Feasibility and reliability of high-resolution phase-contrast quantitative-flow magnetic resonance imaging for assessing femoral head blood flow: a prospective study

Introduction

Ischemic necrosis of the femoral head (i.e., a loss of blood flow to the bone tissue) is caused by a myriad of factors, including corticosteroid use, excessive alcohol intake, and femoral neck fractures (1,2). These factors can precipitate circulatory disruptions, notably to the arterial blood supply and venous congestion (1). The early identification of aberrant blood supply patterns in the femoral head is crucial for initiating timely interventions to mitigate and prevent the progression of ischemic necrosis (3).

Various diagnostic techniques can be used to evaluate blood supply to the femoral head. For example, angiographic methods, such as digital subtraction angiography, high-resolution computed tomography angiography, and contrast-enhanced magnetic resonance (MR) angiography, can be used to visualize the blood vessels supplying the femoral head to a certain degree (4-8). However, these approaches predominantly delineate the morphologic abnormalities of the nutrient arteries. Difficulties may arise in analyzing early hemodynamic changes caused by corticosteroid use and excessive alcohol intake because these are not embolic infarctions (9). Bone scans and dynamic contrast-enhanced magnetic resonance imaging (MRI) can quantitatively detect perfusion anomalies but require the administration of radiotracers and gadolinium-based contrast agents (10). Ultrasonography may reveal deviations in blood circulation in the early phases of femoral head osteonecrosis, preceding the manifestations observed in plain radiography (11-13). Nevertheless, the accuracy and reproducibility of Doppler-based flow assessments are contingent on the operator, and the volumetric flow measurements are influenced by various factors, such as the angle dependence, variability in the estimation of the vessel area, and a lack of consideration of the spatial velocity profile (14-17).

The MR quantitative-flow (Q-flow) sequence, which uses the phase-contrast technique, has the capability to assess a range of hemodynamic parameters. This technique has been widely used in cardiovascular and cerebrovascular assessments (18). Q-flow allows for flow quantification without the need for contrast agent injection, offering a more precise estimation of flow volume compared to ultrasound by delivering a more accurate measurement of the vessel lumen and considering the spatial velocity across the actual vessel lumen (19). Thus, Q-flow is a potentially superior technique for precisely determining blood flow to the femoral head. However, the principal vessels supplying the femoral head [i.e., the medial femoral circumflex artery (MFCA) and the lateral femoral circumflex artery (LFCA), along with their branches] are characterized by small diameters, and the Q-flow faces challenges in accurately measuring small vessels due to errors induced by the partial volume effect (20,21). Consequently, Q-flow faces significant challenges in accurately quantifying the primary blood vessels feeding the femoral head, particularly the deep MFCA, which has a luminal diameter of approximately 1.6–1.8 mm (5).

Currently, research on the use of Q-flow to quantify blood supply to the femoral head is limited. Previous studies have shown the feasibility and accuracy of high-resolution Q-flow for quantifying blood flow in vessels with luminal diameters smaller than 2 mm, with reported errors of less than 6.2% in the phantom experiment (22,23). Therefore, we hypothesized that high-resolution phase-contrast Q-flow could be used to quantify blood flow to the femoral head. The main aims of this study were to use high-resolution phase-contrast Q-flow to quantify the feeding arteries and draining veins of the femoral head, and to ascertain its applicability and effectiveness in clinical settings. We present this article in accordance with the STARD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-411/rc).

Methods

Participants

Between October 2023 and August 2024, this prospective study recruited a convenience sample from Deyang People’s Hospital, Deyang, comprising 10 healthy volunteers (4 men and 6 women), 10 women with systemic lupus erythematosus (SLE) on corticosteroid therapy, and 10 age-matched healthy women for the SLE patients, who served as controls. The characteristics of the study population are presented in Table 1.

The healthy volunteers and controls were selected based on specific criteria that excluded those with a history of conditions affecting femoral head vascularization, such as hip trauma, diabetes mellitus, alcoholism, and steroid hormone use (7). Conventional MRI was performed on the SLE patients to rule out hips with imaging-based ischemic necrosis of the femoral head. Participants with contraindications to MRI, such as individuals with a pacemaker or claustrophobia, were excluded from the study.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Deyang People’s Hospital Ethics Committee (approval No. 2020-04-118-K01), and informed consent was obtained from all the participants before they underwent the Q-flow examination. This study has been registered in the China Medical Research Registration and Filing System (https://www.medicalresearch.org.cn/, No. MR-51-25-001006).

