Neurodevelopmental outcome in preterm infants with intraventricular hemorrhages: the potential of quantitative brainstem MRI

Correletations between neurodevelopmental outcome after 1 yr of life (CCS 1, LCS 1, and MCS 1), 2 yr of life (CCS 2, LCS 2, and MCS 2), and metric relaxation parameters based on MDME sequences.

		CCS 1	LCS 1	MCS 1	CCS 2	LCS 2	MCS 2
Medulla oblongata T1R	Pearson correlation	0.209	0.139	0.271	0.105	0.026	0.157
Medulla oblongata T1R	Sig. (2-tailed)	0.137	0.327	0.052	0.489	0.866	0.297
Medulla oblongata T2R	Pearson correlation	0.191	0.126	0.190	0.036	−0.008	0.008
Medulla oblongata T2R	Sig. (2-tailed)	0.175	0.375	0.178	0.810	0.958	0.956
Medulla oblongata PD	Pearson correlation	0.067	0.075	0.022	−0.062	0.026	−0.034
Medulla oblongata PD	Sig. (2-tailed)	0.636	0.598	0.877	0.683	0.866	0.823
Pontine tegmentum T1R	Pearson correlation	0.149	0.113	0.162	−0.012	−0.112	0.018
Pontine tegmentum T1R	Sig. (2-tailed)	0.293	0.425	0.252	0.935	0.459	0.905
Pontine tegmentum T2R	Pearson correlation	0.199	0.111	0.258	0.039	−0.034	0.048
Pontine tegmentum T2R	Sig. (2-tailed)	0.157	0.434	0.064	0.795	0.820	0.752
Pontine tegmentum PD	Pearson correlation	0.181	0.162	0.083	0.057	0.076	−0.024
Pontine tegmentum PD	Sig. (2-tailed)	0.198	0.252	0.561	0.704	0.617	0.876
Basis pontis T1R	Pearson correlation	0.433^a	0.432^a	0.466^a	0.359^b	0.189	0.327^b
Basis pontis T1R	Sig. (2-tailed)	0.002	0.002	0.001	0.018	0.225	0.032
Basis pontis T2R	Pearson correlation	0.298^b	0.235	0.349^b	0.166	0.155	0.091
Basis pontis T2R	Sig. (2-tailed)	0.037	0.104	0.014	0.287	0.321	0.563
Basis pontis PD	Pearson correlation	0.294^b	0.280	0.188	0.333^b	0.253	0.230
Basis pontis PD	Sig. (2-tailed)	0.040	0.052	0.197	0.029	0.101	0.137
Midbrain T1R	Pearson correlation	0.201	0.184	0.162	0.068	−0.061	0.126
Midbrain T1R	Sig. (2-tailed)	0.152	0.192	0.252	0.652	0.685	0.406
Midbrain T2R	Pearson correlation	0.538^a	0.449^a	0.530^a	0.380^a	0.307^b	0.433^a
Midbrain T2R	Sig. (2-tailed)	0.000	0.001	0.000	0.009	0.038	0.003
Midbrain PD	Pearson correlation	0.305^b	0.434^a	0.300^b	0.293^b	0.225	0.268
Midbrain PD	Sig. (2-tailed)	0.028	0.001	0.031	0.048	0.133	0.072

		CCS 1	LCS 1	MCS 1	CCS 2	LCS 2	MCS 2
Medulla oblongata T1R	Pearson correlation	0.209	0.139	0.271	0.105	0.026	0.157
Medulla oblongata T1R	Sig. (2-tailed)	0.137	0.327	0.052	0.489	0.866	0.297
Medulla oblongata T2R	Pearson correlation	0.191	0.126	0.190	0.036	−0.008	0.008
Medulla oblongata T2R	Sig. (2-tailed)	0.175	0.375	0.178	0.810	0.958	0.956
Medulla oblongata PD	Pearson correlation	0.067	0.075	0.022	−0.062	0.026	−0.034
Medulla oblongata PD	Sig. (2-tailed)	0.636	0.598	0.877	0.683	0.866	0.823
Pontine tegmentum T1R	Pearson correlation	0.149	0.113	0.162	−0.012	−0.112	0.018
Pontine tegmentum T1R	Sig. (2-tailed)	0.293	0.425	0.252	0.935	0.459	0.905
Pontine tegmentum T2R	Pearson correlation	0.199	0.111	0.258	0.039	−0.034	0.048
Pontine tegmentum T2R	Sig. (2-tailed)	0.157	0.434	0.064	0.795	0.820	0.752
Pontine tegmentum PD	Pearson correlation	0.181	0.162	0.083	0.057	0.076	−0.024
Pontine tegmentum PD	Sig. (2-tailed)	0.198	0.252	0.561	0.704	0.617	0.876
Basis pontis T1R	Pearson correlation	0.433^a	0.432^a	0.466^a	0.359^b	0.189	0.327^b
Basis pontis T1R	Sig. (2-tailed)	0.002	0.002	0.001	0.018	0.225	0.032
Basis pontis T2R	Pearson correlation	0.298^b	0.235	0.349^b	0.166	0.155	0.091
Basis pontis T2R	Sig. (2-tailed)	0.037	0.104	0.014	0.287	0.321	0.563
Basis pontis PD	Pearson correlation	0.294^b	0.280	0.188	0.333^b	0.253	0.230
Basis pontis PD	Sig. (2-tailed)	0.040	0.052	0.197	0.029	0.101	0.137
Midbrain T1R	Pearson correlation	0.201	0.184	0.162	0.068	−0.061	0.126
Midbrain T1R	Sig. (2-tailed)	0.152	0.192	0.252	0.652	0.685	0.406
Midbrain T2R	Pearson correlation	0.538^a	0.449^a	0.530^a	0.380^a	0.307^b	0.433^a
Midbrain T2R	Sig. (2-tailed)	0.000	0.001	0.000	0.009	0.038	0.003
Midbrain PD	Pearson correlation	0.305^b	0.434^a	0.300^b	0.293^b	0.225	0.268
Midbrain PD	Sig. (2-tailed)	0.028	0.001	0.031	0.048	0.133	0.072

Significant results are shown in bold.

