Keywords Type A aortic dissection - malperfusion syndrome - ascending aortic replacement
Introduction
Acute Type A aortic dissection (ATAD) is a cataclysmic event requiring emergent surgery.
Perioperative mortality for ATAD is inversely proportional to institutional experience,
ranging from 16.4 to 27.4%, and averaging around 21.6% in the U.S.[1 ]
[2 ]
Malperfusion syndrome significantly compromises the outcomes of ATAD patients and
warrants expeditious diagnosis. Malperfusion usually results from extension of the
dissection flap into the branch vessel, with static or dynamic narrowing or obstruction
of the branch orifice by the flap.[3 ] The subtle nature of compromised end-organ perfusion and ensuing ischemia may result
in diagnostic delays and can result in comparatively higher mortality rates than ATAD
without malperfusion.[4 ] Cases involving renal or mesenteric ischemia are known to have higher (> 50%) postoperative
mortality rates.[3 ]
[4 ]
[5 ] Surgical mortality for patients presenting with any visceral malperfusion has been
recorded to be as high as 43% ± 4%, nearly twice in comparison to the overall ATAD
cohort (25% ± 3%).[5 ]
ATAD complicated by malperfusion usually presents with clinical symptoms of acute
chest pain, syncope, stroke, limb ischemia, abdominal pain, and/or diarrhea.[6 ] However, presumably due to the dynamic nature of end-organ malperfusion, some patients
present without clear signs of malperfusion, and a high index of suspicion is needed
to establish the correct diagnosis in a timely manner, thereby avoiding end-organ
infarction.[7 ] This difficulty in diagnosis and adverse impact on outcome is especially great in
regard to intestinal ischemia. Cerebral and extremity malperfusion are usually more
obvious.
Most surgical centers perform immediate central aortic replacement surgery for ATAD
complicated by malperfusion, reserving peripheral bypass, stent grafting, and fenestration
for patients in whom malperfusion remains uncorrected after central repair. The Michigan
group approaches intestinal malperfusion first, by endovascular means, delaying central
aortic replacement accordingly. Our center applies an acute thoracic emergency protocol
for managing cases presenting to the emergency department (ED) with a high index of
suspicion for aortic dissection/rupture. This protocol prioritizes triage and diagnosis
of such cases to expedite their transfer to the operating room for definitive management.
The importance of emergent surgical intervention for ATAD surgery is well known and
forms the basis of management guidelines for these cases. The aim of this article
is to characterize contemporary outcomes in ATAD patients with and without malperfusion
syndrome who underwent emergent aortic surgeries, with special emphasis on time from
diagnosis to treatment.
Materials and Methods
This study was approved by the Institutional Review Board (HIC#: 2000021950). A single-center,
retrospective review of patients undergoing consecutive ascending aortic replacement
surgery for ATAD from 2008 to 2017 at the Yale New Haven Hospital (YNHH) was performed.
Patients who had complete record of time of admission, time of transfer, and diagnosis
were included in the cohort. Initial screening yielded 128 patients with Type A aortic
dissection and further exclusions were made for patients with chronic ascending aortic
dissection (15 cases), traumatic aortic dissection (7 cases), and incomplete data
(3 cases). The final cohort consisted of 103 acute ATAD cases that were surgically
managed at YNHH. Time of first presentation to a medical facility and time of incision
were recorded for all patients and this interval was defined as the “admission–incision
interval.” Diagnosis of malperfusion syndrome was confirmed using the ED physician's
diagnostic notes and cardiac surgeon's clinical findings based on examination. For
patients who were transferred from outside medical facilities, dissection and malperfusion
was confirmed on their respective primary hospital notes or recorded as an observation
by the surgeon prior to surgical exploration.
Mood's median test, a nonparametric test, was used to compare time from admission
to incision between different groups. Chi-square test was used to examine the difference
in categorical variables between patients with and without malperfusion syndrome.
Multivariate unconditional logistic regression was employed to evaluate the associations
between malperfusion and postoperative outcomes including ventilation over 48 hours,
reoperation for bleeding, postoperative stroke, death during hospitalization, and
30-day mortality. Age, race, sex, and body mass index (BMI) were adjusted in all logistic
models. The log rank method was used to estimate postoperative overall survival between
patients with malperfusion and without malperfusion. All tests were two-sided with
a p -value less than 0.05 being considered significant. All statistical analyses were
performed using SAS 9.4 (SAS Institute Inc., Cary, NC).
Results
The cohort comprised 71 (68.9%) male and 32 (31.1%) female patients. The overall mean
age at the time of surgery was 59.2 ± 14.3 years and BMI was 29.2 ± 6.0 kg/m2 ([Table 1 ]). Mean systolic and diastolic blood pressures measured at initial presentation were
131.4 ± 34.6 and 73.9 ± 22.7 mm Hg, respectively; however, 40 patients were on chronic
beta-blocker therapy or started on anti-impulse therapy at an outside facility.
