The pharmacokinetics of a long-acting OROS hydromorphone formulation.

IMPORTANCE OF THE FIELD: New formulations of opiods can provide round-the-clock pain relief to improve pain management and quality of life for patients with chronic pain. AREAS COVERED IN THIS REVIEW: Information and comments on the pharmacokinetic processes associated with a new once-daily formulation of the potent opiod hydromorphone. WHAT THE READER WILL GAIN: This review presents an overview of data from several small pharmacokinetic studies to gain a better perspective on the pharmacokinetic properties of a new long-acting formulation of hydromorphone. TAKE HOME MESSAGE: The development of advanced oral formulation that deliver analgesic drugs over an extended period provides new solutions to improve pain management and quality of life for patients with chronic pain.

Pharmacokinetic profile of a 24-hour controlled-release OROS formulation of hydromorphone in the presence of alcohol.

OBJECTIVE: The purpose of this study was to investigate the pharmacokinetic properties of a novel, once-daily, controlled-release formulation of hydromorphone (OROS hydromorphone) in the presence of alcohol. RESEARCH DESIGN AND METHODS: In a single-centre, open-label, four-treatment, four-period, four-sequence, crossover study, two groups of 24 healthy subjects (fasted or fed) were randomised to receive four single doses of OROS hydromorphone 16 mg with solutions of either 0%, 4%, 20% or 40% alcohol, and with a naltrexone block. MAIN OUTCOME MEASURES: Plasma samples taken predose and at regular intervals up to 48 h after dosing were assayed for hydromorphone concentrations; a mixed-effect analysis of variance was done on log-transformed data. Bioequivalence was concluded if 90% confidence intervals of treatment mean ratios were between 80% and 125%. RESULTS: Plasma hydromorphone concentrations were slightly higher after dosing with all alcohol treatments in both the fasted and fed subject groups. Median T(max) values were between 12 and 16 h and ranges were similar for all treatments. C(max) values increased after alcohol compared with no alcohol, with the increase slightly lower in the fed state. The greatest mean increase in C(max) observed was 1.3-fold in the fasted state and 1.1-fold in the fed state. Confidence intervals were within 80-125% for AUC but were slightly higher for C(max). CONCLUSIONS: The pharmacokinetics of once-daily OROS hydromorphone were only minimally affected by alcohol, with no dose dumping of hydromorphone. The results indicate that the controlled-release properties of this formulation are maintained in the presence of alcohol.

Effect of OROS controlled-release delivery on the pharmacokinetics and pharmacodynamics of oxybutynin chloride.

UNLABELLED: Aims : Dry mouth is a common side-effect seen with immediate-release oxybutynin (IR-Oxy). Ditropan XL [(Oxy-XL), a controlled-release formulation of oxybutynin chloride, is a once-daily oral dosage form that incorporates the OROS technology. Dry mouth as the pharmacodynamic measure was compared between Oxy-XL and IR-Oxy administration. The steady state stereospecific pharmacokinetics were also established for the two formulations and the kinetic-dynamic relationship of oxybutynin was examined. METHODS: This was a randomized, repeated-dose, double-blind, two-treatment, two-period, crossover study. After a baseline assessment day, volunteers were randomly assigned to one of two treatment sequences and received 4 days of each treatment with a washout period of 7 days between treatments. The treatments were: 1) Oxy-XL 10 mg in the morning and placebo 8 h later, and 2) IR-Oxy 5 mg in the morning and again 8 h later. Volunteers assessed dry mouth severity subjectively using a 100 mm visual analogue scale, VAS (Baseline, treatment days 1 and 4) and objectively by collecting saliva (Baseline and treatment day 4) before dosing and every hour after the morning dose for approximately 16 h. Several blood samples were collected during each treatment, with frequent sampling on day 4 to analyse for plasma R- and S-oxybutynin and R- and S-desethyloxybutynin concentrations. RESULTS: Relatively constant plasma concentrations of oxybutynin and its metabolite were seen over 24 h following Oxy-XL administration with the degree of fluctuation being much lower (P = 0.001; 66% to 81% reduction for the various analytes) than IR-Oxy. Compared with IR-Oxy, Oxy-XL yielded higher (131% and 158% for the R- and S-isomer, respectively) oxybutynin and lower (62% and 78% for the R- and S-isomer, respectively) desethyloxybutynin bioavailability, suggesting reduced first-pass metabolism. Saliva output (area under the effect curve) was significantly higher [P = 0.001; 37% (95% confidence interval: 24, 51%)] with Oxy-XL than with IR-Oxy and, accordingly, dry mouth severity (VAS) integrated over the day was significantly lower with Oxy-XL. The decrease in saliva output and the consequent increase in dry mouth severity correlated with the metabolite R-desethyloxybutynin concentration, and no apparent relationship was observed with the R-oxybutynin concentration. This suggests that the metabolite may contribute to the dry mouth. Therefore, the reduction in metabolite exposure with Oxy-XL may be a possible explanation for the observed decrease in dry mouth severity with OXY-XL compared with IR-Oxy. CONCLUSIONS: Oxy-XL maintains relatively constant plasma drug and metabolite concentrations and minimizes first-pass metabolism of oxybutynin. The metabolite appears to contribute to dry mouth associated with oxybutynin, and following Oxy-XL metabolite exposure is reduced compared with IR-Oxy. Consequently less dry mouth was observed with Oxy-XL as compared with IR-Oxy.

