General data
A total of 51 cases were included in this study. The mean age of these cases was 5.25 years, with a median age of 5.33 years, ranged from four months to ten years and 11 months. The proportion of male patients is higher than female patients as 70.6% (36/51). From 2015 to 2019, the number of cases increased (4,8,6,14,19 respectively; Fig. 1).
Clinical manifestations
The numbers of magnets ingested ranged from two to 24. Ingestion of two magnets accounted for the most as 41.2% (21/51) of the cases (Fig. 2). Among the 51 cases, the specific time of ingestion of six children is unknown. The average visiting time of the other 45 children was 46.75 hours after their ingestions, ranged from two hours to 14 days. Only 19 cases (37.3%) were symptomatic, including 15 cases of abdominal pain, 12 cases of vomiting, two cases of fever and one case of chest pain. Nine patients had positive signs of abdominal tenderness, and four of them had peritoneal irritation signs (Table 1).
Table 1
Clinical features and the outcomes
Symptoms, No. (%)
|
No
|
32(62.7%)
|
Yes
|
19(37.3%)
|
Abdominal pain
|
15(29.4%)
|
Nausea and vomiting
|
12(23.5%)
|
Fever
|
2(3.9%)
|
Chest pain
|
1(2.0%)
|
Signs, No. (%)
|
No
|
42 (82.4%)
|
Yes
|
9 (17.6%)
|
Abdominal tenderness only
|
9 (17.6%)
|
Peritoneal irritation signs
|
4 (7.8%)
|
Radiograph tests, No. (%)
|
X-rays
|
51 (100%)
|
Foreign bodies only
|
49 (96.1%)
|
Bowel obstruction
|
2 (3.9%)
|
CT scans
|
6 (11.8%)
|
Foreign bodies only
|
2 (3.9%)
|
Bowel obstruction
|
3 (5.9%)
|
Perforation
|
1 (2.0%)
|
Treatment, No. (%)
|
No intervention
|
18 (35.3%)
|
|
Endoscopy
|
10 (19.6%)
|
|
Surgery
|
16 (31.4%)
|
|
Endoscopy + surgery
|
7 (13.7%)
|
Outcomes, No. (%)
|
No adverse effects
|
26 (51.0%)
|
Mucosal injury
|
2 (3.9%)
|
Ulcer
|
1 (2.0%)
|
Magnets lodged in gastrointestinal tract
|
2 (3.9%)
|
Fistula
|
1 (2.0%)
|
Perforation
|
3 (5.9%)
|
Multiple perforations, adhesions and necrosis
|
16 (31.4%)
|
Clinical Management
All the patients were treated according to the process we developed and shown in the diagram (Fig. 3). All the 51 children had positive foreign body shadow on abdominal plain radiographs, of which two patients showed intestinal obstruction. Six cases also had CT examinations, which showed that three cases of intestinal obstruction, one case of intestinal perforation, and two cases of gastrointestinal foreign bodies only.
According to the radiograph, 17 cases’ magnets may still be located in the upper digestive tract, so gastroduodenoscopies were performed. By gastroduodenoscopy, the magnets of five cases were removed, of which two cases had no complications, one case had slight mucosal injury and two cases had perforation. In one of these two cases, the location of perforation was duodenum and then the patient was transferred to surgery. The perforation of the other case was in the lower esophagus. After ten days of local drainage and enteral nutrition through Nasojejunal tube, the perforation healed. In six cases, the magnets had passed through the pylorus and entered the lower digestive tract, and the gastric mucosa had mild injury in one of these six cases. In the other six cases, the magnets were still in the upper digestive tract, but they were embedded in the mucosa and could not be removed by gastroduodenoscopy. So, these six cases were transferred to surgery.
In this study, a total of 23 children underwent surgery, including 21 cases of laparotomy and two cases of laparoscopy combined laparotomy. After the operation, four patients stayed in ICU for monitoring for two or three days. 24 patients fasted for an average of 7.5 days ranged from two to ten days. 22 patients had gastrointestinal decompression for an average of 6.8 days ranged from two to nine days. 16 patients had abdominal drainage for an average of 7.8 days ranged from five to nine days. 11 patients had fever after operation for an average of 2.7 days ranged from one to six days. 22 patients used antibiotics for average of 11.8 days ranged eight to 19 days.
The patients were divided into two groups according to whether they were operated or not. Compared with the non-surgical group, the surgical group had more male cases (21/23 vs 15/28, p = 0.004), and the time of visiting was longer (80(72) vs 26(5), hours, p = 0.008). While, there was no significant difference in the mean age and the number of swallowing magnets (Table 2). This suggests that patients should see a doctor as soon as possible after multiple rare-earth digestion. The longer the time is, the more likely patients have to suffer from surgical operation.
Table 2
Difference between surgical and non-surgical group
|
Surgery(n = 23)
|
No surgery(n = 28)
|
P value
|
Age, years
|
4.6(3.8)
|
5.9(6.5)
|
0.155
|
Gender, No (%)
|
|
|
0.004
|
Male
|
21(91)
|
15(54)
|
|
Female
|
2(9)
|
13(46)
|
|
The number of magnets
|
6(4)
|
4(3)
|
0.184
|
Time after swallow, hours
|
80(72)
|
26(5)
|
0.008
|
Outcome
Among the 51 cases, the magnets passed naturally for 47.1% (24/51); were removed by endoscopies for 9.8% (5/51) and by surgeries for 43.1% (22/51). The complications include mucosal injury for two cases, ulcer for one case, magnets lodged in the gastrointestinal tract for two cases, fistula for one case, perforation for three cases and multiple gastrointestinal injuries for 13 cases. The multiple gastrointestinal injuries included perforations, adhesions, and necrosis. In the most serious case, there were 11 intestinal perforations, three intestinal necrosis, and extensive intestinal adhesion. In this study, 39 patients were hospitalized, the average length of stay was 8.8(one-nine) days. No death happened in these cases. We followed up these patients for at least two months. There were no long-term complications such as obstruction or infection.