Recent studies have suggested that bacteria associated with the female reproductive tract - the uterine microbiota - may be important for reproductive health and pregnancy success. Therefore, uterine microbiome research gained much interest in the last few years. However, it is challenging to study late postpartum uterine samples, since they hold a low microbial biomass. Next-generation sequencing techniques are very sensitive for microbial identification, but they cannot make a distinction between actual microbiota and contaminant bacteria or their DNA. Our aim was to test a new method to sample the bovine uterine lumen in vivo, while minimizing the risk of cross-contamination. In order to evaluate this method, we performed a descriptive assessment of the microbial composition of the obtained samples. Transabdominal, laparoscopic sampling of the uterine lumen was conducted in five Holstein-Friesian cows. Uterine fluid from the uterine horns was collected by low-volume lavage. DNA from the samples was extracted using two different DNA extraction methods, and negative controls (sampling blank controls and DNA extraction blank controls) were included. Bacteria were identified using 16S rRNA gene amplicon sequencing. In this proof-of-concept study, no evidence for authentically present uterine microbiota could be found. During laparoscopic sampling, some practical challenges were encountered, and the reliability of low-volume-lavage for the collection of a low microbial biomass could be questioned. By comparing two DNA extraction methods, a significant contamination background could be noticed originating from the DNA extraction kits.