MSACL 2026 Abstract

Investigation of an Unknown Arsenic Peak in Urine by HPLC-ICP-MS

Kathryn A. Smith (1), Bryce Genesi (1), Jessica M. Boyd (1, 2), Kamisha L. Johnson-Davis (1, 2)
(1) ARUP Laboratories, Salt Lake City, UT, (2) Department of Pathology, University of Utah, Salt Lake City, UT

Kathryn Smith, PhD (Presenter) ARUP Laboratories

Relevant Financial Disclosures (within past 24 months, reported on Apr 14, 2026) No relevant financial relationship(s) to disclose. Conflict mitigation NOT REQUIRED.

1. Problem
An unknown arsenic peak eluting near the toxic AsV species was detected at concentrations >5 ppb in approximately 10% of patient urine specimens, raising concern for potential misidentification.

2. Method Information
• 50 μL urine diluted with 950μL diluent + internal standard (Antimony, Sb) solution
• Agilent 1260 Infinity II LC
• Agilent 7700 ICP-MS
• PRP-X100 Anion Exchange HPLC column (150 x 4.6mm, 5 µm), with KrudKatcher Classic HPLC in-Line filter
• MPA – 12 mM Ammonium Carbonate + 2% Ethanol, pH 8.5
• MPB – 55 mM Ammonium Carbonate + 2% Ethanol, pH 8.5
• 12 min analytical step gradient, Flow 1 mL/min
• Column oven 20°C
• 50 μL injection volume

3. Troubleshooting Steps
We first ruled out common spectral and polyatomic interferences (ArCl, Nd²⁺, Sm²⁺) and performed reduction and oxidation experiments, neither of which clarified the identity of the peak. When samples with the unknown peak were sent to a reference laboratory, the peak was incorrectly reported as toxic methylated arsenic. A retrospective review showed that nearly 70% of affected samples originated from a single U.S. region and were associated with occupational surveillance testing, though no additives or unusual collection devices were initially reported. A literature search identified similar unexplained peaks reported by other laboratories, hypothesized as arsenosugar metabolites but unconfirmed due to limited standards. After a prolonged investigative hiatus, renewed investigation including NIST kelp reference material and dietary exposure studies again failed to reproduce the peak. A subsequent review of phenylarsenicals identified p arsanilic acid (ASA) as a possible candidate based on elution behavior. When urine samples were exposed to common urinalysis test strips containing ASA, the unknown peak was reproduced, and HPLC-QTOF-MS confirmed ASA in patient samples, implicating pre-analytical contamination during urine collection as the source.

4. Outcome
The source of the unknown arsenic peak was determined to be p‑arsanilic acid (ASA), likely introduced through pre‑analytical contamination from urinalysis test strips used during urine collection.