A comparison of the effects on saliva output of oxybutynin chloride and tolterodine tartrate.

BACKGROUND: Oxybutynin chloride and tolterodine tartrate are anticholinergic agents used to suppress involuntary bladder contractions in urinary incontinence. They act by inhibiting binding of acetylcholine to the muscarinic receptors in the detrusor muscle of the bladder. The same types of muscarinic receptors are found in the salivary glands; thus anticholinergic agents may decrease saliva production and cause dry mouth, a commonly cited reason for discontinuation of therapy. OBJECTIVE: The primary objective of this study was to compare saliva output, which is an objective measure of dry mouth, in subjects taking immediate- or extended-release oxybutynin, tolterodine, or placebo. METHODS: This was a single-site, single-dose, randomized, double-blind, 4-treatment, 4-period crossover study. Subjects were randomly assigned to 1 of 4 treatment sequences that included extended-release oxybutynin 10 mg, tolterodine 2 mg, immediate-release oxybutynin 5 mg, and placebo. Saliva output was measured objectively before dosing with each treatment and at 0.5, 1, 2, 3, 4, 6, 8, 10, and 12 hours after dosing. RESULTS: Thirty-six healthy adult volunteers (22 women and 14 men) participated in the study. They ranged in age from 19 to 42 years (mean, 27 years). Thirty-one were white, 3 Asian, and 2 black. There were no significant differences in predose saliva output between the 4 study groups. With placebo, saliva output increased throughout the day. Saliva output was maintained at predose levels throughout the day with extended-release oxybutynin. Two hours after dosing with tolterodine and immediate-release oxybutynin, saliva output decreased nearly 0.5 g in specimens collected over 2 minutes. All 3 active treatments were associated with lower saliva output compared with placebo. Extended-release oxybutynin and tolterodine were similar with respect to area under the saliva concentration-time curve but were associated with significantly greater saliva output than was immediate-release oxybutynin (P < 0.01). There were no serious adverse events (AEs) in this study. AEs were similar between treatments, although the incidence of headache was higher in the active-treatment groups than with placebo. CONCLUSIONS: Objective assessment of saliva output in healthy adult volunteers indicated that extended-release oxybutynin and tolterodine had less impact on saliva output than did conventional immediate-release oxybutynin, suggesting that they may yield lower levels of dry mouth.

A semiparametric deconvolution model to establish in vivo-in vitro correlation applied to OROS oxybutynin.

In vitro-in vivo correlation (IVIVC) models may be used to predict in vivo drug concentration-time profiles given in vitro release characteristics of a drug. This prediction is accomplished by incorporating in vitro release characteristics as an input function (A(vitro)) to a pharmacokinetics model. This simple approach often results in biased predictions of observed in vivo drug concentrations, and it can result in rejecting IVIVC. To solve this problem we propose a population IVIVC model that incorporates the in vitro information and allows one to quantify possibly changed in vivo release characteristic. The model assumes linear kinetics and describes the in vivo release as a sum of A(vitro) and a nonparametric function (A(d), a spline) representing the difference in release due to in vivo conditions. The function A(vitro) and its variability enter the model as a prior distribution. The function A(d) is estimated together with its intersubject variability. The number of parameters associated with A(d) defines the model: no parameters indicates perfect IVIVC, a large number of parameters indicates poor IVIVC. The number of parameters is determined using statistical model selection criteria. We demonstrate the approach to solve the IVIVC problem of an oral extended release oxybutynin form (OROS), administered in three pharmacokinetic studies. These studies present a particular challenging case; that is, the relative bioavailability for the OROS administration is >100% compared with that of the immediate-release form. The result of our modeling shows that the apparent lack of IVIVC can be overcome: in vivo concentration can be predicted (within or across data sets) based on in vitro release rate together with a simple form of systematic deviation from the in vitro release.

