Explainable artificial intelligence (XAI) for predicting the need for intubation in methanol-poisoned patients: a study comparing deep and machine learning models.

The need for intubation in methanol-poisoned patients, if not predicted in time, can lead to irreparable complications and even death. Artificial intelligence (AI) techniques like machine learning (ML) and deep learning (DL) greatly aid in accurately predicting intubation needs for methanol-poisoned patients. So, our study aims to assess Explainable Artificial Intelligence (XAI) for predicting intubation necessity in methanol-poisoned patients, comparing deep learning and machine learning models. This study analyzed a dataset of 897 patient records from Loghman Hakim Hospital in Tehran, Iran, encompassing cases of methanol poisoning, including those requiring intubation (202 cases) and those not requiring it (695 cases). Eight established ML (SVM, XGB, DT, RF) and DL (DNN, FNN, LSTM, CNN) models were used. Techniques such as tenfold cross-validation and hyperparameter tuning were applied to prevent overfitting. The study also focused on interpretability through SHAP and LIME methods. Model performance was evaluated based on accuracy, specificity, sensitivity, F1-score, and ROC curve metrics. Among DL models, LSTM showed superior performance in accuracy (94.0%), sensitivity (99.0%), specificity (94.0%), and F1-score (97.0%). CNN led in ROC with 78.0%. For ML models, RF excelled in accuracy (97.0%) and specificity (100%), followed by XGB with sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%). Overall, RF and XGB outperformed other models, with accuracy (97.0%) and specificity (100%) for RF, and sensitivity (99.37%), F1-score (98.27%), and ROC (96.08%) for XGB. ML models surpassed DL models across all metrics, with accuracies from 93.0% to 97.0% for DL and 93.0% to 99.0% for ML. Sensitivities ranged from 98.0% to 99.37% for DL and 93.0% to 99.0% for ML. DL models achieved specificities from 78.0% to 94.0%, while ML models ranged from 93.0% to 100%. F1-scores for DL were between 93.0% and 97.0%, and for ML between 96.0% and 98.27%. DL models scored ROC between 68.0% and 78.0%, while ML models ranged from 84.0% to 96.08%. Key features for predicting intubation necessity include GCS at admission, ICU admission, age, longer folic acid therapy duration, elevated BUN and AST levels, VBG_HCO3 at initial record, and hemodialysis presence. This study as the showcases XAI's effectiveness in predicting intubation necessity in methanol-poisoned patients. ML models, particularly RF and XGB, outperform DL counterparts, underscoring their potential for clinical decision-making.

Prediction of naloxone dose in opioids toxicity based on machine learning techniques (artificial intelligence).

BACKGROUND: Treatment management for opioid poisoning is critical and, at the same time, requires specialized knowledge and skills. This study was designed to develop and evaluate machine learning algorithms for predicting the maintenance dose and duration of hospital stay in opioid poisoning, in order to facilitate appropriate clinical decision-making. METHOD AND RESULTS: This study used artificial intelligence technology to predict the maintenance dose and duration of administration by selecting clinical and paraclinical features that were selected by Pearson correlation (filter method) (Stage 1) and then the (wrapper method) Recursive Feature Elimination Cross-Validated (RFECV) (Stage2). The duration of administration was divided into two categories: A (which includes a duration of less than or equal to 24 h of infusion) and B (more than 24 h of naloxone infusion). XGBoost algorithm model with an accuracy rate of 91.04%, a prediction rate of 91.34%, and a sensitivity rate of 91.04% and area under the Curve (AUC) 0.97 was best model for classification patients. Also, the best maintenance dose of naloxone was obtained with XGBoost algorithm with R2 = 0.678. Based on the selected algorithm, the most important features for classifying patients for the duration of treatment were bicarbonate, respiration rate, physical sign, The partial pressure of carbon dioxide (PCO2), diastolic blood pressure, pulse rate, naloxone bolus dose, Blood Creatinine(Cr), Body temperature (T). The most important characteristics for determining the maintenance dose of naloxone were physical signs, bolus dose of 4.5 mg/kg, Glasgow Coma Scale (GCS), Creatine Phosphokinase (CPK) and intensive care unit (ICU) add. CONCLUSION: A predictive model can significantly enhance the decision-making and clinical care provided by emergency physicians in hospitals and medical settings. XGBoost was found to be the superior model.

