This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (9)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wintermark, M.
Right arrow Articles by Gudinchet, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wintermark, M.
Right arrow Articles by Gudinchet, F.
Related Collections
Right arrow Neurology & Psychiatry
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?
PEDIATRICS Vol. 113 No. 6 June 2004, pp. 1642-1652

Brain Perfusion in Children: Evolution With Age Assessed by Quantitative Perfusion Computed Tomography

Max Wintermark, MD*, Domenico Lepori, MD*, Jacques Cotting, MD{ddagger}, Eliane Roulet, MD{ddagger}, Guy van Melle, PhD§, Reto Meuli, MD, PhD*, Philippe Maeder, MD*, Luca Regli, MD||, Francis R. Verdun, PhD, Thierry Deonna, MD{ddagger}, Pierre Schnyder, MD* and François Gudinchet, MD*

* Department of Diagnostic and Interventional Radiology, University Hospital, Lausanne, Switzerland
{ddagger} Department of Pediatrics, University Hospital, Lausanne, Switzerland
§ Biostatistics Unit, University Institute of Social and Preventive Medicine, Lausanne, Switzerland
|| Department of Neurosurgery, University Hospital, Lausanne, Switzerland
University Institute of Applied Radiophysics, Lausanne, Switzerland


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Objective. The objective of this study was to assess the age-related variations of brain perfusion through quantitative cerebral perfusion computed tomography (CT) results in children without brain abnormality.

Methods. Brain perfusion CT examinations were performed in 77 children, aged 7 days to 18 years. These patients were admitted at our institution for both noncontrast and contrast-enhanced cerebral CT. Only children whose conventional cerebral CT and clinical/radiologic follow-up, including additional investigations, were normal were taken into account for this study (53 of 77).

Results. The average regional rCBF amounts to 40 (mL/100 g per minute) for the first 6 months of life, peaks at ~130 (mL/100 g per minute) at ~2 to 4 years of age, and finally stabilizes at ~50 (mL/100 g per minute) at ~7 to 8 years of age, with a small increase of rCBF values at ~12 years of age. The rCBF in the gray matter averages 3 times that in the white matter, except for the first 6 months of life. The global CBF represents 10% to 20% of the global cardiac output for the first 6 months of life, peaks at ~55% by 2 to 4 years of age, and finally stabilizes at ~15% by 7 to 8 years of age. Specific age-related evolution patterns were identified in the different anatomic areas of the cerebral parenchyma, which could be related to the development of neuroanatomic structures and to the emergence of corresponding cognitive functions.

Conclusions. Quantitative perfusion CT characterization of brain perfusion shows specific age variations. Brain perfusion of each cortical area evolves according to a specific time course, in close correlation with the psychomotor development.

Key Words: child • brain maturation • brain perfusion • psychomotor development • CT • perfusion CT

Abbreviations: MR-PWI, magnetic resonance perfusion-weighted imaging • CT, computed tomography • SPECT, single photon emission computed tomography • PET, positron emission tomography • CBF, cerebral blood flow • MTT, mean transit time • rCBV, regional cerebral blood volume • rCBF, regional cerebral blood flow • ROI, region of interest • BSA, body surface area • ANOVA, analysis of variance • MRI, magnetic resonance imaging

Human brain maturation is incomplete at birth and goes on in the first years of life. During development, the brain undergoes sequential anatomic, functional, and organizational changes necessary to accommodate the complex adjusting (behavior) of a fully mature normal individual. The time-related evolution of the regional brain perfusion and metabolism during childhood may correlate with the development of neuroanatomic structures and the emergence of the corresponding cognitive functions.1–5 The time course of regional brain perfusion and metabolism does indeed match the phenomena of polyneuronal innervation6–10 and elective synaptic stabilization,10–16 known to occur in the period of postnatal brain growth.

Assessment of brain perfusion and its age-related changes during childhood has been addressed since the beginning of brain perfusion imaging.17–19 Several methods have been proposed: Doppler-sonography,20,21 magnetic resonance perfusion-weighted imaging (MR-PWI),22 stable xenon computed tomography (CT),23 single photon emission computed tomography (SPECT) with 133-xenon,5,24,25 99mTc-hexamethylpropylene amine oxime or 99mTc-ethyl cysteinate dimer,26–28 and 15CO2 positron emission tomography (PET).29 Fluorine 18- fluorodeoxyglucose PET has been used to evaluate regional cerebral metabolic rate,3,4 closely related to the local cerebral blood flow (CBF), at least in normal conditions.30

Perfusion CT has recently been implemented in the adult population. It can easily be performed as a complement to conventional noncontrast and contrast-enhanced cerebral CT. It has gained recognition in the early management of acute adult stroke patients and in other cerebrovascular disorders,31–34 because it affords a direct insight into cerebral infarct and penumbra.33,34 Finally, perfusion CT results have been validated as quantitatively accurate in adult patients by comparison with stable xenon CT35,36 and PET.37

We developed dedicated perfusion CT imaging protocols that can be applied successfully to children. The purpose of this study was to assess age-related variations of quantitative cerebral perfusion CT results in children without brain abnormality.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
From September 2001 to October 2002, 82 patients were identified prospectively in the emergency department of our institution. Inclusion criteria were all patients who were between 0 and 18 years of age and admitted for contrast-enhanced cerebral CT. Exclusion criteria were all conventional contraindications to iodinated contrast material, as well as restless children. Patients who were admitted for noncontrast cerebral CT only were not eligible for the present study. This study protocol was approved by the ethical committee for research of the Lausanne University. The principles of the Helsinki Declaration were applied.

