Since bromides are well-known precursors of primary amines, diethylaluminum 12-bromo-1-dodecyl oxide has been prepared and used as an initiator for the ring-opening polymerization of epsilon-caprolactone and L-lactide. Under strictly controlled conditions, the end-functionalization of the polyesters is quantitative and the bromo end-group is easily converted into an azide group whatever the polymeric backbone. The subsequent reduction of the azide into the expected primary amine has been investigated by catalytic transfer hydrogenation (CTH) in DMF and by hydrolysis in the presence of triphenylphosphine in THF, respectively. The hydrolysis reaction (P phi(3)/H2O) is perturbed by a coupling reaction, which involves a protonated secondary amine and leads to a twofold increase in the polyester molecular weight. The CTH method gives rise to the expected omega-NH2 poly(epsilon-caprolactone), in contrast to polylactide which seems to be unstable toward the nascent amine end group. Whatever the polarity of the medium (DMF or THF), aminolysis of polylactides is observed to occur and leads to the formation of an internal amide.