Article Text

Download PDFPDF

Extended report
The Framingham Risk Score underestimates the extent of subclinical atherosclerosis in patients with psoriatic disease
  1. Lihi Eder,
  2. Vinod Chandran,
  3. Dafna D Gladman
  1. Centre for Prognosis Studies in the Rheumatic Diseases, University of Toronto Psoriatic Arthritis Clinic, Toronto Western Hospital, Toronto, Ontario, Canada
  1. Correspondence to Dr Dafna Gladman, Centre for Prognosis Studies in the Rheumatic Diseases 1E-410B, University of Toronto Psoriatic Arthritis Clinic, Toronto Western Hospital, 399 Bathurst Street, Toronto, Ontario, Canada M5T 2S8; dafna.gladman{at}utoronto.ca

Abstract

Aim To investigate the usefulness of carotid atherosclerosis assessment in cardiovascular risk stratification of patients with psoriatic disease compared with the Framingham Risk Score (FRS).

Methods Patients with psoriatic arthritis (PsA) and psoriasis alone (PsC), who had no previous history of cardiovascular disease, chronic kidney disease or diabetes mellitus were recruited. They underwent assessment of their cardiovascular risk factors and the FRS was calculated. Based on the FRS, the participants were classified into low, intermediate and high-risk categories. Ultrasound assessment of the carotid artery was performed, and carotid intima-media thickness (cIMT) and total plaque area (TPA) were measured. Patients were stratified into three ultrasound-based risk categories (low, intermediate and high) according to the severity of atherosclerosis. The extent of reclassification from FRS-based category into US-based risk category was assessed.

Results A total of 226 patients with psoriatic disease were assessed. FRS correlated moderately with TPA (r=0.36) and cIMT (r=0.37) and explained only 19% of their variability. 56.1% of the patients in the FRS-based low to intermediate risk groups were found to have carotid plaques. 55.9% of the patients from the FRS-based intermediate risk category were reclassified into an ultrasound-based high-risk category, while 47.1% of the patients in the FRS-based low-risk category were reclassified into a higher US-based risk group. The extent of reclassification into a higher risk category was particularly high among patients with PsA.

Conclusions Ultrasound assessment of subclinical atherosclerosis may improve risk stratification of patients with psoriatic disease, particularly of those with PsA.

  • Psoriatic Arthritis
  • Cardiovascular Disease
  • Ultrasonography
View Full Text

Statistics from Altmetric.com

Introduction

Cardiovascular (CV) morbidity and mortality are increased in patients with cutaneous psoriasis alone (PsC), and in those with psoriatic arthritis (PsA) compared with the general population.1–3 A young man with severe psoriasis is approximately three times more likely to develop myocardial infarction (MI) compared with an age-matched and sex-matched control,4 while the standardised prevalence ratio for MI and angina pectoris among patients with PsA is 2.57 and 1.97, respectively.2 Vascular imaging studies, including carotid ultrasound and 18F-fluorodeoxyglucose (FDG) positron emission tomography (PET/CT), have shown increased atherosclerotic plaques and vascular inflammation in patients with psoriatic disease.5

While traditional CV risk factors that are prevalent in patients with psoriatic disease play an important role in the pathogenesis, they do not fully explain this excessive risk. Other factors, such as chronic systemic inflammation that has been tightly linked to accelerated atherosclerosis partially explain the increased CV risk.6 Lastly, a deleterious effect on the vessel wall of medications used to treat the underlying diseases cannot be ignored.

