OBJECTIVE. The purpose of this article is to describe the role of cerebral and abdominal sonography with color Doppler sonography, including assessment of multiorgan tissue perfusion, in neonates with hypoxic-ischemic injury.
CONCLUSION. Bedside sonography and color Doppler sonography of the brain and abdominal organs can provide reliable and comprehensive information in asphyxiated neonates with hypoxic-ischemic injury. This article, which includes pathologic correlation, illustrates the major sonographic findings in this critical population.
Perinatal asphyxia is a major contributor to neonatal death and morbidity. Each year, approximately 23% of the 4 million neonatal deaths and 8% of all deaths at younger than 5 years of age throughout the world are associated with signs of asphyxia . Indeed, even at referral centers in developed countries, death or moderate to severe disability occurs in 53–61% of infants diagnosed as having moderate to severe hypoxic-ischemic encephalopathy . Several randomized controlled trials have shown that therapeutic hypothermia is a neuroprotective strategy that improves death and disability in neonates with moderate to severe hypoxic-ischemic encephalopathy, and this treatment has been adopted as the standard of care by most centers around the world .
During asphyxia, gas exchange between the fetus and the placenta is compromised, resulting in fetal hypoxia, hypercarbia, and acidosis. Subsequently, a marked redistribution of blood flow occurs, with an increase in the flow to the brain, heart, and adrenal glands and a decrease in the blood flow to the kidneys, bowel, and skin. Indeed, multiple-organ failure has been reported in 50–60% of neonates with severe perinatal asphyxia . If this process lasts long enough, cerebral blood flow decreases by a combination of abnormal cerebral autoregulation and systemic hypotension, leading to cerebral hypoperfusion and subsequent hypoxicischemic injury . During the postasphyxiated period, an increase in cerebral blood flow may unfold, with the onset generally within the first few hours of life and duration of many hours or days. This chain of events is called reperfusion or the hyperemic phase and is responsible for secondary brain injury.
In neonates with hypoxic-ischemic encephalopathy, high levels of cerebral blood flow measured at 12–24 hours of life have been associated with more severe brain injury . A substantial proportion of asphyxiated infants (35–85%) exhibit predominantly cerebral deep nuclear neuronal involvement , and injury to these areas has been associated with unfavorable neurologic outcome [5, 6]. Therefore, measurements of brain perfusion with dynamic color Doppler sonography during this period may provide information that correlates with reperfusion injury.
Dynamic Color Doppler Sonography for Tissue Perfusion Measurements
Over the past few years, we have used a recently developed software program (Pixelflux, Chameleon-Software) to dynamically quantify the color Doppler signals and obtain tissue perfusion measurements, specifically of the brain and bowel . This color Doppler quantification method provides dynamic blood flow data (during the cardiac cycle) and perfusion velocity in a chosen region of interest (ROI) from a standard color Doppler video, without IV contrast administration. Tissue perfusion is quantified, taking into consideration the amount of blood flow through a specific tissue during a complete cardiac cycle and therefore reflecting the differences between systolic and diastolic perfusion in the small vessels of the specified area. The average values of the flow velocity and area inside the ROI are measured during the cycle and used for the calculation of tissue perfusion intensity (PI):where v is the mean velocity of pixels, A is the area of all color pixels and A ROI is the area of the ROI.
All this calculation is done automatically for images encompassing one full heart cycle, which is also detected automatically by the software .
For all studies, the color Doppler parameters are standardized and kept constant for comparison between studies (color gain, 40; scale, 7.5 cm/s). Color Doppler videos with major motion artifacts are excluded, and the ROI is only determined in areas without artifacts through-out the duration of the video. All neonates who have congenital heart disease or are hemodynamically unstable are excluded. During these studies, it is critical to obtain information on several physiologic parameters, such as oxygenation (SpO2 or Pao2), carbon dioxide levels (Pco2 or TcPco2), blood pressure (systolic, mean, and diastolic), and body temperature (skin and esophageal). Information on the use of medications, such as inotropes, vasoconstrictors, vasodilators, and sedatives, is also important because it can affect tissue perfusion. Clinical or electroencephalography seizures should also be recorded.
In asphyxiated neonates, brain monitoring and assessment are usually done by electroencephalography and MRI [5, 8]. However, ultrasound is an attractive tool given its portability and lack of ionizing radiation. In this article, we will illustrate the most common findings of sonography performed at the bedside to assess the brain and abdominal organs in asphyxiated neonates. We will also describe our experience with the use of dynamic color Doppler sonography. Although the clinical usefulness of the information obtained with dynamic color Doppler sonography has not yet been established in this population, such tissue perfusion measurements have been applied in many different settings with positive and encouraging results. Indeed, tissue perfusion measurements using dynamic color Doppler sonography have been proven useful to describe local inflammatory activity in bowel segments affected by Crohn disease in pediatric patients , assess perfusion of transplanted kidneys , and differentiate stages on metastatic lymph nodes .
As indicated previously, MRI is the standard imaging modality of the brain in neonates with perinatal asphyxia because it provides anatomic and functional information that help determine the severity of the disease and the prognosis. It depicts different patterns of injury, such as watershed injury or involvement of basal ganglia and thalami, as seen in the more severe cases. Although head sonography is thought to be less accurate than MRI, in a recent study, a good correlation between studies and MRI of the brain parenchyma was shown in the assessment of hypoxic-ischemic injury . This suggests that head sonography may be a more effective modality than previously described. Moreover, head sonography remains an excellent screening tool for use in neonates too critically ill to be transported to the MRI suite.
