Sudden infant death syndrome (SIDS) and the toxic gas hypothesis: microbiological studies of cot mattresses.

1. Fifty infants' mattresses were studied to investigate the occurrence of viable fungal and bacterial propagules, with particular reference to Scopulariopsis brevicaulis which had been suggested to be implicated in SIDS cases. A total of 19 SIDS cases mattresses, 1 non-SIDS death, 20 used controls, and 10 new unused controls were examined. 2. Differences were found between SIDS and used controls in the variety of fungal species isolated and the numbers isolated from fillings; bacterial numbers were similar. 3. S. brevicaulis was isolated from only four mattresses, three of which were SIDS cases. It was not found in most of those on which death had occurred. 4. A number of potentially pathogenic or allergenic fungi, including Aspergillus fumigatus, were isolated more frequently from SIDS cases mattresses than new or used controls. 5. Scanning electron microscopy of mattress covers and fillings showed microbial 'biofilms' in the head areas of all SIDS cases examined. This was not seen on other samples. 6. The limited number of mattresses studied and the use of unmatched controls precludes the drawing of any general conclusions as to the significance of the biofilms or other fungi isolated. 7. Reports of the existence of a dimorphism in general growth forms of S. brevicaulis were investigated by growing and transferring authentic strains between a variety of growth media. 8. No 'slimy' state of this fungus was observed and dimorphism was not confirmed.     

Dimethylarsine and trimethylarsine are potent genotoxins in vitro

The mechanism of arsenic carcinogenesis is unclear. A complicating factor receiving increasing attention is that arsenic is biomethylated to form various metabolites. Eleven different arsenicals were studied for in vitro genotoxicity to supercoiled DNA (pBR 322 and phiX174). Five arsenicals showed various degrees of positivity-monomethylarsonous acid, dimethylarsinous acid, monomethylarsine, dimethylarsine, and trimethylarsine. Supercoiled DNA, blotted on nitrocellulose filter paper, was exposed to gaseous arsines by suspending the filter paper above aqueous reaction mixtures of sodium borohydride and an appropriate arsenical. All three methylated arsines damaged DNA; inorganic arsine did not. Arsines were generated in situ in reaction mixtures containing DNA by reaction of sodium borohydride with arsenite, monomethylarsonous acid, dimethylarsinous acid, and trimethylarsine oxide, at pH 8.0. Both dimethylarsine and trimethylarsine (generated from 200 micro M dimethylarsinous acid and trimethylarsine oxide, respectively) damaged DNA in less than 30 min. Under certain conditions, the two most potent genotoxic arsines, trimethylarsine and dimethylarsine, are about 100 times more potent than dimethylarsinous acid (the most potent genotoxic arsenical previously known). There was no evidence to suggest that anything other than the arsines caused the DNA damage. Possible models for the biological production of arsines were examined. The coenzymes, NADH and NADPH, are biological hydride donors. When NADH or NADPH (5 mM) were incubated with dimethylarsinous acid (0-2 mM) for 2 h, DNA damage was increased by at least 10-fold. A possible explanation for this result is that these compounds react with dimethylarsinous acid to generate dimethylarsine. DNA was incubated with a dithiol compound, dithioerythritol (5 mM), and trimethylarsine oxide (0.5 mM) for 2 h, and the reduction of trimethylarsine oxide to trimethylarsine resulted in DNA damage.     

Cellular metabolism of arsenocholine

The biotransformation of arsenocholine and arsenobetaine, which are organic arsenic compounds present in certain aquatic organisms, has been studied in vitro using synthetic reference substances. Incubation of arsenocholine with different liver cell fractions showed arsenocholine to be biotransformed only in presence of the mitochondrial fraction. The biotransformation products were arsenobetaine aldehyde, arsenobetaine, trimethylarsine oxide and trimethylarsine. Arsenobetaine was the major metabolite and it was formed via arsenobetaine aldehyde. Trimethylarsine oxide was formed via a side reaction from arsenobetaine aldehyde. Further reduction of trimethylarsine oxide, produced trimethylarsine. In vitro studies of arsenobetaine, did not show any formation of trimethylarsine oxide or trimethylarsine. Furthermore, cytotoxicity of arsenobetaine or arsenocholine in isolated hepatocytes was not observed.     

