Avascular necrosis of the femoral head (ANFH) is where necrosis, due to various causes, occurs in the femoral head where there has been disruption to the blood supply [1]. ANFH, an intractable condition/disease, has been increasingly detected in recent years mainly due to the increased use of steroids and improvements in radiological technology.

Despite extensive research, in most cases the clear cause of ANFH cannot be identified owing to various complex factors. Sometimes ANFH may be linked to trauma, alcohol, steroid treatment in patients with autoimmune diseases, and several diseases such as hereditary thrombosis, lupus, hemoglobinopathy, and chronic renal disease [1,2]. The pathology of ANFH currently is as follows: accumulated cell stress, fat emboli, and coagulopathy due to various factors leading to lowered blood flow which can cause ischemia and eventually lead to bone necrosis [2].

Treatment can be categorized into surgical and non-surgical intervention. Surgical treatment is performed according to the characteristics of the lesion, stage, type, patient’s age, and occupation. Preservative non-surgical treatment is recommended for younger patients without head collapse [3]. Non-surgical treatments include physiotherapy, weight-bearing restrictions, and medicinal drug treatment (anticoagulants, vasodilators, and phosphates). However, there is no clear evidence that the types of drugs utilized to treat ANFH stop the progress of bone necrosis. Moreover, some drugs have several side effects e.g., osteonecrosis of the mandible with a relatively high proportion of cortical bone and atypical fracture of the femur [3,4].

Clinical research and experimental research studies have reported the efficacy of Korean medicinal treatments such as acupuncture, pharmacoacupuncture, and herbal medicine in treating musculoskeletal diseases [5–8]. In clinical practice, various Korean medicine treatments such as pharmacoacupuncture have been reported to be effective treatments for ANFH [9].

However, to develop a better understanding of the application of Korean medicine treatments for ANFH, we need to identify the components of Korean medicine treatments that contribute to anti-inflammation, pain control, and bone protection which are disease-specific for ANFH. Therefore, this study reviewed rat models of ANFH to assess treatment outcomes using single ingredients and bioactive compounds.

The PubMed database was searched for publications prior to April 26, 2022. The following search string was used: [“Femur Head Necrosis”(Mesh)] AND [“Rats” (Mesh)]. From the 184 results retrieved, research related to natural substances was selected. The Google Scholar database was also searched for publications published before April 26, 2022, up to 99 pages. The search term used was: (Femur Head Necrosis) AND (natural extract) AND (rat). There was no limits on the language.

In this review, the focus was on studies of water- or alcohol-based natural extracts or bioactive compounds of natural substances. In addition to avascular necrosis, general bone necrosis on the femoral head was included. Although in vivo studies of natural extracts were included, studies on complex medications such as herbal medicines, were excluded. In addition, studies involving mice and those performed only in vitro were excluded. Among the selected articles, those that showed insufficient results were excluded.

Data regarding the following parameters were extracted: authors, publication year, animal species, sex, number of animals, induction type/period, administration route/period, morphological observation, anti-inflammatory action, apoptosis, bone formation, angiogenesis, adipogenesis, and safety assessment.

The initial search included 72 articles. After screening titles and abstracts, 45 articles were excluded based on the selection criteria for this study. After a full-text assessment, one more article was excluded. There were 26 articles eligible for review (Fig. 1; Tables 1–3 [10–35]).

Among the rat species used in the studies, Sprague-Dawley (SD) rats (76.9%) were the most commonly used species, followed by Wistar rats (23.1%). Male rats (61.5%) were preferred. The most commonly used method for induction of ANFH was the steroid-only method (46.2%). Among these steroids, dexamethasone (26.9%) was used the most, followed by methylprednisolone acetate (7.7%), prednisolone acetate (7.7%), and methylprednisolone (3.8%). Among the selected articles, eight (30.8%) studies used both endotoxins and steroids. Moreover, all eight articles used lipopolysaccharide (LPS) as the endotoxin, and methylprednisolone (26.9%) was the most commonly used steroid in this combination, though methylprednisolone sodium succinate (3.8%) was also used. In addition, 5 studies used liquid containing alcohol for the induction of ANFH (19.2%), while the surgical method (3.8%) was the least used. Regarding the administration of a single ingredient or bioactive compound, oral gavage and intraperitoneal injection were used in 53.8% and 38.5% of the studies, respectively.

Histological analysis using hematoxylin and eosin (H&E) staining (92.3%) was the predominant stain used for the morphological assessment of ANFH, followed by micro-computed tomography (micro-CT, 42.3%), angiography using micro-CT (11.5%), and dual-energy X-ray absorptiometry (7.7%). The incidence rate (53.8%) and empty lacunae rate (46.2%) were used for histological analysis, which was highly comparable with the determined number of rats with ANFH according to the appraisal standard.

