ADVERTISEMENT

Limited Use of Outpatient Stress Testing in Young Patients With Atypical Chest Pain

Low prevalence of coronary artery disease within this population suggests that younger patients may not require stress testing for chest pain evaluations as long as pretest likelihood is low.
Federal Practitioner. 2018 November;35(6)s:S30-S34
November 6, 2019|Federal Practitioner
LT John C. Chin, MD, MC, USN;CAPT Daniel F. Seidensticker, MD, MC, USN;CDR Andrew H. Lin, MD, MC, USN;Ernest Williams IV, MPH
Author and Disclosure Information

John Chin is an Internal Medicine Resident and Daniel Seidensticker and Andrew Lin are Staff Cardiologists, all at Naval Medical Center Portsmouth. Ernest Williams is an Epidemiologist in the Health Analysis Department, Navy and Marine Corps Public Health Center, Portsmouth, all in Virginia.
Correspondence: John Chin (chinjoh@gmail.com)

Author disclosures
The authors report no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the authors and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the US Government, or any of its agencies.

Demographics and cardiac risk factors (ie, hypertension, hyperlipidemia, diabetes mellitus, and smoking status) were assessed prior to index chest pain evaluations and defined via ICD-9 codes within outpatient records.

Cardiac Testing Outcomes

Patients were initially categorized by the results of baseline electrocardiograms (ECG) and index stress tests (ie, exercise treadmill or stress echocardiography, exercise or Lexiscan myocardial perfusion imaging, dobutamine stress echocardiography). Positive tests were defined as those having electrical or structural ischemic changes. Chronotropic changes were infrequent and nonpathologic and were not counted. Patient endpoints were either additional cardiac testing or negative index stress test without additional testing.

Statistical Analysis

The agreement between both baseline ECG and index stress test as well as index stress test and additional cardiac testing were analyzed using McNemar test and matched-pair odds ratios (ORs) with corresponding 95% CIs. Analyses were stratified by demographics and cardiac risk factors to assess for potential confounding. Analyses were performed using SAS version 9.4 (Cary, NC).

Results

A total of 1,036 patients were evaluated for atypical chest pain and had index stress testing between October 1, 2010 and September 30, 2015. The study cohort was 69% male with a mean (SD) age of 27.3 (4.7) years. More than 60% of the cohort was older than aged > 25 years. 

The most prevalent cardiac risk factor among the study group was smoking (23%), followed by hypertension (15%) and hyperlipidemia (10%) (Table 2).  More than 94% of study patients were referred for index stress testing by their primary care provider.

In the initial testing cohort, exercise treadmill test (59.3%) and exercise echocardiogram (37.1%) were the most common stress testing modalities. The mean (SD) metabolic equivalents (METS) achieved among individuals who performed exercise stress testing was 13.9 (2.8). There were 65 patients who had a positive baseline ECG/negative index stress test, 958 patients had a negative ECG/negative index test, and 8 patients had a negative ECG/positive index test. 

The difference between the first 2 groups (6.27% vs 0.77%) was statistically significant, given χ2 = 44.5; P < .001 (McNemar test); matched-pair OR, 8.125 (95% CI 3.9-16.93, P < .05). There was 93% concordance for the dual negative tests group (Table 3).

There were 102 patients (10%) who performed additional cardiac testing. Among this subgroup, 13 patients (1.3%) had additional testing for further evaluation of a positive index stress test (Table 4) and 89 patients (8.6%) had testing for continuing atypical chest pain despite a negative index stress test. 

The number of additional tests performed exceeded 102 because some patients underwent multiple tests. There were 11 patients that had a positive index stress test/negative additional test, 1 patient had a negative index test/positive additional test, and 88 patients had a negative index test/negative additional test. The difference between the first 2 groups (10.8% vs 0.9%) was statistically significant, (χ2 = 8.33, P < .004 by McNemar test; matched pair OR, 11 [95% CI 1.42-85.2, P < .05]). There was 88% concordance for the dual negative tests group.

Coronary computed tomography angiography (CCTA) demonstrated nonobstructive CAD in 3 patients (0.3%) within the study cohort. There was no obstructive CAD identified in our cohort. Two patients had negative left heart catheterizations (LHC). One of these patients had a negative LHC and a negative Lexiscan after a CCTA showed CAD; all 3 of these additional tests were performed for evaluation of continued chest pain despite negative index stress testing. The positive predictive value of cardiac stress testing for nonobstructive CAD in this low-risk population was 15.4% (2 of 13). Stratification by demographics, CAD risk factors, and cardiac test results revealed no presence of confounding factors during analyses.

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT