Antibiotic prophylaxis for surgery for proximal femoral and other closed long bone fractures

 

Characteristics of included studies [ordered by study ID]

Bergman 1982

---

Methods	RCT Location: University Hospital, Sweden Recruitment period:1974-1977 Adequacy of concealment of assigned allocation: B Intention to treat analysis: 3 Blinding of outcome assessors: 2 Comparability of treatment groups at entry: 1 Use of placebo: 3 Definition of inclusion and exclusion criteria: 1 Assessment of infection: 2 Duration of surveillance: 1 Losses to follow up: None described
Participants	180 analysed (92 women, 88 men, mean age 47 years) Inclusion criteria: undergoing surgery for closed ankle fracture Exclusion criteria: None described
Interventions	a. Dicloxacillin 2g pre-operatively and 6 hourly for 48 hrs b. Benzyl penicillin 3 million international units (IU), same regimen c. Saline placebo, same regimen.
Outcomes	1. Deep wound infection 2. Superficial wound infection 3. Adverse reactions
Notes	This report also contains data for use of the same regimen in 90 open fractures. The total number of individuals given antibiotics was 270 (antibiotic 177, placebo 93); the adverse reaction data are reported and analysed using these denominators.
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Bodoky 1993

---

Methods	RCT. Location: University Hospital, Switzerland. Recruitment period: 1984-1987 Adequacy of concealment of assigned allocation: A Intention to treat analysis: 2 Blinding of outcome assessors: 3 Comparability of treatment groups at entry: 3 Use of placebo: 3 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection 3 Duration of surveillance: 1 Losses to follow up: 45 of 284 (16%)
Participants	239 analysed (female 184, male 55, mean age 77 years) Inclusion criteria: Undergoing internal fixation for intracapsular hip fracture. Exclusion criteria: Pre-existing infection, Antibiotic allergy, Antibiotic prophylaxis for other reason e.g , valve, Immunosuppression, polytrauma.
Interventions	a. Cefotiam 2g induction; 2g once 12 hours later b. Placebo induction once; 12 hours later
Outcomes	1. "Major" wound infection. In the analysis, considered as "deep" 2. "Minor" wound infection In the analysis, considered as "superficial" 3. Urinary tract infection 4. Pulmonary infection 5. Septicaemia 6. Adverse drug effects 7. Infection related death
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A - Adequate

---

Boxma 1996

---

Methods	RCT. Location: Multi-centre study, Netherlands. Recruitment period: 1989-1991 Adequacy of concealment of assigned allocation: A Intention to treat analysis: 3 Blinding of outcome assessors: 3 Comparability of treatment groups at entry: 3 Use of placebo: 3 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection: 2 Duration of surveillance: 2 Losses to follow up:None described
Participants	2195 analysed (female 1132 mean age 65, male 1063, mean age 44) Inclusion criteria: undergoing surgery for closed fracture. Exclusion criteria: External fiation or percutaneous wire fixation, pregnancy, immunosuppressive treatment, known hypersensitivity to cephalosporins, antimicrobial use or symptoms of infection in the week prior to surgery.
Interventions	a. Ceftriaxone 2g given intravenously at induction of anaesthesia for surgery b. Placebo by same regimen
Outcomes	1. Deep wound infection. 2. Superficial wound infection. 3. Respiratory infection 4. Urinary tract infection.
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A - Adequate

---

Boyd 1973

---

Methods	RCT. Location: University Hospital, Boston USA Recruitment period:1966-1969. Adequacy of concealment of assigned allocation: A Intention to treat analysis: 1 Blinding of outcome assessors: 2 Comparability of treatment groups at entry: 3 Use of placebo: 3 Definition of inclusion and exclusion criteria: 2 Assessment of wound infection: 2 Duration of surveillance: 3 Losses to follow up: 68/348 (20%)
Participants	280 analysed (female 203, male 77, age range 40-90+) Inclusion criteria: undergoing operative treatment for proximal femoral fracture, either fixation or endoprosthesis. Exclusion criteria: Pre-existing infection, Antibiotic allergy.
Interventions	a. Nafcillin systemically 500mg Theatre call; 500mg 6 hourly im for 48 hours b. Placebo (glucose in saline ) Theatre call; Same regimen
Outcomes	1. Deep infection (All infected haematomas were treated as deep infections in the analysis.) 2. Superficial infection 3. Adverse drug effects 4. Infection related death
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A - Adequate