In our institution, head sonography is performed in neonates with hypoxic-ischemic encephalopathy, using a 9S4-MHz sector transducer. The most common head sonography findings in neonates with hypoxic-ischemic injury are brain swelling with echogenic subcortical white matter (Fig. 1A), increased cerebral echogenicity with or without loss of gray-white matter differentiation, and basal ganglia involvement (Fig. 2A). Intraventricular bleed (Figs. 3A, 3B, and 3C), although uncommon, has also been described in term neonates with hypoxic-ischemic injury. Head sonography can also assess pulsed Doppler flow velocities and the resistive index of the cerebral arteries .
During the head sonography examination, we also perform dynamic color Doppler sonography (color gain, 40; scale, 7.5 cm/s) with an 11LW4-MHz linear transducer. DICOM color Doppler videos of the basal ganglia blood flow are obtained and recorded in the coronal plane (Figs. S1B and S2B, supplemental videos, can be viewed from the information box in the upper right corner of this article). These videos are later analyzed using dedicated software  to quantify the cerebral perfusion intensity of the area (Figs. 1A, 1B, 1C, 1D, 2A, 2B, 2C, 2D, 2E, and 2F). Data obtained with this technique are currently under investigation, with encouraging results, as a potential marker of reperfusion injury (Faingold R et al, presented at the 2011 annual meeting of the International Paediatric Radiology Congress).
Perinatal asphyxia may also have devastating consequences to the gastrointestinal tract, and gastrointestinal dysfunction has been described in 29% of neonates with perinatal asphyxia . Sonographic evaluation of the intestinal tract can provide information on intestinal appearance, blood flow velocities, and mural perfusion. At our institution, gray-scale abdominal sonographic images of the bowel are acquired with linear transducers ranging from 11 to 18 MHz. The spectrum of sonographic findings varies with the severity of disease and includes normal bowel wall echotexture, bowel wall edema, and presence of sloughed mucosa. Sloughed mucosa is defined as the presence of a halo or echogenic material with echotexture similar to the mucosa within the intestinal lumen (Figs. 4A, 4B, 4C, and 4D).
In severe hypoxic-ischemic injury, a significant decrease in mean blood flow velocities and increase in the resistive index were reported in the superior mesenteric artery using pulse Doppler evaluation . However, end-organ involvement has not been described. Tissue perfusion is a crucial prerequisite for normal function; therefore, quantification of bowel perfusion is important in the assessment of these critical patients. Indeed, intestinal mural perfusion patterns have been described before in patients with Crohn disease  and neonates with necrotizing enterocolitis  using color Doppler sonography and dynamic color Doppler sonography. Intestinal perfusion findings of these studies were classified as preserved intramural perfusion, bowel hyperemia, or decreased bowel perfusion (Figs. 5A, 5B, 5C, 5D, and 5E). During abdominal sonography, we have also obtained DICOM color Doppler videos of the mural blood flow (Figs. S5A and S5D, supplemental videos, can be viewed from the information box in the upper right corner of this article) using an 11LW4-MHz linear transducer (color gain, 40; scale, 7.5 cm/s) to assess intestinal perfusion intensity (Cassia G et al, presented at the 2011 annual meeting of the International Paediatric Radiology Congress). These data are also currently under investigation to evaluate whether accurate assessments of bowel perfusion in asphyxiated infants can help in staging the disease and understanding the mechanisms of intestinal autoregulation.
Perinatal asphyxia is one of the most common causes of acute kidney injury in neonates. The prevalence range has been reported between 30% and 56%, probably an underestimation given the limitations in the diagnostic criteria . Acute kidney injury may develop with or without oliguria or increase in serum creatinine levels. Imaging is not routinely used; however, gray-scale sonography may show parenchymal hyperechogenicity and loss of corticomedullary differentiation. Decreased blood flow velocity in the renal artery with pulse Doppler imaging has been described in severe hypoxic-ischemic injury  (Figs. 6A, 6B, 6C, 7A, and 7B).
Adrenal swelling and thickening have been described in neonates with asphyxia and other causes of perinatal stress . Sonographically, they may be enlarged or may loose their central echogenic stripe (Fig. 8A). Congestion and depletion of cortical lipids are the typical histologic changes of perinatal asphyxia.
Adrenal hemorrhage is relatively uncommon in neonates (Figs. 8B and 8C) but has been described in association with asphyxia, birth trauma, septicemia, and bleeding diathesis. The incidence ranges from approximately 1.7 per 1000 of autopsied neonates to approximately 3% in abdominal ultrasound studies .
Perinatal asphyxia is known to be a possible cause hepatic injury. However, the real incidence of liver injury is not well established because studies have used different definitions on the basis of abnormalities of liver enzymes or autopsy findings. Histologic changes in the liver are seen only with the most severe degrees of asphyxia. One study reported hepatic injury in up to 39% of neonates with asphyxia . Imaging of the liver does not play an important role in the evaluation of these neonates. The liver is usually homogenous or may show geographic hyperechogenic areas (Figs. 9A and 9B).
Sonography of the brain and abdominal organs can provide reliable and comprehensive information in asphyxiated neonates with hypoxic-ischemic injury. Dynamic color Doppler sonography is a simple bedside technique and a promising tool to be used in the assessment of multiorgan perfusion injury, monitoring the response to several drugs or interventions, and helping with the prediction of long-term outcomes in asphyxiated neonates. It also may provide the necessary information to improve the understanding of the changes in cerebral and visceral perfusion occurring in these infants over time.