Imaging Protocol
A unique perfusion CT series was performed immediately after the noncontrast cerebral CT. Four different imaging protocols were used, dedicated to 4 age categories: 0 to 6 months, 6 to 12 months, 1 to 10 years, and 10 to 18 years of age. These imaging protocols are summarized in Table 1. Low-osmolarity nonionic iodinated contrast material (Accupaque 300; Nycomed, Oslo, Norway) was administered intravenously into a peripheral vein using a power injector. Multidetector-row CT technology (Lightspeed CT Unit; General Electric, Milwaukee, WI) afforded the assessment of 2 adjacent 10-mm-thick slices. Two 10-mm-thick slices were preferred to 4 adjacent 5-mm-thick slices, which would have meant lower signal-to-noise ratio for the same acquisition parameters. The 2 studied cerebral slices were selected above the orbits to protect the lenses, running through the basal nuclei, then toward the vertex. They were matched with 2 noncontrast cerebral CT slices. The selection was guided by the clinical history as well as the neurologic examination findings.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Perfusion CT Protocols Dedicated to 4 Age Categories

 
Data Processing
Perfusion CT data consisted of time-contrast enhancement curves registered in each pixel, linearly related to the time-concentration curves for the iodinated contrast material (Fig 1) . These data were filtered both spatially and along with time. The perfusion CT data were analyzed with a perfusion CT software developed at our institution,38,39 the results of which have been validated by comparison with stable xenon CT.35 This software relies on the central volume principle, which is the most accurate for low injection rates of iodinated contrast material.40 The central volume principle uses a mathematic operation called deconvolution (least mean square deconvolution in the case of our software) to calculate a mean transit time (MTT).41–43 This operation requires a reference arterial input function, the selection of which is automatically performed by the perfusion CT software in a region of interest drawn by the user around the anterior cerebral artery. The regional cerebral blood volume (rCBV) map is inferred from a quantitative measurement of the partial size averaging effect, which is absent at the center of the large superior sagittal venous sinus.44–46 Hematocrit measured in each child is taken into account. Finally, a simple equation combining rCBV and MTT values leads to the regional CBF (rCBF) value: rCBF = rCBV/MTT.41–43


Figure 1
View larger version (17K):
[in this window]
[in a new window]
  		Fig. 1. Time-concentration curves extracted from a perfusion CT series in the anterior cerebral artery, in a parenchymal pixel, and at the center of the superior sagittal venous sinus, respectively. Contrast enhancement in Hounsfield units is considered with respect to the time expressed in seconds. The global contrast enhancement, featured by the areas under the curves, is higher in the venous pixel than in the arterial pixel and lower in the parenchymal pixel. This global contrast enhancement is directly related to the percentage of blood in the considered pixel and thus affords the calculation of an rCBV value. The parenchymal time-concentration curve compared with the arterial one, and the venous time-concentration curve compared with the parenchymal one show more and more flattened profiles as well as more and more delayed peak with regard to time. The time delays among the different curves depend on the transit time of the contrast material in the capillary network. It thus affords the calculation of a MTT value for each pixel. Finally, a simple equation combining rCBV and MTT values leads to the rCBF value: rCBF = rCBV/MTT.

 
Data Analysis
The same systematic analysis scheme was applied for each of the maps extracted from perfusion CT data, describing rCBV, MTT, and rCBF values, respectively. Polygonal regions of interest (ROIs; 100-200 mm2) were chosen in the gray and white matter, in the frontal, temporal, parietal, and occipital lobes, for both hemispheres. ROIs were also obtained in bilateral central cortical regions; in bilateral calcarine regions; and in bilateral caudates, lenticulate nuclei, and thalami. Placement of the ROIs in a typical patient is demonstrated in Fig 2. The size of the ROIs (100-200 mm2) was chosen in agreement with previous reports35 demonstrating such ROIs to afford a size-independent assessment of the perfusion CT results. The polygonal shape of the ROIs was adapted to fit the shape of the structure to be evaluated at best. Large vessels were avoided when drawing the ROIs in the cortical gray matter. For rCBV, MTT, and rCBF, an average value was obtained for each hemisphere, as well as a global average value for the whole cerebral parenchyma displayed on the 2 perfusion CT slices.


Figure 2
View larger version (97K):
[in this window]
[in a new window]
  		Fig. 2. Example showing the placement of the ROIs, in the gray (black ROIs) and white (white ROIs) matter, in the frontal (1), temporal (2), parietal (3), and occipital (4) lobes, for both hemispheres. ROIs were also obtained in bilateral central (5) cortical regions; in bilateral calcarine (6) regions; and in bilateral caudates (7), lenticulate nuclei (8), and thalami (9).

 
Calculations were performed for all children as well as separately by gender. For each child, a segmentation process based on adequate thresholding was applied to contrast-enhanced cerebral CT to isolate the brain parenchyma and to calculate its volume. By combining the brain volume with the global average rCBF value measured for the whole cerebral parenchyma displayed on the 2 perfusion CT slices and by taking into account that the density of brain parenchyma is close to unity, an approximation of the global CBF could be obtained.

There is no report of cardiac output in children as a function of age. This is mainly because cardiac output in children varies depending on the body surface area (BSA) rather than on the age. The BSA can be calculated from the height and weight measured in each child according to the following formula47: BSA (m2) = (weight [kg])0.5378 x (height [cm])0.3964 x 0.024265.

The cardiac index,48 defined by the formula Cardiac Index = Cadiac Output/BSA = 4.5 (L/min per m2), is approximately constant along childhood, independent of age.

We thus first calculated the BSA in each child. We then used the constant cardiac index to deduce the global cardiac output from the calculated BSA. Finally, we calculated the ratio of the global CBF to the global cardiac output.

