Hashes are great tools to weed out duplicates

my %seen;
while (my $row = $csv->getline($fh)) {
    next if $seen{ $row->[1] }{ $row->[9] }++;

    ...
}
[download]

This is based on code in perlfaq4