International RA risk tool for CVD comes closer to reality

An international collaboration of experts has developed a cardiovascular risk calculator specifically for rheumatoid arthritis patients that has the potential to become part of routine rheumatology clinical practice in many parts of the world.

The ATACC-RA (A TransAtlantic Cardiovascular Risk Calculator for Rheumatoid Arthritis) consortium will be constructed from data collected at 13 rheumatology centers in 10 countries, Elke Arts said at the annual European Congress of Rheumatology.

The current version, which is still undergoing validation, contains data from eight rheumatology centers in seven countries (Greece, the Netherlands, Norway, South Africa, Sweden, the United Kingdom, and the United States), said Ms. Arts, a rheumatology researcher at Radboud University Medical Centre in Nijmegen, the Netherlands.

So far, ATACC-RA contains pooled data from 3,176 RA patients without cardiovascular disease. Individual data were collected on CV risk factors and outcomes and then combined with RA-specific information, such as disease duration, seropositivity, rheumatoid factor and/or anti-citrullinated protein antibodies, 28-joint Disease Activity Score (DAS28), and acute phase reactants.

During an average of almost 8 years of follow-up comprising 24,733 patient-years, 314 patients developed cardiovascular disease (CVD).

Two possible models came out of multivariable risk score modeling for predicting the RA-specific 10-year risk of CVD. Each incorporated the traditional risk factors of age, sex, current smoking, presence of hypertension, and ratio of total cholesterol to HDL cholesterol, but also either seropositivity or DAS28. The consortium specifically chose to assess potential risk factors that would be easily available to the health professional regardless of the setting (for example, primary, secondary, and tertiary care, and even private practice).

The models that included seropositivity or DAS28 both demonstrated good discrimination and calibration, when compared with the Framingham or SCORE (Systematic Coronary Risk Evaluation) algorithms. The models also showed good concordance with each other (see chart).

"By pooling data from many centers, it appears possible to develop an RA-specific CVD risk algorithm which is more accurate at predicting CVD in people with RA than the currently available risk algorithms, which have been developed for the general population," members of the consortium wrote in response to e-mailed questions.

The consortium is now working to validate the ATACC-RA calculator, with the ultimate aim of validating it in completely independent cohorts. The continued growth of the consortium will help make this a reality within the next couple of years. The end result may be a risk calculator that can be adapted to account for the underlying risk of each patient population.

Once the algorithm is validated, the consortium hopes it will become part of routine rheumatology clinical practice, much the same as the Framingham or SCORE algorithms are used currently in the general population, but more specifically applied to patients with RA.

"The models proved to be quite robust," Ms. Arts said. "If this holds true through the additional analysis and external validation, we may have one that will be applicable to a wide variety of patients all over the world."

Once fully validated, the investigators plan to produce the tool in a user-friendly application for computers or even smartphones.

The investigators had no conflicts of interest to declare.

An international collaboration of experts has developed a cardiovascular risk calculator specifically for rheumatoid arthritis patients that has the potential to become part of routine rheumatology clinical practice in many parts of the world.

The ATACC-RA (A TransAtlantic Cardiovascular Risk Calculator for Rheumatoid Arthritis) consortium will be constructed from data collected at 13 rheumatology centers in 10 countries, Elke Arts said at the annual European Congress of Rheumatology.

The current version, which is still undergoing validation, contains data from eight rheumatology centers in seven countries (Greece, the Netherlands, Norway, South Africa, Sweden, the United Kingdom, and the United States), said Ms. Arts, a rheumatology researcher at Radboud University Medical Centre in Nijmegen, the Netherlands.

So far, ATACC-RA contains pooled data from 3,176 RA patients without cardiovascular disease. Individual data were collected on CV risk factors and outcomes and then combined with RA-specific information, such as disease duration, seropositivity, rheumatoid factor and/or anti-citrullinated protein antibodies, 28-joint Disease Activity Score (DAS28), and acute phase reactants.

During an average of almost 8 years of follow-up comprising 24,733 patient-years, 314 patients developed cardiovascular disease (CVD).