---

Buckley 1990

---

Methods	RCT. Location: University Hospital, Canada. Recruitment period: 1985-1988. Adequacy of concealment of assigned allocation: A Intention to treat analysis: 1 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 2 Use of placebo: 3 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection: 2 Duration of surveillance: 2 Losses to follow up: 40/352 (12%)
Participants	312 analysed (female 225, male 87, mean age 77 years). Inclusion criteria: undergoing hip fracture surgery. Exclusion criteria - Cephalosporin allergy, pathologic fracture, previous surgery on fractured hip, antibiotic treatment with other agent, more than 7 days in hospital pre-op.
Interventions	a. Cefazolin 2g iv induction; 1g 6 hourly iv 3 more doses b. Cefazolin 2g iv induction; placebo (saline) 6 hourly iv 3 doses c. Placebo (saline) iv induction; placebo (saline) 6 hourly iv 3 doses
Outcomes	1. Deep wound infection 2. Superficial wound infection
Notes	Numbers of patients allocated to groups a. multiple dose and c. placebo are reversed in some of the text (including the abstract) compared with the tables in the report. We used the numbers which were consistent with the results (percentages etc) given in the report.
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A - Adequate

---

Burnett 1980

---

Methods	RCT. Location: County hospital, Minneapolis USA. Recruitment period: 1973-1977. Adequacy of concealment of assigned allocation: A Intention to treat analysis: 2 Blinding of outcome assessors: 3 Comparability of treatment groups at entry: 2 Use of placebo: 3 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection: 3 Duration of surveillance: 1 Losses to follow up: 46/307 (15%)
Participants	261 analysed (161 women, 100 men, mean age 72 years) Inclusion criteria: undergoing hip fracture surgery. Exclusion criteria: history of allergy to cephalosporin, active infection requring therapy, history of previous infection about the involved hip.
Interventions	a. Cephalothin 1g iv 4 hourly for 72 hours b. Placebo iv 4 hourly for 72 hours
Outcomes	1. Deep wound infection 2. Urinary tract infection 3. Pulmonary infection 4. Septicaemia 5. Death from infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A - Adequate

---

Ericson 1973

---

Methods	RCT Location: University and regional hospital, Sweden. Recruitment period: 1970-1972. Adequacy of concealment of assigned allocation: B Intention to treat analysis: 1 Blinding of outcome assessors: 2 Comparability of treatment groups at entry: 1 Use of placebo: 3 Definition of inclusion and exclusion criteria: 1 Assessment of wound infection: 2 Duration of surveillance: 2 Losses to follow up: 59/230 (25%) overall; losses for hip fracture sub-group not described. See notes.
Participants	53 analysed. (39 internal fixation of trochanteric fracture, 14 hemiarthroplasty, mean age 70 years). Inclusion criteria: undergoing hip arthroplasty for arthritis, or hip fracture surgery. Exclusion criteria: none described.
Interventions	a. Cloxacillin 1g IM, 1 hour before operation and for three further doses at six hourly intervals. Thereafter 1g orally 6 hourly, with probenecid 1g twice daily until day 14. b. Placebo to same regimen.
Outcomes	1. Wound infection (interpreted as deep infection) 2. Adverse drug reactions
Notes	This report also contains data for use of the same regimen in open fractures. The total number of individuals given antibiotics was 171 (antibiotic 83, placebo 88); the adverse reaction data are reported and analysed using these denominators.
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Garcia 1991

