Article Text

Download PDFPDF

Concise report
Increased arterial wall inflammation in patients with ankylosing spondylitis is reduced by statin therapy
  1. Fleur M van der Valk1,
  2. Sophie J Bernelot Moens1,
  3. Simone L Verweij1,
  4. Aart C Strang1,
  5. Aart J Nederveen2,
  6. Hein J Verberne3,
  7. Michael T Nurmohamed4,
  8. Dominique L Baeten5,
  9. Erik S G Stroes1
  1. 1Department of Vascular Medicine, AMC, Amsterdam, The Netherlands
  2. 2Department of Radiology, AMC, Amsterdam, The Netherlands
  3. 3Department of Nuclear Medicine, AMC, Amsterdam, The Netherlands
  4. 4Departments of Rheumatology Reade, Amsterdam Rheumatology immunology Center, VU University Medical Center Amsterdam, Amsterdam, The Netherlands
  5. 5Department of Clinical Immunology and Rheumatology, AMC, Amsterdam, The Netherlands
  1. Correspondence to Dr Erik S G Stroes, Department of Vascular Medicine, Academic Medical Center, Room F4-211, P.O. Box 22660, Amsterdam 1100 DD, The Netherlands; e.s.stroes{at}


Background Ankylosing spondylitis (AS) is a chronic inflammatory disease with involvement of axial and sacroiliac joints. In addition, patients with AS have increased risk of cardiovascular disease (CVD), which might be attributed to enhanced inflammatory activity of the arterial wall. In the present study, we compared the level of carotid arterial wall inflammation in patients with AS with healthy controls using 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography with CT. As arterial wall inflammation is reduced by statin therapy, we subsequently assessed the effect of 3-month statin therapy on arterial wall inflammation in AS.

Methods and results We included 24 patients with AS (age 44±10, 72% males) without a history of CVD and 20 controls matched for age and gender. Patients with AS had lower high-density lipoprotein cholesterol and increased C reactive protein (CRP) compared with controls. The 10-year CVD risk was 2% in both groups. Notwithstanding, patients with AS had a 20% increase in arterial wall 18F-FDG uptake compared with controls. Three–month atorvastatin 40 mg daily significantly lowered low-density lipoprotein cholesterol (baseline 3.55±1.15 mmol/L, −53%) and CRP (baseline 5.0 (1.5–9.3) mg/L, −58%) with a concomitant decrease of carotid arterial wall inflammation (maximum target-to-background ratio from 1.90±0.30 to 1.67±0.27; p=0.009).

Conclusions Patients with AS and without other CVD risk factors have increased arterial wall inflammation, which decreases upon statin therapy. These subjects are not identified as being at risk in current cardiovascular prevention guidelines. Our data support the need to revise CV disease management in AS, with perhaps a role for early statin therapy.

  • Cardiovascular Disease
  • Inflammation
  • Ankylosing Spondylitis
  • Atherosclerosis
  • Treatment

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Ankylosing spondylitis (AS) is a chronic inflammatory disease, with inflammation of the sacroiliac joints as its most typical symptom.1 Recent population-based studies highlighted an increased morbidity and mortality due to cardiovascular diseases (CVDs) in AS.2 ,3 A potential explanation for accelerated atherosclerosis in AS is the low-grade inflammatory state.4 In patients with rheumatoid arthritis (RA) or psoriasis (PsO), arterial wall inflammation is increased, partially explaining the augmented risk for CVD also found in these diseases.5

18F-fluorodeoxyglucose positron emission tomography with computed tomography (18F-FDG PET/CT) is used to quantify inflammatory activity in atherosclerosis,6 which can be lowered by statins.7–9 Statin therapy also lowers inflammation in non-CVD.10 ,11 In AS, rosuvastatin improved disease activity in patients without elevated levels of low-density lipoprotein (LDL) cholesterol.12 At the level of the arterial wall, rosuvastatin induced atherosclerotic regression (with LDL lowering) in patients with AS13 and improved endothelial function.14 Despite these links between AS and atherosclerosis, recommendations on CVD risk assessment and prevention in AS are limited15; consequently, few patients with AS receive primary CVD prevention.16

In this study, we assessed arterial wall inflammation, using 18F-FDG PET/CT, in patients with AS compared with matched controls. In addition, the effect of 3-month statin therapy on arterial wall inflammation in patients with AS was evaluated.

