J. vet. Phannacol. Therap. 12,209-216, 1989.

Oral clindamycin disposition after single and

multiple doses in normal cats
S. ANTHONY BROWN, T. M. DIERINGER,* R. P. H U N T E R & M. J . ZAYAt
Department of Veterinary Physiology and Pharmacology, *Department of Small Animal
Medicine and Surgery, College of Veterinary Medicine, Texas A&M University, College Station,
TX 77843, and tBiochemistry and Residue Analysis, Agricultural Division, T h e Upjohn
Company, Kalamazoo, MI 4900 1, U.S.A.

Brown, S. Anthony, Dieringer, T., Hunter, R.P. & Zaya, M.J. Oral clindamycin
disposition after single and multiple doses in normal cats. J . vet. Pharmucol.
Therap. 12, 209-216.
Eighteen normal cats were randomly allocated into three treatment groups and
dosed with clindamycin aqueous solution for 10 days at a dosage rate o f (1) 5.5
mg/kg b i d . ; (2) 11 mg/kg b i d . ; o r (3) 22 mg/kg once daily. Serum disposition of
clindamycin was determined after the first and last dose of clindamycin was
given, and was analyzed $ing model-independent pharmacokinetics by both the
trapezoidal rule method and the predictive equation method. Complete blood
counts and clinical chemistries were determined before and after the study. The
trapezoidal rule method produced similar mean results with much less variance
than the predictive equation method. Mean residence time was longer (P < 0.05)
after the high dose (393 f 77 min) than after either the low or medium doses
(276 f 51 and 274 f 45 min, respectively). Oral volume of distribution (Vd(&)
after the high dose (3.06 k 0.92 Ykg) was larger (P < 0.05) than that after the low
or medium doses (1.62 k 0.30 and 1.76 f 0.53 Vkg, respectively). Oral Vd(,,)/F
was significantly smaller (P < 0.01) after the last dose than after the first dose
when analyzed by treatment group. Significant (P < 0.01) decreases in the leuko-
gram and erythrogram were observed, due to the large amount of blood col-
lected for drug analysis. No clinical signs of drug intoxication were observed,
and no drug-related necropsy findings were found.
Dr S . Anthony Brown, Department of VeterinaryPhysiology and Pharmacology, College of
Veterinary Medicine, Texas AUM University, Colkge Station, TX 77843, U.S.A.

INTRODUCTION many of the anaerobic bacterial infections of

Clindamycin is a lincosamide antibiotic which
is effective against gram-positive infections
and anaerobic bacteria (LeFrock et al., 1984).
In dogs, clindamycin is recommended for the
treatment of osteomyelitis caused by Staphylo-
coccus aureus, staphylococcal skin infections,
and other gram-positive soft tissue infec-
tions (Greene et al., 1985; Anonymous, 1987;
Braden et al., 1987, 1988). It has also shown
great promise as an effective antibiotic against
dogs and cats (Berg et al., 1984). The pharma-
cokinetics of clindamycin in humans (LeFrock
et al., 1984) and dogs (Anonymous, 1987)
have been quite well determined. However,
the single-dose and multiple-dose disposition
of oral clindamycin in the serum and tissues of
cats have not been determined, nor has the re-
sponse of cats to chronic clindamycin therapy.
The purpose of this study is to describe the
serum disposition of clindamycin in normal
adult cats, and to describe any untoward

