Page 86 - Read Online

Nardi et al. Postoperative malperfusion in aortic dissection

Table 2: Surgical variables
Variable Overall (n = 214) non-PM (n = 159) PM (n = 55) P value
Arterial cannulation, n (%)
Right axillary 96 (44.9) 69 27 NS
Right or left femoral 96 (44.9) 72 24 NS
Central (into the aorta) 22 (10.2) 18 4 NS
Brain perfusion, n (%)
Monolateral 53 (24.7) 38 15 NS
Bilateral 65 (30.4) 48 17 NS
Not performed 96 (44.9) 74 22 NS
Circulatory arrest, n (%) 125 (58.4) 90 35 NS
Circulatory arrest temperature (°C), mean ± SD 27 ± 2 26.9 ± 2.5 27.1 ± 1.9 NS
Surgical times (min), mean ± SD
CPB 161 ± 82 160.9 ± 87.5 168.8 ± 67.6 NS
Aortic cross clamp 91 ± 47 92.2 ± 46.4 95.2 ± 46.4 NS
Circulatory arrest 38 ± 31 37.6 ± 32.4 39.5 ± 30.8 NS
Surgical procedures, n (%)
AAR 72 (33.6) 53 19 NS
AAR + hemiarch replacement 50 (23.4) 34 16 NS
AAR + arch replacement 38 (17.8) 29 9 NS
Bentall 43 (20.1) 36 7 NS
Bentall + hemiarch replacement 11 (5.1) 7 4 NS
PM: postoperative malperfusion; CPB: cardiopulmonary bypass; AAR: ascending aorta replacement; NS: not significant
Three regions of the aorta (ascending, arch and detect independent predictors of in-hospital mortality
proximal descending) were investigated in each patient and PM logistic regression analyses were performed.
to identify sites of intimal tearing; tears were resected Statistical significance by univariate analyses (P <
whenever possible. Intimal tearing occurred in the 0.10) were required for entry into the multivariate
ascending aorta in 115 (53.7%) patients, in the arch models. Preoperatively analyzed variables included:
in 33 (15.4%) patients, and in the descending aorta in age, gender, arterial hypertension, smoking status,
8 (3.7%) patients. In 58 (27.1%) patients no entry tear body mass index, history of concomitant coronary
was found. Seventy-two patients (33.6%) underwent artery disease, diabetes, chronic kidney disease,
isolated ascending aortic replacement, 50 (23.4%) had previous cardiac surgery, left ventricular dysfunction
hemiarch resection and 38 (17.8%) had both total arch [e.g. left ventricular ejection fraction (LVEF) < 40%],
replacement and ascending aortic replacement. The presence of preoperative malperfusion, and the sites
aortic root was replaced in 43 (20.1%) patients using of entry tears. Intraoperatively analyzed variables
the modified button Bentall technique. Intraoperative included: cannulation sites, techniques for obtaining
data are given in Table 2. brain perfusions, surgical times, whether there was
a need for circulatory arrest, and types of operations
Data collection (i.e. isolated ascending aorta replacement, Bentall
In-hospital mortality events included both intraoperative procedure, ascending aortic replacement extended to
and postoperative mortality within 30 days after surgery. the hemiarch, or total arch replacement). Overall long-
Clinical follow-up visits were performed every 12 term survival (not including operative mortality) was
months in our outpatient control unit; CT-angiography expressed as a mean of the values plus or minus one
and/or echocardiographic data were collected. For standard deviation. Survival analyses were computed
patients living far from this institution who could not using the Kaplan-Meier method; Mantel-Cox log-
participate in regular follow-up visits to the department, rank tests were used to compare survival estimates
clinical status was ascertained by personal interviews between subgroups. Cox regression models were
with the patients and their cardiologists, including the used to evaluate the influence of the variables on time
recording of noninvasive tests. to death in the entire cohort. A P value less than 0.05
was considered statistically significant.
Statistical analyses RESULTS
Statistical analyses were performed using StatView
4.5 programming (SAS Institute Inc., Abacus At presentation, 119 (55.6%) patients exhibited
Concepts, Berkeley, CA, USA). Student’s t-test for preoperative malperfusion: 68 (31.8%) were cerebral
continuous variables and Chi-squared or Fisher’s with neurological symptoms, 38 (17.7%) were renal,
exact tests for categorical variables were used. To and 13 (6.1%) were mesenteric. The most relevant
Vessel Plus ¦ Volume 1 ¦ June 27, 2017 79