Prediction of acute methanol poisoning prognosis using machine learning techniques.

Methanol poisoning is a global public health concern, especially prevalent in developing nations. This study focuses on predicting the severity of methanol intoxication using machine learning techniques, aiming to improve early identification and prognosis assessment. The study, conducted at Loghman Hakim Hospital in Tehran, Iran. The data pertaining to individuals afflicted with methanol poisoning was retrieved retrospectively and divided into training and test groups at a ratio of 70:30. The selected features were then inputted into various machine learning methods. The models were implemented using the Scikit-learn library in the Python programming language. Ultimately, the efficacy of the developed models was assessed through ten-fold cross-validation techniques and specific evaluation criteria, with a confidence level of 95%. A total number of 897 patients were included and divided in three groups including without sequel (n = 573), with sequel (n = 234), and patients who died (n = 90). The two-step feature selection was yielded 43 features in first step and 23 features in second step. In best model (Gradient Boosting Classifier) test dataset metric by 32 features younger age, higher methanol ingestion, respiratory symptoms, lower GCS scores, type of visual symptom, duration of therapeutic intervention, ICU admission, and elevated CPK levels were among the most important features predicting the prognosis of methanol poisoning. The Gradient Boosting Classifier demonstrated the highest predictive capability, achieving AUC values of 0.947 and 0.943 in the test dataset with 43 and 23 features, respectively. This research introduces a machine learning-driven prognostic model for methanol poisoning, demonstrating superior predictive capabilities compared to traditional statistical methods. The identified features provide valuable insights for early intervention and personalized treatment strategies.

Prediction the prognosis of the poisoned patients undergoing hemodialysis using machine learning algorithms.

BACKGROUND: Hemodialysis is a life-saving treatment used to eliminate toxins and metabolites from the body during poisoning. Despite its effectiveness, there needs to be more research on this method precisely, with most studies focusing on specific poisoning. This study aims to bridge the existing knowledge gap by developing a machine-learning prediction model for forecasting the prognosis of the poisoned patient undergoing hemodialysis. METHODS: Using a registry database from 2016 to 2022, this study conducted a retrospective cohort study at Loghman Hakim Hospital. First, the relief feature selection algorithm was used to identify the most important variables influencing the prognosis of poisoned patients undergoing hemodialysis. Second, four machine learning algorithms, including extreme gradient boosting (XGBoost), histgradient boosting (HGB), k-nearest neighbors (KNN), and adaptive boosting (AdaBoost), were trained to construct predictive models for predicting the prognosis of poisoned patients undergoing hemodialysis. Finally, the performance of paired feature selection and machine learning (ML) algorithm were evaluated to select the best models using five evaluation metrics including accuracy, sensitivity, specificity the area under the curve (AUC), and f1-score. RESULT: The study comprised 980 patients in total. The experimental results showed that ten variables had a significant influence on prognosis outcomes including age, intubation, acidity (PH), previous medical history, bicarbonate (HCO3), Glasgow coma scale (GCS), intensive care unit (ICU) admission, acute kidney injury, and potassium. Out of the four models evaluated, the HGB classifier stood out with superior results on the test dataset. It achieved an impressive mean classification accuracy of 94.8%, a mean specificity of 93.5 a mean sensitivity of 94%, a mean F-score of 89.2%, and a mean receiver operating characteristic (ROC) of 92%. CONCLUSION: ML-based predictive models can predict the prognosis of poisoned patients undergoing hemodialysis with high performance. The developed ML models demonstrate valuable potential for providing frontline clinicians with data-driven, evidence-based tools to guide time-sensitive prognosis evaluations and care decisions for poisoned patients in need of hemodialysis. Further large-scale multi-center studies are warranted to validate the efficacy of these models across diverse populations.

