- Email: [email protected]
Epidemiology and management of thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada: A population-based study
McClure et al
Acquired
Epidemiology and management of thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada: A population-based study R. Scott McClure, MD,a Susan B. Brogly, PhD,b,c Katherine Lajkosz, MS,c Darrin Payne, MD,b Stephen F. Hall, MD,c and Ana P. Johnson, PhDc ABSTRACT Objectives: To determine hospital incidence, mortality, and management for thoracic aortic dissections and aneurysms. Methods: A population-based retrospective cohort study of anonymously linked data for residents of Ontario, Canada, was carried out. Incident cases of thoracic aortic dissections and aneurysms were identified between 2002 and 2014. Treatment and mortality trends were assessed.
Conclusions: The incidence of thoracic aortic dissections and aneurysms increased over time but all-cause hospital and late outcomes improved. Gender differences exist. Men incur more disease but women have higher hospital mortality. Surgery was primarily referred to cardiac surgeons. Endovascular therapy was primarily referred to vascular surgeons. (J Thorac Cardiovasc Surg 2018;-:1-11)
From the aDivision of Cardiac Surgery, Libin Cardiovascular Institute, Foothills Medical Center, University of Calgary, Calgary, Alberta, Canada; and bDepartment of Surgery, Kingston General Hospital, and cInstitute for Clinical and Evaluative Sciences, Queen’s University, Kingston, Ontario, Canada. Supported by the Establishment Fund, Queen’s University, Kingston, Ontario, Canada, and an internal grant from the Departmental, Development & Innovation Fund, Queen’s University, Kingston, Ontario, Canada. Parts of this work are based on data and information compiled and provided by the Canadian Institute for Health Information. However, the analyses, conclusions, opinions, and statements expressed herein are those of the authors, and not necessarily those of the Canadian Institute for Health Information.
Incidence proportions for thoracic aortic dissections and aneurysms in Ontario, Canada. Central Message The incidence of thoracic aortic aneurysms and dissections has increased, but hospital mortality and late outcomes have improved. Men incur more thoracic aortic disease but women have higher hospital mortality.
ACQ
Results: There were 5966 aortic dissections (Type A n ¼ 2289 [38%] and Type B n ¼ 3632 [61%]). Overall incidence proportion for aortic dissections was 4.6 per 100,000. There were 9392 thoracic aortic aneurysms with an overall incidence proportion of 7.6 per 100,000. The incidence for both dissections and aneurysms significantly increased over the 12-year study. Only 53% (1204 out of 2289) of Type A dissections underwent surgery. Type B dissection treatment was 83% (3000 out of 3632) medical, 10% (370 out of 3632) surgery, and 7% (262 out of 3632) endovascular. Thoracic aortic aneurysm treatment was 53% (4940 out of 9392) surgery, 44% (4129 out of 9392) medical, and 3% (323 out of 9392) endovascular. Thirty-five percent of known descending thoracic aortic aneurysms (323 out of 924) received a stent graft. Cardiac surgeons performed 87% of the open surgical repairs. Vascular surgeons performed 91% of the endovascular procedures. All-cause 3-year mortality significantly decreased for both aortic dissections (44% to 40%) and aneurysms (30% to 22%). All-cause hospital mortality also decreased. Women had worse outcomes than men.
Perspective The incidence of disease is integral to resource allocation for cost-efficient care. This study demonstrates the incidence of thoracic aortic aneurysms and dissections have increased yet hospital and late outcomes have improved. Men incur more aortic disease yet women have higher hospital mortality. Surgical care is primarily performed by cardiac surgeons. Vascular surgeons perform most endovascular procedures.
See Editorial Commentary page XXX.
Drs McClure and Payne are members of the Canadian Thoracic Aortic Collaborative, a group of cardiovascular clinicians across Canada with a specific interest and expertise in the management of thoracic aortic disease. Read at the 97th Annual Meeting of The American Association for Thoracic Surgery, Boston, Massachusetts, April 29-May 3, 2017. Received for publication May 10, 2017; revisions received Oct 15, 2017; accepted for publication Nov 13, 2017. Address for reprints: R. Scott McClure, MD, Division of Cardiac Surgery, Libin Cardiovascular Institute, University of Calgary, 1403 29th St, NW, Rm C803, Calgary, Alberta, Canada T2N 2T9 (E-mail: [email protected]). 0022-5223/$36.00 Copyright Ó 2018 by The American Association for Thoracic Surgery https://doi.org/10.1016/j.jtcvs.2017.11.105
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
1
Acquired
Abbreviations and Acronyms CCI ¼ Canadian Classification of Health Interventions ICD-10-CA ¼ International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada IRAD ¼ International Registry for Acute Aortic Dissection NACRS ¼ National Ambulatory Care Reporting System TEVAR ¼ thoracic endovascular aortic repair
Scanning this QR code will take you to an appendix for this paper. To view the AATS Annual Meeting Webcast, see the URL next to the webcast thumbnail.
ACQ
Thoracic aortic dissections and thoracic aortic aneurysms are highly resource intensive intermixed disease processes associated with significant morbidity and mortality. Such conditions put profound strain on health care systems. With a clinical presentation that spans the indolent and likely benign (small aortic aneurysm) to that of emergent and life threatening (aortic dissection), proper allocation of health service resources to ensure quality patient care becomes a challenge. Moreover, data to characterize the true burden of thoracic aortic disease are sparse.1-8 Characterizing this burden is critical to ensuring quality care remains cost-efficient. In a single-payer health care system, such as present in Canada, this is prudent to financial sustainability.9,10 We performed a populationbased study to characterize the contemporary burden of disease and describe recent treatment paradigms. MATERIALS AND METHODS Study Design We designed a retrospective cohort study using anonymously linked health information for persons residing in the province of Ontario, Canada. The primary outcomes of interest were the overall and annual calendar year incidence of thoracic aortic dissections and thoracic aortic aneurysms. The cohort encompassed more than 13.5 million (12.2 million in 2002, increased to 13.7 million in 2014) persons of ethnic diversity with the majority being white. The data were housed at the Institute for Clinical Evaluative Sciences. The Institute for Clinical Evaluative Sciences links de-identified, patient-level population-based health information across several administrative databases in Ontario using a unique encrypted health number identifier. A systematic algorithm was constructed to facilitate a distinction between patients who incurred a Stanford type A or type B aortic dissection. Research ethics approval
2
McClure et al
was obtained from the research ethics committees at Queen’s University Kingston, Ontario, Canada, and Sunnybrook Hospital, Toronto, Ontario, Canada. Secondary outcomes included a descriptive characterization of thoracic aortic dissection and aneurysm treatments (medical, endovascular, or surgery), an assessment of surgical and hospital mortality rates, reintervention rates, and specialty involvement (cardiac or vascular surgeons).
Patient Selection All persons residing in the province of Ontario, Canada, between April 1, 2002, and March 31, 2014, were included in the study. Cases were acquired by presentation with the condition of interest to the 231 hospital sites or multitude of ambulatory clinics within the Ontario health care system. A thoracic aortic dissection or thoracic aortic aneurysm was an incident case if no prior diagnosis for that entity had been identified during the previous 5 years. The size criterion to determine whether an aorta was aneurysmal was at the treating physician’s discretion. Marfan, Ehlers-Danlos, and Turner syndrome, as well as connective tissue disorders of unspecified nature, were identified and tabulated. For each incident case, if treatment via endovascular therapy or surgery could not be identified, optimal medical therapy was the presumed treatment. Any reintervention after an initial endovascular or surgical procedure were recorded. Reintervention was defined as any endovascular stenting or surgical therapy > 72 hours after the initial procedure. The specialist on record as the primary operator for each case was captured to delineate cardiac and vascular surgery involvement. All-cause nonoperative hospital mortality (used for patients treated solely with medical therapy during hospital admission. All-cause nonoperative hospital mortality was defined as death occurring in a patient who did not have a surgical or endovascular procedure where the patient’s in-hospital record indicated a discharge status of dead, regardless of cause and length of stay) and operative mortality were assessed (used for patients treated with open surgery or endovascular therapy. Operative mortality was defined as all deaths occurring during the hospitalization in which the operation was performed, even if after 30 days). Follow-up ended on March 31, 2015, or the date of a patient’s death, whichever came first.
Data Collection The compiled dataset for the cohort of interest was the product of several multilinked administrative databases and networks. The Registered Persons Database, the Canadian Institute for Health Information Discharge Abstract Database and Same Day Surgery Database, the National Ambulatory Care Reporting System (NACRS), the Ontario Health Insurance Plan Database, the Ontario Drug Benefit Database, the Office of the Registrar General-Deaths database, and Statistics Canada. The Registered Persons Database collects vital statistics on all permanent residents of Ontario. The Canadian Institute for Health Information Discharge Abstract Database and Same Day Surgery Database record all hospital admissions and discharges along with information on diagnoses and any procedures performed. The NACRS collects data on all ambulatory visits, including outpatient clinics and visits to an emergency department. Diagnoses at presentation are coded using the International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada (ICD-10-CA). Hospital procedures are coded using the Canadian Classification of Health Interventions (CCI) coding system. For the cardiovascular system and pertinent to the study, CCI codes are discernably specific, partitioning surgical procedures based on the segment of aorta involved (ascending aorta, aortic arch, or descending thoracic aorta). Coding also partitions thoracic endovascular stenting procedures based on anatomic positioning within the abdominal or descending thoracic aorta. To cross-reference Canadian Institute for Health Information and NARCS data, the Ontario Health Insurance Plan Database captures fee-for-service billing claims for inpatient and outpatient services. The Ontario Drug Benefit Database records patient
The Journal of Thoracic and Cardiovascular Surgery c - 2018
Acquired
TABLE 1. Aortic dissection classification algorithm Aortic dissection classification algorithm (ICD-10-CA þ CCI codes) If diagnosis of an aortic dissection from ICD-10-CA codes AND Have 1 associated procedural code (CCI code) (i) CCI code ¼ surgery or TEVAR for AscA then type A dissection (ii) CCI code ¼ surgery or TEVAR for AArch then type A dissection (iii) CCI code ¼ surgery or TEVAR for DTA then type B dissection If diagnosis of an aortic dissection from ICD-10-CA codes AND Have 2 or 3 associated procedure codes (CCI codes) (on same day) (i) CCI codes ¼ surgery or TEVAR for AscA þ AArch then type A dissection (ii) CCI codes ¼ surgery or TEVAR for AscA þ DTA then type A dissection (iii) CCI codes ¼ surgery or TEVAR for AscAþ AArch þ DTA then type A dissection If diagnosis of an aortic dissection from ICD-10-CA codes AND Have no associated procedure code (CCI code) ¼ no intervention AND Patient alive at discharge (i) Then patient received medical management therefore type B dissection If diagnosis of an aortic dissection from ICD-10-CA codes AND Have no associated procedure code (CCI code) ¼ no intervention AND Patient died during hospital admission (i) Then assess ORGD for primary cause of death to determine whether type A or B dissection ICD-10-CA, International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada; CCI, Canadian Classification of Health Interventions; TEVAR, thoracic endovascular aortic repair; AscA, ascending aorta; AArch, aortic arch; DTA, descending thoracic aorta; ORGD, Office of the Registrar General - Death Database.
medications. The Office of the Registrar General - Deaths database records in-hospital deaths and cause of death. The systematic algorithm to distinguish type A from type B aortic dissections is shown in Table 1. The algorithm is predicated upon the CCI procedural codes ability to segregate procedures performed on the aorta into ascending, arch, and descending segments. The algorithm exploits the distinctly different treatment most often applied to treat acute type A (surgery) and acute type B (medical therapy or endovascular stenting) aortic dissections. The ICD-10-CA codes to determine the diagnosis of a thoracic aortic dissection or thoracic aortic aneurysm are in Table E1. The CCI codes for the surgical and endovascular procedures performed upon receiving a diagnosis of a thoracic aortic dissection or aneurysm are listed in Table E2. A conservative estimate was derived for thoracic aortic aneurysms, as ICD-10-CA codes I71 (thoracic aortic aneurysm, ruptured) and I715 (thoracoabdominal aortic aneurysm, ruptured) were categorized solely as thoracic aortic dissections. We did not double count such patients to be included in the thoracic aortic aneurysm distribution. Two attending cardiac surgeons independently reviewed all diagnostic and procedure codes.
Statistical Analysis Descriptive data and treatment strategies (medical, endovascular, and surgical) were assessed using analysis of variance and c2 tests for age
and gender differences. Poisson regression models were used to assess patient and hospital mortality trends over time. The incidence proportion was all new cases divided by the population of Ontario reported per 100,000. Incidence proportion was tabulated per year and over the entire population. The population of Ontario was defined as living Ontario residents (as of December 31 of the given year) who had contact with the health care system during the previous 7 years, excluding prevalent cases. All statistical tests were 2-sided. A P value of < .05 was deemed statistically significant. For categorical data resulting in cells<6, absolute numbers are not reported to ensure patient anonymity. All analyses were performed using SAS version 9.4 (SAS Institute Inc, Cary, NC).
