International Journal of Cardiology 207 (2016) 335–340
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Impact of chronic kidney disease on left atrial appendage occlusion for stroke prevention in patients with atrial fibrillation☆ Joelle Kefer a,⁎, Apostolos Tzikas b, Xavier Freixa c, Samera Shakir d, Sameer Gafoor e,w, Jens Erik Nielsen-Kudsk f, Sergio Berti g, Gennaro Santoro h, Adel Aminian i, Ulf Landmesser j, Fabian Nietlispach d,j, Reda Ibrahim k, Paolo Luciano Danna l, Edouard Benit m, Werner Budts n, Francis Stammen o, Tom De Potter p, Tobias Tichelbäcker q, Steffen Gloekler d, Prapa Kanagaratnam r, Marco Costa s, Ignacio Cruz-Gonzalez t, Horst Sievert e, Wolfgang Schillinger q, Jai-Wun Park u, Bernhard Meier d, Heyder Omran v a
Cliniques universitaires Saint-Luc, Brussels, Belgium Interbalkan European Medical Center, Thessaloniki, Greece Hospital Clinic of University of Barcelona, Barcelona, Spain d University Hospital of bern, Bern, Switzerland e CardioVascular Center Frankfurt, Frankfurt, Germany f Aarhus University Hospital, Skejby, Denmark g Heart Hospital, Fondazione C.N.R. Regione Toscana, Massa, Italy h Ospedale Careggi di Firenze, Florence, Italy i Centre Hospitalier Universitaire de Charleroi, Charleroi, Belgium j University Hospital of Zurich, Zurich, Switzerland k Montreal Heart Institute, Montreal, Canada l Ospedale Luigi Sacco, Milan, Italy m Jessaziekenhuis, Hasselt, Belgium n UZ Leuven, Leuven, Belgium o AZ Delta Roeselare, Roeselare, Belgium p OLV Ziekenhuis, Aalst, Belgium q Universitätsmedizin Göttingen, Göttingen, Germany r Imperial College Healthcare NHS Trust, London, United Kingdom s Coimbra University Hospital Centre, Coimbra, Portugal t University Hospital of Salamanca, Salamanca, Spain u Asklepios Hospital Hamburg, Hamburg, Germany v University Hospital of Bonn, Bonn, Germany w Seattle Heart and Vascular, Seattle, WA, USA b c
a r t i c l e
i n f o
Article history: Received 12 October 2015 Received in revised form 26 November 2015 Accepted 1 January 2016 Available online 9 January 2016 Keywords: Left atrial appendage Stroke Bleeding Renal failure Amplatzer cardiac plug
a b s t r a c t Background: Left atrial appendage occlusion (LAAO) using the Amplatzer cardiac plug (ACP) is a preventive treatment of atrial fibrillation related thromboembolism. Aim: To assess the safety and efficacy of LAAO in patients with chronic kidney disease (CKD). Methods: Among the ACP multicentre registry, 1014 patients (75 ± 8 yrs) with available renal function were included. Results: Patients with CKD (N = 375, CHA2DS2-VASc: 4.9 ± 1.5, HASBLED: 3.4 ± 1.3) were at higher risk than patients without CKD (N = 639, CHA2DS2-VASc: 4.2 ± 1.6, HASBLED: 2.9 ± 1.2; p b 0.001 for both). Procedural (97%) and occlusion (99%) success were similarly high in all stages of CKD. Peri-procedural major adverse events (MAE) were observed in 5.1% of patients, 0.8% of death, with no difference between patients with and those without CKD (6.1 vs 4.5%, p = 0.47). In patients with complete follow-up (1319 patients years), the annual stroke + transient ischaemic attack (TIA) rate was 2.3% and the observed bleeding rate was 2.1% (62 and 60% less than expected, similarly among patients with and those without CKD). Kaplan–Meier analysis showed a lower overall survival (84 vs 96% and 84 vs 93% at 1 and 2 yrs. respectively; p b 0.001) among patients with an eGFR b30 ml/min/1.73 m2.
☆ For all the authors: there are no conflicts of interest. The authors take the responsibility for all aspects of reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: Cliniques Universitaires Saint-Luc, University of Louvain, Division of Cardiology, Avenue Hippocrate, 10-2881, 1200 Brussels, Belgium. E-mail address: [email protected] (J. Kefer).
http://dx.doi.org/10.1016/j.ijcard.2016.01.003 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
336
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
Conclusion: LAAO using the ACP has a similar procedural safety among CKD patients compared to patients with normal renal function. LAAO with ACP offers a dramatic reduction of stroke + TIA rate and of bleeding rate persistent in all stages of CKD, as compared to the expected annual risk. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Percutaneous left atrial appendage occlusion (LAAO) is an alternative therapeutic option for stroke prevention in patients with atrial fibrillation (AF). [1–5]. The first randomised trial PROTECT AF [6] showed that LAAO with the WATCHMAN device (Boston Scientific, Marlborough, MA, USA) reduced mortality and stroke rate compared to warfarin, at 5 years follow-up. As compared with the initial attempts [3,7–8], experienced operators have shown a significant reduction of serious periprocedural complication rate around 4–5% [5,9]. Recently, it has been demonstrated that LAAO using the Amplatzer Cardiac Plug (ACP; St. Jude Medical, St Paul, MN, USA), reduces the theoretical risk of both, stroke and bleeding [5] in patients with AF. Renal failure patients are well known to be at high risk of procedural complications and to have a worse outcome after transcatheter interventions [10–11]. The risk of cardiovascular mortality increases in patients with chronic kidney disease (CKD) stage 3 [12]; recently, this category was subdivided in stage 3a and 3b [13]. The prevalence of AF in patients with CKD is high and renal failure represents a main risk factor for stroke, leading to a strong recommendation for anticoagulation therapy [14]. On the other hand, CKD significantly increases the risk of major bleeding in patients with AF [15–16]. Novel anticoagulants were approved for stroke prevention in AF [17–19] but their use is still controversial in the setting of CKD, especially among end-stage renal failure patients (eGFR b30 ml/min/m2) [20]. LAAO, as a non pharmacological approach, could be a potential alternative in patients with AF and CKD. The aim of the present study was to assess the procedural safety of LAAO in patients with AF and CKD, the efficacy on stroke and bleeding prevention, and finally to evaluate the impact of the stages of CKD on the clinical outcome after LAAO using the ACP. 2. Methods Between December 2008 and November 2013, consecutive patients undergoing LAAO using the ACP in 22 centres, were prospectively included in the multicentre ACP registry, published elsewhere [5]. Demographics, baseline characteristics, indications for LAAO, CHA2DS2-VASc and HAS-BLED scores, antithrombotic medication, procedural details, periprocedural adverse events, clinical and echocardiographic follow-up were prospectively collected in a dedicated database. Among this cohort, patients with available serum creatinine and glomerular filtration rate values were included in the current substudy. An informed consent was obtained in each patient; the study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki. 2.1. Renal failure The estimated glomerular filtration rate (eGFR) was calculated with the CKD-EPI formula [21]. By using the creatinine serum level, the age, the gender and the race of the patient, the CKD-EPI equation was expressed as: GFR ¼ 141 minðScr=κ; 1Þα maxðScr=κ; 1Þ−1:209 0:993Age 1:018 ½if female 1:159 ½if black: According to the guidelines [22], CKD was defined as eGFR b60 ml/min per 1.73 m 2 and CKD stages were classified based on the eGFR values: stage 1 = ≥ 90 ml/min/1.73 m 2 , stage 2 = 60 to 89 ml/min/1.73 m 2 , stage 3a = 45 to 59 ml/min/1.73 m 2 , stage 3b = 30 to 44 ml N/min/1.73 m 2 , stage 4 = 15 to 29 ml/min/1.73 m2 and stage 5 = b15 ml/min/1.73 m2. In addition, patients on dialysis and those with previous renal transplantation were noted. 2.2. Procedural success Procedural success was defined as successful implantation of the ACP in the left atrial appendage (LAA).
2.3. Periprocedural complications Periprocedural complications (occuring during 0–7 days after procedure or before hospital discharge, whichever last) included death, myocardial infarction, stroke, transient ischaemic attack (TIA), systemic embolism, air embolism, device embolisation, major bleeding (requiring surgery or transfusion) and cardiac tamponade. Periprocedural major adverse events (MAE) included death, stroke, systemic embolism and complication requiring major surgical or endovascular intervention (major bleeding, tamponade, device migration treated by snare or surgery) occurring between 0.7 days post procedure or before hospital discharge, whichever latest. 2.4. Clinical follow-up Patient survival and occurrence of clinical events during the follow-up were determined by review of medical records or phone contact of patients implanted successfully. Adverse events during follow-up included death (cardiovascular or noncardiovascular), stroke, TIA, systemic embolism and major bleeding. Antithrombotic medication was recorded at the admission date and at last follow-up visit. The recommendation by the manufacturer is to give a low dose of acetylsalicylic acid for a long period with clopidogrel (75 mg/d) for at least 1 month. So, a successful LAAO implies the absence of anticoagulant therapy. Nevertheless, the choice and the duration of antithrombotic therapy was individualised, depending on the patient characteristics and physician preference. 2.5. Efficacy on stroke, TIA and systemic embolism prevention LAAO efficacy on stroke, TIA and systemic embolism prevention was tested by comparing the actual event rate at follow-up with the predicted event rate by the CHA2DS2-VASc score [23]. Individual patient annual risk was recorded and the average annual risk for the whole study population was calculated. The total number of thromboembolic events (stroke, TIA and systemic embolism) during both the periprocedural and follow-up periods was divided by the total patient-years of follow-up and was multiplied by 100 in order to get the actual annual rate of thromboembolism. Thromboembolism reduction was calculated as follows (estimated % − actual% event rate) / estimated % event rate. 2.6. Efficacy on bleeding prevention The total number of major bleeding events during both the periprocedural and followup periods was compared with the events predicted by the HAS-BLED score [24]: bleeding reduction was calculated as follows: (estimated% − actual% event rate) / estimated% event rate. 2.7. Echocardiographic follow-up Implanted patients underwent a transoesophageal echocardiography (TOE) at followup. Residual leak was graded using the width of the Doppler colour jet as trivial (b1 mm), mild (1–3 mm) or significant (N3 mm). Occlusion success was defined as a procedural success without a significant residual leak. The presence of thrombus on the device was noted as well. 2.8. Statistical analysis Continuous variables are presented as mean ± 1 standard deviation. Categorical variables are presented as counts and percentages. Continuous variables were tested by using the independent samples t-test and categorical variables by using the Fischer's exact test. Univariate analysis was done using the Cox proportional-hazards method. Estimates for freedom from the composite of death and MAE were obtained by the Kaplan–Meier estimation method. A p value b0.05 was considered statistically significant. Analyses were performed using the SPSS version 15.0 (SPSS Inc., Chicago, Illinois).