Population-based risk algorithms, such as the Framingham Risk Score (FRS) are widely used to identify individuals at high risk of developing CV events and to determine the aggressiveness of preventive therapy. The FRS incorporates age, gender, smoking, hypertension, total cholesterol and high density lipoprotein (HDL) cholesterol, to derive an estimated risk of developing a CV event (MI, coronary death and angina) within 10 years. Based on the estimated risk, patients are stratified into a low, intermediate and high-risk groups.7 ,8 However, despite its ability to assign an accurate risk at the population level, the FRS often underestimates the actual CV risk at the individual level.9 This problem is augmented in patients with chronic systemic inflammation where traditional CV risk factors explain only a small proportion of the risk, while other contributing factors are not represented in the risk scores.10 ,11 Despite the increased awareness of the importance of risk assessment for CV risk factors in patients with psoriasis and PsA, evidence-based guidelines developed specifically for managing CV risk factors in these patients are lacking. Accurate psoriatic disease-specific CV risk assessment tools are needed, as clinical CV events often occur in patients who are not considered to be at high risk, probably since additional disease-related risk factors play a role.12

Since atherosclerosis can be considered as the cumulative effect of all known and unknown risk factors within an individual, measuring its extent has the potential of improving risk stratification. Several imaging modalities are currently available to identify subclinical atherosclerosis long before the occurrence of CV clinical events. B mode ultrasound is commonly used to measure carotid intima media thickness (cIMT) and carotid total plaque area (TPA), both can predict the occurrence of clinical events in the general population and among patients with inflammatory arthritis.13–16

In summary, subclinical atherosclerosis, including the presence of endothelial dysfunction and increased incidence of abnormally high cIMT and plaques, has been observed in patients with PsA without classic cardiovascular risk factors.17 ,18 These findings highlight the importance of using non-invasive techniques in patients with PsA and psoriasis. In the present study we aimed to investigate the usefulness of carotid atherosclerosis assessment in CV risk stratification of patients with psoriatic disease compared with the FRS-based risk stratification. In other words, we assessed the extent of reclassification from FRS-based category into ultrasound-based risk category in patients with psoriatic disease.

Methods

Study population

In this cross-sectional study, patients with PsA and PsC were recruited from January 2010 to October 2012. Consecutive adult PsA patients who satisfied the CASPAR criteria19 were recruited from the University of Toronto Psoriatic Arthritis cohort that was previously described in detail.20 In this cohort, patients are followed according to a standard protocol, every 6–12 months. PsC patients were recruited from the University of Toronto Psoriasis Cohort.21 This cohort enrols subjects who have a diagnosis of psoriasis confirmed by a dermatologist, and have been assessed by a rheumatologist to exclude a diagnosis of PsA. All participants are followed according to the same protocol as in the PsA cohort and were assessed annually.

For the purpose of this study, patients who had a history of cardiovascular disease, including MI or angina pectoris were excluded. Additionally, since diabetes mellitus and moderate to severe chronic kidney disease are considered cardiovascular disease risk-equivalent, patients who were treated with antidiabetes medications had fasting glucose levels >7 mmol/L, or had glomerular filtration rate (GFR) of less than 60 mL/min/1.73 m2 were also excluded. The study was approved by the University Health Network Research Ethics Board, and all patients gave their informed consent.

Cardiovascular risk factors assessment

Information about smoking status and the use of lipid-lowering agents and antihypertensive medications were recorded. Participants smoking on a daily basis were considered smokers. Height and weight were measured, and Body Mass Index (BMI) was calculated. Blood pressure was measured with a cuff sphygmomanometer. Blood samples were collected after 12 h of overnight fasting and analysed for glucose, total cholesterol, HDL, low density lipoprotein, triglycerides and sedimentation rate. The ATP III charts were used to calculate the FRS for each participant.8 The risk factors considered were age, gender, smoking, systolic blood pressure, use of antihypertensive medications, total cholesterol and HDL cholesterol. From the Framingham chart, an estimated 10-year CHD risk was generated for each patient. Based on that risk, the patient was classified into one of the following groups: low risk (<5%), intermediate risk (6–20%) and high risk (>20%).

Assessment of disease activity

Disease-related information included: age at diagnosis of psoriasis and PsA, current use of non-steroidal anti-inflammatory drugs (NSAID), disease-modifying antirheumatic drugs (DMARD) and antitumour necrosis factor α (TNFα) agents. Tender and swollen joint counts were assessed on physical examination. Current psoriasis activity was determined by the psoriasis area and severity index (PASI).

