April 2012 Journal Watch

Emergency Medicine

Ierland Y et al. Self-referral and serious illness in children with fever. Pediatrics. 2012;129(3):e643-649.
Fever is one of the most common chief complaints presenting to pediatric emergency departments (ED). This study looked at the severity of febrile illnesses in children self-referred to the ED by their parents versus referred by primary care pediatrician. This cross-sectional observational study was performed in two large inner-city hospitals in the Netherlands. 4609 children less than 16 years old who presented to the ED with chief complaint of fever or rectal temperature greater than or equal to 38.5C were included in the study. 38% were referred by pediatricians and 62% were self-referred. Severity of illness was measured with help from the Manchester Triage System, need for diagnostic interventions, therapeutic interventions, and follow-up. The most common presenting problems accompanying fevers were similar in self and pediatrican- referred patients: dyspnea, gastrointestinal complaints, and neurologic symptoms. 43% of pediatrican-referred children and 27% of self-referred children needed interventions in the ED. 46% of pediatrician-referred and 45% of self-referred children were classified as high urgency in the ED. 20-37% of pediatrician-referred children and 11-22% of self-referred children required hospitalization. Children self-referred for fever were overall less severely ill than pediatrican-referred children, but 1 in 4 of self-referred children still required an intervention in the ED. It can be easy when in the ED to stereotype self-referred febrile children as less sick than pediatrician-referred children. Based on the results of this study, listening to parental concerns is important, as 1 in 4 self-referred children required an intervention in the ED.
Submitted by: Brittany Massare
Comment: While the study was cross-sectional without longitudinal follow up, and we can argue about whether these results can be generalized to our own population, Dr. Massare’s teaching point is valid. Most parents do know when their child is ill, and we absolutely need to respect their concern. LF


Ross JL, Quigley A, et al. Growth hormone plus childhood low-dose estrogen in turner’s syndrome. New England Journal of Medicine 2011;364:1230-1242.
The addition of estrogen to human growth hormone therapy in Turner’s Syndrome is controversial. Early estrogen therapy in these patients is thought to decrease average height due to acceleration of plate closure, and it is usually delayed until teenage years despite early evidence of estrogen deficiency in this patients. This study sought to determine if addition of low-dose estrogen would result in an increase in average height for Turner’s patients. Pre-pubertal Turner’s patients (5-13 years old) were randomized to four groups to receive either growth hormone (GH) and estrogen, GH alone, estrogen alone, and only placebo. The dose of estrogen was adjusted depending on the patient’s age and pubertal characteristics to best mimic physiologic dosing. The study showed that GH with adjuvant estrogen produced a statistically significant increase in adult height. It is speculated the adjuvant effect is due to local increase in IGF-1 responsiveness induced by low-dose estrogen at the growth plate level. Although a specialized treatment, this study reflects that low dose estrogen therapy, carefully titrated based on age and pubertal development, can positively affect the adult height of a Turner’s patient. This study also highlights some important principles regarding puberty. First, although ovarian dysfunction is a major problem in Turner’s patients, the study noted that 13% of patients receiving placebo did have pubertal development. This possibility for spontaneous puberty is an important treatment consideration, and may reflect the varying modes of inheritance of the syndrome. Second, the study reflects an important physiological principle that pre-pubertal females still produce low amounts of estradiol with potential effects on growth and development.
Submitted by: Mike Platt
Comment: This is a very interesting study and seems promising. Since there is such a wide range of severity in Turner's syndrome, it would seem that individualized treatment would be necessary. LC

General Surgery

Mehta S et al. Clinical characteristics and risk factors for symptomatic pediatric gall bladder disease. Pediatrics 2012;129:e82-87.
The frequency of pediatric cholecystectomies have increased substantially over the past twenty years and while hemolytic disease has classically been the most common associated comorbidity with pediatric gall bladder disease, it is possible, given the rise in childhood obesity, that risk factors for gall bladder disease in children may now parallel those classically associated with adults. This retrospective chart review assessed demographics, anthropomorphic measurements, surgical indications, and imaging study results of 455 pediatric patients who underwent a cholecystectomy at Texas Children’s Hospital between 2005 and 2008. After excluding patients with hemolytic disease (n=76), symptomatic cholelithiasis (54% of patients), complicated obstructive disease (28%) and biliary dyskinesia (16%) were the primary indications for surgery in this cohort and the majority of patients in all three categories were obese (67, 55, and 51% respectively). Additionally female gender, Hispanic ethnicity, and older age (13-18 years) each presented as independent risk factors in this cohort. In comparison to historical controls, the number of patients with hemolytic disease who underwent cholecystectomy was unchanged. This study strongly suggests the obesity epidemic in children has contributed significantly to the increase in pediatric gallstone disease and further accentuates the need for better interventions to maintain a healthy lifestyle and that pediatric providers should consider gall bladder disease higher on the differential diagnosis of pediatric abdominal pain.
Submitted by: Jason Catanzaro
Comment: Also of interest in this comparison to historical data: biliary dyskinesia (defined as gall bladder ejection fraction <35% on HIDA scan) was a completely new co-morbidity reported in 16% of subjects, and the proportion of cholecystectomies performed as open vs. laparoscopic surgery was dramatically different in the two time periods (15% vs. 96% were laparoscopic in the historical vs. current series respectively). LF