3. Results 3.1. Patients Of the cohort of 1053 patients enrolled in the ACP multicentre registry, serum creatinine and eGFR values were available in 1014 and they were included in the study. Baseline characteristics are listed in Table 1. There were 183 patients categorised in stage 1, 456 in stage 2,
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
337
176 in stage 3a, 119 in stage 3b, 61 in stage 4 and 19 in stage 5. Fourteen patients were on dialysis and 3 other patients had a previous renal transplant. Because of the low number of patients with renal replacement therapy, no further analysis was performed in this group. The mean eGFR value was calculated at 81.8 ± 12.5 ml/min/1.73 m2 in patients with normal renal function (N = 639) and 40.9 ± 13.9 ml/min/1.73 m2 among patients with CKD (N = 375; p b 0.0001). Patients with CKD were older, more frequently female and had a higher prevalence of diabetes mellitus, congestive heart failure, and coronary artery disease. As compared to patients with normal renal function, CKD patients had a significantly higher CHADS2 (3 ± 1.4 vs 2.6 ± 1.3), CHA2DS2-VASc (4.9 ± 1.5 vs 4.2 ± 1.6) and HAS-BLED (3.4 ± 1.3 vs 2.9 ± 1.1) scores; p b 0.0001 for all comparisons.
7 of them were successfully snared, and one was successfully removed surgically. Of the 16 reported tamponades (1.6%), 2 were fatal. Major bleedings (1.2%) were mainly due to a vascular complication at the femoral puncture site (6/14). Patients with an eGFR b60 ml/min/1.73 m2 had not more MAEs than patients with an eGFR N60 ml/min/1.73 m2. More importantly, MAEs did not differ for patients with endstage renal disease (eGFR b30 ml/min/1.73 m2) as compared to patients with an eGFR N30 ml/min/1.73 m2: 3.7 vs 5.2%; p = 0.79. There was no need for new dialysis therapy among the total cohort, especially among patients with pre-existing renal failure. Among the 28 unsuccessfully attempted patients (mainly due to inappropriate size of the LAA), only one died within the periprocedural period due to heart failure.
3.2. Procedure
3.4. Follow-up
Procedural success was similarly high in patients with or without CKD (98% vs 97%; p = 0.23). Procedural characteristics are described in Table 2. There was no difference between groups neither in the mean size of the implanted ACP nor in the need for change of ACP size. The sizes 22 and 24 mm of the ACP device were the most frequently used (20% of cases for both), followed by 26 mm (16%), 30 mm (14%), 28 mm (11%), 20 mm (10%), 18 mm (5%) and 16 mm (3%). The distribution of device sizes was similar between patients with normal renal function and patients with CKD. Patients with normal renal function underwent more frequently combined interventions (31% vs 19%; p = 0.001), mainly driven by the number of coronary angiograms (13% vs 7%; p = 0.003). The mean volume of contrast used was 122 ± 80 ml.
The follow-up was complete in 967 of 986 patients with successful LAAO (98%). The mean duration of follow-up was 498 days (interquartile range 186–753 days) resulting in a total of 1319 patients-years. There were 9 strokes (0.9%), 9 TIAs (0.9%) and 15 major bleedings (1.5%) reported during the follow-up period, without any difference between groups. Overall survival (84% at 1 and 2 yrs) and event-free survival (81% at 1 and 2 yrs) were significantly lower in patients with an eGFR b30 ml/min/1.73 m2 (stages 4 + 5) than in other patients (stage 1 + 2 + 3) for whom overall survival was 96% at 1 yr., and 93% at 2 yrs.; p b 0.001 (Fig. 1), and event-free survival was 92% at 1 yr., and 86% at 2 yrs.; p = 0.007 (Fig. 2). The causes of death were more frequently cardiac among patients in stage 4 + 5 (6.9%) than in stage 1 + 2 + 3 (1.4%; p = 0.007) but never device-related (Table 4). The annual rate of non fatal MAE (stroke, TIA and major bleeding) was not significantly different among patients with endstage renal failure than in patients with an eGFR N30 ml/min/1.73 m2 (4.05% vs 4.5%; p = 1.00). With univariate analysis, eGFR (hazard ratio: 0.986; confidence interval: 0.976–0.995; p = 0.004) and HASBLED score (hazard ratio: 1.43; confidence interval: 1.02–2.00; p = 0.037) were associated with death at follow-up. By multivariare analysis, eGFR b30 ml/min/1.73 m2 was a predictor of total MAEs during followup (hazard ratio: 2.72; confidence interval: 1.51–4.90; p = 0.001),
3.3. Periprocedural complications The rate of periprocedural MAEs was 5.1% in the total cohort without any difference between groups (Table 3). Eight deaths (0.8%) were reported including 2 tamponades, 2 arrhythmias, 1 myocardial infarction, 1 intracranial haemorrhage and 2 device embolisations (one resulting in procedural death and another one treated by surgery but resulting in death). There were 8 additional device embolisations:
Table 1 Baseline characteristics. Characteristics
Age (yrs) Gender BMI Congestive heart failure Hypertension Diabetes Coronary artery disease Carotid disease Previous stroke/TIA Thromboembolism on anticoagulant Previous major bleeding Permanent AF Paroxysmal/persistent AF eGFR MDRD (ml/kg/1.73 m2) eGFR CKD-EPI (ml/kg/1.73 m2) Serum creatinine (mg/dl) Dialysis Renal transplant CHADS2 score CHA2DS2-VASc score HASBLED score
All
mean ± sd M/F mean ± sd n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) mean ± sd mean ± sd mean ± sd n (%) n (%)
No CKD
CKD
Stage 1 + 2
Stage 3a
Stage 3b
Stage 4
Stage 5
N = 1014
N = 639
N = 176
N = 119
N = 61
N = 19
N = 375
74.9 ± 8.4 630/384 27.3 ± 4.8 259 (25) 860 (85) 299 (29) 354 (35) 83 (8) 384 (38) 162 (16) 466 (46) 573 (56) 439 (43) 69.7 ± 27.5 66.8 ± 23.7 1.1 ± 0.8 14 (1) 3 (0.3) 2.8 ± 1.3 4.4 ± 1.6 3.1 ± 1.2
73.1 ± 8.4 425/215 27.1 ± 4.9 124 (19) 536 (84) 152 (24) 183 (28) 50 (8) 264 (41) 99 (15) 282 (44) 361 (56) 278 (43) 86 ± 19 81.8 ± 12.5 0.8 ± 0.1 0 0 2.6 ± 1.3 4.2 ± 1.6 2.9 ± 1.1
77.8 ± 7.1 95/79 27.5 ± 4.8 54 (31) 157 (90) 58 (33) 74 (42) 6 (3) 55 (31) 36 (20) 84 (48) 96 (55) 78 (45) 53.1 ± 4.4 52.6 ± 4.4 1.2 ± 0.1 0 1 (0.5) 3 ± 1.2 4.7 ± 1.5 3.1 ± 1.1
79.2 ± 7.2 67/51 27.6 ± 4.4 40 (34) 110 (93) 49 (41) 62 (52) 15 (13) 40 (34) 0 58 (49) 71 (60) 47 (40) 38.4 ± 4.5 37.3 ± 4.6 1.6 ± 0.2 0 1 (0.8) 3.2 ± 1.3 5.1 ± 1.4 3.5 ± 1.3
76.3 ± 8.3 31/30 27.2 ± 5 34 (56) 54 (88) 28 (46) 26 (43) 9 (15) 16 (26) 4 (6) 26 (43) 31 (50) 30 (50) 24.6 ± 3.8 23.7 ± 3.8 2.3 ± 0.4 2 (3) 1 (1) 3.1 ± 1.5 5.1 ± 1.7 3.7 ± 1.3
76.9 ± 6.6 11/8 27.6 ± 5 7 (36) 13 (68) 8 (42) 9 (47) 3 (16) 9 (47) 3 (16) 13 (68) 14 (74) 5 (26) 9.9 ± 3.1 9.1 ± 2.7 5.4 ± 1.6 12 (63) 0 3 ± 1.7 4.7 ± 1.7 4.3 ± 1.3
77.9 ± 7.3 205/169 27.5 ± 4.7 135 (36) 324 (87) 147 (39) 171 (46) 33 (9) 120 (32) 62 (16) 184 (49) 212 (57) 161 (43) 41.7 ± 13.7 40.9 ± 13.9 1.7 ± 1 14 (3) 3 (0.8) 3 ± 1.4 4.9 ± 1.5 3.4 ± 1.3
p value
Stage 3 + 4 + 5
b0.0001 0.0003 0.21 b0.0001 0.23 b0.0001 b0.0001 0.63 0.003 0.65 0.11 0.94 0.94 b0.0001 b0.0001 b0.0001 0.0001 0.05 b0.0001 b0.0001 b0.0001
CKD = chronic kidney disease; BMI = body mass index; TIA = transient ischaemic attack; AF = atrial fibrillation; eGFR = estimated glomerular filtration rate; MDRD = modification of diet in renal disease; CKD-EPI = chronic kidney disease – epidemiology collaboration.
