Many clinicians and clinician-scientists have serious concerns that clinical studies (i.e. studies with patients) are less favorably accepted than basic ('bench') projects by funding agencies and by promotion committees of university departments. However, this commonly held view has not been previously verified. To test whether bias exists against clinical studies, we compared the acceptance rate of clinical vs. basic studies dealing with drugs and therapeutics in children, which were submitted to a large scientific meeting. Of 197 abstracts reporting on drug/therapeutic studies, submitted to the Society for Pediatric Research in 1993, there were 133 clinical and 64 basic studies. Fifty-nine (44.3%) of the clinical studies were accepted, significantly less than the basic projects (n = 47 or 73.4%, p < 0.0001). A basic paper was 66% more likely to be accepted (95% CI 50.7-82.6 %). This trend was consistent for different groups of drugs/therapeutics, including analgesics, surfactants, corticosteroids, vaccines, hormones, and antiasthmatics. To examine whether the lower rate of acceptance of clinical papers is the result of lower scientific standard, all papers were scored for their quality by raters who were blinded to their acceptance or rejection status. In general, rejected clinical papers scored significantly higher than rejected basic papers (16 +/- 1.8 vs. 14.8 +/- 1.0, p < 0.05). Our study supports the commonly held but previously unproven view that there is a bias against clinical research, in the context of patient-based studies, when compared to basic (bench) research.
Using the isotope-labelled microsphere method, blood flow to the brain, the heart and the kidneys were studied in newborn piglets during nimodipine infusion. Twenty piglets were studied in two different groups. Group 1 (n = 8) was kept normoxic and given a continuous nimodipine infusion (15 micrograms/kg/min). Group 2 (n = 12) was made hypoxemic by breathing 10% O2 for 10 min followed by an identical nimodipine infusion as group 1. In spite of a significant systemic hypotension, nimodipine infusion alone significantly increased blood flow in the brain stem and right cardiac ventricle at 30-60 min of infusion, while blood flow to cerebellum, cerebrum and the left cardiac ventricle did not change. Blood flow to the kidneys decreased significantly. In posthypoxemic piglets nimodipine infusion gave almost similar flow patterns, however, the changes appeared at an earlier time. We conclude that in spite of a significant reduction in blood pressure, cerebral and cardiac blood flow is preserved both in normal and posthypoxemic animals even at high doses of nimodipine. However, because of the decreased blood flow to the kidneys further dose-response studies are needed before clinical use in asphyctic newborns.
OBJECTIVE:To assess the central nervous system side effects of chlorpheniramine and astemizole in children.DESIGN:Prospective, randomized, double-blind cross-over study.SETTING:Children were recruited from the out-patient allergy clinic and from respondents to an advertisement in a local newspaper. The study was conducted in the outpatient clinics of the ambulatory care services of the Children's Hospital of Eastern Ontario.PATIENTS, PARTICIPANTS:Children 8-16 years of age, with isolated allergic rhinitis or hay fever, were eligible for the study. We excluded children with: (1) chronic conditions (specifically asthma, atopic dermatitis, learning disabilities, or current treatment with oral corticosteroid medication); (2) known allergy to the study medications; (3) recent use of any antihistamine medication. One hundred and three children entered the study and 92 completed it.INTERVENTIONS:Children were stratified by age and randomly allocated to treatment with one of the two medication sequences. Over a period of 13 weeks both groups had 1 week of baseline studies, 3 weeks of one study medication, either chlorpheniramine or astemizole, a 6-week wash-out period and then 3 weeks of the other study medication for a second treatment period.MAIN OUTCOME MEASURES:Attention span (continuous performance test), short-term auditory and visual memory (visual aural digit span test), visual memory for geometric shapes (Benton visual retention test), motor coordination and visual-motor integration (grooved pegboard test), tapping speed and fine motor coordination (finger tapping test), physical side effects (such as sleepiness and dizziness), and compliance.RESULTS:One hundred and three patients were enrolled in the study, 92 (89%) completed the study. There were no significant drug effects on the visual retention test and the continuous performance test. On the visual aural digit span test, patients treated with astemizole scored higher than at baseline. There were no clinical or statistical differences in adverse effects between the two medications or between each medication and baseline.CONCLUSIONS:The two antihistamines studied had no adverse effects on the performance of children.
The effects of a single dose of 5 mg.kg-1 of ketamine administered intravenously to 10 critically ill preterm infants prior to epicutaneo-caval catheterization were analyzed using pulsed-wave Doppler ultrasound. The infants weighed between 670 and 1,885 g and their gestational ages ranged from 26 to 33 weeks. Arterial pressure (MAP), cardiac output (CO), transcutaneous oxygen pressure (TcPO2), transcutaneous carbon dioxide pressure (TcPCO2), end-diastolic velocity (EDV), peak systolic velocity (PSV), mean arterial velocity (MAV) of the cerebral anterior artery as well as Pourcelot's resistance index (PRI) were measured before and after injection of the drug. We observed a significant decrease in arterial pressure at 2 min after injection while heart rate and CO did not vary significantly. TcPO2 and TcPCO2, also remained unchanged throughout the period of measurement. EDV, PSV, and MAV did not vary significantly nor did PRI. As this drug provides major comfort to the baby during painful procedures and considerably facilitates difficult thin vessel catheterization, we believe that it may be used in such conditions.
Deficiency of vitamin A (retinol) has been suggested as an important contributing cause in the pathogenesis of bronchopulmonary dysplasia (BPD) in premature newborns with severe lung disease. Although the efficacy of vitamin A supplementation has yet to be clearly confirmed, it is widely employed, at different dosages and schedules, in neonatal intensive care units (NICU). Since in Italy today no suitable formulation of vitamin A is available, the present observational study was designed to define the profile of plasma vitamin A and retinol-binding protein (RBP) concentrations in supplemented infants at risk for BPD admitted to seven Italian NICU. Twelve babies (average 27 weeks of gestation, birth weight 1,008 g), supplemented with vitamin A, were observed with sequential measurements of retinol and RBP up to 28 days of age. At birth retinol and RBP plasma concentrations were both adequate in the infants and half their mothers' levels. During supplementation the levels rose with wide variability according to the differences in dosing and timing in the different units. Plasma levels of retinol and RBP were the same in infants who had BPD and those who did not. A routine standardized therapeutic approach using vitamin A supplementation in Italian NICU will be more than welcome.
Twenty newly diagnosed growth hormone-deficient children (19 males) were randomized to receive methionyl growth hormone (0.3 mg/kg/week) in subcutaneous doses divided daily (n = 12) or 3 times per week (TIW). With the initial dose and at 4-6 weeks after beginning therapy, procollagen type III propeptide (PIIIP) concentrations were determined. Growth velocities were calculated before and at 1, 3, and 6 months after beginning the therapy. Pretreatment growth velocities were 3.66 +/- (SD) 1.45 and 3.79 +/- 0.55 cm/year for the daily and TIW groups, respectively. At 1, 3, and 6 months mean growth velocities increased to 17.2, 10.2, and 9.5 cm/year for the daily group and 9.8, 6.8, and 7.6 cm/year for the TIW group, with differences between groups significant (p < 0.05) at 1 and 3 months. PIIIP concentrations increased significantly (p < 0.05) over 1 month in both groups, from 11.3 to 18.8 ng/ml and from 10.0 to 12.0 ng/ml in the daily and TIW groups, respectively. In addition PIIIP concentrations were significantly higher (p < 0.05) in the daily group at 1 month. A significant correlation was found between PIIIP concentrations at 1 month and the growth velocity at 1 (r = 0.47), 3 (r = 0.60), and 6 (r = 0.67) months. Pretreatment growth velocity was weakly correlated with posttreatment growth velocity at both 1 (r = -0.45) and 3 (r = -0.42) months. We conclude that (1) growth hormone is more effective when administered daily, (2) pretreatment growth velocity and PIIIP plasma concentration at 1 month correlate with 1 month growth velocity, and (3) PIIIP at 1 month provides a good evaluation of 6 months' response to methionyl growth hormone therapy.
