"MATERNAL-FETAL DISPOSITION, FETAL PHARMACODYNAMICS AND COMPARATIVE
PHARMACOKINETICS OF DIPHENHYDRAMINE IN PREGNANT AND NONPREGNANT SHEEP
by
SUN DONG YOO
B.Sc. (Pharm.), Sungkyunkwan University, 1980 M.Sc. (Pharm.), The University of Manitoba, 1985
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
in
THE FACULTY OF GRADUATE STUDIES
(Faculty of Pharmaceutical Sciences Division of Pharmaceutics and Biopharmaceutics)
We accept this thesis as conforming to the required standard.
THE UNIVERSITY OF BRITISH COLUMBIA MAY, 1989
Copyright by Sun Dong Yoo, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.
Department of Pharmaceutical Sciences
The University of British Columbia Vancouver, Canada
Date June 21, 1989
DE-6 (2/88) ABSTRACT
A capillary GC/NPD assay method which provides improved sensitivity and selectivity for diphenhydramine is reported. The assay method involves single drug extraction and splitless sample injection.
Standard curves are linear in the range of 2-320 ng/mL of diphenhydramine, which represents an amount of the drug from -40 pg to
6.4 ng at the detector. This assay method was used in the subsequent studies of the maternal-fetal disposition and comparative pharmacokinetics of the drug in pregnant and nonpregnant sheep.
Pharmacokinetics of the drug were characterized in the nonpregnant sheep after i.v. bolus injection of 25, 50, 100 and 200 mg doses on a crossover basis. The total body clearance and dose- normalized AUC remained unchanged over the dose range studied. There were, however, significant increases in the elimination half-life and volume of distribution after a 200 mg dose, as compared to those after
25 mg dose. The plasma free fraction of the drug (0.229) was independent of drug concentrations over the range of 10-2,000 ng/mL.
The disposition of diphenhydramine in the maternal and fetal plasma, fetal tracheal and amniotic fluids was studied in the chronically catheterized maternal-fetal sheep following maternal i.v. bolus injection or maternal and fetal i.v. infusions. Following maternal i.v. bolus injection, the drug was rapidly transferred from th ewe to the fetus, resulting in significant fetal drug exposure
(fetal/maternal AUC ratio, -0.9). There was no significant difference i i i in the pharmacokinetic parameters between the pregnant and nonpregnant ewes. Plasma free fraction of the drug in the pregnant sheep was not significantly different from that in the nonpregnant sheep, but was smaller than that in the fetus (0.141 vs. 0.277). The total body clearance in the fetus (472.7 mL/min) was smaller than in the mother
(3426.1 mL/min). The transplacental clearance from fetus to mother
(264.4 mL/min) was -3 times higher than that from mother to fetus (82.4 mL/min). Maternal nonplacental clearance (3343.8 mL/min) accounted for
-98% of the maternal total body clearance, whereas fetal nonplacental clearance (208.4 mL/min) accounted for -45% of the fetal total clearance. Therefore, in the ewe, the drug appeared to be eliminated mainly by nonplacental pathways. However, in the fetus, both placental and nonplacental pathways are important for drug elimination. Maternal drug infusions resulted in fetal CNS depressant effects, whereas fetal infusions resulted in stimulation; this difference could be attributed to the different fetal plasma drug concentrations achieved.
Diphenhydramine accumulated both in the amniotic and fetal tracheal fluids. Following injection of diphenhydramine into the amniotic cavity, an average of 63% was eliminated by the mother, of which, -47% was received from the fetus and the rest (16%) from the amniotic fluid via direct diffusion across the uterus. Also, -37% of the total available drug was eliminated by the fetus, which is greater than that obtained after maternal i.v. bolus injection (-5%) or maternal infusion
(-2%) but is comparable to that after fetal infusion (-45%).
Diphenhydramine injected into the amniotic sac, therefore, appeared to be preferentially taken up by the fetus, resulting in much greater degree of fetal drug exposure than maternal exposure. The fetal i v pulmonary extraction ratio of the drug at steady-state averaged -0.08 and the delivery of the drug to the lungs via the pulmonary circulation seemed to be sufficient to account for drug accumulation in the fetal tracheal fluid. V
TABLE OF CONTENTS
Chapter Page
ABSTRACT i i
LIST OF TABLES x
LIST OF FIGURES xi i i
LIST OF SCHEMES xvi
LIST OF ABBREVIATIONS xvi i
ACKNOWLEDGEMENTS xx i
1. INTRODUCTION 1
1.1. General Background 1
1.1.1. Pharmacology and Therapeutics 1
1.1.2. Toxicity 3
1.2. Basic Pharmacokinetics 4
1.2.1. Absorption 4 1.2.2. Distribution 4 1.2.3. Plasma Protein Binding 5 1.2.4. Metabolism 5 1.2.5. Excretion 7 1.2.6. Interactions with Other Drugs 7
1.3. Analytical Methods 9
1.4. Antihistamines in Pregnancy 10
1.4.1. Antihistamine Use in Pregnancy 10 1.4.2. Placental transfer of Antihistamines 11 1.4.3. Safety of Antihistamines in Pregnancy 12 1.4.4. Effects of Drugs on the Fetal Central Nervous System 13 1.5. Rationale and Specific Aims of the Research 16
1.5.1. Rationale 16 1.5.2. Specific Aims 18
2. EXPERIMENTAL 19 vi
2.1. Development of a Capillary GC/NPD Assay Method for Diphenhydramine 19
2.1.1. Materials 19 2.1.2. Preparation of Stock Solutions 19 2.1.3. Instrumentation and Optimization of Gas Chromatographic Conditions 20 2.1.3.1. Equipment 20 2.1.3.2. Optimization of Gas Chromatographic Conditions 20 2.1.4. Drug Extraction Studies 22 2.1.5. Routine Procedures for Diphenhydramine Analysis 22 2.1.5.1. Operating GC Conditions 22 2.1.5.2. Drug Extraction from Biological Samples 23 2.1.5.3. Preparation of the Calibration Curve 23 2.1.5.4. Gas Chromatography-Mass Spectrometry 24 2.1.5.5. Application of the Developed Assay Method to a Placental Transfer Study 25
2.2. Pharmacokinetics of Diphenhydramine after Dose Ranging in Nonpregnant Sheep 25
2.2.1. Animals and Surgical Preparation 25 2.2.2. Drug Administration and Experimental Protocol 26 2.2.3. Pharmacokinetic Analysis 27 2.2.4. Plasma Protein Binding 28 2.2.5 Statistical Analysis 30
2.3. Pharmacokinetics of Diphenhydramine in the Pregnant Ewe and Fetal Lamb: i.v. Bolus Injection Studies 30
2.3.1. Animals and Surgical Preparation 30 2.3.2. Drug Administration and Experimental Protocol 31 2.3.3. Pharmacokinetic Analysis 32
2.4. Transplacental Clearance, Nonplacental Clearance and Fetal Effects of Diphenhydramine in Pregnant Sheep: Maternal and Fetal Infusions to Steady-State 34
2.4.1. Animals and Surgical Preparation 34 2.4.2. Drug Administration and Experimental Protocol 35 2.4.3. Pharmacokinetic Analysis 36 2.4.4. Plasma Protein Binding Studies 38 2.4.5. Recording and Analytical Techniques 38 2.4.6. Statistical Analysis 40
2.5. Recirculation of Diphenhydramine from Amniotic Fluid to the Ewe and Fetal Lamb: Intra-amniotic Bolus Injection Studies 40
2.5.1. Animals and Surgical Preparation 40 2.5.2. Drug Administration and Experimental Protocol 41 2.5.3. Data Analysis 42 2.6. Pulmonary Disposition of Diphenhydramine in the Fetal Lamb: Maternal or Fetal Infusions to Steady-State 42
2.6.1. Animals and Surgical Preparation 42 2.6.2. Drug Administration and Experimental Protocol 43 2.6.3. Pharmacokinetic Analysis 44
RESULTS 45
3.1. Analytical Development 45
3.1.1. Optimization of Capillary GC Conditions 45 3.1.2. Optimization of Drug Extraction Procedures 48 3.1.3. Application of the Developed Assay Method to Placental Transfer Studies in Pregnant Sheep 53
3.2. Pharmacokinetics of Diphenhydramine after Dose Ranging in Nonpregnant Sheep 59
3.3. Pharmacokinetics of Diphenhydramine in the Pregnant Ewe and Fetal Lamb: i.v. Bolus Injection Studies 68
3.4. Transplacental Clearance, Nonplacental Clearance and Fetal Effects of Diphenhydramine in Pregnant Sheep: Maternal and Fetal Drug Infusions to Steady-State 80
3.4.1. Transplacental and Nonplacental Clearances in Maternal-Fetal Unit 80 3.4.1.1. Achievement of Steady-State Diphenhydramine Concentrations in Maternal and Fetal Plasma 80 3.4.1.2. Plasma Protein Binding in Maternal and Fetal Sheep 87 3.4.1.3. Drug Disposition in Fetal Tracheal and Amniotic Fluids 87 3.4.1.4. Transplacental and Nonplacental Clearances 94 3.4.2. Fetal Effects of Diphenhydramine in Pregnant Sheep 101 3.4.2.1. Maternal Diphenhydramine Infusion 101 3.4.2.2. Fetal Diphenhydramine Infusion 105 3.4.2.3. Control Saline Infusion 113
3.5. Recirculation of Diphenhydramine from Amniotic Fluid to the Ewe and Fetus: Intra-amniotic Bolus Injection Studies 115
3.6. Pulmonary Extraction of Diphenhydramine in Fetal Lamb Following Maternal or Fetal Infusion to Steady-State 123
DISCUSSION 129
4.1. Development of a Capillary GLC/NPD Assay Method 129 VI 11
4.2. Pharmacokinetics of Diphenhydramine after Dose Ranging in Nonpregnant Sheep 132
4.2.1. Dose-Linearity of Pharmacokinetics of Diphenhydramine in Nonpregnant Sheep 132 4.2.2. Plasma Protein Binding in Nonpregnant Sheep 134
4.3. Pharmacokinetics of Diphenhydramine in Pregnant Sheep Following i.v. Bolus Administration 136
4.3.1. Placental Transfer of Diphenhydramine and the Extent of Fetal Drug Exposure 136 4.3.2. Disposition of Diphenhydramine in Maternal and Fetal Sheep 136 4.3.3. Comparisons of Diphenhydramine Pharmacokinetics Between Pregnant and Nonpregnant Sheep 137 4.3.4. Extent of Drug Elimination from the Fetal Lamb Following Maternal i.v. Bolus Dosing 137 4.3.5. Drug Accumulation in Fetal Tracheal and Amniotic Fluids 138
4.4. Transplacental Clearance, Nonplacental Clearance and Fetal Effects of Diphenhydramine in Pregnant Sheep: i.v. Infusion to Steady-State 143
4.4.1. Transplacental and Nonplacental Clearances 143 4.4.2. Drug Accumulation in Fetal Tracheal and Amniotic Fluids 151 4.4.3. Drug Disposition in Maternal and Fetal Sheep Following Maternal and Fetal Infusions 152 4.4.4. Fetal Effects of Diphenhydramine in Pregnant Sheep 155
4.5. Drug Recirculation from Amniotic Fluid to the Ewe and Fetal Lamb Following Intra-amniotic Bolus Administration 162
4.5.1. Amniotic Fluid Dynamics 162 4.5.2. Extent of Drug Excretion from Amniotic Fluid by Fetal Swallowing 164 4.5.3. Disposition of Diphenhydramine in the Amniotic and Fetal Tracheal Fluids, Maternal-Fetal Plasma after Drug Injection into Amniotic Cavity 165 4.5.4. Extent of Drug Elimination from the Ewe and Fetus Following Drug Injection into Amniotic Cavity 168
4.6. Pulmonary Disposition of Diphenhydramine in the Fetal Lamb Following Maternal or Fetal Infusions to Steady-State 171 SUMMARY AND CONCLUSIONS 175
REFERENCES 186 X
LIST OF TABLES
Table 1. Calibration curve data for sheep plasma. 56
Table 2. Mean values of arterial Po2, 02 content and hematocrit obtained after i.v. bolus injection of diphenhydramine to 6 nonpregnant sheep. 60
Table 3. Estimates of NONLIN pharmacokinetic constants of diphenhydramine obtained after i.v. bolus injection of 25, 50, 100 and 200 mg doses of the drug to 6 nonpregnant sheep. 63
Table 4. Pharmacokinetic parameters of diphenhydramine obtained after i.v. bolus injections to 6 nonpregnant sheep. 64
Table 5. Free fraction of diphenhydramine unbound to plasma protein determined in samples obtained after i.v. bolus injection of the 200 mg doses to nonpregnant sheep. 70
Table 6. Physical characteristics of individual pregnant sheep used in the maternal i.v. bolus injection study. 71
Table 7. The effect of maternal diphenhydramine administration
on fetal arterial pH, Pco2, Po2, 02 content and hematocrit following 100 mg i.v. bolus. 73
Table 8. Estimates of NONLIN pharmacokinetic constants obtained from maternal plasma diphenhydramine data following a 100 mg i.v. dose of the drug to individual ewes. 76
Table 9. Pharmacokinetic parameters obtained after a 100 mg i.v. dose of diphenhydramine to individual ewes. 77
Table 10. Percentage of the intravenously administered maternal dose (100 mg bolus) of diphenhydramine eliminated by the fetus via nonplacental pathways. 79
Table 11. Physical characteristics of individual pregnant sheep used in the steady-state infusion study. 82
Table 12. Steady-state diphenhydramine concentrations in the ewe and fetus following maternal and fetal drug administrations. 85
Table 13. Maternal to fetal or fetal to maternal diphenhydramine concentration ratios at steady-state following maternal and fetal drug infusion. 86
Table 14. Plasma free fraction of diphenhydramine in the pregnant ewe and fetus at steady-state following maternal or fetal drug administration. 88 xi
Table 15. Values for the observed maximum drug concentrations (Cmax)
of diphenhydramine and time to reach Cmax (Tmax) in tracheal (TF) and amniotic (AF) fluids following maternal and fetal infusions to steady-state. 92
Table 16. Average ratios of the fetal tracheal (TF) to arterial (FA) diphenhydramine concentrations and the elimination half-life (tjj of the drug obtained in maternal (MA) and fetal (FA) plasma, tracheal (TF) and amniotic (AF) fluids obtained after maternal and fetal infusions to steady-state.93
Table 17. Transplacental and nonplacental clearances of diphenhydramine based on total drug concentrations in the ewe and fetus following maternal and fetal infusions to steady-state. 95
Table 18. Total body clearances of diphenhydramine and percent contribution of nonplacental clearances to total drug elimination in the ewe and fetus based on total drug concentration. 96
Table 19. Transplacental and nonplacental clearances of diphenhydramine based on free drug concentrations in the ewe and fetus following maternal and fetal infusions to steady-state. 97
Table 20. Total body clearances of diphenhydramine and percent contribution of the nonplacental clearances to total drug elimination in the ewe and fetus based on free drug concentrations. 98
Table 21. Area under the plasma diphenhydramine concentration vs. time curves in the ewe and fetus and amount of the drug that was eliminated from the mother and fetus via the nonplacental pathway after maternal and fetal infusions. 100
Table 22. Maternal arterial pH, Pco?, P02, O2 content and hematocrit before, during and after diphenhydramine infusion to the ewe and fetus (n = 6). 102
Table 23. Fetal arterial pH, PCO2, P02, Oo content and hematocrit before, during and after diphenhydramine infusion to the ewe and fetus (n = 8). 103
Table 24. Effects of diphenhydramine on the fetal heart rate and arterial pressure calculated over 30 min intervals before, during and after drug infusions to the ewe and fetus (n = 8). 104
Table 25. Overall incidence of low, high and intermediate voltage ECoG episodes in the fetus prior to, during and after infusion of diphenhydramine to the ewe and fetus. 106 xi i
Table 26. Percentage of fetal breathing in low, high and intermediate voltage ECoG prior to, during and after infusion of diphenhydramine to the ewe and fetus (n = 8). 109
Table 27. Percentage of fetal electro-ocular activity (EoG) in low, high and intermediate voltage ECoG prior to, during and after infusion of diphenhydramine to the ewe and fetus. 110
Table 28. Duration of low, high and intermediate voltage ECoG episodes in the fetus prior to, during and after infusion of diphenhydramine to the ewe and fetus (n = 8). Ill
Table 29. Physical characteristics of individual pregnant sheep used in diphenhydramine intra-amniotic bolus injection study. 116
Table 30. Values for the observed maximum drug concentrations (Cmax)
T and time to reach Cmax ( raax) in maternal (MA) and fetal (FA) arterial plasma and fetal tracheal (TF) and amniotic (AF) fluids following injection of 50 mg diphenhydramine hydrochloride into the amniotic cavity 118
Table 31. Average ratios of the maternal arterial (MA), umbilical venous (UV) and fetal tracheal (TF) to fetal arterial (FA) drug concentrations and the elimination half-life (ti.) of the drug obtained after injection of 50 mg diphenhydramine hydrochloride into the amniotic cavity. 120
Table 32. Total area under the drug concentration vs. time curves for the maternal and fetal arterial and umbilical venous plasma, fetal tracheal and amniotic fluids after injection of 50 mg diphenhydramine hydrochloride into the amniotic cavity. 122
Table 33. The steady-state diphenhydramine concentrations and levels in the hour following cessation of the maternal or fetal infusion in fetal pulmonary arterial (PA), pulmonary venous (PV), carotid arterial (CA) blood, fetal tracheal (TF) and amniotic (AF) fluids and maternal arterial blood (MA). 126
Table 34. Fetal pulmonary extraction and the total body clearance of diphenhydramine calculated at steady-state after maternal or fetal drug administration. 128 xi i i
LIST OF FIGURES
Fig. 1. Hydrogen flow rate-NPD response curves: carrier gas (helium) flow rate constant, 1 mL/min; and air flow rate constant, 60 mL/min. 46
Fig. 2. Effect of air flow rate on NPD response and peak width: hydrogen flow rate constant, 3 mL/min; helium flow rate constant, 1 mL/min; and peak width = peak area count/peak height. 47
Fig. 3. Effect of column pressure on NPD response and peak width: hydrogen flow rate constant, 3 mL/min; air flow rate constant, 50 mL/min; and peak width = peak area count/peak height. 49
Fig. 4. Relationship between height equivalent to theoretical plates (H.E.T.P.) vs. linear velocity of the helium carrier gas determined for diphenhydramine. Column head pressures were 7, 9, 11, 13, 15 and 17 psi. 50
Fig. 5. Effect of purge activation time on NPD response: hydrogen flow rate constant, 3 mL/min; air flow rate constant, 50 mL/min; and helium flow rate constant, 1 mL/min. 51
Fig. 6. Effect of the addition of triethylamine (TEA) on extraction of diphenhydramine (hatched bar) and orphenadrine (open bar) with dichloromethane (IAA, isoamyl alcohol). 52
Fig. 7. Extraction time vs. NPD response curves: dichloromethane was used for drug extraction. 54
Fig. 8. Representative capillary gas chromatograms obtained from blank sheep plasma (A) and a plasma sample from a pregnant sheep (B) after administration of 100 mg of diphenhydramine hydrochloride intravenously. 55
Fig. 9. A semilogarithmic plot of the diphenhydramine concentration vs. time profiles obtained in maternal and fetal plasma following a 100 mg i.v. bolus dose to a pregnant ewe. 58
Fig. 10. Average plasma drug concentration vs. time curves obtained after i.v. administration of 25, 50, 100 and 200 mg doses of diphenhydramine hydrochloride to 6 nonpregnant sheep. 61
Fig. 11. Changes in the terminal elimination rate constant (lambdan) of diphenhydramine as a function of dose obtained in 6 nonpregnant sheep. Each symbol represents each animal and the horizontal lines represent mean values. 65
Fig. 12. Steady-state volume of distribution (Vd^*) of diphenhydramine as a function of dose obtained in 6 nonpregnant sheep. 66 xiv
Fig. 13. Relationship between the area under the plasma drug concentration vs. time curve and the i.v. bolus doses of diphenhydramine given on a crossover basis to 6 nonpregnant sheep. 67
Fig. 14. Determination of equilibrium time used in the study of plasma protein binding of diphenhydramine by equilibrium dialysis (n = 3). 69
Fig. 15. Representative capillary gas chromatograms of the maternal (MA) and fetal (FA) arterial plasma, fetal tracheal (TF) and amniotic (AF) fluids obtained from control- and drug- treated pregnant sheep. 74
Fig. 16. Representative semilogarithmic plots of the maternal and fetal arterial plasma, fetal tracheal and amniotic fluid diphenhydramine concentration vs. time profiles following a 100 mg bolus. 75
Fig. 17. Average maternal and fetal plasma diphenhydramine concentration vs. time curves obtained following i.v. infusion of the drug to the ewe (n = 8). 83
Fig. 18. Average maternal and fetal plasma diphenhydramine concentration vs. time curves obtained following i.v. infusion of the drug to the fetus (n = 8). 84
Fig. 19. Representative semilogarithmic plots of the maternal and fetal arterial plasma, fetal tracheal and amniotic fluid diphenhydramine concentration vs. time curves in a pregnant ewe and fetus following drug infusion to the ewe. 90
Fig. 20. Representative semilogarithmic plots of the maternal and fetal arterial plasma, fetal tracheal and amniotic fluid diphenhydramine concentration vs. time curves in a pregnant ewe and fetus following drug infusion to the fetus. 91
Fig. 21. High speed chart recordings of high, low and intermediate voltage ECoG patterns in a fetal lamb. 107
Fig. 22. Representative examples of the recording of breathing, ECoG and electro-oculographic (EoG) activity in the fetal lamb during the control period (A) and during fetal infusion of diphenhydramine (B). 108
Fig. 23. Examples of the large amplitude breathing activity present in the initial phases of diphenhydramine infusions to the fetus. 114
Fig. 24. Average diphenhydramine concentration vs. time curves for maternal and fetal arterial plasma, fetal tracheal and amniotic fluids following administration of 50 mg diphenhydramine hydrochloride into the amniotic cavity into 5 pregnant sheep. 117 XV
Fig. 25. Representative diphenhydramine concentration vs. time curves for maternal and fetal arterial and umbilical venous plasma, fetal tracheal and amniotic fluids following intra- amniotic bolus injection of 50 mg diphenhydramine hydrochloride to a pregnant ewe (ewe no. 287). Inset figure shows the fetal arterial and umbilical venous plasma and fetal tracheal drug concentration vs. time curves over the initial 6 hr period after drug administration. 121
Fig. 26. Plots of the fetal pulmonary arterial (PA), pulmonary venous (PV) and carotid arterial (GA) and maternal arterial (MA) diphenhydramine concentration vs. time profiles following maternal drug infusion. 124
Fig. 27. Representative plots of the fetal pulmonary arterial (PA), pulmonary venous (PV) and carotid arterial (CA) and maternal arterial (MA) diphenhydramine concentration vs. time profiles following fetal drug infusion. 125 xvi
LIST OF SCHEMES
Scheme 1. Chemical structure of diphenhydramine. 1
Scheme 2. A two-compartment open model used in the calculation of transplacental and nonplacental clearances in the maternal-fetal unit (Szeto et al., 1982). 37 xvii
LIST OF ABBREVIATIONS
AAG ctj-acid glycoprotein
AF amniotic fluid
ANOVA analysis of variance
AUC area under the plasma drug concentration vs. time curve
AUMC area under the first moment curve
AUTOAN a decision-making pharmacokinetic computer program
°C "centigrade
Cf steady-state fetal plasma total (bound and unbound) drug
concentration after maternal i.v. infusion
Cf/ steady-state fetal plasma total drug concentration after
fetal i.v. infusion
^f,free steady-state fetal plasma unbound drug concentration after
maternal i.v. infusion cf',free steady-state fetal plasma unbound drug concentration after
fetal i.v. infusion
CI-MS chemical ionization-mass spectrometry
CLff total drug clearance from the fetal compartment
CLfm transplacental clearance of drug from the fetal to maternal
compartment
CLf0 nonplacental clearance of drug from the fetal compartment
CLmf transplacental clearance of drug from the maternal to fetal
compartment
CLmm total drug clearance from the maternal compartment
CLmo nonplacental clearance of drug from the maternal compartment
C/.jg total body clearance of total drug (bound and unbound) xvi i i
cm centimeter
Cm steady-state maternal plasma total (bound and unbound) drug
concentration after maternal i.v. infusion
Cm/ steady-state maternal plasma total drug concentration after
fetal i.v. infusion
Cmax maximum drug concentration
Cm,free steady-state maternal plasma unbound drug concentration
after maternal i.v. infusion
Cm',free steady-state maternal plasma unbound drug concentration
after fetal i.v. infusion
CNS central nervous system
C.V. coefficient of variation
ECoG electrocorticographic
EEG electroencephalograph^
EOG electro-oculographic
EI electron impact
EI-MS electron impact-mass spectrometry
eq. equation
ER extraction ratio
FA fetal arterial
Fig. figure g gram g acceleration of gravity
GC gas chromatography
GC-MS gas chromatography-mass spectrometry
GC/NPD gas chromatography with nitrogen-phosphorus detection
H2 hydrogen gas xix
HC1 hydrochloric acid
He helium gas
H. E.T.P. height equivalent to a theoretical plate
Hg mercury
HPLC high performance liquid chromatography hr hour
I. D. internal diameter i.m. intramuscular
I.U. international unit i.v. intravenous kg kilogram
k0 drug infusion rate to the mother
kQ/ drug infusion rate to the fetus
KH2PO4 monopotassium phosphate
L liter
M molar (moles/liter) m meter
MA maternal arterial mg milligram
Ug microgram mL milliliter mm millimeter min minute m.w. molecular weight
N20 nitrogen oxide
NaOH sodium hydroxide ng nanogram NONLIN computer program for non-linear least squares regression
analysis of pharmacokinetic systems
NPD nitrogen-phosphorus detector
02 oxygen gas
O.D. outer diameter
Pco2 partial pressure of carbon dioxide in the blood pH negative logarithm of hydrogen ion concentration
Po2 partial pressure of oxygen in the blood
PTFE polytetrafluoroethylene
Qum umbilical blood flow r correlation coefficient r^ coefficient of determination
REM rapid eye movement s.c. subcutaneous
S.D. standard deviation sec second tk half-life
tmax time taken to reach maximum drug concentration
TP tracheal pressure t^ retention time of the peak vs. versus
* significantly different (p<0.05) ACKNOWLEDGEMENTS
I would like to sincerely thank Dr. James E. Axelson for his supervision, encouragement and financial support throughout my Ph.D. program. I am deeply indebted to Dr. Dan W. Rurak for his guidance, critical commentary and all the support with the animal experiments.
I would also like to thank my supervisory committee members, Dr.
Frank S. Abbott, Dr. Jim M. Orr and Dr. John G. Sinclair for their encouragement and constructive criticism throughout the program.
I thank Ms. Sandy M. Taylor, Dr. K. Wayne Riggs, Mr. Eddie Kwan and Ms. Marlene Van der Weyde for their help with the animal preparation and experiments and Barbara A. McErlane and David Main for their assistance with drug analysis. I also thank Ms. Grace Chan and Mr.
Matthew R. Wright for their constructive criticism. Thanks also to my colleagues in the laboratory, George R. Tonn, Jing Wang, Krishnaswamy
Yeleswaram for their encouragement and friendship.
Graduate Teaching Assistantship, Graduate Research
Assistantship, University Graduate Fellowship, University Travel Fund and CIDA Travel Grant are gratefully acknowledged. This project was supported by the Medical Research Council of Canada. xxi i
This thesis is dedicated
to
my mother, Yong Ok Ahn and my sister, Dong Ju Yoo for their constant love, encouragement and support from the other side of the world and also to the memory of my father, Tae Seung Yoo. 1
1. INTRODUCTION
1.1. General Background
Diphenhydramine, 2-(diphenylmethoxy)-N,N-dimethylethylamine, is a potent histamine Hj-receptor antagonist that possesses anticholinergic, antiemetic and sedative properties (Douglas, 1985). Diphenhydramine was synthesized in the 1940s (Rieveschl, 1947) and has been in clinical use for over 40 years. This compound is available, without prescription, as an antihistaminic, antiemetic and sedative agent. Diphenhydramine is a weakly basic amine with molecular weight (m.w.) of 255.4 and a pKa of 9.0
(Marshall, 1955; de Roose et al., 1970). Its structural formula is shown in Scheme 1.
Scheme 1. Chemical structure of diphenhydramine.
1.1.1. Pharmacology and Therapeutic Effects
A number of review articles describe the pharmacological actions of diphenhydramine (Loew, 1947; Melville, 1973; Drouin, 1985).
Diphenhydramine appears to block histamine Hj-receptors competitively and 2
reversibly (Rocha e Silva, 1978). Diphenhydramine is capable of antagonizing the ability of histamine to contract smooth muscle in the gastrointestinal tract and lungs of guinea pigs, and to induce vasodilator effects in cats and dogs (Melville, 1973; Douglas, 1985). Diphenhydramine also inhibits the action qf histamine-mediated allergic responses such as wheal formation, edema and increased capillary permeability. It suppresses salivary and lacrimal secretion caused by histamine but fails to antagonize the action of histamine to mediate gastric secretion (Melville, 1973). The action of histamine in the stimulation of gastric acid secretion is thought to be mediated via h^-receptors (Black et al., 1972). Diphenhydramine has a biphasic action on mast cells: at low doses, it inhibits the release of histamine and at higher doses, it induces histamine release (Mota and Dias
Da Silva, 1960; Lau and Pearce, 1985).
Diphenhydramine is known to be effective in relieving allergic symptoms such as hay fever, allergic rhinitis, urticaria, dermatoses and pruritis (Douglas, 1985). It is also known to reduce post-operative nausea and vomiting, and to reduce vomiting due to malignancy or antineoplastic drug use. It is recommended that diphenhydramine be administered as a 25-
50 mg oral dose and as a 10-50 mg i.v. or i.m. dose, with a maximum dose of
400 mg per day (Krogh, 1987). Diphenhydramine is contraindicated in newborn infants, and in patients experiencing an asthmatic attack or hypersensitivity to the drug. Safety has not been proven for use of diphenhydramine in pregnancy and lactation and, as a result, the use of the drug during pregnancy is not recommended (Krogh, 1987).
A good correlation has been reported between plasma 3 diphenhydramine concentrations and the suppression of histamine-induced skin wheal reaction or the sedative effects of the drug in humans
(Carruthers et al., 1978). Following i.v. or oral administration of 50 mg diphenhydramine hydrochloride, antihistamine effects can last for ~7 hr, whereas sleepiness or sedative effects are observed during the first ~3 hr
(Carruthers et al., 1978). Sleep and/or drowsiness seem to occur when plasma drug levels are above -70-100 ng/mL (Albert et al., 1975; Carruthers et al., 1978). It appears that antihistamine effects can be elicited at a therapeutic range of -25-50 ng/mL, without significant sedation.
1.1.2. Toxicity
Studies of acute diphenhydramine toxicity in animals show that the
i.v. LD50 of the drug is 42-46 mg/kg in the rat, 98 mg/kg in mice, 10-11
mg/kg in rabbits and 24-30 mg/kg in the dog. The oral LD50 of diphenhydramine ranges from 500-545 mg/kg in rats and from 164-167 mg/kg in mice (Rieveschl and Gruhzit, 1945; Gruhzit and Fisken, 1947). In these animal species, toxic doses may cause excitement, tremors, convulsions and respiratory and cardiac failure resulting in death (Rieveschl and Gruhzit,
1945; Gruhzit and Fisken, 1947).
In humans, diphenhydramine poisoning has been frequently reported, accounting for -5% of the total number of drug intoxications (Koppel and
Ibe, 1987). Acute symptoms of diphenhydramine poisoning in humans include impaired consciousness, hallucination, mydriasis, tachycardia, seizure and convulsions (wyngaarden and Seevers, 1951; Hausmann et al., 1983). In the reported cases of human diphenhydramine intoxication, approximately 6-40 4
times the therapeutic dose (50 mg) appears to have been ingested. Topical application of diphenhydramine can also elicit toxic symptoms, such as delirium and hallucinations in humans (Filloux, 1986). Acute toxicity of diphenhydramine can be treated by gastric lavage, termination of convulsions with barbiturates, diphenylhydantoin or paraldehyde, and ventilation for cardiorespiratory depression (Reichelderfer et al, 1955;
Hestand and Teske, 1977; Hausmann et a/., 1983).
1.2. Basic Pharmacokinetics
1.2.1. Absorption
Following administration of a single oral dose (50 mg) of diphenhydramine hydrochloride in humans, peak plasma drug concentrations occurred at 2-4 hrs, with maximum drug concentrations of -40-80 ng/mL
(Carruthers et al., 1978; Berlinger et al., 1982). Multiple oral administration of 50 mg q.i.d. to human volunteers resulted in average steady-state plasma drug concentrations of -100 ng/mL (Glazko et al.,
1974). A large first-pass effect has been noted after oral diphenhydramine administration in humans, with approximately 30-60% of the dose being metabolized by the liver before it reaches the systemic circulation (Albert et al., 1975; Carruthers et al., 1978; Berlinger et al., 1982; Meredith et al., 1984; Blyden et al., 1986).
1.2.2. Distribution
Diphenhydramine tissue distribution studies, performed following 5
subcutaneous (s.c.) administration in the rat and guinea pig, showed that the highest concentration was found in lung tissue, followed by spleen, kidney, liver and muscle (Glazko and Dill, 1949a). Hausmann et al. (1983) have reported diphenhydramine concentrations determined in various tissues of a human patient following fatal overdosage of the drug. The highest concentration was also found in the lungs, followed by kidney, liver, spleen, muscle and plasma. The drug concentration in the lungs was -10 times higher than that found in plasma.
1.2.3. Plasma Protein Binding
A preliminary study of diphenhydramine plasma protein binding determined in two human subjects showed that the drug protein binding was extensive (-98%, Albert et al., 1975). Spector et al. (1980) have reported the plasma protein binding of ^-diphenhydramine to be -76% and -85% in
Orientals and Caucasians. Diphenhydramine binds to human serum albumin but the degree of binding is low, suggesting that diphenhydramine may bind to plasma protein(s) other than albumin. However, the plasma protein binding characteristics of the drug have not been extensively studied in either human or animal species.
1.2.4. Metabolism
The liver appears to be the principal site of diphenhydramine degradation in rats, although the lung and kidney are also active (Glazko and Dill, 1949b). Diphenhydramine is readily demethylated by rat liver microsome preparations in vitro (Roozemond et al., 1965). A study 6
employing paper partition chromatography of urine obtained after administration of ^C-diphenhydramine to rats showed that at least six different radioactive compounds were present (Glazko et al., 1949). Drach et al. (1970) later reported the presence of unchanged diphenhydramine and
its N-demethylated metabolites, N,N-didemethyl- and N-demethyl- diphenhydramine, in rat urine. In the rhesus monkey, N-demethylation followed by oxidative deamination to diphenylmethoxyacetic acid is the main metabolic pathway (Drach and Howell, 1968). The mono- and di-demethylated derivatives of diphenhydramine, the N-oxide derivative and the deaminated carboxylic acid derivative have been identified as diphenhydramine metabolites in monkey urine (Drach and Howell, 1968; Drach et al., 1970).
Diphenylmethoxyacetic acid is further conjugated with glutamine in the
Rhesus monkey (Drach and Howell, 1968) or with glycine in the dog (Drach et
al., 1970), but neither the free nor conjugated form was found in rat urine
(Drach et al., 1970). The deaminated carboxylic acid derivative,
diphenylmethoxyacetic acid, has been found to be a major metabolite in
monkey plasma and urine (Drach and Howell, 1968; Drach et al., 1970).
Benzhydrol had been previously suggested to be a metabolite of
diphenhydramine in vitro (Glazko and Dill, 1949b). However, only a trace
amount of benzhydrol was detected in monkey urine (Drach and Howell, 1968).
The metabolic pathway(s) in man has not been fully established but the
identified urinary excretion products for diphenhydramine, measured by gas
chromatography-mass spectrometry (GC-MS), include small amounts of
unchanged drug and a primary amine metabolite, N,N-didemethyl- diphenhydramine, large amounts of a secondary amine, N-demethyl- diphenhydramine, and diphenylmethoxyacetic acid as the major metabolite
(Chang et al., 1974). Diphenylmethoxyacetic acid is excreted as a free or 7 as a conjugated form in humans but the exact nature of the conjugate has not been reported. There appears to be no information on diphenhydramine metabolism in sheep.
1.2.5. Excretion
Following administration of diphenhydramine (10 mg/kg s.c.) to rats, 4-6% of the dose was excreted in urine as unchanged drug (Glazko and
Dill, 1949a). In rabbits, -21% of the administered dose (25-100 mg s.c.) was excreted in urine as unchanged drug (Hald, 1947). In man, urinary diphenhydramine excretion accounts for a small portion (2-4%) of the administered dose (Hald, 1947; Albert et al., 1975), and the sum of the urinary excretion of diphenhydramine and its metabolites accounts for -50-
60% after oral drug administration (Glazko et al., 1974). In contrast,
-33% of the total radioactivity was recovered in urine following s.c. administration of ^C-diphenhydramine to rats with the remainder being recovered in the feces, suggesting the possibility of biliary excretion
(Glazko et al., 1949).
