Left Atrial Appendage Closure During Coronary Bypass Surgery in Patients on Hemodialysis

Abstract

Background: Patients with end-stage renal disease on hemodialysis (ESRD-HD) have a lifelong risk of atrial fibrillation-related stroke. We compared clinical outcomes in ESRD-HD patients undergoing coronary artery bypass grafting (CABG) with and without concomitant left atrial appendage (LAA) closure.

Methods and Results: Of 2,783 consecutive patients undergoing isolated CABG between 2002 and 2020, 242 patients had ESRD-HD with sinus rhythm. The primary outcome was a composite of death and stroke. An inverse probability (IP)-weighted cohort was created based on the propensity score. The 2 IP-weighted groups had well-balanced baseline and surgical backgrounds, with an equivalent follow-up. Five-year stroke-free survival was significantly higher in patients with LAA closure (log-rank test, P=0.035). The adjusted hazard ratio of LAA closure for death and stroke was 0.43 (95% confidence interval [CI] 0.20–0.92; P=0.023). Competing risk analysis showed that LAA closure was significantly associated with a risk reduction of stroke (subhazard ratio 0.26; 95% CI 0.08–0.96; P=0.028). No significant difference was observed in adjusted risk ratios for reoperation for bleeding, new atrial fibrillation, 30-day mortality, and readmission for heart failure.

Conclusions: Concomitant LAA closure during CABG can reduce the risk of death and stroke in ESRD-HD patients with normal sinus rhythm.

The incidence of cardiovascular diseases increases with renal dysfunction, reaching its peak in end-stage renal disease on hemodialysis (ESRD-HD).¹ Among cardiovascular diseases in ESRD-HD patients, the risk of embolic stroke is high,² with most embolic strokes caused by thrombi originating from the left atrial appendage (LAA) due to atrial fibrillation (AF).^3–8

Anticoagulation is a cornerstone in preventing AF-induced stroke in the general population, but the risks and benefits of anticoagulation in ESRD-HD patients remain controversial.^9–15 In interpreting those study results, it also should be noted that most patients had permanent AF. In clinical practice, it is difficult to judge whether to start anticoagulation and how long to continue it, particularly in cases other than permanent AF, such as at first diagnosis or in cases of paroxysmal AF. The same problem is encountered during the postoperative management of ESRD-HD patients who have undergone cardiac surgery. Particularly in the early postoperative period, it is difficult to make judgments due to concerns about the risk of bleeding.

Recently, the Left Atrial Appendage Occlusion Study (LAAOS III), a multicenter randomized trial, showed that LAA closure during cardiac surgery is a useful non-pharmacological therapy to prevent postoperative embolic stroke in patients with AF.¹⁶ Conversely, in patients without AF, contradictory findings have been reported, namely that LAA closure did not reduce the incidence of stroke, but increased postoperative AF and readmission for heart failure.¹⁷ Since 2013, we have adopted LAA closure during CABG for almost all ESRD-HD patients, including those with sinus rhythm, to prevent embolic stroke during follow-up. There was room for further research on this topic because no data have been published on ESRD-HD patients with sinus rhythm. Therefore, the present study compared the short- and mid-term clinical outcomes of ESRD-HD patients with sinus rhythm undergoing CABG with and without concomitant LAA closure.

Methods

This is a retrospective study of single-center data. The Institutional Review Board of Juntendo University (Reference no. R2018-185; February 28, 2019) approved the study, and waived the need for written informed consent because of the retrospective study design. This study was performed in accordance with the principles of the Declaration of Helsinki.

Study Population

Of 2,783 consecutive patients undergoing isolated CABG at Juntendo University Hospital between July 2002 and December 2020, 256 had ESRD-HD. After excluding 14 patients with preoperative AF, LAA closure was concomitantly performed in 102 (42%) patients. The remaining 140 (58%) patients received CABG only.