Comparing Two Naloxone Tapering Methods in Management of Methadone Intoxication; a Quasi-experimental Study.

Introduction: Even though naloxone is the main treatment for methadone poisoning treatment there are controversies about the proper method of its tapering. This study aimed to compare two methods in this regard. Method: This study was a prospective, single-blind pilot quasi-experimental study on non-addicted adult patients poisoned with methadone. Patients were randomly divided into 2 groups. In one group, after stabilization of respiratory conditions and consciousness, naloxone was tapered using the half-life of methadone and in the other group, naloxone was tapered using the half-life of naloxone. Recurrence of symptoms and changes in venous blood gas parameters were compared between groups as outcome. Results: 52 patients were included (51.92% female). 31 cases entered Group A (tapering based on methadone's half-life) and 21 cases entered Group B (tapering based on naloxone's half-life). The two groups were similar regarding mean age (p = 0.575), gender distribution (p = 0.535), the cause of methadone use (p = 0.599), previous medical history (p = 0.529), previous methadone use (p = 0.654), drug use history (p = 0.444), and vital signs on arrival to emergency department (p = 0.054). The cases of re-decreasing consciousness during tapering (52.38% vs. 25.81%; p = 0.049) and after discontinuation of naloxone (72.73% vs. 37.50%; p = 0.050) were higher in the tapering based on naloxone half-life group. The relative risk reduction (RRR) for naloxone half-life group was -1.03 and for methadone half-life group was 0.51. The absolute risk reduction (ARR) was 0.27 (95% confidence interval (CI) = 0.01-0.53) and the number needed to treat (NNT) was 3.7 (95% CI= 1.87- 150.53). There was not any statistically significant difference between groups regarding pH, HCO3, and PCO2 changes during tapering and after naloxone discontinuation (p > 0.05). However, repeated measures analysis of variance (ANOVA), showed that in the tapering based on methadone's half-life group, the number of changes and stability in the normal range were better (p < 0.001). Conclusion: It seems that, by tapering naloxone based on methadone's half-life, not only blood acid-base disorders are treated, but they also remain stable after discontinuation and the possibility of symptom recurrence is reduced.

Montelukast Is Effective in Treating Rhabdomyolysis Due to Intoxication: A Randomized Clinical Trial.

INTRODUCTION: Rhabdomyolysis is a clinical syndrome accompanied with biochemical changes that is diagnosed in some patients with acute chemical or drug poisoning. In this regard, the present study aimed to evaluate the effects of Montelukast in the treatment of intoxication-induced rhabdomyolysis. METHODS: This single-blind randomized clinical trial study was conducted in Loghman Hakim Hospital from March 2021 to March 2022. The study participants were 60 individuals evenly distributed into experimental and control groups. The experimental group received Montelukast plus routine treatment and the control group Creatine phosphokinase (CPK), urea, creatinine, aspartate aminotransferase (AST) and alanine transaminase (ALT) levels were monitored daily in both groups for seven days. The variables of age, gender and history of diabetes mellitus and kidney diseases were recorded. RESULTS: The mean age was 39.9 ± 16.87 and 38.2 ± 16.3 years in the control and intervention groups, respectively. Montelukast significantly (P < .05) reduced CPK levels on days five and seven, urea on days three, four, five and seven, and creatinine on days two to seven. The AST and ALT levels, unlike the control group which has a decreasing trend, increased first in the Montelukast group and then decreased on the sixth and seventh days. CONCLUSION: The results showed that Montelukast effectively reduced CPK, urea and creatinine levels, as well as the recovery time in patients with poison-induced rhabdomyolysis. In other words, Montelukast is effective in the treatment of rhabdomyolysis.  DOI: 10.52547/ijkd.7222.

