effects of diet on pharmacokinetics of phenobarbital in healthy dogs

Epilepsy constitutes 3 to 5% of all diseases seen indogs.1,2 Because a specific cause for epilepsy is often

not identified (idiopathic epilepsy), treatment is direct-ed at controlling recurrent seizures. Phenobarbital isconsidered the drug of choice for the treatment ofepilepsy because of its efficacy, convenient dosing reg-imen, low cost, and relative safety.1 It is routinelyadministered daily for long periods, often for the life ofthe dog.

Pharmacokinetics of phenobarbital may be diffi-cult to predict in individual dogs because of the induc-tion by phenobarbital of the enzymes responsible forits own hepatic metabolism.3-5 Although serum con-centrations of phenobarbital correlate with therapeuticbenefit in some studies, phenobarbital dosage oftendoes not.1,6 Therefore, the dosage of phenobarbitaladministered is often not predictive of serum concen-trations. Variability and unpredictability of serum phe-nobarbital concentrations can have serious clinicalconsequences such as increased frequency or severityof seizures when serum concentrations decrease tosubtherapeutic values. Loss of seizure control can leadto owner frustration and euthanasia of the dog or aug-mentation of the dose to toxic values in an attempt toregain control. Chronic adverse effects of high serumphenobarbital concentrations include increased serumactivities of hepatic enzymes in 50 to 80% of cases,1,7

altered response to adrenocortical function testing in100% of cases studied,7 decreased serum concentra-tions of total and free thyroxine,8 and overt hepatotox-icosis in up to 15% of cases.1,9

Many studies have demonstrated the effects ofdietary manipulation on hepatic metabolism, drug dis-position, and hepatic microsomal enzyme activity inhumans and rodents.10-17 Likewise, it is well-knownthat body composition and metabolic rate influencedrug distribution and metabolism.18,19 Although it isrecognized that many factors may affect the pharmaco-kinetics of phenobarbital,4,5,20 to our knowledge, dietaryinteractions (apart from the effects of feeding and fast-ing21) have not been examined in dogs. Diet formula-tion, body composition, and metabolic rate may haveconsiderable effects on phenobarbital distribution,metabolism, and, subsequently, clinical efficacy andtoxicosis in dogs with epilepsy. The purposes of thestudy reported here were to ascertain the effects of var-ious diets on the pharmacokinetics of phenobarbital inhealthy dogs and to determine potential interactiveeffects of changes in body composition and metabolicrate induced by various diets.

Materials and MethodsDogs—Twenty-seven clinically normal sexually intact

adult female Beagles weighing between 7.7 and 11.6 kg (3.5

JAVMA, Vol 217, No. 6, September 15, 2000 Scientific Reports: Original Study 847

SMALLANIMALS/

AVIAN

Effects of diet on pharmacokinetics of phenobarbital in healthy dogs

Peter J. Maguire, DVM, DACVIM; Martin J. Fettman, DVM, PhD, DACVP; Mary O. Smith, BVMS, PhD, DACVIM; Deborah S. Greco, DVM, PhD, DACVIM; A. Simon Turner, BVSc, MS, DACVS; Judy A. Walton, BS;

Gregory K. Ogilvie, DVM, DACVIM

Objective—To determine effects of various diets onthe pharmacokinetics of phenobarbital and the inter-active effects of changes in body composition andmetabolic rate.Design—Prospective study.Animals—27 healthy sexually intact adult femaleBeagles.Procedure—Pharmacokinetic studies of phenobarbi-tal were performed before and 2 months after dogswere fed 1 of 3 diets (group 1, maintenance diet;group 2, protein-restricted diet; group 3, fat- and pro-tein-restricted diet) and treated with phenobarbital(approx 3 mg/kg [1.4 mg/lb] of body weight, PO, q 12h). Pharmacokinetic studies involved administeringphenobarbital (15 mg/kg [6.8 mg/lb], IV) and collectingblood samples at specific intervals for 240 hours.Effects of diet and time were determined by repeat-ed-measures ANOVA.Results—Volume of distribution, mean residencetime, and half-life (t1/2) of phenobarbital significantlydecreased, whereas clearance rate and eliminationrate significantly increased with time in all groups.Dietary protein or fat restriction induced significantlygreater changes: t1/2 (hours) was lower in groups 2(mean ± SD; 25.9 ± 6.10 hours) and 3 (24.0 ± 4.70)than in group 1 (32.9 ± 5.20). Phenobarbital clearancerate (ml/kg/min) was significantly higher in group 3(0.22 ± 0.05 ml/kg/min) than in groups 1 (0.17 ± 0.03)or 2 (0.18 ± 0.03). Induction of serum alkaline phos-phatase activity (U/L) was greater in groups 2 (192.4± 47.5 U/L) and 3 (202.0 ± 98.2) than in group 1 (125.0± 47.5).Conclusions and Clinical Relevance—Clinicallyimportant differences between diet groups wereobserved regarding pharmacokinetics of phenobarbi-tal, changes in CBC and serum biochemical variables,and body composition. Drug dosage must be reevalu-ated if a dog’s diet, body weight, or body compositionchanges during treatment. Changes in blood variablesthat may indicate liver toxicosis caused by phenobar-bital may be amplified by diet-drug interactions. (J AmVet Med Assoc 2000;217:847–852)

From the Department of Clinical Sciences, College of VeterinaryMedicine and Biomedical Sciences, Colorado State University, FortCollins, CO 80523. Dr. Maguire’s present address is the AnimalCare Center of Sonoma County, 6620 Redwood Dr, Rohnert Park,CA 94928.

