Pediatr Cardiol (2015) 36:1452–1457 DOI 10.1007/s00246-015-1185-6
ORIGINAL ARTICLE
A Smartphone Application to Diagnose the Mechanism of Pediatric Supraventricular Tachycardia Dina J. Ferdman1 • Leonardo Liberman1 • Eric S. Silver1
Received: 4 January 2015 / Accepted: 30 April 2015 / Published online: 10 May 2015 Ó Springer Science+Business Media New York 2015
Abstract Smartphone applications that record a singlelead ECG are increasingly available. We sought to determine the utility of a smartphone application (AliveCor) to record supraventricular tachycardia (SVT) and to distinguish atrioventricular reentrant tachycardia (AVRT) from atrioventricular nodal reentrant tachycardia (AVNRT) in pediatric patients. A prior study demonstrated that interpretation of standard event and Holter monitors accurately identifies the tachycardia mechanism in only 45 % of recordings. We performed an IRB-approved prospective study in pediatric patients undergoing an ablation for SVT. Tracings were obtained by placing the smartphone in three different positions on the chest (PI—horizontal, PII—rotated 60° clockwise, and PIII—rotated 120° clockwise). Two blinded pediatric electrophysiologists jointly analyzed a pair of sinus and tachycardia tracings in each position. Tracings with visible retrograde P waves were classified as AVRT. The three positions were compared by Chi-square test. Thirty-seven patients (age 13.7 ± 2.8 years) were enrolled in the study. Twenty-four had AVRT, and 13 had AVNRT. One hundred and eight pairs of tracings were obtained. The correct diagnosis was made in 27/37 (73 %) with position PI, 28/37 (76 %) with PII, and 20/34 (59 %) with PIII (p = 0.04 for PII vs. PIII and p = NS for other comparisons). A single-lead ECG obtained with a smartphone monitor can successfully record SVT in pediatric patients and can predict the SVT mechanism at least as
& Eric S. Silver [email protected] 1
Division of Pediatric Cardiology, Department of Pediatrics, College of Physicians and Surgeons, Columbia University Medical Center, 3959 Broadway, 2-North, New York, NY 10032, USA
123
well as previously published reports of Holter monitors, along with the added convenience of not requiring patients to carry a dedicated monitor. Keywords Smartphone Pediatric supraventricular tachycardia Atrioventricular nodal reentrant tachycardia Atrioventricular reentrant tachycardia
Introduction Supraventricular tachycardia (SVT) is a common heart rhythm disturbance in the pediatric population. AV reentry tachycardia (AVRT) and AV nodal reentry tachycardia (AVNRT) are the two most common forms of tachyarrhythmias in pediatric patients, and both are due to reentry circuits through the AV node. In AVRT, the most common arrhythmia seen in pediatrics, there is an accessory pathway bridging between the ventricle and the atrium to form the retrograde limb of the circuit in tachycardia. AVNRT is the second most common arrhythmia in children, with an increasing incidence with age into adolescence [9]. AVNRT occurs when the reentry circuit involves two pathways within tissue associated with the AV node itself. Symptoms of SVT may include palpitations, chest pain, discomfort, syncope, or near syncope. For diagnosis and treatment purposes, it is extremely helpful to be able to accurately capture a rhythm disturbance at the time of symptoms. SVT may be diagnosed in the hospital with an ECG or a bedside monitor, or as an outpatient with a monitor, such as a Holter monitor, an event recorder, or an implantable loop monitor. Event recorders have been shown to be more successful in recording the arrhythmia at the time of symptomatic episodes [7]. However, the
Pediatr Cardiol (2015) 36:1452–1457
transtelephonic handheld event monitors that are used in pediatric patients are still not optimally used, as only 48–59 % of symptomatic patients will transmit a recording [3, 16]. SVT can be managed with vagal maneuvers or adenosine for acute termination and/or oral medications for prevention. In older children with refractory SVT, the treatment of choice in pediatrics is catheter ablation [10]. Being able to predict the mechanism of tachycardia prior to beginning an invasive study has potential benefit both for patient counseling and in the technical approach to the study. Some patients with episodes of AVRT may already be known to have an accessory pathway if preexcitation has been noted on a baseline ECG. For the remainder of patients with a normal ECG, both AVRT and AVNRT produce a narrow complex tachycardia that may be difficult to distinguish on a single-lead recording. A prior study demonstrated that interpretation of ECGs obtained solely by standard event and that Holter monitors accurately identify the mechanism of tachycardia in only 45 % of recordings [5]. Smartphone use has become pervasive in daily life [13]. The development of a smartphone application to record a high-quality single-lead ECG makes portable event monitoring more accessible to our patients. This technology has been validated in adults with a normal ECG and those with atrial fibrillation [11]. We sought to use the smartphone application (AliveCor, Inc., San Francisco, CA) in pediatric patients to record SVT and to distinguish AVRT from AVNRT.
