Abstract: Many drugs are used off-label in neonates which leads to large variation in prescribed drugs and dosages in neonatal intensive care units (NICUs). The NeoDose project aimed to develop best evidence dosing recommendations (DRs) for term and preterm neonates using a three-step approach: 1) drug selection, 2) establishing consensus-based DRs, and 3) establishing best evidence DRs. Methods: The selection of drugs was based on frequency of prescribing, availability of a neonatal DR in the Dutch Pediatric Formulary, and the labeling status. Clinical need, pharmacological diversity, and Working Group Neonatal Pharmacology (WGNP) preferences were also taken into account, using a consensus-based approach. For the second step, we requested local dosing protocols from all ten Dutch NICUs and established consensus-based DRs within the WGNP, consisting of neonatologists, clinical pharmacologists, hospital pharmacists, and researchers. In the third step, the consensus-based DRs were compared with the available literature, using standardized PubMed searches. Results: Fourteen drugs were selected for which the local dosing protocols were collected. These protocols differed mostly in total daily dose, dosing frequency, and/or route of administration. Strikingly, almost none of the dosing protocols of these 14 drugs distinguished between preterm and term neonates. The working group established consensus-based DRs, which after literature review needed modification in 56%, mainly in terms of a dose increase. Finally, we established 37 best evidence DRs, 22 for preterm and 15 for term neonates, representing 19 indications. Conclusion: This project showed the successful three-step approach for the development of DRs for term and preterm neonates.

© 2023 The Author(s). Published by S. Karger AG, Basel

Introduction

Many drugs lack a pediatric license, and even more drugs lack a neonatal license. Off-label prescribing in pediatrics is therefore common practice and has the highest prevalence in neonatal care [1]. To improve and maintain safe and effective prescribing, the Dutch Pediatric Formulary (DPF) has been developed [2, 3]. This formulary, launched in 2008 as a freely accessible website (www.kinderformularium.nl), aims to develop best evidence, transparent, and up-to-date drug information for children and neonates, with specific focus on dosing.

This focus on dosing requires a comprehensive review and assessment of the available literature to ultimately develop best evidence dosing recommendations (DRs) [3]. Best evidence, however, only indicates that the DR is supported by the best translation of the available evidence. In the absence of clinical trials, this may even be a consensus-based expert opinion. To do so, all relevant literature was summarized in a risk-benefit analysis (RBA) to obtain a clear overview of the available evidence, to appraise the quality of individual studies, and to compile a DR. The DR was peer reviewed by the National Multidisciplinary Editorial Board and published on the DPF website after adjustments and approval of the board, along with the supporting evidence. The RBA and DRs are regularly updated when new evidence emerges. The DPF has become the national standard for pediatric pharmacotherapy in The Netherlands. Publication of a DR in the DPF therefore assures the translation of new evidence to clinical practice.

However, of all the DRs listed in the DPF, neonatal DRs are underexplored. Mid 2020, the DPF contained 6,426 DRs for 774 drugs; 110 of these DRs were for preterm neonates, representing 73 drugs [4]. For term neonates, the DPF contained 481 DRs for 252 drugs. This lack of neonatal DRs was caused by a paucity of pharmacological studies conducted in neonates [5, 6], resulting in a large variation in drug-dosing practices between neonatal intensive care units (NICUs) [7-9], which poses an increased risk of therapy failure and toxicity [10]. Because of the high incidence of off-label prescribing and the current lack of evidence-based dosing guidelines, establishing best evidence neonatal DRs in every country is essential.

Therefore, the aim of this NeoDose project was to develop neonatal DRs for relevant example drugs and to publish these on the DPF website to make them publicly available. In addition, this paper aimed to encourage the availability of DRs for neonatal drug treatment in each country.

Methods

The NeoDose pilot project was a collaboration of all Dutch NICUs to develop neonatal DRs, performed between January 2018 and December 2020. A Working Group Neonatal Pharmacology (WGNP) was formed with one or two neonatologists per NICU (13 in total, some also trained as clinical pharmacologists), three hospital pharmacists, the DPF chair of the editorial board, two DPF pharmacists preparing the best evidence DRs, and the DPF project manager who collected the local dosing protocols.

