Abstract
Objective:
To determine trends in real-world utilization and in-hospital adverse events from Watchman implantation since its approval by the Food and Drug Administration in 2015.
Background:
The risk of embolic stroke caused by atrial fibrillation is reduced by oral anticoagulants, but not all patients can tolerate long-term anticoagulation. Left atrial appendage occlusion with the Watchman device has emerged as an alternative therapy.
Methods:
This was a retrospective cohort study utilizing data from National In-patient Sample for calendar years 2015–2017. The outcomes assessed in this study were associated complications, in-hospital mortality, and resource utilization trends after Watchman implantation. Trends analysis were performed using analysis of variance. Multivariable adjusted logistic regression analysis was performed to determine predictors of mortality.
Results:
A total of 17 700 patients underwent Watchman implantation during the study period. There was a significantly increased trend in the number of Watchman procedures performed over the study years (from 1195 in 2015 to 11 165 devices in 2017, p < .01). A significant decline in the rate of complications (from 26.4% in 2015% to 7.9% in 2017, p < .01) and inpatient mortality (from 1.3% in 2015% to 0.1% in 2017, p < .01) were noted. Predictors of in-hospital mortality included a higher CHA2DS2-VASc score (odds ratio [OR]: 2.61 per 1-point increase, 95% confidence interval [CI]: 1.91–3.57), chronic blood loss anemia (OR: 3.63, 95% CI: 1.37–9.61) and coagulopathy (OR: 4.90, 95% CI: 2.32–10.35).
Conclusion:
In contemporary United States clinical practice, Watchman utilization has increased significantly since approval in 2015, while complications and in-patient mortality have declined.
Keywords: complications, mortality, national trends, Watchman
1 |. INTRODUCTION
Atrial fibrillation (AF) is the most commonly encountered sustained cardiac arrhythmia in clinical practice and responsible for more than 20% of all embolic strokes.1,2 AF-associated strokes tend to have worse morbidity and mortality when compared to strokes not related to AF.3,4 The left atrial appendage (LAA) is the location for thrombus formation in more than 90% of patients with nonvalvular AF.5 Coumadin and direct oral anticoagulants (DOACs) are the gold-standard therapy for reducing stroke risk in AF patients with risk factors for stroke. However, their utilization is often limited by lack of patient compliance and adverse effects.6–8 Left atrial appendage occlusion (LAAO) using an endocardial Watchman device has shown promising results in mitigating stroke risk when utilized in selected AF patients.9 The landmark PROTECT AF (percutaneous closure of the LAA vs. warfarin therapy for prevention of stroke in patients with atrial fibrillation) trial showed the Watchman device to be noninferior to coumadin in terms of the primary efficacy end-point of stroke, systemic embolism and cardiovascular/unexplained death.10 Subsequently, the PREVAIL (prospective randomized evaluation of the Watchman LAA closure device in patients with atrial fibrillation vs. long-term warfarin therapy) trial confirmed these results and also showed a reduced rate of short-term complications.11 The results of these two trials eventually led to Food and Drug Administration (FDA) approval of the Watchman device in March of 2015. Since FDA approval of the Watchman device, there has been limited real-world data on trends in utilization, complications, and in-hospital mortality from the procedure in contemporary practice.12 The aim of the present study is to assess these parameters from a comprehensive, national United States population database.
2 |. METHODS
2.1 |. Study data
For the purpose of the current analysis, data were derived from the National Inpatient Sample (NIS) for calendar years 2015–2017. The NIS is made possible by a Federal-State-Industry partnership sponsored by the Agency for Healthcare Research and Quality (AHRQ). The NIS is derived from all States for national estimates of healthcare utilization, costs, and outcomes.13 NIS data are compiled annually and therefore the data can be used for analyses of disease trends over time. The NIS approximates 20% of all discharges from all US nonfederal hospitals and provides discharge weights that are used for computation of national estimates. The discharge weights are calculated with in each sampling hospital as the ratio of discharges in the universe (derived from data collected from American Hospital Association survey for non-Health care cost and utilization project [HCUP] hospitals and State Inpatient Databases for HCUP hospitals) to the discharges in the sample hospital. The discharge weight is uniform throughout the sample hospital which implies that estimates of sample means are consistent for both weighted and unweighted encounters. Institutional Review Board approval and informed consents were not required for this study given the deidentified nature of the NIS data set and public availability.