338
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
Table 2 Procedural characteristics. Characteristics
Procedural success Size ACP (mm) First ACP selected implanted Combined intervention PFO closure ASD closure Coronary angiogram Coronary angioplasty TAVI Atrial fibrillation ablation Via transseptal puncture Via PFO
All
n (%) mean ± sd n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%)
No CKD
CKD
Stage 1 + 2
Stage 3a
Stage 3b
Stage 4
Stage 5
Stage 3 + 4 + 5
N = 1014
N = 639
N = 176
N = 119
N = 61
N = 19
N = 375
986 (97) 24.3 ± 3.7 835 (82) 271 (27) 61 (6) 11 (1) 108 (10) 57 (5) 16 (1) 18 (2) 916 (90) 98 (10)
618 (97) 24.2 ± 3.7 529 (83) 199 (31) 50 (8) 7 (1) 82 (13) 37 (6) 8 (1) 15 (2) 566 (89) 73 (11)
175 (99) 24.2 ± 3.7 147 (83) 28 (16) 4 (2) 0 8 (4) 12 (7) 2 (1) 2 (1) 165 (94) 11 (6)
115 (97) 24.2 ± 3.7 92 (77) 24 (20) 5 (4) 3 (2) 6 (5) 5 (4) 4 (3) 1 (1) 107 (90) 12 (10)
59 (97) 24.8 ± 3.5 52 (85) 15 (24) 1 (2) 1 (2) 9 (15) 2 (3) 2 (3) 0 60 (98) 1 (2)
19 (100) 25.6 ± 3.9 15 (79) 5 (26) 1 (5) 0 3 (16) 1 (5) 0 0 18 (95) 1 (5)
368 (98) 24.2 ± 3.7 306 (82) 72 (19) 11 (3) 4 (1) 26 (7) 20 (5) 8 (2) 3 (1) 350 (93) 25 (7)
p value
0.23 0.81 0.66 0.001 0.001 1 0.003 0.88 0.30 0.08 0.02 0.01
ACP = Amplatzer cardiac plug; PFO = patent foramen ovale; ASD = atrial septal defect; TAVI = transcatheter aortic valve implantation; CKD = chronic kidney disease.
but was not a predictor for stroke (hazard ratio: 2.24; confidence interval: 0.27–17.9; p = 0.44), or for major bleeding (hazard ratio: 1.24; confidence interval: 0.16–9.49; p = 0.05). The actual annual stroke + TIA rate (periprocedural and follow-up period) was 2.3%, while the expected annual thromboembolism risk was calculated by the CHA2DS2-VASc score at 5.6%, which translates into 62% risk reduction. The observed annual major bleeding rate (periprocedural and follow-up period) was 2.1% and the annual risk of bleeding estimated by the HASBLED score was 5.3% (60% reduction). Fig. 3 shows the distribution of the event risk reduction stratified by the stages of eGFR. While the risk of bleeding increases with the stage of eGFR, the event reduction was still persistent after LAAO in all stages of CKD, especially in the stage 5 group. Medication at last follow-up was limited to acetylsalicylic acid in 84% of patients, while only 4.7% of them were still under anticoagulant therapy, with no difference between patients with and those without CKD (3.7 vs 3.8%, p = NS). In details, anticoagulant agents used in patients with CKD were vitamin K antagonist (2.5%), NOAC (0.8%) or LMWH (0.2%) and were not different in patients with normal renal function (3.5%, 1.6%, 0.1%, respectively; p = NS). At TOE follow-up (mean 204 ± 199, median 141 days) obtained among 609 patients, thrombus on the device was observed in 27 cases (4.4%) with no difference between groups; they were all treated conservatively with no consequent stroke or other adverse events during follow-up. A residual leak was noted in 71 patients and graded as trivial in 26, moderate in 31, and significant in 11 cases. Therefore, the occlusion success was 98.8%.
4. Discussion The main findings of this paper are: 1. The procedural safety of LAAO with the ACP is similarly high in patients with CKD as in patients with normal renal function, with no impact of the stages of CKD on the periprocedural MAEs. 2. The efficacy on stroke reduction is very high and persistent in all stages of CKD. 3. LAAO with the ACP, allowing to stop anticoagulant after a procedural success, offers a dramatic reduction of bleeding risk even if this theoretical risk increases with the severity of CKD. 4. Overall survival was lower in patients with an eGFR b30 ml/min/m2 (stage 4 + 5) compared to other patients (stage 1 + 2 + 3). However, the rate of non fatal MAEs during follow-up (stroke, TIA, and major bleeding) was not higher among patients with end-stage renal failure. 4.1. Procedural safety This paper confirms that LAAO is a fairly safe procedure even in a high risk population with a reported periprocedural mortality of 0.8%. The rate of MAEs (5.1%) is lower than in the initial ACP European registry [7] or the PROTECT AF study [3], reporting 7.3 and 7.7% of MAEs respectively, but higher than in the more recent publications CAP [9] registry (3.7%) and PREVAIL [25] study (4.2%). The patients included in our study were treated by both experienced teams and in beginners: this may explain that our results range in the middle between early and recent publications. Patients with CKD are well
Table 3 Peri-procedural complications. Characteristics
MAE peri-procedural Death Stroke Device embolisation Device snared Device removed surgically Major bleeding Tamponade Other complications TIAS Air embolisation Total complications
All
No CKD
CKD
p value
Stage 1 + 2
Stage 3a
Stage 3b
Stage 4
Stage 5
Stage 3 + 4 + 5
N = 1014
N = 639
N = 176
N = 119
N = 61
N = 19
N = 375
n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%)
52 (5.1) 8 (0.8) 9 (0.8) 8 (0.8) 7 (0.7) 1 (0.01) 13 (1.2) 14 (1.4)
29 (4.5) 5 (0.8) 4 (0.6) 5 (0.7) 5 (0.7) 1 (0.1) 6 (1) 9 (1.5)
10 (5.6) 0 3 (2) 0 0 0 3 (2) 4
10 (8.4) 2 2 2 1 1 3 1
3 (4.9) 1 0 1 1 0 1 0
0 0 0 0 0 0 0 0
23 (6.1) 3 (0.8) 5 (1.3) 3 (1) 2 (0.5) 1 (0.2) 7 (1.8) 5 (1.6)
0.47 1.0 0.30 1.0 1.0 1.0 0.24 1.0
n (%) n (%) n (%)
4 (0.4) 4 (0.4) 60 (5.9)
2 (0.3) 4 (0.6) 35 (5.4)
1 0 11 (6.2)
1 0 11 (9.2)
0 0 3 (0.5)
0 0 0
2 (0.5) 0 25 (6.6)
0.62 0.30 0.49
CKD = chronic kidney disease; MAE = major adverse event; TIA = transient ischaemic attack.
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
339
Fig. 1. Overall survival after LAAO. Panel A: Kaplan–Meier analysis showing the effect of the eGFR stages on overall survival. Panel B: Kaplan–Meier analysis showing that patients in eGFR stage 4 + 5 had reduced overall survival. eGFR = estimated glomerular filtration rate; LAAO = left atrial appendage occlusion.
known to be at higher risk for procedural complications after transcatheter interventions. Yamamoto et al. [11] showed a higher rate of early complications after TAVI in patients with an eGFR b45 ml/min/1.73 m2. Kitai et al. [26] and Murphy et al. [27] published the negative impact of renal failure on the early outcome after coronary angioplasty. In our series, the rate of periprocedural MAEs remained fairly low in all groups with no significant difference among patients with or without renal impairement. It is remarkable that in patients with CKD stages 4 and 5 (eGFR b30 ml/min/1.73 m2) there was no periprocedural stroke. The fact that LAAO is a transvenous and not a transarterial intervention, with no manipulation of catheter in the aorta, could potentially explain this absence of impact of the renal function on the procedural complications.
4.2. Efficacy on stroke reduction In the present study, the observed annual stroke + TIA rate (periprocedural and follow-up period) was 2.3%, while the expected annual thromboembolism risk was calculated by the CHA2DS2-VASc score at 5.6%, which is 62% risk reduction. Even in patients with CKD who had a higher theoretical stroke risk (6.3%), this reduction of observed events after LAAO persisted (2.7%, reduction of 60%) and was not inferior to that of patients with normal renal function (2.1%, 60% reduction; p = NS). Yamamoto et al. [11] reported a higher stroke rate after TAVI in patients with CKD; it was not the case in our study after LAAO. Importantly, as compared to the PROTECT-AF study, the CHADS2 score of our population was higher (2.8 versus 2.2), but the observed stroke rate was similar (2.3 versus 2.2%/year).
4.3. Impact on bleeding The observed major bleeding rate (periprocedural and follow-up period) was 2.1% and the annual risk of bleeding estimated by the HASBLED score was 5.3% (60% reduction). The bleeding risk increased with the GFR stage from 5.6% in the stage 3a to 9.5% in the stage 5, whereas the reduction of observed events was still persistent in all subgroups of GFR, not inferior among patients with CKD than in patients with normal renal function (−54% vs −64%, p = NS). Harel et al. [28] showed that patients with renal impairement are at higher risk of bleeding complications under vitamin K antagonist; Poli et al. [29] showed that novel anticoagulants are not more efficient in prevention of hemorrhagic complications among patients with CKD. Because of renal elimination, the new oral anticoagulants have a prolonged halflife in patients with CKD, resulting in enhanced antithrombotic efficacy and increased bleeding risk. Importantly, patients with an eGFR b30 ml/min/1.73 m2 were excluded from the 3 studies RELY, ROCKET and ARISTOTLE [17–19]. LAAO using the ACP allows stopping anticoagulant therapy after procedural success. In our study, at last follow-up, only 3.7% of CKD patients remained under anticoagulant therapy whereas 82% of them are on acetylsalicylic acid only. In this population with very high bleeding risk (6.5%), it is remarkable to observe only 3% (−54%) of annual major bleeding events while keeping the efficacy for stroke prevention.
4.4. Overall survival During follow-up, the present study showed a low rate of non fatal MAEs (0.9% stroke, 0.9% TIA and 1.5% major bleedings) and similar between patients with and without CKD. However, patients with an eGFR b 30 ml/min/1.73 m2 presented higher mortality at 2 years compared to those with an eGFR N 30 ml/min/1.73 m2. These cardiac deaths were mainly due to a worsening heart failure and not device
Table 4 Adverse events during follow-up.
Fig. 2. Event-free survival after LAAO. Kaplan–Meier analysis showing that patients in eGFR stage 4 + 5 had reduced event-free survival.
