 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. The Teardrop in Congenital Dislocation of the Hip Diagnosed Late. A Quantitative Study*  , IOWA CITY, IOWAInvestigation performed at The University of Iowa Hospitals and Clinics, Iowa City
We retrospectively reviewed the radiographs of the pelvis and hips of forty-five patients who had unilateral congenital dislocation of the hip treated with closed reduction and application of a cast without subsequent operations. The radiographs were made at the time of the initial diagnosis, two years after the reduction, when the child was ten years old, and at skeletal maturity. The width, shape, and type of the teardrop; the thickness of the acetabular floor; the acetabular index; the center-edge angle; the articulotrochanteric distance; and the Severin class at maturity were measured in the dislocated and contralateral, normal hips. At the time of the initial diagnosis, a well defined teardrop was seen in thirty-six (80 per cent) of the normal hips and in seven (16 per cent) of the dislocated hips. There was no difference in the width of the teardrop in the seven dislocated hips compared with that in the normal hips, although the v-shaped and crossed types of teardrops were more frequent in the dislocated hips. The v shape was not observed in the normal hips but was seen in sixteen dislocated hips two years after the reduction and in twelve dislocated hips when the children were ten years old. The superior and inferior widths of the teardrop of the dislocated hips were significantly greater than those of the normal hips (p < 0.001 and p < 0.05, respectively) when the children were ten years old. The hips with residual acetabular dysplasia had a v-shaped teardrop, widening of the superior width of the teardrop, and thickening of the acetabular floor. These hips, which were usually Severin class IV at the time of skeletal maturity, had a poor prognosis in adult life.
Kahle and Coleman14 as well as Peic23 noted that, in congenital dislocation of the hip, the teardrop was not visible until the femoral head was reduced. However, they did not give any quantitative data to support this statement. Widening of the teardrop has been associated with residual dysplasia and subluxation of the hip after treatment5,7,14, but it has not been associated with the anatomical result at maturity. The association of widening with the anatomical result at maturity may be useful for determining the need for and timing of operative intervention to avoid the sequelae of residual acetabular dysplasia. Parameters such as the acetabular index (at the time of the initial diagnosis and after the reduction) and the thickness of the acetabular floor have not been found to be predictive6,8,9,22,30,35,37. We performed the present study to evaluate the evolution of the teardrop in children from the time of the initial diagnosis of congenital dislocation of the hip to skeletal maturity. The time of the appearance of the teardrop as well as its width, shape, and type were associated with the radiographic appearance of the hip at maturity.
We retrospectively reviewed the charts of 148 children who had been treated at The University of Iowa Hospitals and Clinics for congenital dislocation of the hip that was not diagnosed at birth. We excluded children who had teratological and neuromuscular dislocation or congenital malformation. We reviewed the radiographs of forty-five children (thirty-nine girls and six boys), from this cohort, who had a unilateral congenital dislocation of the hip and were at least six months old at the time of treatment with closed reduction. The left hip was dislocated in twenty-five patients (56 per cent). The mean age at the time of the initial diagnosis was twenty-one months (range, six to sixty-six months). Radiographs were made at the time of the initial diagnosis, two years after the reduction, when the child was ten years old, and at skeletal maturity. Radiographs of the pelvis had to show symmetry of the obturator foramen, iliac wings, and lumbar pedicles to be included in the study2 (Fig. 1). We used these criteria in order to avoid radiographs made with any pelvic tilt or rotation, both of which are known to change the dimensions of the teardrop26,36.
The closed reduction was performed with the child under general anesthesia. A hip-spica plaster cast was worn for three months to maintain the femoral head in a well centered position. After removal of the cast, the child wore an abduction brace at night and during naps. None of the patients had any additional operations. The age of the child when the lateral and medial boundaries of the teardrop were well delineated on the anteroposterior radiographs; the width, shape, and type of the teardrop; the thickness of the acetabular floor22; the acetabular index27,31; and the distance H (the perpendicular distance from the horizontal Hilgenreiner line to the proximal point of the femoral metaphysis30) were determined on the radiographs made at the time of the initial diagnosis, at two years after the reduction, and when the child was ten years old. The type of teardrop was based on its position relative to the ilioischial line3,12 (Fig. 2). The teardrop was considered open when the medial line was lateral to the ilioischial line, closed when the medial line was superimposed on the ilioischial line (Fig. 3-A), crossed when the medial line was medial to the ilioischial line (Fig. 3-B), and reversed when both the medial and the lateral line were medial to the ilioischial line (Fig. 3-C).