Surgical Procedures

The off-pump technique was routinely used through a full median sternotomy. In situ skeletonized arterial grafts were preferably used to avoid aortic manipulation, as described previously.^18,19 The left internal thoracic artery (LITA) was grafted to the left anterior descending artery. The target of the right internal thoracic artery (RITA) was individualized; in most cases, the RITA was used to revascularize the lateral wall as an in situ graft through the transverse sinus. In some cases, the free RITA graft was used as a composite Y graft with the LITA. The inferior wall was revascularized using a saphenous vein or in situ gastroepiploic artery. The severity and location of atherosclerosis and calcification in the ascending aorta were assessed by enhanced computed tomography (CT) scans preoperatively and epiaortic ultrasound during surgery. A sealing device was used when the surgeon judged that partial clamping of the aorta increased the risk of embolism. To assess graft function, the transit-time flow trace was used after each graft had been completed and immediately before chest closure. The LAA was resected and closed using a horizontal mattress suture and over-and-over suture or ligated using a double purse-string suture.²⁰ An intraoperative esophageal echocardiogram confirmed complete closure of the LAA without any residual flow to the remaining structure. In addition, almost all patients underwent enhanced CT postoperatively to investigate the status of residual flow to the LAA. The procedure was performed manually, and no device was used to close the LAA.

Postoperative Management

All patients were monitored with continuous electrocardiographic telemetry for at least 1 week postoperatively. When episodes of arrhythmia were captured by an automatic alarm function, a 12-lead electrocardiogram was performed and assessed by physicians. For postoperative AF, a β-blocker or amiodarone was the first-choice therapy, and electrical cardioversion was considered for patients with postoperative AF lasting more than 24 h or unstable hemodynamics. Oral anticoagulant therapy with warfarin (target international normalized ratio 2.0–3.0) was started within 6 h of the onset and continued after discharge for patients with persistent AF or frequent episodes of paroxysmal AF. Continuous intravenous heparin (target activated partial thromboplastin time 50–80 s) was combined at the physician’s discretion. Patients routinely started aspirin, statins, and β-blockers on the first day after surgery and continued unless contraindicated.

Follow-up

Patients were regularly followed-up at Juntendo University Hospital in an unblinded manner, whereby physicians checked patients’ general condition and evaluated the results of chest X-rays, electrocardiography, and echocardiography. In addition, each patient was given a questionnaire asking about symptoms, cardiac and cerebrovascular events, cardiac-related readmission, cardiac interventions, and medications. Study coordinators telephoned patients who could not visit the hospital for some reason to ask them about their physical condition, symptoms, and adverse events. Several patients were lost to follow-up due to non-compliance with a hospital visit or relocation to other areas. Study coordinators continued telephone contact to follow up with any non-responders. Causes of death were determined on the basis of data obtained from family members, death certificates, hospital records, and autopsy records. Follow-up was closed on June 1, 2021. Two patients were lost to follow-up.

Endpoints, Definitions, and Data Collection

The primary endpoint was a composite of all-cause death and the first occurrence of stroke. Secondary endpoints were new-onset AF, re-operation for bleeding within the first 24 h after surgery, death within 30 days, admission for heart failure, and cardiac-related death (myocardial infarction, heart failure, sudden death). Stroke was defined as a rapid onset of new focal neurological symptoms lasting more than 24 h with any imaging evidence of acute brain ischemia consistent with the symptoms. New-onset AF was defined as an irregular cardiac rhythm without P waves lasting more than 60 min that required further administration of anti-arrhythmic drugs, cardioversion, or anticoagulation therapy. Mediastinitis was defined as a sternal infection requiring refixation of the sternum or omentoplasty.

All data were extracted from the institutional database in which data were prospectively collected following the definitions of the Society of Thoracic Surgeons National Adult Cardiac Database (https://www.sts.org/registries/sts-national-database/adult-cardiac-surgery-database).