RESULTS There were 5966 thoracic aortic dissections (type A n ¼ 2289 [38%] and type B n ¼ 3632 [61%], unknown type n ¼ 45 [1%]). The mean age was 66 years. The cohort was predominantly male (61%) (Table 2). The incidence proportion for all thoracic aortic dissections across the entire population (2002-2014) was 4.6 per 100,000 (95% confidence interval [CI], 4.2/100,000-5.0/100,000). For type A aortic dissections it was 1.9 out of 100,000 (95% CI, 1.7/100,000-2.1/100,000) and for type B aortic dissections it was 2.7 out of 100,000 (95% CI, 2.4/100,000-3.0/100,000) (Table 3). Persons with type A dissections were older, more likely to have hypertension, and more often had a connective tissue disorder relative to persons with type B dissections (Table 2). There were 9392 thoracic aortic aneurysms. The mean age was 67 years and the cohort was again, predominantly male (64%) (Table 2). The incidence proportion for thoracic aortic aneurysms across the entire population (2002-2014) was 7.6 per 100,000 (95% CI, 7.1/100,000-8.0/100,000). Male gender (Figure 1) and increasing age significantly increased the incidence of disease for both aortic dissections and aneurysms (Table 3). Despite more disease in men, total hospital mortality (operative mortality þ all-cause nonoperative hospital mortality) was significantly higher in women relative to men across the 12-year study for type A dissection and thoracic aneurysms. Type B dissections showed a nonsignificant trend (Table 3). For both cohorts, the presence of connective tissue disorders was rare (Table 2). The per-year incidence proportion for thoracic aortic dissections and thoracic aortic aneurysms increased over the 12-year study (aortic dissections from 2.7 to 4.6 per 100,000; P <.0001 and aortic aneurysms from 3.5 to 7.6 per 100,000; P < .0001), with an average increase of 1.8% and 4.7%, respectively (Figure 1). The increased incidence in aortic dissections was driven solely by type A dissections, with an average increase of 4%. No increase in type B dissections was identified. Only 53% of patients (1204 out of 2289) diagnosed with a type A dissection underwent surgery. Relative to patients undergoing surgery, patients with nonoperative treatment
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
3
ACQ
McClure et al
Acquired
McClure et al
TABLE 2. Demographic characteristics of patients diagnosed with a thoracic aortic dissection or thoracic aortic aneurysm in Ontario, Canada (2002-2014)
Characteristic
All (n ¼ 5966)
Percentage
Thoracic aortic dissection Type A Type B (n ¼ 2289) (n ¼ 3632) P value* 38.4
60.9
Age at diagnosis (y) 50 51-74 75
66.0 16.7 1031 (17.3) 2755 (46.2) 2180 (36.5)
68.0 15.2 310 (13.5) 1054 (46.0) 925 (40.4)
64.7 17.3 718 (19.8) 1684 (46.4) 1230 (33.9)
Sex Female Male
2339 (39.2) 3627 (60.8)
907 (39.6) 1382 (60.4)
1411 (38.8) 2221 (61.2)
Diabetes
1244 (20.9)
465 (20.3)
768 (21.1)
Unknown (n ¼ 45)
Thoracic aortic aneurysm (n ¼ 9392)
0.8 <.001 <.001
73.6 18.7 <6 17 (37.8) 25 (55.6)
67.3 14.8 1197 (12.7) 4770 (50.8) 3425 (36.5)
21 (46.7) 24 (53.3)
3373 (35.9) 6019 (64.1)
11 (24.4)
2160 (23.0)
.552
.443
COPD
1195 (20.0)
470 (20.5)
713 (19.6)
.398
12 (26.7)
2047 (21.8)
Hypertension
4489 (75.2)
1773 (77.5)
2682 (73.8)
.002
34 (75.6)
7299 (77.7)
Hyperlipidemia
1639 (27.5)
612 (26.7)
1016 (28.0)
.299
11 (24.4)
3367 (35.8)
58 (1.0)
30 (1.3)
28 (0.8)
.040
0 (0.0)
140 (1.5)
39 (86.7) <6 <6
8064 (85.9) 1318 (14.0) 10 (0.1)
<6 0 (0.0) <6 0 (0.0) 0 (0.0) <6 <6 <6 <6 0 (0.0) 0 (0.0) <6 34 (75.6)
432 (4.6) 567 (6.0) 591 (6.3) 697 (7.4) 737 (7.8) 697 (7.4) 793 (8.4) 863 (9.2) 857 (9.1) 943 (10.0) 932 (9.9) 1030 (11.0) 253 (2.7)
Connective tissue disorder
ACQ
Residence Urban Rural Missing Year of diagnosis 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
.570 5045 (84.6) 913 (15.3) 8 (0.1)
1949 (85.1) 337 (14.7) <6
3057 (84.2) 571 (15.7) <6
328 (5.5) 414 (6.9) 491 (8.2) 464 (7.8) 473 (7.9) 452 (7.6) 487 (8.2) 509 (8.5) 518 (8.7) 516 (8.6) 532 (8.9) 624 (10.5) 158 (2.6)
109 (4.8) 148 (6.5) 175 (7.6) 161 (7.0) 170 (7.4) 155 (6.8) 201 (8.8) 218 (9.5) 216 (9.4) 235 (10.3) 210 (9.2) 257 (11.2) 34 (1.5)
217 (6.0) 266 (7.3) 315 (8.7) 303 (8.3) 303 (8.3) 296 (8.1) 285 (7.8) 289 (8.0) 301 (8.3) 281 (7.7) 322 (8.9) 364 (10.0) 90 (2.5)
<.001
Values are presented as mean standard deviation or n (%). COPD, Chronic obstructive pulmonary disease. *P values are from c2 test for categorical variables or t test for continuous variables. A look-back period of 5 years before index date was used to look at risk factors (eg, diabetes, COPD, hypertension, and hyperlipidemia).
for type A dissections were statistically older, female, and had more comorbid conditions (Table 4). For the 12-year study, operative mortality for type A dissection was 20.6% (Table 4). This remained constant with no significant change over time (P ¼ .43). Survival at 1, 3, 5, and 10 years after successful surgical repair was 80.0%, 73.0%, 65.5%, and 47.5%, respectively. Although operative mortality remained constant, per-year total hospital mortality significantly decreased (from 65.1% in 2002 to 48.4% in 2012; P < .001). Also, patients with diagnosis of a type A dissection made at a cardiac surgery center had improved total hospital mortality relative to patients diagnosed at a noncardiac surgery center requiring transfer for surgery (45.5% vs 62.5%; P < .001), highlighting the influence of time delay on outcomes. 4
For type B dissections, sole medical management was the treatment in 83% of cases (3000 out of 3632), open surgery in 10% (370 out of 3632), and thoracic endovascular aortic repair (TEVAR) in 7% (262 out of 3632). Over the 12 years, all-cause nonoperative hospital mortality for type B dissections treated with medical management was 5.6%. There was no change over time (P ¼ .33). Operative mortality for type B dissection was 19.3% (Table 3). Similar to type A, operative mortality for type B dissections did not change across the study (P ¼ .68). The operative mortality for TEVAR was 14.8%. The first recorded TEVAR for type B dissection was not until 2006. TEVAR use increased from 4% in 2006 to 13% by March 2014. Still, present-day dissemination of TEVAR is certain to be underrepresented by this dataset.
The Journal of Thoracic and Cardiovascular Surgery c - 2018
McClure et al
Acquired
TABLE 3. Age- and sex-adjusted 12-year incidence proportions, hospital mortalities (nonoperative and operative), and all-cause 3-year mortalities for thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada (2002-2014) All
Type B
Thoracic aortic aneurysms
1.89 (1.67-2.13) 2.22 (1.89-2.61) 1.57 (1.30-1.89)
2.67 (2.41-2.96) 3.27 (2.86-3.73) 2.10 (1.79-2.47)
7.56 (7.11-8.04) 10.08 (9.34-10.87) 5.15 (4.64-5.71)
0.83 (0.66-1.04) 8.09 (7.23-9.05) 26.05 (22.99-29.53)
0.27 (0.18-0.40) 3.32 (2.78-3.95) 11.48 (9.51-13.87)
0.56 (0.42-0.74) 4.69 (4.05-5.44) 14.6 (12.3-17.2)
1.33 (1.11-1.60) 14.25 (13.1-15.5) 39.83 (36.0-44.1)
20.29 (17.99-22.80) –
20.58 (17.95-23.49) –
19.3 (14.8-24.8) 14.8 (10.3-20.5)
4.53 (3.88-5.27) 5.38 (3.01-8.87)
–
–
5.60 (4.64-6.69)
1.43 (0.95-2.06)
26.57 (25.10-28.10) 22.60 (20.87-24.43) 32.83 (30.22-35.60)
53.00 (49.76-56.39) 45.65 (41.80-49.75) 64.21 (58.59-70.23)
8.03 (7.00-9.17) 6.88 (5.68-8.26) 9.88 (8.06-12.0)
7.65 (7.01-8.33) 6.04 (5.34-6.81) 10.53 (9.29-11.89)
12.17 (9.83-14.91) 21.10 (19.19-23.15) 39.82 (36.88-42.94)
24.91 (19.30-31.63) 41.96 (37.78-46.48) 76.39 (70.21-82.97)
5.25 (3.43-7.70) 6.20 (4.89-7.75) 11.8 (9.73-14.2)
2.41 (1.51-3.65) 5.13 (4.40-5.94) 12.98 (11.6-14.5)
39.52 (34.18-45.45) 32.86 (26.52-40.26) 48.36 (39.47-58.65)
52.61 (43.65-62.86) 41.67 (31.39-54.23) 67.35 (52.09-85.68)
28.2 (22.2-35.3) 25.2 (17.8-34.5) 32.2 (22.7-44.4)
22.10 (19.2-25.4) 17.80 (14.6-21.5) 29.94 (24.3-36.5)
12.64 (6.31-22.62) 34.05 (26.96-42.44) 59.89 (49.03-72.43)
18.42 (7.41-37.95) 48.28 (36.47- 62.69) 76.32 (57.95-98.66)
8.16 (2.2-20.9) 19.8 (12.6-29.8) 47.5 (35.0-63.0)
4.92 (1.80-10.70) 13.65 (10.5-17.4) 40.87 (34.2-48.5)
Incidence proportion Overall Male Female Age (y) 50 51-74 75
4.58 (4.23-4.95) 5.50 (5.00-6.09) 3.70 (3.27-4.18)
Operative mortality* Surgery Thoracic endovascular aortic repair All-cause nonoperative mortality Medical therapy Total hospital mortalityy Overall Male Female Age (y) 50 51-74 75 All-cause 3-y mortalityz Overall Male Female Age (y) 50 51-74 75
Thoracic aortic dissection Type A
ACQ
Variable
Values for incidence proportion are presented as per 100,000 persons (95% confidence interval). Other values are presented as % (95% confidence interval). *Operative mortality defined as all death occurring during the hospitalization in which open surgery or endovascular procedure was performed, even if after 30 days. yTotal hospital mortality defined as all death occurring during the hospitalization. Combines all open surgery, endovascular, and medically managed mortalities. zPercentage of patients who have died within 3 years of the index date of diagnosis.
For thoracic aortic aneurysms, 53% (4940 out of 9392) underwent surgery, 44% (4129 out of 9392) underwent medical therapy, and 3% (323 out of 9392) underwent TEVAR. Of known descending thoracic aortic aneurysms, 35% (323 out of 924) were treated with TEVAR. In the open surgery cohort, 296 (6%) were thoracoabdominal aneurysms. Operative mortality for open surgery on thoracic aneurysms improved over the 12 years (5.5% to 4.5%; P ¼ .002). Yet, for the subset of thoracoabdominal aneurysm patients, operative mortality was 16.4% with no change over time (P ¼ .11). For medically treated thoracic aneurysms, 10% (441 out of 4278) went on to incur an aortic dissection. With respect to reintervention, 7.8% of surgically repaired aortic dissections and 11.3% of TEVARs required additional treatment to some portion of the thoracic aorta. Median time to re-intervention was substantially shortened with TEVAR (Table 5). Similarly, 4.7% of surgically
repaired thoracic aortic aneurysms and 14.2% of TEVARs required further treatment. Reintervention times were again shorter with TEVAR, although much closer to surgical results relative to aortic dissections (Table 5). Whether reintervention was due to failure at the initial treatment site or progression of disease elsewhere in the thoracic aorta could not be teased out from the dataset. Midterm outcomes mirrored that of the total hospital mortality. Over the 12 years, all-cause 3-year mortality decreased for thoracic aortic dissections (44.2% to 39.5%; P ¼ .04), type A dissections (75.0% to 52.6%; P <.0005) (Figure 2), and thoracic aneurysms (29.6% to 22.1%; P < .0001) (Figure 3). No change was observed for type B dissections (27.8% to 28.2%; P ¼ .299) (Figure 4). For aneurysms, 3-year mortality was significantly higher for women relative to men (Figure 3). For dissections, no clear 3-year mortality gender differences were identified (Figures 2 and 4).
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
5
Acquired
McClure et al
ACQ
FIGURE 1. Incidence proportions for thoracic aortic aneurysms and thoracic aortic dissections in Ontario, Canada (2002-2014). Number of persons at risk is equal to the population of Ontario for the year (12.2 million in 2002, 12.5 million in 2004, 12.7 million in 2006, 13 million in 2008, 13.3 million in 2010, 13.5 million in 2012, and 13.7 million in 2014).
