https://orcid.org/0000-0002-3307-5944
Abstract
Describe and compare incidences across age groups of rehospitalization, repeated aortic surgery, and death in patients who survived surgery and hospitalization for type A aortic dissection.
From Danish nationwide registries, we identified patients hospitalized with Stanford type A aortic dissections (2006–2018). Survivors of hospitalization and surgery on the ascending aorta and/or aortic arch comprised the study population (n = 606, 36 (38.9%) <60 years old (group I), 194 (32.0%) 60–69 years old (group II), and 176 (29.1%) >69 years old (group III)). During the first year, 62.5% were re-hospitalized and 1.4% underwent repeated aortic surgery with no significant differences across age groups (P = 0.68 and P = 0.39, respectively). Further, 5.9% died (group I: 3.0%, group II: 8.3%, group III: 7.4%, P = 0.04). After 10 years, 8.0% had undergone repeated aortic surgery (group I: 11.5%, group II: 8.5%, group III: 1.6%, P = 0.04) and 10.2% (group I), 17.0% (group II), and 22.2% (group III) had died (P = 0.01). Using multivariable Cox regression analysis, we described long-term outcomes comparing age groups. No age differences were found in one-year outcomes, while age > 69 years compared with age < 60 years was associated with a lower rate of repeated aortic surgery [hazard ratio 0.17, 95% confidence interval (CI) 0.04–0.78] and a higher rate of all-cause mortality (hazard ratio 2.44, 95% CI 1.37–4.34) in the 10-year analyses.
Rehospitalisations in the first year after discharge were common in all age groups, but survival was high. Repeated aortic surgery was significantly more common among younger than older patients.
Prognosis in survivors of Stanford type A aortic dissection.
Introduction
Acute Stanford type A aortic dissection is a medical emergency with a high risk of complicating events such as aortic valve insufficiency, aortic rupture (e.g. into the pericardium, the pleural space, or mediastinum), or dissection into the coronary arteries.1 Mortality is 50% without surgical intervention during the first 48 hours, and hence guidelines recommend urgent surgery in all patients.2–4
Survival until discharge is the first point of success after a type A aortic dissection. Long-term survival, good quality of life, and no need for reintervention are secondary, however important, post-discharge goals. Bekkers and colleagues reported a need for re-operation in one out of five patients who had undergone surgery for type A aortic dissection during a study period from 1972–2011.5 During the last two decades, improvements in diagnostics (computed tomography—CT) and surgical techniques have led to a decrease in the mortality from type A aortic dissection.6–8 Because of this development, updated real-world data on survivors of type A aortic dissection are of interest. Therefore, by including data from a nationwide unselected cohort of type A aortic dissection survivors, we aimed to describe and compare, across age groups, rates of rehospitalisations, repeated aortic surgery, and all-cause mortality 1 and 10 years after discharge.
Methods
Data sources
The study was performed using data from the entire Danish population. We linked data from several administrative nationwide registries using the Danish civil registration-number (which all inhabitants are given at birth or immigration to Denmark and which is unique and permanent). We were thereby able to link data about date of birth and death, sex, hospital contacts and diagnoses, prescribed medication, etc. on an individual level. Data from the following registries were included: The Danish Civil Registration System includes information about sex, date of birth, and immigration/emigration status.9 The Danish National Patient Register includes information about all hospital contacts in Denmark. From each hospital contact, type (inpatient, outpatient, emergency department visit), date of hospitalization/discharge, and diagnosis codes given are registered. Diagnoses are registered according to the International Classification of Diseases System version 10 (ICD-10). Further, the registry includes all surgical procedures, registered according to the NOMESCO Classification of Surgical Procedures (NCSP).10 The Danish National Prescription Registry holds information about all fulfilled prescriptions of medication in Denmark. The type of medication is registered according to the Anatomical Therapeutic Chemical classification system (ATC).11 The Danish Registry of Causes of Death includes information about the date and cause(s) of death.12
Study population
We identified all adult patients hospitalized (inpatient hospitalizations) with aortic dissection between 1 January, 2006 and 31 December, 2018. To ensure as complete healthcare data as possible before inclusion, we only included patients who had no history of emigration from Denmark (people living abroad for a period and then returning to Denmark). Throughout the entire manuscript, we will use the Stanford classification of aortic dissections. Patients with type B aortic dissections were excluded. A large proportion of the patients had received a diagnosis code not classifying whether they had type A or B aortic dissection. Among these, we included those who had had urgent surgery on the ascending aorta or the aortic arch. Urgent surgery was defined as surgery during hospitalization. Since urgent surgery is the standard treatment for type A aortic dissection,2 we excluded patients who did not receive urgent surgery. Lastly, we excluded those who died during the hospitalization and grouped the remaining patients into three age groups: age < 60 years, age 60–69, and age > 69.
Outcomes
Primary outcomes included rehospitalization (any cause) and repeated aortic surgery while the secondary outcome was all-cause mortality. Patients were followed from discharge defined as the index date.
Statistics
Comorbidities and concomitant pharmacotherapy were collected at the index date (discharge after hospitalization with type A aortic dissection). Numbers and counts for categorical variables and medians and 25th and 75th percentiles (p25–p75) for continuous variables were presented for different age groups. To meet guidelines from Statistics Denmark and to secure anonymity < 3 was used whenever there were 1 or 2 patients as well as ‘NA‘ was used when the actual number could help identify specific patients. One-year and 10-year rates of rehospitalization, repeated aortic surgery, and all-cause mortality were depicted using cumulative incidence curves, taking the competing risk of death into account. Crude and adjusted hazard ratios (HR) and 95% CI of outcomes (rehospitalization, repeated aortic surgery, and all-cause mortality) were calculated by multivariable Cox regression analysis comparing the defined age groups as well as testing for an association between the outcomes and age as a continuous variable. We calculated HRs one year as well as 10 years after the index date. Variables included in the Cox regression models were sex, calendar year, comorbidities (prior surgery on the aortic valve, aortic valve insufficiency, aortic valve stenosis, prior or current cancer, chronic kidney disease, diabetes, heart failure, hypertension, ischaemic heart disease, prior bleeding event, prior thromboembolic event), and concomitant pharmacotherapy (Adenosine diphosphate (ADP-receptor blockers, aspirin, and anticoagulation therapy).
Other analyses
The number of hospitalizations during the first year after the index date and proportion of days out of hospital during the first year of follow up were calculated for each patient in the study population and for a matched cohort from the background population. Matching was done using exposure density sampling and performed using a previously described method.13 Patients from the study population were matched 1:2 with controls from the background population upon age, sex, and calendar year. Controls were randomly assigned index dates using the index dates of the cases.
Validation study
We validated our definition of type A aortic dissection by reviewing 191 records registered with an ICD-10 code for type A aortic dissection (DI71A) or aortic dissection (any part) (DI710) from a Danish tertiary care center (University Hospital of Copenhagen, Rigshospitalet). Among those classified as having type A aortic dissection and who underwent urgent surgery on the ascending aorta or the aortic arch (N = 126) and those classified as having aortic dissection (any part) who underwent urgent surgery on the ascending aorta or the aortic arch (N = 27), 145 had a clinically documented type A aortic dissection. This corresponded to a positive predictive value of 94.8%. The remaining 5% had either previous type A aortic dissection, aortic aneurism, or had had surgery or trauma resulting in aortic lesions. Among those classified as type A aortic dissection, who did not undergo surgery but survived until discharge, <50% had type A aortic dissection. The quality assessment of the diagnosis of type A aortic dissection was approved by the hospital (University Hospital of Copenhagen, Rigshospitalet).
Ethics
Approval from the Research Ethical Committee System is not required in retrospective registry-based studies in Denmark. The Danish Data Protection Agency approved the use of data for this registry-study (ref.no: 2007–58-0015/GEH-2014–015 I-Suite no: 02 733).
