Introduction

Diabetic foot ulcer (DFU) is a serious complication in patients with diabetes mellitus (DM)1. DFU usually manifests as a neurological ulcer or neuroischemic ulcer and is characterized by hyperinflammation, hypoxia, and persistent infection2,3. The global prevalence of DFU is reportedly 6.3%, varying from 3.0 to 16.6% across different regions and countries4. DFU affects approximately 18.6 million people worldwide each year, with infections accounting for approximately 50–60%, lower extremity amputations accounting for approximately 20%, and 5-year mortality rates exceeding 30%5, which impose significant physical and economic burdens on both individuals and the healthcare system.

Numerous studies have shown that the occurrence and development of DFU are attributed to oxidative stress mediated by reactive oxygen species and free nitrogen radicals6,7, proinflammatory cytokine release8, cellular dysfunction9, pathogenic microorganisms10,11, etc. The entire process involves chronic nonspecific immune-inflammation. Therefore, in recent years, several scholars have employed immunoinflammatory biomarkers, such as the neutrophil‒lymphocyte ratio (NLR), platelet‒lymphocyte ratio (PLR), and lymphocyte‒monocyte ratio (LMR), to predict the risk of peripheral arterial disease, peripheral neuropathy, diabetes-related osteomyelitis, and amputation in DFU patients12,13. The recent guidelines on diabetes-related foot diseases, published by the International Working Group on the Diabetic Foot (IWGDF) and the Infectious Diseases Society of America (IDSA), introduce care and treatment recommendations regarding patient-important evidence and outcomes while also emphasizing the importance of immunoinflammatory biomarker testing14. However, these recommendations principally focus on wound identification, care and disposal, and inflammation management.

The prolonged treatment period, high recurrence rate, and significant morbidity pose challenges to the management of DFU. A novel multidisciplinary management approach that incorporates IWGDF/IDSA recommendations along with a comprehensive health assessment is necessary15. These comprehensive management targets may address aberrant glucose, lipid, and protein metabolism, as well as impaired renal function. Given the pathogenic mechanism of chronic nonspecific immune inflammation, this study aimed to investigate the correlations among cytokines, immunoinflammatory markers, and metabolic disorders related to nutrition, lipids, and kidney function in DFU patients.

Methods

Ethical review

This study was approved by the Ethics Committee of Mianyang Central Hospital in accordance with the principles of the Declaration of Helsinki (Approval No. S2022092, date: Aug. 27, 2022). All participants signed informed consent forms.

Patients

From January 2023 to December 2023, this study enrolled 85 patients with DFU and 53 patients with DM at Mianyang Central Hospital, School of Medicine, University of Electronic Science and Technology of China, including 61 males and 24 females with ages ranging from 42 to 83 years. The inclusion criteria for DFU patients were as follows: (1) all DFU patients met the diagnostic criteria of the IWGDF/IDSA 2023 guidelines14; (2) had a history of diabetes for more than 5 years; and (3) did not have other diabetes-related microvascular complications. The inclusion criteria for DM patients were as follows: (1) all DM patients met the diagnostic criteria of the American Diabetes Association guidelines16; (2) had a history of diabetes for more than 5 years; and (3) without microvascular complications. The exclusion criteria for the both were as follows: (1) had comorbidities of other immune diseases, infectious diseases, malignant tumors, cardiovascular diseases, pulmonary diseases, hepatic diseases, or renal diseases; and (2) with treatment with glucocorticoids within one month. During the same period, a total of 51 healthy individuals who underwent routine physical examinations were recruited as healthy controls (HC). These participants exhibited normal cardiac, pulmonary, hepatic, and renal functions, with no abnormalities detected in their blood and urine test results.