2.2 |. Study population and study design
We analyzed NIS data from January 2015 to December 2017. The study population was selected by using the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) and International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) codes. We selected patients 18 years of age and above for the purpose of our study. Age was further divided into three groups, <65, 65–74, and ≥75. Patients implanted with a WATCHMAN were identified by ICD-9 code of 37.90 and ICD-10 code of 02L73DK. Baseline characteristics and key complications were identified (ICD codes for complications provided in supplement), as previously described.14 Hospital outcomes including inpatient mortality, discharge disposition, length of stay (LOS) and cost of hospitalization (inflation adjusted) were derived. For the computation of hospitalization costs, cost-to-charge ratio files from NIS were utilized.
Complication rate trends were analyzed over the study years. The primary outcome of the study was the prevalence of Watchman implantation over our study period. Additional outcomes analyzed including associated complications, in-hospital mortality and resource utilization (including LOS, cost of stay, and discharge disposition to home, short term care, long-term care, or home with institutional care or home health).
2.3 |. Statistical analysis
Descriptive statistics are presented as frequencies with percentages for categorical variables and as means with standard deviations for continuous variables. Baseline characteristics were compared using a Pearson χ2 test and Fisher's exact test for categorical variables and independent samples t-test for continuous variables. Trends analysis was performed using analysis of variance. Linear regression was used to predict trends over calendar years. Logistic regression was performed to estimate odds ratios (ORs) with 95% confidence intervals (CIs) to determine predictors for mortality. Initially, a binomial logistic regression model was used to identify variables from demographic data (Table 1) that were significantly associated with patient mortality (p < .10). These variables were then subsequently utilized in a multivariable logistic regression model to identify statistically significant predictors of mortality. In the final model, p < .05 was used as cutoff for stepwise forward entry for logistic regression. A type I error rate of <0.05 was considered statistically significant. All statistical analyses were performed using statistical package for social science (SPSS) version 26 (IBM Corp). Discharge weights provided by NIS were used for computation of national estimates. All analyses were done on a weighted sample.
TABLE 1.
Baseline characteristics of the study population of patients undergoing Watchman device implantation, stratified by year of implant from years 2015 to 2017
| Variable no. (%) | 2015 (n = 1195) |
2016 (n = 5340) |
2017 (n = 11165) |
Total (n =17700) |
p Value |
|---|---|---|---|---|---|
| Age, mean (SD), years | 74.2 (8.4) | 75.5 (8.5) | 75.8 (7.9) | 75.6 (8.2) | <.01 |
| Males | 780 (65.30) | 3140 (58.80) | 6685 (59.90) | 10 605 (59.9) | <.01 |
| Females | 415 (34.70) | 2200 (41.20) | 4480 (40.10) | 7095 (40.10) | |
| Age | |||||
| <65 | 155 (13.0) | 445 (8.3) | 830 (7.4) | 1430 (8.1) | <.01 |
| 65–74 | 345 (28.9) | 1770 (33.1) | 3665 (32.8) | 5780 (32.7) | |
| ≥75 | 695 (58.2) | 3125 (58.5) | 6670 (59.7) | 10 490 (59.3) | |
| Elective admission | 960 (81.7) | 4625 (86.9) | 10170 (91.5) | 15 755 (89.5) | <.01 |
| Race and ethnicity | |||||
| Caucasian | 985 (86.0) | 4405 (86.4) | 9260 (85.9) | 14 650 (86.0) | <.01 |
| African American | 45 (3.9) | 175 (3.4) | 475 (4.4) | 695 (4.1) | |
