The values of different scoring systems in predicting the prognosis of patients with paraquat poisoning
Lianjing Liang, Shuyun Xu*, Zhen Jiang and Shiyuan Tang
Department of Emergency, West China Hospital of Sichuan University, Chengdu, Sichuan, PR China
- *Corresponding Author:
- Shuyun Xu
Department of Emergency
West China Hospital of Sichuan University, PR China
Accepted on June 6, 2017
This study aims to explore the values of different scoring systems in predicting the prognosis of patients with paraquat poisoning, and find out the most clinical application value of indicators. The patients with paraquat poisoning were included and treated in the emergency department from January 2016 to October 2016, every patient was underwent different scoring system, and the Receiver Operating Characteristic (ROC) curve was used to calculate the optimal cut-off value of different scoring system, in order to calculate the sensitivity and specificity of different scoring system to compare the value of different scoring system in predicting the prognosis. The COX mode was used to evaluate the scoring system's relationship with the prognosis of patients with paraquat poisoning. A total of 107 cases were included, and there were 37 cases of death. Every scoring system to predict the value of a poor prognosis in patients with paraquat poisoning was not the same. The sensitivity and specificity of lactic combined with Severity Index of Paraquat Poisoning (SIPP) were 0.900 and 0.741 respectively; it was superior to other scoring system. The results of COX model found that the Risk Ratio (RR) value of lactic combined with SIPP was 12.098 (95% CI: 2.473, 59.178), P value was 0.002, and it was superior to other scoring system. Every scoring system can effectively judge the severe degree of the patients with paraquat poisoning and effective predict the prognosis of patients, and the lactic combined with SIPP is superior to other scoring system.
The scoring system, Paraquat poisoning, Prognosis, Value.
Paraquat is a high effective non-contact herbicide. It can take effects faster after spraying, rapidly deactivate into the soil, and it has no residues in the soil. Paraquat has very strong toxicity for human and animals, and it has been common reason for pesticide kill events due to drug misusage or taken oneself that cause acute poisoning, especially in developing countries. In China, there are not official statistical data, but many reports have shown that paraquat poisoning is one of the most common pesticide poisoning. In the emergency department, it has been the second after organophosphorus pesticide poisoning, and is the first in the absolute death in pesticide poisoning [1,2].
Paraquat is the most used defoliant and herbicides in the world wide, and oral lethal dose is 5-15 ml. Paraquat can access human body via cutaneous absorption, respiratory inhalation and alimentary canal, and it has obvious local stimulation and corrosive effect. Oral absorption rate of paraquat is 5%-15%, and it can rapidly achieve blood drug concentration peak after it enters into the bloodstream and it does not bind to plasma protein. Plasma concentration of paraquat is gradually reduced after 15-20 h, and it is excluded from the kidneys. Paraquat poisoning is common in emergency department in China, and its poisoning can involve in many visceral organs, and even lead to multiple organ dysfunction syndrome. The paraquat poisoning mechanism remains unclear. Many studies believe that generated a great amount of oxygen radical can cause lipid peroxidation of cell membranes in tissues after absorption of paraquat, and the phenomenon leads to multiple organ dysfunction, especially for injury of lungs that can cause pulmonary edema ad pulmonary fibrosis. In early stage, acute lung injury or acute respiratory distress syndrome can be detected, and alveoli and pulmonary interstitial fibrosis would be detected in later stage that is the cause of the patient’s death. The fatality rate of paraquat poisoning is near 50-70% [3,4]. Many studies have shown that paraquat poisoning can significantly increase expression of some cell factors and inflammatory factors, such as Tumor Necrosis Factor-alpha (TNF-alpha), Interleukin 6 (IL-6) and Interleukin 10 (IL-10), but inhibit antioxidant factors, such as Superoxide Dismutase (SOD), which further accelerate occurrence and development of pulmonary fibrosis [5-9]. Paraquat lethal dose is low, and paraquat poisoning can cause a very high case fatality rate. However, effective and specific treatment methods lack in clinic, no effective detoxification medicine is found and many treatment methods are explored, and all of these bring great challenges for emergency medical workers. In America,paraquat has been the first pesticide type with absolute number of pesticide poisoning death , and paraquat has been avoided to use in Europe in 2007 due to intense toxicity and high fatality rate [11-13].
