Clinical Pharmacokinetics

, Volume 43, Issue 3, pp 145–156

Pharmacokinetics of Ibuprofen in Children with Cystic Fibrosis

Authors

  • Emily E. Han
    • Laboratory for Pharmacodynamic Research, School of PharmacyUniversity of Southern California
    • Adult Cystic Fibrosis CenterUniversity of Southern California
    • Laboratory for Pharmacodynamic Research, School of PharmacyUniversity of Southern California
    • Adult Cystic Fibrosis CenterUniversity of Southern California
    • Keck School of MedicineUniversity of Southern California
  • Stan G. Louie
    • Laboratory for Pharmacodynamic Research, School of PharmacyUniversity of Southern California
    • Keck School of MedicineUniversity of Southern California
  • Mark A. Gill
    • Laboratory for Pharmacodynamic Research, School of PharmacyUniversity of Southern California
  • Bertrand J. Shapiro
    • Adult Cystic Fibrosis CenterUniversity of Southern California
    • Keck School of MedicineUniversity of Southern California
Leading Article

DOI: 10.2165/00003088-200443030-00001

Cite this article as:
Han, E.E., Beringer, P.M., Louie, S.G. et al. Clin Pharmacokinet (2004) 43: 145. doi:10.2165/00003088-200443030-00001

Abstract

An exaggerated inflammatory response is responsible for the decline of lung function in patients with cystic fibrosis (CF). Ibuprofen is a potent anti-inflammatory agent that demonstrates inhibition of neutrophil activity in vitro at concentrations between 50 and 100 mg/L, whereas lower concentrations result in an increase in inflammatory mediators. Significant decline in the rate of deterioration of pulmonary function and increased nutritional status were observed in children with CF who were administered long-term high-dosage ibuprofen therapy.

As with many other drugs, CF patients appear to exhibit altered pharmacokinetics of ibuprofen (reduced bioavailability, increased volume of distribution, and more rapid clearance) when compared with healthy controls. However, the absence of studies with intravenous ibuprofen as well as protein binding measurements in patients with CF currently limits the ability to compare the pharmacokinetics with those in other populations. Current studies indicate that there is high interpatient variability in ibuprofen pharmacokinetics among CF patients. Some of this variability can be explained by differences in ibuprofen formulation administered.

Therapeutic drug monitoring of high-dosage ibuprofen therapy is recommended because of the biphasic response to inflammatory mediators demonstrated in vitro as well as the high interpatient variability in pharmacokinetics. Due to the differences in absorption characteristics between ibuprofen formulations, the timing of obtaining blood samples for pharmacokinetic analysis is critical. Maximum a posteriori Bayesian analysis has been shown to provide more accurate and precise estimates of the pharmacokinetic parameters of ibuprofen in children with CF, and may also be a useful tool to further investigate the relationship between measures of drug exposure and efficacy/toxicity outcomes.

Cystic fibrosis (CF) is an autosomal recessive genetic disorder that affects nearly 30 000 individuals in the US.[1] CF is caused by a defect in a single gene, the cystic fibrosis transmembrane conductance regulator (CFTR).[2,3] CFTR functions as a chloride channel, which is present predominantly within epithelial cells. Mutations in the CFTR gene target several organ systems resulting in the classic clinical manifestations of chronic airway infection/inflammation, pancreatic insufficiency, meconium ileus and infertility.[4] Pulmonary complications account for over 90% of the morbidity and mortality in patients with CF and therefore serve as the focus of treatment interventions.[1] In particular, chronic airway inflammation leads to bronchiectasis and eventual respiratory failure.

1. Airway Inflammation in Cystic Fibrosis

Inflammation is thought to arise from the persistent endobronchial infections in patients with CF. Defective CFTR in the lung causes decreased chloride secretion with a compensatory increase in sodium and water absorption, resulting in desiccated mucus within the lumen of the airways. This thickened mucus is resistant to clearance by the mucociliary escalator and provides an environment conducive to bacterial growth.[4,5]

The presence of bacteria within the airways stimulates lipopolysaccharide (LPS)-mediated activation of nuclear factor (NF)-κB, a central intracellular regulator of the inflammatory response. NF-κB promotes the transcription of proinflammatory cytokines, including interleukin (IL)-1β, tumour necrosis factor (TNF)-α, IL-6 and IL-8 (see figure 1);[6,7] IL-8 in particular plays a significant role in promoting airway inflammation due to its potent neutrophil chemoattractant activity.[8] In addition, IL-10, which inhibits IL-8, IL-1β and TNF-α expression in normal lung, is either found to be at very low or undetectable levels in CF patients. This suggests that low levels of IL-10 may permit an exaggerated inflammatory response in CF.[7] Elastase produced by neutrophils impairs cellular immunity by cleaving opsonins and receptors necessary for phagocytosis of organisms. In addition, neutrophils degrade elastins, resulting in structural damage to the lung, and stimulate release of additional chemoattractants. The chemoattractants, especially leukotriene (LT) B4 and IL-8, stimulated both by neutrophils and elastase production result in further neutrophil recruitment and perpetuation of the vicious cycle of infection and inflammation (figure 1). Neutrophil turnover may further exacerbate airway obstruction through release of DNA.
https://static-content.springer.com/image/art%3A10.2165%2F00003088-200443030-00001/MediaObjects/40262_2012_43030145_Fig1.jpg
Fig. 1

