Characteristics of unintentional ingestion of oral non-steroidal anti-inflammatory drugs and analgesics in preschool children

Table 1 summarises the circumstances of accidental ingestion of single acute NSAID and analgesic doses in 466 preschool children in 2009–2019 (411 from CPCC records and 55 from CHS hospital records). No difference between the CPCC and CHZ cases was noted, except that CHZ patients were slightly older than the CPCC cases (p<0.001, Mann-Whitney U test).

In 401 of the total 466 cases (86 %) the pharmaceutical form of the medication was known. In 161 (40 %) it was a liquid formulation such as an oral suspension or syrup, in 230 (57 %) an oral tablet or capsule, and in 10 (2 %) effervescent tablets or powders. Liquid forms dominated in cases of dosing errors by parents/caregivers (17 of 19 dosing errors). In terms of proportion, this was a significantly more common situation than with 144 of 382 unsupervised children ingesting a liquid form left within their reach (89 % vs 38 %, respectively; Fisher’s exact test, p<0.001).

In the total study sample, the most frequent drugs ingested were ibuprofen (47 %), followed by paracetamol (20 %), ketoprofen (15 %), and diclofenac (11 %), with no significant difference between the CPCC and CHZ cases (Table 1). Ingested ibuprofen doses were above MDD (30 mg/kg) in 136 of 190 cases (72 %) with reported dose, while paracetamol doses were above the MDD (960 mg) in 39 of 84 such cases (46 %). As expected, ingested ibuprofen and paracetamol doses above the MDD were significantly higher in CHZ than CPCC cases, and so was the proportion of symptomatic patients (Table 2). However, ingestion of a toxic dose above 200 mg/kg of either ibuprofen or paracetamol was reported in only three and two cases, respectively. Consequently, only minor gastrointestinal symptoms (vomiting, diarrhoea) and some laboratory abnormalities were noted with paracetamol (Table 2), but no elevated serum liver enzymes or liver function abnormalities. All of the three symptomatic CHZ patients with determined blood plasma paracetamol levels were at low risk of hepatotoxicity according to the Rumack-Matthew nomogram (12). More specifically, one patient had 53 μg/mL (351 μmol/L) 4 h after and 3 μg/mL (20 μmol/L) 20 h after the estimated ingestion of 135 mg/kg of paracetamol. The second patient had paracetamol levels below the limit of detection [<0.26 μg/mL (<2 μmol/L)] 5 h after the estimated ingestion of 203 mg/kg, and the third patient had 12.59 μg/mL (83 μmol/L) and 1.83 μg/mL (12 μmol/L) four and 23 h after the ingestion of 155 mg/kg, respectively.

Adverse effects of diclofenac in the dose range of 4–25 mg/kg included vomiting, drowsiness, apathy, hyperactivity, pallor, hypothermia, high serum lactate and calcium. With ketoprofen (in the dose range of 4–49 mg/kg), adverse effects included vomiting, facial flushing, high serum lactate, blood urea nitrogen, and acidosis. For naproxen (43 mg/kg) only vomiting was reported.

We also analysed gastric decontamination and treatment information in a subset of 344 CPCC patients on whose behalf a family or emergency physician or nurse consulted CPCC (median time to CPCC call was 40 min, IQR 30 min – 1 h after ingestion). Twenty-seven (8 %) underwent gastric lavage, four of whom also received activated charcoal. Three patients received activated charcoal alone. Other information about symptomatic or supportive treatment was not available for this subset at the time of CPCC calls.

Among the 55 CHZ patients, decontamination by gastric lavage and activated charcoal was done in 15, while 16 received activated charcoal alone. Sixteen symptomatic patients [7 after ibuprofen, three after paracetamol (Table 2), three after diclofenac, and three after ketoprofen ingestions] received supportive treatment, but none required N-acetylcysteine for paracetamol overdose. One patient who developed acidosis after ingestion of 45 mg/kg of ketoprofen received bicarbonates and fluid replacement (normal saline) by intravenous (iv) infusion. The rest received supportive care, mostly in the form of iv fluid replacement (eight patients), or iv infusion of a proton pump inhibitor (six patients).

Our findings call for better prevention of accidental NSAID and non-opioid analgesic ingestion by young children and can serve as real-life examples to inform preventive action. In the great majority (94 %) of cases, the drugs were left within their reach. This is not surprising, given that these medications are often not stored safely (2, 14, 15) and that parents generally do not consider over-the-counter (OTC) medicines dangerous in terms of poisoning (16). Another unsafe habit contributing with up to 20 % of poisonings in children is forgetting to put medication away immediately after use (17).

Another possible scenario of unintentional exposure are dosing errors by parents or caregivers, as was the case in 6 % of our cases. This is similar to the 8 % reported in the USA for children aged 0–6 years overdosed with paracetamol (18). Most of these dosing errors were associated with the liquid form, and only a few cases were owed to parents giving children tablets intended for adults. On the plus side, liquid forms were involved in only one third of unsupervised ingestions, which is about half of what was reported in an earlier study investigating ingestions of OTC drugs by children (19).

Although measures such as child-resistant packaging help to prevent serious poisoning (20–22), more effort is clearly needed to raise awareness that these products are appealing to children and that parents should check the dose carefully when administering liquid formulations.

Children in our study mostly overdosed with ibuprofen and paracetamol, which is in line with data from other countries (5–6). Although our primary focus was not on the symptoms of poisoning in relation to the ingested dose, our findings confirm that exploratory ingestions (which usually involve doses above MDD but below the toxic threshold of 200 mg/kg) rarely lead to symptomatic poisoning. We draw this conclusion from the fact than no paracetamol hepatotoxicity was observed, which suggests that children ingested single doses below its toxic dose of 200 mg/kg (11).

It is interesting to note that doses of ibuprofen and paracetamol ingested by CHZ patients were significantly higher than those ingested by CPCC patients. This is expected, as parents/caregivers usually wait to see if there is any development with their child before they go a hospital and will sooner call a poison control centre for advice. Regarding other NSAIDs, the reported dose ranges of ketoprofen and diclofenac which were associated with symptomatic poisoning, predominately with gastrointestinal symptoms, contribute to the scarce evidence about toxic doses of these medications in children. In adults, a couple of grams of ketoprofen was associated with only minor effects (23).

In conclusion, since unintentional exposure of preschool children to NSAIDs and analgesics predominately occurred because the drug was left within their reach, more preventive effort is clearly needed to address this continuous problem of unsafe storage in Croatia. Our data should be used as a basis for planning preventive actions and as real-life cautionary messages. Also, given the fact that these exposures rarely lead to symptomatic poisoning, parents should be encouraged to first call their local/national poison control centre, emergency department, or their paediatrician to assess the option of home observation before seeking immediate medical attention. This could lower the burden on the healthcare system. On the other hand, it is important to refer to healthcare facility early after significant ingestions, and CPCC continues to support healthcare professionals with information on patient monitoring and treatment. We will continue to follow the situation, since unintentional exposure to substances available at home has increased during the COVID-19 pandemic (24).