Published in the January 2008 issue of Today’s Hospitalist
When revised guidelines for managing non-STEMI and unstable angina patients were released this past summer, Edward McNulty, MD, director of the cardiac cath lab at the San Francisco Veterans Affairs Medical Center, welcomed the news.
After all, it’s been more than five years since the American College of Cardiology/American Heart Association updated those guidelines. And during that time, “there has been a wealth of information,” Dr. McNulty said. “I was eager to see how to synthesize all the recommendations.”
But when he looked at the guidelines that were published in the Aug. 14, 2007, Journal of the American College of Cardiology, he found pages and pages of flow charts.
“It is an excellent document that takes a truly interdisciplinary approach,” he pointed out, “but, like many clinicians out there, I don’t find such complex algorithms to be all that practical.”
That’s why Dr. McNulty set out to simplify those guidelines during a presentation on acute coronary syndromes that he made at the University of California, San Francisco (UCSF) hospital medicine conference last October. In distilling the guidelines down to take-home points, he said, hospitalists are really interested in answers to three questions.
Which patients should be admitted? What anticoagulation strategy should physicians use for non- STEMI or unstable angina patients? And when should doctors ship patients to the cath lab vs. taking a more conservative approach and ordering additional tests?
Dr. McNulty presented actual cases to illuminate the answers to those questions. In the first case, a 35-year old man presents to UCSF’s emergency department reporting vague chest discomfort and muscle soreness after lifting weights for the first time in 10 years. He has no prior history, no history of tobacco or illicit drug use, is taking no medications, and is not tachycardic, although his symptoms are partially reproducible with palpation and movement.
His ECG contains no real evidence of chamber hypertrophy. And although there is what Dr. McNulty called “a hint of abnormality in his T waves,” many physicians, including several cardiologists, read the ECG as normal. Should the patient be discharged or held for further tests?
“Based on his paper presentation, I probably would have sent this patient home,” said Dr. McNulty. “But the guy just didn’t look right to the doctors in the emergency department.”
As for how different symptoms correspond to the possibility of having a myocardial infarction, more classic presentations “such as chest pressure “signal patients at higher risk, said Dr. McNulty, citing research published in the Nov. 23/30, 2005, Journal of the American Medical Association (JAMA).
That same research, however, also ascribed a 0.3 likelihood ratio of an infarct when symptoms can be reproduced with palpation. While it is tempting to exclude a diagnosis of an MI based on whether the symptoms are typical or not, the same study also showed that atypical presentations, such as pain radiating to the right shoulder, conferred the highest risk of developing infarcts, Dr. McNulty pointed out. “And atypical presentations are common.”
Physicians picking up on the T-wave abnormalities would have another risk-stratification tool. Close to one in 10 patients found to have either ST elevation or ST depression on ECG will die in the hospital, according to figures cited by Dr. McNulty that were part of the GUSTO IIb trial, published in the Sept. 12, 1996, New England Journal of Medicine (NEJM). However, T-wave inversions have a 4.9% risk of in-hospital mortality.
“T-wave inversions carry a more benign prognosis,” Dr. McNulty said, “but one out of 20 of these patients died in GUSTO IIb.”
As for the weightlifting patient, Dr. McNulty said that an astute physician decided to order a troponin assay. That test found troponin elevation of 2.5 ng/dl. Dr. McNulty said that according to research published in the Nov. 7, 2006, Journal of the American College of Cardiology, “there is almost a linear relationship between troponin elevation and inpatient mortality. There really is no benign level of elevation of troponin, whether you’re a patient with an acute coronary syndrome or not. Any troponin is bad.”
He did underscore, however, that normal cutpoints for troponin can vary, which makes it impossible to identify any one level as normal. But when you do have elevation, he added, “stop and think about what additional risk stratification may be warranted.”
Ultimately, the patient was found to have a high grade stenosis of the LAD and did well with PCI. The case highlighted the value, Dr. McNulty noted, of troponin assays, particularly with patients with atypical presentations.
But he said an even bigger lesson is the importance of trusting your instincts. “It really came down to he just didn’t look right,” he explained. “That clinical gestalt should trump any kind of algorithm.”
Scoring patient risk
The second case is that of a 65-year-old smoker with hypertension who presents after two weeks of right shoulder tightness and exertional dyspnea while playing golf. He comes to the ED after a minor episode after breakfast.
He takes simvastatin for his high LDL, in addition to aspirin; his heart rate is at the high end of normal; he has a few bibasilar crackles; and his troponin is negative. He has no murmurs or edema, but his ankle-brachial index (ABI) is less than 0.9 bilaterally and his jugular venous pressure is hard to assess. His ECG shows him in sinus, with “possible hints” of J-point elevations.
Dr McNulty used the case to illustrate the utility of clinical risk scores, which should incorporate ECG and physical exam findings when assessing patient risk. This patient, for instance, would score a 3 on a TIMI risk score, racking up points for his age, his use of aspirin and the fact that he has more than two risk factors.
“But I’d assign him a half point for his ECG,” Dr. McNulty said. He also would have liked to assign more points because of some heart failure and crackles, he added, saying that “that fact that he’s having symptoms on medications other than aspirin doesn’t come through in all risk scores.”
Additionally, the low ABI may predict more extensive coronary disease. Even though the TIMI scoring system defines high risk as a score of 4 or more, Dr. McNulty said he would be inclined to consider this patient high-risk, which carries a short-term mortality of between 3% and 5%.
