The Case

A man in his 70s with a past medical history of lymphoma in remission, obesity, hypertension, hyperlipidemia, obstructive sleep apnea, and supraventricular tachycardia was sent to the emergency department (ED) by his primary care physician for a two-week history of nightly episodes of chest pain. The pain woke him from sleep, radiated to the left shoulder, and lasted about 15 minutes before spontaneously improving. He denied any associated dyspnea or diaphoresis during these episodes of pain. He further denied having chest pain during the day, although he lived a sedentary life with minimal exertion. The primary care physician had ordered laboratory testing, which was unremarkable except for a slightly elevated D-dimer, which was normal when adjusted for age.

Upon arrival to the ED, the patient stated that his reason for referral was an elevated lab value (D-dimer), rather than chest pain. The medical record from the primary care physician was not available. The vital signs and physical examination were normal and stable. An electrocardiogram was notable for a new right bundle branch block. Serial troponin assays were normal, as were both a chest x-ray and a ventilation-perfusion scan. An exercise stress test from two years prior was reported as normal, although it was later found to be non-diagnostic due to inadequate exercise capacity. Given the lack of new physical examination findings, normal laboratory values, and a record of normal stress testing, the patient was discharged from the ED. The patient passed away at home two days later. Autopsy showed severe coronary artery disease, without other ostensible causes of death.

The Commentary

By Vijay Agusala, MD, MBA, James Deen, MD and Saul Schaefer, MD

This case represents an unfortunate situation in which a complaint of progressive and accelerating nocturnal angina in a patient with multiple cardiac risk factors was not correctly identified as unstable angina. The evaluation focused instead on possible pulmonary embolism in the setting of a minimally positive D-dimer.

Angina is typically described as a pressure-like or squeezing sensation in the chest, associated with inadequate blood and oxygen supply to the heart muscle resulting in ischemia. Typical angina is defined as substernal chest pain that worsens with exertion and is relieved with rest or nitroglycerin administration.¹ Nocturnal angina is associated with a high pretest probability for obstructive coronary atherosclerosis. While some patients with nocturnal angina do not have obstructive coronary artery disease, making their symptoms likely secondary to vasospasm or variant/Prinzmetal angina, other patients with conditions that predispose to increased nocturnal myocardial oxygen demand, such as sleep apnea with hypoxic episodes, may indeed have obstructive coronary disease.² The increased cardiometabolic demands during hypoxic/apneic episodes may lead to anginal symptoms; these symptoms should be attributed to obstructive atherosclerotic coronary disease until ruled out by imaging or other diagnostic procedures, such as coronary angiography or computed tomography.

The electrocardiogram has an integral role in providing clinically actionable information in a patient who presents with chest pain. A normal electrocardiogram in a patient with angina would suggest a low risk of acute coronary syndrome. However, it is important to remember that electrocardiographic changes are not seen in unstable angina, as there is no permanent injury to the myocardium.³ Any interval changes in the electrocardiogram since a patient’s last contact with the medical system should trigger further investigation, whereas chronic or stable findings imply lower risk. New bundle branch blocks, such as the right bundle branch block that the patient in this case had, ST segment changes, T wave changes, and the presence of Q waves could all indicate prior ischemic events and should raise cardiac ischemia higher on the differential diagnosis. New right bundle branch block in the setting of acute coronary syndrome is an independent predictor of all-cause mortality and may be associated with worse outcomes than new left bundle branch block.⁴ This electrocardiographic finding should have prompted further evaluation and work-up of this patient’s chest pain.

Stress testing is the modality of choice in determining if a patient’s symptoms are likely secondary to cardiac ischemia. Functional capacity is a substantial and prognostic component of several modalities of stress testing, such as treadmill stress testing with continuous electrocardiogram monitoring or treadmill stress testing with echocardiography.⁵ Inadequate functional capacity, as seen in this case, greatly interferes with the sensitivity of exercise stress testing. Myocardial perfusion positron emission tomography stress testing does not rely on functional capacity and is a more suitable method of assessing for cardiac ischemia in patients with impaired exercise capacity.⁵

Approach to Improving Safety & Patient Safety Target

When approaching a patient with a chief complaint of chest pain, thoroughness and attention to detail when taking the history are essential. Gastroesophageal reflux disease, pulmonary embolism, asthma, and pleurisy are often on the differential diagnosis when assessing a patient with nocturnal chest discomfort. Thoughtful questioning can help distinguish between true angina and pathology originating from other organ systems. Symptoms suggesting progressive angina require further evaluation and risk stratification, regardless of the chief complaint that brought the patient to seek medical care.