Inverse Probability of Treatment Weighting

A propensity score was calculated for each patient by fitting a multivariable logistic regression model using LAA closure as the binary dependent variable and the following variables as independent variables:²¹ age, sex, body mass index, serum hemoglobin, smoking history, hypertension, diabetes, malignancy, history of embolic stroke, history of myocardial infarction or percutaneous coronary intervention, AF, number of territories of coronary artery disease, left ventricular ejection fraction, and operation date. The multiple imputation method was used to address missing variables. Inverse probability (IP) of treatment weighting was used to create a weighted cohort of patients to reduce the impact of selection bias and potential confounding (Supplementary Figure).²² Standardized differences were calculated to examine covariate balance after IP weighting, with a difference of ≤10% considered ideal.²³

Statistical Analysis

Bivariate differences were compared using the Chi-squared test or Fisher’s exact test for categorical data, as appropriate, and the unpaired t-test or Mann-Whitney U test for continuous variables after evaluation of the normality of distribution by the Kolmogorov-Smirnov test. Survival was estimated using the Kaplan-Meier method, in which patients still alive were censored at the date of their last follow-up, and was compared using the log-rank test. Cox proportional hazard regression was used to test the association of variables with death. Statistical analysis of time to a non-fatal event considered death a competing risk. The cumulative incidence function was estimated, and Gray’s test was used to determine differences.²⁴ Fine-Gray models were used to estimate subhazard ratios (sub-HRs) in competing risks analysis for time to events, with death as the competing risk.²⁵ Logistic regression calculated odds ratios for in-hospital outcomes (new-onset AF, re-operation for bleeding, death within 30 days). The following variables were included as covariates in multivariable models: age, sex, body mass index, serum hemoglobin, smoking history hypertension, diabetes, history of embolic stroke, history of myocardial infarction, left ventricular ejection fraction <40%, number of coronary diseases, emergency, operation time, number of distal anastomoses, bilateral internal thoracic artery graft, gastroepiploic artery graft, and aorta non-touch. For associations, we used a 2-sided test. P<0.05 was considered statistically significant. All analyses were performed using SPSS version 25.0 (SPSS Inc., Chicago, IL, USA) and R software (R Foundation for Statistical Computing, Vienna, Austria).

Results

Baseline characteristics and perioperative outcomes are summarized in Tables 1 and 2. The mean age was 65 years, and 16% of patients were female. Almost all patients underwent off-pump CABG, and the aorta non-touch technique was used in approximately half the patients. Variables with an absolute standardized difference of >10 in the original cohort were age, body mass index, serum hemoglobin, diabetes, history of myocardial infarction, use of gastroepiploic artery, red blood cell transfusion, and the follow-up period. After IP of treatment weighting, all clinical variables were well balanced (Figure 1).

Table 1. Baseline Characteristics

	Original cohortA				After IP weightingB
	CABG (n=140)	CABG+LAAC (n=102)	P value	ASD	CABG (n=264.4)	CABG+LAAC (n=236.5)	ASD
Age (years)	64±11	64±10	0.16	0.01	64±11	64±10	0.01
<50	17 (12.1)	13 (12.7)		0.01	12.3	12.4	0.03
50–59	18 (12.9)	14 (13.7)		0.03	13.3	13.5	0.01
60–69	58 (41.4)	39 (38.2)		0.17	39.4	38.9	0.01
70–79	41 (29.3)	33 (32.4)		0.19	29.9	31.5	0.04
>80	6 (4.3)	3 (2.9)		0.13	3.9	3.2	0.04
Female sex	21 (15.0)	18 (17.6)	0.58	0.07	16.3	17.0	0.02
BMI (kg/m2)			0.21
<18.5	12 (8.6)	5 (4.9)		0.15	6.5	5.6	0.04
18.5–29.9	124 (88.6)	91 (89.2)		0.02	88.7	88.9	0.01
>30.0	2.9	6 (5.9)		0.15	3.9	4.6	0.04
Serum hemoglobin (g/dL)	11.7±2.0	11.9±1.8	0.20	0.11	11.7±1.9	11.8±1.8	0.05
Medical history
Smoking history	85 (60.7)	58 (56.9)	0.55	0.08	59.1	58.1	0.02
Hypertension	118 (84.3)	84 (82.4)	0.69	0.05	83.2	82.6	0.02
Diabetes	97 (69.3)	76 (74.5)	0.37	0.12	71.6	72.3	0.02
HbA1c (%)	6.1±1.2	6.0±1.1	0.29	0.09	6.0±1.1	6.0±1.1	0.01
Insulin-dependent	46 (32.9)	35 (34.3)	0.81	0.03	32.3	33.3	0.02
Malignancy	18 (12.9)	16 (15.7)	0.53	0.08	13.8	14.5	0.02
History of embolic stroke	30 (21.4)	24 (23.5)	0.70	0.05	22.5	22.9	0.01
Cardiac status
History of MI	77 (55.0)	49 (48.0)	0.28	0.14	51.0	50.4	0.05
History of PCI	48 (34.3)	37 (36.3)	0.75	0.04	34.6	34.9	0.02
CAD			0.13
1-vessel disease	7 (5.0)	5 (4.9)		0.01	5.0	4.9	0.01
2-vessel disease	19 (13.6)	13 (12.7)		0.02	13.1	12.9	0.01
3-vessel disease	65 (46.4)	51 (50.0)		0.07	47.9	48.8	0.02
Left main disease	49 (35.0)	33 (32.4)		0.06	34.0	33.5	0.01
LVEF <40%	38 (27.1)	25 (24.5)	0.65	0.06	26.4	25.7	0.02
EuroSCORE	2.8 [1.9–5.0]	2.7 [2.0–3.6]	0.29	0.09	2.8 [1.8–4.9]	2.7 [1.9–4.1]	0.08
Follow-up duration (years)			0.01
<1	19 (13.6)	16 (15.7)		0.06	15.6	13.4	0.06
1–3	26 (18.6)	25 (24.5)		0.15	19.9	21.7	0.04
3–5	25 (17.9)	33 (32.4)		0.34	23.1	25.2	0.05
>5	70 (50.3)	28 (27.5)		0.48	41.4	39.7	0.04

