이 논문은 2001년 상반기에 국제 독극물 학술지(Toxicology Letter)에 발표될 예정입니다
Acute Paraquat Intoxication in Korea
Kyu-Yoon Hwang, MD, DrPH1, 3 Dong-Ho Yang, MD, Ph.D2, 3 Sae-Yong Hong, MD, Ph.D2, 3 *
1 Department of Preventive Medicine, School of Medicine, Soonchunhyang University 2 Department of Internal Medicine, School of Medicine, Soonchunhyang University 3 Institute of Pesticide Poisoning, Soonchunhyang Chunan Hospital, Republic of Korea
* Address correspondence and reprint requests to Dr. Sae-Yong Hong, Department of Internal Medicine, Soonchunhyang Hospital, 23-20, Bongmyung-Dong, Chunan, Choongnam, 330-100, Republic of Korea Tel: 82-417-570-2121 Fax: 82-417-574-5762 E-mail: syhong@sparc.schch.ac.kr
Abstract Objectives: This study was conducted to describe acute paraquat intoxication and to determine potential factors related to the paraquat fatality. Methods: During 1999, 154 patients with paraquat intoxication were admitted to the Institute of Pesticide Poisoning (IPP), Soonchunhyang, Chunan Hospital. Trained physicians conducted standardized medical emergency procedures according to the Treatment Guideline for Paraquat Intoxication. Exposure to paraquat was assessed by amount of ingested paraquat and a qualitative measure of urine paraquat. Outcomes of paraquat intoxication were determined as recovery and death. Multiple logistic regression was applied to assess the predictors of paraquat fatality. Results: Among all patients, 139 (90.3%) were transferred from other medical facilities to the IPP after mean (SD) 20.1 (2.6) hours of paraquat poisoning. The mean (SD) age was 42.9 (14.6) years old. Intentional ingestion of paraquat contributed 73.4% (113 of 154 patients) of paraquat poisonings and had significantly a higher fatality rate (53.2%) than did accidental ingestion (19.1%) (P < 0.001). Overall paraquat fatality was 43.8%. In multiple logistic regression, the risk of fatality increased significantly with the quantity of ingested paraquat and a positive urinary paraquat test. Age, sex, and time interval between intoxication and medical treatment were not significantly related to fatalities. Conclusion: Our results supported the notion that paraquat is potentially lethal in humans with the risk of fatality directly related to ingested and absorbed.

Keywords: Paraquat, Fatality, Intoxication Introduction Accidental or intentional poisoning by exposure to a noxious substance is the major cause of morbidity and mortality in Korea1. Although accidental mortality by exposure to substance decreased from 7.3 to 2.3/100,000 during 1988 -1997, suicide increased from 8.5 to 14.1/100,000. Pesticides are the second mode (19.9%) of suicide in Korea. Several studies have reported 12.3% - 28.3% of pesticide poisoning among agricultural workers and a possible relation between pesticides and overall mortality of Korean farmers 2-4. Although pesticide poisoning in agricultural communities is due to improper management and the lack of personal protective equipment, approximately 85% of acute pesticide intoxication in Korea is due to suicidal intent 4. Especially, paraquat (1,1-dimethyl-4-4-bipyridium dichloride), Gramoxonea, a wide used, as potent herbicide is notorious as an agent of suicide. Paraquat has been used for 3 decades in Korea causing an estimated 2,000 intoxications annually with 40 - 50% mortality 5. The annual estimated incidence of fatal paraquat poisoning is 20 deaths / million population. Much lower rates have been reported in developed countries such as Japan (11.0 / million population), Ireland (2.7 / million population), Great Britain (0.66 / million population), and the United States (0.004 / million population) 6. This is an analysis of the 154 cases of paraquat intoxication admitted to the Soonchunhyang University Chunan Hospital during 1999 to determine the potential factors predicting fatality. Methods and Materials Study Subjects: One hundred seventy five patients poisoned