| Hispanics | 60 (5.2) | 215 (4.2) | 655 (6.1) | 930 (5.5) | |
| Asian or Pacific Islander | 30 (2.6) | 130 (2.5) | 190 (1.8) | 350 (2.1) | |
| Native American | <10a | 45 (0.9) | 30 (0.3) | 85 (0.5) | |
| CHA2DS2-VASc score | |||||
| 0 | 15 (1.3) | 30 (0.6) | 35 (0.3) | 80 (0.5) | |
| 1 | 50 (4.2) | 165 (3.1) | 425 (3.8) | 640 (3.6) | <.01 |
| 2 | 210 (17.6) | 860 (16.1) | 1705 (15.3) | 2775 (15.7) | |
| 3 | 335 (28.0) | 1640 (30.7) | 3805 (34.1) | 5780 (32.7) | |
| 4 | 340 (28.5) | 1565 (29.3) | 3405 (30.5) | 5310 (30.0) | |
| 5 | 190 (15.9) | 760 (14.2) | 1290 (11.6) | 2240 (12.7) | |
| 6 | 40 (3.3) | 230 (4.3) | 385 (3.4) | 655 (3.7) | |
| 7–9 | 15 (1.2) | 90 (1.7) | 115 (1.0) | 220 (1.2) | |
| Mean CHA2DS2-VASc Score (mean [SD]) | 3.5 (1.3) | 3.5 (1.3) | 3.4 (1.2) | 3.5 (1.2) | <.01 |
| Anemia (blood loss) | 20 (1.7) | 100 (1.9) | 195 (1.7) | 315 (1.8) | .84 |
| Iron deficiency anemia | 170 (14.2) | 670 (12.5) | 1610 (14.4) | 2450 (13.8) | <.01 |
| Congestive heart failure | 410 (34.3) | 1660 (31.1) | 3670 (33) | 5740 (32.4) | <.01 |
| Chronic pulmonary disease | 200 (16.7) | 1165 (21.8) | 2390 (21.4) | 3755 (21.2) | <.01 |
| Coagulopathy | 150 (12.6) | 200 (3.7) | 510 (4.6) | 860 (4.9) | <.01 |
| Coronary artery disease | 55 (4.6) | 405 (7.6) | 930 (8.3) | 1390 (7.9) | <.01 |
| Diabetes | 320 (26.8) | 1240 (23.2) | 2215 (19.8) | 3775 (21.3) | <.01 |
| Diabetes with complications | 55 (4.6) | 405 (7.6) | 1720 (15.4) | 2180 (12.3) | <.01 |
| Hypertension | 910 (76.2) | 4040 (75.7) | 6330 (56.7) | 11 280 (63.7) | <.01 |
| Liver disease | <10a | 135 (2.5) | 290 (2.6) | 435 (2.5) | <.01 |
| Obesity | 195 (16.3) | 620 (11.6) | 1890 (16.9) | 2705 (15.3) | <.01 |
| Paralysis | 20 (1.7) | 195 (3.7) | 350 (3.1) | 565 (3.2) | <.01 |
| Peripheral vascular disorders | 170 (14.2) | 555 (10.4) | 1245 (11.2) | 1970 (11.1) | <.01 |
| Renal failure | 240 (20.1) | 1030 (19.3) | 2540 (22.7) | 3810 (21.5) | <.01 |
| Valvular disease | <10a | 20 (0.4) | 15 (0.1) | 45 (0.3) | <.01 |
| Hospital Location | |||||
| Rural | 15 (1.3) | 40 (0.7) | 160 (1.4) | 215 (1.2) | <.01 |
| Urban nonteaching | 150 (12.6) | 575 (10.8) | 1045 (9.4) | 1770 (10.0) | |
| Urban Teaching | 1030 (86.2) | 4725 (88.5) | 9960 (89.2) | 15 715 (88.8) | |
| Bed size of the hospital | |||||
| Small | 165 (13.8) | 575 (10.8) | 1175 (10.5) | 1915 (10.8) | <.01 |
| Medium | 255 (21.3) | 1070 (20.0) | 2565 (23.0) | 3890 (22.0) | |
| Large | 775 (64.9) | 3695 (69.2) | 7425 (66.5) | 11 895 (67.2) | |
| Census divisions | |||||
| New England | <10a | 180 (3.4) | 350 (3.1) | 540 (3.1) | <.01 |
| Mid-Atlantic | 155 (13.0) | 720 (13.5) | 1410 (12.6) | 2285 (12.9) | |
| East North Central | 195 (16.3) | 715 (13.4) | 1530 (13.7) | 2440 (13.8) | |
| West North Central | 70 (5.9) | 330 (6.2) | 785 (7.0) | 1185 (6.7) | |
| South Atlantic | 235 (19.7) | 1040 (19.5) | 2565 (23.0) | 3840 (21.7) | |
| East South Central | 20 (1.7) | 200 (3.7) | 600 (5.4) | 820 (4.6) | |
| West South Central | 140 (11.7) | 620 (11.6) | 1335 (12.0) | 2095 (11.8) | |
| Mountain | 85 (7.1) | 440 (8.2) | 1175 (10.5) | 1700 (9.6) | |
| Pacific | 285 (23.8) | 1095 (20.5) | 1415 (12.7) | 2795 (15.8) | |
| Payee | |||||
| Medicare/Medicaid | 1025 (85.8) | 4915 (92.2) | 10 015 (90.1) | 15 955 (90.4) | <.01 |
| Private insurance | 150 (12.6) | 365 (6.8) | 920 (8.3) | 1435 (8.1) | |
| Self-pay | 0 | <10a | 60(0.5) | 65 (0.4) | |
| Median income | |||||
| 0–25th | 280 (24) | 930 (17.7) | 2320 (21.1) | 3530 (20.3) | <.01 |
| 26–50th | 210 (18) | 1235 (23.5) | 2815 (25.6) | 4260 (24.5) | |
| 51–75th | 335 (28.8) | 1460 (27.8) | 3175 (28.9) | 4970 (28.5) | |
| 76–100th | 340 (29.2) | 1630 (31.) | 2680 (24.4) | 4650 (26.7) | |
Less than 10 data were not reported as per HCUP recommendations.