Committee on Genetics. Maternal phenylketonuria. Pediatrics 2008;122(2):445-449.
This article was given to me by one of the development/behavior attendings because we saw two children with Attention-Deficit and Hyperactivity Disorder (ADHD) related to maternal phenylketonuria (PKU). Although we see much less PKU now because of newborn screening, we may have female patients of childbearing age with PKU who should be counseled regarding the risks for adverse fetal effects. In summary, PKU is an autosomal recessive disorder that is treated by dietary restriction of phenylalanine (Phe) and supplementation with tyrosine. Data has shown that lifelong treatment is likely necessary to preserve intelligence (despite previous studies showing treatment only until brain myelination is complete). PKU affects brain function by abnormal catecholamine concentrations, myelination, and amounts of amino acids leading to cognitive, emotional and hyperactive disorders. Phe crosses the placenta and can lead to miscarriage, IUGR, global developmental delay and heart defects. Thus, maternal PKU has significant effects whether or not the child has PKU, so strict control of maternal PKU should start before pregnancy or by 8 weeks gestational age at the latest. Dietary control is difficult (bland diet) so poor adherence is a big problem. Maternal Phe concentrations should be monitored every 2 weeks during pregnancy, and the barriers to adherence addressed. Ultimately, the AAP recommends that all females of childbearing age with PKU be counseled regarding the risks of adverse fetal outcomes, contraception if unable/unwilling to maintain appropriate Phe levels, and genetic risks.
Submitted by: Anandhi Gunder
Comment: This is a nice review of why women with PKU need ongoing education about their condition, especially during their childbearing years. LC

Infectious Disease

Lopez-Medina E et al. 2009 Influenza A in infants hospitalized at younger than 6 months. The Journal of Pediatrics 2012;160(4):626-631.
This was a retrospective study of 73 hospitalized infants all less than 6 months diagnosed with Influenza A. The study aimed to look at the clinical features of these infants. The age of infants was from 3-179 days admitted to a tertiary care center in Texas. All infants had laboratory confirmation of Influenza A. Of these infants, 60% had confirmation of H1N1. 68% of patients had a household contact with respiratory signs of infection. All but one had community-acquired Influenza A. The most common initial symptoms were fever and respiratory signs. The next most common clinical sign was poor feeding. Fever was the most common documented reason for hospitalization. About 50% need IV fluids for dehydration. Ten of the patients were also diagnosed with a bacterial infection. The most common co-infection was from the respiratory and urinary tracts. Of the co-infections, one infant had Klebsiella pneumonia bacteremia and one had E.coli meningitis. Five were diagnosed with pneumonia. Admissions to the PICU were more frequent in those patients with co-morbid medical problems. In addition, two deaths occurred, both in patients with complicated past medical history. For treatment, 81% ended up getting IV antibiotics and 84% received. There were no adverse effects reported from the oseltamivir. Most patients had short hospital oseltamivir stays (median=2 days). All but one patient had community acquired influenza A. This study encourages me to continue to emphasize the importance of vaccination in all household contacts of infants to decrease their risk of infection.
Submitted by: Jamee Paoloemilio
Comment: This study should make us all quake in our boots and increase our vigilance regarding young infants diagnosed with Influenza: 14% had an important co-infection. Therefore, a febrile infant with Influenza is no less likely to have serious bacterial illness than one without! LF

