Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) related complications in surgical patients

  • Review
  • Intervention

Authors


Abstract

Background

Risk of methicillin-resistant Staphylococcus aureus (MRSA) infection after surgery is generally low, but affects up to 33% of patients after certain types of surgery. Postoperative MRSA infection can occur as surgical site infections (SSIs), chest infections, or bloodstream infections (bacteraemia). The incidence of MRSA SSIs varies from 1% to 33% depending upon the type of surgery performed and the carrier status of the individuals concerned. The optimal prophylactic antibiotic regimen for the prevention of MRSA after surgery is not known.

Objectives

To compare the benefits and harms of all methods of antibiotic prophylaxis in the prevention of postoperative MRSA infection and related complications in people undergoing surgery.

Search methods

In March 2013 we searched the following databases: The Cochrane Wounds Group Specialised Register; The Cochrane Central Register of Controlled Trials (CENTRAL); Database of Abstracts of Reviews of Effects (DARE) (The Cochrane Library); NHS Economic Evaluation Database (The Cochrane Library); Health Technology Assessment (HTA) Database (The Cochrane Library); Ovid MEDLINE; Ovid MEDLINE (In-Process & Other Non-Indexed Citations); Ovid EMBASE; and EBSCO CINAHL.

Selection criteria

We included only randomised controlled trials (RCTs) that compared one antibiotic regimen used as prophylaxis for SSIs (and other postoperative infections) with another antibiotic regimen or with no antibiotic, and that reported the methicillin resistance status of the cultured organisms. We did not limit our search for RCTs by language, publication status, publication year, or sample size.

Data collection and analysis

Two review authors independently identified the trials for inclusion in the review, and extracted data. We calculated the risk ratio (RR) with 95% confidence intervals (CI) for comparing binary outcomes between the groups and planned to calculated the mean difference (MD) with 95% CI for comparing continuous outcomes. We planned to perform meta-analysis using both a fixed-effect model and a random-effects model. We performed intention-to-treat analysis whenever possible.

Main results

We included 12 RCTs, with 4704 participants, in this review. Eleven trials performed a total of 16 head-to-head comparisons of different prophylactic antibiotic regimens. Antibiotic prophylaxis was compared with no antibiotic prophylaxis in one trial. All the trials were at high risk of bias. With the exception of one trial in which all the participants were positive for nasal carriage of MRSA or had had previous MRSA infections, it does not appear that MRSA was tested or eradicated prior to surgery; nor does it appear that there was high prevalence of MRSA carrier status in the people undergoing surgery.

There was no sufficient clinical similarity between the trials to perform a meta-analysis. The overall all-cause mortality in four trials that reported mortality was 14/1401 (1.0%) and there were no significant differences in mortality between the intervention and control groups in each of the individual comparisons. There were no antibiotic-related serious adverse events in any of the 561 people randomised to the seven different antibiotic regimens in four trials (three trials that reported mortality and one other trial). None of the trials reported quality of life, total length of hospital stay or the use of healthcare resources. Overall, 221/4032 (5.5%) people developed SSIs due to all organisms, and 46/4704 (1.0%) people developed SSIs due to MRSA.

In the 15 comparisons that compared one antibiotic regimen with another, there were no significant differences in the proportion of people who developed SSIs. In the single trial that compared an antibiotic regimen with placebo, the proportion of people who developed SSIs was significantly lower in the group that received antibiotic prophylaxis with co-amoxiclav (or cefotaxime if allergic to penicillin) compared with placebo (all SSI: RR 0.26; 95% CI 0.11 to 0.65; MRSA SSI RR 0.05; 95% CI 0.00 to 0.83). In two trials that reported MRSA infections other than SSI, 19/478 (4.5%) people developed MRSA infections including SSI, chest infection and bacteraemia. There were no significant differences in the proportion of people who developed MRSA infections at any body site in these two comparisons.

Authors' conclusions

Prophylaxis with co-amoxiclav decreases the proportion of people developing MRSA infections compared with placebo in people without malignant disease undergoing percutaneous endoscopic gastrostomy insertion, although this may be due to decreasing overall infection thereby preventing wounds from becoming secondarily infected with MRSA. There is currently no other evidence to suggest that using a combination of multiple prophylactic antibiotics or administering prophylactic antibiotics for an increased duration is of benefit to people undergoing surgery in terms of reducing MRSA infections. Well designed RCTs assessing the clinical effectiveness of different antibiotic regimens are necessary on this topic.