Mechanism of anti-inflammatory action of ibuprofen. IL = interleukin; LT = leukotriene.

2. Anti-Inflammatory Treatment

Anti-inflammatory therapy of CF has received much attention over the past several years due to recognition of the relatively important role that chronic airway inflammation plays in the progressive decline in pulmonary function.

The efficacy and safety of oral corticosteroids was evaluated in a large, multicentre, double-blind, placebo-controlled trial.[9] Patients with CF between 5 and 14 years of age were randomised to receive prednisone 2 mg/kg (high dosage) every other day (n = 95), prednisone 1 mg/kg every other day (n = 94) or placebo (n = 94). Although interim analysis demonstrated significantly greater decline in pulmonary function in those receiving placebo, an increased frequency of cataracts, growth retardation and glucose abnormalities were noted in the high-dosage prednisone group. Due to these adverse events, the study was terminated prematurely.

Because of their more benign safety profile, the focus switched to the usage of nonsteroidal anti-inflammatory drugs (NSAIDs) in an effort to reduce inflammation in CF patients. Ibuprofen in particular is suitable for use in children with CF because of its widespread use in the paediatric population as a whole.

3. Mechanism of Action of NSAIDs

NSAIDs inhibit cyclo-oxygenase (COX) activity, which is necessary for the synthesis of various prostaglandins, including prostaglandin (PG) E2.[10] PGE2 is a potent vasodilator that promotes blood flow to the inflamed region, resulting in local oedema and leucocyte infiltration. As a class, NSAIDs do not affect the lipoxygenase pathway, which is responsible for synthesis of LTs, including LTB4, a potent neutrophil chemoattractant.

Ibuprofen is a nonselective COX-1 and COX-2 inhibitor that exists as two enantiomers, the (R)- and (S)-isomers. When administered, the (R)-isomer is partially inverted to the (S)-isomer. The (S)-isomer is about 160 times more potent than the (R)-isomer as an inhibitor of COX-1 and COX-2.[11,12] However, the inhibition of COX alone does not fully explain all of the anti-inflammatory responses (see figure 1). At high doses, ibuprofen inhibits the lipoxygenase pathway by an unknown mechanism (possibly by interfering with the release of LTB4). Therefore, ibuprofen decreases synthesis of both PGE2 and LTB4 at high doses. In addition, it also inhibits the activation of NF-κB by stabilising the complex between NFκB and its inhibitor IκB, which in turn decreases production of IL-8 and other chemokines responsible for the intense inflammatory process.

Whether the newer NSAIDs (i.e. the selective COX-2 inhibitors) provide similar activity against LTB4 at high doses is not known. In fact, they may be detrimental in that they might promote shunting of arachidonic acid down the lipoxygenase pathway and result in even greater LTB4 production. Until there are data showing that newer NSAIDs decrease neutrophil chemotaxis, their clinical role in CF remains unclear.

4. Pharmacodynamics of Ibuprofen

Ibuprofen has shown to inhibit the migration, adherence, swelling and aggregation of neutrophils in animal models and humans. An important preclinical study that prompted researchers to consider clinical trials of ibuprofen in CF patients was conducted by Konstan et al.[13] To simulate chronic Pseudomonas aeruginosa infection, rats were inoculated with P. aeruginosa embedded in agarose gel and were treated with either 35 mg/kg of oral ibuprofen or placebo every 12 hours. Ibuprofen-treated rats had significantly reduced areal percentage of inflammation caused by P. aeruginosa infection as compared with the placebo-treated group (p < 0.05). Importantly, the reduction of inflammation was not accompanied by an increase in bacterial burden. Better attenuation of inflammation in the peripheral area, away from the site where P. aeruginosa was inoculated, was noted, suggesting that ibuprofen exhibited greater response against inflammation triggered by recruitment of neutrophils rather than inflammation directly stimulated by the infection itself. To further investigate the mechanism of the anti-inflammatory action, neutrophil-rich leucocytes were collected from several rats and stimulated with a calcium ionophore in the presence of increasing concentrations of ibuprofen. Ibuprofen inhibited production of PGE2 at concentrations as low as 5 mg/L. LTB4, interestingly, exhibited a biphasic response; the production increased at lower concentrations but decreased at concentrations greater than 25 mg/L (see figure 2). These data indicate that the dosage of ibuprofen is critical in attaining the target concentration necessary to exert the desired anti-inflammatory response.
https://static-content.springer.com/image/art%3A10.2165%2F00003088-200443030-00001/MediaObjects/40262_2012_43030145_Fig2.jpg
Fig. 2

