Document bxO4o56GJYJ439vgRmk3J2K3

a~ ~ Centrilobular Hepatic Necrosis and Acute Renal Failure in "Solvent Sniffers" RICHARD D. BAERG, M.D., and DANIEL V. KIMBERG, M.D., New York, New York; and Boston, Massachusetts Three teenagers with a history of drug abuse developed acute hepatic injury after the inhalation of Carbona cleaning fluid, a commercial preparation that contains trichloroethylene. Two of these patients had evidence of acute renal injury, presumably due to tubular necrosis. Electrocardiographic and electroencephalographic abnormalities were also noted. Hepatic biopsies in two patients showed acute centrilobular necrosis. Both biopsies also showed evidence of mild centrilobular fibrosis, probably a result of previous injury from exposure to trichloroethylene, Carbona "sniffing" seems to be increasingly popular among adolescents, and its use in a group that has abused other drugs complicates an already difficult situation. Trichloroethylene is a chlorinated hydrocarbon (CHC1=CC1;) with wide industrial and limited medical application. Its low volatility, noninflam mability, and poor water solubility make it a useful solvent for degreasing metals, dry cleaning, and for fat extraction. It is also used in paints, lacquers, adhesives, and in silicone parting agents. Medically its use (Trilene, Great Britain; Trethylene, United States) is currently limited to anesthesia and anal gesia. particularly in obstetrics, where it has been used as a 0.5% mixture with air. The agent has a history of a variety of other medical uses, including the treatment of angina pectoris, trigeminal neuralgia, and emotional disorders and as a wound cleanser and an antihelminthic. These uses have been aban doned as ineffective. Ph^rm Ihe ft Medicine. Columbia University College of chj.icuns and Surgeon's, and the Department of Medicine. Columbia- ercbMcrian Medical Comer, New York, N Y.; and the Department nf ,c iv1"'' Harvard Medical School, and the Gastrointestinal Unit ot the Department of Medicine, Beth Israel Hospital, Boston, Mast. Annalt of Internal Medicina 73:713-720, 1970 In addition to these uses, Carbona cleaning fluid (currently a solution of 44% tr chloroethylene and 56% by volume petroleum distillates) and Carbona Number 10 Special Spot Remover (currently con taining 40% trichloroethylene, 10% 1,1,1-trichloroethane, and 50% by volume petroleum distillates) have recently been popular among "solvent sniffers." The acute and chronic toxic effects of trichloroethyl ene have been incompletely appreciated in "solvent sniffers" in this country (1), despite many reports on the hazards of industrial inhalation. Recent re views provide a more complete understanding of the subject (2, 3). Two patients with acute hepatic and renal damage and another with centrilobuar hepatic necrosis attrib utable to Carbona cleaning fluid* "sniffing" are de scribed. Case Reports CASE 1 Patient T.M., a 15-year-old white boy, was admitted to Presbyterian Hospital, New York, on 18 September 1968 because of icterus. Except for drug abuse, he had been in good health. He had had regular exposure to a variety of model airplane glues, amphetamines, heroin, cocaine, and marijuana. He denied the parenteral ad ministration of drugs, and he did not consume alcoholic beverages. He had "sniffed" Carbona cleaning fluid on one occasion 10 months before admission, after which he noted severe headache. In the month before ad mission he "sniffed" glue occasionally but used none of the other agents. Six days before admission Carbona cleaning fluid was inhaled, causing drowsiness. This was repeated the following day. after which headache, an orexia. and vomiting appeared. The next day the fumes were again inhaled, and later the same day he noted a sore throat, fever, and chills. The following day, 3 days before admission, he was jaundiced and had dark urine. * In Cases 1 and 2 the patients inhaled an older solution of Carbona which contained 70% trichloroethylene and 30% by volume petroleum distillates. 