^aCorrelation is significant at the 0.01 level (2-tailed).

^bCorrelation is significant at the 0.05 level (2-tailed).

Table 1

Correletations between neurodevelopmental outcome after 1 yr of life (CCS 1, LCS 1, and MCS 1), 2 yr of life (CCS 2, LCS 2, and MCS 2), and metric relaxation parameters based on MDME sequences.

		CCS 1	LCS 1	MCS 1	CCS 2	LCS 2	MCS 2
Medulla oblongata T1R	Pearson correlation	0.209	0.139	0.271	0.105	0.026	0.157
Medulla oblongata T1R	Sig. (2-tailed)	0.137	0.327	0.052	0.489	0.866	0.297
Medulla oblongata T2R	Pearson correlation	0.191	0.126	0.190	0.036	−0.008	0.008
Medulla oblongata T2R	Sig. (2-tailed)	0.175	0.375	0.178	0.810	0.958	0.956
Medulla oblongata PD	Pearson correlation	0.067	0.075	0.022	−0.062	0.026	−0.034
Medulla oblongata PD	Sig. (2-tailed)	0.636	0.598	0.877	0.683	0.866	0.823
Pontine tegmentum T1R	Pearson correlation	0.149	0.113	0.162	−0.012	−0.112	0.018
Pontine tegmentum T1R	Sig. (2-tailed)	0.293	0.425	0.252	0.935	0.459	0.905
Pontine tegmentum T2R	Pearson correlation	0.199	0.111	0.258	0.039	−0.034	0.048
Pontine tegmentum T2R	Sig. (2-tailed)	0.157	0.434	0.064	0.795	0.820	0.752
Pontine tegmentum PD	Pearson correlation	0.181	0.162	0.083	0.057	0.076	−0.024
Pontine tegmentum PD	Sig. (2-tailed)	0.198	0.252	0.561	0.704	0.617	0.876
Basis pontis T1R	Pearson correlation	0.433^a	0.432^a	0.466^a	0.359^b	0.189	0.327^b
Basis pontis T1R	Sig. (2-tailed)	0.002	0.002	0.001	0.018	0.225	0.032
Basis pontis T2R	Pearson correlation	0.298^b	0.235	0.349^b	0.166	0.155	0.091
Basis pontis T2R	Sig. (2-tailed)	0.037	0.104	0.014	0.287	0.321	0.563
Basis pontis PD	Pearson correlation	0.294^b	0.280	0.188	0.333^b	0.253	0.230
Basis pontis PD	Sig. (2-tailed)	0.040	0.052	0.197	0.029	0.101	0.137
Midbrain T1R	Pearson correlation	0.201	0.184	0.162	0.068	−0.061	0.126
Midbrain T1R	Sig. (2-tailed)	0.152	0.192	0.252	0.652	0.685	0.406
Midbrain T2R	Pearson correlation	0.538^a	0.449^a	0.530^a	0.380^a	0.307^b	0.433^a
Midbrain T2R	Sig. (2-tailed)	0.000	0.001	0.000	0.009	0.038	0.003
Midbrain PD	Pearson correlation	0.305^b	0.434^a	0.300^b	0.293^b	0.225	0.268
Midbrain PD	Sig. (2-tailed)	0.028	0.001	0.031	0.048	0.133	0.072

		CCS 1	LCS 1	MCS 1	CCS 2	LCS 2	MCS 2
Medulla oblongata T1R	Pearson correlation	0.209	0.139	0.271	0.105	0.026	0.157
Medulla oblongata T1R	Sig. (2-tailed)	0.137	0.327	0.052	0.489	0.866	0.297
Medulla oblongata T2R	Pearson correlation	0.191	0.126	0.190	0.036	−0.008	0.008
Medulla oblongata T2R	Sig. (2-tailed)	0.175	0.375	0.178	0.810	0.958	0.956
Medulla oblongata PD	Pearson correlation	0.067	0.075	0.022	−0.062	0.026	−0.034
Medulla oblongata PD	Sig. (2-tailed)	0.636	0.598	0.877	0.683	0.866	0.823
Pontine tegmentum T1R	Pearson correlation	0.149	0.113	0.162	−0.012	−0.112	0.018
Pontine tegmentum T1R	Sig. (2-tailed)	0.293	0.425	0.252	0.935	0.459	0.905
Pontine tegmentum T2R	Pearson correlation	0.199	0.111	0.258	0.039	−0.034	0.048
Pontine tegmentum T2R	Sig. (2-tailed)	0.157	0.434	0.064	0.795	0.820	0.752
Pontine tegmentum PD	Pearson correlation	0.181	0.162	0.083	0.057	0.076	−0.024
Pontine tegmentum PD	Sig. (2-tailed)	0.198	0.252	0.561	0.704	0.617	0.876
Basis pontis T1R	Pearson correlation	0.433^a	0.432^a	0.466^a	0.359^b	0.189	0.327^b
Basis pontis T1R	Sig. (2-tailed)	0.002	0.002	0.001	0.018	0.225	0.032
Basis pontis T2R	Pearson correlation	0.298^b	0.235	0.349^b	0.166	0.155	0.091
Basis pontis T2R	Sig. (2-tailed)	0.037	0.104	0.014	0.287	0.321	0.563
Basis pontis PD	Pearson correlation	0.294^b	0.280	0.188	0.333^b	0.253	0.230
Basis pontis PD	Sig. (2-tailed)	0.040	0.052	0.197	0.029	0.101	0.137
Midbrain T1R	Pearson correlation	0.201	0.184	0.162	0.068	−0.061	0.126
Midbrain T1R	Sig. (2-tailed)	0.152	0.192	0.252	0.652	0.685	0.406
Midbrain T2R	Pearson correlation	0.538^a	0.449^a	0.530^a	0.380^a	0.307^b	0.433^a
Midbrain T2R	Sig. (2-tailed)	0.000	0.001	0.000	0.009	0.038	0.003
Midbrain PD	Pearson correlation	0.305^b	0.434^a	0.300^b	0.293^b	0.225	0.268
Midbrain PD	Sig. (2-tailed)	0.028	0.001	0.031	0.048	0.133	0.072

Significant results are shown in bold.

