Postnatal Changes in Fetal Hemoglobin and Tissue Oxygenation in Preterm and Term Neonates: A Prospective Observational Study

Abstract

Introduction

Physiological postnatal changes in fetal hemoglobin fraction (FHbF) may affect neonatal tissue oxygenation through shifts in the oxyhemoglobin dissociation curve. We examined cross-sectional associations between FHbF and near-infrared spectroscopy (NIRS)-derived tissue oxygenation, and longitudinal associations between within-infant changes in FHbF and cerebral tissue oxygenation (cStO₂) and oxygen extraction (cFTOE) during the first 3 postnatal weeks in neonates without red blood cell transfusions.

Methods

This prospective observational study included 25 preterm (median [IQR] gestational age 33.6 [1.6] weeks) and 25 term (39.9 [1.4] weeks) neonates in a tertiary-center neonatal unit. Measurements were performed at T1 (days 2–3), T2 (days 5–10), and T3 (days 12–18). cStO₂ was assessed by time-resolved NIRS, peripheral tissue oxygenation by spatially resolved NIRS and peripheral fractional oxygen extraction by venous occlusion. cFTOE was calculated. Total hemoglobin (Hb) and FHbF were measured by co-oximetry. Cross-sectional associations were assessed using Spearman’s rho correlations. Longitudinal trajectories and gestational age-adjusted within-infant change-on-change associations were analyzed using linear mixed-effects models.

Results

No significant cross-sectional correlations were identified between FHbF and cerebral or peripheral NIRS-derived variables at any time point. Over time, Hb and cStO₂ declined, cFTOE increased, while peripheral oxygenation and extraction remained stable. Within-infant Hb changes were not associated with concurrent cerebral NIRS changes. In contrast, FHbF change was associated with change in cStO₂ (T2–T1: β per 1% FHbF, −0.19; 95% CI −0.33 to −0.05) and cFTOE (β 0.22; 95% CI 0.06–0.38).

Conclusion

Within-infant FHbF decline was associated with increased cStO₂ and decreased cFTOE, independent of total Hb.

Introduction

Fetal hemoglobin (HbF) is crucial for fetal oxygen delivery and undergoes marked maturation-dependent changes in its fractional content (FHbF) due to the hemoglobin switch and synthesis of adult hemoglobin (HbA). Owing to its higher oxygen affinity compared to HbA, HbF is optimally suited for oxygen transport in the relatively hypoxic intrauterine environment [1, 2]. The gradual switch from HbF to HbA begins around the 30th week of gestation and progresses at an individual rate for each neonate, independent of the birth event [3, 4]. This transition shifts the oxygen dissociation curve (ODC), thereby modulating tissue oxygen availability [5, 6].

Near-infrared spectroscopy (NIRS) enables noninvasive monitoring of regional tissue oxygenation in various body regions [7]. This technique has been widely applied for monitoring cerebral and peripheral oxygenation in neonatal intensive care units (NICUs). Different NIRS devices employ distinct measurement principles to derive regional tissue oxygenation, either by detecting relative changes in oxy- and deoxyhemoglobin concentrations using spatially resolved spectroscopy or by quantifying their absolute concentrations with time-resolved NIRS technology [8].

Because regional tissue oxygenation reflects a dynamic balance between oxygen delivery, consumption, and perfusion, several studies have focused on physiological variables influencing NIRS-derived tissue oxygenation [9, 10]. Given shifts in the ODC, changes in FHbF may alter arterial-venous oxygen gradients and also influence NIRS-derived parameters. Although oxygenation changes during the immediate fetal-to-neonatal transition have been explored extensively [11, 12], longitudinal data beyond this period remain limited, particularly regarding the relationship between cerebral and peripheral oxygenation and the postnatal changes in FHbF [13].

This study investigated associations between HbF parameters and NIRS-derived tissue oxygenation and oxygen extraction at three predefined time points during the first 3 postnatal weeks. We further assessed longitudinal cerebral and peripheral NIRS changes and whether within-infant FHbF changes were associated with distinct oxygenation and extraction patterns in preterm and term neonates.

Methods

Study Design

This prospective observational pilot study was conducted at a single tertiary NICU between April 2020 and March 2024. The study protocol is provided in online supplementary 1 (for all online suppl. material, see https://doi.org/10.1159/000553520).

