| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Neuroimaging |
1 Department of Radiological Sciences, Division of Diagnostic Imaging, St. Jude
Children's Research Hospital, 332 N Lauderdale St., Memphis, TN
38015-2794.
2 Department of Radiology, University of Tennessee, School of Medicine, Memphis,
TN 38163.
3 Department of Biostatistics, St. Jude Children's Research Hospital, Memphis,
TN 38015-2794.
4 Department of Hematology-Oncology, St. Jude Children's Research Hospital,
Memphis, TN 38015-2794.
5 Department of Pediatrics, University of Tennessee, School of Medicine,
Memphis, TN 38163.
Received December 29, 2003;
accepted after revision March 23, 2004.
Supported in part by the American Lebanese Syrian Associated Charities.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. We performed 55 paired examinations with duplex imaging and nonduplex nonimaging sonography machines on 53 children with sickle cell anemia. Examinations were performed consecutively by three sonographers blinded to the results of the opposing study. Velocities were measured in five clinically relevant vessels.
RESULTS. Time-averaged mean maximum (TAMx) velocity measurements obtained with the duplex equipment were significantly lower than those made with the nonduplex equipment for all vessels except the posterior cerebral artery. The mean differences were 10.9% (p < 0.0001) in the middle cerebral artery, 12.7% (p = 0.002) in the anterior cerebral artery, 2.2% (p = 0.69) in the posterior cerebral artery, 21.0% (p < 0.0001) in the distal internal carotid artery, and 15.3% (p < 0.0001) at the bifurcation of the distal internal carotid artery.
CONCLUSION. If TAMx velocities measured with duplex equipment are used to assign stroke risk in children with sickle cell anemia, we suggest that 180 cm/sec or more should be considered abnormal, and 153–179 cm/sec, as conditional. These values are 10% lower than those obtained from the nonduplex equipment used in the Stroke Prevention Trial in Sickle Cell Anemia study.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The risk of stroke in children with sickle cell anemia is accurately predicted by the intracranial arterial blood velocity as measured by nonduplex transcranial Doppler (TCD) sonography: stroke risk increases with increasing blood velocity [9, 10]. In children with abnormal velocities, first stroke can be prevented by prophylactic treatment with frequent RBC transfusions [7, 10–12]. However, TCD screening has not become the standard of care, except in large sickle cell centers, primarily because the necessary equipment and trained personnel are not available. Although most centers in the United States have sonography equipment, they may not possess the appropriate transducer needed to perform duplex TCD with imaging (TCDI). Because the cost of the imaging transducer is considerably less than that of a nonduplex TCD machine, many centers are likely to use duplex imaging equipment. More important, those performing and interpreting these studies require specific training in TCD sonography techniques, regardless of the equipment used, because these examinations are technically challenging and optimization of velocity measurements requires careful attention to the audible signal, the velocity readouts, and the probe position [13–17].
If stroke risk is to be assigned from measurements obtained by using duplex equipment on the basis of criteria established with nonduplex equipment, it is necessary to consider how velocity measurements obtained with these types of equipment might differ. Previous investigators have found that imaging Doppler sonography equipment produces lower velocity measurements than nonimaging equipment [13–15]. To further validate these findings, we sought to compare velocity measurements obtained by the duplex imaging technique using a Sequoia (Acuson) sonography machine with those obtained by the nonduplex method using a Pioneer TCD (Nicolet Biomedical) sonography machine.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Of 56 patients enrolled, three could not be evaluated because of inadequate temporal bone windows. The 53 patients who could be evaluated included 28 boys and 25 girls, 2–17 years old (median age, 10.0 years). Two patients were enrolled twice; therefore, 55 pairs of TCD and TCDI examinations were performed on 53 patients. All pairs of examinations were performed consecutively but in random order (with regard to which examination was performed first) by two of three sonographers who were blinded to the results of the opposing study. Two sonographers had attended a TCD training course offered by the Medical College of Georgia, and the third performed TCDI examinations under the supervision of the principal investigator. Two sonographers were registered by the American Registry of Diagnostic Medical Songraphers as vascular sonographers, and the third was a pediatric radiologist with 9 years' experience in sonography. The TCDI examinations were performed on an Acuson Sequoia sonography machine with a 3V2 cardiac transducer set at 2.5 MHz. The TCD examinations were performed on a Nicolet Pioneer TCD machine with a 2-MHz transducer. For both types of examination, we used a transtemporal window setting and a 6-mm gate in most cases and obtained measurements at 2-mm increments along the entire length of the middle cerebral artery, beginning at the most superficial depth that allowed an accurate tracing. We obtained measurements at two depths from the distal internal carotid artery and its bifurcation, the anterior cerebral artery, and the posterior cerebral artery. When possible, we obtained two measurements at each depth in all vessels to determine their reproducibility.
