Abstract
BACKGROUND
This study describes in detail the clinical burden of malperfusion associated with acute type A aortic dissection (ATAAD) in a large, national cohort and the effect of treatment strategy on outcomes.
METHODS
All patients undergoing repair of ATAAD between 2017 and 2020 in The Society of Thoracic Surgeons (STS) Adult Cardiac Surgery Database were studied. Malperfusion was defined using STS definitions on the basis of imaging or the surgeon’s evaluation. Multivariable logistic regression was used to analyze the effect of patient and treatment factors on outcomes in patients with and without malperfusion.
RESULTS
A total of 9958 patients undergoing ATAAD repair were studied. Preoperative malperfusion occurred in 27.7% (2748 of 9958) of cases and most often involved the extremity (14.9%; 1484 of 9958), renal (10.2%), or cerebral (9.8%) vascular beds. Operative mortality was much greater among patients with malperfusion (26.8% vs 13.6%; P < .001). After adjustment, coronary malperfusion was associated with the highest odds of mortality (odds ratio, 2.28; 95% CI, 1.85–2.81; P < .001) followed by mesenteric malperfusion (odds ratio, 1.82; 95% CI, 1.45–2.28; P < .001). Cerebral malperfusion was not independently associated with significantly increased odds of mortality (odds ratio, 1.14; 95% CI, 0.94–1.38; P =.18). Partial arch replacement (zone 1 or zone 2) compared with ascending aorta or hemiarch replacement only showed a similar rate of mortality in patients with malperfusion (24.8% vs 26.9%; P = .99) and without malperfusion (11.6% vs 13.6%; P = .54).
CONCLUSIONS
Preoperative malperfusion in ATAAD was common and associated with significant operative mortality, which varied according to the malperfused region. Partial aortic arch replacement, compared with ascending aorta or hemiarch replacement alone, was not associated with increased mortality.
Preoperative malperfusion in the setting of acute type A aortic dissection (ATAAD) occurs when the dissection flap compromises blood flow through an aortic branch vessel, and it has been identified as a key predictor of mortality after surgical intervention.1–8 However, previous reports on preoperative malperfusion in ATAAD were limited by a high degree of variance in fundamental variables, including the incidence of malperfusion (20.5% to 54.2%) and the magnitude of its effect on mortality (8.1% to 28.7%).1–8 Still others have presented series in which malperfusion was not associated with mortality.9–11 The causes of this dispersion include combinations of underpowering, bias from era effects, and single-institution experiences with a diversity of technical approaches, as well as the heterogenous nature of dissection anatomy, patient presentation, and definitions of malperfusion.
To address these limitations, this study describes, in a very large, contemporary, and nationally comprehensive cohort from The Society of Thoracic Surgeons (STS) Adult Cardiac Surgery Database (ACSD), the incidence of malperfusion and its outcomes after surgical repair of ATAAD.
PATIENTS AND METHODS
DATA SOURCE AND STUDY POPULATION.
The STS ACSD is a detailed clinical registry of preoperative characteristics, procedural details, and postoperative outcomes for major procedures in adult cardiac surgery.12 The estimated penetration of the ACSD is 92% to 95% of hospitals performing adult cardiac surgery in the United States, representing all 50 states.13 Data accuracy is maintained through random annual external auditing, which has consistently demonstrated >95% concordance with the medical record.12 In this study, the STS ACSD was queried for all patients aged >18 years who underwent surgical repair of ATAAD from 2017 to 2020 (Supplemental Figure 1). This time period corresponds to version 2.9 of the ACSD, which is the first to collect granular variables specific to aortic surgery, including branch vessel and arch anatomy, dissection extent, malperfusion, and procedural details. Only patients undergoing repair of acute dissection (<2 weeks from symptom onset to surgery) were included. Cases primarily characterized as aortic intramural hematoma rather than dissection were excluded. For this study, data were obtained through the STS Participant User File program, and data analysis was performed at the investigators’ institutions.
MALPERFUSION.
In clinical practice, malperfusion can be identified radiographically or from the constellation of signs and symptoms indicative of end-organ ischemia that is termed malperfusion syndrome. For this analysis, preoperative malperfusion was defined by STS criteria as compromised blood flow through an aortic branch vessel as determined by radiology report or surgeon evaluation.14 We further categorized branch vessel malperfusion into 6 non–mutually exclusive tissue regions: coronary, cerebral, mesenteric, renal, extremity (upper or lower), or spinal. A detailed explanation of how these variables were specified within the STS ACSD is provided (Supplemental Table 1).
