IBU) is administered as a racemic 1:1 ratio mixture of R-IBU and S-IBU. R-IBU undergoes unidirectional conversion to S-IBU in vivo.¹⁴ The reported conversion rate in Caucasian neonates varies greatly, ranging from 0% to 61%.^{11 15} The data concerning enantiomeric conversion in neonates are quite weak and contradictory. Moreover, the metabolic chiral conversion ratio of ibuprofen in Chinese premature infants has not been reported.

Therefore, to improve the efficacy and optimise the dosing of oral ibuprofen for PDA in Chinese preterm neonates, it is essential to investigate both the exposure–response relationship and the PK profile of its enantiomers in patients. The objectives of this study were to (1) establish the relationship between ibuprofen exposure (eg, S-IBU and rac-IBU) and efficacy and safety in Chinese preterm infants routinely treated for hsPDA via population PK modelling; (2) compare the concentration–time profiles of S-IBU, R-IBU and rac-IBU following standard oral dosing; and (3) explore the presence and extent of enantiomeric conversion in this population.

Methods

Trial design and patients

This research was carried out across three neonatal care units (Children’s Hospital of Yongchuan, Woman and Children’s Hospital of Yongchuan and Yongchuan Hospital Affiliated with Chongqing Medical University) from August 2022 to May 2023.

The criteria for enrolment included preterm infants (gestational age (GA) ≤35 weeks and PNA >2 days) with a gastric tube, echocardiographic evidence and clinical symptoms of hsPDA.¹⁶ All patients underwent their first echocardiographic examination within the first 12–24 hours of life. Echocardiographic diagnosis of hsPDA requires the presence of a left-to-right ductal shunt, with either a minimal ductal diameter not less than 1.4 mm/kg or a left atrium/aortic root ratio >1.4.^{17 18} Clinical findings, including heart failure, hypotension, pulmonary overload, increased respiratory support, oliguria or metabolic acidosis were considered the decision for pharmacological PDA closure.^{18 19} The exclusion criteria for infants included major congenital malformations and contraindications to ibuprofen treatment (ie, severe hyperbilirubinaemia, life-threatening infection, active bleeding, necrotising enterocolitis (NEC), a platelet count ≤60 ×10^∧9/L and renal failure, including a urine output <1 mL/kg/hour or a creatinine concentration >88.4 µmol/L).²⁰ Informed consent was obtained from the parents of the enrolled neonates.

Interventions

All the subjects received an ibuprofen suspension (Tianqian, Yangzhou Sanyao Co., Jiangsu, China) at an initial dosage of 10 mg/kg on day 1, followed by 5 mg/kg on days 2 and 3.⁶ The infants underwent cardiac and cranial ultrasound scans within 48–72 hours after the last dose of the treatment. The scans were read by the attending neonatologist. If the DA persists after the first course, patients will receive a repeat course with the same drug and dosage. Moreover, patients who fail to achieve PDA closure with mechanical ventilation after two completed medical treatments should be considered for surgical ligation or occlusion of the ductus.

Ibuprofen suspension was administered via a nasogastric tube infusion. The tube was flushed with approximately 2 mL of saline to ensure complete drug delivery. Blood samples (300–500 µL per sample) were obtained after drug administration. We employed the D-optimal design approach in conjunction with the maximum a posteriori Bayesian prediction method²¹ to optimise the sampling time points. The optimisation results indicated that sample collection at 6 hours could maximise the capture of the elimination phase of ibuprofen in the preterm population. The sampling points per neonate were determined by computer-generated random time, which was defined as one or two sampling times between 30 min and 6 hours after each dose. Blood samples were obtained either together with other clinically indicated laboratory tests, such as blood gas and glucose tests, or through venipuncture attempts. Finally, at least three blood samples were collected from each patient.

