論文ID: CJ-23-0553
Background: Left atrial appendage closure (LAAC) usually requires contrast medium during the procedure. However, patients with chronic kidney disease (CKD) are at high risk of developing contrast nephropathy. This study aimed to assess the safety and feasibility of zero-contrast LAAC in patients with CKD.
Methods and Results: Zero-contrast LAAC was attempted in 15 patients with CKD Stages 3b-5 who were not on hemodialysis. All procedures were performed successfully, without any periprocedural complications. At the 45-day follow-up, no device-related complications or acute kidney disease were observed.
Conclusions: The strategy of zero-contrast LAAC in patients with CKD can be an acceptable option.
In patients with in non-valvular atrial fibrillation (AF) and high bleeding risk, left atrial appendage closure (LAAC) is an alternative therapy to oral anticoagulants for stroke prevention. Chronic kidney disease (CKD) is one of the frequent factors for high bleeding risk in patients with AF.1 To confirm left atrial appendage (LAA) shape, device position, and sealing, LAAC is typically performed using contrast medium. However, patients with CKD undergoing LAAC are at risk of developing contrast nephropathy. Therefore, it is important to reduce or avoid the use of contrast medium to protect the remaining renal dysfunction. Some previous studies reported the feasibility and safety of LAAC without contrast;2,3 however, to date, evidence on zero-contrast LAAC is limited.
The aim of this study was to assess the safety and feasibility of zero-contrast LAAC in patients with CKD from a single center.
From November 2021 to May 2023, 82 consecutive patients underwent percutaneous LAAC using WATCHMAN FLX (Boston Scientific) in Kokura Memorial Hospital (Kitakyushu, Japan). Because of retrospective enrollment of patients into the study, the need for written informed consent was waived. No patients refused to participate in the study when contacted for follow-up. This study was approved by the Ethical Committee of Kokura Memorial Hospital and was conducted in accordance with the Declaration of Helsinki and the Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan. The study was registered to the University Hospital Medical Information Network Clinical Trials Registry (registration number: UMIN000045447).
We attempted zero-contrast LAAC in patients with moderate to severe CKD who were not on hemodialysis (serum creatinine >1.5 mg/dL or CKD Stages 3b-5 based on the Kidney Disease: Improving Global Outcomes [KDIGO] definition4). Before the LAAC procedure, all patients underwent non-contrast electrocardiogram-gated computed tomography (CT). To analyze the morphology and measure the size of the LAA, we used 3 mensio software (Pie Medical Imaging BV). In 2 cases, transesophageal echocardiography (TEE) was additionally performed due to unclear CT images. All LAAC procedures were performed under general anesthesia with TEE guidance. During the procedure, we checked device position, size and sealing using real-time X-ray and TEE measurements. The procedure was performed following a previously described protocol,5,6 except that no contrast medium was used. The devices were released after confirming device position, anchor, size, and seal (PASS criteria).5,6
We evaluated clinical outcomes at hospital discharge and 45 days after LAAC. For follow-up imaging, TEE was mainly performed at 45 days or 4 months, except in 3 patients in whom a small amount of contrast was injected for contrast-enhanced CT because their renal function was relatively good. Outcomes were determined on the basis of the Munich consensus document and previous literature.7 Clinical follow-up information was obtained from the medical records.
Continuous variables are presented as the mean±SD, whereas categorical variables are presented as frequencies with percentages.
Of the 82 patients who underwent LAAC using WATCHMAN FLX, zero-contrast LAAC was attempted in 15 (18.3%). The baseline clinical characteristics of the patients are presented in Table 1A. The mean age was 75.3±6.2 years, and 73.3% of patients were male. The mean serum creatinine concentration was 3.2±1.7 mg/dL, and the estimated glomerular filtration rate (eGFR) was 19.4±9.4 mL/min/1.73 m2. Two (13.3%) patients were CKD Stage 3b, 6 (40.0%) were CKD Stage 4, and 7 (46.7%) were CKD Stage 5 and were yet to undergo hemodialysis. The mean CHA2DS2-VASc and HAS-BLED scores were 4.0±1.1 and 4.0±0.96, respectively.
