Systemic immune‐inflammation index acts as a novel diagnostic biomarker for postmenopausal osteoporosis and could predict the risk of osteoporotic fracture

Abstract Background Postmenopausal osteoporosis (PMOP) is a bone metabolism disorder involving systematic inflammation activation. Blood routine examination is easily available in clinical practice and contains abundant information reflecting the systematic inflammation level. Thus, it is attractive to achieve early diagnosis of PMOP and predict osteoporotic fracture risk just based on the biomarkers in blood routine examination. Methods A multi‐centric prospective cohort study was designed and enrolled postmenopausal women from two independent institutions. All participants underwent the dual‐energy X‐ray absorptiometry (DEXA) scanning for diagnosing PMOP. Blood routine examination was conducted, and the key inflammatory biomarkers such as neutrophil‐to‐lymphocyte ratio (NLR) and systemic immune‐inflammation index (SII) were calculated. PMOP patients were followed up to observe osteoporotic fracture and identify the related risk predictors. Results A total of 92 participants out of 238 enrolled postmenopausal women were diagnosed with PMOP, with a prevalence of 38.66%. The main risk factors identified for PMOP included older age (OR = 2.06, 95% CI = 1.14‐3.72), longer menopause duration (OR = 3.14, 95% CI = 2.06‐4.79), higher NLR (OR = 2.11, 95% CI = 1.37‐3.25), and higher SII (OR = 3.02, 95% CI = 1.98‐4.61). Besides age and menopause duration, SII ≥834.89 was newly identified as a prominent risk factor for discriminating osteoporotic fracture risk in PMOP patients (HR = 3.66, 95% CI = 1.249‐10.71). Conclusion As an easy and economical biomarker calculated from blood routine examination, SII not only acts as a good risk predictor for PMOP diagnosis but also well discriminates the osteoporotic fracture risk, which deserves further investigation and application in clinical practice.


| INTRODUC TI ON
Postmenopausal osteoporosis (PMOP) is a chronic systematic disorder of bone metabolism, which is characterized by bone loss, microstructure deterioration, and prone to fragility fracture. 1,2 Osteoporotic fractures, also known as brittle fractures, are different from fractures that result from violent collisions or unexpected blows; it refers to fractures that occur without trauma or minor trauma. 3 With the aging of the population rapidly increasing, PMOP is becoming prevalent in postmenopausal women in recent years, causing a serious social health problem and heavy economical burdens. 4,5 To date, early detection of PMOP and intervention with protective measures have been the most effective healthcare strategies in PMOP management. Traditional diagnostic approach for PMOP is largely based on the dual-energy X-ray absorptiometry (DEXA) and assessed by bone mineral density (BMD). 2 However, a great number of postmenopausal women are unaware of PMOP and tend not to receive DEXA scanning until some adverse incidents owing to osteoporosis occur, such as bone pain or even bone fracture. Therefore, it is urgent to identify easy and efficient biomarkers to early recognize PMOP among postmenopausal women. 6 It has been well established that PMOP pathogenesis is closely related to body immune dysfunction and systematic inflammation activation. 7,8 Because women would lose the protection of endogenous estrogen after menopause, a mass of inflammatory cytokines increasingly accumulates, such as tumor necrosis factor-alpha, interleukin (IL)-6, IL-12, and IL-17. These inflammatory cytokines could mediate oxidative stress injury, provoke osteoclast, and enhance bone absorbability, thus gradually leading to skeletal remodeling and PMOP. 9 Therefore, it is reasonable to resort to systematic inflammatory biomarkers to early recognize PMOP. For instance, some emerging studies suggested blood neutrophil-to-lymphocyte ratio (NLR). As a simple peripheral blood index which could reflect the systemic inflammatory level, NLR can well discriminate PMOP among postmenopausal women, even being superior to C reaction protein. 10,11 However, to date, scarce study further explored if there are more optimal biomarkers other than NLR in blood routine examination for diagnosing PMOP, and if there are any blood biomarkers could predict fracture risk among PMOP patients.
Given that blood routine examination is easily available, economical and contains abundant useful parameters, it should not be underutilized in PMOP diagnosis and management, which deserves to be further explored. Thus in this study, we established a multicentric cohort consisting of postmenopausal women provided with high-quality data. We mainly aimed to (a) identify more optimal and novel blood biomarkers besides NLR for diagnosing PMOP among postmenopausal women and (b) first explore biomarkers based on blood routine examination for predicting osteoporotic fracture among Chinese PMOP patients.

| Study participants
This study was conducted with a prospective cohort design and en- Hospital of Traditional Chinese Medicine, respectively. There were no significant differences regarding age, menopause duration, body mass index (BMI), and BMD between the participants from the two institutions. This study was approved by the Ethics Committees of the institutions. All participants were fully informed of this study and given the written consent for participation.

| Clinical examinations and follow-up
All participants enrolled in this study underwent the DEXA scanning (HOLOGIC DISCOVERY A). The BMD values of the lumbar spine 2-4 and neck of femur were evaluated. BMD values were presented as mineral amount (g) per scanned area (cm 2 ) and then transformed into T-scores based on corresponding coefficients. According to the PMOP diagnosis criteria defined by the World Health Organization, 12 the participants with a T-score ≤−2.5 were divided into the PMOP patients, while the participants with a T-score ≥−1 were divided into the normal group, and the others with −2.5 ≤T-score ≤−1 were divided into the osteopenia group.
In order to obtain a comprehensive blood routine examination, venous blood samples about 6 mL were collected from all participants after overnight fasting. Then, the blood samples were soon all recorded. NLR, platelet-to-lymphocyte ratio (PLR), and lymphocyte-to-monocyte ratio (LMR) were calculated. Systemic immuneinflammation index (SII) was defined as platelet counts × neutrophil counts/lymphocyte counts. 13 All parameters were then transformed into categorical variables based on the mean or median value.
All participants' baseline and demographic data such as age, menopause duration, and BMI were collected at the time of the enrollment. PMOP patients were subsequently followed up every 4 months by telephone or outpatient visit. Osteoporotic fracture was defined as the fracture caused by the decrease in bone density and bone quality after suffering from osteoporosis, which is a pathological fracture and the most serious consequence of osteoporosis.
The status of osteoporotic fracture and the corresponding time was inquired and recorded. The end of the follow-up was January 2019, and all the enrolled PMOP patients were followed up at least for 2 years.

