Goal-directed fluid therapy for microvascular free flap reconstruction following mastectomy: a pilot study

Abstract

Background

Fluid management of the surgical patient has undergone a paradigm shift over the past decade. A change from ‘wet’ to ‘dry’ to a ‘goal-directed’ approach has been witnessed. The fluid management of patients undergoing free flap reconstruction is particularly challenging. This is typically a long operation with minimal surgical stimulation, and hypotension often ensues. The use of vasopressors in these cases is contra-indicated to maintain adequate flow to the flap. Hypotension is often treated with intravenous fluid boluses. However, aggressive fluid administration to maintain adequate blood pressure can result in flap edema, venous engorgement and, ultimately, flap loss.

Objective

The primary objective of the present study was to determine whether goal-directed fluid therapy, titrated to maintain stroke volume variation ≤13%, with the use of an arterial pulse contour device results in improved postoperative cardiac index (CI) and stroke volume index (SVI) with reduced amounts of intravenous fluid. The primary end points studied were CI, SVI and cumulative crystalloid/colloid administration.

Methods

Twenty female patients undergoing simultaneous microvascular free flap reconstruction immediately following mastectomy were studied. Preoperative and intraoperative care were standardized. Each patient received intra-arterial blood pressure monitoring. In all patients, cardiac output measurement occurred throughout the intraoperative period using the arterial pulse contour device. Control patients had their fluid administered at the discretion of the anesthesiologist (blinded to results from the cardiac output device). Patients in the intervention group had a baseline crystalloid infusion of 5 mL/kg/h, with intravenous colloid boluses to maintain a stroke volume variation ≤13%.

Results

There was no difference in heart rate or mean arterial pressure between groups at the end of the operation. However, at the end of the operation, the intervention group had significantly higher mean (± SD) CI (3.8±0.8 L/min/m² versus 3.0±0.5 L/min/m²; P=0.02) and SVI (51.4±2.4 mL/m² versus 43.3±2.3 mL/m²; P=0.03). This improved CI and SVI was achieved with similar amounts of administered intraoperative fluid (5.8±0.5 mL/kg/h versus 5.0±0.7 mL/kg/h, control versus intervention). The intervention group required less postoperative fluid resuscitation during the early postoperative period (total fluid administered from end of operation to midnight of the operative day, 6.4±1.9 mL/kg/h versus 10.2±3.3 mL/kg/h, intervention versus control, respectively, P<0.01).

Discussion

Goal-directed fluid therapy using minimally invasive cardiac output monitoring resulted in improved end-operative hemodynamics, with less ‘rescue’ fluid administration during the perioperative period.