Contamination Management of Broad-Range Ribosomal DNA PCR: Where Is the Evidence?

We have read with great interest the minireview of Millar and colleagues concerning the risk of contamination of broad-range ribosomal DNA (rDNA) PCR (3). We completely agree with the authors that the high sensitivity of broad-range rDNA PCR could lead to false-positive results. As stated by Millar and colleagues, a critical analysis of the work flow and the obtained results is warranted, as is the introduction of sufficient and appropriate negative and positive controls.

We take issue, however, with the statement that strict segregation of the laboratory work flow is a fundamental requirement for successful broad-range rDNA PCR. The introduction of a segregated work flow, PCR setup cabinets, HEPA filters, and class II safety cabinets greatly increases the PCR costs. The use of different sets of pipettes, the exclusive use of filtered tips, the requirement of sterile gloves, and frequent UV irradiation of the rooms and materials further increase the costs for molecular diagnostics. Moreover, sterilization of materials, the skin, and the PCR room does not necessarily eliminate DNA from dead bacteria. All these measures are surely beneficial for the order books of biomedical supply companies, but the advantages of many of these measures in eliminating contamination of broad-range ribosomal PCRs remain to be proven. Since evidence-based medicine is the ultimate goal in clinical medicine, why not in laboratory medicine (2)?

The magnitude of the problem of contamination of rDNA PCRs is highly variable, and important contamination may come from an unexpected quarter. On the one hand, some authors, such as Corless and colleagues (1), reported major contamination of the samples with 16S rDNA. All kinds of strategies aimed to reduce the number of false positives also affected the yield of the quantitative PCR. On the other hand, in our experience contamination of 16S rDNA TaqMan quantitative PCR was low-level, with an average surplus of only 55 copies and a maximum of 150 surplus copies in a large series of samples (4, 5). In in vivo samples, the amount of 16S rDNA recovered was many times higher than that for the highest negative control (5). The water used was critical: Milli-Cure water generated three times more contamination than double-distilled water. All samples were prepared in the routine laboratory with the utmost “good laboratory practice” care, and quantitative PCR was performed in a separate room. The many other precautions proposed in the review of Millar and colleagues, however, were lacking. In our experience, a critical analysis of the work flow combined with an appropriate cutoff were enough to overcome the very low-level contamination of the samples.

We completely agree with the authors that an appropriate quality control of rDNA PCR is indispensable. However, the publication of these contamination management recommendations in your journal may have a great impact on the organization of molecular diagnostics in many hospitals and research laboratories, amplifying the costs many times. Molecular diagnostics almost becomes a ritual driven by fear of contamination. In our opinion, much more experimental evidence on the cost-benefit relation of all these measures to reduce rDNA contamination is warranted before they can be propagated as the new standard of “good molecular diagnostics practice.”