Lack of effect of food on the pharmacokinetics of an extended-release oxybutynin formulation.

The effect of food on the pharmacokinetics of 15 mg oxybutynin XL was evaluated in a single-dose, randomized, crossover, open-label study in healthy volunteers. A validated, stereospecific, high-performance liquid chromatography assay was used to simultaneously determine the plasma concentrations of R- and S-oxybutynin and active metabolite R- and S-desethyloxybutynin. The mean AUC and Cmax values for each of the four analytes in the fed treatment were within +/- 20% of the fasting treatment values. The 90% confidence intervals for the treatment ratios (fed/fasted) for log-transformed Cmax and AUCinf values for the drug isomers and AUCinf values for the metabolite isomers were all within the 80% to 125% range. Only the ranges for the Cmax values for R- and S-desethyloxybutynin were slightly wider but were well within the 70% to 143% criteria recommended for Cmax when comparing effect of food. Lack of effect of food on oxybutynin XL is consistent with the previous observation that the osmotically controlled formulations are nearly insensitive to the gastrointestinal environment, including food. Oxybutynin XL was well tolerated, and the safety results were comparable whether administered alone or with food. In conclusion, oxybutynin XL administration does not require any caution to be exercised regarding food.

Effect of dosing time on the total intestinal transit time of non-disintegrating systems.

The total gastrointestinal transit time of nondisintegrating tablets may be affected by dosing time; available literature on this topic is inconclusive. OROS systems are nondisintegrating osmotically driven tablets that release drug over a period of time during their transit through the gastrointestinal tract and are excreted intact in the feces. Total transit times following morning administration of OROS systems pooled from various studies (n = 1,163 systems) showed a distribution with peak frequencies clustering around 24 and 48 h and following night administration (n = 80 systems) was found to cluster around 12 and 36 h. The total transit time distribution appears to be different following morning and night administration. However, on reanalyzing the data considering clock time when the tablet was collected rather than time post-administration, most of the difference between the distribution patterns disappeared. This suggested that total transit times after morning or night administration may be related to the bowel movement habits of the study population. Therefore, OROS systems total transit time were compared to the intrinsic bowel movement pattern of the general population reported in the literature and indeed a good correlation was seen between the two. The total transit time appears to be determined by two factors: the defecation frequency and the probability of its inclusion in the defecation event which is related to its location in the GI tract. A tablet is more likely to be excreted if it is further down in the GI tract. The total transit time data for OROS systems suggest that with the morning dosing the tablet is more likely to be excreted in the bowel movement the next morning. With the night time dosing the tablet may not be far enough in the colon to be excreted in the next morning bowel movement and therefore, it is more likely to be excreted the following morning.

Modeling of circadian testosterone in healthy men and hypogonadal men.

The testosterone circadian rhythm has been reported extensively in the literature and has been described by a cosine function. Typically, these data are measured at frequent and regular (e.g., hourly) intervals. However, modeling circadian rhythm with data collected sparsely at irregular intervals and/or data that are not collected at the same time in all individuals has not been reported. The population nonlinear mixed-effects approach can handle such data and also allows covariates to be incorporated into the model. Frequent hourly testosterone concentration data available in the literature for young and elderly healthy volunteers were analyzed first. In the elderly, blunted or completely absent circadian rhythm has been reported, but a full circadian model was significantly better than a model containing one or no circadian component. Therefore, data from both the elderly and young were modeled together, and age was included as a categorical variable (young or elderly). Consistent with literature, the rhythm-adjusted mean testosterone concentrations was lower, and the deviation from the mean, especially to the maximum daily value, was less than half in the elderly (7%) compared to young subjects (16%). The testosterone concentration data measured infrequently and at varying intervals in young normal men and hypogonadal men were evaluated next. Although not measured at regular frequency in each individual, the data were obtained at different clock times for different subjects. Since for population mixed-effects analysis, data from all subjects are pooled, there was enough information to profile the 24-hour circadian cycle. In healthy young subjects, the mean Cnadir, Cpeak, Tnadir, and Tpeak values estimated from the model were 420 ng/dL, 577 ng/dL, 21:42 hours, and 0600 hours, respectively, and were similar to the parameters obtained for the frequently sampled young subjects. In hypogonadal men (testosterone concentrations < 300 ng/dL), the mean testosterone concentrations were much lower than the healthy young or elderly men, and a straight-line model was the best descriptor (i.e., no circadian rhythm was detected). It was also shown that with the application of a transdermal testosterone system, the mean testosterone concentrations in the treated men were within the 95% confidence interval for healthy young men. The results presented here suggest that the advantages of the analysis approach--namely, handling of covariates and handling of sparse, infrequently collected data--can be used in characterizing testosterone circadian rhythm or the lack of it.

