SOURCE: Dept. of Otolaryngology, UTMB, Grand Rounds
DATE: December 13, 1995
RESIDENT PHYSICIAN: Michael Bryan, M.D.
FACULTY: Ronald Deskin, M.D.
SERIES EDITOR: Francis B. Quinn, Jr., M.D.
DISCUSSANT: Ronald Deskin, M.D.
The offending agents can largely be grouped into two categories, strong acids and strong bases. The exposure of the lining of the aerodigestive tract to these substances at the two extremes of the pH scale results in injury through coagulation necrosis (acids) or liquefaction necrosis (bases). Liquefaction necrosis typically occurs through direct extension and deeper injuries are the rule. Coagulative necrosis results in the formation of a somewhat protective coagulum layer at the site of injury and can limit the depth of injury, although full thickness injuries after acid ingestion can occur. The major variable that determine the extent of the injury are the amount and concentration of the ingested material, the form of the material (solid vs liquid), and whether or not emesis and or aspiration has occurred.
Alkaline substances, particularly drain cleaners containing NaOH and/or KOH, have historically accounted for the vast majority (60%-80%) of caustic ingestions. Prior to 1967, solid lye products accounted for most ingestions. Because they are tasteless and odorless, larger amounts of these substances can be ingested before the victim reacts. The solid form tended to result in more oropharyngeal and supraglottic injuries because of the limited ingestion quantities.
In 1967, the rate and severity of caustic ingestion injury increased markedly with the introduction of liquid drain cleaners containing strong alkali concentrations (30-40% NaOH). The severity and frequency of esophageal injury increased then because these agents can be swallowed in significant quantity before the patient reacts to the near immediate burn. Prior to this time, ingestion of solid crystalline or powder forms of these products tended to occur in smaller amounts because it was difficult to ingest much before severe pain ensued. The liquid forms also tended to cause longer more extensive circumferential burns of the esophagus because of its complete contact with the esophageal lumenal surface.
Subsequently, the Federal Hazardous Substance Act Poison Prevention Act, and the Safe Packaging Act imposed restrictions on the concentration (<10%) of these liquid products and established requirements for childproof packaging of hazardous substances. These measures contributed to a subsequent diminishment of injury frequency and severity. Nevertheless, the incredible potential for injury that can occur with even small amounts of low concentrations of alkali material and the continued availability of high concentration solid (powder or crystalline) alkaline drain cleaners warrant an aggressive and vigilant management strategy for these patients. Research has shown that only a brief exposure of dilute (10 seconds @ 3.8% strength) NaOH can injure mucosa and submucosa. Increasing the strength or exposure time increases the injury. 10% solutions of NaOH extend the injury to involve the muscular layers of the esophagus. Transmural injury occurs with a few seconds exposure to 22.5% NaOH.
Acids are more often associated with gastric injury, with less extensive injuries to the esophagus. The reasons for this are the somewhat protective coagulum that is formed at the site of esophageal injury, an inherent ability of the esophagus to withstand acids better than alkalis, and the collection of the acid in the stomach where it is sequestered. The stomach, while accustomed to a physiologic acidic environment, is not able to withstand high quantity exposure to strong acids, and full thickness injuries to the stomach and even the duodenum may result. Other factors that may account for the lower incidence of acid ingestions are that in this country the prevalence of strong acid products is lower than alkaline products, and the pungent odor and noxious taste of these compounds is more likely to repulse the potential victim. However, severe Esophageal injury and perforation have been reported after acid ingestion, and they must be considered as life threatening as alkali ingestions.
Other agents commonly cited as potentially damaging caustic agents include bleaches, "button" type alkaline batteries, phenol containing disinfectant cleaners, certain cosmetic agents (especially hair relaxers), ammonia, and non-phosphate detergents.
Bleaches are relatively pH neutral and are associated with a very low rate of severe injury. They are chlorinated compounds, and through oxygenation reactions form hydrochloric acid. If only household strength solutions (<<3% concentrations) are ingested, superficial burns result in almost all cases. Powdered bleaches are more dangerous because of increased transit time and prolonged contact in the pharynx and esophagus.
Button type alkaline batteries lodged in the esophagus must be treated aggressively with immediate endoscopy and removal because of the high rate of perforation if they remain in the esophagus for even brief periods. The mechanism of injury is thought to be primarily via the leakage of alkaline material from the batteries when a current set up in the esophageal wall between the anode and cathode of the battery. A significant percentage of battery ingestions occur among children who are hearing impaired and ingest the batteries from their own hearing aids.
Caustic injuries secondary to ingestion of cosmetic agents have recently been reported. These are primarily hair relaxers and contain NaOH or Ca(OH)2. For some reason these products have not been required to be distributed in childproof containers, and there have been several literature reports of severe injury due accidental ingestion by toddlers.
