MicroRNAs (miRNAs) are small non-coding RNAs that regulate post-transcriptional gene expression. Recent studies have shown that human disease states correlate with measurable differences in the level of circulating miRNAs relative to healthy controls. Thus, there is great interest in developing clinical miRNA assays as diagnostic or prognostic biomarkers for diseases, and as surrogate measures for therapeutic outcomes. Our studies have focused on miRNAs in human cerebral spinal fluid (CSF) as biomarkers for central nervous system (CNS) diseases. Our objective here was to examine factors that may affect the outcome of quantitative PCR (qPCR) studies on CSF miRNAs, in order to guide planning and interpretation of future CSF miRNA TaqMan® low-density array (TLDA) studies. We obtained CSF from neurologically normal (control) donors and used TLDAs to measure miRNA expression. We examined sources of error in the TLDA outcomes due to (1) nonspecific amplification of products in total RNA, (2) variations in RNA isolations performed on different days, (3) miRNA primer probe efficiency, and (4) variations in individual TLDA cards. We also examined the utility of card-to-card TLDA corrections and use of an unchanged "reference standard" to remove batch processing effects in large-scale studies.