Glutathione modulates recombinant rat arsenic (+3 oxidation state) methyltransferase-catalyzed formation of trimethylarsine oxide and trimethylarsine

Humans and other species enzymatically convert inorganic arsenic (iAs) into methylated metabolites. Although the major metabolites are mono- and dimethylated arsenicals, trimethylated arsenicals have been detected in urine following exposure to iAs. The AS3MT gene encodes an arsenic (+3 oxidation state) methyltransferase, which catalyzes both the oxidative methylation of trivalent arsenicals and the reduction of pentavalent arsenicals. In reaction mixtures containing recombinant rat AS3MT (rrAS3MT) and radiolabeled arsenite, mono- and dimethylated arsenicals and a third radiolabeled product can be resolved by thin-layer chromatography. Hydride generation atomic absorption spectrometry and electrospray ionization mass spectrometry identified the third reaction product as trimethylarsine oxide. The addition of glutathione to reaction mixtures containing radiolabeled arsenite and rrAS3MT increased the yield of methylated and dimethylated arsenicals but suppressed the formation of trimethylarsine oxide. Although a dimethylarsenic-glutathione complex was rapidly converted to trimethylarsine oxide, the addition of a molar excess of glutathione to dimethylarsenic suppressed the production of trimethylarsine oxide. The nonquantitative recovery of radioarsenic from reaction mixtures suggested that AS3MT catalyzed the formation of a volatile arsenical. This volatile species was identified as trimethylarsine. Thus, AS3MT catalyzes the formation of all products in a reaction sequence leading from an inorganic to a volatile methylated arsenical. The regulation of this pathway by intracellular glutathione may be an important determinant of the pattern and extent of formation of arsenicals.     

Evaluation of cot mattress inner foam as a potential site for microbial generation of toxic gases

Recent reports of biovolatilisation of phosphorus and antimony by anaerobic bacteria and of leaching of phosphorus and antimony fire-retardant additives from PVC cot mattress covers, indicate that the polyurethane inner-foam of cot mattresses could be a site for generation of toxic gases of group 15 elements. A toxic gas hypothesis for sudden infant death syndrome (SIDS) involving polyurethane foam of cot mattresses was proposed and tested experimentally. Levels of antimony, phosphorus, arsenic and bismuth were determined at four sites for 44 SIDS and 50 control (no death) cot mattress foams. There was no evidence to suggest that the levels of these elements in cot mattress foam have a causal relation to SIDS. Leaching of antimony trioxide from PVC mattress covers could account for detectable levels of this element in 52% of the cot mattress samples analysed. Volatile forms of antimony, phosphorus, arsenic and bismuth was not detected in the headspace of mixed or monoseptic cultures of anaerobic bacteria containing polyurethane foam. Past microbial activity had given rise to involatile methylated species of antimony in some of the cot mattress foams tested (61%, n = 24). Abiotic oxidation of biogenic trimethylantimony together with physical adsorption of methylantimony forms to the polyurethane foam matrix could account for the apparent absence of "escaped" volatile antimony species in culture headspaces of incubation vial. There was no evidence to suggest that levels of trimethylantimony or total methylantimony forms in cot mattress foams have a causal relation to SIDS.     

Epidemiological factors in postneonatal mortality in New Zealand

Nine hundred and sixty-two postneonatal deaths for 1981-83 were matched to their birth registration forms. Deaths were divided into three categories, sudden infant death syndrome (SIDS) 65.4%, other preventable, 12.8%, and nonpreventable causes, 21.8%, to determine the rates of death as related to known and available risk factors. The risk factor profile for other preventable causes and SIDS was similar, the only exceptions being that other preventable causes showed no north-south gradient and had a higher incidence in the neonatal period (31.2% v 4.7% for SIDS). The most important risk factors for other preventable causes were found to be the mother being Maori (RR 4.35, CI 3.12-6.06), having a low birth weight infant (RR 3.56, CI 2.07-6.13) and being unmarried (RR 3.45, CI 2.47-4.82). These risk factors point to the possibility of selectively targeting of interventions both prenatally as well as postnatally for those who are at high risk.     