Malperfusion syndrome was suspected in 29 (28.2%) patients. Distal lower extremity
pulses were not palpable in 48% of these cases. Other patients with malperfusion had
clinical indications of cerebral and/or upper extremity ischemia. Abdominal malperfusion
was noted as a clinical suspicion in patients with acute abdomen coupled with new-onset
diarrhea and abdominal pain.
Overall, 8 (27.5%) patients with malperfusion underwent a “point-of-care” (POC) heart
echocardiogram by an ED physician and their mean admission-to-incision interval was
4.8 ± 3.3 hours compared with 4.1 ± 2.3 hours for malperfused patients who did not
undergo a POC echo (p = 0.51).
Furthermore, 66 patients also had an intraoperative transesophageal echocardiogram
(TEE) performed before initiation of aortic repair. Echocardiogram confirmed presence
of moderate-large pericardial effusion in 17 cases with 4 patients in the malperfusion
group (13.7%) and 13 (17.5%) in the nonmalperfusion group (p = 0.28). The intraoperative TEE confirmed moderate to severe aortic insufficiency
in 24.1% patients with malperfusion syndrome and 25.6% patients without malperfusion
(p = 0.87).
Confirmation of diagnosis was made on chest computed tomography (CT) scans with a
majority (80.6%) demonstrating Type I DeBakey aortic dissection and the remaining
patients Type II dissections. Presence of a dissection flap was confirmed in the ascending
aorta of 94 patients (91.2%), aortic arch of 82 patients (79.6%), descending aorta
of 61 patients (59.2%), abdominal aorta of 56 patients (54.3%), and iliac arteries
of 39 patients (37.8%). The overall mean admission-to-incision interval was 5.8 ± 4.3
hours (median = 5 hours; [Fig. 1 ]) and the mean duration from admission to CT scan was 1.7 ± 2.5 hours (median = 1
hour).
Fig. 1 Admission–incision interval for all patients.
Patients with malperfusion syndrome had a mean admission-to-incision interval of 4.3 ± 2.5
hours (median: 3.6 hours); those without malperfusion had an average admission-to-incision
interval of 6.3 ± 4.6 hours (median: 5.5 hours; p = 0.02).
There was no significant difference in the surgical procedure (i.e., central repair
and then peripheral revascularization if needed) among ATAD cases with and without
malperfusion ([Table 2 ]). All patients with malperfusion syndrome underwent immediate aortic replacement
surgery as the definitive intervention. Thirteen (12.6%) patients underwent concomitant
aortic valve replacement and ascending aortic replacement. Eleven (84.6%) patients
received a biological valve, whereas 2 (15.4%) received a mechanical valve. Eighteen
(17.5%) patients underwent a composite aortic root (Bentall) replacement and the aortic
valve was resuspended in 14 (13.6%) patients. Hemiarch replacement was performed in
88 (85.4%) cases in addition to ascending aortic replacement, and 7 (6.7%) patients
required complete arch replacement.
Table 1
Preoperative characteristics for patients with and without malperfusion syndrome
Variables
Malperfusion, N = 29 (median, %)
No malperfusion,
N = 74 (median, %)
p -Value
Male
20 (69%)
51 (68.9%)
0.99
Age (y)
58.72 ± 12.1 (41)
59.48 ± 15.16 (53.8)
0.81
Systolic blood pressure (mm Hg)
126.24 ± 32.27 (130)
133.53 ± 35.53 (134)
0.34
BMI (kg/m2 )
29.95 ± 6.74 (30.8)
28.98 ± 5.82 (27.2)
0.47
HCT
39.27 ± 7.85 (40.4)
37.66 ± 6.41 (39)
0.33
Preop creatinine
1.19 ± 0.59 (1.1)
1.24 ± 0.90 (1.1)
0.76
Tamponade
7 (24.1%)
19 (25.7%)
0.87
Prior CAD
4 (13.8%)
10 (13.5%)
0.97
Prior Stroke
2 (6.9%)
3 (4.1%)
0.54
COPD
2 (6.9%)
13 (17.6%)
0.16
Rupture of aorta
2 (6.9%)
5 (6.8%)
0.98
Transferred from outside facility
11 (37.93%)
46 (62.2%)
0.02
Admission–CT interval (min)
1.39 ± 1.33 (0.86)
1.89 ± 2.87 (1.25)
0.36
Admission–incision interval (h)
4.32 ± 2.56 (3.56)
6.37 ± 4.68 (5.50)
0.02
Abbreviations: BMI, body mass index; CAD, coronary artery disease; COPD, chronic obstructive
pulmonary disease; CT, computed tomography; HCT, hematocrit.