Funds provided by the Morris Animal Foundation, the BadisheAnilin and Soda-Fabrik (BASF) Corporation, and Hill’s PetNutrition Inc.

Address correspondence to Dr. Fettman.

847_852.QXD 8/23/2005 1:55 PM Page 847

to 5.3 lb) were acquired from a commercial suppliera andwere housed in groups of 3 at the Colorado State UniversityVeterinary Teaching Hospital. Dogs were acclimated for 2weeks during which they were fed a commercially availablemaintenance dry food.b Water was provided ad libitum.During this period and throughout the subsequent studyperiod, body weights were recorded weekly. Food consump-tion was estimated weekly per pen of 3 dogs by weighing fullfeeders at the start of each week and then reweighing eachfeeder prior to being refilled at the end of the week. The dif-ference represented the weight in kilograms of food con-sumed per week in each pen of 3 dogs. Dogs were deter-mined to be healthy by physical examination, results of CBC,analyses of serum concentrations of glucose, urea nitrogen,creatinine, phosphorus, calcium, total protein, albumin,globulin, cholesterol, total bilirubin, sodium, potassium,chloride, and analyses of serum activities of alanine transam-inase, aspartate transaminase, alkaline phosphatase (ALP),and γ-glutamyltransferase.

Study design—Baseline pharmacokinetics studies ofphenobarbital were performed after the 2-week acclimationperiod. Following a 24-hour period in which food was with-held, dogs were given a single bolus of phenobarbital sodiumc

(130 mg/ml, IV). Blood samples were collected by jugular orcephalic venipuncture at 0, 0.25, 0.50, 0.75, 1, 2, 4, 6, 12, 24,48, 96, 144, 192, and 240 hours after phenobarbital adminis-tration. Blood samples obtained at 24 hours and beyond werecollected after a 24-hour period in which food was withheld.Serum for phenobarbital analysis was obtained from bloodcollected into sterile collection tubesd and allowed to clot atroom temperature; the whole blood was centrifuged, andserum was removed and stored at –20 C until it was assayedfor phenobarbital concentration.

Dogs were randomly assigned to 1 of 3 dietary treatmentgroups and fed 1 of 3 commercially available diets for 2months: maintenanceb (diet 1/group 1; [n = 9]); low proteine

(diet 2/group 2; [9]); or low fat and low proteinf (diet 3/group3; [9]; Table 1). Dogs in each pen (n = 3) were fed ad libitumfrom a group feeder for 1 hour daily each morning. Duringthe 2-month treatment period, phenobarbitalg treatment wasinitiated in each dog at a dose of a single 30 mg tablet (PO, q12 h). As a result, groups 1, 2, and 3 received phenobarbitalorally (q 12 h) in dosages (mean ± SD) of 3.4 ± 0.4 mg/kg(1.5 ± 0.2 mg/lb) of body weight, 3.0 ± 0.3 mg/kg (1.4 ± 0.1mg/lb), and 2.9 ± 0.3 mg/kg (1.3 ± 0.1 mg/lb), respectively.

The 2-month treatment period was also used to train thedogs to lie quietly while they adapted to a plastic facemask tobe used during the indirect calorimetry study. After 2months, CBC and serum biochemical analyses were repeated,and serum phenobarbital concentrations were obtainedapproximately 12 hours following the last phenobarbitaltreatment. The pharmacokinetic study was then repeated inthe same manner, as described (IV bolus injection of 130mg/ml phenobarbital). Phenobarbital was not administeredorally the morning the second pharmacokinetic study wasinitiated or thereafter. Indirect calorimetry and determina-

tion of body composition by dual energy x-ray absorptiom-etry (DEXA) was then performed in each dog.

Serum phenobarbital concentrations were determinedby use of competitive binding fluorescence polarization.h Thelowest measurable concentration by this assay is 1.1 µg/ml,and cross-reactivity with other barbiturates is < 2.3%, where-as cross-reactivity with other drug classes (including benzo-diazepines) is < 1.0%. Coefficients of variation for within-runand between-run target values ≤ 50 mg/ml are 1.64 and1.34%, respectively.

