Article Text
PDF
Abstract
Objective The β2-agonists such as salbutamol are the mainstay of asthma management. Pharmacokinetic–pharmacodynamic (PKPD) models to guide paediatric dosing are lacking. We explored the relationship between salbutamol dose, serum concentration, effectiveness and adverse effects in children by developing a PKPD model.
Design A prospective cohort study of children admitted to hospital with acute asthma, who received intravenous salbutamol.
Setting Children were recruited in two cohorts: the emergency departments of two London hospitals or those retrieved by the Children’s Acute Transport Service to three London paediatric intensive care units.
Patients Patients were eligible if aged 1–15 years, admitted for acute asthma and about to receive or receiving intravenous salbutamol.
Interventions Treatment was according to local policy. Serial salbutamol plasma levels were taken. Effectiveness measurements were recorded using the Paediatric Asthma Severity Score (PASS). Toxicity measurements included lactate, pH, glucose, heart rate, blood pressure and arrhythmias. PKPD modelling was performed with non-linear mixed-effect models.
Main outcomes Fifty-eight children were recruited with 221 salbutamol concentration measurements from 54 children. Median (range) age was 2.9 (1.1–15.2) years, and weight was 13.6 (8–57.3) kg. Ninety-five PASS measurements and 2078 toxicity measurements were obtained.
Results A two-compartment PK model adequately described the time course of salbutamol–plasma concentrations. An EMAX (maximum drug effect) concentration–effect relationship described PASS and toxicity measures. PKPD simulations showed an infusion of 0.5 µg/kg/min (maximum 20 µg/min) for 4 hours after bolus achieves >90% maximal bronchodilation for 12 hours.
Conclusions A paediatric PKPD model for salbutamol is described. An infusion of 0.5 µg/kg/min after bolus achieves effective bronchodilation. Higher rates are associated with greater tachycardia and hyperglycaemia.
- intensive care units, paediatric
- pharmacology
- child health
- paediatric emergency medicine
- respiratory medicine
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Statistics from Altmetric.com
- intensive care units, paediatric
- pharmacology
- child health
- paediatric emergency medicine
- respiratory medicine
WHAT IS ALREADY KNOWN ON THIS TOPIC
There is limited evidence for the use of intravenous β2-agonists in children with asthma. Pharmacokinetic–pharmacodynamic (PKPD) models to guide paediatric salbutamol dose are lacking.
Current paediatric salbutamol dosage regimens are extrapolated from adult PK studies and result in children receiving far more intravenous salbutamol (1–5 µg/kg/min) than the maximum adult dose (20 µg/min).
Salbutamol is associated with adverse effects such as lactic acidosis, hyperglycaemia, ketosis and cardiac arrhythmias.
WHAT THIS STUDY ADDS
A two-compartment disposition model described intravenous salbutamol pharmacokinetics. An Emax concentration–efficacy relationship described effectiveness and toxicity measures.
Blood glucose and heart rate are useful toxicity measurements, with the salbutamol concentration required for maximal therapeutic effect surpassed before toxicity.
Using the PKPD model, an infusion of 0.5 µg/kg/min for 4 hours after 15 µg/kg bolus over 10 min achieved >90% of maximum bronchodilation for 12 hours. A ceiling effect at 20 µg/min is reported.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Current dosage schedules for intravenous salbutamol in children result in salbutamol concentration levels above those required for therapeutic bronchodilation, at the expense of adverse effects.
This study provides a PKPD model on which to base future dosing schedules of intravenous salbutamol in children, thereby maximising effectiveness while minimising toxicity.
Introduction
A child is admitted to the hospital every 20 min with acute asthma.1 The β2-adrenergic receptor agonists such as salbutamol form the mainstay of treatment. National guidelines from the British Thoracic Society recommend that intravenous salbutamol should be considered as second-line treatment in children who fail to respond to nebulised β2-agonists and steroids. An intravenous salbutamol bolus 15 µg/kg body weight over 10 min followed by a maintenance infusion of 1–5 µg/kg/min is recommended,2 although the evidence for this dosing strategy is limited.3 Current paediatric salbutamol dosage regimens are extrapolated from adult pharmacokinetic studies and result in children receiving far more intravenous salbutamol (1–5 µg/kg/min) than the maximum adult dose (20 µg/min). Huge variations in clinical practice are reported.4
Data from pharmacokinetic/pharmacodynamic (PKPD) models on which to base paediatric salbutamol dose are lacking.5 Salbutamol use is associated with adverse effects such as lactic acidosis, hyperglycaemia, ketosis and diastolic hypotension with myocardial injury and cardiac arrhythmias.6–9 Salbutamol beneficial effect and adverse effect ratios have not been explored.
