Document oajJo87Yk60ez7EEV47aOmg8

i TOXICOLOGY OF HYDRC fEP SULFIDE it. .I. Reijktisteitr *.*. 1hq)aitnieirt of Pharinacology , Uiiiversity of Alberta, Eclnionton, Alberta, Canada ' I ~ h G2 H 7 . - -. if -I -.- CYilli(irri C. Hrilber-t !r z.? SI1eI lot I ti. Ror11 Division o f Toxicology, Department of Pharniacology iund `I'herapeutics, University of Calguy, Calgary, Alberta, Canada T2N 4 N I KEY M'OI<I)S: neurolocicd toxicity. lung, airway, clcvulopmcnt irricl growth, forznsic analysis I N`I'RODUCTlON Ilistoi-ii.crl B3cickgroiiml It is allnost 300 years since the first clescription o f hydrogen sulfide (I-IZS) toxici~y( I ) . `!`here have, howover, been f'cw reviews and only one research c o i i t e i e i ~ e( 3 )on I12S toxicity. Numerous governmental agencies concerned with occiipational health o r the cnvironrnent have a t various times prepared ~ I o c i i ~ i i cr~e~Iat(scc1 to regulation 01` I 12S expohiire ( e . g . 3, 4). An excellent history of the eai-lyexperience with HLS appeared in the recent conference proceedings ( 5 ) . A review it1 1984 (6) included a bibliography of Ariio\t 1300 IcfCrCIlCC5, 196 o f which were cited in the text. `Tilt: general o1)iiiioii w;is that sulfide inhibited oxidative enLymcs i n a m m w r similar to I., -e e?::. c. 110 REIFFENSTEIN, HULBERT, & ROTH cyanide, particularly enzymes involved with oxidative phosphorylation, but it must be concluded that additional processes are operating. Therapeutic measures have been suggested that Follow the pattern used for the treatment of cyanide poisoning. However, animal experiments have not proved their validity for postexposure therapy, and clinical reports are still scarce. Recent advances have been made in the diagnosis of H2S poisoning, teratogenicity, and neurological and respiratory effects and have opened new possibilities for therapy. HZS poisoning is still a problem because of widespread environniental and occupational exposure from industrial activities, e.g. paper pulp mills, heavywater production, urban sewers, and farming, to name but a few of the more than 70 identified commercial sources. H2S is the predominant sulfur contaminant of natural gas and ranges in concentration from < I to >90%. At least three epidemiological studies have recently addressed the health of populations exposed to this toxie gas (7-9). Physicochemical Properties H2S is a colorless gas heavier than air (d = I . 19) with a molecular weight of 34.08 (IO). It is the sulfur analog of water. It can be oxidized by a variety of agents to form sulfur dioxide (SO,), sulfates such as sulfuric acid, and elemental sulfur (these products also have toxicological implications). One gram of H2S will dissolve in 242 ml of H20, 94.3 ml of absolute ethanol, or 48.5 ml of diethyl ether at 20C. Because of its lipid solubility, it easily penetrates biological membranes. H2S evaporates from aqueous solutions (vapor pressure = 18.75 X lo5 Pa). An aqueous solution will dissociate (see Equation I ) , yielding a hydrosulfide anion and sulfide ion; the two pKa values are 7.04 and 11.96, respectively. At a physiological pH of 7.4, approximately one-third of H2S exists as the undissociated form and two-thirds as the hydrosulfide anion. Some useful conversion factors for H2S are as follows: 1% volume = 10,000 ppm, 1 mg liter-' = 717 ppm (STP), and 1 ppm = 1.4 mg mW3 +H2S S H' HS- e 2H' + S2- 1. Species Comparisoiz The effects of acute and chronic exposure to H2S in many vertebrates and invertebrates have been investigated from three different points of view: lethality, on commercially valuable species, and mechanism of adaptation. The reader is directed to reviews on this subject (4, 6, 11, 12). Briefly, lethality data for humans, dogs, cows, goats, monkeys, mice, guinea pigs, and rats are very similar, probably because the effect of H2S on eukzryotic cells is similar (see ref. 11 for a discussion). Table 1 shows the effects of HZS HYDROGEN SULFIDE 11 1 `I'able I I luiiiaii physiologic responbes to exposure to hydrogen sulfide" Coiicciilr;itioii o f t ILS .'ppl" 1ng 111-- 0.003 - 0.02 3--10 20-30 30 50 50- I O 0 I O 0 -200 150-200 250-500 500 0.(1042-0.028 4-14 28 -4742 70 70 - 1-10 140-280 2 10-280 350--700 700 500- I O 0 0 700-I400 Physiological responses Odor 1hreslioltl Obvious unpleasant odor Strong offensive odor ("rotten eggs") Sickening sweet odor Conjunctival irritation Irritation of respiratory tract LOSS of smell (olfactory fatigueIb Olfactory paralysis" Pulmonary edema Anxiety, headache, ataxia, dizziness, stimulation of respiration, amnesia, unconsciousness ("knockdown") Respiratory paralysis leading to death, immediate collapse, iicural paralysis, cardiac ai-rhythmias, death on huniiiins. `There are some anomalies in the reported findings. Guinea pigs, but not rats, were reputed to die from exposure to 100 ppm of H2S. This may be related to the fact that guinea pigs have a more extensive nasal labyrinth, are obligate nose breathers, and did not evolve in a high-H2S atmosphere (sewers) like rats did (see 5 ) . ln obligate nose breathers, cellular damage, extoliarion, and miicus secretion cause the nasal passages to plug up, and the imiiiials simply cannot breathe and hence die from lack of oxygen. This effect has been seen in comparisons of nose-breathing and tracheotomized guinea pigs exposed to cigarette smoke (W. C. Hulbert, unpublished data). Birds (canaries) are more sensitive than mammals to H2S: 100 ppm causes 100% mortality. It is unknown whether the mechanism is similar to that in guinea pigs or whether it is due to altered metabolic or neurological function. Extensive studies of the effects of H2S on aquatic vertebrates have been conducted (see rei. 14 for ;Ireview), particularly channel catfish, brown trout, wulleyc, iiorhern pike, blue gill, rainbow trout, and the white sucker. H2S at'iec~sthese species at all stages of development from eggs to adults, and niany et'lects seein related to the ability of the species to express tolerance or atlaptalion ( I 5 ) . Fish reared in sublethal concentrations of HZS exhibited growth enliariced by SO to 200% (14, 16), owing to fungicidal and bactericidal et`fects ot` IflS, siinilnr to the effects of antibiotics normally used with captive fish. llowever, it was also noted that the fish were significantly less active ( 16)arid showed signs of respiratory distress. Histological analysis of the gill lainellae revealed structural alterations of the gill filaments, which were 112 REIFFENSTEIN, HULBERT, & ROI'H shortened and thickened, indicating chronic irritation. Unfortunately, there have been no analyses of muscle and fat content, muscle contractility or enzyme levels, or effects of exercise, factors that may be more significant than the simple enhancement of growth. H2Sbecomes a major problem in fish aquaculture during harvesting ( 17): when anaerobic sediments of fish ponds are disturbed during netting, H2S levels throughout the water column are elevated to the toxic range. Sulfide is abundant in the marine environment (especially near volcanic vents), and many vertebrate and invertebrate species (e.g. crabs, clams, and tubeworms) have evolved strategies for dealing with its presence. Thcse include sulfide detoxification in the body wall, binding and oxidation of' sulfide by blood components and by cytosolic factors, oxidation by mi tochondria, sulfide-insensitive cytochrome c oxidase, and ATP production from sulfide oxidation (see ref. 18 for a review). Target Organ Systems Most organ systems are susceptible to the effects of'H,S; therefore, this toxic gas has often been regarded as a broad-spectrum toxicant. The biological responses to H2S are dependent on the organ system; each system exhibits a different threshold responsiveness, perhaps as a function of concentration, time, or rate of exposure (19). Tissues most susceptible to H2S toxicity are those with exposed mucous membranes and those with high oxygen demands. The effects of prolonged exposure to low Concentrations are not well documented. It has been proposed that the toxicity may be cumulative (20) or noncumulative (21) and that the effects can be completely reversible. Iiecovery from acute intoxication is usually rapid and complete, depending upon exposure; however, some symptoms may persist (9) and some aftereffects may be irreversible as a result OF secondary effects caused by lack of oxygen due to respiratory paralysis and/or pulmonary edema (7, 21). NERVOUS SYSTEM Acute exposure to H2S leads to sudden Fatigue, vertigo, intense anxiety, convulsions, unconsciousness, and respiratory failure. After resuscitation, victims may suffer coordination and psychiatric disturbances, including hallucinations and amnesia. Chronic exposure leads to a variety of physiological and psychological effects (see Table 2). Early neurological studies concerned the increase in respiratory rate seen in moderate H2S exposures; this increase was attributed to stimulation of peripheral chemoreceptors (6). However, little progress had been made in describing the neurological sequelae of high H2S exposure until the advent of current electrophysiological techniques. HYDROGEN SULFIDE 113 13t uitI S1i1jLle CotI tent itI Poisoriitig IIie su~tide(S'. ) concentration present in tissues following poisoning was unknown until recently. A recently developed, extremely sensitive (2 pug liter - I ) method (22, 23), specific to S'- and ideal for analysis of tissue, has det'ined the appropriate concentration of 11's or its salts for experimental use in the brain. The method is 50-fold inore sensitive than earlier methods (24). Surprimgly , both rats and hunians have a relatively high endogenous level of S' : I .57 p g g--' for whole brain and 0.67 p g g-' in the midbrain. It would be of interest to know the S'- levels in brains