Osteonecrosis of the femoral head (ONFH) usually affects middle-aged adults and occurs bilaterally. It particularly affects active adults in the third to fifth decades of life, and 5% to 12% of hips with ONFH require a total hip arthroplasty.1,2
Although various strategies and surgical techniques can be applied to treat the different stages of ONFH, its management continues to be challenging, especially when the femoral head is collapsed. Treatment is based on the patient's age, symptoms, stage, and/or medical status. However, the orthopedic community has not yet adopted a universalized treatment algorithm.3–9
Digital subtraction angiography (DSA) has been widely used clinically and is still considered the gold standard for the evaluation of vascular pathologies.10,11 Changes in the intraosseous blood supply of the femoral head following osteonecrosis have been of much interest in ONFH research. Measuring additional parameters would help to determine the severity of femoral head necrosis. Consequently, it is necessary to select correct and effective parameters to evaluate the blood supply of the femoral head when DSA is performed.
The authors investigated practical and effective parameters to evaluate changes in the intraosseous blood supply of the femoral head and the association between characteristics of the blood supply and the severity of ONFH at different stages by DSA. The authors hope that this knowledge will help to predict the fate of early–middle precollapsed stages of ONFH in the view of the vascular aspect and to avoid collapse of the femoral head by pointing to early surgical interventions.
Materials and Methods
Digital Subtraction Angiography Data
The authors retrospectively reviewed the preoperative DSA data of patients who were diagnosed with ONFH and hospitalized at their facility between January 1, 2012, and January 1, 2017. This study received institutional review board approval.
To be included, patients had to have a diagnosis of ONFH, be 18 years or older, have existing imaging data on which Association Research Circulation Osseous (ARCO) staging could be performed, and have no previous surgical intervention on the affected hip.
Inclusion criteria for DSA involved the following: (1) “effective” images of the femoral artery (FA) and the medial circumflex femoral artery (MCFA) could be obtained for each patient on which each vessel was determined to be intact, without blurring or interruption; (2) the MCFA originated from the deep femoral artery (DFA); and (3) the DSA technique was strictly enforced according to the standardized procedure as described below.
A total of 85 hospitalized patients (81 hips) with ONFH and their images, according to the inclusion criteria, were included in this study. The data of 4 patients with ARCO stage I were not included because such a small sample could result in a greater statistical error. All of the patients were Chinese. There were 60 men and 21 women, with a mean age of 34.6 years (range, 20–68 years). Although 49 patients had bilateral disease, the authors performed superselective arteriography on only the more painful side. All patients were grouped according to existing data, including pathologic examinations, radiographs, computed tomography scans, and magnetic resonance images, by 3 senior orthopedic surgeons (S.W., D.Z., B.W.). A consensus definition was reached, ultimately classifying 31 hips as ARCO stage II, 30 hips as ARCO stage III, and 20 hips as ARCO stage IV.
Digital Subtraction Angiography Technique
The affected extremity was placed in a neutral position with toes upward vertically. This position was maintained throughout the process to ensure that the proximal femur was not rotated. Using the Seldinger12 technique under local anesthesia, a 4F micro angiography catheter was inserted into the FA and the MCFA of the affected limb. The angiography machine (Artis zee; Siemens, Erlangen, Germany) projected at 2 fixed locations invariably. With the hip joint kept in the center of the projection, posteroanterior arteriography of the FA was performed. Then, the direction of the projection was changed to 30° of external rotation to evaluate the affected MCFA. Seven milliliters of iodixanol (16 g/50 mL) was injected at a rate of 3 mL/s into the FA, while a total volume of 4 mL at an injection rate of 0.8 mL/s entered the MCFA. The acquisition rate was 4 frames/s, and a 1-second delay was used for an acquisition duration of 5 to 10 seconds. Vessels, blood distribution, and staining of the affected femoral head were observed.
The intent of this study was to evaluate the intraosseous blood supply of the necrotic femoral head by measuring changes in the parameters of the effective images of patients with different ARCO stages. The images were magnified using image software (Digimizer; MedCalc Software, Ostend, Belgium) and evaluated separately by 2 radiologists. The following data were acquired:
On an effective image of the FA (Figure 1A), the diameters of the DFA, MCFA, and lateral circumflex femoral artery (LCFA) within 2 cm were measured. The relative widths of the MCFA and the LCFA were evaluated using the following ratios:
MCFA/DFA, the ratio of the arterial diameter of the MCFA and the DFA.