3 |. RESULTS
A total of 17 700 patients underwent Watchman implantation from January 2015 to December 2017. The mean age of patients implanted was 75.6 (SD ± 8.2) years. The mean age increased over the study years (74.2 years in year 2015 vs. 75.8 years in year 2017, p < .01). Overall, women constituted 40.1% (n = 7095) of the study cohort, and the majority of patients 86.0% (n = 14 650) were White. Baseline characteristics of the study population are shown in Table 1. Between 2015 and 2017, patients undergoing LAAO in later calendar years were on average older, more commonly female, and more commonly electively admitted for the procedure.
Overall, there was a significant increase in the number of Watchman LAAO procedures in the United States (from 1195 device implants in year 2015 to 11 165 device implants in year 2017, p < .01, Figure 1). Peri-procedural complications associated with Watchman implantation are depicted in Table 2. There was a significant decrease in the rate of complications over the study period (26.4% in year 2015 vs. 7.9% in year 2017, p < .01). The largest decrease in complications over the study years occurred with cardiovascular and neurological complications, with a decreased rate of any cardiovascular complication from 13.8% in year 2015% to 4.7% in year 2017 (p < .01), while the rate of any neurological complication decreased from 7.9% in year 2015 to just 0.9% in year 2017 (p < .01). There were very low rates of device related thrombus at discharge or device embolization during the study period (n < 10 patients; <0.1%). Overall, in-hospital mortality was low at 0.3% (n = 45 patients), and mortality decreased each year from 1.3% in 2015% to 0.1% in 2017 (Figure 2).
FIGURE 1.
Number of Watchman procedures over the study years
TABLE 2.
Complications in patients undergoing Watchman implantation
| Variable no. (%) | 2015 (n = 1195) |
2016 (n = 5340) |
2017 (n = 11165) |
Combined (n = 17700) |
p value |
|---|---|---|---|---|---|
| Overall complications (%) | 315 (26.4) | 440 (8.2) | 885 (7.9) | 1640 (9.3) | <.01 |
| Cardiovascular system | |||||
| Any cardiovascular event/complication | 165 (13.8) | 210 (3.9) | 450 (4.0) | 825 (4.7) | <.01 |
| Percutaneous coronary intervention | <10 (<1.0)a | 20(0.4) | <10 (<0.1) | 35 (0.2) | <.01 |
| Pericardial effusion/Hemopericardium | 65 (5.4) | 160 (3.0) | 325 (2.9) | 550 (3.1) | <.01 |
| Cardiac Arrest/CPR procedure code | <10 (<1.0)a | <10 (<0.2)a | <10 (<0.1) | 20 (0.1)a | <.01 |
| STEMI | <10 (<1.0)a | <10 (<0.2)a | <10 (<0.1)a | 25 (0.1)a | <.01 |
| NSTEMI or type II MI | 50 (4.2) | <10 (<0.2)a | <10 (<0.1)a | 65 (0.4) | <.01 |
| Cardiac Tamponade | <10 (<1.0) | 30 (0.6) | 110 (1.0) | 150 (0.8) | .02 |
| Pericarditis | <10 (<1.0)a | <10 (<0.2)a | 45 (0.4) | 50 (0.3) | <.01 |
| Need for pericardiocentesis | 20 (1.7) | 40 (0.7) | 145 (1.3) | 205 (1.2) | <.01 |
| Cardiogenic Shock | 30 (2.5) | 15 (0.3) | 20 (0.2) | 65 (0.4) | <.01 |
| Systemic | |||||
| Any systemic complication | <10 (<1.0)a | 35 (0.7) | 45 (0.4) | 90 (0.5) | <.01 |
