Metabolic and inflammatory indices in colorectal adenoma: A real-world colonoscopy cohort study

Abstract

BACKGROUND

Colorectal adenomas are precursors of colorectal cancer and have been linked to metabolic and inflammatory dysregulation. The clinical relevance of laboratory-based indices for adenoma detection remains unclear.

AIM

To evaluate the association and comparative diagnostic performance of metabolic and inflammatory indices in patients with colorectal adenomas who underwent colonoscopy.

METHODS

This retrospective cohort colonoscopy-based study included adult patients who underwent complete colonoscopic examination. Patients were classified into adenoma and non-adenoma groups according to histopathological findings. Metabolic indices, including the triglyceride-glucose (TyG) index and the atherogenic index of plasma (AIP), and inflammatory indices, such as the neutrophil-to-lymphocyte ratio, systemic immune-inflammation index, and neutrophil-to-high-density lipoprotein ratio, were calculated. Group comparisons, receiver operating characteristic analyses, and age-adjusted and sex-adjusted logistic regression analyses were performed.

RESULTS

A total of 1570 patients were included in the final analysis cohort, and 395 patients (25.2%) had colorectal adenomas. TyG and AIP were significantly greater in patients with adenomas (P < 0.001). Receiver operating characteristic analyses showed modest discriminatory performance, with TyG having the highest area under the curve. After adjusting for age and sex, the TyG index and AIP remained significantly associated with the presence of adenoma, whereas the inflammatory indices did not retain statistical significance.

CONCLUSION

Metabolic indices, particularly the TyG index and AIP, are significantly associated with the presence of colorectal adenoma, whereas systemic inflammatory indices demonstrate limited diagnostic value. These findings support a potential role of metabolic dysregulation in early colorectal neoplasia.

Key Words: Colorectal adenoma; Triglyceride-glucose index; Atherogenic index of plasma; Insulin resistance; Colonoscopy; Risk stratification

Core Tip: Metabolic dysregulation has been increasingly implicated in colorectal neoplasia; however, the clinical utility of routinely available metabolic and inflammatory indices in colorectal adenoma detection remains uncertain. In this large real-world colonoscopy cohort, the triglyceride-glucose index and the atherogenic index of plasma were independently associated with colorectal adenoma presence after adjustment for age and sex, whereas systemic inflammatory indices demonstrated limited diagnostic value. These findings highlight the potential role of metabolic indices in colorectal adenoma risk stratification and support further prospective validation.

INTRODUCTION

Colorectal adenomas are the main precursor lesions in the adenoma–carcinoma sequence and represent key targets for early detection and prevention[1]. Although colonoscopy remains the gold standard for adenoma detection, identifying individuals at higher risk may increase the efficiency of screening strategies and risk stratification in routine clinical practice[2].

In recent years, increasing evidence has highlighted the role of metabolic dysregulation in colorectal neoplasia. The core components of metabolic syndrome, including insulin resistance, dyslipidemia, central obesity, and impaired glucose metabolism, are associated with colorectal adenoma development through mechanisms involving chronic inflammation, oxidative stress, and altered growth factor pathways[3-5]. These systemic metabolic alterations may influence colorectal carcinogenesis long before the onset of overt malignancy[6].

Several metabolic and inflammatory indices derived from routine laboratory parameters have emerged as potential surrogate markers of metabolic dysfunction. Among these indices, the triglyceride-glucose (TyG) index has been proposed as a simple and reliable indicator of insulin resistance and has been associated with adverse cardiovascular and metabolic outcomes[7,8]. Similarly, lipid-based indices such as the atherogenic index of plasma (AIP) reflect the balance between atherogenic and protective lipoproteins and may provide insight into systemic metabolic risk profiles[9].

Inflammatory markers, including neutrophil-based ratios and composite indices such as the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and systemic immune-inflammation index (SII), have also been investigated in the context of colorectal neoplasia, with heterogeneous and often inconsistent results across studies[10-12]. While inflammation is recognized as a key contributor to colorectal tumorigenesis, the diagnostic and clinical relevance of these indices in adenoma detection remains uncertain[13].