MRI protocol

All the participants were scanned using a 3T scanner (Ingenia; Philips Healthcare) equipped with a 32-channel torso coil covering both hip joints. The relaxation-enhanced angiography without contrast and triggering (REACT) technique was used to visualize the critical femoral vessels for the Q-flow sequence localization. The isotropic REACT technique employed in this study was a flow-independent, Dixon-based fat suppression method that effectively mitigates artifacts arising from both flow variability and inhomogeneous magnetic fields during fat suppression (24). This magnetization-prepared non-balanced dual-echo approach enables the simultaneous imaging of arteries and veins (24). It also enables the visualization of vital femoral nutrient arteries, including the MFCA and LFCA, and their deep branches, as well as corresponding draining veins such as the medial femoral circumflex vein (MFCV) and lateral femoral circumflex vein (LFCV) (25,26).

For the imaging procedure, after a half-hour of rest, the patients were positioned supine with both lower limbs abducted to approximately 30°, as abduction may affect femoral blood flow (12). After the REACT acquisition, the two-dimensional Q-flow slice with an in-plane resolution of 0.4×0.4 (interpolated to 0.2×0.2) mm², oriented perpendicular to the target vessel, was determined based on the maximum intensity projection (MIP) and multi-planar reconstruction (MPR) of the REACT image for flow parameter measurement as depicted in Figure 1. The velocity encoding was set to 50 cm/s for the MFCA and 40 cm/s for the deep MFCA and ascending LFCA, guided by flow velocity measurements obtained by ultrasound (11). The imaging parameters are detailed in Table 2.

Figure 1 Locations of Q-flow scans delineated (as represented by the yellow lines) for the MFCA, ascending LFCA, and deep MFCA. This representation was observed through transverse and coronal thin-layer MIPs (A-C) generated from the REACT sequence, accompanied by a posterior view of the anatomy of the femoral vasculature (D), where a, b, and c are the measurement positions of the MFCA, ascending LFCA, and deep MFCA, respectively. The REACT images carry color-coded arrows, with red indicating the MFCA, green denoting the ascending LFCA, and blue signifying the deep MFCA. In the Q-flow image in full view, the lower right corner shows a zoomed-in view of the area in the yellow dashed box. The red ROIs delineate the target arteries, while the green ROIs indicate the accompanying veins. The measurement position for the MFCA [2] at its origin from the femoral profunda artery [1] elucidates the vascular supply to the femoral head through the superior retinacular artery [4] and inferior retinacular artery [3]. The position for the ascending LFCA [5] was precisely located at the origin of the ascending LFCA, corresponding to the femoral blood supply through the anterior retinacular artery [6]. The position of the deep MFCA was identified subsequent to the branching of the inferior retinacular artery [3] and the descending branch of MFCA [7], reflecting the femoral blood supply through the superior retinacular artery [4]. LFCA, lateral femoral circumflex artery; MFCA, medial femoral circumflex artery; MIP, maximum intensity projection; REACT, relaxation-enhanced angiography without contrast and triggering; ROI, region of interest.

The Q-flow measurement positions for the target vessels were meticulously determined by an imaging technologist with an associate title under the guidance of a deputy chief physician. In measuring the MFCA, the origin of the MFCA was identified anterior to the commencement of the inferior retinacular artery, which signified the blood supply of the femoral head through the superior retinacular artery and inferior retinacular artery (27,28). The position of the deep MFCA was pinpointed following the branching out of the inferior retinacular artery and the descending branch of the MFCA, which indicated the femoral blood supply through the superior retinacular artery (28). In addition, the location of the ascending LFCA was determined at the origin of the ascending LFCA, corresponding to the femoral blood supply through the anterior retinacular artery originating from the ascending LFCA (25,29). The associated veins (the one closest to the measured artery), specifically the lateral and medial circumflex femoral veins, which typically run parallel to their corresponding arteries, were simultaneously measured (26,30). The Q-flow measurement positions are shown in Figure 1.

For each participant, three Q-flow slices were acquired per hip to measure the MFCA, deep MFCA, and LFCA, totaling six Q-flow slices across both hips, with a total Q-flow scan time of approximately 30 minutes per participant. To assess the retest reliability, each healthy volunteer underwent two scans in a single session, separated by a 30-minute interval during which they got up from the MRI table and left the examination room. This allowed for independent positioning and the setup for the second scan. Further, a Q-flow scan was performed on the SLE patients and the controls to assess the efficacy of high-resolution phase-contrast Q-flow in quantifying femoral flow variations.

