JOURNAL CLUB

JOURNAL CLUB

COMMENTARIES BY MICHAEL G. BURKE, MD

Lights and nearsightedness

Children who sleep in a room with a light on during the firsttwo years of life may be at considerably higher risk of developingmyopia than children who sleep in a dark room, a new study shows.Using a questionnaire, investigators queried parents of about480 children from 2 to 16 years of age about how much light theirchildren were exposed to when they slept, both at the child'spresent age and before the age of 2 years, when the eye growsrapidly but myopia usually has not begun.

Parents chose among "room light," "night light,"and "darkness" to describe the lighting conditions inwhich their children slept at night. Myopia was strongly associatedwith light exposure during sleep in the first two years of life,and the stronger the light, the more myopia. Of children who hadslept with a night light on before 2 years of age, 34% were myopic.Of those who slept with a room light on before 2 years of age,55% were myopic. Only 10% of those who slept in darkness beforethe age of 2 were myopic. Reported light exposure at the child'spresent age had no association with nearsightedness (Quinn GEet al: Nature 1999;399:113).

Commentary: This study is limited by its retrospective design,tertiary referral patient population, and dependence on parentalrecall. It still is an interesting initial observation---one thatdeserves to be the focus of further attention. Meanwhile, lastone out of the nursery turns out the lights!

Fast, effective way to deliver topical anesthetic

Iontophoresis, a method of administering lidocaine using aflow of electrical current through the skin, appears to be aneffective way to anesthetize children undergoing minor procedures.Iontophoresis acts within 10 minutes compared with 60 minutesfor EMLA (a topical eutectic mixture of lidocaine and prilocaine),the anesthetic of choice for minor procedures in children.

These conclusions are drawn from a study of the efficacy ofdermal anesthesia by lidocaine iontophoresis in children 7 yearsor older requiring nonemergency peripheral intravenous (PIV) catheterplacement in the emergency department. During a six-month periodinvestigators assigned 22 children to a lidocaine group and 25to a control group. Children in both groups underwent iontophoresis;those in the lidocaine group received 1 mL of 2% lidocaine with1:100,000 epinephrine over a potential PIV site, while the controlgroup received 1 mL of 0.9% saline solution with 1:100,000 epinephrine.

Children in the two groups indicated how uncomfortable theyfelt during during iontophoresis by choosing among five statementsthat ranged from "it did not bother me at all" to "itwas painful." Most patients reported only slight or no discomfort.After PIV placement, children in both groups ranked the pain theyexperienced during the procedure on a scale from 0 (no pain) to10 (worst possible pain). Children in the lidocaine group feltsignificantly less pain than those in the control group. In thelidocaine group, the median pain score was 0.5 compared with 4in the control group (Kim MK et al: Ann Emerg Med 1999;33:395).

Commentary: This technology is being marketed by IOMED underthe product name Numby Stuff. The authors report that the reusableiontophoresis unit costs about $400; single-use iontophoresispads are between $6 and $7.50. The technology may be worth someinvestigation.

Cruising and bruising

A new study throws light on when bruises in children are consistentwith normal activity and when they should spark concern that abuseor illness is responsible. Investigators collected data on 973healthy children younger than 3 years observed by pediatriciansat seven practice sites; 203 (20.9%) of the children had bruises.Frequency of bruising was directly correlated with age and developmentalstage: Bruises were rare in infants and precruisers, but commonamong cruisers and walkers. Within every age group, infants whowalked had more bruises than those who were crawling. Girls andboys were equally likely to be bruised, but African-American infantshad bruises much less frequently than white children.

Bruises were almost always over bony prominences: the frontof the tibia or knee, forehead, and scalp. Bruises on the face(other than the forehead) and trunk were rare, and no childrenhad bruises on the hands or buttocks. Investigators concludedthat bruises in infants younger than 9 months who are not yetwalking should trigger consideration of abuse or illness, as shouldatypically located bruises in toddlers (Sugar NF et al: Arch PediatrAdolesc Med 1999; 153:399).

Commentary: These authors do a nice job quantifying what youalready do when you evaluate a child with bruises: Consider theage and developmental level of the child and the location of thebruise. Of walking children, 51.9% younger than 3 years had atleast one bruise, while only 2.2% of "precruisers" hadany bruises. A smaller prevalence study established similar ratesof bruising in 6- to 12-month-olds, but showed more facial bruisingin presumably unintentional injuries (Carpenter RF: Arch Dis Child1999;80:363).

G-rated films: Smoking and drinking in movies

A review of 50 G-rated children's animated films shows thatmany story plots feature tobacco or alcohol use and fail to communicateverbally that these substances have long-term negative healtheffects. The full-length films in the study were released between1937 and 1997 by five major production companies--Walt Disney,MGM/ United Artists, Warner Brothers Studios, Universal Studios,and 20th Century Fox; the films are all available on video.

Of the 50 films viewed, 34 showed at least one episode of tobaccoor alcohol use; 28, including all seven animated movies releasedin 1996 and 1997, portrayed one or more incidences of tobaccouse, and 25 included examples of alcohol use. Smoking was thevice of choice with 76 characters smoking for more than a cumulativetotal of 45 minutes. Investigators logged alcohol use in 63 characterswho accumulated 27 minutes of screen time. Good characters wereas likely to use tobacco and alcohol as bad or neutral characters.Cigars and wine were the most popular products.

