Article, Emergency Medicine

Salicylate toxicity in the absence of anion gap metabolic acidosis

Unlabelled image

Case Report

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage: www. elsevier. com/ locate/ajem

American Journal of Emergency Medicine 34 (2016) 1328.e1-1328.e3

Salicylate toxicity in the absence of Anion gap metabolic acidosis?

Abstract

Salicylates are a common ingredient in many over-the-counter and prescription medications. Classically, Salicylate poisoning is associated with an elevated anion gap acidosis; however, the absence of an anion gap does not rule out salicylate toxicity.

We present 3 cases of salicylate toxicity with normal anion gap met- abolic acidosis resulting from laboratory interference.

This case series highlights the importance of considering salicylate toxicity in patients with metabolic acidosis even in the absence of an el- evated anion gap.

Salicylates are a common ingredient in many over-the-counter and prescription medications. In 2012, the National Poison Database Service of the American Association of Poison control centers reported more than 300,000 exposures to salicylates alone or salicylate-containing pharmaceuticals. Of reported exposures, 77,959 were treated in a health care facility, and 364 deaths occurred [1].

Salicylate poisoning is classically associated with an elevated anion gap acidosis and results from the accumulation of salicylic acid, lactic acid, and ketone bodies [2]. However, the absence of an anion gap does not rule out salicylate toxicity. We present 3 cases of salicylate tox- icity with normal anion gap acidosis resulting from laboratory interfer- ence. These cases demonstrate the clinical value of considering salicylate toxicity in patients with metabolic acidosis even in the ab- sence of an elevated anion gap.

A 43-year-old woman with history of seizure disorder and glioblasto- ma multiforme status postresection presented to the emergency depart- ment (ED) with chief concern of feeling like she was going to have a seizure. She had not been compliant with her anticonvulsant for several weeks but denies recent seizure. The patient expressed that she had a chronic, unchanged headache and increasing fatigue. Her speech was slurred which was also chronic. She denied gait instability, nausea and vomiting, dizziness, fever and chills, shortness of breath, and chest pain.

Physical examination revealed heart rate of 104 beats per minute, blood pressure of 95/56 mm Hg, respiratory rate of 22 breaths per min- ute, and temperature of 36.5?C. The patient was alert and cooperative. The remainder of the examination was unremarkable with the excep- tion of slurred speech.

Serum chemistry values were as follows: sodium, 140 mmol/L; po- tassium, 3.5 mmol/L; chloride, 118 mmol/L; bicarbonate, 15 mmol/L; Blood urea nitrogen , 25 mg/dL; creatinine, 0.7 mg/dL; and glu- cose, 100 mg/dL. The anion gap was calculated to be 7 mmol/L. All

? This work was completed without the assistance of grants, financial or otherwise.

laboratory values were obtained before administration of intravenous

(IV) fluid or medication. Computed tomography of the brain without contrast revealed no acute findings.

The patient received two 1-L IV normal saline fluid bolus, IV mor- phine sulfate, and lorazepam by mouth for alleviation of headache. A hyperchloremic non-anion gap metabolic acidosis was identified, and after administration of IV fluids, a repeat chemistry was obtained with the following values: sodium, 143 mmol/L; potassium, 3.8 mmol/L; chloride, 121 mmol/L; bicarbonate, 14 mmol/L; BUN, 19 mg/dL; creati- nine, 0.8 mg/dL; and glucose, 83 mg/dL. The anion gap was calculated to be 8 mmol/L. Lactic acid and salicylate levels were obtained in consul- tation with a toxicologist: 0.3 mmol/L and 46 mg/dL, respectively. An ar- terial blood gas (ABG) demonstrated a pH of 7.45, pCO2 of 20 mm Hg, pO2 of 127 mm Hg on room air, and bicarbonate of 14 mmol/L.

Three hours following arrival in the ED, the patient became progres- sively more confused and combative. She could not be reoriented. Re- straints and sedation for patient safety were ordered.