by pesticides were admitted to the Institute of Pesticide Poisoning (IPP), Soonchunhyang University Chunan Hospital from January through December 1999. Of these, 154 (88.8%) patients were intoxicated by paraquat with 139 (90.3%) of paraquat subjects transferred from other medical facilities. Five patients were occupationally intoxicated by paraquat aerosol. All paraquat poisoned patients were admitted to the intensive care unit for further evaluation and treatment. Data Collection: Trained physicians treated patients and recorded all information on the standardized data collection forms. All data were reviewed by two pesticide specialists. Standardized medical emergency procedures were conducted according to the Treatment Guideline for Paraquat Intoxication by IPP (Table 1), a specialized institute for pesticide poisoning in Korea. Briefly, gastric lavage was performed on all subjects seen within 2 hours after ingestion and 100 grams Fuller\'s earth in 200 ml of 20% mannitol was given if intoxication had occurred within 12 hours. Emergency hemoperfusion was performed until urinary paraquat test was negative. All these procedures were conducted with permission and informed consent. Urinary paraquat was measured by Dithionite method 7. Study Variables: Demographic variables such as age, sex, address, social security number, and occupation were enumerated. Exposure to paraquat was assessed by the amount of ingested and urine paraquat test. Time intervals between intoxication and first medical treatment and between first medical treatment and IPP were assessed. The causes of intoxication were divided into three categories; accidental, intentional, and inhalation exposure during agricultural work. Alcoholic intoxication and average alcohol intakes were examined. Health status and past medical history were collected. Recorded all data included vital signs, blood cell counts, liver and renal function, blood gas analyses, and chest X-ray. Outcome was defined as recovery or death. Statistical Analysis: Data analysis was performed with the Stata program (Stata Release 5.0 College Station, Texas). Exploratory data analysis checked distribution of values, expressed as mean and standard deviation for numerical data and proportion (%) for nominal data. Differences in covariates between survivors and deaths were tested by Chi-square test and Odds Ratios were calculated to assess the relations between covariates and outcomes. Multiple logistic analysis was applied to determine potential predictors of outcomes, controlling for possible confounders. Results The mean (SD) age of paraquat intoxication patients was 42.9 (14.6) years old. Females were more frequent than males (57.1% vs 42.9%). Farmers and housewives accounted for 90 of 154 patients (58.5%). Among 54 farmers, 5 (9.3%) were intoxicated by paraquat spray and 17 (31.5%) and 32 (59.3%) ingested paraquat product, accidentally and intentionally, respectively. 88.9% (32 of 36) housewives used paraquat for suicidal purpose. No significant seasonal variation was found. Paraquat was used for suicidal purpose in 72.7% (112/154) and 58.9% (66/112) drank alcohol before ingestion of paraquat. Overall 53.3% drank alcohol before the paraquat intoxication (Table 2). The diagnosis of paraquat poisoning was based on history and clinical features. No confirmed assessment was made for paraquat poisoning in the 5 inhalation patients. Their urine paraquat tests were all negative. All five farmers sprayed paraquat aerosols for 2 - 3 hours without appropriate personal protective equipment and all developed non-specific general symptoms such as gastrointestinal and headache. 90.3% (139 of 154) patients were arrived at the IPP with a mean of 20 hours after their first medical visits. The amount ingested was estimated by history in 148 of 149 patients. 