LCFA/DFA, the ratio of the arterial diameter of the LCFA and the DFA.
d, Diameter of the deep femoral artery (DFA); m, diameter of the medial circumflex femoral artery (MCFA); l, diameter of the lateral circumflex femoral artery (LCFA); MCFA/DFA: m/d; LCFA/DFA: l/d (A). A: AB/AC. Point A, the projective intersection of the MCFA and the femoral neck. Point D, the projective end point of the MCFA. Point C, the crossover point of the AB extension line and the femoral neck. AB is perpendicular to the long axis of the femoral neck. BD is parallel to the long axis (B). Line I: a horizontal line at the upper junction of the femoral head and neck. Line II: the line that connects the upper and lower junctions of the femoral head and neck (C).
On an effective image of the MCFA:
A, the length ratio between the MCFA across the femoral neck and the diameter of the femoral neck (Figure1B, AB/AC). It represents the relative length of the MCFA and indirectly measures the blood supply to the femoral head.
the number of vessels that reached or exceeded line I (a horizontal line at the upper junction of the femoral head and neck, Figure 1C). The superior posterior retinacular arteries above this line, which enter the femoral head at the junction of the femoral head and neck, supply blood to the lateral 60% to 75% of the femoral head.13
the number of vessels that reached or exceeded line II (connecting the upper and lower junctions of the femoral head and neck, Figure 1C) but did not reach line I. The inferior posterior retinacular arteries arise from the MCFA, which courses through the femoral head along the lower edge of the femoral head cartilage, and supply 25% to 50% of the area above this line.13
Data were presented as mean±standard deviation if they had a normal distribution and homogeneous variance. One-way analysis of variance was used to compare the means (standard deviations) of the 3 groups. To evaluate the heterogeneity of the variance data, the authors adopted the rank sum test. All statistical tests were performed using SPSS version 23 software (IBM Corp, Armonk, New York). P<.05 was considered statistically significant.
The mean MCFA/DFA of patients with stage II, III, and IV disease was 0.41±0.08, 0.41±0.12, and 0.39±0.11, respectively. The mean LCFA/DFA of patients with stage II, III, and IV disease was 0.64±0.12, 0.67±0.12, and 0.68±0.13, respectively. The mean ratio of A in patients with stage II, III, and IV disease was 0.85, 0.80, and 0.60, respectively. In patients with stage II, III, and IV disease, the mean ratio of a was 1.84, 1.07, and 0.30 and the mean ratio of b was 1.90, 1.13, and 1.00, respectively.
The MCFA/DFA (P=.841) and the LCFA/DFA (P=.543) were not statistically significantly different between patients with ARCO stage II to IV disease, as calculated by one-way analysis of variance. However, A (P=.040), b (P=.013), and especially a (P=.000) were found to be statistically significantly different by the rank sum test. Moreover, candlestick charts indicated that the distribution of these 3 parameters had a negative relationship with ARCO stage (Figure 2).
Relationship between A (y axis) and Association Research Circulation Osseous (ARCO) stage (x axis) (A). Relationship between a (y axis) and ARCO stage (x axis) (B). Relationship between b (y axis) and ARCO stage (x axis) (C).
Osteonecrosis of the femoral head is an evolutionary process caused by a decrease in the blood perfusion of the femoral head involving marrow necrosis and osteocytic death, reparative process around the necrotic zone, and collapse of necrotic bone and subsequent degenerative arthritis of the hip.14,15 Compared with single-photon emission computed tomography, magnetic resonance imaging, and other noninvasive methods, DSA is an invasive technique. However, it provides a clear, high-resolution image to observe vascular lesions and the position of a stenosis or blockage.16 Some scholars have performed preliminary studies of parameters based on DSA findings. In 1987, Heuck et al17 studied the blood supply of the necrotic femoral head using DSA. A total of 100 hips were studied, being divided into a normal group, a nontraumatic group, and a trauma group. The numbers of vascular alterations, such as stenoses, occlusions, and hypoplasias of the MCFA, posterior branch, and capsular arteries, were counted. The authors suggested that a reduction in the arterial blood supply to the femoral head is a major etiologic factor in traumatic femoral head necrosis. The number of affected arteries was an important parameter in this study. Owing to technical limitations at that time, the observation of minute vessels was difficult, and their diameters, lengths, and branches were not taken into account.