Serum phenobarbital concentrations after IV injectionof phenobarbital were fitted for each dog, using nonlinearleast squares regression to model equations descriptive of asingle-compartment open pharmacokinetic model, by use ofa general pharmacokinetic equation:

Cp = Σ [Ciexp (–λit)]

where Cp is the plasma concentration of phenobarbital, Ci isthe intercept, λi is the slope of each of n exponential terms,exp is the inverse of the natural logarithm, and t is time.Initial values were determined by use of a commerciallyavailable computer software program.i Volume of distribu-tion at steady state (Vdss) was calculated using the equation:

Vdss (L/kg) = dose (mg/kg) X AUMC (min2)/AUC2 (mg/min)

where dose represents amount of phenobarbital administeredIV, AUMC is the area under the moment curve, and AUC isthe area under the curve. Total body clearance (Cl) was cal-culated using the equation:

Cl (ml/min/kg) = Ke (min-1) X Vdss (L/kg)

where Ke is the elimination rate constant. Pharmacokineticmodels were considered appropriate for the data on the basisof the following goodness-of-fit criteria: SE of estimates (val-ues must have been within 25% of the actual value for theestimate), model selection criterion, residual trend analysis,eigenvalues, and correlation coefficients. Pharmacokineticsvariables were calculated separately for each dog, and mean± SD values were calculated for each dietary group in the pre-and posttreatment periods.

Indirect calorimetry studies in the dogs were performedafter a 24-hour fasting period. Energy expenditure was esti-mated by use of an open-flow indirect calorimetry system todetermine oxygen consumption (VO2) and carbon dioxideproduction (VCO2), as described.22,23 Dogs were introduced tothe room and given at least 15 minutes to adapt to the sur-roundings and the plastic facemask collection system. Dogswere positioned in sternal or lateral recumbency on carpetingand acclimated to a thermoneutral quiet environment. After5 to 10 minutes for the system to reach steady state, data wascollected for an additional 15 minutes for calculation ofmean VO2 and VCO2. Resting energy expenditure (REE) wasestimated using the modified Weir formula:

REE (kcal/kg0.75/24 h) = 1.44 (3.9[VO2] + 1.1[VCO2]).

For DEXA analyses, each dog was sedated with 1.0 mg

848 Scientific Reports: Original Study JAVMA, Vol 217, No. 6, September 15, 2000

SMALLANIMALS/

AVIAN

Diet 1b Diet 2e Diet 3f

Nutrient g/100g DM* g/100 kcal g/100g DM* g/100 kcal g/100g DM* g/100 kcal

Protein 21.8 4.8 14.6 3.4 16.7 5.2Fat 21.1 4.6 19.0 4.5 8.80 2.7Carbohydrate 50.4 11.1 60.9 14.3 53.2 16.5Crude fiber 2.50 0.5 1.40 0.3 16.8 5.2

Energy density (kcal/kg) 4195 3938 2927

*Dry Matter basis.

Table 1—Nutrient content of 3 diets fed to healthy adult female Beagles

847_852.QXD 8/23/2005 1:55 PM Page 848

of butorphanolj (10 mg/ml), 0.5 mg of acepromazinek (10mg/ml), and 0.4 mg of atropine sulfatel by subcutaneousinjection. A catheterm was placed in the cephalic vein, anes-thesia was induced with propofoln (5 mg/kg, IV, titrated toeffect), and anesthesia was maintained with isoflurane (1 to2%). Dogs were positioned in sternal recumbency with hindlimbs extended caudally, and DEXA measurements were per-formed by use of a whole body scannero operated in singlebeam mode, as described.24-26 Calibration of the instrumentwas verified by scanning a calibration phantom.Commercially available softwareo was used to analyze scans.Scans took approximately 10 minutes and were performed induplicate if any gross movement was detected. These datawere obtained: mean coefficients of variation for DEXA scan-ning in quadruplicate measurements from each of 2 dogs,total bone mineral density, total bone mineral content, bodyfat content (g), and lean-tissue content (g).

Statistical analyses—Pharmacokinetics data (such asAUC, mean residence time [MRT], Ke, Cl, Vdss, and half-life[t1/2]) for phenobarbital, values derived from CBC and serumbiochemical analyses as well as indirect calorimetry andDEXA data (such as REE and percentage of body fat) weresubject to statistical analysis. A modified Kolmogorov-Smirnov test was used to assess normality of distribution ofdata, and a Bartlett test of homogeneity was used to assessequality of variances among groups to determine the distrib-ution characteristics and whether the actual data could beused for parametric analysis. All parameters were normallydistributed and were compared between groups by 2-wayANOVA for animal and treatment effects with repeated-mea-sures for time effects. The Fisher least significance differencetest was used to identify individual group and time differ-ences. Values of P < 0.05 were considered significant.

ResultsFeed intake was 26.7 ± 4.3, 25.9 ± 2.1, and 32.9 ±

4.4 g/kg/24 h in groups 1, 2, and 3, respectively, and

was significantly higher in dogs fed diet 3 than in thosefed the other 2 diets. Caloric intake was 196.6 ± 31.5,180.8 ± 13.9, and 174.4 ± 22.8 kcal/kg0.75/24 h ingroups 1, 2, and 3, respectively, and was not signifi-cantly different among the 3 groups of dogs.

