Osteoporosis, a systemic skeletal disorder marked by osteopenia and degradation of bone microarchitecture, has emerged as a major global public health concern (1–3). The International Osteoporosis Foundation (IOF) reports that osteoporotic fractures occur every 3 seconds worldwide, with vertebral fractures representing approximately 40% of these fractures. Individuals who have experienced vertebral fractures face a nearly fivefold increased risk of subsequent fractures compared to those with non-vertebral fractures (4). Epidemiological data from China show that the prevalence of vertebral fractures in individuals over 50 years of age is approximately 15%, but only 20%-30% of those with vertebral fractures exhibit typical clinical symptoms, leading to a misdiagnosis rate as high as 60% (5, 6). Epidemiological studies in Europe show that the prevalence of vertebral fractures in people over 50 years of age ranges from 12% to 18%, increasing significantly with age (7, 8). In the United States, data from the National Health and Nutrition Examination Survey (NHANES) indicates that the prevalence of vertebral fractures is approximately 25% in women aged 50 and older, and approximately 10% in men, highlighting the widespread disease burden (9, 10). Although the FRAX tool is commonly used in clinical settings to assess osteoporotic fracture risk, its sensitivity for predicting vertebral fractures is limited, particularly in patients with early-stage osteoporosis who lack additional risk factors (11). This underscores the need for the identification of new biomarkers to enhance the current fracture risk assessment framework.

While the FRAX tool is widely used globally, other validated fracture risk assessment tools, such as QFracture and the Garvan calculator, each have their own characteristics. QFracture, developed in the UK, has the advantage of incorporating a broader range of clinical risk factors and assessing risk without relying on bone mineral density data (12). However, its model is primarily based on UK population data, and its applicability and calibration in other countries may require further validation. The Australian Garvan calculator specifically emphasizes and quantifies the significant impact of fall history and previous fracture counts on fracture risk, providing 5-year and 10-year fracture risk predictions (13). However, it covers relatively fewer clinical risk factors, and its ability to differentiate in older populations may be limited. These tools share several challenges, often struggling to capture and integrate dynamic biological information reflecting a patient’s real-time pathophysiological state. Therefore, exploring novel biomarkers holds promise for providing valuable and readily available supplementary information to existing, relatively static clinical risk assessment systems, potentially enabling more sensitive identification of patients at extremely high fracture risk.

Although osteoporosis can lead to various fracture types, this study focuses specifically on vertebral fractures, primarily due to their unique clinical importance: vertebral fractures are the most common type of osteoporotic fracture, accounting for approximately 40%, and because they often lack typical clinical symptoms in the early stages, the rate of underdiagnosis and misdiagnosis is as high as 60%, thus delaying treatment (14); more importantly, patients with vertebral fractures have a nearly 5-fold increased risk of subsequent fractures, significantly increasing disability and mortality rates, placing a heavy burden on patients’ quality of life and the healthcare system (15, 16). Therefore, focusing on vertebral fractures helps to reveal the predictive value of inflammatory biomarkers in this high-risk but easily overlooked outcome, providing targeted evidence for early intervention.

In recent years, the role of chronic low-grade inflammation in the pathological process of osteoporosis has gained growing interest. As a marker of systemic inflammation, the neutrophil to lymphocyte ratio (NLR) has been shown to be closely linked to an imbalance in bone metabolism (17–19). Research shows that neutrophils activate the RANKL/osteoporosis G signaling axis by secreting IL-17, thereby enhancing osteoclast differentiation (20). Meanwhile, matrix metalloproteinase 9 (MMP-9) released by neutrophils can directly degrade type I collagen, leading to a reduction in trabecular connectivity (21). Lymphocyte subsets, such as regulatory T cells, counteract osteoclast activity by secreting IL-4 and IL-10, thus preserving bone remodeling homeostasis (22). Clinical cohort studies have revealed that peripheral blood NLR in osteoporosis patients is significantly elevated compared to healthy controls, and negatively correlates with femoral neck bone density (23–27). Collectively, these findings suggest that NLR acts as a promising biomarker for predicting vertebral fractures by reflecting the inflammatory status of the bone microenvironment.

