ADC Online First, published on January 27, 2016 as 10.1136/archdischild-2015-309396 Original article
Adverse drug reactions in neonates: a prospective study Ana Belén Rivas,1,2 Luis Arruza,3,4 Enrique Pacheco,5 Antonio Portoles,4,6 Jorge Diz,5 Emilio Vargas2,4 For numbered affiliations see end of article. Correspondence to Ana Belén Rivas Paterna, Nursery Department, UICEC, Hospital Clínico San Carlos, Facultad de Enfermería Fisioterapia y Podología, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Profesor Martín Lagos SN, Madrid 28040, Spain; [email protected] Received 23 July 2015 Revised 28 December 2015 Accepted 29 December 2015
ABSTRACT Aim To investigate the frequency and characteristics of adverse drug reactions (ADRs) in hospitalised neonates to obtain a better understanding of and improvement in neonatal healthcare. Methodology A prospective cohort study. Data were collected on 313 neonates and 2166 drug prescriptions. Clinical characteristics of patients, drugs administered and ADRs were prospectively recorded and analysed. Informed consent was obtained in all cases. Results 116 different ADRs were detected. 17% of the neonates experienced at least one of these ADRs. Systemic antimicrobials and caffeine citrate were the drugs that most commonly caused ADRs. According to the ADR Severity Assessment Scale, 41% were mild, 42% were moderate and 17% were severe. Of the ADRs identified, 11% were classified as ‘certain’ by the Naranjo method and 20% were classified as ‘defined’ by the Karch and Lasagna modified algorithm. Most of the ADRs detected were related to feed intolerance, phlebitis and tachycardia. Most were acute (73%) and lasted between 1 and 7 days (39%). After the occurrence of an ADR, it was necessary to initiate specific treatment in 44 cases, discontinue the drugs involved in 30 cases, and reduce the drug dose in another 30 cases. An association was shown between the number of drugs prescribed and ADR onset. Conclusions There is a high incidence of ADRs in hospitalised newborns, which increases with the number of prescriptions.
BACKGROUND
To cite: Belén Rivas A, Arruza L, Pacheco E, et al. Arch Dis Child Published Online First: [ please include Day Month Year] doi:10.1136/archdischild2015-309396
The difficulties associated with pharmacological research in neonates has resulted in little drug information, which may compromise drug safety and efficacy.1–5 Pharmacovigilance is particularly important in newborns6–8 because of their increased susceptibility to adverse drug reactions (ADRs)9–11 and the frequent overuse of drugs without specific information related to neonates.12 13 ADRs, as established by regional regulations, guidance and practices, concern harmful and unintended responses to a medicinal product.14 Symptoms related to ADRs in newborns are often non-specific.15 Therefore, specific training of healthcare professionals to improve ADR awareness, diagnosis, management and prevention is particularly important.7 16 A large number of studies have confirmed the high incidence of ADRs in adults, which is estimated to be around 35%.17–19 It would therefore be reasonable to expect a higher number of ADRs
What is already known on this topic ▸ The physiological immaturity and continuous state of development of newborns make them more susceptible to drugs and their toxicity. ▸ Neonates have special characteristics that increase their risk of having adverse drug reactions (ADRs) and decrease their ability to tolerate them. ▸ The lack of pharmacological research in neonates has resulted in insufficient drug information, which can compromise drug safety and efficacy.
What this study adds ▸ 17% (one in six) neonates experienced an ADR. ▸ More than one-third of ADRs (44%) needed specific treatment. ▸ Gastrointestinal problems were the most common ADR.
in newborns, although only a few studies have analysed this population.10 20–22 Aranda23 identified 326 neonates with ADRs out of 1200 patients from a neonatal intensive care unit (NICU) in Canada using a system of intensive pharmacological surveillance. Le et al24 performed a retrospective study that analysed hospitalised children and identified an annual incidence of ADRs ranging from 0.4% to 2.3%. More recently, Kunac et al25 explored adverse drug events in paediatric patients and found that most occurred in newborns admitted to the NICU (12.28%). However, both studies only included a small number of neonates. In order to create safer and more effective pharmacotherapy,16 the objective of this study was to investigate the frequency and characteristics of ADRs in hospitalised neonates to obtain a better understanding of and improvement in neonatal healthcare.
METHODS Study design and participants A prospective cohort study was carried out in neonates admitted to the neonatal department at the Hospital Clínico San Carlos in Madrid (Spain) from May 2012 to June 2013.
Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
1
Copyright Article author (or their employer) 2016. Produced by BMJ Publishing Group Ltd (& RCPCH) under licence.
Original article All admissions to neonatal intermediate care and the NICU were eligible for inclusion. Lack of information about the drugs administered to the baby before inclusion in the study was the only exclusion criterion.
ADRs for which the association with the suspected drug was considered to be doubtful or improbable were excluded from the analysis, as they were considered irrelevant to the objective and a potential source of bias in the estimation of ADR incidence.