All blood samples were centrifuged within 0.5 hours. The plasma samples were collected and stored in a –80°C freezer until further analysis. These samples were measured at the Department of Pharmacy, West China Hospital (Chengdu, China). The analysis method we used was based on a verified technique reported by Xiao et al.²²

Pharmacokinetic

PK modelling of rac-IBU (rac-ibuprofen), R-IBU and S-IBU PK modelling was performed via the open source software ADAPT 5.0 (Biomedical Simulations Resource, USC, LA, California, USA).²³ This study uses an analytical solution to predict plasma concentrations on the basis of a one-compartment PK model, incorporating first-order absorption and first-order elimination after the administration of multiple doses to preterm neonates.

The following equation was used to explain the changes in the amount of ibuprofen in the central compartment:

$${\displaystyle {\displaystyle \frac{dX(1)}{dt}=K_a\times X(1)-K_{el}\times X(1)}}$$ (1)

Where X(1) indicates the amounts of rac-IBU, R-IBU and S-IBU in the central compartment; K_a is the absorption rate constant for IBU from the gut to the central compartment; and K_el is the elimination rate constant. Owing to the limited absorption phase data in this study, accurate estimation of the value of K_a proved unattainable. The oral bioavailability was assumed to be 100%, and K_a was fixed at 0.516/hour.²⁴

The PK profile of IBU in plasma (C_p) following oral administration was characterised via the following equation:

$${\displaystyle {\displaystyle C_p=\frac{X(1)}{V/F}}}$$ (2)

In this formulation:

The apparent volume of distribution within the central compartment was denoted as V and the systemic bioavailability was represented as F.

The error variance model used is as follows:

$${\displaystyle {\displaystyle V_i={\left({\sigma }_{\text{inter}}+{\sigma }_{\text{slope}}{y}_i\right)}^2}}$$ (3)

In this formulation:

The estimated variance of the ith observation was indicated as V_i; the intercept and slope parameters of the variance structure were denoted as σ_inter and σ_slope, respectively; and y_i corresponds to the ith predicted value from modelling. The parameters of the variance model (equation 3) were set at σ_inter=0.1 and σ_slope=0.1. The expectation maximisation algorithm was applied to optimise the base PK model via maximum likelihood estimation.

Safety and efficacy

Safety assessments of the study drug included abnormal clinical laboratory results, adverse events (AEs) and death, which were monitored from the beginning of the subject’s participation in the study until discharge. The data included complete renal function, liver function (aminotransferases), bilirubin levels, and albumin levels. The AEs studied primarily included abdominal distension, feeding intolerance, reduced urine output and fluid retention, among others. Transthoracic echocardiography (eg, PDA characteristics, such as DA patency, dimension and left-to-right ductus shunt) was routinely performed in all enrolled patients 2 days after the completion of study drug treatment. Data on participants’ demographics and PDA-related neonatal outcomes (including bronchopulmonary dysplasia, intraventricular haemorrhage and NEC) were collected for comprehensive evaluation.

Statistical analyses

The availability of eligible patients determined the sample size. The demographic data and PK parameters are presented as numbers (percentages) or means (SDs). The biochemical data were compared via Student’s t-test. A p value less than 0.05 was considered to indicate statistical significance. GraphPad Prism V.9.0 (GraphPad Software, San Diego, California, USA) served as the tool for all the statistical analyses.

Results

A total of 14 preterm neonates with hsPDA who met the inclusion criteria were included. All of them received at least one course of ibuprofen treatment. One subject who weighed 1120 g and was of a GA of 29.2 weeks received an unexpected 20 mg/kg loading dose. The birth weight and mean gestation age were 1841.1±840.8 g and 31.8±2.9 weeks, respectively. The demographic information and clinical profiles of the enrolled patients are summarised in table 1.

Table 1. Clinical characteristics of the trial population.