1453
smartphone case directly on the patient’s chest. Cardiac electrical activity from the electrodes is transmitted from the case to the smartphone wirelessly. The smartphone application converts the signal to a digital ECG tracing, which can be viewed in real time and stored (Fig. 1). All recordings were obtained at a paper speed of 25 mm/s with a gain of 10 mm/mV, which is the same as a standard ECG lead. The electrophysiology study proceeded as is standard practice [1]. At the start, the monitor was placed on the patient’s upper left chest to record baseline measurements in sinus rhythm in three different positions: PI—horizontal positioning, PII—rotated 60° clockwise, and PIII—rotated 120° clockwise (Fig. 2). Upon induction of tachycardia during the electrophysiology study, tracings were obtained in the same three positions. For those patients with preexcitation, baseline sinus rhythm tracings were collected
Methods We performed an IRB-approved prospective study in pediatric patients undergoing an ablation for SVT. Informed consent was obtained from all individual participants included in the study. Inclusion criteria were all patients who presented to our center for an electrophysiology study with the preexisting diagnosis of a symptomatic supraventricular arrhythmia or a diagnosis of preexcitation on a resting ECG. Exclusion criteria were those patients presenting for an electrophysiology study for ablation of other less common tachycardia mechanisms (e.g., a long RP tachycardia, an atrial tachycardia, or a ventricular tachycardia) and those with complex congenital heart disease as we wanted to solely focus on distinguishing AVRT and AVNRT in patients with normal cardiac anatomy. We used an iPhone version of the AliveCor monitor, which is a single-channel bipolar ECG recorder. Recording commences with placement of two hands on the two electrodes at the back of the monitor, which is also a case for the smartphone. It can also be obtained by placing the
Fig. 1 AliveCor iPhone case and screen shot (sinus and SVT from the same patient)
123
1454
Pediatr Cardiol (2015) 36:1452–1457
diagnosis obtained during the electrophysiology study. Each tracing of sinus rhythm was compared to the patient’s 12-lead ECG in sinus rhythm for determination of similarity to standard lead placement. Results are expressed as a mean ± standard deviation. Chi-square or Fisher’s exact test was used to compare the differences for categorical variables. T test was used to compare continuous variables. All statistics were obtained using SPSS Statistics. A p value \0.05 was considered significant.
Results
Fig. 2 Lead placement
after the successful ablation. A diagnosis of AVRT or AVNRT was made during the electrophysiology study. Tracings were then randomized as an individual pair of sinus and tachycardia tracings in each position. They were also separately randomized with all three positions for each individual patient reviewed as a group. All tracings were reviewed jointly by two blinded pediatric electrophysiologists for the diagnosis of AVRT versus AVNRT. Tracings with visible retrograde P waves were classified as AVRT, and those without were classified as AVNRT (Fig. 3). All responses were compared to the known
Fig. 3 SVT tracings, a AVRT with visible retrograde P waves (arrow), b AVNRT with no visible retrograde P waves
123
Between June 2013 and February 2014, 37 patients were enrolled in the study (Table 1). Twenty-four had AVRT, and 13 had AVNRT. All patients underwent successful ablation of the arrhythmia substrate. The patients ranged in age from 8 to 18 years (mean 13.7 ± 2.8 years). The AVNRT patients were older than those with AVRT (15.0 ± 1.8 vs. 13.0 ± 2.9 years, p = 0.031) and had greater height and weight (Table 1). All patients had easily obtainable tracings in all positions in sinus rhythm. Thirtyfour patients had easily obtainable tracings in all three positions in tachycardia. The remaining three patients had only two tachycardia tracings obtained due to limitations in sustained tachycardia. One hundred and eight pairs of tracings were obtained. All pairs of tracings were correctly identified as sinus versus tachycardia. For overall arrhythmia diagnosis, the correct diagnosis was made in 27/37 (73 %) with position PI, 28/37 (76 %) with PII, and 20/34 (59 %) with PIII (p = NS) (Table 2). Separating out each arrhythmia, a correct diagnosis of AVNRT was made in 11/13 (85 %) with PI, 10/13 (77 %) with PII, and 10/12 (83 %) with PIII (p = NS). A correct diagnosis of AVRT was made in 16/24 (67 %) with PI, 18/24 (75 %) with PII, and 10/22 (45 %) with PIII (p = 0.04 for PII vs. PIII, p = NS for other comparisons; Table 2). A diagnosis of AVRT using position PI had a sensitivity of 67 % and positive predictive value of 89 %, as compared to PII 75 and 86 % and PIII 45 and 83 %, respectively (Table 3). Using all three lead positions for each patient together did not yield a significant difference in diagnosis, as a diagnosis of AVRT with all leads together had a diagnostic accuracy of 71 %, sensitivity of 68 %, and positive predictive value of 83 %. When comparing our tracings with a standard 12-lead ECG in sinus rhythm, PI was similar to one ECG lead in 27 cases (73 %), PII in 32 cases (86 %), and PIII in 19 cases (51 %). There was variability in the standard leads with
Pediatr Cardiol (2015) 36:1452–1457
1455
Table 1 Patient characteristics
Total n = 37
AVNRT n = 13
AVRT n = 24
p value
Age (year)
13.7 (±2.8)
15.0 (±1.8)
13.0 (±2.9)
0.031
Height (cm)
159.6 (±13.3)
169.3 (±9.3)
154.4 (±12.3)
0.001
Weight (kg)
53.9 (±11.1)
51.1 (±10.7)
0.037
Male
20/37 (54 %)
Table 2 Correct diagnosis by lead position and underlying rhythm AVNRT (n = 13)
AVRT (n = 24)
Overall correct
PI
11/13 (85 %)
16/24 (67 %)
PII
10/13 (77 %)
18/24 (75 %)
28/37 (76 %)
PIII
10/12 (83 %)
10/22 (45 %)
20/34 (59 %)
Totals
31/38 (84 %)
44/70 (63 %)
75/108 (69 %)
Position 27/37 (73 %)
which they were associated; however, the most frequent similarity was PII to standard leads V3 or V4 in 59 % of the tracings.
