Subclinical atherosclerosis in patients with hidradenitis suppurativa treated with TNF inhibitors

Abstract

Background

Hidradenitis suppurativa (HS) is an independent risk factor for the development of subclinical atherosclerosis. Tumour necrosis factor (TNF) inhibitors are effective for the treatment of recalcitrant moderate‐to‐severe HS. However, the effect of treatment with TNF inhibitors on subclinical atherosclerosis in HS patients has not been previously investigated.

Objectives

In this study, we aimed to assess changes in biochemical parameters (fasting blood glucose and lipid levels) and carotid intima‐media thickness (CIMT) values in Hurley stage II and III HS patients undergoing treatment with TNF inhibitors.

Methods

This was a single center prospective study including 30 patients with Hurley stage II and III HS and 30 healthy controls (HCs). Baseline values of biochemical parameters and CIMT were compared to the values recorded after at least 6 months of TNF inhibitor therapy.

Results

CIMT values of the HS patients significantly exceeded those of HCs (for right p = 0.011 and for left p = 0.017). After at least 6 months of TNF inhibitor therapy, there was a statistically significant decrease in fasting blood glucose (p = 0.001), whereas total cholesterol levels significantly increased (p = 0.001). CIMT values also significantly increased (for right p = 0.02 and for left p = 0.01).

Study limitations and conclusions

Small sample size is limitation of the current study. Our study shows that patients with Hurley stage II and III HS undergoing TNF inhibitor therapy are under risk for progression of subclinical atherosclerosis.

1. INTRODUCTION

Hidradenitis suppurativa (HS) is a chronic inflammatory disease presenting with recurrent abscesses, nodules and scars in intertriginous areas. ¹ Multiple comorbidities have been related to moderate‐to‐severe HS such as diabetes, and metabolic syndrome. ²

Carotid intima‐media thickness (CIMT) reliably indicates generalized atherosclerotic process and is a predictor of vascular events. ³ Recent studies evaluating CIMT values in HS reported that moderate‐to‐severe HS is an independent risk factor for premature subclinical atherosclerosis. ⁴ , ⁵ In addition, markers of endothelial dysfunction such as serum levels of angiopoietin‐2 and asymmetric dimethylarginine are increased in patients with HS. ⁶

Tumour necrosis factor (TNF) inhibitors, adalimumab and infliximab, are effective treatment options in recalcitrant moderate‐to‐severe HS. ⁷ , ⁸ However, no previous studies on HS have investigated whether TNF inhibitors affect the cardiovascular risk.

In this study, we aimed to assess the effect of therapy on the progression of subclinical atherosclerosis in Hurley stage 2 and 3 HS patients undergoing treatment using TNF inhibitors. We also aimed to evaluate the difference in biochemical parameters such as fasting blood glucose levels and lipid levels between baseline and last visit.

2. MATERIALS AND METHODS

2.1. Patients

We recruited a total of 30 Hurley stage II and III HS patients and 30 age/sex‐matched healthy controls (HCs) in our single center, prospective study. The study was approved by the institutional review board.

Recruited HS patients received biologic monotherapy (adalimumab or infliximab) for at least 6 months in our tertiary referral dermatology outpatient clinic. Patients were followed‐up until they stopped TNF inhibitor therapy. All patients were irresponsive to conventional treatment modalities such as antibiotics and systemic retinoids or were intolerant to these therapies. Exclusion criteria were individuals with malignancies, diabetes mellitus, chronic kidney disease or any other known systemic inflammatory disorder and patients receiving statins or other lipid‐lowering drugs.

In order to exclude primary nonresponders, we excluded HS patients who undergone TNF inhibitor therapy for less than 6 months. In addition, at least 3 months of treatment are required to see meaningful changes in CIMT values. ⁹

Patients’ demographic and clinical characteristics were recorded. Fasting blood glucose levels, fasting serum lipids including high‐density lipoprotein (HDL), low‐density lipoprotein (LDL) and triglyceride (TG) levels were retrieved from patient records at baseline and measured at the last visit.

2.2. Ultrasound studies

Two independent radiologists (GK and DA) recorded CIMT values using B‐mode ultrasound. Examinations were carried out with patients lying in a supine position, neck extended and rotated at 45 degrees opposite the physician. Measurements were obtained on the common carotid artery at 1 cm from the bifurcation at T wave of the cardiac cycle.