^aCorrelation is significant at the 0.01 level (2-tailed).

^bCorrelation is significant at the 0.05 level (2-tailed).

Correlation between the severity of hemorrhage and neurodevelopmental outcome

Subdivided by severity of hemorrhage according to Papile, the low-grade group (IVH I and II) showed significant correlations only between the T2R determined in the midbrain and the CCS collected at 1 yr corrected age (r = 0.481, P = 0.013; Supplementary Material Table S6).

The high-grade group (IVH III and IV/PVHI) showed significance in the correlation between T2R of the midbrain and CCS after 1 yr (r = 0.506, P = 0.008), LCS after 1 yr (r = 0.444, P = 0.023), and MCS after 1 yr (r = 0.624, P = 0.001) and 2 yr (r = 0.473, P = 0.0023). In this group, there was a significant correlation between PD of the midbrain and LCS after 1 yr (r = 0.431, P = 0.028).

T1R of the basis pontis showed highly significant correlations with CCS after 1 yr (r = 0.507, P = 0.010), LCS after 1 yr (r = 0.581, P = 0.002), and MCS after 1 yr (r = 0.592, P = 0.002; Supplementary Material Table S7).

Correlation between neonatal neuroimaging and development of PHVD

Significant correlations were shown between the measured quantitative metrics of the medulla oblongata, pontine tegmentum, basis pontis, and midbrain, the subsequent development of PHVD, and the need for neurosurgical intervention (Table 2). These correlations were most pronounced for the T2R metrics (medulla oblongata T2R-neurosurgery: P = 0.0002, η² = 0.167; pontine tegmentum T2R-neurosurgery: P < 0.0001, η² = 0.200; midbrain T2R-PHVD: P < 0.0001, η² = 0.197; midbrain T2R-neurosurgery: P < 0.0001, η² = 0.188).

Table 2

Correlation between relaxation parameters and potential occurrence of PHVD or need for neurosurgical intervention.

		Medulla oblongata T1R	Medulla oblongata T2R	Medulla oblongata PD	Pontine tegmentum T1R	Pontine tegmentum T2R	Pontine tegmentum PD	Basis pontis T1R	Basis pontis T2R	Basis pontis PD	Midbrain T1R	Midbrain T2R	Midbrain PD
PHVD	Partial eta squared	0.104	0.077	0.003	0.101	0.122	0.010	0.140	0.048	0.002	0.059	0.197	0.114
PHVD	Sig. (2-tailed)	0.004	0.014	0.624	0.005	0.002	0.395	0.005	0.107	0.764	0.032	<0.0001	0.002
Neurosurgery	Partial eta squared	0.070	0.167	0.006	0.090	0.200	0.020	0.055	0.134	0.000	0.057	0.188	0.176
Neurosurgery	Sig. (2-tailed)	0.019	0.0002	0.483	0.008	<0.0001	0.212	0.086	0.006	0.984	0.035	<0.0001	0.0001

		Medulla oblongata T1R	Medulla oblongata T2R	Medulla oblongata PD	Pontine tegmentum T1R	Pontine tegmentum T2R	Pontine tegmentum PD	Basis pontis T1R	Basis pontis T2R	Basis pontis PD	Midbrain T1R	Midbrain T2R	Midbrain PD
PHVD	Partial eta squared	0.104	0.077	0.003	0.101	0.122	0.010	0.140	0.048	0.002	0.059	0.197	0.114
PHVD	Sig. (2-tailed)	0.004	0.014	0.624	0.005	0.002	0.395	0.005	0.107	0.764	0.032	<0.0001	0.002
Neurosurgery	Partial eta squared	0.070	0.167	0.006	0.090	0.200	0.020	0.055	0.134	0.000	0.057	0.188	0.176
Neurosurgery	Sig. (2-tailed)	0.019	0.0002	0.483	0.008	<0.0001	0.212	0.086	0.006	0.984	0.035	<0.0001	0.0001

Significant results are shown in bold.

Table 2

Correlation between relaxation parameters and potential occurrence of PHVD or need for neurosurgical intervention.

		Medulla oblongata T1R	Medulla oblongata T2R	Medulla oblongata PD	Pontine tegmentum T1R	Pontine tegmentum T2R	Pontine tegmentum PD	Basis pontis T1R	Basis pontis T2R	Basis pontis PD	Midbrain T1R	Midbrain T2R	Midbrain PD
PHVD	Partial eta squared	0.104	0.077	0.003	0.101	0.122	0.010	0.140	0.048	0.002	0.059	0.197	0.114
PHVD	Sig. (2-tailed)	0.004	0.014	0.624	0.005	0.002	0.395	0.005	0.107	0.764	0.032	<0.0001	0.002
Neurosurgery	Partial eta squared	0.070	0.167	0.006	0.090	0.200	0.020	0.055	0.134	0.000	0.057	0.188	0.176
Neurosurgery	Sig. (2-tailed)	0.019	0.0002	0.483	0.008	<0.0001	0.212	0.086	0.006	0.984	0.035	<0.0001	0.0001

		Medulla oblongata T1R	Medulla oblongata T2R	Medulla oblongata PD	Pontine tegmentum T1R	Pontine tegmentum T2R	Pontine tegmentum PD	Basis pontis T1R	Basis pontis T2R	Basis pontis PD	Midbrain T1R	Midbrain T2R	Midbrain PD
PHVD	Partial eta squared	0.104	0.077	0.003	0.101	0.122	0.010	0.140	0.048	0.002	0.059	0.197	0.114
PHVD	Sig. (2-tailed)	0.004	0.014	0.624	0.005	0.002	0.395	0.005	0.107	0.764	0.032	<0.0001	0.002
Neurosurgery	Partial eta squared	0.070	0.167	0.006	0.090	0.200	0.020	0.055	0.134	0.000	0.057	0.188	0.176
Neurosurgery	Sig. (2-tailed)	0.019	0.0002	0.483	0.008	<0.0001	0.212	0.086	0.006	0.984	0.035	<0.0001	0.0001

Significant results are shown in bold.