Sample collection

For each participants, two fasting venous blood samples of 3.0 mL were collected via SST-II and EDTA-2 K vacuum tubes (BD, USA). The samples in SST-II tubes were centrifuged at approximately 1500×g for 10 min. Subsequently, the serum was divided into two portions. One was used to quantitate the levels of routine biochemical marker, such as glucose, protein, lipid, and kidney biomarkers with 2 h. Another was used for cytokine detection. If immediate cytokine detection is not feasible, the serum sample should be stored at 2 ~ 8℃ less than 7 days. The EDTA-2 K anticoagulated samples were used firstly for complete blood count within 2 h, subsequently for quantitation of glycosylated hemoglobin A1c (HbA1c) within 4 h.

Cytokine detection

Cytokine detection was performed on a NAVIOS flow cytometry (Beckman-Coulter, USA), via flow cytometry with a 12-color scheme provided by the Cytokine Multiplex Detection Kit (CatNo: P010100403, CERGER, CellGene Biotech Co., Ltd., CHN). This 12-color scheme could detect 9 interleukins (ILs, including IL1β, IL2, IL4, IL5, IL6, IL8, IL10, and IL12P70), 2 interferons (IFNs, including IFNα and IFNγ), and tumor necrosis factor α (TNFα).

Routine biochemical marker assay and HbA1c measurement

The routine biochemical markers and their detection methods included in the study are as follows: total protein (TP), biuret method; albumin (Alb), bromocresol green method; glucose (Glu), hexokinase method; urea (Ure), urease method; creatinine (Cre), sarcosine oxidase method; triglyceride (TG), glycerol phosphate oxidase method; total cholesterol (T-Chol), cholesterol oxidase method; cystatin C (CysC), high-density lipoprotein cholesterol (HDL-Chol), low-density lipoprotein cholesterol (LDL-Chol), immunoturbidimetric method. All biochemical reagents were provided by Sichuan Maker Biotechnology Co., Ltd., China. All biochemical marker assay was performed on an LST008 automatic biochemical analyzer (HITACHI, JAN). The estimated glomerular filtration rate (eGFR) was calculated via the CKD-EPI CysC-Cr equation17.

The HbA1c level was measured via column chromatography on a Variant II TURBO analyzer (Bio-Rad, USA).

Complete blood count assay and Immunoinflammatory biomarker calculation

The complete blood count was measured using a Sysmex XN-9000 analyzer (Sysmex, JPN) and accompanying special reagent. Both leukocyte differential counts and platelet counts were determined via the sheath-flow method, and the count results were used to calculate the following immunoinflammatory markers: the neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), platelet-to-lymphocyte ratio (PLR), systemic inflammation index (SII), and Prognostic nutritional index (PNI)18,19.

$${\text{SII}} = {\text{platelet count}} \times {\text{ neutrophil count/lymphocyte count}}.$$
$${\text{PNI }} = 5 \times {\text{ lymphocyte count }} + {\text{ Alb}}.$$

Evaluation of malnutrition, dyslipidemia, and kidney dysfunction

The dyslipidemia evaluation was based on the criteria of the American Association of Clinical Endocrinologists and American College of Endocrinology (AACE/ACE) guidelines20: T-Chol ≥ 5.2mmol/L (200 mg/dL), or TG ≥ 1.7mmol/L (150 mg/dL), or HDL-Chol < 1.0mmol/L (40 mg/dL), or LDL-Chol ≥ 3.4mmol/L (130 mg/dL). The malnutrition evaluation was based on a conventional indicator that has been used for decades, namely, a decrease in the serum Alb concentration to 35 g/L21. The kidney dysfunction (KD) evaluation was based on a decrease in the eGFR to 60 mL/min/1.73m222.

Statistical analysis

Statistical analysis was performed via MedCalc software v20.1 (MedCalc, Belgium) and SPSS software v22.0 (SPSS, USA). The continuous data with a nonnormal distribution are presented as the median (P25, P75), whereas the continuous data with a normal distribution are presented as the mean ± SD. The multiple-group difference was analyzed via the independent-samples Kruskal-Wallis test (for nonnormally distributed data), or the one-way ANOVA (for normally distributed data), or chi-square test (for count data). Pairwise comparison of measurement data was used the Bonferroni correction test. The two-group difference was analyzed via the Mann‒Whitney test (for nonnormally distributed data) or the independent-samples t test (for normally distributed data). The bivariate correlation between metabolic disorders/routine biomarkers and cytokines/immunoinflammatory markers was illustrated in a two-matrix heat plot, with a |Spearman correlation coefficient (ρ)| ≥0.3 as the criterion for determining evident correlations23. Independent correlations between metabolic disorders and cytokine/immunoinflammatory marker levels were determined via logistic regression analysis. The predictive performance was analyzed by receiver operating curve (ROC), and determined comprehensively by the area under the curve (AUC), sensitivity and specificity. The restricted cubic spline (RCS) was employed to investigate the linear or nonlinear relationship between cytokines and metabolic disorders. The significance level was set at α = 0.05.

Results

Serum levels of cytokines and kidney/lipid/nutritional/inflammatory biomarkers

There were no statistical differences in sex or age among the DFU, DM, and HC groups (both P > 0.05). Except for IL1β, IL5, INFα, INFγ, and LMR (F/χ2 = 0.430 ~ 4.437, all P > 0.05), other biomarkers demonstrated significant differences among the three groups (F/χ2 = 2.613 ~ 155.622, all P < 0.05) (Table 1). Compared with the HC group, the DFU group had significantly lower levels of IL2, IL4, IL10, IL12P70, IL17, and TNFα, but significantly higher levels of IL6, IL8, NLR, PLR, PNI, and SII (all P < 0.05); Compared with the DM group, the DFU group had reduced levels of IL2, IL10, and IL17, but elevated levels of IL6 and IL8, (all P < 0.05).

Table 1 Basic information and laboratory test data of the participants.
Full size table

Cytokine release, kidney dysfunction, dyslipidemia, and malnutrition in DFU patients

Among these DFU patients, the prevalences of kidney dysfunction, dyslipidemia, and malnutrition were 43.5%, 78.8%, and 82.4%, respectively (Fig. 1). Among the 12 cytokines, only IL4, IL6, IL8, IL10, IL12P70, IL17, and TNFα exhibited overexpression rates of 7.1%, 82.4%, 9.4%, 2.4%, 17.6%, 17.6%, and 7.1%, respectively. Among them, IL-4, IL-10, IL-12p70, IL-17, and TNF-α exhibited a paradoxical pattern in the DFU group, characterized by decreased median levels but increased overexpression rates, indicating significant variation ranges. Furthermore, the overexpression rates of IL6 alone, co-overexpressed with one cytokine, and co-overexpressed with two cytokines were 50.6%, 8.2%, and 23.5%, respectively. However, the simultaneous overexpression of four or more cytokines was not observed.

Fig. 1
figure 1

Cytokine overexpression, kidney dysfunction, dyslipidemia, and malnutrition in DFU and DM patients. *P < 0.05, **P < 0.001. IL, interleukin; IFN, interferon; TNF, tumor necrosis factor. Compared with the DM patients, the prevalences of malnutrition, dyslipidemia, and kidney dysfunction were higher in DFU patients. Of seven overexpressed cytokines, the DFU patients had higher overexpression rates of IL6 and IL17.

Full size image

Compared with the DM patients, the DFU patients had higher prevalences of kidney dysfunction (χ2 = 31.294, P < 0.001), dyslipidemia (χ2 = 16.234, P < 0.001), and malnutrition (χ2 = 26.348, P < 0.001), as well as higher overexpression rates of IL6 (χ2 = 12.209, P < 0.001), IL17 (χ2 = 10.418, P < 0.001), and IL6 co-overexpressed with two cytokines (χ2 = 4.341, P = 0.037). However, there were no statistical differences in the overexpression rates of other cytokines, IL6 alone, and IL6 co-expressed with one cytokine (all P > 0.05).