Résumé scientifique

Prophylaxie antibiotique pour la prévention des complications liées au Staphylococcus aureus résistant à la méthicilline (SARM) chez les patients chirurgicaux

Contexte

Le risque d'infection à Staphylococcus aureus résistant à la méthicilline (SARM) après une intervention chirurgicale.est généralement faible, mais touche jusqu'à 33 % des patients après la réalisation de certains types de chirurgie. L'infection à SARM post-opératoire peut se manifester par des infections du site opératoire, des infections thoraciques ou des infections sanguines (bactériémie). L'incidence des infections à SARM du site opératoire varie de 1 % à 33 % en fonction du type d'intervention chirurgicale pratiquée et du statut de porteur des individus concernés. Le schéma thérapeutique antibiotique prophylactique optimal pour la prévention du SARM après une intervention chirurgicale n'a pas encore été établi.

Objectifs

Comparer les bénéfices et les risques de toutes les méthodes de prophylaxie antibiotique dans la prévention de l'infection à SARM post-opératoire et des complications associées chez les personnes ayant subi une intervention chirurgicale.

Stratégie de recherche documentaire

En mars 2013, nous avons effectué des recherches dans les bases de données électroniques suivantes : le registre spécialisé du groupe Cochrane sur les plaies et contusions ; le registre Cochrane des essais contrôlés (CENTRAL) ; la base des résumés des revues systématiques hors Cochrane (DARE) (The Cochrane Library) ; la base de données d'évaluation de l'économie de la santé (NHS-Economic Evaluation Database) (The Cochrane Library) ; la base de données d'évaluation des technologies de la santé (Health Technology Assessment (HTA)) (The Cochrane Library) ; Ovid MEDLINE ; Ovid MEDLINE (In-Process & Other Non-Indexed Citations) ; Ovid EMBASE ; et EBSCO CINAHL.

Critères de sélection

Nous avons inclus uniquement les essais contrôlés randomisés (ECR) ayant comparé un schéma thérapeutique antibiotique utilisé pour la prophylaxie contre les infections du site opératoire (et d'autres infections post-opératoires) à un autre schéma thérapeutique antibiotique ou à l'absence d'antibiotique, et ayant indiqué le statut de résistant à la méthicilline des organismes mis en culture. Nous n'avons pas limité la recherche d'ECR par langue, statut de publication, année de publication ou effectif.

Recueil et analyse des données

Deux auteurs de la revue ont sélectionné les essais à inclure dans la revue et extrait les données de façon indépendante. Nous avons calculé le risque relatif (RR) avec des intervalles de confiance (IC) à 95 % pour comparer les résultats binaires entre les groupes, et avions prévu de calculer la différence moyenne (DM) avec un IC à 95 % pour comparer les résultats continus. Nous avions prévu d'effectuer la méta-analyse en utilisant à la fois un modèle à effets aléatoires et un modèle à effets fixes. Nous avons extrait les résultats par une analyse en intention de traiter quand cela était possible.

Résultats principaux

Nous avons inclus 12 essais totalisant 4 704 participants dans cette revue. Onze essais ont organisé un total de 16 comparaisons directes de différents schémas thérapeutiques antibiotiques prophylactiques. La prophylaxie antibiotique a été comparée à l'absence de prophylaxie antibiotique dans un essai. Tous les essais étaient à risque élevé de biais. À l'exception d'un essai dans lequel tous les participants étaient positifs pour la transmission par voie nasale de SARM ou avaient précédemment eu des infections à SARM, il ne semble pas que le SARM ait été testé ou éradiqué préalablement à la chirurgie ; il ne semble pas non plus qu'il y ait eu une forte prévalence du statut de porteur de SARM chez les personnes ayant subi une intervention chirurgicale.

Il n'existait pas suffisamment de similarité clinique entre les essais pour réaliser une méta-analyse. La mortalité globale toutes causes confondues dans quatre essais ayant rendu compte de la mortalité était de 14/1 401 (1,0 %) et il n'existait pas de différence significative en termes de mortalité entre les groupes avec intervention et témoin dans chacune des comparaisons individuelles. Il n'y a pas eu d'événement indésirable grave lié aux antibiotiques chez aucune des 561 personnes randomisées dans les sept différents schémas thérapeutiques antibiotiques dans quatre essais (trois essais ayant rendu compte de la mortalité et un autre essai). Aucun des essais n'a rendu compte de la qualité de vie, de la longueur totale du séjour hospitalier ou de l'utilisation des ressources de santé. Globalement, 221/4 032 personnes (5,5 %) ont développé des infections du site opératoire dues à tous les organismes, et 46/4 704 personnes (1,0 %) ont développé des infections du site opératoire dues à SARM.