Reproducible fentanyl doses delivered intermittently at different time intervals from an electrotransport system.

The electrotransport transdermal fentanyl system (ET [fentanyl]), uses a small electrical current to enhance delivery of fentanyl to systemic circulation. Intermittent doses can be administered by periodic application of the current. The purpose of this study was to compare the effects of the frequency of intermittent drug delivery by ET (fentanyl) and compare the drug delivery to systemic circulation by ET (fentanyl) with intravenous administration. The topical safety was also determined for the ET (fentanyl) system. Nine adult male volunteers completed this three-treatment, randomized, 24-h, crossover study. ET (fentanyl) treatments with 200 microA direct current applied for 30 min at frequent (hourly) or infrequent (4-hourly) intervals over a 24-h period were compared. Also, the drug delivery to systemic circulation from ET (fentanyl) was compared with intravenous fentanyl 75 microg infused over 30 min every 4 h over a 24-hour period. The mean serum fentanyl concentration achieved with the hourly ET (fentanyl) regimen was higher than that for the 4-hourly ET (fentanyl) regimen as expected from the higher frequency of drug doses. The amount of fentanyl delivered estimated per dose from the ET (fentanyl) system using the iv fentanyl treatment as the reference was similar for the two ET regimens throughout the dosing period. This indicates consistent drug delivery regardless of the frequency of ET dosing. The majority of subjects reported either no, or barely perceptible, erythema 24 h after removal of the system.

Quantitative characterization of therapeutic index: application of mixed-effects modeling to evaluate oxybutynin dose-efficacy and dose-side effect relationships.

BACKGROUND: Describing a therapeutic index for a drug is important for evaluating safe and effective dosage regimens. Therapeutic index can be evaluated as the relative position of the dose-efficacy and the dose-side effect curves. Oxybutynin XL (Ditropan XL), a once-daily oral controlled-release formulation for oxybutynin chloride, is being developed. Oxybutynin XL efficacy and side-effect data obtained from two parallel-group, randomized, controlled clinical trials were modeled to evaluate the therapeutic index. METHODS: A nonlinear mixed-effects model was used to characterize the oxybutynin dose-efficacy and dose-dry mouth relationship. Weekly urge urinary incontinence episodes, the primary efficacy variable, is a discrete variable (counts) with only non-negative integer values and was therefore modeled as a Poisson variable. The probability of dry mouth severity (the most frequently reported side effect), assessed on a categorical four-point scale, was modeled with a proportional odds model. In the modeling process, it was assumed that the time effect was the same for the active and placebo treatments and that the drug effect was additive. RESULTS AND CONCLUSIONS: The urge urinary incontinence episodes declined log-linearly, and no significant difference was observed between the two formulations. However, there was a trend toward higher efficacy with oxybutynin XL than with immediate-release oxybutynin at the same dose in one study. Dose-dry mouth analysis showed that the probability of dry mouth with an increasing dose was significantly lower with oxybutynin XL than with immediate-release oxybutynin in the second study, and a similar trend was observed in the first study. By combining the dose-urge urinary incontinence and dose-dry mouth relationship, a wider therapeutic index was predicted for oxybutynin XL than for immediate-release oxybutynin.

Pharmacokinetics of an oral once-a-day controlled-release oxybutynin formulation compared with immediate-release oxybutynin.

