Association of dietary inflammatory index with serum asprosin and omentin levels in women with prediabetes

Introduction

Diabetes mellitus (DM) refers to a chronic condition resulting from insufficient synthesis of insulin hormone in the body or inadequate utilization by target tissues (International Diabetes Federation, 2021). Type 2 diabetes (T2DM), which is usually triggered by environmental factors such as nutrition, occurs in over 90% of individuals with diabetes (World Health Organization, 2019). As reported in the 2021 Atlas of the International Diabetes Federation, over the past 21 years, the number of adults with diabetes globally has increased approximately 3.5 times, reaching 537 million. This number is projected to rise by about 46% by 2045. Türkiye has been shown to have the highest population of people living with diabetes (nine million) in Europe for the year 2021 within this age group (International Diabetes Federation, 2021). This situation decreases individuals’ quality of life, increases morbidity, mortality, and the economic burden on countries (Shaw, Sicree & Zimmet, 2010).

Prediabetes refers to a condition with impaired glucose levels preceding diabetes, which increases the risk of progressing to diabetes (Echouffo-Tcheugui & Selvin, 2021). The 2024 National Diabetes Statistics Report of the United States estimates that there are approximately 97.6 million adults with prediabetes (Centers for Disease Control and Prevention, 2024). While some individuals with prediabetes may progress T2DM, others may return to normal glucose levels due to the reversible nature of prediabetes (Tabák et al., 2012). In the treatment of prediabetes, pharmacological treatment and herbal approaches can be applied in addition to basic lifestyle changes such as nutrition and physical activity (International Diabetes Federation, 2017; Ping et al., 2024). Obesity and increased insulin resistance in prediabetes trigger an increase in the levels of pro-inflammatory markers, leading to a state of low-grade chronic systemic inflammation (Hamdy, Porramatikul & Al-Ozairi, 2006; Haffner, 2003). This inflammation contributes to insulin resistance and beta-cell damage (Rohm et al., 2022). It has been noted that inflammatory biomarkers, such as serum IL-6, C-reactive protein (CRP), and TNF-α levels, can be influenced by diet (Shivappa et al., 2017) and may play a part in the development of chronic inflammation (Lontchi-Yimagou et al., 2013). For example, the Mediterranean diet with whole grains, fruits and vegetables, fish, nuts and olive oil was associated with decreased inflammation levels, whereas the Western diet with refined foods, sugar, fried foods, red and processed meat was associated with increased inflammation levels (Farhangi & Najafi, 2018; Mtintsilana et al., 2019; Esposito et al., 2004; Chrysohoou et al., 2004; King, Egan & Geesey, 2003). Furthermore, studies have highlighted the relationship between diet-induced inflammation and diabetes (van Woudenbergh et al., 2013; Minihane et al., 2015).

In light of these developments, the Dietary Inflammatory Index (DII) has been created to assess the effect of diet on chronic disease formation based on the relationship between pro- and anti-inflammatory serum markers, macro and micronutrients, and other dietary components (Shivappa et al., 2014). Energy, carbohydrates, proteins, total fat, saturated fat, trans fat, cholesterol, iron and vitamin B₁₂ have proinflammatory properties, whereas monounsaturated, polyunsaturated fats, omega-3, omega-6, fiber, beta-carotene, folic acid, thiamine, riboflavin, niacin, pyridoxine, retinol, ascorbic acid, tocopherol, vitamin D, magnesium, selenium, zinc, alcohol, caffeine, clove oil, garlic, ginger, onion, turmeric, saffron, green and black tea, flavones, flavonols, flavanones, isoflavones, anthocyanidins, pepper, thyme/marjoram and rosemary have anti-inflammatory properties (Shivappa et al., 2014). According to this index, which can be used in different populations, positive or near-positive values reflect the pro-inflammatory effect of the diet, and negative or near-negative values reflect the anti-inflammatory effect of the diet (Shivappa et al., 2014). One study found a positive relationship between DII and both the risk and severity of diabetes (King & Xiang, 2019). Additionally, a diet that triggers inflammation has been shown to increase the risk of prediabetes in a case-control study (Vahid et al., 2017).

