Introduction
Osteopaenia was initially a qualitative term describing bone that appeared less dense on X-ray examination. Since 1994, it has also taken on a quantitative definition, referring to bone mineral density (BMD) with a T-score between –1.0 and –2.5 standard deviations (SD). Notably, the term osteopaenia was not intended to define a disease, only a range of bone density that included individuals with both low–normal as well as low BMD. It has been shown that the rate of osteopaenia is steadily increasing in both men and women [1]. Among women aged 45–50 years, the prevalence of osteopaenia is 24%, peaking at 66% at age 64–69 years. It then declines slightly to 60% at age 80–84 years as the prevalence of osteoporosis increases. In men, the prevalence of osteopaenia is 28% and 63–64% at ages 45–49 and 64–79 years, respectively. Thus, in the majority of both men and women over 60, BMD results are consistent with osteopaenia [1]. In another study, it was estimated that 15% of peri-menopausal women had been within the BMD range of less than –1.0 SD and that most patients had had osteoporotic fractures [2]. Based on that, it was suggested that a T-score of less than –1.0 SD should be considered as a threshold for treatment in people at increased risk of fractures, in whom it would be most useful to prevent bone mass loss [2,3]. Therefore, in addition to BMD, other risk factors should be analysed, as highlighted in new global and Polish recommendations [4–6].
According to current recommendations for the diagnosis and treatment of osteoporosis, the fracture risk assessment criteria in Poland in women and men over 50 years of age should be divided into 3 groups: very high risk, high risk, and medium/low risk of fractures [6]. It was also emphasised that the diagnosis of osteoporosis based on dual-energy X-ray absorptiometry (DXA) alone is insufficient [6]
The categorisation of a patient into a specific risk group is based on the determination of fracture risk factors, the assessment of the Fracture Risk Assessment Tool (FRAX) calculator [7], and the BMD assessment [8]. Based on the categorisation of fracture risk, further diagnostic and therapeutic decisions should be made. The 3 cut-off points may vary among populations in different countries [4, 5, 9]. In addition to assessing the 10-year risk of osteoporotic fracture at the major location (FRAX MOF), the European recommendations also indicate a cut-off point for the 10-year risk of hip fracture (FRAX HF). Either one or both thresholds can be used to justify pharmacological intervention. In Poland, a category of very high fracture risk has been proposed, based on at least one of the following criteria: 1. fresh low-energy fractures in the major location within < 1 year in a person with a T-score ≤ –1.0 SD; 2. multiple major fractures (≥ 2); 3. fracture during glucocorticosteroid therapy; 4. low T-score < –3.0 SD; and 5. FRAX > 15% for MOF or > 4.5% for HF [6]. For the group of patients at very high risk of fractures, the first-line medication should include an anabolic drug (teriparatide, abaloparatide or romosozumab) followed, in a sequential therapy, by an antiresorptive treatment. The assessment of fracture risk using the FRAX calculator is based on risk factors, i.e. age, female gender, history of osteoporotic fracture [10, 11], history of parental hip fracture, low BMD [12], low BMI, alcohol consumption > 3 alcohol units per day, active smoking, rheumatoid arthritis, diseases causing secondary osteoporosis, or long-term glucocorticosteroid therapy [8, 13, 14]. There is now a new option available for the FRAX Plus calculator, which takes into account such parameters as the type and number of fractures and the number of falls. However, this is a paid option and has no reference to the Polish population. The Screening of Older Women for Prevention of Fracture (SCOOP) trial showed that the proposed treatment, based on a 10-year hip fracture risk assessment, reduced the incidence of hip fractures (but not other fractures) [15].
Even though the knowledge on the pathogenesis of bone mass loss, epidemiology, and therapy has significantly increased for more than 30 years, the definitions of osteoporosis and osteopaenia have not changed. In our retrospective study, we aimed to assess the risk of fractures in patients with newly diagnosed osteopaenia and determine the percentage of patients at high and very high risk of fracture. This is an important issue because the diagnostic guidance, limited to DXA test results only, may misjudge the actual needs for further diagnostics and therapy.