We conducted a prospective cohort study to explore the relationship between salbutamol dose, plasma concentration, effectiveness and adverse effects in children by developing a population PKPD model. We anticipated this model could change dosing recommendations.
Methods
Design
We conducted a prospective cohort study of children admitted to hospital with acute severe asthma, who received intravenous salbutamol during their clinical management. All children were treated according to local policy.
Serial salbutamol concentration levels, effectiveness scores and toxicity measurements were taken as described (see Data sources/measurements section). PKPD modelling was performed using non-linear mixed-effects models (NONMEM7.3, ICON Development Solutions, USA).
Setting
Patients were recruited in two cohorts between September 2014 and May 2017:
The emergency departments at two hospitals in London: St Mary’s and the Royal London Hospitals (ED cohort).
Critically ill children retrieved by the Children’s Acute Transport Service (CATS) to three tertiary paediatric intensive care units (PICUs): Great Ormond Street, St Mary’s and the Royal London Hospitals (CATS cohort).
Participants
Patients were eligible if aged 1–15 years (inclusive), admitted to hospital for acute asthma (defined by the British Thoracic Society guidelines) and about to receive or receiving intravenous salbutamol. Participants were identified by the clinical teams and recruited by research nurses. Patients continued in the study until 8 hours after stopping salbutamol or hospital discharge.
We followed established procedures for obtaining deferred consent in emergency situations.10–12
Data sources/measurements
Study data were collected using REDCap database hosted at University College London.13 Data collected included age, sex, ethnicity, renal function tests; asthma treatments and timing including nebulisers, steroids, magnesium sulfate, aminophylline and intravenous salbutamol; and hospital length of stay, PICU admission and use of mechanical ventilation.
Salbutamol concentrations: Serial blood samples to measure salbutamol concentrations were taken. In the ED cohort, samples were taken at t0 (prior to starting intravenous salbutamol) and tend (prior to stopping the salbutamol infusion). Other samples were taken with routine clinical sampling. We aimed for a minimum of three samples per patient. In the CATS cohort, salbutamol concentrations were taken at tCATS (at first contact with the retrieval team, when the child was usually already on intravenous salbutamol). Further samples were taken via indwelling vascular catheters: prior to a dose change, 30 min after a dose change, once at tend and one in 8 hours after stopping intravenous salbutamol. Samples were analysed by ABS Laboratories, using an liquid chromatography–mass spectrometry method, validated according to Europeans Medicines Agency and Food and Drug Administration regulatory authority criteria.14 15 This assay is validated to a lower limit of 0.1 ng/mL salbutamol.
Genotype data: Blood samples were taken for single nucleoside polymorphism analysis involving the three most common β-2 adrenoceptor genes (ADRB2): Arg16Gly, Gln27Glu and Thr164Ile to see if they influenced the PKPD model.
Effectiveness measurements were recorded in spontaneously breathing patients by the Paediatric Asthma Severity Score (PASS, 0–15), a validated scoring system incorporating wheezing, prolonged expiration and work of breathing. This non-invasive scoring system correlates with peak expiratory flow rate and pulse oximetry.16 Serial measurements were recorded by the research nurse. We did not attempt to measure effectiveness in ventilated patients.
Toxicity-associated physiological measurements recorded included serum lactate, pH, base excess and blood glucose values. Toxicity-associated cardiovascular measurements were heart rate, blood pressure and cardiac arrhythmias.
Sample size
To characterise model parameter–covariate relationships, it is proposed that at least 50 subjects are required in a population PKPD study.17 We aimed to recruit 100 patients.
Ethical approval by the Brent Research Ethics Committee (ref: 14/LO/2103).