1.2.6. Interactions with Other Drugs
Diphenhydramine is known to potentiate effects of other centrally active drugs such aSjethanol, temazepam and butorphanol (Burns and
Moskowitz, 1983; Brademas, 1984; Kargas et al., 1985; Cohen et al., 1987).
Since diphenhydramine can cause sedation and mild euphoria at high doses in man, this compound is subject to abuse (Feldman and Behar, 1986).
Diphenhydramine has been reported to reduce the oral absorption of several 8
drugs such as sodium p-aminosalicylate in rats (Lavigne and Marchand,
1973), acetaminophen and sulfisomidine in rabbits (Ichibagase et al., 1981;
Imamura et al., 1981). This reduction of oral bioavailability was
speculated to be due to delay in the gastric emptying in these species.
When administered together with methaqualone, diphenhydramine can alter methaqualone pharmacokinetics: the elimination half-life was increased by
~2 fold and the total body clearance decreased by one-half in the rat
(Gupta et al., 1982). The alterations in the pharmacokinetics of
methaqualone were attributed to an inhibitory effect of diphenhydramine on
the hepatic microsomal biotransformation of the former drug.
It is not clear whether diphenhydramine induces drug-metabolizing
enzymes. Burns et al. (1963) treated three dogs with daily oral dose of
50 mg/kg of diphenhydramine. Forty days after the diphenhydramine
treatment, they administered 25 mg/kg oral dose of diphenhydramine and
measured the drug plasma concentrations over 7 hr and found that these were
considerably lower than those measured during the control period. These
investigators speculated that chronic diphenhydramine administration
resulted in enzyme induction of its own metabolism. The same group of
investigators (Conney and Burns, 1960) also reported that pretreatment in
rats with diphenhydramine (50 mg/kg/day for 4 days) shortened the duration
of zoxazolamine paralysis and suggested induction of liver microsomal
enzymes, although they did not directly measure the activity of the
zoxazolamine-metabolizing enzyme system for diphenhydramine. There appears
to be no direct evidence for the induction of liver microsomal enzyme
activity by diphenhydramine. 9
1.3. Analytical Methods
A number of analytical methods have been reported for the determination of diphenhydramine in biological fluids, including UV spectrophotometry (Wallace et a7., 1966), fluorescence dye techniques
(Glazko et a7., 1962; Glazko et al., 1974), high performance liquid chromatography (HPLC) (Skofitsch and Lembeck, 1983), GC-MS (Chang et al.,
1974; Carruthers et al., 1978) and gas chromatography (Bilzer and Gundert-
Remy, 1973; Albert et al., 1974; Baugh and Calvert, 1976; Abernethy and
Greenblatt, 1983; Barni Comparini et al., 1983; Chiarotti et al., 1983;
Lutz et al., 1983; Meatherall and Guay, 1984). UV spectrometry and fluorescence dye methods lack sufficient specificity and sensitivity for trace level drug measurement. The HPLC method is convenient since no extraction procedure or derivatization step is required, but sensitivity limitations require the use of relatively large volumes of plasma and injection of the entire sample volume into the chromatograph. GC methods employing conventional packed columns have been used for monitoring plasma concentrations in single dose pharmacokinetic studies (Bilzer and Gundert-
Remy, 1973; Albert et al., 1974; Baugh and Calvert, 1976; Abernethy and
Greenblatt, 1983). These methods require the use of plasma volumes larger than that normally collected during the study of placental transfer of drugs.
GC methods employing glass capillary columns and flame-ionization detection have been reported for the determination of a number of drugs, including diphenhydramine (Barni Comparini et al., 1983; Chiarotti et al.,
1983). The combination of nitrogen/phosphorus-specific detection (NPD) 10 with capillary GC has improved the detection limits for diphenhydramine to
facilitate studies of the pharmacokinetics of the drug in man (Lutz et al.,
1983; Meatherall and Guay, 1984). While these nitrogen selective methods
provide good sensitivity and selectivity, there appeared to be two
significant limitations in these methods: they required large plasma
volumes (1.0-3.0 mL) for extraction and a very small final reconstitution
volume (10 nl) for injection (Lutz et al., 1983; Meatherall and Guay,
1984). Due to these limitations in the existing assay methods, there was
still the requirement for an assay method with improved sensitivity, which
could be routinely applied to placental drug transfer studies where the
sample volume is limited.
1.4. Antihistamines in Pregnancy
1.4.1. Antihistamine Use in Pregnancy
Surveys of drug use in pregnancy published between 1963 and 1987
indicate that on average -20% of pregnant women take antihistamine-
containing preparations at some point during pregnancy (Peckham and King,
1963; Forfar and Nelson, 1973; Hill, 1973; Doering et al., 1978;
Brocklebank et al., 1978; Rayburn et al., 1982; Piper et al., 1987). Among
antihistaminic agents, diphenhydramine appears to be taken by -23% of
pregnant women (Piper et al., 1987). The predominant therapeutic uses of
antihistamines during pregnancy are in the treatment of allergies, nausea
and vomiting, and as cough and cold remedies. Antihistamine therapy
(diphenhydramine or chlorpheniramine) is effective in alleviating symptoms
of pregnancy-related urticarial conditions (Read, 1977; Ahmed and Kaplan, 11
1981). Nausea and vomiting seem to be relatively common problems in early pregnancy. The slightly higher consumption of antihistamines in the first trimester, as opposed to later, is probably due to the use of the drugs in antinausea and antiemetic therapy (Forfar and Nelson, 1973). The common cold appears to be the most frequent respiratory complaint in pregnancy, with an overall incidence of 17% (Peckham and King, 1963). In addition,
nasal congestion resulting from capillary engorgement of the respiratory
tract is common (Pratt, 1981) and would exacerbate the symptoms resulting
from allergic rhinitis and colds. Thus, antihistamines appear to be taken
at all stages of pregnancy. There are limited data on the duration of
antihistamine therapy in pregnancy, although Forfar and Nelson (1973)
report a mean duration of 15 days.
1.4.2. Placental Transfer of Antihistamines
Drugs cross the placenta largely by simple diffusion, and for
lipid soluble molecules at least, there is no significant barrier to the
transfer of compounds with a molecular weight less than -600 (Reynolds,
1979). Most antihistamines have a considerably small molecular weight
(e.g., diphenhydramine m.w.= 255) and, therefore, should readily cross the
placenta. However, there is very little information on the rate and/or
extent of placental transfer of the antihistamines. Brief mention has been made of maternal-fetal transfer of promethazine in humans, but there
appears to be no published quantitative data (Burt, 1971). Placental
transfer of 35S-promethazine seems to occur in the mouse (Hansson and
Schmiterlow, 1961). The placental transfer of cimetidine (Mihaly et al.,
1983; Ching et al., 1985) and ranitidine (Mihaly et al., 1982) has been 12 reported in sheep. Rapid placental transfer occurs but the fetal drug level is considerably lower than that in maternal plasma for both drugs following maternal drug administration. As discussed in the following section, other indirect evidence of placental transfer exists in the form of apparent fetal effects of some antihistamines following administration to the mother.
1.4.3. Safety of Antihistamines in Pregnancy
There is little information on harmful effects of antihistamines in the embryo and fetus, either in terms of the potential of the drugs to interfere with embryogenesis in the first trimester and thus yield teratogenic effects, or the potential to affect fetal functions later in pregnancy. lams and Rayburn (1982) have reported that there are no known harmful effects of antihistamines at any stage of pregnancy. In the case of diphenhydramine, there is some disagreement as to whether the drug elicits harmful effects in the fetus. No teratogenic effects of diphenhydramine have been reported in rats and rabbits (King et al., 1965;
McColl et al., 1967; Schardein et al., 1971). However, other studies reported harmful effects of the drug, such as reduced litter size, fetal body weight, retardation in the development of body organs and alterations in palate development in mice (Naranzo and Naranzo, 1968; Jones-Price et al., 1983). The main source of human data comes from the Collaborative
Perinatal Project (Heinonen et al., 1977), which examined teratogenic drug effects in 50,282 mother-child pairs. Of the antihistamines specifically examined (chlorpheniramine, pheniramine, diphenhydramine, methapyrilene, tripelennamine and bromopheniramine), only bromopheniramine was associated 13 with a higher than expected incidence of fetal anomalies. However, in a retrospective Finnish study (Saxen, 1974) where 559 children with oral clefts were matched with a control group, diphenhydramine intake during the first trimester was greater in mothers of the cleft group. Parkin (1974) gives clinical details of an infant whose mother had taken 150 mg of diphenhydramine hydrochloride daily during pregnancy for treatment of a pruritic rash. On the fifth postnatal day, the infant developed generalized tremulousness and diarrhea, which persisted for several days.
Subsequently, the drug was shown to be present in the blood of the infant on the fifth day. These features are consistent with diphenhydramine withdrawal symptoms.
1.4.4. Effects of Drugs on the Fetal Central Nervous System
Although there is very limited information on the central nervous system (CNS) effects of antihistamines in the fetus, there are numerous reports on the effects of other centrally active agents (Hutchings, 1978;
Dailey et al., 1982). The developing CNS seems particularly vulnerable to toxic damage, partly because of its extended period of development.
Depending on the duration of exposure, dose and toxic potential of the compound in question, the resulting CNS damage can range from gross morphologic anomalies to subtle alterations in CNS function (Hutchings,
1978). These latter anomalies are termed behavioural teratogenic effects and involve deficits in CNS activities such as motor function, learning and memory. Major teratogenic CNS lesions occur mainly during the period of organogenesis, but functional impairment can occur later in pregnancy, particularly during the period of the brain growth spurt (Dobbing, 1976). 14
This period of CNS maturation spans the perinatal period, occurring at different stages of development in different species. In humans, it begins at about midgestation and continues perhaps into the third year of life
(Dobbing, 1976).
The majority of work in behavioural teratology has been carried out with rodent models (Coyle et al., 1980; Kimmel, 1981). A wide variety of behavioural parameters, such as activity, motor coordination, sensory function and learning and memory abilities have been examined in the neonatal, adolescent and adult stages of development in relation to prenatal drug exposure. Two areas of behavioural teratology that have received less attention than the postnatal evaluation of neurologic function are fetal CNS function and behaviour during the actual period of drug exposure, and the relationship between fetal levels of the drug and the observed effects (Kimmel, 1981). Such studies employing small animal models are technically difficult, if not impossible. However, with the development of large animal models, particularly those using pregnant sheep, a large variety of physiological, endocrine and behavioural parameters can be monitored from the undisturbed fetus in utero over the last third of gestation. It is also possible to carry out detailed pharmacokinetic analysis of placental drug transfer and maternal-fetal drug disposition (Szeto et al., 1982a). A number of fetal neurological and behavioural parameters have been recorded in these preparations, including electrocortical activity, eye movements, electromyographic activity in limb, neck and trunk muscles, limb and body movements and breathing activity (Dawes et al., 1972; Natale et al., 1981; Rigatto et a/., 1982).
These various studies have indicated that fetal behavioural states become 15
recognizable by about 0.8 of gestation (term, 145 days), at least as reflected by the appearance of alternating periods of high and low voltage electrocortical activity. The high voltage pattern is similar to quiet sleep after birth, while the bulk of the low voltage activity appears analogous to rapid eye movement (REM) sleep. Also, the fetus may be awake for a brief period of time (Rigatto et al., 1982), although this is disputed (Harding, 1984). Linked to the cycle of electrocortical activity is the episodic occurrence of breathing-like activity, which normally is present only during the low voltage ECoG pattern (Jansen and Chernick,
1983). Limb movements also exhibit an element of periodicity as they occur in high voltage electrocorticographic (ECoG) states or in the transitions between high and low voltage states (Natale et al., 1981).
A number of studies have demonstrated profound effects of centrally active drugs on ECoG and breathing activity in fetal lambs
(Piercy et al., 1977; Jansen et al., 1983; Dawes, 1984; Umans and Szeto,
1985). CNS depressants (e.g., pentothal, meperidine, ethanol) administered to the ewe in doses that have minimal maternal effects suppress fetal breathing activity and low voltage ECoG activity (Boddy et al., 1976;
Ruckebusch et al., 1976; Piercy et al., 1977; Patrick et al., 1985). In contrast, various CNS stimulants result in cortical arousal (low voltage
ECoG activity) and increased fetal breathing activity (Jansen et al., 1983;
Szeto, 1983). Indomethacin and 5-hydroxytryptophan disrupt the normal link between fetal breathing and REM sleep (Kitterman et al., 1979; Quilligan et al., 1981). However, only in a few of these studies, namely those involving methadone and ethanol, were the measurements of fetal behaviour accompanied by a pharmacokinetic analysis of maternal-fetal drug 16 disposition (Szeto et al., 1982a; McLeod et al., 1983; Szeto, 1983; Patrick et al., 1985).
Recent studies with human fetuses, which have employed noninvasive monitoring techniques, particularly real-time ultrasound viewing, have indicated that behavioural states similar to those in the fetal lamb also occur in the human perinatally (Nijhuis et al., 1982). Centrally active drugs can have profound acute effects on fetal CNS function and behaviour in humans (Dailey et al., 1982; Jansen et al., 1983; Szeto, 1983).
However, the postnatal effects of drug-induced prenatal behavioural alterations are not yet known, although Szeto (1983) has suggested that the late postnatal effects of prenatal methadone exposure may be in part, long- term consequences of the direct action of methadone on fetal ECoG activity and cerebral metabolism. It is also possible that disruptions of the normal pattern of fetal respiratory activity could have long-term effects on the respiratory system, since total or partial reduction in breathing activity by phrenic nerve section or experimental growth retardation results in reduced lung weight at birth (Fewell et al., 1981; Malony et al., 1982).
1.5. Rationale and Specific Aims of the Research
1.5.1. Rationale
In the preceeding sections, a case has been made for the following points: 1) there is a relatively common usage of antihistamines during pregnancy, 2) classical histamine Hj-receptor antagonists probably cross 17 the placenta, and may exert depressant or sedative effects in the fetus, 3) there is a shortage of data on the extent of placental transfer and fetal effects of all antihistamines. In regard to the last point, there are serious technical and ethical difficulties in obtaining such data from pregnant humans and, for technical reasons, it is impractical to obtain detailed pharmacokinetic and physiological information from small animal models for pregnancy. Large animal models would seem to be more appropriate, and of these, the pregnant sheep is the most commonly employed. There are clear differences in placental structure and permeability between sheep and humans, but for lipid soluble molecules, the differences are of less importance than for hydrophilic substances. The fetal lamb is more mature at birth than is the human fetus, but in both species there appear to be similar behavioural states in the fetus towards the end of gestation. Hence, the pregnant sheep would seem to be an acceptable model to use in the proposed study, and would provide data that should be useful in assessing the safety of the antihistamine drug use in pregnancy
Diphenhydramine is typical of most histamine Hj-receptor antagonists in that it possesses anticholinergic, antiemetic, antitussive and sedative properties (Douglas, 1985). This compound has been frequently used during pregnancy for the treatment of various allergic disorders
(e.g., allergic rhinitis) and problems specific to pregnancy, such as nausea and vomiting in the first trimester (Forfar and Nelson, 1973) and a pregnancy-related urticarial condition in the third trimester (Ahmed and
Kaplan, 1981). Diphenhydramine is a weakly basic amine compound (pKa =
9.0) and possesses high 1ipophilicity, low molecular weight (255.4) and CNS 18 activity. Therefore, it seems probable that this compound would cross the placenta and cause CNS effects in the fetus.
1.5.2. Specific Aims
The primary objectives of the present investigation are:
1) to develop a capillary gas-liquid chromatographic method
(GC/NPD), with improved sensitivity and selectivity, suitable for studies where biological sample volumes are limited.
2) to examine dose dependency of diphenhydramine pharmacokinetics in nonpregnant sheep.
3) to study the placental transfer and disposition of diphenhydramine in the maternal-fetal sheep following i.v. bolus injection of the drug to the pregnant ewe.
4) to study transplacental and nonplacental clearances of diphenhydramine following separate maternal and fetal drug infusions to steady-state plasma drug concentration.
5) to study the effect of diphenhydramine administration on fetal behavioural states, i.e., electrocortical and electro-ocular activity and breathing movements, and cardiorespiratory variables, i.e., heart rate, blood pressure and blood gas tensions in relation to fetal drug exposure in the latter third of gestation in sheep.
6) to examine the possibility of the recirculation of diphenhydramine within the maternal-fetal unit after bolus injection of the drug into the amniotic cavity in the pregnant sheep.
7) to study the pulmonary extraction of diphenhydramine in the fetal lamb following maternal or fetal drug infusion to steady-state. 19
2. EXPERIMENTAL
2.1. Development of a Capillary GC/NPD Assay Method for Diphenhydramine
2.1.1. Materials
Diphenhydramine hydrochloride, 2-(diphenylmethoxy)-N,N- dimethylethylamine (Parke Davis, Lot No. C429419, Montreal, Quebec) and orphenadrine hydrochloride, N,N-dimethyl-2-[(2-methylphenyl) phenyl- methoxy] ethanamine (Pfaltz and Bauer, Lot No. D352, Stanford, CT) were used as reference standards. Diphenhydramine hydrochloride for injection (Benadryl^ Injectable, Parke Davis, Lot No. AD 613-1,
Brockville, Ontario) was obtained from the Hospital Pharmacy, Health
Sciences Center Hospital, Vancouver, British Columbia. Triethylamine
(TEA) was purchased from Pierce Chemical Co., Rockford, IL. Toluene, dichloromethane, benzene, heptane and hexane, distilled in glass, were purchased from Caledon Laboratories (Georgetown, Ontario) and isoamyl alcohol from Allied Chemical Co. (Morristown, NJ). A solution of 1 M sodium hydroxide was prepared from ACS reagent grade chemical (American
Scientific and Chemical, Seattle, WA). Deionized, distilled water was used in the preparation of stock solution throughout the analysis.
2.1.2. Preparation of Stock Solutions
Diphenhydramine hydrochloride (160 ng/mL, equivalent to base) and the internal standard, orphenadrine hydrochloride (1.0 fig/ml, equivalent to base) were prepared by dissolving these compounds in 20 distilled water. The solutions were stored at 4°C for up to two months.
2.1.3. Instrumentation and Optimization of Gas Chromatographic Conditions
2.1.3.1. Equipment
A Model 5830A Hewlett-Packard (HP, Avondale, PA) gas
chromatograph equipped with a nitrogen/phosphorus-selective detector, a
Model 18835B HP capillary inlet system, a Model 18850 HP integrator
system for peak area integration and quantitation and a 25 m x 0.31 mm
I.D. cross-linked fused-silica capillary column (5% phenylmethyl
silicone, film thickness 0.52 |zm, Hewlett Packard) was used for all
analyses. Thermogreen^ LB-2 silicone rubber septa (Supelco, Bellafonte,
PA) were used to provide low bleed at high inlet temperature. A Model
7671A HP automatic sampler was used for sample injection into the gas
chromatograph in the splitless injection mode.
2.1.3.2. Optimization of Gas Chromatographic Conditions
Samples for the preliminary studies of optimizing GC conditions
were prepared as follows. To 2.0 mL of diphenhydramine stock solution
(160 ng/mL) were added 0.2 mL of the internal standard orphenadrine
solution (1 Mg/mL) and 0.5 mL of 1 M NaOH. To this mixture, 6 mL of
toluene was added and the aqueous phase was extracted by shaking the
mixture for 20 min on a rotary shaker (Labquake Tube Shaker, Model 415-
110, Lab Industries, Berkeley, CA). The sample was centrifuged and then
the aqueous layer was aspirated. The remaining organic layer was 21 transferred to a culture tube and dried under a gentle stream of nitrogen in a 35°C water bath. The residue was reconstituted to a volume of 100 nl with toluene. Aliquots of 2 /zL were injected into the gas chromatograph. Samples were injected in splitless injection mode under GC conditions: injection port temperature, 250°C; oven temperature, 200°C; and detection port temperature, 300°C. Under these
GC conditions, the following instrument parameters were varied and tested for the nitrogen-phosphorus detector response.
1) hydrogen gas (make-up gas) flow: 2.0, 2.5, 2.75, 3.0, 3.25, 3.5,
3.75, 4.0 and 4.25 mL/min at constant rates of the helium carrier
gas (1.0 mL/min) and air (60 mL/min).
2) air (make-up gas) flow: 30, 40, 45, 50, 60 and 80 mL/min at
constant rates of the helium carrier gas (1.0 mL/min) and hydrogen
(3.0 mL/min).
3) column pressure: 8, 10, 13, 15, 18, 20 and 25 psi at constant flow
rates of the hydrogen gas (3.0 mL/min) and air (50 mL/min).
4) purge activation time: 0.1, 0.2, 0.6, 1.0, 1.4 and 1.6 min at
constant flow rates of hydrogen gas (3.0 mL/min), air (50 mL/min)
and helium gas (1.0 mL/min).
The efficiency of a GC column is expressed by the number of theoretical plates (n):
2 n = 16(tR/Wb) (eq. 1)
where tR is the retention time of the peak and Wh is the peak width at base (Rooney, 1981). The theoretical plate is a hypothetical unit of the column in which one equilibrium of the solute occurs between the carrier gas and liquid phase of the column. The hypothetical unit expressed in terms of column length is called the height equivalent to theoretical plate (H.E.T.P.) which can be calculated by dividing the column length by the plate number. In the present study, the number of theoretical plates was calculated for diphenhydramine at column pressures, 7, 9, 11, 13, 15 and 17 psi, and then the van Deemter curve showing the relationship between H.E.T.P. and linear velocity of the helium carrier gas was plotted for this compound.
2.1.4. Drug Extraction Studies
The following extraction parameters were examined.
1) drug extraction efficiency of solvents: heptane, hexane, toluene,
benzene and dichloromethane.
2) effect of the addition of triethylamine to dichloromethane on its
drug extraction efficiency: 0, 0.01, 0.005, 0.1 and 0.5 M
triethylamine.
3) extraction time: 5, 10, 15, 20, 30, 40, 60 and 80 min.
2.1.5. Routine Procedures for Diphenhydramine Analysis
2.1.5.1. Operating GC Conditions
The operating GC conditions for routine sample analysis were: injection port temperature, 250°C; initial oven temperature, 190°C; programming rate, 2°C/min at 1 min and 5°C/min at 8.5 min; final oven temperature, 240°C; nitrogen/phosphorus-selective detector temperature,
300°C; carrier gas (helium) flow rate, 1.0 mL/min; septum purge flow 23
rate, 0.6 mL/min; make-up gas flow rate, 2 mL/min; hydrogen/air flow rate ratio, 3:30.
2.1.5.2. Drug Extraction from Biological Samples
To 0.3-0.5 mL of plasma were added 0.2 mL of internal standard
(1.0 jug/mL) and 0.5 mL of 1 M sodium hydroxide in a 15 mL disposable culture tube with a screw cap lined with polytetrafluoroethylene (PTFE)
(Corning Glass Works, Corning, NY). The aqueous phase was adjusted to a total volume of 2.5 mL (pH -13) with distilled water and mixed on a vortex mixer. To this mixture, 6.0 mL of 0.05 M triethylamine in dichloromethane were added and the aqueous phase was extracted by shaking the mixture for 20 min on a rotary shaker (Labquake Tube Shaker,
Model 415-110, Lab Industries, Berkeley, CA). The samples were placed in a freezer (-20°C) for 5 min and then centrifuged at 2,300 g for 10 min. After centrifugation, the aqueous layer was aspirated and discarded. The remaining organic layer was transferred to a clean 15 mL culture tube and dried under a gentle stream of nitrogen in a 30°C water bath. The residue was reconstituted to a volume of 100 /iL with toluene.
The sample was mixed on a vortex mixer for 10 sec. Aliquots of 2 /zL were used for GC/NPD analysis.
2.1.5.3. Preparation of the Calibration Curve
A 0.5 mL sample of blank sheep plasma was spiked with varying amounts of diphenhydramine (2, 4, 8, 16, 32, 64, 128, 192, 256 and 320 ng) using the prepared stock solution (160 ng/mL) and diluted stock 24
solution (20 ng/mL), and then 0.2 mL of the orphenadrine solution (1.0
/jg/mL) and 0.5 mL of 1 M sodium hydroxide were added. The aqueous phase was adjusted to a total volume of 2.5 mL (pH -13) with distilled water
and the samples were extracted as described above. Quantitative
estimation of diphenhydramine in plasma was accomplished by plotting the
area ratios of diphenhydramine and orphenadrine against the range of
indicated diphenhydramine concentrations. For recovery study, diphenhydramine was dissolved in methanol to provide concentrations
ranging from 2 to 320 ng/mL. The amount of diphenhydramine extracted
from plasma by dichloromethane was calculated using calibration curves
obtained from direct injection of the prepared methanolic solutions.
2.1.5.4. Gas Chromatography-Mass Spectrometry
Capillary GC electron-impact mass spectrometry (EI-MS) and
chemical ionization mass spectrometry (CI-MS) were carried out using a
Model 5987A HP gas chromatograph-mass spectrometer equipped with a 25 m
x 0.32 mm I.D. (0.25 nm film thickness) 0V-1701 fused-silica column
(Quadrex Scientific, New Haven, CT). Methane was used as the ionizing
gas and the splitless injection mode was employed in the study. The
following conditions were used for GC-MS of diphenhydramine and
orphenadrine: initial oven temperature, 50°C; time 1, 0 min; rate,
30°C/min; final oven temperature, 260°C; time 2, 10 min; injection port
temperature, 240°C; carrier gas (helium) flow rate, 1.0 mL/min; electron
multiplier voltage, 2200 V; emission current, 300 /xA. 25
2.1.5.5. Application of the Developed Assay Method to a Placental Transfer Study
A preliminary experiment on the placental transfer of diphenhydramine was carried out in a pregnant ewe chronically implanted with maternal and fetal vascular catheters (Rurak and Gruber, 1983). A
100 mg i.v. dose of diphenhydramine hydrochloride was administered over
a 30 sec period via a maternal femoral venous catheter and the catheter
flushed with 5.0 mL of heparinized saline. Blood samples for
diphenhydramine determination were simultaneously withdrawn from the maternal and fetal femoral arterial catheters at -5, 5, 10, 15, 20, 30,
45, 60, 90, 120, 150, 180, 210, 240, 300 and 360 min. The blood samples
were transferred to heparinized Vacutainer^ collection tubes (Becton-
Dickinson, Mississauga, Ontario) and immediately centrifuged. Following
centrifugation, the plasma was transferred to a 15 mL PTFE-lined screw-
capped disposable culture tube, and stored at -20°C until extracted and
analyzed.
2.2. Pharmacokinetics of Diphenhydramine after Dose Ranging in Nonpregnant Sheep
2.2.1. Animals and Surgical Preparation
Six nonpregnant Dorset or Suffolk ewes of 2-3 years of age,
with a mean body weight of 90 + 16 (S.D.) kg, were used in the study.
The animals were from the same flock (Department of Animal Science,
University of British Columbia) and were raised and fed under the same
conditions. Surgery was performed aseptically using 1-2% halothane 26
(Ayerst Laboratories, Montreal, Quebec) and 60% nitrous oxide
anesthesia, following induction of anaesthesia with 1 g of i.v. sodium pentothal (Abbott Laboratories, Chicago, IL). Silicone rubber catheters, with an O.D. of 2.2 mm (Dow Corning, Midland, MI), were
implanted into the femoral artery and vein. Ampicillin (500 mg, Ayerst
Laboratories, Montreal, Quebec) was given to the ewe i.m. daily for 3 days after surgery. The catheters were flushed daily with 3 mL of heparinized isotonic saline (12 IU/mL). After surgery, the animals were
allowed to recover for at least 3 days prior to experimentation.
2.2.2. Drug Administration and Experimental Protocol
Each animal received 25, 50, 100 and 200 mg doses of diphenhydramine hydrochloride (Benadryl1* Injectable, 50 mg/mL, Parke
Davis, Lot No. 01565-1, Scarborough, Ontario) on a randomized crossover
basis, with each experiment separated by a minimum of a 2 day period for
drug washout. Drug was diluted in isotonic saline to 5 mL and injected
as an i.v. bolus dose via the venous catheter over 30 sec and the
catheter subsequently flushed with 5 mL of isotonic saline. Arterial
blood samples (3.0 mL) were taken via the femoral arterial catheter at
-5, 5, 10, 15, 20, 30, 45, 60, 90, 120, 150, 180, 210, 240, 300 and 360 min following drug administration. Blood samples were transferred to
VacutainerR collection tubes and centrifuged immediately at 3,500 g for
10 min. The plasma was transferred to glass culture tubes and frozen at
-20°C and stored until analysis. Arterial blood samples (0.6 mL) were
collected at -5, 5, 10 and 60 min for the measurement of Po2, Pco2, 02
content and hematocrit. Arterial Po2, Pco2 and pH were measured using 27
IL micro 13 blood gas analyzer (Instrumentation Laboratories, Lexington,
MA) and corrected to 39°C. Blood oxygen content was measured by Lex-C^- con-k oxygen content analyzer (Lexington Instrument, Watham, MA) and hematocrit by a micro-capillary centrifuge (International Equipment Co.,
Needham, MA). Plasma diphenhydramine concentrations were measured by a capillary gas chromatographic method employing nitrogen-phosphorus detection (see section 2.1.5).
2.2.3. Pharmacokinetic Analysis
Plasma drug concentration vs. time data were used for curve
fitting by the computer program AUTOAN (Sedman and Wagner, 1976) and the obtained initial estimates of the pharmacokinetic parameters were
subsequently analyzed using the nonlinear least squares computer program
NONLIN (Metzler et al., 1974). The initial distribution half-life (t%,
lambda^) and the terminal elimination half-life {tk, lambdan) were
calculated by the following equations:
tk, lambda! = 0.693/lambdaj (eq. 2)
tk, lambdan = 0.693/1ambdan (eq. 3)
where lambdaj and lambdan are the largest and the smallest disposition
rate constants obtained from multiexponential plasma drug concentration
vs. time decay curves, respectively (Gibaldi and Perrier, 1982). The
area under the plasma drug concentration vs. time curves, AUCQ, was
calculated using the equation:
AUCQ = AUCQ + AUC* (eq. 4) where t represents the time when the last sample was taken. The first
term, AUCQ was calculated using the trapezoidal rule, and the second 28
term was calculated using the equation:
AUC^ = Ct/lambdan (eq. 5) where C^. is the concentration of the last sample. The AUC obtained from 00
0 to t was >95% of the total AUCQ. Therefore, the possible error in 00
AUCt arising from an estimation of lambdan is minimal. The total body
clearance (CLjg) was obtained from:
C/_TB= Dose/AUCg (eq. 6)
The apparent volume of distribution (ft/area) was obtained from:
= ^area Dose/(lambdan-AUCo) (eq. 7)
The volume of distribution at steady-state [Vdss) was calculated
(Gibaldi and Perrier, 1982) either by an equation which requires curve
fitting:
2 Vdss = Dose'l^C^lambdayVt i^Cz/lambda,-)] (eq. 8)
or by a model-independent method:
2 Vdss = (Dose-AUMCo)/(AUCo) (eq. 9)
oo
where AUMC0 is the total area under the first moment of the plasma
concentration vs. time curve from time 0 to infinity, calculated by the
trapezoidal rule with a correction made for extrapolation to infinity
beyond the last data point:
2 AUMCQ = AUMCg + Ct't/lambdan + Ct/lambdan (eq. 10)
2.2.4. Plasma Protein Binding
Diphenhydramine hydrochloride (Sigma Chem. Co., Lot No. D-3630,
St. Louis, MO) was dissolved in Sorensen's isotonic phosphate buffer
(0.067 M, pH = 7.4) to prepare 10, 25, 50, 100, 250, 500, 1,000, 2,000,
4,000 and 10,000 ng/mL drug solutions. Optimum equilibrium time was 29 determined at the two drug concentrations, 500 and 10,000 ng/mL, by dialyzing the drug-containing buffer solutions against an equivalent volume of blank sheep plasma and by analyzing drug concentration in the plasma compartment at 0.5, 1.0, 2.0, 3.0, 4.0 and 6.0 hr. An optimum equilibrium time of 3 hr was determined and applied for further binding
studies. Dialysis was carried out at 39°C in Plexiglass'* dialysis cells
in which buffer and plasma were separated by cellophane dialysis membrane (Sigma Chem. Co., Lot No. 39C-6131, molecular weight cutoff =
12,000 Daltons, St. Louis, MO). The dialysis membrane was boiled in
distilled water for 1 hr and soaked in isotonic phosphate buffer for 1
hr before it was mounted within the dialysis cell. At the end of the
dialysis, both buffer and plasma were transferred to glass tubes and
kept for drug analysis. The possibility of leakage of protein across
the membrane was tested by adding a drop of the dialysis buffer to 0.1
mL of 3% trichloroacetic acid. The absence of turbidity was an
indicator of membrane integrity. The free fraction of diphenhydramine
was determined in vitro by dialyzing 1.0 mL of the prepared drug
solutions (10 to 10,000 ng/mL), against equivalent volumes of blank
plasma from surgically unaltered nonpregnant sheep. The free fraction
of diphenhydramine was also measured in vivo in the 6 nonpregnant
animals, by analyzing drug concentrations in the 5 and 60 min plasma
samples taken from experiments involving the administration of the 200 mg dose. In two animals, the free fraction of diphenhydramine was
determined in vivo in all the samples collected during the experiments
following administration of the 200 mg dose. The free fraction of the drug was calculated by the equation:
free fraction = Cu/Cp (eq. 11) 30
where Cu and Cp are buffer and plasma concentrations of diphenhydramine after dialysis, respectively.
2.2.5. Statistical Analysis
Statistical analysis was performed by One-way Analysis of
Variance and Tukey multiple comparison test. The significance level was p<0.05. Values are given as the mean + 1 S.D.
2.3. Pharmacokinetics of Diphenhydramine in the Pregnant Ewe and Fetal Lamb: i.v. Bolus Injection Studies
2.3.1. Animals and Surgical Preparation
Four pregnant Dorset ewes were surgically prepared for
experimentation at 122-129 days of gestation. All animals were fasted
overnight prior to surgery. Preoperatively, 3 mg of atropine sulfate
(Glaxo Laboratories, Montreal, Quebec) was administered to the ewe via
the jugular vein 5-10 min before induction of anaesthesia with the i.v.
injection of 1 g of thiopental sodium (Abbott Laboratories, Chicago,
IL). Following intubation, anaesthesia was maintained with halothane
(1.5-2.0%) (Ayerst Laboratories, Montreal, Quebec) and nitrous oxide
(60%) in oxygen.
Aseptic techniques were employed throughout the surgical
procedure. The maternal abdomen was opened by a midline incision, and
access to the fetus was gained through a small incision in a region of 31 the uterus that was free from major blood vessels. Silicone rubber catheters (Dow Corning, Midland, MI) were implanted into the fetal femoral artery, lateral tarsal vein, trachea, and amniotic cavity. The tracheal catheter (2.2 mm O.D.), which was inserted through a small incision 1-2 cm below the larynx and advanced 4 to 5 cm, did not obstruct lung fluid efflux from the airway into the pharynx. Following catheterization of the fetus, the uterine incisions were closed in two layers. Catheters were then implanted in a maternal femoral artery and vein. All catheters were filled with heparinized saline (12 IU/mL) and tunneled subcutaneously and exteriorized through a small incision in the flank of the ewe, where they were stored in a clean denim pouch. The maternal abdominal incision was closed in layers. Amniotic fluid lost during surgery was replaced with irrigation saline (Travenol Canada
Inc., Mississauga, Ontario). Immediately following surgery, 500 mg
Ampicillin (Ayerst Laboratories, Montreal, Quebec) was given to the fetus via the tarsal venous catheter, into the amniotic cavity via the amniotic catheter and to the ewe intramuscularly. Ampicillin was given to the ewe i.m. for the first 3 days after surgery and daily into the amniotic cavity for the duration of the preparation, until delivery.
Following surgery, the ewes were kept in pens in the company of other sheep and were given free access to food and water. Animals were allowed to recover for a minimum of three days before experimentation.
2.3.2 Drug Administration and Experimental Protocol
For each experiment, the ewe was placed in a monitoring cage adjacent to the holding pen in full view of companion ewes and with free 32 access to food and water. A 100 mg dose of diphenhydramine hydrochloride (BenadrylR Injectable, Parke Davis, Lot No. AD 613-1,
Brockville, Ontario) was injected in isotonic saline via the maternal femoral venous catheter over a period of 30 sec, and the catheter was flushed with 5 mL of heparinized saline. Maternal (3.0 mL) and fetal
(1.5 mL) arterial blood samples were withdrawn simultaneously at -5, 5,
10, 15, 20, 30, 45, 60, 90, 120, 150, 180, 210, 240, 300 and 360 min after drug administration. At intervals during the experiment, the fetal blood collected was replaced with an equal volume of blood from the ewe. Amniotic and tracheal fluid samples (1.5 mL each) were also collected simultaneously together with the maternal and fetal blood samples. The blood samples were collected in heparinized collection tubes and immediately centrifuged at 3,500 g for 10 min. The plasma was transferred to a 15 mL disposable glass culture tube with a screw cap lined with polytetrafluoroethylene and frozen at -20° until the time of extraction and assay. Diphenhydramine concentrations in biological fluids were determined using the nitrogen/phosphorus detection-capillary gas chromatographic method (see section 2.1.5).