Adipokines are defined as molecules derived from adipose tissue, playing roles in various biological mechanisms within the body. However, dysfunction in adipose tissue can adversely affect the adipokines it secretes, potentially leading to various diseases. Therefore, understanding the mechanisms of adipokines may offer new options for diagnosing and treating obesity-related diseases (Fasshauer & Blüher, 2015). Asprosin and omentin are two newly identified adipokines, in 2016 and 2006 respectively, both originating from visceral adipose tissue (Luo et al., 2023). Asprosin was discovered during a study focused on individuals diagnosed with neonatal progeroid syndrome (NPS). In the fasting state, it is an adipokine secreted by white adipose tissue that triggers the liver to release glucose (Romere et al., 2016). Asprosin causes inflammation in pancreatic β cells by stimulating cytokine release. This triggers apoptosis in β cells and leads to decreased insulin secretion (Lee et al., 2019). A cross-sectional study found higher plasma asprosin levels in groups with impaired glucose tolerance (IGT) and T2DM compared to a control group with normal glucose tolerance (Wang et al., 2018). Omentin (also known as intelectin) is an adipokine with anti-inflammatory and oxidative stress inhibitory properties first identified from human omental fat (Yang et al., 2006; Zhao et al., 2022). The omentin adipokine has been found to have a positive relationship with insulin sensitivity (Yang et al., 2006). Some studies have reported significantly lower serum omentin-1 levels in individuals with T2DM compared to controls (Zhang et al., 2014; Elsaid et al., 2018), while others have found significantly higher omentin-1 levels in those with T2DM relative to controls (Madsen et al., 2015; Hayashi et al., 2019).

The Glycemic Index (GI) is defined as the comparison of the glycemic response of a carbohydrate-containing food to that of a glucose solution or white bread containing the same amount of carbohydrate (Jenkins et al., 1981). A meta-analysis of studies involving individuals with T2DM has shown that those consuming a diet characterized by a low glycemic index have been more effective in controlling HbA1c and fasting blood glucose (FBG) levels compared to those consuming a high-glycemic index diet or a control group (Ojo et al., 2018). A randomized controlled study found that a diet characterized by a high glycemic index increases serum TNF-α and IL-6 levels (Kelly et al., 2011).

In conclusion, different immune response, such as higher levels of Interleukin-1 alpha, have been described in females vs. males in the context of inflammatory illnesses, such as autoimmune diseases (Tripolino et al., 2024). Prediabetes, a reversible condition, is associated with chronic inflammation (Tabák et al., 2012; Hamdy, Porramatikul & Al-Ozairi, 2006; Haffner, 2003). Therefore, determination of the levels of adipokines such as asprosin and omentin in this period may reflect metabolic and inflammatory change processes. It may also contribute to the identification of new biomarkers for diagnosis (Diao et al., 2024; As’habi et al., 2019). Although the roles of DII, GI and adipokine levels have been investigated separately in individuals with prediabetes, there are few studies in the literature investigating these variables together in individuals with prediabetes. In the light of the existing literature, a case-control study design was preferred in this study to evaluate the possible associations between DII and GI, serum asprosin, omentin, IL-6 and TNF-α among prediabetic women and age- and body mass index-matched healthy female controls. The chosen study design seems to be suitable for determining exposure differences between individuals with prediabetes and healthy (Dey, Mukherjee & Chakraborty, 2020). This study aims to examine of that GI and serum asprosin levels would increase and serum omentin levels would decrease as DII score increased in prediabetic women.

