Jun Gol Song

Association between central venous pressure and blood loss during hepatic resection in 984 living donors

Background: Although low central venous pressure (CVP) anesthesia has been used to minimize blood loss during hepatectomy, the efficacy of this technique remains controversial. We therefore assessed the association between blood loss and CVP during hepatic resection, and examined significant determinants associated with intraoperative hemorrhage during hepatectomy in living donors.

Five-minute parameter of thromboelastometry is sufficient to detect thrombocytopenia and hypofibrinogenaemia in patients undergoing liver transplantation

BACKGROUND Early detection of coagulopathy is important to prevent bleeding during liver transplantation (LT). Rotation thromboelastometry (ROTEM(®)) provides the earliest parameter of clot amplitudes at 5 min (A5). We evaluated whether A5 correlates with platelet count (PLT) and fibrinogen concentration (Fib) and can predict thrombocytopenia and hypofibrinogenaemia in hypocoagulable patients undergoing living-donor LT (LDLT). METHODS A total of 3446 retrospective ROTEM(®) measurements, including 1139 EXTEM, 1182 INTEM, and 1125 FIBTEM, with simultaneously measured PLT and Fib, were analysed during LDLT in 239 patients. The correlations between A5 and maximum clot firmness (MCF) index, PLT, and Fib were calculated. Receiver operating characteristic analysis with area under the curve (AUC) was used to assess A5 thresholds predictive of PLT and Fib. RESULTS The median PLT was 47 000 mm(-3) and the median Fib was 100 mg dl(-1) during LDLT. The A5 parameters of EXTEM (A5EXTEM) and INTEM (A5INTEM) were highly correlated with MCF (r=0.96 and r=0.95, respectively), PLT (r=0.76 and r=0.77, respectively), and Fib (r=0.63 and r=0.64, respectively). A5 of FIBTEM (A5FIBTEM) was also correlated with MCF (r=0.91) and Fib (r=0.75). A5EXTEM thresholds of 15 and 19 mm predicted PLT<30 000 mm(-3) (AUC=0.90) and <50 000 mm(-3) (AUC=0.87), respectively, whereas A5FIBTEM 4 mm predicted Fib<100 mg dl(-1) (AUC=0.86). Biases from A5EXTEM and A5FIBTEM to their MCFs were 16.4 and 1.3 mm, respectively. CONCLUSIONS A5 as an early variable of clot firmness is effective in detecting critically low PLT and Fib. A5 can therefore be a reliable fast index guiding transfusion therapy in hypocoagulable patients undergoing LDLT.

Alterations in QT Interval in Patients Undergoing Living Donor Liver Transplantation

BACKGROUND QT interval prolongation, predisposing to ventricular tachyarrhythmia, has frequently been observed in patients with liver cirrhosis. During liver transplantation (LT) surgery, electrolyte imbalance and hemodynamic instability may affect QT interval changes. We evaluated the alterations in QT parameters at each stage of LT surgery. METHODS We assessed 50 living donor LT recipients without overt heart disease for the corrected QT (QTc) and the interval from peak to the end of the T wave (T(p-e)) automatically using Bazetts formula with LabChart software. QT parameters, laboratory and hemodynamic data were simultaneously collected in the following stages of LT: before anesthetic induction (baseline), pre-anhepatic, anhepatic, 1 hour postreperfusion, and after hepatic artery anastomosis. Recipients were allocated into 2 groups according to their baseline QTc: ≥440 versus <440 msec. RESULTS QTc progressively rose from the pre-anhepatic stage remaining prolonged in each stage of LT surgery compared with the baseline. In the anhepatic stage, 54% of recipients showed marked prolongation of QTc ≥500 msec (522 ± 14), which indicated the potential for a fatal ventricular dysrhythmia: 77% and 36% in groups with QTc ≥440 and <440 msec, respectively. As opposed to changes in QTc, T(p-e) in the anhepatic stage decreased significantly; however, it returned to the baseline level in the neohepatic stage. CONCLUSION A prolonged QTc interval (≥500 msec) was frequently observed throughout the procedure of LT, even among patients with baseline QTc <440 msec, emphasizing the importance of optimizing electrolyte balance and hemodynamic status to reduce greater risk of perioperative arrhythmias.

