| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
1 Departments of Pediatric Radiology (S.G.F.R., M.H.L., M.M.)
2 Pediatric Orthopedic Surgery (A.F.M.D.), Sophia Children's Hospital, Dr Molewaterplein 60, 3015 GJ Rotterdam, the Netherlands
3 Departments of Pathology (J.C.d.H.)
4 Anatomy (C.A.C.E.), Erasmus University, Rotterdam, the Netherlands.
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Six cadaveric specimens were imaged with ultrasonography (US) with special attention to the anterior joint capsule. Subsequently, two specimens were analyzed histologically. These anatomic findings were correlated with the US findings in 58 healthy children and 105 children with unilateral transient synovitis.
RESULTS: The anterior joint capsule comprises an anterior and posterior layer, mainly composed of fibrous tissue, lined by only a minute synovial membrane. Both fibrous layers were identified separately at US in 98 of 116 (84%) hips of healthy subjects and in all hips with transient synovitis. Overall, the anterior layer was thicker than the posterior layer. In transient synovitis compared with normal hips, no significant thickening of both layers was present (P = .24 and .57 for the anterior and posterior layers, respectively). Normal variants include plicae, local thickening of the capsule, and pseudodiverticula.
CONCLUSION: Increased thickness of the anterior joint capsule in transient synovitis is caused entirely by effusion. There is no US evidence for additional capsule swelling or synovial hypertrophy.
Index terms: Hip, anatomy, 44.92 • Hip, infection, 44.26 • Ultrasound (US), guidance, 44.12986
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The aim of this study was to examine the anatomy of the joint capsule of the hip in a plane that is relevant to US imaging and to investigate the appearance of the joint capsule in transient synovitis.
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The study was approved by the institutional review board. In the study of the healthy children, informed consent was obtained from parents and children before the examination.
Cadaveric Study
Six cadavers were available for imaging, anatomic, and histologic examinations.
The cadavers were not fixated during imaging studies. All cadavers were examined with US by using a 7–10-MHz linear array transducer (Ultramark 9 HDI; Advanced Technical Laboratories, Bothell, Wash).
After imaging, two cadavers were deep frozen and sawed, by using a band saw (five teeth per inch), into 5-mm-thick slices parallel to the femoral neck, identical to the US plane, with use of a technique described by Entius et al (11) modified for the hip joint. This was followed by histologic examination of the slices; the tissues were fixed in formalin and embedded in paraffin. Sections 10 µm thick were cut and routinely stained with hematoxylin-eosin.
With US guidance, the remaining four cadavers (eight hips) were injected with 10 mL of saline solution (0.9% NaCl). The needle entered the joint cavity at the level of the labrum to preserve the anatomy of the anterior recess. Anatomic landmarks, such as the layers of the anterior joint capsule, were identified before and after administration of saline solution.
Healthy Children
Fifty-eight asymptomatic children without a history of hip disease were examined with US. Both hips were examined (n = 116). There were 37 boys and 21 girls, aged 1.7–12.5 years (mean, 6.7 years).
Patients with Transient Synovitis
Between January 1994 and May 1997, 105 consecutive patients with transient synovitis were examined with US in a prospective study. Patients with bilateral involvement or fever were excluded. There were 74 boys and 31 girls, aged from 2 to 12.8 years (mean, 6.0 years). The right hip was involved in 57 patients and the left hip in 48.
Patients with an irritable hip were considered to have transient synovitis if US depicted a thickening of the joint capsule of more than 2 mm compared with that in the asymptomatic hip (6,7,10,12) and the complaints subsided completely within 4 weeks without specific therapy and the patient remained symptom free for at least 6 months thereafter.
Methods
The US examinations were performed by one investigator (S.G.F.R.) with use of US equipment with high-frequency 7–10-MHz (Ultramark 9 HDI; Advanced Technology Laboratories) and 7-MHz (model 128 XP10; Acuson, Mountain View, Calif) linear array transducers.