Dans les 15 comparaisons opposant un schéma thérapeutique antibiotique par rapport à un autre, il n'y avait aucune différence significative dans la proportion de personnes qui ont développé des infections du site opératoire. Dans l'unique essai qui a comparé un schéma thérapeutique antibiotique à un placebo, la proportion de personnes qui ont développé des infections du site opératoire était significativement plus faible dans le groupe recevant une prophylaxie antibiotique à base de co-amoxiclav (ou de céfotaxime en cas d'allergie à la pénicilline) par rapport à un placebo (toutes les infections du site opératoire : RR 0,26 ; IC à 95 % 0,11 à 0,65 ; infections à SARM du site opératoire RR 0,05 ; IC à 95 % 0,00 à 0,83). Dans deux essais ayant rapporté des infections à SARM autres que les infections du site opératoire, 19/478 personnes (4,5 %) ont développé des infections à SARM incluant les infections du site opératoire, les infections thoraciques et les bactériémies. Il n'y avait aucune différence significative dans la proportion de personnes qui ont développé des infections à SARM dans n'importe quelle partie du corps dans ces deux comparaisons.

Conclusions des auteurs

La prophylaxie à base de co-amoxiclav réduit la proportion de personnes qui ont développé des infections à SARM comparée à un placebo chez les personnes ne présentant pas de pathologie maligne faisant l'objet d'une gastrostomie (pose d'une sonde d'alimentation gastrique) par voie endoscopique percutanée, bien que cela puisse être dû à la réduction globale de l'infection empêchant ainsi les infections secondaires des plaies dues à SARM. À l'heure actuelle, il n'existe pas d'autres preuves suggérant que l'utilisation d'une association de différents antibiotiques prophylactiques, ou que l'administration d'antibiotiques prophylactiques pendant une durée plus longue soit bénéfique pour les personnes ayant subi une intervention chirurgicale en termes de réduction des infections à SARM. Des ECR bien conçus évaluant l'efficacité clinique de différents schémas thérapeutiques antibiotiques sont nécessaires dans ce domaine.

Plain language summary

Using an antibiotic to prevent MRSA (methicillin-resistant Staphylococcus aureus) infections and related complications in people having surgery

Most bacterial wound infections after surgery heal naturally or after treatment with antibiotics.  Some bacteria are resistant to commonly-used antibiotics, e.g. methicillin-resistant Staphylococcus aureus (MRSA). MRSA infection after surgery is rare, but can occur in wounds (surgical site infections, or SSI), the chest, or bloodstream (bacteraemia), and can be life-threatening. MRSA SSIs occur in 1% to 33% of people having surgery (depending on the type of operation) and result in extended hospitalisation.

Antibiotics can be used individually, or combined, and administered for different durations. To identify the best antibiotic(s), or dose pattern, for preventing development of MRSA infection after surgery, we investigated studies that compared different antibiotics with each other, or with no treatment, to prevent MRSA SSIs. We included only randomised controlled trials (RCTs), and set no limits regarding language, or date, of publication, or trial size. Two review authors identified studies and extracted data independently.

We identified 12 RCTs, with 4704 participants. Eleven trials compared 16 preventative (prophylactic) antibiotic treatments, and one compared antibiotic prophylaxis with no prophylaxis. Generally, MRSA status of the participants prior to surgery was not known.

Four studies reported deaths (14/1401 participants): approximately 1% of participants died from any cause after surgery, but there were no significant differences between treatment groups. Four trials reported on serious antibiotic-related adverse events - there were none in 561 participants. None of the trials reported quality of life, length of hospital stay or use of healthcare resources. Overall, 221 SSIs due to any bacterium developed in 4032 people (6%), and 46 MRSA SSIs developed in 4704 people (1%). There were no significant differences in development of SSIs between the 15 comparisons of one antibiotic treatment against another. When antibiotic prophylaxis with co-amoxiclav was compared with no antibiotic prophylaxis, a significantly lower proportion of people developed SSIs after receiving co-amoxiclav (74% reduction in all SSIs, and 95% reduction in MRSA SSIs).

Two trials reported that 19 participants developed MRSA infection in wounds (SSIs), chest, or bloodstream, but there were no significant differences in the proportion of people who developed them between the two comparisons.

Prophylaxis with co-amoxiclav decreases the proportion of people developing MRSA infections compared with no antibiotic prophylaxis in people without cancer undergoing surgery for feeding tube insertion into the stomach using endoscopy, although this may be due to decreasing overall infection thereby preventing wounds from becoming secondarily infected with MRSA. There is currently no other evidence that either a combination of prophylactic antibiotics, or increased duration of antibiotic treatment, benefits people undergoing surgery in terms of reducing MRSA infections. Well-designed RCTs are necessary to assess different antibiotic treatments for preventing MRSA infections after surgery.