2.3. Statistical methodology

Independent‐samples T‐test or Mann–Whitney U test were used to compare CIMT and the traditional cardiovascular risk factors of two groups, where appropriate. A multiple linear regression model was used to determine clinical and biochemical parameters that associated with increased CIMT in HS patients. Comparisons of first and last biochemical parameters and ultrasonographic measurements of the patients were preformed ussing Wilcoxon Signed Rank Test. Statistical analyses were performed on SPSS version 21.0. Statistical significance (alpha) level was chosen as 0.05.

3. RESULTS

The demographic and clinical characteristics of the 30 recruited patients and 30 HCs can be seen in Table 1. Disease duration of the patients was 9.57 ± 6.61 years (range: 1–25 years). CIMT values of the HS patients exceeded those of HCs (for right p = 0.011 and for left p = 0.017). Among other biochemical and clinical parameters, only blood glucose levels (p < 0.0001), waist circumference (p = 0.008), and BMI (p = 0.01) differed among HS patients and HC Table (1). In the multivariable linear regression model adjusted for age, sex, and traditional cardiovascular risk factors, only the duration of the disease was significantly associated with increased CIMT (r = 0.56; p = 0.012).

TABLE 1.

Demographics, clinical, biochemical, and ultrasonographic findings of the patients and healthy controls with univariate analyses.

Parameters	Patients	Healthy controls	Univariate analyses p value
Age	34.8 ± 13	35.67 ± 9.78	NS
Gender male/female	17/13	15 /15	NS
Current smoker yes/no	21/9	17/13	0,28
Systolic blood pressure (mm/Hg)	125.57 ± 7.61	120.87 ± 8.86	NS
Diastolic blood pressure (mm/Hg)	77.80 ± 10.51	75.20 ± 20.23	NS
Waist circumference (cm)	95.43 ± 14.05	85.30 ± 14.6	0.008
Blood glucose level (mg/dL)	88.73 ± 9.61	76.30 ± 11.14	<0.0001
R‐CIMT (mm)	0.58 ± 0.24	0.46 ± 0.10	0.011
L‐CIMT (mm)	0.61 ± 0.31	0.46 ± 0.11	0.017
Average CIMT	0.60 ± 0.26	0.46 ± 0.10	0.013
BMI (kg/m2)	28.1 ± 4.6	24.97 ± 4.81	0.01
LDL	109.83 ± 36.26	108.4 ± 30.07	NS
HDL	44.87 ± 13.57	50.57 ± 13.28	NS
Total cholesterol	168.6 ± 38.01	174.47 ± 30.51	NS

Abbreviations: CIMT, carotid intima‐media thickness; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; NS, not specific.

*All variables are expressed as mean ± SD unless otherwise specified.

Twenty‐six patients were receiving adalimumab (at HS protocol starting with 160 mg s.c. injection), and four patients were receiving infliximab (at psoriasis protocol as 5 mg/kg i.v. infusion). Mean biologic therapy duration was 14.2 ± 6.7 months (range: 6–24 months). Changes in biochemical parameters and CIMT measurements during the study period can be seen in Table 2. Blood levels of TGs (p = 0.82), LDL cholesterol (p = 0.87), and HDL cholesterol (p = 0.32) did not differ significantly. There was a statistically significant decrease in fasting blood glucose (p = 0.001), whereas total cholesterol levels significantly increased (p = 0.001). CIMT values also significantly increased (for right p = 0.02 and for left p = 0.01).

TABLE 2.

Changes in biochemical parameters and CIMT measurement of HS patients during the study period.

Parameters	Baseline (mean)	Follow‐up (mean)	p Value
Glucose (mg/dL)	88.73	85.77	0.001
Triglycerides (mg/dL)	103.43	102.70	0.82
Total cholesterol (mg/dL)	168.60	172.50	0.001
LDL cholesterol (mg/dL)	109.83	110.70	0.871
HDL cholesterol (mg/dL)	44.87	45.13	0.323
CIMT‐right (mm)	0.587	0.747	0.002
CIMT‐left (mm)	0.617	0.650	0.008

Abbreviations: CIMT, carotid intima‐media thickness; HDL, high‐density lipoprotein; HS, hidradenitis suppurativa; LDL, low‐density lipoprotein; NS, not specific.

