Eur. Radiol. (2001) 11: 567±574 Ó Springer-Verlag 2001
Ü. Aydingöz M. H. Öztürk
Received: 10 April 2000 Revised: 30 June 2000 Accepted: 4 July 2000
)
Ü. Aydingöz ( ) ´ M. H. Öztürk Department of Radiology, Hacettepe University Medical Center, 06100 Ankara, Turkey e-mail:
[email protected] Tel.: + 90-3 12-3 05 29 58 Fax: + 90-3 12-3 12 70 71 Present address: Ü. Aydingöz, Cevher Sok. 2±3, Kurtulus, 06600 Ankara, Turkey
MUSC ULOSKELETA L
MR imaging of the acetabular labrum: a comparative study of both hips in 180 asymptomatic volunteers
Abstract The aim of this study was to determine the MR imaging characteristics of normal acetabular labra on both hips. Three hundred sixty acetabular labra on both hips of 180 asymptomatic volunteers were examined on a 0.5-T MR unit utilizing a superficial coil. T1weighted and T2-weighted gradientecho sequences (TR/TE/FA 169/ 10 ms/40 , and 316/25 ms/40 , respectively) were used. The volunteers were classified into five age groups (10±19, 20±29, 30±39, 40±49, and 50 + years of age). The shape (triangular, flat, round, absent), size, and the presence of intralabral foci of increased intensity (diffuse, linear, central, or basal) were noted on both labra in each volunteer, and symmetry±asymmetry of these features were investigated. The most common labrum shape was triangu-
Introduction The acetabular labrum is a fibrocartilaginous structure attached to the lateral inferior aspect of the acetabulum. It deepens the acetabulum, thus facilitating the hip joint stability. Labrum tears or detachments cause hip pain and are usually secondary to acute hip dislocations and injuries during sporting activities [1, 2]. Other associations with labral lesions include dysplastic hips, LeggCalvØ-Perthes disease, and osteoarthritis [3, 4, 5, 6]. Apart from arthroscopy, radiological methods, such as conventional arthrography and CT arthrography, have been used to investigate labral lesions. In recent years, however, MR imaging and MR arthrography
lar, whereas absence of labrum was the least common condition. A difference of labral shapes between both hips was present in approximately 15 % of volunteers. A size difference of over 25 % between each labrum was noted in approximately one-fourth of volunteers. An age-related pattern of intralabral intensity increase was noted on both MR sequences (more commonly on T1-weighted gradient-echo images). Shape differences and age-related intensity changes of the acetabular labrum in the MR imaging of the hips were confirmed, and the possibility of variability of the labra, with regard to the shape and size, on both hips of the same individual was documented. Key words Hip joint ´ Acetabulum ´ MR imaging
have been increasingly used to evaluate the acetabular labrum [7, 8, 9, 11]. Currently, MR imaging is the modality of choice in non-invasively visualizing this structure. However, normal MR imaging features of the aceTable 1 The age and gender distribution of volunteers Age (years)
Females
Males
Total
10±19 20±29 30±39 40±49 50 + Total
14 30 16 15 15 90
14 27 19 15 15 90
28 57 35 30 30 180
568
A
tabular labrum need to be clarified in order to differentiate the abnormal conditions from normal (sometimes age-related) situations or processes. Few studies have been reported in the literature regarding the normal shape, size, and signal intensity of the labrum in asymptomatic volunteers utilizing conventional spin-echo MR imaging sequences, and both hips have been examined in only several cases [10, 11]. This study aimed to comparatively demonstrate the normal MR imaging features of both acetabular labra in the largest reported series of both hips in asymptomatic volunteers utilizing gradient-echo MR imaging sequences.