Household ammonia is usually lower in concentration than that needed to cause severe mucosal injury ($lt;4.5%), but scattered reports of such injuries do occur. Other household disinfectant cleaners (such as LysolTM) contain phenol, which is a strong alkali and can result in severe injuries when ingested in small quantities.
Non-phosphate detergents have become common because of environmental concerns, but have the potential for very serious injury if ingested. They contain "builders" that are usually silicates or carbonates that cause the pH to be very high, and can cause severe burns of the aerodigestive tract. In powder form, they frequently affect the pharynx and supraglottic areas, and are more apt to cause airway compromise than substances that move easily into the esophagus. Liquid non-phosphate detergents such as dishwashing and laundry detergents also have a high pH, and can cause severe esophageal injuries. Fortunately, non-phosphate detergents have a lower amount of titratable base than NaOH containing substances of similar pH and are therefore slightly less dangerous.
There are a host of other products that can result in esophagogastric injury, but the management principles after their ingestion are shared, with the exception of alkaline battery ingestion which mandates immediate removal unless it has passed through to the stomach in which case it is not removed.
In the absence of gross acute perforation, the ingrowth of newly forming blood vessels and migration of fibroblasts is in ongoing by days 2 through 4 post injury. After 4 to 7 days, there is completion of necrotic tissue sloughing, with early granulation ingrowth. The wound is clearly demarcated, and the risk of perforation is probably greatest at this time. At the end of the second week, the injured areas are filled with granulation tissue. Symptoms tend to resolve around this time and the contractile phase of the healing process begins. The reepithelialization process occurs slowly over the next 4 weeks, while vigorous collagen deposition by fibroblasts is resulting in thickening and scarring of the esophageal wall. The contractile process continues for months and frequently results in stricture formation in circumferential injury sites.
The areas of the aerodigestive tract most susceptible to injury are those where transit is delayed or passage is restricted. These areas correspond to the upper and lower esophageal sphincters, at the level of slight impression by the aortic arch, and the corresponding area of impression by the left mainstem bronchus. Conversely, the difficult the material is to peristaltically propel through the esophagus, the longer it will spend at any one location, and the greater the risk of transmural injury for a given substance. Powdered or crystalline substances are therefore very dangerous in the upper pharynx, supraglottis, larynx, and upper esophagus.
Delayed transit of these materials causes painful burns proximally in many cases and limits the amount of material that is ingested. The liquid compounds can pass through the esophagus and into the stomach by the time the victim begins to react to the injury. Even with rapid transit, the relatively complete exposure of the surface of the esophageal lumen to liquid boluses can result in extensive burns. Reactive emesis results in repeated exposure of the mucosal surfaces to the caustic, exacerbating injuries. Aspiration is obviously another life threatening event whether occurring with initial ingestion or at the time of emesis.
The classification of injury pattern is not universally agreed upon and multiple authors have proposed their own. Probably the most clinically useful one simply borrows from the categorization of cutaneous burns, with first, second, and third degree classifications. First degree injury involves the mucosa and does not result in sloughing or ulceration. This cannot distinguished clinically form more serious injuries, as these patients can still have dysphagia and odynophagia. It is a non-ulcerative esophagitis. Second degree injury results in loss of mucosa, variable loss of submucosa, with possible extension into the muscularis, but some muscularis remains viable and periesophageal tissues are not injured directly.
On examination, the surrounding mucosa is intensely erythematous, and the wound is usually covered with a whitish exudate after sloughing of the mucosa. These injuries almost always produce significant symptoms, and often result in stricture formation if circumferential. In third degree burns, the entire esophageal wall is necrotic. Extension into the periesophageal connective tissue planes is variable. The lumen of the esophagus may be obliterated by necrotic debris and spontaneous perforation can occur. The clinical appearance is of a dusky, red to black peripheral mucosa, with deep ulcerations and no identifiable muscular esophageal wall. These wounds cause symptoms similar to second degree injuries unless perforation occurs resulting in mediastinitis, peritonitis, and/or sepsis.
If a full thickness injury is suspected on the basis of evidence of peritonitis, fever, severe chest pain (especially with a widened or air containing mediastinum on CXR), and/or hypotension, early surgical exploration and intervention is recommended(see "Surgical Intervention" below). Emergency exploratory surgery is warranted if suspicion of perforation is high.
When there is not immediate evidence of perforation, management of the patient from this point forward is not universally agreed upon. In a recent review of the subject, Kikendall asserts that "...the conscientious physician who desires to determine the most appropriate management for these patients can find almost as many different therapeutic recommendations as articles on the subject." One basic principle is that initial intervention is primarily to stabilize and assess the patient. With rare exception, by the time the patient is in the emergency room, it is too late to change the injury pattern or severity.
