A 55-year-old man presents to the emergency department with substernal chest pain. The pain has occurred off and on over the past 2 hours. He has no family history of coronary artery disease. He has no history of diabetes, hypertension, or cigarette smoking. His most recent total cholesterol was 220 mg/dL (HDL, 40; LDL, 155). Blood pressure is 130/70. An ECG obtained on arrival is unremarkable. When he reached the ED, he received a nitroglycerin tablet with resolution of his pain within 4 minutes.
What is the most accurate statement?
A. The chance of CAD in this man over the next 10 years was 8% before his symptoms and is now greater than 20%.
B. The chance of CAD in this man over the next 10 years was 8% and is still 8%.
C. The chance of CAD in this man over the next 10 years was 15% before his symptoms and is now close to 100%.
D. The chance of CAD in this man over the next 10 years was 15% before his symptoms and is now close to 50%.
For years, giving nitroglycerin to patients who present with chest pain has been considered a good therapy, and the response to the medication has been considered a sign that the pain was likely due to cardiac ischemia. Is there evidence that this is true?
The study was a retrospective review of 223 patients who presented to the ED over a 5-month period with ongoing chest pain. They looked at patients who had ongoing chest pain in the ED, received nitroglycerin, and did not receive any therapy other than aspirin within 10 minutes of receiving nitroglycerin. Nitroglycerin response was compared with the final diagnosis of cardiac versus noncardiac chest pain.
Of the patients with a final determination of cardiac chest pain, 88% had a nitroglycerin response, whereas 92% of the patients with noncardiac chest pain had a nitroglycerin response (P = .50).
Deborah B. Diercks, MD, and her colleagues looked at improvement in chest pain scores in the ED in patients treated with nitroglycerin and whether it correlated with a cardiac etiology of chest pain.2 The study was a prospective, observational study of 664 patients in an urban tertiary care ED over a 16-month period. An 11-point numeric chest pain scale was assessed and recorded by research assistants before and 5 minutes after receiving nitroglycerin. The scale ranged from 0 (no pain) to 10 (worst pain imaginable).
A final diagnosis of a cardiac etiology for chest pain was found in 18% of the patients in the study. Of the patients who had cardiac-related chest pain, 20% had no reduction in pain with nitroglycerin, compared with 19% of the patients without cardiac-related chest pain. Complete or significant reduction in chest pain occurred with nitroglycerin in 31% of patients with cardiac chest pain and 27% of the patients without cardiac chest pain (P = .76).
Two other studies with similar designs showed similar results. Robert Steele, MD, and his colleagues studied 270 patients in a prospective observational cohort study of patients with chest pain presenting to an urban ED.3 Patients presenting to the ED with active chest pain who received nitroglycerin were enrolled.
The sensitivity in this study for nitroglycerin relief determining cardiac chest pain was 72%, and the specificity was 37%, with a positive likelihood ratio for coronary artery disease if nitroglycerin response of 1.1 (0.96-1.34).
In another prospective, observational cohort study, 459 patients who presented to an ED with chest pain were evaluated for response to nitroglycerin as a marker for ischemic cardiac disease.4 In this study, presence of ischemic cardiac disease was defined as diagnosis in the ED or during a 4-month follow-up period. Nitroglycerin relieved chest pain in 35% of patients who had coronary disease, whereas 41% of patients without coronary disease had a nitroglycerin response. This study had a much lower overall nitroglycerin response rate than any of the other studies.
Katherine Grailey, MD, and Paul Glasziou, MD, PhD, published a meta-analysis of nitroglycerin use for the diagnosis of chest pain, using the above referenced studies. They concluded that in the acute setting, nitroglycerin is not a reliable test of treatment for use in diagnosis of coronary artery disease.5
High response rate for nitroglycerin in the noncoronary artery groups in the studies may be due to a strong placebo effect and/or that nitroglycerin may help with pain caused by esophageal spasm. The lack of specificity in the pain relief response for nitroglycerin makes it not a helpful test. Note that all the studies have been in the acute, ED setting for chest pain. In the case presented at the beginning of the article, the response the patient had to nitroglycerin would not change the probability that he has coronary artery disease.
Dr. Paauw is professor of medicine in the division of general internal medicine at the University of Washington, Seattle, and he serves as third-year medical student clerkship director at the University of Washington. Contact Dr. Paauw at firstname.lastname@example.org .