We conducted a prospective cohort study in order to determine whether suction drain specimen cultures from orthopaedic surgery predicted an early wound infection. We included 218 consecutive clean orthopaedic operations requiring drains in one unit over a period of 1 year. The suction drain tip, drain fluid and wound discharge specimens were cultured, and the surgical wound was followed up for 3 months. There were six deep and two superficial wound infections. Wound infection was significantly related to positive suction tip culture but not to positive drain fluid culture. Following our methodology for culture, a positive drain tip culture predicts wound infection in 50% and a negative culture virtually excludes the possibility of a deep infection.