Detection of cocaine exposure in the neonate. Analyses of urine, meconium, and amniotic fluid from mothers and infants exposed to cocaine

Cocaine and its metabolites were measured in urine, meconium, and amniotic fluid specimens collected from 30 maternal-infant pairs with histories of prenatal cocaine use. Cocaine, benzoylecgonine, and ecgonine methyl ester were measured by isotope dilution gas chromatography-mass spectrometry. Mothers were interviewed at delivery regarding their cocaine use during pregnancy. There was qualitative agreement between the results of drug determinations in maternal urine, amniotic fluid, infant urine, and meconium. Although all of the mothers in this study admitted to using cocaine during their pregnancy, cocaine or its metabolites were detected only in the 20 cases in which cocaine was used within 3 weeks before delivery. We conclude that when sufficiently sensitive analytic methods are used, maternal urine, infant urine, and meconium analyses yield equivalent results for detection of prenatal cocaine exposure. Importantly, neither meconium nor urinary drug measurements detected cocaine exposure when the last reported use was prior to 3 weeks before delivery.     

Cocaine and its metabolites constrict cerebral arterioles in newborn pigs

We examined the effect of cocaine and several of its metabolites on cerebral arterioles in newborn pigs and evaluated the sympathomimetic properties of each of the compounds as a vasoactive mechanism. After piglets were equipped with closed cranial windows, compounds were suffused over the brain surface and pial arteriolar diameter (base line, approximately 100 microns) was recorded. Cocaine, cocaethylene, norcocaine, ecogonine, benzoylecgonine and ecgonine methylester each caused a dose-dependent (10(-8) M to 10(-4) M) decrease in pial arteriolar diameter: maximum percent reductions in diameter induced by each compound (10(-4) M) were, respectively, 12 +/- 1, 12 +/- 2, 11 +/- 1, 7 +/- 1, 7 +/- 2 and 5 +/- 1. In analyzing the dose-response curves, cocaethylene was the most potent vasoconstrictor, followed by cocaine, norcocaine and then ecogonine, benzoylecgonine and ecgonine methylester. Cerebral vasoconstriction induced by topically applied norepinephrine was enhanced by cocaine, norcocaine and cocaethylene, but not by the other three metabolites. Topical application of phentolamine failed to block vasoconstriction elicited by cocaine or its metabolites, although it did block vasoconstriction elicited by norepinephrine. These observations indicate that cocaine and its metabolites constrict the immature cerebrovasculature by a non-sympathomimetic mechanism.     

Disposition and metabolism of [3H]cocaine in acutely and chronically treated dogs

1. Beagle dogs were chronically treated with cocaine, 5 mg/kg subcutaneously twice daily for 6 weeks, followed by same dose of [3H]cocaine given intravenously. 2. The t1/2 values of cocaine in plasma, liver, spleen and heart, in acutely and chronically treated dogs, were: 1-2, 1-1; 2-2, 1-8; 1-8, 1-3; 2-0, 1-2 h, respectively. In both groups, cocaine disappeared from all areas of the central nervous system 12-24 h after injection but significant amounts of radioactivity due to benzoylnorecgonine and benzoylecgonine persisted in the CNS even 1 week after administration of cocaine. Brain-to-plasma ratios of cocaine were lower in chronically-treated than in acutely-treated dogs 2 and 4 h after injection. 3. Norcocaine, benzoylnorecgonine, benzoylecgonine and ecgonine were metabolites of cocaine in dog brain in both groups. Norcocaine and benzoylnorecgonine were present in higher amounts in brains of chronically treated dogs. Rate of disappearance of norcocaine was similar to cocaine in both groups. 4. The amounts of cocaine excreted in urine and faeces as percentage of dose were 0-9-5-0, 1-1-6 in the acute and 2-2-3-3 and 0-2-0-3 in the chronically treated dogs. Major excretion of radiactivity occurred in urine within 24 h in both groups. Total radioactivity (65% of dose) in urine plus faeces was similar in both groups. 5. Norcocaine, benzoylnorecgonine, benzoylecgonine, ecgonine, norecgonine, ecgonine methyl ester and unidentified compounds were urinary metabolites of cocaine in both groups. Benzoylnorecgonine and ecgonine were excreted in higher amounts and benzoylecgonine and norecgonine in lower amounts in the acute than in the chronically treated dogs. 6. The possible role of persistence of benzoylnorecgonine and benzoylecgonine (which possessed potent stimulant activity intracisternally) in the CNS is discussed.     