Intraclass correlation analysis

The ICC showed almost excellent agreement between both raters for all structures and metrics ranging from 0.844 (95% CI 0.711 to 0.910) to 0.991 (95% CI 0.985 to 0.995; detailed analysis in Supplementary Material Table S5).

Discussion

Our study embarked on addressing the gap in predicting neurodevelopmental outcome in preterm infants with IVHs, a prevalent concern that poses significant challenges for early intervention and therapeutic strategies.

Easily applicable, reliable clinical-radiologic imaging biomarkers in the prognostic assessment of outcome after IVH are relevant for further treatment stratification, therapeutic support, and parental counseling.

A recent study successfully explored the utility of quantitative MRI metrics in predicting neurodevelopmental outcomes in extremely preterm neonates (Schmidbauer et al. 2024).

Here, we focused on early myelinating structures in the neonatal brainstem, which harbors primary corticospinal connections as well as primary neurophysiologic functions, using a rather novel quantitative MRI approach and its relation to future outcome in a comparably large group of preterm neonates with IVH.

Positive correlations between outcomes with T2R and PD values were particularly pronounced in the midbrain. According to the literature, increased relaxometric properties and PD values are considered to indicate lower myelination states (Lee et al. 2018; Schmidbauer et al. 2019). However, the results observed in this study suggest that MRI metrics associated with more advanced states of myelination at term-equivalent age correlate with generally poorer future outcomes in terms of cognitive, language, and motor performance (schematic overview in Table 3).

Table 3

A simplified overview of the quantified T1R, T2R, and PD values and their significance in the studied patient population.

	T1R	T2R	PD
Better neurodevelopmental outcome	▲	▲	▲
Worse neurodevelopmental outcome	▼	▼	▼
Accelerated myelination	▼	▼	▼

Higher T1R, T2R, and PD metrics are based on an accelerated myelination of the investigated regions of the brainstem. This accelerated myelination is associated with poorer neurodevelopmental outcome. Conversely, neonates with lower T1R, T2R, and PD values showed better neurodevelopmental outcome.

Table 3

. https://doi.org/10.1016/j.neuroimage.2018.07.021.

A simplified overview of the quantified T1R, T2R, and PD values and their significance in the studied patient population.

	T1R	T2R	PD
Better neurodevelopmental outcome	▲	▲	▲
Worse neurodevelopmental outcome	▼	▼	▼
Accelerated myelination	▼	▼	▼

Our observations and those of others support the hypothesis that achieving a more mature state of the brainstem and white matter development earlier is associated with poorer neurodevelopmental outcome. This has been explained by studies showing accelerated brain development in preterm neonates in specific neurostructural regions, associated with excessive connectivity and myelination (Alexander et al. 2019; Pittet et al. 2019; Zun et al. 2021). Kim et al. showed that more mature states of myelination of the preterm brain appear to lead to an impaired cognitive outcome (Kim et al. 2022).

As myelination follows a path from caudally to rostrally (Branson 2013), the present study focused on quantitative metrics of the brainstem, as this structure holds relatively large amounts of myelin from 20 gestational wk onward (Tanaka et al. 1995).

Throughout brain maturation, there is evidence suggesting that relaxation times undergo dynamic changes (Lee et al. 2018; Schmidbauer et al. 2022). Notably, the process of myelination seems to play a crucial role in altering T1R/T2R and PD during the fetal period and early infancy (Dubois et al. 2014; Wang et al. 2019).

It is known that T1R already experiences a decline during the pre-myelination stage, which refers to the initial interactions between H₂O and myelin building blocks (such as glycolipids and cholesterol) at the onset of myelinogenesis. However, T2R does not exhibit considerable shortening until fully developed myelin sheaths are formed (Barkovich et al. 1992; Wang et al. 2019). Therefore, lower relaxation time metrics are indicative of more advanced stages of myelination (Lee et al. 2018; Schmidbauer, Dovjak, et al. 2021a).

Consequently, our quantification focused on regions in the brainstem where significant advancements in brain myelination are known to occur during early developmental stages.

The midbrain, which revealed the strongest correlations between quantitative MR measures and long-term outcomes, holds certain structures that appear to play an important role with respect to future neurodevelopment. The present study indicates that the maturation of this region, as observed through quantitative MRI, is strongly associated with neurodevelopmental outcome. Various studies have demonstrated that the midbrain undergoes significant morphological changes in response to ongoing brain maturation, as evidenced by signal intensity changes on MRI. These changes are most likely attributed to ongoing myelination processes and ferric iron deposition. Our findings, however, further indicate that the myelination state of this region is strongly associated with neurodevelopmental outcome (Martin et al. 1991; Counsell et al. 2002).

Several structures in the midbrain undergo early myelination: the inferior colliculi are first detectable on T2-weighted fast spin-echo MR images in preterm neonates as early as the time-equivalent GA 25 onward, the medial longitudinal fasciculi as early as time-equivalent GA 29, and the medial lemnisci as early as time-equivalent GA 30 (Counsell et al. 2002). This region houses nuclei and fiber tracts crucial for the development of linguistic, cognitive, and motor skills, highlighting their significant impact on sensorimotor feedback loops (Duerden et al. 2015; Bissonette and Roesch 2016; Gut and Mena-Segovia 2019; Basso et al. 2021).

