Toxicology Rounds: Salicylate Toxicity: Avoiding a Disastrous Clinical Outcome

Gussow, Leon MD

doi: 10.1097/01.EEM.0000434478.04030.f2
Toxicology Rounds

Dr. Gussow is a voluntary attending physician at the John H. Stroger Hospital of Cook County in Chicago (formerly Cook County Hospital), an assistant professor of emergency medicine at Rush Medical College, and a consultant to the Illinois Poison Center. He is also the editor of his own blog, The Poison Review.

Article Outline

Treating salicylate poisoning is one of the trickiest and most difficult challenges in medical toxicology. The American College of Medical Toxicology recently released “Management Priorities in Salicylate Toxicity” that is full of clinical pearls. I've distilled them into eight key points, and they should help emergency physicians even though they certainly do not represent a complete discussion of all aspects of salicylate toxicity.

Figure. No caption a...
Figure. No caption a...

Watch the units. Salicylate levels are reported in different ways by different clinical laboratories. Our poison center sometimes gets a call for, say, an alarming salicylate level of 90 (therapeutic level=15-30 mg/dL), but the patient is described as asymptomatic. Further investigation reveals that the hospital's lab reports its levels in milligrams per liter. This is equivalent to 9 mg/dL, a subtherapeutic concentration. Be sure to check the units in each case of possible salicylate toxicity. Commonly used units are mg/dL, mg/L, or mmol/L. (10 mg/dL = 100 mg/L = 0.724 mmol/L.)

Do not count on only the anion gap to screen for possible salicylate toxicity. Some laboratory analyzers will report a falsely elevated serum chloride reading when high levels of salicylates are present. The calculated anion gap may be normal in that case.

Correct hypovolemia (which may be severe). “Salicylate-poisoned patients are almost universally volume depleted at the time of presentation to medical care (as much as 4-6 L in most symptomatic adults),” the ACMT said. Many effects of salicylate toxicity contribute to this dehydration, including vomiting, hyperventilation, sodium loss, hyperthermia, and diaphoresis. Elimination of salicylate will be impaired unless this volume loss is corrected, and it will be impossible to alkalinize the urine adequately. The ACMT document does not mention it, but this would be an ideal indication for bedside ultrasound. A small inferior vena cava that collapses with spontaneous inspiration would be an indication for aggressive fluid administration. The rate of fluid repletion can be slowed once the IVC became plumper and less collapsible.

Check levels frequently. Salicylate overdose is notorious for producing erratic absorption, unpredictable pharmacokinetics, and delayed clinical toxicity. The formation of a bezoar (especially after overdose of enteric-coated aspirin) can cause continued absorption and persistently increasing levels. A patient who presents with mild symptoms and only a moderately elevated salicylate level can deteriorate suddenly and unexpectedly. Many toxicologists recommend checking levels every two hours until they are clearly going down and the patient is improving clinically.

Decreasing salicylate levels are not necessarily reassuring. The ACMT noted that “clinical deterioration, even in the setting of a falling serum concentration, is ominous, suggestive of increasing central nervous system (CNS) salicylate concentration.” Evidence suggests that a key predictor of a poor outcome in salicylate overdose is the level of drug in the central nervous system. Even a small decrease in serum pH can cause a huge increase in the amount of drug going into the CNS because CNS uptake of salicylate is exquisitely sensitive to acid-base status. One must see a decreasing level combined with clinical improvement to be convinced that things are heading in the right direction.

The major goal of administering bicarbonate is to avoid acidemia, even more than alkalinizing the urine. “A slightly alkalemic blood pH limits an increase in CNS salicylate concentration by shifting the ionization equilibrium of salicylate in the blood to the charged form and decreasing passage of salicylate into the CNS,” the ACMT document said. Steven Curry, MD, pointed out in his chapter on salicylates in Critical Care Toxicology (Elsevier Mosby, 2005) that “primary efforts should be directed at keeping arterial pH greater than 7.4 in symptomatic patients” after initial hydration. “A reasonable goal,” he said, “is an arterial pH of 7.5 to limit increases in volume of distribution and tissue (e.g., CNS) salicylate concentrations.” Alkalinization of the urine to increase drug elimination is an important but secondary goal.

Consider the benefits and the risks of using multidose activated charcoal (MDAC). Data from human volunteers and animal models suggest that MDAC improves salicylate clearance, but no evidence suggests that this action improves clinical outcome. The patient in cases of severe overdose may have or develop decreased mental status and be unable to tolerate MDAC. MDAC in mild cases would not be necessary to achieve a good outcome. It may be difficult to predict how the case will play out. Certainly it is best to avoid the intervention if clinicians anticipate that the patient's clinical status will deteriorate or if a contraindication to MDAC exists (e.g., inability to protect the airway). Always remember that administering MDAC is never mandatory and never an absolute standard of care.

Endotracheal intubation of a patient with severe salicylate toxicity is an extremely high-risk procedure. Any salicylate overdose patient sick enough to require intubation will hyperventilate to blow off carbon dioxide and avoid extreme acidemia. Interfering with that process — whether by administering sedatives and paralytic agents or by not setting the ventilator to continue hyperventilation after intubation — can be disastrous. The ACMT document suggests that giving a bolus of IV sodium bicarbonate when intubating “in a sufficient quantity to maintain a blood pH of 7.45-7.50 over the next 30 minutes is a reasonable management option.” Unfortunately, this suggestion is not referenced, and I am not aware of evidence that supports it. I also have no idea how to calculate how much sodium bicarbonate would be required to achieve that target pH for half an hour.

The entire ACMT document is worth reading, and can be found on the organization's website:

Click and Connect! Access the links in EMN by reading this issue on our website or in our iPad app, both available on

Back to Top | Article Outline


* Visit Dr. Gussow's blog at

* Follow Dr. Gussow on

* Read all of Dr. Gussow's past columns at

* Comments about this article? Write to EMN at

© 2013 by Lippincott Williams & Wilkins