Despite growing interest in metabolic and inflammatory markers, existing studies evaluating their association with colorectal adenomas are limited by small sample sizes, heterogeneous populations, and inconsistent analytical approaches[14]. Moreover, the comparative diagnostic performance of multiple metabolic and inflammatory indices within a single, large colonoscopy-based cohort has not been sufficiently explored. In addition, previous studies have yielded conflicting results regarding whether these indices remain significantly associated with the presence of colorectal adenoma after adjustment for major demographic confounders.

Therefore, the present study aimed to comprehensively evaluate metabolic and inflammatory indices in a large cohort of patients undergoing colonoscopy, with a particular focus on their association with the presence of colorectal adenoma and comparative diagnostic performance. By analyzing routinely available laboratory parameters, this study sought to clarify the potential clinical utility and limitations of these indices in real-world colonoscopy practice.

MATERIALS AND METHODS

Study design and population

This retrospective cohort observational study was conducted at a tertiary referral center and included adult patients who underwent colonoscopy for screening, diagnostic, or surveillance purposes between January 2018 and December 2023. All colonoscopy procedures were performed by experienced gastroenterologists in accordance with standard clinical practice. Patients with colorectal malignancy, inflammatory bowel disease, prior colorectal surgery, undefined colonoscopy findings, or incomplete demographic records were excluded from the final analysis cohort.

Patients aged 18 years and older who had complete colonoscopic examination reaching the cecum and for whom histopathological evaluation of detected colorectal lesions was available were eligible for inclusion. Individuals were categorized into two groups on the basis of colonoscopic and histopathological findings: (1) Patients with colorectal adenomas; and (2) Patients without adenomatous lesions (non-adenoma group).

Demographic characteristics, including age and sex, as well as relevant clinical and laboratory data obtained within 30 days before colonoscopy, were retrospectively collected from electronic medical records. The study population represents a real-world colonoscopy cohort reflecting routine clinical practice.

Because of the retrospective cohort observational design, no a priori sample size calculation was performed. Instead, all eligible adult patients who underwent complete colonoscopy during the predefined study period and fulfilled the inclusion criteria were included in the analysis. Therefore, the final sample size was determined by the number of eligible patients available in the institutional database rather than by prospective sample size estimation.

Laboratory measurements and index definitions

Fasting venous blood samples obtained within 30 days before colonoscopy were retrospectively evaluated. In routine clinical practice at our center, metabolic and lipid parameters are typically assessed in patients undergoing colonoscopic evaluation, particularly in individuals with metabolic comorbidities or cardiovascular risk factors. Most laboratory measurements were performed before bowel preparation. Laboratory parameters, including fasting plasma glucose, triglycerides, total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein cholesterol, complete blood count, and C-reactive protein levels, were retrospectively retrieved from the hospital laboratory database.

The TyG index was calculated using the following formula: TyG index = ln [fasting triglyceride (mg/dL) × fasting glucose (mg/dL)/2].

The AIP was defined as the base-10 Logarithm of the ratio between triglyceride and HDL cholesterol concentrations: AIP = log10 [triglyceride (mg/dL)/HDL-cholesterol (mg/dL)].

Inflammatory indices were derived from complete blood count parameters. The neutrophil-to-HDL ratio (NHR) was calculated by dividing the absolute neutrophil count by the HDL cholesterol level. The NLR was calculated by dividing the absolute neutrophil count by the absolute lymphocyte count. The PLR was calculated as the ratio of platelet count to lymphocyte count. The SII was calculated according to the following formula: SII = platelet count × neutrophil count/lymphocyte count.

All laboratory measurements were performed using standardized automated analyzers in the hospital’s central laboratory, following internal quality control procedures. Because of the retrospective cohort study design, not all laboratory tests were available for every patient. Missing laboratory values reflected routine clinical practice, in which metabolic and lipid profiles were ordered according to the treating physician’s clinical judgment rather than a standardized research protocol. No imputation procedures were applied, and analyses were performed using complete-case data for each variable.

Colonoscopy and histopathological evaluation

All colonoscopy procedures were performed using standard video colonoscopes by experienced gastroenterologists under routine clinical practice conditions. Bowel preparation quality was assessed during colonoscopy according to routine endoscopic evaluation, and only examinations with adequate bowel cleansing allowing complete mucosal visualization and cecal intubation were included in the analyses.