Moreover, the inferior colliculi house a crucial functional site that governs the auditory pathway. Disruptions in this area can lead to delayed language acquisition, reduced cognitive abilities, and impaired socio-developmental advancement (Ream and Lehwald 2018).

Furthermore, MR measurements of the basis pontis correlated positively with neurodevelopmental outcomes in our study. Highly significant positive correlations were found between the relaxation parameters measured on T1R in the basis pontis and cognitive, language, as well as motor outcomes. This may be attributed to the process of myelination of the tegmentum pontis being almost finished at term, which does not apply to the basis pontis (Brody et al. 1987; Tate et al. 2015). The sharp postnatal increase in oligodendrocyte generation and proliferation of Olig2+ cells is closely associated with the rapid progress of motor skills observed during the early months of life (Brody et al. 1987; Tate et al. 2015). Simultaneously, the development and volumetric enlargement of the corticospinal tract, passing through the base of the pons, occur in conjunction with the aforementioned processes. The premature maturation of these brain structures appears to have a detrimental effect on the flexibility and plasticity of the pyramid pathway, thereby negatively affecting neurodevelopment.

In our study, neonates with low-grade IVH (grades I and II according to Papile) showed a single significant positive correlation between reduced midbrain T2R values and impaired cognitive outcome at 1 yr. By the age of 2, however, these infants had displayed comparable neurodevelopment, pointing toward effective compensatory mechanisms during further development. This underlines that while MDME-based metrics can sensitively predict early developmental challenges, these challenges appear to represent a transient developmental delay rather than lasting adverse outcomes.

For grade III and IV/PVHI hemorrhages, however, MDME-based markers indicate associations with long-term prognosis (in terms of cognitive, language, and motor development) as early as 1 yr of corrected age.

In the past, it was generally believed that IVH grades I and II, which can be difficult to detect on ultrasound (Buchmayer et al. 2022), were associated with favorable outcome, while grades III and IV/PVHI were correlated with adverse results. Our study, however, indicates that a more nuanced distinction is possible with the aid of quantitative MRI, even within the high-grade IVH categories. Novel scoring systems are emerging to assist clinicians in achieving a more refined prognosis (Goeral et al. 2022). Furthermore, our study has illustrated the association between metric correlations in neonatal brainstems and neurodevelopmental outcomes.

The highly significant correlations between relaxation parameters and the later occurrence of PHVD or the need for neurosurgical intervention, respectively, were again most pronounced in the midbrain (Table 2). While brainstem metric measures at term-equivalent age are not yet viable as a clinical decision criterion for neurosurgical procedures, their significant correlations make them valuable candidates for inclusion in future models. These models aim to enhance clinical decision-making and improve outcome predictions.

Clinicians could consider these metrics to identify infants at higher risk of adverse outcome early, enabling the initiation of targeted early intervention programs. Infants exhibiting patterns of accelerated myelination might benefit from closer neurodevelopmental monitoring and early engagement in therapies designed to enhance cognitive, language, and motor skills (Ballabh and de Vries 2021; Sant et al. 2021).

The degree of brainstem involvement indicated by MRI could guide the intensity and nature of therapeutic interventions, ranging from physical therapy to more specialized neurodevelopmental therapies, e.g. cognitive developmental therapy, occupational therapy, speech and language therapy, or neurodevelopmental treatment.

MDME sequences offer the advantage of providing multiple contrasts while providing quantitative maps based on a single scan. Thus, MDME sequences may become an important component for routine MRIs for premature neonates in the future to shorten the overall examination time in this vulnerable patient population.

The MDME sequences displayed substantial interrater reliability in our study and their modality reliability, as reported in the existing literature, aligns with that of conventional sequences (Schmidbauer, Geisl, et al. 2021b). This observation is particularly relevant for neonates, given that their MRI scans frequently suffer from motion artifacts.

These findings underscore the promising capacity of MDME sequences to complement conventional MRI sequences in neonatal imaging.

Strengths and limitations

The excellent and consistent intraclass coefficients between the 2 raters confirm that SyMRI® is a reliable and accurate tool for the quantification of MR metrics in the neonatal brainstem. The presented analysis may be performed rapidly, allowing incorporation into the clinical routine. Future studies could explore the fully automated analysis and interpretation of the examined structures.

However, our study had limitations that should be considered. First, the study focused solely on a restricted number of early-myelinating areas in the brainstem, neglecting the exploration of associations between quantitative metrics in pre-myelinating regions and neurodevelopmental outcomes.

Furthermore, this study did not explore potential correlations between neurodevelopmental outcome and extracranial effects of preterm birth.

In this study, the use of the Bayley Scales of Infant and Toddler Development presents limitations that need consideration. Primarily, the Bayley scales offer quantitative outcomes that may not fully reflect the qualitative aspects of neurodevelopment, particularly in motor skills. Children may achieve acceptable scores yet still exhibit notable developmental challenges, especially in motor function (Burakevych et al. 2017).

Due to the novelty of this technique, MDME sequences are still rarely performed in the daily clinical setting. As a result, neonatal norm values from a (larger) patient population are not yet available. This study further investigated the relationship between T1R/T2R metrics and neurodevelopmental outcome in neonates with IVH but did not include a healthy control population.

Furthermore, outcome parameters were not available for all patients at 2 yr; therefore, the results are based primarily on 1 yr outcome.

Conclusion

Quantitative MRI of the brainstem included in routine diagnostics of preterm infants, offers valuable prognostic information, facilitating the early identification of neonates at risk for complications and unfavorable long-term outcome following IVH.

MDME sequences show the potential to complement or replace conventional sequences in the future, and thus, refine the current gold standard in neonatal MR neuroimaging.

Acknowledgments

We would like to thank Ines Fötschl for her graphic support and Mary McAllister for proofreading our work. Further, we acknowledge the expertise and enthusiastic work of the radiology technicians of the Department of Neuroradiology of the Medical University of Vienna.

