| Home | E-Submission | Sitemap | Contact us |  
Pediatric Emergency Medicine Journal > Volume 7(2); 2020 > Article
가슴불편감으로 응급실을 방문한 청소년의 임상적 특성: 기저질환 인식의 중요성



We aimed to evaluate the clinical features of adolescents who visit emergency departments (EDs) with chest discomfort, and analyze the implications of underlying medical conditions for the development of cardiogenic chest discomfort.


We reviewed the medical records of adolescents (13-18 years) with chest discomfort who visited a tertiary hospital ED in Seoul, Korea from 2014 through 2018. Sex, age, duration and character of the discomfort, symptoms and signs, ED length of stay, abnormal findings of vital signs, chest radiograph, electrocardiogram, and elevated concentrations of cardiac enzymes were reviewed. Final diagnosis was based on cardiac evaluations within 1 year after the index visit. Underlying medical conditions were defined as visits to the cardiology, pulmonology or hematology-oncology clinics at least twice in the preceding year. Initial suspicious clinical findings were defined as palpitation, syncope or high blood pressure. Logistic regression was used to identify predictors for cardiac etiology.


Of the 231 patients, 43 (18.6%) and 69 (29.9%) had underlying medical conditions and initial suspicious clinical findings, respectively. The predictors for cardiac etiology were underlying medical conditions (odds ratio, 4.28; 95% confidence interval, 1.09-16.73), initial suspicious clinical findings (4.77; 1.36-16.77), abnormal electrocardiogram (11.54; 3.22-41.32), and elevated concentration of troponin I (66.52; 5.37-823.55). The patients with cardiogenic chest discomfort had a longer median ED length of stay (281.0 minutes [interquartile range, 215.5-369.0] vs. 199.5 [132.8-298.0]; P = 0.004) and a higher hospitalization rate (48.3% vs. 13.4%; P < 0.001) than those with non-cardiogenic chest discomfort.


It may be necessary to recognize underlying medical conditions and initial suspicious clinical findings in EDs prior to cardiac evaluation in adolescents with chest discomfort.


Approximately 0.3%-0.6% of pediatric patients who visit emergency departments (EDs) present with chest discomfort, and 16% of them make return visits[1-5]. In addition, 13% of pediatric patients who visit EDs have chest pain as an associated symptom[6]. The etiology of pediatric chest discomfort varies, and is frequently non-cardiac[1,7-10]. Only 0.5%-5% of pediatric patients with chest discomfort have cardiogenic causes, such as arrhythmia, pericarditis, and myocarditis[3,8,11,12]. Consistent with this, a previous Korean study on pediatric patients who visited an ED with chest discomfort reported that cardiogenic and idiopathic chest discomforts accounted for 3.8% and 73.6%, respectively[9].
Despite the low rate of cardiogenic chest discomfort in pediatric patients, caregivers’ concerns are serious, resulting in a 5%-39.3% rate of consultation with cardiologists[4,8,10,12]. Accordingly, recent studies have raised concerns on unnecessary consultations and cardiac evaluations related to chest discomfort[1,4,8,12]. However, we found a lack of data on the predictors for cardiogenic chest discomfort among adolescents, especially in those with underlying medical conditions. Hence, we aimed to investigate the rates of diverse etiology of chest discomfort in adolescents in the ED, and more importantly, predictors for cardiac etiology in the adolescents.


1. Study design and setting

Patients aged 13-18 years who visited the ED from 2014 through 2018 with chest discomfort as the chief complaint were included in this study. Patients who were transferred to other institutions were excluded. The patients with chest discomfort underwent routine measurements of initial vital signs, history taking, and physical examination. Cardiac evaluations were performed at each physician’s discretion. This study was approved by the institutional review board of Samsung Medical Center, which waived the requirement of obtaining informed consent due to the retrospective design (IRB no. 2020-03-071).

2. Definitions of terms

Underlying medical conditions were operationally defined as visits to cardiology, pulmonology or hematology-oncology clinics at least twice in the preceding year. Data were collected regarding the diagnoses that formed the reason for visiting the outpatient clinics. Initial suspicious clinical findings were defined as initial palpitation, syncope or high blood pressure because these signs are related to cardiac etiology. Any findings of chest radiographs other than the report of “no active lung lesions” by radiologists and electrocardiographic (ECG) findings other than sinus rhythm with common variations were considered abnormal. Elevated cardiac enzymes were defined as concentrations greater than the 99% cutoff for troponin I (TnI, > 0.04 ng/mL) and for creatine kinase-MB fraction (CK-MB, > 5.0 ng/mL) or a 4-fold increase in follow-up tests[13]. The final diagnosis was established based on cardiac evaluations performed within 1 year after the ED visit. By this diagnosis, the adolescents were divided into the cardiogenic and non-cardiogenic groups.