Developmental exposure to cocaine can produce adverse neurobehavioral and cardiovascular effects. Few animal models of human neonatal exposure have been established. A pharmacokinetic study was therefore conducted to characterize the disposition of cocaine and a major metabolite benzoylecgonine (BE) using piglets as an animal model. Eight piglets (postnatal days 8-9) were instrumented with a jugular cannula for drug administration and blood sampling. One group of subjects (controls) received 6.0 mg/kg of cocaine-HCl (i.v.) and blood samples were drawn over 0-24 h. In another group (labetalol), 0.25 mg/kg labetalol-HCl was coadministered 15 min following cocaine dosing. Plasma levels of cocaine and BE were determined using GC-MS methods. Pharmacokinetics were evaluated by using a model-independent approach and compartmental modeling. For controls model-independent results were as follows: AUC = 148.9 +/- 9.0 mg/l x min, systemic clearance = 0.041 +/- 0.003 liters/min/kg, volume of distribution = 1.543 +/- 0.470 liters/kg, and t1/2 beta = 29.4 +/- 6.8 min. Cocaine followed two-compartment model kinetics with distribution and elimination half-lives of 0.3 +/- 0.1 and 58.0 +/- 18.0 min, respectively. Labetalol significantly decreased systemic clearance to 0.029 +/- 0.004 liters/min/kg. BE kinetics revealed a elimination half-life of 230.0 +/- 83.2 min. The results demonstrate a rapid distribution and metabolism of cocaine to BE followed by a prolonged elimination phase which is extended by labetalol treatment.
Heart muscle is dependent on the entry of calcium from the extracellular fluid to support contraction, and neonatal hearts are particularly sensitive to reductions in transsarcolemmal entry of calcium. Accordingly, this study evaluated the ability of the calcium channel agonist BAY K8644 to prevent or reverse the myocardial depressant effects of halothane or isoflurane in right ventricular papillary muscles from neonatal rabbits. The ability of BAY K8644 to reverse reductions in force (F) and dF/dt (halothane and isoflurane) or prevent reduction (halothane) was studied. Halothane decreased F to 24 +/- 2% of baseline values (p = 0.001). The addition of BAY K8644 reversed F to only 54 +/- 3% of baseline (p = 0.001 vs. baseline and p = 0.002 vs. halothane alone). Isoflurane decreased F to 20 +/- 2% of baseline (p = 0.001) with a return to 45 +/- 4% of baseline with the addition of BAY K8644 (p = 0.0001 vs. baseline and p = 0.0025 vs. isoflurane alone). With BAY K8644 in the bath prior to the addition of halothane, halothane decreased F to 38 +/- 4% of baseline (p = 0.001). dF/dt mirrored changes in F in all studies. These data show that a calcium channel agonist is only partially effective in modulating volatile anesthetic-induced depression in neonatal rabbit ventricular papillary muscle.
We studied the effect of acetylcholine (ACh), 1 x 10(-8) to 5 x 10(-7) M, on electrophysiologic characteristics of the isolated (Langendorf), perfused fetal canine heart. ACh induced concentration-dependent decreases in sinoatrial (SA) rate and recovery from overdrive pacing and in atrioventricular (A-V) conduction. These effects of ACh were greater in mid-gestation than late-gestation hearts. The effects of ACh were potentiated by inhibition of acetylcholinesterase by neostigmine, 1 x 10(-7) M, in the late- but not the mid-gestation fetal heart. Decreasing the pH of the perfusion solution from 7.3 to 6.8 potentiated the response to ACh of SA rate and A-V conduction more in mid- than in late-gestation hearts. The response to ACh of the late-gestation fetal canine heart is more sensitive to cholinesterase inhibition whereas the response of the mid-gestation heart is more sensitive to the action of ACh in the presence of acidosis.
Heart muscle is dependent on the entry of calcium from the extracellular fluid to support contraction, and neonatal hearts are particularly sensitive to reductions in transsarcolemmal entry of calcium. Accordingly, this study evaluated the ability of the calcium channel agonist BAY K8644 to prevent or reverse the myocardial depressant effects of halothane or isoflurane in right ventricular papillary muscles from neonatal rabbits. The ability of BAY K8644 to reverse reductions in force (F) and dF/dt (halothane and isoflurane) or prevent reduction (halothane) was studied. Halothane decreased F to 24 ± 2% of baseline values (p = 0.001). The addition of BAY K8644 reversed F to only 54 ± 3% of baseline (p = 0.001 vs. baseline and p = 0.002 vs. halothane alone). Isoflurane decreased F to 20 ± 2% of baseline (p = 0.001) with a return to 45 ± 4% of baseline with the addition of BAY K8644 (p = 0.0001 vs. baseline and p = 0.0025 vs. isoflurane alone). With BAY K8644 in the bath prior to the addition of halothane, halothane decreased F to 38 ± 4% of baseline (p= 0.001). dF/dt mirrored changes in F in all studies. These data show that a calcium channel agonist is only partially effective in modulating volatile anesthetic-induced depression in neonatal rabbit ventricular papillary muscle.
