Despite advances in the care of newborns, intraventricular hemorrhage (IVH) remains a common complication in preterm infants, especially in those <32 weeks' gestation (1). The overall incidence of IVH is about 36% in preterm infants (2), and the incidence of IVH increases as gestational age and birth weight decrease (3). It occurs in about 45% of extremely preterm infants with a birth weight of <750 g and in about 52% of preterm infants born at <28 weeks' gestation (4, 5). About 50% of cases of IVH occur within 1 day after birth, and about 90% occur within 3 days of life (6).

IVH is graded from I to IV according to the classification of Papile (7). Several studies have shown that III-IV IVH implies more severe pathologic injuries and that it is associated with greater adverse neurological outcomes in preterm infants (8–11) such as cerebral palsy (CP), intellectual ability, and impairment of academic skills. Although I-II IVH is generally considered benign, subtle impairment in subcortical white matter and decreasing cortical gray matter following I-II IVH have been identified by cerebral MRI and might influence the development of processing skills and motor coordination (12, 13). Whether I-II IVH leads to poor neurodevelopmental outcomes in preterm infants remains a controversial issue (14–17), and the purpose of this study was to assess the impact of different grades of IVH (grade I-II IVH and grade III-IV IVH) on neurodevelopmental outcomes and death in very preterm infants.

There were 1,134 preterm infants admitted into the NICUs during the study period, and 55 of them were excluded (14 with congenital or genetic diseases, 3 with congenital cranial malformation, 20 whose parents refused to participate, and 18 who died before cranial ultrasound examination). Thus, 1,079 preterm infants were eligible (625 without IVH, 380 with I-II IVH, and 74 with III-IV IVH). There were 164 (15.2%) infants lost to follow-up, 206 (19.1%) infants who died, and 915 (84.8%) infants who were followed up to 18–24 months corrected of age (Figure 1). Infants lost to follow-up had significantly lower birth weight and higher rates of maternal age ≥35 years, gestational diabetes, sepsis, BPD, and severe retinopathy of prematurity (ROP) compared to infants followed up to 18–24 months of corrected age (p <0.05) (Supplementary Table 1).

In this study cohort, no infants with grade I-II IVH developed hydrocephalus, while the incidence of hydrocephalus in infants with grade III-IV IVH was 15/67 (22.4%). Of nine grade III IVH infants with hydrocephalus (Three of whom died and two of whom developed neurological disabilities), 2 (22.2%) received ventriculoperitoneal shunt. Of 6 grade IV IVH infants with hydrocephalus (Three of whom died and three of whom developed neurological disability), 2 (33.3%) received furosemide therapy to achieve ventricular decompression.

This very preterm cohort showed that mortality and neurodevelopmental outcomes are different according to the severity of IVH in very preterm infants. Preterm infants with I-II IVH and <28 weeks' gestation and those with 28–30 weeks' gestation had similar neurological outcomes and mortality compared to those without IVH. However, III-IV IVH was associated with CP, disability, mortality, and disability + death at 18–24 months of corrected age.

Grade I-II IVH occurs in the germinal matrix or ventricles without ventricular dilatation. The impact of I-II IVH on long-term neurological outcomes in preterm infants has always been a concern because of the conflicting results reported on this issue. In the current study, infants with I-II IVH did not show increased adverse neurological outcomes compared with the non-IVH group after adjusting for confounding factors, and this was consistent with some previous clinical studies (Supplementary Table 3) (10, 16, 34). Even though infants with grade I-II IVH had higher rates of neonatal morbidities such as sepsis, NEC, BDP, severe ROP, and PVL compared with infants with non-IVH (Table 1), BPD and PVL were independent risk factors for disability (Supplementary Table 2). However, it is not grade I-II IVH itself that results in higher rates of neonatal morbidities, and instead multiple perinatal risk factors lead to a simultaneous increase in the incidence of grade I-II IVH, BPD, and PVL in very preterm infants. Reubsaet et al. (16) performed a study with a cohort of 342 preterm infants with a gestational age of 24–32 weeks, and they found that preterm infants with I-II IVH had similar neurodevelopmental outcomes compared with controls at 2 years' corrected age. With a lager sample size, our study came to a similar conclusion at 18–24 months' corrected age. A large sample size, multi-center cohort including 1,472 preterm infants with an extremely low gestational age (<27 weeks' gestation) also found that low-degree IVH had no effect on neurological outcomes (10). Ann Wy et al. (34) enrolled 985 premature infants (gestational age <37 weeks) and found that I-II IVH was not associated with adverse outcomes in intelligence, academic achievement, or problematic behavior at 3, 8, and 18 years of age. Thus, grade I-II IVH might not impair neurological development in preterm infants.