Effect of ibuprofen on production of leukotriene (LT) B4 and prostaglandin (PG) E2 by rat neutrophil-rich leucocytes (reproduced from Konstan et al.,[13] with permission).

The benefit of high-dosage ibuprofen in CF patients was demonstrated in a randomised, double-blind, placebo-controlled trial.[14] Eighty-five CF patients, 5–39 years old, were given high-dosage ibuprofen (16.2–31.6 mg/kg) over a period of 4 years to achieve a target concentration between 50 and 100 mg/L. In an initial phase, plasma concentrations of ibuprofen were obtained every 30 minutes for 3 hours after administration of the drug to enable determination of peak concentration (Cmax) and the time at which it occurred (tmax) in each patient. A target Cmax of 50–100 mg/L of ibuprofen was chosen based on the in vitro data demonstrating inhibition of LTB4 production in this concentration range.[13] The results of this study showed that the use of high-dosage ibuprofen was associated with a slowing of the progressive decline in pulmonary function, as measured by forced expiratory volume in 1 second (FEV1), and improvements in nutritional status, as measured by percentage of ideal bodyweight. The beneficial effects of ibuprofen were greatest in patients who were less than 13 years old at the time of study enrolment. In this subset of patients, FEV1 declined by less than 2% of predicted value compared with a 15% decline in the placebo group over the 4-year treatment period. The results of this study confirmed that the benefits of ibuprofen seen in the animal model could be replicated in CF patients when ibuprofen peak concentrations are maintained between 50 and 100 mg/L.

One limitation to this study is the lack of identification of specific pharmacodynamic goals. All patients in this trial had their dosage of ibuprofen adjusted to achieve similar Cmax (50–100 mg/L). The relationship between various levels of drug exposure (Cmax or area under the concentration-time curve [AUC]) and reduction in airway inflammation (e.g. sputum neutrophil counts) or toxicity (e.g. gastrointestinal ulcers/bleeding, renal insufficiency) is therefore unknown. Due to variability in the clearance of ibuprofen within the population, drug exposure as measured by AUC is likely to vary considerably between individuals despite similarity in Cmax. It would be interesting to know whether there is an improved correlation between the anti-inflammatory action as well as toxicity of ibuprofen in vivo with the level of drug exposure as measured by AUC.

In an attempt to address these pharmacodynamic issues Konstan et al.[15] performed a prospective open-label study of ibuprofen in 16 adult CF patients. Patients were grouped based on the dose of ibuprofen administered, ranging from 2–3 mg/kg up to 20–30 mg/kg. All doses were administered every 12 hours for a period of 10 days. The pharmacodynamic outcome measure was the change in oral mucosal polymorphonuclear neutrophil (PMN) content from baseline as a surrogate for neutrophil migration to the lungs. The results of this study demonstrated a 31% reduction in PMNs in patients in whom peak ibuprofen concentrations exceeded a threshold value of 50 mg/L. In contrast, a 40% increase in PMNs was noted when peak ibuprofen concentrations were below 50 mg/L. A significant relationship between the ibuprofen AUC and PMN migration was also noted with a threshold value of 11.0 mg · min/mL. Unfortunately, since the dosing interval was the same for all patients, both Cmax and AUC were highly correlated with one another precluding the ability to determine which is the better predictor of anti-inflammatory response.

In the absence of definitive pharmacodynamic goals, therapy should be designed to achieve ibuprofen Cmax greater than 50 mg/L in order to attain the clinical improvements noted in the CF trial.[14]

5. Pharmacokinetics of Ibuprofen in Healthy Volunteers

Absorption of ibuprofen in the healthy population differs depending on the formulation of ibuprofen, but the drug is generally rapidly absorbed. Ibuprofen solution is absorbed most quickly (tmax < 0.25 hours), followed by the tablet (tmax ∼2 hours) and other formulations (i.e. suppository). Absorption is delayed when ibuprofen is given with food.[11]

Ibuprofen is extensively bound to plasma proteins. At a plasma concentration of 20 mg/L, ibuprofen is 99% protein bound to albumin.[11,16] In order to determine the relationship between plasma protein binding and resultant total plasma AUC, 15 white men were given one, two or three tablets of ibuprofen 400mg in a 3-week crossover study and ibuprofen suspension 400mg in the fourth week.[17] The results revealed that the total plasma AUC of ibuprofen increased nonlinearly with increasing doses of ibuprofen. The plot of total plasma AUC against dose of ibuprofen plateaued with increasing doses, whereas the relationship between free plasma AUC and dose was linear.

