4 DISCUSSION
In this study, drug prescriptions were evaluated in NICUs of Henan, China. To the best of our knowledge, this is the first large-scale multicenter study conducted in China and the data would be representative of the regional distribution of Henan, demonstrating high rates of off-label use in neonates.
Due to lack of evidence on drug use in neonates and therapeutic alternatives, off-label drug use is widespread in NICUs throughout the world 3-8, 17, 18, with a incidence rate ranging from 14% in an academic Dutch Children’s hospital 17 to 95.6% in a level III NICU in Brazil 4. Regulations and guidelines related to off-label drug use are different in global. For example, the rational off-label drug use is allowed in Unite States, France, Japan and UK,2, 19-21 but is abandoned in India 22. Laws and regulations related to off-label drug use are unavailable in China. Chinese Expert Consensus of Pediatric Off-Label Drug Use was established to regulate pediatric off-label drug use in 2016 by Chinese Journal of Pediatrics with great efforts of Chinese pediatric healthcare experts.11However, there are still barriers to implement the Chinese expert consensus of pediatric off-label drug use in clinical practice,23 resulting in high prevalence for off-label drug use. The percentage of prescriptions found to be off label in our study was 84.62%, which was higher as compared to previous reports ranging from 25.61% to 78.52% in China.12-14
Among the 5813 prescriptions, 63.03% was off label in relation to age. In an Italian multicenter study, the proportion of off-label prescriptions in relation to age was 34.4%.6 The study performed in Germany showed that only 38% had information regarding the use in patients aged less than 1 month in their SmPC among the 102 prescribed drugs.24 In the current study, 76.11% (86/113) of the prescribed drugs are in the absence of information for neonates, leading to the higher prevalence of off-label prescriptions associated with neonatal age. Historically, clinical trials evaluating the safety and efficiency of drugs and published information such as journal articles, practice guidelines, consensus statements in neonates are always unavailable.20Indeed, a large number of policies have been implemented to encourage the clinical evaluation in pediatric population, resulting in more than 500 pediatric labeling changes and the increase of prospective pediatric drug testing.20, 25 Despite this, less than 50% of products had pediatric information in labeling, indicating that labeling with pediatric information is still insufficient.26Additionally, as a vulnerable subpopulation, neonates are always excluded from studies under the incentive legislative measures. Therefore, neonatal information such as safety, efficiency and appropriate dose in labeling is especially scarce.27Between 1997 and 2010, only forty-one studies referred to 28 drugs in neonates were completed resulting in 24 related labeling changes.27 Consequently, numerous efforts are still required to overcome the current limitations and improve the efficiency in neonatal drug evaluation,28 thus increasing the scientific evidence of drug use and alleviating the situation of off-label drug use in this population.
In our study, the abnormal diagnosis such as ‘preterm’ was common (14.29%), which was defined as off label for indication. In addition, the clinical diagnosis was incomplete, for example, phosphocreatine for neonatal respiratory distress syndrome, phenobarbital for acid reflux, fructose diphosphate sodium for pneumonia. Therefore, 47.17% of the prescriptions resulted off label for indication.
Obviously, the exposure to antibiotics is very common in hospitalized neonates. In a one-day point prevalence study of the Antibiotic Resistance and Prescribing in European Children project, 31% (532/1712) of the neonates were exposed to antibiotics, ranging from 19.9% in a general neonatal ward to 39% in a neonatal intensive care setting.29 In the current study, systemic anti-infectives were most prescribed to neonates, which was quite similar to the results of recent studies by Neubert et al24 , de Souza et al 4 and Costa et al 3. However, a great variation regarding the anti-infective drugs prescribed in each NICU has been widely observed. Aminoglycosides such as gentamicin and amikacin, the preferred empirical treatment of suspected neonatal sepsis,30, 31 were most frequently prescribed in UK32, Australia8, USA33, Italy34 and Brazil4. In contrast, our data indicated that beta-lactam antibiotics including piperacillin-tazobactam, cefoperazone-sulbactam and cefotaxime were most prescribed antibiotics. This heterogeneity indicates that neonatologists prescribed antibiotics to the extensive variability in NICUs across globe, due to the lack of consensus regarding empirical antibiotic regimens in neonatal infections. To rationally use antibiotics, different and important aspects such as the pathogens and the resistance pattern, the PK and PD characteristics of antibiotics, and the neonatal pharmacology, should be considered.35
The analysis of off label based on CNFC exhibited a lower rate (65.16%) compared to Mcdex (84.62%) due to the differences of drug information between CNFC and Mcdex. For example, Mcdex recommend aminophylline for the treatment of asthma rather than neonatal apnea, while CNFC does. These results indicated that pediatricians are prone to making therapeutic decisions guided by the available evidence and the benefit for the individual patient in their medical practice, thus adhering to CNFC.
Undoubtedly, this study had some limitations. First, as a retrospective study, detailed demographic information of neonates such as the gestational ages was unavailable, due to lack of corresponding data collected from hospital information system. However, recent studies have shown that a higher proportion of off-label prescriptions were associated with increased postmenstrual age.5, 34Second, off-label conditions were analyzed according to Mcdex database rather than country-specific licensing information in the package leaflets, regardless of the difference from different pharmaceutical enterprises with different trade names, leading to heterogeneity. Third, participating NICUs in this study were from a specialized region in China. Prospective, large sample, multicenter and nationwide studies would be required to understand the status of off-label drug use in neonates in the future.