Deficiency of vitamin A (retinol) has been suggested as an important contributing cause in the pathogenesis of bronchopulmonary dysplasia (BPD) in premature newborns with severe lung disease. Although the efficacy of vitamin A supplementation has yet to be clearly confirmed, it is widely employed, at different dosages and schedules, in neonatal intensive care units (NICU). Since in Italy today no suitable formulation of vitamin A is available, the present observational study was designed to define the profile of plasma vitamin A and retinol-binding protein (RBP) concentrations in supplemented infants at risk for BPD admitted to seven Italian NICU. Twelve babies (average 27 weeks of gestation, birth weight 1,008 g), supplemented with vitamin A, were observed with sequential measurements of retinol and RBP up to 28 days of age. At birth retinol and RBP plasma concentrations were both adequate in the infants and half their mothers' levels. During supplementation the levels rose with wide variability according to the differences in dosing and timing in the different units. Plasma levels of retinol and RBP were the same in infants who had BPD and those who did not. A routine standardized therapeutic approach using vitamin A supplementation in Italian NICU will be more than welcome.
The single-dose pharmacokinetics of dexamethasone were studied in 7 extremely low birth weight infants of mean (+/- SD) gestational age 25.6 +/- 0.5 weeks suffering bronchopulmonary dysplasia. A mean peak dexamethasone concentration of 250.5 +/- 70.7 ng/ml was obtained following an intravenous bolus dose (0.369 +/- 0.04 mg/kg dexamethasone) of dexamethasone sodium phosphate. Dexamethasone was measured in plasma by HPLC. Mean clearance (0.143 +/- 0.028 litres/kg/h) was approximately half that reported previously in children and adults, while the half-life (9.26 +/- 3.34 h) was 2- to 3-fold longer than in these patients. The volume of distribution (1.9 +/- 0.483 litres/kg) was larger than reported in a previous study in adults, but was similar to that determined in pediatric and adult patients in another study.
Neonates, especially preterms, are known to have low glomerular filtration rates (GFR). This may result in elevated trough concentrations during multiple administration of aminoglycosides (AGs), potentially leading to nephro- and ototoxic reactions. The once-daily administration (q.d.) of AGs has been shown to be equally or better tolerated in adults and children than the conventional schedules (twice daily, b.i.d.; thrice daily, t.i.d.), while offering potential pharmacodynamic and nursing advantages. No data, however, are available for neonates. As a consequence, this pilot study was conducted in order to assess the tolerance of the once-a-day administration of amikacin in comparison with the twice daily dose regimen, in relation to the pharmacokinetics of the drug under these two schedules. 22 Male neonates (gestational age > or = 34 weeks; postnatal age < or = 2 days) were randomized to receive amikacin (AK) (15 mg/kg/day) q.d. (n = 10) or b.i.d. (n = 12) together with ampicillin (50 mg/kg/12 h). AK plasma levels were measured at days 1, 3, 5 and 7 of treatment just before the next dose (trough level) and 1 h after completion of infusion (peak level) and after 3 and 6 h only at day 1. Due to the small size of the samples, no difference in efficacy could be assessed and was not the aim per se. Glomerular dysfunction was assessed by creatinine clearance, and tubular injuries by the urinary excretion of proteins (retinol binding protein, beta 2-microglobulin, clara cell protein (P1) and microalbumin), enzymes (N-acetyl-beta-D-glucosaminidase, alkaline phosphatase, alanine aminopeptidase, and gamma-glutamyltransferase), and total phospholipids (TPL) in urine. Ototoxicity was assessed by brainstem auditory evoked potentials (BAEPs) at days 0, 3 and 9 of therapy. Eight healthy neonates served as controls. All patients showed a normal and similar increase of GFR during the first postnatal days. Proteinuria did not increase, but enzymuria and TPL increased significantly during the treatment in both AK groups without significant difference between groups. BAEPs at day 9 were not significantly different between treated and untreated patients. We conclude from this pilot study that, in the absence of more toxicity, the q.d. administration of AK in neonates of > or = 34 weeks of gestational age may be recommended over its bid schedule in view of its potential advantages.
