All Content

PEDIATRIC SPECIAL SECTION: Children’s National Medical Center’sHospitalist Division


Although pediatric resident graduates are well prepared to care for many ill children who require hospitalization, there are clinical, academic, and administrative skills that aren’t a standard part of residency, but are necessary for long-term academic success. The Children’s National Medical Center (CNMC) Hospitalist Fellowship program is structured over three years, through a combination of formal course work, seminars, clinical work, and mentored independent study.

Fellows work under the supervision of faculty in our 20-member Hospitalist Division, with inpatient pediatric services at five hospitals in Washington D.C., and suburban Maryland. A faculty appointment as a clinical instructor of the George Washington University (GWU) School of Medicine and Health Sciences qualifies the fellow to complete a Masters in Public Health, taking 15-18 credits per year. In addition to specific clinical, research, and didactic learning objectives, participation in committees and quality improvement and epidemiologic projects will facilitate study of hospital administration, budget, epidemiology and infection control.

The fellow must be a successful graduate of an American Council of Graduate Medical Education approved pediatric residency.

Clinical Training

Fellows spend approximately three months during the three year fellowship in the pediatric intensive care unit, two weeks in anesthesia, and two months in the emergency department to gain increased confidence, knowledge, and skill in airway management, placement of central lines, evaluation and management of the severely ill/injured child (respiratory failure and shock), conscious sedation, and pain management.

In addition, fellows spend 12 weeks per year on the CNMC hospitalist service. They co-attend with a CNMC hospitalist and lead a team of residents and students to practice an evidence-based approach to the acute evaluation and management of common inpatient problems such as bronchiolitis; asthma; gastroenteritis and dehydration; pneumonia; UTI; ALTE; common head, neck, skin and musculoskeletal infections; metabolic disorders; HIV infection; rheumatologic disease; and child abuse. Fellows also devote eight weeks per year to studying the process of providing coordinated care for the medically complex/technology dependent patients at CNMC and at rehab/subacute care hospital units that focus on improving care for children with multiorgan system disease. They also learn about complications associated with use of technology. Fellows’ elective clinical time may focus on delivery room management skills and well-baby teaching and management issues at one of our community hospital sites with 8,000 deliveries annually.

Research Training

CNMC fellows learn to plan, implement, analyze, and present results of a research study to answer a question relevant to hospitalists. By the end of the fellowship the fellows have an article accepted for publication in a peer-reviewed journal and present findings at a national meeting. To gain knowledge and skill needed to conduct an independent research project the fellows complete coursework in the School of Public Health at GWU and work under faculty mentors at CNMC.

Didactic Training

Four faculty members of our Hospitalist Division have completed the CNMC Master Teacher Certificate program with the GWU School of Education. In seminars taught by master teachers, fellows learn to lead, teach, and develop educational programs effectively. Faculty observe the fellow’s teaching skills in a variety of settings and provide feedback. Fellows have the opportunity to participate and develop computerized and standardized patient simulations to assess residents’ and students’ ability to care for pediatric inpatients.

Contact Information: Pediatric Hospitalist Fellowship Program

Jennifer Maniscalco, MD

Phone: (202) 884-6123

E-mail: [email protected]


In the Literature

Formula Supplementation Prophylaxis for Jaundice in Breastfed Newborns

Gourley GR, Li Z, Kreamer BL, et al. A controlled, randomized, double-blind trial of prophylaxis against jaundice among breastfed newborns. Pediatrics. 2005;116(2):385-391.

Question: In breastfed newborns does oral beta-glucuronidase inhibitors during the first week after birth increase fecal bilirubin excretion and reduce jaundice without affecting breastfeeding deleteriously?

Design: Randomized (randomization table), double-blinded (supplements were prepared by a third party and labeled by a code) controlled trial.

Setting: Clinic.

Patients: 69 newborns, Caucasian, >37 week of gestation who were exclusively breastfed.

Intervention: Patients were randomly assigned to four groups: breastfeeding only (control), L-aspartic acid supplementation (L-aspartic), enzymatically hydrolyzed casein (EHC), and whey/casein-supplement without beta-glucuronidase inhibitor (W/C).

Main outcome measure: Daily transcutaneous and fecal bilirubin levels were obtained. Long-term follow-up assessments included age at first formula feeding, breastfeeding cessation age, and mother’s subjective impression of study.

Main results: Infants in all supplementation groups had significantly lower transcutaneous bilirubin levels in the first week of life than the nonsupplemented infants. The EHC and W/C groups both excreted more bilirubin diglucuronide and bilirubin monoglucuronide than the control group, and the L-aspartic acid had increased bilirubin monoglucuronide excretion compared with the control group. No negative effects from the supplementations were noted.

Conclusions: There is a significant reduction in transcutaneous bilirubin levels, with no negative effects on breastfeeding through the administration of specific supplements that are currently found in infant formulas.

Conflict of interest: None declared.

Commentary: The study describes a possible method of reducing hyperbilirubinemia associated with breastfeeding that employs small-volume supplementations of ingredients found in common infant formulas. The study did not reveal any negative effects on breastfeeding. Though the study was described as blinded the authors do not comment on how the exclusively breastfed infant group was blinded from the researchers and family.

