Thoracic Aortic Dissection in Tetralogy of Fallot: A Review of the National Inpatient Sample Database.

Background Thoracic aortic aneurysm is common in patients with tetralogy of Fallot ( TOF ); the incidence of thoracic aortic dissection ( TAD ) is unknown, but generally considered to be uncommon. The purpose of this study was to determine incidence and risk factors for TAD in TOF patients. Methods and Results This work is a retrospective review of the National Inpatient Sample ( NIS ) database for cases of ascending TAD among all hospital admissions in adults with TOF , 2000-2014. Of 18 353 admissions in TOF patients, 11 (0.06%; 6 per 10 000 admissions) of these were TAD -related admissions. For the TAD -related admissions, mean age was 49.8±7.2 years; aortic surgical interventions were performed during 8 of the admissions, and overall in-hospital mortality was 45% (5 of 11). Risk factors associated with TAD -related admission were age >60 years (odds ratio, 2.41; 95% CI, 1.23-3.25; P=0.013), male sex ( odds ratio, 6.91; 95% CI, 4.85-8.54; P<0.001), and hypertension (odds ratio, 1.74; 95% CI, 1.06-3.19; P=0.037). Conclusions This is the first population-based study of TAD outcomes in patients with TOF , and it showed a low risk of TAD in this population. This has important clinical implication with regard to surveillance and surgical intervention. Given that prevalence of thoracic aortic aneurysm is higher in TOF patients compared with the general population; therefore, making clinical decisions based on outcomes data and practice guidelines derived from patients with degenerative and bicuspid aortic valve-related aortopathies may lead to overtreatment. Further studies are required to better understand the pathobiology of this disease in order to make evidence-based recommendations for surveillance and treatment.

Clinical Perspective

What Is New?

Thoracic aortic dissection is uncommon in adults with tetralogy of Fallot.

Risk factors for thoracic aortic dissection were male sex, older age, and hypertension.

What Are the Clinical Implications?

Tetralogy of Fallot patients with thoracic aortic aneurysm may require less‐frequent imaging follow‐up.

Tetralogy of Fallot patients with thoracic aortic aneurysm who do not have risk factors for thoracic aortic dissection (male sex, older age, and hypertension) may benefit from conservative (nonsurgical) management considering the low event rate in this population.

Introduction

Thoracic aortic aneurysm is a common complication in patients with tetralogy of Fallot (TOF), and previous studies have proposed potential risk factors for aortic aneurysm in this population.1, 2, 3, 4 Some of the risk factors include male sex, right aortic arch, history of palliative shunt, and underlying diagnosis of TOF with pulmonary atresia.1, 2, 3, 4 Thoracic aortic dissection (TAD) is a devastating complication that can occur in patients with aneurysms, and the goal of imaging surveillance is to identify patients who are considered high risk for TAD and identify possible indications for surgical intervention.5, 6, 7

Although several studies have shown a high prevalence of thoracic aortic aneurysm in the TOF population, none of these studies have reported any cases of TAD and, as a result, TAD is generally considered uncommon after TOF repair.8, 9, 10 There are no population‐based studies of TAD incidence and outcomes, and the available data about TAD in TOF patients are derived from case reports.11, 12, 13 In the absence of epidemiological data about the risk of TAD in this population, decisions regarding surgical interventions are based on extrapolations from natural history studies conducted in patients with syndromic, degenerative, and bicuspid aortic‐valve–related aortopathies.5, 6, 7 Considering the importance and anxiety often associated with the potential risk of TAD in TOF patients with thoracic aortic aneurysm, we embarked on a population‐based study to determine incidence of TAD in this population.