^AUnless indicated otherwise, data are presented as the mean±SD, median [interquartile range], or n (%). ^BUnless indicated otherwise, data are presented as the mean±SD, median [interquartile range], or percentages. ASD, absolute standardized difference; BMI, body mass index; CABG, coronary artery bypass grafting; CAD, coronary artery disease; EuroSCORE, European System for Cardiac Operative Risk Evaluation; IP, inverse probability; LAAC, left atrial appendage closure; LVEF, left ventricle ejection fraction; MI, myocardial infarction; PCI, percutaneous coronary intervention.

Table 2. Perioperative Outcomes

	Original cohortA				After IP weightingB
	CABG (n=140)	CABG+LAAC (n=102)	P value	ASD	CABG (n=264.4)	CABG+LAAC (n=236.5)	ASD
Emergency procedure	15 (10.7)	9 (8.8)	0.63	0.06	10.2	9.6	0.02
Off-pump surgery	137 (97.9)	99 (97.1)	0.69	0.05	97.7	97.4	0.02
IABP support	7 (5.0)	5 (4.9)	0.97	0.01	4.9	4.9	0.01
Operation time (min)	282±80	274±75	0.02	0.10	281±78	278±77	0.04
No. distal anastomoses			0.68
1 or 2	49 (35.0)	35 (34.3)		0.01	34.8	34.5	0.02
3 or 4	65 (46.4)	52 (51.0)		0.09	48.7	49.2	0.01
≥5	26 (18.6)	15 (14.7)		0.10	16.7	15.9	0.02
Internal thoracic artery			0.80
Bilateral	75 (53.6)	53 (52.0)		0.03	53.0	52.2	0.02
Single	63 (45.0)	49 (48.0)		0.06	46.9	47.6	0.01
Gastroepiploic artery	32 (22.9)	17 (16.7)	0.24	0.16	20.2	18.2	0.05
Saphenous vein	74 (52.9)	55 (53.9)	0.87	0.02	53.0	53.3	0.01
Sequential grafting	70 (50.0)	51 (50.0)	0.99	0.03	49.8	50.3	0.01
Aorta non-touch	65 (46.4)	45 (44.1)	0.72	0.05	46.2	44.9	0.03
Partial clamping	64 (45.7)	48 (47.1)	0.92	0.03	46.6	46.9	0.01
Sealing device	11 (7.9)	9 (8.8)	0.93	0.03	7.1	6.8	0.01
LAAC technique
Resection	0	85 (83.3)			0	83.0
Ligation	0	17 (16.7)			0	17.0
RBC transfusion	77 (55.0)	63 (61.8)	0.29	0.14	57.6	59.1	0.03
Mediastinitis	1 (0.7)	1 (1.0)	0.82	0.03	0.7	0.9	0.02
ICU stay (days)	2 [1–3]	2 [2–3]	0.82	0.06	2 [1–3]	2 [1–3]	0.05
Hospital stay (days)	11 [8–4]	12 [10–17]	0.30	0.09	11 [8–15]	12 [9–16]	0.04