Carotid ultrasound assessment

A single trained physician (LE) performed all measurements following the study protocol that was previously described in detail.22 Scans were performed with MyLab 30 and MyLab 70 XVision (Esaote, Genoa, Italy) scanners with a linear LA523 7–13 MHz transducer (Esaote). The scan included detailed B-mode images of both right and left common carotid arteries, as well as the carotid bulb, internal carotid and external carotid arteries. The mean cIMT was measured automatically at the far wall of the right and left common carotid arteries at least 1 cm proximally from the origin of the bulb using a real-time radiofrequency-based ultrasound system for carotid imaging (QIMT tool; Esaote). Then the composite mean cIMT was calculated by averaging the common right and left cIMT values. The result was a single cIMT value per person, expressed in micrometres. An atherosclerotic plaque was defined as a localised intimal thickening exceeding 1 mm. The plaque area was measured by tracing the perimeter of each plaque with a cursor.23 TPA was recorded as the sum of all plaques from the clavicle to the jaw in the right and left carotid arteries expressed in mm2. Reading of the ultrasound scans was performed independently from the scanning. The intraobserver intraclass correlation coefficient for TPA was 0.94.

Due to the lack of accepted cut-off points for TPA and cIMT, data from the Tromsø study, a large population-based cohort, was used to determine cardiovascular risk groups. The ultrasound scanning protocol used in that study was similar to the one used in the present study, and to the best of our knowledge is the only population-based study that assessed cIMT and TPA as predictors of hard clinical outcome, the occurrence of first MI among previously healthy individuals.15 TPAs of more 11.7 mm2 for men and more than 9.5 mm2 for women were associated with a relative risk (RR) of 2.34–2.94 for developing MI. Therefore, high, intermediate and low-risk TPA groups were considered as follows: TPA>11.7 (men) or >9.5 (women), TPA=11.7–0.01 (men) or 9.5–0.01 (women) and TPA=0, respectively. cIMT Of more than 860 µm in men and more than 805 µm in women were associated with a RR of 3.23–4.24 of developing MI. Therefore, high-risk cIMT was considered cIMT >860 µm (men) or >805 µm (women). Composite ultrasound-based risk categories were then formed that incorporated both TPA-based and cIMT-based risk categories: high risk (TPA-based or cIMT-based high risk), intermediate risk (TPA-based intermediate), low risk (TPA-low and cIMT low).

Statistical analysis

Continuous data were described as mean and SD and categorical variables as frequencies and percentages. Comparisons between the risk categories were made using two-tailed t tests or analysis of variance (ANOVA) for continuous variables, and by the χ2 test and the Cochrane–Armitage trend test for categorical variables. Pearson correlation coefficients were used to study the relationship between clinical variables and the imaging results. Weighted κ was used to assess the agreement between FRS-based risk categorisation and ultrasound-based risk categorisation. Two-sided p<0.05 was considered statistically significant. The statistical computation was performed using SAS V.9.2.

Results

From 274 patients targeted for FRS categorisation and ultrasound scanning, patients numbering 48 were excluded due to the following reasons: 15 had a history of CV disease, 19 had diabetes mellitus, 5 had chronic kidney disease and 9 patients due to missing data. Thus, the study population consisted of 226 patients with psoriatic disease (123 PsA and 103 PsC) whose characteristics are detailed in table 1. There were 53.5% males and 46.5% females with a mean age of 51.7±11.8 years. Most of them were Caucasians (87.2%). Their mean duration of psoriasis and PsA were 24.1±14.8 and 14.6±11 years, respectively, with a mean PASI of 3.7±4.9. Among the PsA patients, the mean active and swollen joint counts were 2±5.5 and 0.8±1.9, respectively. Eighty (35.4%) patients out of the entire study population were using NSAIDs, 82 (36.3%) were treated with DMARDs and 60 (26.6%) were current users of anti-TNFα agents.