Early accurate evaluation of prognosis can contribute to adopt timely and effective therapeutic measure to reduce mortality [14,15]. Currently, there are many indexes to predict prognosis of patients with paraquat poisoning in clinic, but different indexes have various clinical application values. This study aims to discuss the potential application values of different scoring systems, and to search the potential index with the most clinical application value.
Materials and Methods
Patients with paraquat poisoning were collected in emergency department in our hospital from January 2016 to October 2016. The Early Warning Score (EWS) is a simple physiological scoring system suitable for bedside application. The ability of a modified Early Warning Score (MEWS) to identify medical patients at risk of catastrophic deterioration in a busy clinical area was investigated. The Acute Physiology and Chronic Health Evaluation II (APACHE II) score is a scoring system based on 12 parameters of acute physiology and chronic health. This scoring system has been commonly used for the assessment of the severity and prognosis of diseases, especially in intensive care units. Data of indexes used in APACHE II, EWS and MEWS scoring systems were complete when patients were hospitalized, including heart rate, systolic pressure, respiratory rate, body temperature, and conscious state. Data used in SIPP score were complete, including course of disease (unit: hour) and concentration of paraquat (unit: mg/L); concentration of blood lactic acid (unit: mmol/L). The inclusion criteria were: 1) patients had obvious clinical diagnosis of paraquat poisoning based on history of poison and clinical features, and patients accorded with diagnostic criteria of modern diagnosis and treatment of acute poisoning; 2) ages of patients were more than 14 y old; 3) patients were first confirmed cases in our hospital, and they had not history of gastric lavage, hemodialysis, and special treatment; 4) patients were excluded if they had other kinds of drug poisoning, they died within 24 h after hospitalized, or they had a history of severe liver and brain diseases. This study was conducted in accordance with the declaration of Helsinki. This study was conducted with approval from the Ethics Committee of West China Hospital of Sichuan University. Written informed consent was obtained from all participants.
Selection of parameters
Different indexes used in scoring systems were collected, including heart rate, blood pressure, respiratory rate, body temperature, conscious state, prognostic outcomes (death or survival), course of disease (h), concentration of paraquat (mg/L) and concentration of blood lactic acid (mmol/L). Died patients were recorded, discharged patient were telephone follow-up, and mortality was used to as prognostic indicator in 60 d after poisoning.
ROC curve was used to the optimal cut-off value in estimate every scoring system (including lactic acid, SIPP, acute physiology and chronic health evaluation II (APACHE II), EWS, MEWS, lactic acid+SIPP, lactic acid+APACHEII, lactic acid+EWS, lactic acid+MEWS, lactic acid+APACHEII+SIPP +EWS, and lactic acid+APACHEII+SIPP+MEWS). Every set of data were changed to two-category data, which was used to estimate sensitivity and specificity of different scoring systems in predicting poor prognosis in patients with paraquat poisoning.
COX proportional hazard model was performed in survival analysis as semi-parameter model, and it could reflect hazard rate function at t time when response variable was i. Basis COX model was constructed, including age, sex, and conscious state. Finally, a set of variable of above mentioned 11 variables was added in COX model in turn, and relevant statistical indexes were collected.
Data analysis was performed with SAS 9.4 and SPSS 20.0 software. Measurement data were described using, enumeration data were described using rate or ratio, and P<0.05 was considered as statistical difference.
