0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Article |

Predictors of Death and Neurologic Impairment in Pediatric Submersion Injuries:  The Pediatric Risk of Mortality Score FREE

Gary B. Zuckerman, MD; Patrice M. Gregory, PhD, MPH; Suzanne M. Santos-Damiani, MD
[+] Author Affiliations

From the Departments of Pediatrics (Drs Zuckerman and Santos-Damiani) and Family Medicine (Dr Gregory), Robert Wood Johnson Medical School, New Brunswick, NJ.


Arch Pediatr Adolesc Med. 1998;152(2):134-140. doi:10.1001/archpedi.152.2.134.
Text Size: A A A
Published online

Objective  To evaluate the Pediatric Risk of Mortality Score (PRISM score) as a tool to distinguish which patients presenting to the emergency department (ED) or pediatric intensive care unit (PICU) would survive neurologically intact from those who would die or survive with severe neurologic impairment following a submersion incident.

Design  Retrospective chart review.

Setting  A regional tertiary care ED and PICU.

Materials  Medical records of drowning and near-drowning patients between the ages of 1 day and 18 years who were admitted to the ED or PICU from January 1986 through December 1995.

Main Outcome Measures  For each patient, a PRISM score was calculated based on data collected at the time of arrival to the ED or PICU. A "probability of outcome" was calculated using the PRISM score. Outcomes were defined as survival with intact neurologic function, survival with severe neurologic impairment, or death, and were assessed using the Pediatric Cerebral Performance Category Scale. Scores and probabilities of outcome were compared with the actual outcome for both the ED and PICU patients using the Fisher exact test (2 tailed).

Results  All ED patients with PRISM scores less than 25 or a probability of outcome less than 50% (n=40) survived with intact neurologic function (P<.001). All ED patients with PRISM scores of 25 or greater or a probability of outcome of 50% or greater (n=10) either died or survived with severe neurologic impairment (P< .001). All PICU patients with a PRISM score of 5 or less (n=25) or a probability of outcome less than 4% (n=30) survived with intact neurologic function (P<.05). All PICU patients with a PRISM score greater than 20 or a probability of outcome greater than 50% either died or survived with severe neurologic impairment (P<.05). Outcomes could not accurately be predicted in PICU patients when the PRISM scores were 6 or greater and ≤20 or less (n=10), or when the probability of outcome was 4% or greater and 50% or less.

Conclusions  The PRISM scoring system accurately distinguished ED patients who would survive neurologically intact from those who would die or suffer neurologic impairment. There was not a specific PRISM score or probability of outcome that could distinguish PICU patients who would survive neurologically intact from those who would die or suffer severe neurologic impairment. The PRISM scoring system appeared to be more accurate in distinguishing intact survival from death or neurologic impairment in ED patients than in PICU patients.

DROWNING, defined as death within 24 hours of a submersion incident, and near drowning, defined as survival for at least 24 hours following a submersion incident, represent significant causes of morbidity and mortality in children. Death from submersion incidents is the second-leading cause of accidental death in children,1 with one third of all survivors sustaining significant neurologic damage due to anoxic encephalopathy.2 Predicting neurologic outcome has emerged as a major clinical problem in near drowning.2

Many investigators have attempted to identify prognostic factors related to both survival and neurologic outcome from submersion incidents.313 Most, however, have met with difficulties. There is still no method that accurately predicts which children will survive with intact neurologic function vs those who will either die or survive with severe brain injury. The development of a system to predict prognosis in near-drowning victims is needed. One such system may be the Pediatric risk of Mortality Score (PRISM score).14

The PRISM is a scoring system that predicts mortality in critically ill children.14 The PRISM score is based on the observation that the amount and extent of physiologic dysfunction is related to the patient's mortality risk. The PRISM score has 14 physiologic variables and 23 variable ranges. For each physiologic variable, a point total is provided that directly reflects the contribution of that instability to mortality risk (Table 1). The total PRISM score gives a relative scale of severity of illness. A more precise estimate of mortality risk can be computed from the following equation: Probability of Death = exp (R)/[1 + exp(R)], where R=0.207[PRISM score] − 0.005[age in months] − 4.782.14

To date, the PRISM score has not been used to predict neurologic outcome in pediatric drowning and near-drowning patients, nor has it been used for patients presenting to emergency departments (EDs). The objective of this study was to evaluate the PRISM score as a tool to distinguish which patients presenting to the ED or pediatric intensive care unit (PICU) would survive neurologically intact from those who would die or survive with severe neurologic impairment following a submersion incident. We hypothesized that the equation used to compute the "probability of death" from the PRISM score could also compute a "probability of outcome" which would distinguish those patients presenting to an ED who would die or suffer severe neurologic impairment from those who would survive neurologically intact. We further hypothesized that a probability of outcome computed from the PRISM score would likewise distinguish patients presenting to a PICU who would die or suffer severe neurologic impairment from those who would survive neurologically intact.