Statistical Analysis
From the visual assessment of the graphic representation displaying the age-related evolution of global average rCBF values, we hypothesized that the fitting of a curve must rely on a mathematic model featuring 1 peak in early childhood and possibly a second around puberty. The following general formula provides 1 peak,

Formula 1(1)
where Y is the global average rCBF value, x is age, and the subscript 1 refers to the 1 peak-model. The first factor ({alpha}1 + ß1 · x) represents growth, the second exp(–{gamma}1 · x2) represents exponential decay (thus accounting for a peak), and the last term (A) is the (adult) reference value that is obtained when x grows large (because the exponential decay reduces the first part to 0 for large x).

In an attempt to describe a possible second peak, which would reflect increased rCBF around puberty, a similar equation may be superposed.

Formula 2(1')
Here the decay is centered at {delta}2, which should be near puberty.

Thus, 2 models were fitted: the restricted 1-peak model (1), with 4 unknown coefficients, and the full 2-peak model:

Formula 3(2)
in which 8 unknown parameters had to be estimated.

The estimation of the parameters was performed with nonlinear least square fittings while the models were compared thanks to the generalized F test based on full and restricted sums of squares of errors and the corresponding degrees of freedom. Similarly, the comparisons of left versus right, gray versus white matter, and boys versus girls were performed with the generalized F test, comparing a full model in which each "group" has its own parameters to a restricted model in which both groups share the same coefficients. In the comparison of boys versus girls, the latter model clearly appears as the overall model in which no grouping is considered (n observations). In the other comparisons, the 2 groups are generated by 2 measurements on 2 regions of the brain, which are considered as approximately "independent" (2n observations). Comparisons irrespective of the evolution with age could also be performed. Regions (eg, white vs gray matter), were compared in simple paired analyses (because measurements were taken on the same individual), using a nonparametric approach: the Wilcoxon signed-rank test or the Friedman 1-way analysis of variance (ANOVA) when >2 items were compared. The Wilcoxon rank sum test was applied in the comparison of boys versus girls. The significance threshold was set at .001.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Patients
Eighty-two children were considered for inclusion in this study, from September 2001 to September 2002. Five of them were excluded because of restlessness. Among the 77 remaining, 18 underwent short-term sedation with chloral hydrate or midazolam, whereas 59 were quiet or asleep.

In these children, the clinical indications for the achievement of noncontrast and contrast-enhanced cerebral CT are as follows: 45 patients with head trauma of various severity, 12 patients with headaches, 10 patients with acute seizures, 5 patients with suspected meningitis, 2 patients after surgery for complex congenital heart diseases, 1 newborn with perinatal stroke and acute seizures, 1 patient with sickle cell anemia, and 1 patient with suspected neuromalaria. The final clinical and radiologic diagnoses, including follow-up investigations, were as follows: 53 normal, 5 cerebral contusions, 3 epidural hematomas with fractures, 1 subdural hematoma, 5 cranial fractures without juxtadural or cerebral trauma, 3 bacterial meningitis, 1 herpetic meningoencephalitis and 1 hindbrain encephalitis, 1 sickle cell anemia, 1 neuromalaria, and 3 ischemic brain lesions, including the 2 patients who underwent surgery for complex congenital heart diseases and the newborn with perinatal stroke and acute seizures.

For determining the normal age-related evolution of cerebral perfusion as assessed by perfusion CT, only children who had no family history of neurologic or psychiatric disease and had normal arterial carbon dioxide pressure and arterial oxygen pressure levels in blood gas analyses and whose conventional cerebral CT and clinical/radiologic follow-up, including additional investigations, were normal were included in this study (53 of 77, including 31 patients who were admitted for mild head trauma, 12 patients with headaches, 8 patients who were admitted for febrile seizures, and 2 who were admitted for suspected meningitis).

Our series thus finally consisted of 53 patients, 7 days to 18 years of age (28 boys and 25 girls). Their age distribution was as follows: 7 patients 0 to 6 months, 3 patients 6 to 12 months, 26 patients 1 to 10 years, and 17 patients 10-16 years (median age: 6 years; interquartile range: 1.8-9.3 years).

Imaging Examinations
All perfusion CT examinations were well tolerated, without reported side effects, notably with respect to the intravenous administration of contrast material. Minimal motion artifacts were observed in 5 children: they could easily be corrected by simple registration procedures.

Evolution of Perfusion CT Results Along With Age
Evolution along with age of global average rCBF, rCBV, and MTT values for the whole brain displayed on the 2 perfusion CT slices are represented in Figs 3, 4, and 5, respectively. The equations of the curves fitted to the global average rCBF values with respect to the age (years) are as follows:

Formula 3
for the restricted 1-peak model (r2 = 0.798, P < .001), and

Formula 3
for the full 2-peak model (r2 = 0.819, P < .001).


Figure 3
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 3. Age-related evolution of global average rCBF values.

 

Figure 4
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 4. Age-related evolution of global average rCBV values.

 

Figure 5
View larger version (13K):
[in this window]
[in a new window]
  		Fig. 5. Age-related evolution of global average MTT values.

 
No statistically significant difference could be identified between both curves (P = .262), this statement calling for confirmation in larger series. Thus, the 1-peak curve model was selected to fit the global average rCBF values related to the different age or gender categories and to the different anatomic areas. Results of all of the curve-fitting processes are summarized in Table 2.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Results of the Curve Fitting According to the 1-Peak Model [Y = ({alpha} – ß · x) · – exp (–{gamma} · x2) + A] and of the Comparisons Between the Fitted Curves (SE, r2, SSE, and df)

 
The approximation of the proportion represented by the global CBF compared with the global cardiac output is represented in Fig 6. It shows the same age-dependent pattern as the global average rCBF values. The equation of the fitting curve is as follows.