Variables and patient follow-up At the time of admission, demographic characteristics (gestational age, birth weight, height, head circumference, intrauterine growth) were collected. An intensive pharmacological surveillance protocol was applied from the day of admission to discharge, with daily evaluation of each patient’s charts and notes by the ADR-monitoring team with the purpose of evaluating drug prescriptions and events potentially related to medication. Taking into consideration the International Conference on Harmonisation (ICH) guidelines,14 any deviation of the expected clinical status (signs, symptoms and other clinical and laboratory findings) of the patient was analysed as a potential ADR (regardless of whether it was due to drug–drug interaction, medication error, or other). In addition, when a new drug was prescribed, the following information was collected: name of the drug and anatomical therapeutic chemical (ATC) classification, indication, dosage, pharmaceutical form, route of administration and duration of treatment. When an ADR was suspected, supplementary information was requested from the healthcare professional in charge of the patient and from the parents. The monitoring team evaluated severity, suspected mechanism of ADR development, cause– effect relationship, duration and onset of the symptoms, need for treatment, dose modification or drug discontinuation, and patient outcome. ▸ The severity of the ADR was categorised as mild (uncomplicated primary disease, no treatment required and drug discontinuation not necessary), moderate (signs and symptoms occur but vital organ systems are not affected), or severe (life-threatening symptoms, organ system dysfunction, reduced life expectancy or death).26 ▸ With regard to possible mechanisms involved, type A (augmented) ADRs are those that can be expected from the mechanism of action of the drug, with a clear dose–effect relationship. Type B (bizarre) reactions are idiosyncratic— that is, not related to drug action.27 ▸ Causality was evaluated according to the Naranjo et al28 method and the Karch and Lasagna algorithm.29 These two well-known causality methods were used because of the lack of a validated algorithm for neonates. Both consist of a questionnaire-based score that includes previous bibliographic information, temporal sequence, effect of drug concentrations and dose dependence, and patient history. The scores obtained were pooled according to the association of the ADR with the suspected drug. Probability is scored as definite (≥9), probable (5–8), possible (1–4) or doubtful (0) by the Naranjo algorithm. According to the Karch and Lasagna classification, probability is stratified as conditional (1–3), possible (4–5), probable (6–7) and defined (≥8). When more than one drug caused an ADR, the drug that received the highest score in this causality evaluation system was considered to be the one responsible. All drug orders were recorded daily on the study forms by the ADR team. Data input was verified by the nurse or physician. All the information collected was stored in a database specifically designed for this purpose, and finally a data matrix related to the variables was produced. Thanks to the collaboration of healthcare professionals and parents, all data were correctly recorded. 2
Statistical analysis A sample size of 310 infants was estimated to be necessary based on the reported incidence of ADRs in similar areas (27.17%).23 It was calculated with a 95% CI and a precision of ±5% units without replacement rate, using the GRANMO sample size calculation programme. The frequency distribution of the qualitative variables is presented with 95% CIs. Data distributions for the quantitative variables are expressed as mean or median values. Bivariate comparative analysis between patients with and without ADRs was performed by χ2 test for qualitative variables and by comparison of means for quantitative variables. Data were analysed using SPSS software, and p values ≤0.05 were considered significant.
Ethical considerations The study protocol was approved by the ethics committee of the Hospital Clínico San Carlos, and parental informed consent was obtained in all cases before patient inclusion in the study.
RESULTS The recruitment began in May 2012 and finished in June 2013. A total of 322 newborns were admitted to the neonatal unit during this period. However, nine were not included in the study: informed consent was not obtained from eight parents, and one baby was transferred from another hospital making it impossible to collect all the information about his previous pharmacological treatment. Therefore, a total of 313 patients were studied (140 girls and 173 boys), and 2166 drug prescriptions were analysed. The median age of all patients was 0 days, with 75% of patients ≤1 day, 22% of patients 1–7 days, and 3% of patients 7–29 days. The mean±SD age at admission was 0.9±2.9 days. Patients from 24+0 to 41+6 weeks of gestational age were included in the study (mean gestational age of 36±4.2 weeks and median 37 weeks (Q1–Q3 34–39)). The median birth weight was 2470 g (Q1–Q3 500–4410), and the mean was 2485±868.6 g. Overall, 99 patients were admitted to the NICU and 214 to intermediate care; 210 (67%) were born at term and 10% (32) had intrauterine growth retardation. Sixty-nine (22%) patients did not receive any medication, and the mean number of drugs taken was seven per patient (median=1). Taking into consideration all the prescriptions, the mean duration of drug treatment was 5.5±9.3 days. Overall, 700 (32%) prescriptions were administered as a single dose, and the mean duration of treatment with multiple doses was 8.1 days. The most common interval between doses was ‘every 24 h’ (543, 25%). The route of administration most commonly used was intravenous (1113, 51%), followed by enteral (419, 19%) and intramuscular (259, 12%). At least one ADR was experienced by 54 patients (17%; 20 girls and 34 boys), and a total of 116 different ADRs were detected, which corresponds to 5% of the 2166 drug prescriptions. Of the patients who received medication, 22% experienced an ADR. The majority (94) of the ADRs were experienced in the NICU. Most of the ADRs detected were related to feed intolerance (19), phlebitis (10) and tachycardia (10) (table 1). Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
Original article The ADRs were assessed, and only 11% of the ADRs identified were labelled ‘definitive’, while 57% were ‘probable’ and 32% ‘possible’ according to the Naranjo method. With the Karch and Lasagna modified algorithm, 20% of the ADRs were considered to be ‘defined’, 34% ‘probable’, 28% ‘possible’, and only 18 were considered to be ‘conditional’. For eight of the identified ADRs, there was insufficient information to determine whether they were ADRs or not (ADRs for which the association with the suspected drug was considered to be doubtful or improbable), so they were excluded from the analysis. The majority of ADRs were moderate (49, 42%), 47 (41%) were mild, and 20 were severe, including 3 fatalities. Details of the 20 severe ADRs are shown in table 2. Most ADRs were acute (73%) and lasted between 1 and 7 days (39%). Of the ADRs identified, 87 (75%) were expected according to the drug action spectrum (type A). After the
Table 1 List of the most common adverse drug reactions identified and the causative drugs Adverse drug reaction Feeding intolerance
Phlebitis
Clinical manifestations
Causative drug
N
Presence of gastric residual accompanied by abdominal distension, bilious vomiting, food vomiting or discomfort
Ibuprofen Ferroglycine sulfate Caffeine citrate Paracetamol Ampicillin Multivitamin Vitamin D Zidovudine Vancomycin Fentanyl Ampicillin Cefazolin Cefotaxime Meropenem Trimethoprim sulfamethoxazole Caffeine citrate Salbutamol Dobutamine Dopamine Caffeine citrate Salbutamol Cyclopentolate +phenylephrine Ferroglycine sulfate Fentanyl Linezolid Amphotericin B Micafungin sodium Ampicillin Cefotaxime Caffeine citrate
6 5 2 2 1 1 1 1 3 2 1 1 1 1 1
Caffeine citrate Dobutamine Dexamethasone Rocuronium
3 1 1 1
Meropenem Ampicillin Teicoplanin Paracetamol Ibuprofen Methimizole Caffeine citrate
2 2 1 1 3 1 1
Localised redness and swelling, pain or burning along the length of the vein or vein being hard and cord-like
Tachycardia
Rate >180 beats/min
Central hyperactivity
Irritability, restlessness and jitteriness
Constipation Thrombocytopenia
Jaundice
Hypertension
Liver failure
Gastrointestinal haemorrhage
Time without defecation >48 h or rare and hard stools Platelet count 7.1 mg/dL) (table 1). Ibuprofen was the second most commonly prescribed drug related to ADRs: six patients experienced two or more digestive symptoms associated with this drug (including gastrointestinal haemorrhage, vomiting, constipation, feed intolerance and abdominal discomfort), while three only had gastrointestinal haemorrhage, two patients had necrotising enterocolitis, and another two had renal failure with high levels of serum creatinine. Likewise, ferroglycine sulfate was involved in seven cases of constipation and five cases of digestive disorder (vomiting, abdominal discomfort and feed intolerance). Other medicines associated with ADRs are listed in table 4.
