2scmh_oa.indd

T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e Caffeine Therapy for Apnea of Prematurity Barbara Schmidt, M.D., Robin S. Roberts, M.Sc., Peter Davis, M.D., Lex W. Doyle, M.D., Keith J. Barrington, M.D., Arne Ohlsson, M.D., Alfonso Solimano, M.D., and Win Tin, M.D., for the Caffeine for Apnea of Prematurity Trial Group* Background
Methylxanthines reduce the frequency of apnea of prematurity and the need for me- demiology and Biostatistics (B.S., R.S.R., chanical ventilation during the first seven days of therapy. It is uncertain whether A.O.) and Pediatrics (B.S.), McMaster
University, Hamilton, Ont., Canada; methylxanthines have other short- and long-term benefits or risks in infants with
the Department of Obstetrics and very low birth weight.
Gynaecology, University of Melbourne,
Melbourne, Australia (P.D., L.W.D.); Methods
the Department of Pediatrics, McGill University, Montreal (K.J.B.); the We randomly assigned 2006 infants with birth weights of 500 to 1250 g during the Department of Pediatrics, University of first 10 days of life to receive either caffeine or placebo, until drug therapy for apnea Toronto, Toronto (A.O.); the Department of Pediatrics, University of British Colum- of prematurity was no longer needed. We evaluated the short-term outcomes before bia, Vancouver, Canada (A.S.); and the the first discharge home.
Department of Pediatrics, James Cook
University, Middlesbrough, United Results
Kingdom (W.T.). Address reprint re-quests to Dr. Schmidt at the McMaster Of 963 infants who were assigned to caffeine and who remained alive at a postmen-University Medical Center, Rm. 3N11E, strual age of 36 weeks, 350 (36 percent) received supplemental oxygen, as did 447 1200 Main St. W., Hamilton, ON L8N 3Z5, Canada, or at schmidt@mcmaster.ca.
of the 954 infants (47 percent) assigned to placebo (adjusted odds ratio, 0.63; 95 per-cent confidence interval, 0.52 to 0.76; P<0.001). Positive airway pressure was discon- *Members of the Caffeine for Apnea of tinued one week earlier in the infants assigned to caffeine (median postmenstrual Prematurity Trial Group are listed in the Appendix.
age, 31.0 weeks; interquartile range, 29.4 to 33.0) than in the infants in the placebo group (median postmenstrual age, 32.0 weeks; interquartile range, 30.3 to 34.0; P<0.001). Caffeine reduced weight gain temporarily. The mean difference in weight Copyright 2006 Massachusetts Medical Society. gain between the group receiving caffeine and the group receiving placebo was greatest after two weeks (mean difference, −23 g; 95 percent confidence interval, −32 to −13; P<0.001). The rates of death, ultrasonographic signs of brain injury, and necrotizing enterocolitis did not differ significantly between the two groups.
Conclusions
Caffeine therapy for apnea of prematurity reduces the rate of bronchopulmonary dysplasia in infants with very low birth weight. (ClinicalTrials.gov number, NCT00182312.) Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. c a ffeine ther a p y f or a pne a of pr e m at ur i t y cessation of breathing that lasts for more than 15 seconds and is accompanied by hy- poxia or bradycardia - occurs in at least 85 per- cent of infants who are born at less than 34 weeks of gestation.1 Widely used treatments include the application of continuous positive airway pressure and the prescription of a methylxanthine.2 The methylxanthines - aminophylline, theophylline, and caffeine - reduce the frequency of apnea ities likely to affect lifeexpectancy or neurologic and the need for mechanical ventilation during the first seven days of therapy.3 However, it has re- mained uncertain whether methylxanthines have any additional short- and long-term benefits or risks in preterm infants.3-5 Despite this uncertain- ty, methylxanthines have been the mainstay of the pharmacologic treatment of apnea for the past 25 years.2 Methylxanthines are typically prescribed in very preterm infants until they reach a post- menstrual age of 34 to 35 weeks.6 Drug exposure may last even longer. A recent study from the Neo-natal Research Network of the National Institute of Child Health and Human Development showed that among infants with very low birth weight, 44 percent of those with bronchopulmonary dys-plasia and 21 percent of those without such dis-ease were still receiving methylxanthines at a post- The potential for harm exists because methyl- xanthines are inhibitors of adenosine receptors.8 Adenosine preserves brain ATP levels and pro-tects brain cells during experimental hypoxia and ischemia in a variety of animal models.9-11 Meth- ylxanthines also increase oxygen consumption in preterm infants and may therefore diminish growth.12-15 We conducted this randomized, pla-cebo-controlled, multicenter trial of caffeine to study the short- and long-term efficacy and safety of methylxanthine therapy in infants with very Figure 1. Numbers of Infants Who Were Eligible for the Study and Random-
ly Assigned to Receive Caffeine Citrate or Placebo.
