This was a retrospective observational study conducted at the Neonatal Intensive Care Unit (NICU) of Kagoshima City Hospital, Japan, including neonates admitted between January 2000 and October 2022. Eligible neonates were those diagnosed with HIE who received therapeutic hypothermia and underwent umbilical venous catheter (UVC) placement with appropriate tip positioning for CVP monitoring as part of routine clinical care (2–4). Although informed consent for the use of clinical data was obtained prospectively as part of institutional policy, the research hypothesis and analytical framework were developed retrospectively. Therefore, this study does not conform to a classical case–control design. Instead, the primary objective was to descriptively evaluate CVP dynamics under specific clinical conditions (e.g., high MAP, presence of PPHN, administration of iNO or ECMO), without predefined case or control assignments. This study was approved by the Ethics Committee of Kagoshima City Hospital (Approval No. 2024-06) and conducted in accordance with the Declaration of Helsinki. All included neonates completed therapeutic hypothermia and were discharged.

This study included term or near-term neonates (gestational age ≥ 36 weeks) who were diagnosed with HIE and underwent therapeutic hypothermia in accordance with institutional protocols. All patients had UVCs placed with radiographically confirmed tip positions at the junction of the inferior vena cava and right atrium (T7–T9 level), and CVP was continuously monitored as part of standard intensive care. Patients were excluded if they had major congenital anomalies (such as congenital diaphragmatic hernia, abdominal wall defects, or hydrops fetalis), structural heart disease confirmed by echocardiography, positive blood culture at admission, or if the UVC tip was not appropriately positioned for accurate CVP measurement. Eighteen neonates met the inclusion criteria and were enrolled in the study. All patients required respiratory support and completed therapeutic hypothermia. PPHN was diagnosed based on clinical symptoms and echocardiographic evidence of elevated pulmonary arterial pressure and septal flattening, after ruling out structural heart disease (5). The classification of patients into high and normal MAP groups, and the presence or absence of PPHN, were analyzed retrospectively based on clinical and ventilatory parameters collected during hospitalization (6, 7).

The UVC was introduced via the umbilical vein, then navigated through the left portal vein and ductus venosus, passing either the middle or left hepatic vein before ultimately entering the inferior vena cava (IVC) and terminating in the right atrium. The ideal catheter tip position was confirmed by radiography to be at the junction of the IVC and the right atrium, typically between the T7 and T9 vertebral levels (8). Only those patients with appropriate UVC positioning were included in the study. In full-term neonates with sufficient vessel caliber, a DLP aortic cannula (Medtronic, Minneapolis, MN), originally designed for arterial use, was inserted cranially into the right internal jugular vein and used solely as a drainage cannula for VV-ECMO. The cannula's size was well-suited to the vessel diameter and allowed stable drainage flow. Additionally, a double-lumen catheter (Nipro Corporation, Osaka, Japan) was inserted into the right internal jugular vein and advanced into the internal jugular sinus, serving as both the drainage and return routes. CVP was measured after the VV-ECMO pump flow rate was set to 20 ml/kg/min, as this low flow minimized circuit-related suction effects and allowed for a more physiological and comparable measurement of central venous pressure across patients.

All neonates were intubated and managed with mechanical ventilation. Initial ventilatory support was provided using the synchronized intermittent mandatory ventilation (SIMV) mode. Respiratory acidosis, defined as arterial blood gas findings of pH < 7.2 and PaCO2 > 65 mmHg, persisted despite SIMV settings with a peak inspiratory pressure of 20–21 cmH₂O, positive end-expiratory pressure of 6 cmH₂O, and a respiratory rate of 60–65 breaths/min. If these settings failed to reduce PaCO2 by >10% or FiO2 by >20% within 1 h, the mode was switched to rescue high-frequency oscillatory ventilation (9, 10). In high-frequency oscillatory ventilation, the MAP was initiated at 2–3 cmH₂O above the SIMV setting. The oscillation frequency was set at 12–15 Hz, with an inspiration-to-expiration ratio of 1:1. If adequate PaCO2 clearance could not be achieved at the maximum amplitude, the frequency was reduced to 10 Hz (11). Adjustments were made as needed to optimize CO2 clearance. A high MAP was defined as ≥10 cmH₂O.

CVP was continuously measured using a bedside monitor (CSM-1901; Life Scope G; Nihon Kohden, Tokyo, Japan). The collected data included heart rate (HR) and systolic, mean, and diastolic blood pressure, along with CVP values. Measurements were recorded by continuous monitoring 2–3 h after catheter placement. The mean CVP value was calculated as the average of all continuously recorded CVP data points collected at 1-minute intervals over a 2–3 h monitoring period following catheter placement. CVP values were continuously monitored before and 1 h after initiation of iNO or VV-ECMO. All neonates were monitored using standardized ventilatory and sedation protocols to minimize variability in CVP measurements.

Categorical variables were analyzed using the chi-square or Fisher's exact test, as appropriate. For continuous variables, nonparametric tests were used based on the type of comparison: the Mann–Whitney U test was applied to compare two independent groups, while the Wilcoxon signed-rank test was used for paired comparisons, such as pre- and post-iNO or ECMO initiation. Normality was assessed using the Shapiro–Wilk test, which is more appropriate for small sample sizes. All variables deviated from a normal distribution; therefore, nonparametric tests were applied. Statistical significance was set at p < 0.05 with a 95% confidence interval.