Materials and methods
The retrospective study included 89 postmenopausal patients who were referred for consultation to the Osteoporosis Outpatient Clinic of the Provincial Multispecialist Oncology and Traumatology Centre in Łódź. All the patients, who met the inclusion criteria for the study, were qualified at their first visit to the Osteoporosis Clinic between August 2023 and July 2024. The inclusion criteria were: 1) femoral neck and total hip DXA and/or L1-L4 DXA between T score < –1.0 SD and T-score ≤ –2.5 SD; 2) female patients at least 2 years post-menopause with osteopaenia diagnosed no later than 6 months prior; and 3) patients who had never been treated for osteoporosis. Premenopausal patients, those with active cancer, and those previously treated for osteoporosis were not included in the study. In addition to BMD assessment, data such as age, weight, height, body mass index (BMI), history of parental hip fractures, past fractures (osteoporotic and non-osteoporotic), and comorbidities [hypertension, diabetes mellitus (DM), ischaemic heart disease (IHD), asthma/chronic obstructive pulmonary disease (COPD), or rheumatic diseases] were obtained. Information was collected on current smoking and alcohol consumption of > 3 units per day. All the patients underwent a bone mineral density test using dual-energy X-ray absorptiometry (DXA) scanning with a Hologic Horizon system, and laboratory tests were performed for the following: vitamin D3 (25[OH]D), parathormone (PTH), alkaline phosphatase (ALP), creatinine, glomerular filtration rate (GFR), serum calcium (Ca), and phosphate (P) levels. Based on the collected data, the absolute 10-year fracture risk was assessed in each patient using the Fracture Risk Assessment Tool (FRAX) calculator for the Polish population (FRAX-PL) [16] (https://frax.shef.ac.uk/FRAX/tool.aspx?lang=po). Taking into account BMD, the absolute 10-year fracture risk was calculated for the major osteoporotic fracture (FRAX MOF) and hip fracture (FRAX HF) locations.
Statistical analysis was performed using Statistica software version 13.1 (TIBCO Software 2022). Continuous variables with a distribution different from normal were described using the median along with the 25th and 75th quartiles (Q1–Q3). The distribution for continuous variables was checked using the Shapiro-Wilk W test. Nominal variables were described using counts and percentages. The non-parametric Mann-Whitney U test was used to assess differences between the 2 groups for continuous variables with a distribution different from normal. Statistical significance was assessed at p < 0.05.
Results
The characteristics of the population are shown in Table 1. The median (Q1–Q3) age of the patients included in the study was 68 years, the median FRAX MOF was 13 (Q1–Q3: 10–18), and the FRAX HF was 3.2 (Q1–Q3: 1.4–6.7). The prevalence of fracture risk factors and comorbidities is shown in Table 2. Among the risk factors, glucocorticosteroid intake (47.19%) and parental hip fracture (46.07%) accounted for the highest proportion, followed by smoking (39.33%) and alcohol abuse (2.25%). The total percentage of subjects with at least one fracture in adulthood was 56.6% in the general population. Out of those, 6 (11.54%) hip fractures, 9 (17.31%) vertebral fractures, 23 (44.23%) wrist fractures, and 3 (5.77%) and other osteoporotic fractures were observed (proximal humerus fractures, rib fractures, pelvis fractures), as well as 15 (28,85%) other non-osteoporotic fractures (fractures of the calcaneus, fibula, tibia, sternum, ankle, phalanx of hand, metacarpal bone, foot, patella, and the proximal end of the radius) (Tab. 3).