The standard operating procedure of the DPF for developing DRs first requires evaluating the available literature [3]. Nevertheless, this approach proved ineffective in a previous project on establishment of best evidence DRs for children with an impaired kidney function due to a lack of available relevant literature, ultimately prompting us to rely on consensus. This, along with the general experience that neonatal evidence is also scarce [5, 6], led us to a reversed, three-step approach for the NeoDose pilot project: 1) selection of drugs to be included in the NeoDose project, 2) drafting consensus-based DRs for these drugs, and 3) comparing the available scientific evidence to the consensus-based DRs to establish best evidence DRs. To address gestational age-related differences regarding disposition and effect, the WGNP pragmatically chose to establish one DR for term neonates (born ≥37 weeks of gestation) and one DR for preterm neonates (born <37 weeks of gestation), unless clear evidence was found supporting a more detailed differentiation.

Selection of Drugs

A list of the most commonly used drugs in Dutch NICUs was used for the selection of drugs [7]. From this list, containing 181 drugs, the WGNP selected 14 drugs for the pilot project. Selection was based on frequency of prescribing, the availability of a neonatal DR in the DPF, and the labeling status. Clinical need, pharmacological diversity, and WGNP preferences were also taken into account, using a consensus-based approach.

Consensus-Based DRs

Subsequently, local dosing protocols of the selected drugs were requested from all ten Dutch NICUs (as illustrated with acyclovir in Fig. 1), which were compared and discussed by the WGNP to establish consensus-based DRs. These were published on the DPF website shortly thereafter. For each drug, the degree of agreement between local protocols and consensus-based DRs was determined to gain insight into the variability of prescribing behavior. Additionally, the consensus-based DRs were compared to four international pediatric drug databases (British National Formulary for Children [BNF-c], Lexicomp Pediatric & Neonatal Dosage Handbook, Australasian Neonatal Medicines Formulary [ANMF], and Micromedex).

Fig. 1.

Indications and number of NICU DRs obtained for acyclovir. DR, dosing recommendation; GA, gestational age; HSV, herpes simplex virus; NICU, neonatal intensive care unit; VZV, varicella-zoster virus.

Best Evidence DRs

After establishing consensus-based DRs, DPF pharmacists searched PubMed for each drug separately (online suppl. Table S1; for all online suppl. material, see www.karger.com/doi/10.1159/000528012) and summarized the original relevant literature on pharmacokinetics (PK) and efficacy in an RBA. The PK properties of the drug with respect to gestational and postnatal age were assessed to suggest an optimal dosage. This dosage was compared with the dosages administered in the efficacy studies, simultaneously considering the safety and efficacy of these dosages. This, combined with the clinical practice of the consensus-based DRs, formed the basis for the proposed DR.

The RBAs and proposed DRs were discussed by the WGNP, resulting in best evidence DRs. This approach either confirmed or rejected the already published consensus-based DRs, transforming the consensus-based DRs into best evidence DRs. The best evidence dose was published on the DPF website. To provide insight into how the best evidence DRs compared to local protocols, the degree of agreement between both was determined, as was agreement with BNF-c, Lexicomp, ANMF, and Micromedex.

ResultsSelection of Drugs

We selected 14 drugs for the NeoDose project. Table 1 indicates the reasons for inclusion in this pilot project. Ten of these were within the top 30 most commonly prescribed drugs within Dutch NICUs [7], whereas the remaining four were chosen for other relevant reasons. For ten out of 14, the DPF lacked at least one DR prior to the start of this project. This mostly involved a DR for preterm neonates, except for one of the indications of milrinone, which applied only to term neonates.

Table 1.

Overview of the selected drugs, indications and comparison of local protocols, international pediatric drug databases, consensus-based DRs, and best evidence DRs

Consensus-Based DRs

In preparation for the consensus-based DRs, nine of the ten Dutch NICUs provided their local dosing protocols. Strikingly, most protocols did not distinguish between term and preterm neonates. NICU protocols for these 14 drugs contained a total of 272 DRs for 28 indications (Fig. 2). Table 1 shows for each drug the proportion of NICUs with a protocol, the proportion of different protocols, and the proportion of protocols that corresponded to BNF-c, Lexicomp, ANMF, and Micromedex. The most common differences between protocols involved differences in total daily dose, dosing frequency, and/or route of administration. For example, eight NICUs had a protocol for hydrocortisone used for hypotension, but each protocol differed to some extent; NICU 8 recommended a three times higher maximum dose than NICU 5, while ANMF and BNF-c recommended four to five times higher doses, respectively (Table 2).