Oxybutynin is used for the treatment of urge urinary incontinence. In this randomized, open-label, two-way crossover, multiple-dose study, the pharmacokinetics of a once-daily, controlled-release formulation, OROS oxybutynin chloride, was compared with that of immediate-release (IR) oxybutynin (Ditropan). Thirteen healthy female volunteers received three 5 mg OROS oxybutynin chloride tablets once daily for 4 days or IR oxybutynin 5 mg administered every 8 hours for 4 days. On day 1, with OROS oxybutynin chloride, mean plasma concentrations rose slowly over approximately 6 hours following dosing (mean Cmax 4.2 ng/mL) and remained fairly constant over the 24-hour dosing interval, whereas with IR oxybutynin, mean plasma concentrations rose rapidly within the first hour after dosing (mean Cmax 12.0 ng/mL), then declined. The mean oxybutynin degree of fluctuation was much lower for OROS oxybutynin chloride (78%) than for IR oxybutynin (371%). For both formulations, the plasma concentration-time profiles for the metabolite N-desethyloxybutynin paralleled those of oxybutynin but at higher concentrations. Steady-state oxybutynin concentrations were achieved by day 3 for both formulations. Mean area under the concentration-time curve (AUC) values for both oxybutynin and its metabolite were similar between day 1 and day 4 for each treatment, suggesting time-invariant pharmacokinetics. With OROS oxybutynin chloride, mean relative bioavailability was higher (153%) for oxybutynin and lower (69%) for N-desethyloxybutynin compared with IR oxybutynin. This increased bioavailability may be due to reduced first-pass metabolism; within 3 to 5 hours after dosing, OROS systems are thought to reach the colon, where cytochrome P450-mediated oxidation (oxybutynin's primary metabolic pathway) may be less extensive than in the small intestine. Fewer subjects reported any adverse event with OROS oxybutynin chloride than with IR oxybutynin (including dry mouth, oxybutynin's most frequently reported anticholinergic adverse effect).

Effects of nandrolone decanoate therapy in borderline hypogonadal men with HIV-associated weight loss.

Serum testosterone concentrations are frequently in the low-normal range (lowest quartile, <500 ng/dl) in men with AIDS-wasting syndrome (AWS) and in other chronic wasting disorders. The response of patients in this group to androgen treatment has not been determined, however. Eighteen men with AWS (mean +/- standard error [SE]: 87% +/- 1% usual body weight; CD4 count 90 +/- 24) and borderline low serum testosterone concentrations (382 +/- 33 ng/dl) completed a 21-day placebo-controlled inpatient metabolic ward study comparing intramuscular (i.m.) placebo (n = 7) with low-dose (65 mg/week; n = 4) and high-dose (200 mg/week; n = 7) nandrolone decanoate, a testosterone analogue. Nitrogen balance, stable isotope-mass spectrometric measurement of de novo lipogenesis (DNL), resting energy expenditure, and gonadal hormone levels were measured. Both low-dose and high-dose nandrolone resulted in significant nitrogen retention (33-52 g nitrogen/14 days, representing gains of 0.5 to 0.9 kg lean tissue/week) compared with placebo (loss of 11 g nitrogen/week). This was reflected biochemically in a borderline significant reduction of high DNL (p < .06). Serum testosterone and gonadotropins were suppressed whereas resting energy expenditure was unchanged by nandrolone treatment. In 10 study subjects completing a 12-week open-label follow-up phase, body weight increased by 4.9 +/- 1.2 kg, including 3.1 +/- 0.5 kg lean body mass, and treadmill exercise performance also improved. In summary, nandrolone decanoate therapy in the absence of an exercise program in borderline hypogonadal men with AWS caused substantial nitrogen retention compared with placebo, similar in extent to the nitrogen retention previously achieved with recombinant growth hormone. It is reasonable to expand the criteria for androgen treatment in AWS to include at least patients in the lowest quartile of serum testosterone.

Fentanyl delivery from an electrotransport system: delivery is a function of total current, not duration of current.

