Published online June 1, 2006
PEDIATRICS Vol. 117 No. 6 June 2006, pp. 1893-1900 (doi:10.1542/peds.2005-2179)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Odetola, F. O.
Right arrow Articles by Davis, M. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Odetola, F. O.
Right arrow Articles by Davis, M. M.
Related Collections
Right arrow Neurology & Psychiatry
Right arrowRelated AAP Red Book topics:
Non-Group A or B Streptococcal and...
Pneumococcal Infections

Variation in the Use of Intracranial-Pressure Monitoring and Mortality in Critically Ill Children With Meningitis in the United States

Folafoluwa O. Odetola, MD, MPHa, John M. Tilford, PhDb and Matthew M. Davis, MD, MAPPc

a Department of Pediatrics and Communicable Diseases, Division of Pediatric Critical Care Medicine, and the Child Health Evaluation and Research Unit, University of Michigan, Ann Arbor, Michigan
b Department of Pediatrics, College of Medicine, University of Arkansas for Medical Sciences and Arkansas Children's Hospital, Little Rock, Arkansas
c Division of General Internal Medicine and the Child Health Evaluation and Research Unit, University of Michigan Health System, and Gerald R. Ford School of Public Policy, University of Michigan, Ann Arbor, Michigan


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
OBJECTIVE. Our goal was to describe patient and hospital characteristics associated with the use of intracranial pressure monitors and outcomes in critically ill children with meningitis.

METHODS. This was a retrospective cohort study of children 0 to 17 years of age hospitalized with meningitis and requiring mechanical ventilation using the 1997 and 2000 Kids' Inpatient Database. We generated national estimates of rates of intracranial pressure monitoring and in-hospital mortality by patient and hospital characteristics, and compared in-hospital mortality, hospital length of stay, and total charges for children who received an intracranial pressure monitor with those who did not.

RESULTS. There were an estimated 1067 and 1170 hospitalizations nationally for childhood meningitis requiring mechanical ventilation in 1997 and 2000, respectively. Most (79%) of the hospitalizations involved infants. Overall, intracranial-pressure monitors were used in 7% of hospitalizations for meningitis, with the highest rates in children aged 5 to 17 years and lowest rates in children <1 year. In-hospital mortality was 19.6%, highest in children aged 5 to 17 years and in children with pneumococcal infections. In multivariate regression analyses, intracranial pressure monitor use was positively associated with age, patient volume, and hospitals located in the West census region. In-hospital mortality was associated with increasing age, hospitalization in the year 2000, self-pay/other insurance status, and pneumococcal meningitis. There was no difference in hospital mortality associated with use of intracranial pressure monitors, but both length of stay and log-transformed total hospital charges were significantly higher in the group that received an intracranial-pressure monitor.

CONCLUSION. Intracranial pressure monitoring for the treatment of critically ill children with meningitis varies by census region, the number of cases treated, and patient age. The use of intracranial pressure monitoring was not statistically associated with mortality in this national sample.

Key Words: meningitis • intracranial pressure • critical illness • hospitalization • childhood

Abbreviations: ICP—intracranial pressure • KID—Kids' Inpatient Database • ICD-9-CM—International Classification of Diseases—Ninth Revision—Clinical Modification • NOS—not otherwise specified • CI—confidence interval • OR—odds ratio

Childhood meningitis is associated with significant morbidity and mortality.1,2 Intracranial pressure (ICP) is a crucial determinant of cerebral perfusion and is frequently elevated in patients with clinical meningitis, with the maximal elevation occurring within the first 24 to 48 hours after diagnosis.3,4 Elevated ICP, which causes compression of the brainstem and impairment of cerebral circulation, is an important cause of neurologic damage and mortality in patients with meningitis.5–7