OUTCOMES.
The primary outcome was operative mortality, which is defined by the STS as all deaths occurring during the hospitalization in which the operation was performed, even after 30 days (including patients transferred to other acute care facilities), and all deaths occurring after discharge but before the end of the 30th postoperative day.14 Secondary outcomes included permanent stroke, renal failure, reoperation, deep sternal wound infection, prolonged ventilation, and composite morbidity and mortality, which were defined by STS criteria (Supplemental Table 2).14
STATISTICAL ANALYSIS.
Missing baseline characteristics were imputed using multiple imputation. Missing operative characteristics were imputed to the most common category for categorical variables or median for continuous variables. Missingness was low—only 4 variables (history of stroke, cerebrovascular disease, or carotid stenosis and days from dissection to surgery) had missingness >5%. Standardized mean difference (SMD) was used to describe the baseline and operative characteristics. An SMD of 0.2 indicates a small difference between groups, 0.5 a moderate difference, and 0.8 a large difference. Descriptive univariable analysis was performed using the χ2 test. Multivariable logistic regression modeling was used to analyze an adjusted association between malperfusion and mortality. Model predictors were selected through expert opinion and review of existing literature (Supplemental Table 3), with the multicollinearity threshold based on a variance inflation factor <3.
Statistical analyses were performed using R software version 4.1.1 (R Foundation). The study meets the University of Pennsylvania Institutional Review Board criteria for exemption from review according to 45 CFR 46.104, category 4.
RESULTS
BASELINE CHARACTERISTICS AND PRESENTATION.
The study sample was composed of 9958 patients with ATAAD, and malperfusion occurred in 27.7% (2758 of 9958). Patients with malperfusion tended to be younger, male, have higher body mass index, be a current smoker, and have a history of carotid artery stenosis, cerebrovascular disease, stroke, and peripheral vascular disease. Other baseline characteristics were similar for patients with and without malperfusion (Table 1). Regarding entry tear location, most patients were classified as having a primary entry tear in the proximal aorta (93.4%, 9,300/9,958) (Supplemental Table 4).
TABLE 1.
Baseline Characteristics of the Sample Population
| Characteristics | Entire Sample N = 9958 |
Without Malperfusion n = 7200 |
With Malperfusion n = 2758 |
SMD | P Value |
|---|---|---|---|---|---|
| Demographics and comorbidities | |||||
| Age, y | 60.4 ± 13.7 | 60.9 ± 14.0 | 59.0 ± 12.8 | 0.141 | <.001 |
| Male | 6584 (66.1) | 4621 (64.2) | 1963 (71.2) | 0.150 | <.001 |
| Carotid stenosis | 479 (4.8) | 235 (3.3) | 244 (8.8) | 0.236 | <.001 |
| Previous MI | 1249 (12.5) | 897 (12.5) | 352 (12.8) | 0.009 | .706 |
| Cerebrovascular disease | 1505 (15.1) | 926 (12.9) | 579 (21.0) | 0.218 | <.001 |
| Stroke | 1019 (10.2) | 625 (8.7) | 394 (14.3) | 0.176 | <.001 |
| Peripheral vascular disease | 2062 (20.7) | 1243 (17.3) | 819 (29.7) | 0.296 | <.001 |
| Hypertension | 8250 (82.8) | 5977 (83.0) | 2273 (82.4) | 0.016 | .497 |