Results
Study population
The selection of the study population is depicted in Figure1. From 2006–2018, 1058 adult patients with type A aortic dissection were hospitalized. Of these, 606 (57%) went through surgery on the ascending aorta and/or aortic arch and survived until discharge. Of these, 526 (86.85%) went through surgery within the first 2 days of admission. The study population comprised 236 (39%) patients aged < 60 years, 194 (32%) patients aged 60–69 years, and 176 (29%) patients aged > 69 years (Figure1). Baseline characteristics at the index for the entire study population and divided into age groups are shown in Table 1. Males accounted for 78% of patients aged < 60 years, 74% of patients aged 60–69 years, and 56% of patients aged > 69 years. The proportion of patients who had concomitant aortic valve replacement during dissection surgery was 194 (32.0%) and most frequent in the youngest patients (39.0% in patients aged < 60 years, 30.4% in patients aged 60–69 years, and 25% in patients aged > 69 years, respectively). In general, the proportion of comorbidities increased with increasing age. Bicuspid aortic valve, Marfan syndrome, and Turner's syndrome were most common in the youngest age group. One year after discharge, 13 (6%) of those aged < 60 years, </= 3 of those aged 60–69 years, and none of those aged > 69 years had been diagnosed with Marfan syndrome. The median length of hospital admission for type A aortic dissection was 14 days, differing by one day in between age groups. Supplementary material online, Figure S1 shows the distribution of length of stays among the study population; 12% of the patients were hospitalized >4 weeks.
Flow diagram illustrating the patient selection.
Baseline characteristics at discharge from hospitalisation with type A aortic dissection
| All patients N = 606 | Age <60 N = 236 | Age 60–69N = 194 | Age >69 N = 176 | |
|---|---|---|---|---|
| Demographics | ||||
| Age, median years (p25–p75) | 64 (53–71) | 50 (43.5–55) | 65 (63–67) | 75 (72–78) |
| Males, n (%) | 426 (70.3) | 184 (78.0) | 144 (74.2) | 98 (55.7) |
| Length of stay, median days (p25–p75) | 14 (9–21) | 14 (9–21) | 13 (9–19) | 14 (9.5–24) |
| Type of surgery, n (%) | ||||
| Ascending aorta, only | 349 (57.6) | 122 (51.7) | 115 (59.3) | 112 (63.6) |
| Aortic arch, only | 15 (2.5) | 5 (2.1) | 6 (3.1) | 4 (2.3) |
| Ascending aorta and aortic arch | 47 (7.8) | 17 (7.2) | 14 (7.2) | 16 (9.1) |
| Ascending aorta and aortic valve | 172 (28.4) | 84 (35.6) | 52 (26.8) | 36 (20.9) |
| Ascending aorta, aortic arch, and aortic valve | 22 (3.6) | 8 (3.4) | 6 (3.1) | 8 (4.6) |
| Comorbidities, n (%) | ||||
| Aortic valve insufficiency | NA | 5 (2.1) | <3 | <3 |
| Aortic valve stenosis | 26 (8.0) | 4 (1.7) | 8 (4.1) | 14 (8.0) |
| Aortic valve surgery | 13 (2.8) | 4 (1.7) | 4 (2.1) | 5 (2.8) |
| Bicuspid aortic valve | NA | 5 (2.1) | <3 | <3 |
| Cancer | 44 (7.3) | 6 (2.5) | 10 (5.2) | 28 (15.9) |
| Chronic kidney disease | 34 (5.6) | 5 (2.1) | 14 (7.2) | 15 (8.5) |
| Diabetes | 24 (4.0) | 11 (4.7) | 5 (2.6) | 8 (4.6) |
| Heart failure | 44 (7.3) | 8 (3.4) | 20 (10.3) | 16 (9.1) |