ICP monitoring may permit physicians to target specific therapies to reduce ICP and improve brain perfusion in patients with intracranial hypertension.4–13 Prior studies, predominantly prospective in design, have reported reduced mortality and morbidity associated with ICP-monitor use in childhood meningitis, where the degree of intracranial hypertension is closely correlated with patient morbidity and mortality.6–13 Most of these studies were hampered by small sample sizes and were not intended to test the association of ICP monitoring with outcomes. The use of ICP monitors, as reported in these studies, was prompted largely by clinical findings suggestive of intracranial hypertension and not by radiologic imaging of the brain, because earlier reports had indicated poor correlation between findings on imaging of the brain and the actual presence of intracranial hypertension.6,9–11,14

Despite reported beneficial effects of ICP monitoring and management of intracranial hypertension in childhood meningitis, the indications for ICP-monitor use in meningitis remain unclear, and the association of the use of ICP monitors with outcome in childhood meningitis is also largely undefined. No guidelines presently exist for the use of ICP monitors in patients with meningitis, and patterns of ICP monitoring in children with meningitis are unknown. This study was conducted to describe patient and hospital characteristics associated with the use of ICP monitors and to describe outcomes in critically ill children with meningitis in the United States using a propensity score–based matching algorithm.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Study Design
We conducted a retrospective cohort study of children 0 to 17 years of age hospitalized with meningitis who also received mechanical ventilation using the Kids' Inpatient Database (KID).15,16 The KID, developed by the Agency for Healthcare Research and Quality, includes >2 million pediatric discharge records obtained from 22 states in 1997 and 27 states in 2000, drawn from the 4 major US census regions. The KID is the only national, all-payer database of hospitalizations for children. The database contains 80% of the normal nonnewborn discharges from these states and is nationally representative with the inclusion of discharge weights in analyses. Because of its large size, the KID is the premier database to study hospitalizations involving relatively rare pediatric diseases. Using the database, several studies have been conducted recently to describe national patterns of hospitalization and health care resource use for a variety of pediatric diseases, which would not have been otherwise possible.17–21 Information on patient demographics, hospital characteristics, and diagnosis/procedure codes is included for each hospitalization. Hospital characteristics include total bed size, teaching status, children's hospital status, and regional location as reported by publications available from Agency for Healthcare Research and Quality.15,16

We generated estimates of national rates of ICP monitoring and in-hospital mortality by patient and hospital characteristics. Stepwise logistic-regression analysis was used to identify factors associated with the use of ICP monitoring and in-hospital mortality. We also used a propensity score–based matching algorithm to compare in-hospital mortality, hospital length of stay, and total charges for children who received an ICP monitor with those who did not receive a monitor.

Sample Identification
Children with meningitis were defined using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes for a primary or secondary diagnosis of meningitis (320.0–320.3, 320.7–320.9, 321.0–321.4, 321.8, 322.0–322.2, and 322.9), including bacterial, viral, and fungal meningitis. In addition, we limited the sample to critically ill children by requiring evidence of an ICD-9-CM procedure code for mechanical ventilation using codes 93.90–93.92 and 96.70–96.72. We excluded patients with traumatic brain injury, because guidelines already exist for the use of ICP monitors in those patients.22 Hospitalizations with ICD-9-CM codes for ventriculoperitoneal shunts and other in-dwelling shunts were also excluded to ensure that patients within the study sample had ICP monitoring only during the index hospitalization. Pretransfer hospitalizations, for which the ultimate case disposition was unknown, were excluded, because including pretransfer hospitalizations in the analysis would overestimate the incidence of hospitalizations for meningitis and underestimate the rates of ICP-monitor use and mortality.23 ICD-9-CM procedure codes 01.18 and 02.2 were used to identify children that received an ICP monitor during the hospitalization.

We generated additional information on the etiology of meningitis using the ICD-9-CM diagnosis codes. Codes for pneumococcal, streptococcal, staphylococcal, Gram-negative, and meningitis not otherwise specified (NOS) infections were used to create separate categories. Infections with cell sizes too small for precise estimation, such as fungal, other specific bacterial, other bacterial NOS, and other bacterial diseases, were included in a final category described simply as "other etiology." Comorbidities associated with increased likelihood for the acquisition of meningitis, such as premature birth, splenic dysfunction, cancer, human immunodeficiency virus infection, and immunosuppression from organ or bone marrow transplant, were coded, but only premature birth had an adequate sample size. Finally, a derived "volume" variable was created to describe the number of cases of childhood meningitis per hospital as a measure of the patient caseload within the individual hospitals.24

The main outcomes for the study were in-hospital mortality, patient length of stay, and total hospital charges. These outcomes were assessed in relation to the use of ICP monitoring. The University of Michigan Medical School Institutional Review Board approved the study.

Statistical Analysis
The statistical analysis was aimed first at identifying factors associated with the use of ICP monitoring and in-hospital mortality and, second, to assess whether the use of ICP monitoring was statistically associated with outcomes. Using data pooled from 1997 and 2000, we calculated the number of hospitalizations and the mean rates of ICP monitoring and in-hospital mortality (with their associated 95% confidence intervals [CIs]) according to both patient and hospital characteristics. All of the estimates used the survey commands in Stata for Windows (Stata Corp, College Station, TX), which accounted for the complex survey design. Logistic-regression models for complex survey data were fit to estimate factors associated with the use of ICP monitoring and in-hospital mortality using a backward selection method. Finally, all of the variables were again tested for possible inclusion in the final model. Robust variance estimates were obtained by computing finite population corrections for random sampling without replacement of primary sampling units. Variance estimates accounted for clustering of data at the hospital level by using hospital identifiers as the primary sampling units.

To assess whether use of ICP monitoring was associated with improved outcomes, a propensity score–based matching algorithm for nearest neighbors was created using the pscore suite of programs developed for Stata.25 Propensity analysis aims to identify patients with a similar probability of receiving ICP monitors on the basis of observed clinical characteristics and, thus, helps to adjust for the bias inherent in the decision to treat childhood meningitis with the use of ICP monitors or not.26,27 The propensity scores were created using all of the potential confounders, including age, gender, etiology, and hospital characteristics, and restricting the score to the common support area. We then generated a balanced sample of patients with an ICP monitor and patients without an ICP monitor using the propensity score to identify the nearest neighbors. This algorithm restricted the sample to patients in the common support area. The sample created from matching on nearest neighbors was then used to conduct additional analyses on the main outcome measures using multivariate regression models controlling for potential confounders, including the propensity scores. In these analyses, the independent effect of ICP monitoring on the main outcome measures was our primary interest. We used multivariate logistic regression, negative binomial regression, and multiple linear regression models to assess differences in mortality, length of hospitalization, and (log-transformed) total charges, respectively. Because of the small sample size in the matched sample, analyses conducted with the matched sample do not account for the complex survey design of the KID.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
There were 1067 and 1170 hospitalizations nationally for childhood meningitis requiring mechanical ventilation in 1997 and 2000, respectively. Nearly 80% of the hospitalizations involved infants (<1 year old). Overall, 157 children, or 7% of the study population, received an ICP monitor. The in-hospital mortality rate was 19%, with no statistically significant difference in mortality rates between the 2 years.

For both use of ICP monitors and in-hospital mortality, age was associated with large increases in rates (Table 1). The highest rates of ICP monitoring occurred in children aged 5 to 17 years and lowest in children <1 year of age. In-hospital mortality was highest in children aged 5 to 17 years (34.3%; 95% CI: 28.0–40.6%). There were some differences in the rate of ICP monitoring by etiology of meningitis with streptococcal (nonpneumococcal) infections monitored less often (3.0%; 95% CI: 1.3–4.7%) relative to staphylococcal infections (12.9%; 95% CI: 6.0–19.9%). There were also differences in mortality rates by etiology with pneumococcal infections having the highest rate of mortality (34.8%; 95% CI: 29.2–40.5%). The 95% CIs overlapped for other patient characteristics, such as gender, race, payer status, prematurity, and year.


