Abstract 18: Trends in Median CPR Duration Over Time in Patients With and Without Return of Spontaneous Circulation
Introduction: Survival after in-hospital cardiac arrest (IHCA) is increasing. In the Get-With-The-Guidelines-Resuscitation (GWTG-R) registry, longer median CPR duration in patients not achieving ROSC is associated with higher survival rates at the hospital level. We analyzed trends over time in median CPR duration by hospital in patients who achieved ROSC and those who did not, and stratified this analysis by age, gender and race. Methods: We included adult IHCA cases in GWTG-R from 2001-2017, excluding data from a given hospital and year if fewer than 5 eligible arrests were recorded. A nonparametric test for trend was done to evaluate median CPR duration over time in those with and without ROSC, in all patients and in groups stratified by age (<60, 61-80 and >80 years), gender, and race (white and black). Linear regression was done to evaluate the amount of change per year. Association with survival was tested using Pearsons correlation. Results: Of 359,107 IHCA events, 31,189 were excluded, leaving 327,918 for analysis. Over time, there was a significant increase in median CPR duration in patients who did not achieve ROSC, and a decrease in those who did attain ROSC.(Fig.) These trends persisted when stratified by gender, race and age. Each year was associated with a decrease in median CPR duration of 0.37 min (95% CI -0.41 to -0.33 min) in those with ROSC and an increase of 0.29 min (95% CI 0.25 to 0.33 min) in those without. There was a small but significant correlation between median CPR duration in those without ROSC and adjusted survival by hospital over time (r=0.224, p<0.0001). Conclusions: In the GWTG-R registry, median duration of CPR is decreasing over time in patients achieving ROSC, but increasing in those not achieving ROSC. The increasing trend in CPR duration in those without ROSC correlates positively with the trend in survival. Whether the increase in median CPR duration in those without ROSC is contributing causally to improvements in survival warrants further study.