Abstract
Background
The population-level burden and inequities of aortic dissection (AD) mortality in the United States remain incompletely defined, and contemporary, data-driven forecasts are scarce.
Methods
Using the CDC WONDER Death database (1968–2023), AD deaths were identified by ICD codes (ICD-8/9: 441.0; ICD-10: I71.0). Age-adjusted mortality rates (AAMRs) per 100,000 were stratified by sex, race, age, and Census region. Temporal trends were assessed using Joinpoint regression to estimate annual percent change (APC) and average APC (AAPC). Forecasts through 2033 employed a Bi-GRU model.
Results
We analyzed 175,930 AD-related deaths. Overall, the national AAMR declined by 43 % across 1968–2023 (AAPC −1.10 %, 95 % CI −1.19 to −1.00) but showed a recent upturn (APC +2.00 %, 95 % CI 1.53–2.57). Mortality remained higher in men than women and in Black than White individuals, increased steeply with age, and varied geographically. Bi-GRU forecasts project a modest national decline in AAMR from 1.62 to 1.47 by 2033, with persistent sex (men 1.91 vs women 1.12) and racial (Black 2.32 vs White 1.35) gaps; a slight increase is confined to the South (1.62 → 1.64), while rates in adults ≥85 years improve (10.26 → 9.70).
Conclusions
While U.S. AD mortality has nearly halved over five decades, recent increases and persistent demographic and regional disparities highlight uneven progress. Forecasts to 2033 suggest modest overall declines with persistent sex and racial gaps; targeted hypertension control and regional access to high-volume aortic centers remain priorities.
Keywords: Aortic dissection mortality, Health disparities, Deep learning forecasting
Graphical abstract
1. Introduction
Aortic dissection (AD) is a life-threatening cardiovascular emergency affecting 3–5 per 100,000 annually, with 1–2 % mortality per hour in the first 48 h if untreated. Despite advances, over 15 % of patients still die within 30 days, highlighting its public health burden [1,2]. Surgical treatment began in the late 1960s with DeBakey and Cooley, though early mortality exceeded 30–40 % due to limited diagnostics and bypass technologies [3]. Since then, operative strategies have steadily evolved [4].
From the 1970s to early 1990s, improved imaging (aortography, CT, transesophageal echocardiography) and surgical techniques (e.g., hypothermic arrest, distal perfusion, grafting) reduced perioperative mortality. However, reliable long-term data were scarce until the 1996 launch of the International Registry of Acute Aortic Dissection (IRAD), which reported a 27.4 % hospital mortality rate in 2000 [5].
IRAD and national databases later showed improved outcomes: by 2018, in-hospital mortality for surgically treated type A dissections dropped below 15 % in many high-volume centers. Type B outcomes also improved with thoracic endovascular aortic repair (TEVAR), though population-level impact remains limited [6,7].
From 2013 to 2023, innovations like ECG-gated CT, faster transfers, hybrid ORs, and advanced grafts expanded treatment possibilities. Yet, long-term outcomes remain poor—over 50 % die during follow-up, per a meta-analysis by e Melo et a [8]. Disparities by sex, race, and socioeconomic status persist.
A 2022 population-based study by Nazir et al. in JAHA [9], using CDC WONDER data (1999–2019), found rising age-adjusted mortality from ∼2012, especially among women and non-Hispanic Black populations. However, it lacked broader temporal scope and predictive modeling.
Our study extends this work by: (i) covering 1968–2023 (>175,000 deaths); (ii) identifying inflection points by sex, race, age, and region via joinpoint regression; and (iii) forecasting age-adjusted mortality rates (AAMRs) through 2033 using deep learning, with a focus on equity gaps.
2. Methods
2.1. Study design and data source
We conducted a nationwide, population-based observational cohort study to evaluate long-term mortality trends related to thoracic aortic dissection (AD) in the United States from January 1, 1968, through December 31, 2023. Mortality and population data were obtained from the Centers for Disease Control and Prevention’s Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) Multiple Cause of Death database. Aortic dissection–related deaths were identified using International Classification of Diseases (ICD) codes: ICD-8 code 441.0 (1968–1978), ICD-9-CM code 441.0 (1979–1998), and ICD-10-CM code I71.0 (1999–2023).
Because the CDC WONDER dataset is publicly available and fully de-identified, this study was exempt from Institutional Review Board (IRB) oversight. The study was conducted in accordance with the principles of the Declaration of Helsinki and reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [10].
CDC WONDER mortality data derive from death certificates with a single underlying cause and up to 20 contributing causes; consistent race/Hispanic origin detail is available from 1999 onward. ICD-8/9441.0 and ICD-10 I71.0 do not encode acuity or Stanford type. More granular location-specific codes (e.g., I71.010–I71.019) were introduced in FY2023 and cannot be applied uniformly across 1968–2023 [11].
2.2. Outcome variable
The primary outcome was the annual age-adjusted mortality rate (AAMR) for aortic dissection per 100,000 population, standardized to the 2000 U.S. standard population. Analyses were conducted for the overall population and stratified by sex, race, age group, and U.S. Census region. Crude mortality rates were calculated for age-specific subgroups. Ethnicity (i.e., Hispanic origin) was not included in stratified analyses due to a lack of consistent data prior to 1999.
2.3. Data extraction and variables
All deaths attributed to thoracic aortic dissection between 1968 and 2023 were extracted from the CDC WONDER database. Data were stratified by sex (male, female), race (White, Black), age group (<35, 35–44, 45–54, 55–64, 65–74, 75–84, and ≥85 years), and geography, including U.S. Census regions (Northeast, Midwest, South, West) as well as individual states.
Population denominators were derived from U.S. Census Bureau intercensal and postcensal estimates. Age-adjusted mortality rates were calculated using the direct standardization method to the 2000 U.S. standard population. In compliance with CDC data privacy rules, cells with fewer than 20 deaths were suppressed to preserve statistical reliability and confidentiality.
2.4. Statistical analysis
Crude and age-adjusted mortality rates were calculated per 100,000 population with corresponding 95 % confidence intervals (CIs). Temporal trends in AAMRs from 1968 to 2023 were assessed using the Joinpoint Regression Program (version 4.9.0.0; National Cancer Institute). Log-linear segmented regression models with up to seven joinpoints were selected based on the Bayesian Information Criterion (BIC). Annual percent changes (APCs) and average annual percent changes (AAPCs) were reported with 95 % CIs derived from 5000 Monte Carlo permutations. Statistical significance was defined as a two-sided p-value <0.05.
To forecast mortality trends through 2033, we applied a Bidirectional Gated Recurrent Unit (Bi-GRU) deep learning model to annual AAMR data. The architecture consisted of two Bi-GRU layers (64 and 32 units) followed by dense layers (32 and 1 unit), trained for 50 epochs using the Adam optimizer with mean squared error (MSE) loss. Rolling 15-year lookback windows were used for training, with forecasts evaluated against the last five observed years using MSE, mean absolute error (MAE), root mean square error (RMSE), coefficient of determination (R2), mean absolute percentage error (MAPE), and Pearson’s correlation coefficient (r). Forecasts were generated overall and stratified by sex, race, age group, and region. All analyses and visualizations were conducted in Python using Google Colab.