DISCUSSION This study encompasses an exhaustive compilation and analysis of ADRs diagnosed in a cohort of 313 hospitalised neonatal patients; 2166 prescriptions were analysed using a systematic and intensive pharmacological surveillance protocol. In agreement with previous reports, antibiotics were the group of drugs most commonly prescribed.30 31 In contrast, caffeine citrate was the single drug most frequently used in our study, while previous reports indicated that multivitamins were more commonly prescribed.32 A high incidence of ADRs (17% of the patients) was observed, and 75% of all the ADRs found could be explained by the mechanism of action of the drug (type A) and hence could have been expected and potentially prevented. The incidence of ADRs found in our study is consistent with previous reports.22 23 33 However, it is extremely difficult to compare different studies because the estimation of ADR incidence is greatly influenced by the definitions used, the methodology of detection and classification, and the study setting. In this study, we used the definition established by the ICH guidelines, intensive pharmacology surveillance to detect ADRs, and standard methods to measure cause–effect relationships and the impact of side effects.14 28 29 34–36 Previous reports with similar methodology detected an incidence of 27.1% in NICU patients,23 while, in the present study, an incidence of 44.4% was observed in these particular patients. Nevertheless, the calculated incidence is lower than that published for adults.17–19 This might Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
Original article be explained by the present study not considering ADRs with low causality scores. Other factors that could underestimate ADR incidence in neonates are the difficulty of diagnosis and the lack of specific bibliography and tools. In addition, our bivariate analysis shows that ADR incidence increases with the number of prescriptions per patient, as has already been demonstrated in previous studies carried out in the NICU, where 15–30% of neonates who received more than 10 drugs experienced at least one ADR.23 37 The main limitation of this study is the determination of causality. It is important to note that only 11% of ADRs found were labelled as ‘sure’28 or 20% as ‘defined’.29 However, in order to classify an ADR as certain by applying any of the methods mentioned, it would be necessary to readminister the drug, to determine serum levels, or to administer a placebo, which is not possible or even ethical in most of the cases in neonatal care. In addition, although a high number of ADRs (62%) were considered probable or possible,29 it must be take into consideration that there might be potential bias related to ADR classification, as some true ADRs may have been excluded because of low causality scores, leading to underestimation of the ADR rate. Overall, the results of this analysis are consistent with the few existing publications.22 23 30–33 36–39 The differences detected were expected because of variation in patient characteristics, the system of evaluation carried out, and the pharmacotherapies used.
Data sharing statement Descriptions of the neonatal drugs used can be obtained by sending an email to the main author.
REFERENCES 1 2 3
4 5
6
7
8 9
10
11
CONCLUSIONS In summary, this study demonstrates a high incidence of ADRs in hospitalised neonates, which seems to increase with the number of prescriptions. Further refinement and additional testing are needed to confirm this. The extreme vulnerability of the newborn baby, the frequent need for medications, along with the lack of accurate and reliable information with regard to neonatal pharmacotherapy, as reflected by the incidence of ADRs, makes the development of pharmacological research in neonatology imperative. Author affiliations 1 Nursery Department, UICEC, Hospital Clínico San Carlos, Facultad de Enfermería Fisioterapia y Podología, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain 2 Facultad de Enfermería, Fisioterapia y Podología, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain 3 Neonatology Department, Hospital Clínico San Carlos, Madrid, Spain 4 Departamento de Pediatría, Facultad de Medicina, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain 5 Nursery Department, Facultad de Enfermería Fisioterapia y Podología, Universidad Complutense de Madrid, Madrid, Spain 6 Clinical Pharmacology Department, Hospital Clínico San Carlos, Madrid, Spain Acknowledgements We thank the Facultad de Enfermería, Fisioterapia y Podología (Universidad Complutense de Madrid) for support. We are also grateful to Ana I Soriano for helping with the translation of this paper.
12 13 14
15 16
17 18 19 20
21
22 23 24
Contributors ABR, EV and LA contributed to the design of the study, carried out data collection and analysis, and performed most of the writing. EP contributed to the design of the study and data analysis and revision of the manuscript. AP and JD carried out data collection and data entry and critically reviewed and revised the manuscript. All authors approved the final manuscript as submitted.
25
Funding The help and support of the health professional from the Neonatology Department and Clinical Pharmacology Department of the Hospital Clínico San Carlos were critical in allowing the study to take place.
27
Competing interests None declared.
