Maria Xiberras1, Rob C. van Wijk1, Elke H.J. Krekels1,2, Coen van Hasselt1, Gerdien Zeilmaker-Roest3, Dick Tibboel3, Enno D. Wildschut3, Catherijne A.J. Knibbe1,4, Sebastiaan C. Goulooze1,5
1Division of Systems Pharmacology and Pharmacy, Leiden Academic Centre for Drug Research, Leiden University, 2Certara Inc, 3Intensive Care and Department of Paediatric Surgery, Erasmus University Medical Center–Sophia Children’s Hospital, 4Department of Clinical Pharmacy, St. Antonius Hospital, 5LAP&P Consultants
Introduction: Morphine is commonly used as maintenance and rescue analgesic in children after cardiac surgery [1]. To date, there is very limited knowledge on the exposure-response relationship of morphine in these children. Repeated time to event (RTTE) is a useful tool in which the hazard for a patient to receive rescue morphine is quantified, and which allows to investigate the influence of different factors that can impact the required number of rescue morphine doses as a measure of morphine efficacy. This study aimed to quantify the morphine pharmacodynamics for postoperative pain treatment in children under three years of age after cardiac surgery using rescue morphine data from a multicentre prospective randomised control study on morphine and paracetamol after paediatric surgery [2] analysed by RTTE modelling approach. Methods: Data from the Paediatric Analgesia after Cardiac Surgery (PACS) study were analysed [2]. The PACS study included 194 patients under three years of age who underwent cardiac surgery with use of cardiopulmonary bypass [2]. At the end of surgery, patients received a loading dose of morphine (100 ug/kg) and subsequently were randomized to either continuous intravenous (IV) morphine or IV paracetamol as standard pain treatment. Patients from both groups received rescue morphine if the nurse reported numeric rating scale scores for pain exceeding three. All administrations of additional rescue morphine dosing, i.e. additional bolus doses and/or increases of the continuous morphine infusion, were treated as the event of interest for the RTTE modelling. Individual plasma concentrations of morphine were obtained using individual posthoc parameters from a Bayesian re-estimation with a previously published and validated paediatric morphine population pharmacokinetic model [3,4], the individual morphine dosing information and, when available, observed morphine plasma concentrations. In the absence of observed concentrations, population predictions were used. Here, the hazard to receive rescue morphine was studied in the patients receiving paracetamol as standard pain treatment (N = 92 patients). Additionally, the influence of ‘time since previous rescue morphine event’ as a potential increasing factor on the hazard was tested, as it was hypothesized that the common assumption of independence of events may be violated in this study [5]. Finally, the influence of morphine concentrations on the hazard was explored. NONMEM 7.5.0. (ICON Development Solutions, Ellicott City, MD, USA) was used for the modelling. Results: The RTTE model was developed on data from 92 patients (aged 6 days to 3 years, median 162 days). During the postoperative period (median 48 h (IQR 32–48)), 214 morphine rescue events (median 2 (IQR 0–4) per patient) were observed. In the RTTE model, the hazard of rescue morphine decreased over time for which a Gompertz hazard model was used. A strong increase in hazard for rescue morphine was observed for a short period after an event, which suggests a lack of independence of consecutive rescue morphine events. This finding was implemented in the model by including an increased hazard after each event, which declines exponentially with ‘time since previous rescue morphine event’ with a half-life of 62.4 minutes. A log-linear relationship best characterized the relationship between morphine concentrations and the decrease in the hazard for rescue morphine. Morphine concentrations of 10 and 20 ng/ml resulted in a decrease in hazard of 32.4% and 54.3%, respectively. Without the adjustment for the ‘time since previous rescue morphine event’, the morphine effect was under-estimated by 60.1%. Conclusion: In this study the hazard for morphine rescue events after cardiac surgery in children receiving a loading dose of morphine at the end of surgery followed by intermittent paracetamol maintenance treatment was quantified with a) a decreasing Gompertz hazard with time after surgery, b) an increased hazard after a morphine rescue event and c) a decrease in hazard with increasing morphine concentrations. By quantifying the concentration-effect relationship for morphine, the RTTE model may support (individualised) morphine dosing to treat postoperative pain after paediatric cardiac surgery.
[1] Zeilmaker-Roest GA et al., BMJ Paediatr Open. 2017;1(1):e000046 [2] Zeilmaker-Roest GA et al., Critical Care. 2024;28(1):143. [3] Valkenburg AJ et al., Pediatric Critical Care Medicine. 2016;17(10):930–8. [4] Xiberras M et al., PAGE 32 2024 Abstr 11133 [5] Andersen P and Gill R. Ann Stat. 2015; 10:1100–20
Reference: PAGE 33 (2025) Abstr 11754 [www.page-meeting.org/?abstract=11754]
Poster: Drug/Disease Modelling - Paediatrics