Characteristics	Patients (n=14)
Male, n (%)	11 (79)
Birth weight (g), mean±SD	1841.1±840.8
Gestational age (weeks), mean±SD	31.8±2.9
Gestational-age category, n (%)
≤32 weeks	8 (57)
32–35 weeks	6 (43)
PNA at treatment initiation (days), mean±SD	5.3±3.5
Apgar score at 1 min, mean±SD	8.4±1.3
Apgar score at 5 min, mean±SD	9.3±0.9
Premature rupture of membranes, n (%)	7 (50)
Pregnancy-induced hypertension, n (%)	5 (36)
Caesarean delivery, n (%)	10 (71)
Antenatal glucocorticoids, n (%)	7 (50)
Surfactant treatment, n (%)	8 (57)
High-frequency oscillatory ventilation, n (%)	1 (7)
Non-invasive respiratory support, n (%)	5 (36)
Mechanical ventilation, n (%)	8 (57)
Inspired oxygen, n (%)	33.7±7.6
Vasoactive drugs therapy, n (%)	4 (29)
NRDS, n (%)	8 (57)
Ductal diameter (mm), mean±SD	2.5±0.7
Ibuprofen treatment
Number of ibuprofen courses, mean±SD	1.1±0.3
Total number of ibuprofen doses, mean±SD	3.1±0.9
Total amount of ibuprofen (mg), mean±SD	36.6±15.8

Summary statistics are mean±SD and number (percentage).

PNA, postnatal age; NRDS, neonatal respiratory distress syndrome.

Pharmacokinetics

We collected 42 PK samples. One sample was below the limit of quantification, and one sample was not detected. For data analysis, the drug concentrations in two samples were assigned a value of zero. The fitted concentration–time curves for R-IBU, S-IBU and total rac-IBU in all preterm neonates are shown in online supplemental figure 1. The 72-hour PK profiles of rac-IBU, S-IBU, and R-IBU enantiomers, including individual concentration-time and model-fitted data for these patients (cases 1–14), are shown in Figure 1. The PK parameters obtained from these neonates are shown in table 2. The plasma clearances (CLs) of rac-IBU, R-IBU and S-IBU were 24.22±37.04, 206.1±143.1 and 12.67±26.53 mL/kg/hour, respectively. The areas under the curve (AUC_0–72h) of rac-IBU, R-IBU and S-IBU were 799.8±392.8, 75.9±51.9 and 537±246.4 µg hour/mL, respectively.

Figure 1. Plasma concentration and fitted 72 hours curves of rac-IBU (circles, blue line), R-IBU (upward triangles, green line), and S-IBU (downward triangles, red line) in preterm neonates with haemodynamically significant patent ductus arteriosus following oral ibuprofen administration. Cases 1–14. BW, birth weight; GA, gestational age; PNA, postnatal age; rac-IBU, rac-ibuprofen; R-IBU, R-ibuprofen; S-IBU, S-ibuprofen.

Table 2. Pharmacokinetic profile of oral racemic ibuprofen in premature infants with PDA.

Patients	BW (g)	GA (weeks)	PNA (days）	Dosage (mg/kg)	Predicted clearance (mL/kg/hour)			V (mL/kg)	AUC0–72h (μg hour/mL)
					R	S	Rac		R	S	Rac
1	1090	29.6	2	10/5/5	67.1	<0.0001	11.32	644.2	144.8	808.2	1012.1
2	1650	31.5	3	10/5/5	151.1	3.201	18.59	656	68.7	708.1	810.3
3	2880	35.6	2	10/5/5	148.7	11.97	27.6	213.4	65.9	733	698.2
4	2960	34.3	3	10/5/5	50.81	2.857	15.01	554.7	185.6	797	937.2
5	3050	34.2	4	10/5/5	145.3	<0.0001	12.76	773.4	69.9	677	873.7
6	2970	35.2	4	10/5/5	405.3	<0.0001	0.6983	785.3	24.7	663	1291.4
7	1630	31.6	3	10/5/5	122.2	<0.0001	5.425	879.7	79.3	581.2	973.8
8	955	28.1	11	10/5/5	53.95	<0.0001	<0.0001	1267	152.4	400.7	801.4
9	1120	29.2	6	20/5/5	458.4	<0.0001	<0.0001	1012	31.2	823.9	1647.8
10	2130	34.4	8	10/5/5	251.2	82.96	110.6	666.3	37.6	112.1	169.8
11	2200	34.3	4	10/5/5	237.4	65	105.7	160.9	42.1	153.8	189.2
12	1250	31	2	10/5/5	8.93	<0.0001	<0.0001	1305	99.9	383	766
13	1000	28	10	10/5/5	451.5	<0.0001	0.0004	2009	22.1	263.9	527.9
14	890	28	12	10/5/5	253.6	11.37	31.44	870.1	38.3	411.2	499
Mean	1841.1	31.8	5.3	NA	206.1	12.67	24.22	842.6	75.9	537	799.8
SD	840.8	2.9	3.5	NA	143.1	26.53	37.04	467.6	51.9	246.4	392.8