Comparison of Stroke Volume Variations Derived From Radial and Femoral Arterial Pressure Waveforms During Liver Transplantation

BACKGROUND Stroke volume variation (SVV) is being increasingly used to predict fluid responsiveness. Since radial arterial pressure (RAP) and femoral arterial pressure (FAP) frequently showing discrepancies during liver transplantation (LT), we sought to investigate the effect of differing arterial waveforms on SVV and cardiac output (CO) derived from the Vigileo device, by comparing SVV and CO values derived from RAP (SVV(RAP), CO(RAP)) and FAP (SVV(FAP), CO(FAP)) during LT. METHODS The linear associations and agreements between SVV(RAP) and SVV(FAP) and between CO(RAP) and CO(FAP) were assessed during LT. Hemodynamic variables were measured at nine predefined time points in all 32 recipients, resulting in 288 data pairs. RESULTS Correlations were observed between SVV(RAP) and SVV(FAP) (r = .961) and between CO(RAP) and CO(FAP) (r = .848) at all time points. These correlations between SVV(RAP) and SVV(FAP) (r = .923) and between CO(RAP) and CO(FAP) (r = .902) existed even during the period when mean RAP and FAP values differed (10 minutes after reperfusion). Bland-Altman analysis for SVV(RAP) versus SVV(FAP) and for CO(RAP) versus CO(FAP) showed weak biases (-0.2% and -0.5 L/min) and reasonable limits of agreement (-2.2 to 1.8% and -1.9 to 0.9 L/min). The percentage errors for SVV and CO values were 27.0% and 22.2%. CONCLUSIONS There was no significant difference between SVV(RAP) and SVV(FAP) when measured using the Vigileo device during LT. This finding indicated that SVV obtained using the Vigileo device offered relatively consistent information regardless of the catheterization site.

Factors Associated With Changes in Coagulation Profiles After Living Donor Hepatectomy

BACKGROUND Hepatic resection may be associated with postoperative coagulopathy. However, there is limited information about the predictors affecting coagulopathy after donor hepatectomy. We evaluated the contributors of maximal changes in prothrombin time (PT), activated thromboplastin time (aPTT), and platelet count in the development of postoperative coagulopathy. METHODS We retrospectively analyzed 864 living donors, all of whom received general anesthesia using desflurane, isoflurane, or sevoflurane. A coagulation derangement was defined as one or more of the following events postoperatively: peak PT >1.5 international normalized ratio (INR; highest quartile of PT), peak aPTT >46 seconds (highest quartile of aPTT), or nadir platelet count <100 × 10(9)/L. Factors were evaluated by univariate and multivariate logistic regression analysis to identify predictors of coagulopathy. RESULTS Mean postoperative peak PT, peak aPTT, and nadir platelet count were 1.4 ± 0.2 INR, 43.8 ± 23.7 seconds, and 155.9 ± 37.3 × 10(9)/L, respectively, with 39.4% of donors being at the risk for coagulation derangement. Multivariate logistic regression analysis revealed that predictors of such derangement included anesthesia duration, remnant liver volume, and body mass index (BMI). However, coagulation derangement was not independently associated with age, gender, volatile anesthetics, central venous pressure, fatty change in the liver, estimated blood loss, or intraoperative hypotensive episodes. CONCLUSION We found that long anesthesia duration, low BMI, and small remnant liver volume were predictors of coagulation derangement. These results provide a better understanding of risk factors affecting changes in coagulation profiles after living donor hepatectomy.

Evaluation of Intraoperative Brain Natriuretic Peptide as a Predictor of 1-Year Mortality After Liver Transplantation