Discussion Smartphone use has become pervasive in both medicine and society in general [2, 13, 14]. As more medical applications are being developed, they must be evaluated for efficacy in our patient population. Thus far for pediatric arrhythmias, only applications using the camera flashlight on the smartphone, typically held against the fingertip, have been evaluated to detect heart rate. The accuracy of these smartphone applications to report heart rate decreased with rates of 120–200 bpm [4, 19], limiting its utility when trying to diagnose SVT in a pediatric population. Thus far there is a paucity of data on the utility of this technology for arrhythmia detection and/or diagnosis, especially in pediatric patients. The AliveCor monitor has been used to successfully differentiate sinus rhythm from atrial fibrillation utilizing the single-lead ECG data in adults [11, 17] and has been suggested for use in suspected cases of ST elevation myocardial infarction [21]. There have been no reported use of Table 3 Diagnostic accuracy for AVRT
59.0 (±10.2) 6/13 (46 %)
14/24 (58 %)
0.48
it as an alternative to a standard Holter or event monitor and no reported use of it for arrhythmia diagnosis other than atrial fibrillation. We found the smartphone monitor to be an easy-to-use device to record SVT that is easily distinguishable from sinus rhythm. Looking further at SVT diagnosis, the two most common causes of pediatric SVT are AVRT and AVNRT. The overall incidence of AVRT in pediatric patients with SVT has been reported to be 72–80 % overall; however, when excluding infants, the incidence decreases to 56–68 %, with the incidence of AVNRT at 20–34 % [9, 18, 20]. Our cohort demonstrated similar frequencies, with 65 % of patients with AVRT and 35 % with AVNRT. Our two groups differed slightly in that the AVNRT group tended to be older and bigger in size, which correlates with the higher incidence of AVNRT in older patients. Analysis of 12-lead ECGs of tachycardia has demonstrated several predictors of SVT diagnosis. The presence of a retrograde P wave separated from the QRS complex suggests AVRT, while the presence of a pseudo-r’ deflection in lead V1 and a pseudo-S wave in the inferior leads (II and aVF) suggests the diagnosis of AVNRT [6]. In pediatric patients, Jaeggi et al. [5] analyzed 102 ECGs obtained prior to an electrophysiology study. The authors suggest an algorithm where the first step is the presence of a pseudo-r’ wave at the end of the QRS that was not present on the baseline QRS in sinus rhythm as being consistent with AVNRT. The second step stated that when retrograde P waves were identified, an RP interval [100 ms was more likely to be AVRT. Using these algorithms to predict AVRT versus AVNRT in a 12-lead ECG, they reported 81–91 % accuracy. Given that the baseline QRS width of a typical adolescent is about 90 ms [15], this essentially means that if a retrograde P wave is seen following the QRS, it is more suggestive of AVRT. The AliveCor application records a bipolar chest lead, which is different from the standard unipolar precordial
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
PI
67
84
89
58
PII
75
77
86
62
PIII
45
83
83
45
Three leads combined
68
75
83
56
123
1456
leads on a standard ECG. We found a majority of our tracings resembled the mid-precordial lead tracings, but there was not complete agreement that would make them equivalent to any specific standard lead. Because of this, we chose to focus on identifying a retrograde P wave after the QRS as indicative of AVRT. We obtained similar diagnostic accuracy with our position PI—horizontal positioning, and PII—rotated 60° clockwise, with slightly less accurate results in PIII—rotated 120° clockwise. In a standard 12-lead ECG, right precordial lead V1 has been shown to be the most accurate single lead to identify SVT mechanism, with successful identification in 68 % of AVRT and 62.5 % of AVNRT diagnoses [12]. This correlates with our rates of accurate diagnosis in our precordial lead from the smartphone monitor. Holter monitors typically provide one- to three-channel ECG tracings, which can be effective at diagnosing a rhythm disturbance in general, such as sinus rhythm versus SVT. However, a single study demonstrated that Holter monitors do not provide a high level of diagnostic accuracy in predicting the mechanism of a narrow QRS tachycardia. In the same study by Jaeggi et al. [5], they analyzed Holter monitor tracings and could correctly distinguish the arrhythmia mechanism in only 45 %. A limitation of our study is the controlled setting in which our tracings were obtained. The patients were all in the supine position under varying degrees of sedation. When used in the outpatient setting, we anticipate our patients obtaining the ECG by either placing the monitor in both hands or holding the monitor on their chest by themselves or a caregiver. The quality of the tracing may not be equivalent due to poor contact or motion artifact. As technology improves, there are likely to be more developments implementing smartphones to aid in the monitoring of various physiologic parameters, including cardiac rhythm [8]. With this study, we introduce the use of this smartphone cardiac monitor in diagnosing pediatric SVT. This could potentially be used as a more user-friendly and long-term alternative to transtelephonic event recorders which are more cost-effective than Holter monitors for pediatric SVT diagnosis [7], but are still not optimally used by patients.
