| Objective(s):The objective of this research was to look into the relationship between neutrophils to lymphocytes ratio(NLR)and acute paraquat poisoning(APP)patients at admission,as well as the predictive value of NLR for high-risk(moderate-severe/explosive)APP patients in the early stages.Methods:Patients with acute paraquat poisoning admitted to the Emergency Department of the Second People’s Hospital of Yunnan Province from January 2018 to June 2021 were chosen as the research object,and those who underwent health examination in the hospital’s physical examination center during the same period were chosen as the control.Gender,age,past history,recent infection history,and medication history,route of pesticide exposure,dosage,first gastric lavage time,time of poisoning to our hospital,first blood purification time,clinical symptoms,neutrophil count and lymphocyte count immediately upon admission,liver and kidney function indexes immediately upon admission and dynamic changes during hospitalization,coagulation indexes at admission,dynamin indexes at admission,dynamin indexes at admission,dynamin indexes at admission,The Expert Consensus of Acute Paraquat Poisoning(2022)classified APP patients as mild,moderate-severe,and explosive.Comparing the subjects’general situation and laboratory indexes,analyzing the differences in NLR between groups and the correlation between NLR and the poisoning degree of APP patients,establishing an orderly multi-factor Logistic regression model to screen the independent risk factors of the poisoning degree of APP patients,and obtaining the best cutoff value for early prediction of high-risk(moderate-severe/explosive)APP patients via the receiver operating characteristicResults:1.Subjects basic information:157 healthy controls with a median age of 36 years were included,with 82 males and 75 females.There were 198 APP patients,109 men and 89 women,with a median age of 33 years.Among 198 APP patients,55 had mild symptoms,45 had moderate to severe symptoms,and 98 had explosive symptoms.2.The gender and age distributions in the healthy control and APP groups were not statistically significant(all P>0.05).At admission,the NE and NLR values in the APP group were significantly higher than those in the healthy control group,while the LYM value in the APP group was lower,with statistical differences(all P<0.001).3.Comparative results of acute paraquat poisoning in different degree groups:3.1 There was no significant difference in age,sex,time for first gastric lavage,time for poisoning to our hospital,time for first blood purification,blood potassium,and INR between the three groups(all P>0.05),but there was a significant difference in poison dose,semi-quantitative detection level of urine PQ,BUN,Cr,TB,ALT,NE,LYM,and so on(all P<0.05).The comparison of the two groups revealed that the dose,Cr,NE,and NLR values increased with disease severity(all P<0.05).The levels of urine PQ semi-quantitative detection was significantly higher in patients with moderate-severe type and explosive type than in patients with mild type,LYM was significantly lower than in mild patients,and the differences were statistically significant(all P<0.001),but there was no statistical difference between patients with moderate-severe type and explosive type.3.2 The semi-quantitative detection of toxic dose,BUN,Cr,TB,ALT,NE,NLR,and urine PQ were positively correlated with the degree of acute paraquat poisoning,while LYM was negatively correlated,with NLR being the most correlated with the degree of acute paraquat poisoning.(r_s=0.727,P<0.001).3.3 There was a significant positive correlation between NLR and poison dose and urine PQ semi-quantitative detection level.(r_s was 0.541 and 0.628,all P<0.001).3.4 Ordered multi-classification Logistic regression analysis showed that NLR,urine PQ semi-quantitative detection level and Cr were independent risk factors for acute paraquat poisoning.3.5 The area under curve(AUC)value of NLR is 0.926(95%CI0.890-0.962,P<0.001),which is the highest among all predictive factors,and its best cutoff value is7.785,with 90.9%sensitivity and 81.8%specificity for predicting high-risk APP patients,according to the ROC curve.The AUC of urine PQ semi-quantitative detection level is 0.909(95%CI 0.860-0.959,P<0.001),and when the optimal cutoff value is+++,the sensitivity and specificity for predicting high-risk APP patients are86%and 90.9%,respectively.The AUC for Cr was 0.813(95%CI 0.755-0.872,P<0.001),with 73 umol/L as the best cut-off value.For predicting high-risk APP patients,the sensitivity and specificity were 67%and 87.3%,respectively.The semi-quantitative detection levels of NLR and urine PQ,as well as the AUC of Cr,differed significantly(all P<0.05).NLR and urine PQ semi-quantitative detection grade AUC were not significantly different(P=0.535).When NLR was combined with the urine PQ semi-quantitative detection level,the AUC increased to 0.953(95%CI:0.926-0.980,P<0.001),and the prediction sensitivity for high-risk APP patients was 88.1%,with a specificity of 94.5%.There were statistically significant differences in the combined index,AUC of NLR,and urine PQ semi-quantitative detection level(all P<0.001).Conclusion(s):1.At admission,the NLR of the APP patients was significantly higher than healthy people’s,and the higher the NLR value at admission,the more severe the poisoning degree of APP patients.2.NLR,NE,urine PQ semi-quantitative detection level,Cr and poison dose were all significantly positively correlated with acute paraquat poisoning degree,LYM was negatively correlated with the degree of acute paraquat poisoning,with NLR having the strongest correlation.3.NLR,urine PQ semi-quantitative detection level,and Cr were independent risk factors for acute paraquat poisoning.There are all have better prediction efficiency for high-risk APP patients,with NLR and urine PQ semi-quantitative detection level outperforming Cr.The prediction efficiency is best when NLR is combined with urine PQ semi-quantitative detection level,with high sensitivity and specificity,which can be popularized and applied in primary hospitals. |