of ruminants, since they produce large quantities of H2S. Iiecently bovine brain levels of S'- were reported to be about 5.3 p g g-' (24a), although it is not known how the HPLC nietliocl used compares with the method used for rats and humans (22, 23). At the 50% lethal close (LDSo)of NaI-IS (15 mg kg-l) the level of S2- in rat brain was approximately 3.1 pg g - ' (-75 pM). By contrast, another receiit nietliocl ii)r tissue S'- analysis (25, 26) gave much lower concerilratioiis in the brain. Significant to its effect on respiration, sulfide is selectively taken u p by brain stem compared with other brain areas (27). Iiihalalion of 1600 ppm of H'S and intraperitoneal (i.p.) injection of30 mg of NaHS kg- I (LAD doses within 4 min) produced indistinguishable increases in s 2 - - levels in the brain (28). tjtcttuse of tlicir essential role i n central nervous system (CNS) function, the coilleiit and releilst: of neurotrunsrnitters during acute and chronic exposure to 11,s or siilfic1e salts have been determined. Acute i.p. treatment with NaHS ( 2 x LDso) increased the concentrations of alanine, aspartate, y-aminobutyrate (GABA), glutamate, glutamine, glycine, and taurine selectively in the brain -stein; niinor or 110 changes were seen in other bwin areas (29). This dost: also iiicrcasctl serotonin (5-IIT), dopaniine, epinephrine, and norepinephrine levcls in the blain stem, the only region where all tour amine levels changed ('30).'l'licx chaiiges in catecholunine and 5-kIT levels are due to inhibition of iiioiioiiiiiiiie o x i c l i w (MAO). Acute treatment with sulfide also inhibits acetyl- clioliiieslerase ( 3 I ) and N a ' / K '.ATPase (see Electrophysiological Effects). 12eversal o t M A 0 inhibition was achieved ex vivo by removal of bound sulfidc with the persulfide reagent dithiothreitol (32). Experiniental hanclling of rats produced increases in brain stem glutamate, gliitaiiiiiie, iind taurine levels. Subacute treatment with NaHS ( 0 . 5 ~LDso) resultccl i n ;I reduction of this stress-induced increase (33). This dose of NaHS Iiatl no ci'l'ect on aiiiino acid levels in brain stenis of mice (34). Chronic expostire elepress~dbrain amino acid transmitters (see Reproduction and ikvelop1nentj. Therefore, it appears that degradation of amino acids (and 114 REIFFENSTEIN, HULBERT, & ROTH amines) is inhibited by acute exposure, but that synthesis is also inhibited by chronic exposure. Neurotransmitter Release Release of amino acids has been studied by push-pull perfusion (35)because of evidence that NaHS depresses synaptic transmission presynaptically (3639).Two paradigms were used: (a)NaHS (LD5*)was given i.p. or (b)3-4 p g of NaHS ml-I (the concentration of S2- in the brain after administration of LD50) was included in the perfusion medium. Perfusates from surviving animals revealed that most changes in amino acid release in the hippocampus or caudate-putamen were immediate or delayed increases (40,41). However, in the brain stem reticular nucleus, the only change was a delayed decrease in glycine release (42). These results do not provide evidence that sulfide inhibits transmission by depressing transmitter release. Electrophysiological Efsects Many in vitro neuronal preparations have been used as models in the study of the actions of sulfide, including those discussed below. Ideally, studies of respiratory rhythm generator cells would be desirable, but the technical difficulties of intracellular recording from a sufficient number of these neurons has led to the use of dorsal raphe as a typical midbrain nucleus. The use of the rat hippocampus may relate to the memory losses that are common in survivors of sulfide poisonings. FROG SYMPATHETIC GANGLION In view of the changes in catecholamine levels and in acetylcholinesterase activity (3 1) after administration of sulfide, effects of NaHS in the frog sympathetic ganglion have been investigated by the sucrose gap method (43,44). With this technique, cr2-epinephrine, muscarinic and nicotinic receptors, and Na+/K+ATPase electrogenic pump activity can be studied (45,46). NaHS reversibly depolarized the ganglion, but did not alter the depolarizing effect of nicotine. However, the hyperpolarizing effects of epinephrine and muscarine were both reversibly increased by NaHS. The hyperpolarizing response to pump activation did not change while NaHS was present, but was greatly potentiated after removal of the NaHS, recovering to normal after 45 min. A similar effect occurred in mammalian neurons (see Hippocampal CA 1 Neurons, below). It is remarkable that these neurons were exposed to sulfide for extended periods without being irreversibly damaged. It is tempting to suggest that the sulfide-induced depolarization is due to inhibition of the electrogenic pump, either directly or from inhibition of ATP production; however, this seems unlikely since Na+/K+ATPAse activity in the presence of sulfide was equal to that in the control. c 11 C n n 0 a' ei si CC vt M II (1 CI: N: N; dc ch eit su thi set tar, pla fir1 be trac CO ani' una wa: red; Thc less NaI was at : HYDROGEN SULFIDE 115 CRAYFISH SENSORY NEURON This preparation (Procambarits clarkii) was used to study the effect of sulfide on action potential (AP) generation and conduction, by using extracellular recording (47, 48). Sulfide salts (<lo-' M) caused an initial brief (= 1 min) block of APs, then a prolonged enhancement of AP amplitude, and then another brief inhibition of APs upon washout. Higher concentrations of sulfide caused irreversible changes. In the absence of intracellular studies, the reason for these changes remains to be elucidated. Sulfide did not, however, alter the rate of AP conduction. This is similar to earlier results obtained with frog sciatic nerve, for which large concentrations (1-100 mM) of sulfide only slightly reduced the conduction velocity (49). MOUSE NEUROBLASTOMA CELLS Murine neuroblastoma cells, clone N 1E1 15 derived from sympathetic ganglia, were used to study tetrodotoxin ('M'X)-sensitive Na' channels by the patch-clamp method (50). Ca2+and K+ currents were blocked by Cd+, Cs+ and tetraethylammonium (TEA+). NaHS, even as high as 10 mM, completely failed to alter the TTX-sensitive Na'. channels. As controls for sulfur-containing compounds, trials were also done with taurine and cysteic acid, neither of which alone affected the Na+ cliannels. It was discovered, however, that the combination of NaHS and either amino acid completely and reversibly inhibited the channels. Other sulfur-containing reagents (0.8 m M P-mercaptoethanol and 2 m M dithiothreitol) inhibited Na+ channels by themselves. In in vitro situations it seems unlikely that sulfide will affect APs. However, in vivo, where free taurine levels are normally high and further increased by sulfide, taurine could play a role in H2S depression of CNS function. IIIIJ1'C)CAMPAL C A I NEURONS CA 1 neurons in hippocampal Slices have been studied in current clanip by using potassium acetate-containing intracellular electrodes (36-38; R. J. Baldelli, R. J . Reiffenstein & W. F. CoImers, unpublished data). Slices were treated with 27-200 p M NaHS. The amplitude, duration, and threshold voltage of APs in CAI neurons were unaffected by 80 p M NaHS. The initial response of these neurons to NaHS was a rapid, reversible, concentration-dependent hyperpolarization (IH) and reduction of input resistance. Both effects were maximal at 160 p M NaHS. 'I'he reversal potential for the conductance change was - 100 mV,or slightly less than the calculated E K for these cells. An even more striking effect of Nut-IS was a further hyperpolarization (WOH) that occurred immediately after washout of the NaHS. This also was concentration dependent, being maximal at >200 @IS.ynaptic transmission, measured by extracellularly recorded 116 REIFFENSTEIN, HULBERT, & ROTH population spike and EPSP field potentials, and by intracellular EPSPs, was depressed by NaHS. Pharmacological investigation of the IH and the WOH suggests that these are due to the opening of a K+ channel and to activation of Na+/K+ ATPase, respectively. Maximal responses to 200 p M NaHS were studied (3&38; R. J . Baldelli, R. J. Reiffenstein & W. F. Colmers, unpublished data). Ex- tracellular application of 1 p M TTX (to block evoked transmitter release), 1 mM 4-aminopyridine (4-AP) to block the somatic "A" and "D" K+ currents, or 30 pM muscarine to block the voltage-dependent "M" current, did not alter the IH. However, it was reduced by 50 mM (but not 10 mM) TEA+ (36-38) and by 2 mM Cs+. Extracellular Ba2+ (1 mM), extracellular Ba2+ plus Cs+, and intracellular Cs+ blocked the IH and unmasked a depolarization response to NaHS. Conductance changes were significantly reduced only by Ba2' or intracellular Cs+ (36; R. J . Baldelli, R. J . Reiffenstein & W. F. Colniers, unpublished data). Intracellular release of neither MgATP nor CI- reduced the IH and change in conductance. Thus, it seems reasonable that the 1H is due to increased conductance of a K+ channel. Comparison of the data with two recent compendia of K + channels and their inhibitors (52, 53) makes it relatively easy to determine what that channel is not. It does not involve the fast transient voltage- dependent I*, nor ID, nor the Ca2'-activated nonspecific cation channel (/,), since 4-AP was ineffective, nor ZM, as muscarine did not inhibit it but Cs' did. Because sulfide blocks oxidative phosphorylation (6), the consequent depletion of ATP could activate ATP-sensitive K + conductances, but injection of ATP did not alter the IH. The conclusion that a gKArPis not involved must be tempered as the mechanism of K+-channel control by ATP is not understood; this action of ATP may also be inhibited, given the number of 'enzymes known or inferred to be inhibited by sulfide. TEA+ blocks many K