Table 2.

Differences between hydrocortisone DRs among local protocols and international pediatric drug databases for the treatment of hypotension in term and preterm neonates

Fig. 2.

Schematic overview of number of indications and DRs for all 14 drugs.

During the consensus meeting, 28 identified indications were reduced to 18 as 10 indications either overlapped with other indications or could be merged with another indication (for instance, different wording for same indication). The remaining 18 indications resulted in 36 consensus-based DRs (21 for preterm and 15 for term neonates), six of which corresponded to one of the international pediatric drug databases (Table 3 and Fig. 2). The percentages of NICU DRs that corresponded to the consensus-based DRs are shown in Table 1.

Table 3.

Overview of consensus-based and best evidence DRs

Best Evidence DRs

Evidence was available for all but one of the drugs; for spironolactone indicated for treatment of fluid overload and edema, we could not find studies including neonates. For the remaining drugs, one to 16 useful studies per drug were found (online suppl. Table S2).

The proposed DRs deviated from the consensus-based DRs in 36.1% (13 DRs) of cases; the WGNP approved ten of the new DRs, and the remaining three were slightly modified before being approved. Another seven (19.4%) consensus-based DRs, for which the DPF pharmacists had not proposed a modification, were adapted by the WGNP after review of the RBAs. This leaves 16 (44.4%) consensus-based DRs found to be consistent with the literature. Table 3 visualizes all final modifications.

Finally, 19 indications were established for these 14 drugs, as the WGNP concluded that for hydrocortisone, the indication “bronchopulmonary dysplasia” should be split into treatment and prophylaxis. These 19 indications accounted for 37 best evidence DRs (Table 3), 22 for preterm and 15 for term neonates, which are currently available at the DPF website and also implemented in the local protocols. Five of these best evidence DRs corresponded to one of the international pediatric drug databases. The percentages of NICU DRs that corresponded to the best evidence DRs are shown in Table 1.

Discussion

In the NeoDose project, we established DRs for 14 drugs commonly used in neonatology. The apparent high variability in DRs among different NICUs highlighted the need for harmonization, while the high percentage of consensus-based DRs that required modification showed the need to thoroughly review scientific evidence. This project demonstrated the feasibility for drugs with scarce evidence on neonatal pharmacology of 1) achieving national consensus and 2) combining the available evidence with this consensus to develop best evidence DRs.

The comparison of the collected NICU protocols confirmed the large variability in drug choice and dosages among the Dutch NICUs, which is in agreement with previous reports [7, 11]. These studies indicated that the largest variability was observed among drugs used off-label [7] and among drugs whose PK had been poorly studied [11]. This was fully consistent with our best evidence DRs, as all but one (dobutamine for term neonates) were off-label and for most drugs, only 0–3 PK studies were found. We also noted this large variation among BNF-c, Lexicomp, ANMF, and Micromedex (online suppl. Table S3), confirming a previous report by Ceelie et al. [12] for the pediatric intensive care setting. Additionally, these databases only rarely provided DRs for preterm neonates (online suppl. Table S3).

Remarkably, 56% of consensus-based DRs were modified after review of literature, illustrating the delayed impact of evidence on our practices. Our assumptions about knowledge of literature from the WGNP may have been biased by the presence of the local protocols. The modifications of the consensus-based DRs were either based on consistent evidence or resulted from discussions within the WGNP on inconsistent study results and the implications for clinical practice. For dexamethasone, for example, indicated to prevent bronchopulmonary dysplasia, none of the trials used the same dosage, treatment duration, or inclusion criteria. The study outcomes, however, were quite similar. Besides, an international guideline preferred lower dosages [13], while a meta-analysis had shown that higher dosages were more effective [14]. The local protocols, however, showed more agreement here, with also good clinical experiences. Although this consensus might have been driven by expert guidance, the consensus-based dosing schedule was fully within the reported range of dosing schedules of the clinical trials, both in terms of cumulative dose, duration, and inclusion criteria. Therefore, with studies being inconclusive, it was decided to maintain the consensus-based DR.