Materials and methods

Study population and procedures

The study included 24 patients with AS, fulfilling the 1984 Modified New York Criteria for Ankylosing Spondylitis. Twenty healthy control subjects were matched for age, gender and body mass index. 18F-FDG PET/CT scans were performed as published previously.17 Following imaging, a subgroup of patients with AS received 3-month atorvastatin 40 mg daily, after which a repeat scan was performed. Presence of CV risk factors and use of medication were assessed by questionnaire and physical examination. Date of onset of AS, human leucocyte antigen-B27 status and Bath Ankylosing Spondylitis Disease Activity Index 18 were extracted from medical history. Fasting EDTA blood was obtained to measure baseline parameters. Detailed methods are provided in the online supplementary material.

Statistical analysis

Differences in clinical parameters between controls and patients with AS were assessed using an independent t test or Mann–Whitney U test depending on distribution, and differences in frequencies were assessed with χ2 tests. Differences in target-to-background ratio (TBR) were assessed with a univariate analysis of covariance to account for known cardiovascular risk factors: age, gender and smoking status. Values before and after treatment were compared using a paired t test or Wilcoxon signed-rank test. Unadjusted and adjusted linear regression analysis was performed as described in the online supplementary methods. p Values <0.05 were considered statistically significant. Data were analysed using SPSS V.19.0 (SPSS, Chicago, Illinois, USA).


Study population

In total, 44 study subjects were included, comprising 24 patients with AS (aged 44±10) and 20 matched control subjects (aged 48±6). Also, 5 of the 24 patients with AS did not participate in the intervention study, and 1 patient was excluded due to non-compliance. Baseline characteristics were comparable for the whole AS cohort (see online supplementary table S1) and for those participating in the statin intervention (table 1). Whereas total cholesterol and LDL were not different, high-density lipoprotein cholesterol was lower and C reactive protein (CRP) and leucocytes were increased in patients with AS compared with controls. The 10-year CV risk, assessed with the Framingham Risk (FHR) score, did not differ between patients with AS and healthy controls (average 2% in both groups). AS-specific characteristics are summarised in table 2 and online supplementary table S2.

Table 1

Clinical characteristics of study subjects

Table 2

AS disease-specific characteristics of statin cohort

Increased arterial wall inflammation in AS is reduced by statins

In patients with AS, the whole carotid and the most diseased segment (MDS) showed approximately 20% higher 18F-FDG uptake compared with controls (see figure 1 and online supplementary table S3).

Figure 1

Quantification of the 18F-fluorodeoxyglucose uptake as the mean and maximum target to background ratio (TBR) in and the whole carotid arteries (D and E), and TBR in the most diseased segment (MDS) (F) revealed increased uptake in patients with ankylosing spondylitis (AS) (n=18) compared with control subjects (n=20) (represented in A and B), which decreased after statin therapy (represented in C). **p<0.01, ***p<0.001. CA, carotid artery; JV, jugular vein.

After 3 months of statins, which were well tolerated, we observed a 53% reduction in LDL cholesterol. CRP was reduced by 58%, with the number of patients above upper limit of normal declining from 50% to 24% (table 1). With respect to arterial wall inflammation, 18F-FDG uptake was reduced with 12% in both the whole artery (p=0.009 vs pretreatment) and the MDS (p=0.011) (figure 1).

There was no correlation between baseline TBR and lipids or inflammatory markers in patients with AS (see online supplementary table S4). However, baseline total cholesterol and LDL levels as well as change in LDL after treatment significantly correlated to the change in TBR (ΔTBR), after adjustment for age, gender, smoking status and non-steroidal anti-inflammatory drug use (table 3). Each mmol/L decline in LDL cholesterol resulted in a 26% decrease in TBR (R: 0.488, p=0.065). None of the other parameters correlated with TBR change (see online supplementary table S5).

Table 3

Unadjusted and adjusted linear regression analysis with ΔTBR max as the dependent variable


We report that relatively young patients with AS, with low disease activity, have increased arterial wall inflammation compared with matched control subjects, despite comparable 10-year CVD risk assessed with the FHR score. In addition, we show that in patients with AS, statins reduce both lipids and inflammatory markers, coinciding with a decrease in arterial wall inflammation. These findings highlight the need for improved characterisation of CVD risk in patients with AS. Since statins reduced arterial wall inflammation, these data further support early statin intervention in patients with AS.

