Comparison of automated and manual nucleic acid extraction methods for detection of enterovirus RNA

Automated nucleic acid extraction is an attractive alternative to labor-intensive manual methods. We compared two automated methods, the BioRobot M48 instrument (Qiagen, Inc.) and MagNA Pure (Roche Applied Sciences) methods, to two manual methods, the QIAamp Viral RNA Mini kit (Qiagen) and TRIzol (Invitrogen), for the extraction of enterovirus RNA. Analytical sensitivity was assessed by dilution analysis of poliovirus type 2 Sabin in cerebrospinal fluid. The sensitivity of PCR was equivalent after RNA extraction with QIAamp, BioRobot M48, and MagNA Pure. All 18 replicates of 100 PFU/ml were detected after extraction by the four methods. Fewer replicates of each successive dilution were detected after extraction by each method. At 10(-1) PFU/ml, 17 of 18 replicates were positive by QIAamp, 15 of 18 replicates were positive by BioRobot M48, and 12 of 18 replicates were positive by MagNA Pure; at 10(-2) PFU/ml, 4 of 17 replicates were positive by QIAamp, 2 of 18 replicates were positive by BioRobot M48, and 0 of 18 replicates were positive by MagNA Pure. At 10(-3) PFU/ml, no replicates were detected. Evaluation of TRIzol was discontinued after nine replicates due to a trend of lower sensitivity (at 10(-3) PFU/ml eight of nine replicates were positive at 100 PFU/ml, four of nine replicates were positive at 10(-1) PFU/ml, and zero of nine replicates were positive at 10(-2) PFU/ml). Concordant results were obtained in 24 of 28 clinical specimens after extraction by all methods. No evidence of contamination was observed after extraction by automated instruments. The data indicate that the sensitivity of enterovirus PCR is largely similar after extraction by QIAamp, BioRobot M48, and MagNA Pure; a trend of decreased sensitivity was observed after TRIzol extraction. However, the results of enterovirus PCR were largely concordant in patient samples, indicating that the four extraction methods are suitable for detection of enteroviruses in clinical specimens.