Contrast esophagography is not helpful in the early stages of acute injury unless there is strong suspicion of perforation, or it is anticipated that endoscopy cannot be performed for some reason. Focal partial thickness injury not usually demonstrable by esophagography in the first 24-48 hrs, and it offers little useful information unless there is strong suspicion of severe full thickness injury of extensive length. In this event, a characteristic dilated atonic esophagus can confirm the full thickness injury, or extravasation in cases of perforation. Because of the possibility of perforation, water soluble contrast should be used if contrast esophagography is done.
If surgical intervention is imminent, some advocate proceeding with pre-operative endosopy based on the premise that little additional morbidity can result, and surgery may be be post- poned if the endoscopic examination fails to yield evidence of full thickness injury. Clinical evidence may be so overwhelming that this may not be justified, but in cases where there is less certainty, it is warranted.
All caustic ingestion victims who have any signs or symptoms of injury should undergo endoscopy. Some have advocated no endoscopy for the asymptomatic patient without evidence of oropharyngeal injury. This is based on the belief that in the absence of any symptoms, the injury is almost certainly superficial and no benefit will be gained by endoscopy. In some centers these patients are often treated very conservatively and discharged after 24 hours of observation if no symptoms have appeared. Remembering that up to 20% of patients without any oropharyngeal injury can have esophageal injury, and that there is poor correlation between symptomology and injury in partial thickness burns (which can result in stricture formation), there is a legitimate argument for endoscopy in all cases.
An exception to the universal endoscopy rule of thumb is the patient who has ingested household liquid bleach, in which case the injury is rarely significant. There are very few reports of subsequent complications after household bleach ingestion.
The timing of initial endoscopy is not universally agreed upon, although most authorities recommend initial endoscopy between 24-72 hours after the injury. This is based on the assertion that it takes time (12 to 24 hours) for the injured areas of the esophagus and stomach to demarcate, allowing for meaningful evaluation. However, Wijburg and a few others advocate proceeding with endoscopy as soon as possible. One benefit of immediate endoscopy is the possibility of early discharge in cases where there is no evidence of injury. There are no controlled studies demonstrating the comparative effectiveness and diagnostic accuracy of very early endoscopy versus delaying 1 to 2 days after caustic ingestion injury.
There is no consensus on the role of flexible endoscopy versus rigid endoscopy except that only flexible endoscopy will allow examination of the stomach, and even proximal small intestine if indicated. Flexible endoscopy can also be done without general anesthesia in many cases. Clinical reports support the use of either method in the hands of an experienced endoscopist. Otolaryngologists typically prefer the rigid scopes, while gastroenterologists and general surgeons typically prefer flexible scopes based on their training experiences.
The extent of the examination is another area of controversy. Until recently, most experts advocated stopping the endoscopy at the first significant circumferential burn, for fear producing a perforation if endoscopy proceeded. In a review of a series of 9 patients who suffered accidental caustic ingestion simultaneously, Thompson pointed out that numerous distal severe injuries would not be detected using this protocol. Several authors have since reported routinely successful endoscopy of the entire esophagus and stomach despite circumferential burns in the more proximal esophagus. The key is to only go where there is definite lumen, and not to blindly advance the endoscope at all. Again, proponents of flexible and rigid endoscopy disagree over the best method, but there appears to be growing consensus that if great caution is exercised by an experienced endoscopist, full length endoscopy is possible in all partial thickness cases. Advancing past a full thickness injury is not advisable.
In examining the esophagus, the main purpose is to delineate injury location and severity, since this affects subsequent treatment. The consensus seems to be that the experience and skill of the observer are the most reliable indicator of the accuracy of the endoscopic evaluation.
Periodic contrast esophagography is recommended by most experts, even in those patients with no symptoms, because early detection and intervention are beneficial. Certainly, if symptoms do evolve, more frequent examinations and intervention are indicated. Esophageal dilation is the mainstay of management, and patients may have to undergo many procedures to maintain patency and minimize dysphagia. Both anterograde and retrograde esophageal dilation are widely used. Retrograde is felt to carry less risk of esophageal perforation and is probably better tolerated in an awake patient. It does however necessitate a gastrostomy tube to be placed and maintained.
Some patients fail to respond to dilation and require gastrostomy, or esophagectomy and reconstruction with colonic or jejunal interposition grafts.
The timing of routine follow up contrast studies is tailored to the patient, but a reasonable recommendation is every three to four weeks for several months following caustic ingestion.