Mean (± SD) values for all CBC, serum biochemi-cal analyses, and pharmacokinetic variables prior tophenobarbital treatment (when dogs were consumingthe maintenance diet) were not significantly differentbetween dogs. Following the 2-month treatment peri-od, significant differences were detected in a number ofCBC, serum biochemical, and pharmacokinetic values.These differences were observed within dietary treat-ment groups as a function of time (comparing valuesobtained before and after treatment) and betweendietary treatment groups in the posttreatment period.

Group 3 dogs had significant decreases betweenthe pre- and posttreatment RBC count, PCV, and hemo-globin concentration. Significant differences weredetected in PCV and RBC counts between groups 1 and3 and groups 2 and 3, and in hemoglobin valuesbetween groups 1 and 3 after treatment with pheno-barbital (Table 2).

Cholesterol concentrations in group 2 increased sig-nificantly in the posttreatment period. Posttreatmentcholesterol concentrations were significantly differentbetween groups 1 and 2, and groups 2 and 3. Alkalinephosphatase activities increased significantly with timein all groups and were significantly different betweengroups 1 and 2, and groups 1 and 3 in the posttreatmentperiod. A significant decrease in serum albumin concen-tration was observed in all groups over time, and signifi-cant differences were detected in the posttreatment peri-od between groups 1 and 3 and groups 2 and 3 (Table 2).


SMALLANIMALS/

AVIAN

ReferenceDiet 1b Diet 2e Diet 3f

Variable range Pre Post Pre Post Pre Post

PCV (%) 37�55 50.6 � 4.5* 49.1 � 3.3* 49.2 � 3.7* 49.6 � 3.0* 51.1 � 3.4* 46.1 � 3.5†RBC (�106/�l) 5.5�8.5 7.44 � 0.65* 7.32 � 0.46* 7.25 � 0.42* 7.39 � 0.39* 7.33 � 0.50* 6.71 � 0.59†Hgb (g/dl) 12�18 17.6 � 1.5* 17.6 � 1.0* 17.0 � 1.33* 17.2 � 1.0* 17.9 � 1.3* 16.2 � 1.6†Cholesterol (mg/dl) 130�300 203.0 � 35.5* 232.6 � 40.5* 196.4 � 25.7* 316.2 � 69.1† 182.7 � 32.9* 189.4 � 67.8*ALP (IU/L) 018�141 67.3 � 28.5* 125.0 � 47.5† 49.7 � 10.0* 192.4 � 48.4‡ 54.4 � 12.4* 202.0 � 98.2‡Albumin (g/dl) 2.7�4.5 3.78 � 0.24* 3.44 � 0.19† 3.73 � 0.24* 3.37 � 0.21† 3.79 � 0.16* 03.07 � 0.33‡

*,†,‡Values in the same row with different symbols were significantly (P � 0.05) different. Values are given as mean � SD. Hgb � Hemoglobin. ALP � Alkaline phosphatase.

Table 2—Effects of feeding 1 of 3 diets for 2 months on hematologic and serum biochemical factors in healthy adult Beagles before(pre) and 2 months after (post) receiving daily oral treatment with phenobarbital

Diet 1b Diet 2e Diet 3f

Variable Pre Post Pre Post Pre Post

t1/2 (h) 48.8 � 5.9* 32.9 � 5.2† 47.3 � 4.0* 25.9 � 6.1‡ 47.1 � 6.3* 24.0 � 4.7‡MRT (h) 70.5 � 8.6* 47.5 � 7.5† 68.2 � 5.7* 37.4 � 8.8‡ 68.0 � 9.1* 34.6 � 6.8‡AUC 1.55 � 0.26* 1.49 � 0.30* 1.73 � 0.65* 1.45 � 0.22* 1.96 � 0.29* 1.19 � 0.24†

(�g�h/ml � 103)Vd (L/kg) 0.59 � 0.09* 0.48 � 0.09† 0.53 � 0.07* 0.39 � 0.10‡ 0.53 � 0.10* 0.44 � 0.03†‡Cl (ml/kg/min) 0.15 � 0.02* 0.17 � 0.03† 0.13 � 0.02* 0.18 � 0.03†‡ 0.13 � 0.02* 0.22 � 0.05‡Ke (min�1

� 10�4) 2.39 � 0.30* 3.58 � 0.55† 2.30 � 0.44* 4.67 � 0.11‡ 2.49 � 0.35* 4.97 � 0.93‡

t1/2 � Half life; MRT � Mean residence time; AUC � Area under the curve; Vd � Volume of distribution; Cl � Clearance;Ke � Elimination rate constant.

See Table 2 for remainder of key.

