The role of central venous pressure and type of vascular control in blood loss during major liver resections.

BACKGROUND: Blood loss during liver resection constitutes the primary determinant of the postoperative outcome. Various techniques of vascular control and maintenance of a low central vein pressure (CVP) have been used in order to prevent intraoperative blood loss and postoperative complications. Our study aims at assessing the effects of different levels of CVP in relation to type of vascular control on perioperative blood loss and patient outcome.

METHODS: The records of 102 consecutive patients who underwent a major hepatectomy were retrospectively analyzed. Forty-two patients were operated on with a CVP of 6 mm Hg or more and 60 patients had a CVP of 5 mm Hg or less. The Pringle maneuver was used in 45 patients and selective hepatic vascular exclusion (SHVE) in 57 patients. Blood loss, complications, and mortality were analyzed comparing the two CVP groups in relation to type of vascular control.

RESULTS: The Pringle maneuver is associated with more blood loss when CVP is 6 mm Hg or more compared with CVP 5 mm Hg or less (1,250 mL [250 to 2,850] versus 780 mL [150 to 3,100]; P <0.05). Conversely, blood loss during SHVE is independent of the CVP levels. A significant difference in blood loss between the Pringle maneuver and SHVE was observed, only when CVP was 6 mm Hg or more (1,250 mL [250 to 2,850] versus 680 mL [150 to 1,260]; P <0.05). Hospital stay was also significantly longer in patients operated on with CVP 6 mm Hg or more (15 days [4 to 38] than in patients with CVP 5 mm Hg or less (10 days [4 to 32]; P <0.05).

CONCLUSIONS: Elevated CVP during major liver resections results in greater blood loss and a longer hospital stay. The Pringle maneuver with CVP 5 mm Hg or less is associated with blood loss not significantly different from that with SHVE. The latter, though, has been shown not to be affected by CVP levels and should be used whenever CVP remains high despite adequate anesthetic management.