28
Ethics approval Ethics committee of the Hospital Clínico San Carlos. Provenance and peer review Not commissioned; externally peer reviewed. Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
26
29
Turner MA. Neonatal drug development. Early Hum Dev 2011;87:763–8. Jain L. The challenge of managing drugs safely in the newborn. Clin Perinatol 2012;39:XV–XVI. Mason J, Pirmohamed M, Nunn T. Off-label and unlicensed medicine use and adverse drug reactions in children: a narrative review of the literature. Eur J Clin Pharmacol 2012;68:21–8. Schirm E, Tobi H, de Vries TW, et al. Lack of appropriate formulations of medicines for children in the community. Acta Paediatr 2003;92:1486–9. Walsh J, Mills S. Conference report: formulating better medicines for children: 4th European Paediatric Formulation Initiative conference. Ther Deliv 2013; 4:21–5. European Medicines Agency: Committee for Medicinal Products for Human Use (CHMP). Guideline On Conduct Of Pharmacovigilance For Medicines Used By The Paediatric Population. http://www.ema.europa.eu/docs/en_GB/document_library/ Scientific_guideline/2009/09/WC500003764.pdf (accessed 15 Jul 2015). Juárez OH, Pérez GG, Flores J. Pharmacovigilance and pharmacoepidemiology of drugs in a Mexican pediatric hospital. A proposed guide. Pharmacy World Sci 2007;29:43–6. Sammons HM, Choonara I. What is happening to improve drug therapy in children? Paediatr Int Child Health 2007;17:108–110. Speranza N, Lucas L, Telechea H, et al. Reacciones adversas a medicamentos en niños hospitalizados: un problema de salud pública. Rev Med Urug 2008;24:161–66. Thiesen S, Conroy EJ, Bellis JR, et al. Incidence, characteristics and risk factors of adverse drug reactions in hospitalized children—a prospective observational cohort study of 6,601 admissions. BMC Med 2013;11:237. Gray JE, Goldmann DA. Medication errors in the neonatal intensive care unit: special patients, unique issues. Arch Dis Child Fetal Neonatal Ed 2004; 89:F472–3. Conroy S, Choonara I, Impicciatore P, et al. Survey of unlicensed and off label drug use in paediatric wards in European countries. BrMed J 2000;320:79–82. Kemper EM, Merkus M, Wierenga PC, et al. Towards evidence-based pharmacotherapy in children. Paediatr Anaesth 2011;21:183–9. ICH Guideline E2D; Post-approval safety data management: definitions and standards for expedited reporting, 3.1.1 Spontaneous reports 2004. International Conference on Harmonisation. http://www.ich.org/fileadmin/Public_Web_Site/ICH_ Products/Guidelines/Efficacy/E2D/Step4/E2D_Guideline.pdf (accessed 29 Oct 2015). Dabliz R, Levine S. Medication safety in neonates. Am J Perinatol 2012;29:49–56. World Heath Organization. Promoting Safety of Medicines for Children. http://www. who.int/medicines/publications/essentialmedicines/Promotion_safe_med_childrens. pdf (accessed 15 Jul 2015). Vallano A, Agustí A, Pedrós C. Systematic review of studies assessing the cost of adverse drug reactions. Gac Sanit 2012;26:277–83. BordaI T, Slone D, Jick H. Assessment of adverse reactions within a drug surveillance program. JAMA 1968;205:645–7. Vargas E, Terleira A, Hernando F, et al. Effect of adverse drug reactions on length of stay in surgical intensive care units. Crit Care Med 2003;31:694–8. Impicciatore P, Choonara I, Clarkson A, et al. Incidence of adverse drug reactions in paediatric in/out-patients: a systematic review and meta-analysis of prospective studies. Br J Clin Pharmacol 2001;52:77–83. Pirmohamed M, James S, Meakin S, et al. Adverse drug reactions as cause of admission to hospital: prospective analysis of 18.820 patients. BMJ 2004;329:15–19. Martínez-Mir I, García-López M, Palop V, et al. A prospective study of adverse drug reactions in hospitalized children. Br J Clin Pharmacol 1999;47:681–688. Aranda JV. Factors associated with adverse drug reactions in the newborn. Pediatric Pharmacology New York 1983;3:245–9. Le J, Nguyen T, Law AV, et al. Adverse drug reactions among children over a 10-year period. Pediatrics 2006;118;555–62. Kunac DL, Kennedy J, Austin N, et al. Incidence, preventability, and impact of Adverse Drug Events (ADEs) and potential ADEs in hospitalized children in New Zealand: a prospective observational cohort study. Paediatr Drugs 2009;11:153. Jucker E, Venulet J. Methods of monitoring adverse reactions to drugs. Prog Drug Res 1977;21:231–92. Rawlins MD, Thompson JW. Pathogenesis of adverse drug reactions. In: Davies DM, Ferney RE, Glanville DE, eds. Textbook of adverse drug reactions. 4th edn. New York: Oxford University Press 1991:16–38. Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 1981;30:239–45. Farina JC, Krup P, Tobler HJ. Computer tools in Spontaneous reporting of adverse drug reactions. A multitudinal company approach. In: Ducrot H, Goldberg M,
5
Original article
30 31
32 33 34
6
Hoigné R, et al. Conference on computer aid to drug therapy and to drug monitoring. Amsterdam: North-Holland Publishing Company 1998:117–23. López R, Cabanas MJ, Oliveras M, et al. Utilización de medicamentos en una UCI neonatal: estudio prospectivo. Farm Hosp 2005;29:26–9. Morales-Olivas FJ, Martínez-Mir I, Ferrer JM, et al. Adverse drug reactions in children reported by means of the yellow card in Spain. J Clin Epidemiol 2000;53:1076–80. Feal CB, Barroso PC, Carcelen AJ, et al. Drug use in neonatology units of 6 Spanish hospitals. Farm Hosp 2003;27:69–71. Lazarou J, Pomeraz B, Corey PN. Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies. JAMA 1998;279:1200–5. Real Decreto 577/2013, de 26 de julio por el que se regula la Farmacovigilancia de Medicamentos de Uso Humano. BOE 2013;179:55066–92. http://www.boe.es/boe/ dias/2013/07/27/pdfs/BOE-A-2013-8191.pdf (accessed 16 Jul 2015).
35
36
37
38 39
Real Decreto 1015/2009, de 19 de junio, por el que se regula la disponibilidad de medicamentos en situaciones especiales. BOE 2009;174:60904–14. http://www. boe.es/boe/dias/2009/07/20/pdfs/BOE-A-2009-12002.pdf (accessed 16 Jul 2015). Von Euler M, Eliason E, Ohlen G, et al. Adverse drug reactions causing hospitalization can be monitored from computerized medical records and thereby indicate the quality of drug utilization. Pharmacoepidemiol Drug Saf 2006;15:179–84. Aagaard L, Weber CB, Hansen EH. Adverse drug reactions in the paediatric population in Denmark: a retrospective analysis of reports made to the Danish Medicines Agency from 1998 to 2007. Drug Saf 2010;33:327–39. Bárzaga Z, Novoa D, Caballero M, et al. Adverse drug reactions in children in Camagüey Province, Cuba. Arch Dis Child 2010;95:474–7. Bárzaga Z, López A, Mejías Y, et al. Pharmacovigilance in children in Camagüey Province, Cuba. Eur J Clin Pharmacol 2012;68:1079–84.
Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
Adverse drug reactions in neonates: a prospective study Ana Belén Rivas,1,2 Luis Arruza,3,4 Enrique Pacheco,5 Antonio Portoles,4,6 Jorge Diz,5 Emilio Vargas2,4 For numbered affiliations see end of article. Correspondence to Ana Belén Rivas Paterna, Nursery Department, UICEC, Hospital Clínico San Carlos, Facultad de Enfermería Fisioterapia y Podología, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Profesor Martín Lagos SN, Madrid 28040, Spain; [email protected] Received 23 July 2015 Revised 28 December 2015 Accepted 29 December 2015
ABSTRACT Aim To investigate the frequency and characteristics of adverse drug reactions (ADRs) in hospitalised neonates to obtain a better understanding of and improvement in neonatal healthcare. Methodology A prospective cohort study. Data were collected on 313 neonates and 2166 drug prescriptions. Clinical characteristics of patients, drugs administered and ADRs were prospectively recorded and analysed. Informed consent was obtained in all cases. Results 116 different ADRs were detected. 17% of the neonates experienced at least one of these ADRs. Systemic antimicrobials and caffeine citrate were the drugs that most commonly caused ADRs. According to the ADR Severity Assessment Scale, 41% were mild, 42% were moderate and 17% were severe. Of the ADRs identified, 11% were classified as ‘certain’ by the Naranjo method and 20% were classified as ‘defined’ by the Karch and Lasagna modified algorithm. Most of the ADRs detected were related to feed intolerance, phlebitis and tachycardia. Most were acute (73%) and lasted between 1 and 7 days (39%). After the occurrence of an ADR, it was necessary to initiate specific treatment in 44 cases, discontinue the drugs involved in 30 cases, and reduce the drug dose in another 30 cases. An association was shown between the number of drugs prescribed and ADR onset. Conclusions There is a high incidence of ADRs in hospitalised newborns, which increases with the number of prescriptions.
BACKGROUND
To cite: Belén Rivas A, Arruza L, Pacheco E, et al. Arch Dis Child Published Online First: [ please include Day Month Year] doi:10.1136/archdischild2015-309396
The difficulties associated with pharmacological research in neonates has resulted in little drug information, which may compromise drug safety and efficacy.1–5 Pharmacovigilance is particularly important in newborns6–8 because of their increased susceptibility to adverse drug reactions (ADRs)9–11 and the frequent overuse of drugs without specific information related to neonates.12 13 ADRs, as established by regional regulations, guidance and practices, concern harmful and unintended responses to a medicinal product.14 Symptoms related to ADRs in newborns are often non-specific.15 Therefore, specific training of healthcare professionals to improve ADR awareness, diagnosis, management and prevention is particularly important.7 16 A large number of studies have confirmed the high incidence of ADRs in adults, which is estimated to be around 35%.17–19 It would therefore be reasonable to expect a higher number of ADRs
What is already known on this topic ▸ The physiological immaturity and continuous state of development of newborns make them more susceptible to drugs and their toxicity. ▸ Neonates have special characteristics that increase their risk of having adverse drug reactions (ADRs) and decrease their ability to tolerate them. ▸ The lack of pharmacological research in neonates has resulted in insufficient drug information, which can compromise drug safety and efficacy.
What this study adds ▸ 17% (one in six) neonates experienced an ADR. ▸ More than one-third of ADRs (44%) needed specific treatment. ▸ Gastrointestinal problems were the most common ADR.
in newborns, although only a few studies have analysed this population.10 20–22 Aranda23 identified 326 neonates with ADRs out of 1200 patients from a neonatal intensive care unit (NICU) in Canada using a system of intensive pharmacological surveillance. Le et al24 performed a retrospective study that analysed hospitalised children and identified an annual incidence of ADRs ranging from 0.4% to 2.3%. More recently, Kunac et al25 explored adverse drug events in paediatric patients and found that most occurred in newborns admitted to the NICU (12.28%). However, both studies only included a small number of neonates. In order to create safer and more effective pharmacotherapy,16 the objective of this study was to investigate the frequency and characteristics of ADRs in hospitalised neonates to obtain a better understanding of and improvement in neonatal healthcare.
METHODS Study design and participants A prospective cohort study was carried out in neonates admitted to the neonatal department at the Hospital Clínico San Carlos in Madrid (Spain) from May 2012 to June 2013.
Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
1
Copyright Article author (or their employer) 2016. Produced by BMJ Publishing Group Ltd (& RCPCH) under licence.
Original article All admissions to neonatal intermediate care and the NICU were eligible for inclusion. Lack of information about the drugs administered to the baby before inclusion in the study was the only exclusion criterion.
ADRs for which the association with the suspected drug was considered to be doubtful or improbable were excluded from the analysis, as they were considered irrelevant to the objective and a potential source of bias in the estimation of ADR incidence.