^AUnless indicated otherwise, data are presented as the mean±SD, median [interquartile range], or n (%). ^BUnless indicated otherwise, data are presented as the mean±SD, median [interquartile range], or percentages. ASD, absolute standardized difference; IABP, intra-aortic balloon pumping; ICU, intensive care unit. Other abbreviations as in Table 1.

Figure 1.

Absolute standardized differences in baseline variables before and after the inverse probability of treatment weighting. BMI, body mass index; EuroSCORE, European System for Cardiac Operative Risk Evaluation; IABP, intra-aortic balloon pumping; IP, inverse probability; LV, left ventricle; PCI, percutaneous coronary intervention; RBC, red blood cell.

Primary Outcome

In the IP-weighted cohort, the primary composite of death or stroke occurred in 17.8% of patients with LAA closure, compared with 37.5% of patients without LAA closure (Table 3). The mean duration of follow-up in the IP-weighted cohort was 3.9 and 3.6 years in patients without and with concomitant LAA closure, respectively (P=0.452). The 5-year estimated stroke-free survival was significantly higher in patients with than without LAA closure (log-rank test, P=0.035; Figure 2). LAA closure was significantly associated with a lower risk of death or stroke (hazard ratio [HR] 0.43; 95% CI 0.20–0.92; P=0.023). This relationship was driven by risk reduction of both death (HR 0.48; 95% CI 0.24–0.95; P=0.045) and stroke (sub-HR 0.26; 95% CI 0.08–0.84; P=0.028).

Table 3. Study Outcomes

	Original cohort		IP-weighted cohort		Adjusted RR (95% CI)	P value
	CABG (n=140)	CABG+LAAC (n=102)	CABG (n=264.4)	CABG+LAAC (n=236.5)
Primary outcomes
Stroke and death	52 (37.1)	23 (22.5)	99.2 (37.5)	42 (17.8)	0.43 (0.20–0.92)A	0.023
Stroke	10 (7.1)	2 (2.0)	18.9 (7.1)	4.0 (1.7)	0.26 (0.08–0.84)B	0.028
Death	42 (30.0)	21 (20.6)	80.3 (30.4)	38.0 (16.1)	0.48 (0.24–0.95)A	0.045
Secondary outcomes
Admission for HF	6 (4.3)	4 (3.9)	12.1 (4.6)	9.8 (4.1)	0.98 (0.22–2.91)B	0.872
Cardiac-related death	18 (12.9)	11 (10.8)	35.6 (13.5)	28.8 (12.2)	0.91 (0.15–5.52)B	0.812
Operative outcomes
Reoperation for bleeding	1 (0.7)	2 (2.0)	2.3 (0.9)	3.1 (1.3)	1.52 (0.46–5.03)C	0.809
New-onset AF	38 (27.1)	32 (31.4)	69.7 (26.4)	58.6 (24.8)	0.95 (0.29–3.12)C	0.638
Death within 30 days	4 (2.9)	2 (2.0)	7.0 (2.6)	4.5 (1.9)	0.78 (0.16–3.81)C	0.589

^AHazard ratio (95% confidence interval [CI]). ^BSubhazard ratio (95% CI). ^COdds ratio (95% CI). Relative risks (RR) and 95% CIs were calculated based on the IP-weighted cohort. AF, atrial fibrillation; HF, heart failure. Other abbreviations as in Table 1.

Figure 2.

Stroke-free survival in patients with and without concomitant left atrial appendage (LAA) closure in the original cohort (Left) and in the inverse probability (IP)-weighted cohort (Right). Dashed lines indicate the 95% confidence intervals. CABG, coronary artery bypass grafting.