3. Data collection

We retrospectively reviewed the medical records of the ED visits, hospitalization, and the above-mentioned clinics from the preceding year of the ED visits to 1-year follow-up. Variables obtained were sex, age, duration (day) and character (e.g., squeezing) of chest discomfort, associated symptoms and signs, ED length of stay (minute), underlying medical conditions, initial suspicious clinical findings, abnormal findings of vital signs, chest radiograph, ECG, and elevated cardiac enzymes.

4. Statistical analysis

A chi-square tests and Mann-Whitney U tests were used to evaluate differences in categorical and continuous variables, respectively. Covariates with P < 0.1 on univariable analysis were entered into a multivariable logistic regression model. Multicollinearity between independent variables was assessed by the variance inflation factor (VIF) values, and any variable with a VIF higher than 4 was excluded. A P < 0.05 was considered significant. We used IBM SPSS Statics for Window ver. 25.0 (IBM Corp., Armonk, NY).


1. Clinical characteristics of the patients

From 2014 through 2018, 240 adolescents with chest discomfort visited the ED, of whom, 9 were excluded from the study by the exclusion criterion. The characteristics of the study population are summarized in Table 1. Most patients experienced chest discomfort for less than a day (64.1%), and complained of experiencing the squeezing type of chest discomfort (35.5%). Forty-three patients (18.6%) had underlying medical conditions (see details of the diagnoses in Appendix 1, https://doi.org/10.22470/pemj.2020.00101). Sixty-nine (29.9%) patients presented with initial suspicious clinical findings. Cardiac murmurs or irregular heartbeats were noted in 4.3% of patients. The median ED length of stay was 209 minutes (interquartile range, 133.3-300.0) with a longer median length in the cardiogenic group than in the non-cardiogenic group (281.0 vs. 199.5 minutes; P = 0.004). Hospitalization rate was 17.7% (41 patients) with a higher rate in the cardiogenic group (48.3% vs. 13.4%; P < 0.001). Two patients were hospitalized in the intensive care unit. The sole mortality occurred in a 13-year-old boy with acute myocarditis superimposed on underlying dilated cardiomyopathy. Although he underwent extracorporeal membrane oxygenation for several months while awaiting a heart transplant, the cause of mortality was presumed to be intracranial hemorrhage.

2. Etiology of chest discomfort

Idiopathic origin was most common (45.0%), followed by pulmonary, cardiovascular, gastrointestinal, musculoskeletal, and psychiatric diseases (Fig. 1). In the cardiogenic group, arrhythmia was the most common diagnosis, followed by pulmonary hypertension, cardiomyopathy, myocarditis or pericarditis, vascular disease, vasovagal syncope, and cardiac rhabdomyosarcoma.

3. Comparison between the cardiogenic and the non-cardiogenic groups

No differences in sex, age, and duration of symptoms were found between the 2 groups. However, there were more patients with initial suspicious clinical findings and elevated cardiac enzymes in the cardiogenic group than in the non-cardiogenic group. Abnormal ECGs were also reported more often in the cardiogenic group (Table 2).

4. Independent predictors for development of cardiogenic chest discomfort

Elevated TnI, abnormal ECGs, initial suspicious clinical findings, and underlying medical conditions were entered into a multivariable logistic regression model. The presence of underlying medical conditions was independently associated with the development of cardiogenic chest discomfort (odds ratio, 4.28; 95% confidence interval, 1.09-16.73). The other independent predictors are listed in Table 3. Although CK-MB was entered into the model, the variable was excluded from the final analysis given its VIF higher than 4.