The size, number, and anatomical location of the detected colorectal polyps were recorded. All the identified lesions were removed or biopsied according to standard clinical practice and subsequently subjected to histopathological examination.

Histopathological evaluation was performed by experienced gastrointestinal pathologists who were blinded to the patients’ laboratory and clinical data. Colorectal adenomas were defined on the basis of established histological criteria which included tubular, tubulovillous, and villous adenomas. Hyperplastic polyps and other nonneoplastic lesions were classified as non-adenomatous findings. Sessile serrated lesions (SSLs) and traditional serrated adenomas, when identifiable in the retrospective records, were excluded from the final analyses because of their distinct biological behavior and malignant potential.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics software (version 28.0; IBM Corp., Armonk, NY, United States). Continuous variables were assessed for normality using the Kolmogorov-Smirnov test. Normally distributed continuous variables are expressed as the mean ± SD, whereas nonnormally distributed variables are presented as the median (interquartile range). Categorical variables are summarized as n (%).

Comparisons between patients with colorectal adenomas and those without adenomatous lesions were performed using the independent samples t-test or Mann-Whitney U test for continuous variables, as appropriate. Categorical variables were compared using the χ² test or Fisher’s exact test when applicable. Owing to the retrospective cohort design, the availability of laboratory parameters varied among patients. The available sample sizes for each variable are presented in the corresponding tables, and analyses were performed using complete-case data for each parameter.

To evaluate the diagnostic performance of metabolic and inflammatory indices for determining the presence of colorectal adenoma, receiver operating characteristic (ROC) curve analyses were conducted, and the area under the curve (AUC) values with 95%CI were calculated. Optimal cut-off values were determined using the Youden index.

Univariable and age-adjusted and sex-adjusted logistic regression analyses were performed to assess the association between metabolic and inflammatory indices and the presence of colorectal adenomas. The results are reported as odds ratios with 95%CIs.

To minimize the potential for over-adjustment, the individual laboratory parameters used to derive the composite metabolic indices were not included as additional covariates in the age-adjusted and sex-adjusted regression analyses evaluating those indices. Accordingly, adjustment was limited to the predefined demographic confounders (age and sex).

All the statistical tests were two-sided, and a P value < 0.05 was considered to indicate statistical significance.

RESULTS

Study population and colonoscopy findings

A total of 1603 individuals who underwent colonoscopy were initially evaluated. Patients with colorectal malignancy, inflammatory bowel disease, prior colorectal surgery, undefined colonoscopy findings, or incomplete demographic records were excluded from the final analyses. After the predefined exclusion criteria were applied, 1570 individuals constituted the final analytic cohort for adenoma analyses. Owing to the retrospective cohort design, laboratory parameters were not available for all patients. Therefore, analyses were performed using complete-case data for each variable, and the corresponding sample sizes are presented in the relevant tables.

The median age of the overall cohort was 61 years (interquartile range, 51-70), and 49.2% (n = 773) of the patients were male. Normal colonoscopy findings were observed in 995 patients (63.4%). Nonneoplastic colonoscopy findings included diverticulosis in 52 patients (3.3%) and benign ulceration in 33 patients (2.1%).

Overall, 395 patients (25.2%) were diagnosed with at least one adenomatous polyp. Among patients with adenomatous polyps, tubular adenomas were the most frequently specified histological subtype, followed by tubulovillous and villous adenomas; the adenoma subtype could not be specified in 35 patients because of incomplete retrospective histopathological detail. Non-adenomatous polyps included hyperplastic (4.9%) and inflammatory (1.8%) polyps. The detailed colonoscopic findings and histopathological distributions of the study population are summarized in Table 1. Detailed grading of advanced dysplastic features was not consistently available because of the retrospective cohort design.

Table 1 Baseline characteristics and colonoscopy findings of the study population, median (interquartile range)/n (%).

Variable	Study population (n = 1570)
Age (years)	61 (51-70)
Sex
Male	773 (49.2)
Female	797 (50.8)
Colonoscopy findings
Normal colonoscopy	995 (63.4)
Adenoma present	395 (25.2)
Diverticulosis	52 (3.3)
Benign ulceration	33 (2.1)
Polyp histology
Tubular adenoma	306 (19.5)
Tubulovillous adenoma	47 (3.0)
Villous adenoma	7 (0.4)
Adenoma, subtype not specified	35 (2.2)
Hyperplastic polyp	78 (5.0)
Inflammatory polyp	29 (1.8)
Polyp localization
Rectum	90 (5.6)
Sigmoid colon	106 (6.6)
Descending colon	52 (3.2)
Splenic flexure	7 (0.4)
Transverse colon	63 (3.9)
Ascending colon	35 (2.2)
Cecum	18 (1.1)
Hepatic flexure	4 (0.2)

Percentages were calculated based on the final analytic cohort. Patients with colorectal malignancy, inflammatory bowel disease, prior colorectal surgery, undefined colonoscopy findings, and invalid demographic records were excluded before final analyses. Due to the retrospective cohort design, some patients had multiple non-neoplastic colonoscopic findings. Polyp localization frequencies may exceed the number of adenoma patients because some individuals had multiple lesions involving different colonic segments.

Comparisons between adenoma and non-adenoma groups

Patients with adenoma were significantly older than those without adenoma (P < 0.001) and were more frequently male (61.6% vs 44.3%, P < 0.001).

In terms of metabolic and lipid parameters, patients with adenoma had significantly higher serum glucose, triglyceride, and low-density lipoprotein cholesterol levels, as well as higher TyG index and AIP values, compared with the non-adenoma group (all P < 0.001). In contrast, HDL cholesterol levels were significantly lower in patients with adenoma (P < 0.001). Detailed comparisons of metabolic and lipid parameters are presented in Table 2. Because of the retrospective cohort design, laboratory availability differed across variables, resulting in varying sample sizes for specific analyses. Therefore, the denominators used in each comparison were based on the number of patients with available laboratory data for the corresponding parameter.

Table 2 Comparison of metabolic and lipid parameters according to adenoma status.

Variable	Non-adenoma group (n)	Adenoma group (n)	Mean rank (non-adenoma)	Mean rank (adenoma)	Z value	Effect size (r)	P value
Glucose (mg/dL)	1239	351	728.60	1031.66	-10.920	0.274	< 0.001
Triglycerides (mg/dL)	1208	337	715.33	979.73	-9.620	0.245	< 0.001
HDL cholesterol (mg/dL)	1207	337	801.50	668.63	-4.839	0.123	< 0.001
LDL cholesterol (mg/dL)	1206	337	739.07	889.86	-5.493	0.140	< 0.001
TyG index	1208	337	701.16	1030.52	-11.984	0.305	< 0.001
AIP	1207	337	717.28	970.26	-9.209	0.234	< 0.001

Continuous variables were compared using the Mann-Whitney U test. Higher mean rank values indicate higher levels of the corresponding variable. Effect size r was calculated as |Z|/√n. Analyses were performed using complete-case data for each parameter because laboratory availability varied across patients in this retrospective cohort. TyG: Triglyceride-glucose; AIP: Atherogenic index of plasma; LDL: Low-density lipoprotein; HDL: High-density lipoprotein.

With respect to hematological and inflammatory parameters, neutrophil counts and the NHR were modestly but significantly greater in the adenoma group (P = 0.001 and P < 0.001, respectively). The lymphocyte count, platelet count, NLR, PLR, and SII did not significantly differ between the groups (all P > 0.05). These findings are summarized in Table 3.

Table 3 Comparison of hematological and inflammatory parameters according to adenoma status.

Variable	Non-adenoma group (n)	Adenoma group (n)	Mean rank (non-adenoma)	Mean rank (adenoma)	Z value	Effect size (r)	P value
Neutrophils (103/μL)	1244	352	778.18	870.32	-3.312	0.083	0.001
Lymphocytes (103/μL)	1244	352	789.29	831.04	-1.501	0.038	0.133
Platelets (103/μL)	1244	351	803.07	780.04	-0.827	0.021	0.408
NLR	1244	352	789.85	829.06	-1.409	0.035	0.159
PLR	1244	352	809.36	760.10	-1.770	0.044	0.077
SII	1244	352	794.11	814.01	-0.715	0.018	0.475
NHR	1207	337	740.85	885.86	-5.279	0.134	< 0.001

Continuous variables were compared using the Mann-Whitney U test. Higher mean rank values indicate higher levels of the corresponding variable. Effect size r was calculated as |Z|/√n. Analyses were performed using complete-case data for each parameter because laboratory availability varied across patients in this retrospective cohort. NLR: Neutrophil-to-lymphocyte ratio; PLR: Platelet-to-lymphocyte ratio; SII: Systemic immune-inflammation index; NHR: Neutrophil-to-high-density lipoprotein ratio.

ROC analyses for predicting adenoma presence

ROC curve analyses were performed to evaluate the discriminatory performance of metabolic and inflammatory parameters for the presence of adenoma. Among the evaluated markers, the TyG index demonstrated the highest discriminatory performance, with an AUC of 0.714 (95%CI: 0.684-0.744; P < 0.001). Serum glucose also showed moderate discriminatory performance (AUC = 0.696; 95%CI: 0.666-0.726; P < 0.001), followed by triglycerides (AUC = 0.671; P < 0.001) and the AIP (AUC = 0.664; P < 0.001). HDL cholesterol was inversely associated with adenoma presence (AUC = 0.414; P < 0.001).

The discriminatory performance of inflammatory markers was limited. Among the inflammatory parameters, the NHR demonstrated the strongest discriminatory ability, although the overall performance remained modest. The neutrophil count yielded a modest AUC (AUC = 0.560; P = 0.001), whereas the NHR demonstrated a moderate association (AUC = 0.593; P < 0.001). Other inflammatory indices, including the NLR, PLR, and SII, did not show significant discriminatory ability (all AUCs were close to 0.50).

Using the Youden index, the optimal cut-off value for the TyG index was 8.648, corresponding to a sensitivity of 81.0% and a specificity of 55.0%. For serum glucose, the optimal cut-off was 97.5 mg/dL (sensitivity 65.5% and specificity 68.0%), and for AIP, it was 0.496 (sensitivity 60.7% and specificity 69.0%). The detailed ROC performance metrics are presented in Table 4, and the ROC curves for the TyG index and AIP are shown in Figure 1.

Figure 1 Receiver operating characteristic curves of the triglyceride-glucose index and the atherogenic index of plasma for discriminating colorectal adenoma presence. The triglyceride-glucose index demonstrated the highest discriminatory performance (area under the curve = 0.714), followed by the atherogenic index of plasma (area under the curve = 0.664). Optimal cut-off values were determined using the Youden index. ROC: Receiver operating characteristic; TyG: Triglyceride-glucose; AIP: Atherogenic index of plasma.

Table 4 Receiver operating characteristic analysis of metabolic and inflammatory markers for adenoma detection.

Variable	AUC	95%CI	P value	Optimal cut-off (Youden)	Sensitivity (%)	Specificity (%)
TyG index	0.714	0.684-0.744	< 0.001	8.648	81.0	55.0
Glucose (mg/dL)	0.696	0.666-0.726	< 0.001	97.5	65.5	68.0
Triglycerides (mg/dL)	0.671	0.640-0.702	< 0.001	142.5	63.0	64.2
AIP	0.664	0.633-0.695	< 0.001	0.496	60.7	69.0
HDL cholesterol (mg/dL)	0.414	0.384-0.445	< 0.001	44.5	62.0	41.8
Neutrophils (103/μL)	0.560	0.529-0.591	0.001	4.65	58.0	54.0
NHR	0.593	0.562-0.624	< 0.001	0.087	61.5	56.0
NLR	0.517	0.486-0.548	0.266	-	-	-
PLR	0.522	0.491-0.553	0.177	-	-	-
SII	0.508	0.477-0.539	0.594	-	-	-

Receiver operating characteristic analyses were performed to evaluate the ability of each marker to discriminate adenoma presence. Optimal cut-off values were determined using the Youden index. Analyses were performed using complete-case data for each parameter because laboratory availability varied across patients in this retrospective cohort. Cut-off values were not calculated for parameters without statistically significant receiver operating characteristic discrimination. Because high-density lipoprotein cholesterol was inversely associated with colorectal adenoma, an area under the curve below 0.50 reflects the direction of this association rather than poor discriminatory performance. AUC: Area under the curve; TyG: Triglyceride-glucose; AIP: Atherogenic index of plasma; HDL: High-density lipoprotein; NHR: Neutrophil-to-high-density lipoprotein ratio; NLR: Neutrophil-to-lymphocyte ratio; PLR: Platelet-to-lymphocyte ratio; SII: Systemic immune-inflammation index.

Age-adjusted and sex-adjusted logistic regression analysis

To evaluate whether metabolic and inflammatory indices were associated with adenoma presence after adjusting for age and sex, age-adjusted and sex-adjusted logistic regression analyses were performed. In the adjusted models, the TyG index and AIP remained significantly associated with adenoma presence after adjusting for age and sex. Inflammatory indices, including the NHR, NLR, PLR, and SII, did not retain statistical significance in the adjusted analyses. The results of the age- and sex-adjusted logistic regression analyses are presented in Table 5.

Table 5 Age-adjusted and sex-adjusted logistic regression analyses for adenoma presence.

Variable	Adjusted OR	95%CI	P value
TyG index	1.32	1.10-1.58	0.003
AIP	1.24	1.05-1.47	0.011
NLR	1.05	0.94-1.18	0.340
PLR	0.96	0.86-1.08	0.470
SII	1.04	0.93-1.16	0.480
NHR	1.12	0.99-1.27	0.071

Logistic regression models were adjusted for age and sex. Odds ratios with 95%CIs are presented per 1 standard deviation increase of each variable. TyG: Triglyceride-glucose; AIP: Atherogenic index of plasma; NLR: Neutrophil-to-lymphocyte ratio; PLR: Platelet-to-lymphocyte ratio; SII: Systemic immune-inflammation index; NHR: Neutrophil-to-high-density lipoprotein ratio; OR: Odds ratio.

DISCUSSION

In this large colonoscopy-based cohort, metabolic indices such as the TyG index and the AIP were significantly greater in patients with colorectal adenomas in unadjusted analyses. Importantly, the TyG index and AIP remained significantly associated with adenoma presence after adjustment for age and sex. However, given the retrospective cohort observational design, these findings should be interpreted as associations rather than evidence of a causal relationship. Overall, our results suggest that metabolic indices, particularly the TyG index and AIP, may have potential clinical relevance in identifying individuals at increased risk for colorectal adenomas, whereas most systemic inflammatory indices appear to have more limited predictive value.

The association between elevated metabolic indices and the presence of colorectal adenoma observed in this study is biologically plausible and consistent with prior reports linking insulin resistance and dyslipidemia to early colorectal neoplasia. Given that colorectal adenomas represent precursor lesions in the adenoma-carcinoma sequence, the observed associations between metabolic dysregulation and the presence of adenoma may reflect early biological processes involved in colorectal cancer development. Recent evidence has also highlighted the role of metabolic syndrome and systemic inflammation in early onset colorectal cancer, further supporting the relevance of these pathways. Insulin resistance has been implicated in adenoma development through hyperinsulinemia, increased insulin-like growth factor signaling, chronic low-grade inflammation, and dysregulation of cellular proliferation and apoptosis pathways[5,6,15]. In this context, the TyG index has emerged as a practical surrogate marker of insulin resistance and has been associated with a range of metabolic and gastrointestinal disorders[7,8]. It is also possible that the observed associations reflect shared upstream mechanisms, including obesity, insulin resistance, dietary patterns, or other metabolic disturbances, rather than direct biological effects of the metabolic indices themselves. Therefore, the present findings should be interpreted as evidence of association rather than causation, and prospective studies are warranted to clarify the underlying biological mechanisms and causal pathways.