AUC, area under the curve; BW, birth weight; GA, gestational age; NA, not applicable; PDA, patent ductus arteriosus; V, the volume of distribution.

Safety and efficacy

Figure 2 shows that the aspartate aminotransferase level 48 hours after one course of treatment was significantly lower (p<0.05) than that in infants before treatment. The level of indirect bilirubin after treatment was significantly greater than that before treatment (p<0.05). Overall, AEs were reported in 9 (62%) infants (online supplemental table 1).

Figure 2. The changes in laboratory values in preterm infants with haemodynamically significant patent ductus arteriosus after receiving doses of oral ibuprofen. Shown are the changes in aspartate aminotransferase (AST), indirect bilirubin (IBIL), direct bilirubin (DBIL), alanine aminotransferase (ALT), albumin (ALB), serum creatinine (Cr), total albumin (TLB) and white cell (WBC) levels.

Eight of the 14 patients (57%) achieved DA 48-hour closure after ibuprofen treatment. One patient (7%) did not complete the cardiac ultrasound examination and was transferred to a major children’s medical centre. Therefore, the DA of that newborn was assumed to remain open, which reduces bias. A single mortality case (7%) was recorded in a 31-week-old male preterm infant who developed multiple organ dysfunction syndrome on day 37 of the study, which was considered to be unrelated to the investigational treatment. The incidence rates of other adverse outcomes were presented in table 3.

Table 3. Neonatal outcomes.

Outcomes	Patients (n=14)
PDA related outcome, n (%)
Ductal closure	8 (57)
Surgical ligation	0
Other outcome, n (%)
BPD	4 (29)
IVH	4 (29)
ROP	1 (7)
NEC	0
Sepsis	1 (7)
Impaired renal function, n (%)	5 (36)
Anaemia, n (%)	11 (79)
Pathological jaundice, n (%)	12 (86)
Respiratory infection, n (%)	13 (93)
Respiratory failure, n (%)	11 (79)
Gastrointestinal haemorrhage, n (%)	2 (14)
Pulmonary haemorrhage, n (%)	1 (7)
Hospital transfers, n (%)	3 (21)
Duration of hospitalisation (days), mean±SD	29.3±27.1
Death, n (%)	1 (7)

Results are presented as mean (SD) or number (percentage).

BPD, bronchopulmonary; IVH, intraventricular hemorrhage; NEC, necrotising enterocolitis; PDA, patent ductus arteriosus; ROP, retinopathy of prematurity.

Discussion

In this study, we explored the PK characteristics of the pharmacologic treatment of Chinese preterm infants with a hsPDA with oral ibuprofen enantiomers. The sparse sampling approach used in this study minimised the impact on preterm infants. The data were then analysed via a population modelling method. Our study provides important PK data for optimising the dosing of oral ibuprofen to eliminate PDA shunts in Chinese preterm neonates.