Quantitative measurements

Two readers, reader 1 with 11 years of experience and reader 2 with 6 years of experience in interpreting osteoarticular images, were blinded to the patient and volunteer specifics. The readers delineated the contours of the target arteries and the accompanying veins in the Q-flow images using the Q-flow analysis software package (Philips Healthcare) to quantify the blood flow parameters. The accuracy of vessel delineation was meticulously examined to ensure precise alignment with vessel boundaries on the magnitude, phase, and velocity images. When necessary, proper alignment with the velocity images was achieved through adjustments. Subsequently, the flow parameters, including stroke (the absolute difference between forward and backward flows), regurgitant fraction (RF; representing the fraction of backward to forward flow), mean flux (MF), stroke distance (SD; representing the net distance blood travels in the vessel per beat), mean velocity (MV), and peak systolic velocity (PSV), were automatically computed. The reliability of Q-flow in quantifying blood flow in the femoral head was evaluated based on established methods for reliability assessment as described previously (31,32), including retest, intra-, and inter-rater reliability as presented in Figure 2. Reader 1 measured the scan data of the SLE and the control groups to investigate early blood flow changes.

Figure 2 Analytical approach employed to assess retest, intra-, and inter-rater reliability. Two measurements by Reader 2 were conducted approximately 3 months apart.

Statistical analysis

The retest reliability, and inter- and intra-rater reliability were assessed using the paired-samples t-test (for normal distributions) or the Wilcoxon rank-sum test (for non-normal distributions), intra-class correlation coefficients (ICCs), and a Bland-Altman analysis. An ICC value between 0.61 and 0.80 indicated good reliability, while a value exceeding 0.80 indicated excellent reliability; good or excellent reliability (ICC >0.6) was considered acceptable (33). The quantitative results between the SLE and control groups were analyzed using the Wilcoxon rank-sum test (for non-normal distributions) or the t-test (for normal distributions), with the Shapiro-Wilk test used to ascertain distribution normality. All the statistical analyses were conducted using Rstudio (version 1.4.1717; https://posit.co/).

Results

Study cohort

A total of 20 hips from 10 healthy volunteers (4 men and 6 women) were included in the study to assess the reliability. None of the SLE patients exhibited any ischemic findings on conventional MRI. Thus, 20 hips from 10 SLE patients (mean SLE duration: 0.95±0.44 years, range: 6 months to 1.5 years; mean corticosteroid use: 0.89±0.47 years, range: 3 months to 1.5 years), and 20 hips from 10 aged-matched controls were analyzed to evaluate the efficacy of assessing early blood flow changes. In the hip evaluation, the measurement of the deep MFCVs proved challenging, and had a success rate of only 66.7%. Conversely, nearly all the other vessels were successfully measured (see Table 3).

Retest, intra-rater, and inter-rater reliability

No significant differences in the flow parameters were observed between the two scans in relation to retest reliability, nor between the measurements for intra-rater and inter-rater reliability [P=0.08–1/not applicable (NA)], except for the RF (P=0.01, P<0.01 and P=0.04 for the LFCA, deep MFCA, and LFCV for the retest reliability). The detailed comparative results are presented in Table 4.

The agreement for feeding arteries was not always acceptable for the RF (ICC: 0.01–0.45 and 0.53–0.97 for the retest and intra-rater reliability), but it was acceptable for all the other flow parameters in terms of the retest reliability (ICC: 0.92–0.99, 0.61–0.9, 0.67–0.83, 0.63–0.86, and 0.7–0.86 for the stroke, MF, SD, MV, and PSV, respectively), intra-rater reliability (ICC: 0.97–1, 0.98–0.99, 0.95–0.98, 0.96–0.98, and 0.99–1 for the stroke, MF, SD, MV, and PSV, respectively), and inter-rater reliability (ICC: 1, 0.99–1, 0.93–0.98, 0.92–0.97, and 0.99–1 for the stroke, MF, SD, MV, and PSV, respectively). For the draining veins, the RF, MV, and PSV did not consistently achieve acceptable agreement in terms of retest reliability (ICC: 0.04–0.9, 0.46–0.84, and 0.5–0.73, respectively), while the other parameters consistently exhibited acceptable agreement in terms of the retest reliability (ICC: 0.9–0.97, 0.63–0.75, and 0.75–0.92 for stroke, MF, and SD, respectively), intra-rater reliability (ICC: 0.84–1, 0.87–1 and 0.93–1 for stroke, MF and SD respectively), and inter-rater reliability (ICC: 0.9–0.99, 0.87–0.99 and 0.9–0.99 for the stroke, MF and SD, respectively). The ICC values are illustrated in Figure 3.