Since AUC based on total plasma concentrations was nonlinear with increasing doses, apparent (‘oral’) clearance (CL/F) and bioavailability (F) calculated based on total plasma AUC were dose dependent. In contrast, CL/F and F calculated from free ibuprofen plasma AUC were independent of dose, as expected for a compound that exhibits first-order kinetics.[17] Therefore, ibuprofen binds to protein in a nonlinear fashion, leading to disproportional increases in total plasma AUC with increasing doses.

Ibuprofen exists as a racemic mixture of both (R)- and (S)-isomers. Once racemic ibuprofen is administered, approximately 63% of the (R)-isomer is converted to the (S)-isomer, the pharmacologically more potent conformation.[18] Both (R)- and (S)-ibuprofen are metabolised to two major metabolites, 2-hydroxy-ibuprofen and carboxy-ibuprofen. Phase I metabolism is mediated via cytochrome P450 (CYP) 2C9. These metabolites undergo further metabolism with glucuronide conjugation, which is stereoselective for the (S)-isomer (see figure 3).[11,18,19] These are inactive metabolites that are eliminated through urinary clearance.
https://static-content.springer.com/image/art%3A10.2165%2F00003088-200443030-00001/MediaObjects/40262_2012_43030145_Fig3.jpg
Fig. 3

Metabolic pathway of racemic ibuprofen. About 63% of the (R)-isomer is unidirectionally inverted to the (S)-isomer. Both isomers undergo oxidation via cytochrome P450 (CYP) 2C9 to form two active metabolites, 2-hydroxy-ibuprofen and carboxy-ibuprofen (the two-step formation of carboxy-ibuprofen involves hydroxylation followed by carboxylation). The metabolites are then conjugated via glucuronidation and excreted in the urine.

The mean elimination rate constant in healthy volunteers is similar when determined using total and free plasma concentrations, but there are significant differences between the dosage forms. Even when corrected for plasma protein binding, a slower mean elimination rate was detected following administration of the tablet formulation when compared with the suspension (0.297 and 0.349 h−1, respectively; p < 0.05). This is thought to be due to difficulty in differentiating absorption and elimination phases with the relatively longer absorption time observed with the tablet formulation. The tmax is independent of dose; however, it is shorter with the solution than with the tablet formulations of ibuprofen.[17]

6. Pharmacokinetics of Ibuprofen in Children with Cystic Fibrosis

The pharmacokinetics of a number of compounds have shown to be altered in patients with CF. Pancreatic insufficiency impairs absorption of fat-soluble compounds, as evidenced by low levels of vitamins A, E and D. In addition, enhanced clearance and increased volume of distribution (Vd) have been demonstrated for penicillins, aminoglycosides, theophylline and others.[20,21] Although the mechanism for the altered pharmacokinetics has not been clearly identified, the changes in Vd may be due to differences in body composition. Altered organic ion exchange mechanisms in the kidney have been postulated as a reason for the altered clearance in patients with CF. In addition, mild differences in hepatic drug metabolism have been reported in patients with CF.[22] These alterations are important to consider when initiating therapy in patients with CF in order to ensure the optimal dosage to achieve the desired effect.

Currently, there are four studies available that extensively evaluated the pharmacokinetics of high-dosage ibuprofen in CF patients.[2326] Table I provides a summary of pharmacokinetic parameters from the four studies.
https://static-content.springer.com/image/art%3A10.2165%2F00003088-200443030-00001/MediaObjects/40262_2012_43030145_Tab1.jpg
Table I

Pharmacokinetics of high-dosage ibuprofen therapy in patients with cystic fibrosis

6.1 Absorption

As has been demonstrated with other compounds, CF children exhibit altered absorption of ibuprofen. When CF and healthy children were given similar doses of ibuprofen based on bodyweight, Cmax and AUC in CF children was 27% and 46% (respectively) lower than in healthy children.[23] Greater intersubject variability in the pharmacokinetics of ibuprofen among CF children was also noted, and may in part be explained by differences in dosage forms of ibuprofen administered.