Formula 3


Figure 6
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 6. Age-related evolution of the approximate proportion represented by the global CBF with respect to the global cardiac output.

 
Variations Among Anatomic Areas
The parameters that describe the fitting curves for the left and right hemispheres are summarized in Table 2 and featured in Fig 7. rCBF values are slightly higher in the left hemisphere when compared with the right one. However, this trend was not statistically significant (P = .126). The parameters that describe the fitting curves for the gray and white matter, respectively, are summarized in Table 2 and featured in Fig 8. rCBF values in the gray matter were significantly higher than the ones in the white matter.


Figure 7
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 7. Age-related evolution of rCBF values in left and right hemispheres.

 

Figure 8
View larger version (22K):
[in this window]
[in a new window]
  		Fig. 8. Age-related evolution of rCBF values in gray and white matter.

 
The comparative analysis between different cortical regions focused on the 0- to 6-year age-bracket. The central cortex (peak amplitude = 147.5 mL/100 g per minute, time to peak = 20 months, Friedman 1-way ANOVA average rank = 4.981) is the first one to show a rise in rCBF values. It is followed by the parietal cortex (peak amplitude = 130.5 mL/100 g per minute, time to peak = 27 months, average rank = 4.585), the calcarine cortex (peak amplitude = 155.3 mL/100 g per minute, time to peak = 29 months, average rank = 4.858), the temporal cortex (peak amplitude = 142.0 mL/100 g per minute, time to peak = 29 months, average rank = 4.491), and the basal ganglia (peak amplitude = 141.8 mL/100 g per minute, time to peak = 29 months, average rank = 3.187), with the highest rCBF peak amplitude reached in the calcarine cortex. Finally, the frontal cortex (peak amplitude = 133.6 mL/100 g per minute, time to peak = 30 months, average rank = 3.170) and the occipital cortex (peak amplitude = 124.4 mL/100 g per minute, time to peak = 30 months, average rank = 2.528) in turn show a rise in their rCBF values. The differences measured between the different cortical regions proved statistically significant (P < .001) with the Friedman 1-way ANOVA rank test.

Variations According to Gender
The parameters that describe the fitting curves in boys and girls are summarized in Table 2. No significant difference could be identified between rCBF values in male and female patients (P = .150).


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Perfusion CT obtained in our series of 53 normal children demonstrated a specific evolution pattern of global average rCBF, rCBV, and MTT values with respect to age, with statistically significant differences between the different age brackets (0-6 months, 6-12 months, 1-10 years, and 10-18 years of age) and between the different cortical areas.

The major advantage of the analysis of perfusion CT results with the central volume principle lies in its assessment of 2 additional parameters that characterize brain perfusion—rCBV and MTT—whereas other brain perfusion imaging techniques (stable xenon CT and most nuclear medicine techniques) assess only rCBF. Age-related evolution of rCBV and MTT in children has never been reported in the literature so far. Their potential relationship with postnatal angiogenesis—in the first months of life, the density of the radially oriented primitive vessels decreases, whereas capillary density increases in the cortex throughout development49,50—has not been explored yet. However, absolute rCBF values calculated from perfusion CT data and their evolution with respect to age are in agreement with previously reported values obtained with different imaging techniques: N2O method,18 SPECT5,24,25,27 and 15CO2 PET.29 They are also paralleled by the age-related evolution of glucose metabolism as measured by PET.3,4

Absolute rCBF values are lower at birth than in adulthood because of high rCBV and even higher MTT values as shown by perfusion CT. Their evolution along with age describes a peak at 2 to 4 years of age, characterized by rCBF values exceeding those of adults by a factor of ~2.5. The rCBF values then decrease to reach adult values at ~7 years of age. Such an age-related pattern is in agreement with those previously reported and measured with SPECT.5,25,27 Similar age-related variations were also demonstrated for brain metabolism.3,4

Some authors3,4 have attributed this first peak of regional brain perfusion and metabolism to the phenomena of polyneuronal innervation6–10 and elective synaptic stabilization.10–13 These phenomena are known to occur successively in the period of postnatal brain growth. Polyneuronal innervation relates to an initial process of an excessive production of dendritic spines and synapses.6–10 It is followed by mechanisms that act to retain those pathways in which patterns of external stimuli induce activity and to eliminate possible underactivated connections.10–16 The first peak also coincides with the first period of growth acceleration.51

A second, smaller peak was suspected visually at the age of 12 years, just before stabilization at adult values, but could not be demonstrated as statistically different from the 1-peak curve, although it showed a better correlation coefficient. The relatively low number of patients enrolled in the study is certainly a limitation to the statistical interpretation of the 2-peak model, which involves the estimation of 8 parameters. The existence of the second peak should be assessed in larger series, and its relationship with the growth acceleration in adolescents should be evaluated.

It is to be noted that the peak model selected to fit the data has been chosen to describe better the measured rCBF values. However, its validity and its overall identifiability have to be confirmed in larger series of patients.

