الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 205 |  |  |
| | | from | 205 | | |
Abstract
Acute poisoning is one of the most serious problems worldwide with 90% of the burden of fatal poisoning coming from developing countries. According to the World Health Organization, it is the ninth leading cause of death in young adults from 15 to 29 years old. It is responsible for the loss of over 7.4 million years of healthy life. It also referred that, 0.3 million people die due to various poisonous agents annually. Acute poisoning may affect different body systems including respiratory, cardiovascular and gastrointestinal systems. Diagnosis of acute poisoning is based on history taking, adequate examination and investigations. Treatment of patients with acute toxicity is based on prevention of further exposure, maintaining of open airway, breathing and circulation, decontamination, enhancement of elimination and administration of specific antidote if available. Triage management remains a major challenge, especially in resourcepoor settings in developing countries. It considers a quick assessment technique in the emergency department and depends on certain vital parameters as heart rate, blood pressure, respiratory rate, temperature and oxygen saturation. Predictive scores permit early recognition of the patient’s outcomes. Subsequently, this allows proper treatment of acutely poisoned patients and improves the outcomes. Different studies were done on scoring systems as tools for the prediction of the outcome of acutely poisoned patients. Limitations of these scores as PSS and APACHE II, include having multiple variables and being calculated 24 hours after admission. Moreover, some are not specifically designed to be applied in acute poisoning. The aim of this study was to develop novel scoring systems using vital signs, ABG and consciousness level as a tool of triage for evaluation and outcome prediction of acutely poisoned patients including ICU admission and mortality. The current study was conducted on patients presented with acute poisoning who were admitted to Tanta University Poison Treating Center (TUPTC) during the period from the start of April 2020 to the end of June 2021. Patients who were admitted from the start of April 2021 to the end of March 2021 served as the derivation group and those who were admitted from the start of April 2021 to the end of June 2021 served as the validation group. • Inclusion criteria: Both male and female patients aged 18 years or more admitted within 24 hours of acute poisoning were included. The diagnosis was based on a history of exposure, availability of drug tablets or bottles brought by the patient or his relatives, characteristic clinical manifestations (symptoms and signs) and laboratory investigations if available. • Exclusion criteria: Patients who were excluded from this study were; patients less than 18 years, patients with history of chronic diseases such as cardiac, respiratory, hepatic or renal diseases and patients who received any medical intervention before admission. Furthermore, patients with associated trauma were also excluded. All patients were subjected to: 1. History taking: - Socio-demographic data (age, sex, residence, special habits, occupation and marital state). - Toxicological history including: name, form, route of exposure, mode of poisoning, and time elapsed before hospital admission. 2. Clinical Examination: - Vital signs (pulse, blood pressure, respiratory rate and temperature). - Level of consciousness by Glasgow Coma Scale (GCS). - Routine examinations of different body systems. 3. Laboratory investigations: - Arterial Blood Gases (ABG). The results of this study revealed that: The age of all studied patients included in this study ranged between 18 and 65 years. Females and males were nearly equally presented (51.5 % and 48.5 % respectively). The major part of all studied patients was from Gharbia governorate (82.2%). Addict patients represented 4.6% of all studied patients. The majority of the studied patients were students (27.1%) followed by housewives (25.9%), manual workers (23.3%), idles (17.8%) then employees (6%). Single patients represented the majority of the studied patients (55.8%). As regards toxicological data, pesticides were the most common reported poison that was represented in 57.2% followed by psychotropic drugs that were represented in 9.6% of all included patients. Regarding the form of poison, solid, liquid and gas poisons were encountered in 84.6%, 13.5% and 1.9% of the total studied patients respectively. Ingestion was the route of exposure reported in the majority of studied patients (88.2%). As regards the mode of poisoning, 91.1% of patients alleged suicidal poisoning while 8.9% alleged accidental