Assessment of the oxygen delivery capacity is crucial for the intraoperative management of critically ill patients [1,2]. In this context, measuring the cerebral blood flow (CBF) is a prerequisite for determining the oxygen delivery to the brain [3,4]. However, as measuring CBF is challenging, CBF velocity (CBFV) is measured instead [5]. Although transcranial Doppler (TCD) monitoring enables the measurement of CBFV, its clinical application in pediatric patients is limited because of the difficulty in maintaining an optimal sampling location due to the small vessel size in these patients [6,7].

TCD monitoring has been applied in critically ill neonates and children and the intraoperative management of patients undergoing cardiac and neurosurgical procedures [6,8-12]. However, this monitoring requires specially designed equipment, and there are cases where accurate measurement is difficult due to the movement of the probe. In contrast, transfontanelle ultrasonography (TFU) requires only a standard ultrasound machine with a sector probe, allowing for the measurement of blood flow in the cerebral vessels at the desired location. Notably, the alignment of the direction of ultrasound with the direction of blood flow enhances the accuracy of velocity measurements during TFU [13-16].

Despite being a highly useful approach for anesthetic and intensive care management in pediatric patients, intraoperative point-of-care TFU remains unfamiliar to many anesthesiologists and has been limited to clinical trials. Therefore, the present review explored the procedural aspects and clinical applications of TFU.

Cerebral hemodynamics can be evaluated using two ultrasound techniques: nonduplex TCD and B-mode transcranial color-coded duplex Doppler (TCCD). TCD can analyze spectral waveforms, blood flow direction, and arterial depth to identify cerebral arteries, whereas TCCD enables the visualization of brain tissue and the circle of Willis, along with bidimensional pulsed-wave Doppler for waveform analysis [6]. TFU, as a point-of-care TCCD technique, enables the direct visualization of the brain parenchyma and cerebral arteries using B-mode, duplex, and color-coded imaging. Furthermore, it can be used to assess intracranial hemodynamics by assessing flow velocities and indices, such as pulsatility and resistance indices, of major cerebral arteries through flow tracing using the bidimensional pulsed-wave Doppler technique for waveform analysis of CBF [17].

In healthy infants, blood flows continuously through the cerebral arteries toward the brain throughout the cardiac cycle [18]. However, CBFV may fluctuate because of alterations in intracranial pressure, blood pressure, arterial carbon dioxide levels, temperature, and vascular resistance [1,19].

Additionally, various factors, including age, sex, cardiac output, anemia, partial pressure of carbon dioxide, administration of drugs that affect cerebral vasodilation, and body temperature, can also significantly affect CBFV [1,19]. Considering the importance of patient-specific factors in the assessment of CBFV, normal flow velocity values for healthy children and neonates have been established [6,20-22]. CBFV is higher in children than in adults, rising after birth until 6–8 years and then declining to approximately three-fourths of the maximal velocity by 18 years of age. In 1988, Bode and Wais established the normative values referenced in the recommended protocol for flow velocities in basal cerebral arteries in healthy children (Table 1) and healthy neonates (Table 2) [20].

Specific clinical conditions, such as anesthesia and cardiopulmonary bypass (CPB), also influence CBFV [14]. For instance, continuous monitoring of CBFV with NeoDoppler during anesthesia and surgery demonstrated a significant decrease in the end-diastolic velocity during anesthesia compared with baseline in infants with an open fontanelle undergoing noncardiac surgery [18]. TCD monitoring also revealed reduced diastolic blood flow velocity after anesthesia induction with sevoflurane in children < 6 months of age [23]. Notably, CPB during cardiac surgery in infants also alters CBF, and CBFV before, during, and after CPB differs according to the patient’s body weight (Table 3) [14].