The other control group (W/C) not fed a supplement without a beta-glucuronidase inhibitor yielded the unexpected significant reduction in transcutaneous bilirubin level compared with the control. This may mean that there is at least one other mechanism for the reduction of bilirubin in formula-fed infants than initially proposed by the authors.

In their conclusion, the authors pose a hypothesis explaining this result. They also correctly point out that due to this small study population, larger studies are needed to confirm these results.

This study used transcutaneous bilirubin measurements raising the question of reliability of this method compared to serum bilirubin measurements. A search of Medline using the terms “bilirubin and transcutaneous” produced a technical report.1 One section of this article reviewed available studies on the accuracy of transcutaneous bilirubin measurements concluding that the transcutaneous measurement has a linear correlation to serum bilirubin and its use for hyperbilirubinemia screening is reliable.


Ip S, Chung M, Kulig J, et al. American Academy of Pediatrics Subcommittee on Hyperbilirubinemia. An evidence-based review of important issues concerning neonatal hyperbilirubinemia. Pediatrics. 2004;Jul;114(1):e130-153.

Overall the study showed that there is no significant difference between montelukast and fluticasone with regard to the primary outcome of RFDs.

Comparing Controllers: Montelukast and Fluticasone in Mild Persistent Asthma

Garcia Garcia ML, Wahn U, Gilles L, et al. Montelukast, compared with fluticasone, for control of asthma among 6- to 14-year-old patients with mild asthma: the MOSAIC study. Pediatrics. 2005;116(2):360-369.

Question: In children with mild asthma is the use of an antileukotriene medication as effective as an anti-inflammatory agent in maintaining number of days without the need of a rescue medication or utilization of asthma-related health resources?

Design: Multicenter, double-blind, double-dummy, randomized, parallel-group study. A four-week, single-blind, placebo run-in period was done prior to the initiation of the double-blind portion of the study. This run-in period included three clinic visits. During this time all participants discontinued any asthma controller medications and received image-matching, single-blind, montelukast and fluticasone. A short acting beta-agonist medication (open label) was also provided. Medication usage teaching was done during this time period. There was no placebo control group.

Setting: 104 study sites in 24 countries.

Patients: Patients (six-14 years of age) with mild persistent asthma as defined by the Global Initiative for Asthma (GINA) guidelines and otherwise in good health. Patients were required to have a FEV1 of ≥80% while the beta-receptor agonist was withheld for more than six hours in the first two visits and ≥70% on the third visit of the run-in period.

Intervention: Patients were randomized to receive either montelukast (5 mg oral before bedtime or 10 mg for patients ≥15) or fluticasone (100 mcg through MDI twice a day) and the appropriate placeboes.

Main outcome measures: Percentage of asthma rescue-free days (RFD).

RFD was defined as a day without:

  1. The use of a rescue medication: defined as any day during which a beta-receptor agonist, a systemic corticosteroid, or other rescue medications were used; and
  2. The use of an asthma health resource: defined as an unscheduled visit to a physician, an urgent/emergency care center or a hospital.

Main results: The number of RFDs were statistically similar between both groups. Mean percentage of RFD was 84% in the montelukast group and 86.7% in the fluticasone group.

Conflict of interest: Two of the study authors were employees of Merck and Company, which markets montelukast.

Commentary: This is an extensive study with multiple outcome measures. The study mimics actual clinical experience by using clinically relevant end points as primary outcomes. The study does not have a placebo control group. Overall the study showed that there is no significant difference between montelukast and fluticasone with regard to the primary outcome of RFDs, but there were some differences favoring fluticasone in secondary outcome measures, including number of days of beta-receptor agonist use and improvement of the average score of the control domain of the Pediatric Asthma Therapy Assessment Questionnaire.

Up to 50% of culture negative cases of septic arthritis in patients less than two years of age may be attributed to K. kingae.

K. Kingae Causes Osteomyelitis/Septic Arthritis

Kiang KM, Ogunmodede F, Juni BA, et al. Outbreak of osteomyelitis/septic arthritis caused by Kingella kingae among child care center attendees. Pediatrics. 2005;116;e206-213.

Review by Julia Simmons, MD

Kingella kingae (K. kingae), a gram-negative organism known to colonize the oropharynx, may cause invasive diseases, such as septic arthritis and osteomyelitis in children who were previously healthy. In this article, the authors describe the first reported outbreak of K. kingae causing osteomyelitis/septic arthritis.

In October 2003 the Minnesota Department of Health was notified of an outbreak of K. kingae at a daycare facility. There were two culture positive cases and one culture negative but clinically presumed to be secondary to K. kingae. The children were all in the same classroom (toddler classroom 1) and were between the ages of 17 and 21 months. The previously healthy children presented with fever and limp.