Table 1

Clinical characteristics by FRS-based category (n=231)

Characteristics of patients by FRS category

Table 1 compares traditional CV risk factors and disease-related variables by FRS category. As expected, age, gender and hypertension differed significantly across the three FRS-based risk categories as these are used to construct the FRS score. Other variables that were associated with higher FRS risk category were having PsA compared with PsC and higher triglyceride level.

Carotid IMT and TPA

The mean TPA and cIMT were 12±18 mm2 and 626±126 µm, respectively, while 98 (43.4%) of the patients had no carotid plaques. Forty-eight patients (21.2%) were classified into the ultrasound-based intermediate risk group, while 82 (36.3%) were in the ultrasound-based high-risk group. Only 14 (6.2%) were classified into a cIMT-based high-risk group. Of those, 10 (71.4%) were already classified in the high-risk group based on the TPA criterion, therefore, cIMT-based risk classification did not add much information to that provided by the TPA. FRS correlated moderately with TPA (r=0.36, p<0.0001) and cIMT (r=0.37, p<0.0001), and explained only a small proportion of the variability in atherosclerosis scores (R2=19% for both cIMT and TPA).

Comparison between FRS and ultrasound risk categories

Overall, the agreement between FRS-based and ultrasound-based risk stratification was low, as measured by weighted κ of 0.17 (95% CI 0.10, 0.24). A significant proportion of the patients in the FRS-based low to intermediate risk groups (124/221, 56.1%) were found to have carotid plaque (TPA>0); 55.9% of the patients from the FRS-based intermediate risk category were reclassified into an ultrasound-based high-risk group, while 26.8% of the patients in the FRS-based low-risk category were reclassified into an ultrasound-based high-risk group, and 20.3% were reclassified into an ultrasound-based intermediate risk group (table 2). The extent of reclassification into a higher risk category was particularly high among patients with PsA, where 26/41 (63.4%) of the patients from the FRS-based intermediate risk category were reclassified into an ultrasound-based high-risk group, and 38/77 (49.5%) of the patients in the FRS-based low-risk category were reclassified into an ultrasound-based higher risk group (figures 1 and 2).

Table 2

Framingham risk score groups versus ultrasound-based risk categories

Figure 1

Framingham risk score classification versus ultrasound-based risk groups in PsA.

Figure 2

Framingham risk score classification versus ultrasound-based risk groups in PsC.

Among the patients who were classified into low and intermediate FRS-based risk groups, age and the use of antihypertensive medications and lipid lowering agents were significantly associated with higher ultrasound-based risk category (table 3). Furthermore, patients with PsA tended to be classified into a higher ultrasound-based risk category (p=0.04), as were patients who used NSAIDs (p=0.03), DMARDs (p=0.03) and TNFα blockers (p=0.03), suggesting that these variables were associated with higher burden of atherosclerosis.

Table 3

Clinical characteristics by ultrasound-based category (n=221)

Discussion

Risk stratification, and the management of cardiovascular risk factors in patients with psoriatic disease, are becoming an integral part of patient care. However, there are limited evidence-based data to guide the management of these patients. In this study, we have found that a clinical-based algorithm, the FRS, is limited in its ability to correctly stratify patients into risk groups that would guide the aggressiveness of treatment of CV risk factors. Most of the patients in the FRS-based intermediate risk group, and almost half the patients in the low-risk group were reclassified into a higher risk group after ultrasound assessment. The extent of underestimation of risk by the FRS was even higher in the PsA group.