Two studies examined inflammation (7.7%): interleukin (IL)-1 beta (IL-1β), IL-6, and tumor necrosis factor-alpha were measured in both studies, while p65, toll-like receptor 4, and myeloid differentiation primary response 88 were measured in one study (3.8 %).

Factors related to apoptosis were determined in 14 articles (53.8%), with terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (38.5%) being the most commonly used method. B-cell lymphoma-2 (bcl-2, 26.9%), bcl-2-associated X (26.9%), caspase-3 (26.9%), caspase-9 (11.5%), and cytochrome c (11.5%) were also studied.

Factors related to bone formation were identified in 17 (65.3 %) articles. Alizarin Red S staining (30.8%) and dynamic bone formation using the fluorochrome labeling assay (15.4%) were used in the aforementioned articles. Osteocalcin (38.5%), alkaline phosphatase (42.3%), runt-related transcription factor 2 (38.5%), collagen Type 1 (26.9%), bone morphogenetic protein 2 (19.2%), osteoprotegerin (11.5%), and receptor activator of nuclear factor kappa B ligand (11.5%) were also identified.

Angiogenesis was assessed in 11 articles (42.3%) and angiography using micro-CT (11.5%), and microfilm perfusion (3.8%) were used as the methods. Vascular endothelial growth factor (VEGF, 34.6%) was mainly observed factor.

Factors related to adipogenesis were identified in seven articles (26.9%), in which Oil Red O staining (11.5%) was used. Total cholesterol (15.4%), triglycerides (15.4%), low-density lipoprotein cholesterol (LDL, 15.4%), high-density lipoprotein cholesterol (11.5%), and peroxisome proliferator-activated receptor gamma (PPARγ, 11.5%) levels were measured.

Of all the selected articles, six (23.1%) studies used water- or alcohol-based extraction of single ingredient natural substances, and 20 studies used the bioactive compounds derived from natural substances, including epimedium and icariin (5 cases, 19.2%), antler (2 cases, 7.7%), Rhizoma drynariae and luteolin (2 cases, 7.7%). Twelve articles (46.2%) mentioned the stability of each single ingredient and bioactive compound, and six articles (23.1%) reported that the single ingredient or bioactive compounds used in the study alleviated symptoms of ANFH in a dose-dependent manner.

Rats are commonly used animals in research for many reasons; they grow fast, are easy to handle, and share physiological and metabolic traits with humans [36]. Furthermore, ANFH induced rats clearly display the early stages of ANFH clinical pathology, which are difficult to study in humans [37]. However, rats differ in many aspects depending on the species. For instance, SD rats originate from the selective breeding of Wistar rats [38]. SD rats have different bone characteristics than Wistar rats for example SD rats have a much faster rate of progression of osteoporosis and show more dramatic articular atrophy [39]. For these reasons, SD rats are the most commonly used animals in research. Huang et al [21] and Yue et al [33] studied the estrogen-like activities of icariin in female rats. However, except for gender-specific research, male SD rats are normally used.

There were four methods used to induce ANFH: steroid treatment, steroids with LPS injection, oral administration of ethanol, and surgery. The use of steroids induces histological evidence of the early stages of ANFH [40]. It has been reported by Zhan et al [31] that inducing ANFH with dexamethasone alone takes a long time but has the benefit of having a mechanism similar to that in humans. The incidence rate of ANFH depends on the type, volume, and frequency of steroid administration [36]. Among the studies where steroid-induced ANFH was performed and the incidence rate was reported, subcutaneous injection of methylprednisolone acetate (21 mg/kg) for 6 weeks showed the highest rate of induction of ANFH, Jiang et al [17] reported a 92% rate of induction. Moreover, steroid induction is related to adipogenesis, apoptosis of osteocytes and osteoblast, fat embolism, vascular thrombosis, and oxidative stress [41]. Thus, steroid-only induction is suitable for studies on effect an ingredient or compound has on a particular mechanism.

In the studies included in this review, apoptosis, bone formation, and angiogenesis were reported to be the most common mechanisms in steroid-only induced ANFH studies, and adipogenesis was also observed four times. All studies that confirmed inflammation used steroid-only induced ANFH [25,32].

LPS, also known as an endotoxin, is used along with steroids to induce inflammation [36]. Using a LPS with steroids results in a shorter induction time, a larger area of bone necrosis, and a higher death rate compared with single steroid induction of ANFH [40]. In the studies reviewed, the average induction period for the use of a single steroid was 5.9 weeks. On the other hand, using LPS with steroids took an average of 5.38 days to induce ANFH. Induction of ANFH with LPS and steroids has been reported to result in a relatively high death rate [40]; however, no study in this review has reported death using LPS with steroids.