Ibuprofen is commercially available within the US in tablet, suspension and chewable tablet formulations. The pharmacokinetics of the various formulations of ibuprofen have been investigated in two studies.[25,26] The tmax, in particular, appears to differ significantly depending on the dosage form used.[26] On average, Cmax in patients taking ibuprofen 20–25 mg/kg is achieved within 60–120 minutes and 30 minutes following administration of tablet or suspension formulations, respectively.[25,26] The tmax for chewable tablets would be expected to be intermediate between suspension and tablet formulations due to partial disintegration; however, the tmax was not significantly different from that of the standard tablet formulation at 1.5 hours.[26] Improper chewing of the chewable formulation before swallowing may account for the unexpectedly long tmax.

The tmax is longer in CF children younger than 5 years of age when compared with older children (p < 0.02). One explanation might be that younger children with CF have a greater incidence of gastroesophageal reflux disease (GORD), thus prolonging tmax.[26]

Attempts have been made to describe effects of different doses on tmax and Cmax.[23,25] Dose escalation studies of ibuprofen in CF children suggest that Cmax and AUC increase with increasing doses, although not linearly (see section 6.2), while no change was seen in tmax.[23] This finding is consistent with saturable plasma protein binding. In contrast, others have noted a variable tmax based on different doses administered (p < 0.05, r = 0.594), but noted no significant correlation between Cmax and dose (r = −0.15).[25] Alteration in tmax with dose would suggest a saturable absorption process; however, this was not a dose-escalation study and therefore was not adequately designed to address dose-dependent differences in pharmacokinetics. In addition, there was no effect of sex on tmax in CF children.[26]

Another factor that may affect the pharmacokinetic measures of absorption is differences in bioequivalency between brands of ibuprofen. Because the absolute bioavailability of ibuprofen is not established, most investigators have reported pharmacokinetic parameters normalised for bioavailability.[2326] Bioequivalency testing ensures that various generic formulations will be similar in attainment of plasma concentrations. Two different formulations are considered bioequivalent if the geometric mean ratio of the parameters that measure rate and extent of absorption, Cmax and tmax and Cmax and AUC, respectively, are within the limits of 80–125%.[27] This relatively wide range may contribute to the differences in Cmax and tmax reported. In one study, a higher Cmax and delayed tmax were observed in two cases in which patients received a different brand of ibuprofen. After switching to a brand name ibuprofen, a lowering of dose was necessary to achieve the desired Cmax.[25]

Protein Binding and Volume of Distribution

Ibuprofen is extensively bound (>98%) to plasma albumin and the binding appears to be capacity limited.[11,17] As discussed in sections 5 and 6.1, saturation of protein binding results in a nonlinear relationship between dose and AUC and also between dose and total plasma ibuprofen concentrations; however, when analysed using unbound ibuprofen concentrations the relationships were found to be linear.[17] It is not known to what extent the nonlinearity between dose and total ibuprofen concentrations exists in patients with CF, since few dose-escalation studies have been performed in this population. Fewer still have evaluated unbound concentrations of ibuprofen. Pancreatic insufficiency causes malnutrition in many patients with CF, resulting in lower plasma protein concentrations. Reduced plasma albumin concentrations would be expected to alter the free fraction of ibuprofen due to a lower total ibuprofen concentration, while the unbound concentration would be expected to remain unchanged since ibuprofen is not a high extraction ratio drug.[28] Preliminary studies in patients with CF suggest that a nonlinear relationship between dose, Cmax and AUC does exist in CF children as seen in healthy adults. For example, a 92% increase in mean dose resulted in less than proportional increase in Cmax (74%) and in AUC (86%) [p < 0.05] .[23] Although saturable protein binding is the likely explanation for the nonlinearity, saturable absorption is another possible explanation; however, the tmax was shown to be independent of dose as previously described.[23] To provide a more accurate representation of the efficacy of a dosage regimen, it may be necessary to measure unbound concentrations of ibuprofen, particularly in those patients with reduced plasma protein concentrations.

Another indication of the potential importance of plasma protein binding is the observation of a higher Vd/F in CF patients when compared with healthy children (291 ± 91 vs 158 ± 43 mL/kg, respectively; p < 0.01) when similar doses of ibuprofen were given (13.4 ± 4.1 vs 13.9 ± 7 mg/kg, respectively).[23] Decreased protein binding, resulting in lower total concentrations, is the likely explanation for the greater Varea/F (apparent volume of distribution based on trapezoidal AUC and elimination rate) in CF children. Alternatively, the differences could be attributed to reduced bioavailability or an induction of first-pass clearance. However, CYP2C9 activity, an enzyme responsible for the metabolism of ibuprofen, is not upregulated in patients with CF and therefore induction of first-pass clearance is unlikely to be responsible for the increased Varea/F.[29]

6.3 Elimination

Upon receiving similar doses, the CL/F of ibuprofen in CF has been shown to be markedly increased compared with healthy control subjects (2.3 ± 0.6 vs 1.3 ± 0.2 mL/min/kg, respectively; p < 0.01). The elimination half-life was not found to be different between the two groups of children due to the corresponding increase in Vd in patients with CF, as noted in section 6.2.[23] Several possible explanations exist for the differences in clearance. Since CL/F and Vd/F are uncorrected for bioavailability, the measured clearance may be a result of decreased bioavailability. Alternatively, since CL/F and Vd/F estimations are based on total ibuprofen concentrations, saturable protein binding may partially explain the increased CL/F in CF children.