Statistical methods
Population PKPD modelling
Sequential population PKPD modelling was performed with NONMEM (V.7.3), where a PK model was first developed and individual PK parameters were combined with a PD model to describe the time course of PD observations.18
One-compartment and two-compartment PK models were tested to describe intravenous salbutamol disposition. Nebulised salbutamol administered was included under the assumption that most drug was deposited in the lung over 5 min with estimated inhaled bioavailability (0.09) and absorption rate fixed at 16.7 per hour. An estimated baseline plasma concentration was also tested. Allometric size scaling with body weight on clearance and volume of distribution was added a priori, and a sigmoidal maturation function describing the change of clearance over age was tested.19
PASS scores were treated as continuous data and modelled with a regression model, where salbutamol plasma concentrations were linked with PASS scores.20 Both linear, EMAX and sigmoidal EMAX concentration–effect relationships were tested. Raw data showed a flat concentration–response relationship indicating current clinical salbutamol concentrations achieve the maximum possible bronchodilation effect. As parameters could not be estimated due to O-gradients, EC50 and HILL were fixed to literature values from a dose-ranging study (online supplemental figure 9).21 Since these values were for R-salbutamol, our assay measured total salbutamol, and the salbutamol concentrations were converted to total by adding the R-salbutamol concentration to the predicted S-salbutamol concentrations from a recent study.22
Supplemental material
We examined whether any patient characteristics influenced the estimation of bronchodilation effect (PASS). Covariates analysed were aminophylline, corticosteroids, ipratropium, montelukast, antibiotics, serum creatinine and MgSO4. β-2 adrenoceptor genotyping was visually inspected. The likelihood ratio test or Akaike information criterion were used, respectively, for nested and non-nested models to compare the goodness of fit.23
The following markers of toxicity were modelled: lactate, pH, base excess, blood glucose, heart rate and diastolic blood pressure. Heart rate and diastolic blood pressures were standardised with age-appropriate Z-scores.24 25 Toxicity effects were estimated from circulating salbutamol with first-order equilibration with the plasma compartment. A sigmoidal EMAX model was used to link salbutamol concentration with toxicity effects. Concomitant therapy with corticosteroids was investigated as a covariate on blood glucose PKPD model parameters E0, EMAX and EC50.
PKPD models for plasma concentrations, PASS scores and toxicity measurements were evaluated by prediction-corrected visual predictive checks (pcVPCs) that were generated using Perl-speaks-NONMEM (V.4.8.1). For each pcVPC, 1000 simulations from the PKPD model were performed.26 Whether the median, 2.5th and 97.5th percentiles of observed data lay within the 95% CI of simulated data was visually assessed.
Dose regimen simulation
Using our PKPD model, simulations were conducted in 2000 patients to determine the optimal dose of intravenous salbutamol in children. The simulated patient population had body weights ranging from 5 to 54 kg and were given a bolus dose of 15 µg/kg followed by an infusion of 0.25, 0.5, 1 and 5 µg/kg/min. The target was to propose an infusion dose that achieved >90% of the maximum bronchodilation effect for 12 hours in the majority of patients, with minimum clinical toxicity; in particular of hyperglycaemia (blood glucose >10 mmol/L) and tachycardia (heart rate >2 SD above baseline). We analysed infusion durations up to 24 hours but ultimately focused on a 4-hour infusion as it may obviate the need for out-of-hospital transfer. A 0.5 µg/kg/min infusion was further investigated with and without a maximum dosage of 20 µg/min using Monte Carlo simulations.
Results
Participants
There were 576 patients screened. There were 89 eligible patients, 8 refused or were unable to consent, 1 was previously recruited and 22 were missed. There were 54 of 60 children included in the final analysis (figure 1).
Flow chart of patient recruitment. CATS/PICU, Children’s Acute Transport Service/Paediatric Intensive Care Unit cohort; ED, Emergency Department cohort; iv, intravenous.
Descriptive data
The median (range) age was 2.9 (1.1–15.2) years with 20 (37%) females. There were 40 (74%) children with a known history of asthma or viral-induced wheeze. There were 50 children (93%) given salbutamol inhalers/nebulisation prior to enrolment. In the CATS cohort, 16 children (47%) had received an intravenous salbutamol bolus and 11 children (32%) had commenced on salbutamol infusions prior to enrolment. The median (range) duration of these infusions was 5 (0.5–22.7) hours. At enrolment, the median (range) salbutamol concentrations were 86.65 (1.36–166) and 8.04 (2.47–24) ng/mL in the CATS and ED cohorts, respectively (table 1).