Résumé simplifié

Utilisation d'un antibiotique pour prévenir les infections à SARM (Staphylococcus aureus résistant à la méthicilline) et les complications associées chez les personnes ayant subi une intervention chirurgicale

La plupart des infections bactériennes de plaies suite à une intervention chirurgicale guérissent naturellement ou après traitement par des antibiotiques.  Certaines bactéries présentent une résistance à des antibiotiques couramment utilisés, par exemple Staphylococcus aureus résistant à la méthicilline (SARM). L'infection à SARM après la chirurgie est rare, mais peut survenir dans des plaies (infections du site opératoire, ou SSI), le thorax, ou la circulation sanguine (bactériémie), et peut engager le pronostic vital. Les infections à SARM du site opératoire surviennent chez 1 % à 33 % des personnes ayant subi une intervention chirurgicale (en fonction du type d'opération) et aboutissent à des hospitalisations prolongées.

Les antibiotiques peuvent être utilisés individuellement, ou en association, et être administrés pendant des durées différentes. Pour identifier le(s) meilleur(s) antibiotique(s), ou le meilleur schéma posologique, dans le cadre de la prévention du développement d'une infection à SARM après une intervention chirurgicale, nous avons examiné les études ayant comparé différents antibiotiques les uns par rapport aux autres, ou à l'absence de traitement, pour prévenir les infections à SARM du site opératoire. Nous avons inclus uniquement les essais contrôlés randomisés (ECR), et n'avons imposé aucune limite concernant la langue, ou la date, de publication, ou la taille de l'échantillon de l'essai. Deux auteurs ont indépendamment identifié les études et extrait les données.

Nous avons identifié 12 ECR portant sur 4 704 participants. Onze essais ont comparé 16 traitements antibiotiques préventifs (prophylactiques), et un essai a comparé la prophylaxie antibiotique à l'absence de prophylaxie. En règle générale, le statut de porteur de SARM des participants préalablement à la chirurgie n'était pas connu.

Quatre études ont signalé des décès (14/1 401 participants) : environ 1 % des participants sont décédés d'une cause consécutive à la chirurgie, mais il n'y avait aucune différence significative entre les groupes de traitement. Quatre essais ont rapporté des événements indésirables graves liés aux antibiotiques ; il n'y en a eu aucun parmi 561 participants. Aucun des essais n'a rendu compte de la qualité de vie, de la longueur du séjour hospitalier ou de l'utilisation des ressources de santé. Globalement, 221 infections du site opératoire dues à une bactérie quelconque se sont développées chez 4 032 personnes (6 %), et 46 infections à SARM du site opératoire se sont développées chez 4 704 personnes (1 %). Il n'y avait aucune différence significative dans le développement des infections du site opératoire entre les 15 comparaisons d'un traitement antibiotique par rapport à un autre. Lorsque la prophylaxie antibiotique à base de co-amoxiclav a été comparée à l'absence de prophylaxie antibiotique, il est apparu qu'une proportion significativement plus faible de personnes avaient développé des infections du site opératoire après avoir reçu le co-amoxiclav (74 % de réduction de toutes les infections du site opératoire, et 95 % de réduction des infections à SARM du site opératoire).

Deux essais ont rapporté que 19 participants ont développé une infection à SARM du site opératoire dans les plaies, le thorax, ou la circulation sanguine, mais il n'y avait aucune différence significative au niveau de la proportion des personnes qui les ont développées entre les deux comparaisons.

La prophylaxie à base de co-amoxiclav réduit la proportion de personnes ayant développé des infections à SARM comparée à l'absence de prophylaxie antibiotique chez les personnes ne présentant pas de cancer ayant subi une intervention chirurgicale pour l'insertion d'une sonde d'alimentation dans l'estomac par voie endoscopique, bien que cela puisse être dû à la réduction globale de l'infection empêchant ainsi les infections secondaires des plaies dues à SARM. À l'heure actuelle, il n'existe pas d'autres preuves indiquant que l'une des associations d'antibiotiques prophylactiques, ou qu'une augmentation de la durée du traitement antibiotique, soit bénéfique pour les personnes ayant subi une intervention chirurgicale en termes de réduction des infections à SARM. Des ECR bien conçus sont nécessaires pour évaluer différents traitements antibiotiques pour la prévention des infections à SARM après une intervention chirurgicale.

Notes de traduction

Traduit par: French Cochrane Centre 24th September, 2013
Traduction financée par: Pour la France : Minist�re de la Sant�. Pour le Canada : Instituts de recherche en sant� du Canada, minist�re de la Sant� du Qu�bec, Fonds de recherche de Qu�bec-Sant� et Institut national d'excellence en sant� et en services sociaux.

