Thyroid nodules as predictors of adenomatous colonic polyps: A systematic review and meta-analysis

Abstract

BACKGROUND

Thyroid nodules (TN) are increasingly diagnosed worldwide; investigating the association between TN and colon polyps could be helpful in early detection and management. To our knowledge no meta-analysis has assessed the relationship between TN and adenomatous colonic polyps.

AIM

To assess the association between adenomatous colonic polyps, thyroid-stimulating hormone, and TN.

METHODS

We searched PubMed, MEDLINE, Cochrane Library, EBSCO, and the first 100 articles in Google for articles published in English from inception until April 2025. We included prospective cohorts, retrospective studies, case-control studies, and cross-sectional studies. The keywords thyroid nodules, adenomatous colon polyps, thyroid volume, metabolic syndrome, insulin resistance, and thyroid malignancy were used.

RESULTS

Out of 237 articles, 25 full texts were reviewed, and 5 full texts were included in the final meta-analysis. No relationship was found between TN, colonic polyps, and thyroid-stimulating hormone levels [odd ratio (OR): 1.78, 95% confidence interval (CI): 0.55-5.74, P = 0.33]. Colonic polyps were more common among patients with TN when addressing heterogeneity (OR: 0.42, 95%CI: 0.30-0.52, P < 0.001 and OR: 0.08, 95%CI: 0.70-0.86, P = 0.85).

CONCLUSION

TN were similar among patients with and without adenomatous colonic polyps. However, TN was more common among colon polyps when addressing the heterogeneity. Thyroid-stimulating hormone was not different between those with and without TN. Age, sex, adiposity, and smoking effects might explain the higher rate observed by the included studies. Further studies controlling for the same are needed.

Key Words: Adenomatous colonic polyps; Thyroid stimulating hormone; Thyroid nodule; Predictors; Risk factors

Core Tip: Thyroid polyps are common among the middle age group, and colonoscopy is indicated in individuals older than 50 years. However, it is invasive, making easy reliable methods for risk stratification for colonic polyps screening important. The few studies that assessed the association of thyroid nodules and colon polyps showed contradicting results. The current meta-analysis contributed significantly to the field because it is the first meta-analysis to find an association. Our research challenges the current colonic polyp screening guidelines because physicians might need to include thyroid nodules as a risk factor for colonic polyps.

INTRODUCTION

Thyroid nodules (TN) may be present in nearly two-thirds of the population depending on age. However, only 20% are clinically significant (cancer, pressure symptoms, and thyroid activity in 10%, 5%, and 5% respectively). TN are common, and thyroid cancer occurs in 7%-15% of patients with TN. Clinical examination, thyroid ultrasound, thyroid function tests, fine needle aspiration, and molecular testing are indicated for management and future planning[1,2]. Functioning TN are rarely malignant, but TN with normal or high thyroid-stimulating hormone (TSH) and suspicious ultrasound are candidates for fine needle aspiration. Indeterminate TN require molecular testing in the adult population. However, in children and pregnant females, their role is not validated[3].

The incidence of thyroid cancer is on the rise worldwide, and various environmental factors are to blame including iodine deficiency, vitamin D deficiency, increasing weight, and toxins exposure[4,5]. A case-control study showed the association between vitamin D deficiency and thyroid cancer[6]. Further studies concluded an association between vitamin D and thyroid disease[7-9]. On the other hand, Yildiz et al[10] and Laney et al[11] showed no association. Further studies showed the association between TSH, low vitamin D, and thyroid cancer[12,13].

Colorectal cancer is the third-leading cause of cancer-related deaths globally, and most cancers develop from polyps. The prevalence increases with age, and 12.9% suffered from polyps in a recent survey[14]. A study conducted in Asia found that nearly half were single polyps, and the majority were small and sessile[15]. Early detection of colonic polyps by colonoscopy was shown to reduce this common cancer mortality[16].

TN detection by ultrasonography might be a simple noninvasive measure to assess colonic polys risk and refer those at risk for screening. Given the above and the fact that colonic cancer develops in 5% of patients with colonic polyps[17], we conducted this meta-analysis to assess the relationship between colonic polyps and TN.

MATERIALS AND METHODS

Article selection

We searched PubMed, MEDLINE, Cochrane Library, EBSCO, and the first 100 articles in Google for articles published from inception until April 2025 with adherence to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines.

Inclusion and exclusion criteria

Prospective cohorts, retrospective studies, case-control studies, and cross-sectional studies assessing the relationship between TN and adenomatous colon polyps were included. Articles published on polyps other than adenomatous were excluded. In addition, case reports, series, opinions, editorials, poster presentations, and abstracts were not included.

Literature search

Two authors (Mirghani HO and Alghamdi S) systematically searched the literature for relevant articles that assessed the association of adenomatous colon polyps and TN. The title and abstracts were assessed in addition to their references. The keywords thyroid nodules, adenomatous colon polyps, thyroid volume, and thyroid malignancy were used. Two hundred and thirty-seven articles were eligible, and 128 remained after duplication removal. Of them 25 full texts were reviewed, and 5 full texts were included in the final meta-analysis.

Data extraction

The data were extracted to an Excel sheet detailing the first author’s name, country, study type, year of publication, TN and polyps in cases and control, age, female percentage, body mass index, waist circumference, smoking status, and TSH among patients with/without nodules. The text included three case-control, a prospective cohort, and one retrospective study (Figure 1, Tables 1, 2, and 3).

Figure 1 The relationship of thyroid nodules and colonic polyps following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines.

Table 1 Relationship between thyroid nodules and colonic polyps.

Ref.	Country	Methodology	Polyps and nodules in cases	Polyps and nodules in controls	Results
Duran et al[19], 2012	Türkiye	Case-control	44/66	61/146	More nodules in patients with polyps, P = 0.001
Mousa et al[20], 2019	United Kingdom	Case-control	55/105	24/68	More nodules in patients with polyps, P = 0.017
Ozturk et al[21], 2022	Türkiye	Retrospective	15/20	36/103	More nodules in patients with polyps, P < 0.03
Park et al[22], 2015	Korea	Prospective	63/266	37/277	No significant difference
Yu et al[23], 2021	China	Case-control	542/1379	1035/3135	More nodules in patients with polyps, P < 0.001

Table 2 Risk factors for thyroid nodules among the study group.

Ref.	Age	Females%	Obesity/BMI	Waist	Smoking
Duran et al[19], 2012	53.3 ± 11.4 vs 51.8 ± 11.4	Matched	Not assessed	Not assessed	Not assessed
Mousa et al[20], 2019	54.48 ± 10.01 vs 51.87 ± 10. 86	47.3% vs 52.7%	28.88 ± 7.66 vs 27.99 ± 4. 06	50.6% vs 49.4% high waist	Not assessed
Ozturk et al[21], 2022	61.95 ± 9.01 vs 57.13 ± 14.49	35% vs 69 67%	Not assessed	Not assessed	Not assessed
Park et al[22], 2015	52.08 ± 11.48 vs 47.02 ± 9.26	36.5% vs 55.6%	37.2% vs 27.1%	83.56 ± 9.16 vs 80.20 ± 9.07	16.5% vs 10.5%
Yu et al[23], 2021	50 vs 45	25.1% vs 43.6%	24.5 vs 23.3%	3.24 ± 9.58	39.3% vs 23.5%

BMI: Body mass index.

Table 3 Thyroid stimulating hormone levels among patients with and without thyroid polyps.

Ref.	TSH among patients with nodules	TSH among patients without nodules
Duran et al[19], 2012	1.17 ± 0.95	1.54 ± 1.3
Mousa et al[20], 2019	1.40 ± 0.84	1.72 ± 0.97
Ozturk et al[21], 2022	2.76 ± 2.91	3.24 ± 9.58
Park et al[22], 2015	2.73 ± 7.25	1.69 ± 1.06
Yu et al[23], 2021	1.54	1.64

TSH: Thyroid-stimulating hormone.

Quality assessment of the included studies

The quality of the included studies was assessed using the Newcastle Ottawa Scale[18] (Table 4).

Table 4 Newcastle Ottawa scale risk of bias of the included studies.

Ref.	Country	Selection bias	Comparability	Outcome	Total score
Duran et al[19], 2012	Türkiye	4	2	2	8
Mousa et al[20], 2019	United Kingdom	4	2	2	8
Ozturk et al[21], 2022	Türkiye	4	2	2	8
Park et al[22], 2015	Korea	4	2	2	8
Yu et al[23], 2021	China	4	2	3	9

Statistical analysis

The RevMan system, version 5.4 was used for data analysis. The data were dichotomous and entered manually. Forest plots were generated, and the effect size was estimated as odds ratio (OR) and 95% confidence interval (CI). The random effect model was used due to the significant heterogeneity observed (heterogeneity < 25 was considered mild, and heterogeneity ≥ 50 was considered significant). A P value of 0.05 was considered significant.

RESULTS

The current meta-analysis included three case-control studies, one prospective study, and one retrospective cohort (four from Asia and one from Europe). The studies included 1912 events among 5565 patients[19-23]. No relationship was found between TN and colonic polyps (OR: 1.78, 95%CI: 0.55-5.74, P = 0.33). However, significant heterogeneity was observed (I² = 97%, P < 0.001) (Figure 2A). However, after removing one study to eliminate heterogeneity, colonic polyps were more common among patients with TN (OR: 0.42, 95%CI: 0.30-0.52, P < 0.001, I² = 12%, P for heterogeneity = 0.33) (Figure 2B).

Figure 2 Forest plot. A: The relationship between thyroid nodules and colonic polyps; B: The relationship between thyroid nodules and colonic polyps after addressing the heterogeneity; C: Thyroid-stimulating hormone levels among patients with and without thyroid nodules. CI: Confidence interval.

TSH levels were not different between patients with and without TN (OR: 0.08, 95%CI: 0.70-0.86, P = 0.85). However, substantial heterogeneity was found (I² = 93%, P < 0.001) (Figure 2C).

DISCUSSION

Literature on the link between TN and colonic polyps is scarce. Common risk factors for these two common disorders are age, male sex, and metabolic risk factors[24]. In the present study TN incidence was similar among patients with and without adenomatous colon polyps (OR: 1.78, 95%CI: 0.55-5.74, P = 0.33). However, after removing the heterogeneous study, the relationship became positive (OR: 0.42, 95%CI: 0.30-0.52, and P < 0.001). TSH levels were not different between those with and without TN (OR: 0.08, 95%CI: 0.70-0.86, P = 0.85).

The first study to observe the association of colonic polyps and TN was published by Duran et al[19]. They did not assess body mass index, waist circumference, and smoking, limiting the study. TN were found to be associated with obesity and diabetes[25]. Similarly, the higher rate of TN among patients with adenomatous colonic polyps observed by Mousa et al[20] might be explained by the higher age, male sex, and body mass index (Table 2). High insulin, glucose-dependent insulinotropic peptide, and Insulin-like growth factor 1 (IGF-1) levels are also suggested to mediate the association[26,27]. The results of Park et al[22] included in this study were limited by the older age, male, body mass index, and smoking among patients with colonic polyps. In addition, higher smoking among the colonic polyp arm might bias the results of Yu and Wu[23]. The above observations imply that the higher TN among patients with colonic polyps might be due to various confounding variables (age, male sex, obesity, and smoking status).

The role of the TSH on thyroid growth is conflicting with some showing a positive relationship[28,29] and others failing to show any association[30,31]. In the current meta-analysis TSH levels were not different between patients with and without TN, similar to Layegh et al[32], who conducted a study among patients with obesity and normal body mass index and found no difference in TSH levels among patients with and without TN.

Two out of the five studies included in this analysis reported a higher prevalence of smoking among patients with TN[23,23]. Higher smoking among the TN arm might greatly influence the results[33,34].

Limitations of this study must be acknowledged. The small number of included studies and the observational methodology limited our study. High heterogeneity was also observed.

CONCLUSION

TN were similar among patients with and without adenomatous colonic polyps. TSH levels were not different between those with and without TN. Age, sex, adiposity, and smoking effects might explain the higher rate observed by the included studies. Further studies controlling for the same are needed.