ORIGINAL ARTICLE Analysis of 11-nor-9-carboxy-D 9 -tetrahydrocannabinol in urine samples by hollow fiber-liquid phase microextraction and gas chromatography–mass spectrometry in consideration of measurement uncertainty Sarah Carobini Werner de Souza Eller • Luma Gonc ¸alves Flaiban • Beatriz Aparecida Passos Bismara Paranhos • Jose ´ Luiz da Costa • Felipe Rebello Lourenc ¸o • Mauricio Yonamine Received: 21 May 2014 / Accepted: 2 July 2014 / Published online: 25 July 2014 Ó Japanese Association of Forensic Toxicology and Springer Japan 2014 Abstract Marijuana abuse can be detected by means of toxicological analysis of the most important metabolite 11-nor-9-carboxy-D 9 -tetrahydrocannabinol (THC-COOH) in urine samples. The aim of this study is the establishment of the detailed procedure for analysis of THC-COOH in urine by combination of hollow fiber-liquid phase mic- roextraction (HF-LPME) and gas chromatography–mass spectrometry (GC–MS). The conditions of hydrolysis and extraction were optimized. The method was shown to be very simple and rapid, and a low amount of organic solvent was necessary for extraction. The limit of detection was 1.5 ng/ml. The calibration curves were linear over the specified range (2.0–170 ng/ml; r 2 [ 0.99). The main sources of uncertainty were found to be analyte concen- tration, accuracy, method precision and sample volume. The effect of the analyte concentration on the overall combined uncertainty was most significant. The developed method was successfully applied to a human urine standard reference material at two levels of concentration. The obtained relative combined uncertainty was 8 %, which can be considered acceptable according to international guidelines. The present method seems very useful in clin- ical and forensic toxicology, because of its simplicity, rapidness and inexpensiveness. Keywords Hollow fiber-liquid phase microextraction (HF-LPME)  THC-COOH  GC–MS  Urine  Measurement uncertainty Introduction Cannabis (in the forms of marijuana and hashish) is one of the most widely used illicit drugs worldwide [1–4], in spite of the fact that synthetic cannabinoids are prevailing. The most important member of the class of cannabinoids is D 9 -tetrahydrocannabinol (THC). After cannabis intake, THC is metabolized in the liver and other tissues to numerous metabolites, the main metabolic product of which is 11-nor- 9-carboxy-D 9 -tetrahydrocannabinol (THC-COOH), which is excreted into urine in both free and glucuronide forms [5–10]. These substances can be detected in urine of cannabis users for a period of a week or longer [6]. In fact, urine is the specimen of choice for monitoring illicit drug use, because its collection is physically nonin- vasive, and large volumes of sample are often available [11]. The confirmation and quantification of the presence of THC-COOH in urine requires the use of an advanced analytical approach. The approach used in the analysis of drugs in urine samples recommended by the National Institute on Drug Abuse in the United States (NIDA) is a combination of screening tests (generally immunoassays), followed by confirmation tests based on mass spectrometric techniques [10, 12]. In recent years, some innovative methods have been published in the scientific literature for the determination of THC-COOH in urine samples. Liquid chromatography–mass spectrometry (LC–MS)/LC–MS– MS [2, 6, 8, 11, 13–16] and gas chromatography–mass spectrometry (GC–MS) [1, 3, 10, 17–19] have been the main techniques used for this purpose. In general, S. C. W. de Souza Eller (&)  L. G. Flaiban  B. A. P. B. Paranhos  F. R. Lourenc ¸o  M. Yonamine Faculty of Pharmaceutical Sciences, University of Sa ˜o Paulo, Av. Professor Lineu Prestes, 580 13B, Sa ˜o Paulo, SP 05508-000, Brazil e-mail: sarahcarobini@usp.br J. L. da Costa Forensic Toxicology and Chemistry Laboratory, Criminalistics Institute of Sa ˜o Paulo, Sa ˜o Paulo, SP, Brazil 123 Forensic Toxicol (2014) 32:282–291 DOI 10.1007/s11419-014-0239-4