The patient was started on a sodium bicarbonate continuous infu- sion while preparations were made for hemodialysis given the degree of alteration in mental status. The patient underwent one 4-hour run of hemodialysis with clearing of her mental status and correction of lab- oratory derangements. The day following, the patient admitted to con- suming several handfuls of acetylsalicylic acid (aspirin) for each of the 3 days preceding her presentation in an attempt to harm herself.

A 16-year-old adolescent girl presented to the ED 9 hours following intentional aspirin ingestion following argument with a family member. She reported taking approximately 30 tablets of enteric-coated aspirin in a self-harm attempt. The patient denied any other coingestants and experienced 1 episode of tinnitus that resolved. At the time of her pre- sentation to the ED, she denied any complaints, and physical examina- tion revealed heart rate of 122 beats per minute, blood pressure of 118/63, respiratory rate of 24 breaths per minute, and temperature of

36.7?C. The patient was alert and oriented. The remainder of the exam- ination was unremarkable.

Serum chemistry values at presentation were as follows: sodium, 144 mmol/L; potassium, 4.3 mmol/L; chloride, 114 mmol/L; bicarbon- ate, 21 mmol/L; BUN, 11 mg/dL; creatinine, 0.5 mg/dL; and glucose, 91 mg/dL. The anion gap was calculated to be 9 mmol/L. Initial salicylate level was 43 mg/dL. A venous blood gas revealed metabolic acidosis and respiratory alkalosis (pH of 7.42, pCO2 of 27 mm Hg, pO2 of 38 mm Hg on room air, and bicarbonate of 18 mmol/L). All values were obtained be- fore administration of any IV fluids or medication.

The patient was started on continuous sodium bicarbonate infusion,

and repeat salicylate levels and serum chemistries demonstrated the absence of development of acidemia, renal insufficiency, or worsening of Serum salicylate level. The patient continued asymptomatic. One and half days following presentation, no measurable salicylates were

0735-6757/(C) 2016

1328.e2 S. Bauer, M.A. Darracq / American Journal of Emergency Medicine 34 (2016) 1328.e1-1328.e3

found in the patient’s serum with a normal venous blood gas pH. She was subsequently medically cleared for psychiatric placement.

A 51-year-old woman presented to the ED with 1 day of right ankle pain with a medical history of peripheral vascular disease, chronic vascu- lar ulcer on the right ankle, cirrhosis, and chronic kidney disease. The vas- cular ulcer was under the care of home health nurse with recent dressing change. During initial triage evaluation, the patient was noted to have a heart rate of 124 beats per minute, respiratory rate of 26 breaths per min- ute, and temperature of 36.2?C. She complained of feeling confused and was slow to answer but was answering questions appropriately.

After triage and while awaiting Medical provider evaluation, the pa- tient was noted to have a tonic-clonic seizure of approximately 2-minute duration. Following seizure, vital signs remained unchanged. The patient remained confused and was tangential in her thought process and speech. Serum chemistry values drawn following seizure were as follows: sodium, 144 mmol/L; potassium, 4.2 mmol/L; chloride, 117 mmol/L; bicarbonate, 12 mmol/L; BUN, 22 mg/dL; creatinine, 1.4 mg/dL; and glu- cose, 140 mg/dL. The anion gap was calculated to be 15 mmol/L. The initial elevated anion gap was attributed to seizure. Computed tomogra-

phy imaging of the brain without contrast revealed no acute findings.

While awaiting admission for new-onset seizure, the patient was noted to become progressively more tachypneic with respiratory rates persistently in the high 20s. The patient was placed on Bilevel positive airway pressure with some improvement in respiratory rates. The patient remained confused and tangential and was unable to an- swer questions appropriately. Repeat serum chemistries drawn

7 hours following tonic-clonic seizure were as follows: sodium, 141 mmol/L; potassium, 4 mmol/L; chloride, 118 mmol/L; bicarbonate, 12 mmol/L; BUN, 27 mg/dL; creatinine, 1.5 mg/dL; and glucose, 108 mg/ dL. The anion gap was calculated to be 11 mmol/L. An ABG drawn con- comitantly with chemistry panel revealed the following: serum pH of 7.50, pCO2 of 18 mm Hg, pO2 of 135 mm Hg on room air, and bicarbonate of 14 mmol/L. A toxicologist was consulted, and serum salicylate level was obtained with value of 60 mg/dL. A chest radiograph was obtained, revealing a pattern of pulmonary congestion and alveolar edema.