Reflection spectra of dense amorphous SiO2 in the vacuum-uv region

The interaction of cocaine, its metabolites norcocaine and benzoylecgonine, and cocaethylene, which is formed following a combined cocaine and ethanol exposure, with muscarinic receptor binding and phosphoinositide metabolism was investigated in brain from immature rats. Cocaine and norcocaine inhibited binding of [3H]telenzepine and carbachol-stimulated phosphoinositide metabolism in cerebral cortex, while benzoylecgonine was devoid of any inhibitory activity. Cocaethylene was the most potent inhibitor of both binding and phosphoinositide metabolism. The effect of cocaine was more pronounced at the muscarinic receptors, but a small inhibition of histamine--and serotonin--stimulated phosphoinositide metabolism was also observed.     

A new placental enzyme in the metabolism of cocaine: an in vitro animal model

OBJECTIVE: The aim of this study was to analyze placental metabolism in a genetically controlled in vitro animal model. STUDY DESIGN: Placentas from Sprague-Dawley rats were centrifuged, and microsomes were isolated. Four treatment groups were incubated with cocaine over four time periods: placental microsomes + cocaine, placental microsomes + diisopropyl fluorophosphate (an anticholinesterase) + cocaine, placental microsomes + cocaine + butyrylcholinesterase, and a blank (cocaine only). Gas chromatography was used to quantify cocaine (Limit of quantitation = 19 ng/ml) and metabolites. Gas chromatography/mass spectrometry was used to verify the identity of the metabolites. RESULTS: Butyrylcholinesterase enhanced cocaine metabolism to ecgonine methyl ester. More than 40% of cocaine was metabolized to norcocaine by rat placental when diisopropyl fluorophosphate suppressed cocaine. Norcocaine is produced by hepatic N-demethylase action on methyl-bearing nitrogen in cocaine, suggesting that placenta and liver have this capacity. Gas chromatography/mass spectrometry was essential to the identification of norcocaine, because norcocaine is frequently not identified. CONCLUSIONS: This biotransformation of cocaine to norcocaine may be a primary metabolic pathway induced in the cholinesterase-deficient placenta. This has clinical implications because norcocaine is ninefold more active physiologically than cocaine or ecgononine methylesterase.     

Cocaine and metabolites in the hair of ancient Peruvian coca leaf chewers

Cocaine and its metabolites, benzoylecgonine (BZE) and ecgonine methylester (EME), were found in hair samples from ancient Peruvian coca-leaf chewers dating back to AD 1000. Hair was analyzed by gas chromatography/mass spectrometry (GC/MS) to quantitate the concentrations. The two metabolites were found in higher concentration than the parent drug. The metabolite levels appear to be below that of modern cocaine abusers. Gender does not appear to be a factor in the incorporation of drug into hair.     

Attenuation of cocaine-induced locomotor activity by butyrylcholinesterase.

