Getting to the heart of cardiovascular outcomes in rheumatic disease
Cardiovascular events cause significant morbidity and mortality among patients with rheumatologic diseases. Moreover, there is a robust body of data chronicling these associations.
Given these factors, two byproducts would seem reasonable: One is that rheumatologists should have a broad set of recommendations to follow in managing cardiovascular outcomes in their patients; the second is that the specialty should have a clear-cut understanding of how rheumatologic diseases — and the treatments for those diseases — contribute to or mitigate heart disease risk.
But neither of those things is the case.
“We do not have a comprehensive, data-driven guideline,” Jon T. Giles, MD, MPH, associate professor of medicine in the division of rheumatology at the Columbia University College of Physicians & Surgeons, told Healio Rheumatology. “EULAR has published two guidelines on cardiovascular disease management in rheumatic disease patients, but the level of evidence for most of the recommendations was relatively low, and some of the recommendations have been shown not to improve prediction.”
M. Elaine Husni, MD, MPH, vice chair of the department of rheumatic and immunologic diseases, and director of the Arthritis Center at the Cleveland Clinic, explained why the level of evidence for these guidelines is problematic. “Of course, I would like to see more prospective, placebo-controlled studies following patients recently diagnosed with a rheumatologic disease to determine when and how they experience cardiovascular outcomes, as opposed to the observational and retrospective studies on which current recommendations are based,” she said.
The problem is ethical, according to Husni, who is also a member of the Healio Rheumatology Peer Perspective board. “You can’t withhold treatment or treat a patient with placebo in this situation,” she said.
Despite these limitations, it is generally understood that the pathogenesis of cardiovascular disease in rheumatology is multifactorial and results from the interplay between inflammation, metabolic disease, patient factors such as weight and smoking status, therapeutics and other disease-related issues. But a general understanding and a true ability to parse those myriad influences effectively are two completely different animals.
The good news is that most rheumatologists recognize that their patients are at increased risk for cardiovascular outcomes. However, risk stratification tools used for the general population — including the Framingham and Reynolds risk scores — have proven suboptimal in rheumatology patient populations.
Thus, rheumatologists are encouraged to employ a few general strategies in mitigating and managing these events. One is to simply be vigilant for potential cardiovascular outcomes. Another is to work with a cardiologist and other relevant clinicians to keep their patients healthy. And the third is to do their best to make sense of a robust, but imperfect, body of data.
“Almost all rheumatic diseases have been associated with higher cardiovascular disease risk,” Giles said. “The best studied are RA and lupus, but there is a growing body of literature on psoriasis, PsA and the other spondyloarthropathies. Cardiovascular events are also higher in many of the vasculitides.”
While historical data, including those from Avina-Zubieta and colleagues in Arthritis Care & Research, have estimated that outcomes like myocardial infarction and ischemic stroke may be 50% higher in patients with RA compared with the general population, more recent findings show an improved but still complicated picture.
For example, in their study in Annals of the Rheumatic Diseases, Holmqvist and colleagues reported on secular trends in incident acute coronary syndrome risk in RA vs. controls over 5 decades. The results demonstrated reduction, for the first time, in such risk in patients with RA diagnosis in the present decade.
“Indeed a 40% risk reduction was observed, reflecting improved treatment algorithms and successful awareness campaigns,” George A. Karpouzas, MD, FACR, investigator at The Lundquist Institute, professor of medicine at the David Geffen School of Medicine at UCLA and chief of the division of rheumatology at Harbor-UCLA Medical Center, said in an interview. “Nevertheless, RA patients still incurred a 40% higher acute coronary syndrome risk, which was restricted to those who were seropositive and had at least moderate disease activity at diagnosis.”
Beyond risk for acute coronary syndrome, Karpouzas cited more recent data from Avina-Zubieta and colleagues in Annals of the Rheumatic Diseases — along with those from Ungprasert and colleagues from the International Journal of Rheumatic Diseases and Nicola and colleagues from Arthritis & Rheumatology — demonstrating that clinicians treating patients with RA need to be aware of a range of other outcomes.