BACKGROUND Although brain natriuretic peptide (BNP), a marker of cardiac dysfunction, has been known to predict postoperative mortality, little is known about the postoperative prognostic ability of BNP in liver transplantation (OLT) recipients. We aimed to determine whether intraoperative BNP level can predict 1-year all-cause mortality after OLT. METHODS We retrospectively investigated 525 OLT recipients. BNP and hemodynamic parameters were simultaneously measured 1 hour after induction of anesthesia. Cox regression analysis and receiver operating characteristic curve analysis were performed to determine clinical predictors and optimal cutoff values of post-OLT mortality. RESULTS The 1-year all-cause mortality rate was 9.7% (51/525). Median BNP concentration was significantly higher in nonsurvivors than in survivors (114 vs 56 pg/mL, P < .001). Significant factors in univariate Cox regression analysis were Child-Pugh score, model for end-stage liver disease (MELD) score, logBNP, hemoglobin, creatinine, heart rate, systolic pulmonary arterial pressure, and central venous pressure. In multivariate Cox regression analysis, independent predictors of posttransplant mortality were MELD score and logBNP. However, simultaneously measured hemodynamic parameters did not remain predictors. BNP levels greater than a cutoff of 136 pg/mL (specificity = 83.5%, negative predictive value = 93.6%) were associated with increased post-OLT mortality (log-rank test P < .001). CONCLUSIONS Intraoperative BNP level is an independent predictor of 1-year all-cause mortality after OLT with a high negative predictive value, suggesting that its measurement appears useful in identifying patients at low risk of post-OLT mortality.

Does Stroke Volume Variation Predict Intraoperative Blood Loss in Living Right Donor Hepatectomy

BACKGROUND Although stroke volume variation (SVV) is a valuable index of preload responsiveness, there is limited information about the association between low SVV and increased hepatectomy-related bleeding. We therefore evaluated whether SVV predicts blood loss during living donor hepatectomy. METHODS We evaluated 93 adult liver donors undergoing right hepatectomy for transplantation. Arterial blood pressure, heart rate, body temperature, central venous pressure, SVV, cardiac output, and systemic vascular resistance were measured. Logistic regression and receiver operating characteristic (ROC) curve analyses were performed to determine independent factors and optimal cutoff values of hemodynamic parameters for predicting intraoperative blood loss ≥ 700 mL. RESULTS Of these 93 donors, 36 (38.7%) had blood loss ≥ 700 mL. Univariate logistic regression analysis showed that factors associated with blood loss ≥ 700 mL included heart rate, SVV, cardiac output, and systemic vascular resistance. Multivariate logistic regression analysis revealed that only SVV was an independent predictor of blood loss ≥ 700 mL. ROC curve analysis showed that the optimal cutoff value for SVV predicting blood loss ≥ 700 mL was 6% (area under the curve = 0.64). CONCLUSIONS SVV is a significant independent predictor of blood loss ≥ 700 mL during donor hepatectomy, suggesting that low SVV may provide useful information on intraoperative bleeding in donors undergoing right hepatectomy.

Effect of Right Ventricular Dysfunction on Dynamic Preload Indices to Predict a Decrease in Cardiac Output After Inferior Vena Cava Clamping During Liver Transplantation

BACKGROUND Dynamic preload indices such as stroke volume variation (SVV) and pulse pressure variation (PPV) have yielded false-positive results in patients with right ventricular (RV) dysfunction. We therefore assessed the effect of RV dysfunction on dynamic indices to predict the decrease in cardiac output (CO) during liver transplantation. METHODS Hemodynamic parameters were measured before and after inferior vena cava (IVC) clamping in 52 recipients. The RV dysfunction was defined as an RV ejection fraction (RVEF) ≤ 30%. The area under the receiver operating characteristic curve (AUC) sufficient to detect changes in CO (ΔCO) ≥ 20% after IVC clamping in recipients was calculated. RESULTS Recipients with RVEF ≤ 30% did not show significant increases in SVV or PPV despite having ΔCO ≥ 20%. In recipients with RVEF > 30%, the threshold value and AUC of SVV predicting a decrease in CO were 10% and 0.755 (compared with an AUC of 0.5, P = .011), respectively, whereas those for PPV were 10% and 0.767 (P = .007), respectively. However, in recipients with RVEF ≤ 30%, the threshold value and AUC of SVV were 10% and 0.638 (P = .305), respectively, whereas those for PPV were 12% and 0.684 (P = .159), respectively. CONCLUSIONS These results suggest that dynamic preload indices may not be sufficiently sensitive to detect a CO decrease in liver transplant recipients with RV dysfunction, emphasizing the importance of evaluating RV function when determining the predictability of dynamic indices.