For specialty involvement, cardiac surgeons performed 83% and 88% of open surgical procedures, respectively, for thoracic aortic dissections and thoracic aortic aneurysms. Vascular surgeons performed the clear majority of endovascular procedures, being the primary surgeons on record for 91% of such cases. DISCUSSION Quantifying the true incidence of thoracic aortic dissections and aneurysms is difficult. The acuity and lethality of aortic dissections results in out-of-hospital mortality data that are often imprecise. Moreover, the nebulous size criteria to designate an aorta as aneurysmal results in inconsistent classification and subsequent inexact estimates. Despite such challenges, efforts to characterize disease burden are essential to resource allocation and play an authoritative role in health policy. The present study, a large retrospective population-based hospital analysis of thoracic aortic disease in the most densely populated province in Canada, parallels findings of other contemporary studies.5-8,11 We report an overall incidence proportion of 4.6 per 100,000 for thoracic aortic dissections with an increasing trend over time. The Oxford Vascular Study, a contemporary prospective study of 93,000 persons in the United Kingdom,5 had similar results, reporting an incidence of 5.6 per 100,000 for 6
thoracoabdominal aortic dissections. Whereas our study reports a hospital-based incidence proportion, the Oxford Vascular Study included out-of-hospital deaths, which might explain the lower incidence in our cohort. We also report an incidence proportion of 7.6 per 100,000 for thoracic aortic aneurysms with an increasing trend over time. This too has been observed elsewhere.11 Whether these cumulative reports support a true increase in aortic disease or whether this is all attributable to increased imaging surveillance remains unclear. Not all reports have uniformly observed these increases. A recent retrospective study in Iceland12 notes a lower incidence for aortic dissections—2.53 per 100,000—in keeping with the proportions from seminal studies 3 decades prior. Despite an observed increase in disease, there were improved outcomes with significant decreases to total hospital mortality and improved 3-year survival for aortic dissections and aneurysms. Despite these positive trends, operative mortality for type A dissection was disconcerting at 20%. This is in line with worldwide registry results and highlights the global challenge with this lethal condition.7,13 Type A dissection total hospital mortality significantly decreased over our 12-year study, despite a constant operative mortality of 20%, which implies more patients over time are going for surgery. Time-garnered operative experience by senior surgeons or the creation of dedicated
The Journal of Thoracic and Cardiovascular Surgery c - 2018
Acquired
TABLE 4. Demographic characteristic comparison between operative and nonoperative patients with type A aortic dissections in Ontario, Canada (2002-2014) Characteristic
No surgery (n ¼ 1085)
Surgery (n ¼ 1204)
Age at diagnosis (y) 50 51-74 75
74.9 13.4 64 (5.9) 348 (32.1) 673 (62.0)
61.8 14.6 246 (20.4) 706 (58.6) 252 (20.9)
Sex Female Male
527 (48.6) 558 (51.4)
380 (31.6) 824 (68.4)
Diabetes
258 (23.8)
207 (17.2)
<.001
COPD
296 (27.3)
174 (14.5)
<.001
Hypertension
860 (79.3)
913 (75.8)
.05
Hyperlipidemia
255 (23.5)
357 (29.7)
<.001
Residence Urban Rural Missing
908 (83.7) 176 (16.2) <6
1041 (86.5) 161 (13.4) <6
0 (0.0)
30 (2.5)
65 (6.0) 87 (8.0) 104 (9.6) 96 (8.8) 93 (8.6) 88 (8.1) 105 (9.7) 89 (8.2) 92 (8.5) 83 (7.6) 75 (6.9) 108 (10.0) 0 (0.0)
44 (3.7) 61 (5.1) 71 (5.9) 65 (5.4) 77 (6.4) 67 (5.6) 96 (8.0) 129 (10.7) 124 (10.3) 152 (12.6) 135 (11.2) 149 (12.4) 34 (2.8)
Connective tissue disorder Year of diagnosis 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
P value* <.001 <.001
<.001
.142
<.001 <.001
Values are presented as mean standard deviation or n (%). COPD, Chronic obstructive pulmonary disease. *P values from c2 test for categorical variables and t test for continuous variables
aortic teams at select institutions facilitating successful surgery in more extreme patients may have played a role. Better point-of-care stabilization at hospital admission or more expeditious transfers to aortic surgical centers may also have been factors. Still, with only 53% of patients with type A dissections having undergone surgery, there may be opportunities to better these results. We observed total hospital mortality to significantly improve if transfer to another hospital was not required. More resource allocation for timely transfers may have cost benefits. Ontario’s transport of critical patients has evolved. Eleven centers perform the full gamut of open surgery and TEVAR for thoracic aneurysms and dissections. A province-wide call center facilitates emergent transfers, locating the nearest hospital with the necessary expertise. Aid with transfers is also provided. Yearly refinements to this service
may directly relate to the improved hospital outcomes we observed and warrants study. Others report similar statistics. Iceland also reported a 53% surgery rate for type A dissections.12 The Oxford Vascular Study found 58%.5 The International Registry for Acute Aortic Dissection (IRAD) is the exception, reporting 86% operative candidacy.7 IRAD is a skewed representation because 70% of the cohort is referred from elsewhere. With this, a percent of patients with extreme conditions die on transfer or cannot transfer because of instability. The loss of extremis cases alongside the elevated expertise at IRAD centers accounts for the variance. Palliative, nonoperative treatment of type A dissection occurred more often in women. Other areas of cardiovascular medicine have noted similar patterns, with less referral to advanced therapies for women or referral in a more morbid clinical state.14 Type A dissections are emergent events that eliminate the ambiguity of late referral, so it is of interest that gender remained a significant variable. Women also had more hospital mortality, and for the aneurysm cohort, worse 3-year outcomes. The use of TEVAR was captured and seen to increase over time but its true footprint on the thoracic aorta is underestimated. Hybrid procedures were also captured but in negligible numbers. The higher mortality for TEVAR in patients with type B dissections (14.8%) is likely reflective of a conservative philosophy to its use. It is postulated that TEVAR was reserved for patients with high-risk malperfusion. Broader use over time will improve this statistic. For patients with thoracic aneurysms, operative mortality improved, although little can be inferred from this. We could not reliably separate aneurysms based on ascending, arch, and descending segments. Blind to this distribution, the heterogeneity limits interpretation. Similarly, the extent of the thoracoabdominal aneurysms is unknown.15 Cardiac surgeons had an integral role in the thoracic aortic surgical management of patients but were less involved in TEVAR. Our methods captured only the primary operator on record. In situations where physicians from 2 specialties were present (ie, cardiac and vascular surgeons), capturing only the primary physician on record would result in a somewhat misrepresentation of involvement. Aortic teams utilizing expertise of multiple specialties are thus misrepresented by the analysis. Regardless, with only 8% of TEVAR cases having a cardiac surgeon as the primary, more integration of endovascular training into the cardiac armamentarium is required. Study Limitations As 1 of the largest published population-based studies to assess thoracic aortic disease, in more than 13.5 million persons with complete hospital records due to universal health care and legislated data collection representing
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
7
ACQ
McClure et al
Acquired
McClure et al
TABLE 5. Time to reintervention (>72 hours) in patients undergoing surgical or endovascular repair for thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada (2002-2014)
Treatment group Surgery
Category
n
All
120
> 3-30 d Endovascular
Thoracic aortic dissections Median time to Denominator % re-intervention (d) 1546
7.8
424
n 232
Thoracic aortic aneurysms Median time to Denominator % re-intervention (d) 4940
4.7
7
1546
0.5
8
34
4940
0.7
> 30 d
113
1546
7.3
478
198
4940
4.0
544
All
33
291
11.3
138
46
323
14.2
352
> 3-30 d
6
291
2.1
>30 d
27
291
9.3
ACQ
38.5% of the Canadian populace, our study is very robust and generalizable to the region. Still, there are limitations. Because it is a retrospective study using administrative data, a degree of misclassification is inherent. However, with our cross-referencing across successive databases, this is likely quite negligible. More relevant limitations are the inability to capture out-of-hospital deaths and an inherent shortcoming of the aortic dissection algorithm with respect to nonoperative type A dissection survivors. The accepted distribution for aortic dissection is 60% type A and 40% type B.16 Our distribution was 38% type A and 61% type B. This discrepancy is explainable. Recognizing that nearly 50% of type A dissection patients die before hospital admission,5 our distribution resembles
9.5 239
8.5
6
323
1.9
7
40
323
12.4
449
much more the expected distribution. With Ontario’s expansive landmass and the challenges of remote area access, out-of-hospital death is likely even higher. This remains an unknown factor and warrants study. Although this is the main reason for the distribution discrepancy, the aortic dissection algorithm has an inherent shortcoming that lends to misclassification of a small subset of patients as well. The algorithm assumes nonoperative aortic dissections that survive to hospital discharge are type B dissections. For the overwhelming majority this is correct. Still, a small subset of type A dissections not operated upon will indeed survive to hospital discharge. IRAD had 6.3% (19 out of 303) of type A dissections fit this subset.17 In our study, 2840 patients were classified as
FIGURE 2. Three-year all-cause mortality for type A aortic dissections in Ontario, Canada (2002-2014).
8
370
The Journal of Thoracic and Cardiovascular Surgery c - 2018
Acquired
ACQ
McClure et al
FIGURE 3. Three-year all-cause mortality for thoracic aortic aneurysms in Ontario, Canada (2002-2014).
having type B dissections for being treated medically and surviving to discharge. In a worst-case scenario sensitivity analysis, if 6.3% in fact were nonoperative type A dissection survivors, not type B dissection survivors,
179 patients (3%) would have been misclassified out of 5966 aortic dissections. Adjusting for 50% out-of-hospital death and 179 misclassified patients results in a distribution of 58% type A and 42% type B.
FIGURE 4. Three-year all-cause mortality for type B aortic dissections in Ontario, Canada (2002-2014).
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
9
Acquired
Of note, the shortcoming of the model does not influence the incidence rate for aortic dissections (4.6/100,000). And with 13.5 million persons, has very marginal effects on subset incidence rates for type A and B dissections (type A 2.0/100,000 vs 1.9/100,000, type B 2.6/100,000 vs 2.7/100,000). There is no influence on trend analyses or any operative statistics. We used a conservative method to assess the incidence of thoracic aortic aneurysms. Thoracic aneurysms that ruptured or dissected were analyzed as dissections only. This approach minimized misclassification and maintains data accuracy but at the cost of underestimating the true incidence. Medication therapy following current guidelines was presumably adhered to for medically treated patients but any variance across treating physicians to anti-impulse strategies are not accounted for. Finally, the vast majority of aortic dissections would have been acute, but a small subset of chronic dissections is certain to be present. Codes cannot distinguish between acute and chronic events. Because acute and chronic dissections differ in operative risk, this unavoidable inclusion skews the result slightly relative to a purely acute cohort.
ACQ
CONCLUSIONS The incidence of thoracic aortic dissections and aneurysms has increased, but hospital mortality and 3-year outcomes have improved. Men incur more disease, but women have higher mortality. Surgery is primarily referred to cardiac surgeons. Vascular surgeons perform most TEVARs. Resource allocation for expeditious transfer to aortic centers may have life-saving implications. Webcast You can watch a Webcast of this AATS meeting presentation by going to: https://aats.blob.core.windows.net/ media/17AM/2017-05-02/RM311/05-02-17_Room311_ 1414_McClure.mp4.
Conflict of Interest Statement Authors have nothing to disclose with regard to commercial support. This project was approved by the Institute for Clinical Evaluative Sciences (ICES). ICES is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. The opinions, results, and conclusions reported in this paper are those of the authors and are independent from the funding sources.
10
McClure et al
No endorsement by ICES or the Ontario Ministry of Health and Long-Term Care is intended or should be inferred. Dataset creation and data analysis was performed at ICES Queens by Ms Katherine Lajkosz. Parts of this material are based on data and information compiled and provided by CIHI. However, the analyses, conclusions, opinions, and statements expressed herein are those of the author, and not necessarily those of CIHI. Drs McClure and Payne are members of the Canadian Thoracic Aortic Collaborative (CTAC). CTAC is a group of cardiovascular clinicians across Canada with a specific interest and expertise in the management of thoracic aortic disease.