| Coronary artery disease | 817 (8.2) | 662 (9.2) | 155 (5.6) | 0.137 | <.001 |
| Lung disease | 817 (8.2) | 615 (8.5) | 202 (7.3) | 0.045 | .052 |
| Liver disease | 330 (3.3) | 235 (3.3) | 95 (3.4) | 0.010 | .698 |
| Diabetes | 1166 (11.7) | 872 (12.1) | 294 (10.7) | 0.046 | .048 |
| Dialysis | 212 (2.1) | 167 (2.3) | 45 (1.6) | 0.049 | .04 |
| Previous cardiac surgery | 688 (6.9) | 522 (7.3) | 166 (6.0) | 0.049 | .034 |
| Bicuspid aortic valve | 318 (3.2) | 246 (3.4) | 72 (2.6) | 0.047 | .047 |
| Family history of aortic disease | 949 (9.5) | 667 (9.3) | 282 (10.2) | 0.032 | .155 |
| Connective tissue disorder | 171 (1.7) | 132 (1.8) | 39 (1.4) | 0.033 | .175 |
| Dissection features | |||||
| Transferred | 5315 (53.4) | 3877 (53.8) | 1438 (52.1) | 0.034 | .132 |
| Days from symptom onset to surgery | 0.47 ± 1.25 | 0.51 ± 1.30 | 0.38 ± 1.10 | 0.104 | <.001 |
| Rupture | 1664 (16.7) | 1142 (15.9) | 522 (18.9) | 0.081 | <.001 |
| Cardiogenic shock | 1130 (11.3) | 680 (9.4) | 450 (16.3) | 0.206 | <.001 |
| Malperfused region | |||||
| Coronary | 502 (5.0) | 0 | 502 (18.2) | … | … |
| Cerebral | 972 (9.8) | 0 | 972 (35.2) | … | … |
| Mesenteric | 563 (5.7) | 0 | 563 (20.4) | … | … |
| Renal | 1018 (10.2) | 0 | 1018 (36.9) | … | … |
| Extremities | 1484 (14.9) | 0 | 1484 (53.8) | … | … |
| Spinal | 101 (1.0) | 0 | 101 (3.7) | … | … |
| Number of malperfused regions | |||||
| 0 | 7200 (72.3) | 7200 (100.0) | 0 (0.0) | … | … |
| 1 | 1525 (15.3) | 0 | 1525 (55.3) | … | … |
| 2 | 780 (7.8) | 0 | 780 (28.3) | … | … |
| ≥3 | 453 (4.5) | 0 | 453 (16.4) | … | … |
| Primary entry tear location | |||||
| Proximal | 9300 (93.4) | 6757 (93.8) | 2543 (92.2) | 0.067 | .004 |
| Distal | 658 (6.6) | 443 (6.2) | 215 (7.8) | … | |
| Operative features | |||||
| Proximal extent of surgery | |||||
| Aortic valve resuspension | 7836 (78.7) | 5728 (79.6) | 2108 (76.4) | 0.062 | <.001 |
| Wheat | 204 (2.0) | 143 (2.0) | 61 (2.2) | 0.013 | .527 |
| Bentall | 1752 (17.6) | 1211 (16.8) | 541 (19.6) | 0.060 | .001 |
| VSRR | 109 (1.1) | 82 (1.1) | 27 (1.0) | 0.013 | .563 |
| Other | 57 (0.6) | 36 (0.5) | 21 (0.8) | 0.028 | .162 |
| Distal extent of surgery | |||||
| Ascending aorta or hemiarch only | 8422 (84.6) | 6228 (86.5) | 2194 (79.6) | 0.156 | <.001 |
| Partial arch (zone 1 or 2) | 994 (10.0) | 631 (8.8) | 363 (13.2) | 0.119 | <.001 |
| Total arch | 542 (5.4) | 341 (4.7) | 201 (7.3) | 0.091 | <.001 |
| CPB minutes | 201.0 ± 83.2 | 195.9 ± 80.5 | 214.5 ± 88.5 | 0.220 | <.001 |
| Circulatory arrest | 8220 (82.5) | 5806 (80.6) | 2414 (87.5) | 0.189 | <.001 |
| Circulatory arrest minutes | 30.5 ± 27.4 | 29.0 ± 26.6 | 34.4 ± 29.2 | 0.195 | <.001 |
| Cerebral perfusion strategy | |||||
| Circulatory arrest without cerebral perfusion | 1865 (18.7) | 1339 (18.6) | 526 (19.1) | 0.013 | .607 |
| Antegrade only | 4289 (43.1) | 2977 (41.3) | 1312 (47.6) | 0.127 | <.001 |
| Retrograde only | 1694 (17.0) | 1215 (16.9) | 479 (17.4) | 0.013 | .578 |
| Both antegrade and retrograde, or unknown | 372 (3.7) | 275 (3.8) | 97 (3.5) | 0.016 | .514 |
| No circulatory arrest | 1738 (17.5) | 1394 (19.4) | 344 (12.5) | 0.189 | <.001 |
Values are mean ± SD or n (%). CPB, cardiopulmonary bypass; MI, myocardial infarction; SMD, standardized mean difference, VSRR, valve-sparing root replacement.