Conclusion A single-lead ECG obtained with a smartphone monitor can be used to successfully record SVT in pediatric patients. When the monitor is used in the horizontal position or rotated 60° clockwise, it is similar to a precordial lead and may allow for identification of retrograde P waves to diagnose AVRT. A single tracing with the AliveCor application can predict the SVT diagnosis at least as well as
123
Pediatr Cardiol (2015) 36:1452–1457
previously published reports of Holter monitors along with the added convenience of not requiring patients to carry a dedicated monitor. Acknowledgments AliveCor provided use of the monitor (value $200) to the authors for research purposes. Conflict of interest The authors have no potential conflicts of interest.
References 1. Fogoros RN (2012) Electrophysiologic testing, 5th edn. Wiley, Chichester 2. Franko OI, Tirrell TF (2012) Smartphone app use among medical providers in ACGME training programs. J Med Syst 36(5):3135–3139 3. Goldstein MA, Hesslein P, Dunnigan A (1990) Efficacy of transtelephonic electrocardiographic monitoring in pediatric patients. Am J Dis Child 144(2):178–182 4. Ho C-L, Fu Y-C, Lin M-C, Chan S-C, Hwang B, Jan S-L (2013) Smartphone applications (Apps) for heart rate measurement in children: comparison with electrocardiography monitor. Pediatr Cardiol 35(4):726–731 5. Jaeggi ET, Gilljam T, Bauersfeld U, Chiu C, Gow R (2003) Electrocardiographic differentiation of typical atrioventricular node reentrant tachycardia from atrioventricular reciprocating tachycardia mediated by concealed accessory pathway in children. Am J Cardiol 91(9):1084–1089 6. Kalbfleisch SJ, El-Atassi R, Calkins H, Langberg JJ, Morady F (1993) Differentiation of paroxysmal narrow QRS complex tachycardias using the 12-lead electrocardiogram. J Am Coll Cardiol 21(1):85–89 7. Karpawich PP, Cavitt DL, Sugalski JS (1993) Ambulatory arrhythmia screening in symptomatic children and young adults: comparative effectiveness of Holter and telephone event recordings. Pediatr Cardiol 14(3):147–150 8. Kennedy HL (2013) The evolution of ambulatory ECG monitoring. Prog Cardiovasc Dis 56(2):127–132 9. Ko JK, Deal BJ, Strasburger JF, Benson DW (1992) Supraventricular tachycardia mechanisms and their age distribution in pediatric patients. Am J Cardiol 69(12):1028–1032 10. Kugler JD, Danford DA, Deal BJ et al (1994) Radiofrequency catheter ablation for tachyarrhythmias in children and adolescents. The pediatric electrophysiology society. N Engl J Med 330(21):1481–1487 11. Lau JK, Lowres N, Neubeck L et al (2013) iPhone ECG application for community screening to detect silent atrial fibrillation: a novel technology to prevent stroke. Int J Cardiol 165(1):193–194 12. Liberman L, Pass RH, Starc TJ (2008) Optimal surface electrocardiogram lead for identification of the mechanism of supraventricular tachycardia in children. Pediatr Emerg Care 24(1):28–30 13. Oulasvirta A, Rattenbury T, Ma L, Raita E (2011) Habits make smartphone use more pervasive. Pers Ubiquitous Comput 16(1):105–114 14. Radesky JS, Kistin CJ, Zuckerman B et al (2014) Patterns of mobile device use by caregivers and children during meals in fast food restaurants. Pediatrics 133(4):e843–e849 15. Rijnbeek PR, Witsenburg M, Schrama E, Hess J, Kors J (2001) New normal limits for the paediatric electrocardiogram. Eur Heart J 22(8):702–711
Pediatr Cardiol (2015) 36:1452–1457 16. Saarel EV, Stefanelli CB, Fischbach PS, Serwer GA, Rosenthal A, Dick M (2004) Transtelephonic electrocardiographic monitors for evaluation of children and adolescents with suspected arrhythmias. Pediatrics 113(2):248–251 17. Saxon L (2013) Ubiquitous wireless ECG recording: a powerful tool physicians should embrace. J Cardiovasc Electrophysiol 24(4):480–483 18. Tanel RE, Walsh EP, Triedman JK, Epstein MR, Bergau DM, Saul JP (1997) Five-year experience with radiofrequency catheter ablation: implications for management of arrhythmias in pediatric and young adult patients. J Pediatr 131(6):878–887
1457 19. Wackel P, Beerman L, West L, Arora G (2014) Tachycardia detection using smartphone applications in pediatric patients. J Pediatr 164(5):1133–1135 20. Weindling SN, Saul JP, Walsh EP (1996) Efficacy and risks of medical therapy for supraventricular tachycardia in neonates and infants. Am Heart J 131(1):66–72 21. Wong C-K (2013) iPhone ECG monitoring–the gateway to the new paradigm of STEMI therapy. Int J Cardiol 168(3):2897–2898
123
ORIGINAL ARTICLE
A Smartphone Application to Diagnose the Mechanism of Pediatric Supraventricular Tachycardia Dina J. Ferdman1 • Leonardo Liberman1 • Eric S. Silver1
Received: 4 January 2015 / Accepted: 30 April 2015 / Published online: 10 May 2015 Ó Springer Science+Business Media New York 2015