References 1. Bickerstaff LK, Pairolero PC, Hollier LH, Melton LJ, Van Peenen HJ, Cherry KJ, et al. Thoracic aortic aneurysms: a population-based study. Surgery. 1982;92:1103-8. 2. Svensjo S, Bengtsson H, Berggvist D. Thoracic and thoracoabdominal aortic aneurysm dissection: an investigation based on autopsy. Br J Surg. 1996;83:68-71. 3. Clouse WD, Hallett JW Jr, Schaff HV, Gayari MM, Ilstrup DM, Melton LJ III. Improved prognosis of thoracic aortic aneurysms: a population-based study. JAMA. 1998;280:1926-9. 4. Meszaros I, Morocz J, Szlavi J, Schmidt J, Tornoci L, Nagy L, et al. Epidemiology and clinicopathology of aortic dissection. Chest. 2000;117:1271-8. 5. Howard DPJ, Banerjee A, Fairhead JF, Perkins J, Silver LE, Rothwell PM. Population-based study of incidence and outcome of acute aortic dissection and premorbid risk factor control. 10-year results from the Oxford vascular study. Circulation. 2013;127:2031-7. 6. Mody PS, Wang Y, Geirsson A, Kim N, Desai MM, Gupta A, et al. Trends in aortic dissection hospitalizations, interventions, and outcomes among medicare beneficiaries in United States, 2002-2011. Circ Cardiovasc Qual Outcome. 2014;7:920-8. 7. Pape LA, Awais M, Woznicki EM, Suzuki T, Trimarchi S, Evangelista A, et al. Presentation, diagnosis, and outcomes of acute aortic dissection. 17-year trends from the international registry of acute aortic dissection. J Am Coll Cardiol. 2015;66:350-8. 8. Zimmerman KP, Oderich G, Pochettino A, Hanson KT, Habermann E, Bower TC, et al. Improving mortality trends for hospitalization of aortic dissection in the national inpatient sample. J Vasc Surgery. 2016;64:606-15. 9. Lewis S. A system in name only – access, variation and reform in Canada’s provinces. N Engl J Med. 2015;372:497-500. 10. Wodchis WP, Austin PC, Henry DA. A 3-year study of high-cost users of health care. Can Med Assoc J. 2016;188:182-8. 11. Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection. Increasing prevalence and improved outcomes reported in a nationwide population-based study of more than 14000 cases from 1987 to 2002. Circulation. 2006;114:2611-8. 12. Melvinsdottir IH, Lund SH, Agnarsson BA, Sigvaldason K, Gudbjartsson T, Geirsson A. The incidence and mortality of acute thoracic aortic dissection: results from a whole nation study. Eur J Cardiothorac Surg. 2016;50:1111-7. 13. Conzelmann LO, Weigang E, Mehlhorn U, Abugameh A, Hoffmann I, Blettner M, et al. Mortality in patients with acute aortic dissection type A: analysis of pre- and intraoperative risk factors from the German registry for acute aortic dissection type A (GERAADA). Eur J Cardiothorac Surg. 2016;49:e44-52. 14. Bogaev RC. Gender disparities across the spectrum of advanced cardiac therapies: real or imagined? Curr Cardiol Rep. 2016;11:108-18. 15. Coselli JS, LeMaire SA, Preventza O, de la Cruz KI, Cooley DA, Price MD, et al. Outcomes of 3309 thoracoabdominal aortic aneurysm repairs. J Thorac Cardiovasc Surg. 2016;151:1323-38. 16. Mery CM, Reece TB, Kron IL. Aortic dissection. In: Cohn LH, ed. Cardiac Surgery in the Adult. 4th ed. New York: McGraw-Hill; 2012:999. 17. Tsai TT, Evangelista A, Nienaber CA, Trimarchi S, Sechtem U, Fattori R, et al. Long-term survival in patients presenting with type A acute aortic dissection: insight from the international registry of acute aortic dissection (IRAD). Circulation. 2006;114(1 Suppl):I350-6.
Key Words: aortic dissection, aortic aneurysm, epidemiology, database
The Journal of Thoracic and Cardiovascular Surgery c - 2018
Acquired
Discussion Dr John Elefteriades (New Haven, Conn). Congratulations on your thoughtful, well-conducted investigation in which you took important advantages of the centralized health care system in Canada to make key observations about incidence and treatment of thoracic aortic disease. I have 3 brief questions. Is the increased incidence of thoracic aortic diseases that you found an artifact of increased imaging in recent years or a bona fide increase in the disease itself? Number 2: If you feel it may be a bona fide increase in the disease itself, what is the reason? How can biology or evolution change so quickly as to be detected in a decade? Number 3: How do you suggest that we modify our health care delivery or practice patterns based on your findings? R. Scott McClure. Thank you very much for the questions. We are honored to have you as our discussant, being a leader in this area. To answer your first 2 questions, my personal opinion is that the increased incidence of disease has much more to do with a heightened awareness of the medical condition by the medical community rather
than a true evolutionary increase in disease. To your point, biologic change over a 12-year time frame would be rather unlikely. Unknown environmental or lifestyle factors may be at play but more astute diagnosis of that which was always there seems much more plausible. Improved clinical intuition and exponential use of imaging studies are most likely responsible for the data. With regard to your last question, I think this study highlights importance of allocating more resources to the treatment of thoracic aortic pathology. Because of the stark contrast between the benign aneurysm and the catastrophic aortic dissection, determining the proper amount of funding is difficult. But what we do know is that the incidence of disease for both aneurysms and dissections is higher in the population than previously thought. We need to acknowledge that and adjust resources accordingly. More timely transfer of patients is but 1 area where this may prove beneficial. One interesting statistic that we teased out that I didn’t talk about because of time constraints, is that we found a rather significant difference in outcomes in those patients with a type A dissection diagnosed at a hospital without cardiac surgery on site versus those patients diagnosed at a hospital with cardiac surgery services. So 1 thing that we could do is improve our transfer of patients from point A to point B if and when that diagnosis has been achieved. I think that would go a long way to helping our outcomes.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
11
ACQ
McClure et al
Acquired
McClure et al
TABLE E1. International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada (ICD-10-CA) diagnosis codes ICD-10-CA description Thoracic aortic dissection Dissection of aorta (any part) Injury of thoracic aorta Thoracic aortic aneurysm, ruptured Thoracoabdominal aortic aneurysm, ruptured Aneurysm and dissection of unspecified site Thoracic aortic aneurysm Aneurysm of aorta in disease classified elsewhere (Syphilitic) Thoracic aortic aneurysm, without mention of rupture Thoracoabdominal aortic aneurysm, without mention of rupture Aortic arch syndrome (Takayasu)
ICD-10-CA code I710 S250 I71 I715 I729 I790 I712 I716 M314
ACQ 11.e1
The Journal of Thoracic and Cardiovascular Surgery c - 2018
McClure et al
Acquired
TABLE E2. Canadian Classification of Health Interventions (CCI) surgical and endovascular procedures codes Type of aortic dissection/ aneurysm
CCI code
Description
1HV90WJCFN
Excision total with reconstruction, aortic valve, replacement of valve, aortic root, and ascending aorta (eg, Bentall) using open approach with mechanical valve and synthetic aorta
Surgery
Ascending
1HV90WJXXA
Excision total with reconstruction, aortic valve, replacement of valve, aortic root and ascending aorta (eg, Bentall) using open approach with autograft tissue valve and autograft aorta
Surgery
Ascending
1HV90WJXXD
Excision total with reconstruction, aortic valve, replacement of valve, aortic root and ascending aorta (eg, Bentall) using open approach with xenograft tissue valve (eg, bovine or porcine tissue) and synthetic aorta
Surgery
Ascending
1HV90WJXXK
Excision total with reconstruction, aortic valve, replacement of valve, aortic root and ascending aorta (eg, Bentall) using open approach with homograft tissue valve and homograft aorta
Surgery
Ascending
1HV90WJXXL
Excision total with reconstruction, aortic valve, replacement of valve, aortic root and ascending aorta (eg, Bentall) using open approach with xenograft tissue valve (eg, bovine or porcine tissue) and xenograft aorta
Surgery
Ascending
1IA80GQNRN
Repair, ascending aorta using percutaneous transluminal approach and (endovascular) stent with synthetic tissue (eg, stent graft)
Endovascular
Ascending
1IA80LA
Repair, ascending aorta using open approach without tissue
Surgery
Ascending
1IA80LAXXA
Repair, ascending aorta using open approach with autograft (eg, pericardial patch, omental patch)
Surgery
Ascending
1IA80LAXXK
Repair, ascending aorta using open approach with homograft (eg, arterial homograft)
Surgery
Ascending
1IA80LAXXL
Repair, ascending aorta using open approach with xenograft
Surgery
Ascending
1IA80LAXXN
Repair, ascending aorta using open approach with synthetic tissue (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Ascending
1IA82LA
Reattachment, ascending aorta using open approach
Surgery
Ascending
1IA87LA
Excision partial, ascending aorta using open approach without tissue (eg, anastomosis)
Surgery
Ascending
1IB80GQNRN
Repair, arch of aorta using percutaneous transluminal approach and (endovascular) stent with synthetic tissue (eg, stent graft)
Endovascular
Arch
1IB80LA
Repair, arch of aorta using open approach without tissue
Surgery
Arch
1IB80LAXXA
Repair, arch of aorta using open approach with autograft (eg, pericardial patch)
Surgery
Arch
1IB80LAXXK
Repair, arch of aorta using open approach with homograft (eg, arterial homograft)
Surgery
Arch
1IB80LAXXL
Repair, arch of aorta using open approach with xenograft
Surgery
Arch
1IB80LAXXN
Repair, arch of aorta using open approach with synthetic material (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Arch
1IC50GQOA
Dilation, thoracic (descending) aorta using percutaneous transluminal arterial approach balloon dilator with (endovascular) stent (insertion)
Endovascular
Descending
1IC50GSBD
Dilation, thoracic (descending) aorta using percutaneous transluminal approach with placement/implant of stent and mechanical balloon dilator
Endovascular
Descending
1IC50LANR
Dilation, thoracic (descending) aorta using endovascular stent (eg, ⌠Z O stent] open approach, such as retroperitoneal)
Endovascular
Descending (Continued)
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
11.e2
ACQ
Type of intervention
Acquired
McClure et al
TABLE E2. Continued
CCI code
Description
Type of intervention
Type of aortic dissection/ aneurysm
ACQ
1IC55LANRA
Removal of device, thoracic (descending) aorta open approach surgical repair of defect using autograft of endovascular stent
Surgical reintervention
Descending
1IC55LANRN
Removal of device, thoracic (descending) aorta open approach surgical repair of defect using synthetic material of endovascular stent
Surgical reintervention
Descending
1IC55LANRQ
Removal of device, thoracic (descending) aorta open approach surgical repair of defect using combined sources of tissue of endovascular stent
Surgical reintervention
Descending
1IC80GQNRN
Repair, thoracic (descending) aorta using percutaneous transluminal approach and (endovascular) stent with synthetic tissue (eg, stent graft)
Endovascular
Descending
1IC80LA
Repair, thoracic (descending) aorta using open approach without tissue
Surgery
Descending
1IC80LAXXA
Repair, thoracic (descending) aorta using open approach with autograft (eg, pericardial patch, subclavian flap)
Surgery
Descending
1IC80LAXXK
Repair, thoracic (descending) aorta using open approach with homograft (eg, arterial homograft)
Surgery
Descending
1IC80LAXXL
Repair, thoracic (descending) aorta using open approach with xenograft
Surgery
Descending
1IC80LAXXN
Repair, thoracic (descending) aorta using open approach with synthetic material (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Descending
1IC80LAXXQ
Repair, thoracic (descending) aorta using open approach with combined sources of tissue
Surgery
Descending
1IC80WC
Repair, thoracic (descending) aorta using open approach with fenestration (aneurysm) technique (eg, re-entry operation)
Surgery
Descending
1IC82LA
Reattachment, thoracic (descending) aorta using open approach without tissue
Surgery
Descending
1IC82LAXXA
Reattachment, thoracic (descending) aorta using open approach with autograft
Surgery
Descending
1IC82LAXXN
Reattachment, thoracic (descending) aorta using open approach with synthetic material
Surgery
Descending
1IC82LAXXQ
Reattachment, thoracic (descending) aorta using open approach with combined sources of tissue
Surgery
Descending
1IC87LA
Excision partial, thoracic (descending) aorta using open approach with simple end-to-end anastomosis
Surgery
Descending
1IC87LAXXN
Excision partial, thoracic (descending) aorta using open approach with synthetic material (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Descending
1IC87LAXXQ
Excision partial, thoracic (descending) aorta using open approach with combined sources of tissue
Surgery
Descending
1IC87TQ
Excision partial, thoracic (descending) aorta using open approach with extended end-to-end anastomosis
Surgery
Descending
1ID80QFXXK
Repair, aorta NEC open thoracoabdominal approach using homograft (eg, arterial homograft)
Surgery
Descending
1ID80QFXXN
Repair, aorta NEC open thoracoabdominal approach using synthetic material (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Descending
1ID80QFXXQ
Repair, aorta NEC open thoracoabdominal approach using combined sources of tissue
Surgery
Descending
1ID87QFXXK
Excision partial, aorta NEC using homograft thoracoabdominal approach
Surgery
Descending (Continued)
11.e3
The Journal of Thoracic and Cardiovascular Surgery c - 2018
McClure et al
Acquired
TABLE E2. Continued
CCI code
Description
Type of intervention
Type of aortic dissection/ aneurysm
1ID87QFXXN
Excision partial, aorta NEC using synthetic material thoracoabdominal approach
Surgery
Descending
1ID87QFXXQ
Excision partial, aorta NEC using combined sources of tissue thoracoabdominal approach
Surgery
Descending
ACQ
CCI, Canadian Classification of Health Interventions; NEC, not elsewhere classified.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
11.e4
Acquired
000
Epidemiology and management of thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada: A population-based study R. Scott McClure, MD, Susan B. Brogly, PhD, Katherine Lajkosz, MS, Darrin Payne, MD, Stephen F. Hall, MD, and Ana P. Johnson, PhD, Calgary, Alberta, and Kingston, Ontario, Canada The incidence of thoracic aortic aneurysms and dissections has increased but hospital mortality and late outcomes have improved. Men incur more thoracic aortic disease but women have higher hospital mortality.