A primary enzyme for the metabolism of cocaine is butyrylcholinesterase (BChE). To determine whether the systemic administration of BChE can increase the metabolism of cocaine sufficiently to alter a behavioral effect, rats were tested in a locomotor activity chamber after receiving 17 mg of cocaine per kg intraperitoneally. In rats pretreated intravenously with 5,000 IU of horse serum-derived BChE, the locomotor activity effect was significantly attenuated. BChE pretreatment increased plasma BChE levels approximately 400-fold. When added to rat plasma, this amount of BChE reduced the cocaine half-life from over 5 hr to less than 5 min. BChE altered the cocaine metabolic pattern such that the relatively nontoxic metabolite ecgonine methyl ester was produced, rather than benzoylecgonine. These results suggest that systemic administration of BChE can increase the metabolism of cocaine sufficiently to alter a behavioral effect of cocaine and thus should be investigated as a potential treatment for cocaine abuse. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Evidence that loss of chromosome 18q is associated with tumor progression

Toxicological investigation of suspected cocaine-related deaths routinely involves the identification of cocaine (COC) and its metabolites including benzoylecgonine (BE) and ecgonine methyl ester (EME) in postmortem specimens. We utilized solid-phase extraction followed by gas chromatography/mass spectrometry for the qualitative and quantitative analysis of cocaine and eight cocaine-related analytes. These analytes included anhydroecgonine methyl ester (AEME), a unique product formed during cocaine smoking, and cocaethylene (CE), formed by transesterification of cocaine in the presence of ethanol. Thirteen pairs of postmortem heart blood and urine specimens were analyzed from cases of death due to acute cocaine intoxication, multiple drug intoxication, or other non-drug related causes. COC, EME, and BE were detected in all specimens. The range of concentrations in blood were: COC, 23-2088 ng/mL; BE, 215-9195 ng/mL; and EME, 220-7275 ng/ mL. AEME was identified in 2 blood and 10 urine specimens, and CE was identified in 1 blood specimen and 4 urine specimens. The identification of AEME in the specimens indicated that "crack" cocaine had been smoked, and the presence of CE indicated co-administration of cocaine and ethanol. The presence of these unique cocaine analysis in postmortem specimens provides valuable information regarding the cause and manner of death.     

Ovine fetal-placental cocaine pharmacokinetics during continuous cocaine infusion

OBJECTIVE: To investigate fetal-placental cocaine clearance, and to determine the fetal catecholamine and cardiovascular responses to continuous intravenous cocaine infusion in fetal sheep. METHODS: Eleven pregnant ewes and their fetuses (127 +/- 2 days' gestation; term 150 days) were chronically instrumented. Fetuses received intravenous cocaine at 0.05, 0.1, or 0.2 mg/kg/minute. Fetal cardiovascular and hematologic measurements were made before and serially for 90 minutes after initiation of the cocaine infusion. RESULTS: Steady-state fetal plasma cocaine concentrations were observed by 15 minutes of infusion and averaged 136 +/- 11, 318 +/- 65, and 610 +/- 36 ng/mL, respectively, at each dose. Fetal-placental cocaine clearance rate was independent of dose (337 +/- 39 mL/kg/minute), indicating that it is a first-order pharmacokinetic process. Fetal plasma concentration of benzoylecgonine, a principle cocaine metabolite, increased throughout the study to approximately 25% above cocaine levels by 90 minutes. There were significant increases in fetal heart rate (from 169 +/- 11 to 242 +/- 36 beats per minute), mean blood pressure (from 53 +/- 4 to 63 +/- 5 mmHg), and systolic blood pressure (from 68 +/- 2 to 80 +/- 5 mmHg), with a corresponding increase in catecholamine levels seen in the fetuses infused with 0.2 mg/kg/minute. These changes were not seen in the fetuses given lower doses of cocaine. CONCLUSION: Fetal-placental clearance of cocaine is a rapid, first-order pharmacokinetic process. During prolonged cocaine exposure, plasma benzoylecgonine concentrations accumulate significantly. Significant catecholamine and cardiovascular changes are seen in fetal sheep with a continuous infusion of cocaine at 0.2 mg/kg/minute or greater.     