209
210 S. Anthony Brown et al.
reactions of normal cats to multiple oral doses
of clindamycin. Dosage recommendations will
be made which will achieve peak serum
concentrations four times the MIC and main-
tain serum concentrations of clindamycin
above the MZC for the entire dosing interval
for several bacteria which may produce clini-
cally important infections in cats.
The purpose of this study was: (1) to
determine the disposition of orally adminis-
tered clindamycin in the serum and tissues of
normal cats at three dosage rates for 10 days;
(2) to determine the adverse affects (if any)
and clinical pathological, gross pathological,
and histopathological changes (if any) which
may result from the administration of mul-
tiple oral doses of clindamycin in normal cats
for 10 days; and (3) to determine the most
feasible dosage regimen for giving clindamy-
cin to cats.
MATERIALS AND METHODS
Eighteen normal, healthy domestic shorthair
cats of both sexes were acquired by the Lab-
oratory Animal Resource and Research Facil-
ity at Texas A&M University. These cats were
acclimatized to the environment of the labora-
tory animal facility for 4 weeks prior to the
onset of the experimental study. Each cat was
housed individually with litter and paper re-
placement daily. Each cat was fed a standard
dry cat chow (Purina Cat Chow, Ralston
Purina, St Louis, MO) and water was provided
ad libitum daily.
Prior to the study, all cats were physically
examined and weighed. Complete blood
counts, serum concentrations of urea nitro-
gen, creatinine, albumin, total protein, glu-
cose, bilirubin, calcium, inorganic phos-
phorus, and cholesterol, and serum activities
of alanine aminotransferase, alkaline phos-
phatase, and y-glutamyltransferase were
determined on Days 0 and 10 (the last day of
the study).
Two days before the start of the study (Day
-2) and 2 days before the end of the study
(Day 8), each cat was anesthetized with halo-
thane and an 18-gauge catheter (Venocath,
Abbott Laboratories, North Chicago, IL) was
placed into the jugular vein using an aseptic
technique. The cats were allowed to recover to
the standing position before they were placed
back into their respective cages. The site of
catheter placement was checked daily for
signs of infection, and all catheters were
removed no more than 72 h after placement
in each cat.
The cats were allocated into two blocks of
nine cats each, and the pharmacokinetic study
was performed and analyzed using a random-
ized complete block design with repeated
measures. In each block, nine cats were ran-
domly assigned to one of three treatment
groups: Group 1 - 5.5 mg/g oral clindamycin
b i d . for 10 days; Group 2 - 11 mg/kg oral
clindamycin b i d . for 10 days; and Group 3 -
22 mg/kg oral clindamycin once a day for
10 days. T h e liquid formulation of clindamy-
cin (Antirobe Aquadrops@,The Upjohn Com-
pany, Kalamazoo, MI) was used throughout
the study. Blood samples were taken from
each cat at 0, 20, 40, 60, 90, 120, 150, 180,
240, 360, 480, and 720 min (and 1440 min
from Group 3 cats) after the first oral dose of
clindamycin (Day 0). All cats continued to
receive oral clindamycin according to their
assigned group for 10 days. Blood samples
were obtained at 3 and 12 h after dosage on
Days 3, 5, and 8 of the study for deter-
mination of peak and trough serum clinda-
mycin concentrations. Blood samples were
collected after the last dose of clindamycin
(Day 9) at the same post-administration times
as they were after the first oral dose of
clindamycin. The 2-h blood sample was also
used to determine the neutrophil phagocytic
and bacterial killing capacity for each apimal
compared to that of normal cats evaluated
concurrently. After the last dose of clindamy-
cin, each cat was euthanatized using sodium
pentobarbital prior to complete post-mortem
examination. All blood samples were allowed
to clot at room temperature for 1 h, then
centrifuged, and the serum was removed and
stored frozen at -20°C until they were
shipped on dry ice for analysis.
Serum was analyzed for clindamycin using a
reversephase solid-phase extraction followed
by a microbiological cylinder/plate assay sensi-
tive to 0.05 pg clindamycin/ml of serum.
Briefly, a 1.0-ml serum sample was passed
through a pre-conditioned C18 solid-phase
extraction column U.T. Baker), washed, and
the antibiotic eluted from the column with