Methods

We will make data, analytical methods, and study materials available to other researchers upon request. The National Inpatient Sample (NIS) is the largest all‐payer database of hospital inpatient stays in the United States. The NIS contains discharge data from a 20% stratified sample of community hospitals and is a part of the Healthcare Cost and Utilization Project (HCUP), sponsored by the Agency for Healthcare Research and Quality.14 Information regarding each discharge includes patient demographics, primary payer, hospital characteristics, principal diagnosis, up to 24 secondary diagnoses, and procedural diagnoses. The Mayo Clinic (Rochester, MN) Institutional Review Board approved this study, and informed consent was waived because the study is based on a review of a de‐identified database.

Using the HCUP‐NIS data from 2000 to 2014, adult patients (>18 years) admitted with a primary or secondary diagnosis of TOF (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD‐9‐CM] code 745.2) were identified. Cases of TAD involving the ascending aorta were identified using ICD‐9‐CM 441.01 and 441.1. Patient characteristics (age, sex, race, socioeconomic status, and primary payer) and hospital characteristics (teaching status and location, bed size, and region) associated with each discharge were identified from the HCUP‐NIS database. Deyo's modification of the Charlson Comorbidity Index was used to identify the burden of comorbid diseases (Table S1).15 The primary outcome was to describe the incidence and risk factors for TAD in TOF patients.

As recommended by the HCUP‐NIS, survey procedures using discharge weights provided with the HCUP‐NIS database were used to generate national estimates. Categorical data are expressed as count (%), and continuous data are expressed as mean±SD or median and interquartile range for skewed data. Chi‐square and t tests were used to compare categorical and continuous variables, respectively. Poisson regression was used to analyze trends of TAD‐related admissions over the study duration. Risk factors for TAD were assessed using logistic regression and are expressed as odds ratio (OR) and 95% CI. Two‐tailed P<0.05 was considered statistically significant. All statistical analyses were performed using SPSS software (version 25.0; IBM Corp, Armonk NY).

Results

During the period between January 1, 2000 and December 31, 2014, there were an estimated 18 353 admissions in adults with TOF diagnosis, of which TAD of the ascending aorta (type A dissection) was diagnosed in 11 admissions (0.06%; 6 per 10 000 admissions). Mean age at the time of admission was higher in the TAD‐related admissions compared with other admissions (49.8±7.2 versus 38.4±14.0 years; P<0.001). Tabl shows a comparison of the baseline characteristics of TAD‐related admissions and other admissions. Incidence of TAD‐related admissions was significantly higher in males compared with females (11.8 versus 1.0 per 10 000 admissions; P<0.001) and higher in patients aged >60 years at time of admission (8.6 versus 5.9 versus 1.1 per 10 000 admissions; P<0.001; Figure 1.

Figure 1

Bar graph comparing incidence of TAD‐related admissions, per 10 000 admissions by sex (A) and age group (B). TAD indicates thoracic aortic dissection.

Of the 11 TAD‐related admissions, 8 (73%) underwent aortic surgical intervention during the admission, and in‐hospital mortality occurred in 5 of the 11 (45%) admissions. Risk factors associated with TAD‐related admission were age >60 years (odds ratio, 2.41; 95% CI, 1.23–3.25; P=0.013), male sex (odds ratio, 6.91; 95% CI, 4.85–8.54; P<0.001) and hypertension (odds ratio, 1.74; 95% CI, 1.06–3.19; P=0.037). With regard to hospital characteristics and resource utilization, TAD‐related admissions were more likely to occur in large‐bed‐size hospitals (45%) compared with medium‐bed‐size hospitals (36%) and small‐bed‐size hospitals (18%; P=0.001), and in urban nonteaching hospitals (55%) compared with urban teaching hospitals (46%) and rural hospitals (0%; P=0.004). Mean length of hospital stay was 6.9±6.3 days (4 days; interquartile range, 1–21).

