Early acute kidney injury after transcatheter aortic valve implantation: predictive value of currently available risk scores.

BACKGROUND: Acute kidney injury (AKI) after transcatheter aortic valve implantation (TAVI) is a frequent complication associated with adverse outcomes and mortality. Various scores have been developed to predict this complication in the coronary setting. However, none have ever been tested in a large TAVI population. This study aimed to evaluate the power of four different scores in predicting AKI after TAVI. METHODS: Overall, 1535 consecutive TAVI patients from the observational multicentric "Magna Graecia" TAVI registry were included in the analysis. Of the study population, 235 (15.31%) developed AKI early. The Mehran, William Beaumont Hospital, CR4EATME3AD3, and ACEF scores were calculated retrospectively. RESULTS: The patients who developed TAVI-related AKI had significantly higher absolute values of all risk scores than those who did not. The receiver-operating characteristic analysis also showed a significant correlation between these four scores and AKI, but without a significant difference among all of them (p value = 0.176). Nevertheless, based on their area under the curve values (≤0.604 for all), none had adequate diagnostic accuracy in predicting TAVI-related AKI. Importantly, multivariate analysis identified myocardial revascularization close to the TAVI procedure and implantation of self-expanding prostheses, as well as atrial fibrillation, low-osmolar contrast media administration, corrected contrast medium volume, and any transfusion (p value < 0.05 for all) as independent risk factors for AKI. CONCLUSIONS: Although high values of current AKI risk scores are significantly associated with the development of this complication, these are not sufficiently accurate. Further studies are needed so that a TAVI-dedicated AKI risk score may be created.

Hidden Coronary Atherosclerosis Assessment but Not Coronary Flow Reserve Helps to Explain the Slow Coronary Flow Phenomenon in Patients with Angiographically Normal Coronary Arteries.

The significance of the slow coronary flow phenomenon (SCFph), as visualized in patients (pts) with angiographically normal coronary arteries, is controversial. Absolute coronary flow reserve (CFR) in the left anterior descending coronary artery (LAD), non-invasively assessed by a transthoracic color-guided pulsed-wave Doppler (E-Doppler TTE), is a reliable parameter to assess coronary microcirculatory dysfunction (CMD). Mild and angiographically hidden epicardial atherosclerosis (Hath), as visualized by intracoronary ultrasound (IVUS), which could be the clue to atherosclerotic coronary microvascular involvement, has never been investigated together with CFR in patients. This study was aimed at assessing the value of CFR and HA in explaining the SCFph. Methods. Both non-invasive assessment of CFR in the LAD and corrected TIMI frame count assessment of the coronary contrast runoff were performed in 124 pts with angiographically normal coronary arteries. Among the whole group, 32 patients also underwent intracoronary ultrasounds in the LMCA and LAD, and the maximal plaque burden was assessed (Lesion external elastic (EEM) cross sectional area (CSA)­Lesion Lumen CSA/Lesion EEM CSA * 100). We found that 24 of the 124 pts (group 1) had the SCFph and the remaining 100 had a normal runoff (group 2). CFR, evaluated in both groups, was not significantly different, being 2.79 ± 0.79 (Mean ± SD) in group 1 and 2.90 ± 0.8 in group 2 (p = ns); in the pts also examined by IVUS (32 pts), the SCFph was always associated with hidden atherosclerosis, and a plaque burden of ≥33%. On the contrary, in the normal runoff group, any grade of PB was observed (from no athero to a PB > 70%) and remarkably, 10 pts had no signs of athero or just a minimal plaque burden. This resulted in a ROC curve analysis in which PB < 33% had a high negative predictive value (100%) in ruling out the SCFph. In addition, considering a CFR value < 2.21 as an index of coronary microcirculatory dysfunction, we found CMD in 15 pts (15%) in group 1 and in 7 pts (29%) in group 2 (p = ns). In conclusion, the SCFph is strongly connected to epicardial athero to the extent that the absence of hidden coronary athero has a very high negative predictive power in ruling out SCFph. CFR that is based on an endothelium-independent mechanism remains fairly normal in this condition. An endothelium-dependent microcirculatory constriction at rest due to atherosclerotic involvement of the coronary microvascular network is a possible explanation of the SCFph.