currents, including voltage-dependent "delayed rectifiers" and some Ca' activated K+ channels (52). The cardiac-type inward rectifier (IlR) fits the antagonism data (inhibition by TEA+, Cs', and Ba2+, but not by 4-AI'). None of the procedures listed above, except intracellular Cs' , inhibit the WOH (36-38; R. J. Baldelli, R. J . Reiffenstein & W. F. Colmers, unpublished data). Moreover, the NaHS-induced K+ conductance changes often recovered before the maximum WOH was reached. The Na+/K+ATPase inhibitor strophanthidin (3-30 p M ) did inhibit the WOH; however, this treatment depolarized the neurons, and NaHS caused further depolarization. Thus it is likely that the WOH results from activation of Na+/K+ATPase, just as in the frog sympathetic ganglion. The NaHS-induced depolarization seen in the presence of strophanthidin suggests that sulfide does inhibit Na+/ K+ATPase in this preparation, although this is not so clear when using frog gang1ia. HYDROGEN SULFIDE 17 Neither did the treatments listed above affect the depression of synaptic transniission (36-38), which suggests a presynaptic effect. However, this is not consistent with the release experiments, evidence that depolarization by iontoplioresecl glutarnate pulses can be inhibited (36), or inhibition of glutaiiiaie binding to liippocanipal neuronal membranes (K. Fung, lvl. W. Warenycia, S. 13. Konibian & It. J . lieiffenstein, unpublished data), all of which suggest a pobtbynaptic elfect. IIOIZSAL I<Al'llli N E ~ J I Z O N SEffects of NaHS similar to those obtained in hippocanipal CA I cells have also been observed in voltage-clamped serotonergic ( 5 5 ) dorsal raphe neurons (39). Some cells responded to NaHS with an 11-1 (outward current) followed by a WOH, as in CA1 neurons. ilowever, i n solile of these raphe neurons the 1H was superimposed on a sirll'icle-iiillucecl depolarization (inward current). In both cases, blockade of the 111 with B a 2 + plus Cs revealed an underlying inward current. Some + iicuroiis rrsponded only with initial inward currents, followed by the same W O l l . In all cases the WOH was blocked by strophanthidin. The NatISintluctxl inward currents were also occluded by the strophantliiclin, showing that the depolarization was due to at least partial inhibition of Na+/K'ATPase by sulfide. A few neurons appeared to be unresponsive to sulfide. Althoitgh the mechanisin of action of H2S on neurons is not completely clear, sulfide c;in activate K ' conductances in at least two very different kinds 0 1 iieurons, alii1 these conductances are sensitive to extracellular application of Ha' ' and Cs '. Upon blockade of these currents, all neurons showed a depolarizing response to NakIS, which is caused by a voltage-independent (at least in dorsal raphe neurons) inward current and which may well be due to suppression of outward current generated by the Na '-Kf exchanger. Dorsal raphe ~iciiroiisare variably affected by sulfide, perhaps reflecting the relative contributions of the Na '/K '~ATPase activity and the Cs'-Ba'+-sensitive K+ coiiductaiices i n their resting state. All cells that responded to NaHS also showzcl tlir: outward current response to washout of NaHS. 'This action of ~irll'ideon N a ' / K ' A'll'ase i n iiiaiiinialian CNS neurons is rapidly and com- pletely reversible. Clearly, other types of neurons will have to be sampled before ii generalization about neural inechanisms of sulfide toxicity can be iiratle. At present, the relatively uniform responses of hippocampal CA 1 ~ C L I T O I IdS o suggest that inhibition may be the reason for temporary memory del'icits which occur in high 11,s exposure. The WOI-1, and potentiation of otlier inhibitory mechanisins, may well slow the return of function. I!YI)OXIA A N I ) A NO XIA Are the effects of 1l2S on nervous tissue simply thosc due to inhibitioii of oxidiitive nietabolism? Siinilar nienibrane potential reqionses to those induced by sulfide have been shown in vitro during anoxic 118 REIFFENSTEIN, HULBERT, & ROTH or hypoxic conditions (56-60), but there seem to be pharmacological dis- similarities. The anoxia-induced IH in hippocampal pyramidal neurons was reported to be blocked by 4-AP (0.4 mM) (56). Another group (57) found that Ba2+ (0.5 mM) blocked the 1H but that 4-AP (0.2-1.5 mM) and Csf (2-4 mM) were ineffective (57, 58). Neither group found TEA+ (3-10 mM) effective; however, less than 50 mM TEA+ was ineffective against NaHS (36). It has been variously concluded that the anoxia-induced IH is in part due to a Ca2+-sensitive gK (59) and also due to a muscarine- or carbacholsensitive gK (IM) (60). None of these anoxia data are consistent with the sulfide pharmacology. A reoxygenation hyperpolarization similar to the sulfide WOH has also been observed (57, 58). This also seems to be due to the electrogenic pump, since it was abolished by low K + concentrations or 1 pM ouabain. The action of sulfide is often compared to that of cyanide (6), which does block oxidative metabolism; however, on the basis of comparison of the actions of cyanide and sulfide on frog nerve (49), it has been concluded that sulfide acts by other than metabolic actions. At this time it seems that the processes caused by sulfide and simple anoxia, although producing a similar end point, are not completely identical. In view of the controversy over the actions of cyanide and sulfide and the role of chemically-induced "anoxia," it would appear essential to test the actions of cyanide in a similar manner. RESPIRATORY SYSTEM The effects of toxic gas exposure on the respiratory system have been the focus of intensive investigations (see ref. 61 for a review). One of the hallmarks has been the change in bronchial reactivity to inhaled nonspecific agonists. Although the patterns vary, at some level of exposure there is an increase in bronchial reactivity or hypersensitivity and the expression of asthmalike responses. However, there are few animal studies and even fewer human studies that have examined this very important effect on the pulmonary systern. Clinical Manifestations In reviewing the case studies of accidental exposure, the respiratory complaints are the second major group of symptoms reported after neurological ones. The most prevalent respiratory symptom following accidental exposure to H2S is dyspnea (7). In fact, dyspnea accounted for symptom complaints in 23% of 250 H2S-exposed workers who filed claims with the Workers' Health and Safety Compensation Board in Alberta, Canada. Other prevalent symptomatic complaints in that study were sore throats, coughs, and chest pain. In nine of these workers given pulmonary function tests, three showed an obstructive pattern. Other respiratory signs and symptoms seen less frequently wes fol I 1 rela SlIlY (: pdr dow ;\ssc sucl. syrn ;ind iISSC Srun A min, asyn resis soiin' Wily! (8.4' to 11: the I howc rcsisl 'ril inill sc Icc- I thosc signil It i 3.51I, 11' be iii( with ~"IIIlN tion I( funct i A it irir 'There si1t1cht icy, ai1 HYDROGEN SULFIDE 119 were pulnionary edema, cyanosis, and hemoptysis. One of the complications l'ollowing exposure to H?S is the development of pneumonia, which may be related to the inhibitory effect of l12S on alveolar macrophages and their subscquent alility to inactivate bacteria (63). Only oiie study has evaluated the effects of environmental levels of l-l$ on pu1nioiiai.y liiiictioii (8). This was ;in examination of the effects of living clownwiucl f m i i a natural-gas refinery. Although (lie investigations did not ;issess hypersensitivity by cliallznge with either t12S or a nonspecific agonist such ;is iiictliacholine, they did show that there was an excess of respiratory syiiiptoiiis i n tlic exposed aIea, especially i n children from 5 to I3 years of age arid iii ~icver-siii~keorvser 14 years of age. Unfortunately, their study did not assess t11edirect eflects of H2S, but, rather, those of the combined emissions l'rom the gas refinery. A rccent stiic1y (63) evaluated the effects of inhaling t12S, 2 ppm for 30 mill, on pulrnonary function in a cohort of pulp mill workers who either were asymptomatic lor asthma or had symptoms of asthma. No effect on airway resistance o r spccific airway concliictance i n the asyinptoiiiatic workers was l'oiiiiil. 111 the asthiiiatic subjects there were nonsignificant increases in airways rebistance (26.3%) and decreases in specific airway conductance (X.4%). 'l'he investigators concluded that exposure for ;I relatively short time to 11,s concentrations appreciably higher than those existing in ambient air in the pulp iiiill cloes not cause noticeable effects on respiratory function; however, in 2 01. the 10 asthniatic subjects, changes greater than 30% in both resist il ncc ;Incl coniliictance were fou nd , ind icat ing airflow obstruct ion . 