Materials and Methods

Results

Table 1 displays the basic clinical data for the case and control groups in the study population. The case group’s mean age was 39.10 ± 7.98 years, and the control group’s mean age was 37.57 ± 8.17 years, no statistically significant difference was observed (p = 0.465). No statistically significant difference was found between the case and control groups in height, body weight, BMI, hip circumference, fat mass (%), fat mass (kg), and lean tissue mass (p = 0.601, p = 0.593, p = 0.827, p = 0.616, p = 0.697, p = 0.731, p = 0.662, respectively). Waist circumference and waist/hip ratio were found significantly higher in the case group compared to the control group (p = 0.037, p = 0.011, respectively). In the case group, it was determined that all markers of the glucose metabolism consisting of FBG, HbA1c (%), insulin, HOMA-IR values and LDL-cholesterol, triglyceride, ALT values were significantly higher than those of the control group (p < 0.001, p < 0.001, p < 0.001, p < 0.001, p = 0.001, p < 0.001, p = 0.003, respectively), and HDL-cholesterol values were significantly lower than those of the control group (p = 0.002). It was determined that serum CRP, asprosin, IL-6, TNF-α, DII and GI values were significantly higher in the case group compared to the control group (p < 0.001, p = 0.010, p = 0.005, p < 0.001, p < 0.001, p = 0.001, respectively), and there was no statistically significant difference between the groups in terms of serum omentin values (p = 0.779).

Table 2 displays the correlation between the DII and inflammatory systemic biomarkers, adipokines, GI and glucose metabolism markers in the case and control groups, as well as for the entire study population. In the case group, a significant positive correlation was found between DII and insulin, HOMA-IR, and GI (r = 0.365, p = 0.047; r = 0.440, p = 0.015; r = 0.512, p = 0.004, respectively). For the entire study population, a significant positive correlation was observed between DII and asprosin, IL-6, TNF-α, CRP, HbA1c, FBG, insulin, HOMA-IR, and GI (r = 0.327, p = 0.011; r = 0.367, p = 0.004; r = 0.460, p < 0.001; r = 0.561, p < 0.001; r = 0.544, p < 0.001; r = 0.487, p < 0.001; r = 0.580, p < 0.001; r = 0.646, p < 0.001; r = 0.410, p = 0.001, respectively).

Table 3 shows the relationship between the DII and daily dietary intake of energy, nutrients, and certain flavonoids for both the case and control groups, as well as for the entire study population. A significant negative correlation was found between DII score and energy (r = −0.540, p = 0.002 and r = −0.581, p = 0.001) protein (r = −0.416, p = 0.022 and r = −0.654, p = 0.001), fat (r = −0.414, p = 0.023 and r = −0.409, p = 0.025), carbohydrate (r = −0.538, p = 0.002 and r = −0.524, p = 0.003), dietary fiber (r = −0.780, p < 0.001 and r = −0.858, p < 0.001), vitamin A (r = −0.575, p = 0.001 and r = −0.875, p < 0.001), beta-carotene (r = −0.702, p < 0.001 and r = −0.750, p < 0.001), vitamin E (r = −0.404, p = 0.027 and r = −0.719, p < 0.001), vitamin B6 (r = −0.720, p < 0.001 and r = −0.816, p < 0.001), folic acid (r = −0.768, p < 0.001 and r = −0.824, p < 0.001), vitamin C (r = −0.767, p < 0.001 and r = −0.763, p < 0.001), magnesium (r = −0.682, p < 0.001 and r = −0.812, p < 0.001), iron (r = −0.619, p < 0.001 and r = −0.798, p < 0.001), zinc (r = −0.457, p = 0.011 and r = −0.583, p = 0.001), monounsaturated fat (r = −0.442, p = 0.014 and r = −0.425, p = 0.019), polyunsaturated fat (r = −0.463, p = 0.010 and r = −0.580, p = 0.001), omega-3 (r = −0.452, p = 0.012 and r = −0.555, p = 0.001), omega-6 (r = −0.448, p = 0.013 and r = −0.540, p = 0.002), flavonol (r = −0.687, p < 0.001 and r = −0.730, p = p < 0.001) intakes in both the case group and the control group, respectively. In addition, DII score was found to have a significant negative relationship with flavone only in the case group (r = −0.787, p < 0.001) and anthocyanidin only in the control group (r = −0.417, p = 0.022). In the entire study population, a significant negative relationship was found between the DII and the intake of energy (r = −0.362, p = 0.005), protein (r = −0.519, p < 0.001), carbohydrates (r = −0.332, p = 0.009), dietary fiber (r = −0.852, p < 0.001), vitamins A, C, E and B6 (r = −0.695, p < 0.001; r = −0.782, p < 0.001; r = −0.468, p < 0.001; r = −0.831, p < 0.001) folic acid (r = −0.838, p < 0.001), magnesium (r = −0.799, p < 0.001), iron (r = −0.764, p < 0.001), zinc (r = −0.553, p < 0.001), monounsaturated fats (r = −0.311, p = 0.015), polyunsaturated fats (r = −0.378, p = 0.003), omega-3 (r = −0.594, p < 0.001), omega-6 (r = −0.330, p = 0.010), flavan-3-ol (r = −0.373, p = 0.003), flavone (r = −0.658, p < 0.001), flavonol (r = −0.781, p < 0.001), flavanone (r = −0.389, p = 0.002), anthocyanidin (r = −0.340, p = 0.008), and beta-carotene (r = −0.809, p < 0.001).