Table 3—Effects of feeding 1 of 3 diets for 2 months on pharmacokinetic variables in adult femaleBeagles, as determined after intravenous injection of phenobarbital (15 mg/kg of body weight) before(pre) and 2 months after (post) daily oral treatment with phenobarbital

847_852.QXD 8/23/2005 1:55 PM Page 849

Mean (± SD) fasting serum phenobarbital concen-trations following oral administration of phenobarbitalfor 2 months were 19.2 ± 4.5, 18.2 ± 3.0, and 13.0 ±5.0 µg/ml for groups 1, 2, and 3, respectively. The meanvalue for group 3 was significantly lower than that forgroups 1 and 2.

Mean values for Vdss, MRT, and t1/2 decreased sig-nificantly with time in all groups (Table 3). Percentagedecreases for Vdss were 17.3, 25.9, and 17.2% forgroups 1, 2, and 3, respectively. The t1/2 valuesdecreased by 32.6, 45.2, and 49.0% for groups 1, 2, and3, respectively. The MRT values decreased by 32.6,45.2, and 49.1% in groups 1, 2, and 3, respectively.Total body clearance and Ke increased significantlywith time in each group; Cl increased by 15.7, 25.6,and 40.3% in groups 1, 2, and 3, respectively, whereasKe increased by 33.2, 50.6, and 49.8% in groups 1, 2,and 3, respectively.

Coefficient of variation for DEXA analyses was3.61%. Although significant differences were notdetected between groups in mean body weight at thetime of DEXA analysis, percentage of body fat in group3 was significantly lower than groups 1 and 2 (Table4). Values for VO2, VCO2, and REE were not signifi-cantly different between groups at the end of the study.

DiscussionEpilepsy is an important disease affecting dogs, in

which treatment is directed at controlling recurrentseizures. Phenobarbital has been the drug most com-monly chosen for management of seizures. One of theproblems encountered with phenobarbital use overtime is that the amount of drug administered is oftennot predictive of serum concentration, which can leadto unexpectedly low serum phenobarbital concentra-tions. This may result in loss of seizure control or highserum phenobarbital concentrations that may con-tribute to toxicosis or biochemical alterations that con-found the practitioner’s ability to monitor hepatic func-tion.7

All dogs in our study had significant changes int1/2, MRT, Ke, and Cl between the pre- and posttreat-ment periods during which phenobarbital was admin-istered orally. These changes were likely an effect ofinduction of hepatic biotransformation enzyme activi-ties by phenobarbital, as has been reported.3-5 Thiseffect appeared to be exacerbated in groups 2 and 3, asevidenced by the greater magnitude of changes in t1/2,MRT, and Ke values, and may be attributed to dietary

differences or diet-induced differences in body compo-sition. Studies in rodents and humans demonstratethat manipulation of dietary nutrients such as proteinand fat considerably alter hepatic microsomal enzymeactivity11,12,14 as well as the disposition of drugs metabo-lized hepatically.10,13,15 Although there is some variabili-ty in the specific effects reported, high-protein dietsappear to enhance rates of cytochrome P450-catalyzedoxidative reactions in rodents and humans,13,14,27 where-as protein-deficient diets cause cytochrome P450-asso-ciated enzyme activities to decrease.11,12,28 Dietary lipidsmodulate activity of cytochrome P450 enzymes inrats16,17 and are reported to be essential for the optimalinduction of cytochrome P450 enzymes by phenobar-bital.29,30 However, the changes we observed suggestedthat in dogs, the low protein diet, as well as the low fatand low protein diet, may have increased hepaticenzyme induction and subsequent metabolism of thedrug relative to the effects of the maintenance diet.

Dogs in our study that were fed the low-fat andlow-protein diet consumed significantly more food of alower energy density, which resulted in approximatelythe same caloric intake as dogs fed the maintenancediet or low-protein diet. Although our dogs were nearor below ideal body weight, dogs fed the low-fat andlow-protein diet appeared to lose body fat, comparedwith those fed the other diets. This change in bodycomposition may have contributed to the differenceswe observed in pharmacokinetic variables. Changes inbody composition caused by obesity and its associatedmetabolic effects or restricted food intake duringweight loss can have considerable effects on drugmetabolism in animals and humans.31-37 Changes in Cland AUC were also most notable in group 3 dogs, andthese changes were significantly different from thosefor groups 1 and 2. Increased clearance of phenobarbi-tal in the dogs fed the low-fat and low-protein diet alsomay reflect changes in the activity of the hepatic micro-somal enzymes. Another hypothetical explanationmight be that the significantly decreased body fat andlower serum albumin concentrations in group 3 rela-tive to groups 1 and 2 resulted in a decreased Vdss,thereby accelerating the rate of clearance. If this expla-nation was valid, we would have expected the calculat-ed Vdss for group 3 to be lower than that for group 2,but this was not the case.