The fitting concentration time data of rac-IBU, S-IBU and R-IBU in all preterm neonates (cases 1–14) are depicted in figure 1. The blood concentrations of S-IBU and rac-IBU in the majority of premature infants gradually increased following the second administration. The short-term changes in the plasma levels of S-IBU and rac-IBU are essentially consistent, as shown in online supplemental figure 1. All patients always had higher concentrations of S-IBU than R-IBU during the treatment period. However, ibuprofen is composed of equal amounts of R-IBU and S-IBU, forming a racemic mixture with a 1:1 ratio, which is stored in the human stomach. This may be attributed to a higher enantiomeric R-IBU to S-IBU conversion after oral administration.³

After oral administration, the plasma R-IBU, S-IBU and rac-IBU concentrations over the 72-hour dosing interval showed significant interindividual variation among the patients. The PK parameters from the final model are presented in table 2. Our study revealed that 64.3% of preterm neonates cannot achieve the target rac-IBU AUC_0-72h of 900 mg/hour/L.¹³ One infant who received an accidental overdose maintained drug exposure levels within the therapeutic range, on the basis of prior literature.⁴ Moreover, the exposure results of rac-IBU (in terms of maximum concentration, C_max) varied widely. One of the most significant findings in this study was that the C_max values in 9 out of the 14 cases were less than the target concentration of 20.5 mg/L (a 90% probability of reaching the target exposure), as proposed in the literature and lower than those reported in similar published studies.¹¹ Surprisingly, in five of those nine patients, the ductus arteriosus remained open at the end of treatment. These findings suggests that these infants are at risk for ineffective or subtherapeutic treatment.

We simply estimated that the mean CL_S-IBU value of 12.67±26.53 mL/hour/kg for Chinese preterm infants was greater than the value of approximately 300% reported in previous studies (7.01 mL/hour,¹⁵ 3.98 mL/hour,²⁵ 3.5 mL/hour/kg²⁶). The CL_R-IBU of the PDA infants in this study (206.1±143.1 mL/hour/kg) was slightly greater than that reported in other studies (167.8 mL/hour/kg,²⁶ 189 mL/hour²⁵ and 82.6 mL/hour/kg).¹⁵ We found that CL_rac-IBU and CL_S-IBU increase with increasing GA, which is consistent with the findings reported by Samiee-Zafarghandy et al.¹¹ Like in other studies, CL_R-IBU was always higher than CL_S-IBU. These mixed findings also demonstrated one-way metabolic conversion of R-IBU to S-IBU in patients which is consistent with the findings of a previous study.¹⁵ The mean CL_rac-IBU in Chinese preterm neonates with PDA (24.22±37.04 mL/hour/kg) was lower than the reference values, representing approximately 20% of the paediatric data (Caucasian: 120 mL/hour/kg; Chinese: 140 mL/hour/kg)²⁷ and less than 1% of the adult CL value (3790 mL/hour/70 kg).²⁸

When adult ibuprofen dosing patterns are followed and similar child growth is assumed, Chinese children require higher weight-based doses than Caucasians do. However, the dose-normalised AUC_0-72h or C_max of rac-IBU in our studies was generally low. Fetal ontogeny and ethnic differences may influence the metabolism of the studied drug. First, the biotransformation of ibuprofen is influenced by the developmental physiology of preterm infants and involves metabolic enzymes, such as CYP2C9, CYP2C8 and glucuronosyltransferase.²⁷ CYP2C9 is the main enzyme responsible for the metabolism of ibuprofen, particularly S-IBU, after absorption.²⁷ Second, some studies have reported that the proportion of metabolism via CYP2C9 alleles in Caucasians is greater than that in Asians²⁷ and that mutations in this enzyme are associated with a reduced CL of ibuprofen.²⁹ Moreover, hepatic CYP2C9 is rarely expressed in the immediate postnatal period in very low birth weight infants, but its level increases nonlinearly with GA between 28 and 44 weeks.³⁰ This observation is consistent with the findings of the present study, which demonstrated a rapid growth elimination rate of S-IBU or rac-IBU in CYP2C9 variant infants, who were closely associated with physiological development at this age.

