Useful conclusion? Amrish Baidjoe, RadboudUMC 26 May 2014 I read your paper with great interest. Sensitivity and specificity are interesting subjects when it comes to the detection of low density infections. I am although troubled by your conclusion. In your conclusion you mention that based on the results DBS samples yield a lower amount of positives when compared to whole blood. This seems a reasonable outcome, although we found mixed results. Obviously this depends on stability of PCR targets and such. My biggest issue is that you used different ways of storing your samples. If I read it correctly the whole blood samples are stored at -20 up to the time of extraction (2.5yrs) where the DBS samples are stored at RT for 6-9 months and then stored at -20 for the remainder of time. The large variation of positives could be derived on the storage of the samples. In addition you use 200uL venous blood (from the WB sample) in contrast to 6x3mm punches for DBS, this is less than 200uL (also see: Baidjoe et al. combined DNA extraction and antibody elution...malJ 2013). Based on this inconsistency I believe you can't make a fair comparison and think therefore that part of the conclusion made in this paper is invalid. Competing interests None.