713 SL 035331 Two days later, on 17 September 1968, the patient was seen at the Columbia-Presbyterian Medical Center where icterus was confirmed. On admission to the hos pital the following day, the patient's vital signs were normal. No rash was present. The liver was slightly tender, and its edge was palpable 3 cm below the right costal margin; the spleen tip could also be palpated. The rest of the physical examination, including the neuro logical examination, was within normal limits. Pertinent laboratory tests of hepatic function are summarized in Table 1. Other significant laboratory tests on admission included a hemoglobin of 14.2 g/100 ml; WBC count, 6,160/mm3 with 50% neutrophils, 33% lymphocytes, 8% monocytes, 8% eosinophils, and 1% basophils. A subsequent WBC and differential count on the twelfth hospital day was entirely normal, with 1% eosinophils. A platelet count was 260,000/mm3, and the erythrocyte sedimentation rate (Wintrobe method) was 26 mm in 1 hr. The urinalysis on admission showed a specific gravity of 1.005, pH 6.0, no protein or glu cose, and a negative microscopic examination. Urinalysis on the fifth hospital day showed a specific gravity of 1.025, pH 5.0, 1-1- protein, no glucose, one to two WBC/hpf, and occasional finely granular casts. No ad ditional urinalyses were recorded. Blood urea nitrogen (BUN) was 15 mg/100 ml. Serum albumin was 4.4 g/100 ml; and serum globulin, 2.6 g/100 ml. An elec trocardiogram on admission showed T-wave inversion in V2_,, and T-wave lowering in V4_g; subsequent electro cardiograms were normal. The patient became asymptomatic shortly after ad mission. Tests of hepatic function were initially abnormal but rapidly returned to normal (Table 1). Serum glu tamic-oxalacetic transaminase (SGOT) was normal by the sixth hospital day, whereas serum glutamic-pyruvic transaminase (SGFT) returned to normal more slowly. Six days after admission a percutaneous liver biopsy was performed with a Menghini needle. Microscopic examination of the tissue (Figures 1 and 2) showed ex tensive hepatocellular necrosis with collapse of the reticulin framework, largely restricted to central areas. In these areas hepatocytes were notably absent, and there was preservation of the sinusoidal architecture. There was a moderate mononuclear cell infiltrate in the central areas. Mild centrilobular fibrosis, confirmed by trichrome stains, was present in a few regions. The portal areas were entirely normal, and the lobular archi tectural pattern was not disturbed. Biopsy showed acute toxic centrilobular necrosis superimposed upon a pre vious centrilobular injury. Figure 1, Biopsy of the liver of Patient 1, 12 days after the onset of the most recent episode of Carbona "sniffing" and 6 days after admission to the hospital. See text for description. (Hematoxylin and eosin; original magnification, X 25.2, X 2.5.) The patient had an uneventful recovery' and was dis charged on the thirteenth hospital day. The only medica tion given was a standard multivitamin preparation. Two weeks after the patient was discharged his SGOT, SGPT, and alkaline phosphatase were normal. He con tinues to use marijuana but has avoided other agents. case 2 Patient D.S., a 15-year-old white girl, was admitted to Presbyterian Hospital on 8 August 1967 because of anuria of 3 days' duration. Her pertinent medical history included an uncomplicated tonsillectomy and adenoid- Table l. Hepatic Function Tests in Patient 1 Hospital Day Serum GlutarmcOxalacetic Transaminase* Serum Glutamic- Pyruvic Transaminase* Alkaline Phosphatase t -1 3,800 3,100 34 2 200 890 28 6 50 245 -- 7 40 200 -- 13 33 70 15 * Modified Karmen units, normal, 10 to 40 units. t Kiag-Armstroag units; normal, 10 to 15 units. 714 November 1970 Annal* of Internal Medicine Volume 73 Number 5 Bilirubin Direct Total mg 10G ml 1.07 0.43 -- 0.19 0.15 3.12 1.18 -- 0.64 0.31 Prothrombin Time Patient Control sec 16 14 ___ -- 13 ___ 13 ___ -- SL 035332 ectomy with cyclopropane anesthesia at age 5 years. Urinalysis at that time showed a specific gravity of 1.024, pH 5.5, and two to three WBC/hpf. A routine urinalysis 3 years before the present illness was also normal. Eighteen months before admission the patient began "sniffing" glue to the point of inebriation two to five times weekly, a practice that was discontinued 6 weeks before admission. At that time she began to ingest various amphetamines and barbiturates but did not con sume alcoholic beverages. Five weeks before admission she intentionally lacerated her left wrist with a razor blade. Approximately 1 week later she began to "sniff" Carbona cleaning fluid every other day, and 1 week before admission she began to inhale it nightly, despite tearing, rhinorrhea, and a bifrontal headache. Three days before admission, nausea, vomiting, and right up per quadrant pain appeared and persisted; the pain be came worse with movement. She also noted the absence of urine. The day before admission she