|
Variable (n = 89) |
Median (Q1–Q3) |
Min–Max |
|
Age [y] |
68.00 (62.00–74.00) |
50.00–88.00 |
|
BMI |
26.62 (24.02–29.52) |
19.49–36.57 |
|
Vitamin D [ng/mL] |
29.10 (22.20–36.80) |
7.50–95.10 |
|
PTH [pg/mL] |
59.31 (48.20–65.10) |
14.40–192.00 |
|
Creatinine [mg/mL] |
0.83 (0.69–1.00) |
0.39–1.98 |
|
GFR [mL/min/1.73 m2] |
70.40 (54.90–86.39) |
24.50–162.70 |
|
Ca in serum [mmol/L] |
2.39 (2.31–2.43) |
2.18–2.81 |
|
P in serum [mmol/L] |
1.00 (0.89–1.10) |
0.61–1.38 |
|
ALP [U/L] |
81.00 (69.00–93.00) |
48.00–192.00 |
|
DXA of the femoral neck [T-score] |
–1.70 (–2.20 (–1.40)) |
–2.40–(–0.80) |
|
DXA of the hip [total T-score] |
–1.40 (–1.70 (–0.90)) |
–2.40–0.10 |
|
DXA of the spine [L1–L4 T-score] |
–1.60 (–2.00 (–1.20)) |
–2.40–1.20 |
|
FRAX major osteoporotic fractures |
13.00 (10.00–18.00) |
3.80–47.00 |
|
FRAX hip fractures |
3.20 (1.40–6.70) |
0.40–51.00 |
|
Patients (n = 89) |
n (%) |
|
|
Risk factors |
||
|
Current smoking |
Yes |
35 (39.33%) |
|
Alcohol 3 or more units/day |
Yes |
2 (2.25%) |
|
Glucocorticosteroid therapy |
Yes |
42 (47.19%) |
|
Parental hip fracture |
Yes |
41 (46.07%) |
|
Comorbidities |
||
|
Asthma/chronic obstructive pulmonary disease |
Yes |
17 (19.10%) |
|
Rheumatic diseases |
Yes |
7 (7.87%) |
|
Diabetes mellitus |
Yes |
30 (33.71%) |
|
Arterial hypertension |
Yes |
53 (59.55%) |
|
Ischaemic heart disease |
Yes |
19 (21.35%) |
|
Variable |
n (%) |
|
|
Previous fractures* (N = 89) |
Yes |
52 (58.43%) |
|
No |
37 (41.57%) |
|
|
Hip fracture (n = 52) |
Yes |
6 (11.54%) |
|
Vertebral fracture (n = 52) |
Yes |
9 (17.31%) |
|
Distal radial fracture (wrist) (n = 52) |
Yes |
23 (44.23%) |
|
Other osteoporotic fractures (n = 52) |
Yes |
3 (5.77%) |
|
Other non-osteoporotic fractures (n = 52) |
Yes |
15 (28.85%) |
Fracture risk assessment based on FRAX score
Based on the analysis of fracture risk according to the FRAX calculator and taking into account the BMD of the femoral neck, 39.33% of the patients demonstrated a very high risk of hip fractures (HF FRAX > 4.5%), and 34.83% met the criteria for a very high risk of major osteoporotic fractures (MOF FRAX > 15%). A high fracture risk for HF and MOF was noted in 11.24% and 40.45% of the patients, respectively. In a 10-year scale, the average and low risk of MOF was 17.98% and 6.74%, respectively (Tab. 4).
|
Fracture risk |
HF (n = 89) |
MOF (n = 89) |
|
|
Very high fracture risk for MOF FRAX > 15% for HF FRAX > 4.5% |
Yes |
35 (39.33%) |
31 (34.83%) |
|
No |
54 (60.67%) |
58 (65.17%) |
|
|
High fracture risk for MOF FRAX > 10–15 for HF FRAX > 3–4% |
Yes |
10 (11.24%) |
36 (40.45%) |
|
No |
79 (88.76%) |
53 (59.55%) |
|
|
Medium fracture risk for MOF FRAX from 5% to < 10% |
Yes |
– |
16 (17.98%) |
|
No |
– |
73 (82.02%) |
|
|
Low fracture risk for MOF FRAX < 5% |
Yes |
– |
6 (6.74%) |
|
No |
– |
83 (93.26%) |
|
Fracture risk analysis according to extended criteria
The analysis of fracture risk according to the expanded criteria identified those with very high (53.93%) and high (26.97%) fracture risk and showed a poor proportion of those with medium (16.85%) and low risk (2.25%) (Tab. 5).