Baseline (A) Clinical, (B) Intraprocedural TEE, and (C) Procedural Characteristics (n=15)
| (A) Clinical characteristics | |
| Age (years) | 75.3±6.2 |
| Male sex | 11 (73.3) |
| BMI (kg/m2) | 24.1±2.9 |
| Rhythm at the procedure | |
| Sinus rhythm | 6 (40.0) |
| AF/AFL/AT | 9 (60.0) |
| Comorbidities | |
| Heart failure (NYHA Class III or IV) | 2 (13.3) |
| Prior stroke | 1 (6.7) |
| Prior intracranial bleeding | 2 (13.3) |
| Prior myocardial infarction | 4 (26.7) |
| Prior major bleeding (BARC 3) | 7 (46.6) |
| Chronic kidney disease | |
| Stage 3b | 2 (13.3) |
| Stage 4 | 6 (40.0) |
| Stage 5 | 7 (46.7) |
| Serum creatinine (mg/dL) | 3.2±1.7 |
| eGFR (mL/min/1.73 m2) | 19.4±9.4 |
| CHA2DS2-VASc score | 4.0±1.1 |
| HAS-BLED score | 4.0±1.0 |
| LVEF by TTE (%) | 55.0±11.4 |
| LA diameter by TTE (mm) | 46.0±6.4 |
| (B) Intraprocedural TEE | |
| LAA ostium diameter (mm) | |
| 0° | 22.1±3.8 |
| 45° | 20.0±3.4 |
| 90° | 19.6±3.3 |
| 135° | 20.9±4.1 |
| LAA ostium diameter (3D major axis; mm) | 24.0±3.6 |
| LAA ostium diameter (3D minor axis; mm) | 18.9±3.5 |
| LAA ostium area (3D; mm2) | 370.8±124.6 |
| LAA morphology | |
| Windsock | 6 (40.0) |
| Cactus | 1 (6.7) |
| Cauliflower | 4 (26.7) |
| Chicken wing | 4 (26.7) |
| (C) Procedural characteristics | |
| Anesthesia time (min) | 87.0±19.6 |
| Procedural time (min) | 25.0±13.2 |
| Fluoroscopy duration (min) | 3.8±2.3 |
| Preprocedural mean LA pressure (mmHg) | 14.9±4.1 |
| Device size | |
| 24 mm | 1 (6.7) |
| 27 mm | 4 (26.6) |
| 31 mm | 4 (26.6) |
| 35 mm | 6 (40.0) |
| No. devices use | 1 |
| No. partial recapture | 1.3±2.1 |
| Procedural success | 15 (100.0) |
| Zero-contrast procedural success | 13 (86.7) |
| Procedural complications | 0 (0) |
| TEE assessment after device implantation | |
| Peridevice leak | |
| <3 mm | 3 (20.0) |
| ≥5 mm | 0 (0) |
| Device compression before release (%) | |
| 0° | 19.6±7.9 |
| 45° | 20.0±6.1 |
| 90° | 18.0±5.8 |
| 135° | 17.0±4.4 |
| Residual trabeculation | 0 (0) |
Values are n (%), or mean±SD. 3D, 3-dimensional; AF, atrial fibrillation; AFL, atrial flatter; AT, atrial tachycardia; BARC, Bleeding Academic Research Consortium; BMI, body mass index; eGFR, estimated glomerular filtration rate; LA, left atrial; LAA, left atrial appendage; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; PCI, percutaneous coronary intervention; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography.
Intraprocedural TEE and procedural characteristics data are presented in Table 1B,C. The mean LAA ostium diameter (3-dimensional major axis) measured using TEE was 24.0±3.6 mm. The most common morphology of the LAA was Windsock, which was observed in 40% of cases. The procedural and fluoroscopy times were 25.0±13.2 and 3.8±2.3 min, respectively. No change in device size was noted, and no second device was used. The procedure success rate was 100% in 15 patients, of whom 2 (13.3%) eventually needed a minimum amount of contrast medium during the LAAC procedures. Because TEE images were unclear, 15 mL contrast medium was used in 1 patient used to confirm the bottom and depth of the LAA. Because it was challenging to confirm all circumferential sealing with TEE, 6 mL contrast medium was used in another patient to check for sealing of the 35-mm device. Neither of these patients developed acute kidney disease. There were no significant differences in background, procedure, and outcomes between groups in which contrast was and was not (zero-contrast) used (Table 2). Peridevice leak (PDL) at the end of the procedure was observed in 3 (20.0%) patients, but all leaks were <3 mm. No procedural complications and in-hospital deaths were noted.