had alternating chills and fever. At admission her temperature was 100 F, rectally; pulse, 120/min; and blood pressure, 114/60 mm Hg. No rash, petechiae, or icterus was evident. The pharynx was erythematous, but no exudate was observed. Bi lateral costovertebral angle tenderness on percussion and right upper quadrant tenderness on palpation were noted. The liver, spleen, and kidneys were not palpable. The rest of the physical examination, including the neurological examination, was normal. Laboratory studies on admission were as follows; hemoglobin, 14.3 g/100 ml; WBC count, 13,200/mm3 with 84% neutrophils, 11% lymphocytes, 3% mono cytes, and 2% eosinophils. The erythrocyte sedimenta tion rate (Wintrobe method) was 6 mm in 1 hr. Urinaly sis showed a specific gravity of 1.010, pH 5.0, 3 + protein, O-trace glucose, two to four WBC/hpf, and many RBCs. The serum albumin was 4.3 g/100 ml and serum globulin, 2.1 g/100 ml. An ECG and tests for blood sugar and serum calcium were normal. Alkaline phos phatase was 11 King-Armstrong units. Serial determina tions of other tests of hepatic and renal function are summarized in Table 2. The clinical impression was that the patient had acute tubular necrosis and toxic hepatitis, both secondary to Carbona "sniffing." She was treated with bed rest, a low protein diet (20 g/day), sodium and fluid restric tion, and parenteral vitamin K for correction of the abnormal prothrombin time. As shown in Table 2, the SGOT, bilirubin, and prothrombin time returned to Figure 2. Biopsy of the liver of Patient 1. See text for description. (Laidlaw reticulm stain; original magnification. X 64. X 2.5.) normal within her first 9 days in the hospital. On the twentieth hospital day there was 7% retention of sulfobromophthalein (BSP) 30 min after intravenous injection of 5 mg/kg body weight. Oliguria with a daily urine output in the range of 250 to 328 ml persisted until the fifth hospital day, at which time the urine volume rose to 585 ml/24 hr and the BUN had reached 87 mg/100 ml. Urine volume con tinued to rise and reached a peak of 2,630 ml/24 hr on the eleventh hospital day. Thereafter, the volume de creased to normal, and the patient was discharged on the twentieth hospital day with a BUN of IS mg- 100 ml. A urinalysis on the fourth hospital day showed a specific gravity of 1.008, pH 8.0, 2-i- protein, no glucose, five Table 2. Hepatic and Renal Function Tests in Patient 2 Hospital Day Serum GlutamicOxalacetic Transaminase* Serum Glutamic- Pyruvic Transaminase* 1 2,300 -- 7 29 90 10 25 43 16 45 20 20 -- -- * Modified Karmen units, normal. 10 to 40 units, Bilirubin Direct Total mg 100 ml 0.7 2.1 -- 1.0 ___ -- ___ ___ ---- Prothrombin Time Patient Control sec 24 15 15 -- 13 14 14 - Blood Urea Nitrogen mg 100 ml 33 78 87 24 18 Baers and Kimberg Solvent Sniffers 715 SL 035333 to seven RBCs/hpf< and many WBCs. No subsequent urinalyses were recorded. Despite psychotherapy since discharge, she has con tinued to use marijuana, phenothiazines, barbiturates, and heroin. When seen 12 months after discharge she had shared needles with jaundiced addicts and was noted to be icteric and to have tender hepatomegaly. Hospital ization at that time was declined by the patient. No further follow-up information is available. 3case Patient S.H., an 18-year-old white man. was admitted to Presbyterian Hospital on 16 March 1970 because of of abdominal pain of 5 days' duration. He was examined at age 12 because of a behavioral disorder and was said to have a passive-aggressive personality. A urinalysis, urine culture, and intravenous pyelogram were done at age 13 because of enuresis, and all of these tests were normal. He began "sniffing" Carbona cleaning fluid at age 15, using approximately 4 oz daily for several days in suc cession. Drowsiness and nightmares were the only symptoms noted. In August 1969 he smoked marijuana, ingested barbiturates, and injected LSD and heroin in travenously. In October 1969 he inhaled 4 oz of Carbona cleaning fluid daily for 2 weeks, stopped for several days, and then began using up to 8 oz daily. He acci dentally ingested an unknown quantity of this agent and was treated with gastric lavage at another hospital. He noted no symptoms other than drowsiness after the in gestion. In November 1969 Carbona cleaning fluid was again "sniffed" daily, but this was discontinued shortly before an admission to Presbyterian Hospital on 16 November 1969 for removal of pilonidal cysts. The only symptom related to trichloroethylene toxicity at that time was vague abdominal pain during the week be fore admission. Physical examination in November 1969 showed a blood pressure of 146/90 mm Hg and multiple pilonidal cysts and sinuses. The examination was otherwise nor mal. Laboratory tests included a hematocrit of 50% and a WBC count of T.lOO/mm3 with 4% eosinophils. A urinalysis was normal, and the BUN was 16 mg/100 ml. Tests of hepatic function were abnormal. The total bilirubin was 1.2 mg/100 ml; SGOT was 670 I.U. (normal, 10 to 50); lactic dehydrogenase, 270 I.U. (normal, 90 to 200); alkaline phosphatase, 80 I.U. (nor mal, 30 to 85); serum albumin. 