View this table:
[in this window]
[in a new window]
  		TABLE 1 Use of ICP Monitors and Mortality According to Patient Characteristics

 
The majority of the hospitalizations (61.5%) occurred in children's hospitals with teaching programs (Table 2). ICP-monitor use differed by patient volume, with the hospitals seeing < 5 critically ill children with meningitis having use rates of ~5% and hospitals seeing ≥5 children having higher rates of ~10%. There also were significant variations by census region, with the West having the highest rates of ICP monitoring (12.1%; 95% CI: 7.4–16.8%) relative to the South (4.3%; 95% CI: 2.8–5.8%) and the Northeast (5.2%; 95% CI: 3.9–6.5%). There was no evidence that greater frequency of monitoring was statistically associated with mortality.


View this table:
[in this window]
[in a new window]
  		TABLE 2 Use of ICP Monitors and Mortality According to Hospital Characteristics

 
In multivariate analyses, age, West census region, and patient volume were associated with increased use of ICP monitors (Table 3). The odds of ICP-monitor use for childhood meningitis increased 15% for each 1-year increment in age. In addition, there was more than twofold higher odds of ICP-monitor use in the West census region compared with other regions, while controlling for other factors. Of note, whereas 53% of the study population had either streptococcal (nonpneumococcal) meningitis or meningitis NOS, the odds of having an ICP monitor placed was lower for these patients when compared with other etiologic categories.


View this table:
[in this window]
[in a new window]
  		TABLE 3 Patient and Hospital Characteristics Associated With the Use of ICP Monitoring in a Multivariate Model

 
Factors associated with increased mortality differed from the factors associated with increased use of ICP monitors, with the exception of child age (Table 4). Increasing child age was associated with increased risk of in-hospital mortality, with an increase of 5% for every 1-year increase in child age. Despite much greater frequency of use of ICP monitors in the Western United States, census region was not associated with increased mortality risk after controlling for other factors. Pneumococcal infections, self/other insurance status, and hospitalization in the year 2000 were associated with higher odds of mortality, whereas premature birth and hospitalization in a nonteaching nonchildren's hospital were associated with lower odds of mortality.


View this table:
[in this window]
[in a new window]
  		TABLE 4 Patient and Hospital Characteristics Associated With In-Hospital Mortality in a Multivariate Model

 
Because selection bias may confound the potential for mortality rates in different parts of the country to be influenced by the use of ICP monitoring, we developed a propensity score for the use of ICP monitoring and generated a balanced sample of patients with ICP monitors and those without. Table 5 depicts mortality rates by selected characteristics for 73 children who received an ICP monitor and 73 children who did not. Sixty three (86%) of the 73 children with ICP monitors had ventriculostomy catheters placed. These devices might serve as a conduit for drainage of cerebrospinal fluid in the management of intracranial hypertension.


View this table:
[in this window]
[in a new window]
  		TABLE 5 Characteristics and In-Hospital Mortality for Children Included in Propensity Score–Based Matched Sample According to Use of ICP Monitoring

 
Despite satisfying balancing criteria across observed covariates, it is still possible that the samples might differ in important ways. Thus, we tested whether outcomes differed by ICP-monitoring status in multivariate models that included the propensity score (Table 6). There was no difference in the odds of mortality between the 2 groups (odds ratio [OR]: 1.00; 95% CI: 0.46–2.18). However, length of stay and log-transformed total charges were higher in the monitored group versus the nonmonitored group. Both of these latter models included an indicator variable for whether the patient died or not because of the differential impact of in-hospital death on patient length of stay and total charges.