2.5. Urban–rural proxy (1999–2023)
Deaths are classified into six county-level urbanization strata using the 2013 NCHS Urban–Rural Classification (large central metro, large fringe metro, medium metro, small metro, micropolitan, noncore). However, because these classifications apply only from 1999 onward, we excluded urbanization to maintain geographic consistency across our 1968–2023 study window and avoid overemphasizing a limited two-decade subset in a multi-decade analysis. Additionally, death certificate data lack person-level socioeconomic variables. Socioeconomic status (SES), typically defined by education, income/wealth, and occupation, cannot be directly measured. Although individual education data are available in CDC WONDER from 2018 to 2023, we excluded it due to its limited temporal coverage and lack of comparability within the broader timeframe.
3. Results
3.1. Total aortic dissection-related death
Between 1968 and 1978, a total of 31,060 aortic dissection–related deaths were identified. During this decade, the AAMR declined markedly, from 3.05 (95 % CI, 2.94–3.16) in 1968 to 1.94 (95 % CI, 1.86–2.02) in 1978, representing a 36.2 % reduction. In the subsequent period from 1979 to 1998, 57,927 deaths were identified. The overall AAMR during this era was relatively stable at 1.95 (95 % CI, 1.93–1.96), with a slight decrease from 1.95 in 1979 to 1.91 in 1998 (−2.0 %). Between 1999 and 2020, a further decline was noted; 74,164 aortic dissection–related deaths were identified during this interval, and the AAMR decreased from 1.89 (95 % CI, 1.82–1.95) in 1999 to 1.49 (95 % CI, 1.45–1.54) in 2020, representing a 20.8 % reduction. Most recently, from January 1, 2021, to December 31, 2023, a total of 12,779 aortic dissection-related deaths were recorded, with an overall AAMR of 1.66 (95 % CI, 1.63–1.69). Year-specific AAMRs were 1.65 (95 % CI, 1.60–1.70) in 2021, 1.61 (95 % CI, 1.56–1.66) in 2022, and 1.73 (95 % CI, 1.68–1.78) in 2023.
3.2. Sex-based analysis
Throughout the study period, men consistently showed significantly higher AAMRs than women. Between 1968 and 1978, the AAMR among men was 3.74, more than double the rate in women, which was 1.56. A similar pattern persisted in the following decades: during 1979–1998, the AAMR was 2.73 for men versus 1.34 for women; in 1999–2020, men had an AAMR of 2.06 compared with 1.16 in women. Most recently, between 2021 and 2023, AAMRs remained elevated in men at 2.13, while women had an AAMR of 1.22. Details on confidence intervals are provided in Table 1.
Table 1.
Age-adjusted mortality rates (AAMRs) for aortic dissection by sex, race, and U.S. Census region across study periods.
| Period | Male AAMR (95 % CI) | Female AAMR (95 % CI) | Black AAMR (95 % CI) | White AAMR (95 % CI) | Northeast AAMR (95 % CI) | Midwest AAMR (95 % CI) | South AAMR (95 % CI) | West AAMR (95 % CI) |
|---|---|---|---|---|---|---|---|---|
| 1968–1978 | 3.74 (3.69–3.80) | 1.56 (1.53–1.59) | 2.93 (2.83–3.03) | 2.46 (2.43–2.49) | 2.28 (2.23–2.34) | 2.57 (2.51–2.62) | 2.32 (2.27–2.37) | 3.16 (3.08–3.24) |
| 1979–1998 | 2.73 (2.70–2.76) | 1.34 (1.32–1.36) | 2.35 (2.29–2.41) | 1.90 (1.88–1.91) | 1.85 (1.81–1.88) | 2.00 (1.97–2.03) | 1.80 (1.78–1.83) | 2.26 (2.22–2.30) |
| 1999–2020 | 2.06 (2.04–2.08) | 1.16 (1.15–1.17) | 2.21 (2.17–2.25) | 1.50 (1.48–1.51) | 1.44 (1.42–1.47) | 1.76 (1.74–1.79) | 1.51 (1.49–1.53) | 1.62 (1.60–1.65) |
| 2021–2023 | 2.13 (2.08–2.18) | 1.22 (1.18–1.25) | 2.49 (2.38–2.60) | 1.59 (1.55–1.62) | 1.45 (1.38–1.51) | 1.87 (1.81–1.94) | 1.62 (1.57–1.67) | 1.67 (1.61–1.74) |
3.3. Race-based analysis
Aortic dissection-related mortality varied notably by race. Between 1968 and 1978, Black individuals had a higher AAMR compared to white individuals, with 2.93 versus 2.46, respectively.
This racial variance persisted in subsequent periods. During 1979–1998, the AAMR was 2.35 among Black individuals and 1.90 among White individuals. In 1999–2020, the gap widened with an AAMR of 2.21 for Black individuals compared to 1.50 for White individuals. From 2021 to 2023, this trend persisted, with AAMRs of 2.49 for Black individuals and 1.59 for White individuals. Details on confidence intervals are provided in Table 1.
3.4. Region-based analysis
Region-based variation in aortic dissection–related mortality was evident throughout all study periods. Between 1968 and 1978, the highest AAMR was observed in the Western region (3.16), followed by the Midwest (2.57), the South (2.32), and the Northeast (2.28). Between 1979 and 1998, the West remained to show the highest rate at 2.26, followed by the Midwest at 2.00, the Northeast at 1.85, and the South at 1.80.
From 1999 to 2020, the highest AAMR were recorded in the Midwest (1.76), followed by the West (1.62), South (1.51), and the lowest was in the Northeast (1.44). In the most recent period (2021–2023), this pattern continued, with the Midwest reporting the highest AAMR at 1.87, followed by the West (1.67), the South (1.62), and the Northeast with the lowest rate at 1.45. Details on confidence intervals are provided in Table 1. Heatmaps representing AAMRs for each period across the four regions are presented in Fig. 1.
Fig. 1.
Geographic shifts in age-adjusted mortality rates (AAMR) across U.S. States (1968–2023).
3.5. Age-specific crude mortality rates
Crude mortality rates for aortic dissection rose progressively with age during all study periods. Between 1968 and 1978, the mortality rate was 0.12 in individuals aged 25–34 years and increased steadily to 12.98 among those aged ≥85 years. A similar trend was noted between 1979 and 1998, with rates ranging from 0.15 in the 25–34 age group to 10.35 in those ≥85 years. In the period between 1999 and 2020, mortality rates rose from 0.20 in individuals aged 25–34 to 8.97 in those ≥85 years. From 2021 to 2023, age-specific rates continued to follow this increasing pattern, 0.21 in individuals aged 25–34 years, increasing to 11.04 among individuals aged ≥85 years. Details on confidence intervals are provided in Table 2.
Table 2.
