Brain MR imaging and spectroscopy for outcome prognostication after pediatric cardiac arrest

The LCModel method analyzes an in vivo spectrum as a Linear Combination of Model spectra of metabolite solutions in vitro. By using complete model spectra, rather than just individual resonances, maximum information and uniqueness are incorporated into the analysis. A constrained regularization method accounts for differences in phase, baseline, and lineshapes between the in vitro and in vivo spectra, and estimates the metabolite concentrations and their uncertainties. LCModel is fully automatic in that the only input is the time-domain in vivo data. The lack of subjective interaction should help the exchange and comparison of results. More than 3000 human brain STEAM spectra from patients and healthy volunteers have been analyzed with LCModel. N-acetylaspartate, cholines, creatines, myo-inositol, and glutamate can be reliably determined, and abnormal levels of these or elevated levels of lactate, alanine, scyllo-inositol, glutamine, or glucose clearly indicate numerous pathologies. A computer program will be available.

To describe the short-term outcome of pediatric intensive care by quantifying overall functional morbidity and cognitive impairment, I developed the Pediatric Overall Performance Category (POPC) and the Pediatric Cerebral Performance Category (PCPC) scales, respectively. A total of 1469 subjects (1539 admissions) were admitted to the pediatric intensive care unit of Arkansas Children's Hospital from July 1989 through December 1990. Patients were assigned baseline POPC and PCPC scores derived from historical information and discharge scores at the time of discharge from the hospital (or from the pediatric intensive care unit for patients with multiple hospitalizations). Delta scores were calculated as the difference between the discharge scores and the baseline scores. The changes in POPC and PCPC scores were associated with several measures of morbidity (length of stay in the pediatric intensive care unit, total hospital charges, and discharge care needs) and with severity of illness (pediatric risk of mortality score) or severity of injury (pediatric trauma score) (p less than 0.0001). Interrater reliability was excellent (r = 0.88 to 0.96; p less than 0.001). The POPC and PCPC scales are apparently reliable and valid tools for assessing the outcome of pediatric intensive care.

Significant improvements have been achieved in cardiac arrest resuscitation and postarrest resuscitation care, but mortality remains high. Most of the poor outcomes and deaths of cardiac arrest survivors have been attributed to widespread brain injury. This brain injury, commonly manifested as a comatose state, is a marker of poor outcome and a major basis for unfavorable neurological prognostication. Accurate prognostication is important to avoid pursuing futile treatments when poor outcome is inevitable but also to avoid an inappropriate withdrawal of life-sustaining treatment in patients who may otherwise have a chance of achieving meaningful neurological recovery. Inaccurate neurological prognostication leading to withdrawal of life-sustaining treatment and deaths may significantly bias clinical studies, leading to failure in detecting the true study outcomes. The American Heart Association Emergency Cardiovascular Care Science Subcommittee organized a writing group composed of adult and pediatric experts from neurology, cardiology, emergency medicine, intensive care medicine, and nursing to review existing neurological prognostication studies, the practice of neurological prognostication, and withdrawal of life-sustaining treatment. The writing group determined that the overall quality of existing neurological prognostication studies is low. As a consequence, the degree of confidence in the predictors and the subsequent outcomes is also low. Therefore, the writing group suggests that neurological prognostication parameters need to be approached as index tests based on relevant neurological functions that are directly related to the functional outcome and contribute to the quality of life of cardiac arrest survivors. Suggestions to improve the quality of adult and pediatric neurological prognostication studies are provided.