The basic characteristics of patients
A total of 107 patients were collected in this study (age: 34.36 ± 13.06). There were 37 died patients among the total 107 patients. Concentration of blood paraquat was 46.892 ± 75.328 mg/L, and course of disease was 17.71 ± 44.69 h in the death patients. One hundred patients were survival (26.56 ± 12.01), and concentration of blood paraquat was 3.363 ± 9.780 mg/L, and course of disease was 18.69 ± 26.00 h (Table 1).
|Age (year)||26.56 ± 12.01||34.36 ± 13.06|
|Lactic acid (mmol/L)||2.04 ± 1.05||7.87 ± 5.26|
|Concentration of blood paraquat (mg/L)||3.363 ± 9.780||46.892 ± 75.328|
|Course of disease (h)||18.69 ± 26.00||17.71 ± 44.69|
|Temperature (°C)||36.83 ± 0.69||37.08 ± 0.81|
|heart rate (times/min)||85.86 ± 18.09||90.14 ± 22.87|
|Rebreather (times/min)||19.97 ± 2.29||20.88 ± 3.47|
|Systolic pressure (mmHg)||118.42 ± 15.91||122.45 ± 24.52|
|Diastolic pressure (mmHg)||74.03 ± 11.99||74.18 ± 17.71|
Table 1. The basic characteristics of patients.
Comparisons of the optimal cut-off value, sensitivity and specificity
As shown in Table 2, lactic acid had significant advantage in prediction of poor prognosis in single index rating system, and the optimal cut-off value was 2.8 mmol/L, the relevant sensitivity and specificity were 0.875 and 0.690, respectively. In joint index rating system, lactic acid+SIPP had obvious advantage, and the optimal cut-off value was 0.212 mmol/L, the relevant sensitivity and specificity were 0.900 and 0.714, respectively (Table 2).
SIPP: Severity Index of Paraquat Poisoning; APACHE II: Acute Physiology and Chronic Health Evaluation II; EWS: Early Warning Score; MEWS: Modified Early Warning Score.
Table 2. The optimal cut-off value, sensitivity and specificity in Receiver Operating Characteristic (ROC) curve of each scoring system.
Relationship of scoring system and prognosis using COX model
As shown in Table 3, in single index rating system, there was a strong relationship between lactic acid and poor prognosis after avoiding effects of other factors, and the Risk Ratio (RR) values were 8.866 (95% CI: 2.462, 31.931, P=0.001) (Table 3). In joint index rating system, lactic acid+SIPP and poor prognosis had strong correlation, and the RR values were 12.098 (95% CI: 2.473, 59.178, P=0.002) (Table 3).
|Variable||B||SE||Wald||RR (95% CI)||P|
|Lactic acid||2.182||0.654||11.142||8.866 (2.462,31.931)||0.001|
|Lactic acid+SIPP||2.493||0.81||9.475||12.098 (2.473,59.178)||0.002|
|Lactic acid+APACHEII||1.31||0.521||6.321||3.705 (1.335,10.287)||0.012|
|Lactic acid+EWS||2.248||0.651||11.908||9.470 (2.641,33.954)||0.001|
|Lactic acid+MEWS||2.182||0.654||11.142||8.866 (2.462,31.931)||0.001|
|Lactic acid+APACHEII+SIPP+EWS||2.405||0.707||11.567||11.082 (2.771,44.319)||0.001|
|Lactic acid+APACHEII+SIPP+MEWS||2.405||0.707||11.567||11.082 (2.771,44.319)||0.001|
SIPP: Severity Index of Paraquat Poisoning; APACHE II: Acute Physiology and Chronic Health Evaluation II; EWS: Early Warning Score; MEWS: Modified Early Warning Score.
Table 3. Relationship of scoring system and prognosis of patients using COX model.
This study aims to discuss relationship of different scoring system and prognosis of patients with paraquat poisoning using COX model. We found that different scoring systems can effectively estimate severity levels of patients and predict prognosis of patients, and usage of lactic acid+SIPP has significant advantage than other scoring systems.