Because feeding the low-fat and low-protein dietresulted in a significant decrease in body fat, wehypothesized that group 3 dogs might also have an


SMALLANIMALS/

AVIAN Diet 1b Diet 2e Diet 3

Variable Pre Post Pre Post Pre Post

Body weight (kg) 8.7 � 1.0* 9.8 � 0.9* 10.2 � 1.4* 9.7 � 1.0* 10.7 � 1.1* 9.9 � 1.3*

Post only

Body fat (%) 11.7 � 4.7* 13.2 � 3.8* 7.5 � 2.1†VO2 (ml/kg0.75/min) 12.82 � 2.26* 11.91 � 1.74* 12.52 � 1.41*VCO2 (ml/kg0.75/min) 9.96 � 1.83* 8.44 � 1.76* 9.13 � 1.58*REE (kcal/kg0.75/24 h) 88.0 � 15.5* 80.4 � 11.9* 84.5 � 11.6*

VO2 � Oxygen consumption. VCO2 � Carbon dioxide production. REE � Resting energy expenditure.See Table 2 for remainder of key.

Table 4—Effects of feeding 1 of 3 diets for 2 months to adult female Beagles on body weight, body fat, andindirect calorimetry values before (pre) and 2 months after (post) daily oral treatment with phenobarbital

847_852.QXD 8/23/2005 1:55 PM Page 850

altered resting metabolic rate. If so, this altered meta-bolic rate might result in a change in the pharmacoki-netics of phenobarbital through associated changes indrug distribution or metabolism. However, no signifi-cant diet-induced differences were detected in REE;therefore, changes in pharmacokinetics variables can-not be attributed to alterations in metabolic rate.

It is possible that the different diets may haveaffected the bioavailability of phenobarbital whenadministered orally. Phenobarbital was administered inthe morning, approximately 3 to 4 hours after the dailymeal, and again following a fasting period, approxi-mately 12 hours later. Diet 3 was lower in fat and pro-tein and higher in dietary fiber content, which mayhave altered gastric emptying time or gastrointestinaltransit time. Meals containing cellulose or wheat branincrease the frequency of postprandial contractions indogs and may decrease duodenojejunal flow and pro-long gastrointestinal transit time.38 Fat content of ameal changes intragastric distribution of solid materi-al, and larger amounts may retard gastric emptying.39

The higher fiber content of diet 3 may have impairedthe absorption of phenobarbital by binding to it,increasing the luminal unstirred water layer, or pro-moting goblet cell mucus secretion.10 To our knowl-edge, interprandial effects of meals or diet compositionon phenobarbital absorption in dogs have not beenreported. However, unless prolonged differences inbioavailability cumulatively altered hepatic metabo-lism or whole body distribution of phenobarbital whenadministered orally, it is unlikely that this would play arole in diet-related differences in the pharmacokineticswhen phenobarbital was administered intravenously.The net result of the changes in pharmacokinetics inthe low-protein and low-fat diet group was serum phe-nobarbital concentrations that were significantly lowerthan the other 2 groups. Mean serum phenobarbitalconcentration in group 3 also was below reported ther-apeutic concentrations.1,2

Decreases in serum albumin and cholesterol con-centrations are associated with phenobarbital adminis-tration in epileptic dogs; it is hypothesized that thesedecreases may be the result of phenobarbital-inducedhepatic dysfunction.7 Serum albumin concentrationswere decreased in all diet groups in our study, and thischange was exacerbated in the low-protein and low-fatdiet group. In this group, posttreatment serum albu-min concentrations were significantly lower than thoseof the other groups. Possible explanations include diet-or drug-induced changes in hepatic metabolism orlower protein intake associated with the diet itself.Serum cholesterol concentrations in the low-proteindiet group were significantly increased in the post-treatment period. This finding is in contrast to previ-ously reported findings7 and may represent differencesin dietary cholesterol intake, induction of cholesterolsynthesis, or decreased hepatic removal of cholesterol.Previous studies have demonstrated this effect ofincreased dietary cholesterol intake on serum choles-terol concentrations.40 Altered hepatic metabolism,decreased protein intake, or both also may explain thesignificantly lower PCV, RBC, and hemoglobin concen-trations in the posttreatment period in the low-protein

and low-fat diet group, compared with the other dietgroups. Increases in serum ALP activities that weredetected in the posttreatment period were anticipatedand are likely attributable to phenobarbital induction,as has been reported.7 Increases in ALP activity for thelow-protein and low-fat diet group were greater thanthat of the maintenance group, which suggests a diet-associated effect on hepatocellular enzyme induction.Other studies have demonstrated this effect of restrict-ed dietary protein intake on serum ALP activities.41

