Inconsistencies in dosage practice in children with overweight or obesity: A retrospective cohort study

Abstract Obesity can affect the pharmacokinetics of most drugs, which may result in under‐ or overdosing if traditional pediatric dosing strategies are used. To investigate currently applied dosage strategies in children with overweight or obesity (overweight/obesity), in a clinical treatment facility. In particular, whether dosing guidelines were available and metrics of body size applied. A retrospective cohort study of 200 patients admitted to the Danish Children′s Obesity Clinic. Data were collected from 2007 to 2015. Overweight/obese children 3‐18 years were included if they had at least one drug prescription. Overall there were 658 prescriptions, primarily analgesics, psychotropics, asthma medications, and antibiotics. Except for one prescription, guidelines for dosage of overweight/obese children were not available in the clinic. In one prescription of gentamicin, the dose was adjusted by a metric body size. Otherwise dose was predominately prescribed either by total body weight or as fixed dose by age, in accordance with the recommendations of normal weight children. In drugs with a narrow therapeutic interval, we found large interindividual variations in dosing regimens, that is, for gentamicin, paracetamol, and prednisolone. Reduction of dose to the maximum recommended adult dose was common practice, when the dose calculated by total body weight (ie, mg/kg) exceeded this maximum. This study highlights the shortage of dosing guidelines in overweight/obese children. We found a large interindividual variability in dosage regimens, even in drugs with narrow therapeutic intervals. The clinicians rely on “best practice”, as evidence‐based dosage regimens are missing for many drugs prescribed during childhood.

numerous cancers. 1,2 Consequently, children who are overweight and obese are more likely to receive drug treatment than their normal weight peers. 3,4 1.1 | Drug dosing in normal weight children and challenges in overweight/obese children One of the most common methods for drug dosing in children in pediatric clinical practice is to extrapolate adult dose by body weight, eg mg per kilogram (kg). 5 This approach represents an a priori assumption of a linear relationship between weight and dose, as dose doubles with a twofold increase in the weight of a child. 5 Another common method for dosing in children is based on age, by dividing the pediatric population into subcategories of age. This method introduces an artificial discontinuity in the dose-response relationship across each age category. 5 As a result of the high weight relative to height in children with overweight/obese, a linear dosing by total body weight (TBW) might even exceed the maximum recommended dose in adults, while fixed dosage by age may result in subtherapeutic doses. 5 In addition, the principal pharmacokinetic (PK) determinants of drug dosing, that is, volume of distribution (V d ) and clearance (CL) are both often altered in obese patients. This is partly due to changes in physiology and body composition, with a relatively higher increase in fat (60%) compared to lean tissue (40%) per kg of TBW, as well as lean mass is more hydrated, which is attributed to increased extracellular water. 6 In addition, alterations in drug binding proteins, cardiac output, organ blood flows, and tissue perfusion may influence the PK. 7,8 Even though, the hepatic metabolism and the influence of obesity in children not has been studied extensively. 8 Further, PK data in this subgroup is not required for approval of new drugs by default by the Danish health authorities or the European Medicine Agency (EMA). The lack of clinical trials in children, who are overweight/ obese therefore prompts clinicians to extrapolate dosing strategies from studies in adults with overweight. However, such extrapolation strategies ignore differences in drug disposition (distribution and elimination) characteristics, reflecting children's general degree of immature metabolic pathways and physiological development. In addition, pathophysiological alterations accompanying obesity are assumed to have a similar influence on PK parameters in children, who are overweight/obese as in adults. 5,6,9,10 Ideal body weight (IBW), lean body weight (LBW), adjusted body weight (ABW) and allometric scaling, see Table 1. Whether the metrics are actually used in clinical practice is uncertain.

| Objectives
To investigate currently applied dosing strategies in children, who are overweight/obese, in a clinical treatment facility. In particular, whether dosing guidelines were available and metrics of body size applied.

