Acute kidney injury (AKI) after transcatheter aortic valve replacement (TAVR) has been associated with increased postoperative morbidity and mortality. Long-term outcomes after TAVR with the Edwards SAPIEN valve in patients who develop AKI postoperatively are currently not well described. We retrospectively reviewed 384 consecutive patients undergoing TAVR at 2 institutions from August 2006 to April 2012. AKI was defined and staged according to Valve Academic Research Consortium-2 criteria. The incidence, multivariate predictors, and association of AKI with 3-year mortality were evaluated. Stage 1 AKI occurred in 24.0% of patients (92 of 384), stage 2 in 5.5% (21 of 384), and stage 3 in 8.1% (31 of 384). The overall operative mortality rate was 7.6%, with a mortality of 3.0% in patients with no kidney injury, 7.6% in stage 1, 23.8% in stage 2, and 32.3% in stage 3. The incidence of new postoperative dialysis was 3.1%. Survival at 3 years for no-AKI/stage 1/stage 2/stage 3 was 59.2 ± 3.3%, 43.4 ± 5.2%, 27.8 ± 10.0%, and 25.4 ± 7.9%, respectively. Logistic regression modeling for the combination of stage 2 or 3 AKI after surgery demonstrated that the last preoperative creatinine (for each 1 mg/dl increase, odds ratio = 3.23, 95% CI 1.83 to 5.69; p <0.001) and dye load (for each 10 ml increase, odds ratio = 1.04, 95% CI 1.01 to 1.08; p = 0.006) were significant predictors for AKI. In conclusion, AKI after TAVR is associated with increased postoperative and 3-year mortality. Significant multivariate predictors are potentially modifiable before the procedure.
Up to 30% of patients with severe aortic stenosis (AS) are not eligible for conventional surgical aortic valve replacement because of high surgical risk. Transcatheter aortic valve replacement (TAVR) with the Edwards SAPIEN valve (Edwards Lifesciences, Irvine, California) is approved in the US for the treatment of patients with severe AS who are inoperable because of high surgical risk. It is well established that acute kidney injury (AKI) after any cardiac surgery is associated with increased morbidity and is also an independent predictor of short- and long-term mortality. In an effort to further standardize clinical research to improve comparability and interpretability of study results, the Valve Academic Research Consortium (VARC) recently updated and modified key consensus definitions of important clinical end points for TAVR, including the definition of AKI. Although AKI has been demonstrated as a strong, independent predictor of short-term mortality after TAVR, long-term association between AKI and mortality are not as well studied, particularly when applying VARC-2 criteria. Therefore, the goal of our study was to determine whether an association between AKI as defined by VARC-2 criteria and long-term mortality could be established in patients who underwent TAVR with the Edwards SAPIEN valve.
Methods
A retrospective analysis of 384 TAVR cases performed at Medical City Hospital, Dallas, Texas from August 2006 to April 2012 and at The Heart Hospital Baylor Plano, Plano, Texas from January to August 2012 was performed. The study was approved by both hospitals’ institutional review boards. The incidence of AKI, as defined and staged by VARC-2 criteria ( Table 1 ), was evaluated. Multivariate predictors of AKI and association of AKI with 3-year mortality were examined. Operative mortality was defined as (1) death within 30 days after surgery and (2) death after 30 days during the same index hospitalization.
|
|
|
|
|
|
∗ Patients receiving renal replacement therapy are considered to meet stage 3 criteria irrespective of other criteria.
Summary data are presented as percentages with counts and as means with SDs. These data were analyzed using SAS 9.4 (SAS Institute, Cary North Carolina). The patients in the study were being followed after TAVR and mortality data were available from personal contact, family contact, and clinical providers. Regular scans of obituary columns supplemented data collection. The patient names were also tested against the Social Security Death Master File.
Survival curves to 3 years were created using Kaplan–Meier statistics and odds ratios calculated from logistic regression. A multivariate logistic regression was used to determine if there were preoperative risk factors that predicted an increase in the stage of the acute kidney disease. The patients with “no-AKI” were combined with the “stage 1” patients, and the “stage 2” and “stage 3” patients were grouped together. The logistic regression was used to test for an increase in severity from stage 0/1 to stage 2/3. The independent variables in the model included patient age (years), gender, type of TAVR procedure (transapical [TA] or transfemoral [TF]), amount of dye used during each case (ml), last preoperative creatinine value (mg/dl), and estimated glomerular filtration rate (eGFR; ml/min/1.73 m 2 ).The eGFR was calculated using the Modification of Diet in Renal Disease formula.
Results
The mean age of our study population was 82.13 ± 8.1 years, with 48.7% (182 of 384) men and a mean Society of Thoracic Surgeons Predicted Risk of Mortality (STS-PROM) of 9.42 ± 5.69%. In 60.4% (232 of 384) of the patients, TAVR through a TF access was used (TF-TAVR), the remaining patients underwent TA-TAVR (TA-TAVR). The overall incidence of AKI was 37.5% (144 of 384). Stage 1 AKI occurred in 24.0% of patients (92 of 384), stage 2 in 5.5% (21 of 384), and stage 3 in 8.1% of patients (31 of 384). The incidence of AKI was higher in patients with increased STS-PROM, a history of peripheral vascular disease, TA-TAVR, preoperative renal failure, and increased preoperative creatinine levels. Other preoperative factors did not significantly differ between patients developing AKI and patients without AKI ( Table 2 ). Furthermore, volume of contrast agent used was not significantly different between the groups.