methanol into a small test tube. The sample
was re-dissolved in 1.O ml of 0.1 M potassium
phosphate buffer (pH 8.0), and was ready for
analysis using the microbiological assay.
Further dilutions of the sample (if needed)
were made in buffer (pH 8.0). T h e cylinder/
plate assay used to determine the clindamycin
concentration is similar to the standard mi-
crobial assay method for antibiotics with some
modifications. The assay organism (M. luteus),
ATCC 9341 (concentrated frozen suspen-
sion), is inoculated at 0.02% into Antibiotic
Medium No. 11 (Difco), and a 7.0-ml layer of
medium is added to each Petri dish. A 100-pl
aliquot of sample (and standard solution) was
alternately added to each cylinder and the
plates were incubated at 35°C for 16-18 h.
The concentration of each sample was then
calculated from zone of inhibition diameters
using polynomial regression techniques.
The neutrophil phagocytic capacity and
bacterial killing capacity were determined
using the following protocol. Five milliliters
of cat blood was collected into 200 iu of
preservative-free heparin sodium, and centri-
fuged at 400 g for 5 min at room temperature.
Plasma was removed and replaced with an
equal volume of divalent cation-free Hanks’
Balanced Salt Solution with 15 mM HEPES
(DCFHBSS with HEPES). Five milliliters of
DCFHBSS with HEPES was then added.
Using two tubes, the blood-HEPES mixture
was carefully layered onto 3.33 ml Ficoll-
Hypaque (specific gravity 1.056) in each tube.
The tubes were capped and centrifuged at
400 g for 30 min at room temperature. The
supernatant was collected and discarded, in-
cluding the mononuclear cell band. The
remaining pellet contained only erythrocytes
and granulocytes. Erythrocytes were lysed
with three repeated exposures to red cell
lysing buffer (6 mlltube), and centrifuged
three times at 5°C. Because the supernatant
still contained Ficoll-Hypaque, the first
centrifugation was done at 400 g. The remain-
ing two centrifugations were done at 250gfor
10 min. The cells were washed twice in
DCFHBSS with HEPES (10 ml/tube), and
centrifuged at 250 g for 10 min at room
temperature. 250 pI of DCFHBSS with
HEPES was then added to each pellet, and the
two cell suspensions from each sample were
combined. A 10-pl cell suspension was added
Oral clindamycin in cats 21 1
to 390 pI of 0.03% trypan blue (120 pl
trypan blue 0.1% in physiologic saline solution
plus 270 pl DCFHBSS with HEPES), and
total number of live cells were counted using
a hemocytometer. Each sample was then ad-
justed to a final cell concentration of 5.0 x lo6
cells/ml. Phagocyte-bacterial suspensions were
made by adding 0.5 ml of cell suspension to
0.3 ml of DCFHBSS with HEPES and 0.2 ml
of washed bacteria (Staphylococcus aureus in
tryptose broth, 18-24 hour culture, lo8 CFU/
ml). The mixture was incubated in a shaking
water bath (37°C) for 45 min. A 100-p1 ali-
quot of the suspension is added to 3 ml of
DCFHBSS with HEPES and centrifuged at
400 g for 2 min at 5°C. The supernatant was
discarded, and 1.0 ml of 0.014% Acridine
Orange (ice cold) was added to each sample.
This was vortexed and allowed to stain for
1 min, then 200 p1 of DCFHBSS with HEPES
was added to each sample. While protecting
from the light, a small drop of the suspension
was placed on a slide and covered with a cover
slip. The slide was observed at a magnification
of 1000 oil ernersion power using a fluor-
escence microscope. One-hundred cells were
counted and determined to be phagocytizing
bacteria and whether the bacteria phagocy-
tized were dead or alive. Dead cells were not
counted, and mononuclear cells were differ-
entiated from granulocytes. Control cat blood
was compared with that obtained from cats
being given clindamycin for each run.
Clindamycin disposition was determined
using model-independent calculations, using
both the trapezoidal rule and standard predic-
tive equations (Gibaldi & Perrier, 1982). T h e
initial pharmacokinetic parameters of the pre-
dictive equations were determined using a
computerized curve stripping routine ( R S ~ R I P ,
MicroMath, Salt Lake City, UT). Final predic-
tive functions were fitted to the data by means
of a weighted, non-linear least-squares regres-
sion algorithm conducted on a personal com-
puter (PCNONLIN, Statistical Consultants, Inc.,
Lexington, KY). These final pharmacokinetic
equations were evaluated for goodness-of-fit
by residual trends and Akaike’s information
criterion (Yamoaka et al., 1978). When the
optimum pharmacokinetic function was fit to
the data, the area under the curve (AUC) and
area under the first moment curve (AUMC)
were determined by integrating the predictive
212 S. Anthony Brown et al.