---

Methods	RCT. Location: University hospital, Spain. Recruitment period: 1987-89. Adequacy of concealment of assigned allocation: A Intention to treat analysis: 3 Blinding of outcome assessors: 2 Comparability of treatment groups at entry: 3 Use of placebo: 1 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection: 3 Duration of surveillance: 3 Losses to follow up: No losses described
Participants	1489 (976 female, 513 male, mean age 68 years) analysed. Inclusion criteria: undergoing fracture fixation surgery, of which 305 had a Moores arthroplasty inserted, 697 had an Ender nail inserted, and 487 had another device used to fix a closed fracture. Exclusion criteria: undergoing total joint replacement, known allergy to cephalosporin or penicillin, receiving immunosuppression or antibiotics for other infection, history of infection in the operative field, open fracture.
Interventions	a. Cefonicid 2g iv at induction of anaesthesia b. Cefamandole 2g iv 3 doses ( 30 minutes pre-op, 2 , 8 hours) c. Cefamandole 2g iv 5 doses ( 30 minutes pre-op, 2, 8, 14, 20 hours)
Outcomes	1. Deep infection 2. Superficial infection 3. Urinary tract infection 4. Respiratory infection 5. Infection related death
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A - Adequate

---

Gatell 1984

---

Methods	RCT. Location: University Hospital, Spain. Recruitment period: not given. Adequacy of concealment of assigned allocation: B Intention to treat analysis: 2 Blinding of outcome assessors: 3 Comparability of treatment groups at entry: 1 Use of placebo: 1 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection: 2 Duration of surveillance: 3 Losses to follow up: 16 of 300 (6%)
Participants	284 analysed (169 female, 145 male, mean age 55 years) Inclusion criteria: undergoing internal fixation of a long bone fracture. Exclusion criteria: patients having a total joint replacement or a procedure directly involving the hip joint, patients on immunosuppression or with a history of sensitivity to penicillins or cephalosporins.
Interventions	a. Cefamandole 2g by intravenous injection 30 mins before surgery, and at 2, 8,14 and 20 hours after the start of surgery. b. Placebo using the same regimen as group a.
Outcomes	1. Deep wound infection 2. Superficial wound infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Gatell 1987

---

Methods	RCT. Location: University Hospital, Spain. Recruitment period: not described. Adequacy of concealment of assigned allocation: B Intention to treat analysis: 2 Blinding of outcome assessors: 3 Comparability of treatment groups at entry: 2 Use of placebo: 3 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection: 3 Duration of surveillance: 3 Losses to follow up: 33/750 (4%)
Participants	717 analysed (445 female, 272 male, mean age 65 years). Inclusion criteria: undergoing closed fracture fixation Exclusion criteria: Open fractures, previous total joint replacement, known allergy to penicillin or cephalosporin, immunosuppressive treatment, previous infection in the operative field, already on antibiotics.
Interventions	a. Single dose of cefamandole 2g by intravenous injection 30 minutes before start of surgery and four doses of placebo at 2, 8,14, and 20 hours. b. Cefamandole 2g by intravenous injection 30 minutes before the start of surgery and at 2,8,14, and 20 hours.
Outcomes	1. Deep infection 2. Superficial infection 3. Urinary tract infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Hedstrom 1987

---

Methods	RCT. Location: University Hospitals, Sweden. Recruitment period: 1982-1984 Adequacy of concealment of assigned allocation: A Intention to treat analysis: 1 Blinding of outcome assessors: 3 Comparability of treatment groups at entry: 2 Use of placebo: 3 Definition of inclusion and exclusion criteria : 2 Assessment of wound infection: 2 Duration of surveillance: 1 Losses to follow up: 26/ 147 (18%)
Participants	121 analysed (female 87, male 34, mean age 77 years) Inclusion criteria: undergoing surgery for fixation of trochanteric fracture. Exclusion criteria: decreased renal function, severe senility, known or suspected allergy to penicillins or cephalosporins.
Interventions	a. 1 day of prophylaxis (cefuroxime 750 mg thrice daily on the operative day followed by 6 days of placebo tablets) b. 7 days of prophylaxis(cefuroxime 750 mg thrice daily on the operative day followed by cephalexin 0.5g thrice daily orally for 6 days)
Outcomes	1. Deep infection 2. Superficial infection 3. Adverse effects
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Yes	A - Adequate

