The purpose of this study was to evaluate the use of carrier DNA (i.e., exogenous DNA spike-in) for shotgun metagenome sequencing of ultra-low levels (less than 50 picograms) of metagenomic DNA. We hypothesized that carrier DNA would improve the robustness of library preparation for samples with DNA concentrations below detection by providing a tangible amount of known DNA thereby bringing total DNA concentrations closer to recommended input ranges for metagenomic library kits. We employed adaptive PCR cycling using an iconPCR instrument (N6tec) to allow dynamic thermocycling until sufficient library for sequencing was amplified, regardless of input DNA concentration. Libraries were sequenced and mapped to reference genomes of lambda and mock community organisms, and outcome measures included total reads, on-target reads, evenness of coverage across 10 organisms within each mock community, and PCR duplication rate. We demonstrate that libraries can be prepared down to 50 femtograms of input DNA, but that there is a strong correlation between input DNA concentration and PCR duplication rate. The utility of spiking in carrier DNA is equivocal as it mildly negatively impacts the observed distribution of mock communities and serves as a loss of sequencing output. Although the loss of sequencing capacity due to carrier DNA can be partially offset by reduced loss of data from PCR duplication, carrier DNA spike-in is not recommended for routine library preparation of ultra-low input samples. Adaptive cycling allows for appropriate cycling conditions when input DNA concentrations are below detection.