Paraquat poisoning can cause organ injuries (such as lung, kidney, liver and brain), and the characteristic change is lung damage, mainly including early damaged alveolar epithelial cells, alveolar hemorrhage, edema, inflammatory cell infiltration, and pulmonary interstitial fibrosis in later stage [16-19]. Large doses of poisoning can lead to death due to multiple organ failure, and small doses of poisoning can cause late-onset pulmonary fibrosis with poor prognosis EWS as an important evaluation tool in emergency treatment, it has been widely applied to estimate various kinds of emergency and severe cases, and its scores have close association with prognosis of patients [20-22]. Increased score needs increased medical intervention, but mortality risk also increases during hospitalization. Studies have shown that increased concentration of paraquat in blood and urine leads to higher fatality, and some researchers proposed that SIPP and urine paraquat poisoning severity index are used to predict severity of patients [23,24]. Of these, SIPP=plasma concentration of paraquat when admitted × poisoning to admission time, larger SIPP value means poorer prognosis of patients. However, it is not widely applied in clinic because detection of plasma concentration of paraquat has higher requirements. Improved early warning score is a scoring system based on blood pressure, heart rate, respiratory rate, temperature and consciousness, and it has been widely applied in emergency department and intensive care unit in developed countries due to data are easy to get and calculation is quick and convenient . Higher MEWS scores indicate heavier illness and higher fatality rate. Changes of these indexes contribute to effective intervene and treatment, which has significant clinical significance for reducing fatality rate and improving rescue success rate. Lactic acid detection and APACHEII score may be simple and practical tools to predict the prognosis of paraquat. Liang et al.  showed that the cutoff value of lactic acid was ≥ 2.75 mmol/L when patients were admitted to the hospital, sensitivity and specificity were 0.818 and 0.667, respectively, Youden index was 0.485, and sensitivity and specificity of APACHEII were 0.864 and 0.778, respectively, Youden index was 0.642. There are many indexes to predict prognosis of patients with paraquat poisoning, and these indexes have different values based on different clinical application. In order to survey the best and most clinical application value scoring system, we constructed COX model based on age, sex and conscious state to discuss predictive values of single scoring system and combined scoring system using patients in our hospital.
Acidosis is an important factor influencing the prognosis of paraquat poisoning due to paraquat can damage mitochondrial DNA and further influence relevant respiratory chain peptides. After mitochondria damage, pyruvic acid (metabolic product of glycolysis) forms lactic acid via Lactate dehydrogenase catalysing in cytoplasmic matrix, and simultaneously inhibits lactic acid oxidation and sugar dysplasia, reduces lactic acid removal rate and leads to higher level of serum lactic acid [16-18,27,28]. Arterial blood lactate has higher value to evaluate prognosis, and positive correlation has been validated between serum lactic acid level and SIPP (higher SIPP leads to higher serum lactic acid level) [29,30]. In this study, we found that lactic acid has significant advantage in single index scoring system (the optimal cut-off value is 2.8 mmol/L, sensitivity and specificity are 0.875 and 0.69, respectively). In combined index scoring system, lactic acid+SIPP also shows obvious advantage: the optimal cut-off value is 0.212 mmol/L, sensitivity and specificity are 0.900 and 0.714, respectively. Stronger correlation can be detected between lactic acid and poor prognosis of patients with paraquat poisoning in single index scoring system after excluding effects of other factors, and RR value is 8.866 (95% CI: 2.462, 31.931). Similarly, stronger correlation can be detected lactic acid+SIPP and poor prognosis in combined index scoring system, and RR value is 12.098 (95% CI: 2.473, 59.178). The level of blood lactic acid was positively related to other system score. It plays an important role in judging the conditions and prognosis of paraquat poisoning patients admitted to hospital.