The patient was started on continuous sodium bicarbonate infusion as preparations were made for hemodialysis. The patient underwent one 4-hour run of hemodialysis with clearing of confusion and alter- ation in mentation, normalization of vital signs, and resolution of acido- sis on serum chemistry. Four hours following hemodialysis, repeat serum salicylate levels were undectable. The patient endorsed taking acetylsalicylic acid (aspirin) to assist in the ankle pain and had taken 5 g in the preceding 48 hours.

We present 3 cases of salicylate toxicity in which patients had nor- mal anion gaps and hyperchloremia. Hyperchloremic metabolic acidosis resulting from the loss of serum bicarbonate may occur with diarrhea, with fistula drainage, or from renal acid excretion abnormalities (type I, II, and IV renal tubular acidosis). Administration of large volumes or concentrations of intravenous sodium chloride may also lead to hyperchloremia. Endocrine and oncological causes of hyperchloremia are also described [3]. The patients presented in this case series demon- strated no findings on examination or by history suggestive of these other etiologies of hyperchloremia.

Laboratory interferences are also an important cause of hyperchloremia. Bromide is known to falsely elevate serum Chloride levels because it is mis- taken for chloride by a number of laboratory analyzers. Lithium lowers the anion gap by adding otherwise “unmeasured” cations. Other medications including nonsteroidal anti-inflammatory drugs, phenylbutazone, cyclo- sporine, and chlorothiazide may similarly be mistaken for chloride by some analyzers [4]. Previous case reports have similarly described eleva- tions in serum chloride levels in the presence of salicylates with some ion-selective electrodes. This interference may result from loss of selectivity over the operational life cycle of the electrode and/or competition between salicylates and chloride for binding sites on albumin [5,6,7,8].

Because the anion gap represents the difference between serum- measured cations and anions, any “false” elevation in chloride may

result in a falsely “normal” or “negative” anion gap. There is currently limited published literature describing salicylate toxicity with normal or negative anion gap in salicylate toxicity [9,10].

Jacob and Lavonas describe 2 cases of significant salicylate toxicity with markedly elevated salicylate levels. The first patient presented fol- lowing a reported ingestion of aspirin and alcohol 2 hours before presen- tation. Laboratory studies were as follows: sodium, 139 mmol/L; chloride, 123 mmol/L; serum bicarbonate, 13 mmol/L; and Serum glucose, 102 mg/ dL (calculated anion gap, 3 mmol/L). Her serum salicylate level was 110 mg/dL. An ABG result showed a significant metabolic acidosis and re- spiratory alkalosis (pH of 7.21, pCO2 of 32 mm Hg, pO2 of 152 mm Hg, serum bicarbonate of 12 mmol/L). The second patient presented to the ED following failure to appear at work that day. He denied ingestions other than alcohol and diazepam 12 hours prior. He was noted to be tachypneic, diaphoretic, and drowsy. Laboratory values were as follows: sodium, 142 mmol/L; chloride, 111 mmol/L; serum bicarbonate,

22 mmol/L (calculated anion gap, 9 mmol/L); and serum glucose, 115 mg/dL. His serum salicylate level was 54.6 mg/dL. At that point, the patient admitted to ingesting 150 tablets of 325 mg of aspirin 14 hours be- fore. His ABG showed both metabolic acidosis and a primary respiratory alkalosis (pH of 7.41, pCO2 of 25 mm Hg, pO2 of 93 mm Hg, serum bicar- bonate of 16 mmol/L). He continued to absorb salicylate during the next 11 hours, achieving a highest recorded concentration of 95.3 mg/dL. Both patients underwent hemodialysis and recovered fully [9].

