Although many studies have been done on Perthes disease, the exact pathogenesis of the disease remains unknown. 6–16,18 An episode of ischemia of the capital femoral epiphysis is the basic cause of Perthes disease. 2 However, the cause of this impairment of the blood supply remains unknown. The vascular anatomy of the capital femoral epiphysis may be related to the onset of Perthes disease. Normal vascular anatomy of the femoral head in children was investigated using microangiography of autopsy specimens. 5,17 During growth, there are two different vascular sources in the capital femoral epiphysis. One is the lateral epiphyseal arteries, which supply blood to the extensive area of the capital femoral epiphysis penetrating from the lateral subcapital region. 17 The interruption of blood supply at the stump of the lateral epiphyseal arteries has been reported. 1,6,16 Another source of the nutrient artery of the capital femoral epiphysis is the artery of the ligamentum teres. Trueta 17 reported that the artery of the ligamentum teres does not flow into the epiphysis during the age when Perthes disease most frequently occurs. No differences have been found concerning age, gender, or race in the blood supply from the artery of the ligamentum teres. 4 However, vascular changes of the artery of the ligamentum teres in Perthes disease have not been fully investigated.
In the current study, superselective angiography of the nutrient arteries of the capital femoral epiphysis affected by Perthes disease was done to observe vascular changes. By this technique, the blood supply of the artery of the ligamentum teres in Perthes disease was observed. The aim of the study was to investigate the role of revascularization from the artery of the ligamentum teres in the progression of Perthes’ disease.
MATERIALS AND METHODS
Superselective angiography of the hip with Perthes disease was done on 28 hips of 25 patients. By this technique, the artery of the ligamentum teres was seen in 22 of 28 hips. Twenty-two hips of 20 patients were included in the study. Six hips were excluded from the study because the artery of the ligamentum teres was not seen by the angiographic technique. The 20 patients, at the time angiography was done, ranged from 5 to 14 years of age (mean, 9.5 years). Eighteen patients were boys and two patients were girls. Four patients had bilateral disease and 14 patients had unilateral disease.
Conventional anteroposterior (AP) and lateral radiographs were obtained to determine the stage and involved area of the affected hips. The stages observed at the time of angiography were determined principally by the Waldenstrom staging system. 19 However, flattening of the epiphysis but without apparent fragmentation was classified by the current authors as the initial stage. The stages were initial in nine hips, fragmentation in five hips, and healing in the remaining eight hips. Of nine hips in the initial stage, the capital femoral epiphysis was flattened in seven. The remaining two hips had almost spherical contours.
The exact age at time of onset was difficult to determine because the disease had progressed when the patients were first seen. The Catterall classification 3 was used to radiographically classify the extent of the involved area of the capital femoral epiphysis. If the involved area was not seen clearly at the time of angiography, classification was determined by followup radiographs. Of 22 hips, three were categorized as Catterall Group II, and 19 were Group III or IV. Hips that were classified as Catterall Group III or IV were not clearly distinguished because the stages of the disease had progressed in some cases.
To evaluate normal angiographic findings of the artery of the ligamentum teres, angiography was obtained from four normal hips of four male patients who had leg or knee tumors without any symptoms or any abnormal radiographic findings suggesting Perthes disease. Three of the four patients were 10 years of age and one patient was 12 years of age. Angiographies were done with the patients under general anesthesia or local anesthesia. The contralateral femoral artery was cannulated in the older children, whereas the ipsilateral femoral artery was punctured in the younger children because the bifurcation of the common iliac artery was too acute to allow the cannula to negotiate the angulation. The tip of the catheter was inserted into the origin of the femoral medial circumflex artery, and 6 mL of contrast medium was injected by manual pressure for 4 seconds. If the artery of the ligamentum teres was not filled from the femoral medial circumflex artery, a catheter was inserted into the internal iliac artery. Then, the artery of the ligamentum teres branching from the obturator artery was observed for confirmation.
Angiographic findings were investigated on the AP view. The subtraction method was used to provide good visualization. In the four normal control hips, the artery of the ligamentum teres was seen close to the acetabular fossa toward the proximal direction up to the ligamentum fossa of the femoral head. However, the arteries were not seen in the capital femoral epiphysis (Fig 1 ). In 17 of the 22 affected hips, the artery of the ligamentum teres branched from the femoral medial circumflex artery and was stained by injection into the femoral medial circumflex artery. In five hips, the artery of the ligamentum teres originated from the obturator artery. The artery of the ligamentum teres was detected by injection into the obturator artery in two of the five hips. The artery of the ligamentum teres of three hips was seen through anastomosis between the femoral medial circumflex artery and the obturator artery. The aim of the current study was to investigate the vascularized area of the femoral capital epiphysis in Perthes disease when magnetic resonance imaging (MRI) was not available in the authors’ institution (1983 through 1988). The study was done with informed consent of the patients and their parents.