A total of 20 newborn piglets age 11.5 +/- 0.3 days and weighing 3.7 +/- 0.1 kg were studied under pentobarbital anesthesia. After stabilization following surgical procedures, baseline values for blood gases, base excess (BE), heart rate (HR), aortic pressure (AoP), left-ventricular contractility (LV dP/dtmax), carotid artery flow (CarF) and renal artery flow (RenF) were measured and normal lactic acid 0.2 ml/kg was infused over 1 h and the same parameters repeated. Then sodium bicarbonate (BC, n = 8), Tris-(hydroxymethyl)aminomethane (THAM, n = 6) or dichloroacetate (DCA, n = 6) were infused over 1 h. The doses of BC and THAM were calculated from the standard formula: Mmol = Base deficit x kg x 0.3. DCA was given at a dose of 300 mg/kg. Following lactic acid infusion, pH was 7.00 +/- 0.4 and BE was -20.6 +/- 1.2. Acidosis was associated with a significant (p < 0.05) increase in AoP (+18.6 +/- 7.4%) and decreases in HR (-13.9 +/- 2.7%) and RenF (-43.8 +/- 10.4%). Values of dP/dtmax and CarF were higher during acidosis in all but 3 animals. Following infusion of alkalizing agents pH and BE values were highest with BC and lowest with DCA and the differences were statistically significant (p < or = 0.05). In general, all three alkalizing agents reversed, in part or completely, the changes in cardiovascular parameters associated with acidosis so that following alkali infusion the changes were not statistically significant when compared to baseline values.(ABSTRACT TRUNCATED AT 250 WORDS)
The present study investigated the effect of cocaine (COC) on cerebral circulation (CBF) and oxidative metabolism (CMRO2) in the newborn piglet and aimed to relate pharmacokinetics of cocaine to cerebrovascular effects. COC decreased CBF and CMRO2 from 75 to 64 and 4.27 to 3.91 ml/min/100 g, respectively, at 4 min with reduced flow to all brain regions (p < 0.05) which returned to baseline by 10 min. COC was rapidly metabolized with a t1/2 of 43 min and peak plasma concentration of 1,172 ng/ml. Norcocaine (NOR) appeared in plasma and CSF within 3 min of cocaine administration and remained elevated for the duration of the study along with COC in the CSF. These data show that the timing of the peak plasma COC level is associated with maximal decreased CBF. Further, the stable elevated level of COC and NOR in the CSF suggests that biotransformation does not occur in the brain. As a result, accumulation of these drugs may occur in the brain with successive COC use and affect the developing CNS in a deleterious manner.
Developmental differences in adrenergic responsiveness may cause age-related changes in the cellular response to ischemia. Standard microelectrode techniques were used in isolated young and adult canine Purkinje fibers to determine the effect of simulated ischemia ([K+]o = 10 mM, pH 6.7, pO2 < 25 mm Hg) alone or with adrenergic stimulation on the rhythmic activity in spontaneously beating Purkinje fibers and on transmembrane potentials and delayed afterdepolarizations in paced Purkinje fibers (basic cycle length = 800-300 ms). The adrenergic agonists used were phenylephrine (5 x 10(-8) M) and isoproterenol (1 x 10(-6) M). For all automatic fibers studied, the control maximum diastolic potential in adults (-96 +/- 1 mV, n = 37) and in the young (-98 +/- 1 mV, n = 36) went to -62 +/- 1 mV during ischemia in both groups and returned to -96 +/- 2 mV with reperfusion. The incidence of rhythmic activity (expressed as percent) during ischemia alone was similar at both ages: adults, 22%; young, 25%. The incidence of ectopic activity with phenylephrine superfusion during ischemia for adults was 63%, an effect blocked by prazosin (1 x 10(-6) M) but not by propranolol (2 x 10(-7) M); the incidence for the young was 25%. Isoproterenol caused ectopic rhythms in 86% of young fibers and 17% of adult fibers (p < 0.05 young vs. adult). During reperfusion the return to control rhythm was slower in adults after ischemia alone or ischemia + alpha-adrenergic stimulation with phenylephrine. There were no age-related differences in the transmembrane potential response of paced fibers to ischemia or reperfusion, and there were no delayed afterdepolarizations with interruption of pacing at 800, 500, or 300 ms in either group. These data suggest that age-related differences in adrenergic responses alter the cellular response to an ischemic insult. To the extent that an ectopic beat may initiate an abnormal rhythm, these differences in sensitivity to adrenergic agonists may lead to developmental differences in arrhythmogenic potential during ischemia.