|
Fracture risk (n = 89) |
n (%) |
|
|
Very high fracture risk |
Yes |
48 (53.93%) |
|
No |
41 (46.07%) |
|
|
High fracture risk |
Yes |
24 (26.97%) |
|
No |
65 (73.03%) |
|
|
Medium fracture risk |
Yes |
15 (16.85%) |
|
No |
74 (83.15%) |
|
|
Low fracture risk |
Yes |
2 (2.25%) |
|
No |
87 (97.75%) |
|
Comparing the 2 analyses, it was noted that significantly more subjects (53.93%) were categorised to very high risk of fracture, taking into account the extended criteria rather than on the basis of the FRAX values alone (39.33% for HF and 34.83% for MOF).
Assessment of fulfilment of the criteria by the patients regarding the classification into the very high fracture-risk group and a comparison with FRAX values
Adopting the expanded criteria for fracture risk assessment, it was shown that 53.93% of the patients participating in the study had met at least 1 out of 3 criteria for very high fracture risk. Among those patients, the largest percentage (48.31%) were women meeting the criterion of FRAX > 15% for MOF or FRAX > 4.5% for HF, while 7.87% revealed a history of multiple (≥ 2) major fractures. Because the study included patients with osteopaenia, it is clear that none of them met the criterion of T-score ≤ –3.0 SD (Tab. 6).
|
|
n (%) |
FRAX |
||
|
HF |
MOF |
|||
|
Median (Q1–Q3) |
Median (Q1–Q3) |
|||
|
All patients |
89 (100.00%) |
1.40 (6.70–0.00) |
10.00 (18.00–0.00) |
|
|
Very high fracture risk Fulfilment of 1 out of 3 criteria: |
48 (53.93%) |
6.90 (4.90–12.00) |
17.00 (14.50–24.50) |
|
|
History of multiple fractures ≥ 2 |
Yes |
7 (7.87%) |
2.00 (0.90–3.20) |
11.00 (8.80–18.00) |
|
No |
82 (92.13%) |
3.50 (1.50–7.40) |
13.00 (10.00–19.00) |
|
|
Low T-Score ≤ –3.0 |
Yes |
0 (0.00%) |
– |
– |
|
No |
89 (100.00%) |
3.20 (1.40–6.70) |
13.00 (10.00–18.00) |
|
|
FRAX > 15% for MOF or FRAX > 4.5% for HF |
Yes |
43 (48.31%) |
6.90 (4.90–12.00) |
19.00 (15.00–28.00) |
|
No |
46 (51.69%) |
1.45 (1.10–2.80) |
10.50 (7.40–12.00) |
|
The age structure of the patients with very high and high fracture risk for HF and MOF
Considering the BMD-adjusted FRAX results, the highest proportion of the patients at very high risk of MOF (hip, vertebrae, wrist, humerus, pelvis) was in the 70–75 age range, and for the patients at high risk in the 65–70 age range (Fig. 1A). It was observed that patients at very high risk of MOF had had a statistically significant higher age (Me: 73.00; Q1–Q3: 67.00–76.00) than the patients at high risk (Me: 66.00 Q1–Q3: 61.50–70.50; p = 0.006) (Fig. 1B, Supplementary File — Tab. S1A). The outcome was similar for the patients at very high risk of HF, with the highest percentage in the 70–75 age range, while the high risk of HF was the greatest in the 65–70 age range (Fig. 1C). Here, statistically significant differences were also found between the 2 groups (p = 0.03). It was observed that patients at very high risk of MOF had had a statistically significant higher age (Me: Q1–Q3: 70.00–77.00) than the patients at high risk (Me: 68.50; Q1–Q3: 64.00–70.00), but it should be emphasised that the group at high risk of fractures was less numerous (Fig. 1D, Supplementary File — Tab. S1B). The age analysis of patients at very high and high risk of fracture according to the extended assessment criteria is analogous to the above. The highest number of subjects at very high fracture risk was for the age group 70–75 years (Fig. 1E), while in case of the high fracture risk, it was the age group 65-70 years. In addition, it was shown that patients at very high risk had a statistically significant higher age (Me: 72.50 Q1–Q3: 67.00–76.00) when compared with the high-risk patients (Me: 66.00; 61.50–70.50; p = 0.006) (Fig. 1F, Supplementary File — Tab. S1C).