A retrospective chart review was conducted on all drowning and near-drowning patients between the ages of 1 day and 18 years who were admitted to the ED or PICU of the Robert Wood Johnson University Hospital, New Brunswick, NJ, between January 1, 1986, and December 31, 1995. Each medical record was analyzed for demographic data (age and sex) and incident characteristics (water temperature and location). Submersion incidents were classified as either "warm-water incidents" (water temperature >20°C) or "cold-water incidents" (water temperature ≤20°C) (by history noted in the chart). This distinction was based on prior literature regarding warm and cold-water incidents.15 Exclusion criteria included preexisting neurologic conditions, head trauma or spinal cord injury at the time of the submersion incident, use of barbiturates or steroids at the time of resuscitation and/or assessment, and cold-water submersion incidents.

For each patient, a PRISM score was calculated based on data obtained at the time of arrival to the ED. A probability of outcome was then determined for each patient using the equation that computes the probability of death from the PRISM score. When PRISM scores were calculated, the scores were based on raw data obtained in the ED. The scores did not take into account any variation of prehospital resuscitation (ie, prehospital intubation, intravenous lines, or medications). These patients comprised the "ED group" of the study. The neurologic function of each patient at the time of discharge from the hospital was defined using the Pediatric Cerebral Performance Category Scale (PCPCS) of Fiser16 (Table 2). Patients with a PCPCS score of 3 or less were considered to have survived neurologically intact and those with a PCPCS score of greater than 3 were considered to have severe neurologic impairment. Patients were then assigned to 1 of 2 groups according to their actual outcome. The first group consisted of patients who survived with intact neurologic function. The second group consisted of patients who survived with severe neurologic impairment as well as patients who died.

Table Graphic Jump LocationTable 2. Pediatric Cerebral Performance Category Scale*

All patients from the ED group who were admitted to the PICU made up a second study group. For each of these patients, a new PRISM score was calculated based on new data obtained in the PICU. A probability of outcome was then determined for each in the same manner as with the ED patient group. When PRISM scores and the probabilities of outcome were calculated, the calculations were based on raw data obtained within 24 hours of admission to the PICU. The scores did not take into account any variation of ED resuscitation (ie, intubation, intravenous lines, or medications). These patients comprised the "PICU group" of the study. The neurologic function of each patient at the time of discharge from the hospital was defined using the PCPCS. Patients were again assigned to 1 of 2 groups based on their outcome in the same manner as were the ED patients.

The PRISM scores and probabilities of outcome were compared with the actual outcome using different cutoff points to optimize prediction of outcome for both the ED and PICU groups. Results were analyzed using the Fisher exact test (2 tailed). A P value of .05 or less was considered statistically significant.

The medical records of 50 patients between the ages of 1 day and 18 years were obtained. All 50 patients were admitted directly from the submersion incident site to the ED and represented the ED group in our study. None of the patients were transported from other facilities. Of the 50 patients who presented to the ED, 3 died in the ED and 9 were discharged home or were admitted to a general pediatric ward. The remaining 38 patients were admitted to the PICU and represented the PICU group. None of the patients identified met any of the defined exclusion criteria.

The mean age of the ED patients and PICU patients was 67.3 months and 56.7 months, respectively. The median age of the ED and PICU patients was 39 months and 31.5 months, respectively. In both patient populations, the youngest age was 2 months and the oldest was 238 months. Males outnumbered females in both groups and 100% of all submersion incidents were in warm water (Table 3).

Table Graphic Jump LocationTable 3. Characteristics and Outcome of Patients Presenting to ED and PICU*

Pools represented the most common location for submersion incidents in both ED and PICU patients (80% and 83%, respectively). Private pools accounted for 66% and 67% of pool-related submersion incidents in ED and PICU patients, respectively (Table 3).

Of the 50 patients who presented to the ED, 40 (80%) survived with intact neurologic function, 5 (10%) died, and 5 (10%) survived with severe neurologic impairment. The cause of death in 3 of the 5 children who died was cardiopulmonary arrest, which occurred in the ED. The other 2 children who died were diagnosed as having brain death. The diagnosis of brain death was made during their ensuing hospitalization in the PICU. Of the 38 patients who presented to the PICU, 31 (82%) survived with intact neurologic function, 2 (5%) died, and 5 (13%) survived with severe neurologic impairment (Table 3). The 2 children who died were diagnosed as having brain death.

Seventy percent (35 patients) of the ED patients and 76% (29 patients) of the PICU patients experienced a respiratory arrest at the submersion incident scene. Thirty-two percent (16 patients) of the ED and 34% (13 patients) of the PICU patients experienced a cardiopulmonary arrest at the scene. Cardiopulmonary resuscitation was performed at the scene in 48% (24 patients) of the ED patients and 53% (20 patients) of the PICU patients. Endotracheal intubation was performed at the submersion scene on 18% (9 patients) of the ED patients and 16% (6 patients) of the PICU patients (Table 4).