Secondary Outcome

There was no significant difference in the cumulative incidence of admission for heart failure and cardiac-related death (Table 3, Figure 3). No significant differences were observed in the incidence and adjusted odds ratios for reoperation for bleeding, AF, and death within 30 days between the 2 groups. These relationships did not change after IP of treatment weighting. The causes of death are summarized in Table 4. Of note, there were 4 bleeding events in patients without LAA closure, including intracerebral hemorrhage and intestinal bleeding.

Figure 3.

Secondary outcomes in the inverse probability weighted cohort. Dashed lines indicate 95% confidence intervals. CABG, coronary artery bypass grafting; LAA, left atrial appendage.

Table 4. Causes of Death

	CABG (n=140)	CABG+LAAC (n=102)	P value
Myocardial infarction	7 (5.0)	4 (3.9)	0.69
Heart failure	5 (3.6)	3 (2.9)	0.79
Sudden death	7 (5.0)	5 (4.9)	0.97
Intracerebral hemorrhage	2 (1.4)	0	0.23
Ischemic stroke	3 (2.1)	0	0.14
Aortic dissection	1 (0.7)	0	0.39
Intestinal ischemia	2 (1.4)	1 (1.0)	0.76
Intestinal bleeding	2 (1.4)	0	0.23
Malignancy	4 (2.9)	3 (2.9)	0.97
Pneumonia	9 (6.4)	5 (4.9)	0.62

Data are presented as n (%). CABG, coronary artery bypass grafting; LAAC, left atrial appendage closure.

Discussion

The main finding of the present study was that concomitant LAA closure during CABG was significantly associated with a lower risk of death or stroke without increasing the risk of reoperation for bleeding, 30-day mortality, postoperative AF, or readmission for heart failure in ESRD-HD patients with normal sinus rhythm.

The reason why LAA closure was effective in preventing stroke in ESRD-HD patients may be related to the high prevalence of AF and the high risk of stroke caused by AF. In a recent study of 66 ESRD-HD patients with continuous cardiac rhythm monitoring during a 6-month follow-up, AF occurred in 41% of patients, mostly during and just after the dialysis procedure.²⁶ A sympathetic response to peridialytic fluid and electrolyte flux during dialysis were considered to stimulate AF. Another study using an implantable cardiac monitor in 50 ESRD-HD patients reported that most arrhythmia events identified over 18 months were AF and most likely to occur during the pre-hemodialysis period.²⁷ Thus, ESRD-HD patients frequently develop AF associated with dialysis procedures and, importantly, most are asymptomatic and unidentifiable without continuous cardiac rhythm monitoring. Short-term AF (≥6 min) has been shown to increase the risk of stroke in patients with a CHA₂DS₂-VASc score of >3.²⁸ Approximately 60% of ESRD-HD patients reportedly have a CHA₂DS₂-VASc score of ≥4.²⁹ Such patients are likely to benefit from anticoagulant therapy, but short-term AF is often overlooked because it is subclinical, and patients are rarely monitored by electrocardiography, even during dialysis procedures. We believe this is one reason why LAA closure benefits ESRD-HD patients and should be considered an additional procedure for ESRD-HD patients during CABG.

LAA closure potentially elevates left atrial pressure due to the loss of function as a decompression chamber. It has been reported to increase the risk of postoperative AF,¹⁷ which is inconsistent with our results. Various factors determine left atrial pressure. For example, LAA shape and volume vary significantly among individuals. Moreover, ESRD-HD patients depend on ultrafiltration for postoperative volume control. We believe that the impact of LAA closure on left atrial pressure and AF occurrence is not always uniform. The development of postoperative AF is multifactorial, and assessing the actual effect of a single factor is difficult.

Atrial natriuretic peptide (ANP) released from the LAA increases urine output in response to fluid overload.^30,31 This function could be potentially protective against heart failure, and it has been hypothesized that LAA removal may impair renal clearance of salt and water, increasing the risk of heart failure.^30,31 However, as noted above, ESRD-HD patients completely depend on ultrafiltration for fluid management. Therefore, the drawback of decreased ANP secretion by LAA closure may have been offset without clinical manifestation.