Another example was acyclovir in the treatment of herpes neonatorum, which had so far been prescribed at a dose of 60 mg/kg/day, regardless of gestational and/or postnatal age. All efficacy studies used this dose, despite a population PK study in newborns showing that at this dose, only a minority of term neonates achieved serum levels, shown effective in adults, children, and neonates [15]. The authors considered the risk of undertreatment, possibly resulting in long-term neurodevelopmental impairment or death, to be greater than the risk of toxicity such as the easily diagnosed neutropenia and nephrotoxicity. Although their proposed dose of 80 mg/kg/day had not yet been reported in the literature and clinical experience was lacking, the WGNP agreed that undertreatment was more dangerous than the occurrence of adverse events. They decided on a dose increase from 60 to 80 mg/kg/day for term neonates, with extra alertness to any sign of toxicity.

Although we found relevant literature for most drugs, our standardized PubMed searches had their limitations. The use of PubMed’s filters and MeSH terms may have limited the search results, missing potentially relevant articles. To overcome this limitation, search terms were expanded when only few results were found, but this did not necessarily lead to additional relevant articles (online suppl. Table S1). Furthermore, since we focused on finding the right dose, we considered, but did not establish, the level of evidence of DRs, efficacy, safety, or place in therapy.

Within this project, we pragmatically chose to distinguish only between term and preterm neonates. This simplified representation does not reflect the complex intra- and extrauterine pharmacological maturation during the preterm neonatal period, reflected subsequently by gestational and postnatal age. Although we know that gestational age, postnatal age, and weight are the most important determinants of organ maturation and therefore of drug disposition and efficacy [16], insufficient data were yet available to incorporate all of these factors into a DR. Additionally, the DRs should only be considered as starting dosages. If a neonate needs treatment for a longer period, intermediate dose adjustments are required based on evaluation of efficacy and safety, taking weight as well as ontogeny into account. Unfortunately, it is currently impossible to include such drug-specific complexity in our DRs due to the lack of data, with a few exceptions. For hydrochlorothiazide and midazolam, we were able to further specify gestational age, and for ibuprofen and propofol, postnatal age distinctions were included. In the future, modeling and simulation could support dosing for other drugs in relation to efficacy and safety.

Our study showed that even highly motivated neonatal experts (WGNP members) are not fully aware of all the available literature to draft DRs that reflect the current state of science, illustrating the longtime interval needed between knowledge generation and subsequent translation and application in practice. Nevertheless, drafting consensus-based DRs takes less time than drafting best evidence DRs (two meetings vs. 3 months spread over 3 years). Local protocols add value in this regard and proved indispensable in the preparation of RBAs and resulting dosage proposals, particularly when there is little relevant literature.

In conclusion, this pilot project confirmed the large variation in dosing protocols, thereby emphasizing the need for best evidence-based DRs. Neonatal DRs for 14 drugs have been developed and implemented in the DPF. For more uniform neonatal drug use in The Netherlands and across Europe and to assure implementation, these new DRs have been published on the publicly accessible website of the DPF and its international affiliates in Germany, Austria, and Norway [17-19]. In addition, all NICUs have also updated their protocols with the established best evidence DRs. We now aim to develop best evidence DRs for the other drugs used in neonatology to ensure safe and effective drug use in this vulnerable population.

Acknowledgments

We would like to thank all members of the Working Group Neonatal Pharmacology for their valuable contributions to the development of consensus-based and best evidence-based dosing recommendations.

Statement of Ethics

No ethics approval was required as no human or animal data have been used.

Conflict of Interest Statement

The authors declare that there are no conflicts of interest.

Funding Sources

This project received funding from the Foundation for Quality Funding for Medical Specialists (SKMS) from The Netherlands. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author Contributions

All authors provided significant contributions to the writing of the manuscript and approved the final version. All members of the Working Group Neonatal Pharmacology were involved in the establishment of all dosing recommendations and approved the final version of this manuscript.

Data Availability Statement

Inquiries about the risk-benefit analyses can be directed to the corresponding author.

This article is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC). Usage and distribution for commercial purposes requires written permission. Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.