Crude mortality rates for aortic dissection by age group across study periods (1968–2023).
| Age Group | 1968–1978 Crude Rate (95 % CI) | 1979–1998 Crude Rate (95 % CI) | 1999–2020 Crude Rate (95 % CI) | 2021–2023 Crude Rate (95 % CI) |
|---|---|---|---|---|
| 25–34 years | 0.12 (0.10–0.13) | 0.15 (0.14–0.16) | 0.20 (0.20–0.21) | 0.21 (0.18–0.23) |
| 35–44 years | 0.45 (0.42–0.47) | 0.41 (0.39–0.42) | 0.59 (0.58–0.61) | 0.73 (0.69–0.78) |
| 45–54 years | 1.38 (1.33–1.42) | 1.05 (1.03–1.08) | 1.17 (1.15–1.19) | 1.40 (1.34–1.47) |
| 55–64 years | 3.28 (3.21–3.36) | 2.41 (2.36–2.45) | 1.83 (1.80–1.86) | 1.96 (1.88–2.03) |
| 65–74 years | 6.32 (6.19–6.45) | 4.69 (4.62–4.77) | 2.88 (2.83–2.93) | 2.57 (2.47–2.67) |
| 75–84 years | 9.97 (9.74–10.20) | 7.79 (7.66–7.91) | 5.52 (5.44–5.61) | 5.11 (4.92–5.31) |
| 85+ years | 12.98 (12.46–13.50) | 10.35 (10.09–10.61) | 8.97 (8.80–9.14) | 11.04 (10.56–11.52) |
3.6. State-level variation
Substantial geographic variation in AAMRs for aortic dissection was evident throughout the study period. Several western and non-contiguous states, including Hawaii, Alaska, California, and Nevada, frequently ranked among those with the highest mortality rates, along with selected states in the Midwest and South. In contrast, many northeastern and southern states, such as Massachusetts, Kentucky, Mississippi, and Virginia, were repeatedly among those with the lowest rates. While the ranking of individual states shifted over time, the overall pattern of persistent regional disparities remained consistent. The top and bottom 10 states for each period are presented in Table 3 and illustrated in Fig. 2.
Table 3.
Top 10 and bottom 10 age-adjusted mortality rates (AAMRs) for aortic dissection by U.S. State and period, with 95 % confidence intervals.
| Period | Category | State | AAMR | 95 % CI |
|---|---|---|---|---|
| 1968–1978 | Highest | California | 3.55 | 3.44–3.67 |
| 1968–1978 | Highest | District of Columbia | 3.36 | 2.80–3.92 |
| 1968–1978 | Highest | Delaware | 3.32 | 2.65–4.11 |
| 1968–1978 | Highest | Nevada | 3.17 | 2.47–4.00 |
| 1968–1978 | Highest | Hawaii | 3.13 | 2.50–3.75 |
| 1968–1978 | Highest | Colorado | 2.97 | 2.66–3.28 |
| 1968–1978 | Highest | Idaho | 2.86 | 2.34–3.38 |
| 1968–1978 | Highest | Wyoming | 2.84 | 2.11–3.74 |
| 1968–1978 | Highest | Iowa | 2.84 | 2.60–3.08 |
| 1968–1978 | Highest | Ohio | 2.83 | 2.70–2.97 |
| 1968–1978 | Lowest | Kentucky | 1.71 | 1.53–1.90 |
| 1968–1978 | Lowest | Alaska | 1.76 | 0.80–3.33 |
| 1968–1978 | Lowest | Arkansas | 1.87 | 1.64–2.11 |
| 1968–1978 | Lowest | Tennessee | 2.02 | 1.84–2.20 |
| 1968–1978 | Lowest | Texas | 2.03 | 1.91–2.14 |
| 1968–1978 | Lowest | Oklahoma | 2.03 | 1.81–2.25 |
| 1968–1978 | Lowest | Rhode Island | 2.03 | 1.66–2.40 |
| 1968–1978 | Lowest | Louisiana | 2.05 | 1.84–2.25 |
| 1968–1978 | Lowest | Mississippi | 2.07 | 1.81–2.32 |
| 1968–1978 | Lowest | Montana | 2.07 | 1.65–2.56 |
| 1979–1998 | Highest | Hawaii | 3.03 | 2.70–3.35 |
| 1979–1998 | Highest | Alaska | 2.62 | 1.91–3.50 |
| 1979–1998 | Highest | Wyoming | 2.48 | 2.03–2.94 |
| 1979–1998 | Highest | District of Columbia | 2.39 | 2.03–2.74 |
| 1979–1998 | Highest | California | 2.38 | 2.33–2.44 |
| 1979–1998 | Highest | Nevada | 2.31 | 2.03–2.58 |
| 1979–1998 | Highest | Oregon | 2.27 | 2.12–2.43 |
| 1979–1998 | Highest | Delaware | 2.22 | 1.88–2.56 |
| 1979–1998 | Highest | Washington | 2.2 | 2.08–2.33 |
| 1979–1998 | Highest | Vermont | 2.18 | 1.82–2.54 |
| 1979–1998 | Lowest | Mississippi | 1.56 | 1.42–1.70 |
| 1979–1998 | Lowest | Massachusetts | 1.57 | 1.48–1.66 |
| 1979–1998 | Lowest | New Hampshire | 1.57 | 1.35–1.80 |
| 1979–1998 | Lowest | New Mexico | 1.59 | 1.39–1.79 |
| 1979–1998 | Lowest | Arkansas | 1.6 | 1.47–1.74 |
| 1979–1998 | Lowest | Tennessee | 1.61 | 1.51–1.72 |
| 1979–1998 | Lowest | South Dakota | 1.62 | 1.36–1.88 |
| 1979–1998 | Lowest | Kentucky | 1.63 | 1.51–1.75 |
| 1979–1998 | Lowest | Oklahoma | 1.67 | 1.54–1.80 |
| 1979–1998 | Lowest | North Dakota | 1.68 | 1.41–1.96 |
| 1999–2020 | Highest | Hawaii | 2.54 | 2.33–2.75 |
| 1999–2020 | Highest | District of Columbia | 2.46 | 2.13–2.80 |
| 1999–2020 | Highest | Michigan | 2 | 1.93–2.07 |
| 1999–2020 | Highest | Kansas | 1.88 | 1.75–2.01 |
| 1999–2020 | Highest | Delaware | 1.87 | 1.64–2.10 |
| 1999–2020 | Highest | Ohio | 1.83 | 1.77–1.89 |
| 1999–2020 | Highest | South Carolina | 1.81 | 1.71–1.91 |
| 1999–2020 | Highest | Indiana | 1.81 | 1.72–1.89 |
| 1999–2020 | Highest | Washington | 1.76 | 1.68–1.84 |
| 1999–2020 | Highest | Utah | 1.75 | 1.61–1.90 |
| 1999–2020 | Lowest | Massachusetts | 1.14 | 1.08–1.20 |
| 1999–2020 | Lowest | Mississippi | 1.23 | 1.13–1.34 |
| 1999–2020 | Lowest | Kentucky | 1.29 | 1.21–1.38 |
| 1999–2020 | Lowest | Virginia | 1.31 | 1.24–1.37 |
| 1999–2020 | Lowest | New Jersey | 1.33 | 1.26–1.39 |
| 1999–2020 | Lowest | Connecticut | 1.33 | 1.24–1.42 |
| 1999–2020 | Lowest | New Hampshire | 1.33 | 1.18–1.49 |
| 1999–2020 | Lowest | Maine | 1.37 | 1.22–1.51 |
| 1999–2020 | Lowest | Rhode Island | 1.37 | 1.20–1.54 |
| 1999–2020 | Lowest | Idaho | 1.37 | 1.22–1.53 |
| 2021–2023 | Highest | Hawaii | 3.02 | 2.44–3.61 |
| 2021–2023 | Highest | Oklahoma | 2.42 | 2.09–2.76 |
| 2021–2023 | Highest | Colorado | 2.35 | 2.08–2.63 |
| 2021–2023 | Highest | Idaho | 2.16 | 1.73–2.67 |
| 2021–2023 | Highest | Wisconsin | 2.14 | 1.89–2.38 |
| 2021–2023 | Highest | Vermont | 2.11 | 1.46–2.95 |
| 2021–2023 | Highest | South Carolina | 2.07 | 1.80–2.34 |
| 2021–2023 | Highest | Utah | 2.07 | 1.70–2.44 |
| 2021–2023 | Highest | Washington | 2.05 | 1.83–2.27 |
| 2021–2023 | Highest | Minnesota | 2.04 | 1.79–2.29 |
| 2021–2023 | Lowest | West Virginia | 1.06 | 0.77–1.44 |
| 2021–2023 | Lowest | Massachusetts | 1.13 | 0.97–1.29 |
| 2021–2023 | Lowest | New Jersey | 1.3 | 1.15–1.46 |
| 2021–2023 | Lowest | Alabama | 1.31 | 1.10–1.53 |
| 2021–2023 | Lowest | California | 1.31 | 1.24–1.39 |
| 2021–2023 | Lowest | New York | 1.36 | 1.25–1.46 |
| 2021–2023 | Lowest | Connecticut | 1.36 | 1.12–1.60 |
| 2021–2023 | Lowest | Nevada | 1.39 | 1.11–1.67 |
| 2021–2023 | Lowest | New Hampshire | 1.44 | 1.07–1.89 |
| 2021–2023 | Lowest | Mississippi | 1.45 | 1.16–1.78 |
Fig. 2.