Taken together, combined lactic acid and SIPP may be the best selection due to its obvious advantage. Due to higher requirements of detection of paraquat concentration, lactic acid is the best choice because technique is difficult to achieve. Changes of serum lactic acid have higher value to infer severity and prognosis for acute paraquat poisoning patients. The potential mechanism of increased lactic acid may be damaged cell mitochondria and activated systemic inflammatory response by paraquat, and oxygen supply and demand imbalance after severe poisoning can aggravate formation of lactic acid. However, due to small sample size in this study, further studies should be performed to validate it using larger sample sizes.
Conflicts of Interest
The authors declare no conflict of interest.
This study was supported by application of infrastructure projects by Science and Technology Agency in Sichuan (No. 2013JY0155).
- Frelon JH, Merigot P, Garnier R, Bismuth C, Efthymiou ML. Prognostic factors in acute paraquat poisoning. A retrospective study of cases registered by the Poison Control Center of Paris in 1981. Toxicol Eur Res 1983; 5: 163-169.
- Xie Y, Xu Z, Yang Y. Analysis of 84 cases treatment of paraquat poisoning. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2015; 27: 312-313.
- Dinis-Oliveira RJ, Duarte JA, Sanchez-Navarro A, Remiao F, Bastos ML, Carvalho F. Paraquat poisonings: mechanisms of lung toxicity, clinical features, and treatment. Crit Rev Toxicol 2008; 38: 13-71.
- Zhang W. Progress in the diagnosis and treatment of acute paraquat poisoning. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2015; 27: 242-243.
- Yang S. Clinical significance of dynamic changes in serum inflammatory cell factors after acute paraquat poisoning. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi 2015; 33: 695-697.
- Li H, Liu B, Li P, Feng L, Ma H, Xuan S, Cao Y. Inhibitory effects of curcumin on inflammatory cytokines in rats with paraquat poisoning. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi 2015; 33: 689-692.
- Cao Y, Li J, Ou Z, Yin Z, Chen X, Han Y, Song W. Inhibition of pulmonary nuclear factor -KappaB and tumor necrosis factor -alpha expression by diallyl sulfide in rats with paraquat poisoning. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2015; 27: 274-279.
- Xu J, Zheng J, Zhu J. Tumor necrosis factor-alpha induced protein 6 attenuates acute kidney injury following paraquat poisoning in rats. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2014; 26: 405-408.
- Aires RD, Capettini LS, Silva JF, Rodrigues-Machado Mda G, Pinho V, Teixeira MM, Cortes SF, Lemos VS. Paraquat poisoning induces TNF-alpha-dependent iNOS/NO mediated hyporesponsiveness of the aorta to vasoconstrictors in rats. PLoS One 2013; 8: 73562.
- Tan JT, Letchuman Ramanathan G, Choy MP, Leela R, Lim BK. Paraquat poisoning: experience in hospital taiping (year 2008 - october 2011). Med J Malaysia 2013; 68: 384-388.
- Kim SJ, Gil HW, Yang JO, Lee EY, Hong SY. The clinical features of acute kidney injury in patients with acute paraquat intoxication. Nephrol Dial Transplant 2009; 24: 1226-1232.
- Gawarammana IB, Buckley NA. Medical management of paraquat ingestion. Br J Clin Pharmacol 2011; 72: 745-757.
- Kervegant M, Merigot L, Glaizal M, Schmitt C, Tichadou L, de Haro L. Paraquat poisonings in France during the European ban: experience of the Poison Control Center in Marseille. J Med Toxicol 2013; 9: 144-147.
- Senarathna L, Eddleston M, Wilks MF, Woollen BH, Tomenson JA, Roberts DM, Buckley NA. Prediction of outcome after paraquat poisoning by measurement of the plasma paraquat concentration. QJM Int J Med 2009; 102: 251-259.
- Jones AL, Elton R, Flanagan R. Multiple logistic regression analysis of plasma paraquat concentrations as a predictor of outcome in 375 cases of paraquat poisoning. QJM Int J Med 1999; 92: 573-578.