Abstract Smartphone applications that record a singlelead ECG are increasingly available. We sought to determine the utility of a smartphone application (AliveCor) to record supraventricular tachycardia (SVT) and to distinguish atrioventricular reentrant tachycardia (AVRT) from atrioventricular nodal reentrant tachycardia (AVNRT) in pediatric patients. A prior study demonstrated that interpretation of standard event and Holter monitors accurately identifies the tachycardia mechanism in only 45 % of recordings. We performed an IRB-approved prospective study in pediatric patients undergoing an ablation for SVT. Tracings were obtained by placing the smartphone in three different positions on the chest (PI—horizontal, PII—rotated 60° clockwise, and PIII—rotated 120° clockwise). Two blinded pediatric electrophysiologists jointly analyzed a pair of sinus and tachycardia tracings in each position. Tracings with visible retrograde P waves were classified as AVRT. The three positions were compared by Chi-square test. Thirty-seven patients (age 13.7 ± 2.8 years) were enrolled in the study. Twenty-four had AVRT, and 13 had AVNRT. One hundred and eight pairs of tracings were obtained. The correct diagnosis was made in 27/37 (73 %) with position PI, 28/37 (76 %) with PII, and 20/34 (59 %) with PIII (p = 0.04 for PII vs. PIII and p = NS for other comparisons). A single-lead ECG obtained with a smartphone monitor can successfully record SVT in pediatric patients and can predict the SVT mechanism at least as
& Eric S. Silver [email protected] 1
Division of Pediatric Cardiology, Department of Pediatrics, College of Physicians and Surgeons, Columbia University Medical Center, 3959 Broadway, 2-North, New York, NY 10032, USA
123
well as previously published reports of Holter monitors, along with the added convenience of not requiring patients to carry a dedicated monitor. Keywords Smartphone Pediatric supraventricular tachycardia Atrioventricular nodal reentrant tachycardia Atrioventricular reentrant tachycardia
Introduction Supraventricular tachycardia (SVT) is a common heart rhythm disturbance in the pediatric population. AV reentry tachycardia (AVRT) and AV nodal reentry tachycardia (AVNRT) are the two most common forms of tachyarrhythmias in pediatric patients, and both are due to reentry circuits through the AV node. In AVRT, the most common arrhythmia seen in pediatrics, there is an accessory pathway bridging between the ventricle and the atrium to form the retrograde limb of the circuit in tachycardia. AVNRT is the second most common arrhythmia in children, with an increasing incidence with age into adolescence [9]. AVNRT occurs when the reentry circuit involves two pathways within tissue associated with the AV node itself. Symptoms of SVT may include palpitations, chest pain, discomfort, syncope, or near syncope. For diagnosis and treatment purposes, it is extremely helpful to be able to accurately capture a rhythm disturbance at the time of symptoms. SVT may be diagnosed in the hospital with an ECG or a bedside monitor, or as an outpatient with a monitor, such as a Holter monitor, an event recorder, or an implantable loop monitor. Event recorders have been shown to be more successful in recording the arrhythmia at the time of symptomatic episodes [7]. However, the
Pediatr Cardiol (2015) 36:1452–1457
transtelephonic handheld event monitors that are used in pediatric patients are still not optimally used, as only 48–59 % of symptomatic patients will transmit a recording [3, 16]. SVT can be managed with vagal maneuvers or adenosine for acute termination and/or oral medications for prevention. In older children with refractory SVT, the treatment of choice in pediatrics is catheter ablation [10]. Being able to predict the mechanism of tachycardia prior to beginning an invasive study has potential benefit both for patient counseling and in the technical approach to the study. Some patients with episodes of AVRT may already be known to have an accessory pathway if preexcitation has been noted on a baseline ECG. For the remainder of patients with a normal ECG, both AVRT and AVNRT produce a narrow complex tachycardia that may be difficult to distinguish on a single-lead recording. A prior study demonstrated that interpretation of ECGs obtained solely by standard event and that Holter monitors accurately identify the mechanism of tachycardia in only 45 % of recordings [5]. Smartphone use has become pervasive in daily life [13]. The development of a smartphone application to record a high-quality single-lead ECG makes portable event monitoring more accessible to our patients. This technology has been validated in adults with a normal ECG and those with atrial fibrillation [11]. We sought to use the smartphone application (AliveCor, Inc., San Francisco, CA) in pediatric patients to record SVT and to distinguish AVRT from AVNRT.