Kaul et al describe a 30-year-old man presenting with agitation and diaphoresis with questionable ethanol or levitracetam overdose. Physi- cal examination revealed tachypnea, tachycardia, confusion, and ag- gressive behavior. Serum chemistries were significant for sodium of 146 mmol/L, chloride of 118 mmol/L, CO2 of 21 mmol/L, BUN of 12 mg/dL, creatinine of 1.0 mg/dL, and anion gap of 7 mmol/L. Venous blood gas demonstrated pH of 7.4, bicarbonate of 14.5 mmol/L, PCO2 of 24 mm Hg, and PO2 of 88 mm Hg. Institution specific routine screening for salicylates returned with a value of 45.8 mg/dL. The patient underwent hemodialysis with Complete recovery [10].

Both the case described by Kaul et al and the 2 cases described by Jacob and Lavonas resulted from interferences with the Siemens Dimension Vista chemistry analyzer. Our institution uses the critical care express NOVA an- alyzer without previous interference being reported with this analyzer.

Suspicion for salicylate toxicity should be based on clinical examina- tion and history. The 3 cases presented in the current series and the pa- tients described in other published case reports had findings consistent with salicylate toxicity including tachycardia, tachypnea, confusion, and alterations in mentation. In isolation, the absence of an anion gap acido- sis cannot reliably rule out salicylate toxicity. The presence of metabolic acidosis should prompt the consideration of salicyate toxicity even in the absence of an elevated anion gap.

Stephanie Bauer, DO UCSF-Fresno Department of Emergency Medicine 155 N Fresno St, Fresno, CA, 93701, USA

E-mail address: [email protected]

Michael A. Darracq, MD

UCSF-Fresno Department of Emergency Medicine

Division of Medical Toxicology 155 N Fresno St, Fresno, CA, 93701, USA

Corresponding author

E-mail address: [email protected] http://dx.doi.org/10.1016/j.ajem.2015.12.038

References

  1. Mowry JB, Spyker DA, Cantilena Jr LR, Bailey JE, Ford M. 2012 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 30th annual report. Clin Toxicol 2013;51(10):949-1229.

S. Bauer, M.A. Darracq / American Journal of Emergency Medicine 34 (2016) 1328.e1-1328.e3 1328.e3

  1. Metabolic Acidosis. The Merck manual. (Accessed December 1, 2015, at http://www. merckmanuals.com/professional/endocrine_and_metabolic_disorders/acid-base_ regulation_and_disorders/metabolic_acidosis).
  2. Humes HD, Shayman JA. Approach to the patient with metabolic acidosis or meta- bolic alkalosis. In: HD H, editor. Kelly’s essentials of internal medicine. Philadelphia: Lippincott Williams and Wilkins; 2001. p. 308.
  3. Van Leeuwen AM, Poelhuis-Leth DJ, Bladh ML. Chloride, blood, In: Schnell ZB, Van Leeuwen AM, Kranpitz TR, eds. Davis’s comprehensive handbook of laboratory & di- agnostic tests with nursing implications, Philadelphia: EA Routh JL Davis company, 2013, p 349
  4. Paul WD. Assessment of interference by aspirin with some assays commonly done in the clinical laboratory. Clin Chem 1976;22:837-42.
  5. Wang T, Diamandis EP, Lane A, Baines AD. Variable selectivity of the Hitachi chem- istry analyzer Chloride ion-selective electrode toward interfering ions. Clin Biochem 1994;27:37-41.
  6. Mori L, Waldhuber S. Salicylate interference with the Roche cobas Integra chloride electrode. Clin Chem 1997;43:1249-50.
  7. Zimmer BW, Marcus RJ, Sawyer K. Salicylate intoxication as a cause of pseudohyperchloremia. Am J Kidney Dis 2008;51:346-7.
  8. Jacob J, Lavonas EJ. Falsely normal anion gap in severe salicylate poisoning caused by

laboratory interference. Ann Emerg Med 2011;58(3):280-1.

  1. Kaul V, Imam SH, Gambhir HS, Sangha A, Nandavaram S. Negative anion gap meta- bolic acidosis in salicylate overdose-a zebra! Am J Emerg Med 2013;31(10): 1536.e3-4.

Leave a Reply

Your email address will not be published. Required fields are marked *