The clinical value of the NLR in fracture risk prediction has been established across various disciplines. For instance, in cardiovascular diseases, elevated NLR has been strongly linked to an elevated risk of major adverse cardiovascular events (28). However, its predictive accuracy for osteoporosis-related vertebral fractures remains contentious: a prospective cohort study involving 92 patients with post-menopausal osteoporosis found no significant correlation between NLR and the risk of vertebral fractures (29), while another prospective cohort study (n=80) identified NLR as an independent risk factor for osteoporotic vertebral fractures (OR = 13.229, 95%CI 4.167-41.998) (30). This heterogeneity may be due to differences in study design, including the timing of NLR detection (acute inflammatory phase vs stable phase), fracture diagnostic criteria (radiological diagnosis vs clinical symptoms), and control of confounding factors (such as history of glucocorticoid use). Therefore, evidence-based research is needed to integrate global observational studies, not only to quantitatively assess the overall association strength between NLR and the risk of vertebral fractures in osteoporotic patients, but also to explore the sources of discrepancies in existing research findings through heterogeneity analysis. Clearly identifying and assessing these heterogeneities is of great value in promoting the clinical translation and application of NLR as a predictive biomarker for vertebral fractures.

This latest study aims to integrate global observational research evidence and quantitatively evaluate the relation between NLR and the risk of vertebral fractures in osteoporotic patients. The results will provide important evidence for improving the risk stratification of osteoporotic fractures and formulating individualized intervention strategies, while opening up new perspectives for the translational application of inflammatory mechanisms in bone metabolic diseases.

This study found through a meta-analysis of 6 studies that osteoporotic individuals with high NLR had a notably higher risk of vertebral fractures than those with low NLR, and sensitivity analysis confirmed the stability of this result. However, the continuous variable obtained a negative result, indicating that patients with vertebral fractures and non-vertebral fractures had similar NLR levels. However, it is important to note that the sensitivity analysis of continuous variables revealed that this result was not stable. When the data of Li et al., 2023 were excluded, the results shifted from insignificant to significant, and the heterogeneity decreased significantly. The results also indicated that the NLR levels in individuals with vertebral fractures were significantly higher compared to those with non-vertebral fractures, supporting the results of categorical variables. Therefore, based on the existing evidence, it can be inferred that NLR is indeed an effective inflammatory marker for vertebral fractures in osteoporotic patients, a finding that is consistent with most previous studies (30, 33–35).

The inconsistency between the categorical and continuous variable analyses in this study may stem from a combination of methodological heterogeneity and sample limitations. The NLR cutoff values ​​used in the included studies varied considerably, and the methods used to determine them differed. This heterogeneity in the definition of the cutoff value directly led to high uncertainty in the pooled results of the categorical analysis. While the continuous variable analysis avoided cutoff value selection bias, its statistical power was low due to the limited number of included studies and insufficient total sample size, making it difficult to detect the true effect. The fact that the continuous results became significant after excluding the study by Li et al. (2023) in the sensitivity analysis further indicates that the current evidence system is abnormally sensitive to individual study data, resulting in insufficient robustness of the conclusions. Future research needs to develop unified NLR detection standards, standardize cutoff value determination methods, and expand multi-center sample sizes to effectively improve the comparability and explanatory power of the results.

NLR refers to the ratio between the neutrophil count and lymphocyte count in peripheral blood routine. When there is acute inflammation in the body, the number of neutrophils in peripheral blood increases abnormally, and once the body’s immune defense ability decreases, the lymphocyte count in peripheral blood routine decreases abnormally. Therefore, NLR serves as a direct indicator of the body’s acute inflammatory response and immune defense status (36–38). Recent advances have considerably enhanced our understanding of the prognostic relevance of the NLR. Numerous studies have demonstrated that an abnormally elevated NLR is strongly associated with disease severity, prognosis, and recovery outcomes. Studies have found that peripheral blood NLR>5 is positively correlated with postoperative mortality 5 days after hip fracture surgery (39). Some studies have found that postoperative NLR ≥ 4.7 increases the risk of death in individuals with hip fractures (40). Research has indicated a strong correlation between NLR and inflammatory diseases such as ulcerative colitis, acute appendicitis, cardiovascular disease, colorectal cancer and lung cancer (41–45). Concurrently, some studies also have indicated that NLR surpasses white blood cell and neutrophil counts in predicting adverse outcomes in individuals with cardiovascular diseases and malignant tumors (41, 42, 44).