Table Graphic Jump LocationTable 4. Prehospital Characteristics at Scene*

All of the ED patients with a PRISM score less than 25 survived with intact neurologic function (n=40). These 40 patients had a median PRISM score of 3 and a mean of 4.8 (95% confidence interval [CI], 2.8-6.8). When the PRISM score was 25 or greater, 100% of the ED patients (n=10) either died or suffered severe neurologic impairment. The 10 patients who died or survived with neurologic impairment had a median PRISM score of 37 and a mean score of 36.8 (95% CI, 32.4-41.2). Therefore, a PRISM cutoff score of 25 was found to be statistically significantly related to survival and neurologic function (P<.001) (Table 5).

Table Graphic Jump LocationTable 5. PRISM Score and Probability of Outcome vs Outcome of ED and PICU Patients*

When the computed probability of outcome was less than 50%, 100% of the ED patients (n=40) survived with intact neurologic function. These survivors had a median probability of 1.0% and a mean of 4.0% (95% CI, 1.2-6.9). When the computed probability of outcome was 50% or greater, 100% of the ED patients (n=10) either died or survived with severe neurologic impairment (P<.001) (Table 5). The 10 patients who died or survived with neurologic impairment had a median probability of outcome of 92.2 and a mean score of 87.2% (95% CI, 79.3-95.1).

The 31 patients in the PICU group who survived with intact neurologic function had a median PRISM score of 1.0 and a mean of 2.5 (95% CI, 1.1-3.9). The 7 patients in the PICU group who either died or suffered severe neurologic injury had a median PRISM score of 12 and a mean of 19.9 (95% CI, 9.4-30.3). When the PRISM score was 5 or less, 100% of the PICU patients (n=25) survived with intact neurologic function (P<.05). When the PRISM score was 21 or greater, 100% of the PICU patients (n=3) either died or suffered severe neurologic impairment (P<.05). When the PRISM scores were 6 or greater and 20 or less, we could not distinguish which PICU patients would survive with intact neurologic function from those who would either die or suffer severe neurologic impairment (Table 5).

The 31 survivors in the PICU group had a median probability of outcome of 0.7% and a mean of 1.5% (95% CI, 0.5%-2.5%). The 7 patients in the PICU population who either died or suffered severe neurologic injury had a median probability of outcome of 8.3% and a mean of 37.9% (95% CI, 1.0%-74.9%). When the probability of outcome was less than 4%, 100% of the PICU patients survived with intact neurologic function (P<.05). When the probability of outcome was 50% or greater, 100% of the PICU patients either died or suffered severe neurologic impairment (P<.05). When the probability of outcome was 4% or greater and 50% or less, we could not distinguish which patients would survive with intact neurologic function from those who would either die or suffer severe neurologic impairment (Table 5).

The findings of this study bear on several relevant issues related to drowning and near drowning. These include the difficulties that previous investigators have encountered when attempting to predict survival and neurologic outcome in drowning and near-drowning patients, and the use of a physiologic scoring system to predict survival and neurologic outcome in children who have experienced submersion incidents.

Many investigators have attempted to identify prognostic factors related to both survival and neurologic outcome following submersion incidents in the hopes of identifying which patients may or may not benefit from aggressive resuscitation measures.313 Their studies used submersion incident features such as submersion and resuscitation times3,4; cardiopulmonary function5,6; the Glasgow Coma Scale7; submersion incident–specific scoring systems8; various technologic modalities such as brain computed tomographic scans,9 intracranial pressure monitoring,10 measurements of cerebral consumption of oxygen11 and brainstem auditory evoked potentials12; and laboratory studies such as serum glucose concentrations13 to predict outcome in drowning and near-drowning patients. Based on the above factors, these studies distinguished 2 populations of patients. "Low-risk" patients survived neurologically intact and "high-risk" patients either died or suffered severe neurologic impairment. Most of these studies, however, also identified a third population, a "middle-risk" group of patients, in whom prediction of outcome was extremely difficult if not impossible.

We sought to find whether the PRISM score would accurately distinguish patients presenting to the ED or PICU who would survive neurologically intact from those who would die or survive with severe neurologic impairment following a submersion incident. Selker17 claimed that in order for a physiologic scoring system to accurately predict outcome and be "user friendly" for health care professionals it should have a risk-adjusted mortality predictor that is sufficiently accurate and easy to use in real-time clinical settings, provide the same mortality prediction when used prospectively as when used retrospectively, be based on the data obtained during the early course of a patient's hospital presentation, require only data that are collected in the usual care of patients, use only objective and universally collected information, be accurate at all levels of its severity scale, be compatible with other computerized clinical data systems in general use and, finally, the components of the system must be open for inspection and testing.17

PROPONENTS of the PRISM scoring system claim that as a predictor system the PRISM scoring system meets many of these criteria. The system is easy to use prospectively in real-time clinical settings, and as our study shows, it can be also be used retrospectively. The PRISM scoring system requires only data that are obtained in the usual care of pediatric patients and uses objective and universally collected information. Mortality is predicted based on data from the first full 24 hours after presentation to the intensive care unit (ICU). The PRISM scoring system is compatible with many computerized clinical data systems.