Variables and patient follow-up At the time of admission, demographic characteristics (gestational age, birth weight, height, head circumference, intrauterine growth) were collected. An intensive pharmacological surveillance protocol was applied from the day of admission to discharge, with daily evaluation of each patient’s charts and notes by the ADR-monitoring team with the purpose of evaluating drug prescriptions and events potentially related to medication. Taking into consideration the International Conference on Harmonisation (ICH) guidelines,14 any deviation of the expected clinical status (signs, symptoms and other clinical and laboratory findings) of the patient was analysed as a potential ADR (regardless of whether it was due to drug–drug interaction, medication error, or other). In addition, when a new drug was prescribed, the following information was collected: name of the drug and anatomical therapeutic chemical (ATC) classification, indication, dosage, pharmaceutical form, route of administration and duration of treatment. When an ADR was suspected, supplementary information was requested from the healthcare professional in charge of the patient and from the parents. The monitoring team evaluated severity, suspected mechanism of ADR development, cause– effect relationship, duration and onset of the symptoms, need for treatment, dose modification or drug discontinuation, and patient outcome. ▸ The severity of the ADR was categorised as mild (uncomplicated primary disease, no treatment required and drug discontinuation not necessary), moderate (signs and symptoms occur but vital organ systems are not affected), or severe (life-threatening symptoms, organ system dysfunction, reduced life expectancy or death).26 ▸ With regard to possible mechanisms involved, type A (augmented) ADRs are those that can be expected from the mechanism of action of the drug, with a clear dose–effect relationship. Type B (bizarre) reactions are idiosyncratic— that is, not related to drug action.27 ▸ Causality was evaluated according to the Naranjo et al28 method and the Karch and Lasagna algorithm.29 These two well-known causality methods were used because of the lack of a validated algorithm for neonates. Both consist of a questionnaire-based score that includes previous bibliographic information, temporal sequence, effect of drug concentrations and dose dependence, and patient history. The scores obtained were pooled according to the association of the ADR with the suspected drug. Probability is scored as definite (≥9), probable (5–8), possible (1–4) or doubtful (0) by the Naranjo algorithm. According to the Karch and Lasagna classification, probability is stratified as conditional (1–3), possible (4–5), probable (6–7) and defined (≥8). When more than one drug caused an ADR, the drug that received the highest score in this causality evaluation system was considered to be the one responsible. All drug orders were recorded daily on the study forms by the ADR team. Data input was verified by the nurse or physician. All the information collected was stored in a database specifically designed for this purpose, and finally a data matrix related to the variables was produced. Thanks to the collaboration of healthcare professionals and parents, all data were correctly recorded. 2
Statistical analysis A sample size of 310 infants was estimated to be necessary based on the reported incidence of ADRs in similar areas (27.17%).23 It was calculated with a 95% CI and a precision of ±5% units without replacement rate, using the GRANMO sample size calculation programme. The frequency distribution of the qualitative variables is presented with 95% CIs. Data distributions for the quantitative variables are expressed as mean or median values. Bivariate comparative analysis between patients with and without ADRs was performed by χ2 test for qualitative variables and by comparison of means for quantitative variables. Data were analysed using SPSS software, and p values ≤0.05 were considered significant.
Ethical considerations The study protocol was approved by the ethics committee of the Hospital Clínico San Carlos, and parental informed consent was obtained in all cases before patient inclusion in the study.
RESULTS The recruitment began in May 2012 and finished in June 2013. A total of 322 newborns were admitted to the neonatal unit during this period. However, nine were not included in the study: informed consent was not obtained from eight parents, and one baby was transferred from another hospital making it impossible to collect all the information about his previous pharmacological treatment. Therefore, a total of 313 patients were studied (140 girls and 173 boys), and 2166 drug prescriptions were analysed. The median age of all patients was 0 days, with 75% of patients ≤1 day, 22% of patients 1–7 days, and 3% of patients 7–29 days. The mean±SD age at admission was 0.9±2.9 days. Patients from 24+0 to 41+6 weeks of gestational age were included in the study (mean gestational age of 36±4.2 weeks and median 37 weeks (Q1–Q3 34–39)). The median birth weight was 2470 g (Q1–Q3 500–4410), and the mean was 2485±868.6 g. Overall, 99 patients were admitted to the NICU and 214 to intermediate care; 210 (67%) were born at term and 10% (32) had intrauterine growth retardation. Sixty-nine (22%) patients did not receive any medication, and the mean number of drugs taken was seven per patient (median=1). Taking into consideration all the prescriptions, the mean duration of drug treatment was 5.5±9.3 days. Overall, 700 (32%) prescriptions were administered as a single dose, and the mean duration of treatment with multiple doses was 8.1 days. The most common interval between doses was ‘every 24 h’ (543, 25%). The route of administration most commonly used was intravenous (1113, 51%), followed by enteral (419, 19%) and intramuscular (259, 12%). At least one ADR was experienced by 54 patients (17%; 20 girls and 34 boys), and a total of 116 different ADRs were detected, which corresponds to 5% of the 2166 drug prescriptions. Of the patients who received medication, 22% experienced an ADR. The majority (94) of the ADRs were experienced in the NICU. Most of the ADRs detected were related to feed intolerance (19), phlebitis (10) and tachycardia (10) (table 1). Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
Original article The ADRs were assessed, and only 11% of the ADRs identified were labelled ‘definitive’, while 57% were ‘probable’ and 32% ‘possible’ according to the Naranjo method. With the Karch and Lasagna modified algorithm, 20% of the ADRs were considered to be ‘defined’, 34% ‘probable’, 28% ‘possible’, and only 18 were considered to be ‘conditional’. For eight of the identified ADRs, there was insufficient information to determine whether they were ADRs or not (ADRs for which the association with the suspected drug was considered to be doubtful or improbable), so they were excluded from the analysis. The majority of ADRs were moderate (49, 42%), 47 (41%) were mild, and 20 were severe, including 3 fatalities. Details of the 20 severe ADRs are shown in table 2. Most ADRs were acute (73%) and lasted between 1 and 7 days (39%). Of the ADRs identified, 87 (75%) were expected according to the drug action spectrum (type A). After the
Table 1 List of the most common adverse drug reactions identified and the causative drugs Adverse drug reaction Feeding intolerance
Phlebitis
Clinical manifestations
Causative drug
N
Presence of gastric residual accompanied by abdominal distension, bilious vomiting, food vomiting or discomfort
Ibuprofen Ferroglycine sulfate Caffeine citrate Paracetamol Ampicillin Multivitamin Vitamin D Zidovudine Vancomycin Fentanyl Ampicillin Cefazolin Cefotaxime Meropenem Trimethoprim sulfamethoxazole Caffeine citrate Salbutamol Dobutamine Dopamine Caffeine citrate Salbutamol Cyclopentolate +phenylephrine Ferroglycine sulfate Fentanyl Linezolid Amphotericin B Micafungin sodium Ampicillin Cefotaxime Caffeine citrate
6 5 2 2 1 1 1 1 3 2 1 1 1 1 1
Caffeine citrate Dobutamine Dexamethasone Rocuronium
3 1 1 1
Meropenem Ampicillin Teicoplanin Paracetamol Ibuprofen Methimizole Caffeine citrate
2 2 1 1 3 1 1
Localised redness and swelling, pain or burning along the length of the vein or vein being hard and cord-like
Tachycardia
Rate >180 beats/min
Central hyperactivity
Irritability, restlessness and jitteriness
Constipation Thrombocytopenia
Jaundice
Hypertension
Liver failure
Gastrointestinal haemorrhage
Time without defecation >48 h or rare and hard stools Platelet count 7.1 mg/dL) (table 1). Ibuprofen was the second most commonly prescribed drug related to ADRs: six patients experienced two or more digestive symptoms associated with this drug (including gastrointestinal haemorrhage, vomiting, constipation, feed intolerance and abdominal discomfort), while three only had gastrointestinal haemorrhage, two patients had necrotising enterocolitis, and another two had renal failure with high levels of serum creatinine. Likewise, ferroglycine sulfate was involved in seven cases of constipation and five cases of digestive disorder (vomiting, abdominal discomfort and feed intolerance). Other medicines associated with ADRs are listed in table 4.