equations. The AUC and AUMC, determined TABLE I . Comparison of hematology data before
using both the sum of trapezoids or the in- and after 10 days of clindamycin treatment in cats
tegration method, were used for the calcula-
tion of steady-state volume of distribution Variable First dose Last dose
(V,&F), serum clearance (ClJF), and mean ~~ ~

residence time (MRT),using model-in- WBC ( X lOOO/pI) 14.57 k 5.69* 8.76 f 2.99*
dependent equations (Gibaldi & Perrier, RBC ( X lo6/@) 6.71 k 1.00t 5.01 k 0.74t
1982). A similar pharmacokinetic analysis was hematocrit (%) 30.32 k 4.61* 25.48 f 4.24*
done on the serum concentrations from each hemoglobin (g/dl) 10.62 f 1.61t 8.61 f 1.23t
cat after the last of the multiple oral doses of total protein (g/dl) 7.06 f 0.70 6.78 f 0.73
clindamycin, except that an equation was used neutrophils
that accounted for the accumulation of drug ( X 100O/pl) 9.69 f 5.75* 7.62 f 4.32*
due to multiple dosing (Gibaldi & Perrier,
1982). WBC = total white blood cell count; RBC = total
Normality was determined using Bartlett’s erythrocyte count. *P < 0.01; tP < 0.001.
test for normality. Normally distributed para-
meters are reported as arithmetic means &
standard deviations, whereas half-lives are TABLE 11. Comparison of clinical chemistry data
reported as harmonic means & pseudo- obtained before and after 10 days of clindamycin
standard deviation. Clinical chemistries and treatment in cats
hematologic data were compared by dose
from first to last dose using the paired Variable First dose Last dose
Student’s t-test, whereas pharmacokinetic
values were statistically compared with a two- Cholesterol (mgldl) 94.1 f 21.6t 104.9 f 24.7t
way analysis of variance using repeated meas- SGPT (U/dl) 30.2 f 10.1 36.1 f 27.7
ures, with the treatment groups being dose Albumin (g/dl) 3.36 f 0.31* 3.12 f 0.37*
and time compared by method of calculation. ALP (U/dl) 17.4 f 8.05 18.7 f 8.64
Other pharmacokinetic parameters were cal- Bilirubin (mg/dl) 0.18 f 0.07 0.21 f 0.11
BUN (mg/dl) 23.3 k 5.13 85.2 f 5.97
culated and compared with the parameters Calcium (mEq/l) 8.93 f 0.40 9.01 f 0.51
determined after the initial oral dose using Creatinine (mg/dl) 1.42 k 0.25* 1.33 f 0.25*
the Kruskal-Wallis non-parametric analysis of Glucose (mg/dl) 96.7 f 22.4 82.0 f 43.4
variance. Phosphate (mg/dl) 4.41 f 0.75 4.89 f 0.84
Total protein (g/dl) 6.85 f 0.83 6.55 f 0.65
GGT (U/dl) 2.00 f 0.00 2.00 -+ 0.00
RESULTS
SGPT = serum glutamate pyruvate trans-
All cats remained clinically healthy through- aminase; ALP = serum alkaline phosphatase;
out the study, with the only unusual observa- BUN = serum urea nitrogen; GGT = serum-
tion being profuse salivation and lip-smacking glutamyltransferase. *P < 0.05; tP < 0.01.
at the time the liquid drug formulation was
administered. Overall evaluation of hematol-
ogy data before the first dose and after the last the other hematology or clinical chemistry
dose (Table I) showed significant decreases data changed significantly from the first dose
in erythrocyte count (P < O.OOOl), hemato- to the last dose of clindamycin. Furthermore,
crit (P < 0.01), hemoglobin concentration none of the statistically significant changes in
(P < 0.0001), total leukocyte count (P c either the hemogram or the clinical chemi-
0.001), and polymorphonuclear leukocytes stries were outside the range of normal feline
(P < 0.01). Small but statistically significant values observed by the laboratory.
decreases in serum albumin (P < 0.01) and On necropsy, mild left ventricular hyper-
serum total protein (P < 0.05),and serum trophy was noted in seven of the 18 cats.
creatinine (P C 0.05) were noted, and a Histologic examination revealed mild colitis in
similar small increase in serum cholesterol nine of 18 cats, and mild cholestasis in seven
(P < 0.01) was observed (Table 11). None of of 18 cats. Only mild histologic changes
 Oral clindamycin in cats 213

