4. How reassuring is a recent (< 1 year) negative stress test in managing a patient with chest pain? How about a recent “normal” cath (i.e., < 30% blockage, no intervention)?

In a word (or two): not very.

The problem of what to do with a patient who has had a recent “negative” cardiac work-up and who now presents to the ED with the same complaint of chest pain is a frustrating one for all parties involved.   There is no clear cut answer to this conundrum and there is scant literature addressing the issue, though there has been at least one proposed, but not studied or validated, algorithmic approach to assessing such patients (Lewis, 2010).  Concerningly, some studies have suggested that cardiac morbidity and mortality after a negative ACS work-up may be as high as 14% for adverse cardiac events and 2% mortality at 6 months (Manini, 2002).

In approaching these situations, it is helpful to: 1) have an understanding of the pathophysiology of coronary artery disease and ACS; 2) be familiar with the type of test or study the patient had: What are its diagnostic and prognostic abilities? Its limitations?; and 3) understand the limitations of those statistics as they relate to an individual patient.

Pathophysiology-wise, the key point to remember is that plaques can rupture at any size including when they are considered non-obstructive (less than 50% occluded) (Hackett, 1988). While studies suggest that there is a direct relationship between size of a plaque and its vulnerability to rupture (Rao, 2005), other studies have also shown that lumen-preserving eccentric plaques may be even more vulnerable than those that partially obstruct the lumen (Varnava, 2002; Yamagishi, 2000). This means that what appears to be a patent vessel may in fact be a ticking time-bomb.

Stress tests come in many flavors. For the sake of this question, we’ll use the extreme ends of the spectrum of studies used to assess coronary artery disease burden. On the simplest end we have the electrocardiogram exercise stress test, and on the other end, we have the “gold standard” of cardiac catheterization. Between these two lie many other tests including stress echocardiography, myocardial perfusion imaging (SPECT, PET, etc.), cardiac magnetic resonance imaging, and multi-detector cardiac computed tomography, all with varied sensitivities, specificities, positive and negative predictive values.

Electrocardiogram exercise stress tests (ESTs) are the workhorse of risk stratification of patients with low-and-intermediate-risk chest pain who have normal resting ECGs and who are able to exercise. While they have reasonable sensitivities for detecting three-vessel disease they are relatively poor at identifying single vessel disease-with the exception of left main occlusion, for which they are also sensitive (Detrano, 1989), but of course incredibly high risk to perform. Despite high negative predictive values of 90-99%, which are even better when combined with the Duke Treadmill Score, (Mieres, 2005, Johnson, 2008, Manini, 2010), ESTs do not rule out coronary artery disease and a “negative” test should only be considered valid in ruling out ischemia as a cause of chest pain at the time of the test (Jones, 2010).

Cardiac catheterization, though considered the gold standard in diagnosing coronary artery disease, is also not fail-safe.  It is known to have limitations in identifying lesions, in part due to its two-dimensional nature (Topol, 1995). Even when luminal narrowing is appreciated, the nature and stability of the plaques is not known, and plaques that are eccentric are likely not recognized at all (Yamagishi, 2000). Small plaques may result in a “normal” angiogram, but may rupture the following day. (Of note, intravascular ultrasound has been used to help identify and characterize plaques, but is an additional procedure, not regularly performed with angiographies.) One study showed that 9% of NSTEMIs studied had non-obstructive disease (half of which had no disease seen at all); these patients nonetheless had a 2% death and/or MI rate  and a 12.1% primary end point rate (death, MI, UA, revascularization, stroke) at one year, indicating that lack of “significant” disease is not necessarily reassuring (Bugiardini, 2005).

Patients have been found to have ischemic events even in the absence of any coronary vessel disease. Myocardial infarction with clean coronaries is a frustrating entity to cardiologists. Some causes are believed to be myocardial bridging, coronary tortuosity and microvascular dysfunction (Gaibazzi, 2011).  Syndrome X (normal coronaries with chest pain and either ST changes or a positive stress test) afflicts women more often than men.  Though the prognosis for these patients is generally good, chest pain with non-obstructive disease has also not found to be entirely benign (Bugiardini, 2005).

In considering the weight of a stress test result in evaluating a chest pain patient, it is important to keep in mind that while exercise stress test results are generally considered “valid” for 1-2 years, many of the studies evaluating outcomes of patients, and hence ascribing prognostic value to these tests, only followed patients for 30 days to one year.  Perhaps even more important is to understand that prognostic data applies to a population, not to one individual’s risk of developing ACS at any given time after having a negative study. Hence, even if the likelihood of having ACS within 30 days of a negative test is exceedingly small, it doesn’t mean that the patient in front of you, who had a normal test five days ago, didn’t just rupture a plaque.

In summary then, while stress tests are excellent for prognosticating risk of ACS, and angiograms are strong at determining plaque burden, the ED clinician should be careful not to give too much import to the results of a prior test when facing a patient complaining of chest pain.  If concern for ischemic disease persists, the patient warrants further inpatient evaluation.  Indeed, this seems to be the practice pattern of most ED physicians. According to one prospective cohort study,  ED physicians admitted chest pain patients with and without prior negative stress tests at the same rate. Contributing even more to a brewing sense of nihilism:  both groups had the same cardiac event rate (Nerenberg, 2007).