Table 2
Operative management of patients presenting with and without malperfusion syndrome
Variables
Malperfusion group (N = 29) (%)
Nonmalperfused (n = 74) (%)
p -Value
Root replacement
2 (6.9)
15 (20.3)
0.10
Bentall procedure
5 (17.2)
13 (17.6)
0.96
Valve-sparing procedure
6 (20.7)
8 (10.8)
0.18
Hemiarch replacement
27 (93.1)
61 (82.4)
0.16
Total arch replacement
1 (3.4)
6 (8.1)
0.39
Descending procedure
0
4 (5.4)
0.20
Concomitant CABG
2 (6.9)
5 (6.8)
0.98
DHCA use
20 (69)
37 (50)
0.08
Antegrade cerebral perfusion
8 (27.6)
22 (29.7)
0.82
Retrograde cerebral perfusion
1 (3.4)
14 (18.9)
0.04
CPB time (min)
184.20 ± 46.35 (180)
192.21 ± 50.54 (186)
0.46
X clamp time (min)
90.51 ± 35.34 (86)
111.70 ± 43.40 (108.5)
0.02
Abbreviations: CABG, coronary artery bypass graft; CPB, cardiopulmonary bypass; DHCA,
deep hypothermic circulatory arrest; X clamp, cross clamp.
Unadjusted 30-day mortality for patients presenting with and without malperfusion
was 13.8 and 9.5%, respectively. Multivariate regression analysis (adjusting for age,
sex, BMI, and race) comparing the malperfusion group to patients with no malperfusion
did not reveal a significant difference in perioperative mortality (odds ratio [OR]:
1.53, 95% confidence interval [CI]: 0.40–5.82, p = 0.49) ([Table 3 ]).
Table 3
Postoperative complications in patients with and without malperfusion syndrome
Variables
Malperfusion
No (n = 74)
Yes (n = 29)
p
Unadjusted OR (95% CI)
Adjusted OR (95% CI)[a ]
N
%
N
%
ICU stay (≥5 d)
36
48.6
15
51.7
0.77
1.13 (0.48–2.67)
1.13 (0.47–2.72)
Vent over 48 h
22
29.7
14
48.2
0.07
2.21 (0.91–5.33)
2.15 (0.83–5.56)
Sepsis
4
5.4
1
3.4
1
0.63 (0.07–5.84)
0.55 (0.05–5.45)
Renal failure requiring dialysis
0
0
2
6.9
0.07
Reopen for bleeding
10
13.5
5
17.2
0.62
1.33 (0.41–4.30)
1.43 (0.43–4.70)
Postoperative stroke
6
8.1
5
17.2
0.28
2.36 (0.66–8.45)
2.27 (0.60–8.65)
Postoperative HF
6
8.1
2
6.9
1
0.84 (0.16–4.42)
0.74 (0.13–4.11)
3-y mortality
15
20.2
5
17.2
0.72
0.82 (0.27–2.51)
0.78 (0.25–2.44)
30-d mortality
7
9.4
4
13.7
0.49
1.53 (0.41–5.68)
1.53 (0.40–5.82)
Operative mortality
6
8.1
2
6.9
1
0.84 (0.16–4.42)
0.83 (0.15–4.49)
Reoperation
11
14.8
4
13.7
1
0.92 (0.27–3.15)
0.84 (0.23–3.01)
Redo aortic root operation
3
4.0
2
6.9
0.61
1.75 (0.28–11.07)
2.16 (0.30–15.56)
Abbreviations: BMI, body mass index; CI, confidence interval; HF, heart failure; ICU,
intensive care unit; OR, odds ratio; Vent, ventilation.
a OR adjusted for age, race, sex, and BMI using multivariate regression.
Causes of 30-day mortality for patients with malperfusion were multiorgan failure
(two patients), stroke (one patient), and cardiogenic shock (one patient). Three-year
survival for patients with and without malperfusion on Kaplan–Meier analysis was 79.1
and 74.0%, respectively, with no significant difference in long-term mortality (p = 0.94; [Fig. 2 ]). A proportional hazard analysis did not reveal a significant difference in 3-year
survival for patients with and without malperfusion syndrome (hazard ratio: 1.14;
95% CI: 0.40–3.26).
Fig. 2 Kaplan–Meier (KM) curve demonstrating long-term survival of patients with and without
malperfusion syndrome. No., patients at risk at start of interval; blue, malperfusion;
red, nonmalperfusion group.
Median postoperative intensive care unit stay for all patients was 4 days and the
mean length of hospital stay was 14.3 ± 9 days (median = 9). Five (4.9%) patients
required interval redo aortic surgery (2 root, 2 arch, and 1 descending) during long-term
follow-up.