“Risk of myocardial infarction may be increased by 56%, ischemic stroke by 45%, atrial fibrillation by 30%, heart failure may be increased 2.5-fold and venous thromboembolism is also significantly higher in RA compared to controls,” he said.
Turning to lupus, data from the study by Alhusain and colleagues in the Clinical Medicine Journal showed that clinical cardiovascular disease may be elevated as high as 10%. “Patients with SLE have a five- to six-fold increased risk of developing cardiovascular disease compared to the general population, with higher risk in young patients,” Karpouzas said. “An increased prevalence of subclinical atherosclerosis is seen, with up to 40% of SLE patients having evidence of a carotid plaque, nuclear imaging abnormalities or increased coronary calcification.”
Ted R. Mikuls, MD, MSPH, Umbach Professor of Rheumatology and vice chair of research at the University of Nebraska Medical Center, suggested that cardiovascular events in lupus, in particular, can “put people on their heels,” for one key reason. “You are dealing with young women who you would not normally anticipate having a high risk cardiovascular disease.”
Regarding spondyloarthritis, a meta-analysis from Kim and colleagues published in Annals of the Rheumatic Diseases showed that myocardial infarction (RR = 1.38; 95% CI, 1.18-1.61) and stroke (RR = 2.04; 95% CI, 1.11-3.78) were elevated in spondyloarthropathy patients compared to the general population.
While all the data on various rheumatologic conditions can be illuminating, Mikuls offered a reason why it is so difficult for clinicians to understand disease-specific risks. “I don’t know that we have adequate studies that have definitively compared risk by rheumatic disease state,” he said. “Or, rather, the studies that have looked at that particular question have been limited.”
Mikuls acknowledged that because inflammation is a “shared characteristic” of many rheumatologic diseases, it follows that the burden of cardiovascular events should be higher. “But whether that is attributed to systemic inflammation, metabolic or other factors is unclear,” he said.
Michael S. Garshick, MD, director of the Cardio-Rheumatology Clinic, part of the Center for the Prevention of Cardiovascular Disease at the New York University School of Medicine, offered a possible explanation. “These diseases, including psoriasis/psoriatic arthritis, SLE and RA, many times exhibit higher rates of traditional cardiovascular risk such as hypertension, hyperlipidemia and metabolic syndrome,” he said. “These traditional cardiovascular risk factors can synergistically increase risk in these inflammatory populations.”
While understanding that the diseases may contribute to cardiovascular risk is important, it is only the first part of the equation. The next is understanding which treatments may contribute to risk, and the extent of that contribution.
If there is an overarching issue that pertains to associations between rheumatology drugs and cardiovascular risk, it is that there is frequently conflicting evidence.
For example, findings from a 2019 study by Wilson and colleagues published in Arthritis Care & Research demonstrated a 30% increased risk of stroke or myocardial infarction and a 30% increased risk of death in RA patients taking glucocorticoids.
“The main offender, without question, are corticosteroids,” Giles said. “They have been shown in multiple studies to be associated with atherosclerosis and cardiovascular disease events in proportion to the dose and time of exposure.”
To that point, Karpouzas and colleagues recently published a prospective study of RA patients with baseline and long-term follow-up coronary atherosclerosis assessments in Arthritis & Rheumatology. Results showed that each additional gram of cumulative prednisone dose used independently was associated with a 6% higher number of coronary segments with plaque, 6% greater stenotic plaque severity and 10% higher CAC score at follow-up.
“Despite the fact that physicians generally prescribe glucocorticoids to patients with higher disease activity, our observations highlight the true deleterious effect of glucocorticoids on the vascular wall, rather than confounding by indication,” Karpouzas said.
But Rekha Mankad, MD, a cardiologist and echocardiographer in the department of cardiovascular medicine at the Mayo Clinic, explained the other side of the glucocorticoid coin. “Steroid use is not uncommon in rheumatology because of its potent anti-inflammatory effect,” she said. “So, in one way, they are very beneficial because of their reduction in inflammation, which we believe is a strong driver of heart disease in these patients.”