Among pediatric intensivists who use physiologic scoring systems, the PRISM score has gained acceptance as the most accurate predictor of morbidity and identifier of relative severity of illness in pediatric critical care medicine. To our knowledge, the PRISM score has not been used to predict neurologic outcome in drowning and near-drowning pediatric patients, nor has it been used for patients presenting to EDs.

The PRISM scoring system was highly accurate in distinguishing patients who presented to the ED who would survive with intact neurologic function from those who would either die or survive with severe neurologic impairment following a submersion injury. In predicting outcomes among patients in the PICU, the PRISM scoring system identified populations of patients who would survive neurologically intact and others who would either die or suffer severe neurologic impairment. There did not appear to be a specific probability of outcome that could distinguish those who would survive neurologically intact vs those who would die or suffer severe neurologic impairment. In this regard, the PRISM scoring system showed greater accuracy in distinguishing outcome for patients presenting to the ED than for patients presenting to the PICU following a submersion incident.

The seemingly paradoxical ability of the PRISM scoring system, an ICU scoring system, to predict outcome more accurately in ED patients than in PICU patients touches on a number of controversies regarding physiologic scoring systems. Critics of ICU scoring systems claim that the use of these systems to predict outcome is fundamentally flawed.18 Their concern is that the physiologic data that are collected and then used in the scoring process have been influenced by medical and nursing interventions. The effects of these interventions on the physiologic data (ie, the influence of pressors on blood pressure) are not accounted for by the scoring systems. Interventions then skew the data collected and affect the score that is based on those data. This in turn invalidates the scoring system and the predicted outcome.18 In studies evaluating the ability of the Acute Physiology and Chronic Health Evaluation (APACHE) scoring system, ICU interventions were believed to invalidate the physiologic scores and affect the ability of the APACHE scoring system to predict outcome in medical and surgical adult ICU patients.1922 Selker17 claimed that for this reason, the ideal physiologic scoring system should base its predictions on data from the first minutes of hospital presentation before interventions affect physiologic data.

Others argue, however, that prior therapy does not significantly influence physiologic scores if data are obtained within the first 24 hours of hospitalization.23 Knaus et al24 found no statistically significant differences in APACHE scores when data were obtained either in the first few minutes or within 24 hours of hospitalization in adult ICU patients. In a large multicenter study, Pollack et al25 found that measurement frequency of physiologic variables did not significantly affect the performance of mortality prediction with the PRISM score.

In the present study, most of the data obtained from ED patients were collected within the first few minutes of hospitalization, prior to medical intervention (eg, endotracheal intubation, mechanical ventilation, pressor support). Data from the PICU patients, however, were often obtained after many of these medical interventions had occurred. These medical interventions may have skewed the PRISM scores obtained from data collected in the PICU patients whereas PRISM scores obtained from ED data were unaffected. This may account for the PRISM scoring system's greater accuracy in distinguishing outcome in patients presenting to the ED than in patients presenting to the PICU following a submersion incident. In this regard, Bratton et al26 found that the timing of data collection was critical when attempting to predict outcome in submersion injury patients. These investigators found that a 14-point coma scale that evaluated cortical and brainstem function was extremely reliable at predicting outcome when performed 24 hours after the submersion incident.

Another possible explanation for the discrepancy in the PRISM scoring system's ability to accurately predict outcome in ED patients vs PICU patients is what many believe to be a common limitation among ICU scoring systems. Civetta27 suggested that there are 3 general populations of ICU patients: low risk, medium risk, and high risk. He argued that ICU scoring systems perform well at predicting outcomes in the low- and high-risk populations. Outcomes are not as accurately predicted in the middle-risk population by scoring systems because many of the physiologic variables used in the scoring process are not profoundly abnormal.27 In the present study, 3 populations of PICU patients were in fact identified. Patients with PRISM scores less than 6, or probabilities of outcome less than 3.9% may be considered a low-risk population whose outcomes (survival with intact neurologic function) were accurately predicted by the PRISM scoring system. Patients with PRISM scores greater than 20, or probabilities of outcome greater than 50% may be considered a high-risk population whose outcomes (death or survival with neurologic impairment) were likewise accurately predicted by the PRISM scoring system. Patients with PRISM scores of 6 or greater and 20 or less, or probabilities of outcome of 4 or greater and 50% or less may be considered a medium-risk population. The PRISM scoring system could not accurately predict outcome in this group of patients. Interestingly, there did not appear to be a medium-risk population of ED patients. It is possible that medical interventions in ICUs not only invalidate outcome predictions by affecting physiologic data from which predictions are made, but these interventions also may themselves create a medium-risk population of patients in which accurate outcome prediction becomes extremely difficult if not impossible.