MALPERFUSION.
The pattern and burden of malperfusion were heterogenous (Figure 1). Among all patients undergoing repair of ATAAD, malperfusion of the extremities was most common (14.9%; 1484 of 9958) followed by renal (10.2%), cerebral (9.8%), mesenteric (5.7%), coronary (5.0%), and spinal (1.0%) malperfusion (Figure 2A). Most patients had a single region of involvement (55.3%; 1525 of 2758), although involvement of 2 (28.3%) or 3 or more regions (16.4%) was not uncommon (Figure 2C).
FIGURE 1.
UpSet plot showing the number of patients presenting with all combinations of regional malperfusion. The horizontal black bars in the bottom left quadrant represent the number of instances that each region was affected by malperfusion. The vertical bars composing the central portion of the figure represent the number of patients affected by the unique combination of malperfused regions indicated by the black dots below. The red portion of the vertical bar is the number of patients within this given constellation of regional malperfusion who suffered operative mortality. For example, there were 523 patients with isolated extremity malperfusion and 274 patients with both cerebral and extremity malperfusion.
FIGURE 2.
Among all patients with acute type A aortic dissection, (A) the distribution of malperfusion by region, (B) the incidence of mortality by malperfused region, (C) the distribution of malperfusion by the number of involved regions, and (D) the incidence of mortality by the number of involved regions.
INTERVENTION.
Operative characteristics are displayed in Table 1. In terms of the distal extent of surgery, most patients underwent ascending aorta or hemiarch replacement alone (84.6%; 8422 of 9958). Compared with patients without malperfusion, those with malperfusion more frequently underwent partial arch replacement (zones 1 or 2) (13.2% vs 8.8%; SMD = 0.12) or total arch replacement (7.3% vs 4.7%; SMD = 0.09). Commensurate with this, the average duration of cardiopulmonary bypass (214.5 minutes vs 195.9 minutes; SMD = 0.22) and circulatory arrest (34.4 minutes vs 29.0 minutes; SMD = 0.20) was longer for those with malperfusion. Distribution of the distal extent of surgery by region of malperfusion is displayed in Figure 3. Repair without the use of circulatory arrest was performed in 17.5% (1738 of 9958) of cases. Of these cases, 82.9% (1440 of 1738) involved replacement of the ascending aorta only in terms of their distal extent, and 17.1% (298 of 1738) of them were more extensive. Regarding cerebral perfusion strategy, antegrade cerebral perfusion was most common and was used in 43.1% (4289 of 9958) of all cases and in 52.2% (4289 of 8220) that involved circulatory arrest. Circulatory arrest without cerebral perfusion was used in 18.7% (1865 of 9958) of all cases and 21.1% (1865 of 8220) of all circulatory arrest cases.
FIGURE 3.
Bar chart demonstrating the distribution of the distal extent of surgery chosen on the basis of malperfused regions at presentation and subsequent mortality.
UNADJUSTED OUTCOMES.
Patients with malperfusion experienced significantly higher unadjusted operative mortality compared with patients without malperfusion (26.8% vs 13.6%; P < .001) (Table 2). Patients with malperfusion also experienced significantly higher rates of each secondary outcome with the exception of readmission. Notably, only 49.7% (4954 of 9958) of patients were discharged home after repair, a percentage that declined to 37.4% (1032 of 2758) among patients with malperfusion (P<0.001).
TABLE 2.