Bocquet N et al. Randomized trial of oral versus sequential IV/oral antibiotic for acute pyelonephritis in children. Pediatrics 2012;129:e269-275.
Acute pyelonephritis (APN) may lead to renal scarring, hypertension and other renal dysfunction. Current standard of care to prevent long term sequelae is intravenous followed by sequential oral antibiotic treatment; however it is unknown whether exclusive oral treatment is as efficacious to prevent renal scarring. 171 children aged one to 36 months without vomiting or diarrhea that would preclude oral antibiotic administration presenting to an ED with their first febrile UTI were enrolled in this prospective multicenter trial. In addition to a properly collected urinalysis and culture, these children had an elevated procalcitonin concentration reflecting renal involvement and a renal ultrasound documenting no renal anomalies. Children were randomized to either receive IV ceftriaxone and then oral cefixime to complete a ten day course, while the study group received oral cefixime for ten days. All patients underwent a DMSA scintigraphy, analyzed by radiologists blinded to the treatment groups, within eight days of study inclusion and, if abnormal, six to eight months later. Re-imaging of 119 patients with an initial abnormal scintigraphy, indicating early renal damage, demonstrated no difference in the prevalence of renal scarring (41% control; 44% oral) between the two groups thus supporting exclusive oral antibiotic use for first episodes of APN. While the results are promising, this study lacked sufficient statistical power to properly conclude the safety of an exclusive oral regimen and additionally as all urine cultures demonstrated pan-sensitive E. coli it could not be generalized to all urinary pathogens.
Submitted by: Jason Catanzaro
Comment: I agree with Dr. Catanzaro that this study is interesting, but not applicable to all patients. It shows that oral therapy is useful in an otherwise healthy infant with E.coli UTI. Infants with vomiting and dehydration were excluded from the study. LC

Shah, SS, Aronson, PL, and Lorch SA. Delayed acyclovir therapy and death among neonates with herpes simplex infection. Pediatrics;128(6):1153-1159.
A multicenter retrospective cohort study was done over 6 years with 1086 neonates diagnosed with Herpes Simplex infection (HSV) at discharge. The median age at presentation was 10 days. The investigators compared outcomes of early treatment with intravenous acyclovir, defined as given less than 1 day (24 hours) of hospitalization versus delayed treatment, defined as greater 1 day, but less than 7 days of hospitalization. 262 neonates (24%) had delayed administration. The overall mortality rate was 7.3%. There was a significant difference in mortality between delayed and early groups (9.5% vs. 6.6% respectively). With propensity scores considered, delayed administration was associated with a twofold increase in mortality. As the duration of delayed administration increased, mortality increased for each day, especially in infants less than14 days old. There are several considerations the authors made with respect to the results of their study. First, the dose of acyclovir that was used (high dose standard) was unknown, the timing of the first dose may have been misclassified as delayed if the neonate received a dose prior to transferring to one of the study hospitals, and neurologic morbidity was not addressed. Despite these limitations, this study supports the early use of empiric acyclovir therapy in neonates. This would change my practice by having a low threshold to start acyclovir in neonates as empiric therapy while diagnostic studies are completed.
Submitted by: Emily Wiland
Comment: The authors carefully review the possible study design issues that could have impacted their results; however, the data is quite compelling. The fact that the odds of death increased incrementally with each day of delay in acyclovir initiation speaks strongly in favor of their conclusion that delay is causally associated with increased mortality. LF


Bueno M et al. Breast milk and glucose for pain relief in preterm infants: a noninferiority randomized controlled trial. Pediatrics 2012 129(4):664 -670.
A noninferiority randomized controlled trial was conducted to compare the effect on pain intensity using either expressed breast milk or 25% dextrose during a heelstick. The authors hypothesized that expressed breast milk has at least as much efficacy as the 25% dextrose solution for pain reduction or is worse by an amount less than 10%. A total of 113 late preterm newborns gestational age 34-36 weeks within the first 72 hours of life who had an indication for a heelstick were enrolled. 88 neonates completed the study. Using video taped heelsticks with scorers blinded to solution used, the premature infant pain profile (PIPP) was used to score pain intensity at 30 second increments after heelstick for up to 3 minutes. Incidence of cry and percentage of time spent crying were considered secondary outcomes. Significantly lower pain scores were observed in the 25% dextrose group at all time intervals after lancing. Additionally, significantly more infants in the 25% dextrose group had PIPP scores indicative of minimal pain or absence of pain 30 seconds after lancing. Lower incidence of cry (p = .001) and shorter duration of crying (p = .014) were seen also in the 25% dextrose group. In conclusion, expressed breast milk does not appear to be as effective as 25% dextrose for pain control in late preterm neonates. This study emphasizes the importance of pain control for neonates and shows that 25% dextrose is useful and advantageous over breast milk as an analgesic for minor procedures such as heelstick.
Submitted by: Emily Wiland
Comment: The authors point out in their discussion that slightly higher volumes of breastmilk have been shown to reduce pain (5ml vs. 2ml in this study), and that breastfeeding alone can help reduce pain. So if the infant is stable, allowing the patient to breastfeed through a heelstick can be helpful in reducing pain. LC