The present study also relates to another controversial issue, which is whether a physiologic scoring system such as the PRISM scoring system, intended for use in predicting mortality in a general population of ICU patients, can be used effectively to predict outcomes in patients with specific conditions and disease processes. The study by van Veen et al28 found that the PRISM scoring system was not accurate in predicting mortality in oncologic patients in a pediatric ICU. Zobel et al,29 however, noted that the PRISM scoring system accurately identified patients at increased risk for mortality after cardiac surgery. Madagame et al30 found that the best predictor of death for children requiring mechanical ventilation during therapy for acute bacterial meningitis was the admission PRISM score. Our findings in the ED patients suggest that the PRISM scoring system can accurately predict outcome for a specific condition (submersion incidents). Findings from the group of PICU patients in our study do not, however, support such a conclusion.

To our knowledge, the PRISM scoring system has never been used to predict neurologic outcome in pediatric drowning and near-drowning patients. Many have questioned the ability of ICU scoring systems to predict outcomes other than mortality.31 Ruttimann et al32 found the PRISM scoring system to be accurate in predicting pediatric ICU patient outcome status as independently functional, compromised functional or dead when used in conjunction with the Pediatric Overall Performance Category score. Madagame et al30 found the PRISM score to be the best predictor of functional outcome of children requiring mechanical ventilation during therapy for acute bacterial meningitis. The PRISM scoring system, in the present study, did not predict neurologic prognosis as an isolated outcome measure. When neurologic outcome was combined with mortality or survival, however, the PRISM scoring system accurately predicted outcome in 100% of ED patients and certain PICU patients.

This study is potentially limited by its retrospective design. Of specific concern is whether the accuracy noted by the PRISM scoring system in this retrospective study can be reproduced in a prospective investigation. Selker et al33 developed a physiologic scoring system, the Time-Insensitive Predictive Instrument (TIPI), to assess the likelihood of adult patients having acute cardiac ischemia and thus the appropriateness of admission to a coronary care unit. The TIPI proved to be valid both for real-time clinical use and for retrospective review.33 Whether the PRISM scoring system is "time insensitive" in predicting outcome in patients who experience a submersion incident needs to be determined.

Our study's small sample size may have affected our findings. A larger sample size may have included ED patients in whom the PRISM scoring system would not have been able to accurately distinguish intact survival from death or survival with severe neurologic impairment. It would be worthwhile to perform a large multicenter prospective study to evaluate the accuracy of the PRISM scoring system in predicting outcome in patients presenting to EDs and PICUs following submersion incidents.

Another potential limitation of this study concerns the problems of collecting PRISM data in the ED. Some of the data used to compute PRISM scores are based on laboratory tests that are more often performed in PICUs rather than in EDs. Emergency department patients who were never admitted to the PICU most likely did not have many of these PRISM data measured. This may in turn bias the calculation and interpretation of the PRISM scores of these patients. In a study that investigated the influence of missing values on physiologic scoring systems, Knaus et al24 claimed that if a physiologic measurement is not available, it is overwhelmingly likely that it is normal and would not affect the physiologic score.

To our knowledge, the PRISM scoring system has never been used to predict outcome in pediatric patients in the ED. The present study raises the question of whether there are potential dangers in using a tool that was designed for a PICU setting in a totally different setting. As previously stated, controversies exist as to whether differences in data collection, measurement frequency of physiologic variables, and patient interventions between different ICUs significantly influence physiologic scores. These differences may become strikingly influential if the PRISM scoring system is used in a setting other than a PICU. Large multicenter studies evaluating the accuracy of the PRISM scoring system in EDs may address these controversies and prove worthwhile.

The findings of this study and any investigation that uses ICU scoring systems to predict morbidity and mortality must be treated with caution when assessing outcomes and making decisions regarding individual patients in a "real-time" clinical setting. Prognostic systems will never predict outcome with 100% specificity; high severity scores, therefore, will never indicate absolute irreversibility of disease or impossibility of survival.23 Christensen et al34 claimed that individual outcome cannot be reliably predicted in the ED; therefore, aggressive resuscitation of near-drowning victims should be performed. As knowledge of pediatric cardiopulmonary resuscitation improves, more children survive submersion incidents; however, the number of severely neurologically impaired submersion victims is increasing.3 The PRISM scoring system may provide useful objective data to physicians who must make difficult decisions regarding patient triage and escalation or withdrawal of therapy in critically ill drowning and near-drowning patients. The PRISM scoring system may also prove helpful in efforts to inform families of a patient's probable outcome after a submersion incident. Decisions regarding the escalation or withdrawal of therapy in critically ill submersion injury patients cannot solely be based on the PRISM score or any other physiologic scoring system. Rather, these decisions must be based on a wide array of physiologic, clinical, psychosocial and emotional information that are relevant to each individual patient.