The results of the regression analysis based on the relationship between individuals’ DII scores and biochemical parameters are given in Table 4. In the first model, asprosin in the case group and CRP in the control group were identified as significant predictors in the regression model. It was determined that a one-unit increase in asprosin value increased the DII value by 0.421 units (F = 6.031, p = 0.021, β = 0.421, 95% CI [0.008–0.088], adjusted R² = 0.148), whereas a one-unit increase in CRP value increased the DII value by 0.472 units (F = 8.009, p = 0.009, β = 0.472, 95% CI [0.343–2.141], adjusted R² = 0.195).

The distribution of glycemic index, some adipokines and inflammatory markers according to DII score groups is shown in Table 5. According to the DII score groups, GI, asprosin, IL-6, TNF-α, CRP values were found to be significantly higher in the third group than in the first group (p = 0.004, p = 0.026, p = 0.014, p = 0.002, p < 0.001, respectively). Omentin values were not statistically significant compared to DII groups (p = 0.533).

Discussion

In our study, the case group showed significantly higher levels of DII score, GI value, serum IL-6, asprosin, TNF-α, and CRP, whereas serum omentin levels were higher in the control group, though not significantly. Across the entire study population, a significant positive relationship was found between DII and fasting blood glucose, HbA1c, insulin, HOMA-IR, asprosin, IL-6, TNF-α, CRP levels, and GI. An increase in asprosin in the case group and CRP in the control group was associated with a higher DII score.

In our current study, although the DII value was computed using 44 dietary components, it has been observed that many studies in the literature use fewer parameters, and the number of parameters varies across different studies (King & Xiang, 2019; Vahid et al., 2017; Denova-Gutiérrez et al., 2018). In studies conducted on individuals with T2DM, TNF-α, IL-6 and CRP levels were found to be significantly higher than controls (Toprak et al., 2022; Hu et al., 2004). Our findings consistent with those of the specified studies, as we found that serum TNF-α, IL-6, and CRP levels were significantly higher in the case group compared to the control group. Clinical studies and meta-analyses have shown that asprosin levels are elevated in individuals with T2DM and prediabetes compared to healthy controls (Diao et al., 2024; Li et al., 2018; Naiemian et al., 2020; Wang et al., 2021; Ertuna et al., 2023; Mahat et al., 2024; Ulloque-Badaracco et al., 2024). It has been suggested that asprosin may play a role in the prediabetic pathway (Ertuna et al., 2023). Consistent with these findings, our study also revealed that serum asprosin levels were higher in the case group compared to the control group. However, a study involving T2DM patients reported that plasma omentin-1 levels, before insulin treatment, did not show a significant difference between the case and control groups. In the same study, women were divided into three groups: the control group, a group of obese Type 2 diabetic patients before insulin therapy, and a group after 6 months of insulin therapy. No significant difference in omentin-1 levels was found among these groups (Komosinska-Vassev et al., 2019). A meta-analysis evaluating omentin-1 levels in different types of diabetes found that T2DM patients had lower omentin-1 levels compared to controls (Pan et al., 2019). However, some studies have reported significantly higher plasma omentin levels in T2DM patients compared to non-diabetic controls (Madsen et al., 2015; Hayashi et al., 2019). Similarly, in our study, although there was no significant difference between the case and control groups, serum omentin levels were higher in the case group compared to the control group. It has been stated that the lack of difference may be attributed to the complexity of omentin regulation mechanisms (Komosinska-Vassev et al., 2019). In a previous study, it was reported that the increase in omentin in the case group could be explained as a compensatory response to inflammation caused by chronic hyperglycemia in diabetes (Hayashi et al., 2019). The lack of a significant difference in omentin levels between the case and control groups in our study may be related to this mechanism. In addition, the small sample size may be the cause of the limited statistical relevance of some parameters, including omentin.