Cytokine and Immunoinflammatory marker levels in different metabolic disorders among DFU patients

According to the above analysis, this study focused on the performance of 7 cytokines (IL4, IL6, IL8, IL10, IL12P70, IL17, and TNFα) and 4 immunoinflammatory biomarkers (NLR, PLR, SII, and PNI). Betweem DFU patients with and without kidney dysfunction, four immunoinflammatory biomarkers (NLR, PLR, SII, and PNI: z = 2.504, 3.000, 3.280, and 2.783, P = 0.012, 0.003, < 0.001, and 0.005) as well as four cytokines (IL4, IL6, IL8, and TNFα: z = 3.092, -3.262, -3.790, 2.664, P = 0.002, < 0.001, < 0.001, and 0.008) exhibited significant variations (Fig. 2A). Between DFU patients with and without dyslipidemia, only IL12P70 and IL17 did (z=-1.969 and − 3.073, both P = 0.049 and 0.002) (Fig. 2B). However, none of them showed significant differences between DFU patients with and without malnutrition (z=-1.672 ~ 1.199, all P > 0.05) (Shown in Supplementary Table 1).

Fig. 2
figure 2

Serum biomarker levels among the DFU patients with and without kidney dysfunction or dyslipidemia. *P < 0.05, **P < 0.001. KD, kidney dysfunction; IL, interleukin; IFN, interferon; TNF, tumor necrosis factor; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; SII, systemic inflammation index; PNI, Prognostic nutritional index. Significant differences were found in IL4, IL6, IL8, TNFα, NLR, PLR, SII, and PNI levels between DFU patients with and without kidney dysfunction, while in IL12P70 and IL17 levels between DFU patients with and without dyslipidemia.

Full size image

Spearman correlation between cytokine/immunoinflammatory biomarkers and routine biomarkers of three metabolic disorders

The threshold of |ρ|≥0.3 was employed to determine evident correlations23. For malnutrition-related biomarkers, evident correlations were observed between the LMR and the TP/Alb ratio (both ρ = 0.31, P < 0.001), as well as between the NLR and the Alb level (ρ=-0.32, P < 0.001) (Fig. 3). With respect to dyslipidemia-related biomarkers, evident correlations were observed between IL6 and five lipid biomarkers (T-Chol, TG, HDL-Chol, LDL-Chol, and ApoA1) (|ρ|=0.32 ~ 0.67, all P < 0.001), between IL8 and three lipid biomarkers (HDL-Chol, LDL-Chol, and ApoA1) (|ρ|=0.40 ~ 0.43, all P < 0.001), between IL10/IL12P70 and T-Chol (ρ = 0.35/0.33, both P < 0.001), between IL17 and T-Chol/LDL-Chol (|ρ|=0.32/0.34, both P < 0.001), and between all five immune biomarkers and four lipid biomarkers (T-Chol, TG, HDL-Chol, and LDL-Chol) (|ρ|=0.33 ~ 0.59, all Padj < 0.001). With respect to kidney biomarkers, evident correlations were observed between IL8/IL10/IL12P70/TNFα and Ure (|ρ|=0.30 ~ 0.39, all P < 0.001), between IL17/TNFα and Ure (|ρ|=0.35/0.32, all P < 0.001), between IL8 and CysC (ρ = 0.32, P < 0.001), and between IL8/IL17/TNFα and the eGFR (|ρ|=0.31 ~ 0.36, all P < 0.001). These results suggest a certain correlation between immune inflammation and three metabolic disorders (malnutrition, dyslipidemia, and kidney dysfunction) in DFU patients.

Fig. 3
figure 3

Bivariate correlation coefficients of cytokines and immunoinflammatory biomarkers with biomarkers of malnutrition, dyslipidemia, and kidney dysfunction. × shape, P > 0.05. Six overexpressed cytokines (IL6, IL8, IL10, IL12P70, IL17, and TNFα) and five immunoinflammatory biomarkers demonstrated evident correlations with at least one biomarker related to malnutrition, dyslipidemia, and kidney dysfunction.