A rare case report of lead encephalopathy due to high blood lead level.

Here we report a case of lead poisoning having a serum lead level of 412 mcg dL-1 who presented with decreasing level of consciousness and recurrent seizures. He responded well to treatment with chelation therapy.

Visual impairment following aluminum phosphide poisoning: A rare case.

Key Clinical Message: Aluminum phosphide poisoning may cause rare visual impairment. In a case, a 31-year-old female, visual loss was linked to shock-induced hypoperfusion, causing oxygen lack and cerebral atrophy, emphasizing the need for identifying atypical symptoms. Abstract: This case report describes the multidisciplinary evaluation of a 31-year-old female patient who suffered from visual impairment as a result of aluminum phosphide (AlP) poisoning. Phosphine, which is formed in the body when AlP reacts with water, cannot cross the blood-brain barrier; therefore, visual impairment seems unlikely to be the direct result of phosphine. To our knowledge, it is the first documented report of such impairment due to AlP.

Prediction of acute organophosphate poisoning severity using machine learning techniques.

Poisoning with organophosphate compounds is a significant public health risk, especially in developing countries. Considering the importance of early and accurate prediction of organophosphate poisoning prognosis, the aim of this study was to develop a machine learning-based prediction model to predict the severity of organophosphate poisoning. The data of patients with organophosphate poisoning were retrospectively extracted and split into training and test sets in a ratio of 70:30. The feature selection was done by least absolute shrinkage and selection operator method. Selected features were fed into five machine learning techniques, including Histogram Boosting Gradient, eXtreme Gradient Boosting, K-Nearest Neighborhood, Support Vector Machine (SVM) (kernel = linear), and Random Forest. The Scikit-learn library in Python programming language was used to implement the models. Finally, the performance of developed models was measured using ten-fold cross-validation methods and some evaluation criteria with 95 % confidence intervals. A total of 1237 patients were used to train and test the machine learning models. According to the criteria determining severe organophosphate poisoning, 732 patients were assigned to group 1 (patients with mild to moderate poisoning) and 505 patients were assigned to group 2 (patients with severe poisoning). With an AUC value of 0.907 (95 % CI 0.89-0.92), the model developed using XGBoost outperformed other models. Feature importance evaluation found that venous blood gas-pH, white blood cells, and plasma cholinesterase activity were the top three variables that contribute the most to the prediction performance of the prognosis in patients with organophosphate poisoning. XGBoost model yield an accuracy of 90.1 % (95 % CI 0.891-0.918), specificity of 91.4 % (95 % CI 0.90-0.92), a sensitivity of 89.5 % (95 % CI 0.87-0.91), F-measure of 91.2 % (95 % CI 0.90-0.921), and Kappa statistic of 91.2 % (95 % CI 0.90-0.92). The machine learning-based prediction models can accurately predict the severity of organophosphate poisoning. Based on feature selection techniques, the most important predictors of organophosphate poisoning were VBG-pH, white blood cell count, plasma cholinesterase activity, VBG-BE, and age. The best algorithm with the highest predictive performance was the XGBoost classifier.

Colchicine poisoning treated with hemoperfusion and hemodialysis: A case report.

This case report described an improved case of colchicine poisoning using hemoperfusion and hemodialysis.

Sumatriptan poisoning and its clinical presentation in humans: A case report.

This case report displays some of the possible complications of sumatriptan poisoning, including nephritic syndrome.

Conservative management of massive rivaroxaban overdose: A case report and literature review.

In the cases of acute rivaroxaban overdose, conservative management without prothrombin complex concentrate or other coagulation factors may be sufficient if renal function is normal and there is no bleeding.

Post-mortem Histopathologic Findings of Vital Organs in Critically Ill Patients with COVID-19.

