Translating Best Evidence Into Best CareEDITOR’S NOTE: Journals reviewed for this issue: Archives of Disease in Childhood, Archives of Pediatrics and AdolescentMedicine, British Medical Journal, Journal of the American Medical Association, The Journal of Pediatrics, The Lancet,New England Journal of Medicine, Pediatric Infectious Diseases Journal, and Pediatrics. Heidi Marleau, MLS, Ebling Libraryfor the Health Sciences, University of Wisconsin, contributed to the review and selection of this month’s abstracts.

—John G. Frohna, MD, MPH

H2-Blockers are associated with necrotizingenterocolitis in very low birthweight infantsTerrin G, Passariello A, De Curtis M, Manguso F, Salvia G,Lega L, et al. Ranitidine is associated with infections, necro-tizing enterocolitis, and fatal outcome in newborns. Pediat-rics 2012;129:e40-e5.

Question Among very low birth weight (VLBW) infants, isthe use of ranitidine associated with an increased risk of in-fections, necrotizing enterocolitis (NEC), and fatal outcome?

Design Multicenter, prospective observational study.

Setting 4 neonatal intensive care units in Italy.

Participants 274 VLBW infants with birth weight between401 and 1500 g or gestational age between 24 and 32 weeks.

Intervention Exposure, or not, to ranitidine.

Outcomes The primary outcome was the rate of infectiousdiseases. Secondary outcomes included NEC, death, andlength of hospital stay.

Main Results In this population, 91 infants had taken rani-tidine and 183 had not. The main clinical and demographiccharacteristics did not differ between the 2 groups. Thirty-four (37.4%) of the 91 children exposed to ranitidine and18 (9.8%) of the 183 not exposed to ranitidine had contractedinfections (OR = 5.5, 95% CI, 2.9-10.4, P < .001, numberneeded to harm = 4). The risk of NEC was 6.6-fold higherin ranitidine-treated VLBW infants (95% CI, 1.7-25.0,P = .003) than in control subjects. Mortality rate was signif-icantly higher in newborns receiving ranitidine (9.9% vs1.6%, P = .003, number needed to harm = 12).

Conclusions Ranitidine therapy is associated with an in-creased risk of infections, NEC, and fatal outcome inVLBW infants. Caution is advocated in the use of this drugin neonatal age.

Commentary Gastroesophageal reflux (GER) is a commonphysiologic process in infants, and is even more prevalentin preterm infants. Most cases of infantile GER resolve bythe age of 24 months. Despite this, the use of inhibitors ofgastric acid secretion, including the H2-blocker ranitidine,is commonplace in many preterm infants to treat or asprophylaxis against suspected GER. Multiple observationalstudies have reported an association between the use ofH2-blockers and neonatal sepsis and/or NEC.1-4 This makessense as gastric acid is a major non-immunologic defenseagainst ingested pathogens, and inhibition of gastric acidcan allow pathogenic bacteria to proliferate in the gastroin-

168

testinal tract leading to bacteremia and infection. The studyby Terrin et al is a prospective cohort study that evaluatesthis association. Unique to most cohort studies, a prospectivesample size calculation was done, lending additional credibil-ity to the study. The patient characteristics between the groupexposed to ranitidine and the unexposed group are similarwith regards to important risk factors for infection andNEC, including birth weight, gestational age, sex, Apgarscores, CRIB scores, persistent ductus arteriosus, durationof endotracheal intubation, and central vascular access dura-tion. Many of these potentially confounding risk factors havenot been accounted for in previous studies. Given the multi-ple studies showing an association between the use of H2-blockers and neonatal sepsis and/or NEC, it is unlikely thata randomized controlled trial using H2-blockers in VLBWinfants can be performed with equipoise present. As withmany clinical decisions, the risks and benefits of acid blockersneed to be weighed before they are used. There is no clearbenefit of acid blockers in neonates, particularly in pretermneonates, although there continues to be perceived benefitsamong many subspecialists caring for neonates. As long asacid blockers continue to have perceived benefits in VLBWinfants, many clinicians will continue to use them, believingthe “benefits” outweigh the increased morbidity. The resultsfrom this prospective cohort study provides the most usefuland current information regarding the potential risks ofH2-blocker use in preterm neonates, and they should beused with care in this population.

Jennifer Dalton, MDRobert Schumacher, MDUniversity of MichiganAnn Arbor, Michigan

References

1. Beck-Sague CM, Azimi P, Fonseca SN, Baltimore RS, Powell DA,

Bland LA, et al. Bloodstream infections in neonatal intensive care

unit patients: results of a multicenter study. Pediatr Infect Dis J 1994;

13:1110-6.

2. Graham PL III, BeggMD, Larson E, Della-Latta P, Allen A, Saiman L. Risk

factors for late onset gram-negative sepsis in low birth weight infants hos-

pitalized in the neonatal intensive care unit. Pediatr Infect Dis J 2006;25:

113-7.

3. Guillet R, Stoll BJ, Cotton CM, Gantz M, McDonald S, Poole WK, et al.

Association of H2-blocker therapy and higher incidence of necrotizing

enterocolitis in very low birth weight infants. Pediatrics 2006;117:e137-42.