Figure 3 Comprehensive breakdown of the ICC analysis outcomes for the individual flow parameters in each artery (represented in pink) and vein (represented in cyan). The data are presented with corresponding values, including a 95% CI. CI, confidence interval; ICC, intra-class correlation; LFCA, lateral femoral circumflex artery; LFCV, lateral femoral circumflex vein; MF, mean flux; MFCA, medial femoral circumflex artery; MFCV, medial femoral circumflex vein; MV, mean velocity; PSV, peak systolic velocity; RF, regurgitant fraction; SD, stroke distance.

In the Bland-Altman analysis of retest reliability, the percentages of points outside the 95% agreement threshold were 0–10.53% for arteries and 0–16% for veins, with high RF values at 8.33–16.67%. In relation to inter-rater reliability, the values were 5–11.11% for arteries and 5–16.67% for veins, with a high of 16.67% for the deep MFCV in relation to RF. In relation to intra-rater reliability, the artery values were 0–11.11%, and the vein values were 5–15%, with a high of 15% for the LFCV in terms of the MV. The Bland-Altman analysis results are presented in Figures S1-S3.

Efficacy of assessing early blood flow changes

In relation to the feeding arteries, the SLE group had lower blood flow parameters than the control group. Specifically, significant decreases in the MFCA (in the stroke and SD, P=0.01 and P<0.01), LFCA (in the stroke, MF, SD, MV, and PSV, all P<0.01), and deep MFCA (in the stroke, MF, SD, MV, and PSV, all P<0.01) were observed. In relation to the draining veins, the SLE group had lower flow parameters than the control group for the LFCV (in the SD and MV, P<0.01 and P=0.02) and deep MFCV (in the stroke, MF, SD, and PSV, P<0.01, P=0.02, P=0.01 and P=0.03). These outcomes are detailed in Table 5 and Figure 4. Sample Q-flow images of healthy controls and SLE patients are presented in Figure 5.

Figure 4 Comparative analysis of the results between the SLE and control groups for the flow parameters in each vessel. The red arrows indicate a decrease in the blood flow in the SLE group. Statistical significance was set at *, P≤0.05, **, P<0.01, and ***, P<0.001. LFCA, lateral femoral circumflex artery; LFCV, lateral femoral circumflex vein; MF, mean flux; MFCA, medial femoral circumflex artery; MFCV, medial femoral circumflex vein; MV, mean velocity; PSV, peak systolic velocity; RF, regurgitant fraction; SD, stroke distance; SLE, systemic lupus erythematosus.

Figure 5 Representative mean velocity curves in the healthy controls and SLE patients. Red indicates the target artery, while green indicates the accompanying vein. The control group had higher mean velocity profiles than the SLE group, characterized by greater peak values. LFCA, lateral femoral circumflex artery; MFCA, medial femoral circumflex artery; MFCV, medial femoral circumflex vein; SLE, systemic lupus erythematosus.

Discussion

The evaluation of femoral head blood supply abnormalities is crucial for clinical diagnosis and treatment. This study successfully used the MR phase-contrast Q-flow technique to reliably quantify blood flow to the femoral head. Further, it validated the capacity of Q-flow to assess blood flow changes.

Although bone scans and dynamic contrast-enhanced MRI can monitor early blood supply abnormalities, they require the administration of radiotracers and gadolinium contrast agents (10). Color Doppler can detect changes in blood velocity in early osteonecrosis, but it faces challenges related to accuracy and repeatability, such as insonation angle problems, and an inability to account for spatial velocity profiles (14-17,19). The MR phase-contrast Q-flow technique offers an alternative approach for blood supply quantification, providing the benefits of contrast-free and accurate measurements. To successfully use Q-flow to reliably quantify blood flow to the femoral head, the MRI protocol was optimized. First, limitations related to the partial volume effect were addressed by enhancing the resolution. Previous studies have reported that a resolution of approximately one-quarter of the vessel diameter (pixel/vessel: 1.14/5) allows for the accurate measurement of blood flow parameters (34), and 16 voxels per vessel cross-section can achieve measurement accuracy within a 10% error margin (20,21). Therefore, this study adopted a similar imaging resolution strategy (pixel/vessel: 0.4/1.6, approximately 64 interpolated voxels per cross-section) to ensure the accuracy of the measurement results.