Discussion
The overarching goal of osteoporosis treatment is to reduce the risk of future fractures.
The benefit-risk ratio and the cost-effectiveness of treatment are greatest in patients at high risk of fractures. Although patients with osteoporosis are at higher risk of fractures than those with osteopaenia, the majority of fractures nevertheless occur in the osteopaenic group. In our study, the total number of patients with osteopaenia and with at least one fracture in adulthood was 52 in the entire population, which is as high as 56.6%. A large cohort study analysing fracture risk in
a population of postmenopausal women in South Korea found that more than half of fractures (58.4%) over a 10-year period occurred in women with moderately reduced BMD, including those with osteopaenia, thus justifying targeted efforts to identify early cases and implement prevention strategies for women who are in the osteopaenic range but, at the same time, are at high risk of fractures [17].
The NORA study has shown that the incidence of fractures rises sharply when BMD falls from its normal values down into their osteoporotic range. However, as there were more women with osteopaenia than with osteoporosis, the fracture rate was higher in the group with osteopaenia (52%) than in the smaller group with osteoporosis (18%). For hip fractures, 26% and 51% occurred in women with osteoporosis and osteopaenia, respectively [18]. Similar findings were provided by the Rotterdam study, in which 58% of nonvertebral fractures and 54% of hip fractures occurred in women with osteopaenia, and 63% of nonvertebral fractures and 64% of hip fractures occurred in men with osteopaenia [1]. In contrast, an Australian study confirmed that only 11–20% of fractures were associated with BMD-confirmed osteoporosis in women over 60 years of age [19]. These data suggest that therapeutic interventions, which target only patients with densitometric features of osteoporosis, will have little impact on fracture reduction because the majority of those who develop a fracture will be osteopaenic patients who will remain untreated. Given this observation, an expert group recently concluded that using BMD scores alone in determining thresholds for therapeutic interventions may be problematic [20]. Therefore, in our study we undertook an assessment of fracture risk factors based on FRAX, which helped identify patients at low, medium, high, and very high risk of fractures. We showed that 39.33% of the patients had a very high 10-year risk of hip fractures (HF FRAX > 4.5%) and 34.83% met the criteria for a very high fracture risk for major osteoporotic fractures (MOF FRAX > 15%). In turn, a high fracture risk for HF and MOF was reported in 11.24% and 40.45% of the patients, respectively. Such a high proportion of patients with high and very high fracture risk was probably influenced by the fact that previous fracture (56.6%), glucocorticosteroid intake (47.19%), parental hip fracture (46.07%), and smoking (39.33%) predominated among the risk factors. Different results were obtained in the study by Tomasevic-Todorovic et al. [21], in which the largest proportion of patients with osteopaenia demonstrated a low (63.89%) and medium (36.11%) risk for MOF and a low risk for HF (91.67%). Comparing the individual risk factors, this study showed that the patients with osteopaenia demonstrated a significantly higher risk of fracture when compared to the patients with osteoporosis, due to their lower body weight (t = –2.250, p = 0.05), while the patients with osteoporosis revealed a higher risk of fracture compared to the patients with osteopaenia, due to their higher rates of previous fractures (t = 2.985, p = 0.01). The risk of fracture in women with osteopaenia, without additional risk factors, ranges from very low in young postmenopausal women (T-score –1.2 SD and no risk factors) to very high in older women with lower BMD and additional risk factors. The fracture risk also depends on other important risk factors independent of BMD and age, including fracture history, gender, ethnicity, history of falls, parental hip fracture history, and glucocorticosteroid intake, and it is consequently significantly higher than that based on BMD [3]. Considering the expanded criteria for fracture risk assessment [6], our study additionally showed that significantly more subjects (53.93%) had been classified as being at very high risk of fracture based on the expanded criteria than on the basis of FRAX alone (39.33% for HF and 34.83% for MOF). Among those patients, the greatest percentage (48.31%) were women meeting the criterion of FRAX > 15% for MOF or FRAX > 4.5% for HF, while 7.87% revealed a history of multiple (≥ 2) major fractures. Because the study included patients with osteopaenia, it is clear that none of them met the criterion of T-score ≤ –3.0 SD. These data demonstrate the importance of making therapeutic decisions based on the whole clinical picture of the patient, rather than solely on T-scores.