In conclusion, in the present study, the PRISM scoring system accurately distinguished ED patients who survived neurologically intact from those who died or suffered neurologic impairment. Although the PRISM scoring system identified certain populations of PICU patients who would survive neurologically intact and others who would either die or suffer severe neurologic impairment, there did not appear to be a specific probability of outcome or PRISM score that could predict outcome in PICU patients. Paradoxically, the PRISM score appeared to be more accurate in predicting outcome in ED patients than in PICU patients in this retrospective study of drowning and near drowning. A large multicenter prospective study evaluating the accuracy of the PRISM scoring system in predicting outcome in patients after a submersion is warranted.

Accepted for publication September 29, 1997.

Supported in part by a grant from the Center for Human Development and Developmental Disabilities, Institute for the Study of Child Development, Robert Wood Johnson Medical School, New Brunswick, NJ.

Presented in part at the Society for Pediatric Research/Ambulatory Pediatric Association, Washington, DC, May 9, 1996.

Editor's Note: It's interesting to find that an instrument designed for use in an intensive care unit appears to work more effectively, at least on this problem, in an emergency department. New applications work for all sorts of things.—Catherine D. DeAngelis, MD

Reprints: Gary B. Zuckerman, MD, Division of Pediatric Critical Care Medicine, Department of Pediatrics, Robert Wood Johnson Medical School, One Robert Wood Johnson Place-CN 19, New Brunswick, NJ 08903-0019 (e-mail: zuckerga@umdnj.edu).

Orlowski  JP Drowning, near-drowning, and ice-water drowning. JAMA. 1988;260390- 391
Link to Article
Peterson  B Morbidity of childhood near-drowning. Pediatrics. 1977;59364- 370
Quan  LKinder  D Pediatric submersions: prehospital predictors of outcome. Pediatrics. 1992;90909- 913
Quan  LWentz  KRGore  EJCopass  MK Outcome and predictors of outcome in pediatric submersion victims receiving prehospital care in King County, Washington. Pediatrics. 1990;86586- 593
Biggart  MJBohn  DJ Effect of hypothermia and cardiac arrest on outcome of near-drowning accidents in children. J Pediatr. 1990;117179- 183
Link to Article
Jacobsen  WKMason  LJBriggs  BASchneider  SThompson  JC Correlation of spontaneous respiration and neurologic damage in near-drowning. Crit Care Med. 1983;11487- 489
Link to Article
Dean  JMKaufman  ND Prognostic indicators in pediatric near-drowning: the Glasgow Coma Scale. Crit Care Med. 1981;9536- 539
Orlowski  JP Prognostic factors in pediatric cases of drowning and near-drowning. JACEP. 1979;8176- 179
Link to Article
Romano  CBrown  TFrewen  TC Assessment of pediatric near-drowning victims: is there a role for cranial CT? Pediatr Radiol. 1993;23261- 263
Link to Article
Nussbaum  EGalant  SP Intracranial pressure monitoring as a guide to prognosis in the nearly drowned, severely comatose child. J Pediatr. 1983;102215- 218
Link to Article
Connors  RFrewen  TCKissoon  N  et al.  Relationship of cross-brain oxygen content difference, cerebral blood flow and metabolic rate to neurologic outcome after near drowning. J Pediatr. 1992;121839- 844
Link to Article
Fisher  BPeterson  BHicks  G Use of brainstem auditory-evoked response testing to assess neurologic outcome following near drowning in children. Crit Care Med. 1992;20578- 585
Link to Article
Graf  WDCummings  PQuan  LBrutocao  D Predicting outcome in pediatric submersion victims. Ann Emerg Med. 1995;26312- 319