Summary of findings(Explanation)

Summary of findings for the main comparison. Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: mortality
  1. 1 The risk of bias in the trial was high
    2 The confidence intervals overlapped 1 and/or 0.75 and 1.25. There were fewer than 300 events in total in the intervention and control groups

Mortality
Patient or population: surgical patients
Settings: secondary
Intervention: comparison of different regimens of prophylactic antibiotic regimens
ComparisonsIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Assumed riskCorresponding risk
Control Comparison of different regimens of prophylactic antibiotic regimens
One dose of cefotiam or cefmetazol versus four doses of cefotiam or cefmetazol (elective surgery for colon cancer)See commentSee commentNot estimable275
(1 study)
⊕⊝⊝⊝
very low 1,2
Kanamycin, erythromycin and cefotiam versus cefotiam (surgery for colorectal diseases)See commentSee commentNot estimable143
(1 study)
⊕⊝⊝⊝
very low 1,2
Co-amoxiclav or cefotaxime versus placebo (percutaneous endoscopic gastrostomy) 146 per 1000 79 per 1000
(25 to 251)
RR 0.54
(0.17 to 1.72)
99
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin versus cefuroxime (coronary artery bypass graft without valvular disease) 2 per 1000 5 per 1000
(0 to 50)
RR 2.02
(0.18 to 22.18)
884
(1 study)
⊕⊝⊝⊝
very low 1,2
*The basis for the assumed risk is the control group risk in the study. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence:
High quality: further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: we are very uncertain about the estimate

Summary of findings 2 Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: surgical site infection

Summary of findings 2. Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: surgical site infection
  1. 1 The risk of bias in the trial was high
    2 The confidence intervals overlapped 1 and/or 0.75 and 1.25. There were fewer than 300 events in total in the intervention and control groups

Surgical site infection
Patient or population: surgical patients
Settings: secondary
Intervention: comparison of different regimens of prophylactic antibiotic regimens
ComparisonsIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Assumed riskCorresponding risk
Control Comparison of different regimens of prophylactic antibiotic regimens
Cefamendole versus cefamendole and gentamycin (median sternotomy) Moderate RR 5.08
(0.24 to 105.24)
522
(1 study)
⊕⊝⊝⊝
very low 1,2
5 per 1000 25 per 1000
(1 to 526)
Cefazolin versus cefamendole (median sternotomy) 8 per 1000 27 per 1000
(6 to 131)
RR 3.55
(0.75 to 16.95)
514
(1 study)
⊕⊝⊝⊝
very low 1,2
Cefazolin versus cefazolin and gentamycin (median sternotomy) 32 per 1000 28 per 1000
(10 to 75)
RR 0.87
(0.32 to 2.36)
508
(1 study)
⊕⊝⊝⊝
very low 1,2
Co-amoxiclav or cefotaxime versus placebo (percutaneous endoscopic gastrostomy) 375 per 1000 98 per 1000
(41 to 244)
RR 0.26
(0.11 to 0.65)
99
(1 study)
⊕⊕⊝⊝
low 1
Daptomycin and cefazolin versus cefazolin (vascular surgery) 129 per 1000 39 per 1000
(9 to 177)
RR 0.3
(0.07 to 1.37)
113
(1 study)
⊕⊝⊝⊝
very low 1,2
Kanamycin, erythromycin and cefotiam versus cefotiam (surgery for colorectal diseases) 239 per 1000 110 per 1000
(50 to 242)
RR 0.46
(0.21 to 1.01)
143
(1 study)
⊕⊝⊝⊝
very low 1,2
Levofloxacin versus ofloxacin (breast cancer) 61 per 1000 40 per 1000
(11 to 146)
RR 0.66
(0.18 to 2.39)
181
(1 study)
⊕⊝⊝⊝
very low 1,2
One day of piperacillin, cefazolin, cefmetazole, or cefotiam versus three days of piperacillin, cefazolin, cefmetazole, or cefotiam (elective colorectal surgery) 54 per 1000 57 per 1000
(28 to 116)
RR 1.06
(0.52 to 2.15)
521
(1 study)
⊕⊝⊝⊝
very low 1,2
One dose of cefotiam or cefmetazol versus four doses of cefotiam or cefmetazol (elective surgery for colon cancer) 65 per 1000 51 per 1000
(19 to 134)
RR 0.79
(0.3 to 2.07)
275
(1 study)
⊕⊝⊝⊝
very low 1,2
Pefloxacin versus cefazolin and oxacillin (tibial fracture requiring external fixation) 90 per 1000 67 per 1000
(39 to 115)
RR 0.74
(0.43 to 1.28)
616
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin and cefazolin versus cefazolin (open fractures) 87 per 1000 87 per 1000
(23 to 327)
RR 1
(0.27 to 3.76)
92
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin and cefazolin versus cefazolin (vascular surgery) - suggest add for all comparisons 129 per 1000 125 per 1000
(49 to 323)
RR 0.97
(0.38 to 2.5)
118
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin and cefazolin versus daptomycin and cefazolin (vascular surgery) 39 per 1000 125 per 1000
(27 to 575)
RR 3.19
(0.69 to 14.65)
107
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin versus teicoplanin (paediatric cardiac surgery)No infection in either groupNot estimable22
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin versus cefuroxime (coronary artery bypass graft without valvular disease) 32 per 1000 34 per 1000
(17 to 70)
RR 1.08
(0.53 to 2.21)
884
(1 study)
⊕⊝⊝⊝
very low 1,2
*The basis for the assumed risk is the control group risk in the study. When there were no events in either group, we have indicated so. When there were events in the intervention group but not in the control group, we have used a moderate proportion of 0.5% in the control group. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence:
High quality: further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: we are very uncertain about the estimate