Microbial metabolite of dimethylarsinic acid is highly toxic and genotoxic

Dimethylarsinic acid [DMA, (CH(3))(2)AsO(OH)] causes cancer in the urinary bladder of rats. However, its mechanism of cancer or the ultimate carcinogenic form is not yet known. Rats administered dimethylarsinic acid excrete three unknown arsenic compounds (termed M-1, M-2, and M-3) in urine or feces, and these compounds are presumed to be produced by intestinal bacteria. Escherichia coli A3-6 isolated from a rat yielded two unknown arsenic compounds (M-2 and M-3) from dimethylarsinic acid and M-1 from trimethylarsine oxide (TMAO) in the presence of cysteine (Cys). Contents of M-2 and M-3 varied with cysteine concentration. The cytotoxicity and genotoxicity of the bacteria-free solution of dimethylarsinic acid or trimethylarsine oxide metabolized by E. coli A3-6 were studied using V79 cells. Dimethylarsinic acid (1 mM) metabolized by E. coli A3-6 in the presence of cysteine (1 mM) was highly cytotoxic (50% survival reduction concentration; 2.1 microM As) in V79 cells, and the toxic substance appeared to be M-2. The metabolite solution (at 2.5-10 microM total As) induced c-mitosis and tetraploids, and caused mitotic arrest, since it increased mitotic cells at the cytotoxic dose. The metabolite solution also significantly increased sister chromatid exchange (SCE) and chromosomal aberrations, most of which were chromatid gaps and chromatid breaks. A3-6 converted 96.1% of trimethylarsine oxide to M-1 in the presence of cysteine. This metabolite solution did not exhibit cytotoxicity or genotoxicity. The reported M-2 concentration in urine of rats administered levels of DMA via drinking water known to cause bladder tumors was sufficient to exhibit cytotoxic and genotoxic effects in urinary bladder. Thus, we hypothesize that intestinal bacteria play an important role in carcinogenicity of dimethylarsinic acid.     

Tissue dosimetry, metabolism and excretion of pentavalent and trivalent dimethylated arsenic in mice after oral administration

Dimethylarsinic acid (DMA(V)) is a rat bladder carcinogen and the major urinary metabolite of administered inorganic arsenic in most mammals. This study examined the disposition of pentavalent and trivalent dimethylated arsenic in mice after acute oral administration. Adult female mice were administered [(14)C]-DMA(V) (0.6 or 60 mg As/kg) and sacrificed serially over 24 h. Tissues and excreta were collected for analysis of radioactivity. Other mice were administered unlabeled DMA(V) (0.6 or 60 mg As/kg) or dimethylarsinous acid (DMA(III)) (0.6 mg As/kg) and sacrificed at 2 or 24 h. Tissues (2 h) and urine (24 h) were collected and analyzed for arsenicals. Absorption, distribution and excretion of [(14)C]-DMA(V) were rapid, as radioactivity was detected in tissues and urine at 0.25 h. For low dose DMA(V) mice, there was a greater fractional absorption of DMA(V) and significantly greater tissue concentrations of radioactivity at several time points. Radioactivity distributed greatest to the liver (1-2% of dose) and declined to less than 0.05% in all tissues examined at 24 h. Urinary excretion of radioactivity was significantly greater in the 0.6 mg As/kg DMA(V) group. Conversely, fecal excretion of radioactivity was significantly greater in the high dose group. Urinary metabolites of DMA(V) included DMA(III), trimethylarsine oxide (TMAO), dimethylthioarsinic acid and trimethylarsine sulfide. Urinary metabolites of DMA(III) included TMAO, dimethylthioarsinic acid and trimethylarsine sulfide. DMA(V) was also excreted by DMA(III)-treated mice, showing its sensitivity to oxidation. TMAO was detected in tissues of the high dose DMA(V) group. The low acute toxicity of DMA(V) in the mouse appears to be due in part to its minimal retention and rapid elimination.     

Sudden infant death syndrome among the Auckland Pacific communities 1988-1996: is it increasing?

AIMS: To define ethnic origin and verify the diagnosis of sudden infant deaths among Pacific peoples in Auckland 1988-1996, and to elicit soci-econonic and demographic characteristics. METHODS: Police (P47) and coroner reports were analysed for an ethnic classification and diagnosis. Postneonatal and sudden infant death syndrome (SIDS) register and New Zealand Information Services data were analysed for additional Pacific cases. Rates of Pacific SIDS in Auckland calculated. A Pacific SIDS database was developed and families were tracked. Face to face interviews covering the SIDS event were undertaken with selected families. Data were coded, stratified and a thematic approach to analysis was utilised. RESULTS: There were 52 cases of SIDS and the ethnic origins were: thirteen Samoans, nine Cook Islanders, seven Togans fifteen multiple ethnicity, and eight could not be verified. The annual rates of Pacific SIDS varied from less than one (in 1989) to 4.5 (in 1995) per 1000 Pacific live births. 34 cases (65%) couldd not be contacted and eighteen were traced. Nine in-epth interviews were conducted with caregivers of these cases. All babies had slept in the supine position, seven were breasted, and five of the mothers were non-smokers. Eight babies sept in the same room with their primary caregiver, with seven sleeping in their own bed. All of the mothers had had continuous access to childcare and support from their families, and seven had had previous children. Grief counselling for partners and children was identified as necessary by almost all the mothers. CNCLUSIONS: This preliminary study concludes that the rate of Pacific SIDS increased in 1995 and remains a serious problem. Ethnic misclassification and under reporting of SIDS cases is apparent among Pacific infants. There is a need to establish a national infant mortality database that collects accurate data incorporating standardised ethnic specific categories. Official routine and data sources also need to incorporate standardised ethnic specific categories. A national prospective study is required to study SIDS in Pacific communities as a basis for effective prevention strategies.     