Operative Outcomes
| Outcome | Entire Sample N = 9958 |
Without Malperfusion n = 7200 |
With Malperfusion n = 2758 |
P Value |
|---|---|---|---|---|
| Primary outcome | ||||
| Operative mortality | 1720 (17.3) | 980 (13.6) | 740 (26.8) | <.001 |
| Secondary outcomes | ||||
| Major morbidity or mortality | 5953 (59.8) | 3932 (54.6) | 2021 (73.3) | <.001 |
| Permanent stroke | 1234 (12.4) | 803 (11.2) | 431 (15.6) | <.001 |
| Renal failure | 1582 (15.9) | 891 (12.4) | 691 (25.1) | <.001 |
| Prolonged ventilation | 4647 (46.7) | 3113 (43.2) | 1534 (55.6) | <.001 |
| Reoperation | 1874 (18.8) | 1176 (16.3) | 698 (25.3) | <.001 |
| Postoperative blood products, units | 4.61 ± 10.21 | 4.06 ± 9.47 | 6.04 ± 11.81 | <.001 |
| ICU length of stay, h | 179.0 ± 218.4 | 165.9 ± 194.0 | 214.6 ± 270.8 | <.001 |
| Hospital length of stay, d | 13.5 ± 13.3 | 13.0 ± 12.5 | 14.8 ± 15.1 | <.001 |
| Discharge location | ||||
| Home | 4954 (49.7) | 3922 (54.5) | 1032 (37.4) | <.001 |
| Rehabilitation center | 3321 (33.4) | 2330 (32.4) | 991 (35.9) | <.001 |
| Other or unknown | 1683 (16.9) | 948 (13.2) | 735 (26.6) | <.001 |
| Readmission | 1396 (14.0) | 1036 (14.4) | 360 (13.1) | .09 |
Values are n (%) or mean ±SD. ICU, intensive care unit.
Coronary malperfusion was associated with the highest incidence of mortality (41.2%; 207 of 502) followed by spinal (34.7%; 35 of 101), mesenteric (34.3%; 193 of 563), renal (26.9%; 274 of 1018), cerebral (26.7%; 260 of 972), and extremity (26.5%; 394 of 1484) (Figures 1, 2B). The involvement of a greater number of tissue regions was also associated with a higher incidence of mortality (Figure 2D). For patients with a single involved region, the operative mortality was 24.1% (368 of 1525), for 2 regions it was 26.8% (209 of 780), and for 3 or more regions it was 36.0% (163 of 453).
In terms of the distal extent of surgery, for the entire study population operative mortality was greatest for those patients undergoing total arch replacement (22.7%; omnibus P = .014) (Figure 4, Supplemental Table 5). Conversely, mortality was similar between partial arch replacement and ascending or hemiarch replacement (16.4% vs 17.1%; P > .99). This was true both for patients with malperfusion (24.8% vs 26.9%; P > .99) and without malperfusion (11.6% vs 13.6%; P = .54).
FIGURE 4.
The effect of distal extent of surgery on mortality in (A) all patients with acute type A aortic dissection (ATAAD), (B) patients with malperfusion, and (C) patients without malperfusion.
ADJUSTED OUTCOMES.
The adjusted effects of pre-operative malperfusion on operative mortality are shown in Figure 5 and Supplemental Table 6. Coronary malperfusion was associated with the highest odds of mortality (odds radio [OR], 2.36; 95% CI, 1.91–2.91; P < .001) followed by mesenteric (OR, 1.84; 95% CI, 1.47–2.31; P < .001), extremity (OR, 1.44; 95% CI, 1.22–1.70; P < .001), and renal (OR, 1.34; 95% CI, 1.11–1.62; P = .002). By contrast, cerebral malperfusion (OR, 1.14; 95% CI, 0.94–1.38; P = .18 and spinal malperfusion (OR, 1.42; 95% CI, 0.88–2.25; P = .15) were not significantly associated with excess mortality. The odds of mortality also increased significantly with increasing number of malperfused regions (1 region: OR, 2.01; 95% CI, 1.74–2.33; P < .001; 2 regions: OR, 2.49; 95% CI, 2.06–2.99; P < .001; 3 regions: OR, 3.42; 95% CI, 2.73–4.27; P < .001) (Supplemental Table 7). Compared with ascending and hemiarch replacement only, partial arch replacement was not associated with increased mortality (OR, 1.01; 95% CI, 0.83–1.22; P = .95), whereas total arch replacement was (OR, 1.65; 95% CI, 1.30–2.07; P < .001), in terms of adjusted outcomes. Compared with use of circulatory arrest without cerebral perfusion, the use of antegrade cerebral perfusion (OR, 0.79; 95% CI, 0.68–0.92; P = .003) and retrograde cerebral perfusion (OR, 0.71; 95% CI, 0.59–0.85; P < .001) both showed similarly reduced association with mortality. Model performance was satisfactory, with a C statistic of 0.73 and a Brier score of 0.13 (Supplemental Figure 2). All final model covariates and associated ORs are presented in Supplemental Table 6 and Supplemental Table 8.