'I'licsc: results must be evaluated i n light of two factors. First, no non-pulp i i i i l l workers were iiicliided iii the study. Second, as i t is known that a self sclcctioii takes place i n workers in the pulp and piper industry, such that only tliosc who c u i tolerate he i'uiiies work in the environment (61),this may sigIi i I'iuant1y pre-bias any rzsiiI ts , I t is significant that symptoms of obstructive air flow only occurred in the ilsIllli1;itics. One iiiight speculate that in ;i more normal population, there may be iiioIe responses to the inhalation of I-I2S i n usthniatic subjects, consistent with rheir hypersensitivity to toxic gases. To date, there have been no p1iiioii;ii.y function studies evaluating a cross-section o f the general popula- tiori IO cxl)osiii'c to 11,s. Until tliat is done, the effects 01' 11?S on pulmonary I'u iict i c)Ii ;II id b1-0IIC11i;I 1 react iv i t y i11 11ti nuns w i I I reniain specti1ative. Atiitwl Stiiclies Ir 7Iiel-e have bcen two major ;miniid studies that have examined the acute and siil)clii.oiiicel'lcts o f inhaling 112S on pullnonary function, bronchial reactivity, aiid lung histology (65-70).The subchronic experiments were designed to 120 REIFFENSTEIN, HULBERT, & ROTH .I !' assess the applicability of the current occupational standards for exposure to H2S in the workplace. Currently proposed regulations (71) for exposure of workers to H2S are a time-weighted average-threshold limit value (TWATLV*) of 10ppm for a week consisting of five 8-hr days and a time-weighted average short-term exposure limit (TWA-STEL@)of 15 ppm for 15 min. These levels were initially intended to prevent eye injuries ( l l ) , but the standards were never tested in animals to determine whether they protected against pulmonary injury. SUBCHRONIC LOW-LEVEL INHALATION Subchronic studies (65, 66, 6870) showed that the exposure to H2S at 1, 10 and 100 ppm for 8 hr per day for 5 weeks had no effect on baseline measurements of airways resistance (R,-,a measure of central airway function), dynamic compliance (Ctlyna, measure of ! peripheral airway function), tidal volume, minute volume, or heart rate. It was also found that the maximal changes in R L and Ctlynwith an aerosol methacholine (MCh) challenge were comparable for all groups. This was unexpected in view of reported damage to the nasal mucosal epithelium following H2S exposure (72). It was anticipated that baseline measurements would be elevated, reflecting changes in airway caliber due to injury. Although baseline and maximal responses to MCh were unchanged, there was a leftward shift in the RI*and Cdyndose-response curves (DKC) for some rats: these responded maximally to a 10-fold-lower MCh dose at all H2S exposure levels. Sensitivity (concentration of agonist causing a half-maximal response) and reactivity (rate of the response) varied widely; however, cluster analysis did show distinct groupings of response, with clear separation between animals responding like the controls and hyperreactive individuals (70). Histologic examination of the trachea and lungs in normal and hyperreac- tive rats revealed only subtle differences in structure; neither an inflammatory I infiltrate nor increased numbers of mast cells were seen in the hyperreactive animals. In many rats exposed to H2S, there were two consistent dose-related peculiarities: proliferation of the ciliated cells in the tracheal and bronchiolar epithelium, and a lymphocyte infiltrate of the bronchial submucosa in adtlition to the recognized bronchoalveolar lymphatic tissues. Neither change has been reported to be causally related to bronchial reactivity. The mechanism of the hyperreactivity may be related to increased airway mucosal permeability. Low concentrations of thiols such as HZS and CH3SH are known to markedly increase permeation of macromolecules across the porcine oral mucosa (73). A series of studies on the acute effects of cigarette smoke inhalation have shown that increased bronchial reactivity is associated with increased mucosal permeability (74). The identification of individual rats that were hypersensitive after inhaling f 1 I I I 1 L I t 1 < :1 tl v 1: ti P tl I1 P 111 it SI dt 1' tr: e? 111 (i ti i D th. b in1 ior SPI , HYDROGEN SULFIDE 121 112S Iius significance for accidental hurnan exposure to 1-12s and possibly to other toxic gases. I n some rats the hyperrcactive response occurred after they llild inliiiled only I ppni, which is olily twice tlie conccntration recorded in the city 01' tidiiiontori, Canada, tltiriiig the 1982 gils well blowout 138 k m distant. W Iiztlier prolonging the exposure would recruit more individuals into the IiyperreiicLive subgroup is not known, but the DRC x m i s to imply this. These i'intlings indicate that tlie occupational exposure standards that were estat,li~hecl tipon other criteria (eye damage) may not necessarily protect the p 111I11011ary sy ste111 from cla111age . A ( ' L J I ' Eiii(iii-i2i;vu2I N Ii A i A W o N The acute effects on pulmonary function iiiiii 1)roiicliiiiI reactivity of inhaling moderately high concentrations of H2S for I 11 were investigated by using guinea pigs (67, 68). There were no effects on baseline KI.or Cdylf,or the control, 100-ppm, and 500-ppm groups, but there was a leftward shift in the aerosol MCh DRC that was related to the HIS dose. Sensitivity and reactivity increased 10- and 3-folc1, respectively, for XL and 3uid 4-fold, rcspectively, for Cdyl,.Thus, the central and peripheral airways of the aiiiinols WCIX more sensitive to inhaled MCh, and the slope of the DRC W L I ~stcupcr after exposure to H2S. ?'here are several possible explanations. lhxaiise baseline rneasurements were not significantly different; it is doubtful tliat the startiiig airway caliber is involved. Increased mucosal permeability is possiblz, and its relationship with hyperreactivity is well documented, as tliscussctl above. Another possible mechanism is a change in the smooth iiiiiscle itsell'. Alterutions in the sensitivity of the irritant receptors are also possible: iiltliough not likely, given the lack of effects of H2S on APs and i w v c contluction (see Electroph),siological Effects). '1'111: hypothesis that the increased sensitivity and reactivity were due to iiicrcasctl ~iiiicosalpernieability Iius been tested (75). First, following exposiirc LO I O 0 and 500 ppm 01' 1 12S, the. niucosii became hyperpermeable to iicxtruu (Fl'Kx'-4O; mo1ccul;ir weight, 40,000) in a dose-related way. 'I'ransinission electroil-microscopic analysis showcd precipitation of the dextran i n the intercellular spaces in airway tissues from the animals that had been exposed to II2S but not the controls, corroborating the physiological measurements on dextran in blood. Second, when MCh was injected intravenously ( i . v . ) (bypassing the effect of tlie airway IiiLicosa by delivering the MCh directly to the sinooth muscle), i t was found that there was no difference in the 1)KCs between any o f the control or H2S-exposed animals. This confirmed hat bronchial hyperactivity following the acute exposure to H2S involved 1~y~)er~~~r~i'eao"fbitlhietyairway niucosa, which facilitates the access by iiihaled agonists to the underlying smooth muscle. The mechanism of separation of the tight junctions responsible for the hyperpermeability remains speculative, but is believed to involve alterutions in tlie interaction of the 122 REIFFENSTEIN, HULBERT, & ROTH cytoskeleton at the plasma membrane, influencing the integral components of the tight junction, or focal changes in intracellular and/or intercellular levels of Ca2+, or both. Other pulmonary studies have shown that rats exposed to H2S at 10,200, or 400 ppm for 4 hr develop necrosis of the nasal epithelium followed by exfoliation of ciliated and olfactory mucosal cells but not of the squamous epithelial cells; these cytotoxic effects, and polymorphonuclear exudation and pulmonary edema, were dose related (76). Resolution proceeded much faster in nasal than in olfactory mucosa, as olfactor cells were still exfoliating 44 hr after exposure. This may relate to clinical reports of olfactory fatigue or paralysis. The per-acute effects of inhaled H2S and injected NaI-IS (LDlo+) on the lungs of rats were studied (77). At necropsy, all rats in the H2S group had gross and histological pulmonary edema, characterized by massive extravasation of eosinophilic fluid into the bronchoalveolar space. By contrast, the NaHS group were unaffected. Although the levels of S2- achieved in the brain were identical (28), the effects on the lungs were not. It is not known whether this was due to a concentration difference at the site of action or whether direct access to the alveoli is required. The DRC for rats exposed to H2S is very steep, indicating the possible existence of a sensitive physiologic threshold that, once breached, leads to pulmonary edema and, shortly thereafter, to death (78). Others have shown i that when rats are pretreated with capsaicin, the C-fibers that innervate the central airways are depleted of substance P, thus removing the airway defense properties mediated by tachykinins (79). When these capsaicin-treated animals were exposed to a concentration of H2S that causes 20% mortality in controls, they all died. As well, the capsaicin-pretreated animals died faster than controls and showed more pronounced pulmonary edema. , REPRODUCTION AND DEVELOPMENT Before 1984, the effects of H2S on reproductive processes were not well established. The few reports in the literature originated from studies of the effects of "thermal" mineral waters (containing sulfides). These studies claimed that H2S may be teratogenic and embryotoxic, as well as suppressing the spermatogenic index. The studies lacked scientific merit because they provide few, if any, details of methodology, results, statistical analysis, or adequate controls ( 6 ) . Adverse effects on reproduction following chronic exposure to H2S have been described (80);however, the gas was coadministered with CS2, which is well known to be associated with increased toxicity of reproduction (8 1). HYDROGEN SULFIDE 123 K eptwdi ~i ioti 'I'hese ambiguous effects and uncertainty of the results warranted further reseiircli on l-i2S-induced effects on reproduction and development. Recent well-controlled studies have provided definitive proof that chronic low-dose (<100 ppmj exposures to H2S do not procluce any significant adverse effects on reproduction in rats (82, 83). At parturition, however, it was observed that soiiie tlanis exhibited ii dose-dependent increase in delivery time (dystocia) that could have resulted in loss of fetuses owing to asphyxiation (82). In vitro bludies provided support [hat this effect may be a result of a reduction or blockade of oxytocin receptors (85). I'ups exposed to t12S (<75 ppm) in utero and neonatally to day 21 postpartum developed noriiially, with only a subtle decrease in time of ear detachment and hair growth (82). N o significant differences were seen in growth and wciglit gain, dthough depression of weight gain occurred in adult rats. hleusiircd glucose levels were significantly elevated in rnaternal blood, and beriiiii triglycerides were decreused in pups and dam; however, there was no evidence of alterations in alkaline phosphatase, lactate dehydrogenase, or serutn glut;imate transaminase (83). 