4. DISCUSSION

Chronic inflammation confers a risk for the development of atherosclerosis. ¹⁰ To date, many inflammatory diseases such as chronic psoriasis, rheumatoid arthritis, and systemic lupus erythematosus have been shown to accelerate the subclinical atherosclerosis process. ¹¹ , ¹² , ¹³ Similarly, HS has been linked to increased prevalence of metabolic syndrome and premature atherosclerosis. ² The inflammatory load in HS appears to exceed that of patients with psoriasis. ¹⁴ In our study, compared with HCs, patients with HS had significantly increased CIMT values, as expected. This demonstrated the presence of subclinical atherosclerosis in our study cohort of moderate and severe HS.

Effect of systemic and biologic therapies on CIMT levels was previously evaluated in patients with various diseases. Di Minno et al. compared CIMT values of psoriatic arthritis patients undergoing TNF inhibitor therapy with those undergoing conventional treatment. They reported a decrease in CIMT level of patients receiving TNF inhibitors. ¹⁵ In contrast, Mazzocolli et al. found no change in CIMT levels after 12 weeks of TNF inhibitor therapy, and Ramonda et al. reported progression of atherosclerosis in patients with psoriatic arthritis undergoing TNF inhibitor therapy for 2 years. ¹³ , ¹⁶ Jókai et al. performed a 6‐month longitudinal evaluation of 13 psoriasis patients where TNF inhibitor therapy significantly decreased CIMT levels. ¹⁷ However, Martinez‐Lopez et al. presented a prospective study involving 53 psoriasis patients where statistically significant improvements in CIMT levels were recorded in patients undergoing IL‐12/23 inhibitors and methotrexate but not TNF‐inhibitors. ¹¹ Lastly, a recent article systematically reviewed the studies about the effect of TNF inhibitors on subclinical atherosclerosis and concluded that there is no strong evidence for TNF‐αantagonism to have effects on subclinical atherosclerosis. ¹⁸ Consistent with the recent review, in our study, ultrasonographic evaluation of the CIMT revealed statistically significant progression in moderate to severe HS patients despite treatment with TNF inhibitors for at least 6 months.

There was a significant decrease in glucose levels of our patients. This was consistent with previous studies reporting decreased insulin resistance and glycemia in psoriasis patients receiving TNF blockers. ¹¹ , ¹⁹ However, adalimumab did not significantly affect lipid parameters in patients with psoriasis. ²⁰ Similarly, we did not identify any significant changes in TGs, LDL cholesterol and HDL cholesterol levels. However, serum total cholesterol levels significantly increased in our patients.

Recent studies have reported increased expression of IL‐17 in HS, which have implications for future therapies. ²¹ Moran et al. showed a significantly enriched IL‐17 producing cells in lesional skin from HS patients and demonstrated that lesional skin from TNF inhibitor treated HS patients had substantial reduction of Th17 cells. ²² Serum level of IL‐17 in HS patients was significantly high and correlated with severity of the disease. ²³ Effective treatment of HS with secukinumab was reported in several recent studies. ²⁴ , ²⁵ In a pilot study, IL‐17 inhibition did not significantly affect CIMT values in psoriasis patients. ²⁶ However, data are lacking about the impact of IL‐17 antagonism on vascular structure in patients with HS.

Our study is limited by its small sample size and lack of a patient group treated with conventional therapies or HS patients without any therapy. However, in severe HS, systemic treatments are indicated, and there are nearly no severe HS patients followed without therapy. Thus, the effect of the disease itself on CIMT progression is not known. In our study it is not possible to definitely conclude whether TNF inhibitors have neutral, negative or even positive role on observed increase of CIMT during treatment with TNF inhibitors. Nevertheless, we have observed that there is significant progression of subclinical atherosclerosis in severe HS patients undergoing TNF inhibitor therapy.

5. CONCLUSION

In conclusion, patients with Hurley stage II and III HS, undergoing TNF inhibitor therapy are still under significant risk for progression of subclinical atherosclerosis. We highlight that treatment strategies should address the increased cardiovascular risk in HS. Larger prospective trials should be conducted to confirm the effects of systemic treatments on premature atherosclerosis in HS.

Oba MC, Askin O, Gunver MG, Kocaarslan G, Alis DC, Engin B. Subclinical atherosclerosis in patients with hidradenitis suppurativa treated with TNF inhibitors. Skin Res Technol. 2023;29:1–4. 10.1111/srt.13302

DATA AVAILABILITY STATEMENT

None