Study Infants
Infants with a birth weight of 500 to 1250 g were all clinical centers approved the protocol. Written eligible for enrollment if their clinicians consid- informed consent was obtained from a parent or ered them to be candidates for methylxanthine guardian of each infant. An investigational new therapy during the first 10 days of life. We docu- drug application was filed with Health Canada mented the following reasons why clinicians in- because caffeine is not approved for the treatment tended to use methylxanthines: to prevent apnea, of apnea of prematurity in Canada. Clinical-trial-to treat apnea, or to facilitate the removal of an notification applications were filed in Australia. endotracheal tube. The reasons for exclusion are Appropriate regulatory approvals were obtained listed in Figure 1. The research ethics boards of elsewhere.
Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e Table 1. Baseline Characteristics and Status of the Infants at Randomization and Their Mothers.*
Caffeine Group
Placebo Group
Characteristic
(N = 1006)
(N = 1000)
Antenatal corticosteroids -- no. (%) Infants at birth
Birth weight <10th percentile for gestational age - no. (%)‡ Randomization
saline. This was followed by a daily maintenance A computer-generated randomization scheme was dose of 5 mg per kilogram. If apneas persisted, the used to assign the infants to treatment groups in daily maintenance dose could be increased to a a 1:1 ratio. Randomization was stratified accord- maximum of 10 mg of caffeine citrate per kilo-ing to the study center and balanced in random gram. The maintenance doses were adjusted week-blocks of two or four patients. A designated phar- ly for changes in body weight and could be given macist at each center received a binder contain- orally once an infant tolerated full enteral feed-ing the prespecified sequence of treatment-group ings. The drug was monitored according to its assignments from a statistician at the coordinat- clinical effect only.16 Blood levels of caffeine were ing center who was not otherwise involved in the not measured. Doses of the study drug were held trial. Access to the binder was restricted to selected or reduced for symptoms suggestive of caffeine-pharmacy personnel. The pharmacy study logs induced toxicity (e.g., tachycardia, tachypnea, jit-were retrieved after the completion of recruitment teriness, tremors, and unexplained seizures and to ensure that all randomly assigned infants were vomiting) or for other clinical reasons. The study included in the analysis. Infants were considered drug was discontinued permanently at the discre-to have been randomly assigned at the time the tion of the local clinicians. However, it was recom-first prescription for study drug was signed.
mended to continue therapy with the study drug until the infant had tolerated at least five con- Study Intervention and Cointerventions
secutive days without the use of positive airway As soon as possible after random assignment, eli- pressure.
gible infants received an intravenous loading dose Caffeine citrate for injection was supplied by of either 20 mg of caffeine citrate per kilogram of Sabex. In the single study site in the United States, body weight or an equivalent volume of normal Cafcit (Roxane Laboratories) was used.
Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. c a ffeine ther a p y f or a pne a of pr e m at ur i t y Table 1. (Continued.)
Caffeine Group
Placebo Group
Characteristic
(N = 1006)
(N = 1000)
Infants at randomization
Treatment of documented apnea - no. (%) Facilitation of removal of endotracheal tube - no. (%) Any use of positive airway pressure - no. (%) Therapies between birth and randomization - no. (%) Indomethacin for patent ductus arteriosus Surgical closure of patent ductus arteriosus * Plus-minus values are means ±SD.