Cerebral Doppler arterial waveforms can be analyzed by assessing the peak systolic flow velocity, end-diastolic flow velocity, and pulsatility and resistive indices, which are influenced by various physiological and pathological conditions. Among these parameters, alterations in diastolic flow primarily affect hemodynamics and may indicate reduced cerebral perfusion [17]. Based on the diastolic flow, the flow profiles can be classified as (A) normal flow, (B) increased diastolic flow, (C) decreased diastolic forward flow, and (d) missing diastolic flow, or (E) negative/retrograde diastolic flow [17,24]. Additionally, pathological conditions such as hemodynamically significant patent ductus arteriosus (PDA), asphyxia, and sepsis, may also lead to alterations in blood flow patterns [17]. For instance, diastolic flow in PDA reduces through left-to-right shunting, known as “diastolic steal,” via an open PDA (Fig. 1). Moreover, alterations in diastolic flow after asphyxia are associated with vasoparalysis, loss of autoregulation, molecular stress responses, and metabolic changes [25]. Conversely, systolic flow is less affected by intracranial factors. Instead, it reflects systemic parameters such as cardiac performance, blood volume, and the distribution and condition of larger arteries [17].

The pulsatility index (PI) reflects resistance to blood flow and is calculated by dividing the difference between peak systolic and end-diastolic flow velocities by the mean flow velocity. An elevated PI is associated with increased intracranial pressure [26]. Particularly, a PI > 2.3 (normal < 1.2) in adults corresponds with an intracranial pressure > 22 mmHg [27,28]. Resistive index (RI) is another parameter reflecting the resistance to flow and is calculated as the difference between the peak systolic and end-diastolic flow velocities divided by the peak systolic flow velocity. The reported normal intracranial RI in normal full-term neonates during the first 24 h of life is 0.73 [29]. However, an increase in diastolic flow owing to the shift from fetal to newborn circulation leads to decreased average RI values with age, from 0.77 (± 0.15) in premature infants to 0.73 (± 0.57) in full-term infants, approximately 0.6 in children aged 1 year, and 0.43–0.58 in children aged > 2 years [29,30]. An RI of 0.6–0.9 is considered normal in both full-term and premature infants, whereas an RI of < 0.5–0.6 in any cerebral vessel is associated with a poor neurologic outcome [31].

PI and RI are affected by numerous factors, including flow velocity, blood volume, congenital cardiac anomalies, and peripheral vascular resistance [32,33]. Remarkably, a single measurement of Doppler parameters in the cerebral arteries may not appropriately predict well-being, brain injury, and long-term neurodevelopmental outcomes in an infant and fetus [17,34,35]. For instance, the RI can increase with decreased diastolic flow when the intracranial pressure exceeds systemic pressure due to cerebral swelling, thus obstructing cranial flow during diastole [7]. However, PDA is often the leading cause of increased RI in neonatal intensive care units [32,36]. Therefore, the anatomy must be simultaneously considered when evaluating the CBF indices.

The reliability of Doppler imaging is affected by natural variations in CBFV over time; therefore, multiple measurements of each vessel are crucial for accuracy. This complexity undermines the utility of simple cutoff values, such as the RI, for clinical decisions. Thus, repeated Doppler examinations offer a more detailed view of cerebral perfusion and disease progression than single measurements, emphasizing the need for continuous Doppler assessments to gain a comprehensive understanding of cerebral hemodynamics and guide more effective clinical decisions [35].

Transcranial ultrasound can be performed using five main windows: transfontanelle, transtemporal, trans-orbital, sub-mandibular, and suboccipital. The present review discusses the transfontanelle approach.

Fontanelles, often referred to as “soft spots,” are prominent anatomical features of the skulls of newborns. Among these, the anterior fontanelle is commonly used as an acoustic window for evaluating cerebral vessels in neonates [7]. The mean closure time of the anterior fontanelle ranges from 13 to 24 months. Before closure, cranial ultrasound can pass through these unhindered areas of the cranium to provide high-quality images. Color Doppler ultrasound can facilitate effective visualization of vascular structures including the basilar, internal carotid artery, and anterior and middle cerebral arteries, which form the circle of Willis.