| Hypertension diagnosisa | 386 (63.7) | 135 (57.2) | 137 (70.6) | 114 (64.8) |
| Hypertension medicationa | 260 (42.9) | 84 (35.6) | 83 (42.8) | 93 (52.8) |
| Hypertension in totala | 445 (73.4) | 160 (67.8) | 154 (79.4) | 131 (74.4) |
| Ischaemic heart disease | 115 (19.0) | 26 (11.0) | 41 (21.1) | 48 (27.3) |
| Marfan's syndrome | 7 (1.2) | 7 (3.0) | 0 | 0 |
| Prior bleeding event | 100 (16.5) | 22 (9.3) | 35 (18.0) | 43 (24.4) |
| Prior thromboembolic event | 115 (19.0) | 34 (14.4) | 42 (21.7) | 39 (22.2) |
| Turner's syndrome | <3 | <3 | 0 | 0 |
| Concomitant pharmacotherapy, n (%) | ||||
| ADP-recepter blockers | NA | <3 | 6 (3.1) | 8 (4.6) |
| Anticoagulation therapyb | 81 (13.4) | 29 (12.3) | 19 (9.8) | 33 (18.8) |
| Aspirin | 116 (19.1) | 31 (13.1) | 43 (22.2) | 42 (23.9) |
| All patients N = 606 | Age <60 N = 236 | Age 60–69N = 194 | Age >69 N = 176 | |
|---|---|---|---|---|
| Demographics | ||||
| Age, median years (p25–p75) | 64 (53–71) | 50 (43.5–55) | 65 (63–67) | 75 (72–78) |
| Males, n (%) | 426 (70.3) | 184 (78.0) | 144 (74.2) | 98 (55.7) |
| Length of stay, median days (p25–p75) | 14 (9–21) | 14 (9–21) | 13 (9–19) | 14 (9.5–24) |
| Type of surgery, n (%) | ||||
| Ascending aorta, only | 349 (57.6) | 122 (51.7) | 115 (59.3) | 112 (63.6) |
| Aortic arch, only | 15 (2.5) | 5 (2.1) | 6 (3.1) | 4 (2.3) |
| Ascending aorta and aortic arch | 47 (7.8) | 17 (7.2) | 14 (7.2) | 16 (9.1) |
| Ascending aorta and aortic valve | 172 (28.4) | 84 (35.6) | 52 (26.8) | 36 (20.9) |
| Ascending aorta, aortic arch, and aortic valve | 22 (3.6) | 8 (3.4) | 6 (3.1) | 8 (4.6) |
| Comorbidities, n (%) | ||||
| Aortic valve insufficiency | NA | 5 (2.1) | <3 | <3 |
| Aortic valve stenosis | 26 (8.0) | 4 (1.7) | 8 (4.1) | 14 (8.0) |
| Aortic valve surgery | 13 (2.8) | 4 (1.7) | 4 (2.1) | 5 (2.8) |
| Bicuspid aortic valve | NA | 5 (2.1) | <3 | <3 |
| Cancer | 44 (7.3) | 6 (2.5) | 10 (5.2) | 28 (15.9) |
| Chronic kidney disease | 34 (5.6) | 5 (2.1) | 14 (7.2) | 15 (8.5) |
| Diabetes | 24 (4.0) | 11 (4.7) | 5 (2.6) | 8 (4.6) |
| Heart failure | 44 (7.3) | 8 (3.4) | 20 (10.3) | 16 (9.1) |
| Hypertension diagnosisa | 386 (63.7) | 135 (57.2) | 137 (70.6) | 114 (64.8) |
| Hypertension medicationa | 260 (42.9) | 84 (35.6) | 83 (42.8) | 93 (52.8) |
| Hypertension in totala | 445 (73.4) | 160 (67.8) | 154 (79.4) | 131 (74.4) |
| Ischaemic heart disease | 115 (19.0) | 26 (11.0) | 41 (21.1) | 48 (27.3) |
| Marfan's syndrome | 7 (1.2) | 7 (3.0) | 0 | 0 |
| Prior bleeding event | 100 (16.5) | 22 (9.3) | 35 (18.0) | 43 (24.4) |
| Prior thromboembolic event | 115 (19.0) | 34 (14.4) | 42 (21.7) | 39 (22.2) |
| Turner's syndrome | <3 | <3 | 0 | 0 |
| Concomitant pharmacotherapy, n (%) | ||||
| ADP-recepter blockers | NA | <3 | 6 (3.1) | 8 (4.6) |
| Anticoagulation therapyb | 81 (13.4) | 29 (12.3) | 19 (9.8) | 33 (18.8) |
| Aspirin | 116 (19.1) | 31 (13.1) | 43 (22.2) | 42 (23.9) |
aHypertension diagnosis extracted from the Danish National Patient Registry, hypertension medication extracted from the Danish National Prescription Registry, hypertension in total—either hypertension diagnosis or hypertension medication or both.