indicative of left ventricular hypertrophy
were seen. The severity of the lesions was
independent of treatment group, and no
other post-mortem abnormalities were consis-
tently found.
The serum clindamycin concentration-time
data are depicted in Figs 1 and 2. When
analyzed by two different pharmacokinetic
methods (Table 111), both methods produced
similar means for AUC, vd(ss), Cl,, and MRT,
but the values derived using the trapezoidal
rule were much less variable than those values
derived using the predictive equation method.
Therefore, the calculated parameters using
the trapezoidal rule were used for statistical
comparison (Tables IV and V). Oral vd(,,)/F
was significantly smaller (P < 0.01) after the
last dose than after the first dose when
analyzed by treatment group (Fig. 3). T h e
AUC increased significantly (P < 0.05) with
increasing dose, and MRT was longer
(P< 0.05) after the high dose (393 f 77 min)
than after either the low or medium doses
(276 k 51 and 274 f 45 min, respectively;
Fig. 4). Oral Vd(,,)IF after the high dose (3.06
f 0.92 Vkg) was larger (P < 0.05) than that
observed after the low or medium doses (1.62
f 0.30 and 1.76 f 0.53 l/kg, respectively).
The dose-normalized AUC (AUCldose) tended
to be smaller (P < 0.10) after administration
of the high dose (0.14 f 0.06 kg.min/ml)
compared with that observed after either the
low or medium doses (0.20 f 0.09 and 0.17 f
0.06 kg.min/ml, respectively; Fig. 5).

- 10.0-

0 60 120 180 240 300 360 420 480 540 600 660 720’
Time (min)

FIG. 1. First dose serum clindamycin concentration-time data for three oral dosage regimens: 5.5 mg/kg
b.i.d. ( 0 ) ; 11 mg/kg b i d . (W); and 22 mglkg once daily ( A ) (data are given as mean k 1 SD).

E
2
0.1
0
I I ’ I I I I I I I I
60 120 180 240 300 360 420 480 540 600 660 720
I :
I
‘
‘
lo~ooI 12b0 I 14100

Time ( m i n )

FIG. 2. Last dose serum clindamycin concentration-time data for three oral dosage regimens: 5.5 mg/kg
b.i.d. ( 0 ) ; 1 1 mg/kg b i d . (W); and 22 rng/kg once daily ( A ) (data are given as mean k 1 SD).
214 S. Anthony Brown et al.

T A B L E I I I. Comparison of trapezoidal rule and predictive

equations methods for calculating oral clindamycin pharmacokinetic
values in cats

Variable Trapezoidal rule Predictive equation

AUC (pg.minlm1) 2025 f 1198 2032 f 1291

AUMC (pg.min2/ml) 709911 f 639469 1615288 f 4788375
V d ( d F (Vkg) 2.15 f 0.97 2.66 f 1.15
ClJF (mVmin/kg) 6.98 f 2.93 7.41 f 3.61
MRT (min) 314 f 81 507 f 850
AUCldose (kg.min/ml) 0.17 f 0.07 0.19 f 0.16

AUC = area under the concentra~ion-time curve; AUMC = area

under the first moment curve; V,&F = oral volume of distribution
(volume of distribution divided by bioavailability); C1,lF = oral serum
clearance (serum clearance divided by bioavailability); and MRT =
mean residence time.