poisoning. The pre-hospitalization period ranged from 3 to 24 hours with a median value of 2.5 hours. Regarding vital signs, the median values of pulse, SBP and DBP were 90 beats/minute, 120 mmHg and 80 mmHg respectively for all studied patients. The mean values of the respiratory rate and temperature of all studied patients were 21.5 ± 5.5 breaths/minute, and 37.0 ± 0.3 ºC respectively. The lowest GCS score reported in this study was 3 and the highest score was 15 with a median value of 15 in all studied patients. In the current study, ABG showed the following results; blood pH ranged from 7.05 to 7.65 with a mean value of 7.43 ± 0.08 in all studied patients. Moreover, PaCO2 ranged from 11.4 to 68.7 with a median value of 31.4 mmHg. Furthermore, the range of HCO3 of all studied patients was from 5.8 to 46.0 with a mean value of 21.3 ± 4.9 mEq/L. The oxygen saturation of all studied patients ranged from 18 to 100% and the mean value was 95 ± 10.8%. Regarding outcome, 13.7% of total patients required ICU admission while mortality was represented in 5.8% of all patients included during the period of the study. No significant difference was observed between the derivation group and the validation group regarding socio-demographic data, toxicological data, vital signs, GCS, ABG and outcome. For patients who required ICU, the mean value of age was 28.5 ± 9.7 years and males constituted 53.4% of all ICU admitted patients. The majority of patients were from Gharbia governorate (82.8%). Non-addicts and idles constituted 84.5% and 36.2% of ICU admitted patients respectively. Married patients were 51.7% of ICU admitted patients. No significant difference was found between cases who required and didn’t require ICU admission as regards the age. Moreover, no significant association was observed between ICU requirement and each of gender, residence and marital status in the current study. On the other hand, there was a significant association between the requirement of ICU admission and each of special habits and occupations. According to toxicological data in patients who required ICU admission in the current study, pesticides and opioids were responsible for 62.1% and 19%. Solid, liquid and gas poisons were encountered in 86.2%, 12.1% and 1.7% of the total ICU admitted patients respectively. The oral route of exposure represented the highest percentage 81%. Alleged suicidal poisoning was the mode represented in 94.8% of patients. The prehospitalization period ranged from 1 to 24 hours with a median value of 2 hours. There were significant associations between ICU requirement and each of poison category and route of exposure. Moreover, no significant associations were observed between ICU requirement and each of poison form, delay time and alleged mode of poisoning. The median value of heart rate in patients who required ICU (99 beats/minute) was significantly higher compared to that of patients who didn’t require ICU (90 beats/minute). On the other hand, the median values of SBP and DBP in patients who required ICU (85 mmHg and 55 mmHg respectively) were significantly lower compared to those who didn’t require ICU (120 mmHg and 80 mmHg respectively). The mean value of respiratory rate in patients who required ICU admission (24.2 ± 9.9 breaths/minute) was significantly higher compared to patients who didn’t require ICU admission (21.3 ± 5.3 breaths/minute). Furthermore, the mean value of temperature in patients who required ICU (36.9 ± 0.5 ºC) was significantly lower than cases did not require ICU admission (37.0 ± 0.3 ºC). In addition, the lowest GCS score reported in patients who required ICU admission was 3 and the highest score was 15 with a median value of 15 and there was a significant difference between cases who required and who did not require ICU admission. Regarding ABG, the mean value of blood pH in cases that required ICU admission was significantly lower compared to those who didn’t require ICU (7.35 ± 0.13 and 7.44 ± 0.06 respectively). The mean value of serum HCO3 in patients who required ICU admission (17.6 ± 6.2 mEq/L) was significantly lower compared to patients who didn’t require ICU admission (22.0 ± 4.0 mEq/L). The median values of PaCO2 and PaO2 in patients who required ICU admission were 30 and 87.7 mmHg respectively with no significant difference between both groups. Moreover, it was observed that the mean value of O2 saturation in patients who required ICU admission (80.5 ± 20.2%) was significantly lower compared to those who didn’t require ICU admission (97.6 ± 2.8%). In the current study, ROC curve analysis was used to predict potential factors of ICU requirement in patients with acute poisoning and showed that, each of SBP, DBP, respiratory rate, temperature, GCS, blood pH, serum HCO3 and O2 