Despite their promising outcomes in preclinical models, the genomic effects of using adeno-associated viruses (AAVs) as a delivery strategy for CRISPR-Cas systems are still under investigation. Targeted high-throughput sequencing is the gold standard for characterizing the editing events caused by CRISPR-Cas systems, but the bioinformatic tools currently available do not specifically account for AAV integration. Here, we compare the efficiency of several widely-used tools of detecting and quantifying AAV integrations and demonstrate their impact on a use-case dataset. Our results show that the tested tools tend to underestimate AAV integration rates and AAV capture ratios, and sequencing reads with vector integrations are often discarded or misclassified, while showing similar total editing rates. Therefore, we propose a pre-processing pipeline optimized to retain and quantify all reads with AAV insertions before using the desired genome editing detection tool. Applying this pipeline to a dataset in which different CRISPR-Cas strategies were used to target the same locus for targeted gene disruption, we observe that AAV-delivered single and paired SaCas9 nucleases exhibit higher vector sequence integration rates than paired D10ASaCas9 nickases using the same guides. This study highlights the importance of the analysis pipelines in evaluating AAV-delivered CRISPR-Cas system efficacy and safety and proposes a pre-processing pipeline to overcome the limitations of existing tools for targeted integration analysis.