Discussion:
Using a large multicenter national database, we reported the outcomes of
3,009 infants with truncus arteriosus who underwent surgical repair. The
study provides descriptive but relevant information with regards to
demographics, national trends, in-hospital complications, and outcomes
in infants with truncus arteriosus. The overall mortality in truncus
arteriosus is 10.8%. A recent report from the society of thoracic
surgeons congenital heart surgery database showed a similar mortality
rate of 9.6% in truncus arteriosus repair, which is the second-highest
rate of mortality after Norwood procedure (9). Aiming to identify
predictors of mortality in this disease, a multivariate regression
analysis was used, and we also analyzed the outcomes in subjects with
22q11.2 deletion syndrome. The main findings in this study were as
follows: (1) Patient characteristics that were associated with mortality
were a non-White race, prematurity, and Medicaid insurance. (2) 22q11.2
deletion syndrome was associated with an inverse risk of death despite
having more non-cardiac comorbidities. (3) 22q11.2 deletion syndrome
patients had a longer length of hospital stay and higher cost of
hospitalization.
To address the first finding of insurance disparities, Erickson and
colleagues showed in their study that infants with congenital heart
disease and private insurance are cared for at hospitals with lower
surgical mortality and that non-White children with Medicaid insurance
were less likely to use a hospital with lower mortality rate (10). In
addition, prematurity carries a major risk factor for increased
mortality in infants with congenital heart disease, and preterm birth
rates vary by race and ethnicity. Socioeconomic and ethnic-racial
disparities are a major risk for prematurity (11). Studies have shown
that non-Hispanic Black women are 2.5 times more likely to have preterm
birth versus White women (12). The dynamics and interplay between these
three risk factors (preterm birth, non-White race, and government
insurance) are challenging, and the coexistence of these factors is not
uncommon. The second main finding in this study was the presence of an
inverse risk of death associated with 22q11.2 deletion syndrome. The
baseline characteristics for these patients are similar to subjects
without -22q11.2 deletion syndrome. Many researchers have examined the
operative mortality in truncus arteriosus and some have found that
22q11.2 deletion syndrome does not necessarily increase surgical risk
(13–15), while others reveal that 22q11.2 deletion is an independent
risk factor for late mortality (16, 17). However , the results of our
study are in agreement with the results of a study by Michielon and
colleagues, which found that 22q11.2 deletion syndrome was an inverse
risk for death in patients with truncus arteriosus (18). It could also
be that the complex cardiac pathology and challenging nature of the
repair with its prolonged postoperative course could have much more of
an impact than the impact of the diagnosis of 22q11.2 deletion on
truncus arteriosus. , One could also hypothesize that the cardiac
morphology and anatomy in these patients is different from patients
without 22q11.2 deletion syndrome. A study by McElhinney reviewed
details of cardiac features in truncus patients with 22q11.2 deletion
and showed that patients with 22q11.2 deletion had a higher frequency of
right-sided aortic arch and abnormal aortic arch branching. Most
importantly, interrupted aortic arch, which could impact operative
mortality, occurred at a similar frequency in those with and without
22q11.2 deletion syndrome. In McElhinney’s study, there was a
statistically insignificant trend for a lower incidence of
moderate-severe truncal valve insufficiency in patients with 22q11.2
deletion (11% vs. 33%) which could impact surgical mortality,
suggesting that factors other than chromosome 22q11.2 deletion are the
main determinants of the functional and morphological characteristics of
the truncal valve (19). Previous studies have shown the association with
poor surgical outcomes in subjects with truncal valve regurgitation,
interrupted aortic arch and coronary artery anomalies (20). One of the
limitations of our study is the inability to review patient charts and
cardiac imaging to evaluate for the presence of postoperative truncal
valve regurgitation or the details of coronary anatomy in these
patients. Future studies that also evaluate the cardiac anatomy in
22q11.2 deletion syndrome will likely be able to evaluate our finding of
an inverse death risk that is associated with 22q11.2 syndrome in
truncus arteriosus.
Lastly, this study reported prolonged hospitalization and increased cost
of hospitalization in patients with truncus arteriosus. This is similar
to the study done by Ghimire et al, who showed that the diagnosis of
22q11.1 deletion is associated with longer hospital stay and higher
hospitalization cost (21). The 22q11.2 group had far more non-cardiac
comorbidities and had higher rates of postoperative complications.
Feeding issues were more common in 22q11.2 deletion syndrome, a higher
percentage of these patients had a gastrostomy tube during the index
hospitalization. Airway anomalies were present more frequently in
22q11.2 deletion and this was associated with a higher frequency of
tracheostomy procedure. Immunological dysfunction is not uncommon in
infants with 22q11.2 deletion (22), and it was not surprising to find
higher rates of septicemia and fungal infection in the 22q11.2 deletion
group. Therefore, it was not unexpected to have a prolonged
postoperative hospitalization in patients with 22q11.2 deletion,
modifications in the perioperative care of this subpopulation to address
these complications are warranted to further shorten the postoperative course
and decrease the cost of hospitalization.
Limitations: There are several limitations to this study, the study
relied on the ICD-9 & ICD-10 codes for the diagnosis of truncus
arteriosus and 22q11.2 deletion syndrome, it didn’t describe how the
chromosomal deletion diagnosis was reached. The study span is over a
span of two decades; the field of molecular genetic testing and genomic
profiling has evolved dramatically; this could have affected the
accuracy of the genetic screening method over the years of the study.
The database does not provide details of cardiac anatomy, mainly
coronary artery anomalies, postoperative truncal valve insufficiency, and
aortic arch obstruction. The database lacks information on cardiac
bypass and surgical procedure techniques, which have evolved over the
course of these two decades and therefore is subject to an era effect.
These factors would impact the surgical outcomes in truncus arteriosus.
The current study benefited from the large sample size, some limitations
are inherent to all retrospective studies using administrative
databases. Incorrect or missing data may exist. We rely mainly on ICD
diagnostic and procedure codes to gather information, and we were not
able to refer to specific patient hospital charts to determine the cause
of death. The lack of validation of the data collected by chart review
is a source of potential bias for errors. However, in a large study such
as ours, the patients’ volume likely offset these inaccuracies. Another
limitation of our study is the restriction to the hospitalization
period, and this prevented us from evaluating long-term outcomes.