Numerous studies have indicated that microRNAs (miRNAs) are present and stable in multiple biological fluids, suggesting a great potential as biomarkers for molecular diagnostics and prognostics. Variations in the amount of starting material and isolation method to obtain miRNA may introduce bias and contribute to quantification errors. Given these concerns, we compared five commercially available kits for serum/plasma miRNA isolation to determine whether the plasma miRNA profile varies with the isolation method. We isolated miRNAs in blood plasma from colorectal cancer patients and healthy donors with five commercially available kits: Exiqon, Norgen, Macherey-Nagel, Qiagen, and Zymo Research. First, we assessed the robustness of the RNA isolation process and the quality of isolated miRNAs with the miRCURY microRNA QC PCR Panel (Exiqon), which contains six RNA spike-ins for quality control of RNA isolation (UniSp2, -4, and -5), complementary DNA (cDNA) synthesis (UniSp6 and cel-miR-39-3p), and polymerase chain reaction (PCR) amplification (UniSp3). This panel also includes circulating human miR-103, miR-191, miR-23a, and miR-451. Second, to evaluate the variability in miRNA profiling in relation to the extraction method, we analyzed plasma levels of candidate miRNA biomarkers for colorectal cancer (miR-18a, miR-21, and miR-29a). To determine PCR efficiencies per amplicon and per sample, we used LinRegPCR software. We found that all isolation methods were suitable for extracting miRNA from plasma samples and that all had similar Cq values in the three steps analyzed: RNA isolation, cDNA synthesis, and quantitative reverse transcription (qRT)-PCR. However, although the PCR replicates were excellent, the intersample variability of the spike-ins was unsatisfactorily high and all kits yielded suboptimal PCR efficiencies for some amplicons. Overall, our results underline the great difficulties involved in measuring miRNAs in plasma. The use of spike-ins is critical to control technical factors that affect final miRNA levels. We recommend that researchers investigating circulating miRNAs verify the PCR efficiency for each amplicon because quantification may be influenced by sample and PCR components.