Fig 1.:
Superselective angiography of the femoral medial circumflex artery of the left normal hip of a 10-year-old boy who had a soft tissue tumor of his left thigh. The arteries of the ligamentum teres (black arrows) are seen along the medial acetabulum extending to the medial femoral capital epiphysis. However, these arteries were not stained inside the femoral capital epiphysis. The dotted line shows the contour of the femoral head and acetabulum.
RESULTS
Angiographic Classification of the Artery of the Ligamentum Teres Related to Radiographic Changes of Disease Progression
Patterns of blood supply of the artery of the ligamentum teres seen on the angiograms in the current study correlated with progression of the disease on the AP radiographs and were classified as follows (Fig 2 ): Type I, normal pattern. The artery of the ligamentum teres was present only outside the capital femoral epiphysis without increased vascularity. The capital femoral epiphysis was almost spherical without lateral displacement on the AP radiograph; Type II, absence of the artery of the ligamentum teres in the capital femoral epiphysis. Vascularity of the periphery of the artery of the ligamentum teres was increased, but absence of vascularity was seen in the medial bony epiphysis. The capital femoral epiphysis was almost spherical on the AP radiograph. Lateral displacement was not observed; Type III, vascularity of the periphery of the artery of the ligamentum teres was increased, but vascularity was not observed in the medial bony epiphysis. The capital femoral epiphysis was flattened, and lateral displacement was slight; Type IV, numerous small arteries were distributed in the periphery of the artery of the ligamentum teres outside the bone and medial small portion of the bony epiphysis. However, vascularity in extensive areas of the medial bony epiphysis was absent. The capital femoral epiphysis was flattened. Lateral displacement was slight; and Type V, numerous small arteries were distributed in the extensive area of the medial bony epiphysis, the size in the extraosseous area of the artery of the ligamentum teres was increased. Lateral displacement and flattening of the capital femoral epiphysis were seen.
Fig 2.:
Characteristics of each type of artery of the ligamentum teres as seen by angiography in agreement with the changes seen on the AP radiograph of the disease progression.
Vascularization of the Artery of the Ligamentum Teres of Perthes Disease in Various Stages
In 22 hips with Perthes disease, Type I vascularization of the artery of the ligamentum teres was seen in two hips, Type II in one hip, Type III in seven hips, Type IV in five hips, and Type V in seven hips. In nine hips in the initial stage of Perthes disease, Type III with increased vascularization in only the periphery of the artery of the ligamentum teres was observed in five hips. Type IV was seen in four hips. Hips of Types III and IV had flattening of the capital femoral epiphysis and slight lateral displacement. Type II was seen in only one hip with an almost spherical contour of the capital femoral epiphysis (Fig 3 ). Types I and V were not observed in the initial stage. Of five hips in the fragmentation stage of Perthes’ disease, Type V was seen in three. In these three hips with apparent lateral displacement, distribution of numerous small arteries on the medial epiphysis was seen. One hip of Type III and one hip of Type IV vascularization also were seen. Types I and II vascularization were not observed. In eight hips in the healing stage of Perthes disease, Type V with extensive vascularization of the medial epiphysis was seen most often in four hips with apparent lateral displacement (Fig 4 ). Two hips without lateral displacement and with spherical contours had a Type I pattern. One hip of Type I and one hip of Type IV vascularization also were seen. The summary of the results of vascular changes of the artery of the ligamentum teres of the disease related to the stages of Perthes disease is shown in Table 1 .
TABLE 1: Frequency of Angiographic Types of the Arteries of Ligamentum Teres of Normal Hips and the Radiographic Stage of Perthes Disease
Fig 3A–B.:
Right hip of an 8-year-old boy with evident disease. (A) The capital femoral epiphysis that is almost spherical without lateral displacement but slight collapse (arrow) is see on the medial margin of the epiphysis on the AP radiograph. (B) The artery of the ligamentum teres is evident (arrows) on angiography. The periphery of the artery is vascularized but not stained clearly inside of the epiphysis, indicating a Type II pattern. The dotted line shows the contour of the femoral head. The tip of the catheter is inserted into the femoral medial circumflex artery (1).
Fig 4A–B.:
(A) Anteroposterior radiograph of the right hip of a 7-year-old boy reveals apparent flattening and lateral displacement. The extensive area of epiphysis is dense. (B) The artery of the ligamentum teres (arrows) is observed clearly on the angiogram of the femoral medial circumflex artery (1). The periphery of the artery of the ligamentum teres is well vascularized. Marked vascularization from the artery of the ligamentum teres is evident in an extensive area of medial epiphysis with a Type V pattern. The dotted line shows the contour of the femoral head and acetabulum.