4.1 g 1D0 ml; and globulin. 3.0 g. 100 ml. On the third hospital day the patient's temperature rose to 101 F. surgery vvas post poned, and he was subsequently discharged from the hospital to be followed in the medical clinic. After discharge from the hospital he refrained from using all parenteral agents but was lost to medical fol low-up until his second admission. Between 6 March and 10 March 1970 the patient "sniffed" 4 oz of Car bona cleaning fluid daily. On 7 March he ingested one capsule containing an amphetamine. On 11 March. 5 days before admission, he noted malaise and dull, steady, midabdominal pain. The pain persisted, and the following day it was accompanied by nausea and vomit ing. On 13 March he was seen at the Columbiu-Presbyterian Medical Center Emergency Room, where his physical examination was normal. Hemoglobin. WBC count, and differential count were also normal. Urinaly sis showed a specific gravity of 1.015. 2-- protein, and one to two RBCs 'hpf. No WBCs or casts were seen. He was sent home but returned 3 days later, still com plaining of nausea, and was admitted to the hospital. He denied headache, urinary symptoms, fever or chills. Physical examination showed a blood pressure of 150/100 mm Hg. Other vital signs were normal. The sclerae were icteric; the liver, kidneys, and spleen were not palpable, but there was mild tenderness on per cussion of the right upper quadrant of the abdomen. The neurological examination was normal. Laboratory tests showed: hematocrit, 4-3C&: WBC count. 4,100/mm3 with a normal differential: the initial urinalysis showed a specific gravity of 1.020, no pro':in and one to three WBCs. four to five RBCs. and occasional granular casts/hpf. Subsequent urinalyses showed a specific gravity of 1.017 to 1,020 and up to 20 RBCs and 8 WBCs/hpf. Hyaline and finely and coarsely granular casts were variably present, and 2 + proteinuria reappeared transiently on the tenth hospital day. The urinalysis was entirely normal on the fifteenth hospital day. Serum albumin was 4.2 g/100 ml and globulin, 2.6 g/100 ml. The prothrombin time was 13 sec (control. 12 sec). When tested in the laboratory of Dr. David Gocke, the serum did not contain the hepatitis antigen. The results of other tests of renal and hepatic function are summarized in Table 3. An ECG and serum calcium and phosphorus were normal. An elec troencephalogram done on the fourth hospital day was Table 3. Hepatic and Renal Function Tests In Patient 3 Hospital Day Serum GlutamicOxalacetic Transaminase* Serum Glutamic- Pyruvic Transaminase* Lactic Alkaline Dehydrogenase* Phosphatase f -3 6.120 -- 4,620 2 460 -- 70 4 142 848 ___ 8 63 374 -- 12 50 168 ___ 15 34 74 ___ 17 29 58 * International Units; normal. 10 to JO units t International Units: normal. 90 to 200i units. t International Units; normal, 30 to 85 uruts. 716 November 1970 Annals of Internal Medicine Volume 73 90 69 74 85 52 -- 46 Number J Bilirubin --------------------- " Direct Total Blood Urea Nitrogen mg 100 ml- 4.00 -- 2.00 0.62 1.39 0.41 0.93 0.39 0.77 0.33 0.59 0.36 0.61 19 37 26 21 24 19 21 Serum Creatinine -- 1.6 1.4 1.2 1.2 1.1 1.2 SL 035334 normal at rest, but bursts of medium to high voltage ji, to 3/sec waves were seen bifrontally during hyper ventilation. A liver biopsy with a Menghim needle was also done on the fourth hospital day. Microscopic ex amination of the tissue (Figures 3 and 4) showed ex tensive centrilohular necrosis with collapse of the reticulin framework A nuld polymorphonuclear leukocyte inflammatory reaction tvas associated with the necrotic areas. The sinusoidal architecture was not disturbed. Although a moderate amount of fibrous tissue was seen in manv central areas, the portal areas were quite normal. The biopsy was interpreted as showing acute toxic hepatitis with an older, healed centrilohular lesion, probably the result of previous haiogenated hydrocarbon injury. The patient was treated only with bed rest and had an uneventful recovery . On the seventeenth hospital day the BUN was 21 mg 100 ml, and the serum creatinine had fallen to 1.2 mg/100 ml. On the eighteenth hospital day the BSP retention was 7% 30 min after the injection of 5 mg kg body weight, and the patient was discharged from the hospital. Discussion These patients are presented in detail to illustrate the clinical presentation of individuals with acute Figure 3. Biopsy of the liver of Patient 3. 14 days after the onset of the most recent episode of Carbona "sniffing" and 4 days after admission to the hospital. See text for description. (Mallory trichrome stain; original magnification, >, 25.2, X 2,5.) Figure 4. Biopsy of the liver of Patient 3. See text for description. (Heamtoxyhn and eosin stain; original magnification, X 64, x 2.5.) trichloroethylene inhalation toxicity and to alert physicians who have contact with "solvent sniffers" to the possible toxic effects of this agent. Experimental inhalation of trichloroethylene re sults ir. an estimated 51 to 70% retention of the agent by the lungs (4, 5). The exact fate of the retained trichloroethylene is unclear (3), but ap proximately 70% (4) is metabolized to a variety of substances that are recoverable in the urine. These include trichloroacetic acid (10 to 36%) (4, 5), trichloroethanol (32 to 59% ) (4, 5). and monochloroacetic acid (4%) (4). Small amounts of trichloro acetic acid and trichloroerhanol are also excreted in sweat, salha, and feces (5). The remaining tri chloroethylene is excreted unchanged by the lungs within 48 hr. and only small amounts of the unaltered agent are found in the urine (5). It is still unclear whether trichloroethylene per se or one of its metab olites is responsible for toxicity. Evidence favors trichloroethanol as the most toxic agent, particularly for the central nervous and cardiac conduction sys tems (6), although monochioroacetic acid has been implicated because of its ability to inhibit various enzyme systems (5). Baerg and K/mberg Solvent Sniffar* 717 SL 035335 * A discussion of the toxicity of trichloroethylene inhalation is properly separated into two exposure situations: [1] chronic industrial exposure to low concentrations and [2] acute or recurrent massive inhalation, as occurs in industrial accidents, tri chloroethylene anesthesia, and exposure incurred by "solvent sniffing.'' Chronic exposure in metal de greasers has been reported to result primarily in a neuropsychiatric symptom complex consisting of tremors, giddiness, alcohol intolerance, a neurasthenic syndrome with anxiety states, and bradycardia, as well as increased lacrimation, reddening of the skin, decreased sensitivity of the hands, and insomnia (7). In rats, prolonged trichloroethylene inhalation pro duces neuromuscular depression and possibly a re duction in fear inhibition (8). None of the recent industrial surveys have con firmed the presence of retrobulbar neuritis described earlier (3), although anesthesia of trigeminal nerve endings has been noted (7). Despite this report, the use, so popular at the beginning of the century', of trichloroethylene in the treatment of trigeminal neu ralgia has largely been abandoned. This is in part ow ing to the variable clinical response, particularly in patients with chronic symptoms, and to studies that showed that the effects of trichloroethylene inhalation were a nonspecific result of general anesthesia or sedation (9). Anosmia with presumed olfactory nerve damage has also been reported (10). Anorexia and other functional gastrointestinal complaints have been considered to be manifestations of chronic toxicity; although the cause is unclear, the gastro intestinal disturbances have been attributed to auto nomic nervous system dysfunction (11). Other manifestations of chronic industrial tri chloroethylene inhalation have been even less well defined and are described primarily in case reports. Severe hepatic damage is decidedly unusual, but fatal hepatic necrosis has been reported after chronic in dustrial exposure (12). Abnormalities of hepatic function have been noted by various workers (3), and it has been shown that elevations of serum glutamic-oxalacetic transaminase (SCOT) and aldol ase may be potentiated by pretreatment with ethanol (13). This effect of ethanol has been confirmed ex perimentally in rats (14). Chronic exposure of a variety of