Use of brainstem auditory-evoked response testing to assess neurologic outcome following near drowning in children.

OBJECTIVE

To determine a correlation between serial brainstem auditory-evoked response measurements and ultimate neurologic outcome in pediatric patients who suffered a cardiac arrest resulting from a submersion accident.

DESIGN

Inception cohort, prospective correlational study.

SETTING

A 15-bed pediatric ICU (within a 150-bed tertiary care pediatric hospital) admitting patients of all ages except nonsurgical neonates.

PATIENTS

All previously normal pediatric patients who suffered a cardiac arrest after a submersion accident. Patients evaluated: 111; patients studied: 89.

METHODS

Patients received serial brainstem auditory-evoked response testing within 6 hrs of resuscitation and then once daily for up to 10 days. Brainstem auditory-evoked response measurements included wave I-V interpeak latency, wave V amplitude, and wave I/V amplitude ratio. Upon discharge, patients were evaluated and classified into one of four neurologic outcome groups: normal, handicapped, vegetative, or dead. Patients classified into the handicapped group exhibited mild neurologic deficits after discharge. Vegetative patients were noninteractive with their environment and required full-time caretaker support. Serial brainstem auditory-evoked response measurements from the four outcome groups were compared with brainstem auditory-evoked response measurements obtained from a group of 39 healthy children of comparable age.

MAIN RESULTS

Patients who recovered neurologically intact manifested brainstem auditory-evoked response measurements that were similar to controls. Brainstem auditory-evoked response measurements in the handicapped outcome group were also normal after resuscitation but showed significant reduction in wave V amplitudes over the ensuing days. When compared with controls, patients with a vegetative outcome manifested abnormally prolonged wave I-V interpeak latencies, diminished wave V amplitudes, and large-wave I/V amplitude ratios following resuscitation. However, I-V interpeak latencies normalized within 24 hrs after resuscitation. I-V interpeak latencies were no different than controls until hospital day 3, at which time they became significantly prolonged. Patients who were declared brain dead or died from cardiovascular collapse exhibited very abnormal brainstem auditory-evoked response measurements on admission and until death. Wave V could not be detected on admission in 19/31 patients within this group.

CONCLUSIONS

Brainstem auditory-evoked response testing is useful as an aid in the assessment of neurologic outcome following submersion-induced cardiac arrest. However, standardization of brainstem auditory-evoked response testing and production of normative data are required before this modality can be more widely studied and applied.