An 11-year, single-institution analysis of intra-aortic balloon pump use in cardiac surgery.

OBJECTIVES: The primary objective of this study was to analyze perioperative intra-aortic balloon pump (IABP) insertion in patients undergoing cardiac surgery in the authors' institution from 1995 to 2005 and to propose an explanation for changes in use over this period. A secondary objective was to assess patient variables associated with IABP use.

DESIGN: This is a retrospective study including patients who underwent cardiac surgery between 1995 and 2005.

SETTING: The Cardiothoracic Anesthesia Patient Registry of a single teaching institution was queried to obtain the required information.

PARTICIPANTS: Thirty thousand two hundred sixty-nine cardiac surgery patients.

INTERVENTIONS: Intra-aortic balloon pump insertion before surgery, after cardiopulmonary bypass, or in the cardiovascular intensive care unit was assessed in patients who underwent isolated coronary artery bypass graft surgery, valve surgery, or both. Select patient variables were analyzed for their association with IABP insertion. Transesophageal echocardiography (TEE) examinations, milrinone use, and mortality rates also were determined.

MEASUREMENTS AND MAIN RESULTS: Among 30,269 cardiac surgery patients, 1,310 (4.32%) underwent IABP insertion. Combined preoperative, intraoperative, and postoperative IABP use decreased from 7.8% in 1995 to 3.0% in 2005. Simultaneously, the intraoperative use of milrinone increased from 4.8% to 8.8% and postoperative use increased from 5.2% to 7.8%. The number of intraoperative TEE examinations more than doubled from approximately 1,700 to 3,500. The overall mortality for patients with preoperative, intraoperative, and postoperative IABP insertion was 12.6%, 17.5%, and 47.7%, respectively.

CONCLUSIONS: From 1995 to 2005, preoperative, intraoperative, and postoperative IABP use decreased by approximately 60% in cardiac surgery patients. Simultaneously, the use of TEE and milrinone each doubled. Although a cause-effect relationship cannot be established from the present study's observational data, the trends coincide and may be related.