The imaging resolution of this study (0.4×0.4 mm²) is comparable to the resolutions of previous studies (0.35–0.5 mm) that successfully measured blood flow in vessels smaller than 2 mm in luminal diameter (22,23). In addition, unlike previous studies that focused only on the MFCA and LFCA (11,13), this study extended the measurements to include the deep MFCA (which supplies the femoral head directly) and ascending LFCA (which supplies the femoral head through the anterior retinacular artery). This refined protocol improved the accuracy of femoral blood supply assessment by omitting branches not serving the femoral head, including the descending branches of the MFCA, and the descending and transverse branches of the LFCA (25,27). The measurement positions, located away from the femoral head, may facilitate analysis in cases of femoral neck fracture and core decompression.

High-resolution Q-flow demonstrated acceptable reliability in measuring the femoral head vessels, yielding acceptable retest agreement in all the arterial flow parameters except the RF, along with strong inter- and intra-agreement in this study. However, a few MFCAs and their corresponding deep branches could not be detected by REACT because of the flow variability to the femoral head. For example, the absence of the MFCA may occur if the piriformis branch predominantly supplies blood to the femoral head (5,35). Despite these challenges, the MPR and MIP images of the REACT facilitated the precise localization of the Q-flow scanning plane, ensuring reliable repeatability for the feeding arteries (only the RF did not consistently achieve acceptable agreement). Although these vessels are diminutive in size, high-resolution Q-flow effectively illustrated the lumen, enabling the precise delineation of the regions of interest (ROIs) along the lumen. Consequently, this led to strong inter- and intra-agreement in measuring the target vessels, except for the RF. When arterial and venous vessels are closely situated, delineating ROIs may inadvertently incorporate peripheral pixels from accompanying reversed-flow vessels, which may lead to considerable variability in RF between measurements. Nonetheless, its effect on other parameters is limited because peripheral pixels typically exhibit lower flow rates and contribute less than the central pixel (36). Additionally, successfully measuring the deep MFCVs using Q-flow is a significant challenge due to their extremely sluggish blood flow.

This study further validated the utility of Q-flow in detecting flow reduction in the femoral head of SLE patients on corticosteroids. Corticosteroids induce a femoral intra-osseous compartment syndrome characterized by the hyperplasia of marrow fat cells, intra-osseous hypertension, vascular compression, and intravascular coagulation (9), ultimately resulting in hemodynamic changes in both blood-supplying arteries and draining veins. In addition, corticosteroid-induced osteoporosis may further contribute to reduced perfusion in the femoral head (37,38).

Previous studies have shown that the main blood supply from the superior retinacular arteries is significantly impaired in the early stage of corticosteroid use (39,40). These studies indicate that the vascular injury process begins with impaired blood flow in the true superior retinacular arteries. If perfusion is not restored, the femoral head enters a silent necrotic phase, during which radiographs may still appear normal. In some cases, revascularization may occur through small peripheral neovascularization extending into the ischemic region; however, this process is often disrupted by structural damage, ultimately leading to irreversible osteonecrosis. This sequence typically unfolds over approximately 13.2 months.

Using dynamic MRI, Nakamura et al. (41) showed that within four months of corticosteroid therapy, peak enhancement values in the femoral heads of adult patients recovered from 10.7% to 12.7%, but remained lower than the 20% observed in healthy adults. In addition, an early decreased blood supply was also found in steroid-induced osteonecrosis of the rabbit femoral head (42). Our study found comparable blood flow alterations in the SLE patients who had received hormone therapy for an average of less than 9 months, evidenced by reduced blood flow parameters in the LFCAs, MFCAs, and their deep branches, highlighting the ability of high-resolution Q-flow imaging to detect early changes in arterial blood flow to the femoral head. The early administration of hormones can significantly compromise the blood supply to the femoral head; however, this does not inevitably lead to femoral head necrosis, as the initial blood supply impairment is often followed by a subsequent process of revascularization (39-41). Further, this process may lead to a blood supply that exceeds the physiological blood supply levels of the femoral head (13,43). Consequently, Q-flow is a promising tool for dynamically assessing the recovery of femoral head blood flow.

This study had two limitations. First, it had a small sample size. Second, although we aimed to align the measurement plane perpendicular to the artery, it was not necessarily perpendicular to the accompanying vein, introducing potential inaccuracies in venous measurements. This challenge was at times exacerbated by the presence of multiple accompanying veins, complicating precise venous drainage quantification. Additionally, blood flow is influenced by factors such as age and cardiac function (40). Thus, relative blood flow parameters normalized by these factors may be more practical (e.g., target vessel flow relative to femoral artery flow or cardiac output). However, for longitudinal assessments of specific patients (e.g., for evaluating the effect of core decompression) or when using the healthy side as a reference (e.g., for evaluating unilateral femoral neck fractures), the use of measured raw parameters is also feasible (44).