| Variable | VARC-2 Classification of Kidney Disease | p-Value | |||
|---|---|---|---|---|---|
| No AKI (n = 230) | Stage 1 (n = 92) | Stage 2 (n = 21) | Stage 3 (n = 31) | ||
| Age (years) | 81.96 ± 8.65 | 82.49 ± 6.83 | 82.05 ± 5.68 | 82.13 ± 9.30 | 0.963 |
| Men | 112 (49%) | 47 (51%) | 7 (33%) | 16 (52%) | 0.513 |
| Ejection Fraction (%) | 53.95 ± 12.76 | 54.10 ± 11.98 | 54.52 ± 13.24 | 54.23 ± 8.45 | 0.997 |
| Height (cm) | 167.21 ± 10.68 | 166.08 ± 10.44 | 163.96 ± 9.92 | 167.50 ± 8.89 | 0.544 |
| Weight (kg) | 73.59 ± 18.16 | 75.96 ± 19.81 | 73.05 ± 18.60 | 76.15 ± 15.99 | 0.686 |
| STS-PROM | 8.61 ± 4.82 | 9.96 ± 4.96 | 8.95 ± 3.48 | 13.73 ± 11.44 | < 0.001 |
| Mean Aortic Valve Gradient (mmHg) | 46.25 ± 14.69 | 44.72 ± 14.61 | 47.24 ± 14.24 | 41.90 ± 13.59 | 0.393 |
| Aortic Valve Area (cm 2 ) | 0.60 ± 0.15 | 0.58 ± 0.15 | 0.58 ± 0.12 | 0.62 ± 0.14 | 0.807 |
| Transfemoral TAVR | 158 (69%) | 45 (49%) | 10 (48%) | 19 (61%) | 0.005 |
| Transapical TAVR | 72 (31%) | 47 (51%) | 11 (52%) | 12 (39%) | 0.005 |
| Cerebrovascular disease | 70 (30%) | 31 (34%) | 7 (33%) | 7 (23%) | 0.700 |
| Stroke | 28 (12%) | 13 (14%) | 3 (14%) | 4 (13%) | 0.966 |
| Peripheral vascular disease | 72 (31%) | 37 (40%) | 13 (62%) | 15 (48%) | 0.012 |
| Severe lung disease | 39 (17%) | 13 (14%) | 6 (29%) | 5 (16%) | 0.527 |
| Diabetes mellitus | 85 (37%) | 41 (45%) | 10 (48%) | 12 (39%) | 0.533 |
| Preoperative Myocardial infarction | 49/197 (25%) | 15/81 (19%) | 2/18 (11%) | 3/25 (12%) | 0.236 |
| Blood Hematocrit (%) | 34.64 ± 4.42 | 32.98 ± 5.04 | 33.77 ± 4.85 | 34.10 ± 4.25 | 0.036 |
| Serum Albumin (g/dL) | 3.28 ± 0.38 | 3.27 ± 0.45 | 3.27 ± 0.35 | 3.10 ± 0.31 | 0.246 |
| Preoperative renal failure | 10/214 (5%) | 9/84 (11%) | 0 | 15/28 (54%) | < 0.001 |
| Creatinine (mg/dL) | 1.20 ± 0.57 | 1.40 ± 0.47 | 0.90 ± 0.26 | 2.66 ± 2.15 | < 0.001 |
| Estimated Glomerular Filtration Rate (mL/min/1.73m 2 ) | 52.21 ± 26.76 | 44.97 ± 22.89 | 62.34 ± 21.68 | 30.78 ± 18.19 | < 0.001 |
| On Dialysis | 0 | 0 | 0 | 11 (36%) | < 0.001 |
The overall operative mortality was higher with increased AKI stage ( Table 3 ). The incidence of new postoperative dialysis was 3%. This association between higher AKI stage and mortality remained at 3-year follow-up ( Figure 1 ). Patients with significant AKI (stage 2/3) had increased long-term mortality, which persisted to 3-year follow-up ( Figure 2 ).
| Variable | VARC-2 Classification of Kidney Disease | p-Value | |||
|---|---|---|---|---|---|
| No AKI (n = 230) | Stage 1 (n = 92) | Stage 2 (n = 21) | Stage 3 (n = 31) | ||
| Operative Mortality | 7 (3%) | 7 (8%) | 5 (24%) | 10 (32%) | < 0.001 |
| Contrast dye load (ml) | 166.73 ± 88.16 | 157.20 ± 91.86 | 205.14 ± 169.93 | 186.96 ± 112.25 | 0.153 |
| Total Ventilator time (hours) | 32.90 ± 89.52 | 89.85 ± 191.52 | 211.26 ± 327.65 | 243.78 ± 354.08 | < 0.001 |
| Total ICU Time (hours) | 76.23 ± 83.49 | 163.71 ± 198.92 | 290.30 ± 309.83 | 341.54 ± 345.02 | < 0.001 |
| Intraoperative Blood Products Given | 63 (27%) | 34 (37%) | 10 (48%) | 11 (36%) | 0.121 |
| Creatinine (mg/dL) | 1.21 ± 0.42 | 2.02 ± 0.66 | 2.07 ± 0.61 | 4.80 ± 1.92 | < 0.001 |
| New-onset Renal Failure | 0 | 0 | 8 (38%) | 28 (90%) | < 0.001 |
| Postoperative Renal Failure requiring Dialysis | 0 | 0 | 0 | 12 (39%) | 0.003 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