experimental animals to trichloroethylene inhalation has produced neither significant dysfunc tion nor abnormal hepatic histology (15;. The important acute effects of massive trichloro ethylene inhalation are usually those related to central nervous system depression. The cardinal manifesta tion of such exposure is a depressed level of con sciousness that may be preceded by either mild 718 Nov*mbf isn AnnaJi of Internal Medicma Vo/uitia 73 excitation or euphoria. The usual postrecovery symp tom is headache (16). a symptom experienced by two of the patients in this report. Organic cerebral and cerebellar dysfunction has also been noted (7). The electroencephalographic changes occurring after industrial accidents (17) have also been noted in guinea pigs given trichloroethylene and trichloroethanol intraperitoneally (6). It has been found that trichloroethanol is three times more effective than trichloroethylene in reducing cerebral cortical elec trical excitability and in causing both electroencepha lographic and electrocardiographic abnormalities (6). Another potential hazard of acute trichloroethylene inhalation is the occurrence of cardiac arrhythmias. Atrial flutter with atrioventricular block, nodal rhythm, multifocal premature ventricular contractions, ventricular tachycardia (18), and ventricular fibrilla tion (19) have been seen during trichloroethylene anesthesia or analgesia in humans. Orth and Gillespie (20) noted significant changes in rhythm or rate in 13 of 14 patients anesthetized with this agent, and others (21) have noted'similar abnormalities in 33 of 40 patients, with a 10% frequency of multifocal ventricular tachycardia. Ventricular fibrillation has been considered the cause of death in cases of in dustrial inhalation (22. 23) and may have played a role in the deaths of three "sniffers" (24). Respira tory insufficiency with tachypnea, perhaps secondary to an augmented sensitivity of pulmonary stretch receptors (3), may also have contributed to these fatalities. The transient electrocardiographic ab normality consisting of precordial T-wave inversion in Case 1 has been noted previously (25). Renal damage, such as that noted in Cases 2 and 3, is more unusual than hepatic toxicity. Postanes thesia urinalyses in patients who had received trichlo roethylene have not indicated the presence of renal injury (2, 3); however, one of Cotter's (16) patients with acute massive inhalation had a urinalysis sug gestive of acute tubular necrosis. Furthermore, Gutch, Tomhave, and Stevens (25) reported a pa tient with acute tubular degeneration documented by renal biopsy. Acute renal failure requiring dialysis has also been seen by Litt and Cohen (1) in "solvent sniffers." The effects of experimental inhalation in animals on renal and hepatic function have been variable, perhaps reflecting substantial differences in the species used and in the experimental technique. Al though trichloroethylene inhalation has been reported to cause toxic nephritis in rodents after acute ex posure (2), no histologic evidence of renal damage has been seen in a variety of animals used in other studies (15). Urea clearance is normal in dogs sub- Numbir 5 SL 035336 jected to repeated trichloroethylene anesthesia (20). Fatal hepatic necrosis has been reported in associa tion with massive industrial exposure (26) and has also been noted after trichloroethylene anesthesia (2, 3, 27-30). In the latter instance the association has. often been obscured by severe intercurrent ill ness. Under usual anesthetic circumstances tests of liver function are either unchanged or are only mildly and transiently abnormal (2). Although acute ex posure in dogs affected neither sulfobromophthalein retention nor hepatic histology' (2, 3), subacute ex posure caused pronounced abnormalities in both aspects (20). Abnormal tests of hepatic function (1) or abnor mal hepatic histology (10) have been seen only rarely in "solvent sniffers." All three patients de scribed in the present report had SGOT and serum glutamic-pyruvic transaminase values and a clinical history consistent with an acute toxic hepatitis. In Cases 1 and 3 the liver biopsy demonstrated centrilobular necrosis, a finding characteristic of chlorinated hydrocarbon hepatotoxicity. The presence of fibrous tissue in the centrilobular zones of the biopsies in both cases suggests that repeated exposure may pro duce permanent hepatic alterations, but the func tional significance, if any, of such lesions remains