The risk of acute kidney injury (AKI) is a major issue after percutaneous coronary interventions (PCIs), especially in the setting of ST-elevation myocardial infarction. Preliminary data from large retrospective registries seem to show a reduction of AKI when a transradial (TR) approach for PCI is adopted. Little is known about the relation between vascular access and AKI after emergent PCI. We here report the results of the Primary PCI from Tevere to Navigli (PRIPITENA), a retrospective database of primary PCI performed at high-volume centers in the urban areas of Rome and Milan. Primary end point of this study was the occurrence of AKI in the TR and transfemoral (TF) access site groups. Secondary end points were major adverse cardiovascular events, stent thrombosis, and Thrombolysis in Myocardial Infarction major and minor bleedings. The database included 1,330 patients, 836 treated with a TR and 494 with a TF approach. After a propensity-matched analysis performed to exclude possible confounders, we identified 450 matched patients (225 TR and 225 TF). The incidence of AKI in the 2 matched groups was lower in patients treated with TR primary PCI (8.4% vs 16.9%, p = 0.007). Major adverse cardiovascular events and stent thrombosis were not different among study groups, whereas major bleedings were more often seen in the TF group. At multivariate analysis, femoral access was an independent predictor of AKI (odds ratio 1.654, 95% confidence interval 1.084 to 2.524, p = 0.042). In conclusion, in this database of primary PCI, the risk of AKI was lower with a TR approach, and the TF approach was an independent predictor for the occurrence of this complication.
In the “low-risk intervention era,” acute kidney injury (AKI) after cardiac catheterization is emerging as one of the most important and frequent complications and is correlated with higher in-hospital stay and increased long-term morbidity and mortality. Underlying kidney dysfunction is recognized as the most important risk factor for AKI, and historically, the amount of contrast used during catheterization has been considered the key modifiable factor related to its development. Recently, some authors highlighted the possible contribution of atheroembolism in the development of this complication, with the transfemoral (TF) access potentially leading to the greatest risk because of its proximity to renal arteries. From the theoretical point of view the transradial (TR) approach, following a different path to the coronary arteries, might be associated with lower AKI, as some demographic registries have recently shown. The aim of our study was to evaluate, in the setting of a high-risk intervention-like primary percutaneous coronary intervention (PCI), the possible role of different vascular approaches in the incidence of AKI.
Methods
The Primary PCI from Tevere to Navigli (PRIPITENA) is an investigator-initiated database of primary PCI performed at high-volume centers in the metropolitan areas of Rome and Milan. The main goal of this database is to investigate current ST-elevation myocardial infarction (STEMI) management in highly advanced health systems. Theoretically, all patients living in these urban areas with an STEMI are treated with fast-track primary PCI with early antithrombotic therapy. For the purpose of this analysis, we included all consecutive patients treated between January 2011 and December 2013. Patients were eligible if they presented within 12 hours after the onset of symptoms of cardiac ischemia and had ST-segment elevation of at least 1 mm in at least 2 contiguous leads and if they had complete data for the adjudication of primary and secondary end points. We adopted an all-comer strategy for this study, including patients with cardiogenic shock. All procedures were performed according to current guidelines, and all therapeutic decisions, including antithrombotic regimen, type of stent implanted, and thrombus aspiration, were left at the operators’ discretion. All patients were pretreated according to the standard of centers involved, with aspirin and P2Y12 inhibitors with the recommended loading doses; aspirin was continued indefinitely and P2Y12 inhibitors were recommended for at least 6 months. Unfractionated heparin (initial bolus of 5,000 to 10,000 IU, followed by additional boluses to achieve an activated clotting time >250 seconds) or bivalirudin (with a bolus of 0.75 mg/kg and an infusion of 1.75 mg/kg/h during the procedure) were given after operator’s preference. After the procedure, all patients were treated with saline according to body weight, whereas the use of N-acetylcysteine and/or bicarbonate was left to operator’s discretion.
In patients treated with a TR approach, arterial sheath was removed immediately after procedure. Radial hemostasis was obtained according to the standard of the center involved, using an inflatable wristband or a pressure bandage with elastic sticky straps. In TF patients, the sheath was either removed immediately using vascular closure devices or after the activated clotting time decreased <180 seconds, in case of manual compression. The choice of the approach was left to the operator’s preference. All procedures were performed in a total of 5 centers: 3 centers usually perform primary PCI through a radial access, 1 center through the femoral one, and in 1 center PCIs are performed through either approach depending on operators’ preferences.