Cocaine and benzoylecgonine in serum microsamples of intact and gonadectomized male and female Wistar rats

Tail-tip plasma samples of intact and gonadectomized male and female Wistar rats were analyzed for cocaine and benzoylecgonine. The samples were obtained from immobilized subjects 10 and 30 min following the 1st and 22nd intraperitoneal injections of 10 mg/kg cocaine hydrochloride. Gender differences in plasma cocaine or benzoylecgonine levels were not observed after the first injection of cocaine because many of the samples were below the detection limit. Cocaine plasma levels were much higher after the 22nd injection, but gender differences were not observed either 10 or 30 min following cocaine administration. Plasma levels of benzoylecgonine were higher 30 min than 10 min after cocaine administration in intact and castrated male rats and ovariectomized female rats but not in intact female rats. These data show that, in rats, gender differences in cocaine metabolism may be observed after repeated cocaine administration, but the exact mechanism remains to be elucidated.     

Stereoselective inhibition of human butyrylcholinesterase by phosphonothiolate analogs of (+)- and (-)-cocaine

The hydrolysis of cocaine (benzoylecgonine methyl ester) to ecgonine methyl ester by human butyrylcholinesterase (BuChE; EC 3.1.1.8) has been shown previously to constitute an important means to detoxicate this material to pharmacologically inactive metabolites. The naturally occurring (-)-cocaine is hydrolyzed to ecgonine methyl ester approximately 2000 times slower than the unnatural (+)-cocaine isomer. In good agreement with previous studies, (-)-cocaine bound to human BuChE with relatively good affinity and competitively inhibited the hydrolysis of the spectrophotometric substrate butyrylthiocholine with a Ki value of 8.0 microM. Similarly, (+)-cocaine also showed relatively high affinity for the human BuChE and competitively inhibited butyrylthiocholine hydrolysis with a Ki value of 5.4 microM. The phosphonothiolates corresponding to the transition state analogs for both (-)- and (+)-cocaine hydrolysis were synthesized and tested as inhibitors of human BuChE-catalyzed hydrolysis of butyrylthiocholine. The phosphonothiolate corresponding to the transition state for (-)-cocaine hydrolysis was a competitive inhibitor with a Ki value of 55.8 microM. The phosphonothiolate corresponding to the transition state for (+)-cocaine hydrolysis gave a Ki value of 25.9 microM, but, in addition, it also showed irreversible inhibition with a ki of inactivation of 68.8 min-1 M-1. It is likely that the mechanism-based inhibitor described herein may find use as a mechanistic probe of butyrylcholinesterase action and also possibly aid in the purification of this class of esterases.     

Determination of common drugs of abuse in body fluids using one isolation procedure and liquid chromatography--atmospheric-pressure chemical-ionization mass spectromery

A method for determining opiate agonists (morphine, morphine-3-glucuronide, morphine-6-glucuronide, 6-monoacetylmorphine, codeine, codeine-6-glucuronide, dihydrocodeine, dihydromorphine, buprenorphine, methadone, tramadol, and ibogaine), cocaine and its metabolites (benzoylecgonine and ecgonine methyl ester) and lysergic acid diethylamide in serum, blood, urine and other biological matrices is presented. Aliquots (0.5-1.5 mL) of biological fluids were spiked with appropriate deuterated internal standards and extracted using a common solid-phase extraction method (C18 cartridges). The extracts were subjected to liquid chromatographic-atmospheric-pressure chemical-ionization mass spectrometric examination using selected ion monitoring procedures. These procedures were developed after analysis of full-scan mass spectra of examined compounds. The extraction method appeared very universal; the recoveries were high for almost all drugs and the extracts were very clean. The procedure was applied for routine forensic casework.     