DISCUSSION This study encompasses an exhaustive compilation and analysis of ADRs diagnosed in a cohort of 313 hospitalised neonatal patients; 2166 prescriptions were analysed using a systematic and intensive pharmacological surveillance protocol. In agreement with previous reports, antibiotics were the group of drugs most commonly prescribed.30 31 In contrast, caffeine citrate was the single drug most frequently used in our study, while previous reports indicated that multivitamins were more commonly prescribed.32 A high incidence of ADRs (17% of the patients) was observed, and 75% of all the ADRs found could be explained by the mechanism of action of the drug (type A) and hence could have been expected and potentially prevented. The incidence of ADRs found in our study is consistent with previous reports.22 23 33 However, it is extremely difficult to compare different studies because the estimation of ADR incidence is greatly influenced by the definitions used, the methodology of detection and classification, and the study setting. In this study, we used the definition established by the ICH guidelines, intensive pharmacology surveillance to detect ADRs, and standard methods to measure cause–effect relationships and the impact of side effects.14 28 29 34–36 Previous reports with similar methodology detected an incidence of 27.1% in NICU patients,23 while, in the present study, an incidence of 44.4% was observed in these particular patients. Nevertheless, the calculated incidence is lower than that published for adults.17–19 This might Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
Original article be explained by the present study not considering ADRs with low causality scores. Other factors that could underestimate ADR incidence in neonates are the difficulty of diagnosis and the lack of specific bibliography and tools. In addition, our bivariate analysis shows that ADR incidence increases with the number of prescriptions per patient, as has already been demonstrated in previous studies carried out in the NICU, where 15–30% of neonates who received more than 10 drugs experienced at least one ADR.23 37 The main limitation of this study is the determination of causality. It is important to note that only 11% of ADRs found were labelled as ‘sure’28 or 20% as ‘defined’.29 However, in order to classify an ADR as certain by applying any of the methods mentioned, it would be necessary to readminister the drug, to determine serum levels, or to administer a placebo, which is not possible or even ethical in most of the cases in neonatal care. In addition, although a high number of ADRs (62%) were considered probable or possible,29 it must be take into consideration that there might be potential bias related to ADR classification, as some true ADRs may have been excluded because of low causality scores, leading to underestimation of the ADR rate. Overall, the results of this analysis are consistent with the few existing publications.22 23 30–33 36–39 The differences detected were expected because of variation in patient characteristics, the system of evaluation carried out, and the pharmacotherapies used.
Data sharing statement Descriptions of the neonatal drugs used can be obtained by sending an email to the main author.
REFERENCES 1 2 3
4 5
6
7
8 9
10
11
CONCLUSIONS In summary, this study demonstrates a high incidence of ADRs in hospitalised neonates, which seems to increase with the number of prescriptions. Further refinement and additional testing are needed to confirm this. The extreme vulnerability of the newborn baby, the frequent need for medications, along with the lack of accurate and reliable information with regard to neonatal pharmacotherapy, as reflected by the incidence of ADRs, makes the development of pharmacological research in neonatology imperative. Author affiliations 1 Nursery Department, UICEC, Hospital Clínico San Carlos, Facultad de Enfermería Fisioterapia y Podología, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain 2 Facultad de Enfermería, Fisioterapia y Podología, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain 3 Neonatology Department, Hospital Clínico San Carlos, Madrid, Spain 4 Departamento de Pediatría, Facultad de Medicina, Universidad Complutense de Madrid, Instituto de Investigación Sanitaria del Hospital Clínico San Carlos (IdISSC), Madrid, Spain 5 Nursery Department, Facultad de Enfermería Fisioterapia y Podología, Universidad Complutense de Madrid, Madrid, Spain 6 Clinical Pharmacology Department, Hospital Clínico San Carlos, Madrid, Spain Acknowledgements We thank the Facultad de Enfermería, Fisioterapia y Podología (Universidad Complutense de Madrid) for support. We are also grateful to Ana I Soriano for helping with the translation of this paper.
12 13 14
15 16
17 18 19 20
21
22 23 24
Contributors ABR, EV and LA contributed to the design of the study, carried out data collection and analysis, and performed most of the writing. EP contributed to the design of the study and data analysis and revision of the manuscript. AP and JD carried out data collection and data entry and critically reviewed and revised the manuscript. All authors approved the final manuscript as submitted.
25
Funding The help and support of the health professional from the Neonatology Department and Clinical Pharmacology Department of the Hospital Clínico San Carlos were critical in allowing the study to take place.
27
Competing interests None declared.
