Unweighed groups analysis
Table 1 shows the baseline characteristics of the two groups and the perioperative data. Patients in DHCA+DR group had higher prevalence of diabetes, higher EuroScore II and lower ejection fraction. Total arch replacement and use of frozen elephant trunk (FET) were more frequent in MHCA+ACP. On the contrary aortic valve and mitral surgery were more frequent in DHCA+DR group. CPB time did not differ between groups, but aortic cross clamping time was longer in DHCA+DR group. CA duration was similar in both groups.
Primary endpoints (Table 2). After a median of 14 days (7-34) 14 patients (6.7%) died, 1 (1.7%) in the DHCA+DR group and 13 (8.6%) in MHCA+ACP group, without difference between groups. Causes of death were cardiac in 5 patients and non-cardiac in 9.
NEs prevalence was lower in DHCA+DR group (1.7% versus 10.6%, p=0.034). When only PNDs were considered, there was no difference between the groups (1.7% versus 6%, p=0.193). When ACP was performed, NEs rate was lower when UCP was associated to LCA perfusion, with/our LSA (isolated UCP 16.5%, 12/63, versus 5.1%, 4/78, p=0.025), but PNDs were not different (isolated UCP 8.2%, 6/63, versus 3.8%, 3/78, p=0.258). Seven out of 10 PNDs were embolic of origin, 1 in DHCA+DR and 6 in MHCA+ACP. Death+PND occurred in 21 patients (10%), 2 (3.4%) in DHCA+DR group and 19 (12.6%) in MHCA+ACP group, p=0.046.
KDIGO score was 0 in 83.1% of patients in DHCA+DR group versus 53% in MHCA+ACP (p<0.001). RTT prevalence was higher in MHCA+ACP, 15.2% versus 0, p=0.001. ROC curve analysis showed that, in MHDA+ACP group, RRT was correlated to CA time (cut point 29 min, AUC 0.685, p=0.003) and to CPB time (cut point 127 min, AUC 0.635, p=0.020).
The composite primary endpoint occurred in 3.4% and 21.9% of the patients in DHCA+DR and in MHCA+ACP groups, respectively, p=0.001.
Patients with NEs and need of RRT had higher mortality: 23.5% (4/17, p=0.004) in case of any NE, 30% (3/10, p=0.002) in case of PND and 37.5% (9/24, p=<0.001) in case of RRT.
Secondary endpoints and other results . (Table 2) Time to extubation (590 min versus 938 min, p<0.001) was shorter in DHCA+DR group, as well as ICU LOS (2 versus 4 days, p<0.001), while tracheostomy prevalence was higher in MHCA+ACP (11.3% versus 0, p=0.007). Table 1 supplement shows that time to extubation and ICU LOS remained significantly lower in the DHCA+DR even in patients without composite primary endpoint. Twenty-four-hour bleeding was similar in both groups, but need of transfusion was higher in MHCA+ACP group. It is worthwhile to say that 11.9% of patients in DHCA+DR group and 22.6% in DHCA+ACP group transfused a single unit of blood. In-hospital LOS was similar in both groups.
Weighed logistic regression analysis .
Table 3 compares the risk factors for the primary endpoints, isolated or aggregate, by means of weighed logistic regression analysis. There was no difference between groups in the prevalence of death, any NE and the aggregate of death+PND. On the other side, need of RRT was significantly lower in DHCA+DR groups, as well as the composite primary endpoint. The upper confidence limits were anyway high for all the endpoints due to the non-uniform prevalence of the events explored in the analysis.