- Zhou Q, Jian X, Zhang Z, Wang J, Yu G. The research progress of paraquat poisoning lung molecular mechanism. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi 2015; 33: 72-75.
- Wang Z, Ma J, Zhang M, Wen C, Huang X. Serum Metabolomics in Rats after Acute Paraquat Poisoning. Biol Pharm Bull 2015; 38: 1049-1053.
- Mohamed F, Endre Z, Jayamanne S, Pianta T, Peake P, Palangasinghe C, Chathuranga U, Jayasekera K, Wunnapuk K, Shihana F, Shahmy S, Buckley N. Mechanisms underlying early rapid increases in creatinine in paraquat poisoning. PLoS One 2015; 10: 0122357.
- Liu W, Shan LP, Dong XS, Liu Z. Toll-like receptor 4 implicated in acute lung injury induced by paraquat poisoning in mice. Int J Clin Exp Med 2014; 7: 3392-3397.
- Fenix JB, Gillespie CW, Levin A, Dean N. Comparison of pediatric early warning score to physician opinion for deteriorating patients. Hospital Pediatrics 2015; 5: 474-479.
- Yoo JW, Lee JR, Jung YK, Choi SH, Son JS, Kang BJ, Park TS, Huh JW, Lim CM, Koh Y, Hong SB. A combination of early warning score and lactate to predict intensive care unit transfer of inpatients with severe sepsis/septic shock. Korean J Intern Med 2015; 30: 471-477.
- Bayer O, Schwarzkopf D, Stumme C, Stacke A, Hartog CS, Hohenstein C, Kabisch B, Reichel J, Reinhart K, Winning J. An early warning scoring system to identify septic patients in the prehospital setting: The PRESEP score. Acad Emerg Med 2015; 22: 868-871.
- Xu S, Hu H, Jiang Z, Tang S, Zhou Y, Sheng J, Chen J, Cao Y. APACHE score, Severity index of paraquat poisoning, and serum lactic acid concentration in the prognosis of paraquat poisoning of Chinese Patients. Pediatr Emerg Care 2015; 31: 117-121.
- Min YG, Ahn JH, Chan YC, Ng SH, Tse ML, Lau FL, Chan CK. Prediction of prognosis in acute paraquat poisoning using severity scoring system in emergency department. Clin Toxicol (Phila) 2011; 49: 840-845.
- Patel A, Hassan S, Ullah A, Hamid T, Kirk H. Early triaging using the Modified Early Warning Score (MEWS) and dedicated emergency teams leads to improved clinical outcomes in acute emergencies. Clin Med (Lond) 2015; 15: 3.
- Liang D, Nie S, Liu H, Shao D, Ren Y, Zhang W, Xu B, Sun Z, Sun B, Yang Z, Tang WJ. The research of the correlations among lactic acid, lactate clearance rate and apacheII score with the prognosis of acute paraquat intoxication. J Clin Emergency 2014; 7: 383-386.
- Czerniczyniec A, Lanza EM, Karadayian AG, Bustamante J, Lores-Arnaiz S. Impairment of striatal mitochondrial function by acute paraquat poisoning. J Bioenerg Biomembr 2015; 47: 395-408.
- Milczarek R, Sokolowska E, Rybakowska I, Kaletha K, Klimek J. Paraquat inhibits progesterone synthesis in human placental mitochondria. Placenta 2016; 43: 41-46.
- Zhang J, Zhao Y, Bai Y, Lv G, Wu J. The significance of serum uric acid level in humans with acute paraquat poisoning. Sci Rep 2015; 5: 9168.
- Xu S, Hu H, Jiang Z, Tang S, Zhou Y, Sheng J, Cao Y. Apache score, severity index of paraquat poisoning, and serum lactic acid concentration in the prognosis of paraquat poisoning of chinese patients. Pediatr Emerg Care 2015; 31: 117-121.