April 14, 2015
12 min read

Nonobstructive CAD: Recognition needed, treatment options sought

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

A patient presents with chest pain and is sent to the cath lab. CAD is found, but not enough to be considered obstructive CAD. Traditionally, those patients have been treated for their symptoms. However, it was often assumed that they were at no higher risk for future CV events than people with no CAD.

In recent years, studies have shown that this line of thinking is wrong: People with nonobstructive CAD are at higher risk for future events than those with no CAD.

“In the past, a patient coming in with chest pain and an abnormal stress test might have an angiogram that looked normal, and the patient would be told that everything was fine,” William F. Fearon, MD, associate professor of medicine and director of interventional cardiology at Stanford Health Care, told Cardiology Today. “We have learned that these patients can have abnormalities that are not obvious on the angiogram, and that doing further investigation, particularly invasive testing to examine pressure, flow and other abnormalities in the coronary circulation, can identify the cause for chest pain and then guide treatment strategies.”

However, these procedures have not yet been incorporated into the guidelines or into the clinical practice of many doctors.

“There has been a trend toward more comprehensive coronary physiology assessment at several centers such as Mayo Clinic, but the initial advance that was made is that the European Society of Cardiology has put testing for endothelial function in its guidelines for patients with unexplained chest pain,” Amir Lerman, MD, FACC, FAHA, professor of medicine, chair of research and the director of the chest pain clinic at Mayo Clinic, said in an interview. “Overall, this still remains an unexplored area.”

William F. Fearon, MD, from Stanford Health Care,  discusses the causes, diagnostic practices and treatment options for patients with nonobstructive CAD.

William F. Fearon, MD, from Stanford Health Care, discusses the causes, diagnostic practices and treatment options for patients with nonobstructive CAD.

Image: Norbert von der Groeben; printed with permission from Stanford Health Care

Furthermore, research into optimal prevention and treatment strategies for these patients is lacking. Currently, physicians must rely on their own clinical judgment.

“It is clear that the majority of patients with nonobstructive CAD have an elevated 10-year atherosclerotic CVD risk that is substantially higher than the American College of Cardiology and American Heart Association’s 10-year risk calculator would tell us,” C. Noel Bairey Merz, MD, FACC, FAHA, the Women’s Guild Endowed Chair in Women’s Health; director of the Barbra Streisand Women’s Heart Center; director of the Linda Joy Pollin Women’s Heart Health Program; director of the Preventive Cardiac Center; and professor of medicine at Cedars-Sinai Medical Center, said during a presentation at the ACC Scientific Sessions in March. “My clinical practice is that I dispense statins and aspirin, and if they are symptomatic, anti-anginal and anti-ischemic therapy. Clinical trials are needed to understand the role of preventive therapies in the nonobstructive CAD population.”

Evidence for action

One of the first studies to raise the argument that nonobstructive CAD is not benign was the Women’s Ischemia Syndrome Evaluation (WISE) study. Results indicated that women with nonobstructive CAD have an elevated risk for MI compared with no CAD, and that nonobstructive CAD was often missed using conventional diagnostic methods.

Thomas M. Maddox, MD, MSC, FACC, FAHA

Thomas M. Maddox

The VA CART study published in 2014 found the same phenomenon in a predominantly male cohort of veterans who underwent elective coronary angiography in the Veterans Affairs health care system from 2007 to 2012 (n = 37,674; 22.3% with nonobstructive CAD). Thomas M. Maddox, MD, MSc, FACC, FAHA, cardiologist at the Veterans Administration Eastern Colorado Health Care System, associate director of the VA CART program and associate professor of medicine at the University of Colorado School of Medicine, Aurora, and colleagues found that after adjustment, 1-year MI rates rose with increasing CAD extent, and that veterans with single-vessel nonobstructive CAD (HR = 2; 95% CI, 0.8-5.1), two-vessel nonobstructive CAD (HR = 4.6; 95% CI, 2-10.5) and three-vessel nonobstructive CAD (HR = 4.5; 95% CI, 1.6-12.5) were all more likely to have MI at 1 year vs. those with no apparent CAD.


