Coin swallowing in children is a rather common event, and one that will not escape the experience of most busy clinicians. The clinical approach to a possible swallowed coin has been quite varied in the literature, and ranges from observing the patient at home to routine immediate radiographs in the emergency department.
Protocols are based on institutional bias and personal experience, and there appears to be no consensus in the literature on the most scientific approach. I suggested that all children who possibly swallowed a coin should have an x-ray to determine whether it is lodged in the esophagus, the only place in the GI tract that causes concern. The literature is clear, however, that the presence of a coin in the esophagus is often clinically silent (asymptomatic), and you just can't tell where it is by the child's symptoms or clinical appearance.
Young children or infants who swallow a coin can have airway compression from the esophageal foreign body, simulating epiglottis, croup, asthma, nonspecific vomiting, or feeding abnormalities. There are rare case reports of children being treated by numerous physicians for respiratory or gastrointestinal conditions who are readily cured when the radiograph ferrets out the presence of an unsuspected esophageal coin.
Although the diagnosis of an impacted coin usually can be made with a single AP radiograph, one that includes the stomach and nasal pharynx, multiple coins stacked on top of each other can elude the clinician unless a lateral film also is obtained. Alternatively, if a coin is removed, a follow-up radiograph should be obtained to make sure that it was a single object. When the coin has reached the stomach, it is essentially unprecedented that it will cause problems, and such children are often lost to follow-up and certainly lost to the medical literature.
This month's column discusses various clinical approaches available to the clinician faced with a child with a documented coin in the esophagus. The researcher will quickly become frustrated by suggested approaches from the medical literature, which range from immediate endoscopic removal under general anesthesia to sending the child home to eat a normal meal. Local protocols have developed that are followed religiously without corroborating scientific data, relegating almost any approach to standard of care. The impacted coin in the esophagus remains in the realm of the radiologist, otolaryngologist, gastroenterologist, or pediatrician. This column attempts to place this foreign body also in the realm of emergency medical practice.
The Spontaneous Passage of Esophageal Coins in Children
Soprano JV, et al
Arch Ped Adol Med 1999;153:1073
This is a retrospective chart review of 116 patients (7 months to 13 years) who had x-ray documentation of an esophageal coin. The case was termed “simple” if there was a single coin, if it were present for less than 24 hours, if the child had minimal or no symptoms, and if there was no underlying esophageal pathology. All others were termed “complex.” Simple coins were in the proximal (64%), middle (8%), and distal esophagus (26%). Complex coins were in the proximal esophagus (84%) and middle esophagus (16%).
Follow-up x-rays in the ED revealed that 28 percent of esophageal coins spontaneously passed (no specific time frame or intervention mentioned). The initial site of impaction did not prognosticate whether a coin would pass spontaneously. In “complex” cases (excluded from the “simple” group), none of the coins spontaneously passed. The authors note that regardless of the location of the coin, about one-fourth of simple foreign bodies will pass spontaneously. This suggests that a 12- to 24-hour observation period should be employed before proceeding to invasive procedures.
Comment: This report adds more distance between those who don't even bother to get an x-ray and those who always immediately remove a foreign body when it is found. One wonders how many of the “simple” coins would have passed if they were left alone for a longer period of time or if the child had been allowed to drink or eat. I do not think anyone would eschew coin removal in patients with underlying esophageal injury, such as a previous caustic ingestion.
Any child with respiratory distress should have the coin promptly removed, and if time permits, this should be done in the OR. Using a laryngoscope and forceps in the ED may be your only option in some cases, but this child will be difficult to restrain and unlikely to cooperate with this procedure. While coughing and gagging may be a common initial event, persistent respiratory distress is fortunately very rare and probably limited to a young child and a large coin.
I have personally seen two young children rushed to the ED in significant respiratory distress. One vomited the coin as I attempted to examine him. In the other I was able to flick out the coin from the hypopharynx with my index finger while the father held the child upside down. I have seen no reports of the Heimlich maneuver being used for esophageal coins, probably because they so rarely cause acute respiratory symptoms. In such cases, however, parents will likely first think that the airway is blocked, not merely compressed by the esophageal coin.
The clinical approach to a known esophageal coin had not been prospectively studied in detail. This is fertile ground for emergency medicine research, and I was surprised that I was not able to find a single prospective study evaluating a protocol designed to address this common problem. There are two scenarios where the course of action is clear. In the first instance, the patient is coughing, in respiratory distress, or is drooling and unable to swallow. This condition is due to an upper esophageal coin, and it necessitates urgent removal, either with a Foley catheter or endoscope.