2.3.3. Pharmacokinetic Analysis
Initial estimates of the pharmacokinetic parameters after i.v. administration were obtained by computer analysis using the program
AUT0AN (Sedman and Wagner, 1976). Using the initial estimates provided by AUT0AN, the data was then fitted using the program NONLIN (Metzler et al., 1974). All data points were weighted using the reciprocal of drug concentration. Maternal plasma concentration vs. time curves were 33 fitted using the biexponential equation described as follows:
at - Cp = Ae" + Be ^ (eq. 12) where Cp is the plasma concentration at time t, a and 0 are the apparent distributional and terminal elimination rate constants, and A and B are intercepts of the a and /} phases, respectively. The distributional and
terminal elimination half-lives, ti n and ti, a, were calculated:
tj, a = 0.693/a (eq. 13)
t% ^ = 0.693/0 (eq. 14)
oo
The area under the plasma drug concentration vs. time curves, AUC0, was calculated using the equations 4 and 5. The total body clearance (C/.jg) was obtained by the equation 6. The apparent volume of distribution was calculated by the equations 7 and 9. The percentage of the maternally administered diphenhydramine that is eliminated by the fetus via the nonplacental pathway was calculated by the following equation:
Percentage of administered dose that is eliminated by the fetus
= AUCFo'CLfo-100/(AUCMo-CLmo + AUCFo'CLfo) (eq. 15)
CD 00
where AUCp0 and AUC^0 are the total area under the plasma drug concentration vs. time curves in the fetus and ewe, respectively, and
CLf0 and CLmo are the nonplacental clearances of diphenhydramine from the fetal and maternal compartments, respectively. Mean values (n = 8)
of the CLmo (3343.8 mL/min) and CLfo (208.4 mL/min) for diphenhydramine were obtained from the steady-state drug infusion experiments (Section
3.4.1.4) and used in the calculation. Values in the text and tables are given as the mean ± 1 S.D. 34
2.4 Transplacental Clearance, Nonplacental Clearance and Fetal Effects of Diphenhydramine in Pregnant Sheep: Maternal and Fetal Infusions to Steady-State
2.4.1. Animals and Surgical Preparation
Studies were performed on 12 time-dated pregnant sheep (Dorset and Suffolk breeds). At 119-129 days gestation (term 145 days), surgery was performed aseptically in the animals under halothane (1-2%) and
nitrous oxide (60%) in 02 anaesthesia, following induction of anaesthesia with i.v. sodium pentothal (1 g) and intubation of the ewe.
Access to the fetus was gained through a midline abdominal incision in the ewe and an incision through the uterine wall free from placentomes and major blood vessels. Silicone rubber catheters (Dow
Corning, Midland, MI) were implanted in a fetal femoral artery and lateral tarsal vein, in the fetal trachea and in the amniotic cavity.
In one animal (ewe no. 130), a catheter was also implanted in the common umbilical vein (Rurak and Gruber, 1983). Electrodes of TeflonR-coated stainless-steel wire (Cooper Corp., Chatsworth, CA) were implanted biparietally on the dura to record the fetal electroencephalogram and through the orbital ridge of the zygomatic bone of each eye for electro- ocular recordings. Catheters were also implanted in a maternal femoral artery and vein. All catheters and electrodes were tunneled subcutaneously in the maternal abdominal wall to exit through a small incision in the flank. They were stored in a denim pouch in the ewe's flank when not in use. All vascular catheters were flushed daily with 2 mL of sterile 0.9% NaCl solution containing 12 IU of heparin per mL. 35
Antibiotic prophylaxis was as described in section 2.3.1.
After surgery, the animals were kept in holding pens with other sheep and given free access to food and water. The sheep were allowed to recover for 3-14 days before experimentation. To carry out an experiment, the ewe was transferred to a monitoring cart adjacent to and in full view of companion sheep.
2.4.2. Drug Administration and Experimental Protocol
A total of 20 experiments were carried out on the 12 animals at
126 to 142 days gestation (Table 11). Sixteen of these were carried out on eight of the sheep and involved separate maternal and fetal i.v. infusions of diphenhydramine, given 48 hr apart. The remaining four experiments involved control i.v. infusions of physiological saline to the fetus (n = 3) or ewe (n = 1).
Infusions of drug and saline were carried out using a Harvard pump (Model 944, Harvard Apparatus, Mill is, MA); the volume infusion rate was 0.17 mL/min for 90 min. For drug infusions, diphenhydramine hydrochloride (BenadrylR Injectable, Parke Davis, Lot No. 01565-1,
Scarborough, Ontario) was diluted to the appropriate concentration with
0.9% saline immediately before administration. For maternal infusions, the drug was given as a 20 mg loading dose, followed by a continuous
infusion rate of 0.67 mg/min. The drug was administered to the fetus as a 5 mg loading dose, followed by a continuous infusion rate of 0.17 mg/min. 36
During the drug infusion experiments, paired maternal (2.5 mL) and fetal (1.5 mL) arterial blood, fetal tracheal and amniotic fluids
(1.5 mL each) samples were collected at -5, 5, 15, 30, 45, 60, 75, 90,
95, 100, 105, 110, 120, 135, 150, 180, 210, 240, 270, 300, 330 and 390 min for the estimation of diphenhydramine concentration. In one animal
(ewe no. 130), umbilical venous blood in addition to the maternal and fetal arterial blood was sampled during maternal and fetal infusions.
Samples for maternal and fetal blood gas and pH measurements were collected at -5, 5, 30, 60, 90 and 150 min in both the drug and saline infusion experiments. The total volume of fetal blood removed in the drug infusion experiments was 38 mL, whereas in the control studies, the volume was 5 mL. In all experiments, the volume of fetal blood collected was replaced by an equal volume of maternal blood in three divided and equal doses, two during the experiment (after 45 and 120 min samples) and one at the end. Plasma samples for diphenhydramine analysis were stored at -20°C until analysis.
2.4.3. Pharmacokinetic Analysis
A two-compartment open model was used to describe the disposition of the drug in the maternal-fetal sheep, with drug elimination occurring from both the maternal and fetal compartments
(Scheme 2). 37
CLMF M F
CLfM
• CL MO CL.F O M = MATERNAL COMPARTMENT
F = FETAL COMPARTMENT
Scheme 2. A two-compartment open model used in the calculation of transplacental and nonplacental clearances in the maternal fetal unit (Szeto et al., 1982a).
Transplacental and nonplacental clearances of diphenhydramine were calculated as described by Szeto et al. (1982a):
CL Cm mm - k0/[ ss " Cfss(Cm'ss/Cf'„)] (eq. 16)
CLff = k0'/[Cf'ss - Cm'ss(Cfss/Cmss)] (eq. 17)
CLmf = CLff(Cfss/Cmss) (eq. 18)
CLfm - CLmm(Cm'ss/Cf'ss) (eq. 19)
CLmo - clmm " CLmf CLf0 = CLff - CLfm (eq. 21) where Clmm and CLff are the total clearances from the mother and fetus, respectively, CLmf is the transplacental clearance from the maternal to fetal compartment, CLfm is the transplacental clearance from the fetal to maternal compartment, and CLmo and CLf0 are the nonplacental clearances from the mother and fetus, respectively. By using the values and tne for the infusion rates to the mother (k0) and fetus (kg') steady-state drug concentrations in the mother and fetus following maternal infusion (Cmss and Cfss) and fetal infusion (Cm'ss and Cfss), 38 all the clearances were calculated. Separately, the total body clearance from the mother and fetus, CLmm and CLff, were also calculated by the following equations (Gibaldi and Perrier, 1982): CLmm = k0/Cmss (eq. 22) CLff = k07Cf'ss (eq. 23) 2.4.4. Plasma Protein Binding Studies Protein binding of diphenhydramine in the maternal and fetal plasma was determined by equilibrium dialysis as described in the Section 2.2.4. Briefly, maternal and fetal plasma samples collected at 75 and 90 min from each animal were dialyzed against an equal volume (0.3 mL) of the phosphate buffer solution (Sorensen's isotonic buffer, 0.067 M, pH = 7.4) for 3 hr. Dialysis was carried out at 39°C for maternal plasma and 39.5°C for fetal plasma in Plexiglass'* dialysis cells in which buffer and plasma were separated by cellophane dialysis membrane (Sigma Chem. Co., Lot No. 39C-6131, molecular weight cutoff = 12,000 Daltons St. Louis, M0). The free fraction was calculated by equation 11. 2.4.5. Recording and Analytical Techniques For at least 90 min prior to and for at least 90 min after the infusion period, fetal arterial, amniotic and tracheal pressures, heart rate and electrocortical and electro-ocular activity were recorded continuously on a Beckman R612 recorder (Sensormedics Corp., Anaheim, CA). Fetal heart rate was obtained from the arterial pulse using a 39 cardie-tachometer (Model 9857, Sensormedics Corp., Anaheim, CA). Mean fetal arterial pressure was corrected for intra-uterine pressure, measured from the amniotic catheter, by subtraction. Episodes of fetal breathing activity were detected from the tracheal pressure trace. The minimum breath amplitude analyzed was 1 mm Hg and the minimum duration of breathing activity accepted as a breathing episode was 10 sec. Fetal electrocortical (ECoG) and electro-ocular (EoG) signals were recorded using a type 9806A AC/DC coupler (Sensormedics Corp., Anaheim, CA) with the low and high frequency cutoff frequencies set to approximately 0.5 and 50 Hz, respectively. Recordings of fetal ECoG activity were divided into periods of low, intermediate and high voltage amplitude, with voltage ranges of 25 to 50, 80 to 110, and 125 to 175 MV, respectively. The normal recording speed was 0.1 mm/sec, but occasionally higher speeds (5-10 mm/sec) were used to obtain qualitative information on the frequency characteristics of the ECoG activity. Episodes of rapid eye movements (REM) were detected from the electro-ocular activity trace. Sections of the records where analysis of the ECoG and EoG waveforms was impossible because of maternal or fetal movement were excluded from analysis. The total excluded was 11 min, or 0.2% of the total minutes analyzed. For the entire control, infusion and postinfusion periods, the percentage of time spent in high, low and intermediate ECoG activity and the percentage time with EoG activity was calculated. Maternal and fetal P02, PCO2 and pH were measured using an IL model 1306 automated blood gas analyzer (Instrumentation Laboratories, Lexington, MA) and corrected to temperatures of 39.0 and 39.5°C for the ewe and fetus, respectively. Blood oxygen content was measured by Lex- 40 02-con-k oxygen content analyzer and hematocrit by a microcapillary centrifuge. Measurement of plasma diphenhydramine concentrations was accomplished with the use of a capillary gas-liquid chromatograph with nitrogen/phosphorus selective detection (see section 2.1.5). 2.4.6. Statistical Analysis Statistical analysis was performed using Student's paired t- test for paired data and the unpaired t-test for unpaired data. Changes in the cardiorespiratory states, blood gas status, pH, blood pressures and heart rates during the course of the experiments were tested for statistical significance using 2-way analysis of variance and Duncan's multiple range test. The significance level was p<0.05. Values are given as the mean ± 1 S.D. 2.5. Recirculation of Diphenhydramine from Amniotic Fluid to the Ewe and Fetal Lamb: Intra-amniotic Bolus Injection Studies 2.5.1. Animals and Surgical Preparation Five pregnant Dorset or Suffolk ewes with a mean body weight of 76 ± 8 kg were employed in the study. Surgery was performed at a gestational age of 120 ± 5 days as described previously (see section 2.4.1). Silicone rubber catheters were implanted into the fetal femoral artery, lateral tarsal vein and trachea. Two catheters were placed into the amniotic cavity, one catheter near the hindlimbs and the other near the forelimbs. Catheters were also implanted in a maternal femoral 41 artery and vein. In one animal (ewe no. 287), a catheter was implanted into the umbilical vein as well. Antibiotic prophylaxis was as described previously (section 2.3.1). Animals were allowed to recover for a minimum of 4 days (8 ± 4 days) before experimentation. 2.5.2. Drug Administration and Experimental Protocol For each experiment, a 50 mg dose of diphenhydramine hydrochloride (Benadryl1* Injectable, Parke Davis, Lot No. 01866, Scarborough, Ontario) was diluted in isotonic saline to 5 mL, injected as an i.v. bolus dose via the amniotic catheter placed near the forelimbs over 30 sec and subsequently flushed with 10 mL of isotonic saline. Maternal arterial (3.0 mL), fetal arterial (1.5 mL) and umbilical venous (1.5 mL) blood and fetal tracheal (1.5 mL) and amniotic (1.5 mL) fluid samples were withdrawn simultaneously at -5, 5, 10, 15, 20, 30, 45 min, and 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 5.0, 6.0, 8.0, 10.0, 12.0, 20.0 and 28.0 hr following drug administration. The blood samples were collected in heparinized collection tubes and immediately centrifuged at 3,500 g for 10 min. The plasma was transferred to a 15 mL glass culture tube and frozen at -20°C until analysis. Samples for the measurement of fetal arterial pH, Pco2, P02, O2 content and hematocrit were collected at -5, 5, 10, 30 and 60 min following drug administration. These variables were determined as described in the Section 2.4.5. At intervals during the experiment, the fetal blood collected was replaced with an equal volume of blood from the ewe. 42 2.5.3. Data Analysis The terminal elimination half-life of diphenhydramine in the maternal and fetal arterial and umbilical venous plasma, and tracheal and amniotic fluids was calculated by linear regression analysis of the drug concentration vs. time profile during the terminal elimination phase. The ratio of the fetal tracheal, maternal arterial and umbilical venous to fetal arterial drug concentrations was calculated by dividing drug concentrations in each fluid by corresponding fetal arterial drug concentrations at each sampling time and the average ratio was determined for each animal. Area under the drug concentration vs. time curves in each biological fluid was calculated by eq. 4 and 5. Statistical analysis was performed using Student's paired t-test. The significance level was p<0.05. Values are given as the mean ± 1 S.D. 2.6. Pulmonary Disposition of Diphenhydramine in the Fetal Lamb: Maternal or Fetal Infusions to Steady-State 2.6.1. Animals and Surgical Preparation Three pregnant Dorset or Suffolk ewes were surgically prepared at gestational age of 124 to 129 days. Surgery was performed aseptically on the animals using halothane (1-2%) and nitrous oxide (60%) anesthesia (See Section 2.4.1.). Silicone rubber catheters were implanted in a fetal jugular vein (JV), carotid artery (CA), the main pulmonary trunk (PA), a branch of the left pulmonary vein (PV) and the trachea (TF). Catheters were also implanted in the maternal femoral 43 artery (MA) and vein (MV) and in the amniotic cavity (AF). Ampicillin was given to the ewe and fetus as described in section 2.3.1. Animals were allowed to recover for a minimum of 3 days before experimentation. For each experiment, the ewe was transferred to a monitoring cage adjacent to and in full view of the holding pen. 2.6.2. Drug Administration and Experimental Protocol Experiments were carried on the 3 animals at 128-133 days gestation. Two of the experiments involved fetal drug infusions and one experiment involved maternal drug infusion. Drug infusions were carried out in these animals as described previously in the Section 2.4.2. Briefly, for fetal infusion, diphenhydramine hydrochloride (Benadryl Injectable , Parke Davis, Lot No. 01866, Scarborough, Ontario) was injected as a 5 mg loading dose via the fetal jugular vein catheter, followed by a continuous infusion, with the drug infusion rate of 0.17 mg/min and the infusion volume of 0.17 mL/min for 90 min. For maternal infusion, the drug was given as a 20 mg loading dose via the maternal femoral vein catheter, followed by a continuous infusion, with the drug infusion rate of 0.67 mg/min and the infusion volume of 0.17 mL/min for 90 min. During the drug infusion experiments, fetal pulmonary arterial, pulmonary venous and carotid arterial blood samples (1.5 mL each) were collected simultaneously at -5, 60, 75, 90, 120 and 150 min. In addition, fetal tracheal and amniotic fluids (1.5 mL each) and maternal arterial blood (3.0 mL) were also taken at the same time. The total 44 volume of fetal blood collected was replaced by an equal volume of maternal blood in two divided doses, one after 75 min and one at the end of the experiment. All the samples were stored at -20°C for drug analysis (see section 2.1.5). 2.6.3. Pharmacokinetic Analysis Total body clearances of diphenhydramine in the ewe (CLmm) and fetus (CL^) were calculated by the following equations: CLmm = Maternal drug infusion rate/Cma)SS (eq. 24) CLff = Fetal drug infusion rate/Cca ss (eq. 25) where Cmass and Ccass are the steady-state drug concentrations measured in the maternal femoral arterial and fetal carotid arterial plasma, respectively. Fetal pulmonary extraction ratio (ERpU-|m) of diphenhydramine was calculated by the equation: ERpulm = (cpa,ss " cpv,ss)/cpa,ss (eq.26) where Cpass and CpVSS are the pulmonary arterial and venous drug concentrations at steady-state, respectively. Values are expressed as the mean ± 1 S.D. 45 3. RESULTS 3.1. Analytical Development 3.1.1. Optimization of Capillary GC Conditions The effects of changes in the flow rates of the make-up gases, hydrogen and air, and the carrier gas, helium, on N/P detector response were examined. At constant rates of helium carrier gas flow (1.0 mL/min) and air flow (60.0 mL/min), hydrogen gas flow was increased from 2.0 to 4.25 mL/min (Fig. 1). Although the maximum sensitivity of diphenhydramine and orphenadrine was obtained at hydrogen flow rates, 3.5-3.75 mL/min (Fig. 1), a hydrogen flow rate of 3.0 mL/min was chosen and utilized for further studies since at higher flow rates, the chromatographic baseline became unstable, with increased drift and noise. The effect of changes in air flow rate (range, 30.0 to 80.0 mL/min) on N/P detector response and peak width of diphenhydramine and orphenadrine were examined under constant rates of hydrogen gas flow (3.0 mL/min) and helium gas flow (1.0 mL/min) (Fig. 2). As the air flow rate was increased, the N/P detector response decreased. Peak width remained relatively unchanged over the air flow rate range examined. An optimum air flow rate of 50.0 mL/min was applied for further study since the resolution was better at air flow rates >50.0 mL/min without a appreciable loss of sensitivity (Fig. 2). 46 Fig. 1. Hydrogen flow rate-NPD response curves: carrier gas (helium) flow rate constant, 1 mL/min; and air flow rate constant, 60 mL/min. Error bars represent mean ± 1 S.D. (n = 4, mean of two injections for each sample). 47 *h— 1 1 1 i i ' 30 40 50 60 70 80 Air Flow Rate (ml/min) Fig. 2. Effect of air flow rate on NPD response and peak width: hydrogen flow rate constant, 3 mL/min; helium flow rate constant, 1 mL/min; and peak width = peak area count/peak height. Error bars represent mean ± 1 S.D. (n = 3-4, mean of two injections for each sample). 48 Fig. 3 shows the effect of column pressure on N/P detector response and peak width for diphenhydramine and orphenadrine at constant flow rates of hydrogen (3.0 mL/min) and air (50.0 mL/min). N/P detector response is greatest at column pressure of 13 psi (equivalent to 1.0 mL/min). The van Deemter curves for diphenhydramine and orphenadrine are shown in Fig. 4. The lowest H.E.T.P. value was also obtained at -13 psi. The column pressure of 13 psi, therefore, was chosen to provide optimum helium carrier gas flow rate. The relationship between purge activation time and NPD response for diphenhydramine and orphenadrine is shown in Fig. 5. The purge activation time of 1.0 min allows enough time for the solutes injected into the injection port in splitless mode to be transferred onto the column (Fig. 5). 3.1.2. Optimization of Drug Extraction Procedures Various solvents, including heptane, hexane, toluene, benzene and dichloromethane, in order of increasing polarity, were examined for their relative extraction efficiency. Among the solvents examined, dichloromethane was found to be the most efficient solvent for single- step extraction of diphenhydramine and the internal standard, orphenadrine. Furthermore, the addition of triethylamine in dichloromethane improved the extraction efficiency of both diphenhydramine and orphenadrine by -5 fold over that of dichloromethane alone (Fig. 6). A hexane:isoamyl alcohol mixture (98:2), which has been previously used for diphenhydramine extraction (Abernethy and Fig. 3. Effect of column pressure on NPD response and peak width: hydrogen flow rate constant, 3 mL/min; air flow rate constant, 50 mL/min; and peak width = peak area count/peak height. Error bars represent mean + 1 S.D. (n = 3-4, mean of two injections for each sample). 50 0.500-, 0.400 E 0.300 ui 0.200- x 0.100- 0.000 10 20 30 40 50 Linear Velocity (Cm/sec) Fig. 4. Relationship between height equivalent to theoretical plates (H.E.T.P) vs. linear velocity of the helium carrier gas determined for diphenhydramine. Column head pressures were 7, 9, 11, 13, 15 and 17 psi. 51 45n —i 1 1 1 1 1 1 I i 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Purge Activation Time (min) Fig. 5. Effect of purge activation time on NPD response: hydrogen flow rate constant, 3 mL/min; air flow rate constant, 50 mL/min; and helium flow rate constant, 1 mL/min. Error bars represent mean ± 1 S.D. (n = 3-4, mean of two injections for each sample). 0.0 0.001 0.005 0.01 0.1 0.5 (98:2) TEA Cone (M) in Dichloromethane Hexane:IAA Effect of the addition of triethylamine (TEA) on extraction of diphenhydramine (hatched bar) and orphenadrine (open bar) with dichloromethane (IAA, isoamyl alcohol). Error bars represent mean ± 1 S.D. (n = 4, mean of two injections for each sample). 53 Greenblatt, 1983), resulted in improved extraction efficiency as compared to dichloromethane only, although the extent of increase in extraction efficiency is less than that by dichloromethane with triethylamine (Fig. 6). Fig. 7 shows the relationship between NPD response and extraction time, ranging from 10 to 80 min. Both diphenhydramine and orphenadrine appear to be maximally extracted by -10 min and further sample extractions were conducted with the extraction time of 20 min. 3.1.3. Application of the Developed Assay Method to Placental Transfer Studies in Pregnant Sheep Representative GC chromatograms for maternal plasma obtained from a diphenhydramine-treated pregnant sheep, together with a corresponding control chromatogram is shown in Fig. 8. Extraneous peaks from endogenous constituents did not interfere with the analysis (Fig. 8). The peaks with retention times (t^) of 7.85 and 9.31 min were identified by GC-MS as diphenhydramine and orphenadrine, respectively. The data for a representative calibration curve used in the quantitation of diphenhydramine in maternal and fetal sheep plasma are presented in Table 1. Linearity was observed over the concentration range studied (2-320 ng/mL) with the line of best fit being described by y = 0.0050x + 0.0004 with a correlation coefficient (r) of 0.9996. An average diphenhydramine recovery of 103.2% (coefficient of variation, 5.2%) was obtained over the concentration range studied. Fig. 7. Extraction time vs. NPD response curves: dichloromethane was used for drug extraction. Error bars represent mean ± 1 S.D. (n = 4, mean of two injections for each sample). 55 B in 00 1 *1—' W J L 0 15 15 Fig. 8. Representative capillary gas chromatograms obtained from blank sheep plasma (A) and a plasma sample from a pregnant sheep (B) after administration of 100 mg of diphenhydramine hydrochloride intravenously. The plasma sample from the drug treated sheep contained diphenhydramine (Rj. = 7.85 min) and spiked internal 5 standard, orphenadrine (Rt = 9.31 min). Attenuation 2 , NPD- Voltage 14V. 56 Table 1. Calibration curve data for sheep plasma3. Amount of Diphenhydramine/Orphenadrine Coefficient Diphenhydramine area ratio , of Spiked (ng) (mean ± 1 S.D.)° Variation (%) 2 0.010 + 0.001 10.0 4 0.018 + 0.001 5.6 8 0.037 + 0.003 8.1 16 0.080 + 0.006 7.5 32 0.152 + 0.010 6.6 64 0.324 + 0.025 7.7 128 0.662 + 0.038 5.7 192 0.956 + 0.024 2.5 256 1.238 + 0.047 3.8 320 1.609 + 0.024 1.5 a. Linear regression line for diphenhydramine: y = 0.0500x + 0.0004, r = 0.9996. b. Number of samples, n = 4 (mean of two injections). 57 Pure authentic diphenhydramine samples, diphenhydramine-spiked control sheep plasma and plasma samples collected from diphenhydramine- treated sheep have been examined using GC-MS techniques. The total-ion current mass chromatograms for diphenhydramine obtained from GC-MS studies of these samples all showed a single peak for diphenhydramine in the positive ionization mode, providing evidence for the purity of the compound. Confirmation of the identity of diphenhydramine was obtained using both EI and CI mass spectra from authentic drug and drug-spiked plasma. When Cl-mass spectra were obtained and evaluated throughout the diphenhydramine peak elution from dosed sheep plasma, the results also confirmed the identity of diphenhydramine and peak homogeneity. Prominent ion fragments for diphenhydramine were observed at m/e 167, 256, 209 and 254 in the CI mode and at m/e 58, 73, 167, 165, 45, 168, 44 and 42 in the EI mode. The fragmentation patterns of diphenhydramine were similar to those reported earlier (Costello, 1981). Semilogarithmic plots of diphenhydramine concentration in maternal and fetal sheep plasma vs. time data following an i.v. bolus injection of 100 mg diphenhydramine hydrochloride to the ewe are shown in Fig. 9. Initial estimates of the pharmacokinetic parameters for diphenhydramine after an i.v. dose were obtained by computer analysis using the program AUTOAN (Sedman and Wagner, 1976). The data were later fitted using the program NONLIN (Metzler et al., 1974) and these initial estimates. The decline in the maternal and fetal plasma diphenhydramine concentration with time was observed to follow a biexponential decay described by the following equation: Cp = Ae"at + Be"^, where Cp is the plasma concentration at any time, t. A and B are the intercepts of the 58 —i 1 1 1 1 1 60 120 180 240 300 360 TIME (min) Fig. 9. A semilogarithmic plot of the diphenhydramine concentration vs. time profiles obtained in maternal (•) and fetal (A) plasma following a 100 mg i.v. bolus dose to a pregnant ewe. The terminal elimination half-life was calculated to be 48 and 44 min in the ewe and fetus, respectively. 59 a and fi phases, respectively. The parameters a and fi represent the distribution and terminal elimination rate constants, respectively. The maternal half-life of the a phase (ti,, a) was calculated to be 7 min, indicating rapid distribution of diphenhydramine following administration. A terminal elimination half-life (t^ ^) of 48 min was calculated for diphenhydramine in maternal plasma. The placental transfer of diphenhydramine to the fetus was rapid, with peak fetal plasma concentration being attained in less than 5 min following drug administration. The fetal diphenhydramine decay curve followed a pattern similar to that observed in maternal plasma. Distribution and terminal elimination half-lives of 8 and 44 min were calculated for diphenhydramine in the fetus. 3.2. Pharmacokinetics of Diphenhydramine after Dose Ranging in Nonpregnant Sheep Control values for the mean arterial pH and Pco2 were 7.481 + 0.015 and 34.3 + 1.4 mm Hg, respectively, and no significant changes were observed for these variables over the dose range studied. Changes in mean arterial P02, O2 content and hematocrit are shown in Table 2. Mean arterial P02 decreased significantly at 5 min following injection of 100 and 200 mg doses and thereafter returned to the control range. There were significant increases in 02 content and hematocrit at 5 and 10 min after administration of the 200 mg i.v. dose but these parameters returned to control values by 1 hr following drug administration. Figure 10 shows the average plasma diphenhydramine concentration vs. time curves obtained after i.v. bolus injection of 25, 60 Table 2. Mean values of arterial P02, 02 content and hematocrit obtained after i.v. bolus injection of diphenhydramine to 6 nonpregnant sheep. Diphenhydramine Hydrochloride Dose (mg) Time (min) 25 50 100 200 a Po2 -5 111.2(12.8) 120.8(3.7) 118.8(6.7) 121.5(6.7) 5 114.7(9.0) 110.2(12.5) 106.4(6.1)* 110.5(3.8)* 10 109.4(12.9) 116.7(13.7) 109.7(6.1) 113.1(9.9) 60 113.8(11.4) 111.5(6.7) 115.7(10.5) 116.9(7.1) 02 content -5 12.4(0.9) 13.2(0.7) 12.9(0.8) 12.6(0.6) 5 12.5(1.0) 12.7(1.0) 13.3(0.9) 14.9(1.5)* 10 12.3(0.8) 12.7(0.7) 13.0(1.1) 14.2(1.9)* 60 12.4(1.2) 13.1(0.9) 12.6(1.0) 12.9(1.2) Hematocrit -5 25.7(2.4) 27.4(1.5) 27.1(2.6) 26.8(2.4) 5 25.9(2.8) 27.1(2.5) 28.2(2.1) 31.3(3.8)* 10 26.0(2.1) 26.0(2.2) 28.6(2.8) 30.6(4.6)* 60 26.4(3.4) 29.5(3.7) 26.5(2.9) 28.0(3.3) a. values given as mean + 1 S.D. * p<0.05. 61 Fig. 10. Average plasma drug concentration vs. time curves obtained after i.v. administration of 25 (O), 50 (•), 100 (A) and 200 (•) mg doses of diphenhydramine hydrochloride to 6 nonpregnant sheep. 62 50, 100 and 200 mg dose of the drug to 6 nonpregnant sheep on a crossover basis. Plasma drug concentration vs. time curves were best described by biexponential equations in all but two of the experiments, where triexponential decay was observed in one animal after the 100 mg dose and in another animal after the 200 mg dose. Values for the intercompartmental transfer rate constants are shown in Table 3. Pharmacokinetic parameters calculated are shown in Table 4. The average initial distributional and the terminal elimination half-lives were increased from 5.4 to 8.9 min and from 34.0 to 67.5 min, respectively, as dose increased. The average elimination half-life after a 200 mg dose was significantly longer than after either the 25 or the 50 mg dose: conversely, the terminal elimination rate constant, lambdan, decreased significantly after the 200 mg dose (Fig. 11). There was also an increase in the volume of distribution (2.8 to 5.5 L/kg for Vdss), with increasing dose. The Vdss obtained from the 200 mg dose was significantly greater than that from the 25 mg dose (Fig. 12). However, within the dose range of 25-100 mg, there was no significant difference in either the elimination half-life or the volume of distribution. Values for Vdss were calculated by two different methods, namely, compartmental analysis using intercompartmental microconstants (eq. 8), and moment analysis using AUC and AUMC (eq. 9). VdS5 values calculated by these two methods were in close agreement, with 2-3% variation (Table 4). The average CLJQ (~5 L/hr/kg) remained relatively unchanged regardless of dose (Table 4). Although there were changes in the kinetic parameters of the distribution and elimination half-lives and OO the volume of distribution, there was a proportional increase in AUC0 as oo dose increased (Fig. 13) and therefore dose-normalized AUC0 remained Table 3. Estimates of NONLIN pharmacokinetic constants of diphenhy• dramine obtained after i.v. bolus injection of 25, 50, 100 and 200 mg doses to 6 nonpregnant sheep (mean ± 1 S.D.). Diphenhydramine Hydrochloride Dose (mg) Parameter 25 50 100a 200a Aj (ng/mL) 188.3 275.3 426.7 488.6 (93.9) (69.9) (133.8) (215.6) A2 (ng/mL) 53.4 90.3 142.3 258.8 (29.5) (32.7) (47.8) (73.7) lambda^ (min-*) 0.1545 0.1326 0.1074 0.0861 (0.0638) (0.0633) (0.0549) (0.0399) - lambda2 (min *) 0.0240 0.0203 0.0159 0.0117 (0.0091) (0.0047) (0.0036) (0.0022) 1 6 c k10 (min" ) 0.0696 0.0532 0.0407 0.0270 (0.0241) (0.0102) (0.0049) (0.0097) -1 kj2 (min ) 0.0537 0.0493 0.0392 0.0323 (0.0260) (0.0370) (0.0312) (0.0200) -1 6 k2j (min ) 0.0552 0.0504 0.0433 0.0386 (0.0297) (0.0231) (0.0254) (0.0190) r2 0.992d 0.998 0.995 0.998 (0.017) (0.003) (0.004) (0.003) a. obtained from 5 animals because one set plasma drug concentration- time data fit to triexponential decay. b. statistically different from the value obtained after 100 and 200 mg doses (p<0.05). c. statistically different from the value obtained from 200 mg dose (p<0.05). d. correlation coefficient obtained from NONLIN nonlinear least squares regression analysis. 64 Table 4. Pharmacokinetic parameters of diphenhydramine obtained after i.v. bolus injections to 6 nonpregnant sheep. Diphenhydramine Hydrochloride Dose (mg) Parameter 25 50 100 200 tk, lambdaj (min) 5.4 6.7 7.1 8.9 (2.6)a (4.1) (3.6) (4.6) b tk, lambdan (min) 34.0 36.5 52.3 67.5 (16.5) (12.5) (19.2) (18.7) CZ.TB (L/hr/kg) 5.0 4.8 5.0 4.6 (1.7) (0.7) (0.8) (0.8) ""area (L/k9> 4.1 4.1 6.5 7.3 (2.8) (1.0) (3.3) (1.5) c Vdss (L/kg) 2.8 3.1 4.5 5.5" (1.5) (0.9) (1.9) (1.5) d L k e 2.8 3.0 4.4 5.4 Vdss ( / 9) (1.7) (0.8) (1.7) (1.6) AUCQ (/xg-h/L) 60.4 119.9 229.4 504.4 (12.9) (16.8) (39.4) (116.9) a. values given as mean + (S.D.). b. value different from that obtained after 25 and 50 mg doses (p<0.05). c. calculated by: Dose-AUMCo/(AUCo)2. d. value different from that obtained after 25 mg dose (p<0.05). 2 e. calculated by: Dose-jL (C7-/lambda7-)/[ JL (C7-/lambda7-)] 7=1 7 7 7=1 65 2.0 n • O 1.5 - 1.0 c • A A 0.5 - O 0.0 25 50 100 200 Dose (mg) Fig. 11. Changes in the terminal elimination rate constant (lambdan) of diphenhydramine as a function of dose obtained in 6 nonpregnant sheep. Each symbol represents each animal and the horizontal lines represent mean values. Significantly different from 25 and 50 mg doses (p<0.05). 66 8i CP 6- c o •J—' D • 4- • cn A Q M— A A o • • 0) 2- o > 0 —i 1 1 1 25 50 100 200 Dose (mg) Fig. 12. Steady-state volume of distribution (^cfss) of diphenhydramine as a function of dose obtained in 6 nonpregnant sheep. *Significantly different from a 25 mg dose (p<0.05). 67 Fig. 13. Relationship between the area under the plasma drug oo concentration vs. time curve (AUC0) and i.v. bolus doses of diphenhydramine given on a crossover basis to 6 nonpregnant sheep. Parameters for the linear regression line were: y = 2.54x - 9.52 (r = 0.95). 68 unchanged. Plasma protein binding of diphenhydramine in nonpregnant sheep was determined by equilibrium dialysis. Equilibrium between the plasma and buffer compartments was obtained by 3 hr (Fig. 14). Drug adsorption onto the surface of the dialysis cell and membrane was minimal, with an average recovery of 99 + 5% over the drug concentration range of 10- 10,000 ng/mL. In vitro protein binding experiments showed that there was no significant change in the free fraction over a drug concentration range from 10-2,000 ng/mL, with a mean of 0.332 + 0.057. The free fraction, however, was increased to 0.489 at 4,000 ng/mL and 0.570 at 10,000 ng/mL. The extent of plasma protein binding was also determined in the 6 animals in vivo in 5 and 60 min samples obtained after administration of a 200 mg dose. The average free fraction was 0.229 + 0.080 in the nonpregnant ewes employed (Table 5). In two animals, the free fraction of diphenhydramine was determined in all samples collected after a 200 mg dose. The unbound fraction remained unchanged regardless of drug concentration, with a mean of 0.303 ± 0.033 over the concentration range, 30-567 ng/mL in one animal (ewe no. 3) and 0.170 + 0.020 over the concentration range, 31-783 ng/mL in the other animal (ewe no. 132). 