The UVC used in this study was included as part of routine clinical care and not specifically for research purposes. The study protocol was approved by the institutional review board of Kagoshima City Hospital and was conducted in accordance with the Declaration of Helsinki and the National Guidelines for Human Research Ethics.

In mechanically ventilated patients, high MAP was associated with higher CVP values in our cohort. This may be due to lung overdistension increasing intrathoracic pressure, which could reduce venous return and raise CVP. However, we did not measure right ventricular afterload or cardiac output, so this mechanism remains speculative (12). As shown in Figure 1, high MAP management significantly increased the mean CVP (6 mmHg in the high MAP group vs. 5 mmHg in the normal MAP group, p = 0.03), suggesting that elevated intrathoracic pressure may contribute to impaired venous return and increased CVP.

In neonates with both PPHN and high MAP, CVP was higher than in those without PPHN. One possible explanation is that PPHN may increase pulmonary vascular resistance, leading to elevated right ventricular pressures and CVP (Figure 2) (13). However, right ventricular function and pulmonary arterial pressure were not directly assessed in this study.

A significant reduction in CVP was observed after iNO administration. While we did not assess pulmonary vascular resistance or right ventricular function directly, this reduction may reflect improved hemodynamics associated with pulmonary vasodilation. Further studies with comprehensive cardiac assessments are needed to confirm this hypothesis (Figure 3) (14).

In VV-ECMO, venous blood is drained from the right atrium and oxygenated blood returns to the venous system, effectively replacing the function of the lungs. This allows for the weaning of ventilatory settings and transition to lung-protective ventilation strategies. VV-ECMO may reduce intrathoracic pressure through lung-protective ventilation, which could theoretically improve right ventricular preload and afterload (15). Despite this, our data showed an increase in CVP after ECMO initiation, which may reflect fluid administration, sedation effects, or other hemodynamic shifts. Therefore, while CVP reduction might be expected, our findings suggest a more complex interplay. The observed CVP increase after VV-ECMO initiation is likely due to combined factors, including increased preload from volume resuscitation, myocardial depression due to disease severity or sedation, and possibly direct influence of return cannula flow on the CVP monitoring site. Although return flow in VV-ECMO is generally not expected to increase CVP if the cannula is appropriately positioned, the mechanical interaction between return flow and catheter tip may affect pressure readings, especially in neonates with small vascular structures (Figure 4).

Left ventricular contraction generates the pressure needed to push blood through the capillaries and into the venous system. In this study, LVEF was assessed using M-mode echocardiography; however, this method has inherent limitations. All patients exhibited mildly reduced cardiac motion due to HIE. Nonetheless, initial echocardiography performed after catecholamine and sedative administration showed an EF of ≥45% in all cases, indicating that no patients had heart failure with reduced EF. Furthermore, comparisons between neonates with high MAP and those with normal MAP, as well as those with high MAP with or without PPHN, revealed no significant differences in the EF. CVP measurements were obtained under stable conditions 2–3 h after admission, following catecholamine and sedative administration. Therefore, vis a tergo was considered equivalent across the study groups.

The function of the heart, particularly the right atrium and ventricle, facilitates the return of blood from the venous system. In this study, right atrial and ventricular functions were not specifically evaluated. However, all patients had HIE and were assessed at comparable time points. Furthermore, there were no significant differences in the use of catecholamines (dopamine, dobutamine, and adrenaline), sedatives, or analgesics among the groups. Patients with PPHN were excluded from comparisons between the high and normal MAP groups. Therefore, no significant differences in right heart function were observed, suggesting comparable conditions between the groups. In contrast, comparisons involving high MAP with or without PPHN suggested that the presence of PPHN influences right heart function, contributing to an elevation in CVP.

External pressures, including tissue compression and respiratory factors like intrathoracic or intra-abdominal pressure, can affect venous blood flow and central venous pressure. In this study, all neonates were term births without congenital anomalies or hydrops, and blood cultures obtained at admission were negative, ruling out an infection. Furthermore, there were no significant differences in the use of catecholamines (dopamine, dobutamine, and adrenaline), sedatives, or analgesics among the groups. These factors minimized the impact of vasoconstriction, vasodilation, or inflammation on venous compliance and tone. Patients with diaphragmatic hernias, pneumothorax, meconium aspiration syndrome, intestinal perforation, necrotizing enterocolitis, or abdominal wall defects were excluded, thereby minimizing variations in external pressure from the thoracic or abdominal cavities due to underlying conditions. The effects of intrathoracic pressure resulting from respiratory support were analyzed as potential contributors to CVP by comparing the high MAP and control groups. The findings demonstrated that airway pressure influences CVP levels.

Central venous pressure can also be affected by internal conditions, including the patient's blood volume and fluid status. In this study, water balance and fluid status at the time of admission and measurement were similar across all groups, suggesting that the influence of internal volume on CVP differences was minimal.