BACKGROUND: The scientific evidence concerning pathogenesis and immunopathology of the coronavirus disease 2019 (COVID-19) is rapidly evolving in the literature. To evaluate the different tissues obtained by biopsy and autopsy from five patients who expired from severe COVID-19 in our medical center. METHODS: This retrospective study reviewed five patients with severe COVID-19, confirmed by reverse transcription-polymerase chain reaction (RT-PCR) and imaging, to determine the potential correlations between histologic findings with patient outcome. RESULTS: Diffuse alveolar damage (DAD) and micro-thrombosis were the most common histologic finding in the lung tissues (4 of 5 cases), and immunohistochemical (IHC) findings (3 of 4 cases) suggested perivascular aggregation and diffuse infiltration of alveolar walls by CD4+ and CD8+ T lymphocytes. Two of five cases had mild predominantly perivascular lymphocytic infiltration, single cell myocardial necrosis and variable interstitial edema in myocardial samples. Hypertrophic cardiac myocytes, representing hypertensive cardiomyopathy was seen in one patient and CD4+ and CD8+ T lymphocytes were detected on IHC in two cases. In renal samples, acute tubular necrosis was observed in 3 of 5 cases, while chronic tubulointerstitial nephritis, crescent formation and small vessel fibrin thrombi were observed in 1 of 5 samples. Sinusoidal dilation, mild to moderate chronic portal inflammation and mild mixed macro- and micro-vesicular steatosis were detected in all liver samples. CONCLUSION: Our observations suggest that clinical pathology findings on autopsy tissue samples could shed more light on the pathogenesis, and consequently the management, of patients with severe COVID-19.

Arterial Blood Gas Analysis of Patients with Tramadol-induced Seizure; a Cross Sectional Study.

INTRODUCTION: Tramadol is an active analgesic drug that is commonly used to treat moderate to severe pain. The present study aimed to assess the arterial blood gas (ABG) analysis of patients with tramadol-induced seizure (TIS). METHODS: This prospective cross-sectional study was performed on 50 TIS cases that were referred to emergency department within a maximum of one hour after their last episode of seizure. The results of ABG analysis on admission were collected and their association with dosage and time interval between ingestion and admission was assessed. RESULTS: 50 cases with the mean age of 35.10 ± 9.62 years were studied (80.0% male). The mean dosage of ingestion was 1122.00 ± 613.88 (400 to 3000) mg and the mean time interval between ingestion and admission was 7.16 ± 2.18 hours. ABG analysis on admission showed that 49 (98.0%) patients had pH < 7.35 and PaCO2 > 45 mmHg (respiratory acidosis). There was a significant association between ingestion to admission time interval and both PaCO2 (r = -0.330, p = 0.019), and PaO2 (r = 0.303, p = 0.032). The dose of ingestion was negatively associated with respiratory rate (r = -0.556, p = 0.001), arterial pH (r = -0.676, p = 0.001), and PaO2 (r = -0.514, p = 0.001), but was positively associated with PaCO2 (r = 0.461, p = 0.001). Higher doses of tramadol led to more severe hypercapnia and need for intubation (OR = 1.12, 95% CI: 0.88 - 1.26; p = 0.045). 5 (10.0%) cases needed mechanical ventilation. All patients improved after supportive care with no in-hospital death. CONCLUSION: Based on the findings, 98% of TIS cases had respiratory acidosis. Higher doses of ingested drug and longer time interval between ingestion and admission were associated with severity of ABG disturbances.

Comparison of the efficacy of an infusion pump or standard IV push injection to deliver naloxone in treatment of opioid toxicity.