Antral hyperplasia (AH) induced by prostaglandins (PG) has been described by us recently in 5 infants with cyanotic heart disease receiving the drug. The purpose of the present study was to analyze 14 infants diagnosed as having AH either sonographically or pathologically in an attempt to characterize the dose-response characteristics of this adverse drug reaction, its clinical course and its optimal management. Infants with AH exhibiting large gastric aspirates have received a significantly lower cumulative dose (1,633 +/- 1,266 micrograms/kg) than those presented also with a palpable mass (3,458 +/- 1,703 micrograms/kg), (p < 0.01). While in general there is a dose-related clinical toxicity, variability in the location of the hyperplasia can explain cases of no apparent obstruction despite large cumulative doses of PG. In asymptomatic cases the antral hyperplasia, although visualized, it did not result in gastric outlet obstruction. In all cases followed by us to date, discontinuation of the PG has resulted in resolution of the clinical and sonographic findings. Nasojejunal tube was successfully attempted in several cases, preventing surgery in these very-high-risk infants.
The P-450-associated O-dealkylase activity towards ethoxyresorufin (EROD) and pentoxyresorufin (PROD) was measured in liver microsomes from 1-day-, 1-week-, 4-week-, 6-week-old and adult goats in order to characterize the ontogeny of cytochrome P-450 in this species. The inhibition of these enzyme activities by monoclonal antibodies raised against 3-methyl-cholanthrene-induced (MAb 1-7-1) and phenobarbital-induced (MAb 2-66-3) rat hepatic cytochrome P-450 was used to measure the contribution of the MAb-defined, epitope-specific P-450 to the total activities during these ages of goat development. EROD activity was undetectable until the 1st week of life and increased more than 25-fold by 4 weeks of age. The inhibition of EROD by MAb 1-7-1 increased from 20% in 1-week-old to 70% in adult goats. PROD activity, however, was detectable in the 1-day-old and reached adult levels by 6 weeks of age. The maximal inhibition (40%) of PROD activity by MAb 2-66-3 was demonstrated in 1-day-old goats. The measurement of these enzyme activities and their inhibition by the monoclonal antibodies demonstrated major differences in the ontogeny of these P-450 isozymes in goats. On the other hand, adult goat lung lacked detectable PROD activity, while it expressed approximately one tenth of the EROD activity exhibited by the liver. Over 70% of this activity was inhibitable by MAb 1-7-1.
Indomethacin lowers neonatal cerebral perfusion immediately after intravenous administration. It is important to elucidate whether this reduction is mediated by inhibition of production of prostaglandins, especially prostacyclin, which plays an important role in the autoregulation of the neonatal cerebral vascular bed. We studied changes in cerebral blood flow by serial measurements of temporal mean flow velocity in the anterior cerebral artery (TMFV-ACA), relative cerebral vascular resistance (R-cer), and prostaglandins (measured as changes in thiobarbituric acid reactive substances concentration; TBARS) after a therapeutic dose of 0.1 mg/kg indomethacin administered intravenously for noninvasive closure of patent ductus arteriosus. TMFV-ACA decreased and R-cer increased immediately after the indomethacin administration with a sustained recovery to pre-indomethacin values. The TBARS concentrations, however, did not change during the study period. We conclude that the present study suggests that a therapeutic dose of 0.1 mg/kg of indomethacin has no impact on prostaglandin metabolism.
In order to evaluate parental awareness of the law governing clinical trials in France (Loi Huriet), a study was performed by questionnaire between February and April 1991 in a maternity unit during the days following delivery. The response rate was 59%. 59% of the parents (319/541) were informed of the existence of the law by the media (75%) or their general practitioners (12%). Twenty-one percent (116/541) of the parents would accept the participation of their children in a clinical trial and 74% would refuse. The principal reasons for acceptance were: for the benefit of other children, contribution to medical progress and confidence in physicians. The reasons for refusal were: risk of side effects and unproven efficacy. Parents who would accept had more often received higher education (44%) than parents who would refuse (30%), the latter being less influenced by the explanations of physicians and less willing to accept that a physician should decide for them. Physicians should consider transmitting information directly to parents and indirectly via the media.