Top and bottom 10 U.S. States by age-adjusted mortality rate (AAMR) from aortic dissection, 1968–2023.
3.7. Joinpoint regression of aortic dissection mortality, 1968–2023
Across sex, race, region, and most age groups, mortality from aortic dissection showed an overall pattern of long-term decline, punctuated by intermittent periods of increase and sharper drops in the mid-2000s, followed by more recent upward trends. In both women and men, mortality decreased early in the study period, rose modestly in the 1980s, declined again for over a decade, dropped sharply in the mid-2000s, and then increased in the last decade. Black and White populations followed similar trajectories, with an especially steep mid-2000s decline in Black individuals. All U.S. Census regions experienced early declines, short-term rises or plateaus, mid-period reductions, and renewed increases in recent years.
Age-specific trends were more varied. Middle-aged and older groups generally mirrored the overall population pattern, with early declines, occasional rises, steep mid-2000s drops, and recent increases, resulting in negative long-term AAPCs. In contrast, younger adults showed different profiles: the 25–34 age group had a long-term increase that plateaued in recent years, while the 35–44 group shifted from early decline to mid-period growth and later gains, producing positive AAPCs. These findings highlight persistent mortality reductions over decades but also a reversal toward rising rates in recent years, particularly among younger adults. Detailed APC and AAPC estimates, including p-values and 95 % confidence intervals for each period, are presented in Table 4 and Fig. 3.
Table 4.
Joinpoint regression analysis of mortality trends for aortic dissection by demographic and regional subgroups (age groups: Crude rates), 1968–2023.
| Cohort | Joinpoint Years | APCs (95 % CI; p) | AAPC (95 % CI; p) |
|---|---|---|---|
| Female | 1968, 1979, 1991, 2006, 2011 | −4.07 % (−5.09 to −3.22; p = 0.0024); +1.22 % (0.60–4.52; p = 0.0056); −0.73 % (−1.27 to −0.26; p = 0.0068); −6.28 % (−10.51 to −4.40; p = 0.0044); +2.54 % (1.96–3.24; p = 0.0028) | −0.81 % (−0.91 to −0.68; p < 0.000001) |
| Male | 1968, 1983, 1991, 2006, 2009 | −3.67 % (−4.23 to −3.19; p = 0.0024); +1.21 % (0.01–5.50; p = 0.0488); −1.19 % (−1.76 to −0.75; p = 0.0004); −9.99 % (−11.42 to −5.82; p = 0.0128); +1.44 % (1.03–1.99; p = 0.0059) | −1.37 % (−1.46 to −1.27; p < 0.000001) |
| Overall | 1968, 1981, 1991, 2006, 2010 | −3.92 % (−4.64 to −3.36; p = 0.0036); +1.01 % (0.17–5.49; p = 0.0352); −0.83 % (−1.52 to −0.40; p = 0.0024); −7.71 % (−10.81 to −4.99; p = 0.0188); +2.00 % (1.53–2.57; p = 0.0072) | −1.10 % (−1.19 to −1.00; p < 0.000001) |
| Black | 1968, 1978, 2006, 2009 | −5.75 % (−7.47 to −4.36; p < 0.000001); +0.74 % (0.49–1.11; p < 0.000001); −12.50 % (−14.36 to −6.86; p < 0.000001); +2.54 % (1.99–3.29; p < 0.000001) | −0.79 % (−0.94 to −0.61; p < 0.000001) |
| White | 1968, 1981, 1991, 2005, 2011 | −3.88 % (−4.41 to −3.42; p < 0.000001); +0.95 % (0.31–3.45; p = 0.0064); −0.95 % (−1.46 to −0.54; p = 0.0004); −5.30 % (−8.23 to −4.10; p = 0.0024); +2.01 % (1.53–2.57; p = 0.0012) | −1.16 % (−1.24 to −1.07; p < 0.000001) |
| Northeast | 1968, 1985, 1991, 2006, 2011 | −2.75 % (−3.27 to −2.33; p < 0.000001); +3.27 % (1.34–8.29; p = 0.0020); −1.31 % (−1.84 to −0.75; p = 0.0016); −6.44 % (−10.85 to −4.16; p = 0.0008); +1.98 % (1.26–2.87; p = 0.0040) | −1.04 % (−1.15 to −0.93; p < 0.000001) |
| Midwest | 1968, 1979, 2005, 2010 | −3.91 % (−4.99 to −3.07; p < 0.000001); −0.07 % (−0.30 to 0.25; p = 0.7151); −5.74 % (−10.10 to −3.28; p < 0.000001); +2.12 % (1.52–2.90; p < 0.000001) | −0.87 % (−0.99 to −0.74; p < 0.000001) |
| South | 1968, 1979, 2003, 2012 | −5.10 % (−6.54 to −3.99; p < 0.000001); +0.42 % (0.10–0.85; p = 0.0120); −4.55 % (−7.32 to −3.39; p < 0.000001); +2.73 % (1.84–3.88; p < 0.000001) | −1.08 % (−1.22 to −0.92; p < 0.000001) |
| West | 1968, 1980, 2001, 2012 | −3.90 % (−5.52 to −2.85; p < 0.000001); −0.70 % (−1.13 to 0.44; p = 0.1036); −3.73 % (−7.35 to −2.75; p < 0.000001); +1.87 % (0.92–3.17; p < 0.000001) | −1.51 % (−1.68 to −1.34; p < 0.000001) |
| Age 25–34 | 2003 | +2.04 % (1.61–3.39; p = 0.0040); +0.30 % (−2.20 to 1.08; p = 0.5739) | +1.40 % (1.07–1.82; p < 0.000001) |
| Age 35–44 | 1980, 2006, 2009 | −4.54 % (−6.63 to −2.89; p = 0.0080); +2.56 % (2.02–3.49; p = 0.0316); −4.35 % (−6.74 to 1.98; p = 0.2236); +2.51 % (1.64–4.97; p = 0.0024) | +0.57 % (0.37–0.83; p < 0.000001) |
| Age 45–54 | 1982, 2006, 2011 | −4.10 % (−5.23 to −3.23; p < 0.000001); +1.40 % (1.05–1.96; p < 0.000001); −5.64 % (−10.74 to −2.68; p < 0.000001); +3.33 % (2.50–4.49; p < 0.000001) | −0.27 % (−0.43 to −0.10; p = 0.0020) |
| Age 55–64 | 1981, 1991, 2006, 2009 | −4.48 % (−5.42 to −3.67; p = 0.0092); +1.17 % (−0.85 to 6.64; p = 0.0868); −1.60 % (−2.62 to −0.29; p = 0.0400); −8.30 % (−10.38 to 0.42; p = 0.0560); +1.81 % (1.15–2.70; p = 0.0248) | −1.32 % (−1.45 to −1.18; p < 0.000001) |