Ethnic differences in parent/infant co-sleeping practices in New Zealand

AIM: This study was designed to monitor changes in the prevalence of risk factors for sudden infant death syndrome (SIDS) in the New Zealand population. The behaviour of interest is parent/infant co-sleeping. This paper reports parent/infant co-sleeping arrangements of different ethnic groups in New Zealand. METHODS: A stratified random sample of 6268 infants attending Plunket clinics for their three and six-month visits was taken over the years 1995-1996. Maori and Pacific infants were oversampled. Parents who shared a bed with their infant were asked how they arranged the babies sleeping place according to pre-coded diagrams. Routine parent/infant co-sleeping was defined as "bed sharing at least four nights over the last two weeks". RESULTS: There were 2693 infants who shared the bed with their sleeping parents during at least one of the previous 14 nights. Of these infants, 1060 routinely shared the parents' bed. At three months, 56% of routinely co-sleeping infants slept directly in the bed, 29% slept in a raised position, 3% slept in a carrycot or basket, and 5% in other positions. At six months, 60% of the routinely co-sleeping infants slept directly in the bed with their parents, 23% slept in a raised position, 1% slept in a carrycot or basket, and 7% in other positions. There were significant differences in the co-sleeping locations by ethnicity. CONCLUSION: There is still some ongoing dispute as to whether parent/infant co-sleeping is a risk factor for SIDS. This study has identified differences in the way infants co-sleep with their parents and this can be used to clarify infant care practices in relation to SIDS.     

A comparison of the effects of parental risk markers on pre- and perinatal variables in multiple patient cohorts with fetal alcohol syndrome, autism, Tourette syndrome, and sudden infant death syndrome: an enviromic analysis

The prevalence and magnitude of effect of individual risk markers for specific developmental disorders vary widely across diagnostic category. The four study cohorts for this project were patients from four diagnostic registries in North Dakota for fetal alcohol syndrome (FAS), autism, sudden infant death syndrome (SIDS), and Tourette syndrome. These four cohorts were used to estimate prevalence and magnitude of effect of parental risk markers in patients with developmental disabilities. Cases with North Dakota birth certificates were matched with controls. Using birth certificate data, we then examined five parental risk markers for each cohort and estimated direct and indirect effects for each risk marker by cohort. The authors found two significant paternal risk markers (age in SIDS and education in FAS). Significant maternal markers were age in SIDS, education in FAS, autism, and SIDS. Marital status was a significant risk marker in FAS. Effect sizes were estimated using paired t tests, odds ratios, and population attributable risk (PAR) for both direct and indirect effects for each marker. We estimated both direct and indirect effects to allow for direct comparisons of the differential effect estimates of each of these markers. The direct effect of parental markers differs across diagnostic cohorts of patients. Use of cohorts from similar denominator populations obtained from prevalence studies is a useful methodological tool for estimating the prevalence and magnitude of effect of risk markers.     

Results from the first year of the New Zealand cot death study

New Zealand's high mortality rate from the sudden infant death syndrome (SIDS) prompted the development of the New Zealand cot death study. This report of the preliminary analysis of the first year of the data gives the major identified risk factors. One hundred and sixty-two infants who died from SIDS were compared with 589 control infants, who were a representative sample of all hospital births in the study region. Obstetric records were examined and parental interviews were completed in 96% and 89% of subjects respectively. Data were available for all the variables in this study in 95% of those interviewed, thus 128 cases and 503 controls make up the subjects of this report. As expected we confirmed many risk factors for SIDS including: lower socioeconomic status, unmarried mother, young mother, younger school leaving age of mother, younger age of mother at first pregnancy, late attendance at antenatal clinic, nonattendant at antenatal classes, Maori, greater number of previous pregnancies, lower birth weight, shorter gestation, male infant, admission to neonatal intensive care unit. In addition, however, we identified three risk factors which are potentially amenable to modification. These were the prone sleeping position of baby (odds ratio = 3.53, 95% confidence interval 2.26, 5.54), maternal smoking (1-9 cigarettes/day OR = 1.87, 95% CI = 0.98, 3.54; 10-19/day OR = 2.64, 95% CI = 1.47, 4.74; 20+/day OR = 5.06, 95% CI = 2.86, 8.95) and breast feeding (OR = 2.93, 95% CI = 1.84, 4.67).(ABSTRACT TRUNCATED AT 250 WORDS)     

