Abstract
The study carried out by Leino and Loo (Ann Clin Biochem 2007;44:563–5) 1 highlights the problems of using immunoassay for the detection of buprenorphine in urine. Their conclusions, albeit in this small study, suggest the CEDIA immunoassay used on the Roche platform is suitable for the routine detection of this drug. No false-positive results were reported in this study and only one false-negative result was observed for a urine sample demonstrating norbuprenorphine by LC/MS/MS. Antibody specificity data provided by Microgenics indicates cross-reactivity with some opiates including codeine, dihydrocodeine and hydrocodone but at very high concentrations (100 mg/L). A study published by Pavlic et al. 2 demonstrated false-positive results for the CEDIA buprenorphine assay in urine samples containing morphine (>120 mg/L), methadone (>320 mg/L), codeine (>30 mg/L), dihydrocodeine (60 mg/L) and morphine-3-glucuronide (520 mg/L). The concentrations at which cross-reaction occurs for codeine and DHC were very much lower in this latter study than those quoted by Microgenics.
We evaluated the CEDIA buprenorphine kit (Microgenics GmbH, Germany) adapted to a Beckman LX platform and compared results on 64 urine samples from drug addiction clinics with an established LC/MS/MS method. There were no false-negative results but nine false-positive results were observed for the CEDIA assay compared with LC/MS/MS. Seven of these samples were opiate positive by immunoassay screen (cut-off 300 μg/L), six of which demonstrated heroin abuse with the detection of 6-acetylmorphine (concentrations ranging from 0.09 to 21 mg/L). Five had concentrations of morphine and codeine ranging from 2.5 to >10 mg/L and 0.44 to 8.1 mg/L, respectively. One of the opiate positive urines contained only DHC at a concentration of 30 mg/L. Cross-reactivity of the CEDIA buprenorphine kit with subjects on DHC replacement has also been reported by Bottcher and Beck. 3 Of the two remaining false-positive urine samples in our study one was found to contain methadone at a concentration of 8 mg/L and the other was negative using our immunoassay drugs of abuse screening profile. The latter sample was obtained from a subject begin treated with chlorpromazine. We have no evidence to show cross-reactivity with this drug but the antipsychotic amisulpride is stated to cross-react at 31 mg/L (CEDIA buprenorphine cross-reactivity tables).
Although there appears to be general acceptance within the literature that the CEDIA buprenorphine kit is useful as a general screening method for this drug, drug treatment centres must be made aware of potential false-positive results if their laboratory does not routinely perform confirmatory analyses.