Vol. 161, No. 1 � July 2012

4. Tijerina-Torres CY, Rodriguez-Balderrama I, Gallegos-Davila JA, et al. In-

cidence and risk factors associated with in-hospital neonatal sepsis. Rev

Med Inst Mex Seguro Soc 2011;49:643-8.

Two cochlear impants are better than oneBoons T, Brokx JPL, Frijns JHM, Peeraer L, Philips B,Vermeulen A, et al. Effect of pediatric bilateral cochlear im-plantation on language development. Arch Pediatr AdolescMed 2012;166:28-34.

Question Among prelingually deaf children, does bilateralcochlear implantation, compared with unilateral implanta-tion, result in improved spoken language outcomes?

Design Case-control, retrospective, cross-sectional, multi-center study.

Setting Two Belgian and 3 Dutch cochlear implantation cen-ters.

Participants 25 children with 1 cochlear implant matched(on 10 auditory, child, and environmental factors) with 25children with 2 cochlear implants. Both groups of childrenwere selected from a retrospective sample of 288 childrenwho underwent cochlear implantation before 5 years of age.

Outcomes Performance on measures of spoken languagecomprehension and expression (Reynell Developmental Lan-guage Scales and Schlichting Expressive Language Test). Inthose children who had bilateral implants, the authors alsoevaluated the impact on language development based onthe interval between the first and second implants.

Main Results On the receptive language tests, children un-dergoing bilateral implantation performed significantly bet-ter than those undergoing unilateral implantation (meandifference [95% CI], 9.4 [0.3-18.6]) and expressive languagetests (15.7 [5.9-25.4] and 9.7 [1.5-17.9]). Because the twogroups were closely matched, the authors argue that the dif-ference in performance mainly can be attributed to bilateralimplantation. In addition, for those children with bilateralimplants, a shorter interval between the implants was relatedto higher standard scores. Children undergoing 2 simulta-neous cochlear implantations performed better on the ex-pressive Word Development Test than did childrenundergoing 2 sequential cochlear implantations.

Conclusions The use of bilateral cochlear implants is associ-ated with better spoken language learning. The interval be-tween the first and second implantation correlatesnegatively with language scores. On expressive language de-velopment, there is an advantage to simultaneous versus se-quential implantation.

Commentary Boons et al address the question, “Are two co-chlear implants better than one?” The psychoacoustic litera-ture is clear that normal hearing people and those withhearing aids do better listening with two ears than withone. Why? Binaural hearing allows localization of sound,an important factor in daily communication and safety. Byidentifying the talker, a listener is able to utilize speech read-ing, enhancing comprehension with facial expression, lip

position, breath taking, hand gestures, and body language.Listening with two ears also enhances speech recognition inbackground noise by taking advantage of the head shadow ef-fect (a noise on one side of the skull is partially shielded fromthe opposite ear and the brain focuses attention to the cleanersignal) and binaural central speech processing. Boons et alprovide compelling evidence that, like eye glasses supersedemonocles, two cochlear implants are better than one.

Thomas J. Balkany, MD, FACS, FAAPUniversity of Miami Ear Institute

Miami, Florida

Oral sucrose with facilitated tucking is effectivepain control for preterm infantsCignacco EL, Sellam G, Stoffel L, Gerull R, Nelle M, AnandKJS, et al. Oral sucrose and “facilitated tucking” for repeatedpain relief in preterms: a randomized controlled trial. Pediat-rics 2012;129:299-308.

Question Among preterm infants undergoing repeated heelsticks, is the combination of oral sucrose and facilitated tuck-ing for comfort more effective at relieving pain than eitherintervention alone?

Design Multicenter, randomized controlled trial.

Setting 3 neonatal intensive care units (NICUs) in Switzer-land.

Participants 71 preterm infants between 24 and 32 weeks ofgestation.

Intervention Data were collected during the first 14 days ofthe infant’s NICU stay. Oral 20% sucrose (0.2 mL/kg) wasadministered 2 minutes prior to the procedure. For facili-tated tucking (FT), the infant was held by placing a handon his or her hands and feet and by positioning him/her ina flexed midline position while in either a side-lying, supine,or prone position. Three phases of 5 heel stick procedureswere videotaped (baseline, heel stick, recovery) for eachinfant.

Outcomes Pain responses were graded by videotape reviewaccording to the Bernese Pain Scale for Neonates (BPSN)by four independent, experienced NICU nurses blinded towhether the heel stick was being performed and the heel stickphase with high inter-rater reliability (Cronbach a: 0.90-0.95). The primary outcome was evaluation of pain response,including total pain response and pain response to each com-ponent phase of the heel stick procedure. The secondary out-come was evaluation of pain response based on gestation agegroupings.

Main Results Sucrose alone was significantly more effectivethan FT alone at relieving procedural pain (P < .002). FT incombination with oral sucrose seemed to have an addedvalue in the recovery phase (P = .003) as compared withboth treatment groups alone. There were no differences inpain responses across gestational age.

169