Chest pain triage and risk stratification protocols, when fully implemented, enable systematic evaluation of all patients with chest pain, preventing discharges from the ED or hospital without thorough investigation.⁶ Modern protocols rely heavily on serial high-sensitivity troponin values to either rule in or rule out acute coronary syndrome.⁷ It is important to remember, however, that angina, as opposed to non-ST elevation myocardial infarction (NSTEMI), typically presents with normal troponin levels. Therefore, a patient with risk factors and a compelling history needs further risk stratification/evaluation. A seemingly negative prior stress test is important, but more important is attention to testing parameters that may compromise the sensitivity of the test, such as a lack of adequate functional capacity.

Modern medicine, especially in the ED, can operate on a shift-work basis. It is unclear if transitions of care occurred in this case, but patient volume and turnover demands often affect patient handoffs from one provider to the next. Standardized handoff formats have been developed and can help prevent miscommunications during shift changes.⁸ Effective communication not just within the ED, but also involving referring physicians such as primary care providers, is essential to ensuring that patients are thoroughly evaluated.

System Optimization/Quality Improvement Approach

Several systems-based solutions may help to prevent future cases like this one. First and most importantly, the lack of communication between the outpatient clinic and the ED misdirected this patient’s ED evaluation. While the primary care physician referred the patient to the ED for evaluation of frequent nocturnal chest pain, the primary workup done in the ED was to rule out pulmonary embolism, based on an abnormal laboratory test result that was reported by the patient. A more standardized communication system between the two settings might have affected the workup the patient received and allowed for more thorough evaluation. In our health system, for example, patients who are seen in outpatient clinics but require ED care must be formally referred to the ED, ensuring that patients arrive at the ED with clear written recommendations from the clinicians who know them best.

Second, a standardized risk stratification score like the HEART score helps to inform further evaluation and disposition of patients who present with angina.⁹ The HEART score assigns 0-2 points on each of 5 criteria: history, electrocardiogram, age, number of coronary risk factors, and initial troponin value (Table 1). Patients with elevated HEART scores of 4 or greater have markedly increased of a major adverse cardiac event, defined as acute myocardial infarction, percutaneous coronary intervention, coronary artery bypass graft surgery or death, within 6 weeks of presentation.^10,11 These patients (including the current patient, whose score was 5) should be recommended for inpatient risk stratification with stress testing or cardiac imaging, even with normal high-sensitivity troponin values. Furthermore, as discussed earlier, interval changes in the electrocardiogram should prompt investigation given the increased risk associated with these new abnormalities.

Table 1. HEART Score Criteria

Table adapted from Long et al. (2017)⁹

¹Examples of risk factors include retrosternal pain, pressure, radiation to jaw/left shoulder/arms, duration 5-15 minutes, initiated by exercise/cold/emotion, perspiration, nausea/vomiting, reaction on nitrates within minutes, patient recognizes symptoms. Low risk features of chest pain include: well localized, sharp, non-exertional, no diaphoresis, no nausea or vomiting, and reproducible with palpitation

²Left bundle branch block (LBBB), typical changes suggesting left ventricular hypertrophy (LVH), repolarization disorders suggesting digoxin, unchanged known repolarization disorders.

³Significant ST-segment deviation without LBBB, LVH, or digoxin

⁴Hypertension (HTN), hypercholesterolemia, diabetes mellitus, obesity (BMI >30 kg/m²), smoking (current, or smoking cessation <3mo), positive family history (parent or sibling with cardiovascular disease before age 65).

⁵Use local, regular sensitivity troponin assays and corresponding cutoffs.

Third, non-diagnostic exercise stress tests due to functional limitations, artifacts, or other technical issues should be clearly identified and evident on review of the electronic health record. When a stress test is read as negative for ischemia, any qualifiers affecting the exam result should be just as prominent as the result itself. In this case, the physician who reviewed the patient’s chart was apparently unaware that the recent stress test result was not diagnostic. If this fact had been appreciated and more clearly communicated, another stress testing modality would likely have been pursued, either on the day of the ED visit or shortly thereafter.