The children were examined in the supine position with the hips in neutral position (extension and slight external rotation). An anterior approach along the long axis of the femoral neck was used to visualize the anterior capsule of the hip joint to the best advantage (13,14). A similar approach was used in the cadaver study. In children with synovitis, the contralateral asymptomatic hip was used as the normal reference.
The anterior joint capsule was identified, and the thickness of the capsule was assessed by measuring the maximal distance between the anterior surface of the femoral neck and the posterior surface of the iliopsoas muscle (15).
Moreover, in patients with transient synovitis, the following parameters were examined: (a) identification and measurement of both layers of the anterior joint capsule in both symptomatic and asymptomatic hips, (b) identification and characterization (clear or turbid) of the effusion in the anterior recess of the joint capsule, and (c) evaluation of the anterior contour of the joint capsule.
Statistical Analysis
The difference in thickness of the anterior joint capsule between both sides, as well as the difference in thickness of the various layers of the anterior joint capsule of both hips, were tested by means of the paired t test. The correlation between the thickness of the anterior joint capsule and age was examined by calculating Pearson correlation coefficients. A P value less than .05 was considered statistically significant.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
At the junction of the anterior layer of the joint capsule with the labrum, a slitlike indentation is identified (Fig 2a). On macroscopic inspection, this simulates an artificial tear. Microscopically, however, this indentation is lined by synovial cells and therefore represents an anatomic structure. US examination sometimes shows this structure as a hypoechoic triangular area (caused by a small amount of physiologic joint fluid, Fig 2b). This slitlike indentation is known from arthrographic studies as a small extension of contrast material adjacent to the labrum and represents the superior articular recess.
|
|
|
|
|
|
|
|
|
|
|
|
In the majority of the children (81%), the anterior layer of the joint capsule had the same echogenicity as the posterior layer, being isoechoic to psoas muscle in 48%, hyperechoic in 31%, and hypoechoic in 2%. In 19% of the children, both layers had a different echogenicity; the anterior layer always showed increased echogenicity compared with that of the posterior layer (Fig 8).
|
Patients with Transient Synovitis
Anterior capsule of the hip joint.—The anterior capsule could be visualized in all hips, both symptomatic and asymptomatic. In all symptomatic hips, both layers of the capsule could easily be distinguished from surrounding bone, muscle, and anechoic effusion and could be measured (Table 1, Fig 9).
|
|
Similar to findings in the study of healthy subjects, the thickness of the anterior joint capsule of the asymptomatic hip showed no correlation with age (P = .1). However, the amount of effusion in the symptomatic hip did show a positive correlation with age (P = .001).
In 50 of the symptomatic hips (48%), a local thickening of the posterior layer of the anterior joint capsule, referred to as the "hump," was visible. The thickness of this hump varied considerably, but it was invariably localized at the insertion of the posterior layer near the cartilage of the femoral head (Fig 10a). This phenomenon was almost exclusively observed in hips with effusion. Occasionally, an identical structure was seen on the anterior layer (Fig 10b).
|
|
|
Anterior bulging of the capsule.—The results are shown in Table 2. Convex bulging of the anterior joint capsule has a sensitivity of 94%, specificity of 91%, positive predictive value of 92%, and negative predictive value of 94% for effusion.
|
Additional findings.—In four patients (4%), a linear structure traversing the fluid-filled anterior recess, identical to the structure seen in one of the cadaveric hips (Fig 6a), demonstrated the same echogenicity and texture as did both layers of the anterior joint capsule.
In two patients (2%), a thin-walled cystic protrusion of the joint effusion was seen. This arose from the synovial surface of the anterior layer of the capsule and protruded through the anterior layer into the space between the iliopsoas muscle and anterior border of the joint capsule with a collar-button configuration (Fig 12).
|
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In numerous reports, the thickness of the anterior joint capsule is addressed without a discussion of its separate layers (4,6,7,9,14,20).