Full size image

Independent correlations of cytokines and Immunoinflammatory biomarkers with three metabolic disorders

After adjusting for age and sex, multivariable logistic regression analysis revealed an independent risk association of decreased IL10 levels with malnutrition (OR = 0.29, P = 0.027), elevated IL6 and decreased IL12P70 levels with kidney dysfunction (OR = 1.01 and 0.23, P = 0.008 and 0.046), as well as elevated IL17 and the SII with dyslipidemia (OR = 1.66 and 1.01, P = 0.025 and 0.037) (Table 2). These results suggest that malnutrition, dyslipidemia, and kidney dysfunction in DFU patients might be closely related to immune inflammation.

Table 2 The odds ratio of 7 overexpressed cytokines and 5 Immunoinflammatory biomarkers with three metabolic disorders.
Full size table

However, further ROC analysis revealed a notable significance of IL6 in predicting kidney dysfunction (AUC ± SE = 0.707 ± 0.036, Sensitivity = 91.9% with 95%CI ranged from 78.1 to 98.3%, Specificity = 52.1 with 95%CI ranged from 37.2 to 66.7%) (Fig. 4B), as well as IL17 in predicting dyslipidemia (AUC ± SE = 0.749 ± 0.045, Sensitivity = 64.2% with 95%CI ranged from 51.5 to 75.5%, Specificity = 94.4 with 95%CI ranged from 72.7 to 99.9%) (Fig. 4C). On the contrary, IL10, IL12P70, and SII did not demonstrate predictive value for the corresponding metabolic disorders (all P > 0.05, Fig. 4A–C). Noteworthyly, the combined pattern of IL6 and IL12P70 demonstrated an improved predictive performance compared to IL6 alone for kidney dysfunction (AUC = 0.814 vs. 0.707, z = 2.991, P = 0.003). Nevertheless, the combined pattern of IL17 and SII did not significantly differ compared to IL17 alone for dyslipidemia (AUC = 0.795 vs. 0.749, z = 0.759, P = 0.448).

Fig. 4
figure 4

ROC analysis of independent factors associated with metabolic disorders. ROC, receiver operating curve; KD, kidney dysfunction; IL, interleukin; SII, systemic inflammation index. (A) ROC curve of IL10 for malnutrition; (B) ROC curves of IL6, IL12P70,and both for kidney dysfunction; (C) ROC curves of IL17, SII, and both for dyslipidemia.

Full size image

Linear and nonlinear relationships between cytokines and three complications

RCS analysis revealed a clear linear correlation between IL6 and kidney dysfunction (p for overall = 0.029, p for nonlinearity = 0.255), with a single risk prediction threshold of 30.41 ng/L (Fig. 5A). IL12P70 showed a clear nonlinear correlation with kidney dysfunction (p for overall = 0.003, p for nonlinearity = 0.001), presenting two risk prediction thresholds at 1.50 ng/L and 5.06 ng/L (Fig. 5B). IL17 showed a clear nonlinear correlation with dyslipidemia (p for overall = 0.008, p for nonlinearity = 0.018), with a single risk prediction threshold of 4.56 ng/L (Fig. 5C). However, the relationship between IL-10 and malnutrition, whether linear or nonlinear, was found to be insignificant (p for overall = 0.193, p for nonlinearity = 0.268), and IL-10 merely exhibited a protective effect against malnutrition within a narrow range of 2.56 ng/L-3.74 ng/L (Fig. 5D).