† Race or ethnic group was self-reported.
‡ The 10th percentile for gestational age in a normal population has been reported previously.23 To prevent contamination, the study protocol Ascertainment of this outcome is under way and strongly discouraged the use of open-label meth- is expected to be completed by December 2006. ylxanthines. The use of doxapram - a respiratory The steering committee followed a recommenda-stimulant - was also discouraged because of lim- tion by the external safety monitoring committee ited evidence of its safety and efficacy.17 Nonphar- and agreed to analyze the protocol-specified sec-macologic therapies such as continuous positive ondary short-term outcomes after the completion airway pressure were used as necessary to control of recruitment and the initial hospitalization of apnea.
the study infants, including bronchopulmonary An external safety monitoring committee re- dysplasia, ultrasonographic signs of brain injury, viewed the study data every four to six months necrotizing enterocolitis, retinopathy of prematu-during the enrollment phase. With the exception rity, and growth.
of this committee and the study pharmacists, no Bronchopulmonary dysplasia was defined by one involved in the study or in the care and as- the need for supplemental oxygen at a postmen-sessment of outcomes of the infants was aware of strual age of 36 weeks.18 Cranial ultrasonogra-the individual treatment-group assignments.
phy was recommended between the 14th and the 28th days of life, and between 34 and 36 weeks of Outcomes
postmenstrual age if the infant was still hospital- The primary outcome of this study is a composite ized in the study center at that time. The scans of death, cerebral palsy, cognitive delay, deafness, were read locally, and copies of the written reports or blindness at a corrected age of 18 to 21 months. were sent to the coordinating center. The follow- Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e Table 2. Use of Study Drug, Open-Label Methylxanthines, and Cointerventions.
Caffeine Group
Placebo Group
Variable
(N = 1006)
(N = 1000)
Dose reduction due to suspected caffeine-induced toxicity - no. (%) Postmenstrual age at last use of endotracheal tube - wk* Postmenstrual age at last use of positive airway pressure - wk* Postmenstrual age at last use of supplemental oxygen - wk* * This outcome excludes 52 infants in the caffeine group and 55 infants in the placebo group who died. ing lesions are indicative of brain injury and were ing a higher incidence of patent ductus arteriosus
analyzed as a group: intraparenchymal echodense in the placebo group than in the caffeine group,
lesions, cystic periventricular leukomalacia, poren- the external safety monitoring committee recom-
cephalic cysts, and ventriculomegaly with or with- mended to the steering committee in June 2001
out intraventricular hemorrhage.19 Necrotizing that additional data be collected for all infants
enterocolitis was diagnosed during surgery, at au- whose duct was closed surgically. All decisions
topsy, or by a finding of pneumatosis intestina- pertaining to the diagnosis and treatment of a pat-
lis, hepatobiliary gas, or free intraperitoneal air on ent ductus arteriosus were at the discretion of the
radiography. All stages of retinopathy were re- local clinicians.
corded according to the international classifica-
tion.20,21 Weight and head circumference were re- Statistical Analysis
corded weekly.
Assuming an incidence of death or neurodevelop- The use of drug or surgical therapy to close a mental disability of 20 percent, we needed 1000 patent ductus arteriosus was documented but was infants in each group for the study to have a sta-not a prespecified outcome. However, after notic- tistical power of 80 percent to detect a 25 percent Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. c a ffeine ther a p y f or a pne a of pr e m at ur i t y relative reduction in the risk of the primary out- had doses of the study drug withheld or reduced come. The current analyses of secondary outcomes because of clinical symptoms and signs sugges-were adjusted according to study center with the tive of caffeine-induced toxicity.