B
Materials and methods Study population
C
D
Fig. 1 A±D There are three basic types of labral shape: A triangular; B round; C flat. D Also, there is a possibility of absence of the acetabular labrum
One hundred eighty volunteers (a total of 360 hips) without any history of hip disease, significant trauma, surgery, or radiotherapy to or near the hip joints and any symptoms related to their hips were included in the study. Ninety volunteers were male and 90 were female; age range was 10±73 years (females 10±73 years, males 10±70 years); mean age was 33.9 years (females 33.5 years, males 34.2 years). Informed consent was obtained after the nature of the procedure had been explained. Table 1 shows the gender distribution of volunteers in various age groups. MR technique
Fig. 2 A±C Diagram illustrates the dimensions used in cross-sectional area measurements of various labrum shapes. Labral area measurements were made in midcoronal images in accordance with the labral shape. A In order to make area measurements for triangular labra, which is by far the most common type, the height of the labrum (Distance 1) and the width of the labrum at the level of the labral base (Distance 2) were first measured. B For round labra, Distances 1 and 2 were approximately the same, representing the ªdiameterº of the labrum; however, ellipsoid labra were measured using the appropriate area formula. C For flat labra, Distance 1 represents the length of the ªrectangle,º whereas Distance 2 denotes the width
A
B
Magnetic resonance imaging examinations were performed on a 0.5-T unit (Philips Gyroscan T5-NT, Philips, Eindhoven, The Netherlands) utilizing a receive-only surface coil. Our standard MR examination consisted of obtaining three localizer images of the hips of the patient in supine position, obtained in the coronal, sagittal, and axial planes, followed by eight-slice coronal T1weighted and T2-weighted gradient-echo sequences (TR/TE/ FA = 169/10 ms/40 , and 316/25 ms/40 , respectively) covering both hips (25- or 26-cm fields of view). Other parameters of both sequences were: 5 mm slice thickness with a gap of 0.5 mm; 10 and 4 excitations for T1-weighted and T2-weighted gradient-echo MR images, respectively; and a 256 256 matrix. Images were enlarged in all volunteers to show each hip individually.
C
569
A
B
C
Fig. 4 Percentage of volunteers with intralabral intensity increases on T1-weighted gradient-echo MR images by age. Volunteers with unilateral or bilateral absence of labrum were excluded
D
Fig. 3 A±D Numerous intralabral intensity increases are seen on MR images. These are classified according to their location within the labrum: A central; B basal; C linear; D diffuse. All images are from the T1-weighted gradient-echo sequence
The presence and location of intralabral foci of increased signal intensity (central, basal, linear, or diffuse, according to their pattern and location within the acetabular labrum) were also noted (Fig. 3).
Table 2 Labral shapes on 360 hips in 180 volunteers
Statistical analysis
Shape Triangular Round Flat Absent Total
Males
Females
Total
N
%
N
%
N
%
128 25 21 6 180
71.1 13.9 11.7 3.3 100.0
121 32 24 3 180
67.2 17.8 13.3 1.7 100.0
249 57 45 9 360
69.2 15.8 12.5 2.5 100.0
Radiological evaluation The MR images were interpreted by two radiologists (one a musculoskeletal radiologist). Consensus was reached in all cases. For each sequence and each side the interpreters noted the shape (triangular, round, flat, or absent) of the labrum (Fig. 1). The difference of labrum type on each side of the same volunteer, when present, was also noted. Cross-sectional areas of both labra were calculated on the slice which displayed the largest cross section of the labrum (the fourth or the fifth slice of the eight-slice T1weighted gradient-echo sequence) using the relevant area formula for each labrum type (Fig. 2). Triangular labra were obviously measured using triangle area formula. A flat labrum was considered as a rectangle and was measured accordingly. For round labra area formula of a circle was used; however, when measurements in two axis ± height and width ± were not the same, the shape was considered to be ovoid and the area measurement was made according to the ellipse formula.
Chi-square, Student's t-test, and logistic regression analysis were used in the statistical analyses of the results.
