Background. Viral blips reflecting polymerase chain reaction (PCR) artefacts or transient low-level replication are well described in the human immunodeficiency virus setting. However, the epidemiology of such blips in transplant recipients screened for cytomegalovirus (CMV) with PCR remains uncertain and was investigated in a cohort of solid organ and hematopoietic stem cell recipients. Methods. Eligible recipients had known donor/recipient CMV IgG serostatus, and 3 CMV PCRs ≥. The CMV PCR triplicates (3 consecutive CMV PCRs) were defined; the first CMV PCR was always negative, and the time between the second and third samples was 7 days ≤. A positive second but negative third sample represented a blip. Odds ratio (OR) for factors associated with a triplicate being a blip was estimated by binomial regression adjusted for repeated measurements. Whether blips affected the hazard ratio (HR) for subsequent CMV infection was determined with a Cox model. Results. 851 recipients generated 3883 CMV PCR triplicates. The OR of a triplicate representing a blip decreased with increasing viral load of the second sample (vs 273 IU/mL; >273-910 IU/mL: odds ratio [OR], 0.2; 95% confidence interval [CI], 0.1-0.5; >910 IU/mL: OR, 0.08; 95% CI, 0.02-0.2; P ≤ 0.0002) and increased with intermediary-/low-risk serostatus (vs high risk) (OR, 2.8; 95% CI, 1.2-5.5; P = 0.01). Cumulative exposure to DNAemia in the CMV blips greater than 910 IU/mL indicated increased HR of subsequent CMV infection (HR, 4.6; 95% CI, 1.2-17.2; P = 0.02). Conclusions. Cytomegalovirus blips are frequent; particularly when the viral load of the first positive PCR is < 910 IU/mL, and serostatus risk is intermediary/low. Accumulating blips suggest intermittent low-level replication. If blips are suspected, confirmation of ongoing replication before initiation of treatment is prudent.