FIGURE 5.
Forest plot demonstrating the adjusted effect of preoperative malperfusion and the extent of surgery on (A) mortality and (B) a composite outcome including mortality, stroke, and renal failure.
COMMENT
In this study of the aortic component of the STS ACSD, among the almost 10,000 patients who underwent repair of ATAAD, malperfusion occurred in almost 30% of patients. We found malperfusion to be associated with significantly increased operative mortality, and this effect varied by region of malperfusion, number of affected regions, and the distal extent of surgery. The highest-risk region of malperfusion was coronary malperfusion, followed by mesenteric.
In our large, contemporary study population from the United States, the malperfusion incidence of 27.7% (2758 of 9958) is comparable to that observed in previous reports from other large ATAAD registries, including 33.6% (717 of 2137) in the German Registry for Acute Aortic Dissection Type A (GERAADA), 33.0% (381 of 1159) in the Nordic Consortium for Acute Type A Aortic Dissection (NORCAAD), and 20.5% (103 of 502) in the Emilia Romagna Cardiac Surgery Registry (RERIC).1,6,8 When stratified by malperfused region, the incidence of coronary (5.0% vs 6.4%, 8.1%, and 10.0%), cerebral (9.8% vs 7.9%, 7.8%, and 11.0%), mesenteric (5.7% vs 2.4%, 3.1%, and 6.0%), renal (10.2% vs 2.8%, 5.5%, and 9.0%), extremity (14.9% vs 6.4%, 17.7%, and 13.0%), and spinal (1.0% vs 1.0%, 2.4%, and 2.0%) malperfusion in our analysis was overall similar to that of RERIC, NORCAAD, and GERAADA, respectively.1,6,8
With respect to cerebral malperfusion, no significant independent association with early mortality was identified in this study. Although conflicting observations have been reported, NORCAAD, RERIC, and others have made findings similar to those of this study regarding cerebral malperfusion.1,5,6 Traditionally, cerebral malperfusion in the setting of ATAAD was thought portend a particularly poor outcome. However, anatomic redundancy in cerebral blood supply likely protects many patients from ischemic injury to the brain despite interrupted flow through 1 or more proximal head vessels. Selection bias in the cases that proceed to open surgery likely also plays a role. Indeed, data from the International Registry of Acute Dissection (IRAD) suggests that as many as one-half of patients with ATAAD who present with cerebral malperfusion do not undergo surgery.15 It is likely that many of the patients who do undergo surgery have less severe presentations, such as radiographic-only malperfusion without significant symptoms. Such bias could not be addressed using the STS ACSD. Nonetheless, data from this large, comprehensive registry study indicate that cerebral malperfusion may lack the strong independent association with early mortality that many surgeons and clinicians have traditionally ascribed to it.
A key feature of the management of patients with ATAAD and malperfusion concerns the distal extent of surgery. Although partial or total aortic replacement is associated with greater technical complexity and longer operative duration compared with ascending aorta or hemiarch replacement alone, a more extensive distal operation could promote better resolution of malperfusion and more favorable long-term remodeling through the distal aorta.16,17 Results of this study suggest that a partial arch replacement (zone 1 or zone 2 arch replacement) is performed without an increase in early mortality in patients both with and without malperfusion, whereas total arch replacement was associated with greater mortality in most patients. These findings do not account for biases related to surgeon comfort and experience and patient anatomy that affect treatment selection. Importantly, previous large studies considering the impact of distal extent of surgery on outcomes in ATAAD have failed to distinguish partial arch replacement from hemiarch replacement and total arch replacement in their analyses.18,19 Important also is the advent of hybrid stent devices as an adjunct to traditional hemiarch replacement. Early reports on such devices are quite promising with respect to outcomes in ATAAD with acute malperfusion.20 Overall, further study is warranted to determine the optimal treatment algorithm for patients with malperfusion, perhaps including prospective clinical trials.