Stroke TIA Major bleeding Death Cardiac death
All
Stage 1 + 2 + 3
Stage 4 + 5
N = 967
N = 895
N = 72
9 (0.9) 9 (0.9) 15 (1.5) 61 (6.3) 18 (1.8)
9 (1.0) 8 (0.9) 14 (1.5) 51 (5.6) 13 (1.4)
0 1 (1.3) 1 (1.3) 10 (14) 5 (6.9)
TIA = transient ischaemic attack.
p value
1.00 0.50 1.00 0.01 0.007
340
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
[7]
[8]
[9]
[10]
[11]
[12] Fig. 3. Event (stroke and bleeding) rate after LAAO. Fig. 3 showing the distribution of the event risk reduction stratified by the stages of eGFR. Blue = observed event rate; Pink = expected event rate. LAAO = left atrial appendage occlusion. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
[13]
[14]
related, and probably reflect the expected and well-known higher risk of this population. In contrast, there was no excess of major bleeding observed in these patients who, nevertheless, had a higher theoretical risk at baseline of hemorrhagic complications. The protective effect of LAAO using the ACP is highlighted among CKD patients, directly related to the cessation of anticoagulant therapy in a context of increased risk of bleeding.
[15]
[16] [17]
[18]
5. Conclusions Despite a different risk profile at baseline, LAAO using the ACP has a similar procedural safety among CKD patients and patients with normal renal function. LAAO with ACP offers a dramatic reduction of stroke + TIA rate and of major bleeding rate, which is persistent in all stages of CKD, as compared to the expected annual risk. The protective effect of LAAO with the ACP followed by cessation of anticoagulant therapy, was highlighted in the cohort of renal failure patients at very high bleeding risk, in whom a significant reduction in major bleeding events was observed, while keeping the efficacy in stroke prevention although the vast majority of patients were left untreated by anticoagulants. Patients with end-stage renal failure had a lower overall survival, which was not device- related, with a similar rate of non fatal MAEs during follow-up.
[19]
[20]
[21]
[22]
[23]
[24]
References [25] [1] H. Sievert, M. Lesh, T. Trepels, H. Omran, A. Bartoelli, P. Dela Bella, et al., Percutaneous left atrial appendage transcatheter occlusion to prevent stroke in high-risk patients with atrial fibrillation, Circulation 105 (2002) 1887–1889. [2] P. Block, S. Burstein, P. Casale, P. Kramer, P. Teirstein, D. Williams, et al., Percutaneous left atrial appendage occlusion for patients in atrial fibrillation suboptimal for warfarin therapy, J. Am. Coll. Cardiol. 2 (2009) 594–600. [3] D. Holmes, V. Reddy, Z. Turi, S. Doshi, H. Sievert, M. Buchbinder, et al., Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation: a randomised non-inferiority trial, Lancet 374 (2009) 534–542. [4] J. Camm, G. Lip, R. De Caterina, I. Savelieva, D. Atar, S. Hohnloser, et al., 2012 focused update of the ESC guidelines for the management of atrial fibrillation, Eur. Heart J. 33 (2012) 2719–2747. [5] A. Tzikas, S. Shakir, S. Gafoor, H. Omran, S. Berti, G. Santoro, et al., Left atrial appendage occlusion for stroke prevention in atrial fibrillation: multicentre experience with the amplatzer cardiac plug, EuroIntervention 10 (10) (Jan 22 2015)http://dx.doi. org/10.4244/EIJY15M01_06 (Epub ahead of print). [6] V.Y. Reddy, H. Sievert, J. Halperin, S.K. Doshi, M. Buchbinder, P. Neuzil, et al., PROTECT AF steering committee and investigators. Percutaneous left atrial appendage
[26]
[27]
[28]
[29]
closure vs warfarin for atrial fibrillation: a randomized clinical trial, JAMA 312 (2014) 1988–1998. J.W. Park, A. Bethencourt, H. Sievert, G. Santoro, B. Meier, K. Walsh, et al., Left atrial appendage closure with Amplatzer cardiac plug in atrial fibrillation: initial European experience, Catheter. Cardiovasc. Interv. 77 (2011) 700–706. B. YL, O. H, N. P, T. L, P. M, R. E, et al., Plaato (percutaneous left atrial appendage transcatheter occlusion) for prevention of cardioembolic stroke in nonanticoagulation eligible atrial fibrillation patients: results from the European plaato strudy, EuroIntervention 6 (2010) 220–226. V.Y. Reddy, D. Holmes, S. Doshi, P. Neuzil, S. Kar, Safety of percutaneous left atrial appendage closure: results from the Watchman left atrial appendage system for embolic protection in patients with AF (PROTECT AF) clinical trial and the continued access registry, Circulation 123 (2011) 417–424. M. Wanitschek, M. Pfisterer, A. Hvelplund, S. De Servi, O. Bertel, R. Jeger, et al., Longterm benefits and risks of drug-eluting compared to bare-metal stents in patients with versus without chronic kidney disease, Int. J. Cardiol. 168 (2013) 2381–2388. M. Yamamoto, K. Hayashida, G. Mouillet, T. Hovasse, B. Chevalier, A. Oguri, et al., Prognostic value of chronic kidney disease after transcatheter aortic valve implantation, J. Am. Coll. Cardiol. 62 (2013) 869–877. K. Matsushita, M. van der Velde, B.C. Astor, et al., Association of estimated glomerular filtration rate and albuminuria with all-cause andcardiovascular mortality in general population cohorts: a collaborative meta-analysis. Chronic Kidney Disease Prognosis Consortium, Lancet 375 (2010) 2073–2081. M. van der Velde, K. Matsushita, J. Coresh, et al., Lower estimated glomerular filtration rate and higher albuminuria are associated with all-cause and cardiovascular mortality. A collaborative meta-analysis of high-risk population cohorts, Kidney Int. 79 (2011) 1341–1352. A. Bonde, G. Lip, A.-L. Kamper, P. Hansen, M. Lamberts, K. Hommel, et al., Net clinical benefit of antithrombotic therapy in patients with atrial fibrillation and chronic kidney disease, J. Am. Coll. Cardiol. 64 (2014) 2471–2482. A.S. Go, M.C. Fang, N. Udaltsova, et al., For the ATRIA study investigators. Impact of proteinuria and glomerular filtration rate on risk of thrombo-embolism in atrial fibrillation: the anticoagulation and risk factors in atrial fibrillation (ATRIA), Circulation 119 (2009) 1363–1369. R.G. Hart, E. JW, B. KS, M.M. MS, I. AJ, Stroke prevention in atrial fibrillation patients with chronic kidney disease, Can. J. Cardiol. 29 (Suppl. 7) (2013) S71–S78. S. Connolly, M. Ezekowitz, S. Yusuf, J. Eikelboom, J. Oldgren, A. Parekh, et al., RE-LY steering committee and investigators. Dabigatran versus warfarin in patients with atrial fibrillation, N. Engl. J. Med. 361 (12) (2009) 1139–1151. M. Patel, K. Mahaffey, J. Garg, G. Pan, D. Singer, W. Hacke, et al., Rivaroxaban versus warfarin in atrial fibrillation, N. Engl. J. Med. 365 (2011) 883–891. C. Granger, J. Alexander, J. McMurray, R. Lopes, E. Hylek, M. Hanna, et al., ARISTOTLE committees and investigators. Apixaban versus warfarin in patients with atrial fibrillation, N. Engl. J. Med. 365 (11) (2011) 981–992. H. Reinecke, C. Engelbertz, W.R. Schäbitz, Pre-venting stroke in patients with chronic kidney disease and atrial fibrillation: benefit and risks of old and new oral anticoagulants, Stroke 44 (2013) 2935–2941. K. Matsushita, B.K. Mahmoodi, M. Woodward, J.R. Emberson, T.H. Jafar, S.H. Jee, et al., Levey AS; chronic kidney disease prognosis consortium. Comparison of risk prediction using the CKD-EPI equation and the MDRD study equation for estimated glomerular filtration rate, JAMA 307 (2012) 1941–1951. A. Levey, E. de Jong, J. Coresh, M. El Nahas, B. Astor, K. Matsushita, et al., The definition, classification, and prognosis of chronic kidney disease: KDIGO controversies conference report, Kidney Int. 80 (2011) 17–28. G. Lip, R. Nieuwlaat, R. Pisters, D. Lane, H. Crijns, Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation, Chest 137 (2010) 263–272. G. Lip, L. Frison, J. Halperin, D. Lane, Comparative validation of a novel risk score for predicting bleeding risk in anticoagulated patients with atrial fibrillation: the HASBLED (hypertension, abnormal renal/liver function, stroke, bleeding, history of predisposition, labile INR, elderly, drugs/alcohol concomitantly) score, J. Am. Coll. Cardiol. 57 (2011) 173–180. D.R. Holmes Jr., S. Kar, M.J. Price, B. Whisenant, H. Sievert, S.K. Doshi, et al., Prospective randomized evaluation of the watchman left atrial appendage closure device in patients with atrial fibrillation versus long-term warfarin therapy: the PREVAIL trial, J. Am. Coll. Cardiol. 64 (1) (2014 Jul 8) 1–12. Y. Kitai, N. Ozasa, T. Morimoto, B. Bao, Y. Furukawa, Y. Nakagawa, et al., Kimura T; CREDO-Kyoto registry investigators. Prognostic implications of anemia with or without chronic kidney disease in patients undergoing elective percutaneous coronary intervention, Int. J. Cardiol. 168 (6) (2013 Oct 15) 5221–5228. J.C. Murphy, R.A. Kozor, G. Figtree, P.S. Hansen, H.H. Rasmussen, M.R. Ward, et al., Procedural and in-patient outcomes in patients aged 80 years or older undergoing contemporary primary percutaneous coronary intervention, EuroIntervention 8 (2012) 912–919. Z. Harel, M. Sholzberg, P.S. Shah, K. Pavenski, S. Harel, R. Wald, et al., Impact of glomerular filtration estimate on bleeding risk in very old patients treated with vitamin K antagonists. Results of EPICA study on the behalf of FCSA (Italian federation of anticoagulation clinics), Thromb. Haemost. 107 (6) (2012 Jun) 1100–1106. D. Poli, E. Antonucci, M. Zanazzi, E. Grifoni, S. Testa, W. Ageno, et al., Comparisons between novel oral anticoagulants and vitamin K antagonists in patients with CKD, J. Am. Soc. Nephrol. 25 (3) (2014 Mar) 431–442.