“For a long time, cardiac research only included patients with obstructive lesions, so we were unable to study those who had nonobstructive disease,” Maddox told Cardiology Today. “When we looked at rates of MI and death [for patients in the VA CART cohort] over the following year after that angiogram, we saw that there was this progressive increase. It wasn’t a dichotomous phenomenon where all the MI risk was clustered in the patients with obstructive CAD. There was definitely a [risk for MI] in the patients with nonobstructive CAD.”

A study by Fearon and colleagues published in February demonstrated that occult coronary abnormalities were observed in the majority of patients with angina but no obstructive CAD. In their study of 139 patients (mean age, 54 years; three-quarters women), 77% had a coronary explanation for their angina: 44% had endothelial dysfunction, 21% had microvascular impairment, 5% had fractional flow reserve of 0.8 or less and 58% had a myocardial bridge.

Although understanding the different etiologies is important, it also is significant that 23% of patients had no obvious coronary abnormality to explain the chest pain, so they “can be reassured that the coronary arteries are likely not involved, and hopefully avoid any unnecessary future testing and the potential risks associated with it, as well as unnecessary treatment,” Fearon said.

Carl J. Pepine

In an editorial related to the research published by Fearon and colleagues, Cardiology Today Chief Medical Editor Carl J. Pepine, MD, MACC, wrote that the findings confirmed those of the WISE IVUS substudy. “This information could provide direction for future management (eg, use of statins to prevent atherosclerosis progression). Importantly, it also raises the question of whether atherosclerosis and/or its preceding risk conditions could be linked to these other mechanisms. Certainly this would be true for endothelial dysfunction and also microvascular dysfunction. This notion would broaden our traditional concepts of atherosclerotic CAD to include dysfunction of the smaller coronary vessels and microcirculation.”

He also noted that Fearon’s laboratory had identified multiple potential causes for ischemia in their cohort study and said this represents a shift in the prevailing model for ischemia caused by a flow-limiting stenosis.

The dilemma of MINOCA

Research also has been conducted to identify the characteristics of MI with nonobstructive coronary arteries (MINOCA). A systematic review and meta-analysis published in January by John F. Beltrame, BSc, BMBS, FRACP, PhD, FESC, FACC, FCSANZ, the Michell professor of medicine at University of Adelaide, Australia, and colleagues determined that the prevalence of MINOCA among patients with MI was 6% (95% CI, 5-7; median patient age, 55 years; 40% women), and that compared with those with MI and obstructive CAD, patients with MINOCA were more likely to be younger and female and less likely to have hyperlipidemia.


John F. Beltrame

In addition, all-cause mortality at 1 year was 4.7% (95% CI, 2.6-6.9) for patients with MINOCA, compared with 6.7% (95% CI, 4.3-9) for those with MI and obstructive CAD. Outcome studies for nonobstructive CAD use all-cause mortality instead of CV mortality because doctors who think nonobstructive CAD is benign do not list a CV-related cause of death on death certificates for those patients, Beltrame said. Also, only 24% of patients with MINOCA had a typical infarct on cardiac MRI. Other likely causes of MI were myocarditis (33%), coronary artery spasm (27%) and thrombophilia disorders (14%); 26% had no significant other abnormality detected.

“The most important aspect of this condition is its identification,” Beltrame told Cardiology Today. “We created the MINOCA label so that doctors can flag these patients. Then you have to identify the underlying cause, because they’re different. Why is this important? Because the prognosis is guarded.”


Mechanisms of disease

The root causes of nonobstructive CAD are numerous and, in many cases, differ from those for obstructive CAD. Some have large-vessel dysfunction with spastic angina and some have coronary microvascular dysfunction, Beltrame said during a presentation at the 2015 ACC Scientific Sessions.

