Pediatric Infectious Diseases Newsletter: April 2010


	April 2010

Is That Conjunctivitis Bacterial or Viral?

In general, about a half of pediatric conjunctivitis is caused by bacteria and half by viruses. Differentiating the two without getting a culture can be difficult. Most textbooks and studies offering clinical criteria to distinguish these etiologies do not offer the data to back them up. A recent study attempted to identify children 6 months - 17 years old (median 3 years) at low risk for bacterial conjunctivitis based on clinical findings. Positive cultures, seen in 64.7%, grew Haemophilus influenzae (67.6%), pneumococcus (19.7%), Staphylococcus aureus (8%), and H. parainfluenzae (2.5%). They identified 4 predictors that increased the chance of getting a negative culture: 1) age >6 y, 2) presentation in April-November, 3) no or watery discharge, & 4) no glued eye in the morning. The table below shows the probability that a child's culture will be negative based is upon the number of predictors present. Few children with 4 predictors will have a bacterial infection.

# of predictors	Probability of a negative culture, %
0	11.8
1	21.1
2	45.8
3	76.4
4	92.3

Although most children with conjunctivitis will still need treatment with topical antibiotics, this prediction model may make it easier to select those patients likely to have a viral infection and not need antibiotics.

Meltzer JA, et al. Arch Pediatri Adol Med 2010;164(3):263-7.

Prophylactic Penicillin for Sickle Cell Patients

A recent study revealed that Medicaid infants in Tennessee have poor compliance with receiving penicillin prophylaxis by 3 months of age, as recommended in a guideline. Only 40% had a prescription filled by 3 months of age. Risk of nonadherence rose with increasing number of the following risk factors: maternal age, marital status, education,income, and residence in an urban area. The study was not designed to identify whether prescriptions had been written by primary care providers. The NIH guideline "Management of Sickle Cell Disease" suggests that infants identified to have sickle cell disease on neonatal screening should be placed on penicillin prophylaxis "as early as possible." The recommended regimen is 125 mg PO BID for neonates to children 3 years old, and 250 mg PO BID for 3-5 years old. An alternative regimen is 1.2 million units of Bicillin IM every 3 weeks. Remember to give these kids the full schedule of conjugated pneumococcal vaccines (Prevnar), as well as a dose of polysaccharide pneumococcal vaccine (Pneumovax) at 2 years of age and a booster dose 5 years later.

Warren MD, et al. Arch Pediatr Adol Med 2010;164(3):298-9.
NIH publication #02-2117. The Management of Sickle Cell Disease, 2002.

HSP triggered by Helicobacter pylori

Henoch-Schonlein purpura (HSP) is a vasculitis with a typical clinical presentation of a palpable, purpuric rash that is most prominent on posterior legs and buttocks, and can involve the face and extensor surfaces of the arms. Many children with HSP have had a preceding respiratory or gastrointestinal illness. A few reports have suggested that H. pylori may sometimes trigger HSP. One case report of a child with H. Pylori infection demonstrated the deposition of IgA and C3 complement in dermal blood vessel walls. Others reports have shown that patients with H. pylori infection have elevated levels of serum IgA, lower levels of C3 complement, and more frequent increases of circulating immune complexes and cryoglobulin. Increased serum IgA and decreased C3 levels have also been reported in some patients with HSP. Perhaps, in some patients, H. pylori can cause an aberrant immune response resulting in HSP. In the case report cited below (which case report?), the child's HSP symptoms resolved after starting therapy for H. pylori.

Mytinger JR, et al. Ped Derm 2008;25(6):630-2.
Shin JI, et al. Ped Derm 2009;26(6):768-9.

Take Part in Our Vaccine Survey

A Message from David Berman, D.O.
Pediatric Infectious Disease Program
All Children's Specialty Physician

All physicians are encouraged to complete the brief survey regardless of your opinion regarding childhood vaccination. If enough responses are received, the anonymous results will be presented at the Annual Suncoast Pediatric Conference (Sarasota, Florida) in June 2010.