That said, Mankad acknowledged that long-term steroid use is “fraught” with complications that increase cardiovascular risk. “These risks include more potential for weight gain, elevations in blood pressure and salt retention with a potential for lipid abnormalities,” she said.
If there is another class of drugs with a long history of associations with cardiovascular disease, it is non-steroidal anti-inflammatory drugs (NSAIDs).
“The impact of NSAIDs is a bit unclear, and likely is different between different NSAIDs,” Giles said. “The remaining immunomodulators mostly have data — some more than others — suggesting that they have a protective role.”
Mankad described the use of NSAIDS as “liberal” among rheumatology patients in the past, noting that this class of drugs has been implicated in increased cardiovascular disease, but with an important qualification. “The risk is greater in persons who already have heart disease,” she said.
There may be good news on the horizon, however, according to Mankad. “Now that there are so many other agents that can be used with these conditions, overall use of NSAIDs has gone down, especially high doses being used chronically,” she said.
Mikuls offered a final point summing up both steroid and NSAID use. “The risk seems to outweigh the anti-inflammatory benefit for some patients,” he said. “You should be aware of this if you have someone at high risk for cardiovascular disease and you want to minimize that risk by minimizing exposure to these agents.”
Deeper into the Armamentarium
The prevailing wisdom is that conventional synthetic disease-modifying anti-rheumatic drugs (DMARDs) such as methotrexate and hydroxychloroquine are beneficial for heart disease largely due to their anti-inflammatory effects. However, data clearly outlining these associations also remain elusive.
In the randomized, double-blind, placebo-controlled Cardiovascular Inflammation Reduction Trial (CIRT) published in The New England Journal of Medicine, Ridker and colleagues evaluated low-dose methotrexate as a secondary prevention tool for atherothrombotic events among patients with a history of myocardial infarction or multivessel coronary artery disease who additionally had either type 2 diabetes or metabolic syndrome.
“In this trial, methotrexate did not result in fewer cardiovascular events than placebo and did not reduce levels of interleukin (IL)-1, IL-6, or C-reactive protein,” Karpouzas said.
While Mikuls also tends to think of methotrexate as “cardioprotective,” he underscored the failure of this trial. “The protective effect, if indeed it is true, seems to be unique to people with underlying inflammation as is seen in RA,” he said.
As for hydroxychloroquine, in their study published in Annals of the Rheumatic Diseases, Rempenault and colleagues showed that the drug may benefit the metabolic profile and to a lesser extent cardiovascular events in patients with RA. “This suggests its usefulness combined with other csDMARDs,” Karpouzas said.
As is seen in the move away from NSAIDs, increased use of biologic and biosimilar therapies may push csDMARD use to the sidelines. There seems to be cautious optimism surrounding biologic therapies, according to Garshick. “Meta-analyses have looked at the cardiovascular risk of biologics in the psoriasis/PsA population and they appear safe overall,” he said.
In their study published in Annals of the Rheumatic Diseases, Xie and colleagues conducted a meta-analysis of 26 randomized, controlled trials looking at cardiovascular events in patients with RA treated with JAK inhibitors. Results showed no significant change in risk. However, thromboembolic event risk was elevated among patients treated with both tofacitinib (Xeljanz, Pfizer) and baricitinib (Olumiant, Eli Lilly).
“There has been some concern about the relationship between thromboembolism and high dose JAK inhibitors in the RA population,” Garshick said.
TNF inhibitors have also demonstrated evidence of such a complicated relationship. According to data published in Expert Opinion on Drug Safety, Tocci and colleagues analyzed observational cohort studies, national registry data and metaanalyses and found reason for cautious optimism in light of improved cardiovascular outcomes for these drugs.
However, Mankad noted that more data on the long-term effects of these medications is still needed. “TNF inhibitors have been implicated in congestive heart failure, and thus should be used cautiously in those who have already had this diagnosis,” she said.