In this cohort study, the authors investigated risk factors for invasive disease and for colonization with K. kingae. They also determined the prevalence of asymptomatic colonization at the daycare involving the confirmed cases and at a second daycare facility, which served as the control. Culture samples were analyzed at the Minnesota Department of Health Public Health Laboratory. The laboratory performed confirmation, pulsed-field gel electrophoresis, DNA sequencing, and antimicrobial testing specifically with rifampin, penicillin, and azithromycin.

At the study facility 115/122 (94%) children and 28/29 (97%) adult staff members were cultured. Fifteen (13 %) of the children were colonized. The prevalence of colonization was 45% (9 children) in the toddler 1 classroom, which is the classroom of the confirmed cases. The three case patients had negative oropharyngeal cultures; however, all three had been pretreated with antibiotic therapy. The pulsed-field gel electrophoresis was indistinguishable in the confirmed cases and the colonized students. The authors did not discover any risk factors for invasive disease. This information was gathered from a list of questions regarding past medical history, which the parents discussed on the telephone with the investigators. Rifampin, penicillin, and azithromycin all had low MICs

At the control care center, 45/118 (38%) of people were cultured from the oropharynx. Seven (16%) were colonized. The pulsed-field gel electrophoresis was the same for the control care facility, but varied from the electrophoresis pattern at the study facility.

In conclusion, the authors summarized the findings of K. kingae osteomyelitis/septic arthritis in a daycare facility. There were no identifiable risk factors for invasive disease or for asymptomatic carrier state. It is important to remember that up to 50% of culture negative cases of septic arthritis in patients less than two years of age may be attributed to K. kingae. Thus, the clinician must maintain a high index of suspicion and relay this information to the laboratory so that appropriate measures are taken to ensure the optimal growth environment for the bacteria.

Although the study size was small, the investigators provide initial documentation regarding outbreaks of invasive K. kingae infections. Further studies will be useful to clarify the exact mode of transmission and appropriate duration of antimicrobial therapy.

There are insufficient trials of adequate quality to fully elucidate the most efficacious symptomatic pharmacologic therapy for pediatric patients suffering from migraine headaches.

Systematic Review of Therapy for Pediatric Migraine

Damen L, Bruijn JKJ, Verhagen AP, et al. Symptomatic treatment of migraine in children: a systematic review of medication trials. Pediatrics. 2005;116(2):295-302.

Migraines affect 3%-5% of prepubescent children. The symptoms frequently include unilateral throbbing headache associated with nausea and vomiting. If poorly controlled, the headaches can cause students to miss school and have a decreased quality of life. Symptomatic pharmacologic interventions involve analgesics, triptans, and antiemetics. Nonpharmacologic interventions include dietary changes, biofeedback, and exercise. Treatment is tailored to the patient; however, there are limited studies providing evidence describing the most efficacious pharmacologic treatment for managing symptomatic migraines.

The authors conducted a retrospective literature review to evaluate the evidence from randomized controlled trials (RCT) and clinical controlled trials (CCT) regarding the treatment of symptomatic migraine in children younger than 18. The primary outcome measure was headache clinical improvement defined as a headache decline by >50%. Secondary measures included headache intensity, frequency, duration, and headache index. The authors searched Medline, Embase, PsychInfo, Web of Science, and Cinahl from the start of the databases to June 2004. They used appropriate search terms including “headache,” “migraine,” “child” and variations thereof. Further, they searched reference lists of the articles and the included studies. Two authors independently reviewed the titles and abstracts to determine acceptability in the study. They also used the Delphi list to assess the methodologic quality of the included trials.

Of 3,492 relevant studies, 10 met the inclusion criteria specified by the authors. There were a total of 1,575 patients with a mean age of 11.7 +/- 2.2 years included in the study. A mean of 19.8% of subjects did not complete participation in their respective studies. Overall, the percentage of male and female participants was equal. Six studies were considered to be of high quality by the authors. They found moderate evidence that ibuprofen and acetaminophen were more effective in decreasing headache symptoms one and two hours after ingestion in comparison with a placebo. Further, the nasal spray sumatriptan was more efficacious in diminishing headache than placebo; however, it was associated with more side effects. Intravenous prochlorperazine was more effective than intravenous ketorolac in diminishing symptoms one hour after intake.

While accomplishing a sound systematic review of the available controlled trials, the authors conclude that there are insufficient trials of adequate quality to fully elucidate the most efficacious symptomatic pharmacologic therapy for pediatric patients suffering from migraine headaches. A large randomized prospective trial comparing analgesics, triptans, and antiemetics against placebo is needed. TH

Recommended Reading

SHM Elects Jack Percelay, MD, FAAP, to newly Created Pediatrics Seat on Board of Directors
The Hospitalist
Other Pediatric Literature of Interest
The Hospitalist
Other Pediatric Literature of Interest
The Hospitalist
Pediatric in the Literature
The Hospitalist
SHM Elects Jack Percelay, MD, FAAP, to Newly Created Pediatrics Seat on Board of Directors
The Hospitalist
Other Pediatric Literature of Interest
The Hospitalist
Pediatric in the Literature
The Hospitalist
Pediatric Practice Profile
The Hospitalist
The Hospitalist
Pediatric Hospital Medicine
The Hospitalist
   Comments ()