So, how can we most effectively incorporate the information and tools we currently use to make better and safer decisions for our chest pain patients? First, we need to recognize the limitations of cardiac biomarkers/enzymes. Is myoglobin helpful? No. Myoglobin is very sensitive. Unfortunately, it is associated with a large number of false positives, resulting in unnecessary admissions for patients with a positive test. If you haven’t already, it’s time to pull the myoglobin from your chest pain/cardiac panel. What about CPK MB? MB is a good test. However, it isn’t a great test. MB has an onset of rise between 3-4 hours after injury and is sustained for 24 hours. The onset of rise is similar to that of the troponins. However, troponin I remains elevated for 7 days. Where MB really falls short is with false negatives. 30% of chest pain patients diagnosed with unstable angina due to a negative MB, have a positive troponin I and actually have a Non-STEMI.
Serial troponin measurements cannot be utilized for disposition decisions. Many EDs discharge patients after serial enzymes. So, how many negative sets do you need before you can send your patient home? To rule out AMI, you only need one, provided you sampled a troponin six hours or greater after the onset of their symptoms (Nagurney, J.T., et al, Journal of Emergency Medicine 29(4):409, November 2005). Sampling two, three, four or even ten troponins won’t give you any more information. Any number of negative troponins will not rule out the entire continuum of acute coronary syndrome. In other words, serial negative troponins may rule out AMI. However, negative they cannot rule out unstable angina.
How valuable is the ECG? Well, one negative ECG isn’t very valuable in isolation. Only 57% of AMI patients’ first ECG will be diagnostic and 35% will have non-specific changes (Welch, R.D., et al, JAMA 286(16):1977, October 2001). A second ECG, repeated between 30 and 60 minutes after the first, is recommended. It is also presumed that a negative ECG, while having symptoms, has greater negative predictive value than one obtained when the patient is asymptomatic. I’m sure to all our surprise, this just isn’t true. A symptomatic patient with a non-diagnostic ECG is no less likely to have ACS than an asymptomatic patient with the same ECG (Chase, M., et al, Academic Emergency Medicine 13(10):1034, October 2006).
Does a response to treatment guide us at all? With regard to “GI” remedies (cocktails, H2 blockers, antacids, etc.) and nitroglycerin, the answer is no. Although “GI” cocktails can be used for symptomatic relief, the response to such treatments should never be used for clinical decision-making. Up to 29% of patients with AMI will experience complete pain relief with the administration of antacids (Teece, S., et al, Journal of Emergency Medicine 20:169, March 2003). A positive response to nitroglycerin doesn’t really help us either. The positive predictive value for ACS from a response to nitroglycerin is fairly poor. Interestingly, patients who respond to nitroglycerin are actually more likely to have a non-cardiac etiology for their chest pain. In the American Journal of Cardiology, 2002, it was reported that for 88% of those with ischemic chest pain responded to nitroglycerin, compared to 92% of those with non-cardiac chest pain. Furthermore, the patients with non-cardiac chest pain were more likely to experience complete resolution of their pain.
Age really matters. Although the medicolegal consequences will be greater for missing an AMI in a younger patient, the chances of a patient having ACS under the age of 40 are significantly less than in patients older than 40. This age group represents between 4% and 8% of all AMI patients. Based on Walker, N.J., et al, Academic Emergency Medicine 8:703, 2001, there is a < 1% chance of ACS at 30 days for patients less than 40 years old, with no risk factors and a normal ECG. Myocardial perfusion imaging (MPI) in ED chest pain patients, such as sestamibi nuclear stress testing, is a valuable modality. However, we have to be careful how we apply the results. MPI studies will detect AMI fairly well, 92-100% sensitivity, per a meta-analysis by Ioannidis in the Annals of Emergency Medicine in May of 2001. Unfortunately, the sensitivity for ACS was only 89%. These tests perform best when the patient is experiencing symptoms and when the exercise component is performed in conjunction with a resting study. The sensitivities of resting studies and exercise studies are 71% and 97% respectively (Fesmire, F. M., et al, Annals of Emergency Medicine 38(3):207, September 2001). However, it is fairly risky to send a patient out of the ED for a nuclear stress test while they are experiencing chest pain. Furthermore, exercising someone with possible acute ischemic chest pain isn’t a very good idea either. To avoid sending the wrong patients home, our goal should be to diagnose the entire continuum of ACS and not just AMI. Recognizing the limitations of an ACS evaluation will help you make the right decision. The best test you have is your clinical judgment.