Another factor that contributes to the enhanced clearance of ibuprofen in CF children is increased activity of glucuronidation in this population.[22] As discussed in section 5, ibuprofen exists as a racemic mixture and following administration the majority of the (R)-isomer unidirectionally inverts to the more active (S)-isomer. However, both isomers undergo oxidative metabolism. The metabolites then undergo conjugation via glucuronidation (figure 3). Prior studies indicate that there is increased glucuronidation activity in patients with CF, which is consistent with the overall findings of increased clearance noted in the controlled study by Konstan et al.[23]

The stereoselective pharmacokinetics of ibuprofen have been well described in healthy adults and children, and recently data in CF children have become available. Speculations that children with CF may differ from healthy children in the chiral inversion of ibuprofen secondary to increased clearance in CF children led to a study examining the differences in pharmacokinetic parameters of the two ibuprofen isomers.[30] As with CL/F of the racemic mixture based on total concentrations, CL/F for both isomers was higher in children with CF compared with febrile control children. This finding was also evident after adjustment for bodyweight. In addition, the clearance of the (R)-isomer was more rapid when compared with that of the (S)-isomer, suggesting that the rate of chiral inversion exceeds the elimination rate via oxidative metabolism.

One issue of significance to dosage optimisation is the degree of intersubject variability in the pharmacokinetics of ibuprofen. In efforts to better characterise the intersubject variability, the pharmacokinetics of ibuprofen have been examined to determine their relationship with certain patient characteristics. The clearance of ibuprofen has been shown to correlate significantly with age, weight and body surface area (BSA). When the dosage was normalised to BSA, no age-specific differences were noted; therefore, some researchers advocate the use of BSA in determining the dosage of ibuprofen.[24] Additional studies are needed to determine the relative precision in achieving the target Cmax when dosage is determined according to BSA versus traditional weight-based nomograms.

7. Therapeutic Drug Monitoring

7.1 Blood Sampling Times

High-dosage ibuprofen therapy (20–30 mg/kg every 12 hours) with a target Cmax of 50–100 mg/L is recommended in children with CF for anti-inflammatory therapy. The Cystic Fibrosis Foundation guidelines[31] recommend evaluating ibuprofen blood concentrations at 1, 2 and 3 hours after administration to assess the adequacy of high-dosage ibuprofen therapy. Visual inspection of the serum concentration-time curve is then used to determine the appropriate dosage by using a proportional dosage adjustment.

The initial human study of high-dosage ibuprofen therapy was a dose-escalation study to determine the adequate dosage in CF children to achieve the target Cmax.[23] A mean Cmax of 48 ± 17 mg/L was observed at mean tmax of 66 ± 20 minutes after the tablet dose of 13.4 mg/kg. Subsequent studies measuring ibuprofen plasma concentrations at more frequent time intervals, however, found that there is a high degree of variability in tmax among CF patients. Although the median tmax was 1 hour for patients taking ibuprofen tablets, 33% (4 of 12) of the patients reached tmax in less than 1 hour, and 8% (1 of 12) exceeded 4 hours before reaching the Cmax.[26] In addition, as mentioned in section 5, the tmax varies between dosage forms.[25,26] This suggests that measuring the first sample at 1 hour, as recommended in the guideline, may miss the concentration peak in a significant number of patients, potentially resulting in inappropriate assessment of the adequacy of the dose.

Alternative sampling schemes have been recommended to more specifically address the differences in absorption characteristics between formulations. To adequately assess pharmacokinetic data, samples should be obtained before and 1, 2 and 3 hours after the administration of tablets, and before and 0.5, 0.75 and 1 hour after the administration of suspension.[25,26] However, both of these methods rely on approximation of the Cmax based on visual inspection of the three plasma concentrations obtained. A more precise method of determining the actual Cmax by using D-optimal sampling and maximum a posteriori (MAP) Bayesian estimation has recently been described.[32]

7.2 Optimisation of Sampling Time and Analysis

D-optimal sampling is a method used to maximise information about the parameters within a system by obtaining precisely timed measurements. This method has been applied to the design of sampling schedules for pharmacokinetic experiments. A minimum of one sample is typically required for each parameter to be estimated in the model. The timing of the sample is determined to coincide with the time when information about the parameter to be estimated is maximal.