| Age 65–74 | 1981, 1991, 2003, 2012 | −4.29 % (−5.34 to −3.49; p = 0.0076); +1.05 % (−0.42 to 6.71; p = 0.1008); −1.82 % (−3.90 to −0.66; p = 0.0144); −6.29 % (−11.93 to −4.42; p = 0.0272); +1.62 % (0.41–3.14; p = 0.0248) | −1.97 % (−2.13 to −1.81; p < 0.000001) |
| Age 75–84 | 1980, 1997, 2006, 2011 | −3.81 % (−4.79 to −3.00; p = 0.0084); +0.22 % (−0.34 to 1.68; p = 0.2699); −2.42 % (−3.82 to −0.29; p = 0.0304); −6.98 % (−11.35 to −3.72; p = 0.0328); +1.08 % (0.34–2.03; p = 0.0232) | −1.59 % (−1.71 to −1.46; p < 0.000001) |
| Age ≥85 | 1977, 2006, 2011 | −3.78 % (−11.39 to −1.23; p = 0.0028); −0.50 % (−0.84 to 4.03; p = 0.2653); −5.23 % (−10.73 to −1.34; p = 0.0308); +3.42 % (2.38–4.90; p = 0.0116) | −0.65 % (−0.91 to −0.34; p = 0.0002) |
Fig. 3.
Joinpoint Regression Analysis of Mortality Trends by Age, Region, Race, and Sex (1968–2023). The age panel shows crude mortality rates, while all other panels display age-adjusted mortality rates (AAMR).
3.8. Forecasting results
Between 2024 and 2033, the population-wide age-adjusted mortality rate (AAMR) is projected to decline from 1.62 to 1.47 per 100,000. By race, AAMRs for Black or African American individuals are expected to decrease from 2.44 to 2.32, while rates for White individuals are projected to fall from 1.50 to 1.35. By sex, male AAMRs are forecasted to decline from 1.97 to 1.91, and female rates from 1.23 to 1.12. Regionally, the Northeast is projected to see a decrease from 1.33 to 1.31, the Midwest from 1.80 to 1.77, and the West from 1.48 to 1.45, while the South is expected to see a slight increase from 1.62 to 1.64. Age-specific crude mortality rates show an increase for those aged 25–34 years (0.21–0.22) and 35–44 years (0.73–0.77), but declines for the 45–54 age group (1.38–1.28), 55–64 (1.90–1.82), 65–74 (2.41–2.39), 75–84 (4.09–2.92), and 85 years and older (10.26–9.70). Detailed forecasts are presented in Table 5 and Fig. 4, Fig. 5.
Table 5.
Forecasted changes in aortic dissection mortality rates by demographic and regional subgroups, 2024–2033.
| Category | Subgroup | 2024 Rate | 2033 Rate | Absolute Change | % Change |
|---|---|---|---|---|---|
| Race (AAMR) | Black or African American | 2.44 | 2.32 | −0.12 | −4.92 % |
| Race (AAMR) | White | 1.50 | 1.35 | −0.15 | −10.00 % |
| Sex (AAMR) | Male | 1.97 | 1.91 | −0.06 | −3.05 % |
| Sex (AAMR) | Female | 1.23 | 1.12 | −0.11 | −8.94 % |
| Region (AAMR) | Northeast (Region 1) | 1.33 | 1.31 | −0.02 | −1.50 % |
| Region (AAMR) | Midwest (Region 2) | 1.80 | 1.77 | −0.03 | −1.67 % |
| Region (AAMR) | South (Region 3) | 1.62 | 1.64 | +0.02 | +1.23 % |
| Region (AAMR) | West (Region 4) | 1.48 | 1.45 | −0.03 | −2.03 % |
| Age (Crude) | 25–34 years | 0.21 | 0.22 | +0.01 | +4.76 % |
| Age (Crude) | 35–44 years | 0.73 | 0.77 | +0.04 | +5.48 % |
| Age (Crude) | 45–54 years | 1.38 | 1.28 | −0.10 | −7.25 % |
| Age (Crude) | 55–64 years | 1.90 | 1.82 | −0.08 | −4.21 % |
| Age (Crude) | 65–74 years | 2.41 | 2.39 | −0.02 | −0.83 % |
| Age (Crude) | 75–84 years | 4.09 | 2.92 | −1.17 | −28.61 % |
| Age (Crude) | ≥85 years | 10.26 | 9.70 | −0.56 | −5.46 % |
Fig. 4.
Forecasted Crude Mortality Rates by Age Group (2024–2033). This figure presents historical trends and future projections of crude mortality rates across seven age cohorts: 25–34, 35–44, 45–54, 55–64, 65–74, 75–84, and 85+ years. Solid blue lines represent observed and forecasted rates, while shaded regions depict 95 % confidence intervals. The vertical dashed line at 2024 marks the beginning of the forecast period. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
Fig. 5.
Demographic and Regional Forecasts of Age-Adjusted Mortality Rates (2024–2033) This figure presents historical trends and future forecasts of age-adjusted mortality rates by race (Black, White), gender (Male, Female), overall population, and U.S. Census regions (Northeast, Midwest, South, West). Each plot shows a solid blue line for observed and forecasted data, with shaded areas indicating 95 % forecast confidence intervals. A vertical dashed line at 2024 marks the start of forecast projections. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
4. Discussion
This study presents the most extensive temporal analysis of aortic dissection (AD) mortality in the U.S. to date, covering 1968–2023 and encompassing over 175,000 deaths. Using national mortality data, we evaluated long-term trends shaped by medical innovation, demographic changes, and healthcare inequities. The findings demonstrate both sustained progress and emerging challenges, highlighting the need for renewed efforts in clinical care and public health strategy.
Our application of deep learning reflects the growing use of machine learning in cardiovascular medicine, where such tools have enhanced diagnostic and prognostic capabilities. For instance, ECG-derived models have improved treadmill test performance in predicting obstructive coronary artery disease, and deep-learning approaches have outperformed clinical scores in predicting short-term mortality in acute pulmonary embolism [12,13]. These precedents contextualize our choice of Bi-GRU for mortality forecasting, given its strength in modeling complex temporal patterns in population-level data.