'I '11e clcveIopi 11 g or i i n niat 11re organ isI ii I ;ICk s ni;I ny de fensi ve inec11an i sins such as metabolic processes (86). An incomplete blood-brain barrier and the i iipitl growth characteristics make the brain particularly susceptible to various toxicants (86). Until recently, there was no conclusive evidence that H2S iil~ei.cc1111e developing brain. An isolated report (87) suggested that the i.etaided developnient and listlesmxs of breast-fed infants of mothers work- ing in rayon factories was clue to 11,s. A brief case report (88) described a 20-inonth-old child with encephalopathies that may have been due to chronic 11,s exposure and that reversed spontaneously. We liiive iwcntly docuinenteil that chronic exposurcs to low levels of H2S ( 2 0 -75 p p i ) can procluce siibtle but significant alterations in the architecture arid groLlrth cliaracteristics 01' the cleveloping brain in rats (89). Rat fetuses (in iilcro) itncl neonatiil pups were chronically exposed to low concentrations of 1I2S (50-75 ppiii, 7 hr per day, 7 days per week) from 5 days postconception to 21 days postpartum. The growth patterns of the dendritic fields of developing cerebellar Purkinje cells were evaluated by using a digitizing method of analysis { W ) ,which quantitates growth of the dendritic trees. Exposure to 20 iiiitl 5 0 p p i i of I 12S protluced longer branches, iin increase i n the vertex path Icngtli, aiid variations in the number of branches in particular areas of the tlen~lriticfield. The cells also exhibited a nonsyinnietrical growth pattern at a time when random terminal branching is normally occurring (89). In another a 124 REIFFENSTEIN, HULDEK`I'. s( KOI`H study, changes in the amino acid content of developing rat brain tissue were determined (90). On postnatal day 21, aspartate, glutamate, and GABA levels in the cerebrum and aspartate and GABA levels in the cerebellum were significantly depressed. At this critical time of development, it is possible that a decrease in neurotransmitter contcnt may reflect a cellular loss o r an alteration in the synthesis or release. In preliminary studies we observed that the mean number of cerebellar Purkinje cells was increased by approxirriately 20% (48), suggesting that the decrease in GABA was unrelated to loss of Purkinje cells. It also appears that the normal perinatal loss of Purkinjc cells is suppressed by H2S. Exposure to low concentrations of H2S also rcsultcd in a n initial increase in the level of taurine in the pups, which may have resulted from maternal sources (91). The return to control levels coincided with the approximate time of establishment of the blood-brain barrier (9 I ) , but may be a result of development of capacity to metabolize taurine (90). I n addition, the dams exhibited inhibition of alkaline phosphatase and cytochrome oxitlase in brain (90). These studies provide evidence that chronic exposure to low concentrations of H2S does affect development of the CNS and niay contrihUte to possible long-term abnormalities in motor function and behavior. I t is not yet established whether these effects are reversible in the continued presence of H2S or following removal; a study longcr than 21 (lays postpnrtuni (i.e. at least 90 days) is required. SECONDARY TARGET SYSTEMS Comparatively little attention has bcen focused on other organ systcriis ovcr the past 10 years. Below are summaries of the known effects of H2S on the various organ systems; the reader is referred to previous reviews ( 4 , 6 , 21), or to one of the many regulatory documents, for most bibliography prior to 1982. Recent developments are specifically noted. The broad spectrum of actions and physicochemical properties of H2S predict that all or at least most organ systems will be affected to some degree. In both humans and animals the moist mucous membranes of the eye can be directly exposed to H2S, and at sublethal concentrations the cornea and conjunctiva are usually irritated (keratoconjunctivitis) (6). This was the first reported toxocity of H2S (1). Common complaints following exposure to H2S are listed at the beginning of Table 2. If exposure continues, epithelial cells swell and blister and progress to form vacuoles, which may burst into painful but reversible ulcers on the corneal surface. This condition is often referred to as "sore eye" or "gas eye" (93). In severe cases, ulceration of the cornea has led to scar formation and permanent visual impairment. Recently it has bcen HYDROGEN SULFIDE 125 shown that an increase in epithelial cells collected by an eye wash (conjunctival cells increasing relative to corneal cells) constitutes an objective iiieasure o f ocular irritation by H2S (93). O n e of the inost coninion complaints of individuals exposed to low concentrations of Ills is the unpleasant odor. The odor threshold of H2S is very low coinpared with iiiany other chemicals (2 1); however, at concentrations over 100 ppni there is an apparent loss of the smell sensation, which is said to be due to olfactory fatigue. However, fatigue or paralysis of the olfactory nerve has seldom been referenced to an original source, and this wellaccepted belief should be reexamined ( 5 ) . Prolonged exposure may result in a lower threshold for olfactory fatigue (3), although this also has been disputed ( 5 ) . Nevertheless, the odor of HzS at low doses may not provide an adequate w;iriiii.ig. Siiice sensitivity iippexs to diminish with age, it is likely that older persons i n the workforce will have higher thresholds (2 1). Skiti Altliougli the skin is the largest organ of the body and, to various degrees, will coiiie into direct contact with H2S, there have been few reports on the dermal effects of Ii2S. Some earlier observations at high concentrations have docuiiiented ciiscoloration, spots, and rash. Most anecdotal reports describe reversible skin irritation or allergies that are believed to be a result of exposure to low levels of H2S (<20 ppni), and a few have also documented observations at very high concentrations. Ccirclio~,nscirlalS-ystem Several clinical reports demonstrate the sensitivity of the human cardiovascular system to H2S. Acute exposure to high concentrations has resulted in traiisieiit changes in electrocardiograIms as well as decreases in blood presbtiil:. At low concentrations, H2S may not present any risk to the cardiovascular systein (94). Since cardiac muscle is one of the tissues with a high oxygen deiiiand (95), it is likely that some effects of HZSinay not be directly on the inyocarclium, but secondary to anoxia due to pulmonary edema and CNS el'l'ects. Clinical observations have been well substantiated by animal studies. I t was demonstrated that acute or chronic exposure of rabbits and guinea pigs to 72 p p i i ol' I 12S or inlravenous NulS produced venlricular extrasystoles (6). NaIIS caused arrhythmias and a progressive increase in tension in isolated rat atria ( R . J . Keiffenstein & B . Phipps, unpublished data). Histochemical exuniination of myocardial tissue from H2S-exposed rabbits also revealed enzymatic changes, suggesting a possible interference with oxidative nietabolisni and therefore alteration of ionic conductances in the excitable 126 REIFFENSTEIN, HULBERT, & ROTH tissue. In other studies, chronic exposure to lower concentrations of H2S (10-80 ppm) showed no effect on the heart rate or blood pressure of mice and rats (97). There still exists the possibility that individuals with some form o f cardiovascular disease may be more sensitive to the effects of H2S and thus form a part of society at greater risk of H2S toxicity (9, 98). Hepatic Tissue The few reports describing possible effects of H2S on liver function are conflicting and inadequate (6). Clinical studies have suggested that the incidence of cholecystitis, cholangitis, and cholelithiasis was higher than normal in oil refinery workers. Data from animal studies are less helpful, describing no effects in mice or rats chronically exposed to t12S up to 80 ppni (97) or conversely, decreased bile flow in rats treated with 40 mg of Na2S kg-' (99), enlarged paled livers of mice exposed to 63 ppm for 16 hr, a n d severe hyperemia of monkey liver exposed to 500 ppm (6). Renal System At low concentrations, H2S appears to have little effect on kidney morphology (97) or enzyme activity; however, earlier studies described changes in the color of mouse kidneys exposed to 63 ppm for 16 hr and pathological changes in exposed rat kidney (6). Histochemical studies of H2S-exposed rabbits revealed some reductions in levels of renal enzymes such as succinic dehydrogenase and alkaline phosphatase, as well as enhancement of acid phosphatase (6). These results, similar to those described for liver function, are inconclusive and require more definitive and controlled studies with both acute and chronic paradigms. Gastrointestinal System It is common for individuals exposed to H2S to experience gastrointestinal symptoms such as nausea, vomiting, diarrhea, and pain. The effects appear to be reversible and non-life-threatening in both humans and other animals (97). Hematopoietic System The few documented studies of the effects of H2S on the hematopoietic system have reported variable results. Both increased and