To the Editor: It was with great interest that we read the article by Figueras and colleagues. Congratulations to the authors for the well-designed prospective randomized trial. In fact, intraoperative bleeding in liver surgery is still a major concern, and morbidity and mortality are clearly correlated with the amount of blood loss. Vascular clamping during transection of liver parenchyma may reduce bleeding; therefore, the Pringle maneuver has been traditionally applied during hepatectomy to reduce blood loss. Since the prospective evaluation of vascular occlusion in hepatectomy by Man et al, the Pringle maneuver has become an option for performing liver resections with intermittent clamping. The complication rate, hospital mortality rate, and ICG (indocyanine green) retention at 15 minutes on postoperative day 8 were equivalent in the Pringle and the control group. Man et al included all types of hepatectomies in their study and did not distinguish between major and minor hepatectomies. In addition, ischemic preconditioning as a consequence of intermittent Pringle maneuver provides better intraoperative hemodynamic stability and ischemic tolerance, as illustrated by Chouker et al. The beneficial effect of “ischemic preconditioning” might be related to down-regulation of cytotoxic effects of polymorphonuclear leukocytes. However, ischemic damage still remains a major concern, particularly in patients with underlying acute or chronic liver disease with a significant increase in postoperative morbidity and liver dysfunction rates. A recent paper by Kukita et al showed that the Pringle maneuver in hepatectomy caused remnant liver injury in a pig model by expression of iNOS, a marker related to ischemia/reperfusion injury. The ideal alternative would be to perform liver resection without clamping but with modern equipment that prevents or reduces significant bleeding. Nowadays, minor liver resections are a routine operative procedure in a high-volume surgical center. With the development of modern tools for transection-coagulation such as “CUSA” and water-irrigated forceps, as we routinely use in our department, a selective and safe approach with identification and isolation of the vascular structures is guaranteed during transection of the liver parenchyma. We believe, as previously described, that intraoperative ultrasonographic guidance and low venous pressure are effective methods for preventing bleeding with inflow preservation. In a recently published prospective study, Scatton et al successfully performed 53 major liver resections without clamping in 96% of patients with an acceptable complication rate and no reoperations. The mean bleeding volume was 250.1 591.7 mL. There were 3 transient liver dysfunctions in patients with underlying chronic liver disease, who underwent extended hepatectomies. We perform liver resections with a planned portal triad clamping only in emergency situations. We reviewed 10 consecutive minor liver resections out of 45 liver resections performed in the last 6 months at our center. The 10 analyzed cases of minor noncirrhotic liver resection, including benign and malignant indications, had inclusion criteria similar to those described by the authors. After laparotomy, an intraoperative sonography was performed, and mobilization of the liver lobe/segments was followed by parenchyma dissection. Additionally, the cut surface was continuously coagulated using a water-irrigated forceps. Mean ZVD was 8 2 mm H2O at the beginning of liver transection, and on average 270

PURPOSE:To retrospectively assess the outcome of transjugular intrahepatic portosystemic shunt (TIPS) placement in a nonselected group of consecutive patients.MATERIALS AND METHODS:TIPS placement was attempted in 82 patients. Patients were followed up for at least 3 years according to a standard protocol that included repeated shunt evaluations. Fifty-four patients underwent TIPS placement for variceal bleeding, 24 for refractory ascites, and four for other indications. Recurrent bleeding, effect on ascites, long-term patency, development of encephalopathy, and survival and complication rates were evaluated with Kaplan-Meier survival analysis and Cox multivariate analysis.RESULTS:TIPS placement was successful in 75 patients (91%). Mean follow-up lasted 29.4 months. Primary patency was 22% and 12%, primary-assisted patency was 67% and 46%, and secondary patency was 91% and 91% at 1- and 5-year follow-up, respectively. Nonalcoholic liver disease (P =.007) and increasing platelet counts (P =.006) independently predicted development of shunt insufficiency. The 1- and 5-year rates of recurrent variceal bleeding were 21% and 27%, respectively. In the majority of patients with refractory ascites, a beneficial effect of TIPS placement was observed. The risk for encephalopathy was 25% at 1-month follow-up and 52% at 3-year follow-up. The risk for chronic or severe intermittent encephalopathy was 15% at 1-year follow-up and 20% at 3-year follow-up. Serum creatinine levels (P =.001) and age (P =.02) were independent risk factors. Overall survival rate was 61%, 49%, and 42% at 1-, 3-, and 5-year follow-up, respectively. Age (P =.03), serum albumin level (P =.02), and serum creatinine level (P <.001) were independently related to mortality.CONCLUSION:The risk for definitive loss of shunt function was 17% at 5-year follow-up, indicating that surveillance with shunt revision-when indicated-results in excellent long-term TIPS patency. TIPS placement effectively protects against recurrent bleeding.