INTRODUCTION

Poisoning remains the fourth leading cause of injuries after road traffic accidents, burns, and drowning¹. There is a considerable mortality risk with unintentional poisoning, and most such cases hail from low-middle-income countries². Pediatric poisoning remains a common and preventable cause of morbidity and mortality worldwide³. Hydrocarbons (HCs) account for 5% of all pediatric poisoning admissions⁴ and 90% of HC-related deaths involved children less than 5 years of age⁴. Hydrocarbons are a component of most household chemicals such as cleansing and disinfectant solutions, cooking fuel, insect repellents, etc., which are invariably found in all residences, making them a readily available potential toxic substance. The storage of such products in inappropriate containers, such as soft drink bottles and water storage cans, makes them easily accessible, especially for young, curious children who like mouthing things, resulting in unintentional HC poisoning.

The severity of HC poisoning depends upon the type, viscosity, volatility, and surface tension of the HC Compound⁴. Manifestations are primarily pulmonary, and the risk of aspiration is directly associated with viscosity⁴. Pulmonary complications include atelectasis, pneumonitis, pneumothorax, and pneumatoceles⁴. Cardiotoxicity manifests as dysrhythmias, secondary to HC sensitization of the myocardium to the circulating catecholamines⁴. Gastrointestinal manifestations like throat and abdominal pain, nausea, and vomiting are due to irritant effects⁴. Neurological complications include euphoria, depression, lethargy, convulsions, and coma⁴. However, acute HC poisoning in children can have a variable clinical presentation and outcome depending on the nutritional status, pre-existing co-morbidities, socio-economic factors, cultural beliefs and practices, the healthcare system of the country, etc.

The present study was conducted at a Pediatric Intensive Care Unit (PICU) in a tertiary care center affiliated with a private medical college in South India, catering primarily an urban population belonging to low middle-income socioeconomic class. This study was performed to understand the clinical-demographic profiles and assess the radiological findings and outcome of acute HC poisoning in children.

METHODS

RESULTS

A total of 29 children with hydrocarbon poisoning were admitted during the 18-month study period, out of 1568 admissions. Of the 29 children, 18 (62%) were males and 11 (38%) were females. Although males outnumbered females, there was no statistically significant association between gender and hydrocarbon poisoning (p=0.19). The average age of presentation was 3.3 years and the median age of presentation was 2.6 years (range: 0.5–15 years). The majority of the children (n=17; 58.6%) were aged 2–5 years, followed by <2 years of age (n=8; 27.6%) and >5 years of age (n=4; 13.8%) and the association between age group and poisoning was statistically significant (p=0.01) (Table 1).

All cases belong to the urban population. Of the 29 patients, 18 (62%) were living in a joint family and 11 (38%) were in a nuclear family; however, the difference was not statistically significant (p=0.19). The reason for hydrocarbon poisoning in these children was not associated with the occupation of the parents in all cases, except one where the father was a goldsmith who worked with turpentine in the shed room in their house. The 15-year-old girl mistakenly consumed paint thinner, assuming it to be water at their relative’s house. The common route of poisoning was ingestion and, in all cases, the poisoning was accidental. The most common poisons identified were paint thinner/turpentine (n=22; 75.9%) followed by all-out liquid (insect repellent) (n=6; 20.7%). Baygon (insecticide) poisoning occurred in only 1 (3.4%) case. The peak incidence of HC poisoning was noted during summer months accounting to 48.3% of total cases. However, the difference is not statistically significant (p=0.18) (Table 1).

All 29 children were symptomatic at the time of presentation. Vomiting (69%) and cough (24.1%) were the most common symptoms observed among the admitted cases. Respiratory distress was observed in 6 (20.7%) cases. Drowsiness was observed in 3 (10.3%) cases. Two (6.9%) cases presented with fever and abdominal pain each. None had seizure as a symptom either at the time of presentation or later during the hospital stay. The average time of presentation to ER was one hour thirty minutes with a range of 30 minutes to 2.5 hours, excluding 1 case which is the outlier who presented after 2 days of exposure. A chest X-ray was done on all 29 patients. Chest x-Ray was normal in 26 (89.7%) cases and evidence of pneumonitis in 2 (6.9%) patients, and pleural effusion was evident in 1 (3.4%) patient (Table 2).

The majority of the admissions required minimal supportive care. Respiratory support was required in 12 (44%) patients. Oxygen was needed for 9 (31%) patients and non-invasive ventilator support was needed in 2 (6.9%) patients. Intubation and invasive ventilation were needed in 1 (3.4%) case. Steroid was used in 1 (3.4%) patient who presented with extensive consolidation and pleural effusion. Most (n=25; 86.2%) patients did not receive any antibiotics. Empirical antibiotics were started in 4 (13%) cases but were discontinued in 2 patients later (Table 2).

The mean duration of hospital stay was 33 hours (range: 4–150 hours). There was no mortality. Twenty (70%) patients were discharged within 24 hours, 6 (20%) were discharged between 24 to 72 hours, and only 3 (10%) patients required greater than 72 hours hospital stay. GI decontamination was not done in any case. Topical decontamination was performed in all patients as per the guidelines. All 29 cases had a hydrocarbon severity score of <4 on admission. All 29 cases were discharged home and were well at 28-day follow-up. There was no mortality.

DISCUSSION

Household chemicals are the most commonly reported cause of accidental poisoning in young children. In this study, hydrocarbon poisoning constituted 1.8% of the total 1568 admissions. HC poisoning occurred predominantly in the toddler age group. In our study, the overall incidence of hydrocarbon poisoning in children under 5 years of age was 86.2%. The results are consistent with a study conducted at Benha University, Egypt, by Madbolu et al.⁶ where 76.4% of poisoning cases were seen in children less than 5 years age. In a study conducted in north India by Jayashree et al.⁷, 94% children with HC poisoning were less than 5 years old and only 4 children were older than 5 years of age⁷. Young children are attracted to colorful packages and their immature olfactory function along with their explorative mouthing behavior makes them highly prone to accidental poisoning⁸. In Indian lower socioeconomic background households, kerosene, paint thinner and other daily-use hydrocarbon-based compounds are stored in water bottles or soft drink bottles and are sometimes inadvertently left in easily accessible areas which make young children prone to accidental poisoning.

Male preponderance was seen in this study with males constituting about two-thirds of total cases and similar results have been noted in various other studies as well^2,6,9. The reason for this observation could be the higher degree of hyper-active character or inquisitiveness among boys^10,11.

The incidence of HC poisoning was highest during summer months compared to other seasons and this finding is common to most studies^2,7,11. Excess thirst seen during the hot months prompt children to consume any liquids that are available within their reach mistaking them to be water or cold drinks. In this study, all cases of poisoning were accidental and this could be because almost most HC poisoning occurred in children less than 5 years of age. The 15-year-old girl mistakenly consumed paint thinner assuming it to be water at their relative’s house.

In the present study, we observed that the incidence of paint thinner/turpentine poisoning was predominant involving three-fourths of HC poisoning in children and no cases of kerosene ingestion were observed. The primary reason could be explained by the place of study. This study was conducted in a hospital located in urban regions belonging predominantly to working and lower to middle socio-economic groups. Recent provisions of liquified petroleum gas (LPG) as cooking fuel to lower socio-economic groups at subsidized rates via a flagship scheme by the Indian government and ceasing provision of kerosene via ration have been attributed to fewer incidences of kerosene poisoning. Mosquito repellent and insecticide formed the remainder accounting for 21% and 3.5% cases of HC poisoning, respectively. We found that hydrocarbon poisoning seen in this cohort of children were not related to parental occupation, and hence cannot be perceived as an occupational hazard. In most cases, refurbishing of houses including painting and carpentry resulted in accidental poisoning as the leftover chemicals were stored in the houses and not discarded. This emphasizes the importance of raising community awareness.

The average time to presentation to the hospital emergency was one hour thirty minutes (range: half an hour to 3 hours 15 minutes) of exposure to HC poisoning, excluding 1 child who presented after 2 days of exposure. All cases belonged to urban areas and due to the prime location of the hospital, the average duration for presentation to ER is shorter in comparison to other studies^2,11. All cases except one (28 out of 29) came to the hospital within 2.5 hours of ingestion.

All children had at least one symptom. Gastro-intestinal involvement mainly in the form of vomiting (69%) was the predominant symptom observed followed by respiratory (38%). Neurological involvement was noted in only 13% of children. A few studies showed a higher prevalence of respiratory symptoms followed by gastrointestinal and lastly neurological, but these studies were mainly from Middle Eastern countries^6,9,12. Our study showed predominant gastrointestinal (vomiting) symptoms followed by respiratory and lastly neurological, which is similar to a study from eastern India². This finding could be due to the cultural practice of inducing forced emesis in any patient with alleged poisoning in most parts of India and hence the definite etiology of vomiting occurring as a symptom of poison ingestion or as a treatment methodology based upon the cultural practice, cannot be distinguished. A few studies showed that vomiting after HC ingestion significantly correlated with pneumonitis^5,6,9. We did not find any increase in the risk of pulmonary complications after vomiting, similar to the study by Dudin et al.¹³.

Twelve (41.4%) patients required respiratory support in our study. In a study by Madboly et al.⁶, 86.1% of patients were treated with supplemental oxygen for respiratory distress.

Antibiotic therapy was initiated in 4 patients for suspected secondary infection; however, this was continued in only 2 patients. The incidence of antibiotic use ranged from 31% to 40% in most studies^2,6,9. A South African study of 111 children aged below 5 years showed that secondary bacterial infection is rare (2 in 111) in paraffin poisoning, and all the patients recovered completely without antibiotics¹³. The low rates of antibiotic usage in our study could be due to strict adherence to protocols and most of our patients did not have severe illness as shown by the low hydrocarbon severity scores. This is supported by a review done by Das et al.¹⁴ where it was found that there is no evidence for prophylactic steroids or antibiotics in pediatric acute HC poisoning, but may be warranted in the presence of pre-existing risk factors or pneumonitis and/or established secondary infections. Only one patient who presented after 2 days in severe respiratory distress due to consolidation and bilateral pleural effusion received steroids.

The low HC severity scores of all 29 patients in our study were accompanied by no deaths, which is consistent with previous research indicating extremely low death rates^6,9,14-17. Most patients were discharged home within 24 hours of admission. As part of standard assessment and protocol including early resuscitation with minimal and as-required investigations and the policy of early discharge, the mean duration of stay was less than that in other studies². The low HC severity scores of patients have also contributed to the shorter duration of stay.

CONCLUSIONS

Acute hydrocarbon poisoning in children is a prevalent and preventable household hazard among the low and middle socio-economic classes in India. Majority of such poisonings are accidental in children and are most common in the toddler age group. Following unit protocols for the management of acute HC poisoning may avoid unnecessary investigations, unnecessary use of antibiotics/steroids, and promote early discharge.