Relationship Between Vascularization of the Artery of the Ligamentum Teres and Extent, Location of Lesion of the Femoral Capital Epiphysis in Perthes Disease
Type I pattern with a normal appearance of the artery of the ligamentum teres was observed in only two hips. The two hips were in the healing stage of small involvement and had a spherical capital femoral epiphysis without lateral displacement. In contrast, in 13 hips of Types II, III, and IV vascularization with absence of a medial extensive vascularized area on the capital femoral epiphysis (Fig 3 ), collapse was observed until the medial margin of the bony epiphysis below the medial acetabular roof. The 13 hips did not have apparent lateral displacement, although the stages of the hips varied (initial stage in the nine hips, fragmentation in two hips, and healing in two hips). These 13 hips were classified as Catterall Group III or IV. Seven hips with a Type V pattern had apparent lateral displacement and were of Catterall Group III or IV. In these hips, the medial segment was considered to be viable radiographically and was vascularized by numerous small arteries rising from the arteries of the ligamentum teres (Fig 4 ).
Vascularization of the Arteries of Ligamentum Teres and Age in Perthes Disease
Patients who were younger than 6 years accounted for only three cases (three hips). The type was III in two hips and IV in one hip. In 12 hips of patients between 7 and 10 years of age, Type V vascularization was seen in five hips, Type IV in two hips, Type III in three hips, and Types II and I in one hip each. In seven hips of patients older than 11 years, Type V vascularization was seen in two hips, Type IV in two hips, Type III in two hips, and Type I in one hip. The exact relationship between vascularization from the artery of the ligamentum teres and age could not be determined because the stages of disease varied, although it could be related to the degree of displacement or subluxation of the femoral head.
DISCUSSION
The impairment of blood supply from the nutrient arteries of the capital femoral epiphysis appears to be a major cause of Perthes disease. Angiographic studies of Perthes disease showed interruption of the blood supply of the lateral epiphyseal arteries at their stump. 1,6,16 Impairment of the blood supply of the lateral epiphyseal arteries probably occurs in this area. The lateral epiphyseal arteries are the most important source supplying blood to an extensive area of the capital femoral epiphysis penetrating from the lateral area. However, the artery of the ligamentum teres also supplies blood to the capital femoral epiphysis on the medial portion. Angiographic studies of Perthes disease in vivo 1,6,16 revealed no findings of the arteries of the ligamentum teres. Microangiographic studies using materials obtained at autopsy showed different results concerning the blood supply of the arteries of the ligamentum teres related to the pathogenesis of Perthes disease. Chung 5 reported that the origin of the artery in the ligamentum teres was well filled in 113 of 123 femoral heads of necropsy specimens. However, the artery was present only in the ligament but not in the femoral head in 78 specimens. In 20 specimens, the artery to the ligamentum teres was well perfused and provided one deep vessel to the center of the head. In 15 specimens, two or more deep vessels to the center of the head were present. Chung 5 concluded he was unable to correlate the type of arterial supply with race, age, or gender. In contrast, Trueta 17 reported that from birth to approximately 3 to 4 years of age, the vessels of the ligamentum teres did not contribute to nourishment of the head. From 7 years of age and older, penetration of vessels from the ligamentum teres appeared to be increasingly frequent. Before 6 years of age, penetration of vessels from the ligamentum teres into the epiphysis rarely was seen. However, the microangiographic studies of autopsy specimens showed the distribution of the artery in the ligamentum teres in subjects without Perthes disease. The role of the blood supply from the artery of the ligamentum teres in Perthes disease has not been fully studied.
The superselective angiographic technique in the current study had an advantage in that it showed the blood vessels clearly in the capital femoral epiphysis with Perthes disease in vivo. In the current study, four normal control hips showed that the artery of the ligamentum teres stained the arteries only in the ligament, and not inside the femoral head. Nine hips in the initial stage without apparent lateral displacement showed absence of marked vascularization from the artery of the ligamentum teres on the medial portion of the capital femoral epiphysis (Types II, III, and IV) although the periphery of the artery of the ligamentum teres was well vascularized. In these hips, collapse was extended to the medial margin of the capital femoral epiphysis.
Salter and Thompson 13 speculated that pathologic fractures after potential avascular necrosis caused true Perthes disease. The current authors think that mechanical damage produces impairment of spontaneous revascularization by the artery of the ligamentum teres affected by subchondral fracture and collapse on this area. After the occurrence of subchondral fracture, mechanical and structural instability prohibit penetration of revascularization. Subsequent blockage of revascularization from the artery of the ligamentum teres occurs repeatedly during the repair process. Histologic findings suggest recurrent necrosis is a common finding of Perthes disease in humans. 8,11 Two or more episodes of infarction may occur on the medial capital femoral epiphysis. However, cases with apparent lateral displacement (three of five hips in the fragmentation stage, four of eight hips in the healing stage) were Type V, showing the presence of revascularization consisting of numerous small arteries on the medial extensive area of the capital femoral epiphysis.