Summary of findings 3 Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: MRSA surgical site infection

Summary of findings 3. Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: MRSA surgical site infection
  1. 1 The risk of bias in the trial was high
    2 The confidence intervals overlapped 1 and/or 0.75 and 1.25. There were fewer than 300 events in total in the intervention and control groups

MRSA surgical site infection
Patient or population: surgical patients
Settings: secondary
Intervention: comparison of different regimens of prophylactic antibiotic regimens
ComparisonsIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Assumed riskCorresponding risk
Control Comparison of different regimens of prophylactic antibiotic regimens
Cefamendole versus cefamendole and gentamycin (median sternotomy)No infection in either groupNot estimable522
(1 study)
⊕⊝⊝⊝
very low 1,2
Cefazolin and gentamycin versus cefamendole and gentamycin (median sternotomy)No infection in either groupNot estimable516
(1 study)
⊕⊝⊝⊝
very low 1,2
Cefazolin versus cefamendole (median sternotomy) Moderate RR 3.05
(0.12 to 74.45)
514
(1 study)
⊕⊝⊝⊝
very low 1,2
5 per 1000 15 per 1000
(1 to 372)
Cefazolin versus cefazolin and gentamycin (median sternotomy) Moderate RR 2.98
(0.12 to 72.72)
508
(1 study)
⊕⊝⊝⊝
very low 1,2
5 per 1000 15 per 1000
(1 to 364)
Co-amoxiclav or cefotaxime versus placebo (percutaneous endoscopic gastrostomy) 188 per 1000 9 per 1000
(0 to 156)
RR 0.05
(0 to 0.83)
99
(1 study)
⊕⊕⊝⊝
low 1
Daptomycin and cefazolin versus cefazolin (vascular surgery) 32 per 1000 8 per 1000
(0 to 159)
RR 0.24
(0.01 to 4.94)
113
(1 study)
⊕⊝⊝⊝
very low 1,2
Ertapenem versus cefotetan (elective colorectal surgery) 15 per 1000 9 per 1000
(2 to 37)
RR 0.59
(0.14 to 2.46)
672
(1 study)
⊕⊝⊝⊝
very low 1,2
Kanamycin, erythromycin and cefotiam versus cefotiam (surgery for colorectal diseases) 28 per 1000 55 per 1000
(10 to 294)
RR 1.97
(0.37 to 10.43)
143
(1 study)
⊕⊝⊝⊝
very low 1,2
Levofloxacin versus ofloxacin (breast cancer)No infection in either groupNot estimable181
(1 study)
⊕⊝⊝⊝
very low 1,2
One day of piperacillin, cefazolin, cefmetazole, or cefotiam versus three days of piperacillin, cefazolin, cefmetazole, or cefotiam (elective colorectal surgery)No infection in either groupNot estimable521
(1 study)
⊕⊝⊝⊝
very low 1,2
One dose of cefotiam or cefmetazol versus four doses of cefotiam or cefmetazol (elective surgery for colon cancer) 22 per 1000 22 per 1000
(5 to 107)
RR 1.02
(0.21 to 4.98)
275
(1 study)
⊕⊝⊝⊝
very low 1,2
Pefloxacin versus cefazolin and oxacillin (tibial fracture requiring external fixation) 3 per 1000 13 per 1000
(1 to 113)
RR 3.8
(0.43 to 33.78)
616
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin and cefazolin versus cefazolin (open fractures) Moderate RR 3
(0.13 to 71.78)
92
(1 study)
⊕⊝⊝⊝
very low 1,2
5 per 1000 15 per 1000
(1 to 359)
Vancomycin and cefazolin versus cefazolin (vascular surgery) 32 per 1000 71 per 1000
(14 to 375)
RR 2.21
(0.42 to 11.63)
118
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin and cefazolin versus daptomycin and cefazolin (vascular surgery) Moderate RR 8.21
(0.45 to 148.84)
107
(1 study)
⊕⊝⊝⊝
very low 1,2
5 per 1000 41 per 1000
(2 to 744)
Vancomycin versus teicoplanin (paediatric cardiac surgery)No infection in either groupNot estimable22
(1 study)
⊕⊝⊝⊝
very low 1,2
Vancomycin versus cefuroxime (coronary artery bypass graft without valvular disease) 5 per 1000 5 per 1000
(1 to 32)
RR 1.01
(0.14 to 7.13)
884
(1 study)
⊕⊝⊝⊝
very low 1,2
*The basis for the assumed risk is the control group risk in the study. When there were no events in either group, we have indicated so. When there were events in the intervention group but not in the control group, we have used a moderate proportion of 0.5% in the control group. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence:
High quality: further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: we are very uncertain about the estimate

