Drug confirmation by mass spectrometry: Identification criteria and complicating factors.
BACKGROUND Clinical laboratories are required to establish reference intervals for all the analytes tested, and these are provided along with the test results. In contrast, laboratories testing for pain medications use cutoffs established by the manufacturers of immunoassay reagents. These cutoffs may be inappropriate for monitoring patients being treated for chronic pain with opioid therapy because the cutoffs are set too high. PURPOSE OF THE STUDY To use quantitative urine drug excretion data determined by liquid chromatography-tandem mass spectrometry analysis to calculate cutoffs needed to best determine patient compliance with prescribed medications. METHODS Two methods of calculation were used to estimate cutoffs. First, all positive results for each drug or drug metabolite were ranked from highest to lowest. The lowest value was one-half of the lower limit of quantitation. A nonparametric 2.5 percent estimator was used to establish each cutoff Second, positive results were normalized using creatinine values, resulting in the excretion being expressed as nanograms of excreted drug per gram of creatinine. A nonparametric 2.5 percent estimator was used to establish the cutoff. RESULTS Cutoffs established using these calculations included at least 97.5 percent of the data for the drugs: 7-aminoclonazepam, alpha-hydroxalprazolam, buprenorphine, carisoprodol, hydrocodone, hydromorphone, meperidine, meprobamate, methadone, morphine, oxycodone, oxymorphone, propoxyphene, and tramadol gave cutoffs near the lower limit of quantitation. One exception was with the drug codeine, where the lower limit could not be identified. CONCLUSIONS These cutoffs were significantly lower than those suggested by many immunoassay manufacturers and better identify patient compliance in a representative population of pain patients. The limitation of the study is that only values one-half of our lowest calibrator were used. By using population values, laboratories can establish appropriate cutoffs for best monitoring pain patient medication compliance.