(A) Baseline Characteristics, (B) Clinical Outcomes at Hospital Discharge and 45 Days After LAAC, (C) Images Follow-up After LAAC
| (A) | LAAC with contrast (n=2) |
LAAC without contrast (n=13) |
P value |
|---|---|---|---|
| Clinical characteristics | |||
| Age (years) | 73.0±8.5 | 75.6±6.5 | 0.60 |
| Male sex | 1 (50.0) | 10 (77.0) | 1.00 |
| BMI (kg/m2) | 24.4±3.1 | 24.1±3.1 | 1.00 |
| Rhythm at the procedure | 0.64 | ||
| Sinus rhythm | 0 | 6 (46.1) | |
| AF/AFL/AT | 2 (100.0) | 7 (53.8) | |
| Comorbidities | |||
| Heart failure (NYHA Class III or IV) | 1 (50.0) | 1 (7.7) | 0.60 |
| Chronic kidney disease | 0.83 | ||
| Stage 3b (%) | 0 | 2 (15.4) | |
| Stage 4 (%) | 1 (50.0) | 5 (38.5) | |
| Stage 5 (%) | 1 (50.0) | 6 (46.2) | |
| Serum creatinine (mg/dL) | 2.9±0.3 | 3.2±1.9 | 0.84 |
| eGFR (mL/min/1.73 m2) | 15.0±4.2 | 20.1±10.2 | 0.51 |
| CHA2DS2-VASc score | 4.0±0.0 | 3.9±1.3 | 0.93 |
| HAS-BLED score | 4.5±0.7 | 3.4±1.0 | 0.14 |
| LVEF by TTE (%) | 55.6±12.9 | 55.2±12.2 | 0.96 |
| LA diameter by TTE (mm) | 49.3±0.35 | 45.9±7.0 | 0.53 |
| Intraprocedural TEE | |||
| LAA ostium diameter (mm) | |||
| 0° | 25.5±7.4 | 21.6±3.3 | 0.23 |
| 45° | 19.4±6.4 | 20.1±3.3 | 0.81 |
| 90° | 19.4±5.7 | 19.6±3.4 | 0.96 |
| 135° | 23.0±10.7 | 20.6±3.3 | 0.49 |
| 3D major axis (mm) | 24.9±7.4 | 24.0±3.4 | 0.76 |
| 3D minor axis (mm) | 19.4±8.6 | 18.8±3.0 | 0.84 |
| LAA ostium area (3D; mm2) | 398.5±304.8 | 366.2±105.7 | 0.76 |
| LAA morphology | 0.33 | ||
| Windsock | 2 (100.0) | 4 (30.8) | |
| Cactus | 0 | 1 (7.7) | |
| Cauliflower | 0 | 4 (30.8) | |
| Chicken wing | 0 | 4 (30.8) | |
| Procedural characteristics | |||
| Anesthesia time (min) | 77.0±19.8 | 88.1±20.8 | 0.49 |
| Procedural time (min) | 30.5±9.2 | 23.9±14.3 | 0.54 |
| Fluoroscopy duration (min) | 3.0±0.0 | 3.9±2.6 | 0.64 |
| Mean LA pressure (mmHg) | 12.0±0.0 | 15.3±4.4 | 0.32 |
| Contrast medium (mL) | 10.5±6.4 | 0.0 | <0.001 |
| Device size | 0.73 | ||
| 24 mm | 0 | 1 (7.7) | |
| 27 mm | 1 (50.0) | 3 (23.1) | |
| 31 mm | 0 | 4 (30.8) | |
| 35 mm | 1 (50.0) | 5 (38.5) | |
| No. devices use | 1.0 | 1.0 | NA |
| No. partial recapture | 4.0±4.2 | 0.9±1.7 | 0.06 |
| Procedural success (%) | 2 (100) | 13 (100) | NA |
| TEE assessment at follow-up | |||
| Peridevice leak | 0 | 3 (23.1) | 0.49 |
| < 3 mm | 0 | 3 (23.1) | 0.49 |
| ≥5 mm | 0 | 0 | NA |
| Device compression (%) | |||
| 0° | 25.7±20.7 | 18.6±5.9 | 0.32 |
| 45° | 22.0±15.0 | 19.6±5.1 | 0.63 |
| 90° | 18.8±9.6 | 18.4±5.9 | 0.93 |
| 135° | 16.2±6.3 | 17.0±4.5 | 0.67 |
| Residual trabeculation | 0 | 0 | NA |
| (B) | LAAC with contrast (n=2) |
LAAC without contrast (n=13) |
P value |