Author contributions

Patric Kienast (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Writing—original draft, Writing—review & editing), Victor Schmidbauer (Conceptualization, Formal analysis, Investigation, Software, Writing—review & editing), Mehmet Salih Yildirim (Data curation, Formal analysis), Selina Seeliger (Formal analysis, Investigation, Project administration, Writing—review & editing), Marlene Stuempflen (Software, Validation, Writing—review & editing), Julia Elis (Formal analysis, Funding acquisition, Investigation, Writing—review & editing), Vito Giordano (Software, Supervision, Validation, Writing—review & editing), Renate Fuiko (Supervision, Validation, Writing—review & editing), Monika Olischar (Resources, Supervision, Validation, Writing—review & editing), Klemens Vierlinger (Resources, Validation, Writing—review & editing), Christa Noehammer (Supervision, Visualization, Writing—review & editing), Angelika Berger (Resources, Supervision, Validation, Writing—review & editing), Daniela Prayer (Conceptualization, Validation, Writing—original draft, Writing—review & editing), Gregor Kasprian (Project administration, Resources, Software, Supervision, Validation, Writing—review & editing), and Katharina Goeral (Conceptualization, Data curation, Formal analysis, Funding acquisition, Resources, Supervision, Validation, Writing—review & editing).

Funding

Patric Kienast and Julia Elis are funded by the Vienna Science and Technology Fund (WWTF; LS20-030, PIMIENTO).

Conflict of interest statement: None declared.

References

Adams

Conway

. Eta squared. In:

Michalos

, editors.

Encyclopedia of quality of life and well-being research

Dordrecht

Springer Netherlands

;

2014

. pp.

1965

–

1966

Alexander

Kelly

Adamson

Beare

Zannino

Chen

Murray

Loh

Matthews

Warfield

, et al.

Changes in neonatal regional brain volume associated with preterm birth and perinatal factors

NeuroImage

2019

185

(185):

654

–

663

Andica

Hagiwara

Hori

Kamagata

Koshino

Maekawa

Suzuki

Fujiwara

Ikeno

Shimizu

, et al.

Review of synthetic MRI in pediatric brains: basic principle of MR quantification, its features, clinical applications, and limitations

J Neuroradiol

2019

(

268

–

275

. https://doi.org/10.1016/j.neurad.2019.02.005.

Ballabh

Vries

White matter injury in infants with intraventricular haemorrhage: mechanisms and therapies

Nat Rev Neurol

2021

(

199

–

214

. https://doi.org/10.1038/s41582-020-00447-8.

Barkovich

Kjos

Jackson

Jr,

Norman

Normal maturation of the neonatal and infant brain: MR imaging at 1.5 T

Radiology

1988

166

(

173

–

180

. https://doi.org/10.1148/radiology.166.1.3336675.

Barkovich

Lyon

Evrard

Formation, maturation, and disorders of white matter

AJNR Am J Neuroradiol

1992

(

447

–

461

. https://doi.org/10.1016/j.neuron.2021.01.013.

Basso

Bickford

Cang

Unraveling circuits of visual perception and cognition through the superior colliculus

Neuron

2021

109

(

918

–

937

Bayley

Manual for the Bayley scales of infant development

New York: Psychological Corporation

;

1969

Google Preview

. https://doi.org/10.1016/j.clp.2013.09.005.

Benders

Kersbergen

Vries

Neuroimaging of white matter injury, intraventricular and cerebellar hemorrhage

Clin Perinatol

2014

(

–

Bissonette

Roesch

Development and function of the midbrain dopamine system: what we know and what we need to

Genes Brain Behav

2016

(

–

. https://doi.org/10.1111/gbb.12257.

Branson

Normal myelination: a practical pictorial review

Neuroimaging Clin N Am

2013

(

183

–

195

. https://doi.org/10.1016/j.nic.2012.12.001.

Brody

Kinney

Kloman

Gilles

Sequence of central nervous system myelination in human infancy. I. An autopsy study of myelination

J Neuropathol Exp Neurol

1987

(

283

–

301

. https://doi.org/10.1097/00005072-198705000-00005.

Buchmayer

Kasprian

Giordano

Schmidbauer

Steinbauer

Klebermass-Schrehof

Berger

Goeral

Routine use of cerebral magnetic resonance imaging in infants born extremely preterm

J Pediatr

2022

248

(248):

–

80.e1

. https://doi.org/10.1016/j.jpeds.2022.05.033.

Burakevych

Alsweiler

Wouldes

Harding

Chyld Study T

Bayley-III motor scale and neurological examination at 2 years do not predict motor skills at 4.5 years

Dev Med Child Neurol

2017

(

216

–

223

. https://doi.org/10.1111/dmcn.13232.

Cheng

Ballabh

Recovery of the brain after intraventricular hemorrhage

Semin Fetal Neonatal Med

2022

(

101224

. https://doi.org/10.1016/j.siny.2021.101224.

Cheong

JLY

Anderson

Burnett

Roberts

Davis

Hickey

Carse

Doyle

Victorian Infant Collaborative Study Group

Changing neurodevelopment at 8 years in children born extremely preterm since the 1990s

Pediatrics

2017

139

(

e20164086

. https://doi.org/10.1542/peds.2016-4086.

Childs

Ramenghi

Cornette

Tanner

Arthur

Martinez

Levene

Cerebral maturation in premature infants: quantitative assessment using MR imaging

AJNR Am J Neuroradiol

2001

(

1577

–

1582

Counsell

Maalouf

Fletcher

Duggan

Battin

Lewis

Herlihy

Edwards

Bydder

Rutherford

MR imaging assessment of myelination in the very preterm brain

Am J Neuroradiol

2002

(

872

–

881

. https://doi.org/10.1016/j.wneu.2021.06.043.

