Insulin resistance in hidradenitis suppurativa: a case–control study

The association between chronic inflammatory diseases, such as rheumatoid arthritis and psoriasis, and insulin resistance (IR) has been well established. Hidradenitis suppurativa (HS) is a chronic inflammatory cutaneous disease that affects the apocrine gland‐bearing areas of the body.


Introduction
Hidradenitis suppurativa (HS) is a chronic, inflammatory skin disease of the hair follicle characterized by relapsing painful inflammatory nodules, abscesses and fistula tracts in the apocrine gland-bearing areas of the body, most commonly in the axillae, inguinal and anogenital regions. [1][2][3] HS incidence ranges between four and ten cases per 100 000 population/year and prevalence between 0.2% and 4%. 3 Although its pathogenesis is not completely understood, it is postulated to begin with disturbed keratinization of the follicular infundibulum which results in follicular occlusion. 4 Current knowledge indicates that inflammation induced by abnormal immune response plays an important role in the pathogenesis of HS. 5 To this respect, several investigations have shown an increased expression of proinflammatory cytokines, such as tumour necrosis factor alpha (TNF-a) and interleukin (IL) IL-1b in HS lesions. 6 Moreover, the presence of elevated serum TNF-a 7 and IL-6 8 levels in HS patients has been reported, suggesting a systemic inflammatory activation as a pathogenic mechanism for the disease.
Hidradenitis suppurativa is currently recognized as a systemic inflammatory condition extended beyond the skin, and there is an increasing evidence of an association between HS and other comorbidities, including cardiovascular (CV) risk factors, inflammatory joint disorders, inflammatory bowel disease, endocrinological disorders and psychological disorders. [9][10][11][12] In a recently population-based cohort study, HS was associated with a significantly increased risk of major adverse CV events, including myocardial infarction, ischaemic stroke and CV-associated death. 13 In keeping with these findings, we and others have recently demonstrated an increased prevalence of subclinical atherosclerosis in HS patients. 14,15 Furthermore, HS has also been associated with a greater prevalence of CV risk factors, such as diabetes mellitus (DM), dyslipidaemia, cigarette smoking, metabolic syndrome (MS) and obesity. 16 However, it should be noted that the premature and accelerated development of atherosclerosis in patients with HS is independent of these CV risk factors. 14,15 This fact suggests that HS itself may be an independent risk factor for atherosclerotic CV disease, and that other disease-related factors may be implicated in the premature development of the atherogenic process in HS patients.
Insulin resistance (IR) represents the best predictor of type 2 DM and plays a central role in the high CV risk of the MS. 17 The metabolic changes induced by IR may contribute to the development of accelerated atherosclerosis and CV disease. This occurs by increasing serum glucose and insulin concentrations, but also through mechanisms that involve dyslipidaemia, hypertension and systemic inflammations. 18,19 In this sense, IR is closely related to the presence of a chronic systemic inflammation and has been linked to several chronic inflammatory conditions such as rheumatoid arthritis (RA), 20 systemic lupus erythematosus (SLE) 19 and psoriasis. [21][22][23] Thus, it has also been postulated that IR might explain the increased CV comorbidity associated with these inflammatory diseases.
Taking all of these considerations into account, the aim of this study was to analyse, for the first time to our knowledge, the prevalence of IR in a relatively large, ethnically homogeneous cohort of HS patients. Furthermore, we sought to assess whether there is a correlation between IR and the severity of the HS.

