NeoDoppler: New ultrasound technology for continous cerebral circulation monitoring in neonates

Abstract

Background: There is a strong need for continuous cerebral circulation monitoring in neonatal care, since suboptimal cerebral blood flow may lead to brain injuries in preterm infants and other critically ill neonates. NeoDoppler is a novel ultrasound system, which can be gently fixed to the anterior fontanel and measure cerebral blood flow velocity continuously in different depths of the brain simultaneously. We aimed to study the feasibility, accuracy, and potential clinical applications of NeoDoppler in preterm infants and sick neonates.

Method: Twenty-five infants born at different gestational ages with a variety of diagnoses on admission were included. The probe was placed over the anterior fontanel, and blood flow velocity data were continuously recorded. To validate NeoDoppler, we compared the measurements with conventional ultrasound; agreement was assessed using Bland-Altman plots.

Results: NeoDoppler can provide accurate and continuous data on cerebral blood flow velocity in several depths simultaneously. Limits of agreement between the measurements obtained with the two methods were acceptable.

Conclusion: By monitoring the cerebral circulation continuously, increased knowledge of cerebral hemodynamics in preterm infants and sick neonates may be acquired. Improved monitoring of these vulnerable brains during a very sensitive period of brain development may contribute toward preventing brain injuries.

Fig. 1

The NeoDoppler research set-up. a illustrates the NeoDoppler probe. The NeoDoppler probe transmits plane waves and simultaneously receives echoes from blood vessels at different depths covered by the ultrasound beam. b demonstrates trend curves obtained during a period of 4 h in a preterm infant. The upper panel shows the trend curves for maximum (Vmax, blue), mean (Vmean, red), and end-diastolic velocity (VED, yellow); panel in the middle shows pulsatility index (PI, blue) and heart rate (HR, red); the lower panel shows the quality of the measurements (red). c shows depth-vs-time color M-mode (upper panel), which represents blood flow in several vessels at different depths for a duration of 30 s. The lower panel shows pulsed-wave Doppler velocity waveforms for a specific depth. The green marker in the upper panel, representing the sample volume (SV), indicates the depth where the curve in the lower panel is obtained. One can change the size and position of the sample volume with controls labeled SV depth and SV size. The SV width can be set to cover the artery of interest in each patient. d shows the frequency spectrum where the oscillations in the pulsed-wave Doppler spectrum are represented. The frequency scale, x axis, is presented as oscillations/min and the corresponding amplitude in decibels, y axis, is relative to the mean value of the velocity trace. In this case, the HR of the infant contributes to the major peak at 115/min. The respiratory frequency contributes to the peak at 30/min. The peak at the low frequencies, about 5–7/min, represents the slow observed oscillations in the velocity spectrum, in this case with an amplitude of −35 dB

Fig. 2

Agreement between NeoDoppler and conventional ultrasound for maximum velocity (a) and pulsatility index (b) in all study subjects

Fig. 3

Velocity measurements at different depths simultaneously in a term infant with corresponding frequency spectrums. The figure shows 30 s selected from 6 h of recording with NeoDoppler showing Doppler curves in five different vessels at different depths in a term infant treated with therapeutic hypothermia for hypoxic–ischemic encephalopathy and on mechanical ventilation. The corresponding frequency spectrums at each depth are demonstrated to the right and each of them shows one major peak corresponding to the heart rate of the infant (100/min) and one peak at low frequency (2–5 oscillations/min), which represent the observed slow periodic flow oscillations seen in the Doppler velocity curves. Depths in cm, 1: 1.2–1.6, 2: 1.7–2.2, 3: 2.4–2.7, 4: 2.7–3.1, 5: 3.1–3.5

Fig. 4

Observed oscillations in the Doppler velocity curves during 30 s of recording with corresponding frequency spectrums. The green marker in the color M-mode represents the sample volume at the depth where the Doppler velocity spectrum is obtained. a Cerebral blood flow without oscillations in a vessel at depth 2.2–2.6 cm in one late preterm infant with neonatal sepsis and intestinal perforation, recorded with NeoDoppler postoperatively at the time when the infant was considered hemodynamically unstable. The corresponding frequency spectrum has one major peak at about 135/min corresponding to the heart rate of the infant. There are no other major peaks in the frequency spectrum. b Cerebral blood flow with oscillations in a vessel at depth 1.2–1.5 cm in a term infant with infection. The corresponding frequency spectrum has one peak corresponding to the heart rate of the infant (110/min) and one peak at low frequency (5–7/min), which represents the observed slow periodic flow oscillations seen in the Doppler velocity curves

Fig. 5

Simultaneous velocity measurements of arterial and venous signals. The green marker in the color M-mode, representing the sample volume, indicates the depth where the Doppler velocity curves are obtained and covers both the arterial and the venous signals so that both signals are represented in the Doppler velocity curves. a Measurements of 7 s where the arterial (red) and venous (blue) signals can be measured simultaneously in a term infant with transient tachypnea of the newborn. The pulsed-wave Doppler signal demonstrates no pulsation in the venous signal. b Measurements of 30 s where arterial (red) and venous (blue) signals are measured simultaneously in a late preterm infant with gastroschisis, recorded with NeoDoppler postoperatively. The pulsed-wave Doppler signal demonstrates venous pulsation

Fig. 6

Continuous recording of the pulsatility index (PI) at different depths simultaneously. a Two hour recording in one late preterm infant. Depth 1 (1.5–2 cm): Mean PI 0.82 (SD 0.06). Depth 2 (2.5–3.1 cm): Mean PI 0.84 (SD 0.06). b One hour recording in one very preterm infant. Depth 1 (1–1.5 cm): Mean PI 0.66 (SD 0.03). Depth 2 (1.5–2 cm): Mean PI 0.77 (SD 0.03). c Two and a half hour recording in a term infant treated with therapeutic hypothermia for neonatal encephalopathy and on mechanical ventilation. Depth 1 (1.7–2.1 cm): Mean PI 1.35 (SD 0.13). Depth 2 (2.7–3.5 cm): Mean PI 1.44 (SD 0.16). Depth 3 (3.9–4.2 cm): Mean PI 1.6 (SD 0.10). d Six hour recording in a term infant treated with therapeutic hypothermia for neonatal encephalopathy and on mechanical ventilation. Depth 1 (1.3–1.9 cm): Mean PI 0.96 (SD 0.09). Depth 2 (1.9–2.2 cm): Mean PI 1.07 (SD 0.13). Depth 3 (2.3–3 cm): Mean PI 1.14 (SD 0.12)