Inflammation induced by LPS hinders vascular function, induces apoptosis of osteoblasts, suppresses bone formation, increases bone resorption, and downregulates osteocalcin expression [42]. The use of LPS with steroids is a suitable method for studying materials that have these mechanisms. In this review, studies using LPS with steroids for induction of ANFH have mostly confirmed bone formation and apoptosis.

Rat models of ANFH using ethanol are useful for investigating the cause of disease associated with alcohol and evaluating treatment [40].

It has been reported that 4–6 months of oral gavage administration of over 45% ethanol in rats can induce bone necrosis, but 7 days of administration of 5% ethanol can also induce bone necrosis [43]. The Lieber-DeCarli liquid diet is commonly used for ethanol induction and has the advantage of efficient use of ethanol in animal experiments [43,44]. Among the articles in this review, studies using the Lieber-DeCarli liquid diet used 8% ethanol. Only the study by Yang et al [26] used 5% to 30% ethanol, increasing by 5% every week, but the reason for this was not stated.

Inducing ANFH using ethanol can cause hyperlipidemia. However, it has been reported that this has no relation to proinflammatory reactions via toll-like receptor 4, which plays an important role in alcoholic fatty liver disease [43]. All five studies that used ethanol to induce ANFH studied bone formation-related mechanisms, and apoptosis was studied in four of the articles. Two studies examined the mechanisms of adipogenesis. Research performed by Yu et al [27] proposed that adipogenesis induced by ethanol is related to the AMPK pathway, and that it can be alleviated by chrysophanic acid.

Chemical induction and non-traumatic drug models of ANFH result in a vast area of the femoral head being necrotized, thus have the disadvantage of not being able to precisely imitate the clinical situation [45]. However, surgery-induced necrosis is closely related to blood circulation damage near the femoral head and has the advantage of reflecting the pathology of traumatic ANFH [40].

In this review, only one study by Shi et al [15], used surgery for the induction of ANFH. The vast majority of studies focused only on ANFH induced by steroids or ethanol, this may be a result of the expertise required to perform surgical methods to induce ANFH. The traditional method of inducing ANFH by surgery involves cutting all retinacular blood vessels and the round ligament of the femoral head to induce femoral head dislocation. However, a new model of trauma induced ANFH has been reported whereby partial blood vessel abrasion of the femoral head is performed to interrupt the blood supply [45].

Comparing the 14 studies in this review that reported the incidence rate, the five studies that used steroids only reported an average incidence rate of 81.02%, four studies using steroids with LPS reported an average incidence rate of 79.99%, and four studies using ethanol induction of ANFH reported an average incidence rate of 78%. Therefore, average incidence rate between these methods of inducing ANFH was similar.

Steroid-only induction of ANFH is a suitable model for studying anti-inflammatory and adipogenesis-related mechanisms. It has also been associated with apoptosis, bone formation, and angiogenesis. Steroids with LPS for the induction of ANFH can be used to answer questions concerning anti-inflammatory mechanisms, apoptosis, and bone formation quickly. Ethanol induction of ANFH can be used to observe apoptosis, bone formation, and adipogenesis which may give insight into alcohol induced ANFH caused in patients who are alcoholics. For research on traumatic ANFH, induction surgery is the most suitable method of inducing ANFH.

Rat models of steroid induction of ANFH under hypoxic conditions [46], surgical induction of ANFH considering ageing and bone repair [47], and ANFH induction caused by tobacco smoke [48] have been reported since 2020. By using the appropriate rat model for the induction of ANFH which suits the disease characteristics (mechanism, purpose of study, and bone condition) of interest, the relevant mechanisms of action can be studied in the future.

Single ingredients and bioactive compounds were administered orally, intraperitoneally, intramuscularly, and locally. Different organs and pathways are responsible for the metabolism of each drug; thus, different effects may appear depending on the method of administration. The administration route must always be considered so that an unsuitable administration method is not used [49,50]. Differences in the methods of drug administration were not reviewed in this study.

Following 3–8 weeks post ANFH induction (average 5.7 weeks) the rats were sacrificed but no reason was given in any of the 7 studies for the choice of time point [13,17,22,28,31,32,34].

For morphological observations of induced ANFH micro-CT and angiography, sample observation, dual-energy X-ray absorptiometry, and histological analysis were used. Micro-CT is preferred by many researchers owing to its short scan time and non-invasive features. Furthermore, geometric analysis using morphological parameters, visualized using micro-CT, such as bone mineral density, bone volume, and trabecular thickness, are highly related to actual bone strength data that can be obtained using mechanical tests [51,52]. MRI is superior to normal radiographs because abnormalities in the bone interface can be observed [53]. Use of MRI as a very sensitive method for the early diagnosis of induced ANFH is advisable.