This open-label, parallel study of 28 men was conducted to evaluate the pharmacokinetics and safety of fentanyl delivered by the E-TRANS (fentanyl) electrotransport transdermal system (ALZA Corporation, Palo Alto, CA). The E-TRANS (fentanyl) system provided electrically assisted, transdermal, continuous delivery of fentanyl. Treatments consisted of no current (group A); a constant current of 100 microA for 26 hours plus 4 additional doses at varying currents for varying times during hour 25 (groups B, C, D); a constant current of 100 microA for 26 hours plus 4 additional doses at 1,200 microA over 2.5 minutes during hour 1 (group E); or 500 microA for 0.5 hours and 100 microA for 3.5 hours (group F). No fentanyl was detected in serum when no current had been applied. Mean serum fentanyl concentrations were similar regardless of current duration during hour 25 (treatments B, C, D). Increases in mean serum fentanyl concentrations were significantly lower during additional dosing for treatment E compared with treatments B, C, and D. Serum fentanyl concentrations sufficient for analgesia (1-3 ng/mL) were attained in treatments using the E-TRANS (fentanyl) system with basal current of 100 microA for 26 hours. There were no safety issues after treatment with E-TRANS (fentanyl) system with concurrent opioid antagonist (naltrexone) administration. The only adverse event requiring treatment was a headache (n = 1). The majority of subjects had no or barely perceptible erythema at the application site 24 hours after system removal. Application of E-TRANS (fentanyl) resulted in therapeutically significant serum fentanyl concentrations over a range of applied currents. Overall serum fentanyl concentrations were higher when the skin had been primed by constant-current fentanyl delivery.

Effect of current density on pharmacokinetics following continuous or intermittent input from a fentanyl electrotransport system.

The pharmacokinetics of fentanyl were determined in two open-label crossover studies following 24-h periods of delivery by an electrotransport transdermal system (E-TRANS [fentanyl] system) in young healthy male volunteers. A direct current was applied continuously in study 1 (at 50, 100, and 200 microA; surface area = 5 cm2; n = 8), but in study 2 it was limited to the first 20 min of each hour (at 150, 200, and 250 microA; surface area = 2 cm2; n = 12). The opioid effects of fentanyl were blocked with naltrexone administered every 12 h. With increasing electrical current, the increase in serum fentanyl concentration, amount absorbed, and AUC values were proportional in study 2 but not in study 1. It is hypothesized that the lack of proportionality in study 1 is due to lower current density (microA/cm2) in this study. It appears that for fentanyl, the current density should be about 75 microA/cm2 or greater for a linear relation between current and amount absorbed as seen in study 2. Compared with intravenously infused fentanyl, the serum concentrations resulting from E-TRANS (fentanyl) system application revealed a slightly dampened rate of increase (stratum-corneum barrier effect) and decrease in serum concentrations, and a similar intersubject variability in fentanyl AUC values. Fentanyl pharmacokinetics with either E-TRANS (fentanyl) or intravenous infusion were time-invariant over a 24-h application period, with similar mean half-life values (about 15-18 h). E-TRANS (fentanyl) administration (either continuous or intermittent input) was safe and well tolerated. Adverse effects were mild to moderate; they consisted mainly of local erythema and pruritus (which resolved in most patients within 24 h after system removal) and occasional opioid effects.

Insulin drug delivery systems: rectal gels.

In continuation of a previous study on insulin gels administered deep rectally by means of an applicator, 16 further rectal formulations were evaluated, using the nondiabetic White New Zealand rabbit as animal model. The highest pharmacologic availability, P.A., defined as the ratio of the areas under the % glucose reduction-time curves rectally versus I.V., corrected for dose size and body weight, in the previous experiments was about 29%. In the present study the highest mean P.A. was 42% and 31%, both obtained with Gelucire bases (50/22 and 44/14), pH 8 buffer solution and sodium deoxycholate, using Cab-O-Sil as stiffening agent. Simple gels prepared from gelatine and polyethylene glycol, or Veegum, resulted in mean P.A. of about 23 and 24%, respectively. Addition of Gelucire or sodium deoxycholate did not result in a further increase of P.A. of the latter two gels.

In vitro evaluation of a sustained-release veterinary peroral pellet preparation.

In a preceding in vivo study in horses, wide interindividual variation was found in the extent of bioavailability and time to reach peak concentration after peroral administration of one specific theophylline sustained-release dosage form. The purpose of the present study was to investigate the factors of potency, the pH of dissolution medium, the enzymes in the dissolution medium, and the crushing of the pellets on in vitro performance. The results show a wide variation in potency for the individual units, an increase in release rate with increasing pH, and an increase in release rate if the pellets are crushed. The wide variation in potency explains the variation found in absolute bioavailability, and the increase in release rate when the pellets are crushed explains the differences seen in peak plasma times, since the pellets will be chewed to varying degrees by the horse.