| Anaphylaxis | 0 | 0 | <10 (<0.1) | <10 (<0.1) | .23 |
| Arterial thrombosis | 0 | 15(0.3) | <10 (<0.1) | 25 (0.1) | <.01 |
| Deep venous thrombosis | <10 (<1.0)a | 20 (0.4) | 20 (0.2) | 50 (0.3) | <.01 |
| Septic shock | 0 | 0 | <10 (<0.1) | <10 (<0.1) | .05 |
| Vascular complications | |||||
| Any peripheral vascular complication | <10 (<1.0) | 35 (0.7) | 65 (0.6) | 105 (0.6) | <.01 |
| AV fistula | 0 | <10 (<0.2)a | 15 (0.1) | 25 (0.1) | .28 |
| Pseudoaneurysm | 5 (0.4) | 20 (0.4) | 35 (0.3) | 60 (0.3) | .73 |
| Retroperitoneal bleeding | 0 | 15 (0.3) | 15 (0.1) | 30 (0.2) | .03 |
| Neurological complications | |||||
| Any neurological complication | 95 (7.9) | 55 (1.0) | 100 (0.9) | 250 (1.4) | <.01 |
| Hemorrhagic stroke | <10 (<1.0)a | 15 (0.3) | 50 (0.4) | 70 (0.4) | .28 |
| Ischemic stroke/TIA | 90 (7.5) | 40 (0.7) | 50 (0.4) | 180 (1.0) | <.01 |
| Bleeding and Hematological complications | |||||
| Any bleeding or hematological complication | 155 (13.0) | 160 (3) | 320 (2.9) | 620 (3.6) | <.01 |
| GI bleeding | 20 (1.7) | 145 (2.7) | 300 (2.7) | 465 (2.6) | .1 |
| Hemothorax after procedure | <10 (<1.0)a | <10 (<0.2)a | <10 (<0.1)a | 15 (0.1) | <.01 |
| Need for blood transfusion | 130 (10.9) | 70 (1.3) | 205 (1.8) | 405 (2.3) | <.01 |
| Pulmonary complications | |||||
| Any pulmonary complications | 70 (5.8%) | 45 (0.8%) | 95 (0.8) | 195 (1.1) | <.01 |
| Pneumothorax | 25 (2.1) | <10 (<0.2)a | <10 (<0.1)a | <10 (<0.1)a | <.01 |
| Pleural effusion | 25 (2.1) | <10 (<0.2)a | 55 (0.5) | 90 (0.5) | <.01 |
| Pneumonia bacterial | 25 (2.1) | 20 (0.4) | 40 (0.4) | 85 (0.5) | <.01 |
| Pulmonary embolism | 0 | 15 (0.3) | 0 | 15 (0.1) | <.01 |
| Device related complications | |||||
| Device thrombus | 0 | 0 | 0 | 0 | |
| Device Embolization | 0 | 0 | <10 (<0.1)a | <10 (<0.1)a | <.01 |
Abbreviations: NSTEMI, non-ST segment elevation myocardial infarction; STEMI, ST-segment elevation myocardial infarction; TIA, transient ischemic attack.
Less than 10 data were not reported as per HCUP recommendations.
FIGURE 2.
Trends in mortality after Watchman implant from years 2015 to 2017
Multivariable adjusted predictors of mortality for patients undergoing Watchman implantation are shown in Figure 5. A higher CHA2DS2-VASc score (OR: 2.61 for each 1 point increase in score [95% CI: 1.91–3.57], p < .01), chronic blood loss anemia (OR: 3.63 [95% CI: 1.37–9.61], p < .01) and coagulopathy (OR: 4.90 [95% CI: 2.32–10.35], p < .01) were associated with a higher mortality, whereas a more recent calendar year of implant (OR: 0.28 per year increase [95% CI: 0.19–0.43], p < .01) and history of hypertension (OR: 0.28 [95% CI: 0.14–0.56], p < .01) were associated with a lower odds of mortality at discharge.
FIGURE 5.
Predictors of in-hospital mortality associated with Watchman implantation from years 2015 to 2017
As seen in Table 3, the majority of patients were discharged home or with home healthcare. Both LOS and cost of hospitalization demonstrated a declining trend over the study period (Table 3 and Figures 3 and 4).
TABLE 3.