Deger

Goethe

LoPresti

Lam

Intraventricular hemorrhage in premature infants: a historical review

World Neurosurg

2021

153

(153):

–

Deoni

Mercure

Blasi

Gasston

Thomson

Johnson

Williams

Murphy

Mapping infant brain myelination with magnetic resonance imaging

J Neurosci

2011

(

784

–

791

. https://doi.org/10.1523/JNEUROSCI.2106-10.2011.

Ding

Kucinski

Wittkugel

Goebell

Grzyska

Gorg

Kohlschutter

Zeumer

Normal brain maturation characterized with age-related T2 relaxation times: an attempt to develop a quantitative imaging measure for clinical use

Investig Radiol

2004

(

740

–

746

. https://doi.org/10.1097/00004424-200412000-00005.

. https://doi.org/10.1016/j.neuroscience.2013.12.044.

Dubois

Dehaene-Lambertz

Kulikova

Poupon

Huppi

Hertz-Pannier

The early development of brain white matter: a review of imaging studies in fetuses, newborns and infants

Neuroscience

2014

276

(276):

–

Duerden

Foong

Chau

Branson

Poskitt

Grunau

Synnes

Zwicker

Miller

Tract-based spatial statistics in preterm-born neonates predicts cognitive and motor outcomes at 18 months

AJNR Am J Neuroradiol

2015

(

1565

–

1571

. https://doi.org/10.3174/ajnr.A4312.

Ferrie

Barantin

Saliba

Akoka

Tranquart

Sirinelli

Pourcelot

MR assessment of the brain maturation during the perinatal period: quantitative T2 MR study in premature newborns

Magn Reson Imaging

1999

(

1275

–

1288

. https://doi.org/10.1016/S0730-725X(99)00080-6.

Goeral

Kasprian

Huning

Waldhoer

Fuiko

Schmidbauer

Prayer

Felderhoff-Muser

Berger

Olischar

, et al.

A novel magnetic resonance imaging-based scoring system to predict outcome in neonates born preterm with intraventricular haemorrhage

Dev Med Child Neurol

2022

(

608

–

617

. https://doi.org/10.1111/dmcn.15116.

Gut

Mena-Segovia

Dichotomy between motor and cognitive functions of midbrain cholinergic neurons

Neurobiol Dis

2019

128

(128):

–

. https://doi.org/10.1016/j.nbd.2018.09.008.

Hagiwara

Warntjes

Hori

Andica

Nakazawa

Kumamaru

Abe

Aoki

SyMRI of the brain: rapid quantification of relaxation rates and proton density, with synthetic MRI, automatic brain segmentation, and myelin measurement

Investig Radiol

2017

(

647

–

657

. https://doi.org/10.1097/RLI.0000000000000365.

. https://doi.org/10.1016/j.nicl.2017.08.015.

Hinojosa-Rodriguez

Harmony

Carrillo-Prado

Van Horn

Irimia

Torgerson

Jacokes

Clinical neuroimaging in the preterm infant: diagnosis and prognosis

Neuroimage Clin

2017

(16):

355

–

368

Kang

Choi

Kim

Hwang

Yun

Kim

Sohn

The effect of varying slice thickness and interslice gap on T(1) and T(2) measured with the multidynamic multiecho sequence

Magn Reson Med Sci

2019

(

126

–

133

. https://doi.org/10.2463/mrms.mp.2018-0010.

Kenet

Kuperman

Strauss

Brenner

Neonatal IVH—mechanisms and management

Thromb Res

2011

127

(

Suppl 3

S120

–

S122

. https://doi.org/10.1016/S0049-3848(11)70032-9.

. https://doi.org/10.1097/RLI.0000000000000833.

Kim

Choi

Lee

Jung

Gho

Association of cerebral blood flow and brain tissue relaxation time with neurodevelopmental outcomes of preterm neonates: multidelay arterial spin labeling and synthetic MRI study

Investig Radiol

2022

(

254

–

262

. https://doi.org/10.1016/j.jcm.2016.02.012.

Koo

A guideline of selecting and reporting intraclass correlation coefficients for reliability research

J Chiropr Med

2016

(

155

–

163

Langley

Blake

Coe

Recent review of germinal matrix hemorrhage–intraventricular hemorrhage in preterm infants

Neonatal Netw

2022

(

100

–

106

. https://doi.org/10.1891/11-T-722.

Lee

Choi

You

Lee

Kim

Lee

Cheon

Age-related changes in tissue value properties in children: simultaneous quantification of relaxation times and proton density using synthetic magnetic resonance imaging

Investig Radiol

2018

(

236

–

245

. https://doi.org/10.1097/RLI.0000000000000435.

. https://doi.org/10.1024/1661-4747/a000232.

Macha

Petermann

Bayley scales of infant and toddler development, third edition—deutsche fassung

Z Psychiatr Psychol Psychother

2015

(

139

–

143

. https://doi.org/10.1007/BF00598609.

Martin

Krassnitzer

Kaelin

Boesch

MR imaging of the brainstem: normal postnatal development

Neuroradiology

1991

(

391

–

395

Murphy

Inder

Rooks

Taylor

Anderson

Mogridge

Horwood

Volpe

Posthaemorrhagic ventricular dilatation in the premature infant: natural history and predictors of outcome

Arch Dis Child Fetal Neonatal Ed

2002

(

F37

–

F41

. https://doi.org/10.1136/fn.87.1.F37.

Navidi

De Boissieu

Mannes

Mokhtari

Adamsbaum

Periventricular hemorrhagic infarction (PVHI) associated with intraventricular hemorrhage (IVH) in premature infants: outcome at 2 years of age

Arch Pediatr

2022

(

459

–

466

. https://doi.org/10.1016/j.arcped.2022.05.004.

Papile

Burstein

Koffler

Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm

J Pediatr

1978

(

529

–

534

. https://doi.org/10.1016/S0022-3476(78)80282-0.