Results Labrum shapes of 360 hips in 180 volunteers are shown in Table 2. Triangular shape was the most common, whereas absence of an identifiable labrum was noted in 2.5 % of volunteers (the least common condition). Numerous intralabral intensity increases were noted, more commonly on T1-weighted MR images. Tables 3 and 4 show the age and gender distribution of these on T1-weighted and T2-weighted gradient-echo images, respectively. Cumulative intralabral intensity increases were more frequent in males on both T1-weighted and T2-weighted gradient-echo MR images for all age groups, but this was not statistically significant (p > 0.05). Percentages of volunteers with intralabral intensity increase for each age group are displayed in Figs. 4 and 5 for T1-weighted and T2-weighted gradientecho MR images, respectively (volunteers without visible labrum were excluded from total in the calculation of percentages of intralabral intensity increases). The
570
Table 3 Number of volunteers with intralabral intensity increases on T1-weighted gradient-echo MR images by age and gender Age group 10±19 (years) Gender F Centralb Basalb Linearb Diffuseb Total
20±29 M a
N
%
1 2 2 5
0 3.6 7.1 7.1 17.9
30±39
F a
M a
40±49
F a
N
%
N
%
N
%
1 5 1 2 9
3.6 17.9 3.6 7.1 32.1
4 3 5 12
6.7 5 8.3 0 20.0
4 5 4 2 15
7.5 9.4 7.5 3.8 28.3
M a
50 +
F a
N
%
N
%
1 1 8 2 12
3.1 3.1 25.0 6.2 37.5
1 10 5 2 18
2.8 27.8 13.9 5.6 50.0
M a
F a
M a
N
%
N
%
N
%
2 1 3 3 9
7.1 3.6 10.7 10.7 32.1
3 3 4 1 11
11.1 11.1 14.8 3.7 40.7
1 2 2 11 16
3.4 6.9 6.9 37.9 55.2
N
%a
1 7 5 14 27
3.3 23.3 16.7 46.7 90.0
a These figures represent the percentage of labra with intralabral intensity increases among total labra for the mentioned gender and age groups. Hips without identifiable labrum were excluded from calculations b Intralabral intensity increase pattern
Table 4 Number of volunteers with intralabral intensity increases on T2-weighted gradient-echo MR images by age and gender Age group 10±19 (years) Gender F b
Central Basalb Linearb Diffuseb Total
20±29 M
30±39
F
M
40±49
F
M
F
50 + M
F
M
N
%a
N
%a
N
%a
N
%a
N
%a
N
%a
N
%a
N
%a
N
%a
N
%a
1 2 3
0 0 3.6 7.1 10.7
1 1 2 4
3.6 0 3.6 7.1 14.3
3 5 8
5 0 8.3 0 13.3
2 3 3 1 9
3.8 5.7 5.7 1.9 17.0
1 5 2 8
3.1 0 15.6 6.2 25.0
1 6 2 9
2.8 16.7 5.5 0 25.0
2 1 3 1 7
7.1 3.6 10.7 3.6 25.0
3 2 4 9
11.1 7.4 14.8 0 33.3
1 4 2 6 13
3.4 13.8 6.9 20.7 44.8
1 6 6 6 19
3.3 20.0 20.0 20.0 63.3
a These figures represent the percentage of labra with intralabral intensity increases among total labra for the mentioned gender and age groups. Hips without identifiable labrum were excluded from calculations b Intralabral intensity increase pattern
Table 5 Average labrum area (mm2) on midcoronal T1-weighted gradient-echo MR images by age and gender Age group 10±19 (years) Gender F M
20±29
30±39
40±49
50±59
F
F
F
F
Right Left
8.3 11.2 7.8 11.0
8.6 12.0 8.4 10.4
M
M
7.8 11.4 7.8 10.9
M
8.4 13.5 8.4 8.9
M
10.2 12.7 8.4 10.6
Persons without labrum were excluded in calculating the average areas
frequency of intralabral intensity increases showed a gradual increase with age on gradient-echo MR images (p < 0.05). Labrum area measurements on midcoronal MR images showed a statistically significant difference between the two genders, cross-sectional labrum sizes being greater in males than in females (p < 0.05; Table 5). Labral symmetry on both hips has also been investigated with regard to labrum shape, presence of intralabral intensity changes, and labrum size. Labral symmetry with respect to the labrum shape was observed in the majority of both male and female volunteers (Fig. 6). A
shape difference was seen between labra on both hips in approximately 15 % of the overall study group. There was no statistically significant difference between males and females as to the symmetry or asymmetry of labrum shape on both hips (p > 0.05). Obviously, absent labra on both hips were considered to be ªsymmetrical,º whereas any combination of labral shape with a unilateral absent labrum was considered to be ªasymmetrical.º The frequency of unilateral intralabral intensity increase (in the presence of a contralateral labrum with no intensity increase) was different in various age groups (Fig. 7). This condition, i. e., asymmetrical intralabral intensity increase, was more common between the ages of 40±49 years in both genders and both MR sequences (p < 0.05). Number and percentage of volunteers with a size difference of more than 25 % between their right and left labra were calculated (Figs. 8, 9). ªTwenty-five percentº of size difference was a tentative preference with the thought in mind that more than one-quarter size difference might be discernible to the observer's eyes on MR images of the acetabular labrum. Volunteers without visible labrum (i. e., ªabsent labrumº category) were
571
Fig. 5 Percentage of volunteers with intralabral intensity increases on T2-weighted gradient-echo MR images by age. Volunteers with unilateral or bilateral absence of labrum were excluded
excluded from total in the calculation of percentages of interlabral size differences. Interlabral size difference of more than 25 % was seen in approximately one-fourth of the overall study group (Fig. 8). A size difference of more than 25 % between both acetabular labra on the same individual was especially evident between the ages of 40 and 49 years; Fig. 9); however, this was only true (statistically significant) for males in this age group (p < 0.05). Females in this age group, as is seen in Table 5, which shows the labrum sizes, did not exhibit a size difference.