Efficacy of high dose of recombinant alpha 2b interferon on long term response in chronic hepatitis C and cirrhosis: prospective randomized multicentre study

There is large variability in the rate and extent of fetal damage from cocaine in humans; however, the sources of such variability are not presently known. In order to study the relationship between maternal cocaine pharmacokinetics at the end of pregnancy and maternal or infant cocaine and benzoylecgonine hair concentrations at birth, ten rhesus monkeys were administered cocaine intramuscularly throughout pregnancy. Cocaine and benzoylecgonine hair concentrations were determined at birth and correlated with maternal pharmacokinetics during pregnancy. There were no correlations between either maternal cocaine Cmax or AUC0-infinity and maternal and infant hair cocaine or benzoylecgonine concentrations. There were no significant correlations between maternal hair benzoylecgonine concentrations and either maternal benzoylecgonine AUC0-120 (r = 0.60; P = 0.07) or benzoylecgonine Cmax (r = 0.60; P = 0.07). No correlations existed between infant hair benzoylecgonine concentrations and either maternal benzoylecgonine AUC0-120 (r = 0.30; P = 0.40) or benzoylecgonine Cmax (r = 0.30; P = 0.40). Also, no correlation was found between maternal cocaine dose and maternal or infant cocaine and benzoylecgonine hair concentrations. In comparison to toxicants such as nicotine and carbon monoxide for which there is a good correlation between maternal systemic exposure and neonatal concentrations, the lack of a similar relationship for cocaine is consistent with the role of the placenta in contributing to the variability in the amounts of cocaine reaching the fetus and hence, potentially to the risk of adverse fetal outcome.     

Determination of drugs of abuse in blood

The detection and quantitation of drugs of abuse in blood is of growing interest in forensic and clinical toxicology. With the development of highly sensitive chromatographic methods, such as high-performance liquid chromatography (HPLC) with sensitive detectors and gas chromatography-mass spectrometry (GC-MS), more and more substances can be determined in blood. This review includes methods for the determination of the most commonly occurring illicit drugs and their metabolites, which are important for the assessment of drug abuse: Methamphetamine, amphetamine, 3,4-methylenedioxymethamphetamine (MDMA), N-ethyl-3,4-methylenedioxyamphetamine (MDEA), 3,4-methylenedioxy-amphetamine (MDA), cannabinoids (delta-9-tetrahydrocannabinol, 11-hydroxy-delta-9-tetrahydrocannabinol, 11-nor-9-carboxy-delta-9-tetrahydrocannabinol), cocaine, benzoylecgonine, ecgonine methyl ester, cocaethylene and the opiates (heroin, 6-monoacetylmorphine, morphine, codeine and dihydrocodeine). A number of drugs/drug metabolites that are structurally close to these substances are included in the tables. Basic information about the biosample assayed, work-up, GC column or LC column and mobile phase, detection mode, reference data and validation data of each procedure is summarized in the tables. Examples of typical applications are presented.     

Maternal plasma hypo-osmolality: effects on spontaneous and stimulated ovine fetal swallowing

Fetal swallowing is a major route of amniotic fluid resorption, and thus swallowing activity may alter amniotic fluid volume. Near-term ovine fetal swallowing increases in response to plasma and/or cerebrospinal fluid hypertonicity. As maternal hydration status alters amniotic fluid volume, we hypothesized that maternal plasma hypotonicity may alter fetal swallowing activity. Pregnant ewes (130 +/- 1 d; n = 6) were chronically prepared with maternal and fetal vascular catheters, a fetal esophageal flow probe, and fetal thyrohyoid and nuchal and thoracic esophagus electromyogram electrodes. Spontaneous fetal swallowing and hypertonic saline thresholds for stimulated swallowing were determined prior to and following maternal hypotonicity induced with water loading and intravenous DDAVP (arginine vasopressin V2 agonist). Fetal swallowing thresholds were determined with intracarotid injections (0.15 ml/kg) of increasing sodium chloride concentrations (0.15-1.2 M) at 2-min intervals. Maternal DDAVP infusion significantly decreased mean (+/-SEM) maternal and fetal plasma osmolalities (298 +/- 2-284 +/- 3; 295 +/- 2-278 +/- 3 mOsm/kg, respectively) and sodium concentrations (147.3 +/- 0.4-137.5 +/- 0.9; 142.7 +/- 0.8-133.5 +/- 1.0 mEq/l, respectively), suppressed spontaneous swallowing activity and volume (1.1 +/- 0.2-0.6 +/- 0.1 swallows/min; 0.7 +/- 0.2-0.5 +/- 0.1 ml/min, respectively) and significantly increased the osmotic threshold for swallowing stimulation (0.77 +/- 0.08-1.03 +/- 0.09 M NaCl). We conclude that: (1) maternal, and thus fetal, plasma hypotonicity results in suppression of spontaneous fetal swallowing activity and a decrease in volume swallowed, suggesting that spontaneous fetal ingestive behavior results, in part, from tonic dipsogenic stimulation, and (2) under hypotonic conditions, the intracarotid NaCl injection concentration for swallowing stimulation increases. These results suggest that the reset (lower) maternal plasma osmolality during human pregnancy may serve to minimize fetal ingestive and perhaps arginine vasopressin-mediated antidiuretic responses to acute maternal hypertonicity.     