With the patient potentially headed at some point to the cath lab for invasive assessment, what’s the best anticoagulation strategy to pursue in addition to the aspirin he’s already taking? According to the audience response system, 66% of the physicians in the audience chose enoxaparin alone, while 14% chose heparin alone, 12% picked clopidogrel alone, and only 8% chose either heparin or enoxaparin plus eptifibatide, a IIb/IIIa inhibitor.
While Dr. McNulty admitted that “there are just too many anticoagulation cocktails out there to compare every permutation across the spectrum of clinical risk,” he offered some general observations.
Heparin certainly has a tried-and-true track record, he said, although it’s not convenient to use. Enoxaparin is very convenient and has been found in studies to have a slight advantage in higher-risk patients. But it does appear to carry a cost in terms of more bleeding.
“It is worth noting that these conclusions of an advantage are based on studies with thousands of patients,” he said. “Small absolute risk reductions may not translate to something clinically meaningful.” In addition, enoxaparin creates issues if the patient eventually requires PCI. That’s because few cath labs have the ability to measure the level of anticoagulation provided by enoxaparin.
Clopidogrel is also convenient and can help with a long-term noninvasive approach. “But it is slow in onset and can lead to a potential delay if CABG is needed,” Dr. McNulty said.
In a higher risk patient, especially one destined for the cath lab, IIb/IIIa inhibitors still play an important role in therapy, he continued.
While the drugs are expensive and put patients at higher risk of bleeding, “they reduce upstream and downstream adverse ischemic events,” he pointed out. It may even be possible to use biomarkers to decide which patients benefit from IIb/IIIa inhibitors. Research published in the April 5, 2006, JAMA, found that the IIa/IIIb inhibitor abciximab reduced the risk of combined ischemic endpoints in patients pre-loaded with clopidogrel who have elevated troponin levels. But, Dr. McNulty noted, “They did not find any real advantage with negative troponin.”
The importance of familiarity
He also mentioned that bivalirudin, which has been primarily studied in invasive contexts, “is an old anticoagulant and one that’s being resurrected.” It causes less bleeding and may even obviate the need to use IIb/IIIa inhibitors, he said.
And now there is yet another player: fondaparinux. “Fondaparinux shows less bleeding head-to-head with enoxaparin, and it offers the same convenience in dosing,” said Dr. McNulty. “But it has the same issues with enoxaparin in taking patients to the cath lab.”
Dr. McNulty stressed that with so many options, anticoagulation choices have become daunting. A real strength in the current guidelines, he said, is that they emphasize the importance of institutions developing their own policies, which helps caregivers become familiar with a few of these medications, among the many to choose from.
That way, physicians can administer the drugs in a timely fashion, appropriately dosed, to patients who benefit from them.
Invasive strategies: early vs. conservative
The third case is that of a 39-year-old insulin-dependent diabetic who presents after one month of increasing exertional dyspnea and neck discomfort with minimal exertion. “He actually had an episode of pain walking from the parking lot to the VA for an elective catheterization,” Dr. McNulty said.
The patient previously had an anterior infarction and angioplasty. While he has no prior history of hypertension, his LDL is 170, and he is taking insulin, aspirin, an ACE inhibitor, a statin and a lopressor. On exam, he is found to be obese with a creatinine of 4.9, up from a recent baseline of between 3 and 4, and his troponin is elevated at 3.3 ng/dl. Compared to a prior ECG, the current test shows marked ventricular hypertrophy.
After starting the patient on heparin, a decision needed to be made: Should the patient be sent immediately to the cath lab, or should physicians take a more conservative approach, sending him first for a stress test and then for catheterization if the stress test was abnormal?
Several studies, Dr. McNulty said, have found that an “early invasive approach is better in terms of reducing the combined end point of death, MI and re-admission for another ACS,” he pointed out. That research includes the RITA 3 trial, published in the Sept. 10, 2005, issue of The Lancet, which actually found a late (five-year) mortality advantage with an early, invasive approach.
However, the ICTUS trial didn’t corroborate those findings. Published in the Sept. 15, 2005, NEJM, Dr. McNulty said, researchers found “no difference in either the incidence of heart attack or of dying from another ischemic event” from a strategy of routine, early invasive risk stratification by routine cardiac catheterization.
Look behind the numbers
But in the ICTUS trial, “Participants were included in the trial solely on troponin positivity,” Dr. McNulty pointed out.
And when you look behind the numbers at the number of actual-as-treated revascularizations, the number of patients in the conservative arm of the ICTUS trial who ultimately ended up getting revascularized was not much different than the number in the early invasive arm who went to catheterization straight away. In fact, the percentage of patients who got revascularized in the “conservative” arm of ICTUS was nearly the same as those in the routine, early invasive arm in RITA 3.
The bottom line appears to be that “just as with anticoagulation, high-risk patients benefit from a more aggressive approach,” Dr. McNulty said. The patient, who did go directly to the cath lab, was found to have restenosis in the diagonal branch, as well as a clot and high grade lesion in the proximal RCA. He had a PCI and a drug-eluting stent, and he was discharged on aspirin and clopidogrel for between six and 12 months.
Phyllis Maguire is Executive Editor of Today’s Hospitalist.
How should you score patient risk?
The TIMI risk score for unstable angina and non-STEMI elevation assigns one point each for seven risk factors. Those factors are:
- age greater than or equal to 65;
- history of cardiovascular disease;
- more than two risk factors;
- recurrent severe angina;
- aspirin use;
- positive biomarkers; and
- ST deviation.
But Edward McNulty, MD, director of the cardiac cath lab at the San Francisco Veterans Affairs Medical Center, says that hospitalists need to remember that, while helpful, risk scores and troponin assays do not replace good clinical judgment and experience.
There is almost a linear relationship between troponin elevation and inpatient mortality.