Management of seizures in epileptic dogs withphenobarbital is largely based in part on empirical doseadjustments. Adjustments in dose are made with thegoal of achieving target therapeutic serum concentra-tions of phenobarbital while minimizing toxic effectsthat are more commonly associated with high serumconcentrations of the drug. This effort is easily con-founded when factors that influence the pharmacoki-netics of phenobarbital, such as diet, are not recog-nized by the clinician. Findings of the present studyrevealed that differences in diet or body compositionsubstantially alter the pharmacokinetics of phenobar-bital, and there is a significant effect of diet-drug inter-actions on CBC and serum biochemical concentrationscommonly used to monitor general health.Consequently, diet and body composition changes(obesity or weight loss) should be taken into consider-ation by clinicians when contemplating phenobarbitaldose adjustments, and serum phenobarbital concentra-tions should be monitored closely when changes indiet or body composition develop.

aRidglan Farms Inc, Mt Horeb, Wis.bPrescription Diet Canine c/d, Hill’s Pet Nutrition Inc, Topeka, Kan.cElkins-Sinn Inc, Cherry Hill, NJ.dSarstedt Monovettes, Sarstedt Inc, Princeton, NJ.ePrescription Diet Canine k/d, Hill’s Pet Nutrition Inc, Topeka, Kan.fPrescription Diet Canine w/d, Hill’s Pet Nutrition Inc, Topeka, Kan.gWestern Veterinary Supply Inc, Porterville, Calif.hTDx fluorescence polarization analyzer, Abbott Laboratories, Abbott

Park, Ill.iRSTRIP, MicroMath Scientific Software, Salt Lake City, Utah.jFort Dodge Animal Health, Fort Dodge, Iowa.kVedco Inc, St Joseph, Mo.lThe Butler Co, Dublin, Ohio.mBecton-Dickinson, Sandy, Utah.nAbbott Laboratories, North Chicago, Ill.oHologic QD 1000/W, version 5.71P, Hologic Inc, Bedford, Mass.

References1. Brown SA. Anticonvulsant therapy in small animals. Vet

Clin North Am Small Anim Pract 1988;18:1197–1216. 2. Kay WJ. What is epilepsy? Probl Vet Med 1989;1:495–500.3. Abramson FP. Autoinduction of phenobarbital elimination

in the dog. J Pharm Sci 1988;77:768–770.4. Ravis WR, Nachreiner RF, Pedersoli WM, et al.

Pharmacokinetics of phenobarbital in dogs after multiple oraladministration. Am J Vet Res 1984;45:1283–1286.

5. Pedersoli WM, Wike JS, Ravis WR. Pharmacokinetics ofsingle doses of phenobarbital given intravenously and orally to dogs.Am J Vet Res 1987;48:679–683.

6. Farnbach GC. Serum concentrations and efficacy of pheny-toin, phenobarbital, and primidone in canine epilepsy. J Am Vet MedAssoc 1984;184:1117–1120.

7. Chauvet AE, Feldman EC, Kass PH. Effects of phenobarbi-tal administration on results of serum biochemical analyses andadrenocortical function tests in epileptic dogs. J Am Vet Med Assoc1995;207:1305–1307.


SMALLANIMALS/

AVIAN

Author:Pleaseverifycompanyname infootnotesd and o.

847_852.QXD 8/23/2005 1:55 PM Page 851

8. Kantrowitz LB, Peterson ME, Trepanier LA, et al. Serumtotal thyroxine, total triiodothyronine, free thyroxine, and thy-rotropin concentrations in epileptic dogs treated with anticonvul-sants. J Am Vet Med Assoc 1999;214:1804–1808.

9. Dayrell-Hart B, Steinberg SA, VanWinkle TJ, et al.Hepatotoxicity of phenobarbital in dogs: 18 cases (1985–1989). J AmVet Med Assoc 1991;199:1060–1066.

10. Fettman MJ, Phillips RW. Dietary effects on drug metabo-lism. In: Hand MS, Thatcher CD, Remillard RL, et al, eds. Small ani-mal clinical nutrition. 4th ed. Marceline, Mo: Walsworth PublishingCo, 2000;923–936.

11. Anderson KE, Kappas A. Dietary regulation of cytochromeP450. Annu Rev Nutr 1991;11:141–167.

12. Guengerich FP. Effects of nutritive factors on metabolicprocesses involving bioactivation and detoxication of chemicals.Annu Rev Nutr 1984;4:207–231.

13. Fagan TC, Walle T, Oexmann MJ, et al. Increased clearanceof propranolol and theophylline by high-protein compared withhigh-carbohydrate diet. Clin Pharmacol Ther 1991;50:254–258.

14. Guengerich FP. Influence of nutrients and other dietarymaterial on cytochrome P450 enzymes. Am J Clin Nutr 1995;61(suppl):651S–658S.

15. Pantuck EJ, Pantuck CB, Kappas A, et al. Effects of proteinand carbohydrate content of diet on drug conjugation. ClinPharmacol Ther1991;50:254–258.

16. Mounie J, Faye B, Magdalou J, et al. Modulation of UDP-glucuronyltransferase activity in rats by dietary lipids. J Nutr 1986;116:2034–2043.

17. Yoo JSK, Smith TJ, Ning SM, et al. Modulation of the levelsof cytochromes P450 in rat liver and lung by dietary lipid. BiochemPharmacol 1992;42:2535–2542.