---

Hjortrup 1990

---

Methods	RCT. Location: Regional Hospital, Denmark. Adequacy of concealment of assigned allocation: B Intention to treat analysis: 2 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 1 Use of placebo: 1 Definition of inclusion and exclusion criteria: 2 Assessment of wound infection: 2 Duration of surveillance: 1 Losses to follow up: 15 / 200 (7.5%)
Participants	185 analysed (female 142, male 43, mean age 80 years) Inclusion criteria: Undergoing surgery for hip fracture Exclusion criteria: history of allergy to penicillin, active infection requiring antibiotic therapy, decreased renal function, patients subjected to arthroplasty.
Interventions	a. Single dose Methicillin 2g intravenously at the start of the operation. b. No antibiotic prophylaxis
Outcomes	1. Deep infection 2. Superficial infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Hughes 1991

---

Methods	RCT. Location: University Hospital, Scotland. Adequacy of concealment of assigned allocation: B Intention to treat analysis: 3 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 3 Use of placebo: 1 Definition of inclusion and exclusion criteria: 1 Assessment of wound infection: 1 Duration of surveillance: 1 Losses to follow up: none described
Participants	54 analysed (15 female, 39 male, mean age 44) Inclusion criteria: undergoing surgery for low energy closed fractures Exclusion criteria: none described
Interventions	a.Cefuroxime 1.5g at induction of anaesthesia b. No treatment
Outcomes	1. Deep wound infection 2. Superficial wound infection 3. Urinary tract infection 4. Respiratory tract infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Jones 1987 B

---

Methods	RCT. Location: Prepaid medical care program, Oregon, USA. Recruitment period: 1984-1985 Adequacy of concealment of assigned allocation: B Intention to treat analysis: 2 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 1 (see notes) Use of placebo: 1 Definition of inclusion and exclusion criteria: 2 Assessment of wound infection: 2 Duration of surveillance: 1 Losses to follow up: 122 of 1036 (12%).
Participants	914 in study of which a subgroup of 58 who had undergone fracture surgery were analysed for this review. No demographic details for the fracture subgroup. Inclusion criteria: Main analysis:undergoing elective surgery in a range of specialties. Subgroup- undergoing open reduction and fixation of fracture. Exclusion criteria: Concurrent antibiotic treatment, history of hypersensitivity to cephalosporins or penicillins, pregnancy, preoperative urinary tract infection, renal or hepatic disease.
Interventions	a. Cefazolin 1g by intravenous injection at onset of anaesthesia, and 8 hourly for 24 hrs. b. Cefoxitin 2g by intravenous injection at onset of anaesthesia and 6 hourly for 24 hrs. c. Cefotaxime 1g by intravenous injection at onset of anaesthesia (with second dose if operation time exceeded 2 hours)
Outcomes	1. Wound infection. (defined as drainage of purulent material from the wound) within 30 days of operation.
Notes	Demographic data available only for the whole group, but wound infection data for a fracture fixation subgroup.
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Karachalios 1987

---

Methods	RCT. Location: University Hospital, Greece. Recruitment period: Not described. Adequacy of concealment of assigned allocation: B Intention to treat analysis: 3 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 1 Use of placebo: 1 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection: 1 Duration of surveillance: 1 Losses to follow up: None during the 10 day follow up period.
Participants	200 analysed. Demographic details not provided. Inclusion criteria: undergoing internal fixation of proximal femoral fracture. Exclusion criteria: Known or suspected hypersenitivity to cephalosporins, concomitant or recent antibiotic use, or evidence of pre-existing infection at the time of surgery.
Interventions	a. Ceftriaxone 1g at start of surgery b. Cefotaxime 1g at start of surgery and every 8 hours for 3 days.
Outcomes	1. Deep wound infection. 2. Superficial wound infection 3. Adverse effects
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Kaukonen 1995

---

Methods	CCT. (Quasi-randomised- allocation based on odd or even year of birth) Location: Regional Hospital, Finland. Recruitment period: 1990-1991 Follow up of variable quality depending on hospital length of stay. Adequacy of concealment of assigned allocation: C Intention to treat analysis: 2 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 1 Use of placebo: 1 Definition of inclusion and exclusion criteria: 2 Assessment of wound infection: 1 Duration of active surveillance: 1 Losses to follow up: 13 / 162 (8%).
Participants	149 analysed (female 112, male 37, mean age 76 years) Inclusion criteria: undergoing operative treatment of hip fracture. Exclusion criteria: antibiotic treatment for active infection at the time of admission.
Interventions	a. Single dose prophylaxis with cefuroxime 3g at the start of the operation. b. No antibiotic prophylaxis.
Outcomes	1. Deep wound infection 2. Superficial wound infection.
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	No	C - Inadequate