BACKGROUND: The optimal goal of naloxone infusion in intensive care units is to ameliorate opioid-induced side effects in therapy or eliminate the symptoms of opioid toxicity in overdoses. Accurately monitoring and regulating the doses is critical to prevent adverse effects related to naloxone administration. The present study aimed to compare treatment outcomes when using two methods of intravenous naloxone infusion: an infusion pump or the standard method. METHODS: This study involved 80 patients with signs and symptoms of opioid overdose. The patients were randomly assigned into two groups with respect to intravenous infusion of naloxone by either an infusion pump or the standard method. RESULTS: Comparison of study parameters between the two groups at 12 and 24 hours after intervention showed significantly more compensatory acid-base imbalance in the naloxone infusion pump group. In the group that received naloxone by pump, only one patient experienced withdrawal symptoms, but withdrawal symptoms appeared in 12 patients (30.0%) in the standard intravenous infusion group within 12 hours and in seven additional patients (17.5%) within 24 hours of intervention. In the group receiving pump-based naloxone infusion therapy, no another complications were reported; however in the standard infusion group, the 12-hour and 24-hour complication rates were 55.0% and 32.5%, respectively. The length of hospital stay was 2.85±1.05 and 4.22±0.92 days for the pump and standard infusion groups, respectively (P<0.001). CONCLUSIONS: Naloxone infusion using an infusion pump may be safer with regard to hemodynamic stability, resulting in shorter hospitalization periods, and fewer posttreatment complications.

Inhalational Toxicity of Aluminum Phosphide as an Ongoing Concern; a Report of Two Cases.

Acute aluminium phosphide (ALP) poisoning is an extremely lethal poisoning. Ingestion is usually suicidal in intent, uncommonly accidental and rarely homicidal. Unfortunately, the absence of a specific antidote results in very high mortality and the key to successful treatment is in rapid decontamination and institution of resuscitative measures. Phosphine gas is highly toxic, and fatality is expected even several hours after continuous exposure. However, intensive supportive treatments may be lifesaving in some cases. Here, two cases of accidental inhalation intoxication with ALP are reported. One patient was discharged and another suffered cardiac arrest during treatment.

Characteristics of Mortalities related to Pulmonary Embolism following Multiple Trauma; a Brief Report.

INTRODUCTION: Pulmonary embolism (PE) is introduced as the third major cause of death after trauma in those who survive more than 24 hours. This study aimed to describe the characteristics of mortalities due to trauma related PE in cases referred to the forensic medicine department. METHODS: The present cross sectional study was conducted on medical profiles of cadavers that were registered as trauma related mortality in the dissection department of the Forensic Medicine Organization, Tehran, Iran, during 2011 to 2016. RESULTS: The cause of death for 92 of the 10800 (0.85%) evaluated cadavers was diagnosed as trauma related PE. The mean age of these patients was 58.37 ± 19.39 years (66.3% male). Only 14 (15.2%) hospitalized patients had received anticoagulant agents. The most frequent trauma related PE mortality cases were male (p = 0.003) and aged > 55 years (p = 0.005), with trauma to death time of < 3 weeks (p = 0.004), lower limb injury (p = 0.003), car crash trauma mechanism (p = 0.003), and no anticoagulant prescribed (p = 0.001). CONCLUSION: According to the results of the present study, the prevalence of trauma related PE mortality was 0.85%. It seems that, having a clear anticoagulation therapy protocol in trauma centers could be helpful in decreasing the prevalence of traumatic thromboembolism and its' related mortality.

Main predictors for repetition of suicidal behaviour among women referred to a single public sector tertiary care hospital in Iran.

OBJECTIVE: To assess the main predictors for repetition of suicidal behaviour among women. METHODS: This cross-sectional study was conducted at Loghman Hakim Hospital, Tehran, Iran, in 2014, and comprised women patients. The patients were divided into two groups, i.e. women repeating suicide and women without repeating suicide. Data was collected through a checklist and then analysed with SPSS 20. RESULTS: Of the 300 women, 121(40.3%) repeated suicide and 179(59.7%) did not. The overall mean age was 26.9±9.1 years (range: 14-80 years). High prevalence of psychological drug usage, alcohol use, history of self-mutilation (self-harm), psychotic disturbances, sexual relationships, as well as smoking and opium addition was revealed as major factors in repeated suicidal behaviour in women when compared with other women. The result of multivariate logistic regression model showed two factors of self-mutilation (odds ratio =2.692, p=0.002) and underlying psychotic disorders (odds ratio = 2.780, p<0.001) as main predictors of suicide in women. In this regard, demographic characteristics could not predict repeating suicidal attempts (p>0.05). CONCLUSIONS: The presence of underlying psychotic disorders and self-mutilation were main predictors for repetition of suicidal behaviour.