TAB LE IV. Comparison of first and last dose clindamycin

pharmacokinetic values (using trapezoidal rule calculation) in cats

Variable First dose Last dose

AUC (pg.min/ml) 2057 f 1468 1993 f 892

AUMC (pg.min*/ml) 770526 f 792673 649295 rt 453633
vd(ss)IF (Ilkg) 2.35 f 1.06 1.95 f 0.848
C1JF (mllminlkg) 7.64 F 3.60 6.31 t 1.95
MRT (min) 325 F 91 304 f 71
AUC/dose (kg.min/ml) 0.17 f 0.09 0.17 f 0.05

For key to abbreviations, see Table 111.

TABLE V. Comparison of oral clindamycin pharmacokinetic values obtained from different

dosages of clindamycin in cats

Dose

Variable 5.5 mglkg 11 mglkg 22 mg/kg

AUC (pgminlml) 1107 F 482* 1882 rt 696* 3086 t 1295**

AUMC (pg.min*/ml) 312610 f 160434 540007 f 282235 1277116 f 800659
vd(ss)IF (Vkg) 1.62 f 0.70t 1.76 f 0.53t 3.06 f b.92tt
CISIF (mllminlkg) 6.09 f 3.03 6.78 f 2.92 8.06 f 2.73
MRT (min) 276 f 51* 274 f 45* 393 f 77**
AUClDose (kg.min/ml) 0.20 rt 0.09 0.17 rt 0.06 0.14 F 0.06$

For key to abbreviations see Table 111. *,**Values with different superscripts different
(P < 0.05); t,ttValues with different superscripts different (P < 0.01).
 Oral clindamycin in cats 215

DISCUSSION
4.5L
4.0 T
Oral clindamycin solution appeared to be
unpalatable to the cats, as determined by their
excessive salivation and lip-smacking. The
degree of unpalatability make it more difficult
to administer the drug to the cats toward the
end of the study, since their resistance to
dosing became greater. However, the entire
dose could be administered with a minimum
of difficulty.
Dosage
The changes in the hemogram that were
FIG. 3. The effect on Vd(,,)IF of three dosage observed were not dose-dependent and may
regimens of oral clindamycin (5.5, 11 and 22 mg/ reflect the volume of blood collected from
kg). First dose (0);last dose (B).Last dose smaller each cat during the study. Alternatively, clin-
t.han first dose (P < 0.05). damycin may have a depressive effect on the
bone marrow, causing a relative pancytope-
nia. If the depression of both the erythroid
series and myeloid series was caused by clin-
500 damycin, the changes were mild and may be
of little clinical significance.
The increase in Vd(,,,/F observed with the
high dose, coupled with the decrease in the
AUC/dose for that group, indicate that the
$j 200 bioavailability may be less after the high dose
than after either of the two smaller doses.
I00 Alternatively, the true volume of distribution
may have been larger for the high dose,
-
5. e 1g/kg I I mg/kg
although the reason for that difference is
difficult to explain. The longer MRT observed
Dosage
after the high dose also points to a difference
FIG. 4 . The effect on MRT of three dosage in either the absorption or elimination proces-
regimens of oral clindamycin (5.5, 11 and 22 mg/ ses compared with the lower doses. Elucida-
kg). First dose (0);last dose (69). High dose MRT tion of the precise mechanisms responsible for
different (P < 0.05) than after either low or
medium doses. these changes in disposition require further
studies.
The usual minimum inhibitory concentra-
tions (MZC) for Staphylococcus aureus range
from 0.04 to 0.4 pg/ml; for most anaerobic
bacteria the MIC ranges from 0.1 to 3.1 pg/
ml, with 90-95% of most pathogenic anaero-
bic bacteria inhibited by less than 1.6 pg
clindamycin/ml (LeFrock et al., 1984; Pratt &
Fekety, 1986). The dosage rate of 5.5 mg/kg