Findings in the present cadaveric study show that the anterior joint capsule is composed of two layers. Both layers have a thickness substantial enough to enable US measurements with state-of-the-art equipment. This was confirmed at US in the adult cadavers that demonstrated both layers of the anterior joint capsule (Fig 5). Knowledge of this anatomy is essential for accurate measurements and good interpretation.
The border between the iliopsoas muscle and the joint capsule is visible as a hyperechoic line (Fig 7a), but it does not represent an anatomic structure (Fig 4b). This line should not be interpreted as the fibrous capsule (14,15,18,19,21,22) but merely represents the interface between muscle and joint capsule. The thickness of this line depends on, among other factors, the transducer frequency (23–25).
In previous studies, a wide range of values are reported for thickness of the anterior joint capsule. These values can be divided into two main groups: those with a mean of 5 mm, comparable to that in the present study (6,15,23,26,27), and those with a mean of 2–3 mm (2,4,12,13,28). This discrepancy was already observed by Terjesen and Osthus (7), who found it "difficult to explain, as similar US techniques and anatomic landmarks seem to have been employed." Apparently, this discrepancy is largely caused by unfamiliarity with the anatomy. In the latter studies, the distance was measured between the femoral neck and the interface between both layers of the joint capsule (stripe sign); therefore, only the posterior layer of the anterior joint capsule was measured instead of the total capsule. Even the cartilage of the femoral head was sometimes mistaken for anterior recess (12).
According to Rohrschneider et al (15), there was no statistically significant correlation between age and thickness of the anterior joint capsule in normal hips (in children aged more than 3 years). This finding was confirmed in the present study.
Histologic examination shows that both layers of the joint capsule are composed mainly of collagen fibers lined by only a thin synovial membrane. The thickness of the synovial membrane is approximately 0.025 mm (29), which exceeds the spatial resolution of modern radiologic techniques such as US. Moreover, the synovial membrane of the anterior recess is of the fibrous type. In contrast to the areolar and adipose types of synovial membrane—in which the synovial intima is separated from the fibrous capsule by loose connective tissue or adipose tissue—in the fibrous type, the synovia rests directly on the fibrous layer (30). This also contributes to the inability of US to visualize the synovial membrane as a distinct layer.
In the present study, the anterior layer showed an increased echogenicity compared with that in the posterior layer in 19% of the hips of healthy subjects. This difference in echogenicity may, in part, be attributed to the difference in fiber texture of both layers, as demonstrated in the histologic examination (Fig 4a).
The superior articular recess can be seen at US as a small band of decreased echogenicity adjacent to the labrum at the insertion of the joint capsule at the labrum. This should not be mistaken for a pathologic condition, such as a rupture.
The Anterior Joint Capsule in Transient Synovitis
Results in the study of patients with transient synovitis are compatible with those in the cadaveric study: Both layers of the joint capsule can be readily visualized and measured separately with US in all hips, facilitated by the effusion. In the contralateral normal hip, the thickness of the anterior and posterior layers of the anterior joint capsule could be measured in 79% of the patients and compared with measurements in the symptomatic hip.
In both symptomatic and asymptomatic hips, the anterior layer of the joint capsule was significantly thicker than the posterior layer (Table 1), which is consistent with findings in the cadaveric study. The anterior layer is thicker probably because it is reinforced by the iliofemoral ligament and the orbicular zone.
Moreover, the anterior and posterior layers of the anterior joint capsule showed no statistically significant difference between the symptomatic and asymptomatic sides (Table 1). Synovitis apparently does not cause a measurable thickening of the anatomic components of the joint capsule. Theoretically, the synovial membrane is thickened in synovitis, but because it forms only a minute part of both layers of the anterior joint capsule, its thickening cannot be appreciated at US.