| In-hospital | |||
| All-cause death | 0 | 0 | NA |
| Stroke event | 0 | 0 | NA |
| Any bleeding event | 0 | 1 (7.7) | 1.00 |
| Any procedural complication | 0 | 0 | NA |
| Antithrombotic treatment after the procedure | 0.85 | ||
| At hospital discharge | |||
| DOAC | 1 (50.0) | 4 (30.8) | |
| SAPT | 0 | 1 (7.7) | |
| Warfarin | 1 (50.0) | 2 (15.4) | |
| Warfarin+SAPT | 0 | 2 (15.4) | |
| DOAC+SAPT | 0 | 3 (23.1) | |
| None | 0 | 1 (7.7) | |
| At 45 days | |||
| All-cause death | 0 | 0 | NA |
| Stroke event | 0 | 0 | NA |
| Any bleeding event | 0 | 1 (7.7) | 1.00 |
| Serum creatinine (mg/dL) | 2.8±0.82 | 3.2±2.3 | 0.82 |
| eGFR (mL/min/1.73 m2) | 17.8±9.1 | 23.3±14.7 | 0.62 |
| (C) | |||
| Follow-up images available | 2 (100.0) | 13 (100.0) | NA |
| TEE imaging follow up | 1 (50.0) | 11 (84.6) | 1.00 |
| Peridevice leak at 45 days/4 months | 0/1 | 0/3 | |
| <3 mm | 1 (50.0) | 3 (23.0) | 1.00 |
| ≥5 mm | 0 | 0 | NA |
| Contrast-enhanced CT | 1 (50.0) | 2 (15.4) | 0.85 |
| Device-related thrombosis | 0 | 0 | NA |
Values are n (%), or mean±SD. CT, computed tomography; DOAC, direct oral anticoagulant; LAAC, left atrial appendage closure; NA, not applicable; SAPT, single antiplatelet therapy. Other abbreviations as in Table 1.
Regarding postprocedural antithrombotic regimens, the most common (33.3%) was direct oral anticoagulants only. At 45 days following LAAC, all patients underwent clinical follow-up (Table 3). No patients developed stroke or a device-related complications (device embolization, pericardial effusion, infective endocarditis). At 45 days after LAAC, a Bleeding Academic Research Consortium Type 3a bleeding event had occurred in 1 patient. This patient was 86-year-old man who had hematuria from prostate cancer and needed a transfusion of 2 units of red blood cells. Compared with baseline, there were no changes in serum creatinine concentrations and eGFR at 45 days after LAAC (3.1±2.0 mg/dL and 22.6±13.5 mL/min/1.73 m2, respectively).
Clinical Outcomes at Hospital Discharge and 45 Days After LAAC (n=15)
| In-hospital | |
| All-cause death | 0 (0) |
| Stroke event | 0 (0) |
| Any bleeding event | 1 (6.7) |
| Any procedural complication | 0 (0) |
| Antithrombotic treatment after the procedure | |
| At hospital discharge | |
| DOAC | 5 (33.3) |
| SAPT | 1 (6.7) |
| Warfarin | 3 (20.0) |
| Warfarin+SAPT | 2 (13.3) |
| DOAC+SAPT | 3 (20.0) |
| None | 1 (6.7) |
| At 45 days | |
| All-cause death | 0 (0) |
| Stroke event | 0 (0) |
| Any bleeding event | 1 (6.7) |
| Serum creatinine (mg/dL) | 3.1±2.0 |
| eGFR (mL/min/1.73 m2) | 22.6±13.5 |
Values are n (%), or mean±SD. Abbreviations as in Tables 1,2.