bVitamin K antagonists and direct oral anticoagulants (DOACs)
Baseline characteristics at discharge from hospitalisation with type A aortic dissection
| All patients N = 606 | Age <60 N = 236 | Age 60–69N = 194 | Age >69 N = 176 | |
|---|---|---|---|---|
| Demographics | ||||
| Age, median years (p25–p75) | 64 (53–71) | 50 (43.5–55) | 65 (63–67) | 75 (72–78) |
| Males, n (%) | 426 (70.3) | 184 (78.0) | 144 (74.2) | 98 (55.7) |
| Length of stay, median days (p25–p75) | 14 (9–21) | 14 (9–21) | 13 (9–19) | 14 (9.5–24) |
| Type of surgery, n (%) | ||||
| Ascending aorta, only | 349 (57.6) | 122 (51.7) | 115 (59.3) | 112 (63.6) |
| Aortic arch, only | 15 (2.5) | 5 (2.1) | 6 (3.1) | 4 (2.3) |
| Ascending aorta and aortic arch | 47 (7.8) | 17 (7.2) | 14 (7.2) | 16 (9.1) |
| Ascending aorta and aortic valve | 172 (28.4) | 84 (35.6) | 52 (26.8) | 36 (20.9) |
| Ascending aorta, aortic arch, and aortic valve | 22 (3.6) | 8 (3.4) | 6 (3.1) | 8 (4.6) |
| Comorbidities, n (%) | ||||
| Aortic valve insufficiency | NA | 5 (2.1) | <3 | <3 |
| Aortic valve stenosis | 26 (8.0) | 4 (1.7) | 8 (4.1) | 14 (8.0) |
| Aortic valve surgery | 13 (2.8) | 4 (1.7) | 4 (2.1) | 5 (2.8) |
| Bicuspid aortic valve | NA | 5 (2.1) | <3 | <3 |
| Cancer | 44 (7.3) | 6 (2.5) | 10 (5.2) | 28 (15.9) |
| Chronic kidney disease | 34 (5.6) | 5 (2.1) | 14 (7.2) | 15 (8.5) |
| Diabetes | 24 (4.0) | 11 (4.7) | 5 (2.6) | 8 (4.6) |
| Heart failure | 44 (7.3) | 8 (3.4) | 20 (10.3) | 16 (9.1) |
| Hypertension diagnosisa | 386 (63.7) | 135 (57.2) | 137 (70.6) | 114 (64.8) |
| Hypertension medicationa | 260 (42.9) | 84 (35.6) | 83 (42.8) | 93 (52.8) |
| Hypertension in totala | 445 (73.4) | 160 (67.8) | 154 (79.4) | 131 (74.4) |
| Ischaemic heart disease | 115 (19.0) | 26 (11.0) | 41 (21.1) | 48 (27.3) |
| Marfan's syndrome | 7 (1.2) | 7 (3.0) | 0 | 0 |
| Prior bleeding event | 100 (16.5) | 22 (9.3) | 35 (18.0) | 43 (24.4) |
| Prior thromboembolic event | 115 (19.0) | 34 (14.4) | 42 (21.7) | 39 (22.2) |
| Turner's syndrome | <3 | <3 | 0 | 0 |
| Concomitant pharmacotherapy, n (%) | ||||
| ADP-recepter blockers | NA | <3 | 6 (3.1) | 8 (4.6) |
| Anticoagulation therapyb | 81 (13.4) | 29 (12.3) | 19 (9.8) | 33 (18.8) |
| Aspirin | 116 (19.1) | 31 (13.1) | 43 (22.2) | 42 (23.9) |
| All patients N = 606 | Age <60 N = 236 | Age 60–69N = 194 | Age >69 N = 176 | |
|---|---|---|---|---|
| Demographics | ||||
| Age, median years (p25–p75) | 64 (53–71) | 50 (43.5–55) | 65 (63–67) | 75 (72–78) |
| Males, n (%) | 426 (70.3) | 184 (78.0) | 144 (74.2) | 98 (55.7) |
| Length of stay, median days (p25–p75) | 14 (9–21) | 14 (9–21) | 13 (9–19) | 14 (9.5–24) |
| Type of surgery, n (%) | ||||
| Ascending aorta, only | 349 (57.6) | 122 (51.7) | 115 (59.3) | 112 (63.6) |
| Aortic arch, only | 15 (2.5) | 5 (2.1) | 6 (3.1) | 4 (2.3) |
| Ascending aorta and aortic arch | 47 (7.8) | 17 (7.2) | 14 (7.2) | 16 (9.1) |
| Ascending aorta and aortic valve | 172 (28.4) | 84 (35.6) | 52 (26.8) | 36 (20.9) |
| Ascending aorta, aortic arch, and aortic valve | 22 (3.6) | 8 (3.4) | 6 (3.1) | 8 (4.6) |
| Comorbidities, n (%) | ||||
| Aortic valve insufficiency | NA | 5 (2.1) | <3 | <3 |
| Aortic valve stenosis | 26 (8.0) | 4 (1.7) | 8 (4.1) | 14 (8.0) |
| Aortic valve surgery | 13 (2.8) | 4 (1.7) | 4 (2.1) | 5 (2.8) |
| Bicuspid aortic valve | NA | 5 (2.1) | <3 | <3 |
| Cancer | 44 (7.3) | 6 (2.5) | 10 (5.2) | 28 (15.9) |
| Chronic kidney disease | 34 (5.6) | 5 (2.1) | 14 (7.2) | 15 (8.5) |
| Diabetes | 24 (4.0) | 11 (4.7) | 5 (2.6) | 8 (4.6) |
| Heart failure | 44 (7.3) | 8 (3.4) | 20 (10.3) | 16 (9.1) |