Participants

Preterm and term neonates admitted to the NICU were eligible. Exclusion criteria were known congenital malformations and a family history of hemoglobinopathies. Because of sensor-number and sensor-size limitations, infants with birth weight below 1,500 g were not approached. None received a red blood cell (RBC) transfusion during the study period. Neonates born before 37 completed weeks of gestation were classified as preterm, and those born at or after 37 weeks as term. Given the pilot design, a target sample size of 25 neonates per group was prespecified. Antenatal, perinatal, and postnatal clinical data were recorded, including gestational age (GA), birth characteristics, respiratory support and oxygen therapy, primary diagnoses, and medications.

HbF Analysis and Physiological Monitoring

NIRS measurements were synchronized with routine clinically indicated capillary blood sampling and started immediately afterward. FHbF and total hemoglobin concentration (Hb) were measured using the ABL800 Flex co-oximeter (Radiometer, Denmark). Total fetal hemoglobin concentration (HbFc) was calculated as HbFc (g/dL) = FHbF (%)/100 × Hb (g/dL).

Pulsatile arterial oxygen saturation (SpO₂) and heart rate were continuously monitored using the IntelliVue MP70 monitor (Philips, Netherlands). SpO₂ and NIRS signals were sampled at 1 Hz. Recordings with at least 60 min (≥3,600 data points) of artifact-free data were included and averaged for analysis.

NIRS Measurements and Time Points

Cerebral and peripheral NIRS measurements were recorded simultaneously during resting conditions for up to 2 h at 3 time points: T1, 24–80 h (days 1–3); T2: 120–240 h (days 5–10); T3: 288–432 h (days 12–18) after birth. cStO₂ was measured using a t-NIRS device (Hamamatsu Photonics, Japan) with the sensor applied on the left forehead. cFTOE was calculated as: cFTOE (%) = (SpO₂ − cStO₂)/SpO₂ × 100 [14].

Peripheral tissue oxygenation index (pTOI) was measured using a spatially resolved NIRS device (NIRO-200NX, Hamamatsu Photonics, Japan) with the sensor placed on the right forearm. At each time point, peripheral NIRS monitoring was combined with three venous occlusions (VO) according to a standardized protocol and prespecified quality criteria to derive peripheral fractional oxygen extraction (pFOE) [15–17]. Mean values were calculated from the 3 VOs. All monitoring signals were recorded and stored in alpha-trace (Medical Systems, Austria) for subsequent analysis.

Device Selection

The original protocol-specified t-NIRS for both cerebral and peripheral measurements, including VO. After quality control identified excessive light scattering during peripheral t-NIRS VO measurements, NIRO-200NX was adopted for subsequent peripheral measurements. Only NIRO-200NX-derived pTOI and VO data were analyzed, resulting in higher missing data rates for pTOI and pFOE.

Protocol-Specified Primary and Secondary Analyses

The original study protocol (online suppl. 1) specified the primary analysis as the assessment of correlations between FHbF and HbFc and NIRS-derived tissue oxygenation parameters at T1, T2 and T3. Protocol-defined secondary objective was to assess cohort-based and within-infant longitudinal changes in tissue oxygenation depending on FHbF.

Statistical Analysis

Primary Outcome

Continuous variables were analyzed at the 3 time points. Descriptive statistics were based on all available observations and are reported as mean (SD) or median [IQR], as appropriate. Distributional assumptions were assessed using the Shapiro-Wilk test. Spearman’s rank correlation coefficients assessed associations between primary outcome parameters at each time point.

Secondary Outcomes

Within-infant associations between changes in hemoglobin-related parameters (FHbF and Hb) and concurrent changes in oxygenation and oxygen extraction were assessed using change-on-change analyses. This approach captures individual trajectories rather than cohort-level systematic mean changes. To distinguish changes in Hb from changes in the relative contribution of HbF, FHbF and Hb were modeled as separate predictors. Because the objective was to quantify within-infant longitudinal changes, analyses were not stratified by GA, instead GA was included as a continuous covariate in all models.

Changes between time points (ΔT2–T1) were calculated for predictors (ΔFHbF and ΔHb) and outcomes (ΔSpO₂, ΔcStO₂, and ΔcFTOE). Linear regression models were fitted with ΔFHbF, ΔHb, and GA as predictors. Regression coefficients are reported with 95% CIs and Wald t test p values. Benjamini-Hochberg false discovery rate (FDR) correction was applied within each outcome.