Cocaine benzoyl thioester: synthesis, kinetics of base hydrolysis, and application to the assay of cocaine esterases

The synthesis and characterization of diastereomers of cocaine benzoyl thioester is described. Allococaine benzoyl thioester and allopseudococaine benzoyl thioester were synthesized by the conjugate addition of p-methoxytolyl thiol to ecgonine methyl ester followed by debenzylation and benzoylation. The absolute structure of the hydrochloride salt of the major ecgonine p-methoxytolyl sulfide formed was determined by single-crystal diffraction analysis and used to establish the addition geometry. When placed in aqueous solution, the cocaine benzoyl thioester diastereomers hydrolyzed to give thioecgonine methyl ester. The rate of cocaine benzoyl thioester hydrolysis was carefully investigated spectrophotometrically by using the Ellman reagent. At neutral pH, the hydrolysis of the diastereomers was found to proceed at detectable rates. Upon increasing pH, the rate of hydrolysis of cocaine benzoyl thioester diastereomers was increased and the reaction was catalyzed by basic buffer species. In addition to defining the kinetics of hydrolysis in aqueous solution, cocaine benzoyl thioester was utilized as a highly efficient method to monitor the activity of cholinesterases and pig liver esterase. Use of cocaine benzoyl thioester represents a rapid and sensitive way to screen for cocaine esterase activity.     

Central administration of cocaine produces age-dependent effects on behavior in the fetal rat.

Rat fetuses were exposed to cocaine, lidocaine, or saline on Gestational Day 20 or 21 to provide information about cocaine effects on behavior during prenatal development. Cocaine was administered into the cisterna magna of individual fetal subjects to restrict effects to the CNS. Behavioral effects of cocaine were compared with lidocaine to help distinguish the effects of cocaine on monoamine systems in the brain from its properties as a local anesthetic. Cocaine promoted 3–5 fold increases in fetal motor activity in the absence of explicit sensory stimulation, in contrast to the slight suppressive effects of lidocaine. Cocaine and lidocaine also reduced coordinated behavioral responses to an artificial nipple. The behavioral effects of cocaine administered into the CNS of fetal subjects suggest specific mechanisms of action on developing neural and behavioral systems in the late prenatal period. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Determination and in-depth chromatographic analyses of alkaloids in South American and greenhouse-cultivated coca leaves

Methodology is described for the detection and/or determination of cocaine and minor alkaloids in South American coca as well as in greenhouse- and tropical-cultivated field coca of known taxonomy. Coca leaf from Bolivia, Peru, Ecuador and Colombia were subjected to the determination of cocaine, cis- and trans-cinnamoylcocaine, tropacocaine, hygrine, cuscohygrine and the isomeric truxillines. The greenhouse samples were cocaine-bearing leaves of the genus Erythroxylum and included E. coca var. coca, E. novogranatense var. novogranatense and E. novogranatense var. truxillense, and the alkaloids determined were cocaine, ecgonine methyl ester, cuscohygrine, tropacocaine and the cinnamoylcocaines. The tropical-cultivated coca were E. novogranatense var. novogranatense and E. coca var. coca. Cocaine and minor alkaloids were isolated from basified powdered leaf samples using a toluene extractant, followed by acid-Celite column chromatography. The isolated alkaloids were determined by capillary gas chromatography with flame ionization or electron-capture detection. Methodology is also presented for the isolation and mass spectral analysis of numerous trace-level coca alkaloids of unknown structure.     