TFU requires only a mobile two-dimensional ultrasound machine capable of high-resolution real-time imaging, utilizing a 4–10 MHz phased array probe or a high-frequency linear transducer (6–15 MHz). Using this approach, the entire brain can be scanned for a comprehensive understanding of the overall condition. Imaging of all suspected lesions in both coronal and sagittal planes is crucial, along with routine Doppler imaging to visualize the major veins and arteries [7,37].

The transducer should be positioned at the center of the anterior fontanelle, with the probe marker directed towards the right side of the neonate to capture images in the coronal plane. The brain is then scanned from the frontal to the posterior parietal and occipital lobes by tilting the probe forward and backward. Obtaining symmetrical images is crucial for accurate interpretation. Furthermore, meticulous examination of localized and widespread abnormalities in the cortex, white matter, deep gray matter, and ventricles is imperative. Color Doppler imaging is valuable for visualizing the basilar, internal carotid, and middle (including perforators) and anterior cerebral arteries, as well as for evaluating major venous drainage (Fig. 2).

To perform scans in the sagittal plane, the transducer should be rotated 90° so that the probe marker points toward the infant’s face. Images are captured in the midsagittal plane and two parasagittal planes on either side. Additionally, noting the brain side being imaged is critical while scanning these parasagittal planes. The evaluation involves the midline and adjacent structures including the cingulate gyrus, corpus callosum, tela choroidea, third ventricle, cavum septum pellucidi (including Verga’s ventricle), cavum veli interpositi, cisterns, cerebral aqueduct, fourth ventricle, cerebellum, pons, and cisterna magna (Fig. 3).

The cerebral arteries and veins can also be evaluated using color Doppler imaging through the anterior fontanelle. Images of the transverse sinuses at the cerebellum level can be captured in the coronal plane by locating the circle of Willis, including the internal carotid, middle cerebral, and anterior cerebral arteries near the frontal horns of the lateral ventricles. Moreover, the probe can be tilted backward to observe the basilar artery and nearby jugular veins, and further backward to visualize the internal cerebral and thalamostriate veins.

The flow velocity and RI of the anterior cerebral artery in the sagittal plane can be measured at specific sections, typically below the genu of the corpus callosum, which are referred to as the subcallosal anterior cerebral and pericallosal arteries.

Measuring the baseline CBFV in a hemodynamically stable state is important for monitoring changes in CBF during surgery, as it can serve as a reference for consecutive comparisons. Additionally, maintaining a consistent probe angle in the direction of blood flow is crucial for comparing each measurement.

TCD monitoring and NIRS can provide continuous information on CBF and oxygenation in patients vulnerable to brain injury [40-42]. Point-of-care intraoperative TFU allows for the noninvasive confirmation of adequate CBF when NIRS values decrease. TCD monitoring shows reduced blood flow velocity, or CBF may decline due to decreased blood pressure. In particular, TFU provides a comprehensive real-time approach for monitoring CBF, offering critical information about brain perfusion noninvasively, making it especially useful for vulnerable populations like neonates and infants. However, studies on the impact of intraoperative TFU-guided management on clinical outcomes remain limited. Understanding the relationships among point-of-care TFU measurements, TCD and NIRS data, and neurological outcomes could significantly enhance our ability to predict and prevent adverse events, ultimately improving long-term neurodevelopmental outcomes.

CBF regulation during anesthesia in pediatric patients, particularly neonates, warrants further investigation. The key areas of interest include elucidating the association among blood pressure, PaCO₂, CBF, and cerebral autoregulation. Additionally, comparative studies with other noninvasive cerebral monitoring techniques such as NIRS, magnetic resonance imaging, and electroencephalography, are imperative. Additionally, the effects of pneumoperitoneum and altered CO₂ levels on CBF and the potential influence of microemboli on clinical outcomes represent critical avenues for future research. Moreover, the correlation between the intraoperative application of TFU and the long-term neurodevelopmental outcomes of physiological alterations also requires thorough evaluation.

TFU is a useful tool for evaluating CBF in pediatric patients with open fontanelles.