Our findings indicate that the DII score is positively associated with insulin and HOMA-IR in the case group, and positively associated with FBG, HbA1c, insulin, and HOMA-IR across all participants. Similar results have been observed in some studies involving individuals with T2DM or prediabetes (Mtintsilana et al., 2019; Shu et al., 2022; Zhou et al., 2025). These studies found a significant positive relationship between DII score and FBG, fasting insulin, HbA1c, and HOMA-IR. However, a study involving healthy adults found no significant relationship between DII score and FBG, fasting insulin, or insulin resistance (Moslehi et al., 2016). In a case-control study involving prediabetic individuals, participants with high DII scores had significantly higher fasting plasma glucose (FPG) and HbA1c levels compared to those with low DII scores (Vahid et al., 2017). These differing results in the literature may be attributed to variations in study design and the types of patients selected (Moslehi et al., 2016).

It was previously determined that the DII score was associated with inflammatory markers IL-6, TNF-α and CRP (Shivappa et al., 2014). In addition, a case-control study of obese women with T2DM found a significant positive relationship between DII scores and IL-6, TNF-α, and hs-CRP among all participants (Toprak et al., 2022). However, in our study, no significant relationship was found between DII scores and these inflammatory markers in the case and control groups. This lack of significant association might be attributed to the reduced statistical power due to the small sample size. Nonetheless, when analyzing all participants, a significant positive relationship was identified. Furthermore, in the control group, an increase in serum CRP levels was associated with higher DII scores. Our study also found that IL-6, TNF-α, and CRP levels were significantly higher in the third DII score group (T3) compared to the first group (T1).

In a different study, individuals with high pro-inflammatory DII scores had higher levels of CRP, IL-6, and TNF-α, and lower levels of adiponectin (which has anti-inflammatory properties) compared to those with high anti-inflammatory DII scores (Phillips et al., 2018). In our study, serum asprosin levels were significantly higher in the third DII score group compared to the first group, whereas no significant difference was found for serum omentin levels. Additionally, our findings indicate that an increase in serum asprosin levels was associated with a higher DII score in the case group. To our knowledge, there are no studies investigating the relationship between dietary inflammatory load and serum asprosin levels. However, one study has explained that a positive relationship between the DII score and another pro-inflammatory adipokine suggests that the inflammatory properties of the diet might impact adipose tissue inflammation (Muhammad et al., 2019). In a study involving healthy obese individuals, it was found that those with high pro-inflammatory DII scores had higher levels of pro-inflammatory chemerin compared to those with low DII scores, although there was no significant difference in omentin levels between groups with different DII scores (Mirmajidi et al., 2019). In our study, given that both the case and control groups consisted of mildly overweight and obese individuals, it is thought that the excessive fat accumulation in these individuals might influence the current findings (Nosrati-Oskouie et al., 2021). Further research is needed to investigate the relationship between serum omentin levels and the Dietary Inflammatory Index in prediabetic individuals. In the studies conducted, it has been found that individuals with high DII scores have high pro-inflammatory and low anti-inflammatory food parameters, while those with low DII scores have low pro-inflammatory and high anti-inflammatory food parameters (Denova-Gutiérrez et al., 2018; Mirmajidi et al., 2019; Saboori et al., 2024). Generally, our research results are consistent with the findings of these studies.

