Abstract
Background
Patients with trisomy 21 are at risk of increased pulmonary vascular resistance and pulmonary arterial hypertension (PAH). The risk is further increased in the setting of large ventricular shunts. While trisomy 21 patients with atrioventricular septal defects (AVSD) containing non-restrictive ventricular defects have elevated pulmonary artery systolic pressure pre-operatively given exposure to systemic ventricular pressure, it is difficult to predict those at risk of post-operative PAH following surgical repair.
Objective
To identify pre-operative clinical and echocardiographic factors associated with clinically significant post-operative PAH in patients with trisomy 21 undergoing AVSD repair.
Methods
This was a retrospective single-center case-control study. We defined discharge on medical therapy for PAH following AVSD repair as the primary endpoint and case definition. Cases were chronologically matched 1:3 with patients not discharged home on PAH therapy (control group). Clinical and echocardiographic data were collected. Groups were compared using standard univariate tests.
Results
13 cases and 39 controls were identified. Obstructive sleep apnea (7/13 [53.8 %] vs. 1/39 [2.6 %], p < 0.001) and aspiration (9/13 [69.2 %] vs. 8/39 [20.5 %], p = 0.002) were more frequent in the PAH group. Pre-operative echocardiographic variables, including measures of left-sided chamber enlargement, right ventricular-pulmonary arterial vascular coupling, and presence of pulmonary artery Doppler systolic notching were not associated with the primary outcome. Post-operatively, the eccentricity index was higher in the PAH group (1.57 [1.46, 2.03] vs. 1.44 [1.27, 1.58], p = 0.02). This group experienced longer length of stay (median 16 [IQR 10, 36] vs. 6 [5, 10.5] days, p = 0.002).
Conclusions
Discharge on medical therapy for PAH after AVSD repair is a clinically important endpoint. Few clinical but no echocardiographic factors, including presence of systolic notching in the pulmonary artery Doppler waveform, were associated with this clinical outcome in the setting of limited statistical power. Future work exploring predictors of post-operative PAH following repair of large ventricular shunts with different populations and endpoints will be beneficial.
Highlights
- •
It is hard to predict post-operative PAH in trisomy 21 patients after AVSD repair.
- •
Aspiration and OSA were associated with discharge on pulmonary vasodilator therapy.
- •
PA Doppler notching did not predict clinically significant post-operative PAH.
1
Introduction
Patients with trisomy 21 are prone to the development of increased pulmonary vascular resistance and pulmonary arterial hypertension (PAH). Abnormal lung development in the setting of reduced pulmonary vascular surface area and endothelial dysfunction with higher levels of endothelin-1 and lower levels of nitric oxide predispose trisomy 21 patients to persistent pulmonary hypertension of the newborn and early progression of pulmonary arteriopathy [ ]. Additionally, in this patient population, significant changes in pulmonary vascular resistance can occur within the first 6 months of life in the presence of congenital heart disease [ ]. Patients with large ventricular level shunts, as are often present in atrioventricular septal defects (AVSDs), may have additional risks for development of increased pulmonary vascular resistance and post-operative PAH from exposure of the pulmonary arterial vascular bed to chronic pressure load prior to surgical repair. While all patients with large ventricular shunts will have elevated pulmonary artery systolic pressure prior to surgical septation as the pulmonary arteries are exposed to systemic pressure, it is difficult to predict which patients are at risk for clinically significant PAH post-operatively following closure of the ventricular shunt. Identification of patients at risk for post-operative PAH could be useful to help inform surgical decisions such as whether to leave an atrial shunt to mitigate risk of post-operative pulmonary hypertensive crisis and likelihood of need for inhaled nitric oxide in the acute post-operative period [ ].
Certain echocardiographic measures, such as presence of mid-systolic notching in the right ventricular outflow tract (RVOT) Doppler waveform, have been shown to be an indicator of PAH in septated biventricular hearts [ ]; however, there is little knowledge about utility as a predictor for post-operative PAH in patients with large ventricular shunts. In this study, we sought to identify clinical and echocardiographic factors associated with post-operative PAH requiring medical therapy in patients with trisomy 21 undergoing surgical repairs of AVSDs. We hypothesized that presence of systolic notching in the RVOT or pulmonary artery Doppler waveform would be associated with our primary outcome.
2
Methods
2.1
Study design
We performed a retrospective single-center case-control study on patients with trisomy 21 who underwent AVSD repair between 1/2016 and 11/2022. We chose a priori discharge on medical therapy for PAH following surgical AVSD repair as our primary endpoint indicative of clinically significant post-operative PAH. At our institution, our general practice is to start a pulmonary vasodilator agent when: 1) there are echocardiogram findings of greater than moderately elevated right ventricular pressure along with right ventricular dysfunction and failure to progress clinically or 2) if right ventricular pressures are estimated to be systemic or above systemic beyond the acute post-operative phase with or without right ventricular dysfunction.