obscure. The predominantly centrilobular location of the lesion in the patients with trichloroethylene and other chlorinated hydrocarbon toxicity may reflect alterations in the intralobular circulation (31), dif fering susceptibility of cells in different lobular areas, or zonal differences in the intracellular concentration of the hepatotoxins (32). Although little is known concerning the biochemi cal effects of trichloroethylene or its metabolites on hepatic cell metabolism, there has been a great deal of interest in elucidating the mechanism underlying the toxic action of carbon tetrachloride. It is not unlikely, however, that the hepatotoxic effects of these related chlorinated hydrocarbons are mediated by a similar chain of events. Several mechanisms have been con sidered to account for the effects of carbon tetra chloride, but precise definition of the initiating bio chemical event remains uncertain. This has been the subject of a recent extensive review (32). Carbon tetrachloride causes profound alterations in the mitochondria of hepatic cells, with evidence of swelling, altered permeability, an increase in calcium content, a loss of nicotinamide adenine dinucleotidc (NAD), a decrease in the activity of NAD-linked tricarboxylic acid cycle dehydrogenases, a fall in adenosine triphosphate levels, and uncoupling of oxidative phosphorylation. It seems unlikely, how ever, that mitochondrial injury represents the initial 035337 st* event. Recent studies with carbon tetrachloride have suggested that there is an earlier effect on the endo plasmic reticulum, leading to inhibition of protein synthesis, impaired lipoprotein synthesis, lipid ac cumulation, and ultimate cell necrosis. This damage to the endoplasmic reticulum may be due to carbontetrachloride-related peroxidation of structural lipids (32, 33). Other recent studies (34) have suggested that carbon-tetrachloride-induced lysosomal rupture leading to an early disruption of protein synthesis may be the initiating event in the hepatic injury. Unfortunately, the therapy for trichloroethylene inhalation excess has been less well evaluated than the toxic manifestations (3). Measures such as with drawal of the agent, avoidance of ethanol, and the management of acute hepatic or renal failure, or both, seem obvious. The use of tetraethylthiuram disulfide (disulfiram) has been suggested, especially in those who have ingested trichloroethylene (35). This com pound appears to inhibit the formation of trichloro acetic acid and trichloroethanol from trichloroethy lene, and its administration is associated with an en hancement of the pulmonary excretion of trichloro ethylene, The use of such agents is probably best deferred, however, until more evidence is available concerning the identity of the hepatotoxic and nephro toxic metabolite. While antioxidants such as alphatocopherol may protect against carbon tetrachloride damage in experimental animals (32), the clinical applicability, if any, of this observation is unknown. Trichloroethylene has a sweet smell, and its in halation may produce a sense of elation. It has been suggested that addiction to the inhalation of this agent may occur (3). Carbona "sniffing" seems to be increasingly popular among adolescents, and its use in a group that has abused other drugs can compli cate an already difficult situation. Unless the ex posure is recognized, a mistaken diagnosis of viral hepatitis may be made. A history of "solvent sniffing," the acute onset of symptoms, the possible accompani ment by evidence of renal damage, the absence of the so-called hepatitis antigen, and the relatively rapid resolution of the signs of hepatic injury should help distinguish toxic hepatitis from that owing to viral infection. The complete absence of portal in flammation and the centrilobular location of the damage seen on biopsy of these patients leaves little doubt that they had toxic rather than viral hepatitis. The presence of centrilobular scarring in both biopsy specimens suggests that repeated exposure may lead to permanent hepatic alterations, but the functional significance of such lesions is unknown. ACKNOWLEDGMENTS: Received 4 June 1970; accepted 2J June 1970. Baarg and Klmbarg Solvent Sniffer* 719  Requests for reprints should be addressed to Daniel V. Kimberg, M.D., Beth Israel Hospital, 330 Brookline Ave., Boston, Mass. 02215 References 1. Lm IF, Cohen MI: "Danger . . . Vapor harmful": spotremover sniffing. Yew Eng ] Med 281:543-344, 1969 