ACQ The Journal of Thoracic and Cardiovascular Surgery c - 2018
McClure et al
Acquired
Epidemiology and management of thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada: A population-based study R. Scott McClure, MD,a Susan B. Brogly, PhD,b,c Katherine Lajkosz, MS,c Darrin Payne, MD,b Stephen F. Hall, MD,c and Ana P. Johnson, PhDc ABSTRACT Objectives: To determine hospital incidence, mortality, and management for thoracic aortic dissections and aneurysms. Methods: A population-based retrospective cohort study of anonymously linked data for residents of Ontario, Canada, was carried out. Incident cases of thoracic aortic dissections and aneurysms were identified between 2002 and 2014. Treatment and mortality trends were assessed.
Conclusions: The incidence of thoracic aortic dissections and aneurysms increased over time but all-cause hospital and late outcomes improved. Gender differences exist. Men incur more disease but women have higher hospital mortality. Surgery was primarily referred to cardiac surgeons. Endovascular therapy was primarily referred to vascular surgeons. (J Thorac Cardiovasc Surg 2018;-:1-11)
From the aDivision of Cardiac Surgery, Libin Cardiovascular Institute, Foothills Medical Center, University of Calgary, Calgary, Alberta, Canada; and bDepartment of Surgery, Kingston General Hospital, and cInstitute for Clinical and Evaluative Sciences, Queen’s University, Kingston, Ontario, Canada. Supported by the Establishment Fund, Queen’s University, Kingston, Ontario, Canada, and an internal grant from the Departmental, Development & Innovation Fund, Queen’s University, Kingston, Ontario, Canada. Parts of this work are based on data and information compiled and provided by the Canadian Institute for Health Information. However, the analyses, conclusions, opinions, and statements expressed herein are those of the authors, and not necessarily those of the Canadian Institute for Health Information.
Incidence proportions for thoracic aortic dissections and aneurysms in Ontario, Canada. Central Message The incidence of thoracic aortic aneurysms and dissections has increased, but hospital mortality and late outcomes have improved. Men incur more thoracic aortic disease but women have higher hospital mortality.
ACQ
Results: There were 5966 aortic dissections (Type A n ¼ 2289 [38%] and Type B n ¼ 3632 [61%]). Overall incidence proportion for aortic dissections was 4.6 per 100,000. There were 9392 thoracic aortic aneurysms with an overall incidence proportion of 7.6 per 100,000. The incidence for both dissections and aneurysms significantly increased over the 12-year study. Only 53% (1204 out of 2289) of Type A dissections underwent surgery. Type B dissection treatment was 83% (3000 out of 3632) medical, 10% (370 out of 3632) surgery, and 7% (262 out of 3632) endovascular. Thoracic aortic aneurysm treatment was 53% (4940 out of 9392) surgery, 44% (4129 out of 9392) medical, and 3% (323 out of 9392) endovascular. Thirty-five percent of known descending thoracic aortic aneurysms (323 out of 924) received a stent graft. Cardiac surgeons performed 87% of the open surgical repairs. Vascular surgeons performed 91% of the endovascular procedures. All-cause 3-year mortality significantly decreased for both aortic dissections (44% to 40%) and aneurysms (30% to 22%). All-cause hospital mortality also decreased. Women had worse outcomes than men.
Perspective The incidence of disease is integral to resource allocation for cost-efficient care. This study demonstrates the incidence of thoracic aortic aneurysms and dissections have increased yet hospital and late outcomes have improved. Men incur more aortic disease yet women have higher hospital mortality. Surgical care is primarily performed by cardiac surgeons. Vascular surgeons perform most endovascular procedures.
See Editorial Commentary page XXX.
Drs McClure and Payne are members of the Canadian Thoracic Aortic Collaborative, a group of cardiovascular clinicians across Canada with a specific interest and expertise in the management of thoracic aortic disease. Read at the 97th Annual Meeting of The American Association for Thoracic Surgery, Boston, Massachusetts, April 29-May 3, 2017. Received for publication May 10, 2017; revisions received Oct 15, 2017; accepted for publication Nov 13, 2017. Address for reprints: R. Scott McClure, MD, Division of Cardiac Surgery, Libin Cardiovascular Institute, University of Calgary, 1403 29th St, NW, Rm C803, Calgary, Alberta, Canada T2N 2T9 (E-mail: [email protected]). 0022-5223/$36.00 Copyright Ó 2018 by The American Association for Thoracic Surgery https://doi.org/10.1016/j.jtcvs.2017.11.105
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
1
Acquired
Abbreviations and Acronyms CCI ¼ Canadian Classification of Health Interventions ICD-10-CA ¼ International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada IRAD ¼ International Registry for Acute Aortic Dissection NACRS ¼ National Ambulatory Care Reporting System TEVAR ¼ thoracic endovascular aortic repair
Scanning this QR code will take you to an appendix for this paper. To view the AATS Annual Meeting Webcast, see the URL next to the webcast thumbnail.
ACQ
Thoracic aortic dissections and thoracic aortic aneurysms are highly resource intensive intermixed disease processes associated with significant morbidity and mortality. Such conditions put profound strain on health care systems. With a clinical presentation that spans the indolent and likely benign (small aortic aneurysm) to that of emergent and life threatening (aortic dissection), proper allocation of health service resources to ensure quality patient care becomes a challenge. Moreover, data to characterize the true burden of thoracic aortic disease are sparse.1-8 Characterizing this burden is critical to ensuring quality care remains cost-efficient. In a single-payer health care system, such as present in Canada, this is prudent to financial sustainability.9,10 We performed a populationbased study to characterize the contemporary burden of disease and describe recent treatment paradigms. MATERIALS AND METHODS Study Design We designed a retrospective cohort study using anonymously linked health information for persons residing in the province of Ontario, Canada. The primary outcomes of interest were the overall and annual calendar year incidence of thoracic aortic dissections and thoracic aortic aneurysms. The cohort encompassed more than 13.5 million (12.2 million in 2002, increased to 13.7 million in 2014) persons of ethnic diversity with the majority being white. The data were housed at the Institute for Clinical Evaluative Sciences. The Institute for Clinical Evaluative Sciences links de-identified, patient-level population-based health information across several administrative databases in Ontario using a unique encrypted health number identifier. A systematic algorithm was constructed to facilitate a distinction between patients who incurred a Stanford type A or type B aortic dissection. Research ethics approval
2
McClure et al
was obtained from the research ethics committees at Queen’s University Kingston, Ontario, Canada, and Sunnybrook Hospital, Toronto, Ontario, Canada. Secondary outcomes included a descriptive characterization of thoracic aortic dissection and aneurysm treatments (medical, endovascular, or surgery), an assessment of surgical and hospital mortality rates, reintervention rates, and specialty involvement (cardiac or vascular surgeons).
Patient Selection All persons residing in the province of Ontario, Canada, between April 1, 2002, and March 31, 2014, were included in the study. Cases were acquired by presentation with the condition of interest to the 231 hospital sites or multitude of ambulatory clinics within the Ontario health care system. A thoracic aortic dissection or thoracic aortic aneurysm was an incident case if no prior diagnosis for that entity had been identified during the previous 5 years. The size criterion to determine whether an aorta was aneurysmal was at the treating physician’s discretion. Marfan, Ehlers-Danlos, and Turner syndrome, as well as connective tissue disorders of unspecified nature, were identified and tabulated. For each incident case, if treatment via endovascular therapy or surgery could not be identified, optimal medical therapy was the presumed treatment. Any reintervention after an initial endovascular or surgical procedure were recorded. Reintervention was defined as any endovascular stenting or surgical therapy > 72 hours after the initial procedure. The specialist on record as the primary operator for each case was captured to delineate cardiac and vascular surgery involvement. All-cause nonoperative hospital mortality (used for patients treated solely with medical therapy during hospital admission. All-cause nonoperative hospital mortality was defined as death occurring in a patient who did not have a surgical or endovascular procedure where the patient’s in-hospital record indicated a discharge status of dead, regardless of cause and length of stay) and operative mortality were assessed (used for patients treated with open surgery or endovascular therapy. Operative mortality was defined as all deaths occurring during the hospitalization in which the operation was performed, even if after 30 days). Follow-up ended on March 31, 2015, or the date of a patient’s death, whichever came first.
Data Collection The compiled dataset for the cohort of interest was the product of several multilinked administrative databases and networks. The Registered Persons Database, the Canadian Institute for Health Information Discharge Abstract Database and Same Day Surgery Database, the National Ambulatory Care Reporting System (NACRS), the Ontario Health Insurance Plan Database, the Ontario Drug Benefit Database, the Office of the Registrar General-Deaths database, and Statistics Canada. The Registered Persons Database collects vital statistics on all permanent residents of Ontario. The Canadian Institute for Health Information Discharge Abstract Database and Same Day Surgery Database record all hospital admissions and discharges along with information on diagnoses and any procedures performed. The NACRS collects data on all ambulatory visits, including outpatient clinics and visits to an emergency department. Diagnoses at presentation are coded using the International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada (ICD-10-CA). Hospital procedures are coded using the Canadian Classification of Health Interventions (CCI) coding system. For the cardiovascular system and pertinent to the study, CCI codes are discernably specific, partitioning surgical procedures based on the segment of aorta involved (ascending aorta, aortic arch, or descending thoracic aorta). Coding also partitions thoracic endovascular stenting procedures based on anatomic positioning within the abdominal or descending thoracic aorta. To cross-reference Canadian Institute for Health Information and NARCS data, the Ontario Health Insurance Plan Database captures fee-for-service billing claims for inpatient and outpatient services. The Ontario Drug Benefit Database records patient
The Journal of Thoracic and Cardiovascular Surgery c - 2018
Acquired
TABLE 1. Aortic dissection classification algorithm Aortic dissection classification algorithm (ICD-10-CA þ CCI codes) If diagnosis of an aortic dissection from ICD-10-CA codes AND Have 1 associated procedural code (CCI code) (i) CCI code ¼ surgery or TEVAR for AscA then type A dissection (ii) CCI code ¼ surgery or TEVAR for AArch then type A dissection (iii) CCI code ¼ surgery or TEVAR for DTA then type B dissection If diagnosis of an aortic dissection from ICD-10-CA codes AND Have 2 or 3 associated procedure codes (CCI codes) (on same day) (i) CCI codes ¼ surgery or TEVAR for AscA þ AArch then type A dissection (ii) CCI codes ¼ surgery or TEVAR for AscA þ DTA then type A dissection (iii) CCI codes ¼ surgery or TEVAR for AscAþ AArch þ DTA then type A dissection If diagnosis of an aortic dissection from ICD-10-CA codes AND Have no associated procedure code (CCI code) ¼ no intervention AND Patient alive at discharge (i) Then patient received medical management therefore type B dissection If diagnosis of an aortic dissection from ICD-10-CA codes AND Have no associated procedure code (CCI code) ¼ no intervention AND Patient died during hospital admission (i) Then assess ORGD for primary cause of death to determine whether type A or B dissection ICD-10-CA, International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada; CCI, Canadian Classification of Health Interventions; TEVAR, thoracic endovascular aortic repair; AscA, ascending aorta; AArch, aortic arch; DTA, descending thoracic aorta; ORGD, Office of the Registrar General - Death Database.
medications. The Office of the Registrar General - Deaths database records in-hospital deaths and cause of death. The systematic algorithm to distinguish type A from type B aortic dissections is shown in Table 1. The algorithm is predicated upon the CCI procedural codes ability to segregate procedures performed on the aorta into ascending, arch, and descending segments. The algorithm exploits the distinctly different treatment most often applied to treat acute type A (surgery) and acute type B (medical therapy or endovascular stenting) aortic dissections. The ICD-10-CA codes to determine the diagnosis of a thoracic aortic dissection or thoracic aortic aneurysm are in Table E1. The CCI codes for the surgical and endovascular procedures performed upon receiving a diagnosis of a thoracic aortic dissection or aneurysm are listed in Table E2. A conservative estimate was derived for thoracic aortic aneurysms, as ICD-10-CA codes I71 (thoracic aortic aneurysm, ruptured) and I715 (thoracoabdominal aortic aneurysm, ruptured) were categorized solely as thoracic aortic dissections. We did not double count such patients to be included in the thoracic aortic aneurysm distribution. Two attending cardiac surgeons independently reviewed all diagnostic and procedure codes.
Statistical Analysis Descriptive data and treatment strategies (medical, endovascular, and surgical) were assessed using analysis of variance and c2 tests for age
and gender differences. Poisson regression models were used to assess patient and hospital mortality trends over time. The incidence proportion was all new cases divided by the population of Ontario reported per 100,000. Incidence proportion was tabulated per year and over the entire population. The population of Ontario was defined as living Ontario residents (as of December 31 of the given year) who had contact with the health care system during the previous 7 years, excluding prevalent cases. All statistical tests were 2-sided. A P value of < .05 was deemed statistically significant. For categorical data resulting in cells<6, absolute numbers are not reported to ensure patient anonymity. All analyses were performed using SAS version 9.4 (SAS Institute Inc, Cary, NC).
RESULTS There were 5966 thoracic aortic dissections (type A n ¼ 2289 [38%] and type B n ¼ 3632 [61%], unknown type n ¼ 45 [1%]). The mean age was 66 years. The cohort was predominantly male (61%) (Table 2). The incidence proportion for all thoracic aortic dissections across the entire population (2002-2014) was 4.6 per 100,000 (95% confidence interval [CI], 4.2/100,000-5.0/100,000). For type A aortic dissections it was 1.9 out of 100,000 (95% CI, 1.7/100,000-2.1/100,000) and for type B aortic dissections it was 2.7 out of 100,000 (95% CI, 2.4/100,000-3.0/100,000) (Table 3). Persons with type A dissections were older, more likely to have hypertension, and more often had a connective tissue disorder relative to persons with type B dissections (Table 2). There were 9392 thoracic aortic aneurysms. The mean age was 67 years and the cohort was again, predominantly male (64%) (Table 2). The incidence proportion for thoracic aortic aneurysms across the entire population (2002-2014) was 7.6 per 100,000 (95% CI, 7.1/100,000-8.0/100,000). Male gender (Figure 1) and increasing age significantly increased the incidence of disease for both aortic dissections and aneurysms (Table 3). Despite more disease in men, total hospital mortality (operative mortality þ all-cause nonoperative hospital mortality) was significantly higher in women relative to men across the 12-year study for type A dissection and thoracic aneurysms. Type B dissections showed a nonsignificant trend (Table 3). For both cohorts, the presence of connective tissue disorders was rare (Table 2). The per-year incidence proportion for thoracic aortic dissections and thoracic aortic aneurysms increased over the 12-year study (aortic dissections from 2.7 to 4.6 per 100,000; P <.0001 and aortic aneurysms from 3.5 to 7.6 per 100,000; P < .0001), with an average increase of 1.8% and 4.7%, respectively (Figure 1). The increased incidence in aortic dissections was driven solely by type A dissections, with an average increase of 4%. No increase in type B dissections was identified. Only 53% of patients (1204 out of 2289) diagnosed with a type A dissection underwent surgery. Relative to patients undergoing surgery, patients with nonoperative treatment
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
3
ACQ
McClure et al
Acquired
McClure et al
TABLE 2. Demographic characteristics of patients diagnosed with a thoracic aortic dissection or thoracic aortic aneurysm in Ontario, Canada (2002-2014)
Characteristic
All (n ¼ 5966)
Percentage
Thoracic aortic dissection Type A Type B (n ¼ 2289) (n ¼ 3632) P value* 38.4
60.9
Age at diagnosis (y) 50 51-74 75
66.0 16.7 1031 (17.3) 2755 (46.2) 2180 (36.5)
68.0 15.2 310 (13.5) 1054 (46.0) 925 (40.4)
64.7 17.3 718 (19.8) 1684 (46.4) 1230 (33.9)
Sex Female Male
2339 (39.2) 3627 (60.8)
907 (39.6) 1382 (60.4)
1411 (38.8) 2221 (61.2)
Diabetes
1244 (20.9)
465 (20.3)
768 (21.1)
Unknown (n ¼ 45)
Thoracic aortic aneurysm (n ¼ 9392)
0.8 <.001 <.001
73.6 18.7 <6 17 (37.8) 25 (55.6)
67.3 14.8 1197 (12.7) 4770 (50.8) 3425 (36.5)
21 (46.7) 24 (53.3)
3373 (35.9) 6019 (64.1)
11 (24.4)
2160 (23.0)
.552
.443
COPD
1195 (20.0)
470 (20.5)
713 (19.6)
.398
12 (26.7)
2047 (21.8)
Hypertension
4489 (75.2)
1773 (77.5)
2682 (73.8)
.002
34 (75.6)
7299 (77.7)
Hyperlipidemia
1639 (27.5)
612 (26.7)
1016 (28.0)
.299
11 (24.4)
3367 (35.8)
58 (1.0)
30 (1.3)
28 (0.8)
.040
0 (0.0)
140 (1.5)
39 (86.7) <6 <6
8064 (85.9) 1318 (14.0) 10 (0.1)
<6 0 (0.0) <6 0 (0.0) 0 (0.0) <6 <6 <6 <6 0 (0.0) 0 (0.0) <6 34 (75.6)
432 (4.6) 567 (6.0) 591 (6.3) 697 (7.4) 737 (7.8) 697 (7.4) 793 (8.4) 863 (9.2) 857 (9.1) 943 (10.0) 932 (9.9) 1030 (11.0) 253 (2.7)
Connective tissue disorder
ACQ
Residence Urban Rural Missing Year of diagnosis 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
.570 5045 (84.6) 913 (15.3) 8 (0.1)
1949 (85.1) 337 (14.7) <6
3057 (84.2) 571 (15.7) <6
328 (5.5) 414 (6.9) 491 (8.2) 464 (7.8) 473 (7.9) 452 (7.6) 487 (8.2) 509 (8.5) 518 (8.7) 516 (8.6) 532 (8.9) 624 (10.5) 158 (2.6)
109 (4.8) 148 (6.5) 175 (7.6) 161 (7.0) 170 (7.4) 155 (6.8) 201 (8.8) 218 (9.5) 216 (9.4) 235 (10.3) 210 (9.2) 257 (11.2) 34 (1.5)
217 (6.0) 266 (7.3) 315 (8.7) 303 (8.3) 303 (8.3) 296 (8.1) 285 (7.8) 289 (8.0) 301 (8.3) 281 (7.7) 322 (8.9) 364 (10.0) 90 (2.5)
<.001
Values are presented as mean standard deviation or n (%). COPD, Chronic obstructive pulmonary disease. *P values are from c2 test for categorical variables or t test for continuous variables. A look-back period of 5 years before index date was used to look at risk factors (eg, diabetes, COPD, hypertension, and hyperlipidemia).
for type A dissections were statistically older, female, and had more comorbid conditions (Table 4). For the 12-year study, operative mortality for type A dissection was 20.6% (Table 4). This remained constant with no significant change over time (P ¼ .43). Survival at 1, 3, 5, and 10 years after successful surgical repair was 80.0%, 73.0%, 65.5%, and 47.5%, respectively. Although operative mortality remained constant, per-year total hospital mortality significantly decreased (from 65.1% in 2002 to 48.4% in 2012; P < .001). Also, patients with diagnosis of a type A dissection made at a cardiac surgery center had improved total hospital mortality relative to patients diagnosed at a noncardiac surgery center requiring transfer for surgery (45.5% vs 62.5%; P < .001), highlighting the influence of time delay on outcomes. 4
For type B dissections, sole medical management was the treatment in 83% of cases (3000 out of 3632), open surgery in 10% (370 out of 3632), and thoracic endovascular aortic repair (TEVAR) in 7% (262 out of 3632). Over the 12 years, all-cause nonoperative hospital mortality for type B dissections treated with medical management was 5.6%. There was no change over time (P ¼ .33). Operative mortality for type B dissection was 19.3% (Table 3). Similar to type A, operative mortality for type B dissections did not change across the study (P ¼ .68). The operative mortality for TEVAR was 14.8%. The first recorded TEVAR for type B dissection was not until 2006. TEVAR use increased from 4% in 2006 to 13% by March 2014. Still, present-day dissemination of TEVAR is certain to be underrepresented by this dataset.
The Journal of Thoracic and Cardiovascular Surgery c - 2018
McClure et al
Acquired
TABLE 3. Age- and sex-adjusted 12-year incidence proportions, hospital mortalities (nonoperative and operative), and all-cause 3-year mortalities for thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada (2002-2014) All
Type B
Thoracic aortic aneurysms
1.89 (1.67-2.13) 2.22 (1.89-2.61) 1.57 (1.30-1.89)
2.67 (2.41-2.96) 3.27 (2.86-3.73) 2.10 (1.79-2.47)
7.56 (7.11-8.04) 10.08 (9.34-10.87) 5.15 (4.64-5.71)
0.83 (0.66-1.04) 8.09 (7.23-9.05) 26.05 (22.99-29.53)
0.27 (0.18-0.40) 3.32 (2.78-3.95) 11.48 (9.51-13.87)
0.56 (0.42-0.74) 4.69 (4.05-5.44) 14.6 (12.3-17.2)
1.33 (1.11-1.60) 14.25 (13.1-15.5) 39.83 (36.0-44.1)
20.29 (17.99-22.80) –
20.58 (17.95-23.49) –
19.3 (14.8-24.8) 14.8 (10.3-20.5)
4.53 (3.88-5.27) 5.38 (3.01-8.87)
–
–
5.60 (4.64-6.69)
1.43 (0.95-2.06)
26.57 (25.10-28.10) 22.60 (20.87-24.43) 32.83 (30.22-35.60)
53.00 (49.76-56.39) 45.65 (41.80-49.75) 64.21 (58.59-70.23)
8.03 (7.00-9.17) 6.88 (5.68-8.26) 9.88 (8.06-12.0)
7.65 (7.01-8.33) 6.04 (5.34-6.81) 10.53 (9.29-11.89)
12.17 (9.83-14.91) 21.10 (19.19-23.15) 39.82 (36.88-42.94)
24.91 (19.30-31.63) 41.96 (37.78-46.48) 76.39 (70.21-82.97)
5.25 (3.43-7.70) 6.20 (4.89-7.75) 11.8 (9.73-14.2)
2.41 (1.51-3.65) 5.13 (4.40-5.94) 12.98 (11.6-14.5)
39.52 (34.18-45.45) 32.86 (26.52-40.26) 48.36 (39.47-58.65)
52.61 (43.65-62.86) 41.67 (31.39-54.23) 67.35 (52.09-85.68)
28.2 (22.2-35.3) 25.2 (17.8-34.5) 32.2 (22.7-44.4)
22.10 (19.2-25.4) 17.80 (14.6-21.5) 29.94 (24.3-36.5)
12.64 (6.31-22.62) 34.05 (26.96-42.44) 59.89 (49.03-72.43)
18.42 (7.41-37.95) 48.28 (36.47- 62.69) 76.32 (57.95-98.66)
8.16 (2.2-20.9) 19.8 (12.6-29.8) 47.5 (35.0-63.0)
4.92 (1.80-10.70) 13.65 (10.5-17.4) 40.87 (34.2-48.5)
Incidence proportion Overall Male Female Age (y) 50 51-74 75
4.58 (4.23-4.95) 5.50 (5.00-6.09) 3.70 (3.27-4.18)
Operative mortality* Surgery Thoracic endovascular aortic repair All-cause nonoperative mortality Medical therapy Total hospital mortalityy Overall Male Female Age (y) 50 51-74 75 All-cause 3-y mortalityz Overall Male Female Age (y) 50 51-74 75
Thoracic aortic dissection Type A
ACQ
Variable
Values for incidence proportion are presented as per 100,000 persons (95% confidence interval). Other values are presented as % (95% confidence interval). *Operative mortality defined as all death occurring during the hospitalization in which open surgery or endovascular procedure was performed, even if after 30 days. yTotal hospital mortality defined as all death occurring during the hospitalization. Combines all open surgery, endovascular, and medically managed mortalities. zPercentage of patients who have died within 3 years of the index date of diagnosis.
For thoracic aortic aneurysms, 53% (4940 out of 9392) underwent surgery, 44% (4129 out of 9392) underwent medical therapy, and 3% (323 out of 9392) underwent TEVAR. Of known descending thoracic aortic aneurysms, 35% (323 out of 924) were treated with TEVAR. In the open surgery cohort, 296 (6%) were thoracoabdominal aneurysms. Operative mortality for open surgery on thoracic aneurysms improved over the 12 years (5.5% to 4.5%; P ¼ .002). Yet, for the subset of thoracoabdominal aneurysm patients, operative mortality was 16.4% with no change over time (P ¼ .11). For medically treated thoracic aneurysms, 10% (441 out of 4278) went on to incur an aortic dissection. With respect to reintervention, 7.8% of surgically repaired aortic dissections and 11.3% of TEVARs required additional treatment to some portion of the thoracic aorta. Median time to re-intervention was substantially shortened with TEVAR (Table 5). Similarly, 4.7% of surgically
repaired thoracic aortic aneurysms and 14.2% of TEVARs required further treatment. Reintervention times were again shorter with TEVAR, although much closer to surgical results relative to aortic dissections (Table 5). Whether reintervention was due to failure at the initial treatment site or progression of disease elsewhere in the thoracic aorta could not be teased out from the dataset. Midterm outcomes mirrored that of the total hospital mortality. Over the 12 years, all-cause 3-year mortality decreased for thoracic aortic dissections (44.2% to 39.5%; P ¼ .04), type A dissections (75.0% to 52.6%; P <.0005) (Figure 2), and thoracic aneurysms (29.6% to 22.1%; P < .0001) (Figure 3). No change was observed for type B dissections (27.8% to 28.2%; P ¼ .299) (Figure 4). For aneurysms, 3-year mortality was significantly higher for women relative to men (Figure 3). For dissections, no clear 3-year mortality gender differences were identified (Figures 2 and 4).
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
5
Acquired
McClure et al
ACQ
FIGURE 1. Incidence proportions for thoracic aortic aneurysms and thoracic aortic dissections in Ontario, Canada (2002-2014). Number of persons at risk is equal to the population of Ontario for the year (12.2 million in 2002, 12.5 million in 2004, 12.7 million in 2006, 13 million in 2008, 13.3 million in 2010, 13.5 million in 2012, and 13.7 million in 2014).