4.1. Overall mortality trends
Between 1968 and 1978, the age-adjusted mortality rate (AAMR) for AD fell sharply by 36.2 %, from 3.05 to 1.94 per 100,000. This early decline likely reflects the confluence of key advances: standardization of diagnosis through the DeBakey and Stanford classifications, expanded use of cardiopulmonary bypass facilitating safer surgery, and improved perioperative care, including anesthesia, blood pressure control, and postoperative critical care [14]. Increased emergency physician awareness and faster referrals may have further reduced time to intervention [15,16].
From 1979 to 1998, mortality was relatively stable (−2.0 %), suggesting that earlier gains had reached a ceiling. Although surgical techniques and critical care continued to improve, these advances were incremental. Patients with advanced age or comorbidities—previously inoperable—were increasingly treated, often with persistent high risk of complications. A 20.8 % decline from 1999 to 2020 marked a new era in AD management. The introduction and diffusion of thoracic endovascular aortic repair (TEVAR) transformed care for type B dissections and high-risk patients, reducing perioperative mortality. Simultaneously, cross-sectional imaging (CT, MRI) became routine, enabling earlier and more accurate diagnoses, even in atypical or incidental cases [16,17]. These technological gains were reinforced by the development of standardized care pathways and multidisciplinary “aortic teams,” optimizing triage and intervention [18].
However, from 2021 onward, mortality stagnated, with a slight increase by 2023 (1.73 per 100,000). This reversal likely reflects pandemic-related disruptions: delayed emergency care, reduced imaging and follow-up for chronic disease, and worsened cardiovascular risk profiles due to inactivity, obesity, stress, and limited hospital capacity [19].
4.2. Sex-specific mortality
Throughout the study period, men consistently exhibited higher mortality than women—e.g., 3.74 vs. 1.56 per 100,000 in 1968–1978; 2.13 vs. 1.22 in 2021–2023. This disparity is well-documented and likely stems from both biological and clinical factors. Estrogen’s protective effect on the aortic wall, delayed elastin degradation, and smaller baseline aortic size contribute to lower female risk [20,21]. In contrast, men experience earlier medial degeneration, higher wall stress due to larger aortic dimensions, and are more likely to present with extensive dissections. Moreover, men face greater lifetime exposure to modifiable risk factors such as smoking and hypertension. Women, meanwhile, often dissect at smaller aortic diameters—raising concerns about male-derived surgical thresholds—and are more likely to present atypically, contributing to delayed diagnoses and worse postoperative outcomes despite their lower incidence [[20], [21], [22]].
Both sexes followed similar temporal patterns: sharp mortality declines in the 1970s, plateaus in the 1980s–90s, and renewed declines after 2000. Yet, after 2010, mortality began to rebound. Joinpoint analysis showed post-2011 increases in AAMR for both women (+2.54 %) and men (+1.44 %), indicating systemic factors such as delayed care and population aging likely affect all patients [19,20].
4.3. Racial disparities
Black individuals experienced persistently higher mortality than White individuals throughout the study period—e.g., 2.93 vs. 2.46 per 100,000 in 1968–1978; 2.49 vs. 1.59 in 2021–2023. The disparity widened over time, with Black AAMR nearly 50 % higher than White AAMR by 1999–2020 (2.21 vs. 1.50). These findings reflect broader patterns of excess deaths and years of potential life lost among Black Americans [23,24].
Systemic inequities—rather than biological differences—underpin much of this gap. Contributing factors include reduced access to specialized care, delayed diagnosis, suboptimal hypertension control, and broader social determinants of health [24]. The pandemic further magnified these disparities, reversing progress made in earlier decades [19,23].
4.4. Regional and state-level variation
Regional AAMRs shifted markedly over time. The West had the highest rates in 1968–1978 (3.16 per 100,000), possibly reflecting concentrated high-risk populations and rural care access challenges. By 1999–2020, the Midwest led with an AAMR of 1.76, a pattern persisting through 2021–2023.
State-level differences were more pronounced. In 1968–1978, California recorded the highest AAMR (3.55), while Kentucky had the lowest (1.71). By 2021–2023, Hawaii reported the highest mortality (3.02), and West Virginia the lowest (1.06). These geographic disparities likely reflect complex interactions between demographics, hypertension prevalence, healthcare infrastructure, and access to specialized care [[25], [26], [27], [28]]. Hawaii’s persistently elevated mortality suggests possible population-specific risk factors, such as higher prevalence of heritable conditions or barriers to timely care access [29].
4.5. Age-specific mortality
As expected, crude mortality rates rose steeply with age, from <0.2 per 100,000 in individuals aged 25–34 to >11 per 100,000 in those ≥85 by 2021–2023. This reflects cumulative vascular degeneration, prolonged hypertension exposure, and diminished surgical tolerance due to frailty and comorbidities [14,[30], [31], [32]].
However, younger adults (25–44 years) exhibited concerning trends. Joinpoint analysis revealed mortality rates that plateaued or slightly increased, diverging from the declining trends in older cohorts. This group often includes patients with heritable thoracic aortic disease (HTAD) syndromes (e.g., Marfan, Loeys-Dietz, vascular Ehlers–Danlos) or stimulant-related hypertension and trauma [20,30,33]. Many present without prior diagnosis, making dissection their first clinical manifestation [20,31,33,34].
4.6. Joinpoint regression insights
Joinpoint regression identified five distinct phases in national trends, including steep declines (−3.92 % annually from 1968 to 1981), plateaus, and a sharp −7.71 % drop from 2006 to 2010, followed by a concerning +2.00 % increase after 2010. Men experienced slightly greater overall mortality reductions than women (AAPC −1.37 % vs. −0.81 %).
Among racial groups, Black individuals had early steep declines (−5.75 % through 1978), but sustained increases from 1978 to 2006 (+0.74 %), unlike White populations. Regionally, stagnation was most apparent in the Midwest and West, likely reflecting delayed adoption of surgical advances. The 2006–2011 period marked the steepest decline across most groups, aligning with widespread TEVAR dissemination, before post-2011 reversals tied to broader systemic pressures and the COVID-19 pandemic.
4.7. Implications
The deep learning–based Bi-GRU forecasts project a modest national decline in aortic dissection (AD) mortality over the next decade, with the age-adjusted mortality rate (AAMR) expected to decrease from 1.62 to 1.47 per 100,000 by 2033. However, these overall gains obscure persistent—and in some regions, widening—disparities. The model predicts continued sex-based (1.91 for men vs. 1.12 for women) and racial (2.32 for Black vs. 1.35 for White) mortality gaps, suggesting that structural inequities will persist without targeted intervention. Notably, the South is forecasted to be the only region with an increase in AAMR, reinforcing concerns about regional vulnerabilities related to access to care and hypertension management. Although encouraging improvements are projected for the oldest age group (≥85 years), younger adults (25–44 years) show rising or stagnant trends, indicating emerging epidemiologic concerns. These projections underscore the need for equity-focused clinical and public health strategies. In light of these findings, several actionable steps are warranted for clinicians, health systems, and policymakers. Aggressive prevention through early hypertension detection and control, particularly in younger adults and underserved populations, must be prioritized. Efforts should be made to expand access to high-volume aortic centers and establish standardized emergency referral pathways in regions with persistently high mortality. In addition, enhanced screening and genetic counseling for families with heritable thoracic aortic disorders may help mitigate early-onset dissections. Finally, investments in public health infrastructure, improved diagnostic coding, and integrated clinical-mortality data registries are essential to ensure robust surveillance and inform future interventions.