use of a logistic-regression model that included One hundred ninety infants (9.5 percent) re- terms for treatment and center. The regression co- ceived at least one dose of open-label methyl-efficient associated with treatment in the fitted xanthines, either because of an administrative er-model yielded a point estimate and 95 percent con- ror or intentionally. Of those, 112 infants were fidence interval for the treatment effect expressed switched permanently to open-label methylxan-as an odds ratio. The quotient of the estimated thines at the request of the parents or clinical coefficient and its standard error was used as a staff. A permanent switch to open-label methyl-z-test statistic for the null hypothesis of no treat- xanthines occurred more frequently in the place-ment effect. Mean weight gain and head circum- bo group than in the caffeine group (P = 0.01).
ference were compared between the two groups The administration of positive airway pressure at weekly intervals with the use of Student's t-test. through an endotracheal tube, the use of any Nonparametric tests or Fisher's exact test were positive airway pressure, and oxygen therapy were used as appropriate to analyze the use of the study each discontinued approximately one week ear-drug and cointerventions. All P values are two- lier for infants in the caffeine group than for sided.
infants in the placebo group (P<0.001 for each comparison). In addition, the following cointer- ventions were used less frequently in the caf-feine group than in the placebo group: doxapram, Study Infants
postnatal corticosteroids, and red-cell transfu- The numbers of infants who were deemed eligi- sions (P<0.001 for each comparison). All cointer-
ble for the study and the numbers randomly as- ventions were prescribed at the discretion of the
signed to receive caffeine citrate or placebo are local clinicians.
shown in Figure 1. A total of 2006 infants were
enrolled between October 11, 1999, and October Outcomes
22, 2004 - 994 in Canada, 58 in the United States, Outcomes in the two groups before the first dis-
520 in Australia, and 434 in Europe and Israel. charge home are shown in Table 3. Caffeine sig-
Six infants (four in the caffeine group and two in nificantly reduced the frequency of bronchopul-
the placebo group) did not meet the eligibility cri- monary dysplasia. Of the 963 infants who were
teria but were included in the analysis. All patients assigned to caffeine and who were alive at a post-
were followed to their first discharge home. Long- menstrual age of 36 weeks, 350 (36.3 percent)
term follow-up is under way.22 With the exception received supplemental oxygen, as compared with
of sex, the baseline characteristics of the infants in 447 of the 954 infants (46.9 percent) assigned to
the two groups at birth and of their mothers were placebo (adjusted odds ratio, 0.63; 95 percent con-
similar. The status of the infants at randomiza- fidence interval, 0.52 to 0.76; P<0.001). The rates
tion was also similar in the two groups (Table 1).
of death before the first discharge home, ultraso-nographic signs of brain injury, and necrotizing Study Intervention and Cointerventions
enterocolitis did not differ significantly between The use of the study drug and the use of cointer- the two groups. Adjustments for prespecified and ventions in the two groups are shown in Table 2. prognostically important baseline characteristics Study infants received their first doses of caffeine yielded odds ratios for these outcomes that fur-or placebo at a median postmenstrual age of 28 ther attenuated the associations with the assigned weeks and were weaned off the study drug be- treatment.
fore reaching a median postmenstrual age of 35 During the first three weeks after randomiza- weeks. The median number of days of adminis- tion, infants in the caffeine group gained less tration of the study drug was 37 in the caffeine weight than infants in the placebo group (Fig. 2). group (interquartile range, 24 to 46) and 36 in No significant differences in weight gain were the placebo group (interquartile range, 23 to 46; observed between four and six weeks after ran-P = 0.68). Only 37 infants (1.8 percent) - 23 in the domization. Mean head circumferences in the two caffeine group and 14 in the placebo group - groups remained similar throughout the entire Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e Adjusted
Characteristics
Adjusted
lacia, 16 had porencephalic cysts, and 99 had ventric- 93 had ventriculomegaly (with or without intraventricu- the caffeine group and 838 in the placebo group). A to- ther hospital before their first discharge home. Data on p and 966 in the placebo group). In the caffeine group, caffeine group and one in the placebo group).
enatal administration of corticosteroids, a multiple birth, Unadjusted
1000)
=
(N
1006)
=
(N
Caffeine
Discharge
le 3. Outcomes
Bronchopulmonary dysplasia - no. (%)‡ Drug therapy only for closure of patent ductus arteriosus - Surgical closure of patent ductus arteriosus - no. (%)** The odds ratio has been adjusted for the gestational age and sex of the infant, as well as for the presence or absence of ant and an endotracheal tube at randomization.
