Blood pressure of every patient was monitored using invasive monitorization during the 12 hours before TAVI procedure. Also during this period, mean arterial pressure (mAP) calculated by monitors was entered in patient files. Central venous pressure (CVP) was monitored with catheters implanted by anesthesiologists in all patients before valve implantation. The mPP was calculated using the formula MPP = mAP-CVP.

The Valve Academic Research Consortium-2 (VARC-2) criteria were used to evaluate any complication occuring in TAVI patients^[7^]. VARC-2 recommends that the AKIN system should be used to diagnose AKIN. According to the AKIN system:

AKIN Stage 1: Increase in serum creatinine of 150-199% (1.5-1.99 × increase compared with baseline) or increase of ≥0.3 mg/dL (≥26.4 mmol/L) or urine output <0.5 mL/kg/h for >6 h but <12 h;

Severe AS was diagnosed with echocardiographic methods. The situations of average aortic gradient > 40 mmHg, aortic valve area (AVA) <1 cm², and valve area index (valve area/body surface area) <0.6 cm² were considered to be severe AS^[8^]. The balloon-expandable Edwards Sapien XT valve (Edwards Lifesciences, Irvine, California, USA) was used for TAVI process. Vascular occlusion device (ProStar XL, Abbott Laboratories, North Chicago, Illinois, USA) was used in eligible patients in terms of femoral artery diameter and anatomy. The surgical cut-down method was applied in patients who were unsuitable for using iliac and femoral artery anatomy vascular closure device. Transoesophageal echocardiography and multislice computed tomography (CT) tests were done to determine the diameter of the aortic bioprosthesis implanted. In all patients, clopidogrel 75 mg and acetylsalicylic acid 100 mg were started before TAVI procedure. İohexol (Omnipaque, GE Healthcare), a nonionic low-osmolar monomeric CM, was used as the opaque material. The amount of CM was recorded during all TAVI procedure. Examinations such as CT and coronary angiography that required administration of CM, except TAVI procedures, were performed at least 72 hours before. Daily kidney function tests of all patients were monitored in our centre from admission to discharge. Creatinine levels were checked in the second week, and the first, third, sixth and twelfth month after TAVI procedure (COBAS Integra 400 plus, Roche Diagnostics).

The patients were prospectively followed during one year after TAVI. The informed consent form was obtained from each subject, and the study was conducted in accordance with the Helsinki Declaration. The study protocol was approved by the local ethics committee.

Based on the receiver operating characteristic (ROC) analysis, patients who had lower mPP values (<72 mmHg), which was determined as the threshold value for AKIN development, were included in the high-risk group (HR-G), and those who had higher mPP values (≥72 mmHg) were included in the low-risk group (LR-G). Additionally, characteristics of patients with normal renal functions (NRF) and those who developed AKIN were evaluated.

Statistical analysis was performed using SPSS 17.0 software (SPSS, Chicago, Illinois, USA). The Kolmogorov-Smirnov test was used to assess the normality of distributions. Variables not normally distributed were expressed as medians (interquartile ranges). Normally distributed continuous variables were expressed as a mean ± standard deviation. The means for normally distributed continuous variables were compared by independent-samples t-tests. Skew-distributed continuous variables were compared using a Mann-Whitney U test. Pearsons x² test and Fisher exact test were used to compare categorical variables. Univariate analyses were performed with the variables, such as age (years), left ventricular ejection fraction (%), mean aortic gradient (mmHg), Society of Thoracic Surgeons (STS) score (%), logistic EuroSCORE (%), mPP (mmHg), systolic blood pressure (SBP, mmHg), diastolic blood pressure (DBP, mmHg), DM (%), hypertension (%), coronary artery disease (CAD, %), creatinine (mg/dl), GFR (ml/dk/1.73 m²), amount of CM (ml), pre-TAVI haemoglobin levels (g/dl), and red blood cell transfusion after TAVI (%). The backward stepwise multivariate regression analysis was performed with the variables of mPP, left ventricular ejection fraction, baseline creatinine, GFR, red blood cell (RBC) transfusion, amount of CM, SBP, and DBP; the P values of those was found to be P<0.10, by univariate analyses. A ROC curve analysis was performed to identify the optimal cut-off point of mPP to predict AKIN in patients with severe AS. The area under the curve (AUC) values were calculated as a measure of test accuracy. A two-sided P<0.05 was considered significant within a 95% confidence interval (CI). Kaplan-Meier survival plots were constructed from the index procedure and up to one year after that and compared using the log-rank test. P values <0.05 were accepted as statistically significant.

One hundred and thirty three patients (54.1% females; mean age of 78.1±7.5 years) were enrolled in this study. AKIN was recorded in 30 patients after TAVI (22.6%). Mean values of SBP, DBP, mAP, and mPP were lower in HR-G and AKIN groups. The mean value of CVP was higher in HR-G and AKIN groups. The average mean gradient of the aortic valve was 49.7±11.7 mmHg, the mean AVA was 0.65±0.11 cm², and the mean value of left ventricular ejection fraction was 42 ±14.7%.

A balloon-expandable Edwards SAPIEN XT valve was implanted via transfemoral access in all patients. The mean radiation time was 7.6±3.1 min. Mean CM and duration of procedure were 143.4±22.7 ml and 69.2±28.1 min, respectively. There weren't any significant difference between the two groups according to the numbers and durations of the rapid ventricular pacing.

The values of mPP were significantly lower in AKIN patients (65.9±9.5 vs. 76.3±7.4 P<0.001). Age, gender, DM, hypertension, chronic obstructive pulmonary disease (COPD), CAD, moderate to high grade aortic regurgitation (AR) after TAVI, risk scores for TAVI, haemoglobin, cardiac systolic functions, mean aortic gradient, kidney function tests, amount of CM, previous coronary bypass surgery, and percutaneous coronary intervention (PCI) rates were similar between the two groups. SBP, DBP, mAP, and mPP levels were significantly lower in HR-G group, but CVP level was higher in HR-G group (Table 1).

There was no significant difference between AKIN and NRF groups in terms of CAD, DM, hypertension, and COPD. The AKIN group had longer hospitalisation duration (5.5 vs. 4 days, P=0.014). After TAVI procedure, 2 (2.2%) patients needed permanent pacemaker implantation due to atrioventricular conduction block. Transfusion rates were higher in the AKIN group (13 vs. 22, P=0.032). Patients in AKIN and NRF groups did not show a significant difference in terms of diuretic, renin-angiotensin-aldosterone blocker, and beta-blocker therapy (Table 1). The amount of CM was significantly higher in the AKIN group (148 vs. 138 ml, P=0.028). Eight patients died during the study period. Total mortality was higher in the AKIN group than in the NRF group (16.6% vs. 2.9%, P=0.015; Table 2).

In univariate analysis, mPP, left ventricle ejection fraction, baseline creatinine, GFR, red blood cell transfusion, amount of CM, SBP, and DBP were found to be significantly associated with AKIN (P<0.01 for all parameters). Thus, the multivariate regression analysis was performed with these variables; baseline creatinine, baseline GFR, amount of CM, and mPP were found to be significant predictors of AKIN (Table 3). The ROC analysis of the significant variables in multivariate regression analysis revealed that the cut-off value of mPP was 72 mmHg to predict the development of AKIN (AUC, 0.813; 95% CI, 0.721-0.905; sensitivity, 72%; specificity, 84%; Figure 1). Mortality rates were significantly higher in AKIN patients. Kaplan-Meier survival curves for patients with and without AKIN (NRF group) showed a significantly lower survival rate up to 1 year in the overall AKIN group (16.6% vs. 0.03% log-rank, P=0.02; Figure 2).

Fig. 1

Fig. 2

The main limitation of this study is the small number of patients included. Additional information can be obtained in longer follow-up periods. In our study, TAVI using balloon expandable prosthesis was performed, and it would be useful to conduct similar studies with a self-expandable prosthesis.