The shape of the teardrop, described as u or v, was determined on the basis of the relationship between the superior and the inferior width (Figs. 4-A and 4-B). The superior width was the maximum horizontal distance between the lateral line of the teardrop and the medial line near the triradiate cartilage. The distance from the lateral line to the ilioischial line was also measured when the medial line was poorly defined and when the teardrop was the crossed or reversed type. The inferior width was the maximum horizontal distance between the lateral and the medial line near the obturator foramen (Fig. 5).
The thickness of the acetabular floor22 was the distance between the most inferolateral and the most inferomedial point of the ilium near the triradiate cartilage (Fig. 5). The acetabular angle of Sharp29, the articulotrochanteric distance19, and the center-edge angle31 were measured on the radiographs made at the time of skeletal maturity. The radiographic appearance at maturity was classified according to the anatomical classification system of Severin28. Pairwise comparison of group means was performed with the Mann-Whitney U test. The p value at which the hypothesis was tested was 0.05.
The distance H, the thickness of the acetabular floor, and the acetabular index of the dislocated hips were significantly different from those of the contralateral, normal hips (p < 0.001) on the radiographs made at the time of the initial diagnosis (Table I). There was no association between the different types of teardrops and the age of the patient, and we did not find any specific distribution of closed, crossed, or reversed types of teardrops.
The teardrop was well defined on the initial radiograph of thirty-six (80 per cent) of the normal hips at a mean age of 22.5 months (range, ten to sixty-six months). The teardrop was not well defined in the remaining nine normal hips at a mean age of twelve months (range, six to eighteen months). Seven dislocated hips (16 per cent) had a well defined teardrop at the time of the initial diagnosis at a mean age of twenty-nine months (range, thirteen to sixty months): it was u-shaped in four and v-shaped in three. The teardrop was defined in all of the dislocated hips by a mean age (and standard deviation) of 27 ± 13.5 months. The u shape was seen on the initial radiograph of thirty-five (97 per cent) of thirty-six normal hips (Table I). The closed type was the most frequent type of teardrop in the normal hips (it was seen in twenty-nine [81 per cent]) as well as in the dislocated hips (it was seen in four of the seven hips) (Table I). Two years after the reduction, the distance H, the thickness of the acetabular floor, and the acetabular index of the dislocated hips were still significantly different from those of the normal hips (p < 0.001) (Table I). These differences were probably related to the late diagnosis and treatment and the resulting longer time needed for the acetabular dysplasia to resolve. All of the normal hips had a u-shaped teardrop. Twenty-nine (64 per cent) of the dislocated hips had a u-shaped teardrop, and sixteen (36 per cent) had a v-shaped teardrop. The superior width of the teardrop was significantly greater in the dislocated hips than in the normal hips (p < 0.001). The difference between the inferior widths was not significant (p = 0.05). The distance H, the thickness of the acetabular floor, the acetabular index, and the center-edge angle of the dislocated hips as measured on the radiographs made when the children were ten years old were significantly different from those of the normal hips (p < 0.001) (Table I). The shapes of the teardrop were similar in distribution to those seen two years after the reduction. There were more crossed-type teardrops than closed-type teardrops when the children were ten years old. The superior (p < 0.001) and inferior (p < 0.05) widths of the teardrop were significantly greater in the dislocated hips than in the normal hips. Both the superior and the inferior width were significantly (p < 0.05) greater when the center-edge angle was less than 15 degrees than when the center-edge angle was 15 degrees or more. At maturity, the difference in the mean (and standard deviation) acetabular angle of Sharp29 between the normal hips (40.6 ± 3.5 degrees) and the dislocated hips (44.3 ± 3.8 degrees) and the difference in the mean articulotrochanteric distance between the normal hips (26.0 ± 6.5 millimeters) and the dislocated hips (17.5 ± 13.1 millimeters) were significant (p < 0.001). Forty-three normal hips were Severin class IA, and one was Severin class IB. Seven dislocated hips (16 per cent) were Severin class IA, eight (18 per cent) were Severin class IB, five (11 per cent) were Severin class II, fourteen (32 per cent) were Severin class III, and ten (23 per cent) were Severin class IV. The radiographs made at