“There are a number of different possible etiologies,” Fearon said. “One would be microvascular dysfunction, in which the small vessels of the coronary arteries that we don’t see well on the angiogram can be dysfunctional, and that can lead to abnormalities in cardiac function and increased events. There can also be problems with the epicardial vessels that just aren’t apparent on the angiogram, for example diffuse plaque or atherosclerosis, but they are apparent when one does IVUS or coronary pressure measurements looking at fractional flow reserve. There can be endothelial dysfunction and/or coronary spasm, which can lead to ischemia and future events as well.”

Within the category of coronary microvascular function, there are four common conditions, according to Beltrame. The most studied is cardiac syndrome X, which describes people who have exertional angina, normal angiography and a positive stress test, he said. Microvascular spasm is diagnosed when acetylcholine provocation testing does not induce large vessel coronary spasm but produces chest pain with ischemic ECG changes, thereby implicating a dysfunctional coronary microvasculature; in contrast, patients with microvascular angina have an impaired ability to vasodilate the coronary microvasculature (ie, an impaired coronary flow reserve); and the coronary slow flow phenomenon describes those who have a slow passage of contrast on angiography, reflecting vasoconstriction in the distal microvasculature.

“The problem is that we have a heterogeneous disorder with multiple syndromes and multiple mechanisms, so we need to better define these,” Beltrame said.

Pepine also cautioned about continued use of the term cardiac syndrome X, since others (Vermeltfoort, et al. Clin Res Cardiol. 2010;99:475-481) have documented in an extensive analysis of existing literature that there is no consensus regarding its definition.

Best diagnostic practices

If a patient presents with chest pain but the angiogram shows no evidence of obstructive CAD, physicians should not assume nothing is wrong; instead, they should perform tests to detect forms of nonobstructive CAD, experts told Cardiology Today.

Amir Lerman, MD, FACC, FAHA

Amir Lerman

“If there is no evidence of obstructive disease, proceed to the next step and perform microvascular assessment using administration of adenosine and acetylcholine to determine the epicardial and microvascular endothelial and non-endothelial function,” Lerman said. “You assess the vascular reactivity of the coronary circulation and determine if this circulation can actually increase their flow with increased demand.”

FFR measurement is a good tool to assess the functional status of the coronary vessels, and IVUS can help detect occult coronary disease such as myocardial bridges, according to Fearon. “Further invasive testing is often helpful,” he said.

There is no evidence that invasive screening for nonobstructive CAD in asymptomatic patients is beneficial, Maddox said. However, “in some cases, where people are concerned by a family history of early heart disease, atypical symptoms or other risk factors that would cause them concern, then what we typically have been doing is ordering cardiac CT angiograms. In selected patients you can get an equal, sometimes better, view than traditional angiography, but without the risks that accompany invasive coronary angiography,” he said.

Patients with suspected MINOCA should undergo cardiac MRI, Beltrame said. He noted that cardiac MRI will pick up conditions such as myocarditis, infarcts, takotsubo cardiomyopathy and hypertrophic or dilated cardiomyopathy.

Pepine in his editorial cautioned against assuming the best for patients diagnosed with angina but no obstructive CAD.

“The ‘less than very good’ outcomes of these patients … described recently by many groups, must be recognized so that a near-normal or ‘normal’ angiogram does not drive diagnostic and therapeutic complacency,” he wrote.


Optimal treatment practices

Although there is more recognition that nonobstructive CAD should be treated, there is little evidence about which treatments are optimal.

C. Noel Bairey Merz

C. Noel Bairey Merz

No comprehensive clinical trials on prevention strategies in that patient population have been completed, Bairey Merz, a member of the Cardiology Today Editorial Board, said during her presentation. She noted that Eva Prescott, MD, DMSc, from Bispebjerg Hospital, Copenhagen, Denmark, is leading an ongoing clinical trial to test preventive strategies in patients with nonobstructive CAD, and that she, Judith Hochman, MD,Harmony R. Reynolds, MD, from NYU Langone Medical Center, and David Maron, MD, from Stanford, are attempting to get funding for a study to help determine whether patients with nonobstructive CAD should take preventive medications as well as anti-angina and anti-ischemia therapies.