73.6% (109 of 148) patients ingested more than a mouthful of paraquat (about 15 - 20 ml). Suicidal intent involved on average of 3 mouthfuls, while accidental cases involved 1 mouthful (P < 0.05). Mean (SD) interval time to first medical facility was 9.8 (1.9) hours, but 70.1% (108 of 154) visited a medical facility within 3 hours after intoxication. At the IPP, urinary paraquat was detected in 61.7% (95 of 154). Fuller\'s earth was administered to and emergency hemoperfusion were conducted on 88 (57.8%) and 91 (58.9%) patients, respectively (Table 3). Vital signs and initial laboratory data are summarized in Table 4. 28 of 154 (18.2%) patients had high blood pressure (systolic > 160 mmHg or diastolic > 95 mmHg). Increased WBC counts (above 10,800/mm3) were found in 74 (48.1%). 25 (16.2 %) and 29 (18.8%) patients showed low hemoglobin (< 13 g/dl for male, < 12 g/dl for female) and hematocrit (< 38 % for male, < 37% for female), respectively. Abnormal liver function (both GOT and GPT > 40 IU) was found in 51 (33.1%) patients and bilirubin was elevated (> 1.2 mg/dl) in 83 (53.9%) patients. 34 (22.1%) and 54 (35.1%) patients showed increased BUN (> 22 mg/dl) and creatinine (> 1.2 mg/dl). Acidosis (pH < 7.35) was found in 21 (13.6%). Hypoxia was observed in 35 (22.7%) patients. Abnormal urinary excretion of WBC (> 3/HPF), RBC (> 5/HPF), and protein (> trace) were 22 (15.1%), 48 (32.9%), and 35 (24.0%) patients, respectively. Paraquat fatality increased with the estimated amount ingested (Table 5). Swallowing more than 2 mouthfuls of paraquat was associated with an 82.3% (48 of 55) fatality with 58.29 Odds Ratio (OR) (P < 0.001), compared to patients said to have swallowed less than 1 mouthful. The fatality was significantly higher with 20.9 OR (P<0.0001) for patients with positive urinary paraquat. Suicidal intent resulted in significantly more fatalities (53.2%, 59 of 111) than accidental ingestion (19.1%, 8 of 41). Alcoholic intoxication and emergency treatment intervals were unrelated to fatality (P > 0.05). The overall paraquat fatality rate was 43.8%. In multiple logistic analysis (Table 6), selected study variables were assessed as predictors of paraquat fatality. After controlling for age, sex, and suicide intent, the estimated amount of ingested paraquat remained as a significant predictor of the fatality in the model 1, accounting for 37.3% of the variance. Urinary paraquat excretion significantly related to fatality, accounting for 29.6% of the variance in model 2. These relations were not significantly changed by the addition of other study variables to the models (data not shown). Discussion Although there has been a worldwide interest in paraquat poisoning since the first report in 1966 8, the current study is the first series of paraquat poisonings reported from Korea. In Korea, most commercial paraquat has a high concentration (24.5% w/v) of paraquat dichloride. Paraquat has been publicized as a suicidal agent and continues to be in a large number of suicidal attempts 4. The 154 cases reported here may be the largest a 1-year experience of any hospital. Our paraquat fatality rate in Korea (43.8%) was slightly lower than Yoshika\'s report (60%) in Japan 9. Despite the estimated amounts, 24.5% paraquat ingested, our fatality was relatively lower than previous reports. The case mix of suicidal and accidental may not be comparable, but small amounts even as less than a mouthful were sometimes fatal in our series as in others 10 . The annual production of paraquat in Korea is approximately 800,000 kg and the use of paraquat as Gramoxonea for farm and garden continues to increase 11. During 1988 - 1992, 1,006 acute drug intoxication patients were admitted to our hospital, which serves both suburban and rural communities. Paraquat accounted for 14.2% of poisonings, but 84.5% of paraquat ingestions were of suicidal intent 11 , consistent with data from other countries 6, 12, 13. Proudfoot (1979) suggested that plasma paraquat concentration is a useful measure of assessment of