Link to Article
Pollack  MMRuttimann  UEGetson  PR Pediatric Risk of Mortality (PRISM) Score. Crit Care Med. 1988;161110- 1116
Link to Article
Conn  AWEdmonds  JFBarker  GA Near-drowning in cold fresh water: current treatment regimen. Anaesth Soc J. 1978;25259- 265
Link to Article
Fiser  DH Assessing the outcome of pediatric intensive care. J Pediatr. 1992;12168- 74
Link to Article
Selker  HP Systems for comparing actual and predicted mortality rates: characteristics to promote cooperation in improving hospital care. Ann Intern Med. 1993;118820- 822
Boyd  OGrounds  RM Physiological scoring systems and audit. Lancet. 1993;3411573- 1574
Link to Article
Vassar  MJWilkerson  CLDuran  PJPerry  CAHolcroft  JW Comparison of APACHE II, TRISS, and a proposed 24-ICU point system for prediction of outcome in ICU-treated patients. J Trauma. 1992;32490- 500
Link to Article
McAnena  OJMoore  FAMoore  EEMattox  KLMarx  JAPepe  P Invalidation of the APACHE II scoring system for patients with acute trauma. J Trauma. 1992;33504- 507
Link to Article
Berger  MMMarazzi  AFreeman  JChiolero  R Evaluation of the consistency of Acute Physiology and Chronic Health Evaluation (APACHE II) scoring in a surgical intensive care unit. Crit Care Med. 1992;201681- 1687
Link to Article
Fedullo  AJSwinburne  AJWahl  GWBixby  KR APACHE II score and mortality in respiratory failure due to cardiogenic pulmonary edema. Crit Care Med. 1988;161218- 1221
Link to Article
Becker  RBZimmerman  JE ICU scoring systems allow prediction of patient outcomes and comparison of ICU performance. Crit Care Clin. 1996;12503- 514
Link to Article
Knaus  WAWagner  DPDraper  EA  et al.  The APACHE II prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991;1001619- 1636
Link to Article
Pollack  MMPatel  KMRuttimann  UCuerdon  T Frequency of variable measurement in 16 pediatric intensive care units: influence on accuracy and potential for bias in severity of illness assessment. Crit Care Med. 1996;2474- 77
Link to Article
Bratton  SLJardine  DSMorray  JP Serial neurologic examinations after near drowning and outcome. Arch Pediatr Adolesc Med. 1994;148167- 170
Link to Article
Civetta  JM The clinical limitations of ICU scoring systems. Probl Crit Care. 1989;3681- 695
van Veen  AKarstens  Avan der Hock  CJTibboel  DHahlen  C The prognosis of oncologic patients in the pediatric intensive care unit. Intensive Care Med. 1996;22237- 241
Link to Article
Zobel  GRodl  SRigler  B  et al.  Prospective evaluation of clinical scoring systems in infants and children with cardiopulmonary insufficiency after cardiac surgery. J Cardiovasc Surg (Torino). 1993;34333- 337
Madagame  ETHavens  PLBresnahan  JMBabel  KLSplaingard  ML Survival and functional outcome of children requiring mechanical ventilation during therapy for acute bacterial meningitis. Crit Care Med. 1995;231279- 1283
Link to Article
Sherk  JPShatney  CH ICU scoring systems do not allow prediction of patient outcomes or comparison of ICU performance. Crit Care Clin. 1996;12515- 523
Link to Article
Ruttimann  UEPollack  MMFiser  DH Prediction of three outcome states from pediatric intensive care. Crit Care Med. 1996;2478- 85
Link to Article
Selker  HPGriffith  JLD'Agostino  RB A tool for judging coronary care unit admission appropriateness, valid for both real-time and retrospective use: a time-insensitive predictive instrument (TIPI) for acute cardiac ischemia: a multicenter study. Med Care. 1991;29610- 627
Link to Article
Christensen  DWJansen  PPerkin  RM Outcome and acute care hospital costs after warm water near drowning in children. Pediatrics. 1997;99715- 721