Once swallowed, a coin is usually out of the reach of your fingers or the laryngoscope. When the coin is in the mid- to lower esophagus, it doesn't cause acute airway obstruction, so emergent maneuvers should not be required. In the other scenario, the foreign body is in the stomach or intestines, and these patients can be sent home for elective follow-up. I prefer to have parents check the stool for the coin, and if it is not found within a week, a repeat radiograph is suggested. The parents and family doctor usually forget about it, and assume, probably rightly so, that it was passed unnoticed in the stool.
Children with asymptomatic esophageal foreign bodies present the most difficult treatment scenario. There is relatively good evidence that suggests one can safely wait 24 hours before progressing to invasive treatment. Retrospective data indicate that the majority of coins, probably 70 to 80 percent, will spontaneously pass if left alone for a few hours. Surprisingly, in many studies a coin is removed as soon as the diagnosis is made without waiting to see if it would pass spontaneously.
My approach is to allow the asymptomatic child to eat and drink (usually peanut butter crackers from the vending machine and a carbonated soft drink), and then x-ray the patient again in a few hours, sooner if they tell you the foreign body sensation has gone away. I have tried putting nitroglycerin paste on the skin to promote esophageal relaxation, but I don't know if this is helpful. Using a technique gleaned for an impacted food bolus in the distal esophagus, intravenous glucagon has a theoretical benefit and would likely be helpful in a small number of cases. Glucagon has not been studied carefully for facilitation of the passage of this smooth foreign body, but there should be little downside. The drug would only be theoretically helpful for coins at the GE junction.
If the coin does not pass within three to four hours of ED observation, I will then contact my pediatric/ENT consultant and allow him to make the decision. Most of the time they opt for admission to the hospital with re-evaluation 12 to 24 hours later and subsequent endoscopy if the coin has not passed. (Hopefully they will repeat the x-ray just before they put the child under general anesthesia.) A case can be made for the Foley catheter technique to be used immediately for coins of upper esophagus, but I initially try to have the child drink even with foreign bodies in this area.
Data indicate that the initial location of a coin will not predict whether it will pass spontaneously. If the patient can't drink or continues to vomit, I call for help. Coins in the upper esophagus are a bit more worrisome so if they do not begin to move, I get a bit more nervous. Coins in the mid- to lower esophagus buy you some time and may pass in a few hours, especially after eating or drinking. I am more patient with these children.
While home observation seems strongly supported by the literature, I have not been routinely discharging patients of even reliable parents. I greatly respect esophageal foreign bodies, and it's too easy for the parents to wait three to four days while the asymptomatic coin becomes further imbedded in the esophageal mucosa.
Once you consult ENT or gastroenterology, you are probably guaranteeing esophagosphy in lieu of more conservative measures that seem to be favored by pediatricians. Radiologist and emergency physicians have experience with the Foley catheter technique, but this procedure is tricky and should probably only be used with coins that have been impacted less than 12 hours and are in the upper esophagus. Unlike an impacted bolus of meat in an adult, underlying esophageal pathology is not usually suspected when a coin gets hung up. An evaluation of the esophagus is not indicated for a childhood esophageal coin, but is for adults who cannot pass a bolus of food.
A Literature Based Comparison of Three Methods of Pediatric Esophageal Coin Removal
Pediar Emerg Care 1997;13(2):154
This is a computerized decision analytical model that was developed to determine the cost of a number of esophageal coin removal techniques. While gathering the data, the authors obtained significant information on current practices, complications, and methods of removal. To obtain the data, the authors reviewed 24 papers that reported the clinical course of 1746 esophageal coin removal attempts. The three interventions included endoscopy, Foley catheter removal, or bougineage.
Endoscopic removal, with either a rigid or flexible scope under general anesthesia in the operating room, was the most commonly used intervention. It occurred in approximately two-thirds of the cases. It was 100 percent successful, and there was a small incidence of complications. The most common complication was an exacerbation of previous lung disease, probably asthma, and stridor, which were likely secondary to the scoping procedure or intubation.
The advantage of endoscopy is that the foreign body is removed under direct vision, other foreign bodies may be found, and the underlying integrity of the esophageal mucosa can be directly observed. In addition, it is a controlled situation handled by doctors who would not be expected to break away for an ambulance patient in the middle of the procedure. As expected, the cost of endoscopic removal was the highest, more than four times that of a Foley catheter removal or bougineage.