| Hypertension diagnosisa | 386 (63.7) | 135 (57.2) | 137 (70.6) | 114 (64.8) |
| Hypertension medicationa | 260 (42.9) | 84 (35.6) | 83 (42.8) | 93 (52.8) |
| Hypertension in totala | 445 (73.4) | 160 (67.8) | 154 (79.4) | 131 (74.4) |
| Ischaemic heart disease | 115 (19.0) | 26 (11.0) | 41 (21.1) | 48 (27.3) |
| Marfan's syndrome | 7 (1.2) | 7 (3.0) | 0 | 0 |
| Prior bleeding event | 100 (16.5) | 22 (9.3) | 35 (18.0) | 43 (24.4) |
| Prior thromboembolic event | 115 (19.0) | 34 (14.4) | 42 (21.7) | 39 (22.2) |
| Turner's syndrome | <3 | <3 | 0 | 0 |
| Concomitant pharmacotherapy, n (%) | ||||
| ADP-recepter blockers | NA | <3 | 6 (3.1) | 8 (4.6) |
| Anticoagulation therapyb | 81 (13.4) | 29 (12.3) | 19 (9.8) | 33 (18.8) |
| Aspirin | 116 (19.1) | 31 (13.1) | 43 (22.2) | 42 (23.9) |
aHypertension diagnosis extracted from the Danish National Patient Registry, hypertension medication extracted from the Danish National Prescription Registry, hypertension in total—either hypertension diagnosis or hypertension medication or both.
bVitamin K antagonists and direct oral anticoagulants (DOACs)
Supplementary material online, Table S1 shows baseline characteristics of those who were excluded from the study population, i.e. patients who survived the hospitalization without having undergone urgent surgery and patients who died during the hospitalization regardless of urgent surgery or not.
Rehospitalisations
Figure2 illustrates 1- and 10-year cumulative incidences of rehospitalisations, repeated aortic surgery (taking death into account as a competing risk), and all-cause mortality in patients aged < 60 years, 60–69 years, and > 69 years, respectively. After one year, 63% of patients, and after 10 years, 77% of patients had been readmitted, with no significant difference between age groups. The median time from the index date until the first rehospitalization was 16 days (IQR 5–55 days) and the median number of days hospitalized was 3 days (IQR 2–9 days). Of those patients who were re-hospitalized during the first year of follow up, 52% were admitted for 1–3 days and 69% <1 week. In the adjusted analyses, no differences were found across age groups regarding one-year as well as 10-year rates of rehospitalization vs. no rehospitalization (Figure3). This did not change when age was included as a continuous variable (one-year: HR 1.00, 95% CI 0.99–1.01 and 10 years: HR 1.00, 95% CI 0.99–1.01). Figure4A shows the number of hospitalizations per patient during the first year of follow up while Figure4B shows weeks alive and out of hospital during the first year of follow up (only including those patients with possible follow-up of one year, e.g. index date at least 1 year before 31 December, 2018, N = 520). During the first year, 64% (62%, 62%, and 69% of those aged < 60 years, 60–69 years, and > 69 years respectively) of the study population were hospitalized at least one time (14% of the matched controls, supplementary material online, Figure S2) and 27% were hospitalized at least three times (4% of the matched controls, supplementary material online, Figure S2). During the first year of follow up, 64% spent >51 weeks alive and out of hospital, (93% of the matched controls, supplementary material online, Figure S2) and 4% spent <12 weeks alive and out of hospital (0.5% of the matched controls, supplementary material online, Figure S2). Among cases, the median numbers of days alive and out of hospital during the first year of follow up were 363 days (p25–p75 352–365), 362 days (p25–p75 347–365), and 358 days (p25–p75 339–365) for those aged < 60 years, 60–69 years, and > 69 years, respectively.