This method was recently applied to study the pharmacokinetics of ibuprofen.[32] In this study, individual pharmacokinetic analysis was performed on plasma concentration data obtained from 32 children with CF (22 receiving suspension and 10 receiving tablet formulations). The pharmacokinetics were best described using a one-compartment model with first-order absorption (with a lag time). D-optimal sampling times were then determined for each individual. The optimal sampling strategy for the population was determined from the frequency distribution of the individual D-optimal times. The optimal sampling times were reported to be 0, 0.25–0.5, 1–1.5 and 5 hours for tablets, and 0, 0.25–0.5, 1 and 3–4 hours for the suspension.

MAP Bayesian analysis is a tool that has proven valuable to optimisation of the dosage of aminoglycosides and other compounds in which therapeutic drug monitoring is frequently performed. MAP Bayesian analysis determines the individualised a posteriori parameters using a search algorithm that seeks to minimise the difference between the fitted and measured concentration data as well as the errors between the a priori and a posteriori parameter values. MAP Bayesian analysis is particularly useful when applied to sparse data sets, as in the clinical setting. We recently evaluated the application of MAP Bayesian analysis to determine the pharmacokinetics of ibuprofen in children with CF. MAP Bayesian analysis performed on the plasma concentration data utilising only the four D-optimal times resulted in unbiased and precise estimates of the pharmacokinetic parameters when compared with maximum likelihood analysis performed using seven concentrations. These data demonstrate that precise estimation of the dosage necessary to achieve the desired pharmacokinetic target (Cmax or AUC) can be achieved with four optimally timed blood samples.

Due to variability in clearance, it is possible that patients could have Cmax in the desired range but have very different levels of drug exposure as measured by AUC. Although a definitive relationship between the efficacy and safety of ibuprofen and specific AUC values has not been identified, AUC provides a better assessment of the degree of drug exposure when compared with a Cmax measurement and may therefore be a better predictor of long-term suppression of PGE2 and LTB4 production. The current guidelines recommend Cmax monitoring in order to facilitate outpatient monitoring of therapy; however, the Cmax is relatively insensitive to differences in clearance within the population. Extending the last sample time by 1–2 hours would enable determination of both Cmax and AUC. The dose could then be individualised to achieve the target Cmax and AUC goals to account for differences in both clearance and Vd within the population. MAP Bayesian analysis enables determination of the Cmax and AUC for a given dose regimen, and may therefore provide a useful tool for further exploring the relationship between other measures of drug exposure and efficacy/safety outcomes utilising routine clinical data.

8. Adverse Effects of High-Dosage Ibuprofen

The most frequently reported adverse effects with high-dosage ibuprofen in patients with CF include abdominal pain, conjunctivitis and epistaxis.[14] Currently only one trial has evaluated whether there is a relationship between the occurrence of these adverse effects and various measures of drug exposure (Cmax and AUC).[23] Unfortunately, the two adverse effects experienced most frequently, epistaxis and emotional lability, could not be related to plasma concentrations of ibuprofen, most likely due to the small number of patients enrolled in the trial. A larger sample is needed to establish proper relationships between adverse effects and Cmax or AUC.

Abdominal pain has also been reported in patients taking high-dosage ibuprofen; however, abdominal pain is a manifestation of CF disease itself and alleviation of abdominal symptoms with long-term high-dosage ibuprofen therapy has been reported.[14,23] One case of epistaxis and conjunctivitis was clearly associated with ibuprofen, based on the observation that the condition improved upon discontinuation of the study drug and worsened when the drug was resumed.[14]

Individual cases have reported pyloric channel stricture with the use of high-dosage ibuprofen,[33] and vestibular toxicity with renal failure in a CF patient taking gentamicin and standard doses of ibuprofen for relief of headache.[33,34] Although it has not been reported in these studies, gastrointestinal bleeding is another potentially serious adverse effect with high-dosage ibuprofen.

Larger studies correlating specific Cmax and/or AUC values with toxicities are necessary in order to provide a better understanding of how to individualise each CF patient’s ibuprofen therapy while avoiding toxicity.

9. Food and Drug Interactions

The administration of ibuprofen with food decreases the Cmax by approximately 20% and prolongs the tmax (to 1.5–3 hours) in healthy volunteers.[11] The effect is greater with fatty food. Since CF patients are encouraged to consume fatty foods for maintenance of their nutritional state, this may result in clinically significant changes in Cmax and tmax.