Descriptive data of study participants at enrolment
Following enrolment, 30 children (55%) received further salbutamol nebulisers/inhalers. There were 8 (24%) and 19 (95%) children who received an intravenous bolus in the CATS and ED cohorts, respectively. Salbutamol infusions were run at a median (range) rate of 1.0 (0.1–4) µg/kg/min in the CATS cohort and 1.0 (0.5–2) µg/kg/min in the ED cohort. Salbutamol infusion duration median (range) was 3.33 (0.183–196.2) hours and 5.5 (0.5–48.8) hours in the CATS and ED cohorts, respectively, after enrolment. Concomitant drugs used during the study period are outlined in table 2.
Treatment and outcomes of study patients
Outcome data
We obtained 221 salbutamol plasma concentrations from 54 patients. There were 95 PASS measurements, 280 lactate measurements, 295 pH measurements, 294 base excess measurements, 255 blood glucose measurements, 306 diastolic blood pressure measurements, 358 heart rate measurements (online supplemental figures 1–8) and 20 genotype samples. Mechanical ventilation was required in 34 children (97%) of the CATS cohort and 3 children (15.8%) in the ED cohort for a median duration of 91.6 hours and 99.1 hours, respectively. Median (range) length of hospital stay was 6 (3–34) days and 2 (1–9) days in the CATS and ED cohorts, respectively (table 2).
Main results
Population PK modelling
A two-compartment PK model adequately described the time course of salbutamol plasma concentrations. PK parameters were standardised to a 70 kg person using allometry. Our scaled estimated clearance was 16.3 L/hour/70 kg. Age had no impact on clearance. Other parameters included intercompartment clearance (8.58L/hour/70 kg), central volume (109 L/70 kg) and peripheral volume (69.4 L/70 kg) (table 3).
Parameter estimates from pharmacokinetic and pharmacokinetic–pharmacodynamic effectiveness and toxicity models
Population PD modelling
We found current salbutamol dosing results in salbutamol plasma concentrations that are above the 90% maximum drug effect (EMAX). Parameter values from the EMAX drug effect models are outlined in table 3 for PASS. The concentration that achieved a half maximum response (EC50) and slope (HILL) parameters could not be reliably estimated and were subsequently fixed to literature-based forced expiratory volume in 1 s (FEV1) estimates at 0.15 ng/mL and 3.2, respectively (online supplemental figure 9).
None of the covariates tested: aminophylline, corticosteroids, ipratropium, montelukast, antibiotics, serum creatinine and MgSO4 were significant (p>0.01) on PASS score. As genotype samples were only obtained in 20 patients, no trends were identified.
Blood glucose and heart rate increased with increased salbutamol plasma concentrations. Parameter estimates for these toxicity measurements are outlined in table 3. For both these variables, the concentration–response curve demonstrates that maximum bronchodilation effect is achieved faster than the toxicity effects of hyperglycaemia and tachycardia. Concomitant therapy with corticosteroids or ventilation were not significant covariates on PKPD model parameters. Salbutamol plasma concentration did not correlate with lactate, base excess or diastolic blood pressure.
Model evaluation
Visual plots of PK and PD results (pcVPC) were reasonable for plasma concentrations, PASS and toxicity measurements, suggesting that the developed PKPD models were adequate in describing the observed PKPD (online supplemental figure 10).
Other analyses
Using our PKPD model, we aimed to extrapolate a dose regimen for children. Our target was to achieve 90% of the maximum bronchodilation effect for >12 hours, with minimal toxicity. Using simulations in 2000 virtual patients after a bolus of 15 µg/kg, infusion rates of 0.25–5 µg/kg/min for 4 hours were explored (figure 2). We found that a salbutamol infusion of 0.5 µg/kg/min after bolus with a ceiling rate of 20 µg/min achieved this (online supplemental figure 11).
Schematic representation of PKPD models for salbutamol plasma concentration, PASS and individual toxicity measurement. Simulations to show the effect of 0.25, 0.5, 1 and 5 µg/kg/min infusion for 4 hours after a bolus of 15 µg/kg, on the time course of salbutamol plasma concentration (pharmacokinetics), bronchodilation effect (efficacy) and the two important toxic effects (glucose increase and tachycardia). The virtual population of 2000 patients was weighing 5 to 54 kg. CL, clearance; F, absolute bioavailability; Inf., infusion; IV, intravenous; ka, absorption rate constant; neb., nebulised; PASS, Paediatric Asthma Severity Score; PKPD, pharmacokinetic–pharmacodynamic; Q, intercompartmental clearance; VC, apparent volume of distribution central compartment; VP, apparent volume of distribution peripheral compartment.