---

Luthje 2000

---

Methods	RCT. Location: Multicentre. University and Community Hospitals, Finland. Recruitment period:1994-1998. Adequacy of concealment of assigned allocation: B Intention to treat analysis: 3 Blinding of outcome assessors:1 Comparability of treatment groups at entry: 3 Use of placebo: 1 Definition of inclusion and exclusion criteria: 2 Assessment of wound infection 2 Duration of surveillance: 2 Losses to follow up: None
Participants	224 analysed (female 157, male 67, mean age 77 years). Inclusion criteria: undergoing operative osteosynthesis (internal fixation) of hip fracture. Exclusion criteria: antibiotic treatment for active infection at the time of admission,history of penicillin/ cephalosporin allergy, immunosuppressive medication, open fracture, pathological fracture, under 15 years of age.
Interventions	a. Single dose prophylaxis with cefuroxime 2g IV 1-2 hours preoperatively. b. No antibiotic prophylaxis.
Outcomes	1. Deep wound infection 2. Superficial wound infection. 3. Respiratory infection. 4. Urinary infection. Also reported, but not included in this review, were thrombo-embolic events, decubitus ulcers, fixation failure, and death.
Notes	This paper contains a multivariate analysis of the most important factors predicting complications in this cohort of hip fracture patients, and an economic analysis of the cost-effectiveness of prophylaxis using this regimen.
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B- unclear

---

McQueen 1990

---

Methods	RCT. Location: University Hospital , Scotland. Recruitment period: 1985-1986 Adequacy of concealment of assigned allocation: B Intention to treat analysis: 2 Blinding of outcome assessors: 2 Comparability of treatment groups at entry: 1 Use of placebo: 3 Definition of inclusion and exclusion criteria: 1 Assessment of wound infection: 3 Duration of surveillance: 1 Losses to follow up: Not described
Participants	502 analysed (434 female, 68 male, mean age 79 yrs) Inclusion criteria: undergoing surgical management of hip fracture Exclusion criteria: none described
Interventions	a. Cefuroxime 1.5g by intravenous injection at start of operation b. Placebo by intravenous injection at start of operation
Outcomes	1. Deep wound infection 2. Superficial wound infection 3. Urinary tract infection 4. Pulmonary infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Nelson 1983

---

Methods	CCT. (Quasi-randomised by last digit of hospital number: odd or even). Location: University Hospital, USA. Recruitment period: Not described. Adequacy of concealment of assigned allocation: C Intention to treat analysis: 1 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 1 Use of placebo: 1 Definition of inclusion and exclusion criteria: 1 Assessment of wound infection: 2 Duration of surveillance: 3 Losses to follow up: 6 / 358 (2%).
Participants	103 men and women undergoing hip fracture surgery analysed. Subgroup within a larger study which also included a further 255 joint replacement patients. No demographic details provided. Inclusion criteria: undergoing hip fracture surgery Exclusion criteria: none described
Interventions	a. Intravenous antibiotics for 24 hours (either nafcillin or cefazolin 500mg 6 hourly, choice being surgeon dependent) b. Intravenous antibiotics (same agent and dose as in group a) or three days intravenously, and for a subsequent 4 days orally.
Outcomes	1. Deep wound infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	No	C - Inadequate

---

Nungu 1995

---

Methods	RCT. Location: Multi-centre recruitment from University and regional Hospitals, Sweden. Recruitment period: 1991 - 1993. Adequacy of concealment of assigned allocation: B Intention to treat analysis: 2 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 1 Use of placebo: 1 Definition of inclusion and exclusion criteria: 2 Assessment of wound infection: 3 Duration of surveillance: 2 Losses to follow up: 107/559 (19%)
Participants	559 randomised (412 women, 147 men, mean age 81 years). No demographic details for the 452 analysed. Inclusion criteria: undergoing internal fixation for an extracapsular fracture of the hip. Exclusion criteria: current antibiotic treatment, known allergy to penicillin or cephalosporin.
Interventions	a.Oral prophylaxis with cefadroxil 1g 2hours before surgery and again at 12 hours. b. Intravenous prophylaxis with cefuroxime. Regimen was 750 mg at 8 hour intervals in three participating hospitals, and 1.5g as a single dose in one centre.
Outcomes	1. Deep wound infection 2. Superficial wound infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Paiement 1994