Karpouzas acknowledged that deeper investigation into non-TNF inhibitor biologics also may prove beneficial, citing recent data from Singh and colleagues in Arthritis Care & Research. “Treatment with tocilizumab [Actemra, Genentech] might be associated with a reduced risk of cardiovascular disease events compared with treatment with TNF inhibitors based on this recent meta-analysis.”
Mankad offered a final consideration for clinicians assessing cardiovascular endpoints associated with biologic therapies. “There has also been an implication of biologics resulting in higher cholesterol values,” she said.
With so many questions surrounding disease-specific associations and therapeutics, the rheumatology community would likely benefit from a risk stratification tool to help guide the way in managing this patient population. But this, too, has proven elusive.
Garshick suggested that comparison studies have shown that traditional scores like Framingham and Reynolds underestimate cardiovascular risk in rheumatic conditions ranging from RA to psoriatic disease and SLE. “Even though inflammatory conditions exhibit higher traditional cardiovascular risk factors, which are recognized in traditional cardiovascular risk scores, there is still an independent association between these inflammatory diseases and cardiovascular disease,” he said. “Moreover, in the development and validation of these cardiovascular risk prediction models, inflammatory diagnoses were not accounted for.”
The Reynolds risk score incorporates high sensitivity C-reactive protein into the equation, which Garshick states should theoretically make it more useful than Framingham in the rheumatology setting. “However, this score is still not validated in the rheumatologic patient population,” he said.
With these concerns in mind, the medical community has responded to these inaccuracies with a variety of modifications, according to Garshick. “The American College of Cardiology/American Heart Association now recommends the pooled cohort equation with the incorporation of certain risk enhancers, such as psoriasis/PsA, SLE, RA, high sensitivity CRP and even CAC score to reclassify cardiovascular risk and encourage the earlier initiation of prevention medications such as statin therapy,” he said.
EULAR recommends multiplying risk predication models by 1.5 to upgrade cardiovascular risk in RA. “But this is still probably inaccurate,” Garshick said. “Joint American Academy of Dermatology/National Psoriasis Foundation guidelines suggest something similar for many psoriasis/PsA patients.”
Garshick added that the QRISK 2 score for RA and the QRISK 3 score, which incorporates SLE and steroid use, can also estimate cardiovascular risk. “Unfortunately, given the heterogeneity of inflammatory diseases, exposure and treatment, it is difficult to capture the impact of varying amounts of underlying inflammation on the degree of atherosclerosis and even impact on thrombosis,” he said. “This is also compounded as we are still trying to grasp mechanistic explanations of what is driving cardiovascular risk in these populations.”
One reason that traditional risk scores fail to be effective is that the relationships between traditional cardiovascular disease risk factors and atherosclerotic burden are skewed, particularly in RA and lupus, according to Giles. “RA patients have lower LDL cholesterol, on average, than people without RA, and some of them have very low LDL cholesterol, despite having more atherosclerosis burden,” he said. “It is this group of RA patients that has been shown to have a very high rate of cardiovascular events. This was named the lipid paradox.”
Giles and his group published a paper showing that these patients actually had much more coronary atherosclerosis than could be expected from this level of LDL cholesterol. “Why this is the case is still unknown, but it probably involves the way that circulating lipids change in the setting of autoimmunity and inflammation that make them less predictive,” he said.
As the research community attempts to grapple with these questions, rheumatologists are left to interpret associations and make diagnoses as best they can in the clinic.
In the Clinic
For Husni, until a set of guidelines based on randomized, controlled trial data emerges, sticking to the basics is the best way to keep patients’ hearts healthy. “Treating to target and lowering systemic inflammation is the first and most important step toward mediating cardiovascular disease risk,” she said.
The next step for Husni is to “get serious” about external or modifiable risk factors like weight, cholesterol, stress, smoking and hypertension. She used the term “assess and address” to drive the message home.
That said, Husni was realistic about the limitations rheumatologists face in enacting the assess and address system. “I have 10 minutes per patient nowadays,” she said. “I do not have time to adequately evaluate their disease and still talk to them about smoking cessation and weight management.”