Conclusions

The high-resolution MR phase-contrast Q-flow technique, which had acceptable reliability, was able to detect early changes in arterial blood flow to the femoral head, highlighting its potential clinical applicability.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-411/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-411/dss

Funding: This work was supported by the Sichuan Medical Association (Sichuan Province Medical Research Youth Innovation Project) (No. Q20020).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-411/coif). X.Z. is an employee of Philips Healthcare. All authors report funding from the Sichuan Medical Association (Sichuan Province Medical Research Youth Innovation Project) (No. Q20020). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Deyang People’s Hospital Ethics Committee (approval No. 2020-04-118-K01), and informed consent was obtained from all the participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Mont MA, Salem HS, Piuzzi NS, Goodman SB, Jones LC. Nontraumatic Osteonecrosis of the Femoral Head: Where Do We Stand Today?: A 5-Year Update. J Bone Joint Surg Am 2020;102:1084-99. [Crossref] [PubMed]
2. Zhao D, Zhang F, Wang B, Liu B, Li L, Kim SY, Goodman SB, Hernigou P, Cui Q, Lineaweaver WC, Xu J, Drescher WR, Qin L. Guidelines for clinical diagnosis and treatment of osteonecrosis of the femoral head in adults (2019 version). J Orthop Translat 2020;21:100-10. [Crossref] [PubMed]
3. Liu LH, Zhang QY, Sun W, Li ZR, Gao FQ. Corticosteroid-induced Osteonecrosis of the Femoral Head: Detection, Diagnosis, and Treatment in Earlier Stages. Chin Med J (Engl) 2017;130:2601-7. [Crossref] [PubMed]
4. Wang B, Li L, Wang Y, Wang Z, Li C, Fu W, Qiu X, Zhao D. Digital Subtraction Angiography and Magnetic Resonance Imaging-Based Staging of Circulatory Obstruction in the Femoral Head During Osteonecrosis of the Femoral Head Development. Ann Plast Surg 2020;85:677-84. [Crossref] [PubMed]
5. Zlotorowicz M, Czubak J, Kozinski P, Boguslawska-Walecka R. Imaging the vascularisation of the femoral head by CT angiography. J Bone Joint Surg Br 2012;94:1176-9. [Crossref] [PubMed]
6. Zlotorowicz M, Czubak J, Caban A, Kozinski P, Boguslawska-Walecka R. The blood supply to the femoral head after posterior fracture/dislocation of the hip, assessed by CT angiography. Bone Joint J 2013;95-B:1453-7. [Crossref] [PubMed]
7. Liao Z, Bai Q, Ming B, Ma C, Wang Z, Gong T. Detection of vascularity of femoral head using sub-millimeter resolution steady-state magnetic resonance angiography-initial experience. Int Orthop 2020;44:1115-21. [Crossref] [PubMed]
8. Liao Z, Liu C, Wu B, Ma C, Ming B, Zhou Q, Zhang X, Zhou S, Chen Y. Evaluating blood supply changes in the osteonecrosis of the femoral head using gadobutrol-based steady-state MR angiography. Eur Radiol 2023;33:8597-604. [Crossref] [PubMed]
9. Hines JT, Jo WL, Cui Q, Mont MA, Koo KH, Cheng EY, et al. Osteonecrosis of the Femoral Head: an Updated Review of ARCO on Pathogenesis, Staging and Treatment. J Korean Med Sci 2021;36:e177. [Crossref] [PubMed]
10. Dyke JP, Aaron RK. Noninvasive methods of measuring bone blood perfusion. Ann N Y Acad Sci 2010;1192:95-102. [Crossref] [PubMed]
11. Graif M, Schweitzer ME, Nazarian L, Matteucci T, Goldberg BB. Color Doppler hemodynamic evaluation of flow to normal hip. J Ultrasound Med 1998;17:275-80; quiz 281-2.
12. Yousefzadeh DK, Jaramillo D, Johnson N, Doerger K, Sullivan C. Biphasic threat to femoral head perfusion in abduction: arterial hypoperfusion and venous congestion. Pediatr Radiol 2010;40:1517-25. [Crossref] [PubMed]