maturity were not available for one patient. On the basis of the Severin classification at maturity, we formed two subgroups of dislocated hips and tested the previous parameters. Fifteen hips were Severin class I (A or B) and ten were Severin class IV (Table II). The mean age at the time of diagnosis was eighteen months (range, six to forty months) for the patients who had a Severin class-I hip and twenty-two months (range, twelve to sixty-six months) for the patients who had a Severin class-IV hip. One of the Severin class-I hips and four of the Severin class-IV hips had type-II avascular necrosis15. The inferior width of the teardrop in the Severin class-I hips and the Severin class-IV hips was not significantly different from that in the normal hips at two years after the reduction or when the children were ten years old. The superior width of the teardrop in the Severin class-I and class-IV hips at two years after the reduction (p < 0.05) and when the children were ten years old (p < 0.001 and p < 0.05, respectively) was significantly different from that in the normal hips (Table II). In the dislocated hips that were Severin class I or class IV at maturity, the superior width of the teardrop at two years and when the child was ten years old was greater than that in the normal hips, but this finding had no prognostic value. However, when the children were ten years old, a v-shaped teardrop was more frequent in the Severin class-IV hips (four of ten hips) than in the Severin class-I hips (one of fifteen hips). The acetabular floor was always thicker in the dislocated hips than in the normal hips; however, the difference was more significant for the Severin class-IV hips at the age of ten years (p < 0.001) than it was for the Severin class-I hips at the same age (p < 0.05) (Table II).
The anatomical boundaries of the teardrop in the acetabulum, as seen on anteroposterior radiographs of the pelvis, are a lateral line formed by the cortical surface of the middle third of the acetabular fossa, a medial line formed by the cortical surface of the true pelvic wall, and inferiorly the cotyloid notch of the obturator foramen3,4,10,14,16-18,21,23,24,26,32. The reported age at which the teardrop appears in normal children has ranged from six to twenty-four months23,27,31. Studies have shown how pelvic tilt and rotation can alter the radiographic appearance of the teardrop21,26,27. Its first appearance and the variations during growth in normal children were quantitatively observed by Gusis et al.12. Armbuster et al.3 described different types and shapes of the teardrop in the normal population. In the present study, we examined the relationship between the development of the teardrop from the time of the diagnosis and the radiographic result, based on the Severin classification28, at the time of maturity. All of our patients were managed with closed reduction and application of a cast without any subsequent operative intervention. This allowed us to observe the development of the teardrop in the contralateral, normal hip and to compare it with that in the dislocated hip during childhood and adolescence. A well defined teardrop was seen at the time of the diagnosis of the dislocation in thirty-six of the contralateral, normal hips; the nine hips that did not have a well defined teardrop were in girls who were less than eighteen months old. The age of the child at the time of the appearance of the teardrop has been evaluated in studies of congenitally dislocated hips. Scoles et al.27, in a study of 100 normal hips, fifty of which were in girls, pointed out that only approximately 35 per cent of the hips in the girls had a well defined teardrop when the child was two years old. Peic23 and Tönnis31 suggested that the teardrop is present in most normal children by the age of one year. Kahle and Coleman14 as well as Peic23 suggested that the teardrop is observed only after the femoral head has been reduced. However, there were no quantitative data to support these conclusions. The lateral line, which is poorly defined at the time of the diagnosis of a dislocation, develops after the femoral head has been reduced, as the presence of the head in the acetabulum is said to stimulate the development of the acetabulum11,14,23. The age of the patient could play a role in the spontaneous development of the lateral line when the femoral head remains dislocated. In our study, seven patients whose mean age was twenty-nine months had a well defined teardrop at the time of the diagnosis even though the hip was dislocated and the normal anatomical relationship between the proximal aspect of the femur and the acetabulum was not present. It is possible that these hips were only slightly displaced and thus there was some stimulation of growth of the acetabulum. The teardrop appeared approximately six months after the closed