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Impact of chronic kidney disease on left atrial appendage occlusion for stroke prevention in patients with atrial fibrillation☆ Joelle Kefer a,⁎, Apostolos Tzikas b, Xavier Freixa c, Samera Shakir d, Sameer Gafoor e,w, Jens Erik Nielsen-Kudsk f, Sergio Berti g, Gennaro Santoro h, Adel Aminian i, Ulf Landmesser j, Fabian Nietlispach d,j, Reda Ibrahim k, Paolo Luciano Danna l, Edouard Benit m, Werner Budts n, Francis Stammen o, Tom De Potter p, Tobias Tichelbäcker q, Steffen Gloekler d, Prapa Kanagaratnam r, Marco Costa s, Ignacio Cruz-Gonzalez t, Horst Sievert e, Wolfgang Schillinger q, Jai-Wun Park u, Bernhard Meier d, Heyder Omran v a
Cliniques universitaires Saint-Luc, Brussels, Belgium Interbalkan European Medical Center, Thessaloniki, Greece Hospital Clinic of University of Barcelona, Barcelona, Spain d University Hospital of bern, Bern, Switzerland e CardioVascular Center Frankfurt, Frankfurt, Germany f Aarhus University Hospital, Skejby, Denmark g Heart Hospital, Fondazione C.N.R. Regione Toscana, Massa, Italy h Ospedale Careggi di Firenze, Florence, Italy i Centre Hospitalier Universitaire de Charleroi, Charleroi, Belgium j University Hospital of Zurich, Zurich, Switzerland k Montreal Heart Institute, Montreal, Canada l Ospedale Luigi Sacco, Milan, Italy m Jessaziekenhuis, Hasselt, Belgium n UZ Leuven, Leuven, Belgium o AZ Delta Roeselare, Roeselare, Belgium p OLV Ziekenhuis, Aalst, Belgium q Universitätsmedizin Göttingen, Göttingen, Germany r Imperial College Healthcare NHS Trust, London, United Kingdom s Coimbra University Hospital Centre, Coimbra, Portugal t University Hospital of Salamanca, Salamanca, Spain u Asklepios Hospital Hamburg, Hamburg, Germany v University Hospital of Bonn, Bonn, Germany w Seattle Heart and Vascular, Seattle, WA, USA b c
a r t i c l e
i n f o
Article history: Received 12 October 2015 Received in revised form 26 November 2015 Accepted 1 January 2016 Available online 9 January 2016 Keywords: Left atrial appendage Stroke Bleeding Renal failure Amplatzer cardiac plug
a b s t r a c t Background: Left atrial appendage occlusion (LAAO) using the Amplatzer cardiac plug (ACP) is a preventive treatment of atrial fibrillation related thromboembolism. Aim: To assess the safety and efficacy of LAAO in patients with chronic kidney disease (CKD). Methods: Among the ACP multicentre registry, 1014 patients (75 ± 8 yrs) with available renal function were included. Results: Patients with CKD (N = 375, CHA2DS2-VASc: 4.9 ± 1.5, HASBLED: 3.4 ± 1.3) were at higher risk than patients without CKD (N = 639, CHA2DS2-VASc: 4.2 ± 1.6, HASBLED: 2.9 ± 1.2; p b 0.001 for both). Procedural (97%) and occlusion (99%) success were similarly high in all stages of CKD. Peri-procedural major adverse events (MAE) were observed in 5.1% of patients, 0.8% of death, with no difference between patients with and those without CKD (6.1 vs 4.5%, p = 0.47). In patients with complete follow-up (1319 patients years), the annual stroke + transient ischaemic attack (TIA) rate was 2.3% and the observed bleeding rate was 2.1% (62 and 60% less than expected, similarly among patients with and those without CKD). Kaplan–Meier analysis showed a lower overall survival (84 vs 96% and 84 vs 93% at 1 and 2 yrs. respectively; p b 0.001) among patients with an eGFR b30 ml/min/1.73 m2.
☆ For all the authors: there are no conflicts of interest. The authors take the responsibility for all aspects of reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author at: Cliniques Universitaires Saint-Luc, University of Louvain, Division of Cardiology, Avenue Hippocrate, 10-2881, 1200 Brussels, Belgium. E-mail address: [email protected] (J. Kefer).
http://dx.doi.org/10.1016/j.ijcard.2016.01.003 0167-5273/© 2016 Elsevier Ireland Ltd. All rights reserved.
336
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
Conclusion: LAAO using the ACP has a similar procedural safety among CKD patients compared to patients with normal renal function. LAAO with ACP offers a dramatic reduction of stroke + TIA rate and of bleeding rate persistent in all stages of CKD, as compared to the expected annual risk. © 2016 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Percutaneous left atrial appendage occlusion (LAAO) is an alternative therapeutic option for stroke prevention in patients with atrial fibrillation (AF). [1–5]. The first randomised trial PROTECT AF [6] showed that LAAO with the WATCHMAN device (Boston Scientific, Marlborough, MA, USA) reduced mortality and stroke rate compared to warfarin, at 5 years follow-up. As compared with the initial attempts [3,7–8], experienced operators have shown a significant reduction of serious periprocedural complication rate around 4–5% [5,9]. Recently, it has been demonstrated that LAAO using the Amplatzer Cardiac Plug (ACP; St. Jude Medical, St Paul, MN, USA), reduces the theoretical risk of both, stroke and bleeding [5] in patients with AF. Renal failure patients are well known to be at high risk of procedural complications and to have a worse outcome after transcatheter interventions [10–11]. The risk of cardiovascular mortality increases in patients with chronic kidney disease (CKD) stage 3 [12]; recently, this category was subdivided in stage 3a and 3b [13]. The prevalence of AF in patients with CKD is high and renal failure represents a main risk factor for stroke, leading to a strong recommendation for anticoagulation therapy [14]. On the other hand, CKD significantly increases the risk of major bleeding in patients with AF [15–16]. Novel anticoagulants were approved for stroke prevention in AF [17–19] but their use is still controversial in the setting of CKD, especially among end-stage renal failure patients (eGFR b30 ml/min/m2) [20]. LAAO, as a non pharmacological approach, could be a potential alternative in patients with AF and CKD. The aim of the present study was to assess the procedural safety of LAAO in patients with AF and CKD, the efficacy on stroke and bleeding prevention, and finally to evaluate the impact of the stages of CKD on the clinical outcome after LAAO using the ACP. 2. Methods Between December 2008 and November 2013, consecutive patients undergoing LAAO using the ACP in 22 centres, were prospectively included in the multicentre ACP registry, published elsewhere [5]. Demographics, baseline characteristics, indications for LAAO, CHA2DS2-VASc and HAS-BLED scores, antithrombotic medication, procedural details, periprocedural adverse events, clinical and echocardiographic follow-up were prospectively collected in a dedicated database. Among this cohort, patients with available serum creatinine and glomerular filtration rate values were included in the current substudy. An informed consent was obtained in each patient; the study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki. 2.1. Renal failure The estimated glomerular filtration rate (eGFR) was calculated with the CKD-EPI formula [21]. By using the creatinine serum level, the age, the gender and the race of the patient, the CKD-EPI equation was expressed as: GFR ¼ 141 minðScr=κ; 1Þα maxðScr=κ; 1Þ−1:209 0:993Age 1:018 ½if female 1:159 ½if black: According to the guidelines [22], CKD was defined as eGFR b60 ml/min per 1.73 m 2 and CKD stages were classified based on the eGFR values: stage 1 = ≥ 90 ml/min/1.73 m 2 , stage 2 = 60 to 89 ml/min/1.73 m 2 , stage 3a = 45 to 59 ml/min/1.73 m 2 , stage 3b = 30 to 44 ml N/min/1.73 m 2 , stage 4 = 15 to 29 ml/min/1.73 m2 and stage 5 = b15 ml/min/1.73 m2. In addition, patients on dialysis and those with previous renal transplantation were noted. 2.2. Procedural success Procedural success was defined as successful implantation of the ACP in the left atrial appendage (LAA).
2.3. Periprocedural complications Periprocedural complications (occuring during 0–7 days after procedure or before hospital discharge, whichever last) included death, myocardial infarction, stroke, transient ischaemic attack (TIA), systemic embolism, air embolism, device embolisation, major bleeding (requiring surgery or transfusion) and cardiac tamponade. Periprocedural major adverse events (MAE) included death, stroke, systemic embolism and complication requiring major surgical or endovascular intervention (major bleeding, tamponade, device migration treated by snare or surgery) occurring between 0.7 days post procedure or before hospital discharge, whichever latest. 2.4. Clinical follow-up Patient survival and occurrence of clinical events during the follow-up were determined by review of medical records or phone contact of patients implanted successfully. Adverse events during follow-up included death (cardiovascular or noncardiovascular), stroke, TIA, systemic embolism and major bleeding. Antithrombotic medication was recorded at the admission date and at last follow-up visit. The recommendation by the manufacturer is to give a low dose of acetylsalicylic acid for a long period with clopidogrel (75 mg/d) for at least 1 month. So, a successful LAAO implies the absence of anticoagulant therapy. Nevertheless, the choice and the duration of antithrombotic therapy was individualised, depending on the patient characteristics and physician preference. 2.5. Efficacy on stroke, TIA and systemic embolism prevention LAAO efficacy on stroke, TIA and systemic embolism prevention was tested by comparing the actual event rate at follow-up with the predicted event rate by the CHA2DS2-VASc score [23]. Individual patient annual risk was recorded and the average annual risk for the whole study population was calculated. The total number of thromboembolic events (stroke, TIA and systemic embolism) during both the periprocedural and follow-up periods was divided by the total patient-years of follow-up and was multiplied by 100 in order to get the actual annual rate of thromboembolism. Thromboembolism reduction was calculated as follows (estimated % − actual% event rate) / estimated % event rate. 2.6. Efficacy on bleeding prevention The total number of major bleeding events during both the periprocedural and followup periods was compared with the events predicted by the HAS-BLED score [24]: bleeding reduction was calculated as follows: (estimated% − actual% event rate) / estimated% event rate. 2.7. Echocardiographic follow-up Implanted patients underwent a transoesophageal echocardiography (TOE) at followup. Residual leak was graded using the width of the Doppler colour jet as trivial (b1 mm), mild (1–3 mm) or significant (N3 mm). Occlusion success was defined as a procedural success without a significant residual leak. The presence of thrombus on the device was noted as well. 2.8. Statistical analysis Continuous variables are presented as mean ± 1 standard deviation. Categorical variables are presented as counts and percentages. Continuous variables were tested by using the independent samples t-test and categorical variables by using the Fischer's exact test. Univariate analysis was done using the Cox proportional-hazards method. Estimates for freedom from the composite of death and MAE were obtained by the Kaplan–Meier estimation method. A p value b0.05 was considered statistically significant. Analyses were performed using the SPSS version 15.0 (SPSS Inc., Chicago, Illinois).
