There has been continuing debate about whether use of hetastarch for volume replacement in coronary artery bypass surgery [CABG] increases the risk of postoperative bleeding. A recent meta-analysis of hetastarch use in on-pump CABG concluded that use of hetastarch in these procedures is associated with increased risk, but the safety of hetastarch use in off-pump procedures remains unresolved.

We designed a double-blinded randomized clinical trial to investigate this question. The study was designed as an equivalence trial. Statistical power calculations were performed taking this into consideration. Sealed assignments from a block randomization table developed prior nto initiation of the trial were unsealed in the operating room. These were used to assign patients scheduled for off-pump CABG to receive either 1 L of hetastarch or 1 L of albumin as part of intraoperative volume replacement. Albumin was used for all subsequent intraoperative and postoperative fluid replacement.

The rate of postoperative bleeding was assessed prospectively by monitoring hourly chest tube drainage and number of units of blood products transfused postoperatively in the Intensive Care Unit. Risk was assessed by a Data Safety Monitoring Committee (DSMC) established for this trial. The SAMC was scheduled per protocol to meet after the first 15 subjects (both groups combined) had 1000cc or more of chest tube drainage in the first 12 hours postoperative, and then subsequently either after 15 additional bleeds of this volume or following a schedule set at the discretion of the DSMC.

The trial was continued until 156 patients had been recruited. At that time, 78 participants each had been assigned to the hetastarch and albumin groups. DSMC review at that time determined that use of hetastarch is associated with a risk of postoperative bleeding which is greater than the risk associated with use of albumin and the DSMC accordingly halted the trial.

Recruitment

1. Eligibility screening:

   a. All adult Kaiser Permanente members admitted to Summit Hospital (Oakland, CA) for coronary bypass graft surgery which is planned to be conducted off-pump.

2. Recruitment period: 9/1/2004 through 7/27/05

   1. At recruitment, each subject who agrees to participate is assigned a study number

      1. The study number is assigned sequentially and is totally independent of the medical record number or other personal identifying information
      2. The linkage between the study number and all personal identifying information is kept in a locked file separate from other project data

3. Exclusion criteria:

   1. pregnancy

   2. patients scheduled to undergo on-pump procedures

   3. patients with a history of any of:

      1. cardiac surgery
      2. primary bleeding disorders
      3. end-stage renal disease.

4. Statistical power analysis

   1. This study is designed as an equivalence trial

   2. Target outcome rates are based on those published in Sade RM, Stroud MR, Crawford FA Jr, et al. J Thorac Cardiovasc Surg 1985 May;89(5):713-22.

   3. Statistical power is calculated using the confidence-interval-based technique for normally distributed interval variables.

      1. The confidence interval is set at 95%
      2. The range of equivalence is defined as +100 milliliters
      3. Power is set at 90%

   4. The target sample size which results from this analysis is 165 patients in each arm (i.e., hetastarch and albumin).

5. Analytic strategy

   1. All project analyses are conducted by intention-to-treat

   2. Bivariate comparisons between the study (i.e., hetastarch) and control (i.e., albumin) group measurements use the Fisher exact test or Student's t test for independent samples as appropriate given the nature of the measurement were used to conduct bivariate comparisons of study group measures with control group measures.

      1) Confidence intervals are calculated using the exact techniques

   3. Multivariate comparisons: Analysis of variance and general linear modeling are used to conduct multivariate comparisons of measures between the study and control groups while controlling for covariates.

6. Intervention process

   1. When the patient is in the operating room, the anesthesiologist opens the envelope containing the study group assignment (i.e., hetastarch or albumin).

      1. The attending anesthesiologist administers 1 liter of 6% hetastarch intraoperatively to patients from the study group.
      2. The attending anesthesiologist administers 1 liter of albumin intraoperatively to patients from the study group.

   2. After the first liter of colloid is administered, the patients receive only albumin if more colloid is needed to meet additional requirements for fluid replacement.

   3. All patients receive the remaining standard operative regimen of crystalloids and blood products at the discretion of the anesthesiologist.

   4. All patients are transferred postoperatively to the Intensive Care Unit.

      1. The attending intensivists are blinded to the patient's randomization assignment
      2. The attending intensivists use standard fluid replacement at their clinical discretion.
      3. No hetastarch is administered in the intensive care unit.

7. Measurements

   1. Preoperative

      1. age

      2. gender

      3. body mass index

      4. type of surgery

         a) (elective vs. urgent/emergent)

      5. serum albumin level

      6. serum creatinine level

      7. blood urea nitrogen level

      8. hematocrit

      9. platelet count

      10. prothrombin time

      11. partial thromboplastin time

      12. preoperative administration of anticoagulants

      13. preoperative administration of antiplatelet medications

      14. preoperative administration of thrombolytic agents

   2. Perioperative

      1. identity of the surgeon
      2. identity of the anesthesiologist
      3. lowest body temperature
      4. highest activated clotting time
      5. volume of crystalloids
      6. volume of colloids
      7. volume of cell-saver
      8. number of units of blood products administered

   3. Postoperative

      1. hourly chest tube drainage in the first 12 postoperative hours
      2. volume of crystalloid delivered
      3. volume of albumin delivered
      4. number of units of blood products transfused
      5. whether reoperation for postoperative bleeding was required
      6. duration of mechanical ventilation
      7. duration of stay in the intensive care unit
      8. total length of hospital stay
      9. mortality rates for patients treated in the ICU
      10. mortality rates for patients treated in other inpatient areas of the hospital.

Inclusion Criteria:

• Scheduled coronary bypass graft surgery that is planned to be conducted on adults off-pump (i.e., without use of cardio-pulmonary bypass).

Exclusion Criteria:

• Scheduled coronary bypass graft surgery that is planned to be conducted on adults on-pump (i.e., with use of cardio-pulmonary bypass).
• A history of cardiac surgery
• A history of primary bleeding disorders
• End-stage renal disease