Discussion

Based on the review of a nationally representative database of hospital admissions in the United States, we identified 11 TAD‐related admissions of 18 353 admissions in adults with TOF. All TADs were type A dissections. Incidence of TAD‐related admissions was 6 per 10 000 admissions. Thoracic aortic aneurysm is common in adults with previous TOF repair, and prevalence is reported to range from 28% to 69%.1, 2, 3, 4, 8, 9, 10 The wide variation in prevalence is because of differences in definitions of aneurysm used in these studies. Incidence of TAD after TOF repair is unknown, but it is generally considered to be uncommon. In a recent study of 453 adults with TOF followed at the Mayo Clinic, thoracic aorta dilation was present in 69%, and severe aortic aneurysm defined as aortic dimension >50 mm was present in 9% of patients.16 Total duration of follow‐up in that study was 3700 patient‐years, and there was no case of TAD observed during follow‐up. Several other outcomes studies have reported similar findings of high prevalence of thoracic aortic aneurysm and zero incidence of TAD in the TOF population.1, 2, 3, 4, 8, 9, 10

To the best of our knowledge, there are only 4 cases of TAD in TOF patients reported in the literature. These TADs occurred in a 60‐year‐old man with aortic root dimension of 55 mm; a 36‐year‐old man with mid ascending aorta dimension of 93 mm; an 18‐year‐old man with 22q11 deletion and aorta root dimension of 70 mm; and a 30‐year‐old man with mid ascending aorta dimension of 71 mm.11, 12, 13, 17 All 4 cases were Stanford type A dissection. It is noteworthy who all these reported cases of TAD occurred in males, similar to the sex distribution observed in the current study. In addition, 3 of the 4 reported cases had aortic dimensions that met elective replacement criteria, irrespective of the cause of aortic aneurysm.

In a cross‐sectional study of 1181 patients with bicuspid aortic valve, prevalence of TAD was 5%, and ranged from 4% in patients with aortic dimension of 50 mm to 13% in patients with aortic dimension of 70 mm.18 Other studies of degenerative and bicuspid aortic valve–related aortopathies have reported TAD prevalence of 2.5% to 7%.18, 19, 20, 21, 22 In contrast to these studies, the estimated TAD incidence of 6 per 10 000 (0.06%) in the current study suggests an extremely low risk of TAD in TOF patents. Consistent with our speculation of low risk of TAD, none of the cohort studies of outcomes of aortic aneurysm in TOF patients have reported any TADs.1, 2, 3, 4, 8, 9, 10 The only known cases of TAD in TOF patients are from case reports, and this does not provide an accurate estimate of risk (incidence) of TAD.

Another important finding from the current study was the striking sex difference in the incidence of TAD. Ten of the 11 cases of TAD occurred in males, yielding an incidence of 11.8 cases per 10 000 admissions compared with 1.1 per 10 000 in females. In addition to male sex, the other risk factors for TAD in this study were older age and hypertension. These risk factors are similar to the traditional risk factors for TAD in the general population, suggesting that TOF diagnosis per se does not confer any additional risk beyond these traditional risk factors for TAD.5, 20, 23

Clinical Implications and Future Directions

The current study suggests a low risk of TAD in TOF patients, and this is consistent with data from previous cohort studies in TOF patients. Although the reason for the rarity of TAD is unknown, we speculated that this may be related to the presence of adhesions from previous sternotomies or may reflect a different underlying disease pathogenesis in TOF patients compared with other types of aortopathies. Our results have important clinical implications with regard to imaging surveillance and timing of surgical intervention. Prevalence of thoracic aortic aneurysm is higher in TOF patients compared with the general population; therefore, making clinical decisions based on outcomes data and practice guidelines derived from patients with degenerative and bicuspid aortic valve–related aortopathies may lead to overtreatment. Although surgical aorta replacement can be performed with very low surgical mortality in the high‐volume centers,24 the risks and benefits of this procedure should be carefully considered in patients with multiple previous sternotomies and high probability of future sternotomies for residual/recurrent hemodynamic lesions related to TOF. There is a need for in‐depth genetic and phenotypic characterization of TOF patients with aortic aneurysm and TAD to better understand the pathobiology of this disease, and this can be achieved by creating a multicenter registry of aortic aneurysms and TAD in TOF patients. Appropriate identification and treatment of high‐risk patients is critical because of the high mortality (46% in‐hospital mortality) associated with TAD‐related admissions.