View this table:
[in this window]
[in a new window]
  		TABLE 6 Association of ICP Monitoring With Mortality and Resource Use From Multivariate Models Using Propensity Score–Based Matched Sample

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Despite >5 decades of ICP-monitor use in clinical practice, little consensus exists on the use of these monitors in patients with brain injury and resultant intracranial hypertension triggered by a variety of causes. Guidelines have been published recently regarding the use of ICP monitors in children with traumatic brain injury, with recommendations for their use only in the care of children with severe (Glasgow Coma Score ≤8) but not moderate or mild traumatic brain injury.22 Recent studies of critically ill patients with meningitis who received targeted treatment to reduce acute intracranial hypertension as recorded by continuous ICP-monitor use have described improved morbidity and mortality.4–11,13

In this report on the use of ICP monitors and outcomes in critically ill children with meningitis, we found significant variation in the use of ICP monitors by age of the patient, census region, infectious etiology, and institutional patient volume, without statistically significant differences in mortality. The observed variation by age in the use of ICP monitors for the management of meningitis; however, it runs counter to the epidemiology of meningitis in the study cohort. Whereas ~80% of the study population was <1 year of age, ICP-monitor use occurred mainly among older children 5 to 17 years old. This finding could be related to the reluctance to place ICP monitors in infants because of the belief that open anterior fontanelles could act as an outlet for relief of intracranial hypertension, thereby obviating the need for ICP-monitor use. A bulging anterior fontanelle on clinical examination of an infant with meningitis is universally accepted as a sign of intracranial hypertension prompting immediate management, therefore reducing the likelihood of invasive monitoring of ICP. Infants have been reported to experience intracranial hypertension, however, and it is not known how reliable the clinical examination of the turgor of the anterior fontanelle is in predicting the likelihood of elevated ICP. There also might be increased operator level of comfort placing the monitors in older children and adolescents, compared with infants. Our finding of age variation in the use of ICP monitors with lower likelihood of use among infants corroborates findings from another recent study that described the frequency of ICP monitoring in infants and young toddlers with traumatic brain injury. The authors found that ICP was much less frequently monitored in infants (<1 year) compared with older children 1 to 2 years of age.28 The poor correlation of brain-computed tomography scans with intracranial hypertension6,9–11,14 and the uncommon occurrence of papilledema as a finding on clinical examination in patients with intracranial hypertension4,13 make the diagnosis of raised ICP in children with closed anterior fontanelles often difficult and may lower the threshold for ICP-monitor use in older children with meningitis.

We found differential use of ICP monitoring according to the etiology of the organism. Use of ICP monitoring was higher for pneumococcal, staphylococcal, and Gram-negative organisms compared with nonpneumococcal streptococci and meningitis NOS, despite the latter 2 etiologic categories accounting for the majority of admissions in the sample. The reasons for this finding are unknown.

The reasons for the regional variation in ICP-monitor use that we observed are also unclear. Regional practice variation has been described, often occurring without basis in clinical science or explanation by disease prevalence or illness severity.29 Our finding may reflect variation in physician practice styles originating from clinical mentorship and instruction during residency or fellowship training. The database had no information on physician characteristics (including training and expertise), physician supply, and the availability of ICP-monitoring technology, all of which could contribute to the regional variation described. Certain authors have speculated that the acute management of raised ICP in children, regardless of the etiology, varies widely because of a lack of both evidence and consensus.30 Professional controversies arising from incomplete or ambiguous scientific evidence on the value of specific services have been proposed as a reason for the variation in use rates of medical services.31 Additional study is warranted to elucidate the impact of regional variation in ICP-monitor use on health outcomes for children with meningitis. It was instructive to find that the monitors were placed more often within hospitals with higher patient volume, possibly reflecting greater experience with the placement and subsequent management of these devices.

The factors associated with mortality differed from those associated with the use of ICP monitors. Mortality was highest among children with pneumococcal meningitis, a finding that mirrors the current epidemiology of childhood meningitis.32,33 Delayed presentation for care in childhood meningitis has been associated with poor outcomes.34 It is unknown whether the higher mortality among children with self/other pay compared with other insurance types might be as a result of delayed presentation to the point of definitive care. Mortality was also highest among children hospitalized at teaching (children's and nonchildren's) hospitals where ICP monitors were also likely to be placed. This finding may reflect the admission of sicker children to these institutions. This speculation, however, warrants additional investigation.