The prevalence of connective tissue disease in the ATAAD cohort was less than 2%. This percentage is lower than has been described elsewhere. For example, the prevalence of Marfan disease in the IRAD cohort is greater than 4%.15 In our experience, the presence of connective tissue disease does affect the operative approach to ATAAD, including in the presence of malperfusion. In general, a more aggressive approach is warranted in these patients, including replacement of the aortic root in most cases, given their relatively young age and high likelihood for progression of aortic disease.
This study has several important limitations. First, it is challenging to make a clear distinction between malperfusion, defined as evidence of compromised blood flow through an aortic branch vessel, and malperfusion syndrome, which includes the signs and symptoms of end-organ ischemia resulting from the disruption of blood flow. Moreover, the timing from dissection onset to restoration of blood flow is not captured and thus could not be included in our analysis. Second, and very importantly, the sample consists entirely of patients who underwent an operation and is therefore subject to selection bias. Some institutions may elect to stabilize or transfer patients before proceeding with repair and incur interval mortality. Failure to capture such patients would likely deflate estimates of the incidence of malperfusion and associated mortality.
In conclusion, in this contemporary population of almost 10,000 patients undergoing ATAAD repair with 1 of more than 3000 surgeons at more than 1000 centers, preoperative malperfusion was present in 28% of cases and was associated with operative mortality of 27%. Outcomes varied widely by the region of malperfusion. Partial arch replacement was not associated with increased operative mortality compared with ascending aorta or hemiarch replacement alone in patients both with and without malperfusion. Overall, these data serve as a benchmark for what can be expected when patients with ATAAD complicated by malperfusion undergo surgical repair and suggest that future strategies to improve outcomes should address malperfusion.
Supplementary Material
The Supplemental Material can be viewed in the online version of this article [https://doi.org/10.1016/j.athoracsur.2025.01.002] on http://www.annalsthoracicsurgery.org.
FUNDING SOURCES
Nicholas J. Goel reports financial support provided by the National Institutes of Health.
DISCLOSURES
Hiroo Takayama reports a relationship with May and Samuel Rudin Family Foundation that includes: funding grants. Joseph Bavaria reports a relationship with W.L. Gore & Associates that includes: consulting or advisory; and with Terumo Aortic that includes: consulting or advisory. Anthony L. Estrera reports a relationship with Artivion that includes: board membership; and with W.L. Gore & Associates that includes: consulting or advisory. Hiroo Takayama reports a relationship with Artivion that includes: consulting or advisory. Eric E. Roselli reports a relationship with Artivion that includes: consulting or advisory; with Cook Medical that includes: consulting or advisory; with Edwards Lifesciences that includes: consulting or advisory; with W.L. Gore and Associates that includes: consulting or advisory; with Medtronic that includes: consulting or advisory; and with Bolton Medical España that includes: consulting or advisory. Himanshu J. Patel reports a relationship with Endospan that includes: consulting or advisory. Himanshu Patel reports a relationship with W.L. Gore and Associates that includes: consulting or advisory; and with Medtronic that includes: consulting or advisory. Michael E. Bowdish serves as Chair of the Society of Thoracic Surgeons Adult Cardiac Surgery Database Task Force and Senior Editor of The Annals of Thoracic Surgery. Bradley G. Leshnower reports a relationship with Endospan that includes: consulting or advisory; and with Medtronic that includes: speaking and lecture fees. Nimesh D. Desai reports a relationship with W.L. Gore & Associates that includes: consulting or advisory; and with Terumo Aortic that includes: consulting or advisory.
