Surgical aspects of pediatric cardiac extracorporeal membrane oxygenation

Outcome of 12 drowned children with attempted resuscitation on cardiopulmonary bypass: An analysis of variables based on the “Utstein Style for Drowning”

BACKGROUND

In 2003, the International Liaison Committee on Resuscitation (ILCOR) published the "Utstein Style for Drowning" (USFD) to advance knowledge on the epidemiology, treatment, and outcome prediction after drowning. Applying the USFD and evaluating its data template for outcome analysis, we report here on the largest study published thus far of drowned children (age 0-14) who underwent attempted resuscitation on cardiopulmonary bypass (CPB).

METHODS

We conducted a retrospective review of all drowned children admitted to Göttingen University Hospital between 1/1987 and 12/2005 in sustained cardiopulmonary arrest and resuscitation with CPB. We correlated eight outcome-affecting USFD variables and four additional variables not included in the USFD with potential impact on outcome to four outcome groups: survival, non-survival, survival with full recovery, and failed resuscitation.

RESULTS

Out of 12 children (aged 22 months to 7.5 years), 5 survived to hospital discharge and 7 died in hospital. Two survivors recovered fully and three remained in a vegetative state. In two patients, resuscitation on CPB failed. Both children who fully recovered, compared to the 10 others, had relatively low serum K+ concentrations (2.6 and 3.7 mmol/l versus 5.8+/-3.8 mmol/l [mean+/-S.D.; n=10]), a relatively slow rewarming speed (1.9 and 1.2 degrees C/h versus 3.4+/-1.8 degrees C/h), were female (all three girls survived), received early basic life support (BLS) and showed idioventricular bradycardia. Both children with failed resuscitation had severe hyperkalaemia (11.7 and 13.3 mmol/l versus 10 others, 4.0+/-1.5 mmol/l), were relatively rapidly rewarmed (6.9 and 4.0 degrees C/h versus 10 others, 2.61+/-1.32 degrees C/h), male, and in asystole. We identified no outcome trends for age, pH, or water and core temperatures.

CONCLUSIONS

Most variables relevant for outcome in drowned children can be documented with the use of the USFD. Additional variables not included in the USFD that have emerged from this study and may predict outcome include serum K+ concentration, rewarming speed, and initial cardiac rhythm.

Cold water submersion and cardiac arrest in treatment of severe hypothermia with cardiopulmonary bypass

In the paediatric population, submersion injury with drowning or near-drowning represents a significant cause of morbidity and mortality. This study reviews retrospectively our own experiences and the literature on the use of cardiopulmonary bypass (CPB) to rewarm paediatric victims of cold water submersion who suffer severe hypothermia (<28 degrees C) and cardiac arrest (asystole or ventricular fibrillation). In addition to three children treated at our institution, nine other victims were found in the literature. In this cohort of 12 children aged between 2 and 12 years, there was a tendency to better outcome with lower core temperature at the beginning of extracorporeal circulation (mean temperature in nine survivors, 20 degrees C; in three non-survivors, 25.5 degrees C). The lowest temperature survived was 16 degrees C. Neither base excess, pH nor serum potassium levels were reliable prognostic factors. The lowest base excess in a survivor was -36.5 mmol/l, the lowest pH 6.29. We consider CPB as the method of choice for resuscitation and rewarming of children with severe accidental hypothermia and cardiac arrest (asystole or ventricular fibrillation). Compared with adults, children, especially smaller ones, require special consideration with regard to intravenous cannulation as drainage can be inadequate using femoral-femoral cannulation. In hypothermic children we advocate, therefore, emergency median sternotomy. Until more information regarding prognostic factors are available, children who are severely hypothermic and clinically dead after submersion in cold water--even if for an unknown length of time--should receive cardiopulmonary resuscitation (CPR) and be transported without delay to a facility with capabilities for CPB instituted via a median sternotomy.