During the study period, 271 children presented with the main complaint of foreign body misingestion. Of these, 183 were prospectively evaluated using the scoring system, and 25 children (15 cases of tobacco, 8 of drugs, 1 of liquid soap, and 1 of nail polish) were excluded (Figure). Tobacco or drug misingestion was excluded from the study because of concerns about intoxication, which was not suitable for our scoring system. Some e-cigarettes contain small metal plates to increase the heat transfer efficiency. Such metal plates pose a potential risk of gastrointestinal injury, and children who accidentally ingested them were included in the analysis. The included 158 children, 86 were boys and 72 girls, ranging in age from 5 months to 15 years (median, 2 years and 1 month); 9 were 10 years or older, and 4 (5, 8, 11, and 13 years) had mental retardation. Residents performed all scoring evaluations; 98 children were classified as Class A (follow-up at home), 47 as Class B (consultation with a pediatrician), and 13 as Class C (consultation with a pediatrician and contrast-enhanced CT). After a detailed interview and radiographic examination, some children were presumed to not have ingested a foreign body (Table 2).
Table 2
Results of classification using the scoring system
| Ingested objects | Total | Spontaneous excretion | Enema | Endoscopy | Surgery |
Class A | Glass marble | 8 | 6 | 2 | | |
N = 98 | Coin | 7 | 3 | | 4 | |
| Single magnet | 7 | 7 | | | |
| Plastic piece | 7 | 7 (5)* | | | |
| Paper clip | 4 | 4 | | | |
| Toy | 3 | 3 | | | |
| Absorptive object | 1 | 1 | | | |
| Screw | 1 | 1 | | | |
| Ring | 1 | 1 | | | |
| Fruit pit | 1 | 1 | | | |
| No ingestion | 58 | | | | |
Class B | Button battery | 9 | 3 | | 6 | |
N = 47 | Sharp object | 9 | 1 | 2 | 6 | |
| Coin | 6 | | | 6 | |
| Multiple magnets | 4 | 2 | 1 | 1 | |
| Plastic piece | 2 | 2 | | | |
| Absorptive object | 1 | 1 | | | |
| Stone | 1 | | | 1 | |
| No ingestion | 15 | | | | |
Class C | Button battery | 2 | | | 2 | |
N = 13 | Coin | 2 | | | 2 | |
| Single magnet | 1 | | | 1 | |
| Multiple magnets | 1 | | | | 1 |
| Metal plate | 1 | | | 1 | |
| No ingestion | 6 | | | | |
* Numbers in parentheses indicate children who did not visit our hospital the following day. |
In Class A, 58 of 98 children were not diagnosed with foreign body ingestion. The remaining 40 children had a median age of 3 years and 1 month (range, 8 months to 10 years), and 20 were male. The ingested foreign bodies were commodities and toys, including a glass marble in eight children, a coin in seven, a single magnet in seven, and a plastic piece in seven. All patients were instructed to visit our hospital the day after an overnight observation at home. However, five children did not visit our hospital. Their outcomes were unknown; however, none subsequently visited our hospital for gastrointestinal injuries, suggesting that the foreign body was spontaneously excreted. Four children underwent endoscopic removal of the coin a few days later because the coin remained in the stomach and was unlikely to be spontaneously excreted. The other two children who ingested glass marbles required an enema.
In Class B, 15 of the 47 children were not diagnosed with foreign body ingestion. The remaining 32 children had a median age of 2 years and 11 months (range, 5 months to 15 years), and 13 were male. The ingested foreign bodies included button batteries in nine children, sharp objects in nine, coins in six, and multiple magnets in four. Endoscopic removal was performed in 20 children; the majority of the children who ingested button batteries, sharp objects, or coins underwent endoscopic removal. Enemas were performed in two children who ingested sharp objects and one with multiple magnets. Plastic pieces and absorptive objects are excreted spontaneously. Eventually, all foreign bodies were excreted or removed without gastrointestinal injury.
