Commande :
sort -u -t',' -k1,1 monfichier.csv
Doc : https://unix.stackexchange.com/questions/171091/remove-lines-based-on-duplicates-within-one-column-without-sort