Studies have indicated that the GI of the diet is related to HbA1c, FBG, and glycemic control in individuals with prediabetes or diabetes (Zafar et al., 2019; Zhu et al., 2021). A randomized controlled trial found that a low glycemic index diet prevented adverse inflammatory responses (Gomes, Fabrini & de Alfenas, 2017). In our study, a significant positive relationship was found between the DII score and GI value in women with prediabetes. There is no literature available comparing both indices in individuals with prediabetes. However, according to the literature, both the DII score and GI value are positively associated with glucose metabolism markers and inflammatory markers, as mentioned above. This may help explain the current relationship between the two indices. Overall, differences in the findings of studies in the literature may be attributed to variations in study populations and designs, participant gender and age, disease type, diagnostic criteria, the number of dietary components used in DII value calculations, and sample sizes.

In our study, the participants in both the case and control groups were selected to be comparable with regards to age and BMI, and extensive exclusion criteria were implemented to maintain the homogeneity of the study. Since the study included only female participants, an age restriction was implemented to prevent unknown effects of menopausal physiological changes on asprosin. To calculate the DII score, 44 out of 45 food parameters were used, which is more comprehensive than the majority of the literature. Our study is, as far as we know, the first in Türkiye to investigate the relationship between the DII, glycemic index, and asprosin in women with prediabetes. Additionally, given the limited number of clinical studies examining the relationship between DII scores, prediabetes, and the dietary influence on asprosin, our study is expected to make a significant contribution to the literature in this regard.

Due to the cross-sectional design of the study, causality cannot be inferred. For some of the relationships examined, the small size of the sample may have reduced statistical power. The fact that various secondary factors that may affect omentin levels (e.g., age, BMI, physical activity level or regional dietary customs) could not be controlled in our study constitutes a limitation in the interpretation of the results. The results of the DII and GI calculation may also be affected by potential recall bias due to the individual FFQ assessment method. Since the study was conducted only among pre-menopausal Turkish women, the findings may not be generalizable. Although regression analyses are suitable for hypothesis generation, as confounding variables such as physical activity, or socioeconomic influences cannot be adequately accounted for, replication of these findings with larger samples would confirm the exploratory results and provide external validity.

Conclusions

The association of the increase in DII scores with the increase in serum asprosin levels in prediabetic women may provide preliminary findings on prediabetes-specific biomarkers, while no significant difference between the groups in terms of serum omentin levels may be due to the complexity of the regulation mechanism. However, since the cross-sectional design of the present study limits the establishment of causality, detailed analyses should be performed in prospective studies with larger sample groups to investigate the diagnostic or therapeutic benefits of these adipokines.

In the following process, it is important to design new studies that can evaluate the practical applicability of the DII score calculation system in prediabetes and other different disease processes and whether diets prepared according to this index benefit individuals. It is important to note that given the irreversible nature of diabetes, its serious complications, decreased quality of life, increased morbidity and mortality risks, and the economic burden on individuals and countries, focusing on prediabetes—an earlier, reversible stage of diabetes—and its management is highly valuable. It is thought that designing studies that will clinically investigate the relationship between the current diets, dietary inflammatory index values and dietary interventions to be applied to these individuals, serum asprosin and omentin adipokines and additionally clearly reveal the biological mechanisms underlying this condition will enable strategy development and contribute to the literature gap in the current field. In order to confirm the mediating role of adipokines in the metabolic effects of diet, it is recommended that longitudinal or experimental study designs involving anti-inflammatory dietary interventions be favoured in the future.

Supplemental Information