The rate of empty lacunae, which is the main characteristic of osteonecrosis in histological observations, and the incidence rate (according to histological diagnosis criteria) were measured and evaluated in several studies. Empty lacunae not only showed the degree of osteonecrosis but were also used as an index to compare between studies [54].

Staining with H&E was mainly used for microscopic observation, and bone cell lesions were observed using various histological characteristics such as the number, volume, density, and arrangement of trabeculae, as well as the degree of bone marrow cell debris accumulation. Three studies directly observed a sample of femoral head, and it was confirmed that the ANFH-induced femoral head showed cyanosis [30,31] and brittleness [23].

Toxicity was assessed using change in weight, survival rate, clinical signs, and blood glucose. The survival rate was mainly used to assess toxicity, but only Wang et al [11] reported deaths during the study period. Clinical signs such as activity and behavioral changes, physical appearance (e.g., fluffy hair), and joint exudate were observed [10,12,17,19,27]. It has been reported that toxicity can be assessed more precisely by observing clinical signs such as weight loss (of over 5%), piloerection, half-shut eyes, and motor activity [55].

Factors such as the levels of IL-1β, IL-6, and tumor necrosis factor-alpha have been studied in relation to inflammation, and bcl-2 and bcl-2-associated X have been studied with respect to apoptosis. Typical osteogenic factors, such as collagen Type 1 and osteocalcin, were studied as the factors most related to the osteogenic mechanism, while VEGF and PPARγ were observed to assess angiogenesis and adipogenesis, respectively. The most important signaling pathways in the pathological development of ANFH are osteogenesis-related phosphatidylinositol 3-kinase/protein kinase B signaling, receptor activator of nuclear factor kappa B ligand/receptor activator of nuclear factor kappa B/osteoprotegerin signaling, Wnt/β-catenin pathway, apoptosis-related Wnt/β-catenin, STAT1/caspase3, angiogenesis-related VEGF/FLK1 signaling, and lipogenesis-related PPARγ-mediated signaling pathways. However, the anti-inflammatory cytokines, t helper 17 (Th17) cells and IL-17 [56], and the apoptotic reactive oxygen species/c-Jun N-terminal kinases/c-Jun pathways [57] are recent areas of interest; thus, these mechanisms should be further elucidated in future studies.

Only bioactive compounds were studied after 2015, with the exception of the study by Liu et al [20]. Half of the studies on single ingredients were extracted using water, and there have been no studies on the differential effects of extraction methods. Of the single ingredients and bioactive compounds, Epimedium brevicornum was the most studied, and Cervi Cornu and R. drynariae were studied twice each. Traditional Korean medicine texts state that E. brevicornum displays properties relating to “strengthening sinews and bones” and “warming the kidney and invigorating yang.” Thus, E. brevicornum is widely used to treat knee and back pain, and as a tonic for aphrodisiacs, improving bone strength, and kidney health [11,21,23]. E. brevicornum is also commonly used to treat ANFH and osteoporosis [11,23]. Moreover, through pharmacological studies, E. brevicornum has been reported to have anti-inflammatory, anti-osteoporosis, blood circulation-improving, and anti-aging effects, and is also involved in the phosphatidylinositol 3-kinase/protein kinase B and epidermal growth factor/epidermal growth factor receptor signaling pathways [11,20,21,23,33].

The research articles on induced ANFH included in this review were varied. Of the many proposed mechanisms of ANFH, apoptosis, inflammation, lipid metabolism disorders, insufficient blood supply, and osteogenic/osteoclastic differentiation were the main focuses. In recent studies, Th17 cells and synovial cells in synovial tissue and the interactions between these cells have received much attention as key mechanisms, however, more research is necessary. Discovering a clinically effective treatment is difficult. ANFH is an intractable joint disease that interacts with various pathological mechanisms.

It is necessary to develop a rat model suitable to study the pathophysiology of ANFH. Specifically, to understand the action of immunomodulatory mechanisms and pathways such as those influenced by Th17 cells and their production of IL-17. In addition, it is essential to study the materials that affect Korean medicine pattern identification which is similar to the pathophysiology of ANFH (e.g., essence and blood deficiency, and dampness-heat). Furthermore, research limitations involved in single mechanisms need to be overcome by establishing the basis for diagnosis and treatment using Korean medicine by using next-generation sequencing analysis of tissue samples from ANFH patients.

Flow diagram of the selection of articles.

Summarized Data from Included Articles (2010–2014).

Summarized Data from Included Articles (2015–2018).

Summarized Data from Included Articles (2019–2022).