Reliability of a point-of-care device for international normalized ratio testing during the three surgical phases of orthotopic liver transplantation: a retrospective observational study.

PURPOSE: To investigate the reliability of a point-of-care device, the HEMOCHRON(®) Jr. Signature, for measuring the international normalized ratio (INR) during the three surgical phases of liver transplantation. METHODS: A retrospective review was performed on patients undergoing liver transplantation during July to December 2013. Thirty-one patients who had simultaneous laboratory and point-of-care INR readings from each phase of liver transplant surgery (paleohepatic, anhepatic, and neohepatic) were eligible for inclusion. Bland-Altman analysis, Spearman's rank correlation, and four quadrant plots were used to compare INR results from the point-of-care device (pocINR) with those from the laboratory (labINR). RESULTS: Based on the Bland-Altman analysis, mean biases (95% prediction interval) were 0.10 (0.03 to 0.17), 0.19 (0.12 to 0.27), and 0.21 (0.01 to 0.43) for the paleohepatic, anhepatic, and neohepatic phases, respectively. The pocINR device showed a systematic underestimation of the labINR. The Spearman's rank correlation coefficients (95% confidence interval [CI]) were: Ρ = 0.90 (95% CI 0.80 to 0.95); Ρ = 0.92 (95% CI 0.71 to 0.93); and Ρ = 0.71 (95% CI 0.46 to 0.85), respectively. Direction-of-change analysis between the paleohepatic to anhepatic and the anhepatic to neohepatic phases showed strong concordance of 84% and, also considering the small bias between the measurements, supports the use of the pocINR device in the clinical management of liver transplant surgery. CONCLUSION: Point-of-care INR was accurate prior to hepatic reperfusion, but reliability decreased in the neohepatic phase.

Restrictive Versus Liberal Fluid Regimens in Patients Undergoing Pancreaticoduodenectomy: a Systematic Review and Meta-Analysis.

BACKGROUND: Pancreaticoduodenectomy is associated with significant morbidity and mortality which may be influenced by perioperative fluid management. It remains unclear whether liberal and restrictive fluid regimens impact mortality and morbidity in patients undergoing pancreaticoduodenectomy. METHODS: Medline, EMBASE, Cochrane Library and clinicaltrials.gov were searched for studies comparing restrictive and liberal perioperative fluids in patients undergoing pancreaticoduodenectomy. Both prospective and retrospective studies in those undergoing pancreaticoduodenectomy were eligible for inclusion where the patient outcomes were stratified to restrictive and liberal perioperative fluid management regimens, with mortality as the primary outcome. Following study identification, a systematic review and meta-analysis with trial sequential analysis was completed. RESULTS: Thirteen studies including five prospective trials and eight retrospective analyses totalling 3062 patients were included. Restrictive fluid regimens were associated with a significant reduction in mortality compared to liberal fluid regimens for the overall cohort (odds ratio 0.54; 95% CI 0.31-0.94, p = 0.03). There were no significant differences in complication profile. Subgroup analysis revealed this result was contributed to significantly by retrospective studies. The results of the trial sequential analysis suggest this mortality benefit may be due to a type I statistical error and that further patient numbers are required for definitive conclusions. CONCLUSIONS: Restrictive fluid regimens are associated with a reduction in mortality following pancreaticoduodenectomy. The clinical relevance of this finding needs to be interpreted pragmatically given the lack of association with significant causes of morbidity and in considering the results of the recently published RELIEF study.