We recorded the time-averaged mean maximum (TAMx) velocity, peak systolic
velocity (Vs), and end diastolic velocity
(Vd) and calculated the resistive index (RI) as
follows:
 |
|
The TCD measurements were obtained by using the envelope tracing on the Nicolet Pioneer machine, which automatically encased the waveform over the entire strip (Fig. 2). We postprocessed the TCD recordings by adjusting the gain to maximize the velocity readings while maintaining a tight envelope around the waveform. Because it was not possible to perform angle correction with the TCD equipment, we did not do so for the TCDI measurements.
|
We used the Stroke Prevention Trial in Sickle Cell Anemia Study (STOP) criteria [11], which are based on the velocity (TAMx) measured in the middle cerebral artery and distal internal carotid artery, by the nonduplex (TCD) method, to assign stroke risk as normal (TAMx < 170 cm/sec), conditional (TAMx < 170 cm/sec but < 200 cm/sec), or abnormal (TAMx < 200 cm/sec). To ensure the equivalence of these categories for the duplex (TCDI) method, we adjusted these ranges to appropriate values by performing a regression analysis and determining kappa coefficients for the comparison.
Because multiple measurements were obtained at various depths along each
vessel, the maximum values were used as the primary statistics for analysis.
The relationship between TCD and TCDI is best represented as a relative (or
percentage) difference expressed by
 |
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
In Table 2 and in the data analysis, we included not only patients who had both left and right vessels measured but also those who had only one vessel (left or right) measured. Also, the data set included a few measurements for which the side was not specified. Cases in which the side was unknown were excluded when calculating the numbers presented in Table 1, but all measurements were included in the analysis presented in Table 2, even when the side was unknown. The TCD-based peak systolic velocity and end diastolic velocity measurements were also significantly higher (p < 0.05) than the corresponding TCDI measurements from all vessels except the posterior cerebral artery. We obtained significantly different RI values for the distal internal carotid artery (TCDI value higher by 6.8%; p = 0.002) and the posterior cerebral artery (TCDI value higher by 3.8%; p = 0.05).
On the basis of the STOP criteria, TCD examinations placed 41 patients in
the normal risk category, 12 in the conditional category, and two in the
abnormal category. By the same (unadjusted) criteria, TCDI examinations placed
52 patients in the normal, two in the conditional, and one in the abnormal
category. This poor agreement in stroke risk assignment between the two
techniques was reflected in a low kappa value (0.24), and the difference
between the two sets of results was statistically significant (p =
0.01). The regression analysis of TAMx values obtained from the middle
cerebral artery showed that the TCDI values were (mean ± SD) 12.1%
± 1.2% lower than the TCD values (p < 0.0001). We then
determined the kappa coefficients for agreement between the established TCD
cut points and five proposed TCDI cut points at values centered around a 12.1%
reduction (Table 3). A
reduction of the TCDI cut points by 10% resulted in the best agreement
(highest kappa value) between our data sets for the two techniques (
=
0.68, p < 0.0001) (Table
3). The adjusted categories of velocity measured by the TCDI
method were therefore normal for TAMx less than 153 cm/sec, conditional for
TAMx of 153 cm/sec or greater but less than 180 cm/sec, and abnormal for TAMx
of 180 cm/sec or greater.
|
TAMx measurements obtained by TCD sonography had significantly smaller SDs than did those obtained by TCDI in the middle cerebral artery (p = 0.0001) and bifurcation of the distal internal carotid artery (p = 0.01), but SDs of TAMx measurements were similar from the distal internal carotid artery (p = 0.11), the anterior cerebral artery (p = 0.72), and the posterior cerebral artery (p = 0.44) regardless of the machine used. For all vessels combined, the SDs of TAMx measurements made by TCD sonography were significantly smaller than those of measurements obtained by TCDI (p = 0.0004).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Because of the Doppler shift effect, velocity measurements are most accurate when the angle of insonation between the transducer and the vessel being interrogated is less than 15° [16, 17, 21]. Therefore, a more desirable angle of insonation can be obtained for the middle cerebral artery or posterior cerebral artery than for the other intracranial arteries because the aforementioned vessels tend to course directly toward the transducer when a transtemporal window setting is used.
Because angle correction is not possible with nonduplex equipment and we wished to replicate the STOP scanning protocol, we did not use angle correction in our TCDI measurements. During examination of the distal internal carotid artery and anterior cerebral artery, a necessarily large angle of insonation results in less accurate measurements. This loss of accuracy may explain why we found greater differences between TCD- and TCDI-based measurements in the distal internal carotid artery and anterior cerebral artery than we did between TCD- and TCDI-based measurements in the middle cerebral artery, the bifurcation of the distal internal carotid artery, and the posterior cerebral artery. In addition, the SDs in our TAMx measurements from the middle cerebral artery tended to be smaller than those from most other vessels regardless of the machine used. Others have found that the most reliable velocity measurements are provided by the middle cerebral artery and the least reliable, by the anterior cerebral artery [22–25]. We also found that for all vessels combined, TCD-based measurements have a significantly lower SD than do TCDI-based measurements (p = 0.0004), suggesting that TCD-based results are more reproducible. This result may be at least partially explained by the fact that TCD TAMx measurements are obtained from numerous cardiac cycles, whereas the TCDI measurements are obtained over only two cycles. The greater number of raw data points obtained by TCD would be expected to result in less variability.