Comparison of postoperative clinical outcome after repairing surgery for lumbar spinal stenosis between diabetic and nondiabetic patients.

BACKGROUND: Poorer postoperative outcome is suggested after repairing surgery in diabetic patients with lumbar spinal stenosis in comparison with nondiabetic patients. The present study aimed to compare the clinical outcome of surgery for lumbar spinal stenosis and diabetic and nondiabetic patients to highlight the effect of diabetes on prognosis of this surgical procedure. METHODS: This prospective cohort study is conducted on 25 diabetic patients with lumbar spinal stenosis who were candidate for surgical treatment. A gender, age, and body mass index-matched group including 30 nondiabetic patients with lumbar spinal stenosis was considered as the control. The clinical condition of the patients was assessed based on oswestry disability index (ODI) before and immediately after surgery. RESULTS: There was no difference in baseline ODI index between diabetes and diabetes group (73.68 ± 18.89 vs. 71.20 ± 18.27, P = 0.625), whereas postprocedure ODI was significantly higher in diabetic patients than in nondiabetic group (54.32 ± 19.03 vs. 29.47 ± 18.75, P < 0.001). The multivariable logistic regression analysis could show a difference in postoperative ODI between diabetic and nondiabetic patients with the presence of baseline variables as the confounders (beta = -24.509, P < 0.001). CONCLUSION: Lower improvement in physical ability is expected in diabetic patients after surgery for lumbar spinal stenosis when compared to nondiabetes patients.

The effect of plasmapheresis on treating disseminated intravascular coagulation (DIC) caused by a Hemiscorpius lepturus (Gadim) sting.

INTRODUCTION: The highest mortality from scorpion stings in Iran is due to the stings of a particular type of scorpion known as Hemiscorpius lepturus (H. lepturus, Gadim in local language). The present study aimed at investigating the use of plasmapheresis to treat severe cases of H. lepturus stings. METHOD: This pilot study was a randomized clinical trial conducted from June 2015 to June 2016 in Razi hospital of Ahvaz, Iran. Twenty-nine patients who had been stung by H. lepturus and admitted to ICU because of disseminated intravascular coagulation (DIC) were randomly assigned into control (15 patients, supportive treatments) and plasmapheresis (14 patients, supportive treatments + plasmapheresis) groups, and the patient outcomes were compared between the two groups. FINDINGS: Eighteen patients were female (62%), and the mean of patient age was 24 ± 7. Most of the sting cases had occurred in the torso (15 patients, 52%). Only 10 patients (34%) arrived in the hospital within 12 h of being stung. There was no significant difference between the two groups in terms of the demographic and sting features. In the plasmapheresis group, hemoglobin level was significantly lower, while the PT and INR were measurably higher. In total, the plasmapheresis group experienced 29 sessions of treatment (an average of two sessions for each patient). Overall, 19 patients (66%) expired, whereas 10 patients (34%) experienced recovery with or without complications. The rate of recovery was significantly higher in the plasmapheresis group compared with controls, with eight patients (57%) in the plasmapheresis group surviving compared with two (14%) in the control group (p=.045). The duration of hospitalization was higher in the plasmaphersis group (p < .001). A comparison of the dead and recovered patients' features indicated that the dead patients arrived in the hospital significantly later than the recovered ones, and they also had lower platelet counts. CONCLUSIONS: The findings of this small-scale pilot study show that using plasmapheresis in treating DIC in patients stung by H. lepturus can prevent death and encourage recovery. However, prior to using plasmapheresis as a routine treatment for severe cases of people stung by this scorpion or other similar ones, further controlled studies with a larger sample size are needed.