Hospital outcomes and resource utilization
| Variables no. (%) | 2015 | 2016 | 2017 | total | p value |
|---|---|---|---|---|---|
| Died at discharge | 15 (1.3) | 15 (0.3) | 15 (0.1) | 45 (0.3) | <.01 |
| Discharge disposition | |||||
| Routine/self-care | 810 (67.8) | 4915 (92.0) | 10 235 (91.7) | 15 960 (90.2) | <.01 |
| Short-term hospital | <10a | 0 | <10a | 20 (0.1) | |
| Another type of facility | 150 (12.6) | 155 (2.9) | 285 (2.6) | 590 (3.3) | |
| Home Health Care | 210 (17.6) | 250 (4.7) | 615 (5.5) | 1075 (6.1) | |
| Resource utilization, mean (SD) | |||||
| Length of stay, days | 4.2 (5.6) | 1.5 (2.0) | 1.5 (1.9) | 1.7 (2.5) | <.01 |
| Cost of hospitalization, $ | 150 370 (131 324) | 117 568 (71 736) | 115 803 (65 180) | 118 676 (73 971) | <.01 |
Note: Supplementary 1: ICD-9 and ICD-10 codes used for the study cohort and variables entered in logistic regression model.
Less than 10 data were not reported as per HCUP recommendations.
FIGURE 3.
Mean length of stay during Watchman implantation admissions from years 2015 to 2017
FIGURE 4.
Adjusted mean cost of stay during Watchman implantation admissions from years 2015 to 2017
4 |. DISCUSSION
The main findings of our current investigation are1: Over the study period from 2015 to 2017, there has been a significant increase in the number of Watchman device implantation procedures in the United States (from 1195 in year 2015 to 11 165 devices in year 2017, p < .01).2 There has been a decline in the rate of complications over the study period primarily driven by lower rates of cardiovascular and neurological complications.3 Overall mortality continues to be low during the study period, with an even lower trend towards reduced mortality over the study years (1.3% in year 2015 vs. 0.1% in year 2017, p < .01).4 The total LOS and hospitalization costs after Watchman implantation declined over the study period.
Percutaneous LAAO with the Watchman device provides a viable alternative to oral anticoagulation in select patients based on randomized trials that have shown efficacy and safety of utilizing this approach for stroke risk reduction.10,11 The FDA approved the device for commercial use in United States in March of 2015 and the Watchman implant procedure currently carries a class IIb recommendation in patients with nonvalvular AF at risk for stroke per the latest American College of Cardiology guidelines.15 Our analysis of a contemporary, real-world, national database sampling U.S. practice since FDA approval of Watchman showed consistent increased utilization of device procedures suggesting gradual assimilation of this device implantation procedure in clinical practice. Additionally, our analysis also showed that complications and inpatient mortality associated with implantation of Watchman devices continued to show a downward trend.
Our study showed a significant decline in the overall complication rate over the study period (26.4% in year 2015 vs. 7.9% in year 2017, p < .01). This downtrend was primarily driven by a reduction in cardiovascular and neurological complications over the study period. Cardiac perforation complications including cardiac tamponade were encountered in the landmark PROTECT AF trial where its prevalence was approximately 4.3%.10 Subsequently, with improved operator experience, the incidence of this complication was lowered to 1.9% in the PREVAIL trial, 1.4% in CAP (Continued Access to PROTECT AF), 1.9% in CAP2 (Continued Access to PREVAIL), and 0.3% in EWOLUTION registries.11,16,17 The overall rate of cardiac tamponade in the current study of contemporary Watchman patients was 0.8%. This rate is similar to post-FDA approval study led by Reddy et al. in which the authors reported a nearly similar rate of cardiac tamponade at 1%; still higher than the European EWOLUTION registry which reported cardiac tamponade rate of 0.3%.18 It is also similar to a recently published contemporary registry of Watchman implantations from National Cardiovascular Data Registry which analyzed nearly 38 000 patients.19 It may be expected that with more widespread availability of the Watchman device in U.S. practice that the rate of cardiac tamponade may continue to decline over the coming years. The rate of ischemic stroke/transient ischemic attack (TIA) was 1% in our study cohort with most cases reported in 2015 and a significant downtrend since that time (0.7% in year 2016% and 0.4% in year 2017, p < .01). Vuddanda et al. have shown nearly similar ischemic stroke/TIA rates of about 0.5% when analyzing Watchman implants from year 2016.14 These strokes are presumed to be due to inadvertent air or clot embolization from the transseptal sheath and enhanced physician training should continue to mitigate this risk.