Discussion Lesions of the acetabular labrum are among the causes of hip pain. Magnetic resonance imaging, with its excellent soft tissue contrast and multiplanar capability, is currently the procedure of choice to non-invasively depict this anatomic structure. It is obvious that a thorough understanding of the normal MR imaging characteristics of the acetabular labrum is essential in order to be able to detect pathological conditions. On the other hand, fibrocartilaginous structures have a tendency to show age-related changes in MR imaging owing to compositional alterations over long periods of time, and it is beneficial to study the imaging correlation of the normal ªagingº process of the acetabular labrum. Two distinct studies have been reported in the literature regarding the MR imaging characteristics of the
Fig. 6 Percentage of volunteers with interlabral shape symmetry±asymmetry on MR images by gender. Absent labra on both hips were considered to be ªsymmetrical,º whereas any combination of labral shape with a unilateral absent labrum was considered to be ªasymmetricalº
acetabular labrum [10, 11]. Lecouvet et al. reported the MR imaging features of the labrum in 200 hips of 200 asymptomatic volunteers with an age range of 15±82 years [10]. Cotten et al., on the other hand, explored the same subject in 52 hips in 46 volunteers (both hips of the same individual having been examined in six persons) [11]. There are some discrepancies, however, in the results of these studies, especially with regard to the frequencies of labrum shape. Moreover, comparative MR imaging evaluation of both acetabular labra in the same individual in the literature has been limited only to the six cases in the latter study whose both labra have been examined with MR imaging, and who reportedly displayed identical labral appearance on both hips [11]. It is not an infrequent practice for radiologists to study the other ± presumably normal ± side in evaluating the imaging findings of a certain anatomic structure, especially when having difficulties due to the lack of experience or adequate reported data regarding the imaging features of that particular structure. Our study, therefore, was designed to make, for the first time, a comparative MR imaging evaluation of both acetabular labra on the same asymptomatic individual in a large population, as well as to attempt to shed light into the inconsistencies of the previous reports. Initially, textbook definition of the cross-sectional midcoronal MR imaging shape of the acetabular labrum was triangular [12]. Later, however, the two studies of normal MR imaging features of the acetabular labrum mentioned previously revealed varying percentages of
572
Fig. 7 Percentage of volunteers with unilateral intralabral intensity increase on T1-weighted and/or T2-weighted gradientecho MR images by age and gender. The frequency of unilateral intralabral intensity increase (in the presence of a contralateral labrum with no intensity increase) is different in various age groups. This condition, i. e., asymmetrical intralabral intensity increase, is more common between the ages of 40±49 years in both genders and both MR sequences
Fig. 8 Percentage of volunteers with interlabral size asymmetry of over 25 % on MR images by gender. Volunteers with unilateral or bilateral absence of labrum were excluded
triangular cross-sectional appearance of labrum [10, 11]. Of these, Lecouvet et al.'s [10] report stated a triangular labrum percentage of 66 %, whereas Cotten et al. [11] reported various triangular labrum percentages of 86±94 % depending on the aspect of the labrum (anterior, superior, or posterior). Our study has an overall triangular labrum percentage of 69 % which seems to be more in keeping with Lecouvet et al.'s work, which has a higher number of hips investigated than in the study by Cotten et al. The two previous studies of MR imaging of the labrum in asymptomatic hips reported different but close
percentages of ªabsentº (unidentified) labrum. Lecouvet et al. [10] reported the labrum to be absent in 14 % of their study group, whereas Cotten et al. [11] put this percentage at 10 % at the anterosuperior aspect of the labrum in the presence of a small adjacent superior labrum. Our study has an overall absent labrum rate of 2.5 % on midcoronal MR images, far less common than in Lecouvet et al.'s study group. This may reflect anthropologic variations in study populations, given the fact that the age and gender properties of the two groups do not seem to have significant difference. We cannot speculate about the ªrealº absence of acetabular labrum in our study group, however, because we considered the coronal images only, as did Lecouvet et al., unlike Cotten et al., who also evaluated the anteriormost and posteriormost parts of the labrum on axial images. Still, we