3.3. Pharmacokinetics of Diphenhydramine in the Pregnant Ewe and Fetal Lamb: i.v. Bolus Injection Studies The physical characteristics of individual ewes employed in the study are shown in Table 6. The gestational age at the time of the 14. Determination of equilibrium time used in the study of plasma protein binding of diphenhydramine by equilibrium dialysis (n = 3). Initial buffer concentration: A500 ng/mL, 0 10,000 ng/mL. 70 Table 5. Plasma free fraction of diphenhydramine determined in samples obtained after i.v. bolus injection of the 200 mg dose to nonpregnant sheep. Free Fraction Animal No. 5 min 60 min Average 3 0.284 (567.0) 0.282 (169.1) 0.283 15 0.347 (477.6) 0.324 (111.2) 0.335 128 0.174 (702.7) 0.237 (96.6) 0.205 132 0.179 (782.7) 0.147 (131.6) 0.163 905 0.214 (434.0) 0.318 (109.4) 0.266 906 0.122 (1072.1) 0.124 (183.6) 0.123 mean + S.D. 0.220 ± 0.082 0.239 + 0.086 0.229 ± 0.080 Numbers in brackets are the total drug concentrations (ng/mL) in the plasma sample employed. 71 Table 6. Physical characteristics of individual pregnant Dorset ewes used in the maternal diphenhydramine i.v. bolus injection study. Maternal Gestational Age Fetal Body No. DPHM HC1 Ewe Body (days) Weight of i .v No. Weight in utero Fetuses Dose (kg) (kg)a (mg/kg) Surgery Experiment 104 89.1 122 129 3.0 2 1.122 62 60.0 126 139 2.5 1 1.667 98 61.8 129 140 2.9 1 1.618 110 67.3 122 130 2.1 2 1.486 69.6 125 135 2.6 1.473 (13.4)6 (3) (6) (0.4) (0.246) a. calculated by the equation: log (fetal body weight in utero) = log (birth weight) - (0.0153 x number of days between experiment and birth). b. Numbers in brackets are ± 1 S.D. 72 experiment ranges from 129-140 days (mean 135 days), and the maternal body weight ranged from 60-89 kg (mean 70 kg). Changes in the fetal arterial blood gas status, pH, 02 content and hematocrit are shown in Table 7. Fetal descending aortic Po2, Pco2, pH, 02 content and hematocrit in the control period averaged 17.1 ± 4.3 mm Hg, 49.0 ± 9.0 mm Hg, 7.399 ± 0.108, 8.4 ± 2.5 vol % and 39.9 ± 3.6 vol %, respectively. These values are in the normal range for fetal lambs. In all four fetuses, Po2 fell following drug administration to a low of 12.7 ± 2.9 mm Hg at 5 min, and there was a corresponding fall in blood oxygen content. By 30 min following drug injection, Po2 had returned to control levels. No other changes in fetal blood gas status were observed. Fig. 15 shows representative capillary gas chromatograms obtained from maternal and fetal plasma, fetal tracheal and amniotic fluids of a pregnant sheep following a 100 mg i.v. bolus injection of diphenhydramine hydrochloride. Extraneous peaks from endogenous constituents were negligible and did not interfere with the analysis. Semi logarithmic plots of concentration vs. time data in maternal and fetal plasma, and tracheal and amniotic fluids following administration of a 100 mg i.v. dose to a ewe, are shown in Fig. 16. Each set of maternal plasma concentration vs. time data was best described by a biexponential equation. The intercompartmental transfer rate constants and the pharmacokinetic parameters obtained from the individual ewes after maternal drug administration are shown in Tables 8 and 9. The transfer of the drug to the fetus was rapid, with the 73 Table 7. The effect of maternal diphenhydramine administration on fetal arterial pH, Pco2, Po2, 02 content and hematocrit following 100 mg i.v. bolus. Parameter Control 5 min 10 min 30 min PH 7.399a 7.391 7.394 7.393 (0.108) (0.101) (0.109) (0.096) PcOo 49.0 50.6 49.9 50.1 (mm Hg) (9.0) (5.5) (6.8) (7.1) POo 17.1 * 15.2 17.9 12.7 (mm Hg) (4.3) (2.9) (2.3) (5.3) 02 Content 8.4 5.2* 6.4 8.7 (% v/v) (2.5) (1.5) (2.3) (2.3) Hematocrit 39.9 40.7 41.1 40.7 (% v/v) (3.6) (3.1) (2.9 (3.3) a. mean ± 1 S.D. (n = 4). * P<0.05. 74 LO 00 ro art ro MA FA TF AF ro cn ro CTl LO CO LO 00 LO oo LJJLJ ^ L Lr »J—^ j 15 0 15 0 15 0 15 Fig. 15. Representative capillary gas chromatograms of the maternal (MA) and fetal (FA) arterial plasma, fetal tracheal (TF) and amniotic (AF) fluids obtained from control- and drug-treated pregnant sheep. The samples contained diphenhydramine (Rt = -7.85 min) and spiked internal standard orphenadrine (R^. = -9.31 min). 75 100Ch c £ 100: c (J C o o V 10 c D X) >, x: c x: CL 60 120 180 240 300 360 Time (min) Fig.-16. Representative semi logarithmic plots of the maternal (O) and fetal (•) arterial plasma, fetal tracheal (•) and amniotic (•) fluid diphenhydramine concentration vs. time profiles following a 100 mg i.v. bolus. 76 Table 8. Estimates of NONLIN pharmacokinetic constants obtained from maternal plasma diphenhydramine data following a 100 mg i.v. dose of the drug to individual ewes. Ewe No. Parameter mean ± S.D. 104 62 98 110 A (ng/mL) 764.3 277.6 521.7 1586.7 787.6 ± 568.6b B (ng/mL) 442.6 142.7 265.5 542.3 348.3 ± 178.6 a (min-1) 0.0997 0.0822 0.0567 0.0932 0.0830 ± 0.0189 P (min-1) 0.0145 0.0125 0.0162 0.0112 0.0136 ± 0.0022 1.000 0.998 0.999 0.998 0.999 ± 0.001 -1 k10 (min ) 0.0317 0.0285 0.0307 0.0302 0.0303 ± 0.0013 kj2 (min-1) 0.0368 0.0302 0.0123 0.0380 0.0293 ± 0.0119 -1 k2i (min ) 0.0457 0.0362 0.0298 0.0258 0.0344 ± 0.0088 a. coefficient of regression obtained from NONLIN nonlinear least squares regression analysis. b. Values are expressed as the mean ± 1 S.D. 77 Table 9. Pharmacokinetic parameters obtained after a 100 mg i.v. dose of diphenhydramine to individual ewes. Ewe No. Parameter mean ± 1 S.D. 104 62 98 110 Maternal parameter H, a (min) 7.0 8.4 12.2 7.4 8.8 ± 2.4 H, p (min) 47.8 55.4 42.9 62.1 52.0 ± 8.5 ^area (L/k9) 2.2 9.3 4.1 2.3 4.5 ± 3.3 Vdss (L/kg) 1.6 6.7 2.7 1.7 3.2 ± 2.4 C/1TB (L/hr/kg) 1.9 7.0 4.0 1.5 3.6 ± 2.5 AUCM™ (mg-hr/L) 599.9 240.0 406.0 973.6 554.9 ± 315.5 Fetal parameter _a thf a (min) 7.9 15.5 3.1 8.8 ± 6.3 p (min) 43.8 60.3 37.7 43.3 46.3 ± 9.8 AUCFo (mg-hr/L) 252.6 185.9 561.0 789.0 447.1 ± 280.4 AUCF>UCM; 0.42 0.78 1.38 0.81 0.85 ± 0.40 a. Value indeterminate because concentration vs. time course did not fit biexponential decay. 78 highest fetal plasma cencentrations observed at the earliest sampling time (5 min) in three animals. Three of the four sets of fetal data were described by a biexponential equation, with a distinct but short distribution phase followed by a rapid elimination. In the remaining animal (ewe no. 110), however, peak fetal plasma drug concentration occurred 20 min after drug administration, preventing the identification of a possible short distributional phase preceeding the terminal elimination phase in this fetus. The average fetal drug elimination half-life, determined from the terminal phase (46.3 ± 9.8 min), was not significantly different from that in the ewe (52.0 + 8.5 min). During the entire sampling period, the fetal plasma concentrations closely paralleled that in the mother, with an overall correlation coefficient between maternal and fetal plasma drug concentrations of 0.857. The average fetal to maternal drug concentration ratio was 0.90 ± 0.49, but in one experiment, diphenhydramine concentration in the fetus was slightly, but consistently higher than in the ewe. The ratio of fetal to maternal plasma AUC values obtained from four sheep averaged 0.85 ± 0.40. The mean of the percentage of diphenhydramine administered to the ewe that is eliminated by the fetus via nonplacental elimination pathways (eq. 15) was 5.0% (Table 10). Alternatively, most of the drug (95.0%) administered to the pregnant sheep was eliminated by the ewe via maternal nonplacental pathways. In the tracheal fluid, the average time to reach the maximum drug concentration was 28 min (range, 10-45 min) after drug administration, with an average peak concentration of 551.2 ng/mL (range, 476.7-699.0 ng/mL). The fetal tracheal to arterial plasma drug 79 Table 10. Percentage of the intravenously administered maternal dose (100 mg bolus) of diphenhydramine eliminated by the fetus via nonplacental pathways. a 6 AUCMo-CLmo AUCFo'CLfo AUCF^-CLfo-100/ Ewe No. (mg) (mg) (AUCMo'CLmo^AUCF;.CLfo) 104 120.4 3.2 2.6 62 48.2 2.3 4.6 98 81.5 7.0 7.9 110 195.3 9.9 4.8 111.4 ± 63.3 5.6 ± 3.5 5.0 ± 2.2 a. Mean value (n = 8) for the nonplacental clearance of diphenhydramine from the maternal compartment, CLmo (3343.8 mL/min), was obtained from steady-state infusion experiments (Section 3.4.1.4). b. Mean value ((n = 8) for the nonplacental clearance of diphenhydramine from the fetal compartment, CLfo (208.4 mL/min), was obtained from steady-state infusion experiments (Section 3.4.1.4). 80 concentration ratio was calculated by dividing tracheal drug concentrations by corresponding fetal plasma drug concentrations at each sampling time. The average ratio of the fetal tracheal to arterial plasma diphenhydramine concentrations calculated was 2.7 ± 2.4. Similarly, diphenhydramine concentrations in amniotic fluid rose progressively, with its peak concentration (mean, 113.1 ng/mL) occurring from 1.5 to 3.0 hr after drug administration. Approximately 3.0-3.5 hr after drug dosing, diphenhydramine levels in the amniotic fluid exceeded those in tracheal fluid and maternal and fetal plasma (Fig. 16). The highest diphenhydramine concentration ultimately achieved in amniotic fluid was substantially less than the highest concentrations in fetal plasma in each experiment (mean ratio, 0.34). The average apparent terminal elimination half-life in the tracheal fluid (38.7 ± 1.5 min) determined by linear regression analysis was significantly shorter than those in maternal (52.0 ± 8.5 min) and fetal (46.3 ± 9.8 min) plasma. Diphenhydramine is eliminated from the amniotic fluid more slowly than from the maternal and fetal plasma and tracheal fluid, with an average elimination half-life of 168 min. 3.4. Transplacental Clearance, Nonplacental Clearance and Fetal Effects of Diphenhydramine in Pregnant Sheep: Maternal and Fetal Drug Infusions to Steady-State 3.4.1. Transplacental and Nonplacental Clearances in the Maternal- Fetal Unit 3.4.1.1. Achievement of Steady-State Diphenhydramine Concentrations in Maternal and Fetal Plasma 81 A total of 16 diphenhydramine infusion experiments were carried out in the 8 animals, involving separate maternal and fetal i.v. infusions of the drug, given 48 hr apart. Information on the individual pregnant ewes and fetuses employed in the study is shown in Table 11. Maternal and fetal drug infusion experiments were carried out over the gestational age range of 127-136 (mean, 131 ± 3) and 126-136 (mean, 132 ± 4) days, respectively. Diphenhydramine is rapidly transferred across the placenta from mother to fetus or from fetus to mother, with the highest drug concentrations observed at the earliest sampling time (5 min) after either the maternal or fetal drug administration. Average maternal and fetal plasma drug concentration vs. time curves following maternal and fetal infusions are shown in Fig. 17 and 18, respectively. Plasma drug concentrations reached a plateau by 30-60 min after starting the infusion and from 60-90 min, the drug levels remained relatively constant. Steady-state diphenhydramine concentrations were calculated as the mean of drug concentrations at 60, 75 and 90 min and are shown in Table 12. The average steady-state plasma diphenhydramine concentration after maternal infusion was 212.1 + 67.8 ng/mL in the mother and 36.3 ± 14.4 ng/mL in the fetus, resulting in a fetal to maternal concentration ratio (Cfss/Cmss) of 0.19 ± 0.10 (Table 12 and 13). Following fetal infusions, maternal and fetal steady-state drug concentrations were 31.1 ± 11.6 and 447.6 ± 185.2 ng/mL, respectively, with an average maternal to fetal concentration ratio (Cm'ss/Cf'ss) of 0.08 ± 0.05 (Table 12 and 13). The mean coefficient of variation of the maternal and fetal steady-state drug concentrations between 60-90 min after the start of infusion was 4.1 ± 3.3% and 4.8 ± 3.4%, respectively, for maternal 82 Table 11. Physical characteristics of individual pregnant sheep used in the steady-state infusion study. Maternal Gestational Age Fetal Weight Number Ewe Breed Body (days) in utero (kg) of No. Weight Fetuses (kg) Surgery MI FI MI FI Average Drug Infusion Study 121 Suffolk 66.4 128 134 136 3.3 3.4 3.4 2 125 Suffolk 71.4 123 134 136 1.7 1.8 1.8 3 130 Suffolk 96.4 123 128 126 2.5 2.4 2.5 1 133 Suffolk 88.2 129 136 135 2.2 2.2 2.2 2 138 Dorset 61.4 122 132 134 2.5 2.7 2.6 1 202 Suffolk 89.1 119 132 130 1.5 1.4 1.5 4 204 Dorset 62.7 122 128 130 2.4 2.5 2.5 1 480 Dorset 85.0 120 127 129 1.3 1.3 1.3 2 77.6 123 131 132 2.2 2.2 (2.2) (13.6) (4) (3) (4) (0.7) (0.7) (0.7) Saline Infusion Study 135 Dorset 70.0 122 128 - 1.7 2 213 Suffolk 86.8 123 - 136 2.1 3 215 Dorset 86.4 129 - 142 2.4 1 482 Dorset 68.2 125 - 133 1.1 3 77.9 123 128 135 1.7 1.8 (10.1) (2) (6) (0.6) Numbers in brackets are ± 1 S.D. MI, maternal infusion; FI, fetal infusion. 83 Fig. 17. Average maternal (•) and fetal (O) plasma diphenhydramine concentration vs. time curves obtained following i.v. infusion of the drug to the ewe (n = 8). 84 Fig. 18. Average maternal (•) and fetal (o) plasma diphenhydramine concentration vs. time curves obtained following i.v. infusion of the drug to the fetus (n = 8). 85 Table 12. Steady-state diphenhydramine concentrations in the ewe and fetus following maternal and fetal drug administrations. Ewe Maternal Infusion (ng/mL) Fetal infusion (ng/mL) No. Cm Cm' Cf ss cfss ss ss 121 360.3 (17.3)a 35.5 (5.9) 53.9 (2.6) 658.0 (108.6) 125 215.8 (23.7) 29.6 (5.1) 26.5 (2.9) 697.9 (120.7) 130 185.4 (13.3) 20.8 (6.3) 25.3 (1.8) 274.9 (82.8) 133 207.8 (18.1) 56.0 (12.4) 26.4 (2.3) 367.0 (81.5) 138 197.8 (33.2) 49.6 (16.2) 39.5 (6.6) 509.8 (166.2) 202 152.9 (24.0) 18.3 (6.7) 22.4 (3.5) 323.4 (117.7) 204 236.1 (45.6) 29.1 (11.7) 36.5 (7.1) 192.1 (77.2) 480 140.3 (41.1) 51.2 (13.5) 17.9 (5.3) 557.9 (146.7) x ± S.D. 212.1 ± 67.8 36.3 ± 14.4 31.1 ± 11.6 447.6 ± 185.2 (27.0 ± 11.7) (9.7 ± 4.2) (4.0 ± 2.0) (112.7 ± 32.2) a. Values in brackets are the data based on free drug concentrations. 86 Table 13. Maternal to fetal or fetal to maternal diphenhydramine concentration ratios at steady-state following maternal and fetal drug infusion. Ewe Maternal Infusion Fetal infusion No. Cfss/Cmss Cmss/Cfss Cm'ss/Cf'ss Cf'ss/Cm'ss 121 0.10 (0.34) 10.15 (2.93) 0.08 (0.02) 12.21 (41.76) 125 0.14 (0.22) 7.29 (4.65) 0.04 (0.02) 26.34 (41.62) 130 0.11 (0.47) 8.91 (2.13) 0.09 (0.02) 10.87 (46.00) 133 0.27 (0.69) 3.71 (1.46) 0.07 (0.03) 13.90 (35.44) 138 0.25 (0.49) 3.99 (2.05) 0.08 (0.04) 12.91 (25.70) 202 0.12 (0.28) 8.36 (3.58) 0.07 (0.03) 14.44 (33.60) 204 0.12 (0.26) 8.11 (3.90) 0.19 (0.09) 5.26 (10.87) 480 0.37 (0.33) 2.74 (3.04) 0.03 (0.04) 31.17 (27.68) x ± S.D. 0.19 ± 0.10 6.66 ± 0.98 0.08 ± 0.05 15.89 ± 3.02 (0.39 ± 0.16) (2.97 ± 1.06) (0.04 ± 0.02) (32.83 ± 11.32) Values in brackets are the data based on free drug concentrations. 87 infusions and 6.4 ± 4.2 and 4.5 ± 4.2%, respectively for fetal infusions. 3.4.1.2. Plasma Protein Binding in Maternal and Fetal Sheep The free fraction of diphenhydramine determined at steady-state was significantly higher in the fetus (0.277 ± 0.087) than in the mother (0.141 ± 0.079) (Table 14). The ratio of fetal to maternal free fraction of the drug was greater than unity in all the animals except one, with a mean ratio of 2.37 ± 1.04. The steady-state concentrations of unbound diphenhydramine were calculated accordingly and are shown in parentheses in Table 12. As with total drug concentrations, unbound diphenhydramine concentrations in the mother are still higher than in the fetus following maternal infusion, although the ratio of the fetal to maternal free drug concentrations, CfSS}free/Cmss>free, increased as compared with the Cf$s/Cmss ratio (0.39 vs. 0.19) (Table 13). Similarly, following fetal drug infusions, steady-state unbound drug concentrations in the fetus were higher than in the mother and there was a further decrease in the maternal to fetal free drug concentration as ratio, CmSS5free/Cf'ss?free> compared with the ratio based on the total drug concentration (0.04 vs. 0.08) (Table 13). 3.4.1.3. Drug Disposition in Fetal Tracheal and Amniotic Fluids Representative semilogarithmic plots of the fetal tracheal and amniotic fluid as well as maternal and fetal plasma diphenhydramine concentration vs. time curves obtained after maternal and fetal 88 Table 14. Plasma free fraction of diphenhydramine in the pregnant ewe and fetus at steady-state following maternal or fetal drug administration9. Free Fraction Ewe Fetal/Maternal Ratio No. Ewe Fetus 121 0.048 0.165 3.44 125 0.110 0.173 1.57 130 0.072 0.301 4.18 133 0.087 0.222 2.55 138 0.168 0.326 1.94 202 0.157 0.364 2.32 204 0.193 0.402 2.08 480 0.293 0.263 0.90 x ± S.D 0.141 ± 0. 079 0.277 ± 0.0876 2.37 ± 1.04 a. Free fraction of the drug was determined by equilibrium dialysis in 75 and 90 min maternal and fetal plasma samples obtained following maternal and fetal drug infusions, respectively. b. Significantly different from the maternal value (p<0.05). 89 infusions are shown in Figs. 19 and 20, respectively. During maternal infusion experiments, the average time required to reach the maximum diphenhydramine concentrations (Cmax), Tmax, in the fetal tracheal fluid (95 ± 41 min) was not significantly different from that in the amniotic fluid (113 ± 55 min) (Table 15). The average Cmax in the tracheal fluid (124.7 ng/mL) was significantly higher than that in the amniotic fluid (16.6 ng/mL) (Table 15). The ratio of the fetal tracheal to plasma drug concentrations averaged 3.4 for the maternal infusion experiments (Table 16). However, the highest diphenhydramine concentration ultimately achieved in amniotic fluid following maternal infusion was substantially less than the highest drug concentration in fetal plasma (mean ratio, 0.4 ± 0.2). Following fetal infusions, the time required to reach the Cmax in tracheal fluid (75 ± 31 min) was significantly less than that in amniotic fluid (129 ± 41 min) (Table 15). However, these Tmax values were not significantly different from the corresponding values obtained from maternal infusions. As in the case of maternal infusions, the Cmax in the tracheal fluid (1790.4 ng/mL) was significantly greater than that in the amniotic fluid (139.9 ng/mL) (Table 15). The average fetal tracheal to plasma drug concentration ratio was 5.4 ± 4.5 (Table 16). The Cmax in the amniotic fluid was lower than that in the fetal plasma in each experiment (mean ratio, 0.3 ± 0.3). The average elimination half-life of diphenhydramine in the tracheal fluid obtained after maternal infusion (49 ± 17 min) was not different from that obtained after fetal infusion (56 + 16 min) (Table Fig. 19. Representative semi logarithmic plots of the maternal (o) and fetal (•) arterial plasma, fetal tracheal (A) and amniotic (A) fluid diphenhydramine concentration vs. time curves in a pregnant ewe and fetus following drug infusion to the ewe. 91 5000 A'A"AN ? 1000rhA"A' A A A A c ^•-•-•-•-•-•^ ' ' ^ o / \T>4 c 100 V o c o o Q) boo-o-°-o' c 10. 0.r *E D l_ TJ Infusion >, x: c 0) 60 120 180 240 300 360 420 480 a. Q Time (min) Fig. 20. Representative semi logarithmic plots of the maternal (o) and fetal (•) arterial plasma, fetal tracheal (A) and amniotic (A) fluid diphenhydramine concentration vs. time curves in a pregnant ewe and fetus following drug infusion to the fetus. 92 Table 15. Values for the observed maximum drug concentrations (Cmax) of diphenhydramine and time to reach Cmax (Tmax) in tracheal (TF) and amniotic (AF) fluids following maternal and fetal infusions to steady-state. Maternal Infusion Fetal Infusion Ewe No. (ng/mL) (min) r (ng/mL) (min) ^max Tmax °max ^max TF AF TF AF TF AF TF AF 121 71.9 6.4 95 105 3303.7 41.9 5 110 125 88.7 11.8 95 135 1066.1 98.6 95 90 .a .a 130 31.6 95 551.9 179.9 90 135 133 66.3 9.9 90 150 887.8 35.7 75 100 138 100.4 25.2 135 180 2952.6 142.1 90 90 202 396.7 12.8 5 100 3132.4 128.9 95 150 204 123.3 20.8 135 120 732.7 120.7 90 150 480 118.5 29.4 110 5 1695.9 371.7 60 210 * 124.7 . 16.6 95 113 1790.4* 139.9 75* 129 (113.9)° (8.6) (41) (55) (1161.4) (105.6) (31) (41) a. Amniotic fluid samples have not been taken. b. Numbers in brackets are ± 1 S.D. Significantly different (P<0.05) from the corresponding value in the amniotic fluid. 93 Table 16. Average ratios of the fetal tracheal (TF) to arterial (FA) diphenhydramine concentrations and the elimination half-life (t^) of the drug obtained in maternal (MA) and fetal (FA) plasma, tracheal (TF) and amniotic (AF) fluids after maternal and fetal infusions to steady-state. Maternal Infusion Fetal Infusion Ewe No. (min) (min) TF/FA S TF/FA h MA TF FA AFa MA TF FA AF 121 2.1 (0.9) 51 27 31 1.5 (1.1) 32 37 69 161 125 3.0 (1.1) 39 47 32 1.7 (1.0) 36 50 62 99 130 1.4 (0.3) 51 67 49 1.3 (0.8) 11 64 122 118 133 1.1 (0.3) 63 36 83 2.6 (0.7) 25 39 42 154 138 1.9 (0.9) 66 35 39 - 11.0 (8.3) 36 44 71 - b 202 12.1 (6.0) 33 108 8.9 (4.9) 178 67 90 131 204 3.4 (2.6) 43 69 108 12.0 (15.3) 74 85 73 128 480 2.4 (1.0) 59 61 54 3.9 (2.2) 34 62 51 67 3.4 51 49 63 5.4 53 56 73 123 (3.6) (12) (17) (32) (4.5) (53) (16) (25) (32) a. The t^ in the amniotic fluid following maternal infusion was not determined because amniotic DPHM concentrations were too low to characterize the elimination phase. b. Numbers in brackets are mean ± 1 S.D. 94 16). The elimination half-life of diphenhydramine in the amniotic fluid after maternal infusions was not determined because drug concentrations in this fluid were too low for the characterization of the elimination phase. However, the half-life of diphenhydramine in the amniotic fluid determined following fetal infusions averaged 123 ± 32 min and was significantly greater than that in the tracheal fluid (56 ± 16 min) (Table 16). 3.4.1.4. Transplacental and Nonplacental Clearances Transplacental and nonplacental clearances of diphenhydramine from both mother and fetus were calculated based on the total drug concentrations (Table 17 and 18) and free drug concentrations (Table 19 and 20). The clearance values normalized by the maternal or fetal body weight are also given in parentheses. Based on total drug concentrations, average total body clearances from the mother (CLmm) and fetus {CLff) were 3426.1 and 472.7 mL/min, respectively (Table 18). Fetal total body clearance of diphenhydramine is relatively small (13.8%) as compared with the maternal total body clearance. However, when the CLff was corrected for fetal body weight, the resulting clearance value was -5 fold greater than the maternal body weight corrected CLmm (Table 18). When the total body clearance was calculated by the model-independent method (equation 22 and 23), the values were in very close agreement with those resulting from the compartmental analysis, with an average coefficient of variation of 1.5%. The fetal to maternal transplacental clearance, CLfm (264.4 ± 138.7 mL/min), was ~3 fold greater than the maternal to fetal transplacental clearance, 95 Table 17. Transplacental and nonplacental clearances of diphenhydramine based on total drug concentrations in the ewe and fetus following maternal and fetal infusions to steady-state. Transplacental Clearance Nonplacental Clearance Ewe (mL/min) (mL/min) No. CLmf CLfm CLmo CLfo 121 25.2 (7.4)a 152.8 (44.9) 1840. .2 (27.7) 102. 6 (30.2) 125 32.9 (18.3) 117.9 (65.5) 3072 .6 (43.0) 122. 2 (67.9) 130 68.7 (27.5) 334.4 (133.8) 3564 .6 (37.0) 278. .2 (111.3) 133 124.8 (56.7) 235.3 (107.0) 3146 .8 (35.7) 227. ,8 (103.6) 138 125.4 (48.2) 266.3 (102.4) 3311 .8 (53.9) 233. .8 (89.9) 202 62.2 (41.5) 304.5 (203.0) 4334 .4 (48.7) 215, .2 (143.5) 204 109.5 (43.8) 549.4 (219.8) 2781 .9 (44.4) 339. .0 (135.6) 480 110.3 (84.9) 154.3 (118.7) 4697 .7 (55.3) 148 .0 (113.9) 82.4 ;t 40.5b 264.4 ± 138.7C 3343 .8 ± 890.7 208 .4 ± 80.4C c c (41.0 : ± 24.1) (124.4 ± 60.9) (43.2 ± 9.5) (99.5 ± 36.8) a. Numbers in brackets are the clearances (mL/min/kg) normalized to maternal (CLmo) or fetal (CLfm, CLmf and CLfo) body weight. b. Values are given as mean + 1 S.D. c. Significantly different from maternal value (p<0.05). 96 Table 18. Total body clearances of diphenhydramine and percent contribution of nonplacental clearances to total drug elimination in the ewe and fetus based on total drug concentration. Total Body Clearance Nonplacental Contribution Ewe (mL/min) ('%) No. CL ^ff Maternal Fetal OLmm 121 1865.4 (28.1)a 255.4 (75.1) 98.7 40.2 125 3105.5 (43.5) 240.1 (133.9) 98.9 50.9 130 3633.3 (37.7) 612.6 (245.0) 98.1 45.4 133 3271.6 (37.1) 463.1 (210.5) 96.2 49.2 138 3437.2 (56.0) 500.1 (192.4) 97.6 46.8 202 4396.6 (49.4) 519.7 (346.5) 98.6 41.4 204 2891.4 (46.1) 888.4 (355.4) 96.2 38.2 480 4808.0 (56.6) 302.3 (232.5) 97.7 49.0 3426.1 ± 905.8° 472.7 + 215.7C 97.8 ± 1.1 45.1 ± 4.7C (44.3 ± 9.8) (223.9 ± 95.8)c a. Numbers in brackets are the clearances (mL/min/kg) normalized to maternal (CLmm) or fetal (CLff) body weight. b. Values are given as mean + 1 S.D. c. Significantly different from maternal value (p<0.05). 97 Table 19. Transplacental and nonplacental clearances of diphenhydramine based on free drug concentrations in the ewe and fetus following maternal and fetal infusions to steady-state. Transplacental Clearance Nonplacental Clearance (mL/min) (mL/min) Ewe No. CLmf CLfm CLmo CLfo 121 527. ,7 (155.2) 930. 2 (273.6) 38325.2 (577.2) 617. 1 (181.5) 125 298. ,7 (165.9) 679. ,4 (377.4) 27976.9 (391.8) 708. 6 (393.7) 130 956. ,1 (382.4) 1092. ,7 (437.1) 49308.9 (511.5) 941. 0 (376.4) 133 1428. .6 (649.4) 1059. ,9 (481.8) 36130.0 (409.6) 1025. ,4 (466.1) 138 733. .2 (282.0) 796. .6 (306.4) 19735.8 (321.4) 706. ,0 (271.5) 202 398, .6 (265.7) 832, .9 (555.3) 27611.7 (309.9) 595. .0 (396.7) 204 567, .3 (226.9) 1377, .1 (550.8) 14405.9 (229.8) 834, .0 (333.6) 480 377 .7 (290.5) 593 .1 (456.2) 16037.7 (188.7) 556. .7 (428.2) 661 .0 ± 375.8 920 .2 ± 252.4 28691.5 ± 12067.0 750 .0 ± 170.0 (302.3 ± 158.0) (429.8 ± 104.4) (367.5 ± 132.8) (356.0 ± 91.8) a. Numbers in brackets are the clearances normalized to maternal or fetal body weight (mL/min/kg). b. Values are given as mean + 1 S.D. 98 Table 20. Total body clearances of diphenhydramine and percent contribution of the nonplacental clearances to total drug elimination in the ewe and fetus based on free drug concentrations. Total Body Clearance Nonplacental Contribution (mL/min) (%) Ewe No. CL CLff Maternal Fetal "''-mm 121 38852.9 (585.l)a 1547.3 (455.1) 98.6 39.9 125 28275.6 (396.0) 1388.0 (771.1) 98.9 51.1 130 50265.0 (521.4) 2033.7 (813.5) 98.1 46.3 133 37558.6 (425.8) 2085.3 (947.9) 96.2 49.2 138 20469.0 (333.4) 1502.6 (577.9) 96.4 47.0 202 28010.3 (314.4) 1427.9 (951.9) 98.6 41.7 204 14973.2 (238.8) 2211.1 (884.4) 96.2 37.7 480 16415.4 (193.1) 1149.8 (884.5) 97.7 48.4 29352.5 ± 12252.9° 1668.2 ± 387.2 97.6 ± 1.2 45.2 ± 4.8 (376.0 ± 134.0) (785.8 ± 180.0) a. Numbers in brackets are the clearances normalized to maternal or fetal body weight (mL/min/kg). 6. Values are given as mean + 1 S.D. 99 CLmf (82.4 ± 40.5 mL/min) (Table 17). Average contribution of the placental clearance of the drug accounted for 2.2% of the total drug clearance in the mother (CLmf/CLmm), whereas in the fetus, the average placental clearance accounted for 54.9% of the total clearance from the fetus (CLfm/CLff) (Table 18). In one animal preparation (ewe no. 130), the placental extraction ratio (ER) of diphenhydramine was calculated by the equation: Placental ER = (FA - Umv)/(FA) (eq.27) where FA and Umv are the steady-state fetal femoral arterial and umbilical venous drug concentrations after fetal infusion, respectively. The ER of diphenhydramine across the placenta on the fetal side was calculated to be 0.64. Separately, the extraction ratio of the drug by the fetus was determined in the same animal after maternal drug infusion: Fetal ER = (Umy - FA)/(Umv) (eq. 28) The fetal extraction ratio of diphenhydramine calculated by this equation was 0.33. Absolute amounts of diphenhydramine eliminated from the oo oo maternal compartment (AUC^Q *CLMO) and fetal compartment (AUCp0-CLf0) were calculated for each animal and are shown in Table 21. Following maternal infusion, the amount of diphenhydramine eliminated from the pregnant ewe and fetus averaged 78.6 ± 4.3 and 1.2 ± 0.6 mg, respectively. Whereas, following fetal infusion, the amount of the drug eliminated from the maternal and fetal compartments averaged 14.7 + 6.2 100 Table 21. Area under the plasma diphenhydramine concentration vs. time GO 00 curves in the ewe (AUC^0) and fetus (AUCp0) and amount of the drug that was eliminated from the ewe and fetus via the nonplacental pathway after maternal and fetal infusions. AUCFo.CLfo.100/ Ewe AUC AUCWC LLL AUC AUC ^CL Mo Mo mo Fo F f0 No. AUC CL ( Mo' mo + AUCF;-CLf0) (Mg.h/L) (mg) (/ig-h/L) (mg) Maternal Infusion 121 698.9 77.2 81.4 0.5 0.6 125 416.3 76.7 63.6 0.5 0.6 130 359.2 76.8 57.6 1.0 1.2 133 398.6 75.3 145.8 2.0 2.6 138 445.4 88.5 121.4 1.7 1.9 202 310.2 80.7 58.8 0.8 0.9 204 455.3 76.0 85.0 1.7 2.2 480 276.5 77.9 125.4 1.1 1.4 420.1 78.6 92.4 1.2 1.4 (129.0) (4.3) (34.1) (0.6) (0.8) Fetal Infusion 121 128.2 14.2 1456.9 9.0 38.8 125 57.8 10.7 1562.4 11.5 51.8 130 47.6 10.2 1006.5 16.8 62.3 133 55.5 10.5 681.7 9.3 47.1 138 89.7 17.8 1063.3 14.9 45.6 202 109.7 28.5 926.0 12.0 29.5 204 86.8 14.5 403.9 8.2 36.2 480 38.6 10.9 1179.0 10.5 49.0 76.7 14.7 1035.0 11.5 45.0 (31.9) (6.2) (380.5) (3.0) (10.2) Numbers in brackets are ± 1 S.D. 101 and 11.5 ± 3.0 mg, respectively. The mean percentage of diphenhydramine administered to the ewe that is eliminated by the fetus via the nonplacental pathway, AUCFo'CLfo-100/(AUCMQ-CLmo + AUCFQ-CLfo), averaged 1.4 ± 0.8%. However, in the case of fetal drug infusion, the percentage of the dose administered to the fetus that is eliminated from the fetus via the nonplacental pathway averaged 45.0 ± 10.2%. 3.4.2 Fetal Effects of Diphenhydramine in Pregnant Sheep 3.4.2.1. Maternal Diphenhydramine Infusion Maternal i.v. infusions of diphenhydramine were given to 8 ewes at gestational age ranging from 127-136 (mean, 131 ± 3) days (Table 11). Diphenhydramine concentrations in the ewe and fetus averaged 212.1 ± 67.8 (range, 140.3-360.3) and 36.3 + 14.4 (range, 18.3-56.0) ng/mL, respectively (Table 12). In the control period, maternal arterial pH, Pco2 and Po2 averaged 7.499 ± 0.038, 34.1 ± 2.8 mm Hg and 114.2 ± 14.8 mm Hg, respectively (Table 22). The fetal control values were 7.388 ± 0.037, 44.6 ±3.3 mm Hg and 19.9 ± 3.9 mm Hg, respectively (Table 23). There was a significant rise in fetal Pco2 during the infusion and decreases in 02 content and hematocrit after the infusion (Table 23). Control fetal arterial pressure and heart rate averaged 47.4 ± 3.0 mm Hg and 145.5 ± 18.5 beats/min, respectively (Table 24). There were no changes in either variable during or after the infusion period. 102 Table 22. Maternal arterial pH, Pco2, Po2, 02 content and hematocrit before, during and after diphenhydramine infusion to the ewe and fetus (n = 6). Time pH Pco2 Po2 Oo Content Hematocrit (min) (mm Hg) (mm Hg) (%) (%) Maternal Infusion -5 7.499 + 0.038 34, .1 + 2.8 114.2 + 14.8 13,. 7 + 0.8a 29. 3 + 3. 4 5 7.483 + 0.024 35, .3 + 1.6 115.1 + 15.9 13,. 6 ± 0.7 28. 7 + 3. 7 30 7.487 + 0.034 33, .7 + 2.7 117.9 ± 19.2 13 .2 + 1.1 28. 6 + 1. 7 60 7.503 + 0.034 32, .0 + 2.7 125.5 ± 31.5 13,. 1 + 1.4 29. 9 + 2. 4 90 7.476 + 0.027 35 .5 + 4.1 118.1 ± 30.0 12 .3 + 0.9 28. 4 + 3. 3 p60 7.484 + 0.023 32 .5 + 2.2 123.4 ± 13.1 12 .8 + 0.5 28. 3 + 2. 8 Fetal Infusion -5 7.473 + 0.033 33 .7 ± 4.4 117.4 ± 5.4 13 .0 + 1.76 27. .7 ± 3. .7 5 7.487 + 0.053 34 .0 + 4.0 113.7 ± 5.5 12 .5 + 1.6 26. .7 + 3. ,5 30 7.477 + 0.026 33 .8 + 4.0 123.6 ± 9.3 12 .2 + 1.4 28, .0 + 3. ,4 60 7.477 + 0.025 33 .3 + 3.5 119.2 ± 17.3 12 .6 + 1.6 27, .8 + 3. .2 90 7.488 + 0.043 33 .0 + 3.6 120.7 ± 8.7 12 .6 + 1.6 28, .2 ± 3. ,4 p60 7.491 + 0.066 33 .8 + 6.1 127.3 ± 12.2 12 .7 ± 1.5 28 .1 + 3. .6 a. n = 3 for 02 content measurements in the maternal infusion study. b. n = 5 for 02 content measurements in the fetal infusion study. 103 Table 23. Fetal arterial pH, Pco2, Po2, 02 content and hematocrit before, during and after diphenhydramine infusion to the ewe and fetus (n = 8). Time pH Pco2 Po2 02 Content Hematocrit (min) (mm Rg) (mm Hg) (%) (%) Maternal Infusion -5 7.388 ± 0.037 44.6 + 3.3 19.9 ± 3.9 7.2 + 1.6a 36.4 + 3.1 5 7.367 ± 0.030 47.4 + * 17.8 + 5.0 6.4 + 1.7 36.8 + 3.2 3.8 30 7.371 ± 0.034 45.9 + 2.4 19.4 + 4.0 7.3 + 1.2 35.1 + 3.3 60 7.358 ± 0.069 45.7 + 3.1 18.6 + 3.9 6.8 + 1.3 35.9 + 3.1 90 7.381 ± 0.035 48.1 + 2.9 18.8 + 4.8 6.1 + 1.7 35.5 + 3.0 p60 7.391 ± 0.035 46.4 + 3.2 19.3 ± 4.7 5.7 + 1.0 34.5 + 2.8 Fetal Infusion -5 7.396 ± 0.074 45.8 + 5.5 19.7 + 5.6° 7.5 + 2.5 37.1 + 3.5 5 7.353 ± 0.045* 47.3 + 4.3 16.4 + 5.2* 5.2 + 2.1* 37.6 + 3.9 30 7.360 ± 0.047* 44.9 + 3.6 18.2 + 5.6 6.0 + 2.5* 36.6 + 3.2 60 7.358 ± 0.041* 44.1 + 5.0 18.8 + 5.5 5.8 + 2.0* 36.4 + 2.9* 90 7.380 ± 0.055 44.1 + 2.6 20.0 + 6.2 6.5 + 2.3 36.3 + 1.7* p60 7.375 ± 0.041 42.9 + 3.7* 21.1 ± 6.4 7.0 ± 2.2 34.8 + 2.4* a. n = 4 for 02 content measurements in the maternal infusion study. 6. n = 6 for Po2 measurements in the fetal infusion study. *p<0.05 104 Table 24. Effects of diphenhydramine on the fetal heart rate and arterial pressure calculated over 30 min intervals before, during and after drug infusions to the ewe and fetus (n = 8). Time Intervals Maternal Infusion Fetal Infusion Fetal heart rate (beat/min) -60 ~ -31 min 144 + 17 156 ± 28 -30 - -1 min 147 + 20 152 ± 20 1 ~ 30 min 146 + 14 182 ± 34* 31 ~ 60 min 144 + 17 160 ± 20 61 - 90 min 145 ± 14 159 ± 23 91 - 120 min 144 + 17 155 ± 20 121 ~ 150 min 143 + 17 149 ± 14 Fetal arterial pressure (mm Hq) -60 ~ -31 min 46.5 + 3.2 51.2 ± 6.5 -30 ~ -1 min 48.2 + 2.8 49.9 ± 7.4 1 ~ 30 min 46.9 + 4.5 51.2 ± 7.9 31 - 60 min 46.0 + 5.4 50.9 ± 9.1 61 ~ 90 min 46.8 + 4.8 51.6 ± 7.9 91 ~ 120 min 46.3 + 4.3 50.8 ± 7.6 121 ~ 150 min 45.9 + 5.1 48.6 ± 9.9 *P<0.05 105 In the control period, all fetuses exhibited regular alterations between low and high voltage electrocortical activity (Table 25, Fig. 21-22). There were also brief periods of an intermediate voltage pattern, usually occurring in the transitions between low and high voltage patterns. The percentage of time spent in low, high and intermediate voltage ECoG patterns averaged 55 ± 7, 35 ± 9 and 10 ± 6, respectively (Table 25). Fetal breathing activity occurred for 42 ± 9% of the time and this was largely associated with low voltage ECoG activity, as were REM (Table 26). During the infusion period, there was a significant decline in the percentage of low voltage ECoG pattern (55 ± 7 to 46 ± 12%) (Table 25), the percentage of low voltage ECoG activity containing REM (80 ± 11 to 55 ± 24%) (Table 27), in the overall incidence of fetal breathing (42 ± 9 to 21 ± 15%) (Table 26) and in the amount of breathing during low voltage ECoG activity (67 ± 13 to 36 ± 28%) (Table 26). There were no changes in the average durations of low and high voltage ECoG episodes (Table 28). After cessation of the infusion, all the monitored behavioural and breathing variables returned to their control ranges. All ewes were standing during and after the period of drug administration. There were no discernable alterations in maternal behaviour; in particular there was no evidence of drowsiness. 3.4.2.2. Fetal Diphenhydramine Infusion Fetal i.v. infusions of diphenhydramine were carried out in 8 sheep at gestational ages ranging from 126-136 (mean, 132 ± 4) days (Table 11). Maternal and fetal steady-state diphenhydramine 106 Table 25. Overall incidence of low, high and intermediate voltage ECoG episodes in the fetus prior to, during and after infusion of diphenhydramine to the ewe and fetus (n = 8). % electrocortical activity Drug Infusion Low High Intermediate Maternal Infusion Prior to Infusion 55 + 7 35 + 9 10 + 6 During Infusion 46 + 12* 41 + 13 13 + 14 After Infusion 56 + 9 39 + 11 5 + 3 Fetal Infusion Prior to Infusion 51 ± 10 41 + 10 8 + 4 During Infusion 26 + 20 28 + 21 46 ± 31* After Infusion 50 + 12 34 + 6 17 + 15 *p<0.05 HIGH XX) LOW i^^^M^^ INTBWIEDIATE High speed chart recordings of high, low and intermediate voltage ECoG patterns in a fetal lamb. 108 k. Control Period 5 min TP 25r 4^ (mmHej) o>- ECoG 100(A/C EOG iocw [ |» , i In ^|I#^M*1 B. Fetal Infusion Period TP 25r (mmHg) oL ECoG 10CW [ .•j Uli. «n.l«|.> te|4i EOG 1CXW [ i r Fig. 22. Representative examples of the recording of breathing, ECoG and electro-oculographic (EoG) activity in the fetal lamb during the control period (A) and during fetal infusion of diphenhydramine (B). TP, tracheal pressure. 109 Table 26. Percentage of fetal breathing in low, high and intermediate voltage ECoG prior to, during and after infusion of diphenhydramine to the ewe and fetus (n = 8). % Fetal Breathing in ECoG Episodes Total Drug Infusion Breathing (%) Low High Intermediate Maternal Infusion Prior to Infusion 67 + 13 5 + 4 27 + 23 42 + 9 * •k During Infusion 36 + 5 + 8 32 + 46 21 + 15 28 After Infusion 59 + 25 3 + 4 21 + 34 37 + 15 Fetal Infusion Prior to Infusion 70 + 22 4 + 7 20 + 18 44 + 12 During Infusion 54 + 37 11 + 10 51 + 26* 47 + 22 After Infusion 53 + 24 8 + 8 16 + 15 34 + 13 *p<0.05 110 Table 27. Percentage of fetal electro-ocular activity (EoG) in low, high and intermediate voltage ECoG prior to, during and after infusion of diphenhydramine to the ewe and fetus. % Fetal EOG spent in ECoG Episodes Total EOG Drug Infusion (%) Low High Intermediate Maternal Infusion (n - 5) Prior to Infusion 80 ± 11 0 21 + 19 47 + 10 During Infusion 55 ± 24 0 34 + 33 35 + 13 After Infusion 86 ± 6 5 + 7 5 ± 7 48 + 12 Fetal Infusion (n = 3) Prior to Infusion 84 ± 9 2 + 3 8 + 8 47 + 1 During Infusion 85 ± 13 5 + 9 63 + 21* 53 + 26 After Infusion 87 ± 2 1 + 1 9 + 12 49 + 8 *p<0.05 Ill Table 28. Duration of low, high and intermediate voltage ECoG episodes in the fetus prior to, during and after infusion of diphenhydramine to the ewe and fetus (n = 8). Mean Duration of ECoG Episodes (min) Drug Infusion Low High Intermediate Maternal Infusion Prior to infusion 16.4 ± 2.3 11.6 ± 3.0 5.8 ± 3.6 During Infusion 13.7 ± 5.0 10.8 ± 3.1 4.6 ± 2.1 After Infusion 20.9 ± 4.5 13.9 ± 3.7 6.4 ± 5.6 Fetal Infusion Prior to Infusion 15.0 + 4.0 13.2 + 3.8 5.9 + 1.5 During Infusion 8.8 ± 3.8 10.8 + 7.7 7.9 + 5.6 After Infusion 15.4 + 4.9 11.1 + 2.9 9.1 + 5.2 *p<0.05 112 concentrations averaged 31.1 ± 11.6 (range, 17.9-53.9) and 447.6 ± 185.2 (range, 192.1-697.9) ng/mL, respectively (Table 12). Fetal arterial Po2 fell from 19.7 ± 5.6 to 16.4 ± 5.2 mm Hg in the first 5 min of the infusion and recovered thereafter. There was a decline in both pH and 02 content, which was maintained for the first 60 min of drug administration (Table 23). There was also a decrease in hematocrit following drug administration. Fetal Pco2 was unchanged, except for a slight fall at the end of the experiment. Maternal blood gas variables in the control period averaged 117.4 ± 5.4 mm Hg, 33.7 ± 4.4 mm Hg and 7.473 ± 0.033 for Po2, Pco2 and pH, respectively (Table 22). There were no changes during or after the infusion period. Fetal arterial pressure and heart rate are given in Table 24. There was a significant increase in heart rate during the first 30 min of the infusion period, from 154.0 ± 24.0 to 182.0 ± 34.0 beats/min, with a return to control levels thereafter. Arterial pressure was not significantly changed. Fetal ECoG, electro-ocular and breathing variables are illustrated in Table 25-27. In the control period, the percentages of the time spent in low, high and intermediate voltage patterns were similar to the values obtained before the maternal diphenhydramine infusions. During drug infusion to the fetus, there was a significant fall in the amount of low voltage ECoG pattern (51 ± 10 to 26 ± 20%) and a similar tendency for the amount of high voltage activity to decline (41 ± 10 to 28 ± 21%), although this change was not statistically 113 significant (Table 25). Conversely, there was a marked rise in the incidence of the intermediate voltage pattern, from 8 ± 4 to 47 ± 31%, and also a large rise in the occurrence of REM during intermediate voltage ECoG activity (8 ± 8 to 63 ± 22%) (Table 25 and 27). The increase in the amount of this intermediate voltage pattern resulted in a disruption in the normal cycling between high and low voltage episodes (Fig. 22). This was reflected in a marked fall in the average duration of low voltage episodes during the infusion period, from 15.0 ± 4.0 to 8.8 ± 3.8 min (Table 28). The overall incidence of fetal breathing movements in the control period averaged 44 ± 12% and was not altered during the infusion. However, there was a significant increase in the amount of fetal breathing during intermediate ECoG activity and this was accompanied by a tendency for a fall in the incidence of breathing during the low voltage ECoG pattern (Table 26). Also, in seven of the eight experiments, there was an appearance of large amplitude fetal breathing movements within 1 min of drug administration, which lasted for 5 to 20 min (Fig. 23). In these seven fetuses, arterial Po2 had fallen by 3.9 ± 0.9 mm Hg by 5 min after the start of diphenhydramine infusion. In contrast, in the fetus in which large amplitude breathing activity did not occur, Po2 rose by 1 mm Hg at 5 min. 3.4.2.3. Control Saline Infusion The mean values for fetal blood gases, heart rate, arterial pressure, behavioural and breathing parameters in these four experiments were not different from those found before diphenhydramine infusion. Furthermore, there were no changes in any of these variables during 114 ARTEtlAl nu&sutE HEART BATE | TtACMEU PRESSURE (mmHg) FTO mm mm ECOG 100. J EOG WO, •ft-*. , I 1 "fi'1" ft"1'*' I Fig. 23. Examples of the large amplitude breathing activity present in the initial phases of diphenhydramine infusions to the fetus. The arrow indicates the time at which the infusion was started. The drug was given by simultaneous i.v. bolus injection (5 mg) and infusion (0.17 mg/min for 90 min) to steady-state. 115 infusion of 0.9% saline to the fetus or ewe. 3.5. Recirculation of Diphenhydramine from Amniotic Fluid to the Ewe and Fetus: Intra-Amniotic bolus Injection Studies A 50 mg dose of diphenhydramine hydrochloride was injected as a bolus into the amniotic cavity in 5 time-dated pregnant sheep. Characteristics of individual ewes employed in the study are summarized in Table 29. The gestational age at the time of the experiment ranged from 122-137 days (mean, 129 ± 7 days) and the maternal body weight ranged from 63-85 kg (mean, 76 ± 8 kg). Control values for the fetal arterial blood gases, P02, PCO2 and 02 content averaged 22.3 ± 7.6 mm Hg, 47.3 ± 9.3 mm Hg and 6.8 ±2.5 vol%. Average control fetal arterial blood pH and hematocrit were 7.391 ± 0.056 and 31.2 ± 3.6 vol%. All these variables are within normal range in near-term fetal lamb. Drug administration into the amniotic cavity resulted in no significant change in these variables when examined at 5, 10, 30 and 60 min after drug injection. Fig. 24 shows average drug concentration vs. time curves for maternal and fetal arterial plasma, fetal tracheal and amniotic fluids obtained following drug administration into the amniotic cavity in 5 pregnant sheep. The highest drug concentration in the amniotic fluid was observed at the first sampling time (5 min) in all the animals, with a mean Cmax of 82,467.5 ± 80,918.0 ng/mL (Table 30). The observed mean Cmax in fetal plasma (228.0 ± 101.5 ng/mL) was significantly higher than that in maternal plasma (43.5 ± 15.6 ng/mL) (Table 30). Also, the drug 116 Table 29. Physical characteristics of individual pregnant sheep used in diphenhydramine intra-amniotic bolus injection study. Ewe Breed Maternal Gestational Age Fetal Number No. body (days) Body Weight of Weight in uteror Fetuses (kg) Surgery Experiment (kg) 204 Dorset 62.7 122 137 3.2 1 208 Suffolk 80.9 117 126 1.2 3 279 Dorset 74.5 116 124 1.6 2 283 Dorset 79.1 128 136 3.1 1 287 Dorset 84.5 118 122 1.4 2 76.3 120 129 2.1 (8.4) (5) (7) (1.0) a. Calculated by the equation: log (fetal body weight) = log (birth weight) - (0.0153 x number of days between experiment and delivery). 6. Numbers in brackets are ± 1 S.D. 117 Fig. 24. Average diphenhydramine concentration vs. time curves for maternal arterial (O) and fetal arterial (•) plasma, fetal tracheal (A) and amniotic (A) fluids following administration of 50 mg diphenhydramine hydrochloride into the amniotic cavity into 5 pregnant sheep. 118 Table 30. Values for the observed maximum drug concentrations (Cmax) and time to reach Cmax (Tmax) in maternal (MA) and fetal (FA) arterial plasma and fetal tracheal (TF) and amniotic (AF) fluids following injection of 50 mg diphenhydramine hydrochloride into the amniotic cavity. Ewe (ng/mL) T (min) °max max No. MA FA TR AMN MA FA TR AMN 204 29.9 188.8 769.9 213700.9 90 90 300 5 208 38.6 129.8 3971.4 85991.2 90 120 10 5 279 66.3 150.3 1760.5 81449.8 45 90 45 5 283 52.1 355.5 5952.4 13137.3 45 90 10 5 287 30.4 315.7 1475.4 180585.4 90 90 45 5 43.5 228.0 2785.9 82467.5 72 96 82 5 (15.6) (101.5) (2136.9) (80918.0) (25) (13) (123) (0) Values in brackets are ± 1 S.D. 119 concentration in fetal arterial plasma was consistently and significantly higher than the corresponding value in maternal arterial plasma throughout the sampling time, with the mean maternal to fetal plasma drug concentration ratio of 0.23 ± 0.10 (Fig. 24 and Table 31). The Cmax in tracheal fluid (2,785.9 + 2,136.9 ng/mL) was significantly greater than that either in the maternal or in the fetal plasma. The time taken to reach the Cmax (Tmax) averaged 1.2 ± 0.4 hr in maternal plasma and 1.6 ± 0.2 hr in fetal plasma (Table 30). The average elimination half-life of the drug in maternal plasma (1.7 ± 0.5 hr) was significantly shorter than that in fetal plasma (7.7 + 1.2 hr) (Table 31). The half-life in fetal tracheal and amniotic fluids, 11.0 ± 4.5 hr and 10.8 ± 3.8 hr, respectively, was significantly longer than that in either the fetal or the maternal plasma. In one animal (ewe no. 287), drug concentrations were measured in umbilical venous plasma as well as maternal and fetal arterial plasma, fetal tracheal and amniotic fluid, and the drug concentration-time curves obtained in this animal are shown in Fig. 25. As seen in the other animals, drug concentrations in fetal plasma were significantly higher than in maternal plasma throughout the sampling time. Furthermore, the drug concentration in umbilical venous plasma was higher than in fetal arterial plasma for most of the sampling period and the average difference in concentration was statistically significant (Fig 25). Total area under the drug concentration-time curves for each fluid calculated by the equations 4 and 5 and the amount of the drug eliminated from the maternal and fetal compartments are shown in Table 32. The average AUCpQ is -10 times greater than the CO average AUCMo. However, the amount of the drug eliminated from the 120 Table 31. Average ratios of the maternal arterial (MA), umbilical venous (UV) and fetal tracheal (TF) to fetal arterial (FA) drug concentrations and the elimination half-life (t^) of the drug obtained after injection of 50 mg diphenhydramine hydrochloride into the amniotic cavity. a Ewe Concentration Ratio tk (hr) No. MA/FA UV/FA TR/FA MA FA UV TF AF 204 0.16 - 4.31 2.0 8.3 - 17.2 10.2 (0.09)D (4.39) 208 0.33 - 45.85 1.6 7.8 - 8.2 9.5 (0.19) (113.66) 279 0.33 - 6.10 1.1 9.3 - 9.7 18.8 (0.16) (3.15) 283 0.23 - 21.47 1.3 6.1 - 10.6 9.3 (0.24) (30.55) 287 0.11 2.01 4.18 2.3 7.1 7.5 8.1 7.2 (0.08) (2.31) (3.11) 0.23 - 16.38 1.7 7.7 - 10.8 11.0 (0.10) (17.99) (0.5) (1.2) (3.8) (4.5) a. t\ is determined by linear regression analysis. b. Values are expressed as the mean (± 1 S.D.). 121 0 4 8 12 16 20 24 28 Time (hr) Fig. 25. Representative diphenhydramine concentration vs. time curves for maternal (o) and fetal arterial (•) and umbilical venous plasma (•), fetal tracheal (A) and amniotic (A) fluids following intra-amniotic bolus injection of 50 mg diphenhydramine hydrochloride to a pregnant ewe (ewe no. 287). Inset figure shows the fetal arterial and umbilical venous plasma and fetal tracheal drug concentration vs time curves over the initial 6 hr period after drug administration. 122 Table 32. Total area under the drug concentration vs. time curves for the maternal and fetal arterial and umbilical venous plasma, fetal tracheal and amniotic fluids after injection of 50 mg diphenhydramine hydrochloride into the amniotic cavity. Ewe No. Parameter mean (± S.D.) 204 208 279 283 287 AUCMp 120.2 156. 0 191 .4 159 .0 156.9 156.7 (/zg-hr/L) (25.2) AUCpp 1345.2 1233. 2 1064 .1 1636 .7 2108.6 1477.6 (/zgchr/L) (409.8) - - - - 2987.6 - (/zg-hr/L) 10064.8 10570. 5 6977 .3 52233 .7 10807.0 18130.7 (/zg!hr/L) (19126.3) 127180.0 7711. 5 49540 .1 35263 .4 126182.3 87175.5 (/zg-hr/L) (42849.1) AUCMo-CLmo 24.1 31. 3 38 .4 31 .9 31.0 31.4 (mg) (5.1) 0.6 0. 8 1 .0 0 .8 0.8 AUCyo-CLmf 0.8 (0.1) AUC "-CLf0 16.8 15. 4 13 .3 20 .5 26.4 18.5 (mg) (5.1) AUCFo'CLfm 21.3 19. 6 16 .9 26 .0 33.5 23.4 (mg) (6.5) Mean values (n = 8) for the maternal and fetal nonplacental clearances, CLmo 3343.8 mL/min and CLf0 208.4 mL/min, and the transplacental clearances between the ewe and fetus, CLmf 82.4 mL/min and CLfm 264.4 mL/min, were used in the calculation (see section 3.4.1.4). 123 fetus via the nonplacental pathway (18.5 ± 5.1 mg) was significantly less than that eliminated from the maternal sheep (31.4 ± 5.1 mg). The sum of the amount of drug eliminated from both the maternal and fetal compartments averaged 49.9 ± 6.4 mg. 3.6. Pulmonary Extraction of Diphenhydramine in Fetal Lamb Following Maternal or Fetal Infusion to Steady-State Surgery was performed in 16 time-dated pregnant sheep in the third trimester of gestation, out of which, 4 were employed in a drug infusion study and the rest (n = 12) were excluded since the pulmonary venous catheter malfunctioned. Maternal or fetal diphenhydramine infusion experiments were conducted in the 4 animal preparations over the gestational age ranging from 128-133 days. After the drug infusion experiments, the surgically prepared fetal lambs were subjected to autopsy to confirm that the pulmonary arterial and venous catheters were properly placed. In one surgical preparation, the pulmonary venous catheter tip was found to be positioned in the left atrium and the animal was subsequently excluded from the study. Fig. 26 and 27 show representative plots of the fetal pulmonary arterial, pulmonary venous and carotid arterial drug concentration vs. time profiles following maternal and fetal infusions, respectively. Steady-state drug concentrations were calculated as the mean of the drug concentrations in 60, 75 and 90 min samples (Table 33). During maternal infusion, steady-state drug concentration in carotid arterial plasma was higher than in pulmonary arterial plasma (60.6 vs. 51.2 ng/mL). Drug 124 Fig. 26. Plots of the fetal pulmonary arterial (PA), pulmonary venous (PV) and carotid arterial (CA) and maternal arterial (MA) diphenhydramine concentration vs. time profiles following maternal drug infusion. 125 Fig. 27. Representative plots of the fetal pulmonary arterial (PA), pulmonary venous (PV) and carotid arterial (CA) and maternal arterial (MA) diphenhydramine concentration vs. time profiles following fetal drug infusion. Table 33. The steady-state diphenhydramine concentrations and levels in the hour following cessation of the maternal or fetal infusion in fetal pulmonary arterial (PA), pulmonary venous (PV), carotid arterial (CA) blood, fetal tracheal (TF) and amniotic (AF) fluids and maternal arterial blood (MA). Ewe Route of Sampling Drug Cone. (ng/mL) Drug Cone. (ng/mL) No. Administration Site at Post Infusion Steady-State 30 min 60 min 228 Infusion PA 726.7 323.3 145.0 Fetal PV 621.2 342.7 153.3 CA 521.0 278.2 127.7 TF 2278.6 1474.1 1414.7 AF - - - MA 31.8 16.6 8.0 148 Infusion PA 470.1 200.0 74.9 Fetal PV 451.0 246.6 77.3 CA 437.0 185.2 71.7 TF 2039.3 993.2 593.3 AF 78.9 74.1 86.9 MA - - - 225 Infusion PA 51.2 27.7 13.1 Maternal PV 48.6 23.7 10.4 CA 60.0 28.0 12.9 TF 70.0 53.4 14.9 AF 6.2 10.7 9.3 MA 154.3 36.9 16.9 127 concentration in the pulmonary arterial plasma was consistently higher than in the pulmonary venous plasma throughout the experiment (Fig. 26 and Table 33). In contrast, during fetal infusions, steady-state drug concentration in pulmonary arterial plasma (mean, 598.4 ng/mL) was higher than in carotid arterial plasma (mean, 479.0 ng/mL) (Fig. 27 and Table 33). Diphenhydramine in pulmonary arterial plasma was higher than in pulmonary venous plasma during the fetal infusion period (598.4 vs. 536.1 ng/mL). Following cessation of the fetal drug infusion, however, there was a reversal of the drug concentration gradient, resulting in higher levels in pulmonary venous plasma than in pulmonary arterial plasma (205.0 vs. 186.1 ng/mL) (Table 33). The pulmonary extraction ratio of diphenhydramine by the lung calculated by equation 26 ranged from 0.04 to 0.14 (mean, 0.08 ± 0.06) (Table 34). Following either maternal or fetal drug infusion, fetal tracheal drug concentration was higher than in fetal or maternal plasma, whereas amniotic drug concentrations were lower than the corresponding concentrations in maternal or fetal plasma (Table 33). Maternal and fetal total body clearances were calculated by equations 24 and 25 (Table 34). Maternal total body clearance is more than -10 times greater than fetal clearance (4402.1 vs. 357.7 mL/min). However, when normalized by maternal or fetal body weight, the CLff (188.6 mL/min/kg) is greater than CLmm (53.8 mL/min/kg) (Table 34). 128 Table 34. Fetal pulmonary extraction and the total body clearance of diphenhydramine calculated at steady-state after maternal or fetal drug administration. Ewe No Route of Fetal Pulmonary3 Total Body Clearance Administration Extraction (mL/min) Ratio CLmm CLff 148 Fetal infusion 0.04 - 389.0 (185.2)° 225 Maternal Infusion 0.05 4402.1 (53.8)c 228 Fetal Infusion 0.14 - 326.3 u (191.9)' 0.08 + 0.06 a. calculated by [(PA)-(PV)]xl00/(PA). b. Normalized by the maternal body weight (mL/min/kg). c. Normalized by the fetal body weight (mL/min/kg). 129 4. DISCUSSION 4.1. Development of a Capillary GC/NPD Assay Method Open tubular capillary column GC has been extensively used as an analytical tool in the analysis of complex matrices, such as environmental samples, fossil fuels, food and cosmetics and biological materials (Rooney, 1981). Several glass capillary GC/NPD methods (Chiarotti et al., 1983; Barni Comparini et a/., 1983) have been reported for the determination of diphenhydramine in street drug samples. Although glass capillary columns are considerably more inert than those of metal, they still possess significant column surface activity leading to absorptive interactions with the substances being chromatographed. The recent introduction of inert fused-silica capillary columns minimizes these problems (Lipsky et al. , 1980), providing reproducible and highly sensitive analyses of a variety of drugs and their metabolites in tissues and biological fluids. As a consequence, these methods are more frequently used for pharmacokinetic studies, where small plasma volumes (0.5 mL or less) are encountered. Lutz et al. (1983) reported a GC method for diphenhydramine measurement in human serum using a bonded-phase fused-silica capillary column. Meatherall and Guay (1984) later reported a GC/NPD method for the determination of diphenhydramine employing direct sample injection onto a fused-silica capillary column. These methods provide good sensitivity and selectivity, however, they still required the use of large plasma volumes (1.0-3.0 mL) and a very small final sample 130 reconstitution volume (10 /zL). In the present research, a GC/NPD method with improved sensitivity was developed. Splitless injection was employed to facilitate the automation of sample analysis. This injection mode has been found to be useful for trace analysis of dilute samples without pre-concentration and for samples with components eluting near the solvent peak (Grob and Grob, 1969a, 1969b, 1974). A purge activation time of 1 min was chosen since longer purge activation times did not provide increased peak area counts, suggesting no further transfer of sample onto the column. As isothermal column operation caused peak broadening, multi-step temperature programming was used to improve peak shape and the resolution of the drug and internal standard. Orphenadrine was chosen as an internal standard because of its structural similarity to diphenhydramine. Resolution of diphenhydramine and orphenadrine is excellent, as illustrated in Fig. 9. The developed assay method has been found to show good linearity over the concentration range studied (Table 1). Within-run precision (repeatability) of a representative calibration curve showed good reproducibility with coefficients of variation ranging from 1.5 to 10.0% (Table 1). The sensitivity limit of this assay method (-40 pg at the detector) was found to be suitable for the subsequent pharmacokinetic assessment of plasma diphenhydramine concentration vs. time profiles in chronically catheterized maternal and fetal sheep. 131 Adsorption of diphenhydramine from the organic solvent, by glass surfaces, was observed to be a major source of drug loss during assay development. Solvents of various polarities, including heptane, hexane, toluene, benzene and dichloromethane, were examined for their extraction efficiency. Dichloromethane was found to be the most efficient solvent for single-step plasma extraction of diphenhydramine, providing low-level drug quantitation without any interference from plasma constituents. The addition of triethylamine in dichloromethane (0.01 M) minimized drug adsorption losses during extraction and evaporation processes. The formation of an emulsion between plasma and dichloromethane during extraction occurred, but was effectively broken by placing samples in a freezer (-20°C) for 5 min prior to centrifugation. Since chlorinated solvents cause a reversible nitrogen/phosphorus detector sensitivity loss, toluene was used instead of dichloromethane for the reconstitution of the plasma extracts for sample injection. A significant carry-over of diphenhydramine and orphenadrine from the automatic sampler syringe needle was observed when toluene was used to wash the syringe. Among various solvents examined, acetone was found to eliminate the carry-over problem when the automatic sampler performed four wash cycles. The use of automatic liquid sample introduction has made it possible to perform routine analysis for the diphenhydramine pharmacokinetic studies. 132 4.2. Pharmacokinetics of Diphenhydramine after Dose Ranging in Nonpregnant Sheep 4.2.1. Dose-Linearity of Pharmacokinetics of Diphenhydramine in Nonpregnant Sheep Distribution of diphenhydramine after i.v. bolus injection was rapid in sheep, with a mean distribution half-life of 5-9 min. The average volume of distribution (Vdss) obtained over the dose range studied (2.8-5.5 L/kg) far exceeds the total body water in sheep (-0.6 L/kg) (Macfarlane, 1975). This suggests that despite moderately high plasma protein binding (77%), diphenhydramine distributes extensively into body tissues. In other animal species, such as rats and monkeys, diphenhydramine exhibits a very rapid distribution into tissues, with the maximum tissue uptake occurring at 1-3 min after i.v. injection (Drach et al., 1970; Wagner, 1973). In rats, tissue drug uptake appears to be saturated over the dose range of 4-16 mg/kg, and the elimination half-life tends to decrease with increasing dose (Wagner, 1973). The decrease in the elimination half-life in the rat could be a reflection of the tissue saturation or a decrease in tissue binding of the drug. In the present study, however, the drug dose employed was low (0.2-2.9 m 9Ag) and there was a trend towards an increase in the mean Vdss (from 2.8 to 5.5 L/kg) and elimination half-life (from 34 to 68 min) as the dose was increased. Therefore, it is not likely that tissue saturation occurred in the sheep. The extent of plasma protein binding of the drug did not change over the drug concentration range, 10-2,000 ng/mL in vitro. Since changes in apparent volume of distribution (Vd) are a function of both changes in plasma and tissue binding, it may be 133 possible that tissue drug uptake or binding was increased with no change in plasma protein binding (Gibaldi et al., 1978). OO In the nonpregnant sheep, neither dose-normalized AUC0 nor the total body clearance (C/.Tg) was affected by dose. However, the elimination half-life is dependent upon both C/.Tg and the volume of distribution (0.693IVC//C/.JB) and since C/.Tg remained unchanged while the volume of distribution increased with increasing dose, the net effect was an increase in the elimination half-life. The average C/LTg of approximately 5 L/hr/kg is 1.5 times the liver blood flow (-3.3 L/hr/kg) (Katz and Bergman, 1969) and is close to the total cardiac output (-5 L/hr/kg) (Runciman et al., 1984) in sheep. Diphenhydramine is, therefore, likely to be cleared by the liver as well as by other tissues and organs. Diphenhydramine CZ.Tg in sheep is at least -3 times greater than that found in humans (Spector et al., 1980; Berlinger et al., 1982; Meredith et al., 1984). A similar more rapid clearance in sheep, as compared with humans, has also been reported for a number of other drugs, including etidocaine (Pedersen et al., 1982), lidocaine (Bloedow et al., 1980), meperidine (Szeto et al., 1978) and metoclopramide (Riggs et al., 1988a). In the case of diphenhydramine, the high CZ.Tg in sheep is consistent with a much shorter elimination half-life than in humans since the volumes of distribution of diphenhydramine in these species are similar. In the present study, there was a transient decrease in Po2 at 5 min following injection of either a 100 or a 200 mg dose. There was no change in 02 content and hematocrit over the dose range of 25-100 mg 134 but after 200 mg there were significant decreases in these parameters. The increase in 02 content is likely the result of the increase in hematocrit which in turn may be mediated by alterations in capillary permeability. 4.2.2. Plasma Protein Binding in Nonpregnant Sheep The present study demonstrates that the plasma protein binding of diphenhydramine is independent of the total drug concentration in the range of 10-2,000 ng/mL in nonpregnant sheep plasma in vitro. Appropriate volumes of the plasma samples were used for drug analysis so that drug concentrations fell within the concentration range of the standard curve. The in vivo determination of protein binding in all the samples collected from the two animals dosed with a 200 mg dose also confirmed that the extent of plasma protein binding did not change over the drug concentration range encountered (30-780 ng/mL). Although the free fraction of diphenhydramine appeared to increase at higher drug concentrations (4,000 and 10,000 ng/mL) in vitro, curve fitting of the binding plot was not attempted since the present binding study was not designed to comprehensively characterize the binding parameters. The extent of plasma protein binding of diphenhydramine in the nonpregnant sheep (77%) seems comparable to that reported in humans (76-98%) (Albert et al., 1975; Spector et al., 1980; Meredith et al., 1984). Diphenhydramine binds to human serum albumin but the degree of binding is low (Drach et al., 1970). Furthermore, there is no correlation between the extent of protein binding of the drug and serum albumin concentrations in man (Meredith et al., 1984). Diphenhydramine, like 135 other basic amine drugs such as lidocaine and propranolol (Gillis et al., 1985), may bind to aj-acid glycoprotein, but we have not been able to measure aj-acid glycoprotein concentrations due to lack of commercially available antisera specific to ovine aj-acid glycoprotein. Further investigation is required to determine whether diphenhydramine binds to aj-acid glycoprotein in human or sheep plasma. In summary, diphenhydramine showed nonlinear distribution characteristics over the dose range of 25-200 mg in the nonpregnant sheep. There were increases in the distribution and elimination half- lives and the volume of distribution, with increasing dose. There was, oo however, a proportional increase in AUC0, with no change in C/.jg. Plasma protein binding of diphenhydramine did not change over the concentration range observed. The change in the volume of distribution may be due to changes in tissue drug uptake or binding. 136 4.3 Pharmacokinetics of Diphenhydramine in Pregnant Sheep Following i.v. Bolus Administration 4.3.1. Placental Transfer of Diphenhydramine and the Extent of Fetal Drug Exposure There appears to be no detailed published studies of the placental transfer of histamine Hj-receptor blockers involving serial blood sampling from the mother and fetus in utero. The present investigation demonstrates rapid and extensive maternal-fetal transfer of diphenhydramine in pregnant sheep. The extent of fetal exposure to a drug is more accurately reflected by the fetal to maternal AUC ratio than by a comparison of fetal and maternal plasma drug concentrations at any time, t, after drug administration (Levy and Hayton, 1973). The fetal to maternal AUC ratio of 0.85 obtained for diphenhydramine indicates significant fetal exposure to the drug after maternal i.v. administration. Drug transfer across the placenta appears to occur by simple diffusion. Diphenhydramine has a relatively low molecular weight (m.w. = 255) and high lipid solubility (pKa = 9.0), both of which favor placental transfer by this process (Reynolds, 1979). 4.3.2. Disposition of Diphenhydramine in Maternal and Fetal Sheep Maternal plasma diphenhydramine concentration vs. time curves show a biexponential decay; a similar pattern of decay has also been observed in humans (Albert et al., 1975; Meredith et al., 1984). However, the elimination half-life is shorter in pregnant sheep (0.7-1.0 hr) than in humans (4.1-9.3 hr), (Abernethy and Greenblatt, 1983; 137 Meredith et al., 1984). As discussed previously (section 4.2.1), a similar, more rapid drug clearance in sheep, as compared with humans, exists for a number of drugs including lidocaine (Bloedow et al., 1980), etidocaine (Pedersen et al., 1982), metoclopramide (Riggs et al., 1988a) and meperidine (Szeto et al., 1978; Kuhnert et al., 1980). In the latter case, there are data allowing comparison between pregnant sheep and pregnant women. Hence, in studies of fetal drug effects employing the pregnant sheep model, drug doses higher than normally used in humans may be required. 4.3.3. Comparisons of Diphenhydramine Pharmacokinetics between Pregnant and Nonpregnant Sheep The distribution and elimination half-lives in the pregnant ewes, 9 and 52 min, respectively, are similar to those obtained after 100 mg i.v. bolus injection of the drug to the nonpregnant ewes (8 and 52 min). In the nonpregnant sheep, there is a tendency for higher values of Vdss (6.5 vs. 4.5 L/kg) and CZ.TB (5.0 vs. 3.6 L/h/kg) and for oo a lower value of AUC0 (229.4 vs. 554.9 jtig.h/L) as compared with pregnant sheep (Table 4 and 9). However, when tested for statistical significance (Mann-Whitney test), there is no significant difference in any of the pharmacokinetic parameters between these two groups. 4.3.4. Extent of Drug Elimination from the Fetal Lamb Following Maternal i.v. Bolus Dosing The percentage of diphenhydramine eliminated irreversibly by the fetus via the nonplacental pathways, e.g., hepatic and/or renal 138 elimination, calculated in the present study averaged 5.0 ± 2.2% of the total drug available in the maternal-fetal sheep (Table 10). The sum of the amount of diphenhydramine eliminated by the maternal (111.4 mg) and fetal (5.6 mg) sheep calculated by this method is greater than is anticipated, i.e., the amount of diphenhydramine administered is 87.5 mg as a free base. However, this discrepancy is not surprising considering that there is a large interindividual variability in AUC in these animals (Table 9) and that the CLmo and CLf0 used in the calculation are from a study employing a different group of pregnant sheep. It is also possible that the drug might have been subject to recirculation within the maternal-fetal sheep (see sections 4.3.4, 4.4.3 and 4.5.4 for further discussion). Nevertheless, it seems that -5% of the total drug available in the maternal-fetal sheep was eliminated from the fetal compartment via the nonplacental pathway. When calculated for metoclopramide (data from Riggs et al., 1987b and 1988a), the mean percentage of the amount of the drug that is irreversibly eliminated by the fetus was -2% of the total drug administered to the pregnant sheep, which is less than, but comparable to, the diphenhydramine data (5%). Therefore, it appears that most of the drug administered to the maternal sheep via bolus dosing is eliminated by the mother. 4.3.5 Drug Accumulation in Fetal Tracheal and Amniotic Fluids Following diphenhydramine i.v. bolus injection to the ewe, fetal tracheal drug concentrations were initially lower than those in fetal plasma. However, by -10-45 min after the drug injection, tracheal diphenhydramine concentrations exceeded the corresponding fetal plasma 139 drug concentrations. The average fetal tracheal to plasma drug concentration ratio of 2.8 indicates that a substantial amount of the drug is accumulated in tracheal fluid. In the fetal lamb during the latter third trimester of gestation, fetal tracheal fluid is produced at a constant rate of approximately 4.5 mL/kg/hr from the lungs (Mescher et a7., 1975). The fetal tracheal fluid is likely to be formed as a result of an active transport of chloride ion from plasma and interstitial fluid across the pulmonary epithelium (Olver and Strang, 1974). In some respects, the fluid resembles an ultrafiltrate of plasma, in regards to Na+ and K+ concentrations. However, concentrations of several other electrolytes, such as Cl" and HCO3", are significantly different from those in fetal plasma. Particularly, the total protein content in tracheal fluid (0.3 g/L) is substantially lower than in fetal plasma (62.7 g/L) (Olver and Strang, 1974). The tracheal fluid is known to leave the lungs, passes along the trachea into the pharynx, and the fluid is then either swallowed by the fetus or secreted into the amniotic cavity (Harding et al., 1984). Possible mechanisms responsible for accumulation of diphenhydramine in fetal tracheal fluid include; 1) secretion of the drug across alveolar epithelium after uptake by alveolar cells, 2) secretion across the trachea and 3) diffusion from saliva/amniotic fluid down the airway. Based on the observations that pulmonary extraction ratio of diphenhydramine [(PA-PV)/PA] examined at steady-state averages 0.08 in fetal lambs (section 3.6) and that uptake of the drug by the 140 lungs is sufficient to account for drug accumulation in fetal tracheal fluid (see section 4.6), secretion of the drug across the pulmonary epithelium may be the major route of drug accumulation in the tracheal fluid. However, considering the low rate of tracheal fluid production in the fetus (-4.5 mL/kg/hr), this process alone may not explain the rapid appearance of diphenhydramine at the sampling site, i.e., the trachea, during the initial sampling period. Secretion of drug across the trachea may be an important mechanism for the initial drug accumulation. In humans, diphenhydramine is known to be secreted in saliva, with a mean saliva to plasma drug concentration ratio of 0.33 (Sharp et al., 1983). Assuming that a similar situation exists in the fetal lamb, secretion via salivary glands and diffusion down the airway may play a minor role in the tracheal drug accumulation. Diffusion from amniotic fluid along the airway may be an important route of drug accumulation in the case where the drug is present in high concentrations in the amniotic fluid and the fluid is ingested by the fetus, e.g., intra-amniotic drug administration (see section 4.5.3). In the fetus, diphenhydramine that is accumulated in the tracheal fluid could be either swallowed by the fetus or tranferred into the amniotic sac for further recirculation of the drug within the fetus. In adults, many amine-containing compounds are extensively cleared by the lung through metabolism, excretion or by irreversible binding (Collins and Dedrick, 1982; Hori et al., 1987). Using isolated lung preparations, significant accumulation of several amine compounds, including propranolol, lidocaine, imipramine, amphetamine and methadone, has been demonstrated (Anderson et al., 1974; Roth and Wiersma, 1977). 