For specialty involvement, cardiac surgeons performed 83% and 88% of open surgical procedures, respectively, for thoracic aortic dissections and thoracic aortic aneurysms. Vascular surgeons performed the clear majority of endovascular procedures, being the primary surgeons on record for 91% of such cases. DISCUSSION Quantifying the true incidence of thoracic aortic dissections and aneurysms is difficult. The acuity and lethality of aortic dissections results in out-of-hospital mortality data that are often imprecise. Moreover, the nebulous size criteria to designate an aorta as aneurysmal results in inconsistent classification and subsequent inexact estimates. Despite such challenges, efforts to characterize disease burden are essential to resource allocation and play an authoritative role in health policy. The present study, a large retrospective population-based hospital analysis of thoracic aortic disease in the most densely populated province in Canada, parallels findings of other contemporary studies.5-8,11 We report an overall incidence proportion of 4.6 per 100,000 for thoracic aortic dissections with an increasing trend over time. The Oxford Vascular Study, a contemporary prospective study of 93,000 persons in the United Kingdom,5 had similar results, reporting an incidence of 5.6 per 100,000 for 6
thoracoabdominal aortic dissections. Whereas our study reports a hospital-based incidence proportion, the Oxford Vascular Study included out-of-hospital deaths, which might explain the lower incidence in our cohort. We also report an incidence proportion of 7.6 per 100,000 for thoracic aortic aneurysms with an increasing trend over time. This too has been observed elsewhere.11 Whether these cumulative reports support a true increase in aortic disease or whether this is all attributable to increased imaging surveillance remains unclear. Not all reports have uniformly observed these increases. A recent retrospective study in Iceland12 notes a lower incidence for aortic dissections—2.53 per 100,000—in keeping with the proportions from seminal studies 3 decades prior. Despite an observed increase in disease, there were improved outcomes with significant decreases to total hospital mortality and improved 3-year survival for aortic dissections and aneurysms. Despite these positive trends, operative mortality for type A dissection was disconcerting at 20%. This is in line with worldwide registry results and highlights the global challenge with this lethal condition.7,13 Type A dissection total hospital mortality significantly decreased over our 12-year study, despite a constant operative mortality of 20%, which implies more patients over time are going for surgery. Time-garnered operative experience by senior surgeons or the creation of dedicated
The Journal of Thoracic and Cardiovascular Surgery c - 2018
Acquired
TABLE 4. Demographic characteristic comparison between operative and nonoperative patients with type A aortic dissections in Ontario, Canada (2002-2014) Characteristic
No surgery (n ¼ 1085)
Surgery (n ¼ 1204)
Age at diagnosis (y) 50 51-74 75
74.9 13.4 64 (5.9) 348 (32.1) 673 (62.0)
61.8 14.6 246 (20.4) 706 (58.6) 252 (20.9)
Sex Female Male
527 (48.6) 558 (51.4)
380 (31.6) 824 (68.4)
Diabetes
258 (23.8)
207 (17.2)
<.001
COPD
296 (27.3)
174 (14.5)
<.001
Hypertension
860 (79.3)
913 (75.8)
.05
Hyperlipidemia
255 (23.5)
357 (29.7)
<.001
Residence Urban Rural Missing
908 (83.7) 176 (16.2) <6
1041 (86.5) 161 (13.4) <6
0 (0.0)
30 (2.5)
65 (6.0) 87 (8.0) 104 (9.6) 96 (8.8) 93 (8.6) 88 (8.1) 105 (9.7) 89 (8.2) 92 (8.5) 83 (7.6) 75 (6.9) 108 (10.0) 0 (0.0)
44 (3.7) 61 (5.1) 71 (5.9) 65 (5.4) 77 (6.4) 67 (5.6) 96 (8.0) 129 (10.7) 124 (10.3) 152 (12.6) 135 (11.2) 149 (12.4) 34 (2.8)
Connective tissue disorder Year of diagnosis 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
P value* <.001 <.001
<.001
.142
<.001 <.001
Values are presented as mean standard deviation or n (%). COPD, Chronic obstructive pulmonary disease. *P values from c2 test for categorical variables and t test for continuous variables
aortic teams at select institutions facilitating successful surgery in more extreme patients may have played a role. Better point-of-care stabilization at hospital admission or more expeditious transfers to aortic surgical centers may also have been factors. Still, with only 53% of patients with type A dissections having undergone surgery, there may be opportunities to better these results. We observed total hospital mortality to significantly improve if transfer to another hospital was not required. More resource allocation for timely transfers may have cost benefits. Ontario’s transport of critical patients has evolved. Eleven centers perform the full gamut of open surgery and TEVAR for thoracic aneurysms and dissections. A province-wide call center facilitates emergent transfers, locating the nearest hospital with the necessary expertise. Aid with transfers is also provided. Yearly refinements to this service
may directly relate to the improved hospital outcomes we observed and warrants study. Others report similar statistics. Iceland also reported a 53% surgery rate for type A dissections.12 The Oxford Vascular Study found 58%.5 The International Registry for Acute Aortic Dissection (IRAD) is the exception, reporting 86% operative candidacy.7 IRAD is a skewed representation because 70% of the cohort is referred from elsewhere. With this, a percent of patients with extreme conditions die on transfer or cannot transfer because of instability. The loss of extremis cases alongside the elevated expertise at IRAD centers accounts for the variance. Palliative, nonoperative treatment of type A dissection occurred more often in women. Other areas of cardiovascular medicine have noted similar patterns, with less referral to advanced therapies for women or referral in a more morbid clinical state.14 Type A dissections are emergent events that eliminate the ambiguity of late referral, so it is of interest that gender remained a significant variable. Women also had more hospital mortality, and for the aneurysm cohort, worse 3-year outcomes. The use of TEVAR was captured and seen to increase over time but its true footprint on the thoracic aorta is underestimated. Hybrid procedures were also captured but in negligible numbers. The higher mortality for TEVAR in patients with type B dissections (14.8%) is likely reflective of a conservative philosophy to its use. It is postulated that TEVAR was reserved for patients with high-risk malperfusion. Broader use over time will improve this statistic. For patients with thoracic aneurysms, operative mortality improved, although little can be inferred from this. We could not reliably separate aneurysms based on ascending, arch, and descending segments. Blind to this distribution, the heterogeneity limits interpretation. Similarly, the extent of the thoracoabdominal aneurysms is unknown.15 Cardiac surgeons had an integral role in the thoracic aortic surgical management of patients but were less involved in TEVAR. Our methods captured only the primary operator on record. In situations where physicians from 2 specialties were present (ie, cardiac and vascular surgeons), capturing only the primary physician on record would result in a somewhat misrepresentation of involvement. Aortic teams utilizing expertise of multiple specialties are thus misrepresented by the analysis. Regardless, with only 8% of TEVAR cases having a cardiac surgeon as the primary, more integration of endovascular training into the cardiac armamentarium is required. Study Limitations As 1 of the largest published population-based studies to assess thoracic aortic disease, in more than 13.5 million persons with complete hospital records due to universal health care and legislated data collection representing
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
7
ACQ
McClure et al
Acquired
McClure et al
TABLE 5. Time to reintervention (>72 hours) in patients undergoing surgical or endovascular repair for thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada (2002-2014)
Treatment group Surgery
Category
n
All
120
> 3-30 d Endovascular
Thoracic aortic dissections Median time to Denominator % re-intervention (d) 1546
7.8
424
n 232
Thoracic aortic aneurysms Median time to Denominator % re-intervention (d) 4940
4.7
7
1546
0.5
8
34
4940
0.7
> 30 d
113
1546
7.3
478
198
4940
4.0
544
All
33
291
11.3
138
46
323
14.2
352
> 3-30 d
6
291
2.1
>30 d
27
291
9.3
ACQ
38.5% of the Canadian populace, our study is very robust and generalizable to the region. Still, there are limitations. Because it is a retrospective study using administrative data, a degree of misclassification is inherent. However, with our cross-referencing across successive databases, this is likely quite negligible. More relevant limitations are the inability to capture out-of-hospital deaths and an inherent shortcoming of the aortic dissection algorithm with respect to nonoperative type A dissection survivors. The accepted distribution for aortic dissection is 60% type A and 40% type B.16 Our distribution was 38% type A and 61% type B. This discrepancy is explainable. Recognizing that nearly 50% of type A dissection patients die before hospital admission,5 our distribution resembles
9.5 239
8.5
6
323
1.9
7
40
323
12.4
449
much more the expected distribution. With Ontario’s expansive landmass and the challenges of remote area access, out-of-hospital death is likely even higher. This remains an unknown factor and warrants study. Although this is the main reason for the distribution discrepancy, the aortic dissection algorithm has an inherent shortcoming that lends to misclassification of a small subset of patients as well. The algorithm assumes nonoperative aortic dissections that survive to hospital discharge are type B dissections. For the overwhelming majority this is correct. Still, a small subset of type A dissections not operated upon will indeed survive to hospital discharge. IRAD had 6.3% (19 out of 303) of type A dissections fit this subset.17 In our study, 2840 patients were classified as
FIGURE 2. Three-year all-cause mortality for type A aortic dissections in Ontario, Canada (2002-2014).
8
370
The Journal of Thoracic and Cardiovascular Surgery c - 2018
Acquired
ACQ
McClure et al
FIGURE 3. Three-year all-cause mortality for thoracic aortic aneurysms in Ontario, Canada (2002-2014).
having type B dissections for being treated medically and surviving to discharge. In a worst-case scenario sensitivity analysis, if 6.3% in fact were nonoperative type A dissection survivors, not type B dissection survivors,
179 patients (3%) would have been misclassified out of 5966 aortic dissections. Adjusting for 50% out-of-hospital death and 179 misclassified patients results in a distribution of 58% type A and 42% type B.
FIGURE 4. Three-year all-cause mortality for type B aortic dissections in Ontario, Canada (2002-2014).
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
9
Acquired
Of note, the shortcoming of the model does not influence the incidence rate for aortic dissections (4.6/100,000). And with 13.5 million persons, has very marginal effects on subset incidence rates for type A and B dissections (type A 2.0/100,000 vs 1.9/100,000, type B 2.6/100,000 vs 2.7/100,000). There is no influence on trend analyses or any operative statistics. We used a conservative method to assess the incidence of thoracic aortic aneurysms. Thoracic aneurysms that ruptured or dissected were analyzed as dissections only. This approach minimized misclassification and maintains data accuracy but at the cost of underestimating the true incidence. Medication therapy following current guidelines was presumably adhered to for medically treated patients but any variance across treating physicians to anti-impulse strategies are not accounted for. Finally, the vast majority of aortic dissections would have been acute, but a small subset of chronic dissections is certain to be present. Codes cannot distinguish between acute and chronic events. Because acute and chronic dissections differ in operative risk, this unavoidable inclusion skews the result slightly relative to a purely acute cohort.
ACQ
CONCLUSIONS The incidence of thoracic aortic dissections and aneurysms has increased, but hospital mortality and 3-year outcomes have improved. Men incur more disease, but women have higher mortality. Surgery is primarily referred to cardiac surgeons. Vascular surgeons perform most TEVARs. Resource allocation for expeditious transfer to aortic centers may have life-saving implications. Webcast You can watch a Webcast of this AATS meeting presentation by going to: https://aats.blob.core.windows.net/ media/17AM/2017-05-02/RM311/05-02-17_Room311_ 1414_McClure.mp4.
Conflict of Interest Statement Authors have nothing to disclose with regard to commercial support. This project was approved by the Institute for Clinical Evaluative Sciences (ICES). ICES is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. The opinions, results, and conclusions reported in this paper are those of the authors and are independent from the funding sources.
10
McClure et al
No endorsement by ICES or the Ontario Ministry of Health and Long-Term Care is intended or should be inferred. Dataset creation and data analysis was performed at ICES Queens by Ms Katherine Lajkosz. Parts of this material are based on data and information compiled and provided by CIHI. However, the analyses, conclusions, opinions, and statements expressed herein are those of the author, and not necessarily those of CIHI. Drs McClure and Payne are members of the Canadian Thoracic Aortic Collaborative (CTAC). CTAC is a group of cardiovascular clinicians across Canada with a specific interest and expertise in the management of thoracic aortic disease.