Fig. 5
figure 5

Linear and nonlinear relationship analysis via the RCS. RCS, restricted cubic spline; KD, kidney dysfunction; IL, interleukin; IFN, interferon; TNF, tumor necrosis factor. (A) RCS curve between IL6 and kidney dysfunction, presenting a linear relationship; (B) RCS curves between IL12P70 and kidney dysfunction, presenting a nonlinear relationship of U shape; (C) RCS curve of IL17 for dyslipidemia, presenting a nonlinear relationship of n shape; (D) RCS curve of IL10 for malnutrition, without a linear or nonlinear relationship.

Full size image

Discussion

Owing to the high recurrence, amputation, and mortality rates, DFU significantly impairs patients’ quality of life and mental health. Consequently, the IWGDF/IDSA guidelines recommend a comprehensive set of management strategies aimed at addressing ulcer wounds, associated complications, and inflammation to prevent adverse outcomes. However, various metabolic disorders remain critical aspects that must not be disregarded. From the perspective of chronic immune inflammation pathogenesis, this study revealed an evident correlation between immune inflammation and metabolic disorders among DFU patients. Independent correlation analysis revealed significant associations between decreased levels of IL10 and malnutrition, elevated levels of IL6, decreased levels of IL12P70 with kidney dysfunction, and elevated levels of IL17 and the SII with dyslipidemia. Further RCS analysis ultimately demonstrated distinct linear or nonlinear correlations between these cytokines and corresponding metabolic disorders, except between IL-10 and malnutrition. These findings concerning metabolic disorders may contribute to the development of optimal management strategies for DFU patients.

To date, uncertainty remains regarding the optimal wound healing strategy for DFU patients24. The laboratory tests specifically mentioned in the IWGDF/IDSA 2023 guidelines include C-reactive protein, the erythrocyte sedimentation rate, and procalcitonin; cytokines are not included. Recent studies have shown that elevated levels of the proinflammatory factor TNFα and reduced levels of the anti-inflammatory factors AproA1 and IL10 are involved in DFU development and progression and can serve as grading markers for DFU25,26. Some researchers have investigated the effects of a few ILs on wound healing in animal experiments27,28, emphasizing the link between DFU and immune cell dysfunction29. However, this immune cell dysfunction is nonspecific and affects not only foot inflammation but also other metabolic disorders, ultimately impacting the optimization of management strategies.

Unexpectedly, the DFU patients in this study presented reduced levels of the proinflammatory cytokines IL12P70, IL17, and TNFα. This phenomenon may be attributed to routine hypoglycemic therapy, as therapeutic glucose regulation in diabetic patients has been linked to decreased IL12P70 and IL17 levels30,31,32. Furthermore, it remains to be debated whether IL12 is either a proinflammatory or anti-inflammatory cytokine in DFU patients. IL12P70 is a heterodimer composed of p35 and p40 subunits. The former is also found in the anti-inflammatory cytokine IL35, while the latter is present in the proinflammatory cytokines IL23 and IL2733. This heterogeneous dimeric structure may enable IL12P70 to exert pleiotropic effects, encompassing both proinflammatory and anti-inflammatory responses, contingent upon the inflammatory microenvironment34. However, IL12P70, IL17, and TNFα were overexpressed to a certain extent in cytokine overexpression analysis, suggesting a relatively wide variation range in DFU patients. A certain number of patients still present with high levels of them. Given that therapeutic glucose regulation is associated with decreased proinflammatory cytokine levels in DFU patients, the relationship between cytokine release and metabolic dysfunction needs further investigation to optimize immunotherapy strategies.

Malnutrition is significantly prevalent among DFU patients and has a detrimental effect on patient outcomes35. A recent study validated the efficacy of two nutritional assessment tools, namely, the Global Leadership Initiative on Malnutrition (GLIM) and the Subjective Global Assessment (SGA), in diagnosing immune malnutrition and predicting wound healing36. However, the relationship between malnutrition and immune inflammation remains controversial37,38. Although an evident Spearman correlation between cytokines and malnutrition was not found in DFU patients, our study revealed an independent association between IL10 and malnutrition via logistic regression analysis. Unfortunately, RCS analysis revealed that a narrow range of IL10 exhibited a protective effect against malnutrition. Bracingly, a study on IL10 polymorphisms confirmed that genetic mutations in IL10 could lead to lower serum levels of IL10 and increased susceptibility to malnutrition39. Therefore, we hypothesize that appropriately increasing the serum concentration of IL-10 may help alleviate malnutrition in DFU patients. Its effectiveness needs further validation through clinical trials.