The present study has several limitations. A retrospective single-center study may have increased risks of selection bias and center-specific bias, and limited generalizability. Our surgical series of 242 patients may not provide adequate power for robust statistical comparisons. The difference in the timing of surgery between the 2 groups may have led to changes in clinical routine, quality of surgery, and perioperative management between the comparison groups. More than 90% of patients were followed up after discharge, but the clinical course and the cause of death could not be fully recorded in some patients, potentially affecting the analysis results. The lack of information about anticoagulant therapy and the incidence of AF during the follow-up did not allow us to determine the relative efficacy of LAA closure compared to oral anticoagulation. The follow-up was carried out in an unblinded manner.

In conclusion, concomitant LAA closure during CABG can reduce the risk of death and stroke in ESRD-HD patients with normal sinus rhythm.

Sources of Funding

This work was supported by a Grant-in-Aid for Young Scientists (19K17615) from the Japan Society for the Promotion of Science.

Disclosures

None declared.

IRB Information

This study was approved by the Institutional Review Board of Juntendo University (Reference no. R2018-185; February 28, 2019). Given the retrospective nature of this work, the need for written informed consent was waived.

Data Availability

The deidentified participant data will not be not be shared.

Supplementary Files

Please find supplementary file(s);

https://doi.org/10.1253/circj.CJ-22-0573

References

• 1.   Go AS, Chertow GM, Fan D, McCulloch CE, Hsu C. Chronic kidney disease and the risks of death, cardiovascular events, and hospitalization. N Engl J Med 2004; 351: 1296–1305.
• 2.   Seliger SL, Gillen DL, Longstreth WT, Kestenbaum B, Stehman-Breen CO. Elevated risk of stroke among patients with end-stage renal disease. Kidney Int 2003; 64: 603–609.
• 3.   Genovesi S, Pogliani D, Faini A, Valsecchi MG, Riva A, Stefani F, et al. Prevalence of atrial fibrillation and associated factors in a population of long-term hemodialysis patients. Am J Kidney Dis 2005; 46: 897–902.
• 4.   Go AS, Fang MC, Udaltsova N, Chang Y, Pomernacki NK, Borowsky L, et al. Impact of proteinuria and glomerular filtration rate on risk of thromboembolism in atrial fibrillation: The anticoagulation and risk factors in atrial fibrillation (ATRIA) study. Circulation 2009; 119: 1363–1369.
• 5.   Zimmerman D, Sood MM, Rigatto C, Holden RM, Hiremath S, Clase CM. Systematic review and meta-analysis of incidence, prevalence and outcomes of atrial fibrillation in patients on dialysis. Nephrol Dial Transplant 2012; 27: 3816–3822.
• 6.   Genovesi S, Vincenti A, Rossi E, Pogliani D, Acquistapace I, Stella A, et al. Atrial fibrillation and morbidity and mortality in a cohort of long-term hemodialysis patients. Am J Kidney Dis 2008; 51: 255–262.
• 7.   Stroke prevention in atrial fibrillation study: Final results. Circulation 1991; 84: 527–539.
• 8.   Blackshear JL, Odell JA. Appendage obliteration to reduce stroke in cardiac surgical patients with atrial fibrillation. Ann Thorac Surg 1996; 61: 755–759.
• 9.   Winkelmayer WC. Cardiovascular disease: Still unresolved: Warfarin in ESRD with atrial fibrillation. Nat Rev Nephrol 2014; 10: 244–245.
• 10.   Chan KE, Lazarus JM, Thadhani R, Hakim RM. Warfarin use associates with increased risk for stroke in hemodialysis patients with atrial fibrillation. J Am Soc Nephrol 2009; 20: 2223–2233.
• 11.   Olesen JB, Lip GYH, Kamper AL, Hommel K, Køber L, Lane DA, et al. Stroke and bleeding in atrial fibrillation with chronic kidney disease. N Engl J Med 2012; 367: 625–635.