2. Defalque RJ: Pharmacology and toxicology of trichloro ethylene. A critical review of the world literature. Clin Pharmacol Ther 2:665-688, 1961 3. Smith GF: Trichlorethylene: a review. Brit I Industr Med 23:249-262, 1966 4. Sou&k B, VlachovX D: Excretion of trichloroethylene metabolites in human urine. Brit J Industr Med 17:60-64, 1960 5. BartonISek V: Metabolism and excretion of trichloro ethylene after inhalation by human subjects. Brit J Industr Med 19:134-141, 1962 6. MikiskovI H, Mhuska A: Trichloroethanol in trichloro ethylene poisoning. Brit J Industr Med 23:116-124, 1966 7. BakdodIj Z, VysKoJfit J: The problem of trichloroethylene in occupational medicine. Arch Irtd Health (Chicago) 13: 581-592, 1960 8. Battio K, Geandjean E: Chronic effects of trichloroethylene on rat behavior. Arch Environ Health (Chicago) 7:694^699, 1963 9. Rubinstein HS, Painter E, Harne OG: The neural de pressing effect of trichloroethylene. J Lab Clin Med 24:12381241, 1939 10. James WRL: Fatal addiction to trichloroethylene. Brit J Industr Med 20:47^9, 1963 11. Grandjean E, Munchinoer R, Turrian V, et al: Investiga tions into the effects of exposure to trichloroethylene in mechanical engineering. Brit J Industr Med 12:131-142, 1955 12. Priest RJ, Horn RC: Trichloroethylene intoxication--a case of acute hepatic necrosis possibly due to this agent. Arch Environ Health 11:361-365, 1965 13. Lachnit V, Pietschmann H: Activity of serum glutamicoxaloacetic-transaminase and aldolase in workers exposed to halogenated hydrocarbons. Industr Med Surg 29:523-525, 1960 14. Cornish HH, Adepuin J: Ethanol potentiation of halo- genated aliphatic solvent toxicity. Amer Industr Hyg Ass J 27:57-61, 1966 15. Prendkrgast JA, Jones RA, Jenkins LJ Jr, et al: Effects on experimental animals of long-term inhalation of trichloro ethylene, carbon tetrachloride, 1,1,1-trichloroethane, dichlorodiffuoromethane, and 1,1-dichloroethylene. Toxicol Appt Pharmacol 10:270-289, 1967 16. Cotter LH: Trichloroethvlene poisoning. Arch Industr Hyg (Chicago) 1:319.322, 1950 17. Chalupa B, Synkova J, SEV?ht M: The assessment of elec- troencephalographic changes and memory disturbances in acute intoxications with industrial poisons. Brit J Industr Med 17:238-241, 1960 18. Waters RM, Orth OS, Gillespie DM: Trichlorethylene anesthesia and cardiac rhythm. Anesthesiology 41-5, 1943 19. Bernstine ML- Cardiac arrest occurring under trichloro ethylene analgesia--report of a case with recovery. Arch Surg (Chicago) 68 262-266, 1954 20. Orth OS, Gillespie NA- A further study of trichloroethy lene anaesthesia. Blit J Anaesth 19:161-173. 1945 21. Barnes CG, Ives J: Electrocardiographic changes during Trilene anaesthesia. Proc Roy Soc Med 37:528-532, 1944 22. Bell A: Death from trichloroethylene in a dry-cleaning establishment. .Yen Zeal Med J 50:119-126, 1951 23. Kleinfeld M, Tabershaw IR: Trichloroethylene toxicity-- report of five fatal cases. Arch Industr Hyg (Chicago) 10: 134-141, 1954 24. Bonnichsen R, Maehly AC: Poisoning by volatile com pounds, II. Chlorinated aliphatic hydrocarbons. 1 Forens Sci 11:414-427, 1966 25. Gutch CF, Tomhave WG, Stevens SC: Acute renal failure due to inhalation of trichloroethylene. Ann Intern Med 63: 128-134, 1965 26. Joron GE, Cameron DG, Halpenny GW: Massive necrosis of the liver due to trichloroethylene. Canad Med Ass J 73: 890-891, 1955 27. Annotation in Lancet: Liver necrosis after Trilene anaesthesia. Lancet 1:476, 1944 28. Dodds GH: Necrosis of the liver and bilateral massive suprarenal haemorrhage m the puerperium. Brit Med J 1: 769-770, 1945 29. Annotation in Lancet: Dangers of Trilene anaesthesia. Lancet 1:379-380, 1944 30. Herdman KN: Acute yellow necrosis of the liver following Trilene anaesthesia. Brit Med J 1.689-690, 1945 31. Glynn LE, Himsworth HP. The intralobular circulation in acute liver injury by carbon tetrachloride. Clin Sci 6:235241, 1948 32. Recknagel RO: Carbon tetrachloride hepatotoxicity. Phar macol Rev 19:145-208, 1967 33. Hashimoto S, Glende EA, Recknagel RO: Hepatic lipid peroxidation in acute fatal human carbon tetrachloride poisoning. New Eng ] Med 279:1082-1085, 1968 34. Alpers DH, Isselbacher KJ: The effect of carbon tetra chloride on rat-liver lysosomes. Biochim Biophys Acta 137: 33-41, 1967 35. BartonIJek V, Teisinger J: Effect of tetraethyl thiuram disulphide (Disulfiram) on metabolism of trichloroethylene in man. Brit J Industr Med 19:216-221, J962 720 November 1970 Annals oflntsrnal MsOlcins * Volume 73 Numbers st 035338