Lastly, ED patient volumes have been steadily increasing, motivating the demand for faster patient triage, evaluation, and turnover. In this environment, algorithms such as the HEART score are helpful to reduce provider fatigue and improve ED efficiency¹² while simultaneously allowing for thorough assessment of patients to prevent adverse outcomes.¹³

Take Home Points

• While unstable angina can present in multiple forms and with normal troponin levels, adequate and functionally appropriate evaluation using stress testing methods is necessary for risk stratification in patients with typical pain syndromes.
• While the patient’s understanding of the reason for referral to the ED may be correct, thorough evaluation of the chief complaint, informed by clear communication with the referring clinician, is necessary to prevent suboptimal outcomes.
• Risk stratification scores like the HEART score help identify patients with high pre-test probability for cardiac disease, who can then be considered for more urgent risk stratification testing in the inpatient or observation unit setting.
• Interval development of abnormalities in the electrocardiogram, such as right bundle branch block, are associated with increased risk and should be investigated.
• Institution of standardized patient handoff methods, supported by processes that enable clinical information to be transmitted and received in a timely manner, can facilitate thorough patient evaluation across multiple providers and multiple work shifts.

Vijay Agusala, MD, MBA
Cardiovascular Medicine Fellow
Department of Internal Medicine, Division of Cardiology
UC Davis Health
vagasula@ucdavis.edu

James Deen, MD
Cardiovascular Medicine Fellow 
Department of Internal Medicine, Division of Cardiology
UC Davis Health
jmdeen@ucdavis.edu

Saul Schaefer, MD
Professor 
Department of Internal Medicine, Division of Cardiology
UC Davis Health
sschaefer@ucdavis.edu

References

1. Kloner RA and Chaitman B. Angina and its management. J Cardiovasc Pharmacol Ther. 2017;22:199-209. [Free full text]
2. Franklin KA, Nilsson JB, Sahlin C, et al. Sleep apnoea and nocturnal angina. Lancet. 1995;345:1085-1087. [Available at]
3. Braunwald E. Unstable angina. A classification. Circulation. 1989;80:410-414. [Available at]
4. Alkindi F, El-Menyar A, Rafie I, et al. Clinical Presentations and Outcomes in Patients Presenting With Acute Cardiac Events and Right Bundle Branch Block. Angiology. 2020;71:256-262. [Available at]
5. Joshi PH, de Lemos JA. Diagnosis and management of stable angina: a review. JAMA. 2021;325:1765-1778. [Available at]
6. DeLaney MC, Neth M, Thomas JJ. Chest pain triage: current trends in the emergency departments in the United States. J Nucl Cardiol. 2017;24:2004-2011. [Available at]
7. Morrow DA. Evidence-based algorithms using high-sensitivity cardiac troponin in the emergency department. JAMA Cardiol. 2016;1:379-381. [Free full text]
8. Mardis M, Davis J, Benningfield B, et al. Shift-to-shift handoff effects on patient safety and outcomes. Am J Med Qual. 2017;32:34-42. [Available at]
9. Long B, Oliver J, Streitz, et al. An end-user's guide to the HEART score and pathway. Am J Emerg Med. 2017;35:1350-1355. [Available at]
10. Backus BE, Six AJ, Kelder JC, et al. Chest pain in the emergency room: a multicenter validation of the HEART Score. Crit Pathw Cardiol. 2010;9(3):164-9. [Available at]
11. Backus BE, Six AJ, Kelder JC, et al. A prospective validation of the HEART score for chest pain patients at the emergency department. Int J Cardiol. 2013;168(3):2153-2158. [Free full text]
12. Mahler SA, Riley RF, Hiestand BC, et al. The HEART Pathway randomized trial: identifying emergency department patients with acute chest pain for early discharge. Circ Cardiovasc Qual Outcomes. 2015 Mar;8(2):195-203. [Free full text]
13. Reichlin T, Cullen L, Parsonage WA, et al. Two-hour algorithm for triage toward rule-out and rule-in of acute myocardial infarction using high-sensitivity cardiac troponin T. Am J Med. 2015;128:369-79 e4. [Available at]