Discussion
Key results
A population PKPD model for intravenous salbutamol in children is reported. A two-compartment disposition model adequately described intravenous salbutamol PK. The clearance and volume of distribution estimates in children are comparable with those scaled from adult volunteers. This concurs with a published PK model.22 Dosing per body weight appears appropriate.
Our observed plasma concentrations were substantially higher than levels regarded as toxic in adults, but in-keeping with other reports of children receiving intravenous salbutamol according to current guidelines.22 27 Many patients had plasma salbutamol concentrations associated with near maximum possible effect before receiving any intravenous therapy.
PD modelling was performed for all toxicity variables, although correlation with salbutamol EC50 estimates were only demonstrated for heart rate Z-score and blood glucose. The blood glucose effect was independent of corticosteroids as a covariant. The concentration–response curve demonstrates that the maximum bronchodilation effect is achieved faster than the toxicity effects of hyperglycaemia and tachycardia.
The use of PKPD simulation is well established in drug development.28 Using our PKPD model, simulations showed that after a bolus of 15 µg/kg, an infusion of 0.5 µg/kg/min for 4 hours achieved our target of >90% of the maximum bronchodilation effect for 12 hours in the majority of patients. We report a ceiling effect and a maximum infusion rate of 20 µg/min is adequate. Higher infusion rates did achieve bronchodilation above 90%, but at the cost of increased toxicity.
Limitations
We aimed to recruit 100 patients but fewer patients received intravenous salbutamol, and research nurse availability contributed to recruitment failures. Fifty subjects were suggested for this population PKPD study and we stopped recruitment once that number was achieved. The requirement for three salbutamol concentration samples was also restrictive, with the protocol adjusted to include those with two samples after the first interim analysis (provided five clinical effectiveness measurements were available). The CATS cohort was more data-rich, but PASS scores were only obtained in non-ventilated patients. Height measurements were unavailable in most children (74.1%), which prevented the exploration of fat mass index on PK parameter estimates.
The high salbutamol concentrations at enrolment presented a mathematical challenge for EC50 modelling. This is a common phenomenon in pharmacology where a 10-fold range around the expected EC50 is desirable. The EC50 for effectiveness (ie, PASS) could not be reliably estimated and was fixed to FEV1 literature-based EC50 estimate of 1.15 ng/mL.29
Interpretation
A population PKPD model for intravenous salbutamol in children is described and can be used to inform future randomised controlled trials. Current dose regimens do achieve therapeutic bronchodilation but at the expense of adverse effects. A lower dose of 0.5 µg/kg/min after 15 µg/kg bolus would achieve the desired bronchodilation while minimising toxicity, with a ceiling effect at 20 µg/min. Demonstration that a 4-hour infusion sustained >90% maximum bronchodilation for 12 hours may obviate the need for high-dependency transfer. Understanding the salbutamol plasma–concentration effects on blood glucose and heart rate may allow the titration of ongoing salbutamol infusions to these variables.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics approval
This study involves human participants and was approved by Brent Research Ethics Committee (REC reference 14/LO/2103). R&D approval was obtained at all research sites. Participants gave informed consent to participate in the study before taking part.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Collaborators OSTRICH Study Group: Mark Peters, David Inwald, Ami Parikh, Ian Maconochie and Mario Cortina-Borja.
Contributors SW and PR created the study concept and design, sourced funding and MHRA approval. They formed a collaborative study group and assisted with patient recruitment and data collection. They provided clinical interpretation for the outcomes and wrote and approved the research paper. JFS, BJA, FK, YS and SP assisted with the study design. They performed the data analysis, PKPD modelling and simulation analysis and interpretation. They assisted with writing the research paper.The OSTRICH Study Group comprised the clinical leads for each research centre. They assisted with study design, patient recruitment and data collection. They edited and approved the final research paper. SW accepts full responsibility for the work and the conduct of the study, had access to the data, and controlled the decision to publish, and will act as guarantor.
Funding This study was funded by Great Ormond Street Children’s Charity Starter Grant (protocol no 13CC34 and EudraCT no 2014-002996-27). There was no involvement in the collection, analysis and interpretation of data or in the writing of the report by the funding body.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.