Gastric outlet obstruction is much less common, but can occur (especially after strong acid ingestion) resulting in need for partial gastrectomy
As with most accidents, prevention is the best form of treatment and parents, grandparents should be alert to what is in their kitchen cabinets, etc.. Pediatricians generally do a good job in informing families of these dangers. In the last twenty years a combination of legislation and public awareness has decreased the incidence of these terrible problems.
When you are alerted of a patient coming to or arriving in the emergency room with a caustic ingestion history, some attempts should be made by the family to recover the container and bring the labeled container to the emergency room so that the poison control center can help with identification of the substances and may even be able to tell you the pH of the substance. You can look at the clothes and skin of the patient sometimes and tell how strong the substance was and that there may be skin burns and actually burns through clothing.
The large study in the New England Journal of Medicine over eighteen years with sixty patients at Children's National Medical Center concluded that steroids were of no use but they did note that there were fewer esophageal replacement requirements on the steroid treated group. Our recommendation is to do endoscopy (including direct laryngoscopy, bronchoscopy, and esophagoscopy) within the first forty- eight hours. Waiting twelve to twenty-four hours may help to identify the area and extent of damage. The advantage of a flexible esophagoscope exam may include a good look at the stomach also.
Patients with esophageal burns should be watched through out their lifetime for the chance of increased incidence of esophageal cancer. This cancer usually occurs at an younger age and is equal in males and females as opposed to the more usual esophageal cancer that is more prevalent in males.
With a button battery ingestion, if the battery is not stationary on repeat x-ray then a short course of observation may be safe. When these button batteries are lodged in the nose or ears, the use of ear drops may hasten the release of caustic material from the battery case.
If a patient with a caustic ingestion is seen forty-eight hours or more after the ingestion, the initial diagnostic work-up should not include esophagoscopy but a barium swallow is done instead
Andreoni B; Marini A; Gavinelli M; et al Emergency management of caustic ingestion in adults. Surg Today 1995;25(2):119-24
Arif A; Karetzky MS Complications of caustic ingestion. N J Med 1991 Mar;88(3):201-4
Byrne WJ Foreign bodies, bezoars, and caustic ingestion. Gastrointest Endosc Clin N Am 1994 Jan;4(1):99-119
Doolin EJ Composite reconstruction of the esophagus and hypopharynx after severe caustic injury. Ann Otol Rhinol Laryngol 1994 Jan;103(1):36-40
Ferguson MK; Migliore M; Staszak VM; Little AG Early evaluation and therapy for caustic esophageal injury. Am J Surg 1989 Jan;157(1):116-20
Forsen JW; Muntz HR Hair relaxer ingestion: a new trend. Ann Otol Rhinol Laryngol 1993 Oct;102(10):781-4
Friedman EM Caustic ingestions and foreign bodies in the aerodigestive tract of children. Pediatr Clin North Am 1989 Dec;36(6):1403-10
Gumaste VV; Dave PB Ingestion of corrosive substances by adults. Am J Gastroenterol 1992 Jan;87(1):1-5
Gundogdu HZ; Tanyel FC; Buyukpamukcu N; Hicsonmez A Conservative treatment of caustic esophageal strictures in children. J Pediatr Surg 1992 Jun;27(6):767-70
Kikendall JW Caustic ingestion injuries. Gastroenterol Clin North Am 1991 Dec;20(4):847-57
Riding KH; Bluestone CD Burns and acquired strictures of the esophagus In Pediatric Otolaryngology, third edition; Vol 2. Ed: Bluestone CD, et al W.B. Saunders Co.; Philadelphia, PA; 1996
Rubin MM; Jui V; Cozzi GM Treatment of caustic ingestion. J Oral Maxillofac Surg 1989 Mar;47(3):286-90
Scott JC; Jones B; Eisele DW; Ravich WJ Caustic ingestion injuries of the upper aerodigestive tract. Laryngoscope 1992 Jan;102(1):1-8
Stenson K; Gruber B Ingestion of caustic cosmetic products. Otolaryngol Head Neck Surg 1993 Nov;109(5):821-5
Thompson JN Corrosive esophageal injuries. A study of nine cases of concurrent accidental caustic ingestion. Laryngoscope 1987 Sep;97(9):1060-8
Thompson JN; Browne JD Caustic ingestion and foreign bodies in the aerodigestive tract. In Head and Neck Surgery - Otolaryngology. Ed: Bailey BJ, et al. J.B. Lippincott Co.; Philadelphia, PA; 1993.
Wijburg FA; Heymans HS; Urbanus NA Caustic esophageal lesions in childhood: prevention of stricture formation. J Pediatr Surg 1989 Feb;24(2):171-3