It probably didn't surprise anyone that the National Poison Data System identified salicylate toxicity as the leading cause of clinically preventable poisoning death. (Clin Toxicol 2016;54:441.) After all, exposure to salicylate alone or with other drugs is extremely common: Almost 25,000 cases of salicylate exposure were reported to U.S. poison control centers in 2014, with more than 200 of these resulting in major clinical outcomes. (Am J Emerg Med 2017;35:899.)
Unfortunately, it is easy to make errors in management when treating salicylate poisoning. The drug's toxicology is complex, its pharmacokinetics unpredictable. Patients who look stable and relatively well can deteriorate quickly as salicylate continues to be absorbed from the GI tract long after ingestion. Changes in blood pH facilitate the drug crossing the blood-brain barrier into the central nervous system and into mitochondria where it decouples oxidative phosphorylation, causing metabolic acidosis.
The physician treating a salicylate-poisoned patient is forced to make critical decisions. How much fluid should be administered to correct volume loss and maintain adequate renal perfusion? Is multidose activated charcoal an appropriate and a safe means of GI decontamination? When should the patient be intubated, if at all? Should these patients have hemodialysis?
Toxicologists generally agree that certain clinical findings indicate starting hemodialysis. (Curry SC, Spyres MB.  Salicylates. In Critical Care Toxicology, 2nd Ed. [pp. 1251-1267]. Cham, Switzerland: Springer; Emerg Med Clin North Am 2007;25:333.) These clinical markers represent evidence of significant end-organ toxicity or suggest that standard medical therapy is not working:
Neurotoxicity: Any sign of neurotoxicity is ominous and may presage rapid deterioration and imminent collapse. These findings include lethargy, delirium, agitation, coma, and seizure activity. Altered mental status suggests that nonionized salicylate is crossing the blood-brain barrier and entering the central nervous system, though it may also result from a low level of glucose in the CNS. These patients should receive supplemental dextrose, even if blood glucose is normal. Medical therapy alone, however, may not fend off the impending crash. Failing to initiate hemodialysis at the first sign of salicylate-induced neurotoxicity is a critical error.
Significant persistent metabolic acidosis or electrolyte abnormality: Hemodialysis provides many benefits for patients with acute salicylate toxicity. Salicylate is readily cleared by hemodialysis because it is water-soluble and has a low molecular weight and relatively low protein binding. Hemodialysis also treats chemical imbalances such as metabolic acidosis and hypokalemia. Acidemia can set off a vicious cycle in these patients, causing more nonionized salicylate to enter the mitochondria, leading to increased acidosis, producing more nonionized salicylate that enters the mitochondria and CNS, and on and on. Hemodialysis can effectively interrupt this cycle.
Renal failure: Salicylate at therapeutic doses is metabolized in the liver and eliminated through the kidneys. Hepatic metabolism becomes saturated in overdose, and increasing amounts of salicylate are eliminated in the urine. Renal impairment may defeat efforts to enhance elimination of salicylate by repleting volume losses and alkalinizing the urine, but no real consensus defines renal failure. It would be reasonable in borderline cases to give fluids and closely monitor their effect on creatinine, urine output, serum salicylate concentration, and clinical condition before starting hemodialysis, if no other indications are present, because almost all patients with severe acute salicylate poisoning present with significant prerenal azotemia.
Pulmonary edema: The pulmonary edema and hypoxemia often seen in acute salicylate toxicity usually represent acute respiratory distress syndrome from the direct toxic effect of the drug or congestive heart failure exacerbated by fluid administration from therapeutic purposes. Pulmonary insufficiency will limit options for medical treatment alone without dialysis in either case.
Of course, any evidence of deteriorating clinical condition despite adequate medical therapy suggests the need for hemodialysis.
Many texts and reviews suggest that a serum salicylate level of 100 mg/dL or more in acute overdose should be treated with hemodialysis, even if no other indications are present. I used to think this recommendation was nonsense: Treat the patient, not the number! But my thinking has changed, and now I think starting hemodialysis in patients with levels higher than 100 mg/dL is reasonable.
It is unusual for a patient with acute salicylate toxicity to have a level greater than 100 mg/dL but no change in mental status or hypoxia or other indication for dialysis. The population of well-appearing patients with levels higher than 100 mg/dL must be small. A number of those would certainly go on to require dialysis, and it seems preferable to initiate hemodialysis early on before large amounts of salicylate enter the central nervous system or cross the mitochondrial membrane. The number of acutely poisoned patients who have salicylate levels greater than 100 mg/dL and who can be safely treated without hemodialysis must be exceedingly small. Rather than trying to identify them early on, it might be preferable to treat based on the level alone.
But—and this is a crucial point—a salicylate level less than 100 mg/dL should never be used as justification to avoid starting hemodialysis in a patient with other indications. Clinicians who do so are committing a critical error. They are often called defendants.
Also, always carefully check the units used to report salicylate levels. These levels are most frequently reported as milligrams per deciliter or mg per 100 mL. These are the units I have been using here. Just to make life difficult, salicylate levels are sometimes reported as milligrams per liter. You do not want to start hemodialysis on a patient for a subtherapeutic salicylate level of 120 mg/L, which is equivalent to 12 mg/dL.
Share this article on Twitter and Facebook.
Access the links in EMN by reading this on our website or in our free iPad app, both available at www.EM-News.com.
Comments? Write to us at [email protected].