In theory, Chinese premature neonates have a lower chiral conversion ratio of R-IBU to S-IBU than adults do (25%).³¹ However, the unidirectional conversion ratio appeared to be greater in 11 of the 14 patients in our study (figure 1). CYP2C8 is the main enzyme that contributes to the hydroxylation of R-IBU both in vivo and in vitro.³² However, the fractional contributions of CYP2C8 to R-IBU clearance for Caucasians and the Chinese population were merely 2.8% and 0.81%, respectively.²⁷ Moreover, acyl-CoA synthetase is the key enzyme involved the chiral biotransformation of ibuprofen.³³ This enzymatic activity in neonatal mouse hearts is significantly greater on postnatal day 1 than in the early embryonic stage.³⁴ Therefore, rapid chiral conversion during drug metabolism plays a leading role in these patients, which is consistent with the findings of Padrini et al.¹⁵

In this study, another preterm neonate did not experience any AEs that were considered to be related to oral ibuprofen, despite accidentally receiving a high loading dose. These results are also in agreement with safety data from previous studies.^{35 36} Our study revealed a significant increase in unbound bilirubin levels following oral ibuprofen administration. Bilirubin encephalopathy in premature infants has always been a serious problem. The most important risk factors for neurotoxicity are the level of bilirubin exposure and the duration of time.³⁷ In vitro investigations demonstrated competitive albumin binding between ibuprofen and bilirubin, with IBU clearance showing pronounced protein-binding dependency due to its limited hepatic extraction.³⁸ High ibuprofen concentrations may displace bilirubin from its albumin binding site, thereby increasing free bilirubin levels.³⁹ Theoretically, IBU may increase the risk of jaundice. The competitive binding ability of IBU in preterm infants requires further research.

We observed that the CL of both ibuprofen enantiomers increased with PNA, with GA exerting an additional effect on the CL of S-IBU (figure 1). The accelerated CL in more preterm infants likely contributes to subtherapeutic drug exposure. A cohort study of 263 preterm infants with lower GAs receiving three doses of ibuprofen revealed that the closure rate of PDA was only 21% in these neonates.⁷ The ductus shows increased sensitivity to vasodilatory signals and reduced responsiveness to vasoconstricting signals, potentially indicating a need for increased ibuprofen exposure in neonates with low GAs. Moreover, as the PNA increases, standard doses of ibuprofen may not be as effective for closing hsPDA in extremely premature infants as early treatment.⁴⁰ Hirt et al reported that the decrease in efficacy is most pronounced with increasing PNA, regardless of the dose of ibuprofen indicated by PK data.¹³

There are several limitations in our study. First, clinical trials in premature infants with PDA are difficult to conduct because of practical and ethical issues such as small sample sizes and limited research conditions. The adoption of nonintervention therapeutic approaches has shown a marked upward trend in recent years, as have improvements in hsPDA management strategies for premature infants. Second, our study was a multicentre study with a small sample size and lacked a control group. Therefore, our evaluation of AEs may be unreliable. However, we still carefully recorded the very important changes during the treatment process and provided detailed neonatal intensive care unit results for evaluation and comparison. Finally, the PK parameters obtained in this study, particularly the clearance values, may be overestimated due to the constraints of the sampling schedule which may not have fully captured the terminal elimination phase. However, the PK parameters we obtained are generally consistent with the trends reported by other reports in the Caucasian population.

Conclusions

This study provides an updated profile of the PK, drug exposure and safety of oral ibuprofen for the pharmacological treatment of PDA in Chinese preterm neonates. Our study revealed a significant difference in PK among R-IBU, S-IBU and total R/S-IBU in premature infants. The low efficacy of oral ibuprofen for closing PDA is partly attributed to inadequate dosing, especially in extremely preterm infants. These findings suggest that a high-dose regimen of ibuprofen may be necessary in this vulnerable group to achieve the target exposure. Notably, in the case of an accidental higher dose, the drug exposure likely remained within the therapeutic range. Furthermore, the potential risk of jaundice must be considered. Therefore, future research should consider increasing the dose of ibuprofen to increase the success of pharmacological treatment in eliminating the shunt of hsPDA in preterm neonates. Concurrently, we should further investigate how fetal ontogeny and ancestral differences may influence the metabolism of both types of ibuprofen.

Data availability statement

Data are available upon reasonable request.