The observed significant variation in the use of ICP monitoring was not, however, associated with lower mortality for patients in whom the monitors were used. The finding from the propensity-based analysis still may be the artifact of an observational study, and we were limited by our sample in operationalizing the propensity score method. Prior studies report greater use of invasive devices in critically ill children who eventually do not survive pediatric intensive care.35 Prospective studies, specifically, randomized clinical trials, which control for unobserved differences in the data, are often needed to ascertain a causal relationship when assessing the use of medical devices. This approach in the assessment of the use of ICP monitors for childhood meningitis is likely to be ethically challenging and not practically feasible to execute. Nonetheless, our study findings cannot be interpreted as either justifying or failing to justify the use of ICP monitors in the management of childhood meningitis on the basis of mortality benefit.

This study has other limitations. Because of the retrospective nature of the study, we could not ascertain the actual indication for ICP-monitor use in the study population. We excluded patients with head trauma or those with existing in-dwelling ventriculoperitoneal shunts, which limited the sample size but strengthened the internal and external validity of the study and reduced potential confounding of our results by these factors.

Variation in ICP-monitor use and outcomes could be explained by differences in health status, but we had limited ability to adjust for severity of illness in the study population beyond selecting for children who received mechanical ventilation. In a prior study of outcomes of a large cohort of children admitted to pediatric intensive care units for treatment of meningitis, however, the authors reported significant interinstitutional variation in the use of ICP monitors poorly correlated with coma status, a marker of illness severity.35


   CONCLUSIONS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
The use of ICP monitoring in childhood meningitis was found to vary according to patient and hospital characteristics using a hospital discharge database weighted to reflect a representative sample of discharges from the United States. Findings from a propensity-matched sample revealed no statistical association between the use of ICP monitors for meningitis and mortality. The study findings cannot be used to definitively recommend for or against the use of ICP monitors in the clinical setting.

The relative rarity of childhood meningitis requiring mechanical ventilation may preclude the ability to conduct randomized trials to test the efficacy of ICP monitoring in this population. With recent reports of improved outcomes in childhood meningitis with ICP-monitor use, it may be possible to design prospective studies using matched cohorts with incorporation of pertinent clinical information. This study illustrates the potential for using propensity score–based matching to test the efficacy of ICP monitoring in ongoing observational studies of childhood meningitis in centers that vary widely in their use of this device.


   FOOTNOTES
 
Accepted Nov 30, 2005.

Address correspondence to Folafoluwa O. Odetola, MD, MPH, Department of Pediatrics and Communicable Diseases, Division of Pediatric Critical Care Medicine, 6E07, 300 N Ingalls St, University of Michigan, Ann Arbor, MI 48109. E-mail: fodetola{at}med.umich.edu

This work was presented in part at the annual meeting of the Pediatric Academic Society; May 1–4, 2004; San Francisco, CA.