Discussion
There was no difference in our short- to midterm outcomes for patients with or without
malperfusion syndrome. When compared with the literature, however, the mortality rate
in this cohort is favorable possibly owing to the rapid triage considering that the
only significant difference in management for the malperfusion group was a shorter
admission-to-incision interval.
Current literature describing large multicenter registries reveals > 15% perioperative
mortality for ATAD surgery.[2 ]
[8 ]
[9 ] Most studies have shown temporal improvement in surgical mortality for high volume
aortic centers,[10 ] generally as a result of improved surgical technique, surgeon experience, and technical
advancement over the years.[10 ]
[11 ] However, early mortality after repair of ATAD complicated by malperfusion remains
higher than uncomplicated cases. In a recent, large population-based study comprising
1,159 ATAD patients, Geirsson et al demonstrated using a multivariate regression model
that patients with any type of malperfusion were almost four times more likely to
die in the early (30-day) postoperative period compared with the general cohort (OR:
3.84, 95% CI: 1.87–7.90, p < 0.001).[10 ] Similarly, in a cohort of 197 patients, a multivariate risk model by Rylski et al
demonstrated increased risk of perioperative mortality in patients with ≥ 1 organ
malperfusion (OR: 4.74, 95% CI: 1.63–13.80, p = 0.004).[12 ] Another study, reviewing determinants of adverse outcomes in ATAD surgery found
patients presenting with malperfusion to have higher risk (OR = 2.95, 95% CI: 1.14–7.67,
p = 0.026) of an adverse event (stroke or 30-day mortality).[13 ] Our results demonstrated a nonsignificant difference in early and midterm mortality
provided the diagnosis was established rapidly, allowing for rapid triage to the operating
room.
Some centers, and this is primarily of historical interest, have described patient-specific
staged procedures initially managing malperfusion syndrome (bypass, stenting, and/or
fenestration) and performing a delayed aortic replacement once end-organ ischemia
is resolved.[3 ]
[14 ]
[15 ] Literature describing this algorithm demonstrates equipoise in long-term outcomes
of patients who survived the interval delay in surgery and patients who were operated
immediately.[14 ] However, this approach is associated with interval mortality due to rupture and
ischemia during the intermission between initial intervention and definitive aortic
surgery. On the contrary, none of the patients in our cohort died from frank rupture
of the aorta as a consequence of ATAD.
We practice and advocate immediate central aortic surgery for all patients presenting
with ATAD regardless of malperfusion syndrome. Our results demonstrate that prompt
operation can normalize early and late survival for patients with malperfusion syndrome
to those without malperfusion. Patients with malperfusion were rapidly diagnosed and
triaged to the operating room in our cohort such that their admission-to-incision
interval was significantly shorter than those presenting without malperfusion. This
may be explained by the prompt and early realization of ATAD due to explicit signs
of malperfusion.
Integration of the ATAD protocol was important in generating a coordinated approach
among ED physicians and the cardiac surgery team. The protocol dictated a rapid triage
for patients presenting with a high index of suspicion (chest pain, limb numbness/weakness,
altered mentation, acute abdominal pain/diarrhea), early and consistent use of POC
echocardiogram in the ED, and prioritizing of chest CT scan. A POC echocardiogram
performed by the ED physician is helpful in confirming ATAD for patients who have
low-moderate suspicion and did not appear to significantly increase the admission-to-incision
time interval. Institutional implementation of this protocol was very effective with
the majority of cases being triaged and operated within a 5-hour window after presentation
to the Yale ED ([Fig. 1 ]). The protocol also includes an algorithm to directly transfer patients diagnosed
with ATAD from outside medical facilities to the YNHH operating room.
As the majority of patients in our cohort were operated within a 10-hour window ([Fig. 1 ]), we do not have a wide range of time points to carry out a robust time-to-intervention
based analysis. Further studies with larger sample sizes are required to study the
role of time delay in ATAD surgery to validate the effect of time-to-intervention
on surgical outcomes of patients suffering from malperfusion.
Our sample size was not adequate to demonstrate statistical significance for difference
in early postoperative outcomes. Short-term mortality rates may have achieved significance
with a larger sample size; however, we believe that the results highlight potential
room for improvement in moving all ATAD patients to the operating room in a shorter
time interval as one would for the highest acuity patients.
Conclusions
Our results albeit limited by a small sample size demonstrated a nonsignificant difference
in short-term or midterm mortality rates between the malperfusion group and the nonmalperfused
group. Patients presenting with malperfusion underwent a rapid triage to the operating
room significantly shortening their “admission–incision” interval. This may have been
a major determinant for the comparable outcomes between the two groups; however, further
adequately powered studies are required to elaborate and confirm this potential association.