Osteoarthritis of the hip joint in elderly patients is most commonly atrophic, with low parameters of acetabular dysplasia and possible involvement of osteoporosis

Osteoarthritis (OA) is associated with increased bone formation at a local site, whereas osteoporosis involves decreased bone formation with increased bone resorption, resulting in reduced bone mass [1]. Therefore, an inverse relationship has been identified between OA and osteoporosis [2]. However, excessive bone resorption characteristic of osteoporosis has also been found to occur in the early stages of OA [3], drawing attention to the functional unit consisting of articular cartilage and subchondral bone [4]. Molecular crosstalk between cartilage and subchondral bone contributes to the pathogenesis of OA [5‐8]. Microstructural damage to the subchondral bone and subsequent bone loss by increased remodeling is associated with the pathogenesis of early-stage OA [9, 10]. Anti-osteoporosis drugs such as bisphosphonates are candidate therapeutic drugs with which to prevent the progression of OA [11, 12]. However, the curative effects of these drugs are controversial. The pattern of the biological bone response in OA can be classified as hypertrophic or atrophic based on the presence or absence of osteophytes, respectively [13‐15]. As a result of the heterogeneity of OA, differences in radiographic patterns may be useful in elucidating differences in a patient’s reaction to an anti-osteoporosis drug.

OA is a mechanically induced disease affected by both genetic and acquired factors [4]. Acetabular dysplasia is one of the anatomical risk factors for OA [16, 17], and genetic polymorphism associated with the risk of developmental dysplasia of the hip was recently identified [18]. OA associated with acetabular dysplasia secondary to genetic factors is predicted to have a relatively early onset. In Japan, developmental acetabular dysplasia of the hip is reportedly among the most common causes of hip OA [19, 20]. Meanwhile, the incidence of acetabular dysplasia is lower in elderly patients [21]. The incidence and prevalence of OA after 50 years of age are higher in women than in men, suggesting that estrogen participates in the development of OA [22]. These findings suggest that the etiologic factors of OA in elderly patients may differ from those in young patients and that osteoporosis is involved in the onset of OA in elderly patients. However, few studies have investigated this hypothesis.

We conducted a cross-sectional radiographic study in patients with hip OA and examined the characteristic of hip OA in the elderly, including the association between parameters of acetabular dysplasia of the hip and age as well as the osteoblastic response, to examine the potential involvement of osteoporosis.

The mean Sharp’s angles were 46.1°, 43.9°, 41.1°, and 41.0° in patients aged <60, 60–69, 70–79, and ≥80 years, respectively. The corresponding ARO angles were 23.0°, 20.6°, 16.3°, and 17.0°, respectively, and the corresponding mean AHIs were 61.8, 64.8, 71.9, and 72.1%, respectively (Table 1). As the patients became older, Sharp’s angle and the ARO angle decreased significantly (R = −0.3686, p < 0.01 and R = −0.2603, p < 0.01, respectively) and the AHI increased significantly (R = 0.3292, p < 0.01) (Fig. 2). Namely, as the patients became older, the parameters of acetabular dysplasia decreased. In a comparison according to age group, Sharp’s angle and the ARO angle were significantly smaller in patients aged 70–79 and ≥80 years. The AHI was significantly larger in patients aged 70–79 and ≥80 years (Fig. 3).

The frequency of the atrophic type of OA according to Bombelli classification was 17.5, 40.0, 53.5, and 53.5% in patients aged <60, 60–69, 70–79, and ≥80 years, respectively, showing an increase with age (Fig. 4). The frequency of vertebral body fracture was 11.4% in the hypertrophic type, 12.8% in the normotrophic type, and 25.9% in the atrophic type, with the atrophic type showing the highest frequency (Fig. 5).

The atrophic type of OA occurred in patients of a significantly high age. Sharp’s angle and the ARO angle in the atrophic type were smaller than those in the other types. The AHI in atrophic type was larger than that in the other types. Namely, the parameters of acetabular dysplasia decreased in the atrophic type. On the other hand, the hypertrophic type occurred in patients of a relatively low age, in whom the parameters of acetabular dysplasia increased (Fig. 6).

We conducted a radiographic study of patients with hip OA requiring total hip arthroplasty and evaluated the correlation between parameters of acetabular dysplasia and aging. Additionally, we estimated the osteoblastic response in patients with OA according to Bombelli classification. As our patients become older, the parameters of acetabular dysplasia decreased and the incidence of the atrophic type of OA increased.

Acetabular dysplasia is an anatomical risk factor for hip OA. Reijman et al. [16] reported that acetabular dysplasia (center-edge angle of <25°) was associated with a 4.3-fold increased risk for incident radiographic hip OA. Jingushi et al. [19] reported that Sharp’s angles of >45° or 50° had significantly increased odds ratios for OA (5- or 65-fold greater, respectively) in comparison to <40°.

However, a recent epidemiological study of hip OA in Japan demonstrated that the incidence of acetabular dysplasia was lower in elderly patients [21]. Lane et al. [30] also reported that the association between acetabular dysplasia and OA in white women aged ≥65 years was not determined. In the present study, the indices of acetabular dysplasia in the elderly age groups (70–79 and ≥80 years of age) were significantly lower than those in the younger age groups. Additionally, these acetabular dysplasia parameters were negatively correlated with age.

Anatomical abnormalities such as acetabular dysplasia produce abnormal or incongruous loading on articular cartilage for long periods, resulting in cartilage degeneration at a comparatively younger age [13, 31]. Therefore, the incidence of OA secondary to acetabular dysplasia is high in comparatively young individuals, but low in those who are older. OA secondary to acetabular dysplasia has onset in middle age, and induces a brisk osteoblastic response due to the abnormal mechanical load, resulting in hypertrophy.

Meanwhile, primary OA in hip joints without anatomical abnormalities will maintain essentially normal cartilage under normal conditions of loading at a younger age, but nonmechanical etiological factors such as osteoporosis may cause cartilage failure at a later age. Therefore, primary OA without acetabular dysplasia shows higher incidence in older individuals, and induces a weak osteoblastic response to osteoporosis, resulting in atrophy.

OA is generally characterized by cartilage degradation, new bone spur formation, and osteosclerotic changes of the subchondral bone. However, there is little osteoblastic activity in the atrophic type of OA, so no osteophytes are observed. The atrophic type of OA is associated with a higher risk of joint destruction [32], more rapid disease progression [33], lower mineral density, and higher risk of osteoporotic fracture than other types of OA [34]. Our data showed that the incidence of atrophic OA was higher as the patients became older. In addition, the incidence of vertebral fracture in the atrophic type was higher than in other types of OA. Schnitzler et al. [32] also reported a high prevalence of vertebral fracture in the atrophic type of OA. These views support the concept that osteoporosis is involved in the etiology of the atrophic type of OA.

However, a limitation of our study was the absence of bone mineral density measurements. Because this osteoporosis study was retrospective, we could not collect sufficient data on bone mineral density. Therefore, we inferred that osteoporosis was potentially related to the etiology of atrophic OA by showing the high incidence of vertebral compression fractures in atrophic OA. A prospective epidemiological study to clarify this concept is necessary.

Osteoporosis may induce microstructural damage to the subchondral bone by increased remodeling, resulting in the onset of OA [9, 35]. A prospective randomized study of alendronate treatment for hip OA showed significant improvement in pain but no significant prevention of structural disease progression [36]. More effective curative results may be expected if the patients are limited to the atrophy type strongly influenced by osteoporosis. Microstructural damage to the subchondral bone induces inappropriate crosstalk between osteoblasts and chondrocytes [6‐8]. It is probably important to begin treatment for osteoporosis before such events occur. When planning the optimal OA treatment strategy, it is important that we consider the age of onset and heterogeneity of the etiology.

Various factors, including genetic predisposition, sex, age, obesity, physical activity, joint injury, joint malalignment, and abnormal joint shape, affect the subchondral bone integrity [37]. We statistically demonstrated in this radiographic study that the atrophic type of OA has a predominantly older age at onset and is affected by osteoporosis, whereas the hypertrophic type of OA has a predominantly younger age at onset and is affected by acetabular dysplasia. Because the atrophic type of OA is not accompanied by anatomical abnormalities, it is difficult to predict the onset of atrophic OA before cartilage degradation. No serum biomarkers that are able to demonstrate differences between the atrophic and hypertrophic types of OA have been identified [14]. Additionally, the older onset of atrophic OA may be affected not only by osteoporosis but also by mild acetabular dysplasia [38]. Some studies have reported that the patterns of osteoblastic reaction in the osteoarthritic hip were not related to the general bone mineral density [34, 39]. The uncertain definition of the atrophic type of OA and the lack of appropriate biomarkers are problematic.

This cross-sectional study showed that atrophic OA increases with aging. Various factors other than anatomical abnormalities affect the pathogenesis of OA. Osteoporosis may be a contributor, but this remains controversial. Pathological classification in individual OA cases may be helpful for selection of drug treatment. Further research is needed to understand the difference in the types of OA and heterogeneity of the etiology.

In elderly patients with hip OA, the parameters of acetabular dysplasia decreased and the atrophic type increased as patients become older. The frequency of vertebral body fracture was higher in the atrophic type, suggesting the involvement of osteoporosis in the onset of hip OA.