DOI: 10.1097/01.SA.0000307924.32695.f4
,
Issn Print: 0039-6206
Publication Date: 2008/08/01
Facebook
Twitter
LinkedIn
Email
 |
Checking for direct PDF access through Ovid | |
Excerpt
Duminda N. Wijeysundera,*† Keyvan Karkouti,*† Jean-Yves Dupuis,‡ Vivek Rao,§ Christopher T. Chan,∥ John T. Granton,¶ and W. Scott Beattie*
(JAMA, 297:1801-1809, 2007)
*Department of Anesthesia, Toronto General Hospital and University of Toronto, Toronto; †Department of Health Policy Management and Evaluation, University of Toronto, Toronto; ‡Department of Anesthesia, University of Ottawa Heart Institute, Ottawa; §Division of Cardiac Surgery, Toronto General Hospital and University of Toronto, Toronto; ∥Division of Nephrology, University Health Network and University of Toronto, Toronto; and ¶Division of Respirology and Critical Care Medicine, University Health Network and University of Toronto, Toronto, Ontario, Canada
A serious complication of cardiac surgery is acute renal failure, which when severe enough to require renal replacement therapy (RRT) with hemodialysis or continuous venovenous hemodiafiltration, is independently associated with mortality. A risk index that uses preoperative data to predict RRT would improve clinical management and research design. This retrospective cohort study attempted to derive and validate a simple, accurate predictive index for RRT after cardiac surgery at 2 tertiary care hospitals in Ontario, Canada.
Data on patients undergoing cardiac surgery at the Toronto General Hospital and University of Ottawa Heart Institute were prospectively collected. The study sample included consecutive adults, 18 years and older, who underwent cardiac surgery under cardiopulmonary bypass. The derivation cohort included eligible cases at the Toronto General Hospital between May 1999 and July 2004. One validation cohort consisted of cases at the same institution between August 2004 and December 2005. The other validation cohort included eligible cases at the Ottawa Heart Institute between January 1999 and December 2003. Potential predictor variables included demographic characteristics, preoperative renal reserve, comorbid disease, operative characteristics, and urgency of surgery. Renal reserve was measured using estimated glomerular filtration rate (GFR). The primary outcome was RRT, and indications for RRT included metabolic abnormalities, anuria, and fluid overload.
The derivation cohort included 10,751 patients and the Toronto and Ottawa validation cohorts included 2566 and 6814 patients, respectively, for a total of 20,131. Rates of RRT in the three cohorts were 1.3%, 1.8%, and 2.2%, respectively. The multivariable model for predicting postoperative RRT included predictor variables of preoperative estimated GFR, diabetes mellitus requiring medication, left ventricular ejection fraction (LVEF) percentage, previous cardiac surgery, operative procedure, timing of surgery (elective or urgent), and preoperative intra-aortic balloon pump. The model had good discrimination and calibration. A simplified renal index scoring scheme for estimating risk of postoperative RRT included estimated GFR 31-60 mL/min, 1 point and estimated GFR 30 mL/min or less, 2 points. The following all had 1 point assigned: diabetes mellitus requiring medication, LVEF 40% or less, previous cardiac surgery, procedures other than isolated coronary artery bypass graft or isolated atrial septal defect repair, nonelective procedure, and preoperative intra-aortic balloon pump. The proportions of the derivation cohort with scores of 0, 1, 2, 3, 4, and 5 or higher were 15%, 38%, 28%, 13%, 4%, and 1%, respectively. The cohorts differed in perioperative characteristics including estimated GFR, chronic obstructive pulmonary disease, cerebrovascular disease, LVEF, operative procedure, and cardiopulmonary bypass time. Among 53% of patients with scores 1 or lower (low risk), the risk of RRT was 0.4% compared with a risk of 10% among the 6% of patients with scores considered high risk (≥4). The predictive index had areas under the receiver operating characteristic curve in the derivation, Toronto validation, and Ottawa validation cohorts of 0.81, 0.78, and 0.78, respectively.
Readily available preoperative information can be used to predict RRT after cardiac surgery.
