Electrocardiograms have utility in disease stratification and monitoring in adult pulmonary arterial hypertension (PAH). We examined the electrocardiographic findings that are common in pediatric PAH and assessed for correlation with disease severity and progression. We retrospectively identified patients aged ≤18 years followed at a single institution from January 2001 to June 2012 with catheterization-confirmed diagnosis of idiopathic PAH and PAH secondary to pulmonary developmental abnormalities. Patients with an electrocardiography performed within 60 days of catheterization were included. Primary and secondary outcomes are the prevalence of abnormal electrocardiographic findings at the time of catheterization and the association between electrocardiographic and hemodynamic findings and electrocardiographic changes with disease progression on follow-up catheterization, respectively. Of the 100 electrocardiography-catheterization pairs derived from the 46 patients identified, 93% had an electrocardiographic abnormality: 78% had right ventricular hypertrophy (RVH) and 52% had right axis deviation (RAD) for age. In patients with idiopathic PAH, the presence of RVH and RAD correlated with pulmonary vascular resistance and transpulmonary gradient. RAD and RVH on baseline electrocardiogram was associated with an increased risk of disease progression on subsequent catheterization (odds ratio 11.0, 95% confidence interval 1.3 to 96.2, p = 0.03) after adjusting for PAH subgroup. The sensitivity, specificity, and positive and negative predictive values of RAD and RVH on baseline electrocardiogram for disease progression were 92%, 48%, 33%, and 95%, respectively. In conclusion, electrocardiographic abnormalities are common in pediatric PAH. RAD and RVH on electrocardiogram were associated with worse hemodynamics, whereas their absence is suggestive of a lack of disease progression.
Pulmonary arterial hypertension (PAH) is a sequela to a myriad of congenital heart disease pulmonary developmental abnormalities, other disease states, or may be classified as idiopathic. The incidence of PAH is estimated to be from 2 to 3 per million. PAH is often progressive in nature and can culminate in right ventricular (RV) failure. In infants and children, the presenting symptoms of PAH are typically nonspecific—the most common being dyspnea on exertion and fatigue. Tools currently relied on in the evaluation of PAH include echocardiography, cardiac magnetic resonance imaging, and cardiac catheterization, which require considerable expense, may be limited in availability, and in the case of catheterization, is invasive. In contrast, a 15-lead electrocardiography (ECG) is a relatively economical, accessible, and noninvasive tool. Although the ECG has been found to be an insensitive screening test for PAH in adults, baseline electrocardiographic parameters, including P-wave amplitude, right precordial QRS morphology, and right ventricular hypertrophy (RVH), have been found to be predictive of mortality in adults with PAH. Changes in QRS and T-wave axes on electrocardiogram also have been observed to correlate with disease progression or improvement. In pediatrics, there are limited data delineating the electrocardiographic criteria that may indicate disease severity and progression. This study seeks to elucidate the electrocardiographic parameters that would aid in the assessment and monitoring of PAH in children.
Methods
A retrospective descriptive study was conducted at The Children’s Hospital of Philadelphia. From January 1, 2001, to June 30, 2012, all patients diagnosed with idiopathic PAH (IPAH) and associated PAH secondary to pulmonary developmental abnormalities (APAH-PD) according to the Dana Point classification 1.1 and 3.7, respectively, were identified. Inclusion criteria were as follows: (1) age 0 months to 18 years at the time of first catheterization, (2) ≥1 hemodynamic catheterization to confirm or follow the diagnosis of PAH, and (3) an ECG that is performed within 60 days of catheterization. The ECG performed closest to the date of catheterization was used. Exclusion criteria included patients who had the ECG or catheterization performed at an outside institution or if congenital heart disease or cardiac malposition was present.