Thus, once fracture risk categories have been established in patients with osteopaenia, it is important to consider which therapeutic interventions will be most effective in this group. It seems that the fracture risk threshold for intervention should be similar for both osteopaenic and osteoporotic patients. Various expert groups have provided different recommendations on the indications for antifracture treatment [22, 23]. The variety of recommendations reflects the prevailing uncertainty in this area. The majority of groups accept a T-score of –2.5 SD or less, a fresh fracture, or an increased risk of fracture (however defined) as reasons to offer treatment, with the Bone Health and Osteoporosis Foundation also accepting a 10-year hip fracture risk of ≥ 3% or an MOF risk of ≥ 20% according to US FRAX [24]. In contrast, the data obtained in studies from Manitoba and New Zealand indicate that the MOF risk of 10–15% may be an acceptable indication for treatment with generic bisphosphonates in women over 65 years of age [25, 26]. The following principles are suggested to guide the selection of individual therapeutic thresholds: the proposed treatment regimen should prevent fractures in a randomised controlled trial (or prevention is highly likely based on other trials with similar agents); the treatment is cost-effective in this population (which depends on both the price of the drug and the risk of fractures), and the therapy has an acceptable safety profile [3]. The agents registered for the treatment of osteoporosis mostly meet the first of the above requirements for proven efficacy but have generally been tested in patients with a T-score ≤ –2.5 SD or with a history of fractures. However, there are several studies that indicate that oral bisphosphonates [27, 28], zoledronate [26, 29, 30], oestrogens [31, 32], and possibly denosumab [33, 34] also meet the efficacy criteria for the treatment of osteopaenia and should be considered for fracture prevention in osteopaenic patient populations, where the other 2 criteria are met. There are also claims that while in patients at moderate fracture risk the use of denosumab and anabolic drugs is ruled out for cost reasons, for patients at high fracture risk they should be considered, and for patients with osteopaenia and very high fracture risk an anabolic drug could be a first-line treatment [22].
Our study has several limitations. This study was a retrospective, cross-sectional study; therefore, it was not possible to determine causality. In addition, our sample size was relatively small, and although we obtained statistically significant results, a larger sample size could further confirm the results obtained. In addition, because there was a high proportion of patients taking glucocorticosteroids in the study group, the information on fractures while taking glucocorticosteroids could additionally be included in the eligibility criteria table for the patients at very high fracture risk, as well as the data on fresh low-energy fractures in the major location in the last year, which we omitted due to incomplete data in the medical history of the patients.
In conclusion, our results highlight the fact that a large proportion of patients with osteopaenia are at high and very high risk of fracture, assessed by both FRAX and the extended criteria. Our observations may provide a firm basis to consider which therapeutic options could be adopted in patients with osteopaenia and highlight the importance of categorising fracture risk based on the whole clinical picture and not merely on T-scores. Fracture risk should be assessed via a combination of BMD and clinical risk factors, while the treatment should be considered in patients at an increased risk of fracture.
Data availability statement
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Ethics statement
This retrospective study does not require the consent of the bioethics committee.
Author contributions
K.C. conceived and designed the analysis, collected the data, and wrote the manuscript, M.M-K. contributed data and helped supervise the study, K.P. performed the statistical analysis, M.S. contributed to the interpretation of the results, and E.S. supervised the findings of this work and revised the manuscript critically for important intellectual content. All authors read and approved the final manuscript.
Funding
No funding was received to assist with the preparation of this work.
Acknowledgments
N/A
Conflict of interest
The authors report there are no competing interests to declare.
Supplementary material
Supplementary material is available in the additional attachment