The current study showed a downward trend in in-hospital mortality suggesting improved safety with the device with more operator experience (from 1.3% in year 2015% to 0.1% in year 2017, p < .01). Improvement in mortality rate trends over time seen in the current study of real-world patients was also seen in previous clinical studies, including the CAP2 registry which supplemented the PREVAIL trial and was designed to continue long-term accrual of data, and showed a mortality rate of 0.2% within 7 days of Watchman implant.16 A subsequent study utilizing National Inpatient Sample database by Vuddanda showed a mortality rate of 0.3% at discharge for combined endocardial and epicardial based approaches for LAA occlusion.14 In another post-FDA approval analysis of more than 3800 patients undergoing Watchman implantation from March 2015 to May 2016, Reddy et al. demonstrated procedure related mortality of 0.078%.18 The prospective EWOLUTION registry that enrolled more than 1000 consecutive patients undergoing Watchman implantation outside the United States also showed low procedure related mortality of 0.1%.17 These earlier studies along with our more contemporary data suggest that overall implantation of the Watchman device is associated with low absolute rates of mortality that is decreasing over time.
5 |. LIMITATIONS
Our study has the following key limitations1: The NIS is an administrative claims-based database that utilized ICD codes which may be prone to errors and could introduce information bias due to outcome misclassification. The hard clinical end points, however, are less subjected to error. Additionally, AHRQ quality control measures are routinely instituted that guarantee data integrity.13 Additionally, the ICD-9 code utilized in this study was not specific to the Watchman device and could be referred for any LAA occlusion procedure. Due to the limited magnitude of other research studies of endocardial devices and any epicardial LAA occlusion procedures performed in the United States during the study period,14 we believe that application of this code for the purpose of our study was able to mostly characterize Watchman implants.2 The NIS only captures inpatient admissions and does not provide any information on outpatient encounters. This limitation may result in selection bias; however, our data is well representative of national utilization of Watchman devices performed during in-patient settings; in fact since inpatient hospitalization is often required for reimbursement for the procedure, our results may be more indicative of widespread practice.3,20 The NIS censors data gathering at discharge so long-term outcomes could not be ascertained from the present data set.4 Specific data on potential confounders including medications, as well as operator and intraprocedural characteristics could not be examined from the NIS.
6 |. CONCLUSION
In conclusion, in this large, nationally representative sample of the United States database, there has been a significant increase in the use of Watchman devices since FDA approval in 2015. Between 2015 and 2017, a significantly reduced rate of procedural complications and in-hospital mortality was noted, which appeared primarily driven by reduction in cardiovascular and neurological complications. In later years, LOS shortened and costs of hospitalization for the procedure decreased.
Supplementary Material
Acknowledgments
Disclosures: Dr. Hsu reports receiving honoraria from Medtronic, Abbott, Boston Scientific, Biotronik, Janssen Pharmaceuticals, Bristol-Myers Squibb, Altathera Pharmaceuticals, Zoll Medical, and Biosense-Webster, research grants from Biotronik and Biosense-Webster, and has equity interest in Acutus Medical and Vektor Medical. Other authors: No disclosures.
ABBREVIATIONS:
- AF
atrial fibrillation
- LAA
left atrial appendage
- DOACs
direct acting oral anti-coagulants
- FDA
Food and Drug Administration
- LOS
length of stay
- OR
odds ratio
- NIS
National Inpatient Sample
- ICD-9-CM
International Classification of Diseases, 9th Revision, Clinical Modification
- ICD-10-CM
International Classification of Diseases, 10th Revision, Clinical Modification
Footnotes
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the supporting information tab for this article.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the first author (MBM) upon reasonable request. The data were extracted from Healthcare Cost and Utilization Project National Inpatient Sample (NIS) database (https://www.hcupus.ahrq.gov/nisoverview.jsp).