There are Japanese guidelines for management of large-vessel vasospastic angina, Beltrame said in his presentation. Those give Class I recommendations to lifestyle modifications and calcium-channel blockers and Class II recommendations to long-acting nitrates, nicorandil and rho-kinase inhibitors, he said. The latter two are not available in the United States.

Calcium-channel blockers in a study from 1988 were associated with improved survival and infarct-free survival in patients with vasospastic angina (P = .0024), Beltrame said. Other independent predictors of survival included extent of CAD and evidence of multivessel spasm, he said.

The evidence for nitrates and nicorandil use is less clear, he said. Researchers for an observational study of 1,429 Japanese patients with vasospastic angina assigned calcium-channel blockers did not detect a difference in major adverse CV events between those who were and were not assigned nitrates (HR = 1.28; 95% CI, 0.72-2.28) or between those who were and were not assigned nicorandil (HR = 0.8; 95% CI, 0.28-2.27), but found elevated risk for major adverse CV events in those assigned nicorandil and nitrates, particularly nitroglycerin (HR = 2.14; 95% CI, 1.02-4.47). “The thought is that because nicorandil has nitrate-like properties, this may have to do with nitrate tolerance,” Beltrame said.

When 32 members of the Coronary Vasomotion Disorders International Study (COVADIS) group were polled regarding treatments for coronary microvascular dysfunction, most cited lifestyle modification and calcium-channel blockers as very useful, long-acting nitrates as useful in selected patients, and alpha-beta-blockers, ranolazine, ACE inhibitors and tricyclic antidepressants as occasionally useful, Beltrame said during his presentation.

Small studies have found that atenolol, a beta-blocker, is associated with reduced angina frequency in patients with cardiac syndrome X; ranolazine is associated with improved Seattle Angina Questionnaire scores in those with coronary microvascular dysfunction; and mibefradil, a combined L- and T- calcium-channel blocker, is linked to reduced angina frequency in patients with coronary slow flow syndrome, Beltrame said. Mibefradil was withdrawn in 1998 because of potential for adverse interactions with other drugs; however, other combined L- and T-calcium channel blockers are available in Japan.

Combined L- and T-calcium channel blockers such as mibefradil may work better than conventional L-channel blockers (eg, verapamil, diltiazem, nifedipine and amlodipine) in the treatment of microvascular disorders since there are more T-channels than L-channels in the microvasculature, Beltrame said. “This illustrates an important concept, namely that the pharmacology of the microvasculature differs from that of the large vessels, so that we need more agents that specifically target the microvasculature.”

Until that happens or more conclusive evidence emerges, aspirin and statins may be the default strategies, Maddox said.

He noted that the number needed to treat to prevent one MI in 1 year in patients with single-vessel nonobstructive CAD is approximately 1,000, whereas the number needed to harm via a bleeding event due to aspirin therapy is more than 1 in 3,000.

“If you compare those two outcomes, then clearly the risk–benefit favors the aspirin in these kinds of patients,” he said. “So I would say, for folks who do have identified nonobstructive disease because of an incidental CT or some other imaging, the risk–benefit is now slanted in favor of aspirin and statins.”


Proactive, precise strategies

Currently, physicians should no longer consider stenosis level as a single threshold when deciding whether to further investigate or treat a patient with CAD, experts told Cardiology Today.

“Dichotomous characterization of CAD to risk stratification should be discarded in favor of a graded approach,” Bairey Merz said. “We must be getting our interventional cardiologists to pay much more attention to careful reading of the angiogram, not the yes/no. Recognition of MI risk among patients with nonobstructive CAD should ideally trigger preventive therapy at a minimum and anti-anginal and anti-ischemic therapy at a maximum.”

Adopting a proactive and risk classification strategy for these patients is the best way to reduce future CV events that could have been prevented, Lerman said.