Cohort-level longitudinal changes were analyzed using linear mixed-effects models with a subject-specific random intercept and time as a categorical fixed effect. Prespecified pairwise contrasts were T2 vs. T1, T3 vs. T2, and T3 vs. T1. Effect estimates and 95% CIs were derived using Wald tests, with q < 0.05 (FDR adjusted p) considered significant. All statistical analyses were performed using MATLAB R2024b and Python version 3.13.

Missing Data

Missing data were handled by multiple imputation when suitable for reliable imputation (<30% missing observations per parameter and time point). Estimates were pooled using Rubin’s rule. For peripheral NIRS measures and later contrasts (T3–T2 and T3–T1), missingness exceeded this threshold and was considered too high for valid imputation. Therefore, only the T2–T1 contrast for cerebral NIRS measures was included in change-on-change analyses.

Results

Study Population

Twenty-five moderate-to-late preterm (median GA [IQR], 33.6 [1.6] weeks) and 25 term neonates (median GA [IQR], 39.9 [1.4] weeks) were included. The cohort consisted mainly of clinically stable neonates without cardiovascular support and with low incidence of respiratory support. Participant inclusion is shown in Figure 1 and baseline characteristics in Table 1. Besides GA, significant between-group differences were observed in birth weight and umbilical artery pH, but not in subsequent respiratory support or supplemental oxygen use.

Fig. 1.

Study participants’ flowchart.

Table 1.

Group-specific demographic and clinical characteristics of the included neonates

Parameter	Preterm	Term	p value
N	25	25	​
Gestational age, weeks	33.6 (1.6)	39.9 (1.4)	<0.001
Birth weight, g	1,990 [1,732–2,170]	3,600 [3,400–3,990]	<0.001
Umbilical artery pH	7.28 (0.10)	7.20 (0.10)	0.007
Delivery mode
Vaginal	10 (40)	16 (64)	0.157
Female sex	9 (36)	6 (24)	0.537
APGARa score
1 min	8 [7–9]	8 [7–9]	0.378
5 min	9 [8-9]	10 [9–10]	0.120
10 min	9 [9–10]	10 [9–10]	0.256
Respiratory support n/N
T1	5/25	4/25	1.000
T2	1/22	1/16	1.000
T3	0/19	1/2	0.095
FiO2 in patients on respiratory supportb
T1	0.21 [0.21–0.24]	0.23 [0.21–0.25]	0.680
T2	0.24	0.24	1.000
T3	NA	0.22	NA

Categorical variables (delivery mode, sex) are reported as N (%). Continuous variables (gestational age at birth, birthweight, umbilical artery pH, Apgar score, FiO₂) are reported as mean (SD), median [IQR] as appropriate.

FiO₂, fraction of inspired oxygen; T1, T2, T3, time points 1, 2, 3; NA, not applicable; n/N, number of neonates receiving respiratory support among those measured at the same time point.

^aAPGAR, postnatal vitality scores range from 0 to 10, with higher scores indicating a better adaptation.

^bFiO_2, range from 0.21 to 1.0 with lower value indicating less severe pulmonary disease.

Correlation Analyses

Using Spearman’s rank correlation coefficients, no statistically significant associations were found between the HbF-related parameters (FHbF, HbFc) and cerebral or peripheral NIRS-derived oxygenation (cStO₂, pTOI) or oxygen extraction (cFTOE, pFOE) at any time point (T1, T2, T3). Results are summarized in Table 2.

Table 2.

Correlation analyses for primary outcome measures at each time point (T1, T2, T3)

