Ingestion of cylindrical and button batteries: an analysis of 2382 cases.

During a 7-year period, 2382 cases of battery ingestion were reported to a national registry. Button cells were ingested by 2320 of these patients; 62 patients ingested cylindrical cells. These cases are analyzed to reassess current therapeutic recommendations, hypotheses about battery-induced injury, and strategies for prevention and intervention. Hearing aids were the most common intended use of ingested cells (952 cases, 44.6%); and in 312 (32.8%) of these cases, the battery actually was removed by a child from the child's own aid. Overall, 9.9% of patients were symptomatic. Two children experienced severe esophageal injury following the ingestion of large diameter cells and required repeated dilatation. In this series and in prior reports, most batteries which lodged in the esophagus and caused esophageal injury were large diameter (20 to 23 mm). However, neither battery diameter nor symptom occurrence could be used to detect all patients with esophageal battery position. Outcome was not affected by battery discharge state, but was influenced by chemical system. Lithium cells, with their larger diameters and greater voltage, were associated disproportionately with adverse effects. Mercuric oxide cells were substantially more likely to fragment, compared with other chemical systems. No clinical evidence of mercury toxicity occurred in this series, although one patient demonstrated minimal elevation of blood mercury levels. In this series and 10 prior reports, all patients with elevated blood mercury levels had ingested batteries which fragmented and showed evidence of radio-opaque droplets in the gut. Special monitoring may be advisable for individuals who have ingested 15.6-mm-diameter mercuric oxide cells. Ipecac was administered in 37 cases of button cell ingestion, causing battery expulsion in only one patient. Retrograde battery movement from the stomach to the esophagus necessitating emergent endoscopic retrieval occurred as a complication of ipecac administration in another patient. The use of endoscopic and surgical intervention declined more than fivefold during the study period. Endoscopic retrieval success was a function of battery location, with batteries in the esophagus more readily retrieved than those in more distal gastrointestinal locations (90.0% vs 46.7%, respectively). A current management protocol is presented advocating a noninvasive approach for most cases of button cell ingestion where an esophageal position is excluded. Manufacturers are urged to provide more securely fastened, child-resistant battery compartments on hearing aids as well as other battery-powered products in household use.