The proportion represented by global CBF with respect to global cardiac output features the same pattern as global average rCBF values and is in agreement with previous reports.5,25,27 The global CBF represents >50% of the cardiac output at the peak of 1 to 3 years of age, explaining why the population of this age bracket is at risk for cerebrovascular diseases consecutive to systemic disorders.52

Perfusion CT was sensitive enough to allow for the identification of a specific and statistically significant temporal sequence regarding the successive increase of rCBF values in the various cortical regions. The comparative analysis between different cortical regions focused on the 0- to 6-year age bracket, the latter representing the key period in the acquisition of cognitive functions. Statistical analysis could not show any statistically significant differences between the different cortical regions beyond the age of 6. The temporal sequence as featured by perfusion CT (central cortex, then parietal cortex/calcarine cortex/temporal cortex/basal ganglia, finally frontal cortex/occipital cortex) is in global agreement with previous studies that considered brain perfusion5,27,53 and brain metabolism as assessed by PET.3,4 It may reflect the changes in growth of the different brain regions during development, because the regions of high rCBF coincide with regions of the brain undergoing rapid growth at each period of development.24

The central cortex first shows a rise in its rCBF values in the first months of life. This is consistent with the relatively early morphologic maturation of primary sensorimotor cortex when compared with other cortical areas. Neonatal behavior is primarily dominated by the activity of subcortical brain structures, as featured by the prominent intrinsic brainstem reflexes, such as the Moro, root, and grasp reflexes. Cortical activity is mostly limited to primary sensory and motor areas, whereas visual-motor function is rudimentary.54

The increase of rCBF values in the calcarine, temporal, parietal cortex, and basal ganglia occurs predominantly within the 1- to 4-year age bracket. These structures are of vital importance in the visual-spatial and visual-sensorimotor integration. Their development allows for purposeless limb movements to be replaced by more coordinated reaching movements.54–56

The frontal cortex shows an increase of its rCBF values last. This increase coincides with the appearance of higher cortical and cognitive function. The infant now displays more sophisticated interaction with his or her surroundings. Furthermore, the infant improves his or her performance on the delayed response task, a commonly used neuropsychological paradigm to evaluate the prefrontal lobe integrity.54,57,58

A nonsignificant trend to higher rCBF in the left hemisphere was observed in our series that has been reported previously.27,59,60 It might be related to the development of a left hemisphere dominance, expected to occur at approximately the age of 3 to 4 years.54 However, this can only be speculated because no thorough neuropsychological examination was performed in the children of our series.

At birth, the rCBF in the gray matter is almost equal to that in the white matter (ratio = 1.2), then shows a rapid increase between 6 and 24 months of age, and finally stabilizes at a ratio of 3 between gray and white matter, and this whatever the patient’s age. This may reflect the progressive differentiation between gray and white matter, along with the rapid development of their respective vascularization after birth. The difference in rCBF between gray and white matter after 24 months of age is in agreement with a higher capillary density in the gray matter, presumably in relation to the presence of neurons, whereas only axons are present in the white matter.49,50 It is of interest that the second peak in the rCBF curve at the age of 12 years is accounted for largely by changes in rCBF in the white matter. No statistically significant difference between the time courses of rCBF maturation could be identified between male and female patients.

Some could advocate that perfusion CT in children does not stand the comparison with MR-PWI, mainly because of its limited spatial coverage. However, compared with MR-PWI, perfusion CT shows the major advantage of leading to quantitative results, which is of uppermost importance in children because of the specific age-related variations. Without quantitative assessment, it is impossible to evaluate brain perfusion in an individual child with respect to the normal values for the corresponding age bracket. MR-PWI performed with a bolus of gadolinium-derived chelates affords only a qualitative comparison between the right and left hemispheres, because it provides only with relative CBFs. Indeed, it relies on a susceptibility effect. The latter can be observed only in the vicinity of and not within the artery, making impossible the selection of an appropriate arterial function.61 Phase-contrast and spin-labeling MR-PWI require no exogenous contrast agents but generate a perfusion-related contrast that is usually limited to 1% to 2%, related to a small blood volume fraction and leading to a low signal-to-noise ratio.6 In children, spin-labeling is challenged by the rapid blood flow values, which jeopardize the underlying model of water extraction.62

Although the limited spatial coverage of perfusion CT is a drawback, it does not interfere with the goal of the present study, which was to assess age-related variations of quantitative cerebral perfusion CT results in children without brain abnormality. In the clinical practice, perfusion CT–limited spatial coverage is balanced by the additional information regarding the brain perfusion that it provides and that is not provided by conventional cerebral CT. In our experience, this is especially true in the emergency setting, when magnetic resonance imaging (MRI) is not available or when time or the child’s condition does not allow the specialized anesthetic care often needed by children for an MRI examination. CT, as opposed to MRI, is more widely accessible, is less time-consuming, and does not interfere with the monitoring process in critically ill children. Head trauma, notably with the issue of posttraumatic hyperemia, is one condition that could benefit from the implementation of perfusion CT technique in children and from the quantitative assessment of brain perfusion. Indeed, children with severe trauma undergo cerebral CT and intravenous administration of iodinated contrast for the CT of the chest, abdomen, and pelvis. In such cases, adding a perfusion CT series represents minimal additional contrast material and <20% of the total radiation dose for the admission CT survey.

We acknowledge several limitations to our study. First, the patients who were enrolled in this study were considered as normal on the basis of a normal conventional cerebral CT examination and of a normal clinical/radiologic follow-up, including additional normal investigations. However, they were admitted for suspicion of a pathologic condition, most often mild head trauma or headaches. The repercussion of head trauma on brain perfusion in the acute phase is controversial, especially in children. It was not possible to enroll completely normal children, for obvious ethical reasons such as radiation and contrast administration. Moreover, the selection of children was as strict as possible with respect to the follow-up. In case of head trauma patients, for instance, only those with slight trauma were retained.