Pancreatic enzymes are frequently prescribed to patients with CF to increase absorption of fat and improve nutritional status. There is concern, however, that concomitant administration of high-dosage ibuprofen and pancreatic enzymes may have an additive pharmacodynamic effect. High-dosage ibuprofen alone can induce intestinal ulcers, and the addition of high-dosage pancreatic enzymes has been shown to increase the severity of caecum and colonic ulcers in an animal model resembling early lesions of fibrosing colonopathy.[35] Currently, there is no evidence of an increased prevalence of fibrosing colonopathy from concomitant administration of high-dosage ibuprofen and pancreatic enzymes in humans, but caution should be taken and patients should be routinely monitored for signs and symptoms of gastric pain.

One of the manifestations of CF is GORD, and approximately 20% of CF patients experience heartburn and regurgitation.[36] Treatment and symptomatic relief can be provided with H2 antagonists. Some of these agents inhibit CYP-mediated microsomal drug oxidation, and cimetidine, nizatidine and ranitidine, which are commonly available over-the-counter or by prescription, were evaluated for their potential inhibition of ibuprofen metabolism. No significant interactions were noted, which is consistent with current understanding of the metabolic disposition of ibuprofen.[3739] Ibuprofen is a substrate of CYP2C9, and H2 antagonists are inhibitors of CYP2D6.

H2 antagonists, antacids and proton pump inhibitors can also alter absorption of drugs that require an acidic environment for dissolution (e.g. ketoconazole) by increasing gastrointestinal pH. No significant differences in pharmacokinetic parameters have been noted with coadministration of ibuprofen with these agents in healthy volunteers.[11] However, these agents have not been studied with high-dosage ibuprofen in CF children, and their exact effect in this population is unknown.

Due to the high protein binding exhibited by ibuprofen, competitive protein binding displacement interactions are possible. Since ibuprofen is not a high extraction ratio drug, displacement is unlikely to result in clinically significant differences in activity; however, it will reduce the total concentration in plasma.[28] Therefore, plasma concentrations obtained following initiation of another highly albumin-bound drug should be interpreted cautiously. If possible, determination of unbound concentrations may assist in interpreting the appropriateness of the dosage regimen.

10. Conclusion

The chronic nature of the infection in the lungs of CF patients results in an exaggerated inflammatory response that is the principal cause of decline in lung function. Ibuprofen is a potent anti-inflammatory agent that directly affects neutrophil action, the principal mediator of airway inflammation in patients with CF.

Pharmacodynamic data in vitro and in animal models demonstrate that ibuprofen is effective when administered at high dosages but may paradoxically increase inflammation when given at low dosages. Similar data in patients with CF are currently lacking. In particular, studies to date have been inadequately powered to evaluate the relationship between measures of drug exposure (Cmax or AUC) and efficacy/toxicity outcomes.

The pharmacokinetics of ibuprofen appear altered in patients with CF when compared with healthy controls. In particular, CF patients exhibit a reduced bioavailability. An increased CL/F and Varea/F were noted; however a lack of data for intravenous ibuprofen precludes definitive evidence of altered clearance and Vd in patients with CF. In addition, a high degree of variability in Cmax, tmax, Vd/F and CL/F have been noted in patients with CF. Some of this variability can be explained by differences in ibuprofen formulation administered. The suspension formulation appears to be more rapidly and completely absorbed when compared with the tablet formulations.

Therapeutic drug monitoring of ibuprofen in patients with CF is recommended due to the biphasic response to inflammatory mediators demonstrated in vitro as well as the high interpatient variability in pharmacokinetics. The difference in absorption characteristics between suspension and tablet formulations requires two different sampling schedules. Obtaining samples prior to the dose and 0.25–0.5, 1–1.5 and 5 hours after a tablet dose, or 0.25–0.5, 1 and 3–4 hours after a suspension dose, enables determination of the Cmax as well as other measures of drug exposure (e.g. AUC). MAP Bayesian analysis is a tool that has been utilised extensively to individualise the dosage of aminoglycosides and other drugs commonly monitored in the clinical setting. This method has recently been shown to provide accurate and precise estimates of the pharmacokinetic parameters of ibuprofen in children with CF. Use of MAP Bayesian analysis may also be a useful tool to further investigate the relationship between measures of ibuprofen exposure and efficacy/toxicity outcomes.

A number of new classes of anti-inflammatory agents are currently under various stages of clinical development, including neutrophil elastase inhibitors, leukotriene inhibitors (LBT4), and cytokine modulators; however, until these compounds are found to be more efficacious or exhibit improved safety profiles, high-dosage ibuprofen will remain an important therapy for combating airway inflammation in children with CF.

Acknowledgements

No sources of funding were used to assist in the preparation of this manuscript. The authors have no conflicts of interest that are directly relevant to the content of this manuscript.

Copyright information

© Adis Data Information BV 2004