Abstract 15935: 20,000 Person-Years Later: Long-term Mortality in a Large Drowning Registry
Objective:Relatively little is known about the long-term outcomes of drowning. There is some evidence of long-term neurocognitive dysfunction among small cohorts, but larger epidemiological studies are lacking. We aimed to assess the long-term mortality of drowning victims from a large registry.
Hypothesis: We hypothesized that patients with sequelae of more severe drowning injury are at higher risk of long-term mortality.
Methods: Secondary analysis of an existing drowning registry from Seattle, WA. We included the subset of patients that survived to hospital discharge, tabulating Utstein-style drowning variables with descriptive statistics. We used the National Death Index (NDI) to assess long-term mortality beyond the index event through 2012. Wilcoxon rank-sum and chi-square tests assessed differences between long term survivors and non-survivors. We constructed KM curves, stratified by age, sex, drowning-related cardiac arrest, and mRS at hospital discharge, and compared them with the log-rank test. Cox proportional hazard modeling tested variables associated with long-term mortality.
Results: Of 2,824 subjects in the registry (submersion 1/74 - 7/96), 776 (27%) survived to hospital discharge and were included in our analyses (median age 5 years, IQR 2-15; 68% male). Long term survivors and non-survivors differed by age (4 years, IQR 2-15 vs. 25 years, IQR 6-46; p<0.0001), recreational substance use (4% vs. 18%; p<0.001), pre-existing comorbidities (16% vs. 38%; p=0.01), drowning-related cardiac arrest (7% vs. 25%; p<0.0001), mechanical ventilation (13% vs. 37%; p<0.0001), and seizures (2% vs. 10%; p<0.0001). Only 63 (8%) subjects died during the 19,699 person-year follow-up period. Subjects with age >40 years (p<0.0001), drowning-related cardiac arrest (p<0.0001), and mRS 4-5 (p<0.0001) were at higher risk of long-term mortality. In adjusted analyses, age (HR 1.05; 95% CI 1.03, 1.07) and mRS 4-5 at hospital discharge (HR 8.2; 95% CI 1.8, 36.4) were associated with long-term mortality.
Conclusion: Overall mortality was low during long term follow-up. Age at the index event, drowning-related cardiac arrest, and functional status at hospital discharge were associated with subsequent long term mortality.
Author Disclosures: E.A. Michiels: None. J.C. Reynolds: None. M. Reeves: None. M.N. Ahmadabadi: None.
- © 2015 by American Heart Association, Inc.