---

Methods	RCT Location: Two University Hospitals, USA and Canada Recruitment period: 1989-1990 Adequacy of concealment of assigned allocation: B Intention to treat analysis: 2 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 3 Use of placebo: 3 Definition of inclusion and exclusion criteria: 3 Assessment of wound infection: 2 Duration of surveillance: 3 Losses to follow up:3/125 (2%)
Participants	122 analysed (64 female, 58 male , mean age 43 yrs) Inclusion criteria: Undergoing surgery for isolated closed fracture Exclusion criteria: Aged less than 18 years, concomitant infection, surgery delayed over 72 hours, extensive skin blistering, history of allergy to penicillin or cephalosporin, receiving antibiotic or immunosuppression.
Interventions	a. Cephalotin 1g at least 7 minutes before application of surgical tourniquet, and 6 hourly for 24 hours. b. Placebo, same regimen
Outcomes	1. Deep infection 2. Superficial infection. 3. Adverse drug reactions (none)
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	Unclear	B - Unclear

---

Tengve 1978

---

Methods	CCT (Quasi-randomised -allocation by year of birth). Location: Regional Hospital, Sweden. Recruitment period: 1973-74 Adequacy of concealment of assigned allocation: C Intention to treat analysis: 1 Blinding of outcome assessors: 1 Comparability of treatment groups at entry: 1 Use of placebo: 1 Definition of inclusion and exclusion criteria: 2 Assessment of wound infection: 2 Duration of surveillance: 2 Losses to follow up: 13 / 140 (9%).
Participants	127 analysed (table states 38 women, 89 men, average age 77 years). Female/male ratio is reversed from that expected, and may be an error in the report. Inclusion criteria: undergoing surgery for trochanteric hip fracture Exclusion criteria: none described.
Interventions	a.Cephalothin 2g at start of anaesthesia, at two hours, and every six hours until tolerating oral intake; thereafter 1g cephalexin orally 6 hourly for total of 48 hours/ b. No prophylaxis.
Outcomes	1. Deep infection 2. Superficial infection
Notes
Risk of bias
Item	Authors' judgement	Description
Allocation concealment?	No	C - Inadequate

---

 g = gram
iv = intravenous
im = intra-muscular

 

Characteristics of excluded studies [ordered by study ID]