ECGs were performed and stored electronically using the MUSE Cardiology Information System (GE Healthcare, Waukesha, Wisconsin). Age-related electrocardiographic norms and upper limits were derived from published tables. Electrocardiographic parameters examined were based on those previously explored and defined a priori. Right atrial enlargement was present if (1) the patient was in sinus rhythm and (2) P-wave amplitude in lead II was ≥2.5 mV. RVH by voltage criteria was defined by the presence of either (1) R-wave amplitude in lead V 1 or V 4 R ≥98th percentile for age or (2) S-wave amplitude in lead V 6 ≥98th percentile for age. Right axis deviation (RAD) was identified based on the age-related norms. Strain was considered present if the QRS-T wave angle was ≥90°. Other parameters evaluated include (1) P-wave duration in lead II, (2) R:S ratio in lead V 1 , (3) T-wave amplitude in lead V 1 , (4) QRS duration in leads I, V 4 R, and V 6 , (5) presence or absence of qR or rSR′ pattern in leads V 1 and/or V 4 R, and (6) right bundle branch block.
Catheterization was performed under anesthesia with vasoreactivity testing using 100% oxygen followed by 100% oxygen and 40 ppm of inhaled nitric oxide. The hemodynamic data abstracted were mean pulmonary arterial pressure (mPAP), mean pulmonary capillary wedge pressure, right atrial pressure, cardiac index, and indexed pulmonary vascular resistance (iPVR) at baseline and with vasoreactivity testing. The transpulmonary gradient (TPG) was calculated as TPG = mPAP − mean pulmonary capillary wedge pressure. All 3 of the following hemodynamic data on the first catheterization at our institution were required for the diagnosis of PAH: (1) mPAP ≥25 mm Hg at rest, (2) mPCWP ≤15 mm Hg, and (3) iPVR ≥3 WU/m 2 .
We defined positive responders to pulmonary vasoreactivity testing as ≥20% decrease in mPAP without a change in cardiac index (within 5%). For patients with multiple catheterization-ECG pairs, we defined an increase or decrease in iPVR by 20% between catheterizations as disease progression or improvement, respectively.
The primary outcome was the prevalence of individual electrocardiographic parameters in patients with PAH at catheterization as a total cohort and stratified by Dana Point classification. Each successive catheterization-ECG pair was analyzed individually, and prevalence was calculated as the ratio of the number of electrocardiographic criteria of interest and the total number of catheterization-ECG pairs. Secondary outcomes were (1) association between electrocardiographic and hemodynamic parameters, (2) change in electrocardiographic parameters with disease progression or improvement, and (3) association between baseline electrocardiographic parameters with change in the disease status.
Normally distributed continuous variables are expressed as mean ± SD and were compared using the Student t test or are expressed as median with interquartile range if it is skewed and were compared using Wilcoxon rank-sum test. Categorical electrocardiographic variables are described with frequency counts and percentages and were compared using the chi-square test. Correlation analyses (Pearson or Spearman) were used to assess the relation between continuous electrocardiographic and hemodynamic catheterization variables. Positive predictive value and negative predictive value (NPV), sensitivity, and specificity of baseline electrocardiographic parameters for disease progression were calculated. Change in electrocardiographic findings between baseline and follow-up hemodynamic assessments was assessed using pair t test and Wilcoxon signed-rank test. We used univariate logistic regression analysis with forward selection to test for associations between the secondary outcome and 4 possible covariates: Dana Point classification, age at baseline catheterization, gender, and vasoreactivity drug testing. Covariates demonstrating univariate associations with p values <0.20 qualified for inclusion in the final multivariate logistic regression model that tested the association between the relevant electrocardiographic findings and change in disease. Two-tailed p values ≤0.05 were deemed significant.
All analyses were performed using Stata, version 10.0 (Stata Corp, College Station, Texas). This project was approved by the Institutional Review Board of The Children’s Hospital of Philadelphia.
Result
Twenty-seven patients with IPAH and 39 with APAH-PD were identified during the study time frame. Five patients with IPAH and 15 with APAH-PD did not meet criteria for inclusion. Forty-six patients were ultimately evaluated: 22 (48%) with IPAH and 24 (52%) with APAH-PD. Of those with APAH-PD, 23 (96%) had chronic lung disease of prematurity and 1 (4%) had pulmonary hypoplasia secondary to congenital diaphragmatic hernia. This yielded a total of 100 catheterization-ECG pairs. Demographic and hemodynamic data are summarized in Table 1 .
