Acute kidney injury following orthotopic liver transplantation: incidence, risk factors, and effects on patient and graft outcomes.

BACKGROUND: Liver transplant recipients frequently develop acute kidney injury (AKI), but the predisposing factors and long-term consequences of AKI are not well understood. The aims of this study were to identify predisposing factors for early post-transplant AKI and the impact of AKI on patient and graft survival and to construct a model to predict AKI using clinical variables. METHODS: In this 5-year retrospective study, we analysed clinical and laboratory data from 424 liver transplant recipients from our centre. RESULTS: By 72 h post-transplant, 221 patients (52%) had developed AKI [according to the Kidney Disease Improving Global Outcomes (KDIGO) criteria]. Predisposing factors for development of AKI were female sex, weight (>100 kg), severity of liver disease (Child-Pugh score), pre-existing diabetes mellitus, number of units of blood or fresh frozen plasma transfused during surgery, and non-alcoholic steatohepatitis as the aetiology of end-stage liver disease (P≤0.05). Notably, preoperative serum creatinine (SCr) was not a significant predisposing factor. After fitting a forward stepwise regression model, female sex, weight >100 kg, high Child-Pugh score, and diabetes remained significantly associated with the development of AKI within 72 h (P≤0.05). The area under the receiver operator characteristic curve for the final model was 0.71. The incidence of new chronic kidney disease and requirement for dialysis at 3 months and 1 yr post-transplant were significantly higher among patients who developed AKI. CONCLUSIONS: Development of AKI within the first 72 h after transplant impacted short-term and long-term graft survival.

Central venous thrombosis and perioperative vascular access in adult intestinal transplantation.

BACKGROUND: Venous access is crucial in intestinal transplantation, but a thrombosed venous system may prevent the use of central veins of the upper body. The incidence of venous thrombosis and the necessity to perform alternative vascular access (AVA) in intestinal transplant recipients have not been fully investigated. METHODS: Records of adult patients who underwent intestinal transplantation between January 1, 2001, and December 31, 2009, were reviewed. Contrast venography was performed as pre-transplantation screening. Vascular accesses at the transplantation were categorized as I (percutaneous line via the upper body veins), II (percutaneous line via the lower body veins), and III (vascular accesses secured surgically, with interventional radiology, or using non-venous sites). Categories II and III were defined as AVA. Risk factors for central venous thrombosis and those for requiring AVA were analysed, respectively. RESULTS: Among 173 patients, central venous obstruction or stenosis (<50% of normal diameter) was found in 82% (141 patients). AVA was required in 4.6% (eight patients: four in each category II and III). Large-bore infusion lines were placed via the femoral arteries in all category III patients without complications. Existing inferior vena cava filter and hypercoagulable states were identified as the risk factors for the use of AVA, but not for central venous thrombosis. Outcomes of patients who underwent AVA were similar to those of patients without AVA. CONCLUSIONS: The majority of adult patients undergoing intestinal transplantation had at least one central venous stenosis or obstruction. The recipient outcomes were comparable when either standard vascular access or AVA was used for transplantation.

Pulmonary thromboembolism during adult liver transplantation: incidence, clinical presentation, outcome, risk factors, and diagnostic predictors.

BACKGROUND: Intraoperative pulmonary thromboembolism (PTE) is an often overlooked cause of mortality during adult liver transplantation (LT) with diagnostic challenge. The goals of this study were to investigate the incidence, clinical presentation, and outcome of PTE and to identify risk factors or diagnostic predictors for PTE. METHODS: Four hundred and ninety-five consecutive, isolated, deceased donor LTs performed in an institution for a 3 yr period (2004-6) were analysed. The standard technique was a piggyback method with veno-venous bypass without prophylactic anti-fibrinolytics. The clinical diagnosis of PTE was made with (i) acute cor pulmonale, and (ii) identification of blood clots in the pulmonary artery or observation of acute right heart pressure overload with or without intracardiac clots with transoesophageal echocardiography. RESULTS: The incidence of PTE was 4.0% (20 cases); cardiac arrest preceded the diagnosis of PTE [75% (15)] and PTE occurred during the neo-hepatic phase [85% (17)], especially within 30 min after graft reperfusion [70% (14)]. Operative and 60 day mortalities of patients with PTE were higher (P<0.001) than those without PTE (30% vs 0.8% and 45% vs 6.5%). Comparison of perioperative data between the PTE group (n=20) and the non-PTE group (n=475) revealed cardiac arrest and flat-line thromboelastography in three channels (natural, amicar, and protamine) at 5 min after graft reperfusion as the most significant risk factors or diagnostic predictors for PTE with an odds ratio of 154.32 [95% confidence interval (CI): 44.82-531.4] and 49.44 (CI: 15.6-156.57), respectively. CONCLUSIONS: These findings confirmed clinical significance of PTE during adult LT and suggested the possibility of predicting this devastating complication.

The hyperventilation of cirrhosis: progesterone and estradiol effects.

Progesterone and estradiol are metabolized in the liver and are elevated in patients with cirrhosis. Progesterone stimulates ventilation by activating progesterone receptors in the central nervous system; estradiol may facilitate progesterone's actions by increasing progesterone receptors. This study evaluated whether progesterone and estradiol contribute to the respiratory alkalosis common in cirrhotic patients. Arterial blood gases and progesterone and estradiol levels were obtained in 50 patients with cirrhosis. Multiple linear regression revealed a statistically significant correlation between PaCO2 and progesterone and estradiol (r = .54, P < .05). Patients with severe hyperventilation (PaCO2 < or = 30 mm Hg) had statistically higher levels of progesterone and estradiol than did patients with mild hyperventilation (30 < PaCO2 < or = 35) or normal ventilation (PaCO2 > 35) (P < .05). Although the progesterone levels were two orders of magnitude lower than those associated with hyperventilation in pregnant patients, the increased ventilatory effect may be because of the altered blood-brain barrier (BBB) present in cirrhotic patients. Progesterone and estradiol appear to contribute to the hyperventilation in cirrhotic patients.