Summary of findings 4 Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: overall MRSA infections

Summary of findings 4. Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: overall MRSA infections
  1. 1 The risk of bias in the trial was high.
    2 The confidence intervals overlapped 1 and/or 0.75 and 1.25. There were fewer than 300 events in total in the intervention and control groups.

Overall MRSA infections
Patient or population: surgical patients
Settings: secondary
Intervention: comparison of different regimens of prophylactic antibiotic regimens
ComparisonsIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Assumed riskCorresponding risk
Control Comparison of different regimens of prophylactic antibiotic regimens
One dose of cefotiam or cefmetazol versus four doses of cefotiam or cefmetazol (elective surgery for colon cancer) 29 per 1000 22 per 1000
(5 to 97)
RR 0.77
(0.17 to 3.36)
275
(1 study)
⊕⊝⊝⊝
very low 1,2
Kanamycin, erythromycin and cefotiam versus cefotiam (surgery for colorectal diseases) 56 per 1000 111 per 1000
(35 to 353)
RR 1.97
(0.62 to 6.26)
143
(1 study)
⊕⊝⊝⊝
very low 1,2
*The basis for the assumed risk is the control group risk in the study. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence:
High quality: further research is very unlikely to change our confidence in the estimate of effect
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate
Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate
Very low quality: we are very uncertain about the estimate

Background

Description of the condition

Methicillin-resistant Staphylococcus aureus (MRSA) was first discovered in 1961 (Barber 1961; Jevons 1961; Knox 1961), and outbreaks of infection have been reported since the 1970s (Klimek 1976; O'Toole 1970). MRSA infection is associated with significant mortality and morbidity. In the European Union member states plus Norway and Iceland, MRSA infections cause an estimated one million extra hospital stays and cost an estimated EUR 600 million (ECDC 2009a). In the USA, an estimated 125,000 hospitalisations occur each year in relation to MRSA infections (Kuehnert 2005). While there has been a decrease in the incidence of MRSA in some countries such as the USA (Kallen 2010), probably because of measures to combat MRSA infections (ECDC 2009b), there has been an increase in the incidence of MRSA infections in Nordic countries (Skov 2005). Methicillin (meticillin is the International Nonproprietary Name but we have used 'methicillin' in this review as it is the more commonly used name in our experience) resistance is a marker of resistance to some beta-lactam antibiotics (penicillin and the cephalosporin group of antibiotics, which are some of the most commonly used antibiotics in the patients) (Otter 2011). In addition to beta-lactam antibiotics, MRSA may be resistant to many other commonly-used antibiotics such as erythromycin, clindamycin, gentamycin, ciprofloxacin, and fusidic acid (Otter 2011). So, even though the antibiotic methicillin is not commonly used itself, methicillin resistance indicates resistance to a wide range of antibiotics. There are currently concerns that farm animals may become reservoirs of MRSA, and a source of a major epidemic of MRSA outside hospitals (Wulf 2008).

The incidence of MRSA infection after surgery is usually low, but can be up to 33% in certain types of surgery, such as pancreatic surgery (pancreatoduodenectomies) (Sanjay 2010). Post-operative MRSA infection can present as surgical site infections (SSI), chest infections, or bloodstream infections (bacteraemia) (Fraser 2010; Reddy 2007; Sanjay 2010). Nosocomial (hospital-acquired) MRSA transmission is believed to be due to cross-contamination from healthcare workers whose hands become colonised transiently while performing patient care activities on people colonised or infected with MRSA (Boyce 1994). Healthcare workers who are persistent nasal carriers of MRSA (i.e. who carry MRSA without suffering from infection) may also act as the source of infection for MRSA (Boyce 1994). Generally, air-borne transmission is not considered to be a mode of transmission for MRSA. Large burn wounds in people colonised or infected with MRSA may act as reservoirs of MRSA (Boyce 1994).