prognosis 14. Since our previous study reported that a qualitative urine paraquat test indicates absorption and correlates well with plasma paraquat 15, we employed urine paraquat to assess paraquat exposure and define treatment. In this study, urine paraquat was shown to be a significant prognostic index. There was a significant dose-response relation between estimated paraquat exposure and positive urine paraquat with fatality. Suicidal intent was significantly associated with fatality in crude analysis, but disappeared after adjusting for the ingested amount. Age was not significantly associated with paraquat fatality. Ethanol consumption, time interval to medical treatment, and other possible explanatory variables did not significantly modify fatality as predicted by multiple logistic analysis. This is probably a consequence of potent lethality of paraquat. It was concluded that paraquat fatality is primarily linked to the amount ingested and absorbed. References 1. Annual Report on the Cause of Death Statistics: Based on Vital Registration, National Statistical Office, Republic of Korea, 1997. 2. Lim HS, Zong MS. A Survey on the Damage done to the Farmers by Agrochemicals in Rural Area of Korea. Korean J of Preventive Medicine 1892;15(1):205-209. 3. Oh HC, Nam CM, Lee SH. A Cohort Study on the Relationship between Pesticide Use and Mortality, and Cancer Mortality. Korean J of Preventive Medicine 1991;24(3): 390-399. 4. Hong SY. Guidebook of Pesticide Toxicology. Seoul: Korea Medicine, 1998. 5. Hong SY, Hwang KY, Yang DH, Cho SR, Han CS, Park YH, Chang SK. Effect of vitamin C on Plasma Total Antioxidant Status in Patients with Paraquat Intoxication. (in preparation). 6. Onyon LJ, Volans GN. The Epidemiology and Prevention of Paraquat Poisoning. Human Toxicology 1987;6(1):19-29. 7. Goulding R, Volans GN, Crome P, Widdop B. Paraquat Poisoning. Bri Med J 1976;1:42. 8. Bullivant CM. Accidental Poisoning by Paraquat. Br Med J 1966;I:1271-1273. 9. Yoshioka T, Sugimoto T, Kinoshita N, Shimazu T, Hiraide A, Kuwagata Y. Effects of Concentration Reduction and Partial Replacement of Paraquat by Diquat on Human Toxicity: A Clinical Survey. Human & Environmental Toxicology 1992;11:241-245. 10. Bismuth C, Garnier R, Dally S, Fournier P. Prognosis and Treatment of Paraquat Poisoning: A Review of 28 Cases. Journal of Toxicology - Clinical Toxicology 1982;19(5):461-474. 11. Lee JS, Jeong MK, Kim TJ, Kim JB, Paek JK, Yang DH, Hong SY. Clinical Observation of Paraquat Poisoning. Korean J Medicine 1994;47(1):93-100. 12. Tan CT. Suicidal Poisoning Deaths in Singapore 1975-1984. Annals of the Academy of Medicine (Singapore) 1987;16:300-302. 13. Tinoco R, Tinoco R, Parsonnet J, Halperin D. Paraquat Poisoning in Southern Mexico: A Report of 25 cases. Archives of Environmental Health 1993;48(2):78-80. 14. Proudfoot AT, Stewart MS, Levitt T, Widdop B. Paraquat Poisoning: Significance of Plasma-paraquat Concentrations. The Lancet 1979;ii: 330-332. 15. Hong SY, Yang DH, Sabapathy NN. Significance of Plasma Paraquat Concentration in Paraquat Poisoning. Korean J Medicine 1995;48(4):480-485. Table 1. Summary of treatment guideline for paraquat intoxication depending on the time lag after paraquat ingestion Emergency Treatment Protocol 1. Gastric lavage, if within 2 h of ingestion 2. Urine test for paraquat (6 h interval until negative) 3. Fuller\'s earth 100 gm in 20 ml of 20% mannitol every 4 hr for 12hr 4. InjectionL-glutathione 50 mgVitamin C 3,000 mgMethyl prednisolone 500 mgLasix 100 mg15% mannitol 250 ml 4. Emergency continuous arteriovenous hemofiltration if admitted within 12 h after ingestion or urine test for paraquat is positive 5. LaboratoryCBCUrine analysisLiver function test, electrolytesRenal function testX-rayArterial blood gas analysisECG 6. Decreased PO2: Re-breathing in a closed tent 7. Superoxide dismutase inhalation by nebulizer