saturation were significantly valid to predict ICU requirement. On the other hand, pulse was not significantly valid to predict ICU requirement. To construct a scoring system that could predict ICU requirement in acute poisoning, backward stepwise binomial logistic regression was done and identified seven independent variables that contributed significantly in prediction of ICU requirement. These variables were O2 saturation, SBP, blood pH, GCS, serum HCO3, temperature and respiratory rate and the accuracy of this score was assessed by using ROC curve analysis. The discriminatory power of the ICU score in the derivation group was excellent (AUC: 0.979) at cut off >1. Moreover, it showed 100% sensitivity and 83.9% specificity at cut-off point more than 1. To evaluate the accuracy of the proposed score, it was internally validated on a new set of patients (the validation group) using ROC curve analysis. At cut-off value >1 point, it showed excellent performance (AUC: 0.953) with 94.1% sensitivity and 81.7% specificity. The proposed score could be displayed as a tool of triage to identify poisoned patients who are going to need ICU admission. For socio-demographic data in non-survivor cases, the mean value of age was 28.8 ± 11.1 years. The majority of patients were from Gharbia governorate (84.6%), non-addict cases were 100%, housewives were 34.6% and married cases were 57.7%. No significant association was found between mortality and socio-demographic data of all studied patients. Regarding toxicological data in non-survivor cases, the majority were intoxicated by pesticides (92.3%). All non-survivors were intoxicated by ingestion route, solid form of poison and alleged suicidal poisoning. No significant associations were observed between mortality and each of the poison category, route of exposure, form of poison and mode of poisoning. The pre-hospitalization period of non-survivors, it ranged from 1 to 5 hours with a median of 1.5 hours. Significant difference was noticed between survivors and non-survivors. Regarding vital signs reported in non-survivor cases, the median value of heart rate was 80 beats/minute with no significant difference noticed between survivors and non-survivors. The median values of SBP and DBP in non-survivors (70 mmHg and 40 mm/Hg respectively) were significantly lower compared to the median value of SBP and DBP in survivors (120 mmHg and 80 mmHg respectively). The mean value of respiratory rate in non-survivors (26.3 ± 6.4 breaths/minute) was significantly higher compared to the mean value of respiratory rate in survivors (21.3 ± 5.2 breaths/minute). Moreover, the mean value of temperature in survivors (37.0 ± 0.3 ºC) was significantly higher than in non-survivors (36.8 ± 0.3 ºC). The GCS ranged between 3 to 15 in non-survivors with a median value of 15 and there was no significant difference between survivors and non-survivors. Regarding the ABG in non-survivor cases, the mean values of blood pH and serum HCO3 in survivors (7.43 ± 0.08 and 14.2 ± 3.9 mEq/L respectively) were significantly higher than their mean values in non survivors (7.37 ± 0.12 and 21.8 ± 4.3 mEq/L respectively). In addition, the median value of PaCO2 in survivors (31.7 mmHg) was significantly higher than the median value in non-survivors (22.5 mmHg). Moreover, the median value of PaO2 in non-survivors was 100 mmHg with no significant difference was observed in comparison with survivors. Furthermore, the current results showed that O2 saturation was significantly lower in non-survivors than in survivors with mean values 87.4 ± 8.9% and 95.8 ± 9.4% respectively. In the current study, ROC curve analysis was used to predict potential factors of mortality in patients with acute poisoning and revealed that, each of SBP, DBP, respiratory rate, temperature, blood pH, PaCO2, serum HCO3 and O2 saturation were significantly valid to predict mortality. On the other hand, pulse was not significantly valid to predict mortality. To construct a scoring system to predict mortality in acute poisoning, backward stepwise binomial logistic regression was done in the present study and identified three independent variables (O2 saturation, SBP and serum HCO3) that contributed significantly in prediction of mortality. The accuracy of the proposed mortality score was assessed by using ROC curve analysis and showed that, its discriminatory power in the derivation group was excellent (AUC: 0.952) at cut off >1. Moreover, it showed 91.3% sensitivity and 94.9% specificity at cut-off point more than one. The proposed mortality score was internally validated on a new set of patients (the validation group) using ROC curve analysis. At cut-off value >1 point, it showed excellent performance (AUC: 0.965) with 87.5% sensitivity and 93.8% specificity.