Study participants and protocol
This was a cross-sectional, case-control study that included 76 patients with HS and 61 age and gender-matched controls. The HS patients were recruited from our Dermatology outpatient clinic at the University Hospital Marques de Valdecilla (Santander, Northern Spain). The diagnosis of HS was always performed by dermatologists based on clinical findings, including a history of recurrent inflamed and non-inflamed nodules, abscesses and sinus tracts involving of typical skin areas. The control group consisted of hospital medical staff and of subjects who attended the dermatology department due to skin disorders other than HS, such as melanocytic naevus, warts or epithelioma.
The exclusion criteria for both groups were as follows: (i) age <18 years; (ii) documented history of major adverse cardiovascular events; (iii) type 1 or type 2 DM; (iv) chronic kidney or liver diseases; (v) diseases that might influence glucose metabolism (such as Cushing syndrome, thyroid disorders, polycystic ovary syndrome. . .); (vi) treatment with drugs that might affect carbohydrate metabolism (i.e. systemic corticosteroids, retinoids, cyclosporine, hypoglycaemic drugs. . .) in the previous 6 months; and (vii) concomitant inflammatory disorders, such as cutaneous diseases (psoriasis or atopic dermatitis among others), inflammatory bowel disease, inflammatory arthritis (RA or ankylosing spondylitis), or autoimmune or connective tissue diseases (scleroderma, SLE. . .).
The study protocol was approved by the local institutional ethics committee, and all the participants provided informed written consent.
The severity of HS was assessed by the HS Physician Global Assessment (HS-PGA), 24 which includes six stages (scale 0-5) ( Table 1). According to HS-PGA and as we have previously described in detail, 14 HS was classified as moderate-severe-very severe (PGA > 3) and as minimal-mild HS (PGA < 3). The Hurley grade of disease severity for each zone involved at the time of clinical examination was also evaluated. Moreover, information on previous treatment for HS was also assessed in all the patients.
All the participants provided information on their demographic features, past medical history and data on current and Table 1 Hidradenitis suppurativa-Physician 0 s Global Assessment (HS-PGA) No abscesses, no draining fistulae, no inflammatory nodules, and no noninflammatory nodules  25 waist circumference >102 cm in men or >88 cm in women; hypertriglyceridaemia ≥150 mg/dL or lipid lowering treatment; HDL-c <40 mg/dL in men and <50 mg/dL in women; BP ≥130/85 mmHg or current use of medication for hypertension; fasting plasma glucose ≥110 mg/dL or use of antidiabetic drugs.
The homeostatic model assessment for IR (HOMA-IR) was calculated using this formula: fasting insulin level (lIU/ mL) 9 fasting glucose level (mg/dL)/405. IR was defined as an elevated HOMA-IR value of >2.5, based on the original HOMA research. 26

Statistical analysis
Results were expressed as number (percentage), mean AE standard deviation (SD) or median and interquartile range (IQR), as appropriate. Mann-Whitney U-test was performed to compare quantitative variables and chi-squared test or Fisher exact test, to compare qualitative variables. Multivariable general linear regression models, using HOMA-IR levels as dependent variable, were performed to assess the relationship between HS and IR. A P value <0.05 was considered statistically significant. All the analyses were performed with the Stata V.12/SE package (Stata Corp, College Station, Texas, USA).

Results
A total of 137 participants, 76 patients with HS and 61 control subjects were included in the study. Table 2 shows the main demographic, clinical and laboratory findings of patients and controls. Patients and controls were similar in age and sex (51% and 49% women, respectively, P = 0.80). According to HS-PGA, 32 (42.1%) patients were classified as having minimal-mild HS (HS-PGA score < 3) and the remaining 44 (57.9%) as having moderate-severe-very severe HS (HS-PGA score ≥ 3). Nineteen HS patients (25%) were classified as Hurley stage I, 44 (57.8%) as stage II and 13 (17.1%) as stage III. Twenty-three HS patients (30.2%) were on anti-TNF-a agents.
There was no significant statistical difference between patients and controls in terms of height and triglyceride levels. The HS group had significantly higher weight, BMI, waist perimeter, systolic and diastolic BP than controls. Moreover, HS group showed lower serum HDL levels than the control group (P < 0.05).
Noteworthy, fasting serum glucose and insulin levels, as well as HOMA-IR values, were higher in HS patients than in controls (Fig. 1). As seen in the Table 2, IR, defined as a HOMA-IR value greater than 2.5, was observed in 43% of cases and 16 % of controls (P = 0.001). Besides, the prevalence of MS was three times more common in HS patients than in control subjects.
In the general linear regression multivariable analysis after adjusting for age, sex and BMI, HS remained as a significant factor for HOMA-IR [mean (SE) 2.51(0.18) vs 1.92(0.21); P = 0.04].
To determine whether the severity of the HS had any impact on the association between HS with both IR and MS, we divided the HS patients into two subgroups, according to their HS-PGA score (>3 or <3). There was no significant association between disease severity and the prevalence of the MS and/or the HOMA-IR value. Although IR was higher in patients with moderate-severe HS (52.3%) than in those with HS-PGA score <3 (31.3%), this difference did not reach statistical significance (P = 0.06) ( Table 3).