REFERENCES
- 1.Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics-2016 update: a report from the American Heart Association. Circulation. 2016;133:e38–e360. [DOI] [PubMed] [Google Scholar]
- 2.Chugh SS, Havmoeller R, Narayanan K, et al. Worldwide epidemiology of atrial fibrillation: a global burden of disease 2010 study. Circulation. 2014;129:837–847. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Saposnik G, Gladstone D, Raptis R, Hart RG. Atrial fibrillation in ischemic stroke: predicting response to thrombolysis and clinical outcomes. Stroke. 2013;44:99–104. [DOI] [PubMed] [Google Scholar]
- 4.Seet RC, Zhang Y, Wijdicks EF, Rabinstein AA. Relationship between chronic atrial fibrillation and worse outcomes in stroke patients after intravenous thrombolysis. Arch Neurol. 2011;68:1454–1458. [DOI] [PubMed] [Google Scholar]
- 5.Alkhouli M, Noseworthy PA, Rihal CS, Holmes DR Jr. Stroke prevention in nonvalvular atrial fibrillation: a stakeholder perspective. J Am Coll Cardiol. 2018;71:2790–2801. [DOI] [PubMed] [Google Scholar]
- 6.Hayden DT, Hannon N, Callaly E, et al. Rates and determinants of 5-year outcomes after atrial fibrillation-related stroke: a population study. Stroke. 2015;46:3488–3493. [DOI] [PubMed] [Google Scholar]
- 7.Oldgren J, Healey JS, Ezekowitz M, et al. Variations in cause and management of atrial fibrillation in a prospective registry of 15,400 emergency department patients in 46 countries: the RE-LY Atrial Fibrillation Registry. Circulation. 2014;129:1568–1576. [DOI] [PubMed] [Google Scholar]
- 8.Hsu JC, Maddox TM, Kennedy KF, et al. Oral anticoagulant therapy prescription in patients with atrial fibrillation across the spectrum of stroke risk: insights from the NCDR PINNACLE Registry. JAMA Cardiol. 2016;1:55–62. [DOI] [PubMed] [Google Scholar]
- 9.Holmes DR Jr, Alkhouli M, Reddy V. Left atrial appendage occlusion for the unmet clinical needs of stroke prevention in nonvalvular atrial fibrillation. Mayo Clin Proc. 2019;94:864–874. [DOI] [PubMed] [Google Scholar]
- 10.Holmes DR, Reddy VY, Turi ZG, 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. 2009;374:534–542. [DOI] [PubMed] [Google Scholar]
- 11.Holmes DR Jr, Kar S, Price MJ, 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. 2014;64:1–12. [DOI] [PubMed] [Google Scholar]
- 12.Reddy VY, Doshi SK, Kar S, et al. 5-Year outcomes after left atrial appendage closure: from the PREVAIL and PROTECT AF trials. J Am Coll Cardiol. 2017;70:2964–2975. [DOI] [PubMed] [Google Scholar]
- 13.Agency for Healthcare Research and Quality. Overview of the national inpatient sample (NIS) Rockville: AHRQ. https://www.hcupus.ahrq.gov/nisoverview.jsp. Accessed on January 21, 2020. [Google Scholar]
- 14.Vuddanda VLK, Turagam MK, Umale NA, et al. Incidence and causes of in-hospital outcomes and 30-day readmissions after percutaneous left atrial appendage closure: A US nationwide retrospective cohort study using claims data. Heart Rhythm. 2019;17. [DOI] [PubMed] [Google Scholar]
- 15.January CT, Wann LS, Calkins H, et al. 2019AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. Heart Rhythm, 2019(16):e66–e93. [DOI] [PubMed] [Google Scholar]
- 16.Holmes DR Jr, Reddy VY, Gordon NT, et al. Long-term safety and efficacy in continued access left atrial appendage closure registries. J Am Coll Cardiol. 2019;74:2878–2889. [DOI] [PubMed] [Google Scholar]
- 17.Boersma LVA, Schmidt B, Betts TR, et al. Implant success and safety of left atrial appendage closure with the WATCHMAN device: periprocedural outcomes from the EWOLUTION registry. Eur Heart J. 2016;37:2465–2474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Reddy VY, Gibson DN, Kar S, et al. Post-approval U.S. experience with left atrial appendage closure for stroke prevention in atrial fibrillation. J Am Coll Cardiol. 2017;69:253–261. [DOI] [PubMed] [Google Scholar]
- 19.Freeman JV, Varosy P, Price MJ, et al. The NCDR left atrial appendage occlusion registry. J Am Coll Cardiol. 2020;75:1503–1518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Centers for Medicare and Medicaid Services Coverage with Evidence Development (Left Atrial Appendage Occlusion) https://www.cms.gov/Medicare/Coverage/Coverage-with-Evidence-Development/LAAC. Accessed on March 16, 2020.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the first author (MBM) upon reasonable request. The data were extracted from Healthcare Cost and Utilization Project National Inpatient Sample (NIS) database (https://www.hcupus.ahrq.gov/nisoverview.jsp).