Water fluoridation and the sudden infant death syndrome.

AIMS: To determine whether exposure to fluoridated water supplies prenatally or postnatally at the time of death increases the risk of sudden infant death syndrome (SIDS). METHODS: A nationwide, case-control study, with infant's water fluoridation status determined from census area unit information for mother's usual address at the time of the infant's birth, infant's usual address at the time of death / nominated sleep and address where infant died / was at nominated sleep. SIDS risk associated with fluoride exposure postnatally was assessed according to method of infant feeding (breast or reconstituted formula), for the two days prior to infant's death / nominated sleep. RESULTS: Infants exposed to fluoridated water supplies during pregnancy were not at increased risk for SIDS, adjusted odds ratio (OR) 1.19 (95% confidence interval (CI) 0.82, 1.74). For breast-fed infants at the time of death / nominated sleep, fluoridated water exposure was not associated with an increased risk for SIDS, adjusted OR 1.09 (95% CI 0.66, 1.79). Similarly, 'fluoridated' formula feeding, when compared with 'unfluoridated' formula feeding, showed no increased risk of SIDS, adjusted OR 1.25 (95% CI 0.73, 2.13). There was no evidence of an interaction between fluoridation and infant feeding for the last two days (chi2 = 0.171, df = 1, p = 0.68). CONCLUSION: Exposure to a fluoridated water supply prenatally or postnatally at the time of death did not affect the relative risk for SIDS.     

Hexachlorobenzene stimulates uroporphyria in low affinity AHR mice without increasing CYP1A2

Adult female Fisher 344 rats received drinking water containing 0, 4, 40, 100, or 200 parts per million of dimethylarsinic acid or 100 parts per million of arsenate for 14 days. Urine was collected during the last 24 h of exposure. Tissues were then taken for analysis of dimethylated and trimethylated arsenicals; urines were analyzed for these arsenicals and their thiolated derivatives. In dimethylarsinic acid-treated rats, highest concentrations of dimethylated arsenic were found in blood. In lung, liver, and kidney, concentrations of dimethylated arsenic exceeded those of trimethylated species; in urinary bladder and urine, trimethylated arsenic predominated. Dimethylthioarsinic acid and trimethylarsine sulfide were present in urine of dimethylarsinic acid-treated rats. Concentrations of dimethylated arsenicals were similar in most tissues of dimethylarsinic acid- and arsenate-treated rats, including urinary bladder which is the target for dimethylarsinic acid-induced carcinogenesis in the rat. Mean concentration of dimethylated arsenic was significantly higher (P<0.05) in urine of dimethylarsinic acid-treated rats than in arsenate-treated rats, suggesting a difference between treatment groups in the flux of dimethylated arsenic through urinary bladder. Concentrations of trimethylated arsenic concentrations were consistently higher in dimethylarsinic acid-treated rats than in arsenate-treated rats; these differences were significant (P<0.05) in liver, urinary bladder, and urine. Concentrations of dimethylthioarsinic acid and trimethylarsine sulfide were higher in urine from dimethylarsinic acid-treated rats than from arsenate-treated rats. Dimethylarsinic acid is extensively metabolized in the rat, yielding significant concentrations of trimethylated species and of thiolated derivatives. One or more of these metabolites could be the species causing alterations of cellular function that lead to tumors in the urinary bladder.     

