P>The aim of this in vitro study was to test the effect of different post systems on fracture strength of roots with re-attached fragments. Root canals of eighty extracted single-rooted human teeth were instrumented (ProFile) and randomly divided into two groups. The roots in the first group were vertically cracked, and the fragments were re-attached using Super Bond C&B (Sun Medical, Tokya, Japan). The roots in the second group were kept sound. Obturation of the roots was performed with MetaSEAL (Sun Medical) and gutta-percha. Post spaces were prepared, and the roots were restored with one of the followings: UniCore (Ultradent), Everstick (Stick Tech), Ribbond (Ribbond), ParaPost (Coltene/Whaledent) (n = 10). Four mm high build-ups were created (Clearfil DC Bond Core; Kuraray, Tokyo, Japan). Compressive loading of the samples was performed after 24 h (1 mm min-1). Mean load necessary to fracture each sample was recorded (Newton) and statistically analysed (One-way anova, t-tests). ParaPost showed the highest fracture strength among the roots with re-attached fragments (P < 0.05). UniCore and ParaPost systems showed similar fracture strength in the sound roots (P > 0.05). Re-attached fragments significantly reduced the fracture strength of roots in UniCore group (P = 0.000). Ribbond post showed mostly repairable fractures. Metal post (ParaPost) showed the highest fracture strength in the roots with re-attached fragments; however, fracture pattern was 41% non-repairable. Re-attached fragments significantly reduced the fracture strength of the roots in UniCore group. Prefabricated posts showed similar fracture strength in the sound roots. Customized post systems EverStick and Ribbond showed mostly repairable failure after loading in sound roots or roots with re-attached fragments.