One- and 10-years age-stratified cumulative incidences of rehospitalization, repeated aortic surgery, and all-cause mortality.
Number of events, person years (PY), incidence rates (IR), and crude and adjusted hazard ratios (HRs) of rehospitalisation, repeated aortic surgery, and all-cause mortality (one- and 10-years outcomes).
(A) Number of hospitalisations per patient during the first year after discharge. (B) Weeks alive and out of hospital during the first year after discharge.
Among the first rehospitalisations, causes were dominated by 127 (33%) classified as caused by diseases from the circulatory system (ICD-10 ‘DI’) and 78 (21%) classified as caused by factors influencing health status and contact with health services (ICD-10 ‘DZ’) such as ‘Medical observation and evaluation for suspected diseases and conditions, ruled out’ and ‘Follow-up examination after treatment for malignant neoplasms’. See supplementary material online, Table S3 for further specifications.
Repeated aortic surgery and all-cause death
After 1 and 5 years of follow up, 1% and 4%, respectively, of the survivors had gone through repeated aortic surgery, with no significant difference between the age groups. After 10 years of follow up, 8% had repeated aortic surgery (12% of those aged < 60 years, 9% of those aged 60–69 years, and 2% of those aged > 69 years, P = 0.04). In the long-term analysis including 10 years of follow-up, older age (>69 years) compared with younger age (<60 years) was associated with a statistically significant lower rate of repeated aortic surgery (Figure3). Increasing age, continuously measured, showed neither a statistically significant association with one-year [adjusted HR 0.98, 95% CI (0.92–1.04)] nor 10-year (HR 0.97, 95% CI 0.94–1.00) rates of repeated aortic surgery.
One year after discharge, 3, 8, and 7% of those aged < 60 years, 60–69 years, and > 69 years, respectively, had died and after 10 years, 10, 17, and 22% of those aged < 60 years, 60–69 years, and > 69 years, respectively, had died. Of all patients who died during follow-up, 72% were registered as dead due to a cardiovascular cause and 7% due to cancer. In the adjusted analysis, increasing age was significantly associated with an increased rate of all-cause mortality during 10 years of follow-up (Figure3). Increasing age, continuously measured, showed a statistically significant association with one-year and 10-year rates of all-cause mortality (HR 1.05, 95% CI 1.01–1.08 and HR 1.04, 95% CI 1.02–1.06, respectively).
Discussion
In this study, we explored one-year as well as 10-year outcomes among 606 survivors of hospitalization and surgery for type A aortic dissection. Our main findings were as follows: first, during the initial year of follow up, 63% of the study population were re-hospitalized at least one time with no significant differences across age groups. However, the time spent alive and out of hospital was >51 weeks during the first year of follow-up in almost two-thirds of the patients. Secondly, the rates of repeated aortic surgery were low (1% during the first year of follow-up and 8% during 10 years of follow up) with an association between increasing age and a lower rate of repeated aortic surgery.
Rehospitalisations
Almost two-thirds of the study population were re-hospitalized at least one time during the first year after their aortic dissection with no significant difference across age groups. Further, more than a quarter of the patients had been hospitalized at least three times during the first year of follow up. The median time from discharge to first rehospitalization was ∼2 weeks and the median length of this second hospital stay was 3 days (IQR 2–9 days) with more than two-thirds of those hospitalized, being admitted for <1 week.
The majority (almost two-thirds) of the study population spent >51 weeks out of hospital and alive during the first year of follow-up. Further, <4% spent <3 months out of hospital and alive. These are encouraging numbers. To our knowledge, no previous studies have evaluated rehospitalisations in survivors of type A aortic dissections and in this study, we were not able to appropriately elucidate reasons for rehospitalisations. Thus, future studies grouping and evaluating rehospitalisations in this patient group could be of major interest in the process to obtain better care for patients with type A aortic dissections.