| Total (n = 100) | IPAH (n = 54) | APAH-PD (n = 46) | p-Value | |
|---|---|---|---|---|
| Age at catheterization (months) | 51.5 (18.5–202.0) | 129 (30.0–202.0) | 19 (5.0–59.0) | <0.0001 ∗ |
| Female | 27 (59.0%) | 16 (72.7%) | 11 (45.8%) | <0.001 ∗ |
| Days between ECG and catheterization † | 1 (−48 to 42) | 1 (−13 to 24) | 3.5 (−25 to 29) | 0.23 |
| Mean pulmonary arterial pressure (mmHg) | 42 (21–83) | 53.5 (26–82) | 32 (22–62) | <0.0001 ∗ |
| Mean pulmonary capillary wedge pressure (mmHg) | 11 (5–16) | 11 (6–15) | 11 (8–15) | 0.98 |
| Mean right atrial pressure (mmHg) | 7 (2–13) | 7 (2–12) | 8 (5–13) | 0.24 |
| Cardiac index (L/min/m 2 ) | 3.8 (2.2–6.6) | 3.6 (2.4–5.5) | 4 (2.5–6.5) | 0.06 |
| Indexed pulmonary vascular resistance (Woods units/m 2 ) | 7.9 (2.2–27.9) | 9.9 (3.6–27.9) | 4.8 (2.4–11.7) | <0.0001 ∗ |
| Positive vasoreactivity testing ‡ | 15 (30.0%) | 5 (17.9%) | 10 (45.5%) | 0.04 ∗ |
† Negative and positive values represent before and after the day of catheterization, respectively.
‡ Total sample size of patients with vasoreactivity testing is 50 (IPAH: n = 28; APAH-PD: n = 22).
Sinus rhythm was present in all the electrocardiograms collected. Table 2 summarizes the measured electrocardiographic parameters. The greater median heart rate of patients with APAH-PD may be related to their younger age as the heart rate is inversely related to age (ρ = −0.59, p <0.0001). Of the 100 ECG-catheterization pairs, 93 had at least 1 electrocardiographic abnormality. Voltage criteria for RVH were commonly seen in both subgroups. RAD was likewise common particularly in patients with IPAH. The combination of RVH and RAD was also observed more frequently in patients with IPAH.
| Electrocardiographic Variable | Total (n = 100) | IPAH (n = 54) | APAH-PD (n = 46) | p-Value |
|---|---|---|---|---|
| Heart rate (bpm) | 97 (67–166) | 88 (69–136) | 118 (79–165) | <0.0001 ∗ |
| Right atrial enlargement | 18 (18.0%) | 8 (14.3%) | 10 (21.7%) | 0.37 |
| P duration (msec) | 90 (60–132) | 100 (80–132) | 80 (60–100) | <0.0001 ∗ |
| V1 QRS duration (msec) | 81.3 ± 17.1 | 88.2 ± 15.4 | 73.4 ± 15.6 | <0.0001 ∗ |
| V1 R > 98 th percentile † | 40 (41.7%) | 22 (43.1%) | 18 (40.0%) | 0.76 |
| V1 R/S ratio † | 2.1 (0.3–19.2) | 1.78 (0.3–10.1) | 2.6 (0.8–18.8) | 0.03 ∗ |
| V4 QRS duration † (msec) | 75.8 ± 18.3 | 83.4 ± 16.3 | 66.3 ± 15.3 | <0.0001 ∗ |
| V4R > 98 th percentile † | 60 (62.5%) | 30 (56.6%) | 30 (69.8%) | 0.19 |
| Right bundle branch block | 3 (3.1%) | 1 (2.0%) | 2 (4.3%) | 0.49 |
| rSR′ or qR pattern V1 or V4R | 57 (57.0%) | 26 (49.1%) | 31 (67.4%) | 0.05 ∗ |
| V6 QRS duration (msec) | 74 (50–108) | 80 (60–100) | 62 (50–84) | <0.0001 ∗ |
| V6 S > 98 th percentile † | 55 (56.7%) | 25 (47.2%) | 30 (68.2%) | 0.04 ∗ |
| Right axis deviation | 52 (52.0%) | 36 (66.8%) | 16 (34.7%) | 0.001 ∗ |
| QRS-T axis > 90 degrees | 32 (32.0%) | 23 (42.6%) | 9 (19.6%) | 0.01 ∗ |
| Right ventricular hypertrophy | 78 (78.0%) | 40 (74.1%) | 38 (82.6%) | 0.30 |