141 Following oral or s.c. administration, accumulation of diphenhydramine in the lungs of adult guinea pig, rat and human has also been reported (Glazko and Dill, 1949a; Hausmann et al., 1983). The cell type in lung responsible for accumulation of these drug substances has not been identified for all drugs although pulmonary endothelial cells appear to be involved in the uptake of a number of basic amine drugs (Roth, 1985). With reference to the pulmonary clearance of drugs, the fetus would appear to differ from the adult in at least two respects. Firstly, the excretion of nonvolatile, intact drug via lung fluid may be an additional route of elimination available prenatally. Secondly, the marked accumulation of diphenhydramine occurs despite a very low rate of pulmonary blood flow and receipt of a small fraction of combined ventricular output as opposed to the adult, where the entire cardiac output is delivered to the lung. To date, metoclopramide as well as diphenhydramine are the only drugs that have been reported to be excreted into the fetal tracheal fluid, although the extent of tracheal drug excretion tends to be much higher for metoclopramide than the latter drug (fetal tracheal to arterial plasma drug concentration ratio, 15.1 vs. 2.8) (Riggs et al., 1987a). At present, there are no data to explain the greater accumulation of metoclopramide into the tracheal fluid as compared to diphenhydramine. Tracheal fluid pH is lower than that in plasma (6.23 vs. 7.39), but ion trapping would not explain the difference, given the fact that both diphenhydramine and metoclopramide are weakly basic, with pKa values of ~9. The higher plasma protein binding of diphenhydramine (-70%) as compared with metoclopramide (-40%, Riggs et al., 1988a) may have been involved. The pulmonary disposition of diphenhydramine in the fetal lamb will be further discussed in the 142 following section (see section 4.6). Following i.v. bolus injection of diphenhydramine to the pregnant ewe, drug concentrations in the amniotic fluid rose progressively, reaching the peak concentration (mean = 113.1 ng/mL) by 1.5-3.0 hr after drug administration. However, the highest diphenhydramine concentration in the amniotic fluid is -1/3 of the highest drug concentration in the fetal plasma, indicating less drug accumulation in this fluid as compared with tracheal fluid. Diphenhydramine is more slowly eliminated from the amniotic fluid (t^, ^ = 168 min) than from either tracheal fluid (39 min) or fetal plasma (46 min). During the third trimester of gestation in both sheep and humans, the fetus is known to swallow both amniotic and tracheal fluid, and probably salivary secretions as well (Harding et a7., 1984). For example, the fetus swallows amniotic fluid at an average rate of -300- 1,000 mL/day in late ovine pregnancy (Harding et al., 1984; Tomoda et al., 1985) and -750 mL/day in near term human pregnancy (Lotgering and Wallenburg, 1986). Therefore, it would appear that diphenhydramine excreted into amniotic fluid may be swallowed by the fetus, or otherwise transferred from the amniotic cavity by diffusion across the amnio- allantois into the fetal or maternal circulation. The recirculation of diphenhydramine from the amniotic fluid will be discussed further in a subsequent section (4.5). 143 4.4 Transplacental Clearance, Nonplacental Clearance and Fetal Effects of Diphenhydramine in Pregnant Sheep: i.v. Infusion to Steady-State 4.4.1. Transplacental and Nonplacental Clearances The pharmacokinetics of diphenhydramine were studied in the chronically instrumented maternal-fetal sheep model following maternal and fetal drug infusions to steady-state. As seen in Figs. 17 and 18, it is apparent that steady-state drug concentrations were obtained in both maternal and fetal plasma by 30-60 min after drug administration. Drug infusions, however, did not result in drug equilibrium between the mother and fetus at steady-state: the Cfss/Cmss ratio is smaller than unity on the basis of either the total drug concentration (0.19 ± 0.10) or the free drug concentrat ion (0.39 ± 0.16), indicating that fetus has an ability to eliminate diphenhydramine via nonplacental pathways. The present data indicate that the nonplacental clearance accounts for an average of 45.1 ± 4.7% of the total clearance of diphenhydramine from the fetus, which is greater than that reported for metoclopramide (-24%) (Riggs et al., 1987b). The average fetal to maternal total drug concentration ratio at steady-state following maternal infusion is strikingly smaller than the average fetal to maternal concentration ratio obtained after maternal i.v. bolus injection (0.19 vs. 0.90). Similar situations, where the total fetal to maternal drug concentration ratio after i.v. infusion is smaller than that after i.v. bolus injection, exist for a number of drugs, including metoclopramide (0.55 vs. 0.70), methadone (0.14 vs. -0.30) and morphine (0.13 vs. 0.25) (Szeto et al., 1981; Szeto et al., 1982b; Golub et al., 1986; Riggs et 144 al., 1988a). Therefore, comparisons of the extent of relative fetal drug exposure, determined as the Cfss/Cmss ratio or fetal/maternal oo concentration or AUC0 ratio (Mihaly and Morgan, 1984), should always be made under similar experimental protocols. There are limited data available on the placental transfer of other antihistamines. In the case of cimetidine and ranitidine, the fetal to maternal steady-state total drug concentration ratio after maternal infusion is -0.03 in the pregnant sheep (Mihaly et al., 1982; Ching et al., 1985). Both cimetidine and ranitidine are hydrophilic in nature and the extent of protein binding of these drugs (-30-40%) is lower than that of diphenhydramine (-72-86%) in sheep. However, the greater lipophilic nature of diphenhydramine seems to favour more extensive fetal drug exposure as compared to the latter two drugs. Diphenhydramine is known to bind to human serum albumin but the degree of binding is low (Drach et a/., 1970). It is not known whether the drug binds to aj-acid glycoprotein in human or animal plasma. But the degree of binding of the drug does not seem to change over a wide range of drug concentrations, i.e., -98% at -100 and -1100 ng/mL in man (Albert et al., 1975) and -70% over 30-2,000 ng/mL in sheep. The average free fraction of diphenhydramine in the pregnant ewe (0.141 ± 0.079) is significantly lower than that in the fetal lamb (0.277 ± 0.087). A higher free fraction in the fetal sheep as compared with pregnant ewe has been reported for a number of other basic lipophilic drugs, such as propranolol, methadone and lidocaine (Hill et al., 1986). Lower binding of these drugs in the fetus as compared with mother has 145 been speculated to be due to the lower aj-acid glycoprotein concentrations in the fetus as compared with mother. We have not measured aj-acid glycoprotein concentrations in sheep and, therefore, do not have data to support the correlation between the protein concentration and the extent of protein binding of the drug. There was no significant difference in the free fraction of the drug between pregnant ewes (0.141 ± 0.079) and the nonpregnant ewes (0.229 ± 0.080). No correlation was observed between gestational age and the fetal nonplacental clearance of diphenhydramine or the fetal nonplacental contribution (%) to the total fetal drug clearance. This lack of correlation suggests that the development of the functional metabolic capacity of diphenhydramine did not vary appreciably over 126- 136 days gestation in the fetus. The fetal nonplacental clearance of the drug, CLf0 (208.4 mL/min) is -6% of the maternal nonplacental clearance, CLmo (3343.8 mL/min). The CLf0 normalized by the fetal body weight (99.5 mL/min/kg) is -20% of the fetal combined ventricular output (-530 mL/min/kg), whereas the CLmo normalized by the maternal body weight (43.2 mL/min/kg) is -30% of the maternal cardiac output (-150 mL/min/kg) in sheep (Rudolph and Heymann, 1970; Rosenfeld, 1977). Therefore, despite the fact that the weight normalized CLf0 is -2 times greater than the weight normalized CLmo, diphenhydramine appears to be cleared more efficiently from the mother than from the fetus. Placental clearance for several substances has been frequently determined by use of the Fick's diffusion principle (Meschia et al., 1967). Meschia et al. (1967) have defined the placental clearance as 146 the rate of drug diffusion across the placenta divided by the transplacental concentration gradient, and have formulated the concept of flow-limited and diffusion-limited placental clearance. Flow-limited substances have a high placental permeability and their placental transfer is primarily determined by the rate of drug delivery to the placenta. The permeability of the placenta to diffusion-limited substances is low and transfer is determined primarily by placental structural characteristics. This method can be used for drugs that are metabolically inert but are cleared rapidly by the placenta, resulting in a measurable transplacental concentration gradient. More recently, Anderson et al. (1980) reported a technique for the measurement of placental clearance of drugs without any assumption regarding placental metabolism or plasma protein binding, as long as drug transfer across the placenta is diffusion-limited. The placental clearance calculated by this method is equivalent to CLmf. These two methods are based on one common assumption, that the rate of placental clearance is the same in either direction across the placenta. The unidirectional fetal- maternal placental clearance has been reported for a number of drugs, such as acetyl salicylic acid (Anderson et al., 1980), a glucose analogue, a-methyl-D-glucopyranoside (Stacey et al., 1978), antipyrine, 3 H20, urea (Meschia et al., 1967) and ethanol (Bonds et al., 1980). Recently, Wang et al. (1986), using Szeto's compartmental method, demonstrated that there is no significant difference between CLmf (60.8 mL/min) and CLfm (58.4 mL/min) for acetaminophen in sheep, suggesting a unidirectional transplacental clearance. However, situations where CLmf is different from CLfm exist for a number of other drugs such as methadone (Szeto et al, 1982a) and metoclopramide (Riggs et al., 1987b). 147 In the case of diphenhydramine, the CLmf (82.4 mL/min) is significantly smaller than CL^ (264.4 mL/min). Diphenhydramine is highly lipophilic and its placental transfer may be affected largely by the perfusion characteristics of the placenta and, to a lesser extent, by the anatomic properties of the placenta or the placental permeability, i.e., it may 3 be flow-limited. A number of substances such as antipyrine, H20 and ethanol have been previously classified by the Fick principle [Infusion Rate/(Cf'ss - Cm'ss)] as being flow-limited, with the transplacental clearance of these compounds ranging from -80-160 mL/min/kg (Meschia et al., 1967; Walker, 1977; Bonds et al., 1980; Owens et al., 1986). On the other hand, urea is classified as a diffusion-limited substance, with a low transplacental clearance (-20 mL/min/kg) (Meschia et al., 1967). When a modified equation of the Fick principle [(Cfss x CLfm)/(Cfss - Cm'ss)] was applied, the transplacental value for diphenhydramine averaged -130 mL/min/kg, again suggesting that the placental clearance of diphenhydramine is flow-limited. In near-term sheep, maternal placental blood flow comprises -15% of maternal cardiac output, whereas fetal placental blood flow is -40% of fetal combined ventricular output (Rudolph and Heymann, 1970; Rosenfeld, 1977). Therefore, with diphenhydramine, CLfm may be higher than CLmf because of the greater proportion of cardiac output delivered to the placenta in the fetus compared to the ewe. One potential limitation of the two-compartment model proposed by Szeto et al. (1982a) is that it does not account for the possibility of placental drug metabolism in the transplacental clearance terms, CLmf and CLfm. According to this model, both CLmf and CLfm are simply the 148 transfer clearances and, if there is any drug metabolism occurring within the placenta, it will be reflected in the fetal or maternal nonplacental clearance, CLf0 and CLmo. On the other hand, the placental clearance calculated by the model-independent method using the Fick principle will represent the metabolic clearance within the placenta as well as the transfer clearance across the placenta (Meschia et al., 1967; Wang et a7. 1986). Therefore, if there is any drug metabolism occurring within the placenta, the placental clearance calculated by the compartmental method may be an underestimate as compared to the placental clearance calculated by the model-independent method. In case of acetaminophen, the transplacental clearance (CLmf or CLfm) calculated by the two methods was similar, and placental metabolism was, therefore, postulated to be negligible (Wang et al., 1986). The ER of diphenhydramine across the placenta on the fetal side (0.64) calculated in one experiment in the present study (eq. 23) is ~5 times greater than that of acetaminophen (0.12). Assuming a similar rate of umbilical blood flow in the two studies (-500 mL/min, Wang et al., 1986), the placental clearance of diphenhydramine from the fetal side, CLp (ER x umbilical blood flow), must be -5 times greater than that of acetaminophen. In fact, the average CLfm calculated for diphenhydramine (264.4 mL/min) is - 5 fold greater than the CLfm reported for acetaminophen (58.4 mL/min). The fetal ER of diphenhydramine (0.33) calculated by the eq. 24 represents the fraction of the drug received by the fetus that is cleared by nonplacental mechanisms, such as fetal hepatic, renal, pulmonary and other possible routes but excluding placental uptake/metabolism. In this animal (no. 130), therefore, the difference (-11%) between the nonplacental contribution (45.4%) 149 calculated by the compartmental method and the percentage fetal extraction ratio (33%) determined after maternal infusion may represent a proportion of the percentage contribution of the placental metabolism to the total fetal clearance of diphenhydramine. However, further investigation is required to determine the contribution of the placental metabolism to the total fetal drug clearance. Placental and nonplacental clearances of a number of drugs, including morphine, methadone (Szeto et al., 1982a and 1982b), acetaminophen (Wang et al., 1986) and metoclopramide (Riggs et al., 1987b), have been determined at steady-state in the chronically instrumented pregnant sheep preparation. Methadone, metoclopramide and diphenhydramine are all weakly basic compounds with high 1ipophilicity (pKa 9.0-9.3). These drugs are known to bind to plasma protein more extensively in the mother than in the fetus. The extent of protein binding of diphenhydramine (86% in mother and 72% in fetus) tends to be higher than that of methadone (75% in mother and 58% in fetus) and metoclopramide (49% in mother and 39% in fetus) (Szeto et al., 1982c; Riggs et al., 1988a). Respective values for the CLff, CLfm, CLmf for diphenhydramine (472.7, 264.4 and 82.4 mL/min) are comparable to those for methadone (470.8, 312.8 and 93.5 mL/min) (Szeto et al., 1982a). Since the steady-state Cfss/Cmss ratio after maternal infusion is a function of both CLmf and CLff (Cfss/Cmss = CLmf/CLff) and CLmf and CLff for diphenhydramine and methadone are similar, the Cf$s/Cmss ratio for diphenhydramine (0.19) closely approaches the Cfs$/Cmss ratio for methadone (0.18) (Szeto et al., 1982b). Riggs et al. (1987b) reported placental and nonplacental clearances of metoclopramide in sheep, as 150 body weight normalized values. The CLfm for diphenhydramine (124.4 mL/min/kg) is similar to the CLfm for metoclopramide (125.1 mL/min/kg) but the CLf0 for diphenhydramine (99.5 mL/min/kg) is ~ 2.5 fold greater than for metoclopramide (40.0 mL/min/kg), resulting in the greater CLff for diphenhydramine (223.9 mL/min/kg) than metoclopramide (165.1 mL/min/kg). On the other hand, CLmf for metoclopramide is ~2 fold greater than that for diphenhydramine (79.0 vs. 41.0 mL/min/kg) and the resulting Cfss/Cmss ratio for metoclopramide (0.55) is much higher than that for diphenhydramine (0.19). In contrast, morphine and acetaminophen are much less lipid soluble than methadone, metoclopramide and diphenhydramine. The extent of protein binding in mother and fetus is about the same for morphine and acetaminophen, with the degree of binding being <5% for the former and -95% for the latter. The transplacental clearances, CLfm and CLmf, and the total fetal clearance, CLff for these drugs are lower than for the more lipophilic drugs discussed above. The CLff for the more lipophilic drugs, methadone, metoclopramide (assuming fetal body weight of 2.5 kg) and diphenhydramine (range, -410-470 mL/min) is at least 2 fold greater than that for morphine (81.0 mL/min) and acetaminophen (184.0 mL/min). The higher CLff for the more lipophilic drugs seems to be a result of greater CLfm of these drugs (range, 264.4—312.8 mL/min) as compared to the CLfm of acetaminophen (58.4 mL/min) and morphine (58.4 mL/min). Therefore, it is likely that the 1ipophilicity/hydrophobicity of a drug is a major determinant for the transplacental clearances, CLfm and CLmf, whereas CLf0 reflects the metabolic capacity in the fetus. In regards to the latter view, it is interesting to note that acetaminophen, morphine and metoclopramide are metabolized mainly by glucuronidation 151 and sulfation and diphenhydramine and methadone are metabolized primarily by N-demethylation. The low CLf0 for acetaminophen (22.6 mL/min), morphine (125.6 mL/min) and metoclopramide (-100 mL/min) as compared to diphenhydramine (208.4 mL/min) and methadone (158.0 mL/min) may reflect the functional differences in the metabolic capacity in the fetus. Placental metabolism of acetaminophen is negligible in near-term sheep (Wang et al., 1986), whereas diphenhydramine could be metabolized by the placenta, as demonstrated indirectly by the noncompartmental extraction method (equation 27) in one animal experiment (ewe no. 130) in the present study. Diphenhydramine is also known to be excreted into fetal lamb lung fluid (Riggs et al., 1987a). Preliminary studies of the fetal pulmonary disposition of diphenhydramine indicate that extraction by the left fetal lamb lung is -8% at steady-state (see section 3.6). It is also possible that tissues other than fetal lungs and placenta, such as fetal liver, kidneys and/or intestine are responsible for the fetal diphenhydramine elimination. In conclusion, diphenhydramine is cleared from mother and fetus by placental and nonplacental pathways. Further work is required to determine the primary pathways of the drug elimination, e.g., fetal hepatic metabolism, renal and pulmonary excretion 4.4.2. Drug Accumulation in Fetal Tracheal and Amniotic Fluids Following drug infusions to the ewe and fetus, diphenhydramine was found to be excreted into the fetal tracheal and amniotic fluids (Figs. 19 and 20). In all the animals employed, the average fetal 152 tracheal to arterial drug concentration ratio was greater than 1.0 following either the maternal or fetal drug infusion (Table 16). The average ratio of the fetal tracheal to arterial drug concentrations after maternal infusion (mean, 3.4 ± 3.6) was comparable to that obtained after fetal drug infusion (mean, 5.4 ± 4.5). Also, these ratios are comparable to the ratio after i.v. bolus injection of the drug to pregnant sheep (mean, 2.7 ± 2.4). The Cmax in the tracheal fluid is significantly greater than the Cmax in the amniotic fluid following either the maternal infusion (mean, 124.7 vs. 16.6 ng/mL) or the fetal infusion (mean, 1790.4 vs. 139.9 ng/mL) (Table 15). Diphenhydramine concentrations in the tracheal fluid are higher than the corresponding concentrations in the amniotic fluid for most of the sampling period (see Figs. 19 and 20), hence it is possible that diphenhydramine excreted into fetal lung fluid in part accounts for the drug accumulation in the amniotic fluid since the tracheal fluid is continuously produced at a constant rate from the lungs and flows out, and is either swallowed or secreted into the amniotic sac (see section 4.3.4). 4.4.3. Drug Disposition in Maternal and Fetal Sheep Following Maternal and Fetal Infusions Following infusions of the drug to the ewe, an average of 78.6 ± 4.3 mg of diphenhydramine, as a free base, was calculated to be oo - eliminated from the maternal sheep (AUCMo CLmo) and 1.2 ± 0.6 mg of diphenhydramine, as a free base, was eliminated from the fetal lamb (AUCFQ-CLfo) (Table 21). Therefore, an average of 1.4% of the total 153 drug available in the maternal-fetal sheep was eliminated by the fetus [AUCFo-CLfo'100/(AUCMo-CLmo + AUCFo'CLfo)] and the rest (98.6%) was eliminated by the ewe [AUCMQ-CLrno-100/(AUCMQ-CLmo + AUCFQ-CLfo] (Table 21). On the other hand, following drug infusions to the fetus, an average of 14.7 ± 6.2 mg and 11.5 ± 3.0 mg of diphenhydramine, as a free base, was eliminated from the ewe and fetus, respectively (Table 21). Therefore, in the case of fetal drug infusion, an average of 45.0% of the total drug available in the maternal-fetal sheep was eliminated by the fetal lamb and 55.0% of the total drug by the maternal sheep (Table 21). Previously, maternal and fetal transplacental and nonplacental clearances of the drug were calculated (eq. 16-21) by use of the values of the steady-state maternal-fetal drug concentrations following maternal and fetal infusions and their infusion rates (section 3.4.1.4). The average percent contribution of maternal nonplacental clearance to maternal total drug clearance, CLmo/(CLmo + CLmf), was calculated to be 97.8%, whereas the average percent contribution of the fetal nonplacental to fetal total drug clearance, CLf0/(CLf0 + CLfm), is 45.1% (Table 18). There is a good agreement between the percent contribution of the maternal nonplacental to maternal total drug clearance [CLmo-100/(CLmo + CLmf) = 97.8%, Table 18] and the percentage of the total amount of the drug that was eliminated from the mother via the nonplacental pathway following maternal drug infusion, calculated by use of AUC and nonplacental clearance values [AUCMQ'CLmo'100/(AUCMQ-CLmo + 00 AUCFo'CLf0) = 98.6%, from Table 21]. Similarly, there is also a good agreement between the percent contribution of the fetal nonplacental to 154 fetal total drug clearance [CLfo'100/(CLfo + CLfm) = 45.1%, Table 18] and the percentage of the total amount of the drug that was eliminated from the fetus via the nonplacental pathway following fetal drug infusion [AUCFo'CLfo'100/(AUCMo'CLmo + AUCFo'CLfo) = 45.0%, Table 21]. Therefore, employing the two different methods, it has been demonstrated that the percentage of the total diphenhydramine available in the maternal-fetal sheep that was eliminated by the fetal lamb averaged -1-2% following maternal infusions and -45% following fetal drug infusions. The amount of diphenhydramine eliminated from the ewe and fetus via the nonplacental pathway averaged 78.6 mg and 1.2 mg, respectively, for the maternal infusions (sum, 79.8 mg) and 14.7 mg and 11.5 mg, respectively, for the fetal infusions (sum, 26.2 mg) (Table 21). It should be noted that the sum of the amount of the drug eliminated from oo both the ewe and fetus via the nonplacental pathway, AUC^0'CLmo + OO AUCp0'CLf0, exceeded the administered dose. In the case of maternal infusion, the calculated amount of diphenhydramine that was eliminated by the maternal-fetal unit was 79.8 mg, whereas the administered dose was 70.0 mg (as a free base). Similarly, in the case of fetal drug infusion, the calculated amount of the drug that was eliminated by the maternal-fetal unit (26.2 mg) exceeded the administered dose (17.5 mg as a free base). As stated in the proceeding sections (sections 4.3.2 and 4.4.2), it is probable that the drug administered to the pregnant ewe or fetus is subject to recirculation via fetal swallowing of the tracheal and amniotic fluids and via drug transfer from the chorio-allantois back to the fetal or maternal circulation. Therefore, it is expected that 155 the calculated amount of total diphenhydramine available in the maternal-fetal sheep is greater than the dose administered. Also, it is not surprising that the calculated amount of drug eliminated by the ewe and fetus compared to the dose actually administered is greater following the fetal drug infusion (149.8%) as compared to the maternal infusion (114.0%) since following fetal infusion, a greater amount of the drug could be accumulated in the fetal tracheal and amniotic fluids for further drug recirculation. 4.4.4. Fetal Effects of Diphenhydramine in Pregnant Sheep In the last quarter of gestation, the fetal lamb exhibits regular alterations between episodes of high voltage, low frequency and low voltage, high frequency ECoG activity (Dawes et al., 1972). The former activity bears many similarities to quiet sleep after birth, whereas the latter pattern, the bulk of which is associated with REM and rapid irregular breathing movements, appears equivalent to REM or paradoxical sleep. Various investigators have reported brief periods (1-3 min) of an aroused state, with low voltage ECoG activity, lack of REM and vigorous breathing and body movements; this may represent wakefulness (Ruckebusch et al., 1977; Harding, 1984). A transitional state, between high and low voltage episodes, and vice versa, with an intermediate voltage amplitude has also been reported (Natale et al., 1981). In the present investigation, we obtained ECoG results similar to these findings, although we did not attempt to identify periods of fetal arousal. 156 The fetal effects of diphenhydramine were examined at two different concentration ranges, achieved by separate infusions of the drug to the ewe and fetus. The fetal steady-state concentration of the drug during the maternal infusions averaged 36.3 ng/mL. This level is below that (<50 ng/mL) associated with discernable CNS effects in adult humans (Carruthers et al., 1978). The fetal plasma diphenhydramine concentration during fetal infusion averaged 447.6 ng/mL. This is approximately 3 times the initial plasma diphenhydramine concentration resulting from i.v. injection of 50 mg of the drug to human subjects (Carruthers et al., 1978; Blyden et al., 1986), but well below the level of ~5 Mg/mL found in a male subject who died from a diphenhydramine overdose (Hausmann et al., 1983). There do not appear to be any reports of the drug levels in children who have had convulsive episodes resulting from diphenhydramine poisoning. The two modes of diphenhydramine administration used in the present study resulted in two distinct patterns of fetal behavioural effects, which we ascribe to the markedly different fetal plasma drug concentration achieved. During infusions of the drug to the ewe, there were reductions in the incidence of low voltage ECoG activity and the amount of REM associated with the low voltage activity, i.e., a reduction in the amount of REM sleep. Coupled with this was a fall in the amount of breathing activity. This latter effect is similar to those reported for CNS depressants such as chloralose, pentobarbital, diazepam and ethanol, which involve temporary suppression of fetal breathing movements (Dawes et al., 1972; Boddy et al., 1976; Piercy et al., 1977; Patrick et a7.,1985). In terms of the fetal ECoG effects of 157 diphenhydramine during maternal infusions, these appear to be similar to, but less striking than those elicited by pentobarbital administration, where there is a marked reduction in the incidence of the low voltage pattern and increase in the amount of high voltage activity (Boddy et a7., 1976). In contrast, with ethanol administration to pregnant sheep, there is a replacement of both high and low voltage ECoG activities by an intermediate voltage pattern (Patrick et a7., 1985), a situation similar to the results obtained during infusion of diphenhydramine to the fetus. The fetal ECoG effects of diazepam and chloralose do not appear to have been reported. In adults, the predominant central effect of diphenhydramine is sedation, reflected by inability to concentrate, altered coordination and sleepiness (Carruthers et a7., 1978; Roth et al., 1987). The assessment of the CNS effects in adults, both humans and animals, has relied largely on behavioural tests and subjective assessments of drowsiness by the experimental subjects (Carruthers et al., 1978; Winter, 1985). The results from studies of the EEG effects of the drug in humans (Goldstein et a7., 1968; Fink and Irwin, 1979) are consistent with a sedative effect of the drug, although Roth et al. (1987) found that when diphenhydramine was given in the evening the percentages of the various sleep stages during the night were unaltered from the placebo results. However, these behavioural and EEG data from humans cannot be readily compared with the information obtained from the fetal lamb. The main reason for this is the great difference in sleep state pattern in the fetus and adult. In the fetus, the awake state rarely if ever occurs (Harding, 1984), whereas in the adults, the assessment of 158 the CNS effects of diphenhydramine has been accomplished by administering the drug to awake subjects. However, in view of the similarities of the fetal effects of diphenhydramine during the maternal infusions to those elicited by other CNS depressant substances, it seems reasonable to conclude that, at the plasma levels achieved during these experiments (36 ng/mL), the drug elicited fetal sedative effects. Whether higher fetal drug levels, in the range associated with sedative actions in adults, would result in greater effects appears worthy of investigation. When diphenhydramine was infused directly to the fetus, a distinctly different pattern of fetal behavioural effect was observed. These comprised a dramatic increase in the incidence of an intermediate voltage ECoG pattern at the expense of both REM and quiet sleep. The net effect was a disruption in the normal cycling of ECoG activity between REM and quiet sleep by a large increase in the amount of a transitional ECoG state. The overall incidence of fetal breathing activity was unaltered, although there was an increase in the amount during the intermediate ECoG pattern and the appearance of large amplitude breathing movements immediately after the initial bolus injection, when the plasma drug levels were at their highest. There was also modest and transient hypoxemia, acidemia and tachycardia. The average fall in Po2 (3.3 mm Hg) and pH (0.043) is similar to that observed during vigorous fetal activity and breathing movements (Harding et a7., 1983; Rurak and Gruber, 1983). These blood gas changes are probably a consequence of an increase in fetal oxygen consumption resulting from the large amplitude breathing activity, as similar 159 changes and an increase in 02 consumption are observed during vigorous fetal breathing, both when occurring spontaneously (Rurak and Gruber, 1983) and when stimulated by hypercapnia (Rurak et al., 1986). The rise in fetal heart rate is similar to the tachycardia observed in adult dogs given an i.v. infusion of diphenhydramine at a rate (4 mg/kg/min) close to that (-5 mg/kg/min) used in the current study, and has been attributed to the anticholinergic actions of the drug (Hirschowitz and Molina, 1983). Alternatively, potentiation of the effects of endogenous histamine by diphenhydramine could be involved in the tachycardia, inasmuch as in the isolated guinea-pig heart, diphenhydramine at 10"^ to 10"^ M potentiates the positive chronotropic effects of low but not high doses of histamine (Levi and Kuye, 1974) and, in the current study, fetal steady-state concentrations of the drug averaged 1.8 x 10"° M during the fetal infusions. However, the cardiac effects of histamine and its antagonists are complex and vary between species (Levi et al., 1982), so that further study is needed to understand the cardiac effects of diphenhydramine in the fetal lamb. The ECoG changes during the infusion of diphenhydramine to the fetus are similar to that observed after acute ethanol administration to pregnant sheep (Patrick et al., 1985). However, the effects of these two agents on fetal breathing activity are different, because with ethanol, fetal breathing activity is suppressed for 9 hr after drug administration (Patrick et al., 1985). Other agents have stimulatory effects in the fetal lamb, involving the occurrence of breathing activity in low and high voltage ECoG states, as in the case of 160 indomethacin and other prostaglandin synthetase inhibitors (Kitterman et al., 1979) and 5-hydroxytryptophan (Quilligan et al., 1981). Methadone, on the other hand, elicits a hyperactive state involving suppression of both quiet and REM sleep and continuous breathing and body movements (Szeto, 1983). The effects of all these agents are unlike those after fetal infusion of diphenhydramine and so the effects of the latter would appear to be unique, at least in terms of the agents examined to date. During the fetal infusions, we obtained no evidence for the convulsive and seizure activity that occurs in children subject to diphenhydramine overdose (Wyngaarden and Seevers, 1951; Reyes-Jacang and Wenzl, 1969) and adult cats given high doses of the drug (Faingold and Berry, 1972). Epileptiform seizure activity can be induced in the fetal lamb with lidocaine (Teramo et al., 1974); it is also observed in some fetuses before intra-uterine death (Chapman et al., 1978). However, it was not present at any point during the period of infusion of diphenhydramine to the fetus, in spite of the relatively high drug levels obtained. Whether such activity can be elicited with higher doses remains to be determined. After cessation of diphenhydramine administration, with both maternal and fetal infusions, fetal behavioural parameters quickly returned to their control ranges, so that the values in the immediate post-infusion period were not different from those in the control period. This is probably a reflection of the rapid rate of clearance of the drug from the maternal and fetal circulations. The relevance of the present observations in the fetal lamb to the human fetus is unclear at present, as there appear to be no reports 161 on the fetal behavioural effects of diphenhydramine or other antihistamines in humans. In this regard, however, it is worth noting that the effects of ethanol on fetal breathing activity are very similar in the sheep and human (McLeod et a7., 1983; Patrick et al., 1985). As mentioned previously, diphenhydramine elicited behavioural effects in the fetal lamb at plasma drug levels lower than those associated with CNS effects in adult humans. It would also appear that the fetal lamb is more sensitive to the CNS actions of diphenhydramine than is the pregnant ewe as no behavioural modifications were noted in the latter, even during maternal administration of the drug, when the maternal plasma levels of the drug were much higher than those needed to elicit CNS effects in adult humans. This situation is similar to the effects of pentobarbital in pregnant sheep, where doses of the drug (4 mg/kg) which have no obvious effects on the ewe cause suppression of REM sleep and breathing activity in the fetus (Boddy et al., 1976). Similarly, ethanol at a dose of 0.25 mg/kg in pregnant women markedly suppresses fetal breathing activity with minimal effects on the mother (McLeod et al., 1983). If diphenhydramine elicits behavioural effects in the human fetus, this could have two important consequences, particularly if the drug was effective at lower concentrations in the fetus than in the adult. The first results from the fact that the current diagnostic tests of human fetal well-being, such as the nonstress test (Brown and Patrick, 1981) and the fetal biophysical score (Manning et al., 1980) monitor elements of fetal behaviour, such as body and breathing movements and movement-associated heart rate changes. If these monitored variables are affected by diphenhydramine-induced alterations in fetal behaviour, there could be a misdiagnosis of fetal distress, 162 particularly if the physician were unaware that the patient was taking the drug. The second potential consequence is that prolonged exposure of the fetus to the drug could result in long term alterations in CNS function. This is suggested by the observations of Parkin (1974) on apparent withdrawal symptoms in a newborn infant whose mother had been treated with 150 mg of diphenhydramine daily for an unspecified period of time during pregnancy. In summary, the effects of the antihistamine, diphenhydramine, have been investigated in the fetal lamb at two different concentration ranges, achieved by separate i.v. infusions of the drug to the ewe and fetus. With low fetal diphenhydramine concentrations, sedative effects were observed, comprising reduction in the amounts of REM sleep and fetal breathing activity. In contrast, with higher drug levels, there was transient vigorous breathing activity, hypoxemia, acidemia and tachycardia and a marked increase in the amount of a transitional ECoG pattern. Moreover, fetal CNS effects were observed at plasma concentrations of the drug below those associated with sedation in adult humans. Given the fairly high frequency of use of antihistamines during pregnancy, further examination of their effects on the fetus, in both animals and humans, is warranted. 