Piccolo

Marchignoli

Pisani

Intraventricular hemorrhage in preterm newborn: predictors of mortality

Acta Biomed

2022

(

e2022041

. https://doi.org/10.23750/abm.v93i2.11187.

. https://doi.org/10.1016/j.nicl.2019.102014.

Pittet

Vasung

Huppi

Merlini

Newborns and preterm infants at term equivalent age: a semi-quantitative assessment of cerebral maturity

Neuroimage Clin

2019

(24):

102014

Ream

Lehwald

Neurologic consequences of preterm birth

Curr Neurol Neurosci Rep

2018

(

. https://doi.org/10.1007/s11910-018-0862-2.

Rothman

No adjustments are needed for multiple comparisons

Epidemiology

1990

(

–

. https://doi.org/10.1097/00001648-199001000-00010.

Sant

Hotwani

Palaskar

Naqvi

Arora

Effectiveness of early physiotherapy in an infant with a high risk of developmental delay

Cureus

2021

(13):

e16581

. https://doi.org/10.7759/cureus.16581.

. https://doi.org/10.3174/ajnr.A6945.

Schmidbauer

Dovjak

Geisl

Weber

Diogo

Yildirim

Goeral

Klebermass-Schrehof

Berger

Prayer

, et al.

Impact of prematurity on the tissue properties of the neonatal brain stem: a quantitative MR approach

AJNR Am J Neuroradiol

2021a

(

581

–

589

Schmidbauer

Geisl

Cardoso Diogo

Jengojan

Perepelov

Weber

Goeral

Lindenlaub

Klebermass-Schrehof

Berger

, et al.

Validity of SyMRI for assessment of the neonatal brain

Clin Neuroradiol

2021b

(

315

–

323

. https://doi.org/10.1007/s00062-020-00894-2.

Schmidbauer

Geisl

Diogo

Weber

Goeral

Klebermass-Schrehof

Berger

Prayer

Kasprian

SyMRI detects delayed myelination in preterm neonates

Eur Radiol

2019

(

7063

–

7072

. https://doi.org/10.1007/s00330-019-06325-2.

Schmidbauer

Yildirim

Dovjak

Goeral

Buchmayer

Weber

Diogo

Giordano

Mayr-Geisl

Prayer

, et al.

Different from the beginning: WM maturity of female and male extremely preterm neonates—a quantitative MRI study

AJNR Am J Neuroradiol

2022

(

611

–

619

. https://doi.org/10.3174/ajnr.A7472.

Schmidbauer

Yildirim

Dovjak

Goeral

Buchmayer

Weber

Kienast

Diogo

Prayer

Stuempflen

, et al.

Quantitative magnetic resonance imaging for neurodevelopmental outcome prediction in neonates born extremely premature—an exploratory study

Clin Neuroradiol

2024

. https://doi.org/10.1007/s00062-023-01378-9.

. https://doi.org/10.1159/000528981.

Steiner

Schwarz

Kasprian

Rittenschober-Boehm

Schmidbauer

Fuiko

Olischar

Klebermass-Schrehof

Berger

Goeral

Brain biometry reveals impaired brain growth in preterm neonates with intraventricular hemorrhage

Neonatology

2023

120

(

225

–

234

Szpecht

Szymankiewicz

Nowak

Gadzinowski

Intraventricular hemorrhage in neonates born before 32 weeks of gestation—retrospective analysis of risk factors

Childs Nerv Syst

2016

(

1399

–

1404

. https://doi.org/10.1007/s00381-016-3127-x.

Tanaka

Mito

Takashima

Progress of myelination in the human fetal spinal nerve roots, spinal cord and brainstem with myelin basic protein immunohistochemistry

Early Hum Dev

1995

(

–

. https://doi.org/10.1016/0378-3782(94)01608-R.

Tate

Lindquist

Nguyen

Sanai

Barkovich

Huang

Rowitch

Alvarez-Buylla

Postnatal growth of the human pons: a morphometric and immunohistochemical analysis

J Comp Neurol

2015

523

(

449

–

462

. https://doi.org/10.1002/cne.23690.

Valdez Sandoval

Hernandez Rosales

Quinones Hernandez

Chavana Naranjo

Garcia Navarro

Intraventricular hemorrhage and posthemorrhagic hydrocephalus in preterm infants: diagnosis, classification, and treatment options

Childs Nerv Syst

2019

(

917

–

927

. https://doi.org/10.1007/s00381-019-04127-x.

van de

Bor

Guit

Schreuder

Wondergem

Vielvoye

Early detection of delayed myelination in preterm infants

Pediatrics

1989

(

407

–

411

. https://doi.org/10.1542/peds.84.3.407.

. https://doi.org/10.1016/j.semperi.2022.151594.

Vohr

Neurodevelopmental outcomes of premature infants with intraventricular hemorrhage across a lifespan

Semin Perinatol

2022

(

151594

Walani

Global burden of preterm birth

Int J Gynaecol Obstet

2020

150

(

–

. https://doi.org/10.1002/ijgo.13195.

Wang

Ledig

Hajnal

Counsell

Schnabel

Deprez

Quantitative assessment of myelination patterns in preterm neonates using T2-weighted MRI

Sci Rep

2019

(

12938

. https://doi.org/10.1038/s41598-019-49350-3.

Warntjes

Leinhard

West

Lundberg

Rapid magnetic resonance quantification on the brain: optimization for clinical usage

Magn Reson Med

2008

(

320

–

329

. https://doi.org/10.1002/mrm.21635.

World Medical Association

World Medical Association declaration of Helsinki: ethical principles for medical research involving human subjects

JAMA

2013

310

(

2191

–

2194

. https://doi.org/10.1001/jama.2013.281053.

Yakovlev

PLA

. The myelogenetic cycles of regional maturation of the brain. In:

, editors.

Regional development of the brain in early life

Oxford

Blackwell

;

1967

. pp.

–

Google Preview