Two possible models came out of multivariable risk score modeling for predicting the RA-specific 10-year risk of CVD. Each incorporated the traditional risk factors of age, sex, current smoking, presence of hypertension, and ratio of total cholesterol to HDL cholesterol, but also either seropositivity or DAS28. The consortium specifically chose to assess potential risk factors that would be easily available to the health professional regardless of the setting (for example, primary, secondary, and tertiary care, and even private practice).

The models that included seropositivity or DAS28 both demonstrated good discrimination and calibration, when compared with the Framingham or SCORE (Systematic Coronary Risk Evaluation) algorithms. The models also showed good concordance with each other (see chart).

"By pooling data from many centers, it appears possible to develop an RA-specific CVD risk algorithm which is more accurate at predicting CVD in people with RA than the currently available risk algorithms, which have been developed for the general population," members of the consortium wrote in response to e-mailed questions.

The consortium is now working to validate the ATACC-RA calculator, with the ultimate aim of validating it in completely independent cohorts. The continued growth of the consortium will help make this a reality within the next couple of years. The end result may be a risk calculator that can be adapted to account for the underlying risk of each patient population.

Once the algorithm is validated, the consortium hopes it will become part of routine rheumatology clinical practice, much the same as the Framingham or SCORE algorithms are used currently in the general population, but more specifically applied to patients with RA.

"The models proved to be quite robust," Ms. Arts said. "If this holds true through the additional analysis and external validation, we may have one that will be applicable to a wide variety of patients all over the world."

Once fully validated, the investigators plan to produce the tool in a user-friendly application for computers or even smartphones.

The investigators had no conflicts of interest to declare.

An international collaboration of experts has developed a cardiovascular risk calculator specifically for rheumatoid arthritis patients that has the potential to become part of routine rheumatology clinical practice in many parts of the world.

The ATACC-RA (A TransAtlantic Cardiovascular Risk Calculator for Rheumatoid Arthritis) consortium will be constructed from data collected at 13 rheumatology centers in 10 countries, Elke Arts said at the annual European Congress of Rheumatology.

The current version, which is still undergoing validation, contains data from eight rheumatology centers in seven countries (Greece, the Netherlands, Norway, South Africa, Sweden, the United Kingdom, and the United States), said Ms. Arts, a rheumatology researcher at Radboud University Medical Centre in Nijmegen, the Netherlands.

So far, ATACC-RA contains pooled data from 3,176 RA patients without cardiovascular disease. Individual data were collected on CV risk factors and outcomes and then combined with RA-specific information, such as disease duration, seropositivity, rheumatoid factor and/or anti-citrullinated protein antibodies, 28-joint Disease Activity Score (DAS28), and acute phase reactants.

During an average of almost 8 years of follow-up comprising 24,733 patient-years, 314 patients developed cardiovascular disease (CVD).

Two possible models came out of multivariable risk score modeling for predicting the RA-specific 10-year risk of CVD. Each incorporated the traditional risk factors of age, sex, current smoking, presence of hypertension, and ratio of total cholesterol to HDL cholesterol, but also either seropositivity or DAS28. The consortium specifically chose to assess potential risk factors that would be easily available to the health professional regardless of the setting (for example, primary, secondary, and tertiary care, and even private practice).

The models that included seropositivity or DAS28 both demonstrated good discrimination and calibration, when compared with the Framingham or SCORE (Systematic Coronary Risk Evaluation) algorithms. The models also showed good concordance with each other (see chart).

"By pooling data from many centers, it appears possible to develop an RA-specific CVD risk algorithm which is more accurate at predicting CVD in people with RA than the currently available risk algorithms, which have been developed for the general population," members of the consortium wrote in response to e-mailed questions.

The consortium is now working to validate the ATACC-RA calculator, with the ultimate aim of validating it in completely independent cohorts. The continued growth of the consortium will help make this a reality within the next couple of years. The end result may be a risk calculator that can be adapted to account for the underlying risk of each patient population.

Once the algorithm is validated, the consortium hopes it will become part of routine rheumatology clinical practice, much the same as the Framingham or SCORE algorithms are used currently in the general population, but more specifically applied to patients with RA.

"The models proved to be quite robust," Ms. Arts said. "If this holds true through the additional analysis and external validation, we may have one that will be applicable to a wide variety of patients all over the world."

Once fully validated, the investigators plan to produce the tool in a user-friendly application for computers or even smartphones.

The investigators had no conflicts of interest to declare.