4.5. Drug Recirculation from Amniotic Fluid to the Ewe and Fetal Lamb Following Intra-amniotic Bolus Administration 4.5.1. Amniotic Fluid Dynamics 163 In the last third of gestation in sheep, the fetus is surrounded by amniotic fluid with an average volume of 300-1,000 mL (Wintour et al. 1978; Tomoda et a/., 1985). Primary factors controlling the amniotic fluid volume are thought to be fetal urine and lung fluid production, fetal swallowing and transplacental fluid exchange (Mescher et al 1975; Harding et al., 1984; Tomoda et al., 1985; Ross et al., 1988; Wlodek et al., 1988). Near term, total fetal urinary output rate averages -700-1,300 mL/day (Hill and Lumbers, 1988; Wlodek et al., 1988) and approximately 40-60% of the fetal urine output, or -300-800 mL of fetal urine drains into the amniotic cavity via the urethra and the rest into the allantoic cavity via the urachus (Ross et al., 1988; Wlodek et al., 1988). Fetal tracheal fluid is produced at a relatively constant rate (4.5 mL/kg'hr, Mescher et al., 1975) and therefore, in a 2 kg fetus, more than 200 mL of the fetal tracheal fluid enters the pharynx per day, and either drains into the amniotic cavity or is swallowed by the fetus. A substantial quantity of fluid (mean, 300-1,100 mL/day) is swallowed by the fetal lamb in utero in the last third of gestation (Bradley and Mistretta, 1973; Harding et al., 1984; Tomoda et al., 1985). This appears to be mainly a mixture of amniotic and fetal tracheal fluids, although it is possible that saliva and secretions of the nasal mucosa are also swallowed (Harding et al., 1984). In the proceeding sections (sections 3.3 and 3.4.1.3), it has been demonstrated that diphenhydramine is excreted into the fetal tracheal and amniotic fluids following either maternal or fetal drug i.v. administration. One potential consequence of these findings is 164 recirculation of the drug from the tracheal and amniotic fluids back to the fetus, via several possible routes, i.e., re-uptake from the lungs, swallowing of amniotic or lung fluids and uptake from amniotic fluid by the fetal membranes. To examine this possibility, we administered diphenhydramine into the amniotic cavity and monitored drug concentrations in the amniotic and fetal tracheal fluids, fetal and maternal arterial plasma in 5 chronically catheterized maternal-fetal sheep at 122-137 days gestation. 4.5.2. Extent of Drug Excretion from Amniotic Fluid by Fetal Swallowing Tomoda et al (1985) measured the disappearance rate of radioiodinated (*^I) serum albumin from amniotic fluid following injection into the amniotic cavity. The disappearance rate of the labelled albumin from the amniotic fluid averaged 4.9 ± 2.7 %/hr (n = 17), and the disappearance of the radioactivity was accounted for by fetal swallowing since the labelled albumin disappears only through fetal swallowing: it is swallowed, digested and absorbed into the fetal circulation as free iodine (Tomoda et al., 1985). The free iodine is readily transferred from the fetus to the ewe across the placenta and eventually all the radioactivity of the labelled serum albumin can be recovered in maternal urine (Pritchard, 1965), with negligible amount of the activity found in fetal urine (Tomoda et al., 1985). Therefore, the possibility of recirculation of the labelled albumin via fetal urination into the amniotic cavity is excluded. When the disappearance rate was calculated for diphenhydramine by use of the same principle (Lingwood 165 and Wintour, 1984; Tomoda et al., 1985), i.e., disappearance rate expressed as percent per hour, disappearance rate (%/hr) = (1-10s)* 100, where s is the slope of the linear regression line of the terminal linear phase of the logarithm of the amniotic drug concentration vs. time data, the average disappearance rate was calculated to be 6.7 ± 2.0 %/hr. The ratio of the labelled albumin to diphenhydramine disappearance rate, 4.9/6.7, gives an estimation (-73%) of the proportion of the total rate of loss of diphenhydramine that can be accounted for by fetal swallowing. However, it should be noted that the disappearance rate from the amniotic fluid calculated for diphenhydramine (6.7 %/hr) may not be accounted for by fetal swallowing alone and if the drug is subject to recirculation within the maternal- fetal sheep, the disappearance rate would be underestimated. Also, the calculated percentage of the total rate of loss from amniotic fluid accounted for by fetal swallowing (73%) may be an overestimation if the drug is transferred by direct diffusion across the placenta from the amniotic fluid to the maternal circulation or if the drug is taken up by the fetus via the fetal circulation in the chorio-allantois and the placenta and via diffusion across the fetal skin. 4.5.3. Disposition of Diphenhydramine in the Amniotic and Fetal Tracheal Fluids, Maternal-Fetal Plasma after Drug Injection into Amniotic Cavity In all the experiments, the drug concentration in the fetal arterial plasma was consistently higher than the corresponding value in the maternal plasma throughout the sampling time, with an average fetal to maternal plasma drug concentration ratio being -4.5. Szeto et al., 166 (1978) reported that, following intra-amniotic administration of meperidine in pregnant sheep, the drug was transferred more extensively to the mother than to the fetus. At this time, it is not clear what is responsible for the difference in the transfer pattern of these two drugs. In our present study, diphenhydramine appeared to be taken up preferentially by the fetus and then further transferred from the fetus to the mother. Similar observations were made for metoclopramide: following intra-amniotic administration, the drug was found to be much more extensively transferred to the fetal compartment than maternal compartment (Riggs et a7., 1988b). Further discussion on the extent of drug transfer, i.e., approximate estimates of the amount of drug transferred from the amniotic fluid to the fetus or to the mother across the placenta by direct diffusion, is made in the following section (see section 4.5.4). The apparent elimination half-life of the drug in the amniotic fluid (11.0 ± 4.5 hr) is comparable to that in the fetal tracheal fluid (10.8 ± 3.8 hr) but is significantly longer than that observed in fetal (7.7 ± 1.2 hr) or maternal (1.7 ± 0.5 hr) plasma (Table 31). It is worth noting that, in the present study, the drug was given extravascularly into the amniotic cavity, where drug absorption took place from the amniotic fluid into the maternal-fetal circulation. The absorption process is reflected in the half-life of the drug in amniotic fluid (11.0 hr), which is significantly longer than the elimination half-life in fetal (7.7 hr) or maternal (1.7 hr) plasma. All of the half-lives obtained for maternal and fetal plasma, tracheal and amniotic fluid in the intra-amniotic bolus experiments are -2-10 times greater 167 than the corresponding half-lives obtained after maternal or fetal i.v. administration (Table 9, 16 and 31). It would appear that in the case of intra-amniotic drug injection experiments, the half-life of diphenhydramine in all four fluids has been primarily influenced by the absorption of drug across fetal and maternal membranes, 7*.e., absorption is the rate-limiting step. All these half-lives are, therefore, likely an overestimation of the "true" half-life if the drug was transferred from the amniotic cavity to the maternal and fetal compartments over a prolonged period of time. The fetal tracheal to plasma drug concentration ratio calculated in the intra-amniotic injection study averaged 16.4 ± 18.0 (range, 4.2-45.9): this ratio is greater than that obtained after either the maternal i.v. bolus injection (2.7 ± 2.4), maternal i.v. infusion (3.4 ± 3.6) or fetal i.v. infusion (5.4 ± 4.5). Fetal plasma Cmax obtained following drug injection into the amniotic cavity (mean, 228.0 ± 101.5 ng/mL) is lower than the steady-state fetal plasma levels after fetal drug infusion (447.6 ± 112.7 ng/mL) and greater than that after maternal drug infusion (36.3 ± 14.4 ng/mL). Therefore, the greater extent of drug accumulation in the fetal tracheal fluid following intra-amniotic injection, as compared to maternal or fetal i.v. administration, can not be explained by extraction across the pulmonary epithelium alone. It is possible that diphenhydramine in the amniotic fluid could have diffused directly into the fetal tracheal fluid in the airway due to a large concentration gradient between the amniotic and tracheal fluids (Figs. 24 and 25), as the amniotic fluid enters the pharynx due to fetal swallowing. However, further study is 168 required to investigate the pathways and mechanism for drug accumulation into fetal lung fluid. 4.5.4. Extent of Drug Elimination from the Ewe and Fetus Following Drug Injection into Amniotic Cavity co Using the maternal and fetal AUCQ values combined with the previous estimates of maternal and fetal transplacental (CLmf and CLfm) and nonplacental clearances (CLmo and CLf0), approximate estimates of various routes of drug elimination from amniotic fluid were obtained. Following drug injection into the amniotic cavity, an average of 18.5 ± 5.1 mg of diphenhydramine (free base) was eliminated from the fetus oo (AUCp0*CLf0) and 31.4 ± 5.1 mg of the drug eliminated from the ewe CO (AUC^0'CLmo) (Table 32). Therefore, the percentage of the total drug available in the maternal-fetal sheep that was eliminated by the fetus, AUCFo*CLfo-100/(AUCMo-CLmo + AUCFo'CLfo), averaged 37.1% . This (37.1%) is greater than that after maternal i.v. bolus injection (5.0%) or maternal i.v. infusion (1.4%) and is less than, but comparable to, that after direct infusion of the drug to the fetus (45.0%) (see sections 4.3.3 and 4.4.3). The amount of diphenhydramine eliminated by the mother that was transferred directly from the amniotic fluid, AUCMQ*CLmo - AUCpQ*CLfm, averaged 8.0 mg (range from -2.1 to 21.5 mg). Hence, an average of 26% (range from -7 to 56%) of the total drug eliminated by the mother was CO CO transferred directly from amniotic fluid [(AUCMo'CLmo - AUCp0*CLfm)•100/ oo AUC^Q'CL^] . The rest of the total drug eliminated by the mother (74 ± 169 20%) was received from the fetus (AUCFo-CLfm-100/AUCM"-CLmo). Also, it appears that approximately 63 ± 8% of the total drug available in the 03 maternal-fetal sheep was eliminated by the mother [AUCMo*CLmo*100/ CO 00 (AUC^Q'CLCTQ + AUCp0'CLf0)], of which, an average of 47% was received from the fetus (eq. 29) and the rest (16± 18%) was received from the amniotic fluid via direct diffusion across the placenta (eq. 30). Percentage of total drug available in the maternal-fetal sheep eliminated by the mother received from the fetus = AUCpo'CLfrn-100/(AUCMo-CLrno + AUCpo'CLfo) (eq. 29) Percentage of total drug available in the maternal-fetal sheep eliminated by the mother transferred directly from the amniotic fluid = (AUCM^CLmo - AUCF^CLfm)-100/(AUCMo-CLmo + AUCpo'CLfo) (eq. 30) Overall, direct diffusion across the uterus appears to play a minor role in the transfer of diphenhydramine from the amniotic fluid to the mother. Rather, it seems that the transfer of substances from the maternal circulation to the amniotic fluid is mediated through the fetus. In the case of ampicillin administration to pregnant women, accumulation of drug in the amniotic fluid occurred, and by 2-3 hr amniotic levels exceeded those of either maternal or fetal plasma (Bray et ai., 1966). However, when the experiment was repeated with a dead fetus, only a small amount of ampicillin (less than -20% of that observed in the case of living fetus) accumulated in the amniotic fluid. These results suggesting that ampicillin is largely transferred indirectly from the maternal circulation to the amniotic fluid via the fetus and to a lesser extent by direct diffusion across the placenta are consistent with those obtained from the present diphenhydramine intra- amniotic injection study. In one animal (ewe no. 287), uptake of diphenhydramine by chorio-al1antoic membranes was calculated by the following equation: Amount of drug taken up by chorio-al1antoic membranes = 2 ^t7-+l - t7-)[(UVt7-+1 + UVtf) - (FAty+1 + FAti)]Qum/2 (eq.31) where UVt7- and FAt7- are the drug concentrations in umbilical venous and fetal arterial plasma at sampling time t7- and Qum is umbilical blood flow. With a fetal body weight of 1.4 kg (Table 29) and Qum of 230 mL/min'kg (Cohn et al., 1982), the total uptake by the membrane (eq. 31) was 14.2 mg. Further, the percentage of total drug loss from amniotic fluid due to chorio-allantoic uptake (eq. 32) calculated in this animal was 33%. Percent of total drug loss from amniotic fluid due to chorio-allantoic uptake = amount taken up by chorio-allantoic membrane*100/(administered dose - amount remaining in amniotic fluid at the end of the sampling time) (eq. 32) where the amount of the drug taken up by chorio-allantoic membranes was 14.2 mg (calculated by eq. 31), the amount of the drug administered is 43.8 mg as a free base (equivalent to 50 mg as hydrochloride salt) and the amount of the drug remaining in the amniotic fluid was 0.2 mg, calculated by multiplying the drug concentration in the amniotic fluid at the last sampling time 28 hr (214.3 ng/mL) and an assumed amniotic fluid volume of 1,000 mL (Tomoda et al., 1985). The higher drug concentration observed in umbilical venous plasma than in fetal arterial plasma is consistent with uptake of the drug from amniotic fluid by the chorio-allantois and delivery to the fetus via the umbilical venous circulation. Overall, the data suggest preferential uptake of 171 diphenhydramine from amniotic fluid by the fetus, resulting in a much greater degree of fetal drug exposure than maternal exposure. 4.6. Pulmonary Disposition of Diphenhydramine in the Fetal Lamb Following Maternal or Fetal Infusions to Steady-State In both sheep and humans, the pattern of blood circulation during fetal life is different from that after birth. In adults, systemic venous blood proceeds from the right atrium to the right ventricle and to the lungs, and the oxygenated blood from the lungs circulates to the left atrium and ventricle and is then distributed in the body. In contrast, fetal blood circulation is characterized by the presence of blood shunts, viz, the ductus venosus, foramen ovale and ductus arteriosus. In the fetal lamb, only a portion (-45%) of the oxygenated umbilical venous blood passes through the liver and the remainder (-55%) bypasses the hepatic circulation via the ductus venosus (Reuss and Rudolph, 1980). Approximately 37% of umbilical venous/inferior vena caval blood bypasses the right heart and flows into the left atrium via the foramen ovale, whereas only a small portion (-2%) of superior vena caval blood entering the heart is delivered to the left atrium via the foramen ovale (Reuss and Rudolph, 1980; Rudolph, 1985). This blood flow pattern appears to favor greater oxygen supply to the fetal myocardium and the brain. Also, in the fetal lamb, right ventricular output is -67% of combined ventricular output. But only -12% of the right ventricular output (-7% of the combined ventricular output) is delivered to the fetal lungs, the rest is directed via the ductus arteriosus to the descending aorta (Rudolph, 1985). 172 In the present study, uptake of diphenhydramine by the fetal lamb lung has been assessed by measuring pulmonary arterio-venous concentrations at steady-state following maternal or fetal drug infusion. During maternal infusion, the drug concentration in carotid arterial plasma was higher than in pulmonary arterial or venous plasma due to preferential distribution of drug-rich inferior vena caval blood through the foramen ovale into the left heart and ascending aorta, and the drug concentration in the pulmonary arterial plasma was higher than in pulmonary venous plasma throughout the sampling period. (Fig. 26, Table 33). In contrast, during the fetal infusion, drug concentration in pulmonary artery was higher than in carotid artery as the drug was delivered from the jugular vein to the right heart and pulmonary artery with the rest of superior vena caval blood return (Fig. 27, Table 33). But again, diphenhydramine concentration in pulmonary artery was higher than in pulmonary vein. Following cessation of the fetal drug infusion, there was a reversal of the drug concentration, indicating net release of intact diphenhydramine from the lung into the fetal circulation (Table 33). As described in the preceeding sections, we have found marked accumulation of diphenhydramine in the fetal tracheal fluid following maternal i.v. bolus injection or infusion, fetal infusion or intra- amniotic injection in pregnant sheep, and have suggested that the excretion of drug via this route may be an important mechanism of drug elimination in the fetus. Overall, the delivery of diphenhydramine to the lungs via the pulmonary circulation would appear to be sufficient to account for the levels observed in tracheal fluid. This argument is 173 based on rough calculations done in the animals employed. It has been assumed that the lung fluid volume is 30 mL/kg (Kitterman, 1984), lung fluid production rate 4.5 mL/hr'kg (Mescher, 1975) and a lung blood flow of 38.5 mL/min'kg (Rudolph and Heymann, 1970). Using the measured drug concentrations in pulmonary arterial and venous plasma and tracheal fluid, the total amount of drug taken up by the lungs over the infusion period [(PA-PV) x pulmonary blood flow x infusion time] averaged 293.9 /ig. The amount of drug present in tracheal fluid at steady-state (87.8 fig) was calculated as the product of total lung fluid volume times tracheal drug concentration, and the amount excreted from the lungs into amniotic cavity (19.7 /xg) as the product of lung fluid production rate x infusion time x tracheal drug concentration. The drug taken up by the lungs (293.9 /ig) may be either present in tracheal fluid (87.8 /ig) or excreted into the amniotic cavity (19.7 /ig), and the remainder (186.4 /ig) is bound/metabolized by lung tissue. Therefore, it appears that during the infusion period, roughly -30% of the total drug delivered to the lungs is accounted for by drug accumulation in the tracheal fluid, -7% by fluid secretion out of the lungs and -63% by cellular binding/metaboli sm. Total drug clearance in the maternal (CLmm, 4402.1 mL/min) and fetal (CLff, 357.7 mL/min) compartments obtained in the present study are within the range observed previously (Table 18), e.g., 1865.4-4808.0 mL/min for CLmm and 240.1-888.4 mL/min for CLff. The fetal pulmonary extraction ratio of diphenhydramine averaged 0.08 ± 0.06. Assuming that fetal lungs receive -7% of the fetal combined ventricular output (-550 mL/min/kg) in near-term sheep (Rudolph and Heymann, 1970), fetal 174 pulmonary blood flow is calculated to be 77 mL/min (mean fetal body weight, 2.0 kg). Fetal pulmonary clearance of diphenhydramine calculated as the product of pulmonary blood flow times pulmonary drug extraction ratio averages 6.2 mL/min at steady-state, which comprises -2% of the fetal total clearance (mean CLff = 357.7 mL/min, Table 33) or -3% of the fetal nonplacental clearance (mean CLf0 = 208.4 mL/min, Table 17). The drug secreted into the tracheal fluid appears to be eliminated from the fetus into the amniotic fluid or recirculated into the fetal circulation via back extraction of the drug from the lungs and/or by fetal swallowing of the tracheal and amniotic fluids. Pulmonary extraction ratio of drugs determined after i.v. bolus injection could be overestimated if drug concentrations are measured during the distribution phase. A steady-state approach may be favored for assessing the lung drug clearance. In summary, the pulmonary extraction ratio of diphenhydramine averaged 0.08 in the fetal lamb in utero. The drug clearance by the fetal lung accounts for -2% of the total drug clearance from the fetus. Even with the low rate of pulmonary blood flow in the fetus, the delivery of the drug to the lung via the pulmonary circulation can account for the accumulation of the drug in the fetal lung fluid. 175 5. SUMMARY AND CONCLUSIONS A nitrogen-phosphorus detection-gas chromatographic assay method (GC/NPD), which provides improved sensitivity and selectivity for diphenhydramine, is reported. A 25 m x 0.31 mm cross-linked, 5% phenylmethyl silicone-coated fused-silica capillary column (film thickness, 0.52 /im) was used for all analyses. The assay procedure involved a single extraction step, with no sample derivatization. The splitless injection mode was employed in the assay, with a 2 /iL sample introduced by an automatic liquid sampler. Standard curves, using orphenadrine as an internal standard, were linear in the range of 2-320 ng/mL of diphenhydramine. This represents an amount of diphenhydramine from -40 pg to 6.4 ng at the detector. Chromatographic separation of diphenhydramine and orphenadrine was excellent, with no interference from endogenous constituents. The assay method was used in the subsequent studies of the pharmacokinetics of diphenhydramine in the nonpregnant and pregnant sheep. Pharmacokinetics of diphenhydramine were characterized in nonpregnant sheep (n = 6) after i.v. bolus injection of 25, 50, 100 and 200 mg doses of diphenhydramine hydrochloride on a crossover basis. Plasma drug concentration vs. time profiles exhibited multiexponential characteristics. The average distribution half-life (t^a) increased from 5 to 9 min and the elimination half-life (t^ ^) from 34 to 68 min, as the dose was increased. There was also an increase in the volume of distribution (l^cfss) from 3 to 6 L/kg, with increasing dose. The elimination half-life and the volume of distribution after a 200 mg dose 176 were significantly greater than those obtained after a 25 mg dose. There was, however, a linear increase in the area under the plasma drug concentration-time curves (AUC), as dose was increased. The average total body clearance (C/.-rg, ~5 L/hr'kg) remained unchanged regardless of dose. The free fraction of the drug determined in the nonpregnant sheep plasma by equilibrium dialysis averaged 0.229 ± 0.080, and the extent of drug binding to plasma protein was independent of the drug concentration encountered (30-780 ng/mL) in the nonpregnant sheep in vivo. Concentration-independent binding of the drug was also confirmed by in vitro binding studies over the drug concentration range of 10-2,000 ng/mL. Therefore, it appears that changes in the volume of distribution are likely to be a result of changes in tissue uptake or binding of the drug, as a function of dose. The placental transfer and pharmacokinetics of diphenhydramine were studied in the chronically catheterized pregnant ewe and fetus following i.v. bolus administration of a 100 mg dose to the ewe (n = 4) over the gestational age range of 122- 129 days (term 145 days). Serial samples of the maternal and fetal arterial blood, fetal tracheal and amniotic fluids were collected for drug analysis. Rapid placental transfer was observed with peak fetal plasma concentrations occurring within 5 min after injection. The fetal to maternal AUC ratio averaged 0.85, indicating significant fetal exposure to the drug. The average elimination half-life in the ewe (52 min) was not significantly different from that obtained in the fetus (46 min). The maternal total body clearance and the volume of distribution were 3.6 L/hr-kg and 3.2 L/kg, respectively. There was no significant difference in any of these pharmacokinetic parameters between the pregnant and nonpregnant sheep. Approximately -95% of the dose 177 injected to the ewe appeared to be eliminated by the mother via maternal nonplacental pathways, with only a small portion of the dose (-5%) eliminated by the fetus. There was a significant accumulation of the drug in the fetal tracheal fluid, with the mean ratio of the fetal tracheal to plasma drug concentrations of 2.8. Diphenhydramine was also accumulated in the amniotic fluid, but to a lesser extent than in tracheal fluid (amniotic fluid to fetal plasma drug concentration ratio, -0.3). In the fetus, diphenhydramine that is accumulated in the tracheal fluid could be either swallowed by the fetus or transferred into the amniotic cavity for further recirculation of the drug within the fetus. Transplacental and nonplacental clearances of diphenhydramine were determined by use of a general two-compartment open model, with drug elimination occurring from both compartments. The drug was given by simultaneous i.v. bolus injection and infusion to steady-state to the chronically catheterized pregnant ewe and, in a separate experiment, to the fetus (n = 8), approximately 48 hrs apart. Average steady-state plasma drug concentrations after maternal infusion was 212.1 ± 67.8 ng/mL in the mother and 36.3 ± 14.4 ng/mL in the fetus, resulting in a fetal to maternal concentration ratio of 0.19 ± 0.10. Following fetal infusions, maternal and fetal steady-state drug concentrations were 31.1 ± 11.6 ng/mL and 447.6 ± 185.2 ng/mL, respectively. The plasma free fraction of diphenhydramine in the pregnant ewe (0.141 ± 0.079) is not significantly different from that determined in the nonpregnant sheep (0.229 ± 0.080), but is significantly smaller than that in the fetus (0.277 ± 0.087). The total fetal clearance (CLff, 472.7 ± 215.7 mL/min) was relatively small as compared to the total maternal clearance (CLmm, 3426.1 ± 905.8 mL/min). 178 The transplacental clearance from fetus to mother (CLfm, 264.4 ± 138.7 mL/min) was ~3 fold higher than that from mother to fetus (CLmf, 82.4 ± 40.5 mL/min). Maternal nonplacental clearance (CLmo, 3343.8 ± 890.7 mL/min) accounted for 97.8 ± 1.1% of the total maternal clearance whereas the fetal nonplacental clearance (CLf0, 208.4 ± 80.4 mL/min) accounted for only 45.2 ± 4.8% of the total fetal clearance. Therefore, diphenhydramine administered to the ewe appears to be eliminated mainly by nonplacental pathways, viz, hepatic, renal and pulmonary metabolism in the mother. However, in the case where the drug is given to the fetus, both the fetal nonplacental and placental clearances may be important for elimination of the drug from the fetus. The maternal and fetal elimination half-life of the drug obtained after maternal infusion (51 ± 12 min and 63 ± 32 min, respectively) are similar to those after fetal infusions (53 ± 53 and 73 ± 25 min, respectively). Diphenhydramine was accumulated in the fetal tracheal fluid, with an average fetal tracheal to plasma drug concentration ratio of 3.4 ± 3.6 for maternal infusions and 5.4 ± 4.5 for fetal infusions. The elimination half-life of the drug in tracheal fluid averaged 49 ± 17 min after maternal infusion and 56 ± 16 min after fetal infusion. These values are significantly shorter than that calculated in the amniotic fluid (123 ± 32 min). All these elimination half-lives were comparable to those obtained after maternal i.v. bolus injection. Diphenhydramine concentrations in the tracheal fluid were higher than the corresponding amniotic fluid drug concentrations. It is probable that the excretion of drug in tracheal fluid in part accounts for the accumulation of the drug in the amniotic fluid since the tracheal fluid is continuously produced at a 179 constant rate from the lungs and flows out, and is either swallowed by the fetus or secreted into the amniotic sac. The effects of diphenhydramine on fetal behavioural states, i.e., fetal electrocortical (ECoG), electro-ocular (EoG) and breathing activities, blood gas status, arterial pressure and heart rate were investigated in the fetal lamb following maternal and fetal drug infusions to steady-state. During maternal drug infusion, there were declines in the percentage of low voltage ECoG pattern (from 55 to 46%), the percentage of low voltage activity containing rapid eye movements (from 80 to 55%), the overall incidence of fetal breathing (from 42 to 21%) and in the amount of breathing during low voltage ECoG activity (from 67 to 36%). These sedative-like effects occurred at fetal plasma drug concentrations (-36 ng/mL) lower than those resulting in discernible central nervous system effects in adults. Drug infusion to fetus achieved higher fetal plasma drug levels (-448 ng/mL) and resulted in a transient decline in arterial P02 and pH, associated with transient tachycardia and vigorous breathing movements during the initial portion of the infusion. There was also a significant fall in the amount of low voltage ECoG pattern (from 51 to 26%) and marked increases in the amount of intermediate voltage pattern (from 8 to 46%). The two routes of diphenhydramine administration, therefore, resulted in two distinct patterns of fetal behavioural effects, which could be attributed to the markedly different fetal plasma drug concentrations achieved. Given the high use of diphenhydramine during pregnancy, further examination of the effects of the drug on the fetus, in both animals and humans, is warranted. 180 The possibility of drug recirculation from fetal tracheal and amniotic fluids back to the fetus via several possible routes, i.e., swallowing of amniotic or lung fluids and uptake from amniotic fluid by the fetal and maternal membranes, was examined. A 50 mg dose of diphenhydramine hydrochloride was injected into the amniotic cavity in pregnant sheep (n = 5) over the gestational age of 122-137 days. The peak plasma drug concentration obtained in the fetus (228.0 ± 101.5 ng/mL) is significantly higher than that in the ewe (43.5 ± 15.6 ng/mL). Also the plasma drug concentration in fetus was consistently higher than the corresponding value in the ewe throughout the sampling period, with the mean maternal to fetal ratio of 0.23 ± 0.10. The elimination half-life of the drug in the amniotic fluid (11.0 ± 4.5 hr) was comparable to that in the fetal tracheal fluid (10.8 ± 3.8 hr) but was significantly longer than that observed in fetal (7.7 ± 1.2 hr) or maternal (1.7 ± 0.5 hr) plasma. These elimination half-lives were -2-10 times greater than those observed after maternal or fetal i.v. administration. The elimination half-lives following the intra-amniotic injection would be expected to be influenced primarily by the absorption of drug across fetal and maternal membranes, 7.e., absorption is the rate-limiting step. In the fetus, the tracheal to plasma drug concentration ratio after intra-amniotic injection averaged 16.4 ± 18.0 (range 4.2-45.9). This ratio was greater than that obtained after either the maternal i.v. bolus injection (2.7 ± 2.4), maternal i.v. infusion (3.4 ± 3.6) or fetal i.v. infusion (5.4 ± 4.5). Therefore, when the drug was injected into the amniotic cavity, it appeared to be taken up by direct diffusion from the amniotic fluid to fetal tracheal fluid due to a large concentration 181 gradient between the two fluids. The percentage of total amount of the drug available in the maternal-fetal sheep that was eliminated by the fetus (37.1%) was greater than that after maternal i.v. bolus injection (5.0%) or maternal i.v. infusion (1.4%) and was somewhat less than, but comparable to, that after direct infusion of the drug to the fetus (45.0%). It appears that 63 ± 8% of the total drug available in the maternal-fetal sheep after intra-amniotic injection was eliminated by the mother, of which, an average of 47% was received from the fetus and the rest (16%) from the amniotic fluid via direct diffusion across the uterus. Direct diffusion of the drug across the uterus, therefore, appears to play a relatively minor role in the transfer of the drug from the amniotic fluid to the mother. Overall, the data suggest preferential uptake of the drug from amniotic fluid by the fetus, resulting in a much greater degree of fetal drug exposure than maternal exposure. The pulmonary disposition of diphenhydramine in the fetal lamb was studied in the chronically instrumented pregnant sheep (n = 3) over the gestational age of 128-133 days. The drug was given to the ewe via the maternal femoral catheter or to the fetus via the fetal jugular vein by simultaneous i.v. bolus injection and infusion to steady-state. During maternal infusion, the steady-state drug concentration in carotid arterial plasma was higher than that in pulmonary arterial plasma due to preferential distribution of drug-rich inferior vena caval blood through the foramen ovale into the left heart and ascending aorta. Drug concentration in the pulmonary arterial plasma was consistently higher than in the pulmonary venous plasma throughout the maternal infusion experiment. In contrast, during fetal infusion, the steady-state drug concentration in 182 fetal pulmonary arterial plasma (598.4 ng/mL) was higher than in carotid arterial (479.0 ng/mL) or pulmonary venous plasma (536.1 ng/mL) as the drug was delivered from the jugular vein to the right heart and pulmonary artery with the rest of superior vena caval blood return. Following cessation of the fetal drug infusion, however, there was a reversal of the drug concentrations, resulting in higher drug concentration in pulmonary venous plasma than in pulmonary arterial plasma, which indicated net release of the intact diphenhydramine from the lung into fetal circulation. The average percent extraction by the left fetal lamb lung, calculated by use of the steady-state pulmonary arterial and venous concentrations, averaged 0. 