Given persistent excess mortality in men and Black populations and regional stalls, priorities include (i) aggressive hypertension detection/control in high-burden counties; (ii) regionalized networks for rapid transfer to high-volume aortic centers; (iii) telemedicine-enabled follow-up in rural areas; and (iv) linkage of clinical registries with mortality data to target gaps in timeliness and access.
4.8. Strengths
The primary strength of this study lies in its unprecedented scope and temporal breadth, spanning over five decades and encompassing more than 175,000 deaths attributed to aortic dissection across the United States. By leveraging nationally representative mortality data, our findings reflect the true population-level burden rather than the experience of select tertiary centers. The study’s granular stratification by sex, race, age group, census region, and state provides a uniquely detailed picture of mortality dynamics, allowing for the identification of vulnerable subgroups and regions most in need of targeted interventions. Furthermore, the application of joinpoint regression modeling enabled precise identification of periods of accelerated decline, stagnation, or reversal in mortality trends, offering valuable context for linking epidemiologic shifts with historical advances in diagnostics, therapeutics, and healthcare delivery.
4.9. Limitations
Despite its strengths, this study has several important limitations. Although death certificate data are comprehensive, they are prone to misclassification and diagnostic inaccuracies—especially in earlier decades when imaging technologies were less accessible and reporting practices were inconsistent. Transitions between ICD coding systems over the study period may have introduced classification inconsistencies that subtly affect trend estimates. Additionally, the dataset lacks clinical detail: it does not distinguish between type A and type B dissections (Stanford classification), nor does it capture whether cases were managed surgically or medically, or whether it is acute or chronic, or include relevant comorbidities such as connective tissue disorders, the severity of hypertension, or atherosclerotic burden. Data on Hispanic origin are consistently available only from 1999 onward, and subgroup estimates may be biased by small-cell suppression or misclassification. Finally, residual confounding from unmeasured socioeconomic factors and disparities in healthcare access may contribute to some of the observed differences, particularly the persistent mortality gap across racial and geographic groups.
5. Conclusion
Over five decades, U.S. aortic dissection mortality has nearly halved due to advances in imaging, intervention, and care systems. Yet, gains remain uneven, with persistent—and sometimes widening—disparities by sex, race, and region, alongside a recent national uptick. Forecasts to 2033 predict only modest further declines, with some high-burden areas facing stagnation or reversal. Sustained progress requires both continued medical innovation and targeted public health strategies to ensure equitable access to hypertension control, specialized care, and emergency referral systems. Without structural reforms, preventable deaths will persist despite technological advances.
Credit author statement
Abdalhakim Shubietah: Conceptualization, Methodology, Original draft writing, Data acquisition, and Joinpoint analysis. Mohamed S. Elgendy: Methodology, Review and editing, and Figure design. Abubakar Nazir: Original draft writing. Ahmed Ahmed: Literature review, Manuscript review, and Editing. Ameer Awashra: Original draft writing. Mustafa Alkhawam: Original draft writing. Sarah Saife: Critical review, Editing, and Figure design. Hamza A. Abdul-Hafez: Original draft writing. Mohamed Saad Rakab: Critical review, Editing, Supervision and Figure visualization. Mohammed AbuBaha: Original draft writing. Mohammed Tareq Mutar: Oversaw software development, Statistical analysis, and Data interpretation. Ahmed Emara: Literature review, Manuscript review, Supervising the Project, and Editing.
Ethics approval and consent to participate
The study used publicly available, de-identified data and was exempt from institutional review board approval.
Clinical trial number
Not applicable.
Availability of data and materials
The datasets analyzed during the current study are publicly available from the CDC WONDER database http://wonder.cdc.gov.
Funding
This research received no specific grant from any funding agency.
Acknowledgments
None.
Contributor Information
Abdalhakim Shubietah, Email: hakeemraqi@gmail.com.
Mohamed S. Elgendy, Email: dr.elgendy.mo@gmail.com.
Abubakar Nazir, Email: abu07909@gmail.com.
Ahmed Ahmed, Email: ahmedrsahmed1@gmail.com.
Ameer Awashra, Email: ameer.awashra7@gmail.com.
Mustafa Alkhawam, Email: mustafa.alkhawam@gmail.com.
Sarah Saife, Email: sarahsafe508@gmail.com.
Hamza A. Abdul-Hafez, Email: hamzaakrm12@gmail.com.
Mohamed Saad Rakab, Email: mohamedrikab2000@gmail.com.
Mohammed AbuBaha, Email: moabubaha@gmail.com.
Mohammed Tareq Mutar, Email: mohammed.tareq1600c@comed.uobaghdad.edu.iq.
Ahmed Emara, Email: emara9055@gmail.com.
References
- 1.DeMartino R.R., Sen I., Huang Y., et al. Population-based assessment of the incidence of aortic dissection, intramural hematoma, and penetrating ulcer, and its associated mortality from 1995 to 2015. Circ. Cardiovasc. Qual. Outcome. 2018;11(8) doi: 10.1161/CIRCOUTCOMES.118.004689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Harris K.M., Nienaber C.A., Peterson M.D., et al. Early mortality in type A acute aortic dissection: insights from the international registry of acute aortic dissection. JAMA Cardiol. 2022;7(10):1009–1015. doi: 10.1001/jamacardio.2022.2718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chiu P., Miller D.C. Evolution of surgical therapy for stanford acute type A aortic dissection. Ann. Cardiothorac. Surg. 2016;5(4):275–295. doi: 10.21037/acs.2016.05.05. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Altobaishat O., Bataineh O.A., Ibrahim A.A., et al. Single arterial cannulation vs. dual arterial cannulation during acute type A aortic dissection repair: a systematic review and meta-analysis. J. Cardiothorac. Vasc. Anesth. 2025;39(1):244–255. doi: 10.1053/j.jvca.2024.10.022. [DOI] [PubMed] [Google Scholar]
- 5.Hagan P.G., Nienaber C.A., Isselbacher E.M., et al. The international registry of acute aortic dissection (IRAD): new insights into an old disease. JAMA. 2000;283(7):897–903. doi: 10.1001/jama.283.7.897. [DOI] [PubMed] [Google Scholar]
- 6.Evangelista A., Isselbacher E.M., Bossone E., et al. Insights from the international registry of acute aortic dissection: a 20-Year experience of collaborative clinical research. Circulation. 2018;137(17):1846–1860. doi: 10.1161/CIRCULATIONAHA.117.031264. [DOI] [PubMed] [Google Scholar]
- 7.Benedetto U., Sinha S., Dimagli A., et al. Decade-long trends in surgery for acute type A aortic dissection in England: a retrospective cohort study. Lancet Reg. Health Eur. 2021;7 doi: 10.1016/j.lanepe.2021.100131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gouveia E.Melo R., Mourão M., Caldeira D., et al. A systematic review and meta-analysis of the incidence of acute aortic dissections in population-based studies. J. Vasc. Surg. 2022;75(2):709–720. doi: 10.1016/j.jvs.2021.08.080. [DOI] [PubMed] [Google Scholar]