Parameter	N	FHbF (%)		Hb (g/dL)		HbFc (g/dL)
		ρ	p value	ρ	p value	ρ	p value
Oxygenation
SpO2 (%) at T1	48	−0.02	0.887	−0.19	0.199	−0.11	0.468
SpO2 (%) at T2	37	0.01	0.941	−0.30	0.071*	−0.11	0.531
SpO2 (%) at T3	21	−0.59	0.005	−0.15	0.510	−0.27	0.235
cStO2 (%) at T1	48	−0.15	0.312	0.11	0.444	−0.01	0.935
cStO2 (%) at T2	36	−0.11	0.539	−0.20	0.252	−0.15	0.369
cStO2 (%) at T3	21	−0.31	0.177	0.30	0.190	0.06	0.795
pTOI (%) at T1	35	0.20	0.255	−0.15	0.399	0.01	0.968
pTOI (%) at T2	26	0.11	0.601	0.03	0.881	0.12	0.568
pTOI (%) at T3	15	−0.17	0.557	0.25	0.370	0.13	0.648
Oxygen extraction
cFTOE (%) at T1	48	0.13	0.379	−0.16	0.263	−0.01	0.943
cFTOE (%) at T2	36	0.14	0.404	0.07	0.701	0.11	0.521
cFTOE (%) at T3	21	−0.17	0.463	−0.32	0.152	−0.24	0.299
pFOE (%) at T1	35	−0.02	0.896	0.41	0.014	0.21	0.219
pFOE (%) at T2	28	−0.24	0.227	−0.14	0.464	−0.25	0.198
pFOE (%) at T3	14	−0.17	0.555	−0.26	0.376	−0.15	0.620

Bold represents statistically significant result. N, number of measurements; ρ, Spearman’s rank correlation coefficients; FHbF, fraction of fetal hemoglobin, Hb, total hemoglobin concentration; HbFc, total fetal hemoglobin concentration; SpO₂, pulsatile arterial oxygen saturation; cStO₂, cerebral regional tissue oxygenation; pTOI, peripheral tissue oxygenation index; cFTOE, cerebral fractional tissue oxygen extraction; pFOE, peripheral fractional oxygen extraction.

*Result with a trend to statistical significance.

Within-Infant Longitudinal Changes

Results of the T2–T1 change-on-change analyses are shown in Table 3 and Figure 2a, b. A larger decline in FHbF was associated with a greater increase in cStO₂ (β = −0.19; 95% CI −0.33, −0.06; q = 0.018) and a greater decrease in cFTOE (β = 0.22; 95% CI 0.08, 0.37; q = 0.018). No significant associations were observed between changes in FHbF and SpO₂, or between changes in Hb and any evaluated oxygenation or oxygen extraction parameter.

Table 3.

Within-infant change-on-change analyses for T2-T1

Parameter change (Δ)	Beta	95% CI	p value	q (FDR adjusted)
Δ FHbF on ΔSpO2	0.03	−0.09, 0.14	0.660	0.660
Δ FHbF on ΔcStO2	−0.19	−0.33, −0.06	0.006	0.018
Δ FHbF on ΔcFTOE	0.22	0.08, 0.37	0.004	0.018
Δ Hb on ΔSpO2	0.12	−0.30, 0.55	0.579	0.660
Δ Hb on ΔcrStO2	0.33	−0.12, 0.78	0.161	0.321
Δ Hb on ΔcFTOE	−0.26	−0.76, 0.24	0.317	0.476

CI, confidence interval; FDR, false discovery rate adjustment; FHbF, fraction of fetal hemoglobin; Hb, total hemoglobin concentration; SpO₂, pulsatile arterial oxygen saturation; cStO₂, cerebral regional tissue oxygenation; cFTOE, cerebral fractional tissue oxygen extraction.

Fig. 2.

Association between within-infant changes in FHbF from time point 1 (T1) to time point 2 (T2) and changes in cerebral tissue oxygenation-ΔcStO₂(a) and changes in cerebral fractional tissue oxygen extraction-ΔcFTOE (b).

Cohort-Level Longitudinal Changes

GA-adjusted time point comparisons (T2-T1, T3-T1, T3-T2) for hemoglobin parameters (Hb, FHbF), noninvasive monitoring variables (SpO₂, cStO₂, pTOI) and oxygen extraction variables (cFTOE and pFOE) are available in Figure 3 and online supplement 2, Table S1. Over the first three postnatal weeks, Hb decreased significantly across all evaluated contrasts (all q < 0.001; Fig. 3a). In contrast, FHbF showed no significant cohort-level longitudinal change (Fig. 3b). Among oxygenation parameters, cStO₂ decreased significantly over time (T2-T1, q = 0.008, T3-T1, p = 0.044; Fig. 3c). Without FDR adjustment, SpO₂ also decreased significantly between T1 and T2 (p = 0.031). No significant changes in pTOI were observed across all time points, with or without FDR adjustment. Changes in oxygen extraction were significant only before FDR adjustment, with both cFTOE and pFOE increasing over time. Cohort-level longitudinal changes in other assessed physiological variables are reported in online supplement 2, Table S2.

