J Korean Soc Emerg Med.
1998 Sep;9(3):437-444.
Analysis of predictive factors in the Assessment of Mear-drowning in Children
Abstract
- Drowning is the second most common cause of accidental death of children in Korea. But the study of near-drowning is very rare in Korea. The purpose of our study is to determine the factors that may influence survival on the basis of submersion time, consciousness state upon admission to hospital, the kinds of water, laboratory findings and neurological outcomes. The survey was performed by a retrospective cohort study on 28 near drowning victims of less than 15 years of age who were admitted to the emergency room in the Chosun University Hospital between May 1988 & May 1997. The results were as follows 1) Sex distribution was 22 males and 6 females. 2) Submersion time was (5 minutes in 9 cases, 5-9 minutes in 6 cases,10-14 minutes in 5 cases,15-19 minutes in 3 cases, > or =20 minutes in 5 cases. 3) Comatose patients upon arrival were 6/7 cases in the death group(86%) and 2/21 cases in the improved group(10%). they had unfavorable outcomes(P<0.05). 4) The first pH value was mean 7.02)0.12 in the death group and mean 7.31 +/-0.13 in the improved group. there was a statistically significant difference between the death and the improved group(P<0.05). 5) The patients who had increased blood glucose concentration were all 7 cases in the death group and 12/22 cases in the improved group(54%). The mean blood glucose concentration was mean 424.7235.6mg/dl in the death group and mean 140.182.7mg/dl in the improved group. There was a statistically significant difference between the death and the improved group(P<0.05). 6) The patients who had pulmonary edema upon arrival were all 7 cases in the death group and 1/21 cases in the improved group(18%). They had unfavorable outcomes(P<0.05). 7) The patients who had been submerged more than 15 minutes were all 7 cases in the death group and 6/21 cases in the improved group(5%). They had unfavorable outcomes(P<0.05). We conclude that pediatric victims of near-drowning can be assigned to high or low likelihoods of unfavorable outcomes with the use of five variables, comatose mentation upon arrival, decreased initial blood pH, increased initial blood glucose concentration, pulmonary edema, and maximum submersion time estimated longer than fifteen minutes. This prediction rule may be useful if it can be validated in another cohort.