| Right ventricular hypertrophy + right axis deviation | 48 (48.0%) | 34 (63.0%) | 14 (30.4%) | 0.001 ∗ |
| Right ventricular hypertrophy + right axis deviation + right ventricular strain | 28 (28.0%) | 22 (40.7%) | 6 (13.0%) | 0.002 ∗ |
| Right ventricular hypertrophy + right axis deviation + right ventricular strain + right atrial enlargement | 7 (7.0%) | 7 (13.0%) | 2 (4.3%) | 0.13 |
| Any abnormality | 93 (93.0%) | 49 (90.7%) | 44 (95.7%) | 0.34 |
P-wave amplitude (ρ = 0.30 to 0.51, p ≤0.03), R-wave amplitude in lead V 1 (ρ = 0.43 to 0.58, p ≤0.002) and V 4 R (ρ = 0.43 to 0.59, p ≤0.001), S-wave amplitude in lead V 6 (ρ = 0.29 to 0.40, p ≤0.04), and QRS axis (ρ = 0.43 to 0.57, p ≤0.001) were significantly correlated with mPAP, TPG, and iPVR in patients with IPAH. However, individual electrocardiographic parameters were not observed to be significantly associated with hemodynamic findings for patients with APAH-PD. Observation of RAD and RVH simultaneously on electrocardiograms was associated with greater mPAP, TPG, and iPVR in patients with IPAH only ( Table 3 ).
| Electrocardiographic Finding | Present | Absent | p-Value | |
|---|---|---|---|---|
| Hemodynamic variable | ||||
| Right atrial enlargement † | Mean right atrial pressure (mmHg) | 7 (6–8) | 7 (2–12) | 0.89 |
| Mean pulmonary arterial pressure (mmHg) | 72 (69–75) | 50 (26–72) | 0.001 ∗ | |
| Transpulmonary gradient (mmHg) | 61 (60–62) | 38 (15–61) | 0.0006 ∗ | |
| Cardiac index (L/min/m 2 ) | 3.7 (3.0–4.3) | 3.6 (2.6–5.5) | 0.93 | |
| Indexed pulmonary vascular resistance (Wu/m 2 ) | 21.9 (21.0–22.7) | 9.4 (3.6–20.3) | 0.01 ∗ | |
| Right axis deviation ‡ | Mean right atrial pressure (mmHg) | 7 (4–11) | 8 (5–10) | 0.86 |
| Mean pulmonary arterial pressure (mmHg) | 59 (34–82) | 39 (26–49) | 0.0001 ∗ | |
| Transpulmonary gradient (mmHg) | 49 (22–68) | 26 (15–38) | <0.0001 ∗ | |
| Cardiac index (L/min/m 2 ) | 3.6 (2.4–5.5) | 3.7 (3.1–4.3) | 0.81 | |
| Indexed pulmonary vascular resistance (Wu/m 2 ) | 11.8 (5.4–27.9) | 6.7 (3.6–10.7) | 0.003 ∗ | |
| Right ventricular hypertrophy § | Mean right atrial pressure (mmHg) | 7 (4–11) | 8 (6–11) | 0.73 |
| Mean pulmonary arterial pressure (mmHg) | 57 (32–82) | 37 (27–40) | 0.001 ∗ | |
| Transpulmonary gradient (mmHg) | 48 (22–68) | 23 (16–36) | 0.0009 ∗ | |
| Cardiac index (L/min/m 2 ) | 3.6 (2.4–5.5) | 3.7 (2.7–4.2) | 0.81 | |
| Indexed pulmonary vascular resistance (Wu/m 2 ) | 10.4 (4.6–27.9) | 7.7 (4.7–10.7) | 0.05 | |
| Right axis deviation + right ventricular hypertrophy ¶ | Mean right atrial pressure (mm Hg) | 7 (4–11) | 8 (5–11) | 0.68 |
| Mean pulmonary arterial pressure (mm Hg) | 59 (40–82) | 38.5 (26.0–60.0) | 0.0001 ∗ | |
| Transpulmonary gradient (mm Hg) | 49 (32–68) | 26 (15–45) | <0.0001 ∗ | |
| Cardiac index (L/min/m 2 ) | 3.6 (2.4–5.5) | 3.7 (2.7–4.4) | 0.86 | |
| Indexed pulmonary vascular resistance (Wu/m 2 ) | 12.7 (5.4–27.9) | 8.2 (3.6–10.8) | 0.004 ∗ | |
† Present: n = 8; absent: n = 46.
