Abstract 3914: Serious Renal Dysfunction After Percutaneous Intervention Can Be Predicted
Background: A prediction rule for determining the post-PCI risk of developing contrast-induced nephropathy (≥25% or ≥0.5 mg/dL change in creatinine, CR) has been reported. However, little work has been done on predicting patient-specific risk for developing more serious renal dysfunction (SRD: new dialysis, ≥2.0 (mg/dL) absolute increase in CR, or a ≥50% increase in CR). We hypothesized that pre-procedural patient characteristics could be used to predict the risk of post-PCI SRD.
Methods: Data were prospectively collected on a consecutive series of 11,498 patients undergoing PCI in Northern New England from 2003–2005. Patients on dialysis prior to the PCI were excluded from the analysis (n=357), leaving 11,141 in our study cohort. Multivariate logistic regression model was used to identify a combination of patient characteristics most predictive of developing post-PCI SRD. The ability of the model to discriminate was quantified using a bootstrap validated C-Index (Area Under ROC). Its calibration was tested with a Hosmer-Lemeshow statistic.
Results: SRD occurred in 0.74% of patients (83/11,141) with an associated in-hospital mortality of 19.3% v 0.9% in those without SRD. In a multivariate model, pre-procedure patient characteristics associated with SRD included: age≥80, gender, priority, congestive heart failure, diabetes, baseline CR, and intra-arterial balloon pump. Seventy-six percent of the risk was estimated by diabetes, congestive heart failure, and baseline CR. The SRD prediction model was significant with chi-square 210.5, p-value < 0.0001. The model discriminated well, ROC 0.87 (95%CI: 0.82–0.91). The model was well calibrated according to the Hosmer-Lemeshow test. It performed well for relevant patient subgroups: diabetics - ROC 0.69; congestive heart failure - ROC 0.68; and baseline CR - ROC 0.72.
Conclusions: Though infrequent, the occurrence of SRD following PCI is associated with a very high mortality. We have developed a robust clinical prediction rule to determine which patients are at high risk for SRD. Use of this model may help physicians perform targeted interventions to reduce this risk.