The effects of neonatal ethanol and/or cocaine exposure on isolation-induced ultrasonic vocalizations

Isolation-induced ultrasonic vocalizations (USVs) are emitted by young rat pups when isolated from their dam and conspecifics. These USVs play an important role in maternal/offspring interactions, and have been used as an indicator of response to stress and isolation. This study examined the effects of neonatal ethanol and/or cocaine exposure on USVs in neonatal rats. The neonatal exposure paradigm serves as a model for the "human third trimester of pregnancy" in terms of CNS development. There were five treatment groups including an artificially reared (AR) ethanol-exposed group (6 g/kg/day), an AR cocaine-exposed group (60 mg/kg/day), an AR ethanol- and cocaine-exposed group (6 g/kg/day+60 mg/kg/day), an AR isocaloric control, and a normally reared control. Both groups that received ethanol took longer to vocalize, and displayed fewer vocalizations than non-ethanol-exposed pups when tested on clean bedding (Experiment 1) or on chips from the nest of a lactating dam (Experiment 2). These results suggest that neonatal ethanol exposure alters the pup's immediate response to isolation. This could have direct effects on maternal/infant interactions, and might help explain some of the long-term effects of ethanol exposure on social behaviors.     

Postneonatal mortality review in Auckland: two years experience

Postneonatal deaths in the Auckland Region in 1984 and 1985 were reviewed. There were 134 deaths and most deaths could be placed into four broad categories, namely sudden infant death syndrome (SIDS, 80 60%), congenital anomalies (24, 18%), infections (9, 7%) and problems arising in the perinatal period (8, 6%). There was good agreement with the cause of death as recorded by the National Health Statistics Centre (98.5%) Potentially preventable causes of death were infrequent (14, 10%), but notable factors were present in 90% of SIDS. For SIDS cases the following notable factors were identifiable: young mothers, Maori, low socioeconomic status, poor accommodation, frequent changes of address, maternal smoking, previous postneonatal death, poor antenatal care, male infant, low birth weight, twin, poor infant weight gain.     

NICOTINE EXCITES CARDIAC VAGAL NEURONS VIA THREE SITES OF ACTION

1. Nicotine is involved in many cardio-respiratory diseases, including hypertension and sudden infant death syndrome (SIDS), which is the most common cause of death in infants between 1 month and 1 year of age. While the aetiology of SIDS remains largely unknown, recent clinical studies suggest maternal cigarette smoking is a major risk factor in SIDS and an abnormality of cardio-respiratory control, particularly a centrally mediated slowing of the heart that precedes or accompanies apnoea, is involved. 2. Because the sites, mechanisms of action and diverse receptor types of nicotine within the central nervous system are controversial and poorly understood, in the present study we examined the effects of nicotine on specific brainstem neurons that control heart rate. Cardiac vagal neurons were identified in an in vitro slice preparation using a retrograde fluorescent tracer and were studied using both whole-cell and perforated patch-clamp electrophysiological techniques. 3. We have found there are different pre- and post-synaptic nicotinic receptors that have dramatic effects on glutamatergic neurotransmission as well as directly activating vagal cardio-inhibitory neurons.     

Tissue distribution and urinary excretion of inorganic arsenic and its methylated metabolites in C57BL6 mice following subchronic exposure to arsenate in drinking water

The relationship of exposure and tissue concentration of parent chemical and metabolites over prolonged exposure is a critical issue for chronic toxicities mediated by metabolite(s) rather than parent chemical alone. This is an issue for AsV because its trivalent metabolites have unique toxicities and relatively greater potency compared to their pentavalent counterparts for many endpoints. In this study, dose-dependency in tissue distribution and urinary excretion for inorganic arsenic and its methylated metabolites was assessed in female C57Bl/6 mice exposed to 0, 0.5, 2, 10 or 50 ppm arsenic (as arsenate, AsV) in their drinking water for 12 weeks. No adverse effects were observed and body weight gain did not differ significantly among groups. Urinary excretion of arsenite monomethylarsonous acid (MMA(III)), dimethylarsinous acid (DMA(III)), dimethylarsinic acid (DMAV), and trimethylarsine oxide (TMAO) increased linearly with dose, whereas AsV and monomethylarsonic acid (MMAV) excretion was non-linear with respect to dose. Total tissue arsenic accumulation was greatest in kidney > lung > urinary bladder > skin > blood > liver. Monomethyl arsenic (MMA, i.e. MMA(III)+MMAV) was the predominant metabolite in kidney, whereas dimethylarsenic (DMA, i.e., DMA(III)+DMAV) was the predominant metabolite in lung. Urinary bladder tissue had roughly equivalent levels of inorganic arsenic and dimethylarsenic, as did skin. These data indicate that pharmacokinetic models for arsenic metabolism and disposition need to include mechanisms for organ-specific accumulation of some arsenicals and that urinary metabolite profiles are not necessarily reflective of target tissue dosimetry.