3. Results 3.1. Patients Of the cohort of 1053 patients enrolled in the ACP multicentre registry, serum creatinine and eGFR values were available in 1014 and they were included in the study. Baseline characteristics are listed in Table 1. There were 183 patients categorised in stage 1, 456 in stage 2,
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
337
176 in stage 3a, 119 in stage 3b, 61 in stage 4 and 19 in stage 5. Fourteen patients were on dialysis and 3 other patients had a previous renal transplant. Because of the low number of patients with renal replacement therapy, no further analysis was performed in this group. The mean eGFR value was calculated at 81.8 ± 12.5 ml/min/1.73 m2 in patients with normal renal function (N = 639) and 40.9 ± 13.9 ml/min/1.73 m2 among patients with CKD (N = 375; p b 0.0001). Patients with CKD were older, more frequently female and had a higher prevalence of diabetes mellitus, congestive heart failure, and coronary artery disease. As compared to patients with normal renal function, CKD patients had a significantly higher CHADS2 (3 ± 1.4 vs 2.6 ± 1.3), CHA2DS2-VASc (4.9 ± 1.5 vs 4.2 ± 1.6) and HAS-BLED (3.4 ± 1.3 vs 2.9 ± 1.1) scores; p b 0.0001 for all comparisons.
7 of them were successfully snared, and one was successfully removed surgically. Of the 16 reported tamponades (1.6%), 2 were fatal. Major bleedings (1.2%) were mainly due to a vascular complication at the femoral puncture site (6/14). Patients with an eGFR b60 ml/min/1.73 m2 had not more MAEs than patients with an eGFR N60 ml/min/1.73 m2. More importantly, MAEs did not differ for patients with endstage renal disease (eGFR b30 ml/min/1.73 m2) as compared to patients with an eGFR N30 ml/min/1.73 m2: 3.7 vs 5.2%; p = 0.79. There was no need for new dialysis therapy among the total cohort, especially among patients with pre-existing renal failure. Among the 28 unsuccessfully attempted patients (mainly due to inappropriate size of the LAA), only one died within the periprocedural period due to heart failure.
3.2. Procedure
3.4. Follow-up
Procedural success was similarly high in patients with or without CKD (98% vs 97%; p = 0.23). Procedural characteristics are described in Table 2. There was no difference between groups neither in the mean size of the implanted ACP nor in the need for change of ACP size. The sizes 22 and 24 mm of the ACP device were the most frequently used (20% of cases for both), followed by 26 mm (16%), 30 mm (14%), 28 mm (11%), 20 mm (10%), 18 mm (5%) and 16 mm (3%). The distribution of device sizes was similar between patients with normal renal function and patients with CKD. Patients with normal renal function underwent more frequently combined interventions (31% vs 19%; p = 0.001), mainly driven by the number of coronary angiograms (13% vs 7%; p = 0.003). The mean volume of contrast used was 122 ± 80 ml.
The follow-up was complete in 967 of 986 patients with successful LAAO (98%). The mean duration of follow-up was 498 days (interquartile range 186–753 days) resulting in a total of 1319 patients-years. There were 9 strokes (0.9%), 9 TIAs (0.9%) and 15 major bleedings (1.5%) reported during the follow-up period, without any difference between groups. Overall survival (84% at 1 and 2 yrs) and event-free survival (81% at 1 and 2 yrs) were significantly lower in patients with an eGFR b30 ml/min/1.73 m2 (stages 4 + 5) than in other patients (stage 1 + 2 + 3) for whom overall survival was 96% at 1 yr., and 93% at 2 yrs.; p b 0.001 (Fig. 1), and event-free survival was 92% at 1 yr., and 86% at 2 yrs.; p = 0.007 (Fig. 2). The causes of death were more frequently cardiac among patients in stage 4 + 5 (6.9%) than in stage 1 + 2 + 3 (1.4%; p = 0.007) but never device-related (Table 4). The annual rate of non fatal MAE (stroke, TIA and major bleeding) was not significantly different among patients with endstage renal failure than in patients with an eGFR N30 ml/min/1.73 m2 (4.05% vs 4.5%; p = 1.00). With univariate analysis, eGFR (hazard ratio: 0.986; confidence interval: 0.976–0.995; p = 0.004) and HASBLED score (hazard ratio: 1.43; confidence interval: 1.02–2.00; p = 0.037) were associated with death at follow-up. By multivariare analysis, eGFR b30 ml/min/1.73 m2 was a predictor of total MAEs during followup (hazard ratio: 2.72; confidence interval: 1.51–4.90; p = 0.001),
3.3. Periprocedural complications The rate of periprocedural MAEs was 5.1% in the total cohort without any difference between groups (Table 3). Eight deaths (0.8%) were reported including 2 tamponades, 2 arrhythmias, 1 myocardial infarction, 1 intracranial haemorrhage and 2 device embolisations (one resulting in procedural death and another one treated by surgery but resulting in death). There were 8 additional device embolisations:
Table 1 Baseline characteristics. Characteristics
Age (yrs) Gender BMI Congestive heart failure Hypertension Diabetes Coronary artery disease Carotid disease Previous stroke/TIA Thromboembolism on anticoagulant Previous major bleeding Permanent AF Paroxysmal/persistent AF eGFR MDRD (ml/kg/1.73 m2) eGFR CKD-EPI (ml/kg/1.73 m2) Serum creatinine (mg/dl) Dialysis Renal transplant CHADS2 score CHA2DS2-VASc score HASBLED score
All
mean ± sd M/F mean ± sd n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) mean ± sd mean ± sd mean ± sd n (%) n (%)
No CKD
CKD
Stage 1 + 2
Stage 3a
Stage 3b
Stage 4
Stage 5
N = 1014
N = 639
N = 176
N = 119
N = 61
N = 19
N = 375
74.9 ± 8.4 630/384 27.3 ± 4.8 259 (25) 860 (85) 299 (29) 354 (35) 83 (8) 384 (38) 162 (16) 466 (46) 573 (56) 439 (43) 69.7 ± 27.5 66.8 ± 23.7 1.1 ± 0.8 14 (1) 3 (0.3) 2.8 ± 1.3 4.4 ± 1.6 3.1 ± 1.2
73.1 ± 8.4 425/215 27.1 ± 4.9 124 (19) 536 (84) 152 (24) 183 (28) 50 (8) 264 (41) 99 (15) 282 (44) 361 (56) 278 (43) 86 ± 19 81.8 ± 12.5 0.8 ± 0.1 0 0 2.6 ± 1.3 4.2 ± 1.6 2.9 ± 1.1
77.8 ± 7.1 95/79 27.5 ± 4.8 54 (31) 157 (90) 58 (33) 74 (42) 6 (3) 55 (31) 36 (20) 84 (48) 96 (55) 78 (45) 53.1 ± 4.4 52.6 ± 4.4 1.2 ± 0.1 0 1 (0.5) 3 ± 1.2 4.7 ± 1.5 3.1 ± 1.1
79.2 ± 7.2 67/51 27.6 ± 4.4 40 (34) 110 (93) 49 (41) 62 (52) 15 (13) 40 (34) 0 58 (49) 71 (60) 47 (40) 38.4 ± 4.5 37.3 ± 4.6 1.6 ± 0.2 0 1 (0.8) 3.2 ± 1.3 5.1 ± 1.4 3.5 ± 1.3
76.3 ± 8.3 31/30 27.2 ± 5 34 (56) 54 (88) 28 (46) 26 (43) 9 (15) 16 (26) 4 (6) 26 (43) 31 (50) 30 (50) 24.6 ± 3.8 23.7 ± 3.8 2.3 ± 0.4 2 (3) 1 (1) 3.1 ± 1.5 5.1 ± 1.7 3.7 ± 1.3
76.9 ± 6.6 11/8 27.6 ± 5 7 (36) 13 (68) 8 (42) 9 (47) 3 (16) 9 (47) 3 (16) 13 (68) 14 (74) 5 (26) 9.9 ± 3.1 9.1 ± 2.7 5.4 ± 1.6 12 (63) 0 3 ± 1.7 4.7 ± 1.7 4.3 ± 1.3
77.9 ± 7.3 205/169 27.5 ± 4.7 135 (36) 324 (87) 147 (39) 171 (46) 33 (9) 120 (32) 62 (16) 184 (49) 212 (57) 161 (43) 41.7 ± 13.7 40.9 ± 13.9 1.7 ± 1 14 (3) 3 (0.8) 3 ± 1.4 4.9 ± 1.5 3.4 ± 1.3
p value
Stage 3 + 4 + 5
b0.0001 0.0003 0.21 b0.0001 0.23 b0.0001 b0.0001 0.63 0.003 0.65 0.11 0.94 0.94 b0.0001 b0.0001 b0.0001 0.0001 0.05 b0.0001 b0.0001 b0.0001
CKD = chronic kidney disease; BMI = body mass index; TIA = transient ischaemic attack; AF = atrial fibrillation; eGFR = estimated glomerular filtration rate; MDRD = modification of diet in renal disease; CKD-EPI = chronic kidney disease – epidemiology collaboration.
