Acute kidney injury (AKI) after liver transplantation has been reported to be associated with increased mortality. Recently, machine learning approaches were reported to have better predictive ability than the classic statistical analysis. We compared the performance of machine learning approaches with that of logistic regression analysis to predict AKI after liver transplantation. We reviewed 1211 patients and preoperative and intraoperative anesthesia and surgery-related variables were obtained. The primary outcome was postoperative AKI defined by acute kidney injury network criteria. The following machine learning techniques were used: decision tree, random forest, gradient boosting machine, support vector machine, naïve Bayes, multilayer perceptron, and deep belief networks. These techniques were compared with logistic regression analysis regarding the area under the receiver-operating characteristic curve (AUROC). AKI developed in 365 patients (30.1%). The performance in terms of AUROC was best in gradient boosting machine among all analyses to predict AKI of all stages (0.90, 95% confidence interval [CI] 0.86–0.93) or stage 2 or 3 AKI. The AUROC of logistic regression analysis was 0.61 (95% CI 0.56–0.66). Decision tree and random forest techniques showed moderate performance (AUROC 0.86 and 0.85, respectively). The AUROC of support the vector machine, naïve Bayes, neural network, and deep belief network was smaller than that of the other models. In our comparison of seven machine learning approaches with logistic regression analysis, the gradient boosting machine showed the best performance with the highest AUROC. An internet-based risk estimator was developed based on our model of gradient boosting. However, prospective studies are required to validate our results.

When CVCs are inserted to a depth derived by adding the length between the needle insertion point and the clavicular notch and the vertical length between the clavicular notch and the carina, the CVC tip can be reliably placed near the carina level.

P Pu ur rp po os se e: : When patients are moved from the supine to the lateral decubitus position, the double-lumen endobronchial tube (DLT) is often displaced. The aim of this study was to determine whether a DLT is displaced when there are no movements of the head and neck.

Acute hypotension after reperfusion of the liver graft occurs frequently during liver transplantation. A randomized, prospective trial was performed to test the effects of epinephrine and phenylephrine pretreatments for attenuating postreperfusion syndrome (PRS). Ninety-three adult liver recipients were randomly allocated to receive an intravenous bolus of 10 lg of epinephrine, 100 lg of phenylephrine, or normal saline (the control group) at the time of graft reperfusion. The occurrence of PRS, the use of vasoactive drugs, and the postoperative courses were compared. The epinephrine and phenylephrine groups showed PRS less frequently (39% and 48%) than the control group (77%, P ¼ 0.006) as well as higher mean arterial pressures (MAPs) immediately after reperfusion (P < 0.05). An overshoot of MAP was observed in one-third of the pretreated patients with minimal heart rate changes. Only 2 patients in each pretreatment group showed an increase in MAP that was greater than 20% of the baseline value. The intraoperative epinephrine and dopamine requirements were significantly lower in both pretreatment groups. Perioperative laboratory data, postoperative stays, and in-hospital mortality rates were similar for the 3 groups. In conclusion, pretreatment with 10 lg of epinephrine or 100 lg of phenylephrine significantly reduces the occurrence of PRS and vasopressor requirements without immediate or delayed adverse effects in adult liver transplantation. Liver Transpl 18:1430-1439, 2012. V C 2012 AASLD.Received April 30, 2012; accepted July 7, 2012.Acute systemic hypotension frequently occurs immediately after reperfusion of the liver graft during orthotopic liver transplantation surgery. If a greater than 30% decrease in the mean arterial pressure (MAP) lasting more than 1 minute is observed within 5 minutes after reperfusion, postreperfusion syndrome (PRS) is diagnosed. 1 The reported incidence of PRS varies greatly (12%-81%) with the study design. 2-5 Typically, PRS is handled once it occurs instead of being proactively prevented because of its unpredictability and unclear underlying mechanism. However, because of the high incidence of PRS and its associated adverse effects, it seems reasonable to search for preventive measures. 2,4,[6][7][8] The piggyback technique and liver graft flushing are proven surgical prophylaxis methods for reducing PRS. 9-11 Another approach with varying degrees of success is pharmacological pretreatment, which is focused on blocking presumed causes of PRS such as ischemia/reperfusion cascades and their final products. 5,[12][13][14][15] However, previously tested drugs such as nafamostat mesilate, methylene blue, and aprotinin are neither familiar nor currently available to most anesthesiologists. Therefore, the anticipatory treatment of hypotension with vasoactive agents such as epinephrine and phenylephrine (rather than counteracting specific mediators of PRS) seems more practical. 15,16 Unfortunately, the only published study addressing the prophylactic use of adrenergic agonists to preve...