In many articles dealing with transient synovitis, both layers of the anterior joint capsule are either not identified or not mentioned, probably due to use of inadequate equipment or unfamiliarity with the anatomy of the anterior joint capsule. However, those articles that do describe both layers erroneously identify them as synovium or synovial hypertrophy or thickening (3,9,14,18,28). According to results in the present study, neither of the layers of the anterior joint capsule are diffusely thickened in transient synovitis. Moreover, "synovium or synovial membrane" is a misnomer because the joint capsule is composed of a fibrous capsule with only a minute layer of synovial membrane.
Because (a) both layers of the anterior joint capsule are not thickened, and (b) the effusion was always discernible, it seems more rational to detect effusion by visualizing the effusion itself rather than by relying on indirect signs such as differences between hips in measurements of the joint capsule (4–10,28). In cases of poor US visualization of the joint capsule, this indirect method can be a good alternative, although it should be realized that it fails when bilateral effusions are present.
Hump of the Posterior Layer
This phenomenon was observed almost exclusively in hips with effusion, probably because the effusion facilitates demarcation of the synovial surface. This hump could be interpreted as debris or flocculation, but several facts argue against this. (a) In some patients, we were able to demonstrate vessels in this hump at Doppler US; and (b) in the supine position, the hump does not migrate to the most dependent section of the recess. The stripe sign phenomenon does not offer enough anatomic detail to confirm the presence of such a local thickening in hips without an effusion. However, evidence of a hump was present in one asymptomatic hip, and moreover we observed this structure in one cadaveric hip (Fig 6b). Apparently this local hump is not restricted to synovitis but should be regarded as a normal anatomic landmark representing the insertion of the capsule in the femoral neck. Marchal et al (13) observed an identical local thickening in a patient with septic arthritis, but findings in the present study demonstrate that this sign certainly is not pathognomonic for septic arthritis.
Effusion
Because both layers of the anterior joint capsule are not thickened in transient synovitis, the widening of the anterior recess can be attributed solely to effusion. In all symptomatic hips, the effusion could be readily discriminated from the layers of the anterior joint capsule and measured separately.
In nine patients, the effusion was not completely clear. None of these patients had clinical signs of septic arthritis, and all had an uneventful recovery. This phenomenon was also observed by Dörr et al (18) and Marchal et al (13) in 21% and 10% of their patients, respectively, and Zawin et al (3) demonstrated that turbid effusion is not diagnostic for infection. We found that the mean duration of complaints in patients with turbid effusion was considerably longer than in the patients with a clear effusion (7.2 vs 4 days, P < .001). This may be the result of progressive accumulation of cellular debris in effusion of longer duration. Obviously, the presence of turbid effusion is associated with the duration of the disease, but it has no further diagnostic value.
Surprisingly, the amount of fluid in transient synovitis shows a positive correlation with age (P < .001). Assuming that the pathophysiologic mechanism of transient synovitis is age independent, this correlation must be attributed to geometric factors. Because the anterior recess is larger in older children, the same pressure will induce more expansion, according to the physical law of Laplace.
A small layer of joint fluid in the anterior recess of the asymptomatic hip was present in 12 patients (11%), with a mean thickness of 1.0 mm and a maximum of 1.6 mm (Fig 7b). This finding is similar to that of Rohrschneider et al (15), who found this in 12% of healthy children, with a maximum thickness of 1.5 mm. Therefore, 2 mm seems to be a sound threshold to differentiate a pathologic from a physiologic effusion. This is in keeping with a difference of more than 2 mm between both symptomatic and asymptomatic anterior joint capsules that is considered pathologic in several reports (7,8,12–15,19,28). The stripe sign, representing the collapsed anterior recess, can be of additional value in excluding small amounts of effusion in the joint. This sign is visible only when both layers lie close together, that is, in the absence of effusion (Figs 5, 7a).