Synthesis and ligand binding studies of 4'-iodobenzoyl esters of tropanes and piperidines at the dopamine transporter

Four analogs and two homologs of cocaine, designed as potent cocaine antagonists, were synthesized. The SN2 reaction between ecgonine methyl ester (13) or appropriately substituted piperidinol (19, 21) and appropriately substituted 4-iodobenzoyl chloride gave 4-iodobenzoyl esters of tropanes and piperidines (5-8). 2'-Hydroxycocaine (9) was obtained from 2'-acetoxycocaine (12) by selective transesterification with MeOH saturated with dry HCl gas. 2'-Acetoxycocaine (12) was synthesized from acetylsalicyloyl chloride (23) and ecgonine methyl ester (13). The binding affinities of these compounds were determined at the dopamine transporter for the displacement of [3H]WIN-35428. An iodo group substitution at the 4'-position of cocaine decreased dopamine transporter binding potency, while a hydroxy or acetoxy group at the 2'-position exhibited increased binding potency for the dopamine transporter compared to cocaine (10- and 3.58-fold, respectively). 2'-Hydroxylation also enhanced the bidning potency of 4'-iodococaine (5) by 10-fold. Replacement of the tropane ring with piperidine led to poor binding affinities.     

Cell age distribution of erythrocytes at the incidence of cerebral stroke in stroke-prone spontaneously hypertensive rats, and their glutathione peroxidase activity

OBJECTIVE: A recently described mathematical model of human amniotic fluid dynamics used known and estimated rates of fetal fluid production (lung liquid and urine) and composition (osmolality) to enable calculation of previously unmeasured routes of amniotic fluid resorption, including fetal swallowing and intramembranous (across the amnion) water flow. This "osmolar" model assumed that only free water resorption occurred across the intramembranous route. We hypothesized that intramembranous flow also may include solutes and electrolytes because significant concentration gradients exist between amniotic fluid and fetal plasma. We used mass balance analysis to determine the direction and magnitude of intramembranous sodium flux and to assess the ability of a newly described "sodium" model to predict changes in amniotic fluid volume in response to changes in intramembranous electrolyte flow. Mathematical modeling was used to predict changes in amniotic fluid volume in response to changes in intramembranous electrolyte flow. STUDY DESIGN: Model predictions were calculated using published values for human amniotic fluid and fetal urine composition and volume. Ovine studies were used to derive lung fluid volumes and composition. Fetal swallowing and intramembranous flow were independently determined using net amniotic fluid osmolar (osmolality model) and sodium (sodium model) balance. Differences between osmolality and sodium model predictions were normalized to calculate the net intramembranous sodium flux, assuming a net balance of intramembranous osmotic solute flow. RESULTS: Both sodium and osmolality models predicted swallowed volume to be greater than intramembranous flow until 28 to 32 weeks' gestation, after which the relationship reversed. However, the sodium model predicted greater intramembranous flow and lower swallowing rates compared with the osmolality model at all gestational ages. Osmolar mass balance required daily intramembranous sodium flux into the amniotic fluid, which increased with gestational age. Furthermore, assuming stable swallowing and intramembranous water flow, the model predicts that 5% increases or decreases in amniotic fluid solute concentrations caused by intramembranous flux result in polyhydramnios or oligohydramnios, respectively. CONCLUSION: Sodium and osmolality models demonstrate similarities in determinations of amniotic fluid dynamics. However, mass balance equations demonstrate a net intramembranous flow of sodium into the amniotic fluid under normal conditions. Mathematical modeling suggests that small alterations in daily intramembranous sodium flux may evoke large changes in amniotic fluid volume.