Mutations in the spatacsin gene have recently been identified as the genetic cause of autosomal-recessive spastic paraplegia (SPG) with thin corpus callosum, mapping to chromosome 15p13-21. While several nonsense and frameshift mutations as well as splice mutations have been identified, large genomic deletions have not yet been found, potentially due to the absence of an efficient analysis tool. After complete sequencing of 12 autosomal recessive hereditary spastic paraplegia patients with suggestive clinical signs, we were able to define nine SPG11 cases but were left with three patients in which only one SPG11 mutation could be identified by direct sequencing. In these patients, we performed high-resolution comparative genomic hybridization using a predesigned human chromosome 15 tiling array with an average spacing of 100 bp. Data analysis suggested heterozygous genomic deletion within the spatacsin gene in all three patients. In one patient, a relatively small genomic deletion (8.2 kb) could be validated by quantitative polymerase chain reaction (PCR) and long-range PCR, allowing the diagnosis of the deletion of exons 31 through 34. For two patients, quantitative PCR validation could not confirm a genomic deletion. As high density tiling arrays are available for the entire human genome, we suggest this approach for the screening of heterozygous genomic deletions in candidate genes down to a few kilobases.