analyzed the whole set of coronal images to identify a labrum and reported an ªabsentº labrum only after that analysis. Our study is the first to extensively compare both acetabular labra on the same asymptomatic individual. Previously, both labra in the same person without symptoms related to the hip joint were examined in only 6 of the 46 volunteers in Cotten et al.'s study who were reported to have identical findings on both sides [11]. Our study, however, disclosed an acetabular labral shape asymmetry of approximately 15.6 %. The presence of intralabral intensity increases in asymptomatic volunteers has already been described [10, 11]. Lecouvet et al. found an age-related increase of the frequency of intralabral high-intensity signal in their study which utilized T1-weighted MR images only [10]. Cotten et al., however, did not report an age-related pattern of intralabral intensity increase, although they did find this occurrence to be more common on T1weighted (and proton-density weighted) images. Our study showed an age-related gradual increase of intralabral high-intensity signal on both T1-weighted and T2-
573
Fig. 9 Percentages of volunteers with labral size asymmetry of a threshold of 25 % on MR images by age. Volunteers with unilateral or bilateral absence of labrum were excluded
weighted images. The higher frequency of intralabral intensity increase on T1-weighted gradient-echo images in our study is in keeping with the findings of Cotten et al. [11]. The absence of an age-related pattern in these authors' study, however, has been attributed to a poor correlation between intralabral signal alterations and histologic evidence of degeneration reported in a previous study [7]. Hodler et al., in their study of the MR arthrographic evaluation of the acetabular labrum with histologic correlation, have speculated that the increased intensity at the base of the labrum could be due to the presence of small fibrovascular bundles or an irregular zone of labral insertion which could possibly contribute to signal alterations in the absence of degeneration [7]. Our results, however, do not reveal a significantly higher number of young asymptomatic volunteers showing ªbasalº intensity increase on their labrum; therefore, we believe that it is safe to conclude that there actually is an age-related intensity increase both on T1-weighted and T2-weighted gradient-echo MR images of the acetabular labrum. We found the asymmetrical (unilateral) intralabral intensity increase to be significantly more common between the ages of 40 and 49 in both genders on both
gradient-echo MR sequences. This can represent varying rates of degeneration of each acetabular labrum in the same individual possibly due to asymmetrical weight-bearing. We have also made normative measurements of acetabular labrum sizes on midcoronal images by calculating cross-sectional areas. Not surprisingly, males exhibited significantly greater results, reflecting probably their overall bigger body size and possibly increased workload of the hip joint. It is of interest that our study revealed a cross-sectional size difference of over 25 % between two labra on each hip of the same individual in approximately one-quarter of the study group. The 25 % figure has been selected to represent a threshold for observers to visually discern the size differences between both labra in the same individual. The size difference of over 25 % between both labra in the same asymptomatic individual may reflect the consequences of slight differences of weight bearing on each side possibly due to postural abnormalities. There are some limitations of our study. The inclusion of both hips in the same sequence resulted in a field of view of 25 or 26 cm with a yield of limited resolution of available images. Smaller surface coils can be utilized to image a single hip (and labrum) with higher resolution and smaller fields of view which can be more effectively achieved with higher magnetic field strength (especially 1.5 T) MR systems. Moreover, our study has not evaluated the anterior and posterior aspects of the acetabular labrum. The variable appearance of the labrum in coronal and axial MR images have been documented [11]. Failure to identify a labrum on midcoronal MR images of the hip does not necessarily imply its complete absence. Finally, variability in hip positioning or slice selection could lead to important size differences between both labra of the same individual. To overcome this, we positioned the patients, coil, and slices with utmost care. In conclusion, our study confirms the previous reports of variability in acetabular labral shape and agerelated changes in its signal intensity. Moreover, the possibility of variability of the labra, with regard to the shape and size, on both hips of the same individual is documented.