Fig. 3.

Cohort-based changes over time in primary outcome measures presented in box plots. a Total hemoglobin (Hb; g/dL). b Fraction of fetal hemoglobin (FHbF; %). c Cerebral regional tissue oxygenation (cStO₂; %).

Discussion

This prospective observational study examined whether HbF parameters are related to NIRS-derived tissue oxygenation in preterm and term neonates during the early postnatal weeks. Cross-sectional analyses did not identify significant associations between FHbF or HbFc and NIRS parameters at individual time points. However, longitudinal change-on-change analyses provided additional insight into within-infant adaptation. A decline in FHbF during the first two postnatal weeks was associated with increasing cStO₂ and decreasing cFTOE, whereas changes in Hb were not associated with these outcomes. This suggests that postnatal changes in hemoglobin composition may contribute to cerebral oxygenation dynamics beyond hemoglobin concentration alone.

Cerebral Oxygenation

The observed change-on-change associations between FHbF and cerebral oxygenation metrics are physiologically plausible and likely reflect heterogeneous hemoglobin transitions during the first two postnatal weeks. Although FHbF is expected to decline with maturation, trajectories were not uniform across the cohort, with some neonates showing stable or even increasing FHbF (Fig. 2b) [18]. Among neonates with declining FHbF, cStO₂ increased while cFTOE decreased, consistent with replacement of HbF by HbA and a rightward shift of the ODC, which reduces oxygen affinity and facilitates tissue oxygen unloading [1, 4, 19]. Under stable cerebral metabolic demand and preserved cerebrovascular adaptation, this shift may allow cerebral oxygen requirements to be met with a smaller arterial-venous saturation difference, resulting in higher venous-weighted cStO₂ and lower cFTOE (online suppl. 2-Fig. S1) [20]. Conversely, infants with increasing or persistently higher FHbF demonstrated lower cStO₂ and higher cFTOE, compatible with a greater extraction gradient in the setting of higher hemoglobin-oxygen affinity [3, 5]. Although FHbF did not change significantly at cohort level, within-infant changes were associated with cerebral oxygenation changes, reflecting individual postnatal adaptations rather than a uniform cohort-wide trend (online suppl. 2, Fig. S2c).

Previous studies have highlighted the role of HbF in neonatal cerebral oxygenation. During immediate fetal-to-neonatal transition in preterm neonates, lower total HbFc has been associated with lower cStO₂ and higher cFTOE [21]. In the present study, correlation analyses showed no significant associations between HbFc and either cStO₂ or cFTOE. This difference may reflect the physiological context: during immediate transition, SpO₂ is still rising and pulmonary gas exchange, cardiac output, and cerebrovascular autoregulation may not be established. In contrast, during the first two postnatal weeks, SpO₂ is usually maintained within the target range of 90–95%, and, in the context of preserved cerebral autoregulation, cStO₂ and cFTOE may not correlate with HbFc or FHbF. However, longitudinal changes in FHbF may induce shifts in the ODC and thereby influence accompanying changes in cStO₂ (online suppl. 2, Fig. S1).

Since physiological changes in FHbF and NIRS parameters have not been studied previously, our findings can only be compared with RBC transfusion studies. In a recent study, very preterm neonates receiving adult A-RBC transfusions, resulting in a FHbF decline, demonstrated higher cStO₂ and lower cFTOE compared with neonates receiving cord blood transfusions, in whom FHbF increased [22]. This aligns with our findings, although our cohort captured physiological FHbF changes without transfusion. However, without within-infant change-on-change analyses, it remains uncertain whether differences in cStO₂ can be causally attributed to ODC. Positive pre-transfusional FHbF and cStO₂ correlations in that cohort, may reflect altered cerebrovascular autoregulation in the presence of anemia and greater prematurity, in contrast to our cohort of clinically stable more mature neonates without anemia.

Although the effect size of physiological FHbF decline on cStO₂ was relatively small, with each 1% decrease in FHbF corresponding to a 0.19% increase in cStO₂ and 0.22% decrease in cFTOE, larger FHbF changes, induced by RBC transfusion, may have more clinically relevant effects. Consistently, A-RBC transfusion studies in anemic preterm neonates reported higher cStO₂ and lower cFTOE [23, 24]. Transfusions cause abrupt changes in hemoglobin concentration and composition, whereas our study reflects gradual endogenous FHbF change under relatively stable clinical conditions. Despite this difference, directionally similar NIRS-derived changes support the plausibility of FHbF-related shifts in cStO₂ and cFTOE.

