Li YQ, Kuai WT, Chen L, Zeng MH, Tao XM, Han JX, Wang YK, Xu LX, Ge LY, Liu YG, Li S, Xu L, Mi YQ. Predicting colorectal adenomatous polyps in patients with chronic liver disease: A novel nomogram. World J Gastroenterol 2025; 31(2): 99082 [DOI: 10.3748/wjg.v31.i2.99082]
Corresponding Author of This Article
Yu-Qiang Mi, MD, Professor, Department of Hepatology, Tianjin Second People’s Hospital, No. 7 Sudi South Road, Xuefu Street, Nankai District, Tianjin 300192, China. yuqiangmi68@163.com
Research Domain of This Article
Infectious Diseases
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Co-corresponding authors: Liang Xu and Yu-Qiang Mi.
Author contributions: Li YQ and Kuai WT conducted formal analysis and investigation, they contributed equally to this article, they are the co-first authors of this manuscript; Li YQ wrote the original draft; Mi YQ, Xu L, and Li S designed the study and acquired funding; Li YQ and Xu L developed the methodology; Mi YQ and Xu L contributed equally to this article as the co-corresponding authors; Chen L, Zeng MH, Tao XM, Wang YK, and Han JX participated in review and editing; Li YQ, Xu LX, and Ge LY collected the data; Liu YG provided pathological guidance for colorectal polyps; Li S provided endoscopic guidance.
Supported by the National Natural Science Foundation of China, No. 62375202; Natural Science Foundation of Tianjin, No. 23JCYBJC00950; Tianjin Health Science and Technology Project Key Discipline Special, No. TJWJ2022XK034; and Research Project in Key Areas of Traditional Chinese Medicine in 2024, No. 2024022.
Institutional review board statement: This study has been reviewed and approved by the Ethics Committee of Tianjin Second People’s Hospital.
Informed consent statement: The requirement for written informed consent was not needed due to retrospective design of the study.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Data sharing statement: The data involved in this study can be obtained from the corresponding author.
Open-Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Yu-Qiang Mi, MD, Professor, Department of Hepatology, Tianjin Second People’s Hospital, No. 7 Sudi South Road, Xuefu Street, Nankai District, Tianjin 300192, China. yuqiangmi68@163.com
Received: July 15, 2024 Revised: October 29, 2024 Accepted: November 11, 2024 Published online: January 14, 2025 Processing time: 155 Days and 19.9 Hours
Abstract
BACKGROUND
Colorectal polyps are commonly observed in patients with chronic liver disease (CLD) and pose a significant clinical concern because of their potential for malignancy.
AIM
To explore the clinical characteristics of colorectal polyps in patients with CLD, a nomogram was established to predict the presence of adenomatous polyps (AP).
METHODS
Patients with CLD who underwent colonoscopy at Tianjin Second People’s Hospital from January 2020 to May 2023 were evaluated. Clinical data including laboratory results, colonoscopy findings, and pathology reports were collected. Key variables for the nomogram were identified through least absolute shrinkage and selection operator regression, followed by multivariate logistic regression. The performance of the model was evaluated using the area under the receiver area under curve, as well as calibration curves and decision curve analysis.
RESULTS
The study enrolled 870 participants who underwent colonoscopy, and the detection rate of AP in patients with CLD was 28.6%. Compared to individuals without polyps, six risk factors were identified as predictors for AP occurrence: Age, male sex, body mass index, alcohol consumption, overlapping metabolic dysfunction-associated steatotic liver disease, and serum ferritin levels. The novel nomogram (AP model) demonstrated an area under curve of 0.801 (95% confidence interval: 0.756-0.845) and 0.785 (95% confidence interval: 0.712-0.858) in the training and validation groups. Calibration curves indicated good agreement among predicted and actual probabilities (training: χ2 = 11.860, P = 0.157; validation: χ2 = 7.055, P = 0.530). The decision curve analysis underscored the clinical utility of the nomogram for predicting the risk of AP.
CONCLUSION
The AP model showed reasonable accuracy and provided a clinical foundation for predicting the occurrence of AP in patients with CLD, which has a certain predictive value.
Core Tip: This study is the first effort to develop and validate a predictive model for assessing the risk of adenomatous polyps in patients with chronic liver disease. Based on six risk factors, the developed adenomatous polyps model will function as an effective means of identifying and stratifying patients undergoing management for chronic liver disease, thereby helping to reduce the risk of colorectal cancer in these individuals.
Citation: Li YQ, Kuai WT, Chen L, Zeng MH, Tao XM, Han JX, Wang YK, Xu LX, Ge LY, Liu YG, Li S, Xu L, Mi YQ. Predicting colorectal adenomatous polyps in patients with chronic liver disease: A novel nomogram. World J Gastroenterol 2025; 31(2): 99082
Chronic liver disease (CLD) affects an estimated 1.5 billion individuals globally, necessitating diverse healthcare interventions owing to its multi-organ and systemic impact. Although the prevalence of viral hepatitis has been decreasing in high-income nations, CLD remains a significant healthcare burden[1,2]. Non-alcoholic fatty liver disease (NAFLD) has a worldwide prevalence of 30%, and has emerged as a major contributor to liver-related morbidity and mortality[3]. Colorectal cancer (CRC) is a highly malignant cancer of the gastrointestinal tract. As reported in the World Health Organization’s 2020 Global Cancer Statistics, CRC ranks as the third most common cancer globally and is the second leading cause of cancer-related deaths[4]. The development of CRC occurs through a well-established “adenoma-cancer” sequence in which precursor lesions, primarily adenomatous colon polyps, undergo genetic and epigenetic alterations over a period of 10-15 years before progressing to CRC[5]. Potential risk factors for colorectal polyps include age, sex, smoking, alcohol consumption, high-fat diet, metabolic syndrome (MS), and gallbladder disease[6-8]. CLD has been reported to promote lesions in colorectal adenoma-carcinoma sequence[9]. Primary sclerosing cholangitis, NAFLD, and viral liver disease are linked to an increased likelihood of colorectal lesions, particularly adenomas and hyperplastic polyps[10-13]. However, the relationship between the contributing factors for colorectal polyps in patients with CLD and the potential impact of different CLD etiologies on colorectal polyp detection rates remains poorly understood. Effective predictive models for adenomatous polyps (AP) in patients with CLD are unavailable. Therefore, this study aimed to identify the risk factors for colorectal polyps in this population more efficiently to reduce the need for invasive colonoscopies.
MATERIALS AND METHODS
Study population
The study included participants recruited from individuals undergoing colonoscopy at Tianjin Second People’s Hospital from January 2020 to May 2023. Initially, 1693 patients with CLD were enrolled. The exclusion criteria were as follows: 12 patients were excluded due to age (either aged < 18 years or > 80 years), 468 patients due to poor bowel preparation and incomplete pathological analysis, 115 patients with familial adenomatous polyposis or inflammatory bowel disease, and 86 patients due to a history of diseases such as acquired immunodeficiency syndrome, tuberculosis, and syphilis. In addition, 98 patients had incomplete medical records and laboratory tests; 20 were excluded because of a family history of CRC in first-degree relatives, and 24 were excluded because of unexplained liver disease or an unclear diagnosis. Finally, 870 patients were enrolled in the study: 347 without polyps, 274 with non-AP, and 249 with AP. Figure 1 illustrates the flowchart of the study design.
Figure 1 Flowchart of this study.
AP: Adenomatous polyps.
Data collection
Data regarding age, sex, smoking status, alcohol consumption, body mass index (BMI), hypertension, diabetes, and MS were obtained from the patients’ medical records. Smoking was categorized as the intake of at least one cigarette daily for a minimum of one year, and alcohol consumption was defined as having at least one alcoholic drink per week over the course of one year. Blood samples were obtained from the participants after an eight-hour fasting period via venipuncture of the antecubital vein. The samples were analyzed using an automatic biochemical analyzer (Model 7180; HITACHI, Japan) to assess a range of biochemical parameters. All analyses were conducted in a single laboratory to ensure the methodological consistency and reliability of the results. The biochemical parameters evaluated comprised total high-density lipoprotein (HDL), low-density lipoprotein, cholesterol, triacylglycerol, fasting plasma glucose (FPG), serum total protein, serum albumin (ALB), platelet count, determine serum iron, serum ferritin (SF), transferrin, serum thyroid stimulating hormone (TSH), free T4, free T3, gamma-glutamyl transferase (GGT), aspartate aminotransferase, and alanine aminotransferase.
Liver disease etiology diagnosis
Population demographics and clinical data were extracted from the medical records of hospitals using standard protocols. All patients with CLD underwent routine examinations to identify common etiologies, such as hepatitis C virus (HCV) infection, hepatitis B virus (HBV) infection, alcoholic liver disease (ALD), autoimmune hepatitis (AIH), primary biliary cholangitis (PBC), and drug-induced liver injury (DILI). Patients with HBV or HCV were identified through hepatitis B surface antigen testing and positive serum anti-HCV antibodies (confirmed by HCV RNA)[14,15]. ALD was defined as having a long-term history of alcohol consumption, typically exceeding 5 years, with an equivalent ethanol intake of ≥ 40 g/day for men and ≥ 20 g/day for women, or a history of heavy drinking within 2 weeks with an equivalent ethanol intake of > 80 g/day[16]. AIH was diagnosed based on simplified criteria that utilized autoimmune markers and liver biopsy[17]. PBC was diagnosed based on cholestatic liver biochemistry, autoimmune antibody reactivity, and/or liver biopsy[18]. DILI was diagnosed using the Roussel Uclaf Causality Assessment Method scale and/or liver biopsy[19]. Experienced radiologists performed abdominal ultrasonography using ultrasound scanners. The diagnosis of fatty liver disease was based on the presence of an enhanced hepatorenal echo, characterized by differences in echo patterns between the liver and kidney parenchyma, attenuation of the deep ultrasound beam, blurring of vascular structures, and a straightened hepatic vein lumen[20]. Metabolic dysfunction-associated steatotic liver disease (MASLD) is characterized by the presence of hepatic steatosis alongside a cardiometabolic risk factor, without any identifiable cause: (1) BMI ≥ 23 kg/m2; (2) Fasting serum glucose ≥ 5.6 mmol/L or 2-hour post-load glucose level ≥ 7.8 mmol/L or glycated hemoglobin ≥ 5.7% or on specific drug treatment; (3) Blood pressure ≥ 130/85 mmHg or specific drug treatment; (4) Plasma triglycerides ≥ 1.70 mmol/L or specific drug treatment; and (5) Plasma HDL < 1.0 mmol/L for men and < 1.3 mmol/L for women or specific drug treatment[21].
Diagnosis and classification of colorectal polyp
Experienced endoscopists performed the colonoscopies, and bowel preparation was performed on all study participants by adept gastroenterologists before the colonoscopies, utilizing a standardized protocol with an OLYMPUS equipment (Tokyo, Japan), the procedure aligned with the protocols for diagnostic colonoscopies[22]. The number of polyps, including both single and multiple polyps, were recorded. After detection during colonoscopy, all abnormalities were subjected to biopsy, with pathological confirmation conducted in accordance with contemporary clinical guidelines, specifically the World Health Organization Classification of Digestive Tumors: Fifth Edition[23]. Pathological diagnosis: Pathological assessments were performed by two seasoned pathologists, supported by at least one chief pathologist, to ensure rigorous examination and interpretation. Immunohistochemistry was used to stain the pathological specimens. Pathological findings from these examinations led to the categorization of polyps into adenomatous and non-adenomatous types based on their histological characteristics.
Statistical analysis
Statistical analyses were conducted using SPSS version 25.0 software. For the groups without polyps, AP, and non-AP, variables that adhered to or closely approximated a normal distribution were compared using one-way analysis of variance. Variables that did not satisfy the normality assumption were assessed using the Kruskal-Wallis test. Post-hoc comparisons were conducted using the least significant difference method and Tamhane’s T2 tests. For the training and validation groups, normally distributed continuous data were analyzed with the t-test, whereas non-normally distributed data were evaluated using the Mann-Whitney U test. Descriptive statistics are presented as mean ± SD for variables with normal distribution and as median with interquartile range (25th-75th percentile) for those with non-normal distribution. All count data are presented as the number of subjects and percentages, with comparisons made using the χ2 test. Bonferroni correction was implemented as multiple pairwise comparisons where necessary. This predictive model was developed using R Software v.4.0.2. Initially, the least absolute shrinkage and selection operator regression method was used to filter variables and select predictors. Subsequently, a multivariate logistic regression was used for further variable selection. The nomogram was constructed using the “rms” package in R Software. Statistical significance was set at P < 0.05.
RESULTS
Baseline demographics and laboratory findings
A total of 870 participants were included in this study, of whom 347 (39.9%) were diagnosed with non-polyps, 249 (28.6%) with AP, and 274 (31.5%) with non-adenomatous (including hyperplastic and inflammatory) polyps. The cohort had an average age of 53.21 ± 10.67 years. The detection rate of AP in male patients was 32.9% (160/486), whereas that in female patients was 22.1% (85/384). Compared to the non-polyp group, the AP group showed a higher proportion of men, older age, elevated serum iron and SF levels, higher BMI, and greater proportions of alcohol consumption and smoking. Significant differences were found in hypertension, MS, and FPG, HDL, ALB, TSH, and GGT levels (P < 0.05). However, no significant differences were observed between non-adenomatous and AP regarding MS, FPG, ALB, and GGT. Additionally, no significant correlations were found among the three groups with respect to diabetes, liver cirrhosis, and cholesterol, triacylglycerol, low-density lipoprotein, platelet count, serum total protein, transferrin, free T4, free T3, alanine aminotransferase, and aspartate aminotransferase levels (P > 0.05) (Table 1).
Table 1 Clinical and laboratory data of patients with chronic liver disease.
Baseline features of the training and validation groups
In total, 596 eligible patients, including those in the non-polyp and AP groups, were enrolled in the AP cohort, which had an average age of 53.02 years. 249 patients were identified as having AP: 40.5% and 45.1% in the training and validation groups. No significant differences were found between the training and validation groups according to the statistical analysis (P > 0.05) (Table 2).
Table 2 Baseline characteristics of adenomatous polyp cohort in training and validation group.
Variables
Total (n = 596)
Training group (n = 432)
Validation group (n = 164)
χ2/t/Z
P value
Male, n (%)
324 (54.4)
90 (54.9)
234 (54.2)
0.024
0.876
Age (year), mean ± SD
53.02 ± 11.14
53.04 ± 11.24
52.96 ± 10.91
0.089
0.929
BMI (kg/m²), mean ± SD
25.66 ± 3.94
25.78 ± 3.94
25.31 ± 3.93
1.323
0.186
Alcohol, n (%)
179 (30.0)
127 (29.4)
52 (31.7)
0.302
0.583
Smoking, n (%)
193 (32.4)
145 (33.6)
48 (29.3)
1.002
0.317
Diabetes, n (%)
154 (25.8)
120 (27.8)
34 (20.7)
3.080
0.079
Hypertension, n (%)
165 (27.7)
120 (27.8)
45 (27.4)
0.007
0.934
MS, n (%)
96 (16.1)
69 (16.0)
27 (16.5)
0.021
0.884
Cirrhosis, n (%)
285 (47.8)
215 (49.8)
70 (42.7)
2.392
0.122
FPG (mmo/L), mean ± SD
6.15 ± 2.07
6.23 ± 2.16
5.93 ± 1.85
-1.573
0.116
HDL (mmo/L), mean ± SD
1.33 ± 0.40
1.33 ± 0.41
1.33 ± 0.38
0.105
0.916
CHO (mmo/L), median (IQR)
4.54 (3.82, 5.38)
4.57 (3.81, 5.40)
4.41 (3.85, 5.41)
-0.714
0.458
TG (mmo/L), mean ± SD
1.42 ± 1.12
1.44 ± 1.22
1.35 ± 0.82
-0.826
0.409
LDL (mmo/L), mean ± SD
2.81 ± 0.90
2.75 ± 0.82
2.83 ± 0.93
-0.979
0.382
TP (g/L), mean ± SD
71.54 ± 7.98
71.29 ± 7.76
72.19 ± 8.51
1.239
0.216
ALB (g/L), mean ± SD
40.75 ± 6.14
40.83 ± 6.07
40.52 ± 6.34
-0.551
0.582
SI (μmol/L), mean ± SD
22.36 ± 10.67
22.37 ± 10.15
22.33 ± 11.97
-0.040
0.968
SF (ug/L), median (IQR)
165.00 (70.25, 303.25)
163.00 (76.00, 304.00)
174.00 (54.50, 295.00)
-0.721
0.471
TRF (g/L), mean ± SD
2.75 ± 0.89
2.75 ± 0.91
2.76 ± 0.83
0.203
0.840
TSH (mIU/L), median (IQR)
1.85 (1.33, 2.83)
1.94 (1.30, 2.89)
1.72 (1.34, 2.66)
-1.435
0.151
FT4 (pmol/L), mean ± SD
17.10 ± 14.18
17.31 ± 9.77
16.57 ± 9.62
-0.698
0.485
FT3 (pmol/L), mean ± SD
4.98 ± 1.00
5.01 ± 1.02
4.90 ± 0.94
-1.245
0.214
ALT (IU/L), median (IQR)
33.20 (19.40, 78.57)
32.20 (19.10, 76.57)
34.75 (19.65, 79.25)
-0.610
0.542
AST (IU/L), median (IQR)
33.75 (22.35, 72.47)
33.75 (22.02, 65.60)
34.10 (22.62, 78.75)
-0.491
0.624
GGT (IU/L), median (IQR)
53.85 (24.35, 116.77)
54.55 (24.97, 100.57)
50.05 (22.07, 163.45)
-0.274
0.708
Development of prediction AP model
Predictive factors for the identification of AP were explored. Variables were collected from the patients. Initially, the collected clinical data, including demographics, laboratory data, and liver etiology variables were incorporated. Least absolute shrinkage and selection operator regression analysis identified nine potential predictive factors with non-zero coefficients: Age, male sex, BMI, alcohol consumption, and overlapping MASLD, SF, ALB, and TSH levels (Figure 2). Subsequently, these nine predictive factors were examined using a multivariate logistic regression. Compared to individuals without polyps, age [odds ratio (OR) = 1.069, 95% confidence interval (CI): 1.044-1.094, P < 0.001], male sex (OR = 3.217, 95%CI: 1.902-5.444, P < 0.001), BMI (OR = 1.098, 95%CI: 1.025-1.175, P = 0.008), history of alcohol consumption (OR = 1.956, 95%CI: 1.065-3.593, P= 0.031), overlapping MASLD (OR = 2.301, 95%CI: 1.384-3.826, P = 0.008), and SF (OR = 1.001, 95%CI: 1.000-1.010, P = 0.047) were identified as independent risk factors for AP (Table 3). The regression equation of the model was derived as logit, AP model = -6.188 + 0.066 age + 1.16 male sex + 0.093 BMI + 0.067 alcohol consumption + 0.833 other liver diseases (OLDs)-MASLD + 0.001 SF.
Figure 2 Least absolute shrinkage and selection operator regression analysis identified potential predictive factors.
A: The least absolute shrinkage and selection operator model was optimized by selecting parameters based on the criteria method, which added one standard error to the minimum error. This process utilized five-fold cross-validation, resulting in λ = 0.03902279 and log(λ) = -3.241, satisfying the one standard error criteria for a simpler and more regularized model; B: The regression coefficients varied as λ increased, smaller λ values were associated with larger coefficients for most features, while larger λ values resulted in coefficients being reduced towards zero or reaching zero, a total of nine variables with non-zero coefficients were identified.
Table 3 The multivariate logistic regression analyses of the risk for model.
Variables
B
Standard error
Wals
P value
Expect (B)
95%CI
Age
0.066
0.012
30.283
< 0.001
1.069
1.044-1.094
Male
1.169
0.268
8.971
< 0.001
3.217
1.902-5.444
BMI
0.093
0.035
7.113
0.008
1.098
1.025-1.175
Alcohol consumption
0.671
0.310
4.677
0.031
1.956
1.065-3.593
OLDs-MASLD
0.833
0.259
10.312
0.001
2.301
1.384-3.826
SF
0.001
0.001
2.627
0.047
1.001
1.000-1.010
Smoking
0.311
0.293
1.129
0.288
1.365
0.769-2.423
ALB
0.033
0.021
2.499
0.114
0.968
0.929-1.008
TSH
0.160
0.083
3.738
0.053
0.852
0.724-1.002
Validation of prediction AP model
The above-mentioned equation was used to construct a column chart, providing a user-friendly and simple nomogram for predicting AP probability (Figure 3A). Each variable was represented on its respective axis to assess the risk of developing AP in patients with CLD. Vertical lines were extended from each value on the upper scale to identify the corresponding scores. These scores were then aggregated and mapped onto a total-score scale. Ultimately, the total score was projected vertically onto the lower axis to ascertain the personalized risk for AP. To determine the discriminatory power of the predictive AP model, in the training group, the area under curve (AUC) was found to be 0.801 (95%CI: 0.756-0.845) (Figure 3B). The best specificity was 0.704, corresponding to a sensitivity of 0.791 and accuracy of 0.739. In the validation group, the AUC was 0.785 (95%CI: 0.712-0.858) (Figure 3C). Internal validation for different etiologies of liver disease showed the receiver operating characteristic curve results (Figure 3D). The AUC for patients with PBC reached its peak at 0.970 (95%CI: 0.964-0.990), while the AUC for HBV registered the lowest value at 0.769 (95%CI: 0.710-0.827).
Figure 3 Nomogram and performance metrics for predicting the risk of adenomatous polyps.
A: Nomogram; B: In the training group, the area under curve (AUC) was 0.801 (95% confidence interval: 0.756-0.854). The impact on AUC from age, gender, body mass index, alcohol consumption, serum ferritin, and overlapping metabolic dysfunction-associated steatotic liver disease was 0.64, 0.65, 0.61, 0.62, 0.62, and 0.60, respectively; C: In the validation group, the AUC was 0.785 (95% confidence interval: 0.712-0.858); D: Internal validation among chronic liver disease patients yielded the following AUCs: Hepatitis B virus: 0.769, hepatitis C virus: 0.800, alcoholic liver disease: 0.892, drug-induced liver injury: 0.812, primary biliary cholangitis: 0.970, autoimmune hepatitis: 0.900. BMI: Body mass index; OLDs-MASLD: Other liver diseases overlapping metabolic dysfunction associated steatotic liver disease; SF: Serum ferritin; AUC: Area under curve; CI: Confidence interval; HBV: Hepatitis B virus; HCV: Hepatitis C virus; ALD: Alcoholic liver disease; DILI: Drug-induced liver injury; PBC: Primary biliary cholangitis; AIH: Autoimmune hepatitis.
Evaluation of prediction AP model
Additionally, nomogram accuracy was evaluated using a calibration plot. The calibration curve indicated no significant difference in the comparison of the estimated and observed probabilities of AP occurrence in both the training and validation groups (χ2 = 11.860, P = 0.157) and validation groups (χ2 = 7.055, P = 0.530) (Figure 4A and B). To determine the clinical applicability of the AP model, a decision curve analysis was conducted, and net benefits were calculated based on the relative risks of false positives and false negatives, along with the difference between the proportions of individuals with true-positive results and those with false-positive results. The resultant curve demonstrated a significant net benefit for patients with AP, encompassing risk threshold probability ranges of 1.8%-80.2% in the training group and 1.3%-70.1% in the validation group (Figure 4C and D).
Figure 4 Calibration and decision curve analysis for predicting adenomatous polyps.
A: The training group (χ2 = 11.860, P = 0.157); B: The validation group (χ2 = 7.055, P = 0.530); C and D: Decision curve analysis for the adenomatous polyps (AP) nomogram, blue line and orange line corresponds to the risk nomogram results for the training group and validation group, respectively. Black line: The assumption of no patient having AP; gray line: Assumes that all patients experienced AP.
Variety in liver disease etiologies
Subsequent subgroup analysis, as illustrated in Figure 5, showed a significant enhancement in the detection rate of AP to 28.8% when MASLD coexisted with HBV infection, compared with 21.6% without MASLD (P < 0.001). Additionally, compared to OLDs and pure MASLD, the detection rate of AP in MASLD overlapping with OLDs was significantly higher (P < 0.001). In patients with ALD, the prevalence of AP was significantly higher than that of non-AP (61.1% vs 24.3%, P < 0.001). No significant statistical differences were observed in patients with HCV compared to those with DILI; However, the percentage of AP was greater in patients with concurrent MASLD than in those without MASLD. A statistical difference was observed in patients with AIH, potentially because only two patients with AIH without concurrent MASLD were included (Table 4).
This study explored the clinical characteristics of colorectal polyps and the detection rates of adenomatous and non-AP in patients with CLD. Additionally, a personalized prediction nomogram for incident AP in Chinese adults with CLD was developed and validated. The prediction model incorporated six parameters: Age, sex, BMI, alcohol consumption, OLDs-MASLD, and SF. Model assessment and internal validation consistently demonstrated strong predictive performance of the nomogram. First, we found that 28.6% of the participants were diagnosed with AP, whereas 31.5% were diagnosed with non-AP. The occurrence of colorectal polyps varies according to geographic location and ethnicity, exhibiting the highest levels in Western nations, while Africa and Southern Asia report the lowest rates. In Asian countries, a prevalence of approximately 20% is observed in populations with average-risk colorectal polyps[24,25], with older adults and men exhibiting higher prevalence rates. Yang et al[26] reported that the prevalence of colorectal adenomas among average-risk individuals was 34.5% in men and 20.0% in women, with an increasing trend observed over the study period. An earlier study from Germany indicated that 26% of patients with CLD had AP, which focused on alcoholic and viral liver diseases, including 73 cases (23.6%) of ALD and 63 cases (23.1%) of viral liver diseases[9]. This study included a notable number of patients with HBV-related liver diseases comprising 417 patients (47.9%). Differences in the distribution of liver disease etiologies between studies could potentially contribute to variations in the prevalence of AP.
Unhealthy lifestyle habits, such as smoking and alcohol consumption, are more likely to be associated with the presence of polyps in patients with CLD. In this study, the training group showed a 1.95-fold increased risk of AP in patients who consumed alcohol than those without polyps. Bailie et al[8] reported a 2.5-fold increased risk of serrated polyps in smokers and a 1.33-fold increased risk in alcohol consumers (95%CI: 1.17-1.52). The underlying mechanisms are unclear; however, the inflammatory stimulus of alcohol may contribute to the development of polyps. Long-term feeding of mice with ethanol has been shown to enhance inflammation and elevate pro-inflammatory cytokines and chemokines, thereby promoting the occurrence of colorectal tumors[27]. In the multivariate analysis, smoking was not established as a standalone contributing factor for AP. This might be explained by the stronger influence of other variables such as age, sex, BMI, alcohol consumption, OLDs-MASLD, and SF in the multivariate analysis, which weakened the significance of smoking as an independent risk factor. In addition, several studies have shown that BMI is linked to an increased risk of developing AP in men and women[28-30]. The hypothesis suggests that being overweight can be viewed as a condition of chronic mild inflammation, and that obesity is connected to the onset of CRC through increased chronic low-grade inflammation and increased adipokine production[31,32].
When OLDs coexisted with MASLD, the combined condition became an independent risk factor. Previous reports have indicated that the association between MAFLD and cancer risk varies by site, with an 1.14 times increased risk for CRC[33]. The exact pathophysiological mechanisms connecting MASLD and colorectal AP are not completely understood. Patients with NAFLD have lower numbers of CD4+ and CD8+ T cells in their blood circulation than healthy participants[34]. Similarly, patients with CRC exhibited significant differences in the abundance of these T cells. Low CD8+ T cell levels are associated with the advancement of original and metastatic colorectal adenomatous carcinoma[35,36]. This suggests that NAFLD, which is characterized by low levels of CD4+ and CD8+ T cells, may create an advantageous setting for the growth of colonic tumor cells. Related to NAFLD, tumor necrosis factor alpha has been reported to promote epithelial-mesenchymal transition in colon carcinoma cells[37]. Moreover, dysregulation of endoplasmic reticulum homeostasis and elevated endoplasmic reticulum stress are associated with epithelial-mesenchymal transition, which are conditions present in NAFLD[38]. Additionally, the gut microbiota significantly influences the regulation of immune responses. An altered composition of gut microbiota has been connected to the onset of NAFLD, including an increased Bacteroidetes to Firmicutes ratio and decreased abundance of butyrate-producing ruminococcaceae[39]. This suggests that enterotoxins produced by Bacteroides fragilis can trigger significant colonic inflammation in mice, thereby mimicking inflammation-driven colon carcinogenesis observed in humans[40]. The persistence of such bacterial populations in NAFLD and early CRC could help identify individuals with a background of fatty liver disease at increased risk for colorectal adenomas. A report published in 2023 suggested that serum microRNA levels could serve as biomarkers for diagnosing NAFLD in Chinese patients with colorectal polyps[24]. Further studies are required to fully investigate the association between MASLD and colorectal adenomas.
Studies indicate that irregular iron metabolism, whether due to genetic mutations or high dietary iron consumption, is a risk factor for CRC[41,42]. Abnormal iron status has been a topic of concern in CLD and is connected to the accelerated development of liver fibrosis[43,44]. The occurrence of AP may be associated with the inflammatory state of the liver. However, the study found no association between cirrhosis and AP. Interestingly, compared with patients with polyps, those without polyps had higher TSH levels (P = 0.011). As previously reported in a population of 139426 individuals, hypothyroidism was linked to a 22% reduction in CRC risk (OR = 0.78; 95%CI: 0.65-0.94; P = 0.08), and the strongest inverse relationship between rectal cancer and hypothyroidism was particularly evident in patients aged > 50 years[45]. Additional research is needed to ascertain if hypothyroidism may offer protection against the development of polyps in patients[45]. The advantages of this study include the comprehensive recording of histopathological information that confirmed the diagnosis of colorectal polyps based on patient pathology reports. Nonetheless, this study had several limitations. First, this was a retrospective analysis conducted at a single institution. The limited number of patients, particularly in the subgroup analysis, may have influenced the statistical power and reliability of the findings. Selection and information biases may also have occurred. In the future, we intend to conduct a prospective multicenter study that includes diverse regions and ethnicities to further validate the predictive performance of the nomogram model, thereby enhancing its generalizability. Second, establishing a causal relationship between MASLD and the risk of colorectal polyps is challenging when relying on cross-sectional studies. Moreover, the establishment and evaluation of this model were performed internally owing to the availability of data, enabling us to refine the model in a more controlled setting. However, this limitation affects the generalizability of the findings to other institutions. Finally, certain potential factors related to AP, such as detailed dietary habits, exercise patterns, medication history, and patient genetic information, were not included in our retrospective analysis because we were limited to the data accessible in the medical database. Future model development should consider these potentially influential factors.
CONCLUSION
To the best of our knowledge, this is the first study to create and validate a nomogram for predicting AP in patients with CLD. We established a simple model and internally validated it, thereby demonstrating its utility as a risk assessment tool. The predictive model developed is valuable for assisting in the evaluation of whether an endoscopic examination is necessary in patients with CLD.
ACKNOWLEDGEMENTS
The authors thank Shuang Li for providing guidance on the endoscopic diagnosis and treatment of colorectal polyps.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade B, Grade B
Novelty: Grade B, Grade B
Creativity or Innovation: Grade B, Grade B
Scientific Significance: Grade B, Grade B
P-Reviewer: Wang WJ; Yan JX S-Editor: Bai Y L-Editor: A P-Editor: Zhang XD
Bech JM, Terkelsen T, Bartels AS, Coscia F, Doll S, Zhao S, Zhang Z, Brünner N, Lindebjerg J, Madsen GI, Fang X, Mann M, Afonso Moreira JM. Proteomic Profiling of Colorectal Adenomas Identifies a Predictive Risk Signature for Development of Metachronous Advanced Colorectal Neoplasia.Gastroenterology. 2023;165:121-132.e5.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 6][Cited by in F6Publishing: 7][Article Influence: 3.5][Reference Citation Analysis (0)]
Li W, Chen Z, Chen H, Han X, Zhang G, Zhou X. Establish a Novel Model for Predicting the Risk of Colorectal Ademomatous Polyps: a Prospective Cohort Study.J Cancer. 2022;13:3103-3112.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Le C, Liu C, Lu B, Zhou X, Jiamaliding Y, Jin T, Dai S, Li J, Ding K, Xiao Q. Association between Hepatitis B virus infection and liver metastasis in colorectal cancer.MedComm (2020). 2024;5:e584.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Chen CW, Shan Cheng J, Chen TH, Kuo CJ, Ku HP, Chien RN, Chang ML. Different Metabolic Associations of Hepatitis C With Colon and Rectal Cancers: A 9-Year Nationwide Population-Based Cohort Study.Clin Colorectal Cancer. 2024;.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Zachou K, Azariadis K, Lytvyak E, Snijders RJALM, Takahashi A, Gatselis NK, Robles M, Andrade RJ, Schramm C, Lohse AW, Tanaka A, Drenth JPH, Montano-Loza AJ, Dalekos GN; International Autoimmune Hepatitis Group (IAIHG). Treatment responses and outcomes in patients with autoimmune hepatitis and concomitant features of non-alcoholic fatty liver disease.JHEP Rep. 2023;5:100778.
[PubMed] [DOI][Cited in This Article: ][Cited by in F6Publishing: 3][Reference Citation Analysis (0)]
Ahmad J, Barnhart HX, Bonacini M, Ghabril M, Hayashi PH, Odin JA, Rockey DC, Rossi S, Serrano J, Tillmann HL, Kleiner DE; Drug-Induced Liver Injury Network. Value of liver biopsy in the diagnosis of drug-induced liver injury.J Hepatol. 2022;76:1070-1078.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 28][Cited by in F6Publishing: 25][Article Influence: 8.3][Reference Citation Analysis (0)]
Rinella ME, Lazarus JV, Ratziu V, Francque SM, Sanyal AJ, Kanwal F, Romero D, Abdelmalek MF, Anstee QM, Arab JP, Arrese M, Bataller R, Beuers U, Boursier J, Bugianesi E, Byrne CD, Castro Narro GE, Chowdhury A, Cortez-Pinto H, Cryer DR, Cusi K, El-Kassas M, Klein S, Eskridge W, Fan J, Gawrieh S, Guy CD, Harrison SA, Kim SU, Koot BG, Korenjak M, Kowdley KV, Lacaille F, Loomba R, Mitchell-Thain R, Morgan TR, Powell EE, Roden M, Romero-Gómez M, Silva M, Singh SP, Sookoian SC, Spearman CW, Tiniakos D, Valenti L, Vos MB, Wong VW, Xanthakos S, Yilmaz Y, Younossi Z, Hobbs A, Villota-Rivas M, Newsome PN; NAFLD Nomenclature consensus group. A multisociety Delphi consensus statement on new fatty liver disease nomenclature.Hepatology. 2023;78:1966-1986.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 612][Cited by in F6Publishing: 772][Article Influence: 386.0][Reference Citation Analysis (0)]
Maida M, Dahiya DS, Shah YR, Tiwari A, Gopakumar H, Vohra I, Khan A, Jaber F, Ramai D, Facciorusso A. Screening and Surveillance of Colorectal Cancer: A Review of the Literature.Cancers (Basel). 2024;16.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Ng L, Sin RW, Cheung DH, Leung WK, Man AT, Lo OS, Law WL, Foo DC. Serum microRNA Levels as a Biomarker for Diagnosing Non-Alcoholic Fatty Liver Disease in Chinese Colorectal Polyp Patients.Int J Mol Sci. 2023;24:9084.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Siripongpreeda B, Mahidol C, Dusitanond N, Sriprayoon T, Muyphuag B, Sricharunrat T, Teerayatanakul N, Chaiwong W, Worasawate W, Sattayarungsee P, Sangthongdee J, Prarom J, Sornsamdang G, Soonklang K, Wittayasak K, Auewarakul CU. High prevalence of advanced colorectal neoplasia in the Thai population: a prospective screening colonoscopy of 1,404 cases.BMC Gastroenterol. 2016;16:101.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 12][Cited by in F6Publishing: 16][Article Influence: 1.8][Reference Citation Analysis (0)]
Shukla PK, Chaudhry KK, Mir H, Gangwar R, Yadav N, Manda B, Meena AS, Rao R. Chronic ethanol feeding promotes azoxymethane and dextran sulfate sodium-induced colonic tumorigenesis potentially by enhancing mucosal inflammation.BMC Cancer. 2016;16:189.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 21][Cited by in F6Publishing: 24][Article Influence: 2.7][Reference Citation Analysis (0)]
Bai H, Xu Z, Li J, Zhang X, Gao K, Fei X, Yang J, Li Q, Qian S, Zhang W, Gao X, Tang M, Wang J, Chen K, Jin M. Independent and joint associations of general and abdominal obesity with the risk of conventional adenomas and serrated polyps: A large population-based study in East Asia.Int J Cancer. 2023;153:54-63.
[PubMed] [DOI][Cited in This Article: ][Cited by in F6Publishing: 2][Reference Citation Analysis (0)]
Jia Y, Li D, You Y, Yu J, Jiang W, Liu Y, Zeng R, Wan Z, Lei Y, Liao X. Multi-system diseases and death trajectory of metabolic dysfunction-associated fatty liver disease: findings from the UK Biobank.BMC Med. 2023;21:398.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Tedesco D, Thapa M, Gumber S, Elrod EJ, Rahman K, Ibegbu CC, Magliocca JF, Adams AB, Anania F, Grakoui A. CD4(+) Foxp3(+) T cells promote aberrant immunoglobulin G production and maintain CD8(+) T-cell suppression during chronic liver disease.Hepatology. 2017;65:661-677.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 9][Cited by in F6Publishing: 10][Article Influence: 1.3][Reference Citation Analysis (0)]
Kuwahara T, Hazama S, Suzuki N, Yoshida S, Tomochika S, Nakagami Y, Matsui H, Shindo Y, Kanekiyo S, Tokumitsu Y, Iida M, Tsunedomi R, Takeda S, Yoshino S, Okayama N, Suehiro Y, Yamasaki T, Fujita T, Kawakami Y, Ueno T, Nagano H. Correction: Intratumoural-infiltrating CD4 + and FOXP3 + T cells as strong positive predictive markers for the prognosis of resectable colorectal cancer.Br J Cancer. 2019;121:983-984.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 6][Cited by in F6Publishing: 6][Article Influence: 1.0][Reference Citation Analysis (0)]
Mlecnik B, Tosolini M, Kirilovsky A, Berger A, Bindea G, Meatchi T, Bruneval P, Trajanoski Z, Fridman WH, Pagès F, Galon J. Histopathologic-based prognostic factors of colorectal cancers are associated with the state of the local immune reaction.J Clin Oncol. 2011;29:610-618.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 689][Cited by in F6Publishing: 760][Article Influence: 54.3][Reference Citation Analysis (0)]
Takasago T, Hayashi R, Ueno Y, Ariyoshi M, Onishi K, Yamashita K, Hiyama Y, Takigawa H, Yuge R, Urabe Y, Oka S, Kitadai Y, Tanaka S. Anti-tumor necrosis factor-alpha monoclonal antibody suppresses colorectal cancer growth in an orthotopic transplant mouse model.PLoS One. 2023;18:e0283822.
[PubMed] [DOI][Cited in This Article: ][Cited by in F6Publishing: 2][Reference Citation Analysis (0)]
Sharma R, Zhao W, Zafar Y, Murali AR, Brown KE. Serum hepcidin levels in chronic liver disease: a systematic review and meta-analysis.Clin Chem Lab Med. 2024;62:373-384.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Ryan JD, Armitage AE, Cobbold JF, Banerjee R, Borsani O, Dongiovanni P, Neubauer S, Morovat R, Wang LM, Pasricha SR, Fargion S, Collier J, Barnes E, Drakesmith H, Valenti L, Pavlides M. Hepatic iron is the major determinant of serum ferritin in NAFLD patients.Liver Int. 2018;38:164-173.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 51][Cited by in F6Publishing: 57][Article Influence: 8.1][Reference Citation Analysis (0)]