Previous studies described an increased carotid intima-media thickness in patients with AS,15 ,19 which diminished upon rosuvastatin therapy as detailed in the Rosuvastatin in Rheumatoid Arthritis, Ankylosing Spondylitis and Other Inflammatory Joint Diseases (RORA-AS) study.13 Here, we show an increased inflammatory activity in the arterial wall in young patients with AS, as was previously reported for RA and PsO.5 ,20 Also, we report that after 3 months of statins, arterial wall inflammation was significantly reduced, accompanied by LDL and CRP lowering. Since increased carotid 18F-FDG uptake has shown to predict future CVD events,17 this suggests that increased CVD risk in patients with AS is reduced by statins.

We also evaluated correlations between (change in) inflammatory/lipid measures and arterial 18F-FDG uptake. Interestingly, although CRP significantly decreased upon statin treatment, neither baseline CRP nor the decline in CRP after treatment predicted the change in TBR. This corroborates previous observations, showing no clear relation between CRP levels and arterial wall inflammation at baseline,17 ,21 or after statins.22 Conversely, LDL cholesterol did significantly predict change in TBR at baseline, with a concomitant correlation for decrease in LDL and decrease in TBR. These data suggest that LDL lowering, even within physiological baseline levels, dictates the decline in arterial wall inflammation. Interestingly, we previously showed that LDL lowering by apheresis also decreased arterial wall inflammation, without affecting CRP.23 Moreover, in a subset of patients this effect was found after a single session of apheresis. This very rapid effect (arterial wall inflammation was measured at a median of 3 days after apheresis) may be at least partially attributable to improved endothelial function, which was also shown as an effect of statins in a substudy of the RORA-AS study.14

Several limitations need to be addressed. First, this pilot study lacked a placebo group; hence, we cannot address the natural course of arterial wall inflammation in AS. However, the decline in arterial wall inflammation was primarily mediated by LDL lowering, which is not expected in a placebo arm.10 Second, we provided short-term treatment, whereas CVD develops over years. However, in previous studies, sustained effects were found at 18 months13 ,14 and arterial wall inflammation can be expected to follow this course. Finally, our study provides no outcome data concerning CV events, only providing surrogate markers in a limited (Caucasian) population, restricting wider extrapolation. However, 18F-FDG PET seems to be of prognostic value in assessing CVD17 and can therefore be a valuable biomarker in diseases with relatively low numbers of patients, rendering outcome studies difficult and time consuming.

Collectively, we show that patients with AS have increased arterial inflammatory activity. The European League Against Rheumatism formulated recommendations for cardiovascular risk screening and management in patients with RA, AS and psoriatic arthritis.15 In patients with RA, with the most compelling evidence for increased CV risk, annual risk assessment is recommended and risk must be multiplied by a 1.5 factor when at least two out of the following characteristics are present: disease duration of >10 years, rheumatoid factor (RF) or anti-citrullinated peptide (CCP) positivity or the presence of certain extra-articular manifestations. Our results support a similar recommendation in AS as we show elevated arterial wall inflammation in young patients with AS, despite normal LDL levels, who according to current guidelines do not qualify for CV prevention. Hence, this arterial inflammatory activity in AS underscores the need to revise CV disease management in AS, aiming to evaluate the need for early statin therapy comparable to RA.


The authors thank MF Lam and ME Hemayat for their assistance with 18F-FDG PET/CT.


View Abstract


  • Handling editor Tore K Kvien

  • FMvdV and SJBM contributed equally to this paper.

  • Contributors FMvdV and SJBM contributed equally to the article. FMvdV, DLB and ESGS designed the study. FMvdV, SJBM, SLV and ACS performed the research. MTN and DLB assisted in patient recruitment. AJN and HJV provided analytic tools. FMvdV, SJBM and ESGS drafted the paper. All authors critically reviewed the manuscript.

  • Funding This work was supported by a European Horizon 2020 grant (PHC-03-2015: 667837: REPROGRAM) and the Dutch Heart Foundation (CVON-2011-19: GENIUS).

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval Institutional review board of the AMC, Amsterdam, the Netherlands,.

  • Provenance and peer review Not commissioned; externally peer reviewed.