Figures

Tables

Table Graphic Jump LocationTable 2. Pediatric Cerebral Performance Category Scale*
Table Graphic Jump LocationTable 3. Characteristics and Outcome of Patients Presenting to ED and PICU*
Table Graphic Jump LocationTable 4. Prehospital Characteristics at Scene*
Table Graphic Jump LocationTable 5. PRISM Score and Probability of Outcome vs Outcome of ED and PICU Patients*

References

Orlowski  JP Drowning, near-drowning, and ice-water drowning. JAMA. 1988;260390- 391
Link to Article
Peterson  B Morbidity of childhood near-drowning. Pediatrics. 1977;59364- 370
Quan  LKinder  D Pediatric submersions: prehospital predictors of outcome. Pediatrics. 1992;90909- 913
Quan  LWentz  KRGore  EJCopass  MK Outcome and predictors of outcome in pediatric submersion victims receiving prehospital care in King County, Washington. Pediatrics. 1990;86586- 593
Biggart  MJBohn  DJ Effect of hypothermia and cardiac arrest on outcome of near-drowning accidents in children. J Pediatr. 1990;117179- 183
Link to Article
Jacobsen  WKMason  LJBriggs  BASchneider  SThompson  JC Correlation of spontaneous respiration and neurologic damage in near-drowning. Crit Care Med. 1983;11487- 489
Link to Article
Dean  JMKaufman  ND Prognostic indicators in pediatric near-drowning: the Glasgow Coma Scale. Crit Care Med. 1981;9536- 539
Orlowski  JP Prognostic factors in pediatric cases of drowning and near-drowning. JACEP. 1979;8176- 179
Link to Article
Romano  CBrown  TFrewen  TC Assessment of pediatric near-drowning victims: is there a role for cranial CT? Pediatr Radiol. 1993;23261- 263
Link to Article
Nussbaum  EGalant  SP Intracranial pressure monitoring as a guide to prognosis in the nearly drowned, severely comatose child. J Pediatr. 1983;102215- 218
Link to Article
Connors  RFrewen  TCKissoon  N  et al.  Relationship of cross-brain oxygen content difference, cerebral blood flow and metabolic rate to neurologic outcome after near drowning. J Pediatr. 1992;121839- 844
Link to Article
Fisher  BPeterson  BHicks  G Use of brainstem auditory-evoked response testing to assess neurologic outcome following near drowning in children. Crit Care Med. 1992;20578- 585
Link to Article
Graf  WDCummings  PQuan  LBrutocao  D Predicting outcome in pediatric submersion victims. Ann Emerg Med. 1995;26312- 319
Link to Article
Pollack  MMRuttimann  UEGetson  PR Pediatric Risk of Mortality (PRISM) Score. Crit Care Med. 1988;161110- 1116
Link to Article
Conn  AWEdmonds  JFBarker  GA Near-drowning in cold fresh water: current treatment regimen. Anaesth Soc J. 1978;25259- 265
Link to Article
Fiser  DH Assessing the outcome of pediatric intensive care. J Pediatr. 1992;12168- 74
Link to Article
Selker  HP Systems for comparing actual and predicted mortality rates: characteristics to promote cooperation in improving hospital care. Ann Intern Med. 1993;118820- 822
Boyd  OGrounds  RM Physiological scoring systems and audit. Lancet. 1993;3411573- 1574
Link to Article
Vassar  MJWilkerson  CLDuran  PJPerry  CAHolcroft  JW Comparison of APACHE II, TRISS, and a proposed 24-ICU point system for prediction of outcome in ICU-treated patients. J Trauma. 1992;32490- 500
Link to Article
McAnena  OJMoore  FAMoore  EEMattox  KLMarx  JAPepe  P Invalidation of the APACHE II scoring system for patients with acute trauma. J Trauma. 1992;33504- 507
Link to Article
Berger  MMMarazzi  AFreeman  JChiolero  R Evaluation of the consistency of Acute Physiology and Chronic Health Evaluation (APACHE II) scoring in a surgical intensive care unit. Crit Care Med. 1992;201681- 1687
Link to Article
Fedullo  AJSwinburne  AJWahl  GWBixby  KR APACHE II score and mortality in respiratory failure due to cardiogenic pulmonary edema. Crit Care Med. 1988;161218- 1221
Link to Article
Becker  RBZimmerman  JE ICU scoring systems allow prediction of patient outcomes and comparison of ICU performance. Crit Care Clin. 1996;12503- 514
Link to Article
Knaus  WAWagner  DPDraper  EA  et al.  The APACHE II prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991;1001619- 1636
Link to Article
Pollack  MMPatel  KMRuttimann  UCuerdon  T Frequency of variable measurement in 16 pediatric intensive care units: influence on accuracy and potential for bias in severity of illness assessment. Crit Care Med. 1996;2474- 77
Link to Article
Bratton  SLJardine  DSMorray  JP Serial neurologic examinations after near drowning and outcome. Arch Pediatr Adolesc Med. 1994;148167- 170
Link to Article
Civetta  JM The clinical limitations of ICU scoring systems. Probl Crit Care. 1989;3681- 695
van Veen  AKarstens  Avan der Hock  CJTibboel  DHahlen  C The prognosis of oncologic patients in the pediatric intensive care unit. Intensive Care Med. 1996;22237- 241
Link to Article
Zobel  GRodl  SRigler  B  et al.  Prospective evaluation of clinical scoring systems in infants and children with cardiopulmonary insufficiency after cardiac surgery. J Cardiovasc Surg (Torino). 1993;34333- 337
Madagame  ETHavens  PLBresnahan  JMBabel  KLSplaingard  ML Survival and functional outcome of children requiring mechanical ventilation during therapy for acute bacterial meningitis. Crit Care Med. 1995;231279- 1283
Link to Article
Sherk  JPShatney  CH ICU scoring systems do not allow prediction of patient outcomes or comparison of ICU performance. Crit Care Clin. 1996;12515- 523
Link to Article
Ruttimann  UEPollack  MMFiser  DH Prediction of three outcome states from pediatric intensive care. Crit Care Med. 1996;2478- 85
Link to Article
Selker  HPGriffith  JLD'Agostino  RB A tool for judging coronary care unit admission appropriateness, valid for both real-time and retrospective use: a time-insensitive predictive instrument (TIPI) for acute cardiac ischemia: a multicenter study. Med Care. 1991;29610- 627
Link to Article
Christensen  DWJansen  PPerkin  RM Outcome and acute care hospital costs after warm water near drowning in children. Pediatrics. 1997;99715- 721

Correspondence

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Submit a Comment

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Topics
PubMed Articles