References 1. Shea KP, Kalamchi A, Thompson GH (1986) Acetabular epiphysis-labrum entrapment following traumatic dislocation of the hip in children. J Pediatr Orthop 6:215±219 2. Ikeda T, Awaya G, Suzuki S, Okada Y, Tada H (1988) Torn acetabular labrum in young patients. J Bone Joint Surg Br 70-B:13±16
3. Dorrell JH, Catterall A (1986) The torn acetabular labrum. J Bone Joint Surg Br 68-B:400±403 4. Klaue K, Durnin CW, Ganz R (1991) The acetabular rim syndrome: a clinical presentatiton of dysplasia of the hip. J Bone Joint Surg Br 73-B:423±429
5. Suzuki S, Kasahara Y, Seto Y, Futami T, Furukawa K, Nishino Y (1994) Arthroscopy in 19 patients with Perthes' disease. Acta Orthop Scand 65:581±584 6. Altenberg AR (1977) Acetabular labrum tears: a cause of hip pain and degenerative arthritis. South Med J 70:174±175
574
7. Hodler J, Yu JS, Goodwin D, Haghighi P, Trudell D, Resnick D (1995) MR arthrography of the hip: improved imaging of the acetabular labrum with histologic correlation in cadavers. AJR 165:887±891 8. Czerny C, Hofmann S, Neuhold A, Tschauner C, Engel A, Recht MP, Kramer J (1996) Lesions of the acetabular labrum: accuracy of MR imaging and MR arthrography in detection and staging. Radiology 200:225±230
BOOK RE VIEW Allan P. L., Dubbins P. A., Pozniak M. A., McDicken W. N. (Editors): Clinical Doppler ultrasound. London Edinburgh: Churchill Livingstone, 2000, 293 pages illustrated, £ 44.95, ISBN 0-443-05549-1 This is a comprehensive guide to Doppler ultrasound, edited by four leading specialists in the field of ultrasound and Doppler techniques. All the contributors are wellqualified European and American experts. Like many books covering aspects of an imaging modality, there is an introductory presentation of physics principles and the practice of Doppler techniques. A large amount of space is devoted to a detailed description of the most frequently occurring artefacts. Actual safety statements are given. A second chapter describes the prin-
9. Czerny C, Hoffmann S, Urban M, Tschauner C, Neuhold A, Pretterkllieber M, Recht MP, Kramer J (1999) MR arthrography of the adult acetabular capsular-labral complex: correlation with surgery and anatomy. AJR 173:345±349 10. Lecouvet FE, Vande Berg BC, Malghem J, Lebon CJ, Moysan P, Jamart J, Maldague BE (1996) MR imaging of the acetabular labrum: variations in 200 asymptomatic hips. AJR 167:1025±1028
11. Cotten A, Boutry N, Demondion X, Paret C, Dewatre F, Liesse A, Chastanet P, Fontaine C (1998) Acetabular labrum: MRI in asymptomatic volunteers. J Comput Assist Tomogr 22:1±7 12. Stoller DW, Maloney WJ (1993) The hip. In: Stoller DW (ed) Magnetic resonance imaging in orthopaedics and sports medicine. Lippincott, Philadelphia, p 57
ciples of fluid mechanics applied to blood flow, and blood flow measurement. These two chapters are both clear and concise, and would be useful for those undertaking Doppler studies so they are aware of the basic principles of the techniques they are using. There is a section, albeit short, on ultrasound contrast agents and new modalities such as harmonic modes. At the end of the book, an appendix is devoted to system controls and their use. The main part of the book is divided into twelve chapters dealing with the main fields of application of Doppler imaging, including cervical vessels and transcranial Doppler; peripheral arteries and veins; aorta and inferior vena cava; liver, native and transplanted kidneys; prostate, penis and scrotum; female pelvis and pregnancy. These chapters are organized in a similar, didactic format that allows the reader to progressively move from anatomy, to Doppler technique, normal ultrasound findings,
and assessment of diseases. Finally, the pros and cons, and the accuracy of Doppler imaging in relation to other techniques ± mainly CT ± are discussed. The book has many attractive features, including well-organized and easy-to-follow text, selected references, and useful imaging protocols for various clinical circumstances. Numerous tables summarize clinical indications and Doppler diagnostic criteria. The quality of the drawing, the abundance of diagrams and illustrations, most of them in colour, provide clear, wellannotated examples of the major Doppler findings. The book is attractively organized in a practical fashion. It is highly recommended to anyone who is interested or active in the field of Doppler techniques, and should be of interest to a large range of disciplines. This 293-page book is available at a reasonable price. M. Claudon, Vandoeuvre