Tong et al18 observed and analyzed 39 cases (64 hips), classifying the state of the superior metaphyseal artery into 5 types: type I, normal; type II, vascular interruption; type III, minor penetration branches; type IV, irregular penetration branches; and type V, maximum penetration branches. They reached 2 conclusions. First, type II is most commonly observed, and type IV and type V are rarely found in patients with Ficat stage 0, I, and II femoral heads. Patients with Ficat stage III femoral heads mostly showed type IV and type V changes. Second, ONFH has different vascular types because of its different etiologies. Type IV and type V are more common in traumatic cases, but cases of steroid- and alcohol-induced ONFH generally lead to type II and type III disease. This is the main method for classifying blood flow according to the distribution of the superior metaphyseal artery and lacked quantifiable parameters in this study.
In recent years, drawing a reference line has become a valuable method for counting the number of blood vessels. The position of the reference line is crucial for accurately assessing vessels. Chen and Pang19 investigated the effects of different types of femoral neck fractures on the blood supply to the femoral head. The numbers of retinacular arteries above line a (connecting the upper and lower junctions of the femoral head and neck), line b (from the lower junction of the femoral head and neck to the apex of the greater trochanter), and line c (from the initial part of the MCFA to the apex of the greater trochanter) (Figure 3A) were counted, and the relative length of the MCFA was measured based on 50 cases of DSA findings. They concluded that higher femoral neck fractures are more seriously displaced and have fewer vessels in the femoral head, especially above line a. Shortly after that study, the locations of the lines were disputed. In 2010, Xing20 studied the DSA of 220 patients with ONFH, believing that line b does not reflect the blood supply to the necrotic femoral head because it is underlined through the starting point of the MCFA and the end point of the LCFA. Combined with the experience of his department, it is more reasonable to mark line B—a parallel line over the midpoints of lines A and C through the midpoints of the superior and inferior retinaculum trunks. Line A is the same as line a; line C is a reference line and almost runs from the lesser trochanter to the greater trochanter (Figure 3B). Xing thought that the number of vessels that reached or exceeded line B was significantly different between Ficat stages. However, these reports did not explore the relationship between the width of important vessels and the severity of disease.
Line a: at and below the upper junction of the femoral head and neck. Line b: from the lower junction of the femoral head and neck to the apex of the greater trochanter. Line c: from the initial part of the medial circumflex femoral artery to the apex of the greater trochanter (A). Line A: same as line a. Line B: a parallel line over the midpoint of lines A and C, through the midpoints of the superior and the inferior retinaculum trunks. Line C: almost from the lesser trochanter to the greater trochanter and used as a positioning line (B).
In this study, the superior and the inferior retinaculum were evaluated with a and b, respectively. These 2 parameters are the number of vessels that can be seen entering the femoral head on the effective image of the MCFA, with a representing the number of vessels that reached or exceeded line I and b representing the number of vessels that reached or exceeded line II but did not reach line I. The blood supply of the femoral head mainly comes from the MCFA,21–23 which arises from the DFA in 57% of cases and accounts for 65% to 80% of the blood to the femoral head.24–26 The superior retinaculum of the MCFA supplies the lateral 60% to 75% of the femoral head. The superior epiphyseal arteries anastomose in the superolateral region above line I, with fewer secondary anastomoses than in other regions of the femoral head. This region represents the most crucial area where collapse and resulting osteonecrosis occur.2,27,28 Moreover, the inferior retinaculum above line II supplies 25% to 50% of the blood to the femoral head. It is especially important when arteries cannot reach line I.
A represents the relative length of the MCFA. With an increase of A, more arteries of the retinaculum are retained. It indirectly shows the blood supply of the femoral head. To prevent the changes of determining points in Figure 1B, caused by rotation of the proximal femur and viewing plane, the DSA technique strictly followed the standardized procedure described above. Rotation error was controlled by keeping the patient's position constant, and plane consistency was controlled by projecting at a fixed location.
Finally, the authors preliminarily studied the MCFA/DFA and the LCFA/DFA as vascular width parameters. To avoid errors and reduce the interference caused by the uneven distribution of contrast, they used the ratio method and measured the initial part of the arteries. It is more comprehensive and accurate to estimate the blood supply of the femoral head with MCFA/DFA, LCFA/DFA, A, a, and b.
The basic pathologic change characteristic of ONFH is bone necrosis due to a decrease in the blood perfusion of the femoral head. However, the underlying cause of the circulatory defect in osteonecrosis varies and may involve both local and systemic changes. The femoral head is supplied by vessels arising mainly from the MCFA. Other vessels with lesser contribution arise from the LCFA, obturator artery, superior gluteal artery, and the first perforating branch of the DFA. On the basis of the authors' research, the trunk of the MCFA is generally located outside of the necrotic lesion, so the decreased relative length of the MCFA is more likely the cause of the disease together with other factors. It is unclear whether the decreased number of vessels or the compromised intravascular factors lead to necrosis of the femoral head and determine disease progression. But, when ONFH progresses, especially when the femoral head collapses, the vessels will be interrupted because of elevated intraosseous pressure and/or compression of fractured surrounding trabeculae. However, the number of intraosseous vessels to the necrotic lesion can provide a reference value for disease evaluation.
