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<0.01). Overall, women constituted 40.1% (n=7,095) 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 US (from 1,195 device implants in year 2015 to 11,165 device implants in year 2017, p < 0.01, 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 < 0.01). 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 < 0.01), while the rate of any neurological complication decreased from 7.9% in year 2015 to just 0.9% in year 2017 (p < 0.01). 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).
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<0.01), chronic blood loss anemia (OR 3.63 [95% CI 1.37-9.61], p<0.01) and coagulopathy (OR 4.90 [95% CI 2.32-10.35], p<0.01) 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<0.01) and history of hypertension (OR 0.28 [95% CI 0.14-0.56], p<0.01) were associated with a lower odds of mortality at discharge.
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).