This outcome is for infants who were alive at a postmenstrual age of 36 weeks (963 in the caffeine group and 954 in the place This outcome is for infants who were examined for retinopathy in the 35 study centers where the infants were enrolled (822 in tal of 531 infants (53 percent) in the caffeine group and 490 infants (49 percent) in the placebo group were transferred to ano retinal examinations performed after those transfers will be collected at the 18-month follow-up.
This outcome is for infants who underwent cranial ultrasonography at least once after randomization (967 in the caffeine grou 33 infants had intraparenchymal echodense lesions, 24 had cystic periventricular leukomalacia, 6 had a porencephalic cyst, and lar hemorrhage); in the placebo group, 41 infants had intraparenchymal echodense lesions, 37 had cystic periventricular leukoma ulomegaly (with or without intraventricular hemorrhage).
Fourteen infants in each group received ibuprofen, and 282 in the caffeine group and 372 in the placebo group received indome This outcome excludes infants who underwent surgical closure of a patent ductus arteriosus before randomization (five in the Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. c a ffeine ther a p y f or a pne a of pr e m at ur i t y six weeks (data not shown). In a post hoc analysis, the infants randomly assigned to caffeine were significantly less likely to undergo therapy (in par- ticular, surgery) to close a patent ductus arte- riosus than were the infants in the control group We designed this large, multicenter, randomized, placebo-controlled trial of caffeine to resolve the long-standing uncertainty about the short- and long-term efficacy and safety of methylxanthine Mean Weight Change from Birth (g)
therapy for apnea of prematurity.3-5 This article focuses on secondary short-term outcomes; we re- port these early because of their clinical relevance, pending ascertainment of the primary outcome (a composite of death, cerebral palsy, cognitive delay, Weeks since Randomization
deafness, or blindness at a corrected age of 18 to Figure 2. Mean Change in Body Weight during the First Four Weeks
21 months). We found that caffeine substantially after Random Assignment to Caffeine or Placebo.
reduced the frequency of bronchopulmonary dys- The mean difference in weight gain was −16 g after one week (95 percent plasia. Caffeine had the potentially adverse effect confidence interval, −25 to −7), −23 g after two weeks (95 percent confi- of diminishing weight gain for the first three dence interval, −32 to −13), and −13 g after three weeks (95 percent confi- weeks after the start of therapy but had no sig- nificant effects on the rates of death before the first discharge home, ultrasonographic signs of brain macologic therapies, including the use of assisted injury, and necrotizing enterocolitis.
ventilation through an endotracheal tube, nasal Previous data have established that methylxan- continuous positive airway pressure, and supple- thines reduce the frequency of apnea,3 but their mental oxygen.2effects on the duration of assisted ventilation and We stratified the random assignment to caf- their other short- and long-term benefits and risks feine or placebo according to study center, and we (including effects on bronchopulmonary dysplasia, adjusted for center in the analysis to eliminate the brain injury, and growth) have remained uncer- bias caused by the variability of practice among tain.3,4 One published trial enrolling a total of the clinical sites. We found that infants in the 85 infants randomly assigned to either caffeine placebo group required each of three levels of re-or placebo reported 4 patients with necrotizing spiratory support - positive airway pressure with enterocolitis in the caffeine group, as compared an endotracheal tube in place, any positive airway with 2 in the placebo group.24 When the Food and pressure, and supplemental oxygen - for one Drug Administration approved caffeine citrate more week than did infants in the caffeine group. (Cafcit) for the treatment of apnea of prematurity, We speculate that the increased incidence of bron-a warning was included in the label about the pos- chopulmonary dysplasia among infants treated sible association between the use of methylxan- with placebo was caused mainly by their longer thines and the development of necrotizing entero- exposure to positive airway pressure.