We included the patients who underwent AVSD repairs between 0 and 12 months of age with a biventricular circulation. For purposes of this study, we used the term PAH to specify pulmonary hypertension attributable to increased pulmonary vascular resistance and not the hyperkinetic pulmonary hypertension, which is present in all patients with shunts exposing the pulmonary arteries to systemic pressure. Our exclusion criteria included presence of other significant congenital heart disease other than patent ductus arteriosus (PDA), patients with pressure-restrictive or closed ventricular defects, patients who were on pulmonary vasodilator therapy pre-operatively, and unbalanced AVSDs that required single ventricle palliation.
Cases were defined as patients who were discharged home on pulmonary vasodilator therapy for PAH (PAH group) following their index operation. Cases were matched 1:3 with patients not discharged home on PAH therapy (control group) based on surgical dates. Specifically, controls were chronologically selected as the ones with trisomy 21 and AVSD repair immediately preceding the cases. The study was approved by the Institutional Review Board at Children’s Healthcare of Atlanta (IRB number: STUDY00001595).
2.2
Biometric, clinical, laboratory and baseline echocardiography data
Biometric and clinical data were extracted from the medical record. Aspiration was identified clinically by a speech language pathologist and supported by additional testing as appropriate. Similarly, obstructive sleep apnea (OSA), laryngomalacia, bronchopulmonary dysplasia, hypothyroidism, Hirschsprung’s disease and duodenal atresia were diagnosed pre-operatively in our study by expert assessment and supported by additional testing as appropriate. We used presence of an acid blocker agent, such as a histamine H2-receptor antagonist or a proton pump inhibitor, as a proxy for gastroesophageal reflux disease (GERD). We also evaluated records for other significant issues related to the endocrine, pulmonology, and neurology systems that were identified prior to surgery.
Similarly, laboratory data was extracted from the medical record. All patients presenting from the ambulatory setting typically had a pre-operative visit where pre-operative labs were obtained, from which we extracted blood cell counts and measures of hepatic and renal function. For those that were inpatient before their operation, the pre-operative labs were obtained the day before or the morning of the operation.
Echocardiographic data was extracted from the patients’ pre-operative transthoracic echocardiogram, which at our center is a complete study typically performed 1–3 days prior to the patients’ operations. Left ventricle (LV) and right ventricle (RV) systolic function and the degree of regurgitation of the atrioventricular valves were derived from the echocardiogram reports. Presence of notching of the pulmonary artery Doppler waveform, the acceleration time (AT) as well as acceleration time/ejection time (AT/ET) ratio, tricuspid annular plane systolic excursion (TAPSE), the left atrium to aorta (LA/AO) ratio and the end diastolic dimension (EDD) of the left ventricle were retrospectively calculated by a single user who was blinded to group assignment. Notching of the pulmonary artery Doppler waveform was defined as presence of a notch in the Doppler waveform derived from interrogation of the RVOT or at the level of the pulmonary valve or main pulmonary artery if RVOT Doppler was not available [ ] ( Fig. 1 ). The AT and AT/ET ratio was determined by measuring the acceleration times and ejection times from an RVOT or pulmonary Doppler waveform. TAPSE was calculated from 2-dimensional 4-chamber images according to previously described methods [ ]. Right ventricular-pulmonary arterial coupling was assessed using the ratio of TAPSE to AT and TAPSE to the AT/ET ratio to adjust for ejection time [ ]. The LA/AO ratio and EDD measurements were performed according to published standards [ ].
2.3
Operative data, outcomes and post-operative echocardiography data
The duration of cardiopulmonary bypass cross-clamp and whether the patients needed a second bypass run or extracorporeal membrane oxygenation (ECMO) were extracted from the medical record. Hospital length of stay, intubation duration, use of nitric oxide, and 30-day mortality were collected from the medical record. At our center, at least one complete post-operative transthoracic echocardiogram is obtained before discharge. Typically, this echo cardiogram examines all cardiac structures and is obtained close to date of discharge. Echocardiographic data were extracted from the complete post-operative study when multiple post-operative studies were available. The systolic eccentricity index and TAPSE from the post-operative echocardiogram were calculated by a single user blinded to group assignment. We evaluated long-term use of pulmonary vasodilator therapy in the case group.
2.4
Statistical analysis
Continuous data are presented as medians (interquartile range), and categorical data are presented as N (%). Continuous data were compared using the Mann-Whitney U test. Categorical data were compared using Fisher’s exact test. Discharge on pulmonary vasodilator therapy was chosen a priori to distinguish cases and controls. Following primary analysis, a post-hoc analysis was completed by dividing groups according to presence of absence of notching of the RVOT or pulmonary artery Doppler waveform on pre-operative echocardiograms to evaluate whether this finding was associated with pulmonary vasodilator therapy at discharge, a lower TAPSE score, or a higher systolic eccentricity index on the post-operative echocardiogram. A p-value of <0.05 was considered significant. To provide a more conservative estimate of significance, Bonferroni correction was used for multiple comparisons. Given small sample size, a multivariable analysis could not be completed.