Discussion
The current study shows that patients with HS have a significantly higher prevalence of IR compared to age-and sexmatched controls. Moreover, and in accordance with other previous reports, [27][28][29] we found in our series of HS patients a significantly higher prevalence of MS than in the control group. Moreover, patients with HS were more obese and had higher prevalence of hyperglycaemia, lower HDL-cholesterol levels and raised systolic and diastolic BP than control subjects. Nevertheless, no correlation was observed between the severity of HS and both IR and MS.
Our results suggest that HS may predispose to the development of IR. In this sense, we consider that the high prevalence of obesity found in our HS patients might have influenced the obtained results, as it is well known that obesity can impair the effects of insulin. However, even after adjustment for BMI, age and sex, the HOMA-IR value remained significantly higher in HS patients than in controls. This fact suggests that HS itself may be an independent risk factor for the development of IR.
A possible explanation for the link between HS and IR could be the presence of chronic inflammation that occurs due to persistent secretion of several proinflammatory cytokines, such as TNF-a and IL-6. 7,8 In this regard, TNF-a is considered nowadays one of the major pathogenic factors for HS as demonstrated by the efficacy of the anti-TNF-a agent adalimumab in the treatment of this disease. 3,24 Furthermore, it is known that TNF-a plays a key role in the impairment in glucose tolerance and insulin sensitivity. 30,31 Thus, TNF-a is able to induce insulin signalling defects by acting on adipocytes and muscle cells, impair insulin signalling through inhibition of tyrosine kinase activity of the insulin receptor, and suppress the secretion from adipocytes of adiponectin, an anti-inflammatory molecule that also regulates insulin sensitivity. 31,32 To this respect, Malara et al. 33 recently reported that patients with HS have significantly decreased the serum adiponectin levels.
On the other hand, it has been shown that IL-6 induces IR in hepatocytes. 34 In this scenario, the increased inflammatory burden in HS could alter glucose metabolism and induce IR. IR may in turn contribute to endothelial dysfunction, leading subsequently to atherosclerosis and finally to long-term clinical events, such as myocardial infarction or stroke.
Eddy et al. 35 observed that IR prevention may reduce myocardial infarction risk by 42% in young adults. Thus, it is worth emphasizing that several studies have demonstrated a beneficial effect of the TNF-a blockade on the mechanisms of accelerated atherogenesis in patients with chronic inflammatory diseases, such as RA or psoriasis, including specifically the effect of these agents on IR. 20,36 In this regard, it should be noted that about 30% of our patients were on adalimumab therapy. Anti-TNF-a therapy may therefore have improved the insulin sensitivity in these patients, underestimating the actual prevalence of IR in HS. On the other hand, it is also noteworthy that several studies have shown that metformin, an antihyperglycemic agent used for treating type 2 DM, could be beneficial in the treatment of HS. 37,38 Metformin decreases hepatic glucose production and improves insulin sensitivity by increasing peripheral glucose uptake and utilization. Although metformin in HS acts by a mechanism yet unknown, it has been suggested that the beneficial effects of this drug might be, at least in part, through lowering the IR that is present in some patients with HS. 38 Our study has the inherent limitations of its cross-sectional design. Despite these limitations, we think that it provided important insights into the association of HS and IR that may have clinical implications for the overall management of these patients.
In conclusion, our study provides evidence that patients with HS have a significant higher prevalence of IR than controls. There is growing evidence that HS is not only a cutaneous disease but also a chronic systemic inflammatory disorder with increased CV risk. Thus, it is important to control known modifiable CV risk factors according to current guidelines, as well as  monitoring serum glucose levels in patients with HS. Besides its clinical benefit, treatment with anti-TNF-a agents can provide an effective treatment strategy for the prevention and control of the development of IR. Nevertheless, further investigations are needed to elucidate the beneficial effects of biological therapy on IR in patients with HS.