This study shows that the adolescents with underlying medical conditions were approximately 4.3 times more likely to develop cardiogenic chest discomfort than those without the conditions. In addition, the presence of initial suspicious clinical findings was associated with an approximately 4.8-fold increase in developing the symptom.
Previous studies have shown that an idiopathic chest discomfort is the most common diagnosis among the pediatric population with a 0.5%-5% rate of cardiac etiology[3,8,11,12]. Although idiopathic origin was also most common in our study, we noted a higher rate of cardiac etiology and lower rates of musculoskeletal and psychiatric origins, compared to the equivalent values in the previous studies[1,5,8,9,14]. This difference in etiology might stem from a higher proportion of patients with underlying medical conditions in this present study. None of the 43 patients with underlying medical conditions had an acute illness. Most previous studies were based on the general population or excluded patients with underlying medical conditions[6,8,12,14,15].
The initial suspicious clinical finding is a potential predictor for cardiogenic chest discomfort. This operational definition was based on the most diligently recorded symptoms, such as syncope. Massin et al.[1] reported dyspnea and palpitation as significant findings for cardiogenic chest discomfort at the ED. Other studies have also reviewed various manifestations, such as palpitation, dyspnea, syncope, and dizziness, in pediatric patients and adults aged up to 22 years with chest discomfort[8,10,12]. However, there is a lack of data on the role of the symptoms as predictors for cardiac etiology. Additionally, there is little data on whether high blood pressure is a predictor for cardiac etiology in adolescents. However, a previous study has shown that left ventricular hypertrophy and vascular remodeling seen with hypertension could also increase the cardiovascular risk in pediatric patients[16].
Assays for cardiac enzymes are traditionally used in EDs as simple methods to seek cardiac etiology. Although concentration of TnI was not measured in some patients in this study (Table 2), elevated TnI was more frequently observed in the cardiogenic group (25.9% vs. 0.8%). Conversely, several studies have reported no differences in TnI concentration between pediatric patients with and without cardiogenic chest discomfort[15,17-20]. However, these findings should be cautiously interpreted because elevated TnI is also found in chest trauma, sepsis, intoxication or substance abuse[21].
Abnormal findings of ECG should be interpreted carefully with consideration of other explanations. ECG is easily obtained and repeated at EDs[22]. In the present study, abnormal ECGs were more frequently found in the cardiogenic group, which is consistent with a previous report[14]. However, abnormal ECGs also occur in other conditions, such as poisoning due to caffeine or energy drinks[18-20,23]. The abnormal ECGs can be seen after correction of cardiac anomalies without clinical implications, such as right bundle branch block after correction of ventricular septal defect[24].
Given the single-center and retrospective design, this study has several limitations. First, patients were not systemically evaluated and followed up. This flaw caused missing data, especially the description of symptoms and signs. Second, underlying medical conditions were operationally defined as a history of visits to the specific clinics given the medical environment of tertiary hospitals. This coarse definition may be less generalizable to other settings. However, the use of the definition helped to screen patients with significant underlying medical conditions.
In conclusion, in addition to elevated concentration of TnI and abnormal ECGs, the underlying medical conditions and initial suspicious clinical findings may be associated with the development of cardiogenic chest discomfort in adolescents with chest discomfort.

Conflicts of interest

No potential conflicts of interest relevant to this article were reported.


No funding source relevant to this article was reported.

Fig. 1.
Etiology of chest discomfort (n = 231). Idiopathic origin was most commonly reported (n = 104 [45.0%]). Others were pulmonary (n = 42 [18.2%]), cardiovascular (n = 29 [12.6%]), gastrointestinal (n = 25 [10.8%]), musculoskeletal (n = 16 [6.9%]), and psychiatric diseases (n = 10 [4.3%]). The cardiovascular diseases included arrhythmia (n = 10 [34.5%]), pulmonary hypertension (n = 6 [20.7%]), cardiomyopathy (n = 4 [13.8%]), myocarditis or pericarditis (n = 3 [10.3%]), vascular diseases (n = 3 [10.3%]), vasovagal syncope (n = 2 [6.9%]), and cardiac rhabdomyosarcoma (n = 1 [3.4%]).
Table 1.
Clinical characteristics of the study population
Characteristic Value (N = 231)
Boy 144 (62.3)
Age, y 15.8 (14.0-17.0)
Chest discomfort duration, d
 < 1 148 (64.1)
 1-10 45 (19.5)
 > 10 38 (16.5)
Chest discomfort character
 Unknown 44 (19.0)
 Squeezing 82 (35.5)
 Sharp 68 (29.4)
 Respiratory 17 (7.4)
 Heartburn 10 (4.3)
 Beating 10 (4.3)
 Palpitation 37 (16.0)
 Syncope 4 (1.7)
 Cough/sputum/rhinorrhea 63 (27.3)
 Dyspnea 56 (24.2)
 Wheezing 6 (2.6)
 Postnasal drip 2 (0.9)
 Abdominal discomfort 13 (5.6)
 Nausea/vomiting/diarrhea 37 (16.0)
 Anxiety 14 (6.1)
 Hyperventilation 3 (1.3)
 Back pain 2 (0.9)
 Myalgia 1 (0.4)
 Murmur or irregular heartbeat 10 (4.3)
 Chest wall tenderness 12 (5.2)
 Abnormal lung sound 39 (16.9)
 Epigastric tenderness 20 (8.7)
Vital sign
 High blood pressure 36 (15.6)
 Tachycardia/bradycardia 122 (52.8)
 Fever 15 (6.5)
Initial suspicious clinical findings*
 Palpitation 37 (16.0)
 Syncope 4 (1.7)
 High blood pressure 36 (15.6)
Underlying medical conditions
 Cardiology 22 (9.5)
 Pulmonology 6 (2.6)
 Hematology-oncology 16 (6.9)