colitis.25 The absence of any discernible effect of Caffeine appeared to reduce the frequency of caffeine on the incidence of necrotizing enteroco- a patent ductus arteriosus that was judged by the litis in our study is reassuring.
clinical staff to require closure with drug or sur- We used a placebo to ensure masking of the gical therapy. This finding was unexpected and assignment of the study drug. However, apneas post hoc and must be interpreted cautiously. We that persisted despite optimal use of the study did not include patent ductus arteriosus among drug were treated if they required more than the short-term outcomes in our study protocol be-mild stimulation. Clinicians were instructed to cause it had not been suggested previously that control such apneas with all necessary nonphar- methylxanthine therapy altered the rate of ductal Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. T h e n e w e n g l a n d j o u r n a l o f m e d i c i n e closure in the preterm infant. However, we re- cidence of drug-induced toxicity is low. The rec-corded the use of drug or surgical therapy for ognition that bronchopulmonary dysplasia is an closure of a patent ductus, prompting the exter- important risk factor for neurosensory impair-nal safety monitoring committee to monitor this ment in early childhood26 suggests the potential outcome. Our study protocol did not mandate se- for long-term benefits of caffeine therapy in in-rial echocardiography in all study patients. There- fants with very low birth weights. However, infor-fore, it is uncertain whether caffeine promoted the mation on short-term outcomes is insufficient to closure of a patent ductus arteriosus or whether assess the overall efficacy and risk of neonatal in-the clinical staff were more likely to look for and terventions.27,28 Follow-up of our study cohort to close a patent ductus arteriosus in the placebo the corrected ages of 18 to 21 months and 5 years, group than in the caffeine group because the in- currently in progress, is needed before one can fants in the placebo group required positive air- confidently recommend the standard use of meth-way pressure and supplemental oxygen for a lon- ylxanthine therapy for apnea of prematurity.
ger period than did infants assigned to caffeine. Supported by the Canadian Institutes of Health Research and Additional studies are needed to clarify the effect the National Health and Medical Research Council of Australia, No potential conflict of interest relevant to this article was In summary, caffeine therapy for apnea of pre- reported.
maturity reduces the incidence of bronchopulmo- We are indebted to the study pharmacists, to the nursing and medical staff at all the participating hospitals, and to Sabex in nary dysplasia. Except for a temporary reduction Boucherville, Que., Canada, for providing the caffeine citrate in weight gain, caffeine has no apparent short- injection for the study.
term risks. At the doses used in this trial, the in- appendix
The following investigators, research nurses, and hospitals participated in the neonatal phase of the Caffeine for Apnea of Prematurity Trial (study sites are listed according to the number of infants they enrolled): McMaster University Medical Center, Hamilton, Ont., Canada - B. Schmidt, J. D'Ilario, J. Cairnie; Royal Women's Hospital, Melbourne, Australia - P. Davis, L. Doyle, B. Faber; Women's College Hospital, Toronto - E. Asztalos, L. Golec; Women's and Children's Hospital, Adelaide, Australia - R. Haslam, C. Barnett, L. Goodchild, R. Lontis; Mercy Hospital for Women, Melbourne, Australia - S. Fraser, J. Keng; Centre Hospitalier Universitaire de Que-bec, Quebec, Canada - A. Bairam, S. Ferland, L. Laperriere; Ottawa Hospital, Ottawa - M. Blayney, D. Davis, J. Frank; British Colum-bia Children's Hospital, Vancouver, Canada - A. Solimano, A. Singh, M. Chalmers, K. Ramsay; Academic Medical Center, Amsterdam - M. Offringa, D. Nuytemans, E. Vermeulen; Meir General Hospital, Kfar-Saba, Israel - S. Arnon, A. Chalaf; Mount Sinai Hospital, Toronto - A. Ohlsson, K. Nesbitt; Royal University Hospital, Saskatoon, Sask., Canada - K. Sankaran, S. Morgan; the Brooklyn Hospital Center, Brooklyn, N.Y. - M. LaCorte, P. LeBlanc, A. Braithwaite; Soroka University, Beer Sheva, Israel - A. Golan, T. Barabi; the Canberra Hospital, Canberra, Australia - G. Reynolds, B. Dromgool, S. Meskell; Foothills Hospital, Calgary, Alta., Canada - D. McMillan, D. Schaab, L. Spellen; St. Boniface, Winnipeg, Man., Canada - R. Alvaro, A. Chiu, C. Porter, G. Turner; University Hospital Maastricht, Maastricht, the Netherlands - T. Muler; Kingston General Hospital, Kingston, Ont., Canada - M. Clarke, J. Parfitt; Hotel Dieu Grace Hospital, Windsor, Ont., Canada - C. Nwaesei, L. Kuhn; Ludwig Maximilian University, Munich, Germany - A. Schulze, P. Pudenz, M. Muller; Astrid Lindgren Children's Hospital, Stockholm - H. Lagercrantz, M. Bhiladvala, L. Legnevall; Victoria General Hospital, Victoria, B.C., Canada - D. Matthew, W. Amos, S. Tulsiani; Kaplan Medical Center, Rehovot, Israel - E. Shinwell, R. Levine; Royal Victoria Hospital, Montreal - K. Barrington, T. Kokkotis; James Cook University Hospital, Middlesbrough, United Kingdom - S. Sinha, W. Tin, S. Fritz; University of Sherbrooke, Sherbrooke, Que., Canada - H. Walti, D. Royer; Royal Maternity Hospital, Belfast, Northern Ireland - H. Halliday, D. Millar, A. Berry; Basel Children's Hospital, Basel, Switzerland - H. Fahnenstich, K. Philipp; Moncton Hospital, Moncton, N.B., Canada - R. Canning; Royal Victoria Infirmary, Newcastle, United Kingdom - U. Wariyar, W. Tin, S. Fritz; University Hospital Zurich, Zurich, Switzerland - H. Bucher, J.-C. Fauchere; Neonatal Nursing Initiative, Stockton on Tees, United Kingdom - W. Tin, S. Fritz; University Hospitals of Geneva, Geneva - R. Pfister, V. Launoy; University of Tuebingen, Tuebingen, Germany - C. Poets, P. Urschitz-Duprat; Steering Committee - B. Schmidt (chair), K. Barrington, P. Davis, L.W. Doyle, A. Ohlsson, R.S. Roberts, A. Solimano, W. Tin; External Safety Monitoring Committee - M. Gent (chair), W. Fraser, E. Hey, M. Perlman, K. Thorpe; Consultant Pharmacist - S. Gray; Coordinating and Methods Center in Hamilton, Ont., Canada - R.S. Roberts, C. Chambers, L. Costan-tini, E. McGean, L. Scapinello.
Refe renc e s
1. Barrington K, Finer N. The natural premature infants: sound practice, disas-
pact of a physiologic definition on bron- history of the appearance of apnea of pre- ter, or fruitless byway? J Pediatr 1999;135: chopulmonary dysplasia rates. Pediatrics maturity. Pediatr Res 1991;29:372-5.
2. Martin RJ, Abu-Shaweesh JM, Baird 5. Millar D, Schmidt B. Controversies sur-
8. Fredholm BB. Astra Award Lecture:
TM. Apnoea of prematurity. Paediatr Respir rounding xanthine therapy. Semin Neo- 3. Henderson-Smart DJ, Steer P. Methyl-
6. Eichenwald EC, Aina A, Stark AR. Ap-
xanthine treatment for apnea in preterm nea frequently persists beyond term gesta- 9. Thurston JH, Hauhard RE, Dirgo JA.
infants. Cochrane Database Syst Rev 2001; tion in infants delivered at 24 to 28 weeks. Aminophylline increases cerebral meta- bolic rate and decreases anoxic survival in 4. Schmidt B. Methylxanthine therapy in 7. Walsh MC, Yao Q, Gettner P, et al. Im-
young mice. Science 1978;201:649-51.
Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved. c a ffeine ther a p y f or a pne a of pr e m at ur i t y 10. Boutilier RG. Mechanisms of cell sur-
vival in hypoxia and hypothermia. J Exp fants. Cochrane Database Syst Rev 2004;4: 11. Dunwiddie TV, Masino SA. The role 18. Shennan AT, Dunn MS, Ohlsson A, e35. (Accessed April 21, 2006, at http://
and regulation of adenosine in the central
monary outcomes in premature infants: 2/e35.) prediction from oxygen requirement in 24. Erenberg A, Leff RD, Haack DG, Mos-
12. Gerhardt T, McCarthy J, Bancalari E. the neonatal period. Pediatrics 1988;82:
Effect of aminophylline on respiratory 527-32.
citrate for the treatment of apnea of pre- center activity and metabolic rate in pre- 19. Pinto-Martin JA, Riolo S, Cnaan A, maturity: a double-blind, placebo-con-
mature infants with idiopathic apnea. Holzman C, Susser MW, Paneth N. Cra- nial ultrasound prediction of disabling 644-52.
13. Milsap RL, Krauss AN, Auld PA. Oxy-
and nondisabling cerebral palsy at age 25. New drug application: CAFCIT (NDA)
gen consumption in apneic premature two in a low birth weight population. Pe- infants after low-dose theophylline. Clin diatrics 1995;95:249-54. [Erratum, Pedi- 26. Schmidt B, Asztalos EV, Roberts RS,
14. Carnielli VP, Verlato G, Benini F, 20. An international classification of reti-
et al. Metabolic and respiratory effects of nopathy of prematurity. Pediatrics 1984;74: plasia, brain injury, and severe retinopa- theophylline in the preterm infant. Arch 127-33.
thy on the outcome of extremely low-birth- 21. An international classification of reti-
weight infants at 18 months: results from nopathy of prematurity. II. The classifica- the trial of indomethacin prophylaxis in 15. Bauer J, Maier K, Linderkamp O, tion of retinal detachment. Arch Oph-
Hentschel R. Effect of caffeine on oxygen thalmol 1987;105:906-12. [Erratum, Arch 27. Allen MC. Preterm outcomes research:
consumption and metabolic rate in very Ophthalmol 1987;105:1498.]
a critical component of neonatal intensive low birth weight infants with idiopathic 22. Schmidt B. Methylxanthine therapy care. Ment Retard Dev Disabil Res Rev
apnea. Pediatrics 2001;107:660-3.
for apnea of prematurity: evaluation of 2002;8:221-33.
16. Pesce AJ, Rashkin M, Kotagal U. Stan-
treatment benefits and risks at age 5 years 28. Ambalavanan N, Whyte RK. The mis-
dards of laboratory practice: theophylline in the international Caffeine for Apnea of match between evidence and practice: and caffeine monitoring. Clin Chem 1998; Prematurity (CAP) trial. Biol Neonate 2005; 17. Henderson-Smart D, Steer P. Doxa-
23. Kramer MS, Platt RW, Wen SW, et al. Copyright 2006 Massachusetts Medical Society.
POWERPOINT SLIDES OF JOURNAL FIGURES AND TABLES
At the Journal's Web site, subscribers can automatically create PowerPoint slides
of Journal figures and tables. Click on a figure or table in the full-text version
of any article at www.nejm.org, and then click on PowerPoint Slide for Teaching.
A PowerPoint slide containing the image, with its title and reference citation, can
then be downloaded and saved.
Downloaded from www.nejm.org at HAUPTBIBLIOTHEK UNIVERSITAET on February 21, 2008. Copyright 2006 Massachusetts Medical Society. All rights reserved.

Source: http://www.neonatologie.usz.ch/Documents/LehreUndForschung/Publikationen/Schmidt_2006_Multicenterstudy_CaffeinetherapyforapneaofPrematuritiy.pdf