decreased erythrocyte counts were recorded in animals exposed to 1-50 ppm and to very high (>900 ppm) levels of H2S, An increase in a variety of hematological parameters, but a decrease in erythrocyte numbers in mice exposed to H2S have been reported (100). In contrast, other studies (97) observed no changes in the hematological parameters following chronic exposure to H2S at 10-80 ppin. A decrease in enzymatic activities associated with heme synthesis occurs in humans exposed to H2S plus methylmercaptan during wood pulp production HYDROGEN SULFIDE 127 ( 1 0 1 , 102). Results of interaction of 1HLSwith henioglobin to produce sulfheiiioglobin iiIe also contradictory ( 5 , 6). I t I 1 11111t 1 e S y s t ettl `I'liere ;ire a few studies which suggest that H2S interferes with the iminune system. I t wiis concluded i n one study involving SfqiIzylococcus challenge in rats exposud to 45 ppm of FILSthat secondary infection due to depression of iiiaciupliuge function may occur following I-12S exposure (62). This was receiitly conl'irnied in an epidemiology study (8). T h e possibility of allergic or enhanced anaphylactic response has also been iinplicated in one early study (6), but the response in rabbits was opposite to that in guinea pigs; this creates some doubt about the validity of the results and their interpretation. bkloct-it1e S)i.sterii I'ossiblc: altct`ation of endocrine functions has also been suggested by a clecreuse ill milk production in cows exposed to 20-50 ppm of HZS and by a 50% increase in plasma cortisol levels i n goats at 100 ppni. Dose-dependent lesioiis 01' rat thyroid gland following administration of 14-28 ppin of H2S havu also beuii reported (6). ~~s))c.liolosic.cEifLiects Dehaviorul and psychological effects of H2S (see Table 2) have been discussed in several earlier studies, and there is a recent report of persistent cogiiitivt: iinpairiiieiit of three patients following acute exposure to H2S (103). A recent case of "knockdown" (unconsciousness) resulted in permanent retrograde amnesia (I. M . 0. Vicas, personal communication). The offensive odor is often interpreted as ciungerous or life-threatening, and this could induce ;I vx-iety o f both psychological and neurophysiologic111 reactions. C(ircirI o g e ~eisis `I'hcre tias been very little activity in research concerning the carcinogenic, iiiulageiiic, o r teratogenic effects of H2S in huinans and other animals; the reporkil guriotoxic effects niay be limited to cytotoxicity (6). Further investigations in validated test systems are obviously required. CLINICAL APPLICATlONS Lliagiiosis of exposure to 1 12Sis usually a matter of circumstance. There are a witlc variety o t symptoms (Table 2), not all of which may be present. 128 REIFFENSTEIN, HULBERT, & ROTH Table 2 Clinical symptoms after H2S exposilre' "Felt ill" Visual "fogging" Conjunctivitis Photophobia Tearing Eye pain Nausea Vomiting Anorexia ~~~ ~ Headache Insoninia Sore throatlcough Chest pain Dyspnea Hcrnoptysis Lethargy Abnormal peripheral reflexes Weakness of extremities ~ aConipiletl from several s~iitrces(2, 7 , 108). Depression Irritability Amnesia Disequilibrium Convulsions Pulmonary edema Cyanosis Unconsciousness Bradycardia Forensic Detection of H2S Poisoning Confirmation of sublethal exposure has been difficult until the recent development of a simple method for detecting blood sulfide levels (105). Even so, the procedure is too long to provide rapid diagnosis. Sulfide levels are still elevated in blood samples taken 2 hr after exposure ( 101, 102). In atlclition, some enzyme levels remain depressed long after the exposure (101, 102). After lethal exposures, the brain sulfide content can now be measured for forensic purposes (22, 23). To date this procedure has been used to determine the cause of death in four cases of suspected H2S poisoning (106); one of these was determined not to directly involve H2S. The use of dithiothreitol in this test should improve the differentiation between normal and poisoned individuals, although this has only been tried in animals thus far (32). Therapeutic Munagernent of H2S Poisorzing The initial events in recovery are the most important, and immediate removal to fresh air is paramount (but inadequately equipped rescuers often become victims). Most victims, even though they may be unconscious, appear to recover spontaneously if they are breathing. If they are not, assisted breathing should be instituted immediately, with full cardiopulmonary resuscitation if there is no heartbeat (107). Given the low level of H2S exhaled, there would appear to be no danger to the rescuer in the use of mouth-to-mouth respiration (107). Only then should other measures be instituted. Two such approaches have been advocated: scavenging sulfide with methemoglobin formed by administration of nitrites, and administration of hyperbaric oxygen. NITRITES Nitrite does protect against subsequent poisoning in animals (108), but there are few cases in which treatment after exposure can be shown to have affected the outcome. One early survey (109) suggested that nitrites were of little use. Four rei cent case reports (1 10-1 13) give conflicting evi- HYDROGEN SULFIDE 129 dence 01. the value o f nitrites. It has been suggested ( 5 ) that nitrites may be of IISC only if given within minutes after the 112S exposure. OXYGLN Although nitrite did protect against sulfide toxicity, the same study h w e d that 100% 0 2 at atmospheric pressure had no beneficial effect (108). 'I'here are two case reports of the use of hyperbaric O2 therapy ( 1 12, 113) (in which nitrites were also used). In the first there was evidence of pulmonary edema, when the hyperbaric O2 undoubtedly increased O2 delivery. In the latter case, after extensive but ineffective treatment with nitrites, I2 treatiiients with hyperbaric oxygen were given over 6 days before the patient was asyniptoniatic. This cannot be distinguished from normal recovery. One other patient (1. M . 0. Vicas, personal communication) was unconscious tor several hours and regained consciousness during hyperbaric oxygen therapy; however, this individual had severe, apparently permanent, retrograde amnesia. Although this therapy is modeled after treatment for cyanide poisoning ( 1 13), it is still unclear whether hyperbaric oxygen affects the outconie ot 1I2Spoisoning (107). Even if the increased partial pressure of the oxygen conipetitively reactivated oxidative cytochromes, as has been suggested ( I 13), it is possible that the action of ATP thus produced is still inhibited by sull'ide. l ~ I ~ I ~ S ~ J l . It<~lilAl C~ El iNTS Some enzymes remain inhibited f;Lr beyond the Lillie to apparent recovery, e.g. some blood enzymes returned to normal only 2 months after the H2S poisoniiig (101, 102). This suggests that some of the sult'itle reniains firinly bound and is not removed by lowering the plasma sulf'idc:levels by scavenging or metabolism. It should be possible to actively retilove this 1IS..- ion by persulfide reagents. Dithiothreitol given 20 niin bcl'orc Nal-IS pmvided significant protection in rats (32, 44); however, death was too rapid for this agent to provide resuscitation if given after the NaHS. Dithiothreitol administration increases the amount of sulfide recovered from the brain tissue of poisoned animals and reverses the inhibition of M A 0 by tlS- (32). It also restores contractile function in oxidant-injured cardiac Iiiuscle ( 1 14). 'I'his might be cloubly useful, since the 1975 report of increased cur.dio\iascular niortality among workers exposed to H2S (98) has recently I)eeii conllniicd (9), and there are a number of reports of prolonged arrhythrnias lidlowing exposure of animals to H2S (6; see also Cardiovascular Systeni). Interventions that actively remove sulfide from the sites causing inhibition of- enzymes, or altered control of ion channels, should hasten recovery. Antidotes suitable for use in the field, especially in situations in which allnost iiiiinediate death occurs, are unlikely to be found ( 5 ) . However, i n ciiscs of exteiidetl unconsciousness, when the victirns are hospitalized while still alive, this approach should be considered. ' 130 REIFFENSTEIN, HULBERI', & KOTH SUMMARY Significant progress had been made in determining the action o f sulfide on the primary target organs. It is reasonably clear that sulfide causes both K'channel-mediated hyperpolarization of neurons and potentiation of other inhibitory mechanisms. It is not clear whether these processes are similar to those that occur in anoxia. Changes in perinatal and adult brain neurotransmitter content and release may be related to clinical impairment of cognition. H2S exposures at concentrations below the current occupational limits cause physiological changes in pulmonary function, thus suggesting that asthmatics are at risk. Studies of fetal and neonatal brain tissue have shown an abnormal development, and the long-term consequences of these neuronal changes have not yet been assessed. Finally, new approaches to therapy are required, such as the use of agents that actively remove sulfide from its sites of action. This may prove more useful in preventing some of the long-term adverse sequelae than the use of nitrites and hyperbaric 0 2a,lthough the latter should be used in cases of pulmonary edema. ACKNOWLEDGMENTS We thank the various agencies which have supported our research: Alberta Occupational Health and Safety; Alberta Lung Association; Alberta Heritage Foundation for Medical Research; Medical Research Council of Canada; and members of the Canadian Petroleum Association. We are also indebted to the many participants in the International Conference on Hydrogen Sulphide Toxicity, Banff, 1989. Literature Cited 1. Ramazzini, B. 1713. Diseases of Workers. (Transl. from the Latin text De Morhis ArtiJciiin by W . C. Wriglit, 1940) Chicago: Univ. Chicago Press. 