This study identified age-related variations among rCBV, MTT, and rCBF values extracted from perfusion CT data. These variations may be associated with the evolution of brain perfusion with age, as already reported previously with other imaging techniques3,4,18 and SPECT.5,24,25,27 However, the design of the present study cannot rule out the influence of the different imaging protocols applied to the 4 age categories. The rationale for the different protocols in the different age groups is as follows: the 4 protocols can mainly be distinguished by the total duration of data acquisition and the delay between the beginning of intravenous contrast material administration and data acquisition, which have been adapted to the distinctive cardiovascular hemodynamics of each age category. The acquisition parameters have been lowered as much as possible to reduce the patients’ radiation dose while keeping an acceptable signal-to-noise ratio. The radiation dose involved in each protocol is limited, because it represents 26% to 63% of a conventional noncontrast and contrast-enhanced cerebral CT. The amount of iodinated contrast material injected reaches the lower limit (1 mL/kg) of the standard range accepted in children, which is up to 3 mL/kg for intravenous urograms.63 Finally, the selected injection rates were the maximal ones acceptable in children, taking into account the peripheral vein diameter and resistance, as well as the gauge of the venous catheters used. For the same ethical reasons as mentioned above, it was not possible to apply different protocols successively in the same patients just for comparison’s sake. The possible repercussions of the short-term sedation required in 18 patients should also be taken into consideration as a potential limitation of this study, although dedicated imaging studies have previously suggested that the effect of premedication on CBF is limited or at least insignificant.64 Finally, we acknowledge a relatively small number of patients aged 1 to 3 years (5 patients). This is an important limitation of the study given that it is at this age that the relationship between rCBF values and age changes most dramatically.


   CONCLUSIONS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Characterization of brain perfusion through quantitative perfusion CT shows specific age variations, underlined by a neuroanatomic developmental background. Brain perfusion of each cortical area evolves according to a specific time course, in correlation with the development of cognitive functions. Perfusion CT results with respect to rCBF are in agreement with previous reports using different imaging techniques. However, perfusion CT provides with original data regarding MTT and rCBV, possibly related to postnatal angiogenesis. Perfusion CT thus represents a routine clinical tool that affords a direct insight into child brain perfusion. Despite its spatial coverage, perfusion CT represents an alternative to MRI for the quantitative assessment of brain perfusion disorders in children, such as brain trauma.


   ACKNOWLEDGMENTS
 
We acknowledge the competence of L. de Palma as a research assistant in the Department of Diagnostic and Interventional Radiology. We thank M. Rousselle for help in editing this manuscript.


   FOOTNOTES
 
Received for publication Jun 2, 2003; Accepted Oct 20, 2003.

Reprint requests to (M.W.) Department of Diagnostic and Interventional Radiology, University Hospital (CHUV), 1011 Lausanne, Switzerland. E-mail: max_wintermark{at}hotmail.com


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Brody H. Organization of the cerebral cortex. III. A study of aging in the human cerebral cortex. J Comp Neurol.1955; 102 :511 –556[CrossRef][Web of Science][Medline]
  2. Lassen NA. Cerebral blood flow tomography with xenon-133. Semin Nucl Med.1985; 15 :347 –356[CrossRef][Web of Science][Medline]
  3. Chugani HT, Phelps ME. Maturational changes in cerebral function in infants determined by 18FDG positron emission tomography. Science.1986; 231 :840 –843[Abstract/Free Full Text]
  4. Chugani HT, Phelps ME, Mazziotta JC. Positron emission tomography study of human brain functional development. Ann Neurol.1987; 22 :487 –497[CrossRef][Web of Science][Medline]
  5. Chiron C, Raynaud C, Maziere B, et al. Changes in regional cerebral blood flow during brain maturation in children and adolescents. J Nucl Med.1992; 33 :696 –703[Abstract/Free Full Text]
  6. Dobbing J, Sands J. Quantitative growth and development of human brain. Arch Dis Child.1973; 48 :757 –767[Abstract/Free Full Text]
  7. Oppenheim RW. Naturally occurring cell death during neural development. Trends Neurosci.1985; 8 :487 –493[CrossRef][Web of Science]
  8. Purves D, Lichtman JW. Elimination of synapses in the developing nervous system. Science.1985; 228 :799 –809[Abstract/Free Full Text]
  9. Brody BA, Kinney HC, Kloman AS, Gilles FH. Sequence of central nervous system myelination. J Neuropathol Exp Neurol.1987; 46 :283 –301[Web of Science][Medline]
  10. Changeux JP, Dranchin A. Selective stabilization of developing synapses as a mechanism for the specification of neuronal networks. Nature.1976; 264 :705 –712[CrossRef][Medline]
  11. Huttenlocher PR. Synaptic density in human frontal cortex—developmental changes and effects of aging. Brain Res.1979; 163 :195 –205[CrossRef][Web of Science][Medline]
  12. Huttenlocher PR, de Courten C, Gary LJ, van der Loos H. Synaptogenesis in human visual cortex—evidence for synapse elimination during normal development. Neurosci Lett.1982; 33 :247 –252[CrossRef][Web of Science][Medline]
  13. Cowan WM, Fawcett JW, O’Leary DDM, Stanfield BB. Regressive events in neurogenesis. Science.1984; 225 :1258 –1265[Abstract/Free Full Text]
  14. Farkas-Bargeton E, Diebler MF, Rosenberg B, Wehrle R. Histochemical changes of the developing human cerebral neocortex. Neuropediatrics.1984; 15 :82 –91[Web of Science][Medline]
  15. McDonald JW, Johnston MV. Physiological pathophysiological roles of excitatory amino acids during central nervous system development. Brain Res Rev.1990; 15 :41 –70[CrossRef][Medline]
  16. Easter SS Jr, Purves D, Rakic P, Spitzer NC. The changing view of neural specificity. Science.1985; 230 :507 –511[Abstract/Free Full Text]
  17. Baird HW, Garfunkel JM. A method for the measurement of cerebral blood flow in infants and children. J Pediatr.1953; 42 :570 –575[CrossRef][Web of Science][Medline]
  18. Kety SS. Human cerebral blood flow and oxygen consumption as related to aging. J Chronic Dis.1956; 3 :478 –486[CrossRef][Medline]
  19. Kennedy C, Sokoloff L. An adaptation of the nitrous oxide method to the study of the cerebral circulation in children; normal values for cerebral blood flow and metabolic rate in childhood. J Clin Invest.1957; 36 :1130 –1137
  20. Bode H, Wais U. Age dependence of flow velocities in basal cerebral arteries. Arch Dis Child.1988; 63 :606 –611[Abstract/Free Full Text]
  21. Horiuchi I, Sanada S, Ohtahara S. Developmental and psychologic changes in cerebral blood flow velocity. Pediatr Res.1993; 34 :385 –388[Web of Science][Medline]
  22. Ball WS Jr, Holland SK. Perfusion imaging in the pediatric patient. Magn Reson Imaging Clin N Am.2001; 9 :207 –230[Medline]
  23. Adelson PD, Clyne B, Kochanek PM, et al. Cerebrovascular response in infants and young children following severe traumatic brain injury: a preliminary report. Pediatr Neurosurg.1997; 26 :200 –207[Web of Science][Medline]
  24. Ogawa A, Sakurai Y, Kayama T, Yoshimoto T. Regional cerebral blood flow with age: changes in rCBF in childhood. Neurol Res.1989; 11 :173 –176[Medline]
  25. Suzuki K. The changes of regional cerebral blood flow with advancing age in normal children. Nagoya Med J.1990; 34 :159 –170
  26. Pupi A, Castagnoli A, De Cristofaro MT, Bacciottini L, Petti AR. Quantitative comparison between 99mTc-HMPAO and 99m-Tc-ECD: measurement of arterial input and brain retention. Eur J Nucl Med.1994; 21 :124 –130[Web of Science][Medline]
  27. Barthel H, Wiener M, Dannenberg C, et al. Age-specific cerebral perfusion in 4- to 15-year-old children: a high-resolution brain SPET study using 99mTc-ECD. Eur J Nucl Med.1997; 24 :1245 –1252[CrossRef][Web of Science][Medline]
  28. Schiepers C, Verbruggen A, Casaer P, De Roo M. Normal brain perfusion pattern of Technetium-99m-ethylcysteinate dimer in children. J Nucl Med.1997; 38 :1115 –1120[Abstract/Free Full Text]
  29. Takahashi T, Shirane R, Sato S, Yoshimoto T. Developmental changes of cerebral blood flow and oxygen metabolism in children. AJNR Am J Neuroradiol.1999; 20 :917 –922[Abstract/Free Full Text]
  30. Kennedy C, Grave GD, Jehle JW, Sokoloff L. Blood flow to white matter during maturation of the brain. Neurology.1970; 20 :613 –618[Free Full Text]
  31. Eastwood JD, Lev MH, Provenzale JM. Perfusion CT with iodinated contrast material. AJR Am J Roentgenol.2003; 180 :3 –12[Free Full Text]
  32. Lev MH, Segal AZ, Farkas J, et al. Utility of perfusion-weighted CT imaging in acute middle cerebral artery stroke treated with intra-arterial thrombolysis—prediction of final infarct volume and clinical outcome. Stroke.2001; 32 :2021 –2027[Abstract/Free Full Text]
  33. Wintermark M, Reichhart M, Thiran JPH, et al. Prognostic accuracy of cerebral blood flow measurement by perfusion computed tomography, at the time of emergency room admission, in acute stroke patients. Ann Neurol.2002; 51 :417 –432[CrossRef][Web of Science][Medline]
  34. Wintermark M, Reichhart M, Maeder P, et al. Comparison of admission perfusion computed tomography and qualitative diffusion- and perfusion-weighted magnetic resonance imaging in acute stroke patients. Stroke.2002; 33 :2025 –2031[Abstract/Free Full Text]
  35. Wintermark M, Thiran JPH, Maeder P, Schnyder P, Meuli R. Simultaneous measurements of regional cerebral blood flows by perfusion-CT and stable xenon-CT: a validation study. AJNR Am J Neuroradiol.2001; 22 :905 –914[Abstract/Free Full Text]
