Introduction

Skin tags are small, soft, pedunculated, flesh to dark brown colored, benign skin tumors commonly occurring on the neck and axilla. They are thought to be relatively common skin lesions and it has been reported that they might reflect insulin resistance states. Age, obesity, atherogenic lipid profile are also associated with skin tags [1–9]. Moreover, obesity, dyslipidemia, hypertension, hyperglycemia, hyperinsulinemia, hyperleptinemia are known to be associated with insulin resistance. It is well known that insulin resistance and low-grade inflammation are two major causes of atherosclerosis. Therefore, increased high-sensitive C-reactive protein (hs-CRP) and uric acid, and insulin resistance, hyperleptinemia, dyslipidemia, obesity and hypertension play an important role in the pathogenesis of atherosclerosis and cardiovascular disease [10–13].

Two recent studies showed that skin tags are commonly associated with overt diabetes and impaired glucose tolerance [5, 14]. Although there have been a few reports that the presence of skin tags is associated with diabetes mellitus, hypertension, obesity and atherogenic lipid profile [1–9, 14], no data in the literature show that the presence of skin tags is associated with elevations of serum hs-CRP, uric acid, free fatty acid and leptin level in patients with normal glucose tolerance. The aim of this study was to evaluate the frequency of hypertension, dyslipidemia, obesity and to investigate insulin resistance, serum hs-CRP, uric acid and serum leptin levels in normal glucose tolerant patients with skin tags who were admitted to our dermatology outpatient clinic.

Patients and methods

We evaluated consecutive 113 normal glucose tolerant patients [67 (59.3%) female and 46 (40.7% male)] with skin tags who were admitted to our dermatology outpatient clinic from January to December 2004. All patients were examined for skin tags by the same dermatologist. The clinical diagnosis of skin tag was confirmed by histological examination. The control group consisted of 31 healthy subjects without skin tags [18 (58%) female and 13 male (42%)]. Sex and age distribution of the patients in the groups was similar. All subjects were in good health, with no known diseases including cancer, diabetes mellitus, acromegaly or gastrointestinal diseases.

All subjects underwent oral glucose tolerance test to exclude diabetes mellitus and impaired glucose tolerance. Body weight, height, fat percentage, fat weight and body mass index (BMI), insulin resistance and serum leptin, lipids, uric acid, free fatty acid and hs-CRP levels were measured.

Height, weight and waist circumference were measured by the same person with the subjects wearing light clothing but not shoes. BMI was calculated as weight in kilograms divided by the square of the height in meters. Overweight was defined as a BMI 25–30 kg/m2, and obesity was defined as a BMI of ≥30 kg/m2. Body fat percentage and fat weight were measured by bio-electrical impedance analysis (OMRON BF 302, Omron Healthcare, Hoofdorp, the Netherlands).

All blood pressure measurements were made with calibrated mercury manometers (Riester, Germany) in the right arm with the patient in a sitting position after a rest of 5 min. Hypertension was defined as a systolic/diastolic blood pressure of 140/90 mmHg or higher, and/or a current antihypertensive treatment.

Dyslipidemia was defined in presence of at least one of following conditions; raised plasma triglycerides (>200 mg/dl), total cholesterol (>200 mg/dl), LDL-cholesterol (>130 mg/dl), low HDL-cholesterol (<35 mg/dl for men and <45 mg/dl for women) and/or a current antilipidemic treatment.

Fasting plasma insulin levels were measured with Roche Modular Analytics E170 immunoassay system (Roche Diagnostics GmbH, Mannheim, Germany) using electrochemiluminescence immunoassay method (ECLIA). Fasting plasma glucose cholesterol and triglyceride levels were determined by enzymatic colorimetric assay method (Roche Diagnostics GmbH, Mannheim, Germany). Insulin resistance was estimated using the homeostasis model assessment (HOMA-IR) derived from the following equation [HOMA-IR = (fasting plasma glucose level × fasting plasma insulin level)/22.5] [15]. Fasting serum leptin levels were measured by two-site immunoradiometric assay method (Active Human Leptin IRMA, DSL-23100, USA). Hs-CRP levels were measured by nephelometric method on BNII nephelometer (Dade Behring, Germany). Serum non-esterified free fatty acids were measured by enzymatic colorimetric method on the Modular autoanalyser (Roche Diagnostics, Switzerland). In this method (Wako NEFA C, ACS-ACOD), free fatty acids in serum, when treated with acyl-CoA synthetase (ACS) in the presence of ATP, magnesium and CoA, form the acyl-CoA. Then, acyl-CoA is oxidized by added acyl-CoA oxidase (ACOD) to produce hydrogen peroxide that in the presence of added peroxidase allows to form a purple-colored adduct with an absorption at 550 nm. Hence, the amount of free fatty acid in the sample can be determined from the optical density measured at 550 nm. Glycated hemoglobin (HbA1c) was measured by “Turbidimetric inhibition immunoassay” (TINIA) method (HBAIC II, Roche Diagnostics GmbH, Mannheim, Germany).

Statistical analysis

Statistical analysis was performed using SPSS statistical software (SPSS 10.0 for Windows, standard version). The results were presented as mean ± SD. Univariate analysis of variance and Mann–Whitney U tests were performed to compare the groups. Correlation analysis was done by Spearman test. A P value <0.05 was considered statistically significant.

Results

The number of skin tags ranged from 1 to 48 (mean = 16.1 ± 20.1, median = 8). Disseminated and localized skin tags were detected in 23 (20.4%) and 90 (79.6%) patients, respectively. Skin tags were detected on the neck in 77 (68.1%) patients, on the back of the upper body in 30 (26.5%) patients, in the axillae in 23 (20.4%) patients, on the extremities in 6 (5.3%) patients. The frequency of smoking was 30.1%, and history of diabetes in the family was 37.2%, and the history of hypertension in the family was 49.6% in patients with skin tags. A total of 61 (53.9%) and 38 (33.6%) patients with skin tags were overweight and obese, respectively. The frequency of hypertension was 30.1%, dyslipidemia 59.3%, insulin resistance 21.2% and metabolic syndrome %39.8.

BMI (P < 0.001), waist circumference (P < 0.001), systolic blood pressure (P = 0.003), diastolic blood pressure (P = 0.023), body fat percentage (P = 0.016), body fat weight (P = 0.007) were all higher in patients with skin tags than in healthy subjects, whereas serum leptin levels were not different (P = 0.16). Similarly, serum insulin (P = 0.002), glucose (P < 0.001), free fatty acid (P = 0.002), hs-CRP (P = 0.001), uric acid (P = 0.001), total cholesterol (P = 0.018), LDL-cholesterol (P = 0.023), triglyceride (P = 0.001) and HbA1c (P < 0.001) were higher in patients with skin tags. Serum HDL-cholesterol levels were lower in patients with skin tags (P = 0.026; Table 1). After adjusting for BMI, the differences in serum insulin, free fatty acid and HOMA-IR between the two groups were diminished (P > 0.05) whereas the differences in serum glucose (P = 0.001), HbA1c (P = 0.03), LDL (P = 0.04), triglyceride (P = 0.01), uric acid (P = 0.035) and hs-CRP levels (P = 0.045) between two groups were maintained.

Table 1 Demographic and biochemical parameters in patients with skin tags and control groups
Full size table

The number of skin tags and distribution of the skin tags were not related to any parameters. Patients with skin tags showed a positive correlation between leptin levels and glucose (r = 0.25, P = 0.033), insulin (r = 0.5, P < 0.001) and HOMA-IR (r = 0.47, P < 0.001). A positive correlation between BMI and leptin (r = 0.42, P < 0.001), hs-CRP (r = 0.31, P = 0.026), insulin (r = 0.36, P = 0.03), free fatty acid (r = 0.42, P < 0.001) and HOMA-IR (r = 0.32, P = 0.009) was also observed in patients with skin tags.

Discussion

Skin tags, which appear to be associated with some endocrine diseases, are skin growths histologically characterized by a papillomatous acanthotic pattern in the epidermis [16]. Although it is not completely confirmed, in patients preselected for gastrointestinal complaints, skin tags appear to be more prevalent in those with colonic polyps. However, skin tags are also part of ageing process and it is quite natural to find more tags in older patients. Skin tags are most commonly found on the neck and in the axillae [2]. Skin tags were detected on the neck in 77 (68.1%) patients, on the back of the upper body in 30 (26.5%) patients, in the axillae in 23 (20.4%) patients, on the extremities in 6 (5.3%) of the patients in our study.

Increased hs-CRP, insulin resistance, dyslipidemia, obesity and hypertension play an important role in the pathogenesis of atherosclerosis and cardiovascular disease [11–13, 17, 18]. The first purpose of this study was to evaluate the relationship between skin tags and obesity, hypertension, dyslipidemia and insulin resistance. Dermatological changes have been reported in patients with obesity due to insulin resistance (acanthosis nigricans and skin tags) [3, 5, 19, 20]. Skin tags commonly develop in individuals with obesity and were observed in up to 74% of patients in one study; their prevalence correlated positively with the severity of the obesity [3, 4]. Similarly, 53.9 and 33.6% of patients with skin tags were overweight and obese, in our study. Only 12.5% of patients with skin tags had a normal BMI. We found no correlation between the number of skin tags and BMI.

Only patients with skin tags and normal glucose tolerance were included in the present study, because the relationships between skin tags, insulin resistance and type 2 diabetes mellitus have been established in previous studies. Recently, overt diabetes mellitus has been found in 26.3–45% of patients, and impaired glucose tolerance has been found in 7.9–19.2% of patients with skin tags [7, 8, 14, 19, 20]. Although some researchers have hypothesized that skin tags may be a marker for insulin resistance [1, 6, 19, 20], Mathur et al. [5] did not accept this suggestion. As in the study by Erdogan et al. [19], we calculated insulin resistance by using HOMA-IR formula. It was significantly higher in our patients with skin tags than that in the controls. Although we selected patients with normal glucose tolerance; HOMA-IR, glucose and HbA1c were higher in patients with skin tags than in the control group. Because of our patients with skin tags have a higher BMI, we adjusted the HOMA-IR value for BMI. After this adjustment, there was no difference between the two groups. We thought that obesity plays an important role for the higher levels of HOMA-IR. On the other hand, after adjusting for BMI, serum glucose and HbA1c were higher in patients with skin tags than in the control group. The frequency of insulin resistance (HOMA-IR > 2.5) was 21.2% in patients with skin tags. This may reflect the selection of patients with normal glucose tolerance.

There have been a few reports in the literature that the presence of skin tags is associated with hypertension and an atherogenic lipid profile [1, 2, 10, 20]. Hypertension was found in 30.1% and dyslipidemia in 59.3% of patients with skin tags in our study. We also detected higher total cholesterol, LDL-cholesterol, triglyceride levels and lower HDL-cholesterol in patients with skin tags. Increased free fatty acid levels play an important role in the pathogenesis of atherosclerosis and cardiovascular disease [11–13, 17, 18]. Higher free fatty acid concentrations can also induce muscle and hepatic insulin resistance as well as endothelial dysfunction [21–24]. However, a possible relationship between skin tags and serum free fatty acid levels in subjects with normal glucose tolerance has not been investigated yet. Elevated free fatty acid levels were a novel finding of our study. Although we detected higher free fatty acid levels in patients with skin tags than the control group, after adjustment for BMI, this difference diminished. We conclude that obesity plays an important role in raising levels of serum free fatty acids.

The major purpose of this study was to evaluate the relationship between skin tags and serum leptin, free fatty acid and hs-CRP levels. To the best of our knowledge, this is the first investigation of a relationship between skin tags and these parameters in subjects with normal glucose tolerance. Elevated hs-CRP is another novel finding of our study. Recent studies showed that atherosclerosis is a chronic inflammatory process [25, 26]. C-reactive protein is a prototypic marker of inflammation, and many studies showed that hs-CRP confers a risk for atherosclerosis and coronary heart disease [11–15, 17, 18, 27]. Even after adjusting for BMI, serum hs-CRP was higher in patients with skin tags when compared with the control group in our study. Skin tags may be markers of a state of inflammation, which correlates with increased risk of atherosclerosis and coronary heart disease.

Serum uric acid is strongly associated with atherosclerosis and cardiovascular risk factors [28–30]. This is the first study to examine the relationship between serum uric acid levels and skin tags. We detected higher serum uric acid levels in patients with skin tags. Elevated serum uric acid levels indicate that these patients have greater risk for atherosclerosis and coronary heart disease.

Elevated plasma leptin levels or hyperleptinemia correlate with insulin resistance and other markers of the metabolic syndrome including obesity, hypercholesterolemia and hypertension. In addition to metabolic effects of leptin, it plays a role in the pathogenesis of atherosclerosis [31, 32]. To the best of our knowledge, a possible relationship between skin tags and serum leptin levels in subjects with skin tags has not been investigated. In this study, we found a non-significant correlation between leptin levels and skin tags number.

Our results suggest a correlation between the presence skin tags and markers of the metabolic syndrome and increased risk of cardiovascular disease. However, further studies are required to clarify the effects of BMI on these findings in patients with skin tags and full-matched control group for BMI.

In conclusion, we propose that skin tags are cutaneous findings frequently associated with the risk factors (obesity, hypertension, dyslipidemia, insulin resistance, elevated hs-CRP, etc.) for metabolic syndrome and heart disease. Given the higher frequency of overweight and/or obesity, dyslipidemia and elevated hs-CRP, glucose, HbA1c, uric acid levels in patients with skin tags, these patients should be carefully evaluated for metabolic syndrome and heart disease.