- 9.Nazir S., Ariss R.W., Minhas A.M.K., et al. Demographic and regional trends of mortality in patients with aortic dissection in the United States, 1999 to 2019. J. Am. Heart Assoc. 2022;11(7) doi: 10.1161/JAHA.121.024533. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ghaferi A.A., Schwartz T.A., Pawlik T.M. STROBE reporting guidelines for observational studies. JAMA Surg. 2021;156(6):577–578. doi: 10.1001/jamasurg.2021.0528. [DOI] [PubMed] [Google Scholar]
- 11.Q&A: using 2023 ICD-10-CM codes for aortic dissections and ruptures | ACDIS. https://acdis.org/articles/qa-using-2023-icd-10-cm-codes-aortic-dissections-and-ruptures
- 12.Yilmaz A., Hayıroğlu M.İ., Salturk S., et al. Machine learning approach on high risk treadmill exercise test to predict obstructive coronary artery disease by using P, QRS, and T waves’ features. Curr. Probl. Cardiol. 2023;48(2) doi: 10.1016/j.cpcardiol.2022.101482. [DOI] [PubMed] [Google Scholar]
- 13.Cicek V., Orhan A.L., Saylik F., et al. Predicting short-term mortality in patients with acute pulmonary embolism with deep learning. Circ. J. 2025;89(5):602–611. doi: 10.1253/circj.CJ-24-0630. [DOI] [PubMed] [Google Scholar]
- 14.Yuan X., Mitsis A., Nienaber C.A. Current understanding of aortic dissection. Life. 2022;12(10):1606. doi: 10.3390/life12101606. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.The American heart association emergency cardiovascular care 2030 impact goals and call to action to improve cardiac arrest outcomes: a scientific statement from the American heart association | circulation. https://www.ahajournals.org/doi/10.1161/CIR.0000000000001196 [DOI] [PMC free article] [PubMed]
- 16.Hong M.S., Feezor R.J., Lee W.A., Nelson P.R. The advent of thoracic endovascular aortic repair is associated with broadened treatment eligibility and decreased overall mortality in traumatic thoracic aortic injury. J. Vasc. Surg. 2011;53(1):36–43. doi: 10.1016/j.jvs.2010.08.009. [DOI] [PubMed] [Google Scholar]
- 17.Smith-Bindman R., Miglioretti D.L., Larson E.B. Rising use of diagnostic medical imaging in A large integrated health system. Health Aff. 2008;27(6):1491–1502. doi: 10.1377/hlthaff.27.6.1491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.The aortic team model and collaborative decision pathways for the management of complex aortic disease: clinical practice update from the Canadian cardiovascular society/canadian society of cardiac surgeons/canadian society for vascular surgery/canadian association for interventional radiology - canadian journal of cardiology. https://onlinecjc.ca/article/S0828-282X(23)01574-X/fulltext [DOI] [PubMed]
- 19.Lu J., Shen Y., Liu X., et al. Increased prevalence of cardio-cerebrovascular risk factors during the COVID-19 pandemic lockdown: a large, single center, cross-sectional study. BMC Med. 2025;23(1):414. doi: 10.1186/s12916-025-04193-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Aortic aneurysms in a general population cohort: prevalence and risk factors in men and women | European heart journal - Cardiovascular imaging | Oxford Acad. Accessed August 10, 2025.https://academic.oup.com/ehjcimaging/article/25/9/1235/7658327. [DOI] [PubMed]
- 21.Yang W., Wu S., Qi F., et al. Global trends and stratified analysis of aortic aneurysm mortality: insights from the GBD 2021 study. Front. Cardiovasc. Med. 2025;12 doi: 10.3389/fcvm.2025.1496166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sex-based difference in aortic dissection outcomes: a multicenter study. https://www.mdpi.com/2308-3425/10/4/147 [DOI] [PMC free article] [PubMed]
- 23.Caraballo C., Massey D.S., Ndumele C.D., et al. Excess mortality and years of potential life lost among the black population in the US, 1999-2020. JAMA. 2023;329(19):1662–1670. doi: 10.1001/jama.2023.7022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.African Genetic Ancestry, Structural and social determinants of health, and mortality in black adults | equity, diversity, and inclusion | JAMA Network Open | JAMA Network.Accessed August 10, 2025. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2833889. [DOI] [PMC free article] [PubMed]
- 25.Richman L., Pearson J., Beasley C., Stanifer J. Addressing health inequalities in diverse, rural communities: an unmet need. SSM - Popul. Health. 2019;7 doi: 10.1016/j.ssmph.2019.100398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ahmad F., Sridhar A., Hoover S., Henning-Smith C. Context matters: geographic and age differences explain high heterogeneity in social isolation. Wellbeing Space Soc. 2025;8 doi: 10.1016/j.wss.2025.100257. [DOI] [Google Scholar]
- 27.Buresch J.M., Medgyesi D., Porter J.R., Hirsch Z.M. Understanding how population change is associated with community sociodemographics and economic outcomes across the United States. Front Hum. Dyn. 2024;6 doi: 10.3389/fhumd.2024.1465218. [DOI] [Google Scholar]
- 28.Demographic and regional trends of mortality in patients with aortic dissection in the United States, 1999 to 2019 | J. Am. Heart Assoc. Accessed August 10, 2025.https://www.ahajournals.org/doi/10.1161/JAHA.121.024533. [DOI] [PMC free article] [PubMed]
- 29.Cardiovascular disease mortality among Native Hawaiian and Pacific Islander adults aged 35 years or older, 2018 to 2022 | Ann. Intern. Med.Accessed August 10, 2025. https://www.acpjournals.org/doi/10.7326/M24-0801. [DOI] [PMC free article] [PubMed]
- 30.Acute aortic dissection: pathogenesis, risk factors and diagnosis | Swiss Med. Wkly. Accessed August 10, 2025. https://smw.ch/index.php/smw/article/view/2356.
- 31.Early mortality in type A acute aortic dissection: insights from the international registry of acute aortic dissection | cardiology | JAMA cardiology | JAMA Network. Accessed August 10, 2025. https://jamanetwork.com/journals/jamacardiology/fullarticle/2795672. [DOI] [PMC free article] [PubMed]
- 32.Mortality and burden related with aortic aneurysms and dissections. The importance of information and education - ScienceDirect.Accessed August 10, 2025. https://www.sciencedirect.com/science/article/pii/S0146280624000239?via%3Dihub. [DOI] [PubMed]
- 33.Emara A., Emara M., Gadelmawla A.F., Murad M.R., Aboeldahab H., Elgendy M.S., Hassanin M.S., Aldemerdash M.A., Othman A.M., Khaled M., Shubietah A., Assaassa A., Bapat V.N. Transcatheter versus surgical treatment in aortic stenosis with coronary artery disease: a meta-analysis of time-to-event data on 162,305 patients. Heart Lung. 2025 Oct 8;75:184–197. doi: 10.1016/j.hrtlng.2025.09.018. Epub ahead of print. PMID: 41067128. [DOI] [PubMed] [Google Scholar]
- 34.Emara A., Emara M., Aldemerdash M.A., Hemmeda L., Gadelmawla A.F., Saber A., Ellebedy M., Zordok M., Harris K.M., Bapat V.N., Beckmann E., Brilakis E.S., Megaly M. Impact of aortic root abscess on outcomes in infective endocarditis and predictors of in-hospital mortality: a meta-analysis. Heart Lung. 2025 Oct 30;75:337–346. doi: 10.1016/j.hrtlng.2025.10.009. Epub ahead of print. PMID: 41172899. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets analyzed during the current study are publicly available from the CDC WONDER database http://wonder.cdc.gov.