The best magazine
Discussion
The key finding of our work is that the traditional definition of CIN is preferable to the new definition that has been used in most clinical trials. We evaluated the 2 most commonly used definitions of CIN in clinical and research settings and demonstrate that extending the traditional definition by adding a cohort of patients who had a ≥0.25% relative increase in serum Cr did not help identify patients who were at a greater risk for adverse events. Because the traditional definition captures those patients who meet the new definition and have worse outcome, there is little reason to consider a change to this paradigm.
Our work corroborates and significantly extends prior work in the field. Classical studies have used ≥0.5mg/dL as the definition of CIN. In the natural history study by Rihal et al, the incidence and risk factors for acute renal failure (ARF) after coronary intervention were evaluated in a single-center cohort study. Using the definition of an absolute rise of ≥0.5 mg/dL, 3.3% of patients were found to experience ARF after PCI. Individuals experiencing ARF were at increased risk for mortality during the hospitalization or during follow-up. The definition of acute kidney injury using absolute vs relative increases in serum Cr has recently been examined. This study evaluated absolute and relative changes in serum Cr in clinical scenarios with different baseline renal function using differential equations. The time over which the serum Cr rose was also evaluated as a variable, depending on the baseline renal function as well. The simulation studies performed by these authors support our use of absolute rise as a better marker than relative change in Cr. This is especially true when dealing with a population with baseline renal dysfunction, which is a major risk factor for CIN, who experience transient reductions in Cr clearance, which is often the case in contrast exposure. Recently, Harjai and colleagues evaluated the impact of CIN on 6-month major adverse cardiovascular events and all-cause mortality using contemporary definitions of CIN in a single-center post-PCI. This study excluded subjects who experienced in-hospital mortality. Rates of CIN were similar to our findings, with 5.2% experiencing a ≥0.5-mg/dL rise in serum Cr. A rise in Cr of ≥0.5 mg/dL was found to have the largest prognostic impact on 6-month major adverse cardiovascular events and all-cause mortality. Similarly, Budano et al evaluated 2 definitions of absolute and relative change in Cr at predicting outcomes in a single center after both diagnostic angiography and percutaneous intervention. These authors also found that an absolute increase of ≥0.5 mg/dL was a better predictor of short- and long-term outcomes than a ≥0.25% increase in Cr, which further questions the clinical relevance of the second group. Our study was larger and involved patients treated at multiple institutions and by a large number of operators, thus increasing its generalizability.
The new definition of CIN has been routinely used by investigators evaluating different strategies for reducing CIN. Multiple trials and meta-analyses have suggested that CIN can be reduced by hydration with sodium bicarbonate–based solutions and N-acetylcysteine, but none have shown a reduction in NRD. It is possible that disparities in the definition of CIN used in different studies could be, in part, blamed for this. An interesting disconnect between a reduction in CIN and a lack of reduction in hard events has emerged, with none of these studies individually or in meta-analysis demonstrating a beneficial reduction in NRD, whereas reductions in CIN have been consistently demonstrated. Furthermore, these studies have verified marked heterogeneity, and the ideal therapeutic strategy for prevention of CIN continues to be debated. It is likely that variability in the choice of definitions used is, in part, responsible for this heterogeneity and the lack of a reduction in NRD despite often dramatic reductions in CIN. These and studies like them should be reanalyzed using a ≥0.5-mg/dL rise in Cr as the definition of CIN to allow better prediction of hard clinical end points. If practitioners, researchers, and regulatory agencies adopted this common definition, future studies would be better primed to make an impact on patient outcomes and institutional policies.
We further evaluated the optimal cutoff for a rise in serum Cr, and our analysis suggests that a shift to a more sensitive cutoff would result in a decrease in specificity and may have the potential to adversely impact health care use without clearly impacting hard events. However, unlike the impact of using a relative increase in Cr, the use of this alternate definition does increase the sensitivity for in-hospital death, albeit without any change in identifying patients with NRD. Further study is warranted to determine if ≥0.4 mg/dL is a more appropriate clinical definition. Other determinants of acute kidney injury have been evaluated and may prove better than absolute change in Cr at predicting and, ultimately, defining acute kidney injury. Further investigation of these molecules (cystatin C, neutrophil gelatinase-associated lipocalin, and kidney injury molecule-1) is necessary in patients exposed to intravenous contrast. If validated, these markers could serve as the ideal surrogates for study of investigational agents being evaluated for reduction in CIN.
Our findings, thus, have 2 important implications. First, they provide evidence that validates the long-standing definition of CIN based on measurable and meaningful clinical outcomes. This provides an opportunity for reanalysis of current data as well as standardization for future study. Second, our study should be of interest to regulatory agencies and investigators evaluating strategies for reduction in CIN. Consistent definition and reporting of CIN will ensure that continued progress in the field is accompanied by clinically meaningful returns.
Source: ...