reduction in the other dislocated hips, and this probably represents the response to the normal stimulation of the acetabular ossification and growth by the reduced femoral head13,25. With growth, there was progressive narrowing of the teardrop in the normal hips12. We found that, with time, there was progressive widening of the teardrop in the dislocated hips; the increase was first in the superior width (at two years) and then in the inferior width (when the children were ten years old). This widening could have been caused by a growth disturbance of the acetabulum, as has been seen in experimental dislocation of the hip13. Papavasiliou and Piggott22 also described a thickening of the acetabular floor in congenital dislocation of the hip. The absence of a normal relationship between the acetabulum and the femoral head could be responsible for this failure of normal growth25. Furthermore, if the pelvis is rotated and tilted when the radiograph is made, the apparent width of the teardrop may increase26. A triplane pelvic deformity (lateral tilt, anterior bending, and rotation) has been demonstrated experimentally in unilateral dislocation of the hip1, and pelvic asymmetry has been described in unilateral congenital dislocation of the hip2. Pelvic deformity can also increase the width of the teardrop in congenital dislocation of the hip, and narrowing of the teardrop has been observed after rotational pelvic osteotomies for congenital dislocation of the hip7. All of the normal hips in the present study had a u-shaped teardrop by two years after the reduction. However, the teardrop in sixteen of the dislocated hips was v-shaped because of the widening of the teardrop superiorly. According to a study of 150 children12, the closed teardrop is the most frequently seen type in normal hips at an early age, but it gradually evolves into the crossed and reversed types in adolescents. In the dislocated hips in our study, the teardrop had a more crossed shape when the children were ten years old. This could be a result of the increased distance between the lateral line and the ilioischial line secondary to the growth disturbance and the pelvic deformity2. The radiographs made at maturity showed major differences between the normal and the dislocated hips, with regard to the acetabular angle of Sharp29 and the articulotrochanteric distance. None of the normal hips were Severin class IV (indicating a poor result), compared with ten of the dislocated hips. We analyzed the teardrop when the children were ten years old, at which time most of the acetabular growth was complete and the potential for spontaneous remodeling was minimum20. At this age, superior widening of the teardrop was a common feature in both Severin class-I and Severin class-IV dislocated hips. Nevertheless, the width was greater in the hips with a rating of Severin class IV at maturity than in those with a rating of Severin class I. The v-shaped teardrop and the difference in the thickness of the acetabular floor at the age of ten years (Table II) suggest that the hips will have a poor radiographic result at skeletal maturity and a poor functional result in adult life33,34. The teardrop is not v-shaped in normal hips, and only one Severin class-I hip had a v-shaped teardrop when the child was ten years old. This shape could have been due to poor remodeling of the central aspect of the acetabulum and good peripheral development, resulting in a satisfactory radiographic and functional outcome. We speculate that the initial closed reduction in most of these patients was probably not concentric because of the disparity between the acetabulum and the femoral head that is present in patients who were not managed at birth. Therefore, normal anatomical congruency and stability of the hip could never be fully achieved. A normal femoral head is essential for normal acetabular growth and development1,13. Despite studies on the growth and development of dislocated hips, investigators have not yet identified predictors with which to determine the need for and timing of reconstructive operations and thus to avoid the adverse sequelae of untreated acetabular dysplasia6,8,9,33,37. The continued presence of a v-shaped teardrop with widening of its superior width and thickening of the acetabular floor are suggestive of residual acetabular dysplasia. Most hips with such findings are Severin class IV at skeletal maturity and thus are associated with a poor prognosis in adult life5. In these patients, reconstructive procedures should be considered in an attempt to avoid the adverse sequelae of persistent dysplasia and subluxation of the hip.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding sources were Fondo de Investigation Sanitaria (93/7018.Madrid, Spain) and the St. Giles Foundation of New York City.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |