Skin tags and acanthosis nigricans in patients with hepatitis C infection in relation to insulin resistance and insulin like growth factor-1 levels.

BACKGROUND: Skin tags (ST) are papillomas commonly found in the neck, axillae of middle-aged and elderly people AIM: Insulin and insulin-like growth factor (IGF-1) levels are affected by hepatitis C virus (HCV) infection and both of them may be implicated in the etiopathogenesis of ST and acanthosis nigricans (AN) through their proliferative and differentiating properties. So, the aim of this work was to evaluate the impact of HCV infection on ST and AN through the estimation of insulin resistance and IGF-1.
MATERIALS AND METHODS: PARTICIPANTS WERE ARRANGED INTO FOUR GROUPS: (ST +ve / HCV +ve) 23 subjects, (ST+ / HCV -ve) 19 subjects, (HCV -ve / ST-ve) 20 subjects and (ST-ve /HCV +ve) 22 subjects. Age, ST size, color, number, AN, fasting glucose, fasting insulin, insulin resistance, IGF-1, HCV-antibodies (Ab) were recorded.
RESULTS: The mean number of ST in Group 1 was half the number of ST in Group 2 (11.0±9.3 / 22.3±14.0) (P=0.005). The difference in insulin resistance between the same groups was non-significant (13.1±10.6 / 9.0±5.5) (P=0.441) while the difference in IGF-1 was statistically significant (218.6±46.2 /285.4±32.8) (P=0.002). The multivariate logistic regression for the variables revealed that insulin resistance is the only factor affecting the occurrence of ST (OR=1.096, P=0.023). Multivariate regression analysis for the variables showed that HCV was borderline but not a significant factor affecting the number of ST (Beta=-0.409, P=0.053). The number of patients with AN was doubled in Group 2 in comparison to Group 1 but this was non significant 3(13%) / 6(32%) (P=0.2800).
CONCLUSION: HCV is associated with a significant decrease in the ST number and in the serum level of IGF-1 together with an obvious decrease in the occurrence of AN. Our results may point to the entrant effect of insulin resistance and IGF-1 in ST and AN development. The current study suggests the evaluation of IGF-1-lowering agents in the control of ST and AN especially in the females with polycystic ovary and in the prevention of the recurrence of ST after surgical removal.

Introduction

Skin tags (acrochordons) are small flesh-colored to dark brown sessile or pedunculated papillomas that commonly occur on the neck, in the axillae and on the eyelids, and manifest less often on the trunk and in the groin. They have the same incidence in both sexes.[1] Histologically, ST are composed of loose collagen fibers and dilated capillaries.[2] ST have been reported to be associated with many diseases and conditions including diabetes mellitus,[3] obesity,[4] acromegaly,[5] Crohn's disease,[6] aging,[7] child abuse,[8] organ transplantation[9] and colonic polyps.[10] It was also recorded with pregnancy,[11] with human papilloma virus,[12] with increased mast cell count[13] and with increased androgen, α and β estrogen receptor levels.[14]

Recent studies have suggested that chronic hepatitis C is not nearly a hepatic viral infection, but is a disease which exhibits features of metabolic liver disease such as insulin resistance, high prevalence of steatosis, increased prevalence of impaired glucose tolerance, type two diabetes mellitus and changes in lipid metabolism irrespective of the presence of cirrhosis.[15] High glucose levels as well as high insulin levels have a fibrogenetic effect whether in the liver or in the skin.[16] Insulin is a well-established growth-promoting hormone. Hyperinsulinemia elevates serum concentrations of free insulin-like growth factor (IGF-1), while simultaneously reducing IGF-binding protein 3 (IGFBP-3). Binding of IGF-1 to its receptors on keratinocytes causes epidermal hyperplasia. On the other hand, insulin-mediated reductions in IGFBP-3, which is a ligand for the nuclear retinoid X receptor alpha, may decrease the transcription of anti-proliferative genes normally activated by the body's endogenous retinoids. These endocrine shifts alter cellular proliferation and growth, which may manifest as cutaneous papillomas, ST or AN in the skin.[17]

Growth hormone (GH) and IGF-1 play an important role in epidermal homeostasis. GH is secreted from the pituitary gland and binds to GH-receptor, expressed on most peripheral cells of the body.[18] GH induces hepatic synthesis and secretion of IGF-1, the mediator of growth. IGF-1 exerts its major effect on proliferation, while having an effect similar to insulin on differentiation.[18] Excess and deficiency of GH and IGF-1 are associated with profound alterations of the function and the structure of connective tissues[19] as IGF-1 is considered a mitogenic and anti-apoptotic agent for fibroblasts and keratinocytes.[20] It was found that IGF-1 synthesis is disturbed in patients with chronic hepatitis C and its level in serum reflects the severity of the liver fibrosis.[21]

In literature, insulin and IGF-1 levels are affected by hepatitis C virus (HCV) infection and both of them may be implicated in the etiogenesis of ST and AN through their proliferative and differentiating properties. So, the aim of this work was to evaluate the impact of HCV infection on ST and AN through the estimation of insulin resistance and IGF-1.

Materials and Methods

This study was conducted from 2007 to 2009 after obtaining the permission of the local research Committee. The study included 84 participants collected from the outpatient clinic of dermatology and from the outpatient and internal medicine department of a university hospital. Patients were divided into four groups:(ST +ve /HCV +ve) Group 1 included 23 subjects, (ST +ve/ HCV -ve) Group 2 included 19 subjects, (ST -ve /HCV -ve) Group 3 included 20 subjects, and (ST -ve /HCV +ve) Group 4 included 22 subjects. History-taking highlighted age, marital status, family history of diabetes, and history of cardiac, hepatic, gastrointestinal or endocrinal disorders other than HCV. Individuals associated with medical conditions such as diabetes mellitus or pregnancy were excluded from the study. ST parameters including size (small less than 5 mm-big more than 5 mm), count, number, color, AN, fasting blood sugar, fasting insulin, insulin resistance, IGF-1 level and serum HCV-Ab were recorded.

Estimation of fasting serum insulin

Serum Insulin was measured by Enzyme Linked Immunosorbent Assay (ELISA) using DRG Human Insulin EIA-2935. This kit provides a method for the quantitative determination of insulin in human serum.

Detection of insulin resistance by HOMA

HOMA beta cell index was used as beta cell function index. HOMA beta cell index (μU/mmol) =20× fasting insulin / (fasting glucose – 3.5).[22]

Detection of HCV-Ab in serum

HCV-Ab in serum was detected using ELISA for the cutoff determination of antibodies against HCV in human serum or plasma (ABBOTT Murex). This kit employs solid-phase, indirect ELISA assay for detection of antibodies to HCV.

Quantitation of IGF-1 in serum

For Quantitation of IGF-1 in serum, DRG Leptin Enzyme Immunoassay Kit (DRG Instruments GmbH, Germany) was used.

Statistical Analysis

Data was coded and entered using the statistical package SPSS Version 15. Data were summarized using mean and standard deviation for quantitative variables and percent for qualitative variables. Comparison between groups was done using Chi-square tests for qualitative variables and analysis of variants (ANOVA) and multiple comparison post hoc test for normally distributed quantitative variables while non-parametrical Kruskal-Wallis test and Mann-Whitney test were used for quantitative variables not normally distributed. Correlation was done to test for linear relation between quantitative variables. P values less than 5% were considered as statistically significant.

Results

In order to evaluate the ST characteristics and AN in HCV-infected subjects, the basic studied characteristics were recorded in Table 1.

Table 1

Clinical and lab findings among the four different groups

Comparison of the age between Group 1 and Group 2 was statistically significant (P=0.033). Also, the comparison between Group 2 and Group 4 was statistically significant (P=0.002) [Table 1].

Comparison of the insulin level between Group 1 (49.1±32.3) and Group 3 (19.3±8.5) was statistically significant (P=0.002). Comparison of the insulin level between Group 2 (36.9±22.4) and Group 3 was also statistically significant (P=0.015) [Table 1].

Comparison of the insulin resistance between Group 1 (13.1±10.6) and Group 3 (5.1±2.2) was statistically significant (P=0.023). Comparison of the insulin resistance between Group 2 (9.0±5.5) and Group 3 was also statistically significant (P=0.018) [Table 1].

Comparison of IGF-1level in Group 1 (218.6±46.2) to Group 2 (285.4±32.8) was statistically signifi cant (P=0.002). Comparison of IGF-1level in Group 1 to Group 3 (280.8±23.5) was also significant (P=0.003). Comparison of IGF-1level in Group 2 to Group 4 (P<0.001) as well as to Group 3 and Group 4 (218.1±48.7)was statistically significant (P<0.001) [Table 1].

Comparison of ST numbers in Group 1 (11.0±9.3) to Group 2 (22.3±14.0) was statistically significant (P=0.005) [Table 1]. Comparison of the mean number of small size ST between Group 1 and Group 2 was statistically non-significant (P=0.201), and was also non significant as regards the big size ST (P=0.907) [Table 1]. Comparison of ST color between Group 1 (flesh 7 / 30.4%, hyperpigmented 2 / 8.7% and mixed-color 14 / 60.9%) and Group 2 (flesh 5 / 26.3%, hyperpigmented 1/ 5.3% and mixed-color 13 / 68.4%) was non-significant (P=0.892). Multivariate logistic regression for the variables associated with the occurrence of ST among the study population was recorded in Table 2. Multivariate regression analysis for the variables affecting the number of ST among the study population was recorded in Table 3.

Table 2

Multivariate logistic regression for the variables associated with the occurrence of ST among the study population

Table 3

Multivariate regression analysis for the variables affecting the number of ST among the study population

The percent of cases of AN in the four groups was recorded in Table 1. The percent of AN cases 3 (13%) / 6 (32%) was higher in Group 2 than in Group 1, this was statistically non-significant (P=0.280) [Table 1]. Insulin resistance in the AN-positive group was higher than in the AN-negative group (10.7±6.1 / 8.7±7.8) though statistically non-significant (P=0.104). IGF-1 was nearly equal in both groups (249.0±44.7 / 248.3±51.8) (P=0.977) [Table 4].

Table 4

IR and IGF-1 in AN-positive and negative groups

Discussion

To our knowledge, this is the first record of a significant decrease in ST number in association with a significant low IGF-1 serum level in HCV-infected subjects.

Skin tags, insulin resistance and IGF-1

Unpredictably, the mean number of ST in Group 1 (ST +ve /HCV +ve) was half the number of ST in Group 2 (ST +ve /HCV -ve) (11.0±9.3 / 22.3±14.0). This was statistically significant (P=0.005). Now the question is: Why was the ST number lower in the HCV-infected group in comparison to the HCV-free group although HCV infection is associated with insulin resistance? In the literature, IGF-1 exerts its major effect on proliferation, while having an effect similar to insulin on differentiation.[18] Thus we suggest that when IGF-1 synthesis is significantly decreased in patients with HCV (as known in the literature), insulin resistance alone may not induce the ST number as like when IGF-1 is normal (an assumption to be solidified by a study on a greater number of patients). Further, ST usually remains without change after insulin resistance normalization.[23] Multivariate regression analysis for the variables affecting the number of ST showed that HCV was borderline but statistically not significant (P=0.053) (this may be due to the low number of participants) [Table 3].

In a recent study Jowkar et al.[24] estimated insulin level and IGF-1 level in non-diabetic subjects with/without ST. They found that the insulin level in subjects with ST was significantly higher in comparison to the control group (18.3±11.2 / 9.7±3.4). On the other hand, IGF-1 in subjects with ST was not significantly higher in comparison to the control group (309.2±102.7 / 296.4±110.4). In our study, on comparing Group 2 (ST+ve / HCV-ve) and Group 3 (ST-ve / HCV-ve) for insulin level (36.9±22.4 / 19.3±8.5), results were similarly significant (P=0.015) and on comparing IGF-I in the same groups (285.4±32.8 / 280.8±23.5) the results were similarly non-significant (P=0.214). The same study[24] revealed and confirmed the importance of insulin in the pathogenesis of ST in every person and concluded that the role of IGF-1 in ST may be questionable. In the current study, the multivariate logistic regression for the variables associated with the occurrence of ST revealed that insulin resistance is the only factor affecting the occurrence of ST [Table 2], however, The significant decrease in ST number in association with a significant decrease in serum IGF-1 level may point to the entrant effect of insulin resistance and IGF-1 in ST development.

Acanthosis nigricans

At the therapeutic level, a randomized trial showed that the use of the oral hypoglycemic drug (rosiglitazone 4 mg/ once daily) in patients with AN resulted in a significant decrease in insulin level but this was not associated with a significant clinical improvement.[25] Again, IGF-1 exerts its major effect on proliferation, while having an effect similar to insulin on differentiation.[18] The obvious doubling of AN in Group 2 in comparison to Group 1 may reflect the entrant effect of IGF-1 with insulin resistance in AN development as in skin tags. Thus we look forward for IGF-1-lowering agents (such as tamoxifen, fenretinine and octreotide)[26] that may help in the management of AN (associated with polycystic ovary)[27] and in the prevention of the recurrence of ST after surgical removal. It is noteworthy that systemic modulation of IGF-1 is fortunately associated with relatively few side-effects in human patients.[28]

Conclusion

HCV is associated with a significant decrease in ST number and in the serum level of IGF-1 together with an obvious decrease in the occurrence of AN. Our results may point to the entrant effect of insulin resistance and IGF-1 in ST and AN development. The evaluation of IGF-1-lowering agents in the control of both diseases (with polycystic ovary) may be the step forward.

Recommendations

The evaluation of IGF-1-lowering agents in the management of AN and ST is being studied now in our department.