Quantification of Both Platelet Count and Fibrinogen Concentration Using Maximal Clot Firmness of Thromboelastometry During Liver Transplantation.

BACKGROUND Rotation thromboelastometry (ROTEM®) is increasingly used in liver transplantation (LT). Of the ROTEM® parameters, maximum clot firmness (MCF) of EXTEM (MCFEXT) and INTEM (MCFINT) are influenced by both platelet count (PLT) and fibrinogen concentration (FIB), whereas MCF of FIBTEM (MCFFIB) is solely influenced by FIB. We aimed to determine whether using MCFs of thromboelastometry could reliably predict both PLT and FIB and to evaluate their relations in patients with thrombocytopenia and hypofibrinogenemia during LT. METHODS A total of 4100 retrospective ROTEM® assays with simultaneous standard laboratory tests performed during LT were analyzed in 295 patients. The optimal cut-off values of PLT and FIB according to the ROTEM® transfusion guideline were determined by area under the curve (AUC) of receiver operating characteristic (ROC) curve analysis. RESULTS MCFEXT and MCFINT showed good correlation with platelet count (r = 0.79 and 0.80, respectively, P < .001) and with fibrinogen concentration (r = 0.67 and 0.66, respectively, P < .001). MCFFIB and fibrinogen concentration were highly correlated (r = 0.84, P < .001). Additionally, PLT and FIB were calculated mathematically: PLT (/μL) = 14827 + 3.93 (MCFEXT)(2.5); FIB (mg/dL) = 63 + 0.00082 (MCFEXT)(3.0); FIB (mg/dL) = 29 + 13.3 MCFFIB. MCFEXT <35 mm predicted PLT of 43 × 10(3)/μL (AUC = 0.89) and FIB of 91 mg/dL (AUC = 0.78), whereas MCFEXT <45 mm predicted PLT of 52 × 10(3)/μL (AUC = 0.89) and FIB of 121 mg/dL (AUC = 0.86), MCFFIB <8 mm predicted FIB of 128 mg/dL (AUC = 0.94). MCFINT showed almost the same cut-off values as MCFEXT. CONCLUSIONS Both PLT and FIB can be reliably quantified by MCFs of thromboelastometry, reducing the needs for additional laboratory tests to know values of thrombocytopenia and hypofibrinogenemia in patients undergoing LT.

Factors Affecting Intraoperative Changes in Regional Cerebral Oxygen Saturation in Patients Undergoing Liver Transplantation

BACKGROUND Regional oxygen saturation (rSO(2)) is a sensitive marker of cerebral hypoperfusion during liver transplantation. However, bilirubin absorbs near-infrared light, resulting in falsely low rSO(2) values. We sought to determine whether rSO(2) values vary in response to bilirubin concentrations during liver transplantation and to assess whether rSO(2) changes were associated with factors reflecting cerebral oxygen delivery in patients with hyperbilirubinemia. METHODS Measurements of rSO(2) values continuous cardiac output (CO), mean arterial pressure, central venous pressure, body temperature, arterial blood gas analysis, and laboratory parameters were simultaneously performed at 1 hour after the surgical incision (baseline) and at 3 predetermined times during the anhepatic and neohepatic phases in 95 end-stage liver disease patients including 67 males of Child A/B/C/29/29/37 categories respectively. Relationships between changes in parameters were evaluated by correlation and multivariate regression analyses. RESULTS The 273 measurements revealed changes in rSO(2) (range, -18% to 40%) to correlate significantly with alterations in hemoglobin (Hb), serum glucose, lactate, prothrombin time, pH, partial arterial CO(2) pressure (PaCO(2)), and CO, but not with serum total bilirubin (TB). Multivariate linear regression analysis revealed that changes in Hb, CO, PaCO(2), and pH were independent of rSO(2) changes during liver transplantation. CONCLUSIONS Our findings showed that rSO(2) changes were independently associated with factors reflecting cerebral oxygen delivery, such as Hb, CO, PaCO(2), and pH, whereas rSO(2) values did not correlate with changes in bilirubin concentrations, indicating that rSO(2) changes reveal cerebral oxygen balance regardless of TB levels among patients undergoing liver transplantation.