We retrospectively reviewed 35 autopsy cases where death had resulted from homicidal sharp force injuries and compared cases where the injuries had been inflicted by family members or relatives (relative group) with cases where the injuries had been inflicted by an unrelated person (stranger group). We reviewed the age and sex of the victims, the number of stab wounds, the site of the stab wounds, the presence of defence wounds, the detection of alcohol and other drugs and the mental status of the victims and perpetrators. We found the following tendencies: (a) a female victim was more frequently killed by a relative than by a stranger; (b) the percentage of cases receiving a single stab wound and the percentage of cases receiving more than ten stab wounds were both unexpectedly higher in the relative group than in the stranger group, and (c) in the stranger group, when there were no defence wounds, the victim had usually consumed alcohol, whereas when there were neither defence wounds nor alcohol intake, the case usually fell into the relative group. These tendencies will contribute towards our forensic appraisement in autopsy cases resulting from sharp force injuries.