In Class C, six of the thirteen children were not diagnosed with foreign body ingestion. The remaining seven children had a median age of 1 year (range, 9 months to 12 years), and five were boys. The ingested foreign bodies included a button battery in two children, a coin in two, a metal plate from an e-cigarette in one, a single magnet in one, and multiple magnets in one. All children underwent invasive treatment, whereas contrast-enhanced CT was not performed in any of the children, despite our scoring system being designed to consider contrast-enhanced CT. Endoscopic removal was performed in six children. The ingested objects were removed without causing gastrointestinal injury. Surgical removal was necessary in a child who have ingested multiple magnets, complicated by gastrointestinal perforation.
The rate of invasive treatment was compared between Class A (observation) and Classes B and C (considering treatment). Invasive treatment (endoscopic and surgical removal) was performed in four of 84 children in Class A and 27 of 60 children in Classes B and C (Table 3). The rate of invasive treatment was significantly higher in Classes B and C than in Class A (P < 0.001).
Table 3
Direction by the scoring system | Total | Invasive treatment | No invasive treatment | P value |
Observation at home (Class A) | 98 | 4 Endoscopy 4 | 94 Spontaneous excretion 34 Enema 2 No ingestion 58 | < 0.001 |
Consideration of interventions (Classes B and C) | 60 | 27 Endoscopy 26 Surgery 1 | 33 Spontaneous excretion 9 Enema 3 No ingestion 21 | |
Table 4 compares the results of our scoring system with those of the NASPGHAN algorithm recommendations. Eleven of the seventy nine patients underwent treatments different from those recommended by the NASPGHAN algorithm. According to the NASPGHAN algorithm, observations were recommended for five children (Cases 1, 3, 8, 10, and 11). Four of them were consequently treated with glycerin enemas, and one underwent colonoscopy. In the third patient, colonoscopic removal was necessary because the magnet had stagnated in the colonic folds. In contrast, the NASPGHAN algorithm indicated endoscopy in six children (Cases 2, 4, 5, 6, 7, and 9). Only one of them (Case 4) received a glycerin enema, while the others were observed with no treatment. Endoscopic removal was difficult in two children with button battery ingestion and in one with multiple battery ingestion because the objects were found in the intestines.
Table 4
List of cases in which the treatment was different from the NASPGHAN algorithm
Case | Age (yrs) | Class | Ingested objects | Treatment | NASPGHAN algorithm | Outcome |
1 | 2 | A | Glass marble | GE | Observation | Excreted |
2 | 1 | A | Absorptive object | Observation | Endoscopy | Excreted |
3 | 4 | A | Glass marble | GE | Observation | Excreted |
4 | 2 | B | Multiple magnets | GE | EGD | Excreted |
5 | 2 | B | Button battery | Observation | EGD | Excreted |
6 | 0 | B | Button battery | Observation | EGD | Excreted |
7 | 3 | B | Pick (used for lunch box) | Observation | EGD | Excreted |
8 | 1 | B | Earrings | GE | Observation | Excreted |
9 | 2 | B | Absorptive object | Observation | EGD | Excreted |
10 | 0 | B | Metal plates | GE | Observation | Excreted |
11 | 0 | C | Single magnet | Colonoscopy | Observation | Removed |
GE: Glycerin enema, EGD: esophagogastroduodenoscopy |
Finally, we retrospectively classified 88 children who were not evaluated using our scoring system despite suspicion of foreign body ingestion. Among them, ingestion of foreign bodies was not confirmed in 73 children; eight children were classified as Class A, and the other seven were classified as Class B. Five of the seven children retrospectively classified as Class B had undergone endoscopic removal of foreign bodies. The objects tested were button batteries (three children), metallic ball (three), plastic piece (two), and single magnets, multiple magnets, absorptive objects, screws, fish bones, rings, and nostril retainers (one child each).