In contrast to those studies, however, several other studies have shown adverse neurodevelopmental outcomes in preterm infants with I-II IVH (Supplementary Table 3) (15, 35, 36). It has been reported that I-II IVH is followed by subtle impairment in subcortical white matter development, which has relevant impacts on the development of processing skills and fine motor coordination, especially in preterm infants with low gestational age (12, 13). Patra et al. (35) found that I-II IVH was associated with adverse neurological outcomes in extremely preterm infants at 20 months of corrected age even without any white matter abnormalities. The adverse neurological outcomes in infants assessed at the first years of life are more likely to be associated with cognitive and neurological impairment at school age (37). With a routine long-term follow-up, Klebermass-Scherhof et al. (36) found that I-II IVH was associated with significant adverse neurological outcomes at 1, 2, 3, and 5.5 years of age. Hollebrandse et al. (9) showed that I-II IVH was associated with a significantly higher rate of CP in school-aged children born extremely preterm, but not with any difficulties in academic skills or motor dysfunction. Some studies support the finding that reduced cortical gray matter following I-II IVH may result in neurological disabilities in preterm infants (36, 38). The reduction in cortical gray matter might be due to glial precursor cell loss during migration, and the absence of astrocytes and oligodendrocytes might then disrupt cerebral cortical development and myelination (36).

Severe IVH with ventricular dilatation or parenchymal involvement is well-known to be related to poor outcomes in preterm infants (39). In our study, III-IV IVH increased the incidence of neurological disability, which was consistent with other studies (15, 36, 40, 41). Ventricular dilatation, periventricular edema, and increased intracranial pressure following the pathological process of III-IV IVH all contribute to the development of neurological disability (15, 42). Oxygen deprivation due to persistent inhibition of cerebral blood flow, hemoglobin-mediated oxidative injury, and cell death following III-IV IVH also contribute to the development of neurological disability (43–45). In this study, grade III IVH and grade IV IVH were combined into a single entity because both groups of IVH are associated with poor outcomes. There was a tendency for a higher rate of disability or death in grade IV IVH compared to grade III, but the difference was not statistically significant. This might be related to lower numbers of infants with grade IV IVH. Due to different pathogenic mechanisms, grade IV IVH presumably represents a different and more severe form of injury compared to grade III.

Mortality in preterm infants with IVH is associated with the severity of bleeding, and III-IV IVH is one of the main causes of death in preterm infants (46). In our study, preterm infants with III-IV IVH had a significant increase in mortality compared to those without IVH. Preterm infants with III-IV IVH had a higher rate of post-hemorrhagic hydrocephalus, and some of them required surgical shunts to decrease intracranial pressure, all of which contribute to the development of neurological disability and death (47). However, there are currently no effective therapies for neonatal IVH. Fortunately, in the past few decades antenatal steroids (48), magnesium sulfate (49), mesenchymal stem cells (50), and erythropoietin (30) have shown promise as therapeutic agents for neuroprotection against death and neurological disability in preterm infants with severe IVH.

There were some limitations in this study. First, preterm infants who were lost to follow-up had higher rates of neonatal complications such as sepsis, BPD, and ROP than those who completed follow-up (Supplementary Table 1), which might lead to more preterm infants with poor neurological outcomes being lost to follow-up. Second, the follow-up at 18–24 months of corrected age limited the power of our study. Subsequent follow-ups to school age or even longer for comprehensively evaluating the impact of IVH on neurological outcomes in preterm infants are needed.

Our study indicated that preterm infants of <30 weeks' gestational age with I-II IVH had similar neurological outcomes and mortality compared to those without IVH, while III-IV IVH was associated with CP, disability, death, and disability + death at 18–24 months of corrected age. Effective strategies are required to protect against III-IV IVH and reduce adverse neurological outcomes. Further follow-up to school age is also needed in preterm infants with IVH.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by the Ethics Committee of the Third Affiliated Hospital of Zhengzhou University. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

YW and CZ designed the study. YW, JS, XZ, WK, WL, YY, SZ, and FX were involved in data collection. YW, JS, XW, and CZ analyzed the data and wrote the paper. All authors have read and approved the final manuscript.

This study was supported by National Key Research and Development Program of China (2018YFC1004604), the National Nature Science Foundation of China (U21A20347), Swedish Research Council (2018-02267), ALF (ALFGBG-965197), The Royal Society for Science and Knowledge in Gothenburg (2020-476, 2021-496), and Stiftelsen Edit Jacobsons Donationsfond (2021-102).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.