With these time constraints, most rheumatologists will hand a patient off to the cardiology team. Hearing messages about diet and exercise from as many health care providers as possible is critical to mitigating cardiovascular risk, according to Husni. But she urged rheumatologists to have these conversations whenever possible.
Moving away from the clinic, a final point for Husni was a call to arms to the basic and translational science research communities. “We need to work on understanding the mechanism of how low-grade, systemic inflammation can lead to heart disease,” she said. “I have lifelong RA patients who drink and are overweight who never experience a cardiovascular event, and patients who eat healthy and exercise regularly who do. We need to understand why.”
Mankad expanded on this point. “Although we call these conditions risk enhancers, we really do not know how much of an enhancer it is in each individual patient,” she said.
If there is another area ripe for exploration, it is further recognition that other parts of the heart, and not just the arteries, can be affected by this chronic inflammation, according to Mankad. “Studies to better identify which of these patients may be at risk for heart failure in the future is important.”
Looking toward the future, if there is another area Husni would like to see studied, it pertains to biomarkers that may predict cardiovascular risk in rheumatology patients. “We are looking at paraoxonase 1, which can be used to show that HDL cholesterol is not working as well as it should,” she said. But she is realistic about the impact that biomarkers may have in the clinic in the near future. “The whole field is still in its infancy.”
- Alhusain A, et al. Clin Med (Lond). 2013;doi:10.7861/clinmedicine.13-4-395.
- Avina-Zubieta JA, et al. Arth Care Res. 2008;doi:10.1002/art.24092.
- Avina-Zubieta JA, et al. Annals Rheum Dis. 2012;doi:10.1136/annrheumdis-2011-200726.
- Goodson NJ, et al. Ann Rheum Dis. 2009;doi:10.1136/ard.2007.076760.
- Holmqvist M et al. Ann Rheum Dis. 2017;doi:10.1136/annrheumdis-2016-211066.
- Kim J, et al. Ann Rheum Dis. 2019;doi:10.1136/annrheumdis-2019-eular.6597.
- Karpouzas GA, et al. Arthritis Rheumatol. 2020;doi:10.1002/art.41122.
- Nicola PJ, et al. Arthritis Rheumatol. 2005;doi:10.1002/art.20855.
- Rempenault C, et al. Ann Rheum Dis. 2018;doi:10.1136/annrheumdis-2017-211836.
- Ridker PM, et al. N Engl J Med. 2019;doi:10.1056/NEJMoa1809798.
- Singh S, et al. Arth Care Res. 2020;doi: 10.1002/acr.23875.
- Tocci G, et al. Expert Opin Drug Saf. 2016;doi:10.1080/14740338.2016.1218469.
- Ungprasert P, et al. Int J Rheum Dis. 2017;20:343-441.
- Williams B, et al. Eur Heart J. 2018;doi:10.1093/eurheartj/ehy339.
- Wilson JC, et al. Arth Care Res. 2019;doi:10.1002/acr.23611.
- Xie W, et al. Ann Rheum Dis. 2019;doi:10.1136/annrheumdis-2018-214846.
- For more information:
- Michael S. Garshick, MD, can be reached at 530 1st Ave., HCC, Suite 4F New York, NY 10016; email: email@example.com.
- Jon T. Giles, MD, MPH, can be reached at 630 W. 168th St., New York, NY 10032; email: firstname.lastname@example.org.
- M. Elaine Husni, MD, can be reached at 500 Euclid Ave. A50, Cleveland, OH 44195; email: email@example.com.
- George A. Karpouzas, MD, FACR, can be reached at 1124 W. Carson St., Building E4-R17A, Torrance, CA, 90502; email: firstname.lastname@example.org.
- Rekha Mankad, MD, can be reached at 9974 214th St. W., Lakeville, MN 55044; email: email@example.com.
- Ted R. Mikuls, MD, MSPH, can be reached at 986270 Nebraska Medical Center, Omaha, NE 68198-6270; email: firstname.lastname@example.org.