Silbergleit MD, Durkalski V, Lowenstein D, et al. Intramuscular versus intravenous therapy for prehospital status epilepticus. N Engl J Med 2012;366:591-600.
This study was a randomized non-inferiority trial to assess intramuscular (IM) midazolam to intravenous (IV) lorazepam for the termination of status epilepticus in a prehospital EMS setting. 893 patients were enrolled (total of 145 patients < 21 years of age). Each group of paramedics had a study kit containing an IM autoinjector and pre-filled IV syringe. Each kit was randomized by the central pharmacy to contain either IM midazolam or IV lorazepam with the other injector/syringe containing a placebo. Subjects were enrolled after assessment by the EMS staff and opening of the study kit. Each patient received the IM auto-injection followed by IV access and administration of the IV syringe. Primary outcome measure was termination of seizures prior to Emergency Department (ED) arrival with secondary outcomes of time from kit opening to seizure termination, time from medication administration to seizure termination, length of hospitalization, and adverse events. Study results demonstrated that 73.4% of patients had seizure termination prior to ED arrival in the IM treatment group compared to 63.4% in the IV treatment group (P<0.001 for non-inferiority). The IM treatment group also had a relative risk of 0.88 for hospital admission compared to the IV treatment group (95% CI 0.79-0.98). Treatment failure, endotracheal intubation, recurrent seizures, hypotension, and length of hospital stay were similar between treatment groups. While IM treatment occurred sooner than IV administration this was offset by the faster onset of action of IV treatment resulting in a nonsignificant difference from treatment to seizure termination. While this study is limited by including a limited number of pediatric patients and two different benzodiazepines it appears that IM midazolam is a valid treatment for cessation of status epilepticus in an EMS setting. This is especially important in pediatric patients where obtaining IV access is potentially more difficult, especially in a seizing child.
Submitted by: Nathan Beins
Comment: It is probably useful as a first responder to have more than one option for treating a seizure, and this data shows that IM midazolam can be effective for patients with difficult IV access. LC


Reich B, Zalut T, Weiner SG. An international evaluation of ultrasound versus computed tomography in the diagnosis of appendicitis. International Journal of Emergency Medicine. 2011 Oct;4:68.
As the harmful effects of computeed tomography (CT) radiation become more apparent in the medical literature, a push to find safer and equally effective diagnostic tools is getting more and more attention. Ultrasound (US) has become increasingly popular due to its portability, lower cost, and great safety profile. Despite these advantages there is debate over its use, especially in the diagnosis of appendicitis. In this article, investigators compared the effectiveness of CT with radiologist-operated US in the evaluation of acute appendicitis both in the U.S. and Israel where US is used more often. A retrospective chart review comparing two teaching hospitals in each country with 79 patients in the U.S. evaluated by CT and 197 patients in Israel evaluated by US. CT had a sensitivity of 100% and the negative appendectomy rate in patients with positive CT was 0%. US had a sensitivity of 68.4% with a negative appendectomy rate in patients with positive US was 5.5%. Of the US patients, 23.4% had follow-up CT scans. Median time from CT order to completion was 185min compared to 20min for US. Length of stay in the ED was 533min in CT patients and 387min in US group. Ultimately, while US was quicker to perform and lacked radiation/contrast, CT had far superior sensitivity and PPV. In this study US where preformed by highly-trained radiologists, and the article failed to address operator dependent errors US is subject to. In the end the article supports a “first-pass” approach to save time, money, and limit radiation by using US first, and if non-diagnositc, followed by CT. As ultrasound technology gets better and better I feel that these statistics will only improve. For now it is important to use both your clinical judgement as well as ultrasound first if there is suspicion for acute appendicitis. CT scan should always be an option however, and you must weigh both the short and long term consequences of using such a modality before ordering it.
Submitted by: Justin Greiwe
Comment: Most of us agree that too many children are getting too many abdominal CTs- the good news is that ultrasound seems to be getting better and better, with negative appendectomy rates dipping below those of patients diagnosed by history and physical alone. Other investigators have found that the impact of imaging on negative appendectomy rates varies by age and gender, with no impact of imaging recently reported for boys >age 5 years (Bachur et al Pediatr 2012). LF

Need to Refer a Patient?

Click here for Patient Referrals

Browse Services A-Z

Maps and Directions

Click here for directions