Although many risk factors have been reported for ONFH, the pathogenesis is believed to be closely related to direct or indirect injury to the vascular supply of the femoral head.14 Femoral head collapse, acetabular involvement, and secondary osteoarthritis are inevitable results of the disease. Through their research, the authors have found that DSA can be used as one element to predict the fate of precollapsed femoral head. For young patients with obvious clinical symptoms, a large lesion, or anterolateral weight-bearing areas affected in ARCO stage II, the authors suggest that DSA could be performed to evaluate the intraosseous vascular condition so as to predict the collapse of the femoral head. When relevant parameters indicate a high risk of insufficient blood supply to the femoral head, surgical interventions include core decompression, vascularized or nonvascularized bone grafting, and osteotomy to delay the need for total hip arthroplasty.
Because of the limited sample for each stage, the authors could not draw a definite conclusion regarding the relationship between the evaluation parameters and ARCO stage and then accurately predict disease progression. More quantitative research of relevant parameters is needed. Further, the associations between the authors' findings and other classifications of ONFH have not been confirmed. Finally, it remains uncertain whether these parameters are suitable for other populations.
This preliminary study showed that the relative length of the MCFA and the numbers of vessels above line I and line II were related to the severity of ONFH. Digital subtraction angiography can be an important vascular indicator together with other imaging technology of the fate of precollapsed ONFH and provides a new option for supplementing ARCO staging.
- Mont MA, Jones LC, Sotereanos DG, Amstutz HC, Hungerford DS. Understanding and treating osteonecrosis of the femoral head. Instr Course Lect. 2000;49:169–185.
- Scaglione M, Fabbri L, Celli F, Casella F, Guido G. Hip replacement in femoral head osteonecrosis: current concepts. Clin Cases Miner Bone Metab. 2015;12(suppl 1):51–54.
- Lieberman JR, Berry DJ, Mont MA, et al. Osteonecrosis of the hip: management in the 21st century. Instr Course Lect. 2003;52:337–355.
- Gou WL, Lu Q, Wang X, et al. Key pathway to prevent the collapse of femoral head in osteonecrosis. Eur Rev Med Pharmacol Sci. 2015;19(15):2766–2774.
- Cheng EY, Thongtrangan I, Laorr A, Saleh KJ. Spontaneous resolution of osteonecrosis of the femoral head. J Bone Joint Surg Am. 2004;86(12):2594–2599. doi:10.2106/00004623-200412000-00002 [CrossRef]
- Glueck CJ, Freiberg RA, Sieve L, Wang P. Enoxaparin prevents progression of stages I and II osteonecrosis of the hip. Clin Orthop Relat Res. 2005;435:164–170. doi:10.1097/01.blo.0000157539.67567.03 [CrossRef]
- Hungerford DS, Jones LC. Asymptomatic osteonecrosis: should it be treated?Clin Orthop Relat Res. 2004;429:124–130. doi:10.1097/01.blo.0000150275.98701.4e [CrossRef]
- Jergesen HE, Khan AS. The natural history of untreated asymptomatic hips in patients who have non-traumatic osteonecrosis. J Bone Joint Surg Am. 1997;79(3):359–363. doi:10.2106/00004623-199703000-00006 [CrossRef]
- Steinberg ME, Larcom PG, Strafford B, et al. Core decompression with bone grafting for osteonecrosis of the femoral head. Clin Orthop Relat Res. 2001;386:71–78. doi:10.1097/00003086-200105000-00009 [CrossRef]
- Deuschl C, Göricke S, Gramsch C, et al. Value of DSA in the diagnostic workup of pulsatile tinnitus. PLoS One. 2015;10(2):e0117814. doi:10.1371/journal.pone.0117814 [CrossRef]
- Maj E, Cieszanowski A, Rowinski O, Wojtaszek M, Szostek M, Tworus R. Time-resolved contrast-enhanced MR angiography: value of hemodynamic information in the assessment of vascular diseases. Pol J Radiol. 2010;75:52–60.