Awareness of the increased cardiovascular risk in patients with psoriasis and PsA resulted in recommendations for risk stratification and management of CV risk factors in recently published guidelines for rheumatologists and dermatologists.24 ,25 However, paucity of data that specifically assessed cardiovascular risk stratification in psoriatic patients precluded making any specific recommendations. Studies in the general population have shown that the FRS and other clinical algorithms often underestimate the cardiovascular risk when tested against imaging findings of subclinical atherosclerosis and clinical events.9 ,26 ,27 In rheumatoid arthritis, a task force of the European League Against Rheumatism (EULAR) has proposed to adapt CV risk management calculated according to the systematic coronary risk evaluation (SCORE) function by application of a multiplier factor of 1.5 in those high-risk patients (seropositivity, prolonged disease and extra-articular manifestations).28 However, this modification does not result in a clinically meaningful improvement of risk stratification, thus imaging of atherosclerosis, genetic testing and laboratory biomarkers are being evaluated as strategies for use in combination with traditional CV risk factors to improve the identification of patients at high CV risk.29

Imaging of subclinical atherosclerosis can serve as a supplementary measure to improve risk stratification. In the general population, the presence of carotid artery plaques, and to a lesser extent increased cIMT, were found to significantly augment the predictive ability of clinical risk scores by reclassifying patients into higher or lower risk groups.13 Systemic inflammation is tightly linked to atherosclerosis which results in acceleration of the latter by means that are only partially mediated through traditional CV risk factors. In recent years, it has become clear that inflammation, triggered by an autoimmune response against plaque antigens, such as oxidised LDL, heat shock proteins and, possibly, citrullinated proteins,30–33 leads to atherosclerotic plaque formation and progression. An increase in autoreactive Th1 cells and reduced activity of regulatory T cells resulting in release of proinflammatory cytokines, such as TNFα, IL-1β and IL-6, are shared pathogenic pathways for the development of atherosclerosis, arthritis and psoriasis.34–36 Therefore, it is likely, although not specifically tested in this study, that the problem of underestimation of CV risk is augmented in patients with systemic inflammatory conditions, such as psoriasis and PsA. Indeed, we have found that higher proportions of PsA patients were reclassified into higher risk categories after considering ultrasound information. This finding may be explained by the higher systemic inflammatory burden that entails both skin and joint inflammation in patients with PsA. The suppression of inflammation by immune modulating agents, such as methotrexate, TNFα and IL-6 blockers represent a promising new target for treatment of cardiovascular disease in the general population (CIRT trial),37 and among patients with chronic inflammatory conditions.

Our study was limited by the lack of accepted ultrasound cut-off points for subclinical atherosclerosis risk categories. Different scanning protocols limit arriving at a consensus. The presence of plaques is considered to be a more advanced disease, and has shown a better predictive ability compared to cIMT.23 ,38 Since we have used the TPA that quantifies the extent of plaques, instead of merely recording the presence of plaque, we adopted more stringent criteria for the high-risk group that were based on a population-based study of more than 6000 individuals and provided gender-related cut-off points for both TPA and cIMT that were associated with clinical events.15 Nevertheless, this approach would have to be assessed in longitudinal studies in patients with psoriatic disease. Furthermore, plaque vulnerability that may also affect the risk of developing clinical events, was not assessed in this study. Finally, the lack of clinical outcome data prevents firm conclusions as to the usefulness of ultrasound in cardiovascular risk stratification of patients with psoriatic disease.

In summary, this is the first study that assessed the usefulness of carotid atherosclerosis assessment in risk stratification of patients with psoriatic disease. The results of the study suggest that ultrasound assessment of subclinical atherosclerosis may improve risk stratification of patients with psoriatic disease, particularly of those with PsA. Given the widespread use of ultrasound in rheumatology, this method may be more useful compared with other less accessible and more costly imaging modalities. This approach should be further investigated through longitudinal studies to determine the correlation between ultrasonographic findings and clinical cardiovascular events.

References

View Abstract

Footnotes

  • Handling editor Tore K Kvien

  • Contributors All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. All authors were likewise involved in the study conception and design, acquisition of data, as well as analysis and interpretation of data. DDG had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

  • Funding The Psoriatic Arthritis programme is funded, in part, by The Arthritis Society, Canadian Institutes of Health Research and the Krembil Foundation. The study was funded by a grant from Abbott Laboratories.

  • Competing interests None.

  • Ethics approval University Health Network Research Ethics Board.

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

  • Data sharing statement All authors were involved in the study conception and design, acquisition of data, as well as analysis and interpretation of data.

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.