3
Results
3.1
Biometric, clinical, laboratory and baseline echocardiography data
We identified 13 cases discharged on pulmonary vasodilator therapy who were matched chronologically with 39 controls who were not discharged on medical therapy for PAH following surgical AVSD repair. Baseline biometric data and frequency of PDAs were similar between groups ( Table 1 ). The PAH group had a borderline significantly higher weight/length percentile (40%ile vs. 12%ile p = 0.053), although other measurements of size were similar between groups.
| Group | PAH group (N = 13) | Control group (N = 39) | p | |
|---|---|---|---|---|
| Age (days) | 142 (128, 173) | 147 (129, 183) | 142 (126.5, 166.5) | 0.54 |
| Male sex | 24/39 (61.5) | 8/13 (61.5) | 16/39 (41.0) | 0.34 |
| GA (weeks) | 38 (36, 38) | 37 (36, 38) | 38 (36, 38) | 0.60 |
| Weight (kg) | 5.2 (4.6, 5.8) | 5.4 (4.6, 5.8) | 5.1 (4.5, 5.7) | 0.47 |
| Weight percentile (%) | 17 (4, 47) | 17 (4, 46) | 16 (4, 46) | 0.92 |
| Length (cm) | 60 (56, 62) | 58 (54, 63) | 60 (57, 62) | 0.39 |
| Length percentile (%) | 36.5 (5, 63) | 25 (1, 48) | 38 (7, 76) | 0.13 |
| Weight/length percentile (%) | 23 (2, 77) | 40 (14, 92) | 12 (1, 60) | 0.053 |
| BSA (m 2 ) | 0.28 (0.25, 0.30) | 0.28 (0.25, 0.31) | 0.28 (0.26, 0.29) | 0.87 |
Among pre-operative clinical factors, oropharyngeal aspiration (9/13 [69.2 %] vs. 8/39 [20.5 %], p = 0.002) and obstructive sleep apnea (7/13 [53.8 %] vs. 1/39 [2.6 %], p < 0.001) were more frequently observed in the PAH group ( Table 2 ). There were trends toward higher pre-operative oxygen use in the PAH group (6/13 [46.2 %] vs. 6/39 [15.4 %], p = 0.051) and more frequent use of acid suppressing agents before surgery (11/13 [85 %] vs. 21/39 [54 %], p = 0.057). Otherwise, there were no significant differences in conditions related to the endocrine, pulmonology, neurology systems, and other forms of congenital malformations between the two groups. The laboratory data was also similar between the two groups ( Table 2 ).
| PAH group (N = 13) | Control group (N = 39) | p | |
|---|---|---|---|
| Vital signs | |||
| HR | 145 (130, 150) | 136 (131, 148) | 0.62 |
| RR | 38 (32, 52) | 40 (34, 48) | 0.47 |
| SpO2 (%) | 95 (92, 96) | 96 (92, 100) | 0.45 |
| >1 diuretic agent | 6/13 (46.1) | 11/39 (28.2) | 0.31 |
| On oxygen pre-op | 6/13 (46.2) | 6/39 (15.4) | 0.051 |
| Tube feeds pre-op | 10/13 (77) | 24/39 (62) | 0.50 |
| Acid blockers pre-op | 11/13 (85) | 21/39 (54) | 0.057 |
| Hirschsprung’s disease | 1/13 (8) | 0/39 (0) | 0.25 |
| Duodenal atresia | 1/13 (8) | 1/39 (3) | 0.44 |
| Hypothyroidism | 3/13 (23.1) | 8/39 (20.5) | 1.0 |
| Laryngomalacia | 3/13 (23.1) | 2/39 (5.1) | 0.09 |
| Aspiration | 9/13 (69.2) | 8/39 (20.5) | 0.002 ⁎ |
| OSA | 7/13 (53.8) | 1/39 (2.6) | <0.001 ⁎ |
| BPD | 2/13 (15.4) | 1/39 (2.6) | 0.15 |
| CDH | 0/13 (0.0) | 0/39 (0.0) | |
| TEF | 0/13 (0.0) | 0/39 (0.0) | |
| Meconium aspiration | 1/13 (7.7) | 1/39 (2.6) | |
| Creatinine (mg/dL) | 0.30 (0.20, 0.40) | 0.30 (0.26, 0.35) | 0.99 |
| Creatinine normal a | 11/13 (85) | 35/39 (90) | 0.63 |
| AST (N = 49) (U/L) | 40 (32, 46) | 38 (31, 55.5) | 0.79 |
| AST normal a | 13/13 (100) | 35/36 (97) | 0.63 |
| ALT (N = 49) (U/L) | 31 (25, 39) | 35.5 (21.5, 48) | 0.81 |
| ALT normal a | 11/13 (85) | 31/36 (86) | 1.00 |
| Hematocrit (%) | 41 (39, 46) | 42 (39, 44) | 0.74 |
| Polycythemia? | 6/13 (46) | 15/39 (38) | 0.75 |
| Platelets (thou/μL) | 315 (280, 350) | 324 (267, 369) | 0.88 |
| Neutrophils (%) | 37 (34, 41) | 34 (28, 49.5) | 0.85 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