REFERENCES
- 1.Zindovic I, Gudbjartsson T, Ahlsson A, et al. Malperfusion in acute type A aortic dissection: an update from the Nordic Consortium for Acute Type A Aortic Dissection. J Thorac Cardiovasc Surg. 2019;157: 1324–1333.e6. [DOI] [PubMed] [Google Scholar]
- 2.Geirsson A, Szeto WY, Pochettino A, et al. Significance of malperfusion syndromes prior to contemporary surgical repair for acute type A dissection: outcomes and need for additional revascularizations. Eur J Cardiothorac Surg. 2007;32:255–262. [DOI] [PubMed] [Google Scholar]
- 3.Augoustides JG, Geirsson A, Szeto WY, et al. Observational study of mortality risk stratification by ischemic presentation in patients with acute type A aortic dissection: the Penn classification. Nat Clin Pract Cardiovasc Med. 2009;6:140–146. [DOI] [PubMed] [Google Scholar]
- 4.Trimarchi S, Nienaber CA, Rampoldi V, et al. Contemporary results of surgery in acute type A aortic dissection: the International Registry of Acute Aortic Dissection experience. J Thorac Cardiovasc Surg. 2005;129: 112–122. [DOI] [PubMed] [Google Scholar]
- 5.Girdauskas E, Kuntze T, Borger MA, Falk V, Mohr FW. Surgical risk of preoperative malperfusion in acute type A aortic dissection. J Thorac Cardiovasc Surg. 2009;138:1363–1369. [DOI] [PubMed] [Google Scholar]
- 6.Pacini D, Leone A, Belotti LMB, et al. Acute type A aortic dissection: significance of multiorgan malperfusion. Eur J Cardiothorac Surg. 2013;43: 820–826. [DOI] [PubMed] [Google Scholar]
- 7.Patrick WL, Yarlagadda S, Bavaria JE, et al. The Penn classification system for malperfusion in acute type A dissection: a 25-year experience. Ann Thorac Surg. 2023;115:1109–1117. [DOI] [PubMed] [Google Scholar]
- 8.Czerny M, Schoenhoff F, Etz C, et al. The impact of pre-operative malperfusion on outcome in acute type A aortic dissection. J Am Coll Cardiol. 2015;65:2628–2635. [DOI] [PubMed] [Google Scholar]
- 9.Chiu P, Tsou S, Goldstone AB, et al. Immediate operation for acute type A aortic dissection complicated by visceral or peripheral malperfusion. J Thorac Cardiovasc Surg. 2018;156:18–24.e3. [DOI] [PubMed] [Google Scholar]
- 10.Grimm JC, Magruder JT, Crawford TC, et al. Differential outcomes of type A dissection with malperfusion according to affected organ system. Ann Cardiothorac Surg. 2016;5:202–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Girardi LN, Krieger KH, Lee LY, et al. Management strategies for type A dissection complicated by peripheral vascular malperfusion. Ann Thorac Surg. 2004;77:1309–1314. [DOI] [PubMed] [Google Scholar]
- 12.Bowdish ME, D’Agostino RS, Thourani VH, et al. The Society of Thoracic Surgeons Adult Cardiac Surgery Database: 2020 update on outcomes and research. Ann Thorac Surg. 2020;109:1646–1655. [DOI] [PubMed] [Google Scholar]
- 13.Jacobs JP, Shahian DM, Grau-Sepulveda M, et al. Current penetration, completeness, and representativeness of The Society of Thoracic Surgeons Adult Cardiac Surgery Database. Ann Thorac Surg. 2022;113:1461–1468. [DOI] [PubMed] [Google Scholar]
- 14.STS SCA Data Specifications v2.9 Training Manual. The Society of Thoracic Surgeons; 2019. Accessed April 1, 2024. https://www.sts.org/sites/default/files/ACSD_TrainingManualV2-9_November2019.pdf [Google Scholar]
- 15.Pape LA, Awais M, Woznicki EM, 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;66:350–358. [DOI] [PubMed] [Google Scholar]
- 16.Yang B, Norton EL, Shih T, et al. Late outcomes of strategic arch resection in acute type A aortic dissection. J Thorac Cardiovasc Surg. 2019;157:1313–1321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Trivedi D, Navid F, Balzer JR, et al. Aggressive aortic arch and carotid replacement strategy for type A aortic dissection improves neurologic outcomes. Ann Thorac Surg. 2016;101:896–903. [DOI] [PubMed] [Google Scholar]
- 18.Berretta P, Patel HJ, Gleason TG, et al. IRAD experience on surgical type A acute dissection patients: results and predictors of mortality. Ann Cardiothorac Surg. 2016;5:346–351. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Conzelmann LO, Weigang E, Mehlhorn U, 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–e52. [DOI] [PubMed] [Google Scholar]
- 20.Szeto WY, Fukuhara S, Fleischman F, et al. A novel hybrid prosthesis for open repair of acute DeBakey type I dissection with malperfusion: early results from the PERSEVERE trial. J Thorac Cardiovasc Surg. Published online August 6, 2024. 10.1016/j.jtcvs.2024.07.059 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.