Background: Dried blood spots (DBS) may be valuable in the diagnosis of congenital cytomegalovirus (CMV) infection. However, the 2007 European Quality Control for Molecular Diagnostics (QCMD) proficiency testing programme showed that CMV DNA detection in DBS was lacking sensitivity in a considerable number of participating laboratories.

Objective: To compare DNA extraction methods for DBS for detecting CMV. Sensitivity and applicability of the methods for high-throughput usage were assessed.

Study design: Guthrie cards were spotted with CMV DNA-positive whole blood (n=15). DNA was extracted from the DBS using different extraction methods, followed by CMV amplification by means of real-time PCR.

Results: Significant differences between the extraction methods with respect to the sensitivity were found. Optimal sensitivity was achieved when samples were tested in triplicate, demonstrating that the methods in general operated around their detection limits. Triplicate testing using the protocol by Barbi et al. [Barbi M, et al. Cytomegalovirus DNA detection in Guthrie cards: a powerful tool for diagnosing congenital infection. J Clin Virol 2000;17:159-65], representing the most sensitive methods, resulted in sensitivities of 100%, 86%, and 50% for DBS with CMV DNA loads of 5-4, 4-3, and 3-2log(10)copies/ml, respectively. This indicates that sensitivity limitations apply in the clinically relevant concentration range. Few methods appeared suitable for 96-well format high-throughput testing.

Discussion: When considering universal neonatal screening for congenital CMV infection, an assay which is both sensitive and applicable for high-throughput testing is required. The protocol by Barbi et al. and the BioRobot Universal System appear appropriate candidates currently available for 96-well format application in neonatal screening using DBS.