The authors have indicated they have no financial relationships relevant to this article to disclose.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Wald E, Bergman I, Taylor G, Chiponis D, Porter C, Kubek K. Long-term outcome of group B streptococcal meningitis. Pediatrics. 1986;77 :217 –221[Abstract/Free Full Text]
  2. Sell S. Long term sequelae of bacterial meningitis in children. Pediatr Infect Dis J. 1983;2 :90 –93
  3. Goitien KJ, Tamir I. Cerebral perfusion pressure in central nervous system infections in infancy and childhood. J Pediatr. 1983;103 :40 –43[CrossRef][ISI][Medline]
  4. Minns RA, Engleman HM, Stirling H. Cerebrospinal fluid pressure in pyogenic meningitis. Arch Dis Child. 1989;64 :814 –820[Abstract]
  5. Goitein KJ, Amit Y, Musaffi H. Intracranial pressure in central nervous system infections and cerebral ischemia of infancy. Arch Dis Child. 1983;58 :184 –186[Abstract]
  6. Lindvall P, Ahlm C, Ericsson M, Gothefors L, Naredi S, Koskinen L-OD. Reducing intracranial pressure may increase survival among patients with bacterial meningitis. Clin Infect Dis. 2004;38 :384 –390[CrossRef][ISI][Medline]
  7. Rebaud P, Berthier JC, Hartemann E, Floret D. Intracranial pressure in childhood central nervous system infections. Intensive Care Med. 1988;14 :522 –525[CrossRef][ISI][Medline]
  8. Pesso JL, Floret D, Cochat P, Dumont C. Suppurated pneumococcal meningitis in infants and children: complications and prognostic factors [in French]. Pediatrie. 1988;43 :263 –267[ISI][Medline]
  9. Grande PO, Myhre EB, Nordstrom CH, Schliamser S. Treatment of intracranial hypertension and aspects on lumbar dural puncture in severe bacterial meningitis. Acta Anaesthesiol Scand. 2002;46 :264 –270[CrossRef][ISI][Medline]
  10. Rennick G, Shann F, de Campo J. Cerebral herniation during bacterial meningitis in children. BMJ. 1993;306 :953 –955[ISI][Medline]
  11. Ibsen LM. Radiological case of the month. Arch Pediatr Adolesc Med. 2002;156 :293 –294[Free Full Text]
  12. Warren HS Jr, Gonzalez RG, Tian D. Case records of the Massachusetts General Hospital: weekly clinicopathological exercises. Case 38–2003: a 12-year-old girl with fever and coma. N Engl J Med. 2003;349 :2341 –2349[Free Full Text]
  13. Nugent SK, Bausher JA, Moxon ER, Rogers MC. Raised intracranial pressure: Its management in Neisseria meningitides meningoencephalitis. Am J Dis Child. 1979;133 :260 –262[Abstract]
  14. Schoeman JF, Le Roux D, Bezuidenhout PB, Donald PR. Intracranial pressure monitoring in tuberculous meningitis: clinical and computerized tomographic correlation. Dev Med Child Neurol. 1985;27 :644 –654[ISI][Medline]
  15. Healthcare Cost and Utilization Project. Kids' Inpatient Database, 1997 (CD-ROMs). Rockville, MD: Agency for Healthcare Research and Quality; 2000
  16. Healthcare Cost and Utilization Project. Kids' Inpatient Database, 2000 [CD-ROMs]. Rockville, MD: Agency for Healthcare Research and Quality; 2002
  17. Reynolds MG, Holman RC, Curns AT, O'Reilly M, McQuiston JH, Steiner CA. Epidemiology of cat-scratch disease hospitalizations among children in the United States. Pediatr Infect Dis J. 2005;24 :700 –704[CrossRef][ISI][Medline]
  18. Holman RC, Curns AT, Belay ED, Steiner CA, Schonberger LB. Kawasaki syndrome hospitalizations in the United States, 1997 and 2000. Pediatrics. 2003;112 :495 –501[Abstract/Free Full Text]
  19. Killingsworth JB, Tilford JM, Parker JG, Graham JJ, Dick RM, Aitken ME. National hospitalization impact of pediatric all-terrain vehicle injuries. Pediatrics. 2005;115 (3). Available at: www.pediatrics.org/cgi/content/full/115/3/e316
  20. Guthery SL, Hutchings C, Dean JM, Hoff C. National estimates of hospital utilization by children with gastrointestinal disorders: analysis of the 1997 kids' inpatient database. J Pediatr. 2004;144 :589 –594[CrossRef][ISI][Medline]