Repeated aortic surgery
Repeated aortic surgery was not common. Approximately 1% of the study population underwent repeated aortic surgery during the initial year of follow up and <10% died in all age groups. Surgery is recommended in all patients with type A aortic dissection,2 and our results indicate that patients surviving initial surgery for type A aortic dissection, independently of age, have a relatively good prognosis in the first year. However, it is important to keep in mind, that type A aortic dissection is a thoracic catastrophe with a high in-hospital mortality.4 In 2008, Shrestha et al. found high rates of neurological complications in elderly patients (>80 years) treated with surgical intervention for type A aortic dissections.14 In 2013, Tang et al. showed better emotional health scores among elderly patients (>70 years) vs. younger patients treated with surgery for type A aortic dissections.15 Obviously, studies evaluating quality of life in survivors of type A aortic dissections would be of great interest, and could potentially help clinicians in the choice of treatment especially older patients presenting with aortic dissection.
After 10 years of follow up, 8% of the study population (taking death into account as a competing risk) had undergone repeated aortic surgery. This is a significantly lower rate than reported by Bekkers et al. for patients who underwent surgery for type A aortic dissection from 1972–2011 (one out of five were reoperated).5 This likely reflects improvements in surgical treatment of type A aortic dissections over the years. Repeated aortic surgery was significantly less common among older patients. A lower rate of repeated aortic surgery in the elderly may not necessarily reflect a lower need for re-surgery, but more likely clinical decisions not to operate due to a higher predicted risk for elderly going through surgery compared with younger. A couple of studies have shown better survival in elderly patients with type A aortic dissections who underwent primary surgical intervention compared with no surgical intervention.14,15 However, to the best of our knowledge, there is a lack of studies investigating post-discharge surgical complications and the need for repeated surgery in older patients. It remains to be explored whether the risk of further complications supersede the benefits of repeated surgery in older vs. younger patients with surgical indications remain unexplored, and future studies exploring these areas could be of great interest for clinicians handling this patient group. Further, we found a higher proportion of patients who had received aortic valve replacement and in general more intensive aortic repair during the urgent surgery for their type A aortic dissection in the younger patients. Whether this reflects a more severe pathology in the younger patients, or a higher proportion of survivors among the youngest patients undergoing more complex surgery was not explored in this study. However, more complex surgery in younger patients could contribute to a higher need for reintervention.
Limitations
First of all, we used data from administrative registries and did not have access to information about symptoms, extent of dissection, perioperative complications, frailty etc. Secondly, the retrospective and observational study design only allowed us to look at associations, and thus, no causative relationships could be drawn. Lastly, we found high rates of rehospitalisations with many different causes registered in the registry (The Danish National Patient Register). Review of the medical records from the rehospitalisations may had helped us understand the reasons for the rehospitalisations better, However, we did not have permission to access the journals. Positively, we had access to an unselected national cohort of survivors of type A aortic dissection with several years of follow up from high-quality registries.
We validated the diagnosis of type A aortic dissection and found a high positive predictive value (94.8%). To avoid patients with chronic aortic dissection and patients with short life expectancy, we only included patients who underwent surgery on the aorta (ascending part or aortic arch) during the hospitalization, since this is the standard treatment for type A aortic dissections.2 Further, our validation study showed, that <50% of those who were given a diagnosis code for type A aortic dissection, but did not undergo urgent surgery, actually had type A aortic dissection. This may also explain the low rate of patients undergoing surgery for their type A aortic dissection in this study.
Conclusions
In this nationwide cohort of survivors of type A aortic dissection, rehospitalisations were common, irrespectively of age. Re-hospitalisation tended to be of short length and only a minority spent more than a week hospitalised in the first year after their dissection. Repeated aortic surgery during the first year of follow-up was not common. However, younger age was associated with higher rates of recurrent aortic surgery and after 10 years of follow up, one out of 10 patients aged < 70 years had had repeated aortic surgery. Thus, this study in general shows uplifting results for those who survive their aortic dissection, however, studies exploring quality of life in this patient group are warranted.
Acknowledgment
None.
Funding
None.
Conflicts of interest
AG: None
LK: Speakers honorarium from AstraZeneca, Boehringer, Bayer, Novartis and Novo.
DEH: board member, Danish Society of Cardiology.
MVA: None.
MWP: None.
CTP: None.
KK: None.
PS: Leadership or fiduciary role in Bioguard-MI and speakers honorarium from Biotronik.
MS: None.
ELF: None.
Data availability
The data underlying this article cannot be shared publicly due to the privacy of individuals that participated in the study. The data will be shared on reasonable request to the corresponding author.