---

Study	Reason for exclusion
Benson 1983	CCT. Open fracture management. One comparison is of antibiotic management.
Braun 1987	RCT. Open extremity fractures only. Comparison: cloxacillin and placebo.
Centulio 1988	RCT. Comparison of single dose of ceftriaxone with 3 days of ceftriaxone. Uncemented joint replacement used in hip surgery. 13 hip fracture patients included but no usable data for this subgroup.
De Benedictis 1984	RCT. Comparison: cefonicid versus cefazolin. Joint replacement patients only.
Dellinger 1988	RCT. Open fracture management. Comparison between one day of antibiotic prophylaxis with a cephalosporin, and five days (2 groups - each a different cephalosporin).
Evrard 1988	RCT. Comparison of 5 days of cefazolin versus two days of cefamandole .. Hip replacement data only.
Gunst 1984	RCT. Comparison: parenteral cefamandole and no antibiotic. Total hip replacement data only.
Henley 1986	RCT. Comparison: intravenous cefamandole versus placebo. 130 of 743 participants had fracture surgery, but no usable data for this group are available.
Hill 1981	RCT. Comparison: cefazolin by injection over 5 days versus placebo. Total joint replacement patients only.
Johnson 1988	RCT. Antibiotic management in open tibial fractures.
Jones 1985	RCT. Participants 30 patients undergoing lower limb fracture surgery. Conducted to evaluate levels of three cephalosporins in serum, muscle and haematoma. No usable data on clinical outcomes.
Jones 1987 A	RCT. Comparison: single dose cefoperazone versus two doses of cefotaxime. Participants included subgroup undergoing joint replacement and other unspecified orthopaedic procedures. No usable data on fracture fixation.
Knapp 1996	RCT. Antibiotic management of open fractures caused by gunshot wounds.
Lidwell 1984	RCT. Participants: hip and knee replacement. Randomised to clean air or not.
Mauerhan 1994	RCT. Comparison one day of cefuroxime compared with three days of cefazolin. Total joint replacement only.
Mollan 1994	RCT. Comparison: single dose teicoplanin versus 4 doses of cephamandole over 24 hrs. Total joint replacement patients only.
Patzakis 1977	RCT. Comparison of antibiotic regimens in the management of open fractures.
Pavel 1974	RCT. Comparison: cephaloridine versus placebo. Participants were undergoing clean orthopaedic surgery, the procedures being unspecified. It is likely that closed fracture patients were amongst these, but there are no usable data.
Periti 1989	RCT comparing single dose prophylaxis using ceftriaxone with multi-dose prophylaxis using cefamandole. Only 9 of 883 participants underwent fracture surgery, and no usable data were available.
Periti 1992	RCT. Comparison: single dose teicoplanin versus multiple dose cefazolin. All participants underwent total hip or knee replacement.
Pollard 1979	RCT. Comparison: 3 doses of cephaloridine versus 14 days of flucloxacillin. Joint replacement patients only.
Schulitz 1980	RCT. Joint replacement patients only.
Vainionpaa 1988	RCT comparing narrow-spectrum agent (cloxacillin) with a broad spectrum agent (cefamandole). Joint replacement patients only included.
Van Meirhaeghe 1989	RCT. Comparison: narrow spectrum antibiotic (flucloxacillin) versus broad spectrum (cefazolin) in clean orthopaedic surgery. A proportion (not described) were fracture fixation but a wide range of other clean orthopaedic operations were included. No usable data for this review.
Visuri 1976	CCT (alternation). Narrow spectrum (dicloxacillin) versus broad spectrum (ampicillin) in patients undergoing total joint replacement.
Wymenga 1992	RCT. Single dose versus multiple dose of cephalosporin in hip replacement. Some hip fracture patients are included but no data are available for this subgroup.

---

 

Characteristics of studies awaiting assessment [ordered by study ID]

Chiu 1993

---

Methods	?RCT
Participants
Interventions
Outcomes
Notes	Abstract only. Authors contacted for more information. No reply to date.

---

 

Comparison 1. Antibiotic agent (mulitple dose) versus placebo or no treatment

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Deep wound infection

10

1896

Risk Ratio (M-H, Fixed, 95% CI)

0.36 [0.21, 0.65]

---

1.1 Hip fracture fixation

4

526

Risk Ratio (M-H, Fixed, 95% CI)

0.22 [0.07, 0.75]

---

1.2 Hip endoprosthesis

1

14

Risk Ratio (M-H, Fixed, 95% CI)

0.56 [0.03, 11.57]

---

1.3 Unspecified hip fracture procedure

3

770

Risk Ratio (M-H, Fixed, 95% CI)

0.47 [0.21, 1.04]

---

1.4 Operative management of other or unspecified closed fracture

3

586

Risk Ratio (M-H, Fixed, 95% CI)

0.35 [0.10, 1.24]

---

 2 Superficial wound infection

7

1366

Risk Ratio (M-H, Fixed, 95% CI)

0.48 [0.28, 0.81]

---

 3 Urinary tract infection

2

500

Risk Ratio (M-H, Fixed, 95% CI)

0.66 [0.43, 1.00]

---

 4 Respiratory infection

2

500

Risk Ratio (M-H, Fixed, 95% CI)

0.81 [0.41, 1.63]

---

 5 Adverse drug effects

2

882

Risk Ratio (M-H, Fixed, 95% CI)

1.83 [0.96, 3.50]

---

5.1 Gastro-intestinal symptoms

2

441

Risk Ratio (M-H, Fixed, 95% CI)

2.05 [0.99, 4.25]