References 1. Bickerstaff LK, Pairolero PC, Hollier LH, Melton LJ, Van Peenen HJ, Cherry KJ, et al. Thoracic aortic aneurysms: a population-based study. Surgery. 1982;92:1103-8. 2. Svensjo S, Bengtsson H, Berggvist D. Thoracic and thoracoabdominal aortic aneurysm dissection: an investigation based on autopsy. Br J Surg. 1996;83:68-71. 3. Clouse WD, Hallett JW Jr, Schaff HV, Gayari MM, Ilstrup DM, Melton LJ III. Improved prognosis of thoracic aortic aneurysms: a population-based study. JAMA. 1998;280:1926-9. 4. Meszaros I, Morocz J, Szlavi J, Schmidt J, Tornoci L, Nagy L, et al. Epidemiology and clinicopathology of aortic dissection. Chest. 2000;117:1271-8. 5. Howard DPJ, Banerjee A, Fairhead JF, Perkins J, Silver LE, Rothwell PM. Population-based study of incidence and outcome of acute aortic dissection and premorbid risk factor control. 10-year results from the Oxford vascular study. Circulation. 2013;127:2031-7. 6. Mody PS, Wang Y, Geirsson A, Kim N, Desai MM, Gupta A, et al. Trends in aortic dissection hospitalizations, interventions, and outcomes among medicare beneficiaries in United States, 2002-2011. Circ Cardiovasc Qual Outcome. 2014;7:920-8. 7. Pape LA, Awais M, Woznicki EM, Suzuki T, Trimarchi S, Evangelista A, et al. Presentation, diagnosis, and outcomes of acute aortic dissection. 17-year trends from the international registry of acute aortic dissection. J Am Coll Cardiol. 2015;66:350-8. 8. Zimmerman KP, Oderich G, Pochettino A, Hanson KT, Habermann E, Bower TC, et al. Improving mortality trends for hospitalization of aortic dissection in the national inpatient sample. J Vasc Surgery. 2016;64:606-15. 9. Lewis S. A system in name only – access, variation and reform in Canada’s provinces. N Engl J Med. 2015;372:497-500. 10. Wodchis WP, Austin PC, Henry DA. A 3-year study of high-cost users of health care. Can Med Assoc J. 2016;188:182-8. 11. Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection. Increasing prevalence and improved outcomes reported in a nationwide population-based study of more than 14000 cases from 1987 to 2002. Circulation. 2006;114:2611-8. 12. Melvinsdottir IH, Lund SH, Agnarsson BA, Sigvaldason K, Gudbjartsson T, Geirsson A. The incidence and mortality of acute thoracic aortic dissection: results from a whole nation study. Eur J Cardiothorac Surg. 2016;50:1111-7. 13. Conzelmann LO, Weigang E, Mehlhorn U, Abugameh A, Hoffmann I, Blettner M, et al. Mortality in patients with acute aortic dissection type A: analysis of pre- and intraoperative risk factors from the German registry for acute aortic dissection type A (GERAADA). Eur J Cardiothorac Surg. 2016;49:e44-52. 14. Bogaev RC. Gender disparities across the spectrum of advanced cardiac therapies: real or imagined? Curr Cardiol Rep. 2016;11:108-18. 15. Coselli JS, LeMaire SA, Preventza O, de la Cruz KI, Cooley DA, Price MD, et al. Outcomes of 3309 thoracoabdominal aortic aneurysm repairs. J Thorac Cardiovasc Surg. 2016;151:1323-38. 16. Mery CM, Reece TB, Kron IL. Aortic dissection. In: Cohn LH, ed. Cardiac Surgery in the Adult. 4th ed. New York: McGraw-Hill; 2012:999. 17. Tsai TT, Evangelista A, Nienaber CA, Trimarchi S, Sechtem U, Fattori R, et al. Long-term survival in patients presenting with type A acute aortic dissection: insight from the international registry of acute aortic dissection (IRAD). Circulation. 2006;114(1 Suppl):I350-6.
Key Words: aortic dissection, aortic aneurysm, epidemiology, database
The Journal of Thoracic and Cardiovascular Surgery c - 2018
Acquired
Discussion Dr John Elefteriades (New Haven, Conn). Congratulations on your thoughtful, well-conducted investigation in which you took important advantages of the centralized health care system in Canada to make key observations about incidence and treatment of thoracic aortic disease. I have 3 brief questions. Is the increased incidence of thoracic aortic diseases that you found an artifact of increased imaging in recent years or a bona fide increase in the disease itself? Number 2: If you feel it may be a bona fide increase in the disease itself, what is the reason? How can biology or evolution change so quickly as to be detected in a decade? Number 3: How do you suggest that we modify our health care delivery or practice patterns based on your findings? R. Scott McClure. Thank you very much for the questions. We are honored to have you as our discussant, being a leader in this area. To answer your first 2 questions, my personal opinion is that the increased incidence of disease has much more to do with a heightened awareness of the medical condition by the medical community rather
than a true evolutionary increase in disease. To your point, biologic change over a 12-year time frame would be rather unlikely. Unknown environmental or lifestyle factors may be at play but more astute diagnosis of that which was always there seems much more plausible. Improved clinical intuition and exponential use of imaging studies are most likely responsible for the data. With regard to your last question, I think this study highlights importance of allocating more resources to the treatment of thoracic aortic pathology. Because of the stark contrast between the benign aneurysm and the catastrophic aortic dissection, determining the proper amount of funding is difficult. But what we do know is that the incidence of disease for both aneurysms and dissections is higher in the population than previously thought. We need to acknowledge that and adjust resources accordingly. More timely transfer of patients is but 1 area where this may prove beneficial. One interesting statistic that we teased out that I didn’t talk about because of time constraints, is that we found a rather significant difference in outcomes in those patients with a type A dissection diagnosed at a hospital without cardiac surgery on site versus those patients diagnosed at a hospital with cardiac surgery services. So 1 thing that we could do is improve our transfer of patients from point A to point B if and when that diagnosis has been achieved. I think that would go a long way to helping our outcomes.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
11
ACQ
McClure et al
Acquired
McClure et al
TABLE E1. International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Canada (ICD-10-CA) diagnosis codes ICD-10-CA description Thoracic aortic dissection Dissection of aorta (any part) Injury of thoracic aorta Thoracic aortic aneurysm, ruptured Thoracoabdominal aortic aneurysm, ruptured Aneurysm and dissection of unspecified site Thoracic aortic aneurysm Aneurysm of aorta in disease classified elsewhere (Syphilitic) Thoracic aortic aneurysm, without mention of rupture Thoracoabdominal aortic aneurysm, without mention of rupture Aortic arch syndrome (Takayasu)
ICD-10-CA code I710 S250 I71 I715 I729 I790 I712 I716 M314
ACQ 11.e1
The Journal of Thoracic and Cardiovascular Surgery c - 2018
McClure et al
Acquired
TABLE E2. Canadian Classification of Health Interventions (CCI) surgical and endovascular procedures codes Type of aortic dissection/ aneurysm
CCI code
Description
1HV90WJCFN
Excision total with reconstruction, aortic valve, replacement of valve, aortic root, and ascending aorta (eg, Bentall) using open approach with mechanical valve and synthetic aorta
Surgery
Ascending
1HV90WJXXA
Excision total with reconstruction, aortic valve, replacement of valve, aortic root and ascending aorta (eg, Bentall) using open approach with autograft tissue valve and autograft aorta
Surgery
Ascending
1HV90WJXXD
Excision total with reconstruction, aortic valve, replacement of valve, aortic root and ascending aorta (eg, Bentall) using open approach with xenograft tissue valve (eg, bovine or porcine tissue) and synthetic aorta
Surgery
Ascending
1HV90WJXXK
Excision total with reconstruction, aortic valve, replacement of valve, aortic root and ascending aorta (eg, Bentall) using open approach with homograft tissue valve and homograft aorta
Surgery
Ascending
1HV90WJXXL
Excision total with reconstruction, aortic valve, replacement of valve, aortic root and ascending aorta (eg, Bentall) using open approach with xenograft tissue valve (eg, bovine or porcine tissue) and xenograft aorta
Surgery
Ascending
1IA80GQNRN
Repair, ascending aorta using percutaneous transluminal approach and (endovascular) stent with synthetic tissue (eg, stent graft)
Endovascular
Ascending
1IA80LA
Repair, ascending aorta using open approach without tissue
Surgery
Ascending
1IA80LAXXA
Repair, ascending aorta using open approach with autograft (eg, pericardial patch, omental patch)
Surgery
Ascending
1IA80LAXXK
Repair, ascending aorta using open approach with homograft (eg, arterial homograft)
Surgery
Ascending
1IA80LAXXL
Repair, ascending aorta using open approach with xenograft
Surgery
Ascending
1IA80LAXXN
Repair, ascending aorta using open approach with synthetic tissue (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Ascending
1IA82LA
Reattachment, ascending aorta using open approach
Surgery
Ascending
1IA87LA
Excision partial, ascending aorta using open approach without tissue (eg, anastomosis)
Surgery
Ascending
1IB80GQNRN
Repair, arch of aorta using percutaneous transluminal approach and (endovascular) stent with synthetic tissue (eg, stent graft)
Endovascular
Arch
1IB80LA
Repair, arch of aorta using open approach without tissue
Surgery
Arch
1IB80LAXXA
Repair, arch of aorta using open approach with autograft (eg, pericardial patch)
Surgery
Arch
1IB80LAXXK
Repair, arch of aorta using open approach with homograft (eg, arterial homograft)
Surgery
Arch
1IB80LAXXL
Repair, arch of aorta using open approach with xenograft
Surgery
Arch
1IB80LAXXN
Repair, arch of aorta using open approach with synthetic material (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Arch
1IC50GQOA
Dilation, thoracic (descending) aorta using percutaneous transluminal arterial approach balloon dilator with (endovascular) stent (insertion)
Endovascular
Descending
1IC50GSBD
Dilation, thoracic (descending) aorta using percutaneous transluminal approach with placement/implant of stent and mechanical balloon dilator
Endovascular
Descending
1IC50LANR
Dilation, thoracic (descending) aorta using endovascular stent (eg, ⌠Z O stent] open approach, such as retroperitoneal)
Endovascular
Descending (Continued)
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
11.e2
ACQ
Type of intervention
Acquired
McClure et al
TABLE E2. Continued
CCI code
Description
Type of intervention
Type of aortic dissection/ aneurysm
ACQ
1IC55LANRA
Removal of device, thoracic (descending) aorta open approach surgical repair of defect using autograft of endovascular stent
Surgical reintervention
Descending
1IC55LANRN
Removal of device, thoracic (descending) aorta open approach surgical repair of defect using synthetic material of endovascular stent
Surgical reintervention
Descending
1IC55LANRQ
Removal of device, thoracic (descending) aorta open approach surgical repair of defect using combined sources of tissue of endovascular stent
Surgical reintervention
Descending
1IC80GQNRN
Repair, thoracic (descending) aorta using percutaneous transluminal approach and (endovascular) stent with synthetic tissue (eg, stent graft)
Endovascular
Descending
1IC80LA
Repair, thoracic (descending) aorta using open approach without tissue
Surgery
Descending
1IC80LAXXA
Repair, thoracic (descending) aorta using open approach with autograft (eg, pericardial patch, subclavian flap)
Surgery
Descending
1IC80LAXXK
Repair, thoracic (descending) aorta using open approach with homograft (eg, arterial homograft)
Surgery
Descending
1IC80LAXXL
Repair, thoracic (descending) aorta using open approach with xenograft
Surgery
Descending
1IC80LAXXN
Repair, thoracic (descending) aorta using open approach with synthetic material (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Descending
1IC80LAXXQ
Repair, thoracic (descending) aorta using open approach with combined sources of tissue
Surgery
Descending
1IC80WC
Repair, thoracic (descending) aorta using open approach with fenestration (aneurysm) technique (eg, re-entry operation)
Surgery
Descending
1IC82LA
Reattachment, thoracic (descending) aorta using open approach without tissue
Surgery
Descending
1IC82LAXXA
Reattachment, thoracic (descending) aorta using open approach with autograft
Surgery
Descending
1IC82LAXXN
Reattachment, thoracic (descending) aorta using open approach with synthetic material
Surgery
Descending
1IC82LAXXQ
Reattachment, thoracic (descending) aorta using open approach with combined sources of tissue
Surgery
Descending
1IC87LA
Excision partial, thoracic (descending) aorta using open approach with simple end-to-end anastomosis
Surgery
Descending
1IC87LAXXN
Excision partial, thoracic (descending) aorta using open approach with synthetic material (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Descending
1IC87LAXXQ
Excision partial, thoracic (descending) aorta using open approach with combined sources of tissue
Surgery
Descending
1IC87TQ
Excision partial, thoracic (descending) aorta using open approach with extended end-to-end anastomosis
Surgery
Descending
1ID80QFXXK
Repair, aorta NEC open thoracoabdominal approach using homograft (eg, arterial homograft)
Surgery
Descending
1ID80QFXXN
Repair, aorta NEC open thoracoabdominal approach using synthetic material (eg, polytetrafluoroethylene felt, polyethylene terephthalate, nylon, and acrylic)
Surgery
Descending
1ID80QFXXQ
Repair, aorta NEC open thoracoabdominal approach using combined sources of tissue
Surgery
Descending
1ID87QFXXK
Excision partial, aorta NEC using homograft thoracoabdominal approach
Surgery
Descending (Continued)
11.e3
The Journal of Thoracic and Cardiovascular Surgery c - 2018
McClure et al
Acquired
TABLE E2. Continued
CCI code
Description
Type of intervention
Type of aortic dissection/ aneurysm
1ID87QFXXN
Excision partial, aorta NEC using synthetic material thoracoabdominal approach
Surgery
Descending
1ID87QFXXQ
Excision partial, aorta NEC using combined sources of tissue thoracoabdominal approach
Surgery
Descending
ACQ
CCI, Canadian Classification of Health Interventions; NEC, not elsewhere classified.
The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -
11.e4
Acquired
000
Epidemiology and management of thoracic aortic dissections and thoracic aortic aneurysms in Ontario, Canada: A population-based study R. Scott McClure, MD, Susan B. Brogly, PhD, Katherine Lajkosz, MS, Darrin Payne, MD, Stephen F. Hall, MD, and Ana P. Johnson, PhD, Calgary, Alberta, and Kingston, Ontario, Canada The incidence of thoracic aortic aneurysms and dissections has increased but hospital mortality and late outcomes have improved. Men incur more thoracic aortic disease but women have higher hospital mortality.
ACQ The Journal of Thoracic and Cardiovascular Surgery c - 2018
McClure et al