549 pp. Reprinted 1964 in History Med., Vol. 23 2. Prior, M. G., Roth, S. H., Green, F. H. Y., Hulbert, W. C., Reiffenstein, R., eds. 1990. Proc. Int. Coif. Hydrogen Sulphide Toxicity, Banff, 1959. Edmonton: Sulphide Res. Network. 231 pp. 3. National Institute for Occupational Safety and Health. 1977. NIOSH Criteria for a Recommended Staridard . . . Occupational Exposure to Hydrogen Suijide. DHEW (NIOSH) Publ. No. 77158. Cincinnati: US Dept. Health, Educ. Welf. 4. Report on H2S Toxicity. 1988. Edmonton: Alberta Health. 65 pp. 5 . Smith, R. P. 1990. Hydrogen sulphide: Overview, history, theories and unknowns. See Ref. 2, pp. 1-13 6. Beauchamp. R. O., Bus, J. S., Popp, J. A., Boreiko, C. J., Andjelkovich, D. A . 1984. A critical review o f the literature on hydrogen sulfide toxicity. CRC Crir. RPV. Toxicol. I3:25-97 7. Arnold, I. M. F., Dufresne, R. M., Alleyne, B. C., Stuart, P. J. W. 1985. Health implication of occupational ex- posures to hydrogen sitlfide. J. Occrcp. Men. 27~373-76 8. Dales, R. E., Spitzer, W. 0.. Suissa. S., Schechter, M. T., Tousignant, P., Stinmetz, N. 1989. Respiratory health of a population living downwind from nat- ural gas refineries. Am. Rev. Respir. Dis. 139595-600 9. Jappinen, P., Tola, S. 1990. Car- diovascular mortality aniong pulp mill workers. Br. J. Iiid. Med. 47:259-62 IO. Budavari, S., O'Neil, M. J., Sniith, A . , Hcckclnian, P. E.. ctls. Tire Mrrck f i i dex. Rahway, NJ: Merck & C o . . Inc. 2302 pp. 1 lth ed. I I . Smith, R. S., ed. Subcommittee on Hy- I HYDROGEN SULFIDE 131 drogcn Sulfide, National kscarch and ruiiien tluid by ion-interaction re- Council. 1979. Ifytlrogetr Sulfide. Balti- versed-phase high-performance liquid inore: University Park Press. 183 pp. chromatography. J . ChmzLitogr. 526: 12. National I<cscarch Council. 1977. H v - 5-KL-45 droxcrt Sidjide. Washington, DC: Natl. 25. Nagaia, T . , Kage, S., Kimura, K . , Acad. Sci. I49 pp. Kudo, K . 1987. Analysis of sulfide in 13. Delcted i n ~ I J O ~ blood using extractive alkylation and gas 14. Smith, L. L , , Oseid, D. M . 1974. Effect chromatography. In Proc. 24th I n r . uf hydrogen sulfiilc on developiuent and Meet. I t i t . Assoc. Foretisic T u s ~ Led~. . survival of eight freshwater fish species. 111 7'11~1 h l y L,g> tlisrory oJ'Fidi,cd. J . G . R. Jones, P. P. Singer, pp. 35256. Ednionton: Univ. Alberta Press. 554 S. 11. Blaxtcr, pp. 417-30. New York: I'P. Springer-Verlag 26. Naguta, T . , Kage, S . , Kimura, K., 15. Hochaclik;i, 1'. W., Soinero, C . N . 1979. Srt.tfre~iesof I3iochetllicd ` 4 l l q ~ - Kudo, K . , Noda, M . 1990. Sulfide concentrations in postmortem niarnmnlian 1 i w . Pliiladelpliia: Saunders. 358 pp. tissues. J . Furetisic Sei. 35:706-12 16. I<eynoltls, I:. A , , Haines, T. A . 19x0. 27. Warenycia, M . W., Coodwin, L. R., Effects of chronic exposure to hydrogen Benishin, C. G . , Reiffenstein, R. 1.. sulfide on newly hatched brown trout Francom, D. M . , et al. 1989. Acute hy- SllI/tlU t/`f1llLfL . Emirut1. POlllll. 22: I I- drogen sulfide poisoning: Measurement 17 of brain sulfide levels demonstrates se- 17. E.' r o l ~ r a l l b , I.. , cielnens, 11. 1'. 19x2. lective uptake of sulfide by the brain- I'liysiological and bioctieinical effects of StClll. ~ i O ~ ' / l ~ l Pl l/ l. l J l - t / f t f ~ O3~8.1973- acutc exposure 01' fish to hydrogen sul- 81 fide. c'ottfp. f ~ i O < ~ / l ~P~h/ )t :Js.io/. 7 I : 183- 2X. Reiffenstein, R. J . , Goodwin, L., Prior, 90 M . , Lopez, A. 1988. Sulphide concen- I X . V ~ ~ t eI<r. , D., Bagariniw, T. 1990. De- tration in brain after H2S inhalation and toxit'icaiioii and exploitation of hydrogen NaHS injection. Proc. C u i . Fed. Bioi. sulpliide by niariiie organisms. See Ref. Soc. 3 I :I6 I (Abstr.) 2 , pp. 99-1 Ih 29. Kombian, S. E., Warenycia, M . W., 19. Llvanb, ('. 19h7. `llic toxicity o f hydro- blele, F. C., Reiffenstein, R. J. 1988. gcii wlphide xiid otlicr sulphiiles. Q. J . Effects o f acute intoxicahi with hydro- E y . ~J/f),.\iOl5. L:'B 1-48 gcii sulfide on central amino acid 20. Savoliriiicii, t i . , Tenhuneii, It., Elo- transniitter systems. NeicroTu,vicology v ~ a ,E., `rossavainen, A. 1980. 9587-96 Cuii~ulativebiochemical effects o f re- 30. Warenycia, bl. W., Smith, K. A., peated c1inic;il hydrogen sulphide in- Blashko, C. S., Konibian, S. B., Reif- toxication in iiiousc brain. ftti. Arch, fenstein, I<. J . 1989. Monoamine ox- O C C f l / ) . li/fb~itY~tlt.IL'LfIlI1 46:87-92 idase inhibition as a sequel of hydrogen 21. Class, I>. C. 1990. A review of the sulfide intoxic;ition: increases in brain health effects of hydiagen sulphicle catcchol;imine and 5-hydroxytrypta- ex]XJSLlL'L?. A / f t l . OL`Cfl/). ffyg. 34:323- mine. Arch. Tuxicol. 63: I3 1-36 27 31. Maneckjee, A. 1985. Hydrogrti sir/- 22. Chotlwin, L. I<., Francom, D., Dieken, pliide tosicit):: it1 virro effects UIZ F. I > . , Taylor, 1. D., Warenycia, M . W . , cliolitiergic tiervoiis system cutnputzetifs. et al. 1989. I)eteriiiination of sulfide i n i ~ ~ . ~ t ~ l ~ ~ / l orlr /i~ rtl ~nircosti~tii~c ~ret -~ ~ e brain tissue by gas dialysisiioii ceptor .sires. M S c thesis. Antigonish, I ~lir~)niatogral,hyp: ost-mortem studies NS: St. Francis Xavier Univ. 158 pp. ? a i d two case reports. J . A n d . ?b.ricd. 32. Warenycia, M . W., Coodwin, L. R . , 13:105-9 Franconi, D. M . , Dieken, F. P., Koin- I 23. Goodwin, L. R . , Francom, D., Wareny- biaii, S. B., Reiffenstein, R. J . 1990. cia, h l . W. 1990. Hydrogen sulfide Dithiothreitol liberates non-acid labile analysis in biosiiniples by gas-perincable sulfide from brain tissue of HIS- D niciiibrane and ion chroiiiatograpliy. See poisoned animals. Arch. Toxicol. Ref. 2, pp. 29-40 64:650-55 24. McAnalley, B. H., Lowry, W. T., Oli- 33. Warenycia, M. W., Kombian, S. B., ver, It., Carriott, J . C . 1979. De- Reiffenstein, It. J . 1990. Stress-induced termination of inorganic sulfide and increases in brainstem amino acid levels cyanide i n blood using specific ion elec- are prevented by chronic sodium hydro- trodes: application to the investigation of sulfide treatment. NeiiroTo.rico/ogy 11 : Iiytlrogeii sulfide and cyanide poisoning. 93-98 J . Atltrl. Tu.I-~L'3o:~1 .I 1-14 33. Reiffenstein, R. J., Warenycia, bl. W. 24a , Savage, J . C . , Could, D. H. 1990. 1987. The effects of chronic exposure to Dcterniination of sulfide in brain tibsue hydrogen sulphide on miino acid con- I. 1 t 132 REIFFENSTEIN, HULBERT, RC ROTH tent of the brainstem. Br. J. Pharrnacol. 91:315P (Abstr.) 35. Hudson, L. P., Reiffenstein, R. J., Warenycia, M. W. 1987. A method for quick and simple construction of pushpull cannulae and their use in recovery of hippocampal amino acids. J. Pharrnacol. Methods 1899-93 36. Baldelli, R. J. 1990. Eflects of sulfide on rat hippocampal neurons. MSc thesis. Univ. Alberta, Edmonton. 139 pp. 37. Baldelli, R. J., Reiffenstein, R. J., Colmers, W. F. 1989. Hydrogen sulfide inhibits rat hippocampal CA 1 pyramidal cells and synaptic potentials in vitro: mechanism of toxicity? Soc. Nerrrosci. Abstr. 151348 (Abstr.) 38. Baldelli, R. J., Reiffenstein, R. J., Colmers, W. F. 1990. Hydrogen sulfide increases potassium conductance in CA I pyramidal cells in rat hippocampus. Proc. West. Pharmacol. Soc. 33:272 (Abstr.) 39. Kombian, S. B., Reiffenstein, K. J., Colmers, W. F. 1991. Effect of hydrogen sulfide on membrane and synaptic conductances in dorsal raphe in vitro. SOC. Neurosci. Abstr. 17:1464 40. Kombian, S. B., Reiffenstein, R. J. 1990. Alteration of hippocampal CAI release of amino acid neurotransmitters following poisoning with NaHS. Proc. West. Phartnacol. SOC. 33:279 (Abstr.) 41. Reiffenstein, R. J., Kombian, S. 9. 1990. Hydrogen suphide alters amino acid neurotransmitter release in the caudate-putamen. Proc. Can. Fed. Biol. SOC. 33:49 (Abstr.) 42. Kombian, S. B., Warenycia, M. W., Reiffenstein, R. J. 1989. The effect of hydrogen sulfide on amino acid neurotransmitter release pattern in rat brainstem reticular nucleus. SOC. Neurosci. Abstr. 15:1347 (Abstr.) 43. Kombian, S. 9.. Zidichouski, J., Kehoe, M., Reiffenstein, R. J. 1988. An electrophysiological study of hydrogen sulfide toxicity in frog sympathetic ganglion. SOC.Neurosci. Abstr. 14:1083 (Abstr.) 44. Reiffenstein, R. J. 1990. Neurochemical and neurophysiological effects of high dose hydrogen sulphide. See Ref. 2, pp. 117-38 45. Weight, F. F. 1983. Synaptic mechanisms in amphibian sympathetic gan- 527 pp. 46. Smith, P.A. 1984. Does the electrogen- ic Na+/K+ pump play a role in the neuronal effects of catecholamines? Trends Pharmacol. Sci. 10:422-25 47. Roth, S. H., Mathison, [ I . , Hayden, I,. J . , Hannah. R. S. 1989. The effects of low concentrations of hydrogen sulphide on the discharge activity of an isolated sensory neuron. Prnc. Cari. Fed. Biol. Soc. 32: 118 (Abstr.) 48. Roth, S. H., Hannah, R . S. 1990. The neurophysiological effects of low concentrations of hydrogen sulphide on the developing and mature nervous systems. See Ref. 2, pp. 139-56 49. Beck, J. F., Donini, J. C . , Maneckjee, A. 1983. The influence of sulfitle and cyanide on axonal function. To.ricology 26:37-45 50. Warenycia, M. W., Steele, J. A , , Kar- pinski, E., Reiffenskin, R. J. 1989. fly- drogen sulfide in cornhination with taurine or cysteic acid reversihly aholishes sodium currents in neuroblastoma cells. NeuroToxicology 10:191-200 51. Deleted in proof 52. [,link, R . R. 1988. The intrinsic elcc- trophysiological propertics of ni:iiiini:ilian neurons: insights into ccntral ncrvous system function. Sriertre 242: 1654-64 53. Watson. S., Abbott, A . 