Both chronic kidney disease and DFU are common complications of diabetes mellitus, with a significant epidemiologic link between their risk, occurrence, and prognosis40,41. Our findings revealed that IL6 and IL12P70 are independently associated with kidney dysfunction in DFU patients, with IL6 levels greater than 30.41 ng/L playing a risk role and IL12P70 levels ranging from 1.50 ng/L to 5.06 ng/L playing a protective role. The genetic mutation of IL6 has been reported to increase susceptibility to kidney dysfunction and is associated with atherosclerosis42,43. Additionally, elevated serum levels of IL6 have been shown to contribute to the progression of renal anemia44. However, the association of IL12P70 with kidney dysfunction has rarely been reported. Some studies have reported no significant association between IL12P70 levels and kidney dysfunction in patients with diabetic kidney disease45, whereas a controversial link exists between IL12P70 levels and arterial stiffness in patients with chronic kidney disease46,47. These findings from other disease studies support the feasibility of anti-IL6/pro-IL12 intervention strategies to delay the progression of kidney dysfunction. However, such strategies may increase the risk of side effects due to the multiple homeostatic roles of IL6 and IL12P7048. Therefore, further investigations are necessary to establish a comprehensive benefit‒risk intervention in the context of DFU.

Dyslipidemia can increase the risk of DFU49 and continues to affect disease progression and patient outcomes50. However, little is known about the relationship between IL17 and dyslipidemia in DFU patients. Our findings revealed an independent association of IL17 and the SII with dyslipidemia in DFU patients, and IL17 levels greater than 4.56 ng/L indicated a significant risk for dyslipidemia. Recent studies have also confirmed a significant positive association between the SII and dyslipidemia in the NHANES (2015–2020) population51 as well as in patients with type 2 diabetic peripheral neuropathy52. Similarly, IL17 is positively associated with dyslipidemia in other diseases, such as nonalcoholic fatty liver disease and obstructive sleep apnea syndrome53,54. Therefore, targeting IL17 and/or the SII for lipid-lowering strategies may have universal applicability.

Taken together, our findings represent the first preliminary and systematic investigation into the independent associations and dose-response curves between multiple cytokines and major metabolic disorders in DFU patients. To date, limited studies have examined some cytokine and inflammatory biomarker expression in certain DFU complications, demonstrating that elevated IL-6 and IL10 levels correlate with the risk of anemia55 and peripheral neuropathy56. Our findings go even further to indicate an independent risk of immune inflammation for multiple metabolic disorders in DFU patients. Further detailed exploration of the relationship between cytokines and metabolic disorders, particularly the determination of cytokine risk thresholds for specific metabolic disorders, could provide valuable evidence to inform future immunotherapy strategies for DFU.

Limitations: Positive conclusions from small-sample, single-center studies should be confirmed by large-sample, multicenter studies. Conventional methods for evaluating malnutrition may introduce bias into our conclusions. Management strategies derived from related studies should be subjected to animal experiments before clinical trials to assess the benefit‒risk profile of changes in the cytokine network.

Conclusions

Given the pathogenesis of chronic nonspecific immunoinflammatory injury in DFU, we conducted this study to investigate the correlations between cytokine release and metabolic dysfunctions among patients with DFU. The findings revealed the significant roles of different cytokines in malnutrition, renal dysfunction, and dyslipidemia. Further identifying the comprehensive benefit‒risk profile of these cytokines may facilitate the development of immunotherapy, thereby advancing optimal management strategies.