---

5.2 Skin reactions

2

441

Risk Ratio (M-H, Fixed, 95% CI)

1.26 [0.30, 5.22]

---

 

Comparison 2. Antibiotic agent (single dose) versus placebo or no treatment

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Deep wound infection

7

3500

Risk Ratio (M-H, Fixed, 95% CI)

0.40 [0.24, 0.67]

---

1.1 Unspecified hip fracture procedure

5

1251

Risk Ratio (M-H, Fixed, 95% CI)

0.68 [0.28, 1.66]

---

1.2 Operative management of other or unspecified closed fracture

2

2249

Risk Ratio (M-H, Fixed, 95% CI)

0.32 [0.17, 0.60]

---

 2 Superficial wound infection

7

3500

Risk Ratio (M-H, Fixed, 95% CI)

0.69 [0.50, 0.95]

---

 3 Urinary tract infection

4

2975

Risk Ratio (M-H, Fixed, 95% CI)

0.63 [0.53, 0.76]

---

 4 Respiratory infection

4

2975

Risk Ratio (M-H, Fixed, 95% CI)

0.46 [0.33, 0.65]

---

 

Comparison 3. Single dose short-acting versus multiple dose - same agent

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Deep wound infection

2

921

Risk Ratio (M-H, Fixed, 95% CI)

7.89 [1.01, 61.97]

---

1.1 Unspecified hip fracture procedure

1

204

Risk Ratio (M-H, Fixed, 95% CI)

Not estimable

---

1.2 Operative management of other or unspecified closed fracture

1

717

Risk Ratio (M-H, Fixed, 95% CI)

7.89 [1.01, 61.97]

---

 2 Superficial wound infection

1

717

Risk Ratio (M-H, Fixed, 95% CI)

4.82 [1.08, 21.61]

---

 3 Urinary tract infection

1

717

Risk Ratio (M-H, Fixed, 95% CI)

1.81 [1.01, 3.23]

---

 

Comparison 4. Single dose long acting versus any multiple dose regimen

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Deep wound infection

3

1747

Risk Ratio (M-H, Fixed, 95% CI)

0.57 [0.20, 1.64]

---

1.1 Unspecified hip fracture procedure

2

505

Risk Ratio (M-H, Fixed, 95% CI)

0.66 [0.12, 3.59]

---

1.2 Operative management of other or unspecified closed fracture

2

1242

Risk Ratio (M-H, Fixed, 95% CI)

0.52 [0.13, 2.02]

---

 2 Superficial wound infection

2

1689

Risk Ratio (M-H, Fixed, 95% CI)

1.01 [0.35, 2.93]

---

 3 Urinary tract infection

1

1489

Risk Ratio (M-H, Fixed, 95% CI)

0.69 [0.37, 1.32]

---

 4 Respiratory infection

1

1489

Risk Ratio (M-H, Fixed, 95% CI)

0.31 [0.04, 2.48]

---

 

Comparison 5. Operative day only versus longer prophylaxis

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Deep wound infection

2

224

Risk Ratio (M-H, Fixed, 95% CI)

1.10 [0.22, 5.34]

---

1.1 Unspecified hip fracture procedure

2

224

Risk Ratio (M-H, Fixed, 95% CI)

1.10 [0.22, 5.34]

---

 2 Superficial wound infection

1

121

Risk Ratio (M-H, Fixed, 95% CI)

0.57 [0.17, 1.93]

---

 

Comparison 6. Oral versus parenteral administration of antibiotic agent

---

Outcome or subgroup title

No. of studies

No. of participants

Statistical method

Effect size

---

 1 Deep wound infection

1

452

Risk Ratio (M-H, Fixed, 95% CI)

0.29 [0.01, 7.07]

---

1.1 Unspecified hip fracture procedure

1

452

Risk Ratio (M-H, Fixed, 95% CI)

0.29 [0.01, 7.07]

---

 2 Superficial wound infection

1

452

Risk Ratio (M-H, Fixed, 95% CI)

0.17 [0.02, 1.47]

---

 3 Urinary tract infection

1

452

Risk Ratio (M-H, Fixed, 95% CI)

1.10 [0.62, 1.94]

---