‡ Present: n = 36; absent: n = 18.
§ Present: n = 40; absent: n = 14.
Thirty-three of the 46 patients (72%) had serial catheterizations, and 21 of the 33 (64%) had >1 follow-up hemodynamic catheterization. This generated a total of 54 consecutive catheterization-ECG pairs. The median number of days between catheterizations was 366 (interquartile range 56 to 868) for the 2 groups combined. Twelve (22%) had an increase and 25 (46%) had a decrease in iPVR by ≥20%. There was no difference in the number of days between catheterization between patients with IPAH and those with APAH-PD (p = 0.48), and a similar number of patients from the 2 groups had 1 and 2 serial catheterizations (p = 0.50) and change in iPVR (p = 0.85). As the number of patients who progressed on follow-up catheterization was small, both PAH subgroups were analyzed together.
In the 54 catheterization-ECG pairs, the interval between catheterizations was similar between those whose disease improved or remained unchanged and worsened (351 vs 374 days, p = 0.36). Twenty-five patients demonstrated a significant improvement in hemodynamic findings at follow-up catheterization ( Table 4 ). Results of the vasodilatory drug testing on baseline catheterization were available for 50 of the 54 catheterization-ECG pairs, and a positive response was seen on 15 evaluations (30%). However, all 11 patients with worsened hemodynamics at follow-up had a negative vasoreactivity drug test result at baseline catheterization, which precluded our ability to assess vasodilatory drug testing as a covariate.
| Baseline Catheterization | Follow-Up Catheterization | p-Value | |
|---|---|---|---|
| Disease worsened (n = 12) | |||
| Mean pulmonary arterial pressure (mmHg) | 44.5 (29.0–60.5) | 62.0 (33.5–74.5) | 0.11 |
| Transpulmonary gradient (mmHg) | 34.5 (16.5–50.5) | 52.5 (27.5–64.0) | 0.01 ∗ |
| Cardiac index (L/min/m 2 ) | 3.9 (3.6–4.3) | 3.8 (2.8–4.4) | 0.24 |
| Indexed pulmonary vascular resistance (WU/m 2 ) | 6.9 (4.3–11.2) | 12.3 (8.3–24.1) | 0.008 ∗ |
| Disease unchanged (n = 17) | |||
| Mean pulmonary arterial pressure (mmHg) | 42.0 (38.0–54.0) | 42.0 (34.0–64.0) | 0.92 |
| Transpulmonary gradient (mmHg) | 32.0 (27.0–44.0) | 36.0 (20.0–55.0) | 0.87 |
| Cardiac index (L/min/m 2 ) | 3.6 (3.0–4.2) | 3.3 (2.6–3.8) | 0.38 |
| Indexed pulmonary vascular resistance (WU/m 2 ) | 8.9 (5.0–11.1) | 9.6 (5.0–11.1) | 0.92 |
| Disease improved (n = 25) | |||
| Mean pulmonary arterial pressure (mmHg) | 54.0 (33.0–62.0) | 32.0 (24.0–53.0) | <0.0001 ∗ |
| Transpulmonary gradient (mmHg) | 45.0 (24.0–49.0) | 22.0 (14.0–42.0) | <0.0001 ∗ |
| Cardiac index (L/min/m 2 ) | 3.6 (2.8–4.5) | 3.9 (3.2–4.8) | 0.17 |
| Indexed pulmonary vascular resistance (WU/m 2 ) | 9.7 (7.2–14.5) | 4.4 (2.8–8.2) | <0.0001 ∗ |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
