Many microsurgeons experience inadvertent back-wall stitch as a cause of immediate anastomotic failure. To investigate this factor as a possible cause of delayed arterial anastomotic failure that does not appear in the operation room, a purposeful back-wall stitch (PBWS) model of microarterial anastomosis was applied in various configurations on femoral and carotid arteries of rats. Carotid (n = 28) and femoral (n = 28) artery groups were equally divided into 4 different subgroups according to the type of PBWS. Control subgroups had no PBWS. Thirty-degree, 60-degree, and 90-degree subgroups had PBWS located at 30, 60, and 90 degrees, respectively. Patencies were assessed at 60 minutes and 24 hours. Immediate milking test demonstrated normal anterograde refilling in all anastomoses. PBWS increased thrombosis in femoral arteries, while it did not increase it in the carotid at either evaluation times. The only significant difference was between 90 degrees PBWS and the control subgroup irrespective of the vessel factor in 24 hours. Histologic examination supported absence of thrombosis, as suggested by surgical examination. This may show that small-sized arteries are more inclined to thrombosis compared with larger ones and the thrombosis risk increases as the inadvertent back-wall stitch is more centrally located. Contrary to general notion that the inadvertent back-wall stitch causes immediate thrombosis, thromboses later than 1 hour, and even patent anastomoses in 24 hours, were observed in femoral and carotid artery groups. This study suggests that the inadvertent back-wall stitch should also be considered as a possible cause of late anastomotic problems.