About one-third of the patients had their coins removed by the Foley catheter technique. This was typically performed under fluoroscopic guidance by a radiologist. It included the passage of an uninflated Foley catheter beyond the coin, inflation of the balloon with contrast material, and withdrawal of the balloon in attempts to pull the coin into the mouth. The disadvantage of this technique is that it does not allow for direct inspection of the esophagus, and may cause inadvertent esophageal injury. However, in this series, there were only sporadic complications, and it is the procedure of choice in some institutions. The specific complications of Foley catheter removal were not detailed.
A small number of children (83 of 1746) underwent bougineage removal. This consisted of blind but careful passage of a lubricated esophageal dilator into the esophagus with an attempt to push the coin into the stomach. This appeared to be a rather haphazard technique. Some would argue that this is a poorly thought out technique that has the potential for esophageal injury. However, in this series there were no complications reported.
The authors note the personal and institutional bias for the method chosen for the removal of esophageal coins. One shortcoming of this study is that it did not indicate where in the esophagus the coin was lodged, how long the doctors waited, or if other manipulations were used. Certainly endoscopic removal in the operating room would be expected to be the most expensive method, but it is preferred because it is well known and well controlled. Complications with all techniques were minor, and usually resulted from manipulation of the airway, rather than by the coin removal procedure itself. The authors therefore conclude that all three methods appeared quite safe; however, they note that reticence to publish unfavorable results may have limited the number of failed procedures or complications. Overall the retrospective nature of all the papers that were reviewed limited the interpretation of the data.
Comment: When approaching the child with a esophageal coin, a number of caveats must be emphasized. Unfortunately, most literature does not specifically clarify the area of the obstruction or length of impaction. It is interesting to note that no child was observed for a prolonged period of time, and apparently they were not allowed to eat or drink. It appears that the reflex action of the surgical subspecialties is to immediately remove an esophageal coin, even though experience dictates that a good percentage of them will pass within a few hours, and it is safe to observe children for such a length of time.
Most institutions have developed a protocol for the treatment of children with suspected or known esophageal coins. Emergency clinicians should work with other departments, and come to some agreement on a rational prospective approach. It seems most reasonable to first document the presence or absence of an esophageal coin. Once a coin's location has been found in the esophagus, the level of impaction and length of time it has been in the esophagus will dictate the proper approach.
It is difficult to argue with routine endoscopic removal under general anesthesia, but this may be overly aggressive in a large number of cases. If the parents bring the child in within an hour or two of swallowing the coin, it makes sense to allow nature to take its course with a hope that it will pass spontaneously. As long as the coin is not causing any airway distress, it is reasonable to allow the child to drink a carbonated beverage and some solid food in an effort to remove it with the help of peristalsis. The feeding and observation technique can probably persist for 12 to 24 hours, and such observation can occur at home or in the hospital. If a coin has not passed within 24 hours, it is unlikely to do so, and some type of removal must be undertaken.
The Foley catheter technique is somewhat difficult to master, requires a fair amount of time, and is best reserved for coins in the upper esophagus. This is one technique I will give to a radiologist. I would argue against consideration of the bougineage technique in the ED because it is a blind procedure that has a potential for esophageal perforation. It just doesn't seem to make sense with today's technology. Endoscopic removal appears to be the most logical choice for stubborn coins although it may require transfer to a pediatric hospital. Granted, esophagoscopy is quite expensive and requires general anesthesia and tracheal intubation. The obvious advantage of endoscopy is that it allows removal under direct vision, and the safety, high success rate, few complications, and evaluation of the esophagus probably outweighs the costs issue.
It would be nice to see a prospective study that allows for a more reasonable period of observation and some less invasive techniques, particularly allowing the child to eat. Some of those coins that are urgently removed with an endoscope probably would have passed if the clinician were more patient.
Any child in pain should obviously be given analgesics, preferably judicious parenteral narcotics. I have not seen recommendations on the use of nitroglycerin or glucagon, but these interventions have been modestly successful for the treatment of an impacted food bolus.
I personally have not seen enough cases to report a series, but I have had more than half of my patients with a coin in the esophagus pass it spontaneously — if they are allowed to eat or drink. This statistic is in general agreement with the retrospective literature. Concerns that a small amount of food or carbonated soda would cause aspiration or contraindicate subsequent endoscopic removal are probably overdone. As long as one does not suggest that they eat a full-course meal, I see no downsides. Most children will tell you soon enough whether they will tolerate eating and drinking, and will refuse to do so if it causes them much distress. One would not expect such a child to aspirate any more than they would aspirate when they are vomiting with gastroenteritis.