The primary end point of the study was the occurrence of AKI in the TR and TF groups after propensity score matching. Secondary end points were major adverse cardiovascular events (MACEs), major and minor bleedings, and stent thrombosis (propensity-matched population); the same end points were also adjudicated for the total population. AKI was defined as an increase of >25%, or an absolute increase of 0.5 mg/dl or more, in serum creatinine from the baseline value in the first 72 hours after the procedure. MACEs were defined as death, reinfarction, and target lesion revascularization. Bleedings were classified according to the Thrombolysis in Myocardial Infarction (TIMI) classification as major and minor. Stent thrombosis was categorized after the Academic Research Consortium classification.
Continuous variables were expressed as mean ± SD and were compared by independent samples Student t test if normally distributed or by Mann-Whitney U test if not normally distributed. Categorical variables were expressed as count and percentages and were compared by chi-square test or Fisher’s exact test. We performed propensity score matching to adjust for different clinical and procedural characteristics between patients treated by the 2 approaches. A nonparsimonious propensity score was calculated for each patient by logistic regression analysis, entering in the model the following variables: age, gender, body mass index, left ventricular ejection fraction (LVEF), systolic arterial pressure, glycoprotein IIb/IIIa inhibitors, intra-aortic balloon pump, contrast volume, history of smoke, hypertension, dyslipidemia, diabetes, previous myocardial infarction, previous coronary artery bypass intervention, previous PCI, previous transient ischemic attack or stroke, hematocrit value, and basal glomerular filtration rate assessed by the Cockroft-Gault formula, treated vessel. We performed a nearest neighbor 1:1 matching; to exclude bad matching, we imposed a caliper of 0.2 of the SD of the logic of the propensity score. In the matched population, categorical and continuous variables in radial and femoral groups were compared by chi-square and t tests. Moreover, binary logistic regression analysis was performed to assess the influence of clinical and procedural characteristics on the occurrence of in-hospital AKI. The following variables, based on the result of univariate analysis (p <0.1) and clinical relevance, were entered in a stepwise forward conditional model, with p-in = 0.05 and p-out = 0.10: age, gender, body mass index, hematocrit, LVEF, vascular access, diastolic arterial pressure, glycoprotein IIb/IIIa inhibitors, intra-aortic balloon pump, history of smoke, hypertension, dyslipidemia, diabetes, previous stroke, or transient ischemic attack. We included in the model both the propensity score for the selection of vascular access and the incidence of TIMI bleedings (major + minor) because the latter covariate might affect the development of AKI and was, indeed, significantly associated with AKI at univariate analysis. The goodness of fit of the model was assessed by the Hosmer-Lemeshow test. All analyses were performed with SPSS 19.0 software (SPSS, Chicago, IL). Propensity score matching was performed with propensity score matching for SPSS, version 3.0 (F. Thoemmes, 2011). A p <0.05 was considered statistically significant.
Results
The total study sample consisted of 1,330 patients who underwent primary PCI from the PRIPITENA database, 836 treated with a TR and 494 with a TF approach. We excluded 240 patients because of incomplete data regarding study end points or in case of early death (before adjudication of AKI: 32 patients). Patient characteristics, categorized by access site, are listed in Table 1 . TF patients were older, more often female and dyslipidemic, had a slightly worse LVEF, and a lower baseline glomerular filtration rate. On the contrary, TR patients were more often treated with bivalirudin and had fewer final TIMI 3 flow at the end of the procedure. Notably, TF interventions were associated with lower contrast use (188.9 ± 71.6 vs 207.0 ± 73.6, p <0.0001). The incidence of AKI after primary PCI in the whole population was lower in the TR cohort than that in the TF (8.1% vs 16.9%, p <0.0001). After multivariate analysis with binary logistic regression, the independent predictors of AKI are listed in Table 2 . The femoral cohort was associated with a 60% increase in the risk of AKI and resulted as a strong predictor for its occurrence. The propensity-matched population adjusted for baseline covariates consisted of 225 patients for each group. Histograms with estimates of standardized differences before and after matching are reported in Figure 1 . After matching, both groups were similar in terms of demographic, clinical, and procedural characteristics ( Table 3 ). The incidence of AKI, primary study end point, was lower in patients treated with TR primary PCI (8.4% vs 16.9%, p = 0.007) ( Figure 2 ). The incidence of MACE and stent thrombosis did not differ significantly. In contrast, major bleedings were more often seen in TF patients ( Figure 3 ).