Thank you for your participation.

www.allkids.org/vaccinesurvey

Once-Daily Dosing (ODD) of Gentamicin

The aminoglycoside antibiotics are often employed in the treatment of Gram-negative infections. A standard dosing regimen has been to divide the total daily dose into 3 doses given every 8 hours. The last 20 years has seen increasing interest in ODD of aminoglycosides, first in adults and subsequently in children. Such dosing has the potential to increase antibacterial activity and decrease nephrotoxicity. Although the dosing standards have been developed for adults, the published data on ideal doses in children has been less clear. In a recent pharmacokinetic study in children without cystic fibrosis over 3 months of age, investigators extrapolated the following once-daily doses to achieve peak gentamicin serum levels of >20 mcg/ml:

Age	Dose
3 month to <2 years	9.5 mg/kg
2 to <8 years	8.5 mg/kg
8-18 years	7 mg/kg

Such dosing may not be applicable to patients with renal insufficiency. Further studies will have to be done to confirm these results and to establish safety and efficacy.

McDade EJ, et al. Pharmacotherapy 2010;30(3):248-53.

Evaluation of Inpatient Admissions and Potential Antimicrobial and Analgesic Errors in Overweight Children

The prevalence of overweight and obesity in American children has increased steadily over the last several decades. The Centers for Disease Control and Prevention (CDC) and Institutes of Medicine recognize children as overweight and obese if their body mass index (BMI) exceeds the 85th and 95% percentile, respectively. A retrospective study of children aged 5-12 years who were admitted to a children's hospital over a period of 6 months was conducted to determine the percentage of admissions of children with a BMI greater than or equal to the 85th percentile for age and sex and the mean error rate per admission in the overweight versus control group.

A total of 839 admissions representing 699 patients were included. The overweight group included 278 (33.1%) admissions. Comparison of overall mean error rate per admission revealed a statistically significant increase in dosing errors for overweight patents (0.4 ± 0.6 vs. 0.3 ± 0.6; p=0.030), with underdose errors occurring more frequently than overdose errors (0.3 ± 0.6 vs. 0.5; p=0.010). The antimicrobial agents associated with the most risk of underdosing in the overweight group included cefazolin, ceftriaxone, vancomycin, amoxicillin and cefpodoxime, The antimicrobial agents associated with the most risk of overdosing in the overweight group included gentamicin, tobramycin and ceftazidime.

In this study, overweight children accounted for one-third of admissions and appeared to be at greater risk for dosing errors, specifically for under- and overdosing of antimicrobials. This study did not focus on clinical outcomes of under- or overdosing; however, clinicians should be aware that overweight and obese children could be at an increased risk for therapeutic failures or adverse effects. Strategies to decrease potential dosing errors in overweight children include providing an assessment in children who weigh 40kg or more - the clinician should assess whether or not weight-based dosing is appropriate for that specific medication and indication. Secondly, clinicians should be familiar with adult dosing recommendations so as not to exceed, maximum adult doses.

Miller JL, Johnson, PN, Harrision, DL et al. Annals of Pharmacotherapy.2010; 44:35-42

Can Deer Ticks Transmit Bartonella henselae?

Bacteria of the genus Bartonella, which encompasses over 20 different species, is responsible for a large variety of animal and human infections. Recently, several studies have demonstrated Bartonella spp. DNA by PCR in ticks.

Because of these findings, some healthcare providers believe that B. henselae is present in and, therefore, transmissible, by the deer tick (Ixodes scapularis). In addition, there is also a concern that coinfection of Borrelia burgdorferi, the causative agent of Lyme disease and Bartonella henselae, the causative agent of cat scratch disease may occur.

A list reasons why deer ticks are unlikely or not proven to be responsible for transmission of Bartonella henselae is as follows:

Typical cat-scratch disease after a recognized deer tick bite has not been observed.
There is an absence of reports of typical lymph node findings of cat-scratch disease proximal to the bite tick.
Cat-scratch disease has a different seasonal pattern from that of Lyme disease.
The seasonality of cat-scratch disease, in (most cases in temperate regions occur in autumn and early winter when peak breeding of cat fleas and birth of kittens occur) is different than the seasonality for Lyme disease (which occurs predominantly in summer with some cases in spring and autumn).
Appropriate seroepidemiologic studies have not been done.
Vector competence of ticks for B. henselae in an animal system has not been proven.
No convincing evidence of B. henselae in deer ticks has been reported.
The Bartonella species present in Peromyscus leucopus (the white-footed mouse) is not B. henselae.
The US cases with convincing evidence of B. henselae infection after a tick bite occurred in areas where Lyme disease is not endemic