| Statistical analysis
All statistical analyses and graphics were conducted in SPSS 22.0 and GraphPad Prism 7.0. Continuous data were expressed as mean ± standard deviation (SD) and examined by Student's t test.
Categorical data were expressed as absolute number with percentage and examined by the chi-square test. Univariate logistic analysis was employed to preliminarily screen the potential risk factors for PMOP. Factors with a P value less than .05 in the univariate analysis were sent into a forward stepwise multivariate logistic analysis to identify independent risk factors for PMOP. Odds ratio (OR) or hazard ratio (HR) with 95% confidence interval (CI) was used for measuring the strength of the association. Kaplan-Meier curve was depicted to explore the association of blood markers with bone fracture. The Cox proportional hazard regression analysis was used to identify the independent risk factors for fracture. All P values were two-sided, and P < .05 was considered statistically significant.

| Participants' baseline characteristics
The baseline characteristics of the study participants were presented in Table 1. Among the 238 postmenopausal women, 92 patients were diagnosed as PMOP, with the PMOP prevalence of 38.66%. The average age of the PMOP patients (67.2 ± 7.2 years) was significantly older than that of the osteopenia participants (57.3 ± 8.6 years) and the normal participants (54.9 ± 7.9 years) (P < .05). The menopause duration significantly increased from the normal group (6.7 ± 4.3 years) to the osteopenia group (11.8 ± 5.1 years) and then to the PMOP group (18.4 ± 6.9 years) (P < .05). The BMI of the normal group (24.9 ± 1.1 kg/m 2 ) was significantly higher than that of the PMOP group (22.0 ± 1.3 kg/m 2 ) (P < .05). Both BMD and T-score

| Risk factors for PMOP patients
Univariate analyses were conducted to identify the parameters associated with PMOP diagnosis, and the results were presented in As presented in Table 3, in the subsequent multivariate analysis, age older than 60 years was identified as an independent risk factor for PMOP diagnosis (adjusted OR = 2.06, 95% CI = 1.14-3.72).
Menopause duration over than 12 years conferred postmenopausal women a high risk for PMOP (adjusted OR = 3.14, 95%

| Risk factors for osteoporotic fracture in PMOP patients
The median follow-up time was 38. 8  TA B L E 3 Multivariate logistical regression analysis of risk factors for PMOP among postmenopausal women PMOP patients with age ≥60 years significantly tend to have osteoporotic fracture than those with age <60 years (P < .05; Figure 1A).
PMOP patients with duration of menopause ≥12 years were also more likely to occur fracture (P < .05) ( Figure 1B). High BMI ≥ 23 kg/ m 2 can play a protective effect on PMOP patients against fracture risk ( Figure 1C). Although lower albumin level and higher NLR level seemed to increase the risk of fracture, the differences were not adequately achieve statistical significance (both P > .05; Figure 1D,E).
SII displayed an excellent ability to discriminate high fracture risk patients or low fracture risk patients (P < .05; Figure 1H). The other blood routine biomarkers, such as PDW, MCH, MCHC, and RDW, were shown to have no significant differences regarding fracture risks (all P > .05; Figure 1I-L, respectively).
In the subsequent hazard analysis, the HR of age ≥60 years for  Over the past decade, increasing studies have reported that many blood routine examination-derived biomarkers, such as NLR, PLR, and LMR, whose levels could be closely related to systemic inflammation and immune response status. 18 These biomarkers are proved to be well associated with various infectious diseases, oncological diseases and autoimmune diseases. [19][20][21] In recent years, following NLR, PLR, and LMR, SII has been discovered as an emerging indicator for these diseases and shed great predictive and diagnostic potential. 22 However, to date, whether SII could also help  Although the novel findings as mentioned above, there were inevitably some limitations in our study. First, although we enrolled a great number of postmenopausal women as the participants, the number of our target population, namely PMOP patients, was still not abundant. The small sample size of our target population may restrict some further deep and meaningful subgroup analysis. Second, the follow-up time in our study seemed to be not adequately long to observe the outcome event of osteoporotic fracture incidence. Only a few PMOP patients occurred fracture until the endpoint, and this might lead to some fluctuation of the estimated HR values. Therefore, in the future, more PMOP patients should be enrolled and long followed in order to yield more meaningful and insightful findings. Third, PMOP is a systemic disease involving in multiple body disorders. Currently, we have adequately explored the inflammatory signs. However, there must be some other aberrant blood routine signs reflecting the disease.

| D ISCUSS I ON
Therefore, in the future, it is attractive and valuable to investigate more markers such as the blood lipid markers and blood coagulation markers in PMOP.
In summary, the present study newly identified that higher SII acts as a significant risk predictor for PMOP diagnosis among postmenopausal women. More than that, SII could also well discriminate the osteoporotic fracture risk in PMOP patients. Because SII is an easy and economical blood routine examination-derived biomarker, in the future clinical practice, it may play an important role in PMOP screening and prevention.