“This should be a significant priority in the next few years from a CV standpoint,” he said. “We focus a lot on MI and obstructive disease, but we tend to neglect the large number of patients, particularly women, who are coming to our cath lab already with symptoms of ischemia and do not have obstructive disease. They deserve a more comprehensive evaluation of why they arrived in the cath lab. We’ve showed in the past that their quality of life is improved after you diagnose and treat them. We need to take these patients seriously and try to help them overcome their symptoms of ischemia and reduce their events.” – by Erik Swain


Bairey Merz CN, et al. J Am Coll Cardiol. 2006;doi:10.1016/j.jacc.2004.12.084.

Bairey Merz CN and Beltrame JF. Myocardial ischemia with nonobstructive CAD: Fact or fancy? Presented at: American College of Cardiology Scientific Sessions; March 14-16, 2015; San Diego.

Beltrame JF, et al. J Am Coll Cardiol. 2004;doi:10.1016/j.jacc.2004.03.055.

Gierach GL, et al. J Am Coll Cardiol. 2006;doi:10.1016/j.jacc.2005.02.099.

Handberg E, et al. J Am Coll Cardiol. 2006;doi:10.1016/j.jacc.2005.10.023.

Khuddus MA, et al. J Interv Cardiol. 2010;doi:10.1111/j.1540-8183.2010.00598.x.

Lee BK, et al. Circulation. 2015;doi:11.1161/CIRCULATIONAHA.114.012636.

Maddox TM, et al. JAMA. 2014;doi:10.1001/jama.2014.14681.

Pasupathy S, et al. Circulation. 2015;doi:10.1161/CIRCULATIONAHA.114.011201.

Pepine CJ. Circulation. 2015;doi:10.1161/CIRCULATIONAHA.115.015553.

Pepine CJ, et al. J Am Coll Cardiol. 2006;doi:10.1016/j.jacc.2005.09.023.

Shaw LJ, et al. J Am Coll Cardiol. 2006;doi:10.1016/j.jacc.2005.01.072.

Shaw LJ, et al. J Am Coll Cardiol. 2006;doi:10.1016/j.jacc.2005.03.080.

Takahashi J, et al. Eur Heart J. 2015;doi:10.1093/eurheartj/ehu313.

Yasue H, et al. Circulation. 1988;doi:10.1161/01.CIR.78.1.1.

For more information:

C. Noel Bairey Merz, MD, FACC, FAHA, can be reached at Cedars-Sinai Heart Institute, 127 S. San Vincente Blvd., Suite 3206, Los Angeles, CA 90048; email: noel.baireymerz@cshs.org.

John F. Beltrame, BSc, BMBS, FRACP, PhD, FESC, FACC, FCSANZ, can be reached at The Queen Elizabeth Hospital, 28 Woodville Road, Woodville, SA 5011 Australia; email: john.beltrame@adelaide.edu.au.

William F. Fearon, MD, can be reached at 300 Pasteur Drive, H2103, Stanford, CA 94305; email: wfearon@stanford.edu.

Amir Lerman, MD, FACC, FAHA, can be reached at Mayo Clinic, 200 First St. SW, Rochester, MN 55905; email: lerman.amir@mayo.edu.

Thomas M. Maddox, MD, MSc, FACC, FAHA, can be reached at 1055 Clermont St., Denver, CO 80220; email: thomas.maddox@va.gov.

Carl J. Pepine, MD, MACC, can be reached at the Cardiology Today office, 6900 Grove Road, Thorofare, NJ 08086; email: carl.pepine@medicine.ufl.edu.

Disclosures: Bairey Merz reports receiving consultant fees/honoraria from Gilead and Kaiser and serving on the speakers’ bureau for Practice Point Communications and Vox Media. Beltrame reports receiving research grants from AstraZeneca and Servier. Fearon reports receiving research support from Medtronic and St. Jude Medical and speakers’ honoraria from Medtronic. Lerman reports consulting for Itamar. Maddox and Pepine report no relevant financial disclosures.