18. Ducharme MP, Slaughter RL, Edwards DJ. Vancomycinpharmacokinetics in a patient population: effect of age, gender, andbody weight. Ther Drug Monit 1994;16:513–518.

19. Dunn TE, Ludwig EA, Slaughter RL, et al. Pharma-cokinetics and pharmacodynamics of methylprednisolone in obesity.Clin Pharmacol Ther 1991;49:536–549.

20. Schwartz-Porsche D. Management of refractory seizures. In:Kirk RW, Bonagura JD, eds. Current veterinary therapy XI.Philadelphia: WB Saunders Co, 1992;986–991.

21. Thurman GD, McFayden ML, Miller R. The pharmacoki-netics of phenobarbitone in fasting and non-fasting dogs. J S Afr VetAssoc 1990;61:86–89.

22. Walters LM, Ogilvie GK, Salman MD, et al. Repeatability ofenergy expenditure measurements in clinically normal dogs by use ofindirect calorimetry. Am J Vet Res 1993;54:1881–1885.

23. Ogilvie GK, Salman MD, Kesel ML, et al. Effect of anesthe-sia and surgery on energy expenditure determined by indirectcalorimetry in dogs with malignant and nonmalignant conditions.Am J Vet Res 1996;57:1321–1326.

24. Grier SJ, Turner AS, Alvis MR. The use of dual-energy x-rayabsorptiometry in animals. Invest Radiol 1996;31:50–62.

25. Toll PW, Gross KL, Berryhill SA, et al. Usefulness of dual-energy x-ray absorptiometry for body composition measurement inadult dogs. J Nutr 1994;124:2601S–2603S.

26. Turner AS, Mallinckrodt CH, Alvis MR, et al. Dual-energyx-ray absorptiometry in sheep: experiences with in vivo studies. Bone1995;17 (suppl 4):381S–387S.

27. Andersen KE, Conney AH, Kappas A. Nutritional influ-ences on chemical biotransformation in people. Nutr Rev1982;40:161–171.

28. Campbell TC, Hayes JR. Role of nutrition in the drugmetabolizing enzyme system. Pharmacol Rev 1974;26:171–197.

29. Norred WP, Wade AE. Effects of dietary lipid ingestion onthe induction of drug metabolizing enzymes by phenobarbital.Biochem Pharmacol 1973;22:433–436.

30. Marshall WJ, McLean AEM. A requirement for dietarylipids for induction of cytochrome P450 by phenobarbitone in ratliver microsomal function. Biochem J 1971;122:569–573.

31. Reidenberg MM. Obesity and fasting—effects on drugmetabolism and drug reduction in obese subjects enhances carba-mazepine elimination. Clin Pharmacol Ther 1992;51:501–506.

32. Caraco Y, Zylber-Katz E, Berry EM, et al. Significant weightreduction in obese subjects enhances carbamazepine elimination.Clin Pharmacol Ther 1992;51:501–506.

33. Abernethy DR, Greenblatt DJ, Divoll M, et al. The influenceof obesity on the pharmacokinetics of oral alprazolam and triazolam.Clin Pharmacokinet 1984;9:177–183.

34. Shum L, Jusko WJ. Theophylline disposition in obese rats.J Pharmacol Exp Ther 1984;228:380–386.

35. Yuk J, Nightingale CH, Sweeney K, et al. Pharmacokineticsof nafcillin in obesity. J Infect Dis 1988;157:1088–1089.

36. Fleming RA, Eldridge RM, Johnson CE, et al. Disposition ofhigh-dose methotrexate in an obese cancer patient. Cancer1991;68:1247–1250.

37. Schwartz AE, Matteo RS, Ornstein E, et al. Pharmacokineticsof sufentanil in obese patients. Anesth Analg 1991;73:790–793.

38. Bueno L, Praddaude F, Fioramonti J, et al. Effect of dietaryfiber on gastrointestinal motility and jejunal transit time in dogs.Gastroenterology 1981;80:701–707.

39. Heddle R, Collins PJ, Dent J, et al. Motor mechanisms asso-ciated with slowing of the gastric emptying of a solid meal by anintraduodenal lipid infusion. J Gastroenterol Hepatol 1989;4:437–447.

40. Julien P, Fong B, Angel A. Composition, morphology anddistribution of high-density lipoproteins in plasma and peripherallymph: effect of feeding cholesterol and saturated fat. BiochimBiophys Acta 1988;960:275–285.

41. Davenport DJ, Mostardi RA, Richardson DC, et al. Protein-deficient diet alters serum alkaline phosphatase, bile acids, proteinsand urea nitrogen in dogs. J Nutr 1994;124:2677S–2679S.


SMALLANIMALS/

AVIAN

847_852.QXD 8/23/2005 1:55 PM Page 852

effects of diet on pharmacokinetics of phenobarbital in healthy dogs

Documents