13. Hassan SZ, Abdalla MA, El-Azizi HM, Mohamed NN. Ultrasonographic evaluation of haemodynamic flow changes to the hip in systemic lupus erythematosus patients: Correlation with risk factors for osteonecrosis. The Egyptian Rheumatologist 2018;40:161-6.
14. Krishnamurthy U, Yadav BK, Jella PK, Haacke EM, Hernandez-Andrade E, Mody S, Yeo L, Hassan SS, Romero R, Neelavalli J. Quantitative Flow Imaging in Human Umbilical Vessels In Utero Using Nongated 2D Phase Contrast MRI. J Magn Reson Imaging 2018;48:283-9. [Crossref] [PubMed]
15. Du Y, Ding H, He L, Yiu BYS, Deng L, Yu ACH, Zhu L. Quantitative Blood Flow Measurements in the Common Carotid Artery: A Comparative Study of High-Frame-Rate Ultrasound Vector Flow Imaging, Pulsed Wave Doppler, and Phase Contrast Magnetic Resonance Imaging. Diagnostics (Basel) 2022;12:690. [Crossref] [PubMed]
16. Applegate GR, Thaete FL, Meyers SP, Davis PL, Talagala SL, Recht M, Wozney P, Kanal E. Blood flow in the portal vein: velocity quantitation with phase-contrast MR angiography. Radiology 1993;187:253-6. [Crossref] [PubMed]
17. Blanco P. Volumetric blood flow measurement using Doppler ultrasound: concerns about the technique. J Ultrasound 2015;18:201-4. [Crossref] [PubMed]
18. Markl M, Schnell S, Wu C, Bollache E, Jarvis K, Barker AJ, Robinson JD, Rigsby CK. Advanced flow MRI: emerging techniques and applications. Clin Radiol 2016;71:779-95. [Crossref] [PubMed]
19. Maier IL, Hofer S, Joseph AA, Merboldt KD, Tan Z, Schregel K, Knauth M, Bähr M, Psychogios MN, Liman J, Frahm J. Carotid artery flow as determined by real-time phase-contrast flow MRI and neurovascular ultrasound: A comparative study of healthy subjects. Eur J Radiol 2018;106:38-45. [Crossref] [PubMed]
20. Tang C, Blatter DD, Parker DL. Accuracy of phase-contrast flow measurements in the presence of partial-volume effects. J Magn Reson Imaging 1993;3:377-85. [Crossref] [PubMed]
21. Greil G, Geva T, Maier SE, Powell AJ. Effect of acquisition parameters on the accuracy of velocity encoded cine magnetic resonance imaging blood flow measurements. J Magn Reson Imaging 2002;15:47-54. [Crossref] [PubMed]
22. Promelle V, Bouzerar R, Milazzo S, Balédent O. Quantification of blood flow in the superior ophthalmic vein using phase contrast magnetic resonance imaging. Exp Eye Res 2018;176:40-5. [Crossref] [PubMed]
23. Ambarki K, Hallberg P, Jóhannesson G, Lindén C, Zarrinkoob L, Wåhlin A, Birgander R, Malm J, Eklund A. Blood flow of ophthalmic artery in healthy individuals determined by phase-contrast magnetic resonance imaging. Invest Ophthalmol Vis Sci 2013;54:2738-45. [Crossref] [PubMed]
24. Yoneyama M, Zhang S, Hu HH, Chong LR, Bardo D, Miller JH, Toyonari N, Katahira K, Katsumata Y, Pokorney A, Ng CK, Kouwenhoven M, Van Cauteren M. Free-breathing non-contrast-enhanced flow-independent MR angiography using magnetization-prepared 3D non-balanced dual-echo Dixon method: A feasibility study at 3 Tesla. Magn Reson Imaging 2019;63:137-46. [Crossref] [PubMed]
25. Seeley MA, Georgiadis AG, Sankar WN. Hip Vascularity: A Review of the Anatomy and Clinical Implications. J Am Acad Orthop Surg 2016;24:515-26. [Crossref] [PubMed]
26. Xiao J, Yang XJ, Xiao XS. DSA observation of hemodynamic response of femoral head with femoral neck fracture during traction: a pilot study. J Orthop Trauma 2012;26:407-13. [Crossref] [PubMed]
27. 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 Br 2011;93:1471-4. [Crossref] [PubMed]
28. Lazaro LE, Klinger CE, Sculco PK, Helfet DL, Lorich DG. The terminal branches of the medial femoral circumflex artery: the arterial supply of the femoral head. Bone Joint J 2015;97-B:1204-13. [Crossref] [PubMed]