1991. Receptor nomenclature supplement. Tt.crrd.si'lzor- VUZCO/.S C ~ .12(Suppl.):3 1-33 54. Deleted in proof 55. Williams, J. T., Colmers, W. F., Pan, Z. Z. 1988. Voltage- and ligandactivated inwardly rcctifying currents in dorsal raphe neurons in vitro. J. Neuro- sci. 8:3499-3506 56. Hansen. A. J., Hounsgaard, I . , Jahnsen. H. 1982. Anoxia increases potasciuni conductance in hippocampal nerve cells. Acto Plt),.~iolS. c a d . I15:301-10 57. Fujiwara, N., Higashi, H . , Shinio.ji, K., Yoshimura, M. 1987. Effects o f hypoxia on rat hippocampal neurones in vitro. J. Plysiol. 384: I3 1-5 I 58. Leblond, J., Krnjevif, K. 1989. Hypox- ic changes in hippocampal neurons. J. Neurophysiol. 62: 1-14 59. Kmjevif, K . , Xu, Y. Z. 1989. Dan- trolene suppresses the hyperpolarization or outward current ohservetl (luring anoxia in hippocampal neiirons. (`on. ./. Physiol. Pharnmcol. 67: 1602-4 60. KmjeviC, K., Xu, Y. 1990. Mechanisms underlying anoxic hyperpolarization of hippocampal neurons. Con. J. Physiol. Pharnmcol. 68: 1609-I 3 61. Man, S. F. P., Hulbert, W. C. 1988. Airway repair and adaptation to inhalation injury. In Lung Biology in Hrcrlth and Disease, ed. L. J . Lenfant, 34: 1-49. New York: Dekker. 565 pp. 62. Rogers, R. E., Ferin, J . 1981. Effect of hydrogen sulfide on bacterial inactiva- HYDROGEN SULFIDE 133 tion i n the rat lung. Arch. E/iviro/t. thesis. Univ. Alberta, Edmonton, 150 Meiil!ti 36:261-64 PP. 63. Jiippinen, P., Vilkka, V., Marttila, O., 76. Lopez, A., Prior, hi., Yong, S., Albas- Haahtela, T. 1990. Exposure to hydro- S a m , bl., Lillie, L. E. 1987. Biochemic- gen sulphide and respiratory function. al and cytologic alterations in the res- B r . J . I t i d . h l e d . 37:824-28 piratory tract of rats exposed for 3 hours 64. Ferris, B. G . 1978. Health effects of to hydrogen sulfide. Futitkurtr. A p p l . exposure to low levels of regulated air l'o.rico1. 91753-62 pollutants: a critical review. J . Air Pul- 77. Lopez, A., Prior, M. G . , Reiffenstein, l l i t . C`otitt.01 ASSOL`.28:482-96 R . J . , Goodwin, L. R. 1989. Peracute 65. Ilulbcrt, W. C . , Prior, M .G . , Pieroni, pneumotoxicity of hydrogen sulfide P . , Florence, 2. 1989. Hyperresponsive- and sodium hydrosulfide on the lungs of riess iii rats after 5 weeks exposure to rats. Fiitidrittr. Appl. Toxicol. 12367- hydrogen wlptiide. Cliti. Itrvesr. M e d . 73 12(Suppl. 0): D89 (Abstr.) 78. Prior, bl. C . , Sharma, A. K . , Yong, S., 66. lfulbcrt, W. C., Prior, M . , Pieroni, P. Lopez, A. 1988. Concentration-time in- 19x9, Eflect of I , I O and 100ppni l?S teractions in hydrogen sulfide toxicity in on baseline airway function and bron- rats. Co/i. J . Vet. Res. 52375-79 chial reactivity in Fisclier F-344 ruts. 79. Green, I;. H. Y . , Prior, M. G . , De- FASEB J . 3:AI 190 (Abstr.) Sanctis, G . , Balu, A , , Lopez, A., Kelly, 67. Yarig, Z. J . , Hulbert, W. C. 1990. W., Yee, D., Fick, G. 1989. blediation Acute hydrogen sulfide exposure in- of hydrogen sulphide-induced pulmo- creases airway reactivity to inhaled nary injury by tachykinins. Proc. Can. niethacholine in guinea pigs. Phrrrtnu- Fed. Biol. SOC. 31:l 17 (Abstr.) colugisr 32: I58 (Abstr.) 80. Burilyak, !., Vasil'eva, I. 1974. Anti- i 6X. Hulbert, W . C., Yang, Z. J . , Trebil- mitotic and cytogenetic activity of car- cock, M. K . , Donelon, J., Boyd, W. A. bon bisulphide and hydrogen sulphide in 1990. Effect of hydrogen sulphide on the sniall concentrations. Cytol. Ceaer. pulnionary system. See Ref. 2, pp. 177- (USSR) 8:24-27 (In Russian) 203 81. Saillenfait, A. M., Bonnet, P., de- 69. 1-lulbert, W. C., Prior, M. C . , Pierone, Ceaurriz, J. 1989. Effects of inhalation i'., llciiry, l'., Kelly, W. 1989. Effect 0 1 exposure to carbon disulfide and its cxpostire to Iiydrogen sulpliide on pul- conhination with hydrogen sulfide on nionary function in rats. Proc. Cuii. embryonal and fetal development in rats. Fed. Biol. SOL..32: 116 (Abstr.) Tuxicol. Le!!. 48:57-66 70. Florence, Z.,tiulbert, W., Prior, M . , 82. Hayden, L. J., Goeden, H., Roth, S. H. Pieroni, 1'. 1989. Statistical analysis of 1990. Growth and development during pulinonary reactivity i n rats exposed to sub-chronic exposure to low levels of hydrogen sulphide. Pruc. C m . Fed. hydrogen sulfide. Toxicol. Itid. Hcdtlz lliol. Soc. 32: I16 (Abstr.) 6:389-401 71. Lioy, P. L., Lippniann, M., eds. 1988. 83. Hayden, L. J . , Goeden, H., Roth, S. H. 7'o.v-i~Air Pullittcitir Guirieliiies: Re\ietv 1990. Exposure to low levels of hydro- ut'Rec.e/ir Progt-ess wid Problettis. Cin- gen sulfide elevates circulating glucose cinnati: Ani. Cotif. Cov. Ind. tiyg. 67 in maternal rats. J . Tuxicol. E/rviron. 11'' ' 72. Lopez, A , , Prior, M . , Yong, S., Lillie, We1rltlr 3 1 :27-34 84. Deleted in proof I-., Lefebvre, M. 1988. Nasal lesions 85. Hayden, L. J., Franklin, K. H., Roth, S. i n rats exposed to hydrogen sulfide for H., Moore, C . J . 1989. luhibition of l'our hours. hi.J . Vet. Res. 39: I 107- oxytocin-induced but not angiotensin- I I induced rilt uterine contractions follow- 73. Ng, W., Toiuetich, 1. 1984. Effect o f ing exposure to sodium sulfide. Lijk Sei. hytlrogen sulfide arid methyl nicrcaptan 45:7557-60 on pernieability of oral niucosa. J . D ~ I I I . 86. Calli, C. L., Manso, L., Spencer, 1'. S., Res. 63:994-97 eds. 1988. R~cetitArlvcirices in Nervous 74. Hulbert, W. C., Walker, D. C., Jack- x m , A. s., Hogg, J. C. 1981. Airway Sysietrt Toxicology. New York: Plenum. 391 pp. Ixrniciibility to horseradish peroxidxe 87. Glebova, K . 1950. Permissible tolerance iii guiiica pigs: the repair pliaae after of hydrogen sulphide in the atmosphere. injury by cigarette smoke. Atti. Rev. Re- Gig. SLUI;!4.:19-21 (In Russian) spir. Dis. 123:320-26 88. Gaitonde, U. B., Sellar, R. J. 1987. 75. Yarig, Z. J . 1991. The toxic eflects oJ Long-terni exposure to hydrogen SUI- t i d t q y t i .siiljkle e.vposiire 011 r l i c respi- phide producing subacute encephalopa- rutory sysretti 01 g i i i w L i pigs. PIID thy in a child. Br. Med. J . 294:614 134 REIFFENSTEIN, HULBERT, & ROTH 89. Hannah, R. S., Roth, S. H . 1991. Chronic exposure to low concentrations of hydrogen sulfide produces abnormal growth in developing cerebellar Purkinje cells. Neurosci. Lett. 1223225-28 90. Hannah, R. S., Hayden, L. J., Roth, S. H. 1989. Hydrogen sulfide exposure al- ters the amino acid content in developing rat CNS. Neurosci. Lett. 99:323-27 91. Hannah, R. S., Bennington, R., Roth, S. H. 1990. A relationship between hy- drogen sulfide exposure and taurine levels in maternal rats. Proc. West. Pliarmacol. SOC.33: 177-79 92. Gosselin, R. E., Hodge, H., Smith, R., Gleason, M. 1976. Hydrogen sulphide. Clinical Toxicology of Commercial Products, 111: 169-73. Baltimore, Wil- liams & Wilkins 93. Prior, M. G., Lefcbvre, M., Yee, D., Fritz, D. 1991. Objective measures of ocular irritation as a measure of hydro- gen sulfide exposure. Proc. Can. Fed. Bioi. Soc. 3 4 9 5 (Abstr.) 94. Bhambhani, Y., Singh, M. 1991. Effects of hydrogen sulphide inhalation during exercise in healthy men. J. Appl. Physiol. In press 95. Ammann, H. M. 1986. A new look at physiologic respiratory response to H2S poisoning. J. Hazard. Mater. I3:369-74 96. Deleted in proof 97. Toxigenics, Inc. 1983. 90-day vapor in- halation toxicity study of fide in Fischer 344 ratsl. h. y.d.roSgpernagsiiircl-- Dawley rats1 . . . [and] B6C3Fl mice. CIIT Docket no. 22063, 32063, and 42063. Research Triangle Park, NC: Chem. Ind. Inst. Toxicol. 98. Tolonen, M., Hernberg, S., Nurminen. M., Tiitola, K. 1975. A follow-up study of coronary heart disease in viscose rayon workers exposed to carbon disul- phide. Br. J. lnd. Med. 32:l-10 99. Sava, G.. Bertole, G., Giraldi, T. 1980. Effects of sodium sulphide on bile flow in the anesthetized rat. Riv. Fartnacol. Ter. 1 1:15-20 100. Tenhunen, R., Savolainen, H.. Jap- pinen, P. 1983. Changes in haem synthesis associated with occupational exposure to organic and inorganic sul- phides. Clin. Sci. 64:187-9I 101. Jappinen, P. 1990. Hydrogen sulphide analysis in non-fatal hydrogen sulphide poisoning. See Ref. 2, pp. 41-48 102. Jappinen, P., Tenhunen, R. 1990. Hy- drogen sulphide poisoning: blood sul- phide concentration and changes in haem metabolism. Br. J . Ind. Med. 471283-85 103. Wasch, H. H., Estrin, W. J.. Yip, P., Bowler, R.. Cone, J. E. 1989. Prolongation of the P-300 latency associated with hydrogen sulfide exposure. Arch Neirrol. 46:902-4 104. Deleted in proof 105. Lindell, H., Jappinen, P., Savolainen, H. 1988. Determination of sulphide in blood with an ion-selective electrode by pre-concentration of trapped sulphide in sodium hydroxide solution. Analyst 1 13:839-40 106. Warenycia, M. W., Reiffenstein, R. J., Goodwin, L. R. 1990. Forensic analysis of human fatalities: postmortem detec- tion of brain sulphide after hydrogen sulphide. See Ref. 2, pp. 49-59 107. Herbert, F. A. 1990. Treatment of hydrogen sulphide intoxication. See Ref. 2, pp. 211-15 108. Smith, R. P., Kruszyna, R., Kruszyna, H . 1976. Management of acute sulfide poisoning: effects of oxygen. thiosulfate, and nitrite. Arch. Ertvirori. Hcnlrli 31:166-69 109. Burnett, W. W., King, E. G., Grace, M., Hall, W. F. 1977. Hydrogen sulphide poisoning: review of five year's experience. Can. Med. Assoc. J . 1 17:1277-80 1 IO. Huang, C.-C., Chu, N.-S. 1987. A case of acute hydrogen sulfide (H2S) intoxication successfully treated with nitrites. J. Formosan Med. Assoc. 86: 1018-20 I l l . Andeau, F. M., Gnanaharan, C., Davey, K. 1985. Hydrogen siilphide poisoning: associated with pelt processing. N . Z . Men. J. 98:145-47 112. Whitecraft, D. D., Bailey, T. D., Hart, G. B. 1985. Hydrogen sulfide poisoning treated with hyperbaric oxygcn. J. Enierg. Med. 3:23-25 113. Smilkstein, M . J., Bronstein, A . C., Pickett, H. M.. Rumack. B. H. 1985. Hyperbaric oxygen therapy for severe hydrogen sulfide poisoning. J. Enierg. Med. 3:27-30 114. Eley, D. W., Korecky, B., Fliss, H . 1989. Dithiothreitol restores contractile function to oxidant-injured cardiac niiiscle. Ani. J. Pliysiol. 257:H1321-25