1453
smartphone case directly on the patient’s chest. Cardiac electrical activity from the electrodes is transmitted from the case to the smartphone wirelessly. The smartphone application converts the signal to a digital ECG tracing, which can be viewed in real time and stored (Fig. 1). All recordings were obtained at a paper speed of 25 mm/s with a gain of 10 mm/mV, which is the same as a standard ECG lead. The electrophysiology study proceeded as is standard practice [1]. At the start, the monitor was placed on the patient’s upper left chest to record baseline measurements in sinus rhythm in three different positions: PI—horizontal positioning, PII—rotated 60° clockwise, and PIII—rotated 120° clockwise (Fig. 2). Upon induction of tachycardia during the electrophysiology study, tracings were obtained in the same three positions. For those patients with preexcitation, baseline sinus rhythm tracings were collected
Methods We performed an IRB-approved prospective study in pediatric patients undergoing an ablation for SVT. Informed consent was obtained from all individual participants included in the study. Inclusion criteria were all patients who presented to our center for an electrophysiology study with the preexisting diagnosis of a symptomatic supraventricular arrhythmia or a diagnosis of preexcitation on a resting ECG. Exclusion criteria were those patients presenting for an electrophysiology study for ablation of other less common tachycardia mechanisms (e.g., a long RP tachycardia, an atrial tachycardia, or a ventricular tachycardia) and those with complex congenital heart disease as we wanted to solely focus on distinguishing AVRT and AVNRT in patients with normal cardiac anatomy. We used an iPhone version of the AliveCor monitor, which is a single-channel bipolar ECG recorder. Recording commences with placement of two hands on the two electrodes at the back of the monitor, which is also a case for the smartphone. It can also be obtained by placing the
Fig. 1 AliveCor iPhone case and screen shot (sinus and SVT from the same patient)
123
1454
Pediatr Cardiol (2015) 36:1452–1457
diagnosis obtained during the electrophysiology study. Each tracing of sinus rhythm was compared to the patient’s 12-lead ECG in sinus rhythm for determination of similarity to standard lead placement. Results are expressed as a mean ± standard deviation. Chi-square or Fisher’s exact test was used to compare the differences for categorical variables. T test was used to compare continuous variables. All statistics were obtained using SPSS Statistics. A p value \0.05 was considered significant.
Results
Fig. 2 Lead placement
after the successful ablation. A diagnosis of AVRT or AVNRT was made during the electrophysiology study. Tracings were then randomized as an individual pair of sinus and tachycardia tracings in each position. They were also separately randomized with all three positions for each individual patient reviewed as a group. All tracings were reviewed jointly by two blinded pediatric electrophysiologists for the diagnosis of AVRT versus AVNRT. Tracings with visible retrograde P waves were classified as AVRT, and those without were classified as AVNRT (Fig. 3). All responses were compared to the known
Fig. 3 SVT tracings, a AVRT with visible retrograde P waves (arrow), b AVNRT with no visible retrograde P waves
123
Between June 2013 and February 2014, 37 patients were enrolled in the study (Table 1). Twenty-four had AVRT, and 13 had AVNRT. All patients underwent successful ablation of the arrhythmia substrate. The patients ranged in age from 8 to 18 years (mean 13.7 ± 2.8 years). The AVNRT patients were older than those with AVRT (15.0 ± 1.8 vs. 13.0 ± 2.9 years, p = 0.031) and had greater height and weight (Table 1). All patients had easily obtainable tracings in all positions in sinus rhythm. Thirtyfour patients had easily obtainable tracings in all three positions in tachycardia. The remaining three patients had only two tachycardia tracings obtained due to limitations in sustained tachycardia. One hundred and eight pairs of tracings were obtained. All pairs of tracings were correctly identified as sinus versus tachycardia. For overall arrhythmia diagnosis, the correct diagnosis was made in 27/37 (73 %) with position PI, 28/37 (76 %) with PII, and 20/34 (59 %) with PIII (p = NS) (Table 2). Separating out each arrhythmia, a correct diagnosis of AVNRT was made in 11/13 (85 %) with PI, 10/13 (77 %) with PII, and 10/12 (83 %) with PIII (p = NS). A correct diagnosis of AVRT was made in 16/24 (67 %) with PI, 18/24 (75 %) with PII, and 10/22 (45 %) with PIII (p = 0.04 for PII vs. PIII, p = NS for other comparisons; Table 2). A diagnosis of AVRT using position PI had a sensitivity of 67 % and positive predictive value of 89 %, as compared to PII 75 and 86 % and PIII 45 and 83 %, respectively (Table 3). Using all three lead positions for each patient together did not yield a significant difference in diagnosis, as a diagnosis of AVRT with all leads together had a diagnostic accuracy of 71 %, sensitivity of 68 %, and positive predictive value of 83 %. When comparing our tracings with a standard 12-lead ECG in sinus rhythm, PI was similar to one ECG lead in 27 cases (73 %), PII in 32 cases (86 %), and PIII in 19 cases (51 %). There was variability in the standard leads with
Pediatr Cardiol (2015) 36:1452–1457
1455
Table 1 Patient characteristics
Total n = 37
AVNRT n = 13
AVRT n = 24
p value
Age (year)
13.7 (±2.8)
15.0 (±1.8)
13.0 (±2.9)
0.031
Height (cm)
159.6 (±13.3)
169.3 (±9.3)
154.4 (±12.3)
0.001
Weight (kg)
53.9 (±11.1)
51.1 (±10.7)
0.037
Male
20/37 (54 %)
Table 2 Correct diagnosis by lead position and underlying rhythm AVNRT (n = 13)
AVRT (n = 24)
Overall correct
PI
11/13 (85 %)
16/24 (67 %)
PII
10/13 (77 %)
18/24 (75 %)
28/37 (76 %)
PIII
10/12 (83 %)
10/22 (45 %)
20/34 (59 %)
Totals
31/38 (84 %)
44/70 (63 %)
75/108 (69 %)
Position 27/37 (73 %)
which they were associated; however, the most frequent similarity was PII to standard leads V3 or V4 in 59 % of the tracings.