  36. Furukawa M, Kashiwagi S, Matsunaga N, et al. Evaluation of cerebral perfusion parameters measured by perfusion CT in chronic cerebral ischemia: comparison with xenon CT. J Comput Assist Tomogr.2002; 26 :272 –278[CrossRef][Web of Science][Medline]
  37. Kudo K, Terae S, Katoh C, et al. Quantitative cerebral blood flow measurement with dynamic perfusion CT using the vascular-pixel elimination method: comparison with H215O positron emission tomography. AJNR Am J Neuroradiol.2003; 24 :419 –426[Abstract/Free Full Text]
  38. Wintermark M, Cuisenaire O, Thiran JPH, et al. Perfusion-CT determination of penumbra and infarct in acute stroke patients: from rCBF and rCBV maps to a clinical prognosis. Radiology.2001; S221 :647
  39. Wintermark M, Cuisenaire O, Thiran JPH, et al. Perfusion-CT determination of penumbra and infarct in acute stroke patients: from cerebral perfusion maps to a clinical prognosis. Eur Radiol.2002; S12 :460
  40. Wintermark M, Maeder P, Thiran JPH, Schnyder P, Meuli R. Quantitative assessment of regional cerebral blood flows by perfusion CT studies at low injection rates: a critical review of the underlying theoretical models. Eur Radiol.2001; 11 :1220 –1230[CrossRef][Web of Science][Medline]
  41. Axel L. Cerebral blood flow determination by rapid-sequence computed tomography. Radiology.1980; 137 :679 –686[Abstract/Free Full Text]
  42. Axel L. A method of calculating brain blood flow with a CT dynamic scanner. Adv Neurol.1981; 30 :67 –71[Medline]
  43. Axel L. Tissue mean transit time from dynamic computed tomography by a simple deconvolution technique. Invest Radiol.1983; 18 :94 –99[CrossRef][Web of Science][Medline]
  44. Ladurner G, Zilkha E, Iliff LD, et al. Measurement of regional cerebral blood volume by computerized axial tomography. J Neurol Neurosurg Psychiatry.1976; 39 :152 –155[Abstract/Free Full Text]
  45. Zilkha E, Ladurner G, Linette D, et al. Computer subtraction in regional cerebral blood-size measurements using the EMI-scanner. Br J Radiol.1976; 49 :330 –334[Abstract/Free Full Text]
  46. Ladurner G, Zikha E, Sager WD, et al. Measurement of regional cerebral blood volume using the EMI 1010 scanner. Br J Radiol.1979; 52 :371 –374[Abstract/Free Full Text]
  47. Haycock GB, Schwartz GJ, Wisotsky DH. Geometric method for measuring body surface area: a height-weight formula validated in infants, children and adults. J Pediatr.1978; 93 :62 –66[Web of Science][Medline]
  48. Shann F. DRUG DOSES. 11th ed. Parkville, Australia: Victoria Collective Pty Ltd;2001
  49. Schade JP, van Groenigen WB. Structural organization of the human cerebral cortex. Acta Anat.1961; 47 :74 –111[Web of Science][Medline]
  50. Norman MG, O’Kursky JR. The growth and the development of microvasculature in human cerebral cortex. J Neuropathol Exp Neurol.1986; 45 :222 –232[Web of Science][Medline]
  51. Largo RH, Walli R, Duc G, Fanconi A, Prader A. Evaluation of perinatal growth. Presentation of combined intra- and extrauterine growth standards for weight, length and head circumference. Helv Paediatr Acta.1980; 35 :419 –436[Web of Science][Medline]
  52. Fullerton HJ, Chetkovich DM, Wu YW, et al. Deaths from stroke in US children, 1979 to 1998. Neurology.2002; 59 :34 –39[Abstract/Free Full Text]
  53. Lou H, Henriksen L, Greisen G, Schneider S. Redistribution of cerebral activity during childhood. Brain Dev.1990; 12 :301 –305[Web of Science][Medline]
  54. Illingworth RS. The Development of the Infant and Young Child. Abnormal and Normal. 8th ed. Edinburgh: Churchill Livingstone; 1987
  55. Bronson G. The postnatal growth of visual capacity. Child Dev.1974; 45 :873 –890[CrossRef][Web of Science][Medline]
  56. Von Hofsten C. Developmental changes in the organization of prereaching movements. Dev Psychobiol.1984; 20 :378 –388[CrossRef]
  57. Kagan J. Do infants think? Sci Am.1972; 226 :74 –82[Web of Science][Medline]
  58. Goldman-Rakic PS. The frontal lobes: uncharted provinces of the brain. Trends Neurosci.1984; 7 :408 –414[CrossRef][Web of Science]
  59. Podreka I, Goldenberg G, Baumgartner C, et al. HMPAO brain uptake in young normal subjects: gender differences and hemispheric asymmetries. J Cereb Blood Flow Metab.1989; 9 :202
  60. Chiron C, Jambaque I, Nabbout R, Lounes R, Syrota A, Dulac O. The right brain hemisphere is dominant in human infants. Brain.1997; 120 :1057 –1065[Abstract/Free Full Text]
  61. Sorensen AG, Reimer P. Cerebral MR Perfusion Imaging. Principles and Current Applications. Stuttgart, Germany: Thieme Verlag; 2000
  62. Oguz KK, Golay X, Pizzini FB, et al. Sickle cell disease: continuous arterial spin-labeling perfusion MR imaging in children. Radiology.2003; 27 :567 –574
  63. Slovis TL, Sty JR, Haller JO. Imaging of the Pediatric Urinary Tract. Philadelphia, PA: WB Saunders; 1989: 1–43
  64. Werner C. Effects of analgesia and sedation on cerebrovascular circulation, cerebral blood volume, cerebral metabolism and intracranial pressure. Anaesthesist.1995; 44(suppl 3) :S566 –S572
PEDIATRICS (ISSN 1098-4275). ©2004 by the American Academy of Pediatrics

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Facebook Facebook   Add to Reddit Reddit   Add to Technorati Technorati   Add to Twitter Twitter    What's this?


This article has been cited by other articles:


Home page
 Am. J. Neuroradiol.Home page
J.M. Pollock, C.T. Whitlow, A.R. Deibler, H. Tan, J.H. Burdette, R.A. Kraft, and J.A. Maldjian
Anoxic Injury-Associated Cerebral Hyperperfusion Identified with Arterial Spin-Labeled MR Imaging
AJNR Am. J. Neuroradiol., August 1, 2008; 29(7): 1302 - 1307.
[Abstract] [Full Text] [PDF]

Home page
 BrainHome page
K. A. Strauss, J. Lazovic, M. Wintermark, and D. H. Morton
Multimodal imaging of striatal degeneration in Amish patients with glutaryl-CoA dehydrogenase deficiency
Brain, July 1, 2007; 130(7): 1905 - 1920.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (9)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wintermark, M.
Right arrow Articles by Gudinchet, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wintermark, M.
Right arrow Articles by Gudinchet, F.
Related Collections
Right arrow Neurology & Psychiatry
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?