338
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
Table 2 Procedural characteristics. Characteristics
Procedural success Size ACP (mm) First ACP selected implanted Combined intervention PFO closure ASD closure Coronary angiogram Coronary angioplasty TAVI Atrial fibrillation ablation Via transseptal puncture Via PFO
All
n (%) mean ± sd n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%)
No CKD
CKD
Stage 1 + 2
Stage 3a
Stage 3b
Stage 4
Stage 5
Stage 3 + 4 + 5
N = 1014
N = 639
N = 176
N = 119
N = 61
N = 19
N = 375
986 (97) 24.3 ± 3.7 835 (82) 271 (27) 61 (6) 11 (1) 108 (10) 57 (5) 16 (1) 18 (2) 916 (90) 98 (10)
618 (97) 24.2 ± 3.7 529 (83) 199 (31) 50 (8) 7 (1) 82 (13) 37 (6) 8 (1) 15 (2) 566 (89) 73 (11)
175 (99) 24.2 ± 3.7 147 (83) 28 (16) 4 (2) 0 8 (4) 12 (7) 2 (1) 2 (1) 165 (94) 11 (6)
115 (97) 24.2 ± 3.7 92 (77) 24 (20) 5 (4) 3 (2) 6 (5) 5 (4) 4 (3) 1 (1) 107 (90) 12 (10)
59 (97) 24.8 ± 3.5 52 (85) 15 (24) 1 (2) 1 (2) 9 (15) 2 (3) 2 (3) 0 60 (98) 1 (2)
19 (100) 25.6 ± 3.9 15 (79) 5 (26) 1 (5) 0 3 (16) 1 (5) 0 0 18 (95) 1 (5)
368 (98) 24.2 ± 3.7 306 (82) 72 (19) 11 (3) 4 (1) 26 (7) 20 (5) 8 (2) 3 (1) 350 (93) 25 (7)
p value
0.23 0.81 0.66 0.001 0.001 1 0.003 0.88 0.30 0.08 0.02 0.01
ACP = Amplatzer cardiac plug; PFO = patent foramen ovale; ASD = atrial septal defect; TAVI = transcatheter aortic valve implantation; CKD = chronic kidney disease.
but was not a predictor for stroke (hazard ratio: 2.24; confidence interval: 0.27–17.9; p = 0.44), or for major bleeding (hazard ratio: 1.24; confidence interval: 0.16–9.49; p = 0.05). The actual annual stroke + TIA rate (periprocedural and follow-up period) was 2.3%, while the expected annual thromboembolism risk was calculated by the CHA2DS2-VASc score at 5.6%, which translates into 62% risk reduction. The observed annual major bleeding rate (periprocedural and follow-up period) was 2.1% and the annual risk of bleeding estimated by the HASBLED score was 5.3% (60% reduction). Fig. 3 shows the distribution of the event risk reduction stratified by the stages of eGFR. While the risk of bleeding increases with the stage of eGFR, the event reduction was still persistent after LAAO in all stages of CKD, especially in the stage 5 group. Medication at last follow-up was limited to acetylsalicylic acid in 84% of patients, while only 4.7% of them were still under anticoagulant therapy, with no difference between patients with and those without CKD (3.7 vs 3.8%, p = NS). In details, anticoagulant agents used in patients with CKD were vitamin K antagonist (2.5%), NOAC (0.8%) or LMWH (0.2%) and were not different in patients with normal renal function (3.5%, 1.6%, 0.1%, respectively; p = NS). At TOE follow-up (mean 204 ± 199, median 141 days) obtained among 609 patients, thrombus on the device was observed in 27 cases (4.4%) with no difference between groups; they were all treated conservatively with no consequent stroke or other adverse events during follow-up. A residual leak was noted in 71 patients and graded as trivial in 26, moderate in 31, and significant in 11 cases. Therefore, the occlusion success was 98.8%.
4. Discussion The main findings of this paper are: 1. The procedural safety of LAAO with the ACP is similarly high in patients with CKD as in patients with normal renal function, with no impact of the stages of CKD on the periprocedural MAEs. 2. The efficacy on stroke reduction is very high and persistent in all stages of CKD. 3. LAAO with the ACP, allowing to stop anticoagulant after a procedural success, offers a dramatic reduction of bleeding risk even if this theoretical risk increases with the severity of CKD. 4. Overall survival was lower in patients with an eGFR b30 ml/min/m2 (stage 4 + 5) compared to other patients (stage 1 + 2 + 3). However, the rate of non fatal MAEs during follow-up (stroke, TIA, and major bleeding) was not higher among patients with end-stage renal failure. 4.1. Procedural safety This paper confirms that LAAO is a fairly safe procedure even in a high risk population with a reported periprocedural mortality of 0.8%. The rate of MAEs (5.1%) is lower than in the initial ACP European registry [7] or the PROTECT AF study [3], reporting 7.3 and 7.7% of MAEs respectively, but higher than in the more recent publications CAP [9] registry (3.7%) and PREVAIL [25] study (4.2%). The patients included in our study were treated by both experienced teams and in beginners: this may explain that our results range in the middle between early and recent publications. Patients with CKD are well
Table 3 Peri-procedural complications. Characteristics
MAE peri-procedural Death Stroke Device embolisation Device snared Device removed surgically Major bleeding Tamponade Other complications TIAS Air embolisation Total complications
All
No CKD
CKD
p value
Stage 1 + 2
Stage 3a
Stage 3b
Stage 4
Stage 5
Stage 3 + 4 + 5
N = 1014
N = 639
N = 176
N = 119
N = 61
N = 19
N = 375
n (%) n (%) n (%) n (%) n (%) n (%) n (%) n (%)
52 (5.1) 8 (0.8) 9 (0.8) 8 (0.8) 7 (0.7) 1 (0.01) 13 (1.2) 14 (1.4)
29 (4.5) 5 (0.8) 4 (0.6) 5 (0.7) 5 (0.7) 1 (0.1) 6 (1) 9 (1.5)
10 (5.6) 0 3 (2) 0 0 0 3 (2) 4
10 (8.4) 2 2 2 1 1 3 1
3 (4.9) 1 0 1 1 0 1 0
0 0 0 0 0 0 0 0
23 (6.1) 3 (0.8) 5 (1.3) 3 (1) 2 (0.5) 1 (0.2) 7 (1.8) 5 (1.6)
0.47 1.0 0.30 1.0 1.0 1.0 0.24 1.0
n (%) n (%) n (%)
4 (0.4) 4 (0.4) 60 (5.9)
2 (0.3) 4 (0.6) 35 (5.4)
1 0 11 (6.2)
1 0 11 (9.2)
0 0 3 (0.5)
0 0 0
2 (0.5) 0 25 (6.6)
0.62 0.30 0.49
CKD = chronic kidney disease; MAE = major adverse event; TIA = transient ischaemic attack.
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
339
Fig. 1. Overall survival after LAAO. Panel A: Kaplan–Meier analysis showing the effect of the eGFR stages on overall survival. Panel B: Kaplan–Meier analysis showing that patients in eGFR stage 4 + 5 had reduced overall survival. eGFR = estimated glomerular filtration rate; LAAO = left atrial appendage occlusion.
known to be at higher risk for procedural complications after transcatheter interventions. Yamamoto et al. [11] showed a higher rate of early complications after TAVI in patients with an eGFR b45 ml/min/1.73 m2. Kitai et al. [26] and Murphy et al. [27] published the negative impact of renal failure on the early outcome after coronary angioplasty. In our series, the rate of periprocedural MAEs remained fairly low in all groups with no significant difference among patients with or without renal impairement. It is remarkable that in patients with CKD stages 4 and 5 (eGFR b30 ml/min/1.73 m2) there was no periprocedural stroke. The fact that LAAO is a transvenous and not a transarterial intervention, with no manipulation of catheter in the aorta, could potentially explain this absence of impact of the renal function on the procedural complications.
4.2. Efficacy on stroke reduction In the present study, the observed annual stroke + TIA rate (periprocedural and follow-up period) was 2.3%, while the expected annual thromboembolism risk was calculated by the CHA2DS2-VASc score at 5.6%, which is 62% risk reduction. Even in patients with CKD who had a higher theoretical stroke risk (6.3%), this reduction of observed events after LAAO persisted (2.7%, reduction of 60%) and was not inferior to that of patients with normal renal function (2.1%, 60% reduction; p = NS). Yamamoto et al. [11] reported a higher stroke rate after TAVI in patients with CKD; it was not the case in our study after LAAO. Importantly, as compared to the PROTECT-AF study, the CHADS2 score of our population was higher (2.8 versus 2.2), but the observed stroke rate was similar (2.3 versus 2.2%/year).
4.3. Impact on bleeding The observed major bleeding rate (periprocedural and follow-up period) was 2.1% and the annual risk of bleeding estimated by the HASBLED score was 5.3% (60% reduction). The bleeding risk increased with the GFR stage from 5.6% in the stage 3a to 9.5% in the stage 5, whereas the reduction of observed events was still persistent in all subgroups of GFR, not inferior among patients with CKD than in patients with normal renal function (−54% vs −64%, p = NS). Harel et al. [28] showed that patients with renal impairement are at higher risk of bleeding complications under vitamin K antagonist; Poli et al. [29] showed that novel anticoagulants are not more efficient in prevention of hemorrhagic complications among patients with CKD. Because of renal elimination, the new oral anticoagulants have a prolonged halflife in patients with CKD, resulting in enhanced antithrombotic efficacy and increased bleeding risk. Importantly, patients with an eGFR b30 ml/min/1.73 m2 were excluded from the 3 studies RELY, ROCKET and ARISTOTLE [17–19]. LAAO using the ACP allows stopping anticoagulant therapy after procedural success. In our study, at last follow-up, only 3.7% of CKD patients remained under anticoagulant therapy whereas 82% of them are on acetylsalicylic acid only. In this population with very high bleeding risk (6.5%), it is remarkable to observe only 3% (−54%) of annual major bleeding events while keeping the efficacy for stroke prevention.
4.4. Overall survival During follow-up, the present study showed a low rate of non fatal MAEs (0.9% stroke, 0.9% TIA and 1.5% major bleedings) and similar between patients with and without CKD. However, patients with an eGFR b 30 ml/min/1.73 m2 presented higher mortality at 2 years compared to those with an eGFR N 30 ml/min/1.73 m2. These cardiac deaths were mainly due to a worsening heart failure and not device
Table 4 Adverse events during follow-up.
Fig. 2. Event-free survival after LAAO. Kaplan–Meier analysis showing that patients in eGFR stage 4 + 5 had reduced event-free survival.