- Seldinger SI. Catheter replacement of the needle in percutaneous arteriography: a new technique. Acta Radiol Suppl (Stockholm). 2008;434:47–52. doi:10.1080/02841850802133386 [CrossRef]
- Liu Y, Li M, Zhang M, et al. Femoral neck fractures: prognosis based on a new classification after superselective angiography. J Orthop Sci. 2013;18(3):443–450. doi:10.1007/s00776-013-0367-4 [CrossRef]
- Hong JM, Kim TH, Kim HJ, Park EK, Yang EK, Kim SY. Genetic association of angiogenesis- and hypoxia-related gene polymorphisms with osteonecrosis of the femoral head. Exp Mol Med. 2010;42(5):376–385. doi:10.3858/emm.2010.42.5.039 [CrossRef]
- Hamilton TW, Goodman SM, Figgie M. SAS weekly rounds: avascular necrosis. HSS J. 2009;5(2):99–113. doi:10.1007/s11420-009-9107-x [CrossRef]
- Jin H, Xia B, Yu N, et al. The effects of autologous bone marrow mesenchymal stem cell arterial perfusion on vascular repair and angiogenesis in osteonecrosis of the femoral head in dogs. Int Orthop. 2012;36(12):2589–2596. doi:10.1007/s00264-012-1674-7 [CrossRef]
- Heuck A, Reiser M, Schmucker F, et al. Selective digital subtraction arteriography in necrosis of the femoral head. Skeletal Radiol. 1987;16(4):270–274. doi:10.1007/BF00361467 [CrossRef]
- Tong P, Xiao LW, Xu M. The image of the digital subtraction angiography (DSA) about pre-treating and post-treating of osteonecrosis of femoral head. J Bone Joint Injury. http://en.cnki.com.cn/Article_en/CJFDTOTAL-GGJS200005001.htm
- Chen XJ, Pang QJ. The clinical research on blood supply of the femoral head after the fracture of femoral neck by DSA. Modern Pract Med. http://www.en.cnki.com.cn/Article_en/CJFDTotal-NBYX200903007.htm
- Xing FF. Study of the digital subtraction angiography about different stages and different causes of avascular necrosis of the femoral head [master's thesis]. Guangzhou, China: Guangzhou University of Chinese Medicine; 2010.
- Zlotorowicz M, Szczodry M, Czubak J, Ciszek B. Anatomy of the medial femoral circumflex artery with respect to the vascularity of the femoral head. J Bone Joint Surg Am. 2011;93(11):1471–1474. doi:10.1302/0301-620X.93B11.26993 [CrossRef]
- Gautier E, Ganz K, Krügel N, Gill T, Ganz R. Anatomy of the medial femoral circumflex artery and its surgical implications. J Bone Joint Surg Br. 2000;82(5):679–683. doi:10.1302/0301-620X.82B5.10426 [CrossRef]
- Grose AW, Gardner MJ, Sussmann PS, Helfet DL, Lorich DG. The surgical anatomy of the blood supply to the femoral head: description of the anastomosis between the medial femoral circumflex and inferior gluteal arteries at the hip. J Bone Joint Surg Br. 2008;90(10):1298–1303. doi:10.1302/0301-620X.90B10.20983 [CrossRef]
- Dewar DC, Lazaro LE, Klinger CE, et al. The relative contribution of the medial and lateral femoral circumflex arteries to the vascularity of the head and neck of the femur: a quantitative MRI-based assessment. Bone Joint J. 2016;98(12):1582–1588. doi:10.1302/0301-620X.98B12.BJJ-2016-0251.R1 [CrossRef]
- Al-Talalwah W. The medial circumflex femoral artery origin variability and its radiological and surgical intervention significance. Springerplus. 2015;4:149. doi:10.1186/s40064-015-0881-2 [CrossRef]
- Zhao D, Xiaobing Y, Wang T, et al. Digital subtraction angiography in selection of the vascularized greater trochanter bone grafting for treatment of osteonecrosis of femoral head. Microsurgery. 2013;33(8):656–659. doi:10.1002/micr.22179 [CrossRef]
- Choi HR, Steinberg ME, Cheng E. Osteonecrosis of the femoral head: diagnosis and classification systems. Curr Rev Musculoskelet. 2015;8(3):210–220. doi:10.1007/s12178-015-9278-7 [CrossRef]
- Shen J, Wei H, Yu Q, Cheng L. Three-dimensional computerized tomography-assisted identification of necrotic volume, distribution, shape and prognosis of collapse in ONFH. Advances in Computed Tomography. 2016;5(1):1–18. doi:10.4236/act.2016.51001 [CrossRef]