Bulging of the Capsule
Some authors consider anterior bulging of the joint capsule the best criterion for effusion (1–3). However, the contour of the joint capsule depends on rotation, being concave in exorotation and convex in endorotation (15). Moreover, joint capsules of normal hips can have a straight or convex contour (15). This is also true in the present study, in which 9% of the asymptomatic hips showed anterior bulging in the absence of effusion, making this sign less reliable. Six percent of the symptomatic hips had no convex contours, despite the fact that a substantial amount of effusion was present. On the other hand, a concave border of the anterior joint capsule seems to be a reliable indicator for the absence of effusion, because in the present study, none of these hips showed joint fluid at US.
Plica
In six patients (6%), a platelike structure traversed the effusion, mostly running from the hump of the posterior layer to the anterior layer. The structure and echogenicity were identical for both layers of the anterior joint capsule. This structure was also demonstrable in one of the cadaveric hips after intraarticular injection of saline solution. It probably represents a plica, comparable to the well-known plicae in the knee joint.
None of the patients with such a plica had previous complaints, and all had an uneventful recovery.
Diverticulum
The cystic protrusion of joint effusion through the anterior layer of the anterior recess probably represents a herniation of the synovial membrane through a defect in the fibrous capsule, which creates a pseudodiverticulum of the synovial membrane. The thin wall also suggests it is a pseudodiverticulum of the synovial membrane rather than a real diverticulum of the joint capsule.
Conclusions
Findings in this study show that the anterior capsule of the hip joint is composed of anterior and posterior layers that were visualized separately at US in 84% of healthy children. Both layers are composed mainly of fibrous tissue (representing the fibrous capsule) and lined by only a minute layer of synovial membrane, which is too thin to be visualized separately at US. No measurable thickening of both layers of the anterior joint capsule is present in transient synovitis. Therefore, enlargement of the anterior joint capsule in transient synovitis is caused solely by the presence of effusion.
It is more rational to detect effusion by visualizing the effusion itself than by relying on indirect signs such as contours of the capsule or differences in joint capsule measurements between hips. These indirect methods can be good alternatives only in cases of poor US visualization of the joint capsule, although the latter method fails when bilateral effusions are present.
Knowledge of the US anatomy of the capsule of the hip joint is essential for future studies, especially since state-of-the-art US equipment with high-frequency transducers allows visualization of the hip joint in greater detail. Correct identification of both layers, especially the posterior layer, will prevent erroneous measurements and interpretation.
Moreover, findings in this study describe normal variants that can be visualized with state-of-the-art US equipment. Sonologists should be aware of these findings to prevent misinterpretation of normal variants mistaken for pathologic conditions.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, S.G.F.R.; study concepts and design, S.G.F.R.; definition of intellectual content, S.G.F.R., M.M.; literature research, S.G.F.R.; clinical studies, S.G.F.R., M.H.L., A.F.M.D.; experimental studies, S.G.F.R., C.A.C.E.; data acquisition, S.G.F.R., C.A.C.E., J.C.d.H.; data analysis, S.G.F.R.; statistical analysis, S.G.F.R.; manuscript preparation, S.G.F.R.; manuscript editing, S.G.F.R., M.H.L.; manuscript review, S.G.F.R., M.M.
Received March 10, 1998;
revision requested May 12, 1998; revision received July 27, 1998;
accepted September 1, 1998.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. N. Weybright, J. A. Jacobson, K. H. Murry, J. Lin, D. P. Fessell, D. A. Jamadar, M. Kabeto, and C. W. Hayes Limited Effectiveness of Sonography in Revealing Hip Joint Effusion: Preliminary Results in 21 Adult Patients with Native and Postoperative Hips Am. J. Roentgenol., July 1, 2003; 181(1): 215 - 218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G. F. Robben, M. H. Lequin, A. F. M. Diepstraten, W. C. J. Hop, and M. Meradji Doppler Sonography of the Anterior Ascending Cervical Arteries of the Hip: Evaluation of Healthy and Painful Hips in Children Am. J. Roentgenol., June 1, 2000; 174(6): 1629 - 1634. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