28
Ethics approval Ethics committee of the Hospital Clínico San Carlos. Provenance and peer review Not commissioned; externally peer reviewed. Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
26
29
Turner MA. Neonatal drug development. Early Hum Dev 2011;87:763–8. Jain L. The challenge of managing drugs safely in the newborn. Clin Perinatol 2012;39:XV–XVI. Mason J, Pirmohamed M, Nunn T. Off-label and unlicensed medicine use and adverse drug reactions in children: a narrative review of the literature. Eur J Clin Pharmacol 2012;68:21–8. Schirm E, Tobi H, de Vries TW, et al. Lack of appropriate formulations of medicines for children in the community. Acta Paediatr 2003;92:1486–9. Walsh J, Mills S. Conference report: formulating better medicines for children: 4th European Paediatric Formulation Initiative conference. Ther Deliv 2013; 4:21–5. European Medicines Agency: Committee for Medicinal Products for Human Use (CHMP). Guideline On Conduct Of Pharmacovigilance For Medicines Used By The Paediatric Population. http://www.ema.europa.eu/docs/en_GB/document_library/ Scientific_guideline/2009/09/WC500003764.pdf (accessed 15 Jul 2015). Juárez OH, Pérez GG, Flores J. Pharmacovigilance and pharmacoepidemiology of drugs in a Mexican pediatric hospital. A proposed guide. Pharmacy World Sci 2007;29:43–6. Sammons HM, Choonara I. What is happening to improve drug therapy in children? Paediatr Int Child Health 2007;17:108–110. Speranza N, Lucas L, Telechea H, et al. Reacciones adversas a medicamentos en niños hospitalizados: un problema de salud pública. Rev Med Urug 2008;24:161–66. Thiesen S, Conroy EJ, Bellis JR, et al. Incidence, characteristics and risk factors of adverse drug reactions in hospitalized children—a prospective observational cohort study of 6,601 admissions. BMC Med 2013;11:237. Gray JE, Goldmann DA. Medication errors in the neonatal intensive care unit: special patients, unique issues. Arch Dis Child Fetal Neonatal Ed 2004; 89:F472–3. Conroy S, Choonara I, Impicciatore P, et al. Survey of unlicensed and off label drug use in paediatric wards in European countries. BrMed J 2000;320:79–82. Kemper EM, Merkus M, Wierenga PC, et al. Towards evidence-based pharmacotherapy in children. Paediatr Anaesth 2011;21:183–9. ICH Guideline E2D; Post-approval safety data management: definitions and standards for expedited reporting, 3.1.1 Spontaneous reports 2004. International Conference on Harmonisation. http://www.ich.org/fileadmin/Public_Web_Site/ICH_ Products/Guidelines/Efficacy/E2D/Step4/E2D_Guideline.pdf (accessed 29 Oct 2015). Dabliz R, Levine S. Medication safety in neonates. Am J Perinatol 2012;29:49–56. World Heath Organization. Promoting Safety of Medicines for Children. http://www. who.int/medicines/publications/essentialmedicines/Promotion_safe_med_childrens. pdf (accessed 15 Jul 2015). Vallano A, Agustí A, Pedrós C. Systematic review of studies assessing the cost of adverse drug reactions. Gac Sanit 2012;26:277–83. BordaI T, Slone D, Jick H. Assessment of adverse reactions within a drug surveillance program. JAMA 1968;205:645–7. Vargas E, Terleira A, Hernando F, et al. Effect of adverse drug reactions on length of stay in surgical intensive care units. Crit Care Med 2003;31:694–8. Impicciatore P, Choonara I, Clarkson A, et al. Incidence of adverse drug reactions in paediatric in/out-patients: a systematic review and meta-analysis of prospective studies. Br J Clin Pharmacol 2001;52:77–83. Pirmohamed M, James S, Meakin S, et al. Adverse drug reactions as cause of admission to hospital: prospective analysis of 18.820 patients. BMJ 2004;329:15–19. Martínez-Mir I, García-López M, Palop V, et al. A prospective study of adverse drug reactions in hospitalized children. Br J Clin Pharmacol 1999;47:681–688. Aranda JV. Factors associated with adverse drug reactions in the newborn. Pediatric Pharmacology New York 1983;3:245–9. Le J, Nguyen T, Law AV, et al. Adverse drug reactions among children over a 10-year period. Pediatrics 2006;118;555–62. Kunac DL, Kennedy J, Austin N, et al. Incidence, preventability, and impact of Adverse Drug Events (ADEs) and potential ADEs in hospitalized children in New Zealand: a prospective observational cohort study. Paediatr Drugs 2009;11:153. Jucker E, Venulet J. Methods of monitoring adverse reactions to drugs. Prog Drug Res 1977;21:231–92. Rawlins MD, Thompson JW. Pathogenesis of adverse drug reactions. In: Davies DM, Ferney RE, Glanville DE, eds. Textbook of adverse drug reactions. 4th edn. New York: Oxford University Press 1991:16–38. Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 1981;30:239–45. Farina JC, Krup P, Tobler HJ. Computer tools in Spontaneous reporting of adverse drug reactions. A multitudinal company approach. In: Ducrot H, Goldberg M,
5
Original article
30 31
32 33 34
6
Hoigné R, et al. Conference on computer aid to drug therapy and to drug monitoring. Amsterdam: North-Holland Publishing Company 1998:117–23. López R, Cabanas MJ, Oliveras M, et al. Utilización de medicamentos en una UCI neonatal: estudio prospectivo. Farm Hosp 2005;29:26–9. Morales-Olivas FJ, Martínez-Mir I, Ferrer JM, et al. Adverse drug reactions in children reported by means of the yellow card in Spain. J Clin Epidemiol 2000;53:1076–80. Feal CB, Barroso PC, Carcelen AJ, et al. Drug use in neonatology units of 6 Spanish hospitals. Farm Hosp 2003;27:69–71. Lazarou J, Pomeraz B, Corey PN. Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of prospective studies. JAMA 1998;279:1200–5. Real Decreto 577/2013, de 26 de julio por el que se regula la Farmacovigilancia de Medicamentos de Uso Humano. BOE 2013;179:55066–92. http://www.boe.es/boe/ dias/2013/07/27/pdfs/BOE-A-2013-8191.pdf (accessed 16 Jul 2015).
35
36
37
38 39
Real Decreto 1015/2009, de 19 de junio, por el que se regula la disponibilidad de medicamentos en situaciones especiales. BOE 2009;174:60904–14. http://www. boe.es/boe/dias/2009/07/20/pdfs/BOE-A-2009-12002.pdf (accessed 16 Jul 2015). Von Euler M, Eliason E, Ohlen G, et al. Adverse drug reactions causing hospitalization can be monitored from computerized medical records and thereby indicate the quality of drug utilization. Pharmacoepidemiol Drug Saf 2006;15:179–84. Aagaard L, Weber CB, Hansen EH. Adverse drug reactions in the paediatric population in Denmark: a retrospective analysis of reports made to the Danish Medicines Agency from 1998 to 2007. Drug Saf 2010;33:327–39. Bárzaga Z, Novoa D, Caballero M, et al. Adverse drug reactions in children in Camagüey Province, Cuba. Arch Dis Child 2010;95:474–7. Bárzaga Z, López A, Mejías Y, et al. Pharmacovigilance in children in Camagüey Province, Cuba. Eur J Clin Pharmacol 2012;68:1079–84.
Belén Rivas A, et al. Arch Dis Child 2016;0:1–6. doi:10.1136/archdischild-2015-309396