08 ± 0.06 (range, 0.04-0.14) in utero. Thus, despite the low rate of pulmonary blood flow in the fetus, the delivery of the drug to the lungs via the pulmonary circulation would appear to be sufficient to account for the drug accumulation observed in the fetal tracheal fluid. Briefly, the findings from the present study may be summarized as follows: 1. A capillary GC/NPD assay method with improved sensitivity developed for diphenhydramine (lower sensitivity limit, -40 pg at the detector) was found to be suitable for the pharmacokinetic assessment of the drug in sheep. 2. Dose dependency of the pharmacokinetics of diphenhydramine was examined in the nonpregnant sheep over the dose range of 25-200 mg: a) There were significant increases in the elimination half-life (34 to 68 min) and the volume of distribution (3 to 6 L/kg), with increasing dose, b) There was, however, a proportional increase in AUC, with no change in CZ.TB (5 183 L/hr/kg). c) The average free fraction of diphenhydramine unbound to plasma protein (0.229) was independent of plasma drug concentrations in vivo. 3. Disposition of diphenhydramine in the maternal-fetal sheep was studied after a 100 mg maternal i.v. bolus dose: a) There was a rapid and extensive placental transfer of the drug from the ewe to the fetus, as reflected by the average fetal to maternal AUC ratio of 0.85. b) There were no significant differences in the distribution and elimination half- lives of the drug between the pregnant ewe and fetal lamb, c) There was no significant difference in the pharmacokinetics of the drug between pregnant and nonpregnant ewes, d) Diphenhydramine accumulated in amniotic fluid and, to a greater extent, in fetal tracheal fluid. Accumulation of drug in these fluids may play a role in fetal drug elimination and/or recirculation in the pregnant sheep. 4. Transplacental and nonplacental clearances of diphenhydramine were studied following separate maternal and fetal infusions to steady-state: a) The total body clearance of the drug in the ewe (CLmm, 3426.1 mL/min) was significantly greater than that in the fetus (472.7 mL/min), but when normalized to body weight, the fetal clearance value was higher (223.9 vs. 44.3 mL/min). b) The fetal to maternal transplacental clearance (CLfm, 264.4 mL/min) was -3 fold greater than the maternal to fetal transplacental clearance (CLmf, 82.4 mL/min). c) The average contribution of the placental clearance to the total drug clearance accounted for -2% in the ewe and -55% in the fetus, d) Therefore, the drug administered to the ewe appears to be cleared mainly via maternal nonplacental pathways, whereas in 184 the fetus, both the fetal nonplacental as well as fetal to maternal transplacental clearances are important. 5. Fetal effects of diphenhydramine were studied following maternal and fetal infusions to steady-state: a) With low fetal drug concentrations (-40 ng/mL), sedative-like effects were observed in the fetus, comprising reduction in the amounts of REM sleep and fetal breathing activity, b) With higher drug levels (-450 ng/mL), there was transient vigorous breathing activity and a disruption in the normal cycling of ECoG activity, as reflected in a marked increase in the amount of a transitional ECoG pattern. 6. The possibility of recirculation of diphenhydramine from the amniotic fluid to the ewe and fetus was studied after intra-amniotic drug administration: a) The drug given into the amniotic cavity appears to be taken preferentially into the fetal circulation (-84% of the total drug) via the chorioallantois and/or fetal swallowing, of which -37% is eliminated via fetal nonplacental pathways and -47% eliminated by the fetal to maternal transplacental pathway, b) Approximately 63% of the total drug was eliminated by the ewe, of which -16% was received from the amniotic fluid via direct diffusion across the uterus; the remaining portion (-47%) was received from the fetus via the fetal to maternal placental pathway. 7. The pulmonary extraction of diphenhydramine in the fetal lamb was studied following maternal or fetal infusion to steady-state: a) The pulmonary extraction ratio averaged 0.08 in the fetal lamb in utero. b) Even with the low rate of pulmonary blood flow in the fetus, delivery of 185 the drug to the lung via the pulmonary circulation can account for the accumulation of the drug in the fetal tracheal fluid. The placental permeability to lipid soluble compounds does not differ greatly in different species and, therefore, diphenhydramine is likely to be transferred from mother to fetus in human pregnancy. Whether similar patterns of the disposition of the drug would occur in humans is not known. The results from the pregnant sheep model may not be directly extrapolated to human pregnancy and further studies of the effect of the drug on human fetal behaviour are warranted. 6. REFERENCES Abernethy, D.R.; Greenblatt, D.J.: Diphenhydramine determination in human plasma by gas-liquid chromatography using nitrogen- phosphorus detection: application to single low-dose pharmacokinetic studies. J. Pharm. Sci. 72:941-943 (1983). Ahmed, A.R.; Kaplan, R.: Pruritic urticarial papules and plaques of pregnancy. J. Am. Acad. Dermatol. 4:679-681 (1981). Albert, K.S.; Sakmar, E.; Morais, J.A.; Hallmark, M.R.; Wagner, J.G.: Determination of diphenhydramine in plasma by gas chromatography. Res. Commum. Chem. Pathol. Pharmacol. 7:95-103 (1974). Albert, K.S.; Hallmark, M.R.; Sakmar, E.; Weidler, D.J.; Wagner, J.G.: Pharmacokinetics of diphenhydramine in man. J. Pharmacokin. Biopharm. 3:159-170 (1975). Anderson, M.W.; Orton, T.C.; Pickett, R.D.; Eling, T.E.: Accumulation amines in the isolated perfused rabbit lung. J. Pharmacol. Exp. Ther. 189:456-466 (1974). Anderson, D.F.; Phernetton, T.M.; Rankin, J.H.G.: The measurement of placental drug clearance in near-term sheep: Indomethacin. J. Pharmacol. Exp. Ther. 213:100-104 (1980). Barni Comparini, I.; Centiti, F.; Pariali, A.: Simultaneous determination of narcotics, adulterants and diluents in street samples by means of gas chromatography with capillary columns. J. Chromatogr. 279:609-613 (1983). Baugh, R.; Calvert, R.T.: A rapid method for the determination of diphenhydramine in plasma. Br. J. Clin. Pharmacol. 3:1062-1064 (1976). Berlinger, W.G.; Goldberg, M.J.; Spector, R.; Chiang, C; Ghoneim, M.M Diphenhydramine: kinetics and psychomotor effects in elderly women. Clin. Pharmacol. Ther. 32:387-391 (1982). Berger, F.M.: The chemistry and mode of action of tranquilizing drugs. Ann. N.Y. Acad. Sci. 67:685-700 (1956). Bilzer, W.; Gundert-Remy, U.: Determination of nanogram quantities of diphenhydramine and orphenadrine in human plasma using gas- liquid chromatography. Eur. J. Clin. Pharmacol. 6:268-270 (1973). Bloedow, D.C.; Ralston, D.H.; Hargrove, J.C: Lidocaine pharmacokineti in pregnant and nonpregnant sheep. J. Pharm. Sci. 69:32-37 (1980). Blyden, G.T.; Greenblatt, D.J.; Scavone, J.M.; Shader, R.I.: 187 Pharmacokinetics of diphenhydramine and a demethylated metabolite following intravenous and oral administration. 0. Clin. Pharmacol. 26:529-533 (1986). Boddy, K.; Dawes, G.S.; Fischer, R.L.; Pinter, S.; Robinson, J.S.: The effects of pentobarbitone and pethidine on foetal breathing movements in sheep. Br. J. Pharmac. 57:311-317 (1976). Bonds, D.R.; Anderson, S.; Meschia, G.: Transplacental diffusion of ethanol under steady state conditions. J. Dev. Physiol. 2:409-416 (1980). Brademas, M.E.: Nonmedical use of butorphanol and diphenhydramine. J. A. M. A. 252:1010 (1984). Bradley, R.M.; Mistretta, CM.: Swallowing in fetal lamb. Science 179:1016-1017 (1973). Bray, R.E.; Boe, R.W.; Johnson, W.L.: Transfer of ampicillin into fetus and amniotic fluid from maternal plasma in late pregnancy. Am. J. Obstet. Gynecol. 96:938-942 (1966). Brocklebank, J.C.; Ray, W.A.; Federspiel, C.F.; Schaffner, W.: Drug prescribing during pregnancy. Am. J. Obstet. Gynecol. 132:235-244 (1978). Brown, R.; Patrick, J.: The nonstress test: how long is enough? Am. J. Obst. Gynecol. 141:646-650 (1981). Burns, J.J.; Conney, A.H.; Koster, R.: Stimulatory effect of chronic drug administration on drug-metabolizing enzymes in liver microsomes. Ann. N.Y. Acad. Sci. 104:881-893 (1963). Burns, M.; Moskowitz, H.: Effects of diphenhydramine and alcohol on skills performance. Eur. J. Clin. Pharmacol. 17:259-266 (1980). Burt, R.A.P.: The foetal and maternal pharmacology of some of the drugs for the relief of pain in labour. Brit. J. Anaesth. 43:824-836 (1971). Carruthers, S.G.; Shoeman, D.W.; Hignite, C.E.; Azarnoff, D.W.: Correlation between plasma diphenhydramine level and sedative and antihistamine effects. Clin. Pharmacol. Ther. 23:375-382 (1978). Chang, T.; Okerholm, R.A.; Glazko, A.J.: Identification of diphenhydramine (BenadrylR) metabolites in human subjects. Res. Commun. Chem. Pathol. Pharmacol. 9:391-404 (1974). Chapman, R.L.K.; Dawes, G.S.; Rurak, D.W.; Wilds, P.L.: Intermittent breathing before death in fetal lambs. Am. J. Obstet. Gynecol. 131:894-898 (1978). 188 Chiarotti, M.; Carnevale, A.; Giovanni, N.: Capillary gas chromatographic analysis of illicit diamorphine preparations. Forensic Sci. Int. 21:245-251 (1983). Ching, M.S.; Jones, D.B.; Morgan, D.J.; Mihaly, G.W.; Hardy, K.J.; Smallwood, R.A.: Fetal exposure to cimetidine following chronic administration to pregnant sheep. Res. Commun. Chem. Pathol. Pharmacol. 50:139-142 (1985). Cohen, A.F.; Hamilton, M.J.; Peck, A.W.: The effects of acrivastine (BW 825C), diphenhydramine and terfenadine in combination with alcohol on human CNS performance. Eur. J. Clin. Pharmacol. 32:279-288 (1987). Cohn, H.E.; Piasecki, G.J.; Jackson, B.T.: The effect of ^-adrenergic stimulation on fetal cardiovascular function during hypoxemia. Am. J. Obstet. Gynecol. 144:810-816 (1982). Collins, J.M.; Dedrick, R.L.: Contribution of lungs to total body clearance: linear and nonlinear effects. J. Pharm. Sci. 71:66-70 (1982). Conney, A.H.; Burns, J.J.: Biochemical pharmacological considerations of zoxazolamine and chlorzoxazone metabolism. Ann. N.Y. Acad. Sci. 86:167-177 (1960). Costello, C.E.: CRC handbook of mass spectra of drugs, Ed. by I. Sunshine, CRC Press, Boca Raton, FL, pp. 37 and 110, 1981. Coyle, I.; Wayner, M.J.; Singer, G.: Neural and behavioural teratology, Advances in the study of Birth Defects, Vol. 4, Ed. by T.V.N. Persaud, MTP Press, Lancaster, pp. 111-133, 1980. Dailey, P.A.; Baysinger, C.L.; Levinson, G.; Snhider, S.M.: Neurobehavioural testing of the newborn infant. Clinics in Perinatol. 9:191-214 (1982). Dawes, G.S.; Fox, H.E.; Leduc, B.M.; Liggins, G.C.; Richards, R.T.: Respiratory movements and rapid eye movement sleep in the foetal lamb. J. Physiol. 220:119-143 (1972). Dawes, G.S.: The central control of fetal breathing and skeletal muscle movements. J. Physiol. 346:1-18 (1984). Dobbing, J.: The biology of human fetal growth, Ed. by D.F. Roberts and A.M. Thomson, Taylor & Francis Ltd., London, pp. 137-147, 1976. Doering, P.L.; Stewart, R.B.: The extent and character of drug consumption during pregnancy. J. A. M. A. 239:843-846 (1978). Douglas, W.W.: The pharmacological basis of therapeutics, Ed. by A.G. Gilman, L.S. Goodman, T.W. Rail and F. Murad, MacMillan, New York, pp. 604-638, 1985. 189 Drach, J.C.; Howell, J.P.: Identification of diphenhydramine urinary metabolites in the rhesus monkey. Biochem. Pharmacol. 17:2125-2136 (1968). Drach, J.C.; Howell, O.P.; Borondy, P.E.; Glazko, A.J.: Species differences in the metabolism of diphenhydramine (BenadrylR). Proc. Soc. Exp. Biol. Med. 135:849-853 (1970). Drouin, M.A.: HI histamines: perspective on the use of the conventional and new agents. Ann. Allergy 55:747-752 (1985). Faingold, C.L.; Berry, C.A.: Effects of antihistamine agents upon the electrographic activity of the cat brain: a power spectral density study. Neuropharmacology 11:491-498 (1972). Faber, J.J.: Steady-state methods for the study of placental exchange. Fed. Proc. 36:2640-2646 (1977). Feldman, M.D.; Behar, M.: A case of massive diphenhydramine abuse and withdrawal from use of the drug. J. A. M. A. 255:3119-3120 (1986). Fewell, J.E.; Lee, CC; Kitterman, J.A.: Effects of phrenic nerve section on the respiratory system of fetal lambs. J. Appl. Physiol. 51:293-297 (1981). Filloux, F.: Toxic encephalopathy caused by topically applied diphenhydramine. J. Pediatrics 108:1018-1019 (1986). Fink, M.; Irwin, P.: CNS effects of the antihistamines diphenhydramine and terfenadine (RMI 9918). Pharmacopsychiatry 12:335-344 (1979). Forfar, J.O.; Nelson, M.M.: Epidemiology of drugs taken by pregnant women: Drugs that may affect the fetus adversely. Clin. Pharmacol. Ther. 14:632-642 (1973). Gibaldi, M.; Levy, G.; McNamara, P.J.: Commentaries: Effect of plasma protein and tissue binding on the biologic half-life of drugs. Clin. Pharmacol. Ther. 24:1-4 (1978). Gibaldi, M.; Perrier, D.: Pharmacokinetics, Ed. by J. Swarbrick Marcel Dekker, New York, pp. 45-111, 199-219, 319-353, 1982. Gill is, A.M.; Yee, Y-G.; Kates, R.E.: Binding of antiarrythmic drugs to purified human alpha-l-acid glycoprotein. Biochem. Pharmacol. 34:4279-4282 (1985). Glazko, A.J.; Dill, W.A.: Biochemical studies on diphenhydramine (Benadryl). II. Distribution in tissues and urinary excretion. J. Biol. Chem. 179:403-408 (1949a). Glazko, A.J.; Dill, W.A.: Biochemical studies on diphenhydramine (Benadryl). IV. Degradation of benadryl by tissue enzymes. 190 J Biol. Chem. 179:417-422 (1949b). Glazko, A.J.; McGinty, D.A.; Dill, W.A.; Wilson, M.L.; Ward, C.S.: Biochemical studies on diphenhydramine. J. Biol. Chem. 179:409-416 (1949). Glazko, A.J.; Dill, W.A.; Fransway, R.L.: Determination of diphenhydramine blood levels using a new fluorescent dye-salt procedure generally applicable to basic organic compounds. Fed. Proc. 21:269 (1962). Glazko, A.J.; Dill, W.A.; Young, R.M.; Smith, T.C.; Ogilvie, R.I.: Metabolic disposition of diphenhydramine, Clin. Pharmacol. Ther. 16:1066-1076 (1974). Goldstein, L.; Murphree, H.B.; Pfeiffer, CC: Comparative study of EEG effects of antihistamines in normal volunteers. J. Clin. Pharmacol. 8:42-53 (1968). Golub, M.S.; Eisele, J.H.; Anderson, J.H.: Maternal-fetal distribution of morphine and alfentanil in near-term sheep and rhesus monkeys. Dev. Pharmacol. Ther. 9:12-22 (1986). Grob, K.; Grob, G.: Splitless injection on capillary columns, Part I. The basic techniques; steroid analysis as an example. J. Chromatogr. Sci. 7:584-586 (1969a). Grob, k.; Grob, G.: Splitless injection on capillary columns, Part II. Conditions, limits, practical realization. J. Chromatogr. Sci. 7:587-591 (1969b). Grob, K.; Grob, K. Jr.: Isothermal analysis on capillary columns without stream splitting: The role of the solvent. J. Chromatogr. 94:53-64 (1974). Gruhzit, O.M.; Fisken, R.A.: A toxicologic study of two histamine antagonists of the benzhydryl alkamine ether group. J. Pharmacol. Exp. Ther. 89:227-233 (1947). Gupta, K.P.; Bhargava, K.P.; Ali, B.: Modification of methaqualone pharmacokinetics by diphenhydramine. J. Pharm. Pharmacol. 34:744-746 (1982). Hald, J.: The excretion of diphenhydramine hydrochloride (dimethyl - aminoethylbenzhydryl ether hydrochloride) in urine of rabbits and man. Acta Pharmacol. Toxicol. 3:296-302 (1947). Hansson, E.; Schmiterlow, C.G.: A comparison of the distribution, excretion and metabolism of a tertiary (promethazine) and a quaternary (Aprobit) phenothiazine compound labelled with S35. Arch. Int. Pharmacodyn. 131:309-324 (1961). Harding, R.; Siggers, J.N.; Wickham, P.J.: Fetal and maternal influences on arterial oxygen levels in the sheep fetus. J. Dev.Physiol. 191 5:267-276 (1983). Harding, R.: Fetal physiology and medicine, Ed. by R.W. Beard and P.W. Nathanielsz, Marcel Dekker, New York, pp. 255-285, 1984. Harding, R.; Bocking, A.D.; Sigger, J.N.; Wickham, P.J.D.: Composition and volume of fluid swallowed by fetal sheep. Quart. J. Exp. Physiol. 69:487-495 (1984). Hausmann, E.; Wewer, H.; Wellhoner, H.H.; Weller, J.: Lethal intoxica• tion with diphenhydramine. Arch. Toxicol. 53:33-39 (1983). Heinonen, O.P.; Slone, D.; Shapiro, S.: Birth defects and drugs in pregnancy, Ed. by O.P. Heinonen, D. Slone and S. Shapiro, Publishing Sciences Group, Littleton, pp. 322-334, 1977. Hestand, H.E.; Teske, D.W.: Diphenhydramine hydrochloride intoxication. J. Pediatr. 90:1017-1018 (1977). Hill, R.M.: Drugs ingested by pregnant women. Clin. Pharmacol. Ther. 14:654-659 (1973). Hill, M.D.; Craft, J.B.Jr.; Abramson, F.P.: Elevated plasma glycoproteins in fetal sheep do not bind basic lipophilic drugs. J. Pharm. Sci. 75:881-882 (1986). Hill, K.J.; Lumbers, E.R.: Renal function in adult and fetal sheep. J. Dev. Physiol. 10:149-159 (1988). Hirschowitz, B.I.; Molina, E.: Anticholinergic potency of diphenhydramine measured against Bethanecol in the gastric fistula dog. J. Pharmacol. Exp. Ther. 226:171-173 (1983). Hori, R.; Okumura, K.; Yoshida, H.: Binding of basic drugs to rat lung mitochondria. Pharm. Res. 4:142-146 (1987). Hutchings, D.E.: Studies on the development of behaviour and the nervous system. Early influences, Vol. 4, Ed. by G. Gottlieb, Academic Press, New York, pp. 7-34, 1978. lams, J.D.; Rayburn, W.F.: Drug therapy in obstetrics and gynecology, Ed. by W.F. Rayburn and F.P. Zuspan, Appleton-Century-Crofts, Norwalk, pp. 9-17, 1982. Ichibagase, H.; Imamura, Y.; Shiozu, K.: Effect of simultaneous administration of drugs on absorption and excretion. XI. Effect of diphenhydramine on sulfisomidine absorption in rabbits. Chem. Pharm. Bull. 29:887-891 (1981). Imamura, Y.; Wang, L.H.; Lee, C.S.; Perrin, J.H.; Shiozu, K.; Ichibagase, H.: Acetaminophen-diphenhydramine interaction in rabbits. Int. J. Pharmaceutics, 8:277-284 (1981). Jansen, A.H.; Chernick, V.: Development of respiratory control. Physiol. Rev. 63:437-483 (1983). Jansen, A.H.; Ioffe, S.; Chernick, V.: Drug-induced changes in fetal breathing activity and sleep state. Can. J. Physiol. Pharmacol. 61:315-324 (1983). Jones-Price, C; Kimmel, C.A.; Ledoux, T.A.; Reel, J.R.; Langhoff- Paschke, L.; Marr, M.: Teratologic evaluation of diphenhydramine hydrochloride in CD-I mice. Teratology 27:53A (1983) . Kargas, G.A.; Kargas, S.A.; Bruyere, H.J.; Gilbert, E.F.: Perinatal mortality due to interaction of diphenhydramine and temazepam. N. Engl. J. Med. 313:1417-1418 (1985). Katz, M.L.; Bergman, E.N.: Simultaneous measurements of hepatic and portal venous blood flow in the sheep and dog. Am. J. Physiol. 216:946-952 (1969). Kimmel, C.A.: Developmental Toxicology, Ed. by CA. Kimmel and J. Buelke-Sam, Raven Press, New York, pp. 321-331, 1981. King, C.T.G.; Weaver, S.A.; Narrod, S.A.: Antihistamines and teratoge- necity in the rat. J. Pharmacol. Exp. Ther. 147:391-398 (1965). Kitterman, J.A.; Liggins, G.C; Clements, J.A.; Tooley, W.H.: Stimulation of breathing movements in fetal sheep by inhibitors of prostaglandin synthesis. J. Dev. Physiol. 1:453-466 (1979). Kitterman, J.A.: Fetal lung development. J. Dev. Physiol. 6:67-82 (1984) . Koppel, C; Ibe, K.: Clinical symptomatology of diphenhydramine overdose: An evaluation of 136 cases in 1982 to 1985. J. Toxicol. Clin. Toxicol. 25:53-70 (1987). Krogh, C.M.E.: Compendium of pharmaceuticals and specialties, Ed. by CM.E. Krogh, Southam Murray, Toronto, pp. 265, 1987. Kuhnert, B.R.; Kuhnert, P.M.; Prochaska, A.L.; Sokol, R.J.: Meperidine disposition in mother, neonate, and pregnant females. Clin. Pharmacol. Ther. 27:486-491 (1980). Lau, H.Y.A.; Pearce, F.L.: Dual effect of antihistamines on rat peritoneal mast cells: induction and inhibition of histamine release. Agents Actions 16:176-177 (1985). Lavigne, J.; Marchand, C: Inhibition of the gastrointestinal absorpti of p-aminosalicylate (PAS) in rats and humans by diphenhydramine. Clin. Pharmacol. Ther. 14:404-412 (1973). Levi, R.; Kuye, J.O.: Pharmacological characterization of cardiac histamine receptors: sensitivity to Hl-receptor antagonists. Eur. J. Pharmacol. 27:330-338 (1974). 193 Levi, R.; Owen, A.A.; Trzeciakowski, J.: Pharmacology of histamine receptors, Ed. by C.R. Ganellin and M.E. Parsons, Wright.PSG., London, pp. 236-297, 1982. Levy, G.; Hayton, W.L.: Fetal pharmacology, Ed. by L. Boreus, Raven Press, New York, pp. 29-39, 1973. Lingwood, B.E.; Wintour, E.M.: Amniotic fluid volume and in vivo permeability of ovine fetal membranes. Obstet. Gynecol. 64:368-372 (1984) Lipsky, S.R.; McMurray, W.J.; Hernandez, M.: Fused silica capillary columns for gas chromatographic analyses. J. Chromatogr. Sci. 18:1-9 (1980). Loew, E.R.: Pharmacology of antihistamine compounds. Physiol. Rev. 27:542-573 (1947). Lotgering, F.K.; Wallenburg, C.S.: Mechanisms of production and clearance of amniotic fluid. Seminars in Perinatol. 10:94-102 (1986). Lutz, D.; Gielsdorf, W.; Jaeger, H.: Quantitative determination of diphenhydramine and orphenadrine in human serum by capillary gas chromatography. J. Clin. Chem. Clin. Biochem. 21:595-597 (1983). Macfarlane, W.V.: Blood of Sheep; Ed. by M.H. Blunt, Springer-Verlag, New York, pp 1-27, 1975. Malony, J.E.; Bowes, G.; Brodecky, V.; Dennett, X.; Wilkinson, M.; Walker, A.: Function of the future respiratory system in the growth retarded fetal sheep. J. Dev. Physiol. 4:279-297 (1982). Manning, F.A.; Plat, L.D.; Sipos, L.: Antepartum fetal evaluation: Development of a fetal biophysical profile score. Am. J. Obstet. Gynecol. 136:787-792 (1980). Marshall, P.B.: Some chemical and physical properties associated with histamine antagonism. Brit. J. Pharmacol. 10:270-278 (1955). McColl, J.D.; Robinson, S.; Globus, M.: Effect of some therapeutic agents on the rabbit fetus. Toxicol. Appl. Pharmacol. 10:244-252 (1967). McLeod, W.; Brien, J.; Loomis, C; Carmichael, L.; Probert, C; Patrick, J.: Effect of maternal ethanol ingestion on fetal breathing movements, gross body movements, and heart rate at 37 to 40 weeks' gestational age. Am. J. Obstet. Gynecol. 145:251-257 (1983). Melville, K.I.: Histamine and antihistamines, Ed. by D. Bovet, Pergamon, Oxford, pp. 127-171, 1973. 194 Meatherall, R.C.; Guay, D.R.P.: Isothermal gas chromatographic analysis of diphenhydramine after direct injection onto a fused-silica capillary column. J. Chromatogr. 307:295-304 (1984). Meredith, C.G.; Christian, CD.; Johnson, R.F.; Madhavan, S.V.; Schenker, S.: Diphenhydramine disposition in chronic liver disease. Clin. Pharmacol. Ther. 35:474-479 (1984). Mescher, E.J.; Platzker, A.C.G.; Ballard, P.L.; Kitterman, J.A.; Clements, J.A.; Tooley, W.H.: Ontogeny of tracheal fluid, pulmonary surfactant, and plasma corticoids in the fetal lamb. J. Appl. Physiol. 39:1017-1021 (1975). Meschia, G; Battaglia, F.G.; Bruns, P.D.: Theoretical and experimental study of transplacental diffusion. J. Appl. Physiol. 22:1171-1178 (1967). Metzler, CM.; Elfring, G.K.; McEwen, A.: Package of computer programs for pharmacokinetic modeling. Biometrics, 30:562-563 (1974). Mihaly, G.W.; Morgan, D.J.; Marshall, A.W.; Smallwood, R.A.; Cockbain, S.; MacLellan, D.; Hardy, K.J.: Placental transfer of ranitidine during steady-state infusions of maternal and fetal sheep. J. Pharm. Sci. 71:1008-1010 (1982). Mihaly, G.W.; Jones, D.B.; Morgan, D.J.; Ching, M.S.; Webster, L.K.; Smallwood, R.A.; Hardy, K.J.: Placental transfer and renal elimination of cimetidine in maternal and fetal sheep. J. Pharmacol. Exp. Ther. 227:441-445 (1983). Mihaly, G.W.; Morgan, D.J.: Placental drug transfer: Effects of gestational age and species. Pharmacol. Ther. 23:253-266 (1984). Mota, I.; Dias Da Silva, W.: The anti-anaphylactic and histamine- releasing properties of the antihistamines. The effect on the mast cells. Br. J. Pharmac. 15:396-404 (1960). Naranzo, P.; Naranzo,, E.: Embryotoxic effects of antihistamines. Arzneim. Forsch. 18:188-195 (1968). Natale, R.; Clewlow, F.; Dawes, G.S.: Measurement of fetal forelimb movements in the lamb in utero. Am. J. Obstet. Gynecol. 140:545-551 (1981). Nicholson, A.N.: Antihistamines and sedation. Lancet 2:211-212 (1983). Nijhuis, J.G.; Prechtl, HFR; Martin, C.B.; Bots, R.S.G.M.: Are there behavioural states in the human fetus? Early Human Dev. 6:177-195 (1982). Olver, R.E.; Strang, L.B.: Ion fluxes across the pulmonary epithelium 195 and the secretion of lung liquid in the foetal lamb. J. Physiol. 241:327-357 (1974). Owens, J.A.; Falconer, J.; Robinson, J.S.: Effect of restriction of placental growth on umbilical and uterine blood flows. Am. J. Physiol. 250:R427-R434 (1986). Parkin, D.E.: Probable Benadryl withdrawal manifestations in a new born infant. J. Pediatr. 85:580. (1974). Parry, R.; Calvert, R.T.: The distribution of diphenhydramine and four of its analogues in rabbits and implication for structure- activity relationship studies in drug design. Biopharm. Drug Dispos. 3:255-265 (1982). Patrick, J.; Richardson, B.; Hasen, G.; Clarke, D.; Wlodek, M.;Bousquet, J.; Brien, J.: Effects of maternal ethanol infusion on fetal cardiovascular and brain activity in lambs. Am. J. Obstet. Gynecol. 151:859-867 (1985). Peckham, C.H.; King, R.W.: A study of intercurrent conditions observed during pregnancy. Am. J. Obstet. Gynecol. 87:609-624 (1963). Pedersen, H.; Morishima, H.O.; Finster, M.; Arthur, G.R.; Covino, B.G.: Pharmacokinetics of etidocaine in fetal and neonatal lamb and adult sheep. Anesth. Anal. 61:104-108 (1982). Piercy, W.N.; Day, M.A.; Neims, A.H.; Williams, R.L.: Alteration of ovine fetal respiratory-like activity by diazepam, caffeine and doxapram. Am. J. Obstet. Gynecol. 127:43-49 (1977). Piper, J.M.; Baum, C; Kennedy, D.L.: Prescription drug use before and during pregnancy in a Medicaid population. Am. J. Obstet. Gynecol. 157:148-156 (1987). Pratt, W.R.: Allergic diseases in pregnancy and breast feeding. Ann. Allergy 47:355-360 (1981). Pritchard, J.A.: Deglution by normal and anencephalic fetuses. Obstet. Gynecol. 25:289-297 (1965). Quilligan, E.J.; Clewlow, F.; Johnston, B.M.; Walker, D.W.: Effect of 5-hydroxytryptophan on electrocortical activity and breathing movements of fetal sheep. Am. J. Obstet. Gynecol. 141:271-275 (1981). Rayburn, W.; Wible-Kant, J.; Bledsoe, P: Changing trends in drug use during pregnancy. J. Reprod. Med. 27:569-575 (1982). Read, M.D.: A new hypothesis of itching in pregnancy. Practitioner 218:845-847 (1977). Reichelderfer, T.E.; Livingston, S.; Auld, R.M.; Peck, J.L.: Treatment of acute Benadryl (diphenhydramine hydrochloride) intoxication with severe central nervous system changes and recovery. J. Pediatr. 46:303-307 (1955). Reyes-Jacang, A.; Wenzl, J.: Antihistamine toxicity in children. Clin. Pediatr. 8:297-299, (1969). Reynolds, F.: Placental transfer, Ed. by G.V.P. Chamberlain and A.W. Wilkinson, Pitman Medical, Bath, pp. 166-181, 1979. Rieveschl, G.; Gruhzit, O.M.: A toxicologic study of histamine antagonist /J-dimethylaminoethyl benzhydryl ether hydrochloride (benadryl). Fed Proc. 4:150 (1945). Rieveschl, G.: Dialkylaminoalkyl benzhydryl ethers. U.S. Patent 2,421,714 (1947), Chem. Abstr. 41:5550h (1947) Rigatto, H.; Blanco, C.E.; Walker, D.W.: The response to stimulation of hindlimb nerves in fetal sheep, in utero, during the different phases of electrocortical activity. J. Dev. Physiol. 4:175-185 (1982). Riggs, K.W.; Rurak, D.W.; Yoo, S.D.; McErlane, B.A.; Taylor, S.M.; McMorland, G.H.; Axelson, J.E.: Drug accumulation in the lung fluid of fetal lamb after maternal or fetal administration. Am. J. Obstet. Gynecol. 157:1286-1291 (1987a). Riggs, K.W.; Axelson, J.E.; McErlane, B.A.; Taylor, S.M.; Kwan, E.; McMorland, G.H.; Rurak, D.W.: Metoclopramide: placental and nonplacental clearances in maternal and fetal sheep following infusions to steady-state. The Japan-United States Congress of Pharmaceutical Sciences, December 1987b, Honolulu, Hawaii. Riggs, K.W.; Axelson, J.E.; Gruber, N.C.; McErlane, B.A.; McMorland, G.H.; Rurak, D.W.: Metoclopramide pharmacokinetics in pregnant and nonpregnant sheep. J. Pharm. Sci. 77:373-379 (1988a). Riggs, K.W.; Yoo, S.D.; Axelson, J.E.; Taylor, S.M.; McErlane, B.E.; Rurak, D.W.: Recirculation of drugs from amniotic fluid to the fetus in pregnant sheep. The 12th Annual Ontario-Quebec Perinatal Investigators Meeting, November 1988b, Kingston, Ontario. Rocha e Silva, M.: Handbuch der Experimentellen Pharmakologie, Vol. 18, Pt. 2, Ed. by M. Rocha e Silva, Springer-Verlag, Berlin, pp. 295-332, 1978. Rooney, T.A.: High resolution gas chromatography, Ed. by R.R. Freeman, company, place, pp. 7-29, 141-182, 1981. de Roos, A.M.; Rekker, R.F.; Nauta, W.Th.: The base strength of substituted 2-(diphenylmethoxy)-N,N-dimethyl ethyl amines. Arzneim. Forsch. 20:1763-1765 (1970). Roozemond, R.C.; Vegt, G.J.B.; Hespe, W.; Nauta, W.Th.: In vitro 197 precipitation of liver microsomes by deptropine and some structurally related compounds: A cause for the discrepancies observed between in vivo and in vitro N-demethylation. Biochem. Pharmacol. 14:699-702 (1965). Rosenfeld, C.R.: Distribution of cardiac output in ovine pregnancy. Am. J. Physiol. 232:H231-H235 (1977). Ross, M.G.; Ervin, M.G.; Rappaport, V.J.; Youssef, A.; Leake, R.D.; Fisher, D.A.: Ovine fetal urine contribution to amniotic and allantoic compartments. Biol. Neonate 53:98-104 (1988). Roth, R.A.; Wiersma, D.A.: Role of the lung in total body clearance of circulating drugs. Clin. Pharmacol. 4:355-367 (1977). Roth, R.A.: Biochemistry, physiology and drug metabolism-implications regarding the role of the lungs in drug disposition. Clin. Physiol. Biochem. 3:66-79 (1985). Roth, T.; Roehrs, T.; Koshorek, G.; Sicklesteel, J.; Zorick, F.: Sedative effects of antihistamines. J. Allergy Clin. Immunol. 80:94-98 (1987). Ruckebusch, Y.; Guajoux, M.; Eghbali, B.: Placental transfer of central nervous system depressants in sheep. Eur. J. Pharmacol. 37: 193-193 (1976). Ruckebusch, Y.; Gaujoux, M.; Eghbali, B.: Sleep cycles and kinesis in the foetal lamb. Electroencephal. Clin. Neurophysiol. 42: 226-237 (1977). Rudolph, A.M.; Heymann, M.A.: Circulatory changes during growth in the fetal lamb. Circ. Res. 26:289-299 (1970). Rudolph, A.M.: Distribution and regulation of blood flow in the fetal and neonatal lamb. Circ. Res. 57:811-821 (1985). Ruess, M.L.; Rudolph, A.M.: Distribution and recirculation of umbilical and systemic venous blood flow in fetal lambs during hypoxia. J. Dev. Physiol. 2:71-84 (1980). Runciman, W.B.; Ilsley, A.H.; Mather, L.E.; Carapetis, R.; Rao, M.M.: A sheep preparation for studing interactions between blood flow and drug disposition. I: Physiological profile. Br. J. Anaesth. 56:1015-1027 (1984). Rurak, D.W.; Gruber, N.C.; Increased oxygen consumption associated with breathing activity in fetal lambs. J. Appl. Physiol. 54:701-707 (1983). Rurak, D.W.; Cooper, CC; Taylor, S.M.: Fetal oxygen consumption and Po2 during hypercapnia in pregnant sheep. J. Dev. Physiol. 8:447-459 (1986). Rurak, D.W.; Yoo, S.D.; Kwan, E.; Taylor, S.M.; Riggs, K.W.; Axelson, J.E.: Effects of diphenhydramine in the fetal lamb following maternal or fetal administration. J. Pharmacol. Exp. Ther. 247:271-278 (1988). Saxen, I.: Cleft palate and maternal diphenhydramine intake. Lancet 1:407 (1974). Schardein, J.L.; Hentz, D.L.; Petrere, J.A.; Kurtz, S.M.: Teratogenesi: studies with diphenhydramine HC1. Toxicol. Appl. Pharmacol. 18:971-976 (1971). Schomburg, G.; Behlau, H.; Dielmann, R.; Weeke, F.; Husmann, H.: Sampling techniques in capillary gas chromatography. J. Chromatogr. 142:87-102 (1977). Sedman, A.J.; Wagner, J.G.: "AUTOAN. A dicision-making pharmacokinetic computer program", Publication Distribution Service, Ann Arbor, MI, 1976. Sharp, M.E.; Wallace, S.M.; Hindmarch, K.W.; Peel, H.W.: Monitoring saliva concentrations of methaqualone, codeine, secobarbital, diphenhydramine and diazepam after single oral doses. J. Anal. Toxicol. 7:11-14 (1983). Skofitsch, G.; Lembeck, F.: Serum levels of dimenhydrinate. Arzneim. Forsch. 33:1674-1676 (1983). Spector, R.; Choudhury, A.K.; Chiang, C; Goldberg, M.J.; Ghoneim, M.M Diphenhydramine in orientals and Caucasians. Clin. Pharmacol. Ther. 28:229-234 (1980). Stacey, T.E.; Weedon, A.P.; Haworth, C; Ward, R.H.T.; Boyd, R.D.H.: Fetomaternal transfer of glucose analogues by sheep placenta. Am. J. Physiol. 234.-E32-E37 (1978). Szeto, H.H.; Mann, L.I.; Bhakthavathsalan, A.; Liu, M.; Inturrisi, CE Meperidine pharmacokinetics in the maternal-fetal unit. J. Pharmacol. Exp. Ther. 206:448-459 (1978). Szeto, H.H.; Clapp, J.F.; Larrow, R.W.; Hewitt, J.; Inturrisi, C.E.; Mann, L.I.: Disposition of methadone in the ovine maternal-fetal unit. Life Sci. 28:2111-2117 (1981). Szeto, H.H.; Umans, J.G.; Rubinow, S.I.: The contribution of transplacental clearances and fetal clearance to drug disposition in the ovine maternal-fetal unit. Drug Metab. Dispos. 10:382-386 (1982a). Szeto, H.H.; Umans, J.G.; McFarland J.: A comparison of morphine and methadone disposition in the maternal-fetal unit. Am. J. Obstet. Gynecol. 143:700-706 (1982b). Szeto, H.H.; Umans, J.G.; Umans, H.R.; McFarland, J.W.: The relationsh 199 between maternal and fetal plasma protein binding of methadone in the ewe during the third trimester. Life Sci. 30:1271-1279 (1982c). Szeto, H.H.: Effects of narcotic drugs on fetal behavioural activity: acute methadone exposure. Am. J. Obstet. Gynecol. 146:211-216 (1983). Teramo, K.; Benowitz, N.; Heymann, M.A.; Kahanpaa, K.; Siimes, A.; Rudolph, A.M.: Effects of lidocaine on heart rate, blood pressure, and electrocorticogram in fetal sheep. Am. J. Obstet. Gynecol. 118:935-949 (1974). Tomoda, S.; Brace, R.B.; Longo, L.D.: Amniotic fluid volume and fetal swallowing rate in sheep. Am. J. Physiol. 249:R133-R138 (1985). Tozzi, S.: The mechanism of action of histamine on the isolated rat uterus. J. Pharmacol. Exp. Ther. 187:511-517 (1973). Umans, J.G.; Szeto, H.H.: Precipitated opiate abstinence in utero. Am. J. Obstet. Gynecol. 151:441-444 (1985). Wagner, J.G.: A modern view of pharmacokinetics. J. Pharmacokin. Biopharm. 5:363-401 (1973). Walker, D.W.: Placental clearance of antipyrine and urea during hypoxia and adrenaline infusions in sheep. J. Physiol. 271-.65P-66P (1977). Wallace, J.E.; Briggs, J.D.;Dahl, E.V.: Determination of diphenhydramine and certain related compounds by ultraviolet spectrophotometry. Anal. Chem. 38:831-834 (1966). Wang, L.H.; Rudolph, A.M.; Benet, L.Z.: Pharmacokinetic studies of the disposition of acetaminophen in the sheep maternal-fetal unit. J. Pharmacol. Exp. Ther. 238:198-205 (1986). Winter, J.C: Sedation and the stimulus properties of antihistamines. Pharmacol. Biochem. Behav. 22:15-17 (1985). Wintour, E.M.; Barnes, A.; Brown, E.H.; Hardy, K.J.; Horacek, I.; McDougall, J.G.; Scoggins, B.A.: Regulation of amniotic fluid volume and composition in the ovine fetus. Obstet. Gynecol. 52:689-693 (1978). Wlodek, M.E.; Challis, J.R.G.; Patrick, J.: Urethral and urachal urine output to the amniotic and allantoic sacs in fetal sheep. J. Dev. Physiol. 10:309-319 (1988). Wyngaarden, J.B.; Seevers, M.H.: The toxic effects of antihistamine drugs. J.A.M.A. 145:277-282 (1951). 200 Yang, F.J.; Brown, A.C.III; Cram, S.P.: Splitless sampling for capillary-column gas chromatography. J. Chromatogr. 158:91-109 (1978).