Similarly, lipid-related abnormalities characterized by elevated triglyceride levels and reduced HDL cholesterol have been linked to colorectal adenoma formation. The AIP, which integrates these lipid parameters, reflects an adverse metabolic milieu that may facilitate colorectal neoplastic processes[9,16]. Nevertheless, studies evaluating metabolic markers in relation to colorectal adenomas have frequently been limited by small sample sizes and heterogeneous study designs[14], highlighting the importance of large, real-world colonoscopy-based cohorts to better delineate these associations.

Although systemic inflammatory indices such as the NLR, SII, and related composite markers have been extensively investigated in colorectal neoplasia, their role in colorectal adenoma detection remains incompletely defined. Previous studies have suggested that low-grade systemic inflammation may contribute to early colorectal tumorigenesis through immune dysregulation and alterations in the tumor microenvironment; however, evidence supporting a consistent association between inflammatory indices and the presence of adenoma is limited and heterogeneous[10-12,17].

In screening- and colonoscopy-based populations, inflammatory markers are influenced by multiple confounding factors, including metabolic status, obesity, insulin resistance, subclinical infections, and comorbid inflammatory conditions, which may partly explain the inconsistent findings reported across studies[18]. The persistence of significant associations for the TyG index and AIP after adjustment further supports the hypothesis that metabolic dysfunction may play a more prominent role in early colorectal neoplasia than generalized systemic inflammatory responses do. These considerations likely account for the modest and variable associations observed for inflammatory indices in the present study.

From a clinical perspective, the present findings suggest that metabolic indices, particularly the TyG index and AIP, may contribute to colorectal adenoma risk stratification, although their discriminatory performance remains insufficient for standalone diagnostic use. Accordingly, these markers should not replace colonoscopy but may provide complementary information when interpreted alongside established clinical risk factors and screening tools. In the future, metabolic and inflammatory indices could be incorporated into composite risk prediction models, including fecal immunochemical testing, although this approach requires validation in large prospective studies[19,20].

The strengths of this study include its relatively large sample size, colonoscopy-based design, and comprehensive evaluation of multiple metabolic and inflammatory indices within a real-world clinical cohort. In addition, the use of a rigorously cleaned patient-based analytic cohort and standardized exclusion criteria strengthened the internal consistency of the analyses. The simultaneous assessment of several routinely available laboratory markers allowed for a pragmatic comparison of their associations with colorectal adenomas in everyday practice. Nevertheless, several limitations should be acknowledged. Although identifiable sessile serrated lesions and traditional serrated adenomas were excluded from the analyses, incomplete retrospective characterization of serrated lesions may still represent a limitation. The retrospective cohort and single-center nature of the study may limit causal inference and generalizability to other populations. The inclusion of patients who underwent colonoscopy for different clinical indications (screening, diagnostic evaluation, and surveillance) may have introduced selection bias. Unfortunately, because the specific indication for colonoscopy was not consistently available in the retrospective database, subgroup analyses according to colonoscopy indication could not be performed. Although this limitation precluded subgroup analyses according to colonoscopy indication, the inclusion of a heterogeneous real-world population may enhance the generalizability of our findings to routine clinical practice. Future prospective studies with systematically recorded colonoscopy indications should further evaluate whether the observed associations differ across clinical subgroups. Potential residual confounding factors, including smoking status, body mass index, lifestyle-related variables, and medication use, particularly lipid-lowering therapies, could not be fully accounted for because of the retrospective cohort design. Additionally, the timing of blood sample collection in relation to bowel preparation could not be fully standardized because of the retrospective cohort design, which may have influenced certain laboratory parameters. Furthermore, laboratory parameters were assessed at a single time point, which may not reflect longitudinal metabolic or inflammatory changes. Despite these limitations, our results provide clinically relevant insight into the relationship between metabolic dysregulation and the presence of colorectal adenoma, particularly by demonstrating significant associations for the TyG index and AIP after adjusting for age and sex.

CONCLUSION

Metabolic indices, particularly the TyG index and AIP, are significantly associated with the presence of colorectal adenoma, although their standalone diagnostic performance remains limited. These findings support the potential role of metabolic dysregulation in early colorectal neoplasia and suggest that metabolic indices may contribute to future risk stratification strategies when interpreted alongside established clinical and endoscopic approaches.