Follow-up imaging was performed in all patients (Table 4). TEE follow-up imaging was performed in 12 patients, of whom 2 and 10 underwent TEE at 45 days and 3 months, respectively. Contrast-enhanced CT was performed in 3 patients at the 4-month follow-up because renal function had improved in 2 patients with diuretic adjustment and treatment of multiple myeloma, and the other patient started on hemodialysis 107 days after LAAC because of worsening diabetic nephropathy. Minor leaks of <3 mm PDL were noted in 4 (26.7%) patients, all of whom were followed up at 4 months. No device-related thrombus was observed on follow-up imaging.
Imaging Follow-up After LAAC (n=15)
| Follow-up images available | 15 (100.0) |
| TEE imaging follow-up | 12 (80.0) |
| Peridevice leak at 45 days/4 months | 0 (0)/4 (26.7) |
| <3 mm | 4 (26.7) |
| ≥5 mm | 0 (0) |
| Contrast-enhanced CT | 3 (20.0) |
| Device-related thrombosis | 0 (0) |
Values are n (%). Abbreviations as in Tables 1,2.
In this case series, we investigated the feasibility of zero-contrast LAAC in patients with CKD. Thirteen of 15 (86.7%) patients underwent zero-contrast LAAC from the preoperative examination to the end of the LAAC procedure. No major complications were observed in any of the procedures. In the Japanese LAAC registries, renal failure, which is defined as eGFR <60 mL/min/1.73 m2, was found in 50.0–73.9% of patients.7,8 Nombela-Franco et al reported that 22.6% (80/355) of patients were CKD Stages 3b-5, the incidence of acute kidney injury was 9% following LAAC, and patients with worse baseline renal function were at higher risk of the development of acute kidney injury.9 Moreover, patients who developed acute kidney injury had worse clinical outcomes.9 Therefore, the use of contrast media should be avoided as much as possible in patients with CKD.
In our 15 patients, no procedural complications were noted, and all procedures were successful. The previous WATCHMAN FLX registry studies reported a mean procedure time ranging from 37.9 to 44.8 min, a mean fluoroscopy time of 10.3 min, and a mean contrast medium volume of 59.5 mL.5,10 In the present study, the procedure and fluoroscopy times were comparable to those of a routine LAAC procedure that uses contrast medium. Device size was determined on the basis of intraoperative TEE measurements; in the present study, the device size was unchanged. The improvement of TEE images, in addition to 2-dimensional TEE quality and 3-dimensional TEE function, enables us to provide detailed LAA measurements and guidance during the procedures. Furthermore, the evolution of devices made zero-contrast LAAC possible. The FLX ball allowed WATCHMAN FLX devices to safely move forward within the LAA. The WATCHMAN FLX is safer than previous generations of devices.11 Therefore, when TEE images are unclear or the implanter does not feel comfortable, the flexibility to use contrast medium may be a significant factor in ensuring patient safety.
The limitations of this study include its non-randomized design and small sample size. Zero-contrast LAAC was started after experience with 40 cases of WATCHMAN FLX in Kokura Memorial Hospital. Overall, this is the first study from Japan on a zero-contrast LAAC procedure.
This study demonstrated that zero-contrast LAAC in patients with CKD is feasible and safe. Therefore, zero-contrast LAAC could be an acceptable option in patients with CKD. Although non-contrast CT can be used to evaluate the LAA, TEE imaging is important both pre- and intraoperatively.
None.
This study did not receive any specific funding.
M.F. is a proctor for Boston Scientific. The other authors have no conflicts of interest to declare.
This study was approved by the Institutional Review Board of Kokura Memorial Hospital (#21090201).