Limitations

The NIS is an administrative database and relies on accuracy of diagnosis codes. Second, we did not have data about thoracic aorta dimensions at time of admission, the incidence of bicuspid aortic valves and genetic syndromes in these patients, and other potential risk factors, such as rate of progression of aortic aneurysm and family history of TAD, in these patients. Finally, incidence of TAD may be underestimated in this study because the patients who had TAD and died before arrival to the hospital would have been excluded from this study based on the sampling technique of the NIS. All these factors could have potentially influenced the observed findings in this study.

Conclusions

In the current study, we reviewed data from 18 353 admissions in TOF patients and reported a low incidence of TAD of 6 per 10 000 admissions (0.06%). Risk factors for TAD were male sex, older age, and hypertension. The low event rate in this study suggests that using the practice guidelines for degenerative and bicuspid aortic valve–related aortopathies to decide on the timing of intervention may lead to overtreatment in this population because of the high prevalence of thoracic aortic aneurysm in TOF patients. Further studies are required to better understand the pathobiology of this disease in order to make evidence‐based recommendations for surveillance and treatment.

Sources of Funding

Dr Egbe is supported by National Heart, Lung, and Blood Institute (NHLBI) grant K23 HL141448‐01.

Disclosures

None.

Table S1. Administrative Coding Used for Identification of Comorbidities and Procedures

Click here for additional data file.

Table 1

Baseline Characteristics of Admissions With and Without TAD

Characteristic		TAD‐Related Admission (N=11)	Others (N=18 342)	P Value
Age, y		49.8±7.2	38.4±14.0	<0.001
Female sex		1 (9.1)	9872 (53.8)	0.003
Race	White	10 (90.9)	10 225 (55.8)	0.001
	Black	1 (9.1)	1891 (10.3)
	Hispanic	···	1658 (9)
	Asian	···	408 (2.2)
	Native American	···	67 (0.4)
	Others	···	510 (2.8)
	Missing	···	3583 (19.5)
Primary payer	Medicare	3 (27.3)	5136 (28)	0.011
	Medicaid	4 (36.4)	4459 (24.3)
	Private	4 (36.4)	7119 (38.8)
	Uninsured	···	1004 (5.5)
	No charge	···	103 (0.6)
	Others	···	521 (2.8)
Quartile of median household income for ZIP code	0 to 25th	4 (36.4)	4151 (22.6)	0.006
	26th to 50th	3 (27.3)	4591 (25)
	51st to 75th	4 (36.4)	4588 (25)
	75th to 100th	···	5012 (27.3)
Hospital teaching status and location	Rural	···	1485 (8.1)	0.004
	Urban non‐teaching	6 (54.6)	4353 (23.7)
	Urban teaching	5 (45.5)	12 504 (68.2)
Hospital bed size	Small	2 (18.2)	2203 (12)	0.001
	Medium	4 (36.4)	3618 (19.7)
	Large	5 (45.5)	12 521 (68.3)
Hospital region	Northeast	3 (27.3)	3801 (20.7)	0.004
	Midwest	4 (36.4)	3794 (20.7)
	South	2 (18.2)	6731 (36.7)
	West	2 (18.2)	4016 (21.9)
Charlson Comorbidity Index		1.4±1.3	1.3±1.1	0.101
Comorbidities	Hypertension	6 (54.6)	3266 (17.8)	<0.001
	Hyperlipidemia	4 (36.4)	1258 (6.9	<0.001
	Chronic kidney disease	2 (18.2)	974 (5.3)	0.03
	Atrial fibrillation	4 (36.4)	2802 (15.4)	0.001

Represented as percentage or mean±SD. TAD indicates thoracic aortic dissection.