• 12.   Carrero JJ, Evans M, Szummer K, Spaak J, Lindhagen L, Edfors R, et al. Warfarin, kidney dysfunction, and outcomes following acute myocardial infarction in patients with atrial fibrillation. JAMA 2014; 311: 919–928.
• 13.   Shen JI, Montez-Rath ME, Lenihan CR, Turakhia MP, Chang TI, Winkelmayer WC. Outcomes after warfarin initiation in a cohort of hemodialysis patients with newly diagnosed atrial fibrillation. Am J Kidney Dis 2015; 66: 677–688.
• 14.   Shah M, Avgil Tsadok M, Jackevicius CA, Essebag V, Eisenberg MJ, Rahme E, et al. Warfarin use and the risk for stroke and bleeding in patients with atrial fibrillation undergoing dialysis. Circulation 2014; 129: 1196–1203.
• 15.   Yoon CY, Noh J, Jhee JH, Chang TI, Kang EW, Kee YK, et al. Warfarin use in patients with atrial fibrillation undergoing hemodialysis. Stroke 2017; 48: 2472–2479.
• 16.   Whitlock RP, Belley-Cote EP, Paparella D, Healey JS, Brady K, Sharma M, et al. Left atrial appendage occlusion during cardiac surgery to prevent stroke. N Engl J Med 2021; 384: 2081–2091.
• 17.   Mahmood E, Matyal R, Mahmood F, Xu X, Sharkey A, Chaudhary O, et al. Impact of left atrial appendage exclusion on short-term outcomes in isolated coronary artery bypass graft surgery. Circulation 2020; 142: 20–28.
• 18.   Hirose H, Amano A, Takanashi S, Takahashi A. Skeletonized bilateral internal mammary artery grafting for patients with diabetes. Interact Cardiovasc Thorac Surg 2003; 2: 287–292.
• 19.   Hirose H, Amano A, Takahashi A. Aortic nontouch off-pump complete revascularization using 3 in situ arterial conduits: Bilateral internal mammary arteries and gastroepiploic artery. Heart Surg Forum 2004; 7: E122–E125.
• 20.   Kato TS, Iwamura T, Endo D, Yokoyama Y, Oishi A, Shimada A, et al. Left atrial appendage closure reduces the incidence of postoperative cerebrovascular accident in patients undergoing cardiac surgery. Circ J 2015; 79: 2591–2597.
• 21.   Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med 2015; 34: 3661–3679.
• 22.   Rosenbaum PR, Rubin DB. The central role of propensity score in observational studies for causal effects. Biometrika 1983; 70: 41–55.
• 23.   Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med 2009; 28: 3083–3107.
• 24.   Gray R. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann Stat 1988; 16: 1141–1154.
• 25.   Fine JP, Gray RJ. A proportional hazard model for the subdistribution of a competing risk. J Am Stat Assoc 1999; 94: 496–509.
• 26.   Roy-Chaudhury P, Tumlin JA, Koplan BA, Costea AI, Kher V, Williamson D, et al. Primary outcomes of the Monitoring in Dialysis Study indicate that clinically significant arrhythmias are common in hemodialysis patients and related to dialytic cycle. Kidney Int 2018; 93: 941–951.
• 27.   Wong MC, Kalman JM, Pedagogos E, Toussaint N, Vohra JK, Sparks PB, et al. Temporal distribution of arrhythmic events in chronic kidney disease: Highest incidence in the long interdialytic period. Heart Rhythm 2015; 12: 2047–2055.
• 28.   Kaplan RM, Koehler J, Ziegler PD, Sarkar S, Zweibel S, Passman RS. Stroke risk as a function of atrial fibrillation duration and CHA₂DS₂-VASc score. Circulation 2019; 140: 1639–1646.
• 29.   Schamroth Pravda M, Cohen Hagai K, Topaz G, Schamroth Pravda N, Makhoul N, Shuvy M, et al. Assessment of the CHA₂DS₂-VASc score in predicting mortality and adverse cardiovascular outcomes of patients on hemodialysis. Am J Nephrol 2020; 51: 635–640.
• 30.   Wilber DJ. Neurohormonal regulation and the left atrial appendage: Still more to learn. J Am Coll Cardiol 2018; 71: 145–147.
• 31.   Lakkireddy D, Turagam M, Afzal MR, Rajasingh J, Atkins D, Dawn B, et al. Left atrial appendage closure and systemic homeostasis: The LAA HOMEOSTASIS study. J Am Coll Cardiol 2018; 71: 135–144.