lo.\
0.I
every 12 h in cats provides average peak
serum concentrations greater than four times
the MZC for most S. aureus and highly suscep-
tible anaerobic bacteria, and maintains serum
0.oL concentration above their MZC for 75% of the
Dosage dosing interval. The dosage rate of 11 mg/kg
FIG. 5. The effect on ACJCldose of three dosage
every 12 h in cats will produce average peak
regimens of oral clindamycin (5.5, 1 1 and 22 mg/ serum concentrations greater than four times
kg). First dose (0);last dose (D). the MZC for moderately susceptible anaerobic
216 S. Anthony Brown et al.
bacteria, with serum concentrations persisting
above the M l C of those bacteria for at least
75% of the dosing interval. T he dosage rate of
22 mg/kg every 24 h in cats will produce
average peak serum concentrations greater
than four times the MZC for most pathogenic
anaerobic bacteria, and will maintain serum
concentrations in the average cat above the
MIC of most pathogenic bacteria (< 1.6 pg/
ml) for the first 12 h. Although breakthrough
recrudescence of infections may be more
likely with once-a-day therapy, such break-
through infections have not been reported.
Finally, serum concentrations will probably be
maintained above the M l C for most S . aureus
bacteria (< 0.4 pg/ml) for 75% of the dosing
interval if 11 mg clindamycin/kg every 24 h is
administered as the oral solution.
Oral clindamycin solution therefore ap-
pears to be safe in cats, with minor palata-
bility problems. Serum concentrations can be
maintained above the MIC for most S . aureus
infections using either 5.5 mg/kg every 12 h or
1 1 mg/kg every 24 h, whereas 1 1 mg/kg every
12 h or 22 mg/kg every 24 h produces serum
concentrations above the MIC for many
anaerobic bacteria. Further clinical studies are
needed to confirm the appropriate oral clin-
damycin dosage regimen for treatment of
bacterial infections in cats.
ACKNOWLEDGMENTS
This study was supported by a grant from
T h e Upjohn Company, Kalamazoo, Michi-
gan, and is published as Journal Paper No.
TA-24046 from the Texas Agricultural
Experiment Station. T h e authors thank
Mrs P.A. Pierce for her technical assistance.
REFERENCES
Anonymous. (1987) Antirobe@ Capsules and
Antirobe Aquadrops@ Liquids. Upjohn Vetenn-
aty Report. The Upjohn Company, Kalamazoo,
Michigan.
Berg, J.N., Scanlan, C.M., Buening, G.M., Fales,
W.H., 'Schmidt, D.A. & Wilson, J.C. (1984)
Clinical models for anaerobic bacterial infections
in dogs and their use in testing the efficacy of
clindamycin and lincomycin. American Journal of
Veterinay Research, 45, 1299-1306.
Braden, T.D., Johnson, C.A. & Gabel, C.L. (1987)
Posologic evaluation of clindamycin, using a
canine model of post-traumatic osteomyelitis.
American Journal of Veterinuty Research, 48, 1101-
1105.
Braden, T.D., Johnson, C.A. & Wakenell, P. (1988)
Efficacy of clindamycin in the treatment of
Staphylococcus aureus osteomyelitis in dogs. Journal
of the American Veterinuty Medical Association, 192,
172 1-1725.
Gibaldi, M. & Perrier, D., eds (1982) P h a m c o k i n e -
tics, 2nd edition. Marcel Dekker, Inc., New York.
Greene, C.E., Cook, J.R. Jr. & Mahaffey, E.A.
(1985) Clindamycin for treatment of Toxoplusma
polymyositis in a dog. Journal of the American
Vetm'naty Medical Association, 187, 63 1-634.
LeFrock, J.L., Molavi, A. & Prince, R.A. (1984)
Clindamycin. In Antimicrobial Therapy. Eds Ris-
tuccia, A.M. and Cunha, B.A. pp. 235-247.
Raven Press, New York.
Pratt, W.B. & Fekety, R., eds (1986) The Antimicro-
bial Drugs.pp. 200-205. Oxford University Press,
New York.
Yarnoaka, K., Nakagawa, T . & Uno, T. (1978)
Application of Akaike's information criterion
(AIC) in the evaluation of linear pharmacokinetic
equations. Journal of Pharmacokinetics and Bio-
phamceutics, 6 , 165-175.