29. Netter FH. Atlas of human anatomy, 7th edn. Philadelphia, PA: Elsevier; 2019.
30. Caggiati A, Bergan JJ, Gloviczki P, Eklof B, Allegra C, Partsch H. Nomenclature of the veins of the lower limb: extensions, refinements, and clinical application. J Vasc Surg 2005;41:719-24. [Crossref] [PubMed]
31. Morgan AG, Thrippleton MJ, Stringer M, Jin N, Wardlaw JM, Marshall I. Repeatability and comparison of 2D and 4D flow MRI measurement of intracranial blood flow and pulsatility in healthy individuals and patients with cerebral small vessel disease. Front Psychol 2023;14:1125038. [Crossref] [PubMed]
32. Chen X, Qin L, Pan D, Huang Y, Yan L, Wang G, Liu Y, Liang C, Liu Z. Liver diffusion-weighted MR imaging: reproducibility comparison of ADC measurements obtained with multiple breath-hold, free-breathing, respiratory-triggered, and navigator-triggered techniques. Radiology 2014;271:113-25. [Crossref] [PubMed]
33. Chang KH, Lee YH, Chen CY, Lin MF, Lin YC, Chen JH, Chan WP. Inter- and Intra-Rater Reliability of Individual Cerebral Blood Flow Measured by Quantitative Vessel-Flow Phase-Contrast MRI. J Clin Med 2020;9:3099. [Crossref] [PubMed]
34. Dyvorne HA, Knight-Greenfield A, Besa C, Cooper N, Garcia-Flores J, Schiano TD, Markl M, Taouli B. Quantification of hepatic blood flow using a high-resolution phase-contrast MRI sequence with compressed sensing acceleration. AJR Am J Roentgenol 2015;204:510-8. [Crossref] [PubMed]
35. Zlotorowicz M, Czubak-Wrzosek M, Wrzosek P, Czubak J. The origin of the medial femoral circumflex artery, lateral femoral circumflex artery and obturator artery. Surg Radiol Anat 2018;40:515-20. [Crossref] [PubMed]
36. Koerte I, Haberl C, Schmidt M, Pomschar A, Lee S, Rapp P, Steffinger D, Tain RW, Alperin N, Ertl-Wagner B. Inter- and intra-rater reliability of blood and cerebrospinal fluid flow quantification by phase-contrast MRI. J Magn Reson Imaging 2013;38:655-62. [Crossref] [PubMed]
37. Weinstein RS. Glucocorticoid-induced osteoporosis and osteonecrosis. Endocrinol Metab Clin North Am 2012;41:595-611. [Crossref] [PubMed]
38. Griffith JF, Yeung DK, Tsang PH, Choi KC, Kwok TC, Ahuja AT, Leung KS, Leung PC. Compromised bone marrow perfusion in osteoporosis. J Bone Miner Res 2008;23:1068-75. [Crossref] [PubMed]
39. Atsumi T, Kuroki Y. Role of impairment of blood supply of the femoral head in the pathogenesis of idiopathic osteonecrosis. Clin Orthop Relat Res 1992;22-30.
40. Atsumi T, Kuroki Y, Yamano K, Muraki M. Revascularization in nontraumatic osteonecrosis of the femoral head. Clin Orthop Relat Res 1996;168-73. [Crossref] [PubMed]
41. Nakamura J, Ohtori S, Watanabe A, Nakagawa K, Inoue G, Kishida S, Harada Y, Suzuki M, Takahashi K. Recovery of the blood flow around the femoral head during early corticosteroid therapy: dynamic magnetic resonance imaging in systemic lupus erythematosus patients. Lupus 2012;21:264-70. [Crossref] [PubMed]
42. Wang G, Zhang CQ, Sun Y, Feng Y, Chen SB, Cheng XG, Zeng BF. Changes in femoral head blood supply and vascular endothelial growth factor in rabbits with steroid-induced osteonecrosis. J Int Med Res 2010;38:1060-9. [Crossref] [PubMed]
43. Bluemke DA, Petri M, Zerhouni EA. Femoral head perfusion and composition: MR imaging and spectroscopic evaluation of patients with systemic lupus erythematosus and at risk for avascular necrosis. Radiology 1995;197:433-8. [Crossref] [PubMed]
44. Hirata T, Konishiike T, Kawai A, Sato T, Inoue H. Dynamic magnetic resonance imaging of femoral head perfusion in femoral neck fracture. Clin Orthop Relat Res 2001;294-301. [Crossref] [PubMed]