| Variable | Radial Approach (n = 836) | Femoral Approach (n = 494) | p |
|---|---|---|---|
| Age (years ± SD) | 63.1 ± 12.9 | 64.8 ± 12.9 | 0.022 |
| Body mass index (Kg/m 2 ± SD) | 27.5 ± 4.4 | 27.8 ± 5.1 | 0.365 |
| Smoker | 487 (58.7%) | 238 (54.1%) | 0.066 |
| Hypertension ∗ | 487 (58.7%) | 266 (60.6%) | 0.524 |
| Hyperlipidemia † | 305 (36.8%) | 196 (44.5%) | 0.004 |
| Family history of coronary artery disease | 146 (17.6%) | 138 (31.7%) | 0.000 |
| Diabetes mellitus | 178 (21.4%) | 110 (22.9%) | 0.286 |
| Previous myocardial infarction | 106 (12.7%) | 73 (15.2%) | 0.122 |
| Previous coronary bypass | 12 (1.4%) | 11 (2.5%) | 0.177 |
| Previous PCI | 78 (9.4%) | 56 (12.8%) | 0.063 |
| Peripheral vascular disease | 53 (8.7%) | 27 (8.4%) | 0.494 |
| Previous transient ischemic attack/stroke | 28 (3.8%) | 19 (4.3%) | 0.652 |
| Left ventricular ejection fraction (% ± SD) | 47.3 ± 9.6 | 45.9 ± 10.4 | 0.018 |
| Systolic blood pressure (mm Hg ± SD) | 131.2 ± 24.8 | 123.6 ± 27.3 | 0.000 |
| Basal glomerular filtration rate (as per Cockroft-Gault formula) (ml/min ± SD) | 92.4 ± 35.4 | 81.4 ± 36.5 | 0.000 |
| Glomerular filtration rate <60 ml/min | 123 (17.2%) | 118 (25.9%) | 0.000 |
| Preventive N-acetylcysteine | 324 (38.7%) | 204 (41.3%) | 0.459 |
| Preventive bicarbonate | 285 (34.1%) | 164 (33.2%) | 0.872 |
| Hematocrit (% ± SD) | 42.8 ± 4.7 | 43.1 ± 4.9 | 0.426 |
| Creatinine (mg/dl ± SD) | 1.0 ± 0.5 | 1.1 ± 0.6 | 0.028 |
| Narrowed coronary arteries: | 0.012 | ||
| 1 | 407 (48.7%) | 180 (40.2%) | |
| 2 | 263 (31.5%) | 158 (35.3%) | |
| 3 | 166 (19.9%) | 110 (24.6%) | |
| Infarct related artery: | 0.683 | ||
| Left anterior descending | 397 (47.5%) | 223 (45.1%) | |
| Circumflex | 134 (16.0%) | 78 (15.8%) | |
| Right | 239 (35.0%) | 182 (36.8%) | |
| Left main trunk | 8 (1.0%) | 6 (1.2%) | |
| Venous graft | 4 (0.5%) | 4 (0.8%) | |
| Glycoprotein IIb/IIIa inhibitor use | 442 (52.9%) | 247 (50.0%) | 0.170 |
| Bivalirudin | 154 (18.4%) | 15 (3.0%) | 0.000 |
| Thrombus aspiration | 322 (39.1%) | 195 (39.5%) | 0.9 |
| Intra-aortic balloon pump use | 21 (2.5%) | 48 (9.7%) | 0.000 |
| Number of stents used (n ± SD) | 1.3 ± 0.7 | 1.5 ± 0.8 | 0.000 |
| Contrast load (ml ± SD) | 207.0 ± 73.6 | 188.9 ± 71.6 | 0.000 |
| TIMI flow post: | 0.056 | ||
| 0 | 14 (2.0%) | 1 (0.45) | |
| 1 | 6 (0.9%) | 1 (0.4%) | |
| 2 | 41 (6.0%) | 6 (2.6%) | |
| 3 | 628 (91.1%) | 224 (96.6%) |
∗ Blood pressure >140/90 mm Hg.
| OR | 95% C.I. | p | |
|---|---|---|---|
| Age | 1.027 | 1.009–1.044 | 0.003 |
| Femoral access | 1.654 | 1.084–2.524 | 0.020 |
| Basal hematocrit | 0.920 | 0.883–0.959 | 0.000 |
| Glomerula filtration rate, ml/min | 0.993 | 0.986–1.000 | 0.040 |
| TIMI bleedings (major + minor) | 3.936 | 1.783–8.686 | 0.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
