Pre-Op Index Predicts Renal Failure After Cardiac Surgery

TORONTO, April 24 -- A simplified scoring system using readily available preoperative information predicts acute renal failure and need for replacement therapy after cardiac surgery, researchers here reported.

The predictive index for these heart-surgery patients identified those at high or low risk for renal replacement therapy with hemodialysis or continuous venovenous hemodiafiltration, according to a report in the April 25 issue of the Journal of the American Medical Association.

Among the 6% of patients with high-risk scores, the risk of renal replacement therapy was 25 times greater than that of the 53% of patients with low-risk scores, Duminda N. Wijeysundera, M.D., of the University of Toronto, and colleagues found.

Although high-risk kidney patients constitute less than 2% of patients having open-heart surgery, patients needing postoperative renal replacement therapy use 12% of intensive-care-unit resources, as measured by length of stay, Dr. Wijeysundera said.

Accurate preoperative risk stratification for risk of renal replacement therapy after cardiac surgery would inform clinical decision making and also research design, the investigators said.

To provide accurate and generalizable prognostic information, the predictive index should remain stable when the index is externally validated in different data sets, geographic locales, and time periods, the researchers wrote.

"This simple predictive index may facilitate preoperative risk stratification for [renal risk therapy], and thereby improve clinical decision making, communication of perioperative risk, resource allocation, and research design," they concluded.

To develop a predictive index, the researchers studied a retrospective cohort of 20,131 cardiac-surgery patients at two hospitals in the province of Ontario. Patients 18 or older underwent cardiac surgery under cardiopulmonary bypass.

The derivation cohort included 10,751 patients at Toronto General Hospital (1999-2004).

Two validation cohorts included 2,566 patients at Toronto General Hospital (2004-2005) and 6,814 patients at the Ottawa Heart Institute (1999-2003). The validation hospitals differed in locale and data-collection methods.

Renal replacement rates in the derivation cohort and the Toronto and Ottawa validation cohorts were 1.3% (N= 139), 1.8% (N=45), and 2.2% (152), respectively, the researchers reported.

The multivariable predictors of renal replacement therapy included eight predictors and were scored from zero to eight points. These included an estimated glomerular filtration rate less than or equal to 30 mL/min, which was assigned two points.

Other components, assigned one point each, were estimated glomerular filtration rate 31 to 60 mL/min, diabetes requiring medication, ejection fraction less than or equal to 40%, previous cardiac surgery, procedure other than coronary-artery-bypass grafting, intra-aortic balloon pump, and being a nonelective case.

Among the 53% of patients with low-risk scores (?1), the renal-replacement risk was 0.4%. By comparison, this risk was 10% among the 6% of patients with high-risk scores (?4), the researchers reported.

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.

When these cohorts were stratified on the basis of index scores, likelihood ratios for renal replacement were more concordant than the observed rates.

The predictive index of this simplified scoring system, which used readily available, clinically sensible preoperative information, compared favorably with other perioperative risk indices with regard to discriminating between high- and low-risk patients, the researchers said.

The principal strength of the index is its combination of simplicity and accuracy. The index used only eight components, almost all equally weighted, which should promote its use in clinical care and research settings, the investigators said.

It also differed from recent comparable indices by measuring preoperative renal function using estimated glomerular filtration rates, found in studies to be superior to serum creatinine as a predictor of clinical outcomes, they said.

Low-risk kidney individuals, about 55% of cardiac surgery patients, would require no specific modifications in usual management. Intermediate-risk patients might benefit from limited use of renal-protective interventions, for example, strict control of intraoperative hematocrit and off-pump surgery.

High-risk patients might benefit from more aggressive use of potential renal-protective interventions. In addition, the researchers said, identifying these high-risk patients might assist intensive care units in planning use of post-op renal-therapy resources.

Clinical trials, they said, might improve selection of intermediate-to-high-risk patients by using this index in their inclusion criteria.

The researchers noted several study limitations. The association between glomerular filtration and renal replacement observed in this study, they said, should not be extrapolated to alternative prediction equations, such as the Modification of Diet in Renal Disease formula or cystatin-C-based equations.

In addition, they said, the data were limited to in-hospital outcomes, therefore precluding long-term implications. Given that starting renal replacement therapy is based on clinical judgment, consulting nephrologists might have also modified their threshold for in-hospital renal therapy on the basis of their clinical judgment.

Finally, the investigators said, the index should be validated in other geographic regions to further characterize its generalizability.

The calibration of this index across cohorts improved generalizability when used to estimate likelihood ratios for renal replacement as opposed to predicted event rates, the researchers said.