| Study participants
Two hundred consecutively enrolled overweight/obese children aged 3-18 years, entering the chronic care multidisciplinary intervention program at the Children 0 s Obesity Clinic. Only children treated with at least one drug were included. Children had been referred from their general practitioners (GPs), school-and community-based doctors, or pediatricians. Exclusion criteria were a Body Mass Index (BMI) Standard Deviation Score (SDS) ≤ 1.28, which corresponds to the 90th percentile according to the Danish age and sex-adjusted references. 11

| Outcomes and measures
The following data were collected for each specific drug: Indication for treatment according to the International Classification of Diseases version 10 (ICD-10), 12  Distribution of all drug prescriptions in accordance to the ATC classification system is illustrated in Figure 1. Totally 658 prescriptions were registered, of these were 203 registered at the first visit.
A majority of the prescriptions (46%) were respiratory drugs (ATC group R), followed by drugs acting on the nervous system (ATC group N; 27%) , for example, central nervous system stimulants and analgesics, and drugs acting on the alimentary tract and metabolism (ATC group A; 9%), mainly laxative drugs and proton pump

| Dosing strategies
Distribution of the individual drugs prescribed at ATC-level five, and the dose regimen used in the follow up period is presented in

| DISCUSSION
This study gives an overview of clinical practice in dosing of overweight/obese children in different pediatric settings. We found no specific dosing guidelines available for overweight/obese children in the investigated clinics for any of the recorded drugs. As limited data are available in overweight/obese children for most drugs, the results of this study will most likely apply to other pediatric settings in Denmark as well as internationally.

| Antibiotics
In line with other studies, 16,17 we found a noticeable prescription of antibiotics (ATC group J), of which, it is well known that in addition to therapeutic failure, inappropriate dosage can lead to antibiotic resistance, and drug-related toxicity. Nonetheless, we found only one overweight/obese child dosed by an adjusted metric of body size, which was gentamicin dosed by ABW in treatment of pyelonephritis. It is generally accepted that the initial empiric dose (loading-dose) is critical to achieve an early high peak concentration of gentamicin that directly correlates with maximal killing of the pathogen. 18 One PK study of gentamicin has been performed in overweight/obese children, 18   Despite their widespread use, there is very little information regarding dosing penicillin in obese patients. 20 We found phenoxymethyl penicillin prescribed as a fixed dose in accordance with adult regimen in 7 out of 8 children. Consequently, a 6-year-old child had an approximately fourfold higher daily dose per kg body weight, than the 17-year adolescent (88 mg kg À1 day À1 vs 21 mg kg À1 day À1 ).
For penicillin's in which time of the drug concentration above the minimum inhibitory concentration (T > MIC) is important, it follows that increasing doses or frequency will improve the PD. Furthermore, the therapeutic index for penicillin's is wide, and it therefore seems reasonable to dose penicillin by TBW in overweight/obese children. However, this has to be investigated further, and PKPD studies of penicillin in children, who are overweight/obese are therefore in demand. 16,20

| Analgesics
We found a large interindividual dosing of paracetamol, and the capping dose was common practice. In a recent study by Rongen et al., 21 the pharmacokinetics of paracetamol in morbidly obese adult patients was investigated, with a specific emphasis on Cytochrome (CYP) 2E1 mediated CL. Paracetamol plasma concentrations were significantly lower in obese patients. Thus, an increased dose of paracetamol may be anticipated to achieve a better pharmacodynamic response in obese patients. However, the induced CYP2E1activity may also worsen the safety profile of paracetamol due to higher concentrations of the toxic metabolite N-acetyl-p-benzoquinone (NAPQI). As paracetamol is the most prescribed drug in pediatric wards internationally, 22 PKPD studies in overweight/obese children is of outmost importance to clarify this question.