In addition, inflammation is the most important stress response of the body in the development of osteoporosis. Regardless of the type of osteoporosis, the common feature is that it is accompanied by systemic or focal inflammatory response. The inflammatory response in the body will enhance the activity of osteoclasts, leading to increased bone resorption, breaking the original bone homeostasis in the body, and thus accelerating the development of osteoporosis (46, 47). Barbour et al. (48) found that abnormally elevated inflammatory markers increased the risk of hip fracture in elderly women. These studies collectively highlight the significant role of the body’s inflammatory response in the pathogenesis of osteoporosis. Furthermore, a large body of clinical studies have confirmed that the peripheral blood NLR ratio plays a crucial role in fracture risk. In a study involving 438 individuals, the bone density of the lumbar spine and femoral neck decreased with the increase of NLR (49). An increased neutrophil-to-lymphocyte ratio (NLR) after a hip fracture has been recognized as an indicator of postoperative mortality and cardiovascular events (39). Atlas et al. (50) observed that NLR levels were elevated in patients admitted to the intensive care unit (P = 0.007) as well as in those who succumbed (P = 0.007) within 5 days following hip fracture surgery. The significant associations observed in this meta-analysis support the role of the aforementioned inflammatory mechanisms in vertebral fractures.

When assessing the specificity of NLR as a predictive biomarker for vertebral fractures, the potential confounding effects of common complications in elderly patients with osteoporosis, such as cardiovascular disease, diabetes, and chronic kidney disease, on NLR levels must be fully considered. These complications are often accompanied by a chronic low-grade inflammatory state, which can independently cause neutropenia or lymphopenia, leading to non-specific increases in NLR values ​​and thus obscuring the true causal relationship between NLR and vertebral fracture risk. For example, patients with cardiovascular disease have higher underlying inflammation levels, and their elevated NLR may reflect vascular inflammation processes rather than simple bone metabolic abnormalities. Therefore, future research needs to use rigorous multivariate adjusted models or stratified analyses targeting specific comorbidity populations to control for these confounding factors in order to more accurately assess the specific predictive ability of NLR for vertebral fractures and avoid overestimating its predictive value.

This study has several limitations that require careful consideration. First, the sample size is relatively small and the geographical distribution of the studies is highly concentrated, with most studies involving Asian populations, which limits the generalizability of the results to a wider population. Second, there are significant differences in the selection of cutoff values ​​for NLR among the studies, and the lack of a unified standard may introduce classification bias and increase heterogeneity. Furthermore, the diagnostic criteria for vertebral fractures are inconsistent across different studies, relying on imaging examinations in some cases and combining clinical symptoms in others; this inconsistency in judgment criteria may affect the pooling and comparison of results. Statistically, most original studies did not adequately adjust for potential confounding factors, such as patient comorbidities, medication history, and lifestyle, which may interfere with the observed associations. Finally, the limited number of included studies, and the tendency for positive results to be more easily published, may introduce publication bias and potentially affect the estimation of the pooled effect size. Despite these limitations, this meta-analysis provides preliminary evidence consolidation of the association between NLR and the risk of vertebral fractures, and further validation with more high-quality, large-sample prospective studies is needed.

This systematic review and meta-analysis is the first to comprehensively assess the association between NLR and the risk of vertebral fracture in patients with osteoporosis. The results showed that, based on categorical variable data, a higher NLR was significantly associated with an increased risk of vertebral fracture; however, continuous variable data did not show a significant difference in NLR levels between the fracture and non-fracture groups. This inconsistency highlights the significant heterogeneity among studies and its underlying limitations, including a limited number of included studies, insufficient overall sample size, geographical limitations due to the majority of studies focusing on Asian populations, inconsistent NLR cutoff values ​​used across studies, differences in diagnostic criteria for vertebral fracture, and insufficient control over potential confounding factors. These factors suggest that the strength of the current evidence should be viewed with caution. Future research should focus on large-scale, multicenter prospective cohort studies, prioritizing the standardization of NLR measurement procedures, combining NLR with commonly used clinical indicators to assess its incremental predictive value, including broader populations to validate the generalizability of the results, and conducting long-term follow-up to clarify the long-term prognostic significance of NLR. In conclusion, the clinical application value of NLR as a potential biomarker for vertebral fractures has begun to emerge, but its exact clinical efficacy still needs to be further verified by higher-quality studies.