J . Blanco-Morillo et al .: J Extra Corpor Technol 2023 , 55 , 30 – 38 33
Figure
3 . Minimized extracorporeal circuit applied during haematic antegrade repriming . Lines ( dimensions ): 1 – Post-reservoir line ( 3 / 8 x 35 cm ); 2 – Pre-oxygenator line ( 3 / 8 40 cm ); 3 – proximal arterial line ( 3 / 8 25 cm ); 4 – medial arterial line ( 3 / 8 55 cm ); 5 – distal arterial line ( sterile ) ( 3 / 8 135 cm ); 6 – distal venous line ( sterile ) ( 3 / 8 105 cm ); 7 – proximal venous line ( 3 / 8 40 cm ); 8 – arterial recirculation line ( clamp included ) 1 / 8 ( double male Luer-lock 70 cm ).
sulfate was administered , aiming to reduce the incidence of fibrillation .
Statistical analysis
Baseline characteristics were described using means and standard deviations for continuous variables , as well as proportions for dichotomous ones , being compared with Student ’ s T- test and Pearson ’ sChi 2 test , respectively .
To estimate the effect of treatment on the outcome variates , a PSM ( with kernel matching ) model was applied [ 17 ]. Thus , a logistic regression model was used to estimate the propensity score to be treated with the HAR technique considering the covariates : sex , weight , age , body surface area , presence of diabetes mellitus , and type of surgery as well as the preoperative hematocrit , left ventricular ejection fraction , stroke and logistic EUROSCORE . Then , the balance between groups was assessed . A covariate was considered as balanced if the standardized difference was < 10 % and the variance rate was within the range of 0.5 and 2 points [ 18 , 19 ]. Afterward , the average treatment effect on treated ( ATT ) was estimated by applying the bootstrap method ( with 100 , 1000 , and 2000 iterations to verify the convergence of the model ) to obtain 95 % confidence intervals [ 17 – 19 ]. Then , the ATT estimation , regarding transfusion and ICU stay , was considered to calculate the economic impact of the HAR methodology . Statistical analyses were performed using Stata v . 14 , StataCorp LP .
Results
Two hundred and twenty-five patients treated with HAR were compared to 210 controls . No significant differences were found between groups in terms of preoperative risk ( Euroscore Log .: CG = 7.5 %, HG = 6.8 %, p = 0.42 ). Despite the differences in sex distribution ( Sex ( male ): GC = 59.5 % vs . HG = 71.1 %; p = 0.01 ), no other significant differences were found in terms of weight , body surface area , preoperative Hb , comorbidities , or type of surgery ( Table 1 ).
Regarding the PSM model , covariates appeared to be properly balanced ( Table 2 ). ATT indicated that the treatment group ’ s exposure to any blood product until discharge was lower ( GlobalBP : CG = 66.75 % vs . GH = 6.88 %, p < 0.001 ). Within the first 24 h , the exposure to RBC ( RBC24 : CG = 52.60 % vs . HG = 5.05 %, p < 0.001 ), plasma ( FP24 : CG = 11.22 % vs . HG = 0.92 %, p < 0.001 ) and platelets ( PT24 : CG = 32.07 % vs . HG = 3.21 %, p < 0.001 ), was significantly lower for treated . In the period within the first 24 h and hospital discharge , requirements of RBC ( RBC > 24 : CG = 38.55 % vs . HG = 4.59 %, p < 0.001 ) and plasma ( FP > 24 : CG = 4.45 % vs . HG = 0.46 %, p < 0.05 ) wasalso found to be lower for patients receiving HAR ( Table 3 ).
The requirement of mechanical ventilation after 10 h was significantly reduced for treated , ( MV > 10 h : CG = 26.51 % vs . HG = 12.62 %, p = 0.005 ). Similarly , extension in ICU stay after 2 days was lowered in the treatment group ( ICU > 2 d : CG = 47.47 % vs . HG = 31.19 %, p = 0.003 ) ( Table 3 ).