| Ibuprofen
We found ibuprofen primarily prescribed as a fixed dose of 400 mg.
Reduced peak, increased V d has been found for ibuprofen in adults with obesity, 23 indicating dose may have to be increased. Ibuprofen has not been investigated in children overweight/obese. Due to common severe adverse effects such as gastrointestinal bleeding; increasing dose by extrapolation of adult data cannot be recommended as standard in this population, without prior investigations.

| Medication for obstructive airways
We found that all the registered drugs in asthma treatment were prescribed as fixed dose by age. Lower therapeutic responsiveness

| Antiepileptic drugs
We found one and four prescriptions of diazepam and lamotrigine, respectively. Both antiepileptic drugs are widely used in pediatric patients. Diazepam is a highly lipophilic drug, which has been found to have an increased V d in obese adult patients. 28 As V d is the primary PK parameter of importance when calculating loading dose, the initial dose may have to be increased in overweight/obese children.
Opposite, lamotrigine is used for maintenance treatment of epilepsy.
CL is therefore the primary PK parameter of importance when dosing. However, the effect of obesity on lamotrigine PK has not been investigated.
The lack of sufficient data on how to dose children who are overweight and or obese in acute care, such as antibiotics, anti-epileptics, and drugs with narrow therapeutic interval, that is, gentamicin, paracetamol, and diazapam is concerning.
In the light of the prevalence of overweight/obese, especially in children has increased over the last several decades. At present, approximately 31.8% of US children aged 2-19 years are considered overweight or obese. 29 Among European countries, overweight/ obese in children below the age of ten ranges from more than 40% in southern Europe to less than 10% in northern Europe. 29 In Denmark, the combined prevalence of childhood overweight/obese is 10%-12%, in preschool children, 30 increasing to 15%-22% during adolescence. 31 Given prevalence is based on age-and gender-specific BMI. However, definitions of overweight/obese and the reference populations differ between studies and countries. 29

| Main therapeutic classes
In accordance with previous epidemiological studies, we found a high use of respiratory drugs, for example, inhaled steroids and short-acting beta-2 agonists 3 and drugs acting on the nervous system. In group N, we found that both methylphenidate and antipsychotics such as quetiapine and aripiprazole were highly prevalent. Of notice, the use of antipsychotic agents is considered very problematic in this population, due to side effects such as weight gain and metabolic syndrome. 32 48 In this study, we found that the capping dose was common practice, when the dose exceeded the maximum recommended adult dose. This strategy seems pragmatic, but relies, however, on a precautionary approach rather than evidence, and may not always be applicable. For example, carbamazepine, phenytoin, digoxin, and propofol should be dosed more frequently and in higher dose per kg body weight in normal weight children than in adults. 17 On the other hand, there exists a well-known nonlinear relationship between bodyweight and physiological factors such as clearance (CL). Thus, providing maximum adult dose even though limited, may result in supratherapeutic doses. 5

| Strengths and limitations
A major strength of this study is that through manually review of patient records, we were able to define the chosen dosage strategy for each prescription, and included a long-term period of follow up.
Some obvious limitations are the retrospective design and the method of chart audit with missing data and an underestimation of drug prescription by GPs during follow up. Further, the pharmacodynamics (PD) was not investigated in this study, and due to the retrospective nature of the study, we did not have valid or even the necessary measurements ie visual analog scores (VAS) in accordance to measure effects of analgesics.

| CONCLUSION
This study highlights the shortage of dosing guidelines in children, who are overweight/obese. We found a large interindividual variability in dosing regimens, even in drugs with narrow therapeutic intervals. The clinicians rely on "best practice", as evidence-based dosage regimens are missing for many drugs prescribed during childhood.

| Perspective
Especially prospective, PK/PD and randomized studies of different dosage strategies are needed to provide information on optimal dosing in children, who are overweight/obese. Meanwhile we encourage the associations of pediatricians in collaboration with clinical pharmacologists and other health care professionals, for example, pediatric pharmacists to standardize dosage guidelines. Individualized dosing strategies and PKPD modeling may be useful tools in this population.

DISCLOSURES
None declared.