Table 5. Antibiotic prophylaxis for the prevention of wound infections in long-bone open fractures: patients, interventions, and outcomes.

			Desirable effect		Undesirable effect		Desirable effect
Sample	Treatment	Control	Outcome	Outcome quality	Outcome	Outcome quality	Treatment arm % (95%-CI)	Controls % (95%-CI)	Difference %(95%-CI)	RR (95%-CI)
Long-bone open fractures	48-hour course, dicoxacillin or penicillin - 60 patients	Placebo - 30 patients	Wound infection (not specifically osteomyelitis)	Weak, Minor prognostic relevance	Antibiotic- resistant bacteria selection, not investigated by the study	Robust, Clinically relevant for the patient/crucial for the health-care organization	6.7 (2.6 to 15.9)	20 (9.5 to 37.3)	–13.3 (–31.1 to 0.7)	0.33 (0.1 to 1.09)
Long-bone open fractures	cloxacillin 10-day course - 43 patients	Placebo - 44 patients	Wound infection (not specifically osteomyelitis)	Weak, Minor prognostic relevance	Antibiotic- resistant bacteria selection, not investigated by the study	Robust, Clinically relevant for the patient/crucial for the health-care organization	4.7 (1.3 to 15.5)	27.3 (16.3 to 41.8)	–22.6	0.17 (0.04 to 0.72)