Several factors may partially explain why significant differences exist between TCD- and TCDI-based velocity measurements. The transducer used for TCD examinations has a smaller "footprint" and is lighter and easier to manipulate than the TCDI probe. These features are advantageous, especially when the temporal bone window setting is small because very small differences in probe position and angle can result in significant differences in velocity measurements. Also, the TCD machine is equipped with a small hand-held remote control pad that can be easily managed to record velocity measurements with one hand while the other hand manipulates the Doppler probe. This feature allows the TCD examiner to sit comfortably throughout the examination. In contrast, the TCDI examiner must be able to reach the keyboard with one hand while holding the probe in the other. This may require the examiner to stretch across the patient, if standing at the bedside, or to twist toward the keyboard, if sitting at the head of the table. In addition, when acquiring measurements with the Acuson Sequoia, the examiner must freeze the image and then scroll along the entire wavestrip to identify the highest velocity before recording the measurements. The Nicolet TCD machine provides a continuous readout of velocity measurements and allows instantaneous storage of measurements without a delay in the study. These factors may result in examiner fatigue during TCDI that could affect the accuracy of the TCDI examination. Not surprisingly, considering these factors, Neish et al. [15] found that TCD examinations were performed more quickly than TCDI examinations: the average TCD examination took 29 min and the average TCDI, 43 min.
Inherent differences in signal processing between these two types of equipment may also account for the differences we found. In the future, comparative studies using phantoms may improve our understanding of differences between imaging and nonimaging Doppler sonography.
Nevertheless, TCDI does offer several advantages. The equipment is readily available, although a probe capable of performing transcranial Doppler sonography is also required. Imaging offers enhanced operator confidence because one can visualize the vessel being examined and know with certainty from which vessel measurements are obtained. Consequently, TCDI may require a shorter learning period. In addition, TCDI may reveal unexpected vascular findings such as an aneurysm or vascular malformation [25].
In applying the STOP criteria for stroke risk to our TCD- and TCDI-based
measurements, we found 12 cases (22%) of discordance between the two sets of
values. In every case, the TCDI result placed the patient in a lower stroke
risk category than did the TCD result (11 patients classified as conditional
using TCD examination were classified as normal using TCDI, and one patient
whose TCD result was abnormal was classified as conditional by TCDI). Because
middle cerebral artery measurements are probably the most accurate, as
discussed previously, and because the STOP study found that the highest
velocity measurement was obtained from the middle cerebral artery in more than
90% of cases (Adams R, personal communication), it is appropriate to determine
stroke risk on the basis of velocity measurements from the middle cerebral
artery. We found that a reduction of 10% in the cut points of the middle
cerebral artery TAMx values measured by TCDI resulted in the best agreement
between TCD-based and TCDI-based stroke risk assignments for our cohort.
Therefore, on the basis of our results and those of others, we recommend that
when assigning stroke risk on the basis of blood velocity in the middle
cerebral artery (but not other vessels) measured by TCDI, a TAMx of 153 cm/sec
or greater should be considered conditional and a TAMx of 180 cm/sec or
greater, abnormal. Jones et al.
[14] similarly recommended
that a TCDI-based TAMx of 180 cm/sec or more be considered abnormal. Applying
these adjusted values to our study cohort results in six (11%) rather than 12
discordant cases (22%) and an improvement in agreement between data sets from
low ( = 0.24) to moderate ( = 0.68).
Stroke risk is a continuum. Patients with conditional results should have a repeat TCD or TCDI examination within 6 months, and those with abnormal results should be reevaluated within 1 week for confirmation [9–11, 14]. Those with values just below the abnormal cut point also deserve close and early follow-up examinations. Patients with conditional or abnormal TCD or TCDI results should be evaluated with MRI and MR angiography, which can reveal ischemic brain injury or arterial stenoses and occlusions [7, 9, 10, 26–29]. These imaging studies provide a baseline assessment that may prove to be particularly valuable for comparison purposes if the patient subsequently experiences an overt or suspected clinical central nervous system event. Patients with two abnormal TCD or TCDI results are candidates for chronic RBC transfusion, but the decision to begin transfusion therapy should be made only after careful consideration of the associated risks and benefits. Problems with venous access, compliance, cost, and complications of transfusions must be weighed against the risk of irreversible brain damage secondary to stroke [10].
Acknowledgments
We thank Robert Adams for reviewing this manuscript and sonographers April
Chapman and Stacey Glass for technical assistance.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Krejza, W. Rudzinski, M.A. Pawlak, M. Tomaszewski, R. Ichord, J. Kwiatkowski, D. Gor, and E.R. Melhem Angle-Corrected Imaging Transcranial Doppler Sonography versus Imaging and Nonimaging Transcranial Doppler Sonography in Children with Sickle Cell Disease AJNR Am. J. Neuroradiol., September 1, 2007; 28(8): 1613 - 1618. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