Only 10 films showed that wooziness or coughing can resultfrom tobacco use or being around smoke; only seven depicted theimmediate ill effects of alcohol use, such as drunkenness, passingout, hiccups, losing balance, or falling over. None of the filmsaddressed the long-term effects of these substances on health,or the possibility and consequences of addiction (Goldstein AO:JAMA 1999;281:1131).

Commentary: Our children are constantly exposed to images ofalcohol and tobacco use--in advertising, live-action entertainmentand, it now is clear, in animated motion pictures. One solutionis to regulate that exposure. (Have you noticed that billboardsadvertising tobacco are coming down?) Another option is to educatechildren about how they are being manipulated by the media. TheMedia Sharp program of the American Academy of Pediatrics providespediatricians and educators with video and written informationfor teaching children and parents how to analyze messages in themedia. You can get more information from AAP's Division of PublicEducation.

Viral or bacterial: Does band count help?

A new study suggests that measuring the immature neutrophils(bands) in the peripheral blood smears of young febrile childrendoes not help to distinguish patients with bacterial infectionsfrom those with viral infections, as some observers have suggested.Investigators found that the band count is similar in these twogroups of children, whether represented as the percentage of whiteblood cells in the peripheral blood smear, the absolute band count,or the band-neutrophil ratio. Patients with bacterial infectionshad a higher absolute neutrophil count than patients with respiratoryviral infections, however.

Investigators studied prospectively 100 febrile children 2years or younger who had laboratory-documented bacterial or respiratoryviral infection. They obtained a complete blood cell count fromall patients and performed a manual differential count of theperipheral blood smear. In making their determination, investigatorsadjusted for the child's age, temperature, and general appearance(Kuppermann N et al: Arch Pediatr Adolesc Med 1999; 153:261).

Commentary: I am not yet ready to say that the band count iscompletely useless. As with any test, however, the band countfor each patient needs to be considered in the context of theclinical picture as a whole.

Costs of feeding infants formula

Because breastfed infants have lower rates of illness thanthose fed with formula, babies who are formula-fed are likelyto cost the health-care system more than their breastfed peers.In a study that supports this hypothesis, investigators estimatedthe actual costs of care for formula-fed and breastfed infantswith regard to three common conditions in the first year of life:lower respiratory tract illness, otitis media, and gastrointestinalillness. They calculated the frequency of office visits and hospitalizationsfor the three illnesses in about 2,000 children who were neverbreastfed, partially breastfed, or exclusively breastfed duringthe first three months of life.

For each 1,000 infants, those who were fed formula in the firstyear of life had 2,033 more office visits, 212 more days of hospitalization,and 609 more prescriptions for these three illnesses than infantsexclusively breastfed for at least three months. These additionalhealth- care services cost the managed care health system an estimated$331 to $475 for each formula-fed infant during the first yearof life. Investigators concluded that health-care plans are likelyto realize substantial savings, as well as to provide improvedcare, by supporting and promoting exclusive breastfeeding (BallTM et al: Pediatrics 1999; 103[4]:870).

Commentary: Follow the money! These authors make a conservativeestimate of health-care cost savings associated with breastfeeding.Extension of their work to other diagnoses, costs of lost parentalwork days, and costs of the formula would make the numbers evenmore striking. In the words of the authors, "Health insuranceplans would be wise to support breastfeeding initiatives thatimprove patient care, infant health, and the financial bottomline."

Briefly noted

Differentiating Lyme meningitis from viral meningitis. Practitionerscan easily confuse Lyme meningitis (LM) with viral meningitis(VM). Serologic tests may be negative or indeterminate early ininfection, when disease disseminates to the nervous system. Withthis in mind, investigators conducted a retrospective analysisof the records of 22 children with LM or VM to determine how thetwo conditions can be differentiated early in their course inthe absence of erythema migrans, which is pathognomonic for LM.Certain key points, they found, suggest the diagnosis:

• Cranial neuropathy, especially peripheral seventh nerve palsy, is strong evidence of LM.
• Papilledema is more common in children with LM than in those with VM.
• Headache, neck pain, and malaise are equally likely in patients with LM and VM, but last longer in patients with LM.
• Fever that is present at the time of diagnosis is more likely in VM than LM.
• Cerebrospinal fluid pleocytosis (a greater than normal number of cells) is less pronounced in LM than in VM, especially the neutrophilic component (Eppes SC et al: Pediatrics 1999;103[5]: 957).

Pertussis-like epidemic caused by C pneumoniae. Researchersin Japan have reported on an epidemic of pertussis-like illnessassociated with Chlamydia pneumoniae infection. They describea middle school of 230 students from 12 to 15 years old, 136 (59%)of whom developed illness with severe coughing. One child developedpneumonia, nine had bronchitis, and 126 (93%) had upper respiratorytract infections. Mean duration of cough was 17.4 days in patientswith upper respiratory tract infections and 30.4 days in patientswith lower respiratory disease. Serology and cultures for Bordetellapertussis, Bordetella parapertussis, Mycoplasma pneumoniae, Chlamydiatrachomatis, Chlamydia psittaci, or viruses were negative. Of46 patients tested by serology and polymerase chain reaction forC pneumoniae, 40 tested positive (Hagiwara K et al: Pediatr InfectDis 1999;18:271). For more about C pneumoniae infection, see HammerschlagMR: Contemporary Pediatrics 1999;(16)5:57.