Discussion Smartphone use has become pervasive in both medicine and society in general [2, 13, 14]. As more medical applications are being developed, they must be evaluated for efficacy in our patient population. Thus far for pediatric arrhythmias, only applications using the camera flashlight on the smartphone, typically held against the fingertip, have been evaluated to detect heart rate. The accuracy of these smartphone applications to report heart rate decreased with rates of 120–200 bpm [4, 19], limiting its utility when trying to diagnose SVT in a pediatric population. Thus far there is a paucity of data on the utility of this technology for arrhythmia detection and/or diagnosis, especially in pediatric patients. The AliveCor monitor has been used to successfully differentiate sinus rhythm from atrial fibrillation utilizing the single-lead ECG data in adults [11, 17] and has been suggested for use in suspected cases of ST elevation myocardial infarction [21]. There have been no reported use of Table 3 Diagnostic accuracy for AVRT
59.0 (±10.2) 6/13 (46 %)
14/24 (58 %)
0.48
it as an alternative to a standard Holter or event monitor and no reported use of it for arrhythmia diagnosis other than atrial fibrillation. We found the smartphone monitor to be an easy-to-use device to record SVT that is easily distinguishable from sinus rhythm. Looking further at SVT diagnosis, the two most common causes of pediatric SVT are AVRT and AVNRT. The overall incidence of AVRT in pediatric patients with SVT has been reported to be 72–80 % overall; however, when excluding infants, the incidence decreases to 56–68 %, with the incidence of AVNRT at 20–34 % [9, 18, 20]. Our cohort demonstrated similar frequencies, with 65 % of patients with AVRT and 35 % with AVNRT. Our two groups differed slightly in that the AVNRT group tended to be older and bigger in size, which correlates with the higher incidence of AVNRT in older patients. Analysis of 12-lead ECGs of tachycardia has demonstrated several predictors of SVT diagnosis. The presence of a retrograde P wave separated from the QRS complex suggests AVRT, while the presence of a pseudo-r’ deflection in lead V1 and a pseudo-S wave in the inferior leads (II and aVF) suggests the diagnosis of AVNRT [6]. In pediatric patients, Jaeggi et al. [5] analyzed 102 ECGs obtained prior to an electrophysiology study. The authors suggest an algorithm where the first step is the presence of a pseudo-r’ wave at the end of the QRS that was not present on the baseline QRS in sinus rhythm as being consistent with AVNRT. The second step stated that when retrograde P waves were identified, an RP interval [100 ms was more likely to be AVRT. Using these algorithms to predict AVRT versus AVNRT in a 12-lead ECG, they reported 81–91 % accuracy. Given that the baseline QRS width of a typical adolescent is about 90 ms [15], this essentially means that if a retrograde P wave is seen following the QRS, it is more suggestive of AVRT. The AliveCor application records a bipolar chest lead, which is different from the standard unipolar precordial
Sensitivity (%)
Specificity (%)
PPV (%)
NPV (%)
PI
67
84
89
58
PII
75
77
86
62
PIII
45
83
83
45
Three leads combined
68
75
83
56
123
1456
leads on a standard ECG. We found a majority of our tracings resembled the mid-precordial lead tracings, but there was not complete agreement that would make them equivalent to any specific standard lead. Because of this, we chose to focus on identifying a retrograde P wave after the QRS as indicative of AVRT. We obtained similar diagnostic accuracy with our position PI—horizontal positioning, and PII—rotated 60° clockwise, with slightly less accurate results in PIII—rotated 120° clockwise. In a standard 12-lead ECG, right precordial lead V1 has been shown to be the most accurate single lead to identify SVT mechanism, with successful identification in 68 % of AVRT and 62.5 % of AVNRT diagnoses [12]. This correlates with our rates of accurate diagnosis in our precordial lead from the smartphone monitor. Holter monitors typically provide one- to three-channel ECG tracings, which can be effective at diagnosing a rhythm disturbance in general, such as sinus rhythm versus SVT. However, a single study demonstrated that Holter monitors do not provide a high level of diagnostic accuracy in predicting the mechanism of a narrow QRS tachycardia. In the same study by Jaeggi et al. [5], they analyzed Holter monitor tracings and could correctly distinguish the arrhythmia mechanism in only 45 %. A limitation of our study is the controlled setting in which our tracings were obtained. The patients were all in the supine position under varying degrees of sedation. When used in the outpatient setting, we anticipate our patients obtaining the ECG by either placing the monitor in both hands or holding the monitor on their chest by themselves or a caregiver. The quality of the tracing may not be equivalent due to poor contact or motion artifact. As technology improves, there are likely to be more developments implementing smartphones to aid in the monitoring of various physiologic parameters, including cardiac rhythm [8]. With this study, we introduce the use of this smartphone cardiac monitor in diagnosing pediatric SVT. This could potentially be used as a more user-friendly and long-term alternative to transtelephonic event recorders which are more cost-effective than Holter monitors for pediatric SVT diagnosis [7], but are still not optimally used by patients.