  21. Lewis CW, Carron JD, Perkins JA, Sie KC, Feudtner C. Tracheotomy in pediatric patients: a national perspective. Arch Otolaryngol Head Neck Surg. 2003;129 :523 –529[Abstract/Free Full Text]
  22. Adelson PD, Bratton SL, Carney NA, et al. Guidelines for the acute medical management of severe traumatic brain injury in infants, children, and adolescents. Chapter 5. Indications for intracranial pressure monitoring in pediatric patients with severe traumatic brain injury. Pediatr Crit Care Med. 2003;4 :S19 –S24[CrossRef][Medline]
  23. Westfall JM, McGloin J. Impact of double counting and transfer bias on estimated rates and outcomes of acute myocardial infarction. Med Care. 2001;39 :459 –468[CrossRef][ISI][Medline]
  24. Tilford JM, Simpson PM, Green JW, Lensing S, Fiser DH. Volume-outcome relationships in pediatric intensive care units. Pediatrics. 2000;106 :289 –294[Abstract/Free Full Text]
  25. Becker SO, Ichino A. Estimation of average treatment effects based on propensity scores. Stata J. 2002;4 :358 –377
  26. Rosenbaum P, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70 :41 –55[Abstract/Free Full Text]
  27. Rubin DB. Estimating causal effects from large data sets using propensity scores. Ann Intern Med. 1997;127 :757 –763[Abstract/Free Full Text]
  28. Keenan HT, Nocera M, Bratton SL. Frequency of intracranial pressure monitoring in infants and young toddlers with traumatic brain injury. Pediatr Crit Care Med. 2005;6 :537 –541[CrossRef][Medline]
  29. Wennberg JE. Unwarranted variations in healthcare delivery: implications for academic medical centers. BMJ. 2002;325 :961 –964[Free Full Text]
  30. Segal S, Gallagher AC, Shefler AG, Crawford S, Richards P. Survey of the use of intracranial pressure monitoring in children in the United Kingdom. Intensive Care Med. 2001;27 :236 –239[CrossRef][ISI][Medline]
  31. Wennberg JE. Dealing with medical practice variations: a proposal for action. Health Aff (Millwood). 1984;3 :6 –32[Medline]
  32. Prober CG. Central nervous system infections. In: Behrman R, ed. Nelson Textbook of Pediatrics. 16th ed. Philadelphia, PA: WB Saunders Co; 2000:756
  33. Grimwood K, Nolan TM, Bond L, Anderson VA, Catroppa C, Keir EH. Risk factors for adverse outcomes of bacterial meningitis. J Paediatr Child Health. 1996;32 :457 –462[ISI][Medline]
  34. Hailu M, Muhe L. Childhood meningitis in a tertiary hospital in Addis Ababa: clinical and epidemiological features. Ethiop Med J. 2001;39 :29 –38[ISI][Medline]
  35. Odetola FO, Bratton SL. Characteristics and immediate outcome of childhood meningitis treated in the pediatric intensive care unit. Intensive Care Med. 2005;1 :92 –97[CrossRef]
PEDIATRICS (ISSN 1098-4275). ©2006 by the American Academy of Pediatrics



This article has been cited by other articles:


Home page
 Arch Pediatr Adolesc MedHome page
J. M. Robbins, T. M. Bird, J. M. Tilford, J. A. Reading, M. A. Cleves, M. E. Aitken, C. M. Druschel, and C. A. Hobbs
Reduction in Newborns With Discharge Coding of In Utero Alcohol Effects in the United States, 1993 to 2002
Arch Pediatr Adolesc Med, December 1, 2006; 160(12): 1224 - 1231.
[Abstract] [Full Text] [PDF]

Home page
 PediatricsHome page
M. A. Fishman
Variation in the use of intracranial-pressure monitoring and mortality in critically ill children with meningitis in the United States.
Pediatrics, June 1, 2006; 117(6): 2279 - 2280.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Odetola, F. O.
Right arrow Articles by Davis, M. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Odetola, F. O.
Right arrow Articles by Davis, M. M.
Related Collections
Right arrow Neurology & Psychiatry
Right arrowRelated AAP Red Book topics:
Non-Group A or B Streptococcal and...
Pneumococcal Infections