Mechanical loading was concentrated in the collapsed lateral epiphysis below the lateral acetabulum with lateral displacement because of an apparent incongruence between the acetabulum and the capital femoral epiphysis. The authors assumed the decreased mechanical pressure at the medial portion of the capital femoral epiphysis occurred at that time. Thus, revascularization from the artery of the ligamentum teres may occur at the affected medial capital femoral epiphysis. Revascularization from the artery of the ligamentum teres of Perthes disease occurs as follows: After an ischemic episode of the capital femoral epiphysis, revascularization from the artery of the ligamentum teres starts to penetrate the involved medial portion of the capital femoral epiphysis. However, this vascular penetration is impaired by subchondral fracture and collapse at the weightbearing portion, corresponding with the medial margin of the acetabular roof. After progression of the lateral displacement, mechanical pressure is concentrated below the lateral acetabular roof. At that time, mechanical pressure at the medial epiphysis is decreased. Thus, revascularization consists of newly formed small arteries that start to penetrate into this area. After completion of healing, the blood supply of the artery of the ligamentum teres is not important for nourishment of the capital femoral epiphysis. The authors think normal vascular anatomy of the artery of the ligamentum teres is not related to the onset of the disease.
References
1. Atsumi T, Yamano K, Muraki M, et al: The blood supply of the lateral epiphyseal arteries in Perthes’ disease. J Bone Joint Surg 82B:392–398, 2000.
2. Calver R, Venugal V, Dorgan J, et al: Radionuclide scanning in the early diagnosis of Perthes disease. J Bone Joint Surg 63B:379–382, 1981.
3. Catterall A: The natural history of Legg-Calve-Perthes disease. J Bone Joint Surg 53B:37–53, 1971.
4. Catterall A, Pringle J, Byers PD, et al: Perthes’ disease: Is the epiphyseal infarction complete. J Bone Joint Surg 64B:276–281, 1982.
5. Chung SMK: The arterial supply of the developing proximal end of the human femur. J Bone Joint Surg 58A:961–970, 1976.
6. DeCamargo FP, DeGodog Sr RM, Tovo R: Angiography in Perthes disease. Clin Orthop 191:216–220, 1984.
7. Henard DC, Calandruccio RA: Experimental production of roentgenographic and histological changes in the capital femoral epiphysis following abduction, extension and internal rotation of the hip. Surg Forum 22:442–444, 1971.
8. Inoue A, Freeman MAR, Vernon RB, et al: The pathogenesis of Perthes’ disease. J Bone Joint Surg 58B:453–461, 1976.
9. Kemp HBS: Perthes’ disease in rabbits and puppies. Clin Orthop 209:139–159, 1986.
10. Kleinman RG, Bleck EE: Increased blood viscosity in patients with Legg-Perthes disease: A preliminary report. J Pediatr Orthop 2:131–136, 1981.
11. Mckibben B, Ralis Z: Changes found in a case of Perthes’ disease at necropsy: A case report. J Bone Joint Surg 56B:438–447, 1974.
12. Salter RB: Clinical and Laboratory Research Studies on Legg-Perthes Disease. The Hip. Proceedings of the First Scientific Meeting of The Hip Society, St Louis, CV Mosby 4–13, 1973.
13. Salter RB, Thompson GH: Legg-Calve-Perthes’ disease: The prognostic significance of the subchondral fracture and a two-group classification of the femoral head involvement. J Bone Joint Surg 66A:479–499, 1984.
14. Sanchis M, Zahir A, Freeman MAR: The experimental simulation of Perthes’ disease by consecutive interruptions of blood supply to the capital femoral epiphysis in the puppy. J Bone Joint Surg 55A:335–342, 1973.
15. Sutherland AD, Savage JP, Paterson DC, et al: The nuclide bone-scan in the diagnosis and management of Perthes’ disease. J Bone Joint Surg 62B:300–306, 1980.
16. Theron J: Angiography in Legg-Calve-Perthes disease. Pediatr Radiol 135:81–92, 1980.
17. Trueta J: The normal vascular anatomy of the human femoral head during growth. J Bone Joint Surg 39B:358–394, 1957.
18. Vegter J, Lubsen CHC: Fractional necrosis of the femoral head epiphysis after transient increase in joint pressure: An experimental study in juvenile rabbits. J Bone Joint Surg 69B:530–535, 1987.
19. Waldenstrom H: The first stages of coxa plana. J Bone Joint Surg 20:559–566, 1938.