Table 2. General characteristics of 154 cases of paraquat intoxication, Korea, 1999 No. of Subjects % Age (year)< 2020 - 2939 - 3940 - 4950 - 59> 60Mean ± SD 6302933312542.9 ± 14.6 3.919.518.821.420.116.2 SexMaleFemale 6688 42.957.1 Occupation Farmer Housewife Unemployed Self-employed Office worker Student Employee Unknown 5436191311966 35.123.412.38.47.15.83.93.9 Intoxication Intentional ingestion Non-intentional ingestion Inhalation* 113365 73.423.43.2 Season Spring Summer Fall Winter 31552840 20.135.718.226.0 Alcohol Past History Non-drinker Regular drinker 7579 48.751.3 Ethanol Intoxication No Yes 7282 46.853.3 Total 154 100 * Five farm workers were intoxicated by paraquat spray Table 3. Emergency medical evaluation of paraquat intoxication cases No. of Subjects % Amount (mouthful) (n = 148)*< 11 - 2> 2Mean ± SD (ml)95% confidence interval 38545554.2 ± 8.637.2 - 71.2 25.736.537.2 Time interval to medical facility (h)< 11 - 23 - 11> 12 Mean ± SD95% confidence interval 575118289.8 ± 1.96.3 - 13.6 37.033.111.718.2 Transfer interval to IPP (h) (n = 139)< 33 - 56 - 1112 - 2324 - 47 > 48 TotalMean ± SD**95% confidence interval 39301720103813920.1 ± 2.614.9 - 25.3 17.321.612.29.47.227.3100 Urinary paraquat excretionNegative - trace++++++++++ 5920141348 38.313.09.18.431.2 Fuller\'s earth treatment No Yes 6588 42.257.8 Emergency hemoperfusion No Yes 6391 40.958.9 * Five inhalation exposed patients and one ingested patients without known amount were not included. ** An extreme outlier was excluded (n = 138) Table 4. Initial vital signs and laboratory results Mean (SD) 95% CI Vital signs Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Pulse (bpm) Respirations /min 130.5 (25.1)83.0 (14.9)82.3 (14.8)22.2 (6.9) 126.5 - 134.580.6 - 85.379.9 - 84.621.1 - 23.3 Hematologic index WBC /mm3 Hemoglobin (g/dl) Hematocrit (%) 13135 (621)14.0 (0.1)42.1 (0.4) 11909 - 1436213.7 - 14.341.2 - 43.0 Liver function GOT (IU) GPT (IU) Bilirubin (mmol/L) 58.1 (8.4)47.9 (5.8)28.6 (3.2) 41.4 - 74.836.5 - 59.422.2 - 34.9 Renal function BUN (mmol/L) Creatinine (mmol/L) 6.8 (0.5)137.9 (12.4) 5.9 - 7.8113.2 - 160.9 Arterial blood gas analysis pH PCO2 (mmHg) PO2 (mmHg) 7.42 (0.01)33.6 (0.6)94.0 (1.8) 7.41 - 7.4332.3 - 34.890.5 - 97.6 Urine analysis WBC, No. (%) RBC, No. (%) Proteinuria, No. (%) 22 (15.1)48 (32.9)35 (24.0)
Table 5. Fatality and Odds ratios of acute paraquat intoxication in Korea, 1999 No. of subjects No of expired (%) Odds Ratio 95% CI Amount of paraquat (mouthful) *< 11 - 2> 2 385455 4 (10.6)14 (25.9)48 (82.3) -2.9858.29- 0.89 - 9.8915.81 - 214.87 Urine paraquatNegativePositive 5994 5 (8.5)62 (63.0) -20.93 7.61 - 57.49 Suicidal intent NoYes 42111 8 (19.1)59 (53.2) -4.82 2.05 - 11.35 SexFemaleMale 6687 28 (42.4)39 (44.8) -1.10 0.58 - 2.10 Ethanol intoxicationNoYes 7182 31 (43.7)36 (43.9) -1.01 0.53 - 1.92 Time interval to medical facility (h)< 11 - 2> 3 565146 28 (50.0)27 (52.9)12 (26.1) -1.132.00 0.53 - 2.400.66 - 6.08 Transfer interval to IPP (h)< 66 - 24> 24 693737 33 (47.8)13 (35.1)19 (51.4) -0.530.66 0.23 - 1.990.31 - 1.40 First treatment at IPPNoYes 13815 63 (45.7)4 (26.7) -0.43 0.13 - 1.43 Total** 153 67 (43.8) * Estimated amount ingested, 147 patients ** One case not able followed to outcome. IPP: Institute of Pesticide Poisoning Table 6. Multiple logistic analysis of exposure to paraquat on paraquat fatality in Korea, 1999 Covariates Odds ratio 95% CI P-value Model r2 Model 1: Age (year) Sex (0 = female, 1 = male) Suicidal intent (0 = No, 1 = Yes) Amount of ingested paraquat (0 = below 1, 1 = 1 - 2 mouthfuls) (0 = below 1, 2 = above 2 mouthfuls) 0.991.661.823.1554.47 0.96 - 1.020.66 - 4.200.59 - 5.580.93 - 10.6914.13 - 210.06 0.490.280.290.07<0.01 37.3 Model 2: Age (year) Sex (0 = female, 1 = male) Suicidal intent ( 0 = No, 1 = Yes)Urine paraquat (0 = Negative, 1 = Positive) 0.992.043.4918.98 0.97 - 1.020.87 - 4.771.22 - 9.996.63 - 54.32 0.690.100.02<0.01 29.6 Only patients who ingested paraquat (n = 147) were included in the models.