Values are expressed as median (interquartile range) or number (%).

* Seven patients had both palpitation and high blood pressure, and one had both palpitation and syncope.

One patient visited both the cardiology and hematology-oncology clinics.

Table 2.
Comparison of the cardiogenic and the non-cardiogenic groups
Variable Cardiogenic group (N = 29) Non-cardiogenic group (N = 202) P value
Boy 15 (51.7) 129 (63.9) 0.224
Age, y 15.8 (14.0-17.0) 15.7 (14.0-17.0) 0.954
Chest discomfort duration < 1 d 20 (69.0) 128 (63.4) 0.680
Initial suspicious clinical finding 15 (51.7) 54 (26.7) 0.009
Underlying medical condition* 15 (51.7) 28 (13.9) < 0.001
 Cardiology 14 (48.3) 8 (4.0) < 0.001
 Pulmonology 0 (0) 6 (3.0) 1.000
 Hematology-oncology 1 (3.4) 15 (7.4) 0.700
Abnormal chest radiograph 2 (6.9) 29/190 (15.3) 0.388
Abnormal electrocardiogram 17 (58.6) 8/151 (5.3) < 0.001
Elevated troponin I 7/27 (25.9) 1/126 (0.8) < 0.001
Elevated creatine kinase-MB fraction 6/27 (22.2) 1/126 (0.8) < 0.001

Values are expressed as median (interquartile range) or number (%).

* One patient visited both the cardiology and hematology-oncology clinics.

The denominators represent the numbers of patients undergoing the tests.

Table 3.
Multivariable logistic regression model for predictors for cardiogenic chest discomfort
Variable Univariable
Odds ratio P value Odds ratio P value
Elevated troponin I 43.75 (5.11-374.76) 0.001 66.52 (5.37-823.55) 0.001
Abnormal electrocardiogram 25.32 (9.08-70.66) < 0.001 11.54 (3.22-41.32) < 0.001
Initial suspicious clinical findings 2.94 (1.33-6.49) 0.008 4.77 (1.36-16.77) 0.015
Underlying medical conditions 6.66 (2.90-15.28) < 0.001 4.28 (1.09-16.73) 0.037

Values are expressed as point estimates and 95% confidence intervals.


1. Massin MM, Bourguignont A, Coremans C, Comte L, Lepage P, Gerard P. Chest pain in pediatric patients presenting to an emergency department or to a cardiac clinic. Clin Pediatr (Phila) 2004;43:231–8.
crossref pmid
2. Massin MM, Montesanti J, Gérard P, Lepage P. Spectrum and frequency of illness presenting to a pediatric emergency department. Acta Clin Belg 2006;61:161–5.
crossref pmid
3. Thull-Freedman J. Evaluation of chest pain in the pediatric patient. Med Clin North Am 2010;94:327–47.
crossref pmid
4. Geggel RL. Conditions leading to pediatric cardiology consultation in a tertiary academic hospital. Pediatrics 2004;114:e409–17.
crossref pmid
5. Selbst SM. Approach to the child with chest pain. Pediatr Clin North Am 2010;57:1221–34.
crossref pmid
6. Khairandish Z, Jamali L, Haghbin S. Role of anxiety and depression in adolescents with chest pain referred to a cardiology clinic. Cardiol Young 2017;27:125–30.
crossref pmid
7. Cava JR, Sayger PL. Chest pain in children and adolescents. Pediatr Clin North Am 2004;51:1553–68, viii.
crossref pmid
8. Saleeb SF, Li WY, Warren SZ, Lock JE. Effectiveness of screening for life-threatening chest pain in children. Pediatrics 2011;128:e1062–8.
crossref pmid
9. Chun JH, Kim TH, Han MY, Kim NY, Yoon KL. Analysis of clinical characteristics and causes of chest pain in children and adolescents. Korean J Pediatr 2015;58:440–5.
crossref pmid pmc pdf
10. Sert A, Aypar E, Odabas D, Gokcen C. Clinical characteristics and causes of chest pain in 380 children referred to a paediatric cardiology unit. Cardiol Young 2013;23:361–7.
crossref pmid
11. Patel S, Sedaghat-Yazdi F, Perez M. Management of pediatric chest pain, palpitations, syncope, and murmur presenting to the emergency department. Clin Pediatr Emerg Med 2018;19:328–39.
12. Friedman KG, Kane DA, Rathod RH, Renaud A, Farias M, Geggel R, et al. Management of pediatric chest pain using a standardized assessment and management plan. Pediatrics 2011;128:239–45.
crossref pmid
13. Ferguson JL, Beckett GJ, Stoddart M, Walker SW, Fox KA. Myocardial infarction redefined: the new ACC/ESC definition, based on cardiac troponin, increases the apparent incidence of infarction. Heart 2002;88:343–7.
crossref pmid pmc
14. Drossner DM, Hirsh DA, Sturm JJ, Mahle WT, Goo DJ, Massey R, et al. Cardiac disease in pediatric patients presenting to a pediatric ED with chest pain. Am J Emerg Med 2011;29:632–8.
crossref pmid
15. Yoldaş T, Örün UA. What is the significance of elevated troponin I in children and adolescents? A diagnostic approach. Pediatr Cardiol 2019;40:1638–44.
crossref pmid pdf
16. Weaver DJ Jr. Hypertension in children and adolescents. Pediatr Rev 2017;38:369–82.
crossref pmid
17. Bhargava V, Dasgupta G, Aly AM. The role of troponin I in the evaluation of chest pain in children and adolescents. EC Pediatr 2016;3:382–9.

18. Sanchis-Gomar F, Pareja-Galeano H, Cervellin G, Lippi G, Earnest CP. Energy drink overconsumption in adolescents: implications for arrhythmias and other cardiovascular events. Can J Cardiol 2015;31:572–5.
crossref pmid
19. Nowak D, Jasionowski A. Analysis of the consumption of caffeinated energy drinks among Polish adolescents. Int J Environ Res Public Health 2015;12:7910–21.
crossref pmid pmc
20. Acra PE, Pérez MT. The evaluation of adolescent chest pain: a screening ECG or PSC-17? Curr Opin Pediatr 2017;29:414–9.
crossref pmid
21. Brown JL, Hirsh DA, Mahle WT. Use of troponin as a screen for chest pain in the pediatric emergency department. Pediatr Cardiol 2012;33:337–42.
crossref pmid pdf
22. Mohan S, Nandi D, Stephens P, MʼFarrej M, Vogel RL, Bonafide CP. Implementation of a clinical pathway for chest pain in a pediatric emergency department. Pediatr Emerg Care 2018;34:778–82.
23. Verghese GR, Friedman KG, Rathod RH, Meiri A, Saleeb SF, Graham DA, et al. Resource utilization reduction for evaluation of chest pain in pediatrics using a vovel Standardized Clinical Assessment and Management Plan (SCAMP). J Am Heart Assoc 2012;1:jah3–e000349.
24. Karadeniz C, Atalay S, Demir F, Tutar E, Ciftci O, Ucar T, et al. Does surgically induced right bundle branch block really effect ventricular function in children after ventricular septal defect closure? Pediatr Cardiol 2015;36:481–8.
crossref pmid pdf


Appendix 1.

Diagnoses of patients with underlying medical conditions
Editorial Office
The Korean Society of Pediatric Emergency Medicine
Chungmu-building office 213, 197, Toegye-ro, Jung-gu, Seoul 04557, Korea
TEL: +82-2-3674-7888   FAX: +82-2-3674-7889   E-mail: pemjpublisher@gmail.com
About |  Browse Articles |  Current Issue |  For Authors
Copyright © Korean Society of Pediatric Emergency Medicine.                 Developed in M2PI
Close layer
prev next