Conclusion A single-lead ECG obtained with a smartphone monitor can be used to successfully record SVT in pediatric patients. When the monitor is used in the horizontal position or rotated 60° clockwise, it is similar to a precordial lead and may allow for identification of retrograde P waves to diagnose AVRT. A single tracing with the AliveCor application can predict the SVT diagnosis at least as well as
123
Pediatr Cardiol (2015) 36:1452–1457
previously published reports of Holter monitors along with the added convenience of not requiring patients to carry a dedicated monitor. Acknowledgments AliveCor provided use of the monitor (value $200) to the authors for research purposes. Conflict of interest The authors have no potential conflicts of interest.
References 1. Fogoros RN (2012) Electrophysiologic testing, 5th edn. Wiley, Chichester 2. Franko OI, Tirrell TF (2012) Smartphone app use among medical providers in ACGME training programs. J Med Syst 36(5):3135–3139 3. Goldstein MA, Hesslein P, Dunnigan A (1990) Efficacy of transtelephonic electrocardiographic monitoring in pediatric patients. Am J Dis Child 144(2):178–182 4. Ho C-L, Fu Y-C, Lin M-C, Chan S-C, Hwang B, Jan S-L (2013) Smartphone applications (Apps) for heart rate measurement in children: comparison with electrocardiography monitor. Pediatr Cardiol 35(4):726–731 5. Jaeggi ET, Gilljam T, Bauersfeld U, Chiu C, Gow R (2003) Electrocardiographic differentiation of typical atrioventricular node reentrant tachycardia from atrioventricular reciprocating tachycardia mediated by concealed accessory pathway in children. Am J Cardiol 91(9):1084–1089 6. Kalbfleisch SJ, El-Atassi R, Calkins H, Langberg JJ, Morady F (1993) Differentiation of paroxysmal narrow QRS complex tachycardias using the 12-lead electrocardiogram. J Am Coll Cardiol 21(1):85–89 7. Karpawich PP, Cavitt DL, Sugalski JS (1993) Ambulatory arrhythmia screening in symptomatic children and young adults: comparative effectiveness of Holter and telephone event recordings. Pediatr Cardiol 14(3):147–150 8. Kennedy HL (2013) The evolution of ambulatory ECG monitoring. Prog Cardiovasc Dis 56(2):127–132 9. Ko JK, Deal BJ, Strasburger JF, Benson DW (1992) Supraventricular tachycardia mechanisms and their age distribution in pediatric patients. Am J Cardiol 69(12):1028–1032 10. Kugler JD, Danford DA, Deal BJ et al (1994) Radiofrequency catheter ablation for tachyarrhythmias in children and adolescents. The pediatric electrophysiology society. N Engl J Med 330(21):1481–1487 11. Lau JK, Lowres N, Neubeck L et al (2013) iPhone ECG application for community screening to detect silent atrial fibrillation: a novel technology to prevent stroke. Int J Cardiol 165(1):193–194 12. Liberman L, Pass RH, Starc TJ (2008) Optimal surface electrocardiogram lead for identification of the mechanism of supraventricular tachycardia in children. Pediatr Emerg Care 24(1):28–30 13. Oulasvirta A, Rattenbury T, Ma L, Raita E (2011) Habits make smartphone use more pervasive. Pers Ubiquitous Comput 16(1):105–114 14. Radesky JS, Kistin CJ, Zuckerman B et al (2014) Patterns of mobile device use by caregivers and children during meals in fast food restaurants. Pediatrics 133(4):e843–e849 15. Rijnbeek PR, Witsenburg M, Schrama E, Hess J, Kors J (2001) New normal limits for the paediatric electrocardiogram. Eur Heart J 22(8):702–711
Pediatr Cardiol (2015) 36:1452–1457 16. Saarel EV, Stefanelli CB, Fischbach PS, Serwer GA, Rosenthal A, Dick M (2004) Transtelephonic electrocardiographic monitors for evaluation of children and adolescents with suspected arrhythmias. Pediatrics 113(2):248–251 17. Saxon L (2013) Ubiquitous wireless ECG recording: a powerful tool physicians should embrace. J Cardiovasc Electrophysiol 24(4):480–483 18. Tanel RE, Walsh EP, Triedman JK, Epstein MR, Bergau DM, Saul JP (1997) Five-year experience with radiofrequency catheter ablation: implications for management of arrhythmias in pediatric and young adult patients. J Pediatr 131(6):878–887
1457 19. Wackel P, Beerman L, West L, Arora G (2014) Tachycardia detection using smartphone applications in pediatric patients. J Pediatr 164(5):1133–1135 20. Weindling SN, Saul JP, Walsh EP (1996) Efficacy and risks of medical therapy for supraventricular tachycardia in neonates and infants. Am Heart J 131(1):66–72 21. Wong C-K (2013) iPhone ECG monitoring–the gateway to the new paradigm of STEMI therapy. Int J Cardiol 168(3):2897–2898
123