Stroke TIA Major bleeding Death Cardiac death
All
Stage 1 + 2 + 3
Stage 4 + 5
N = 967
N = 895
N = 72
9 (0.9) 9 (0.9) 15 (1.5) 61 (6.3) 18 (1.8)
9 (1.0) 8 (0.9) 14 (1.5) 51 (5.6) 13 (1.4)
0 1 (1.3) 1 (1.3) 10 (14) 5 (6.9)
TIA = transient ischaemic attack.
p value
1.00 0.50 1.00 0.01 0.007
340
J. Kefer et al. / International Journal of Cardiology 207 (2016) 335–340
[7]
[8]
[9]
[10]
[11]
[12] Fig. 3. Event (stroke and bleeding) rate after LAAO. Fig. 3 showing the distribution of the event risk reduction stratified by the stages of eGFR. Blue = observed event rate; Pink = expected event rate. LAAO = left atrial appendage occlusion. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
[13]
[14]
related, and probably reflect the expected and well-known higher risk of this population. In contrast, there was no excess of major bleeding observed in these patients who, nevertheless, had a higher theoretical risk at baseline of hemorrhagic complications. The protective effect of LAAO using the ACP is highlighted among CKD patients, directly related to the cessation of anticoagulant therapy in a context of increased risk of bleeding.
[15]
[16] [17]
[18]
5. Conclusions Despite a different risk profile at baseline, LAAO using the ACP has a similar procedural safety among CKD patients and patients with normal renal function. LAAO with ACP offers a dramatic reduction of stroke + TIA rate and of major bleeding rate, which is persistent in all stages of CKD, as compared to the expected annual risk. The protective effect of LAAO with the ACP followed by cessation of anticoagulant therapy, was highlighted in the cohort of renal failure patients at very high bleeding risk, in whom a significant reduction in major bleeding events was observed, while keeping the efficacy in stroke prevention although the vast majority of patients were left untreated by anticoagulants. Patients with end-stage renal failure had a lower overall survival, which was not device- related, with a similar rate of non fatal MAEs during follow-up.
[19]
[20]
[21]
[22]
[23]
[24]
References [25] [1] H. Sievert, M. Lesh, T. Trepels, H. Omran, A. Bartoelli, P. Dela Bella, et al., Percutaneous left atrial appendage transcatheter occlusion to prevent stroke in high-risk patients with atrial fibrillation, Circulation 105 (2002) 1887–1889. [2] P. Block, S. Burstein, P. Casale, P. Kramer, P. Teirstein, D. Williams, et al., Percutaneous left atrial appendage occlusion for patients in atrial fibrillation suboptimal for warfarin therapy, J. Am. Coll. Cardiol. 2 (2009) 594–600. [3] D. Holmes, V. Reddy, Z. Turi, S. Doshi, H. Sievert, M. Buchbinder, et al., Percutaneous closure of the left atrial appendage versus warfarin therapy for prevention of stroke in patients with atrial fibrillation: a randomised non-inferiority trial, Lancet 374 (2009) 534–542. [4] J. Camm, G. Lip, R. De Caterina, I. Savelieva, D. Atar, S. Hohnloser, et al., 2012 focused update of the ESC guidelines for the management of atrial fibrillation, Eur. Heart J. 33 (2012) 2719–2747. [5] A. Tzikas, S. Shakir, S. Gafoor, H. Omran, S. Berti, G. Santoro, et al., Left atrial appendage occlusion for stroke prevention in atrial fibrillation: multicentre experience with the amplatzer cardiac plug, EuroIntervention 10 (10) (Jan 22 2015)http://dx.doi. org/10.4244/EIJY15M01_06 (Epub ahead of print). [6] V.Y. Reddy, H. Sievert, J. Halperin, S.K. Doshi, M. Buchbinder, P. Neuzil, et al., PROTECT AF steering committee and investigators. Percutaneous left atrial appendage
[26]
[27]
[28]
[29]
closure vs warfarin for atrial fibrillation: a randomized clinical trial, JAMA 312 (2014) 1988–1998. J.W. Park, A. Bethencourt, H. Sievert, G. Santoro, B. Meier, K. Walsh, et al., Left atrial appendage closure with Amplatzer cardiac plug in atrial fibrillation: initial European experience, Catheter. Cardiovasc. Interv. 77 (2011) 700–706. B. YL, O. H, N. P, T. L, P. M, R. E, et al., Plaato (percutaneous left atrial appendage transcatheter occlusion) for prevention of cardioembolic stroke in nonanticoagulation eligible atrial fibrillation patients: results from the European plaato strudy, EuroIntervention 6 (2010) 220–226. V.Y. Reddy, D. Holmes, S. Doshi, P. Neuzil, S. Kar, Safety of percutaneous left atrial appendage closure: results from the Watchman left atrial appendage system for embolic protection in patients with AF (PROTECT AF) clinical trial and the continued access registry, Circulation 123 (2011) 417–424. M. Wanitschek, M. Pfisterer, A. Hvelplund, S. De Servi, O. Bertel, R. Jeger, et al., Longterm benefits and risks of drug-eluting compared to bare-metal stents in patients with versus without chronic kidney disease, Int. J. Cardiol. 168 (2013) 2381–2388. M. Yamamoto, K. Hayashida, G. Mouillet, T. Hovasse, B. Chevalier, A. Oguri, et al., Prognostic value of chronic kidney disease after transcatheter aortic valve implantation, J. Am. Coll. Cardiol. 62 (2013) 869–877. K. Matsushita, M. van der Velde, B.C. Astor, et al., Association of estimated glomerular filtration rate and albuminuria with all-cause andcardiovascular mortality in general population cohorts: a collaborative meta-analysis. Chronic Kidney Disease Prognosis Consortium, Lancet 375 (2010) 2073–2081. M. van der Velde, K. Matsushita, J. Coresh, et al., Lower estimated glomerular filtration rate and higher albuminuria are associated with all-cause and cardiovascular mortality. A collaborative meta-analysis of high-risk population cohorts, Kidney Int. 79 (2011) 1341–1352. A. Bonde, G. Lip, A.-L. Kamper, P. Hansen, M. Lamberts, K. Hommel, et al., Net clinical benefit of antithrombotic therapy in patients with atrial fibrillation and chronic kidney disease, J. Am. Coll. Cardiol. 64 (2014) 2471–2482. A.S. Go, M.C. Fang, N. Udaltsova, et al., For the ATRIA study investigators. Impact of proteinuria and glomerular filtration rate on risk of thrombo-embolism in atrial fibrillation: the anticoagulation and risk factors in atrial fibrillation (ATRIA), Circulation 119 (2009) 1363–1369. R.G. Hart, E. JW, B. KS, M.M. MS, I. AJ, Stroke prevention in atrial fibrillation patients with chronic kidney disease, Can. J. Cardiol. 29 (Suppl. 7) (2013) S71–S78. S. Connolly, M. Ezekowitz, S. Yusuf, J. Eikelboom, J. Oldgren, A. Parekh, et al., RE-LY steering committee and investigators. Dabigatran versus warfarin in patients with atrial fibrillation, N. Engl. J. Med. 361 (12) (2009) 1139–1151. M. Patel, K. Mahaffey, J. Garg, G. Pan, D. Singer, W. Hacke, et al., Rivaroxaban versus warfarin in atrial fibrillation, N. Engl. J. Med. 365 (2011) 883–891. C. Granger, J. Alexander, J. McMurray, R. Lopes, E. Hylek, M. Hanna, et al., ARISTOTLE committees and investigators. Apixaban versus warfarin in patients with atrial fibrillation, N. Engl. J. Med. 365 (11) (2011) 981–992. H. Reinecke, C. Engelbertz, W.R. Schäbitz, Pre-venting stroke in patients with chronic kidney disease and atrial fibrillation: benefit and risks of old and new oral anticoagulants, Stroke 44 (2013) 2935–2941. K. Matsushita, B.K. Mahmoodi, M. Woodward, J.R. Emberson, T.H. Jafar, S.H. Jee, et al., Levey AS; chronic kidney disease prognosis consortium. Comparison of risk prediction using the CKD-EPI equation and the MDRD study equation for estimated glomerular filtration rate, JAMA 307 (2012) 1941–1951. A. Levey, E. de Jong, J. Coresh, M. El Nahas, B. Astor, K. Matsushita, et al., The definition, classification, and prognosis of chronic kidney disease: KDIGO controversies conference report, Kidney Int. 80 (2011) 17–28. G. Lip, R. Nieuwlaat, R. Pisters, D. Lane, H. Crijns, Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation, Chest 137 (2010) 263–272. G. Lip, L. Frison, J. Halperin, D. Lane, Comparative validation of a novel risk score for predicting bleeding risk in anticoagulated patients with atrial fibrillation: the HASBLED (hypertension, abnormal renal/liver function, stroke, bleeding, history of predisposition, labile INR, elderly, drugs/alcohol concomitantly) score, J. Am. Coll. Cardiol. 57 (2011) 173–180. D.R. Holmes Jr., S. Kar, M.J. Price, B. Whisenant, H. Sievert, S.K. Doshi, et al., Prospective randomized evaluation of the watchman left atrial appendage closure device in patients with atrial fibrillation versus long-term warfarin therapy: the PREVAIL trial, J. Am. Coll. Cardiol. 64 (1) (2014 Jul 8) 1–12. Y. Kitai, N. Ozasa, T. Morimoto, B. Bao, Y. Furukawa, Y. Nakagawa, et al., Kimura T; CREDO-Kyoto registry investigators. Prognostic implications of anemia with or without chronic kidney disease in patients undergoing elective percutaneous coronary intervention, Int. J. Cardiol. 168 (6) (2013 Oct 15) 5221–5228. J.C. Murphy, R.A. Kozor, G. Figtree, P.S. Hansen, H.H. Rasmussen, M.R. Ward, et al., Procedural and in-patient outcomes in patients aged 80 years or older undergoing contemporary primary percutaneous coronary intervention, EuroIntervention 8 (2012) 912–919. Z. Harel, M. Sholzberg, P.S. Shah, K. Pavenski, S. Harel, R. Wald, et al., Impact of glomerular filtration estimate on bleeding risk in very old patients treated with vitamin K antagonists. Results of EPICA study on the behalf of FCSA (Italian federation of anticoagulation clinics), Thromb. Haemost. 107 (6) (2012 Jun) 1100–1106. D. Poli, E. Antonucci, M. Zanazzi, E. Grifoni, S. Testa, W. Ageno, et al., Comparisons between novel oral anticoagulants and vitamin K antagonists in patients with CKD, J. Am. Soc. Nephrol. 25 (3) (2014 Mar) 431–442.