The physiological relevance of FHbF for cerebral oxygenation should be interpreted cautiously. We do not suggest that FHbF is a dominant determinant of cStO₂, but rather a modifying factor within established contributors, including arterial oxygen content, Hb, cerebral blood flow, and cerebral metabolic demand [10]. NIRS-derived cStO₂ reflects the balance between cerebral oxygen delivery and consumption, not mitochondrial oxygen delivery directly. Thus, cStO₂ and cFTOE should be interpreted as markers of cerebral oxygen supply-demand balance. Although Hb declined substantially, it was not significantly associated with cStO₂ or cFTOE, suggesting that hemoglobin composition, rather than Hb concentration alone, may contribute to early postnatal cerebral oxygenation changes.

Clinical Implications

The observed associations between physiological FHbF changes and cerebral oxygenation are hypothesis-generating and may be relevant to vulnerable populations. Extremely preterm neonates often receive A-RBC transfusions, causing abrupt hemoglobin composition shifts that may exceed immature cerebrovascular autoregulatory capacity. Although causality cannot be inferred, these findings support further investigation of transfusion-related hemoglobin changes using cerebral NIRS.

Peripheral Oxygenation

In contrast to cerebral oxygenation, no significant longitudinal changes in peripheral tissue oxygenation or oxygen extraction were observed. Since change-on-change analyses were not feasible because of missing data, FHbF-related effects on peripheral tissue compartment remain unclear.

Strengths and Limitations

Strengths include the prospective design, repeated paired measurements, and within-infant analytical approach minimizing inter-individual confounding. Limitations include the modest sample size, not inclusion of infants with birth weight <1,500 g, limiting generalizability to very low birth weight, very preterm, or critically ill neonatal populations. Reduced data availability at later time points precluded later change-on-change analyses. Moreover, FHbF was estimated by capillary co-oximetry without repeat or reference-method confirmation. Therefore, preanalytical and analytical variability may have influenced individual FHbF trajectories. Finally, although HbF and HbA differ slightly in near-infrared absorption characteristics, previous neonatal data suggest that the resulting error is small and negligible [25]. Accordingly, findings should be considered hypothesis-generating and require confirmation in less stable neonatal populations exposed to the RBC transfusions.

Conclusion

In this prospective observational study, within-infant postnatal FHbF changes were associated with concurrent cerebral oxygenation and oxygen extraction changes in stable preterm and term neonates without red blood cell transfusion. FHbF decline was associated with increased cStO₂ and decreased cFTOE, whereas FHbF increase showed opposing associations. Confirmation in larger cohorts of extremely preterm neonates is needed.

Acknowledgments

The authors are grateful to Evelyn Ziehenberger, our study nurse, for her careful execution of all measurements and her unwavering commitment to maintaining high-quality standards.

Statement of Ethics

This study protocol was reviewed and approved by the Ethics Committee of Medical University of Graz, Approval No. EK 32-232 ex 19/20 and registered at ClinicalTrials.gov (NCT04802629). Written informed consent was obtained from parents of the participants prior to inclusion in the study.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

Ena Suppan and Gerhard Pichler: substantial contributions to conception and design, acquisition of data, analysis and interpretation of data. Thomas Suppan and Bernhard Schwaberger: substantial contributions to data analysis and interpretation. Christina Wolfsberger, Christoph Schlatzer, Daniel Pfurtscheller, and Melina Winkler: substantial contributions to acquisition of data and interpretation of data. Ena Suppan, Thomas Suppan, Bernhard Schwaberger, Christina Wolfsberger, Nariae Baik-Schneditz, Christoph Schlatzer, Daniel Pfurtscheller, Melina Winkler, and Gerhard Pichler: drafting the article or revising it critically for important intellectual content. Ena Suppan, Thomas Suppan, Bernhard Schwaberger, Christina Wolfsberger, Nariae Baik-Schneditz, Christoph Schlatzer, Daniel Pfurtscheller, Melina Winkler, and Gerhard Pichler: final approval of the version to be published.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

The data that support the findings of this study are not publicly available due to privacy reasons but are available from the corresponding author upon request.