The Centers for Disease Control and Prevention (CDC) criteria for SSIs, published by Horan et al (Horan 1992), provide definitions of nosocomial SSIs. Interested readers may refer to this document for definitions, but broadly speaking, infections confined to skin and subcutaneous tissues are superficial SSIs; those involving the fascia and muscles are deep SSIs; and those involving organs or spaces other than the incision are organ space SSIs (Horan 1992). The incidence of MRSA SSIs in developed countries varies from 1% to 25% depending upon type of surgery and the carrier status of the individual (i.e. whether MRSA colonisation was present prior to the surgery) (Harbarth 2008a; Reddy 2007; Ridgeway 2005; Sanjay 2010; Shukla 2009). The role of universal MRSA screening and contact precautions in hospitalised patients is controversial. Guidelines published by the Society for Healthcare Epidemiology of America (SHEA) recommend routine screening and contact precautions (Muto 2003); others suggest that MRSA screening targeted at people at high risk of MRSA colonisation, such as those requiring intensive care, people with chronic wounds, and nursing home residents is more cost-effective than universal MRSA screening (Creamer 2011; Kang 2012). Others suggest that the contact precautions are not necessary for decreasing post-operative MRSA infection rate, provided that the people are isolated after screening (Spence 2011); yet others suggest that MRSA screening is not effective in decreasing post-operative MRSA SSIs (Harbarth 2008b). The risk factors, other than the type of surgery and carrier status of the individual, include emergency surgery, prolonged duration of surgery, contaminated surgery, immunosuppression, and the presence of co-morbidities such as diabetes mellitus, renal insufficiency, and ischaemic disease (Harbarth 2008a). MRSA SSIs are associated with increased mortality in people undergoing cardiac operations such as coronary artery bypass graft surgery and cardiac valve surgery (Reddy 2007). Cardiac surgical patients with MRSA SSIs had an in-hospital mortality of 12.9% compared with an in-hospital mortality of 3% in uninfected cardiac surgery patients (Reddy 2007). In people undergoing vascular surgery, such as aortic aneurysm repair, carotid endarterectomy, and vascular bypass procedures, the presence of post-operative MRSA infection resulted in a four-fold increase in the in-hospital mortality (Cowie 2005). Patients who developed MRSA infections also stayed longer in hospital than those who did not develop MRSA infections (Chemaly 2010; Cowie 2005; Fraser 2010; Sanjay 2010; Shukla 2009).

The incidence of post-operative MRSA infection can vary with the type of surgery; such infection is usually rare, but post-operative chest infections can be found in up to 15% of pancreatic surgeries (Sanjay 2010), and can result in bacteraemia in up to 5% of the pancreatic surgical patients (Sanjay 2010). MRSA bacteraemia is associated with a 30-day mortality of about 28% to 38% (Lamagni 2011; Lewis 2011; Wang 2010), and a one-year mortality of about 55% (Kaye 2008).

Description of the intervention

The Oxford English Dictionary defines an antibiotic substance as, "One of a class of substances produced by living organisms and capable of destroying or inhibiting the growth of micro-organisms especially used for therapeutic purposes. Synthetic organic compounds having similar properties are also called antibiotics" (OED 2011). Prophylactic antibiotics (to prevent infection) require administration before the infection sets in. A variety of antibiotics, including beta-lactams (penicillin derivatives, cephalosporins),glycopeptide antibiotics (e.g. vancomycin, teicoplanin), clindamycin, trimethoprim-sulfamethoxazole (TMP-SMX), a tetracycline (doxycycline or minocycline), linezolid, daptomycin, telavancin, rifampicin, gentamycin, and fluoroquinolone, all work against MRSA (Liu 2011). Different antibiotics are administered in different ways, with the common routes being oral, intravenous, and topical (surface) administration (Liu 2011). Antibiotics can be given as a single agent or in combinations (Liu 2011). Antibiotics can be given pre-operatively, during the operation, post-operatively, or a combination of the above. Prophylactic antibiotics are usually started just before surgery, and can vary between a single dose or multiple doses for a short period of time after the surgery (Saginur 2000). An antibiotic regimen can be considered as a particular method of administration of antibiotic that includes route of administration, dose administered, and whether single or multiple antibiotics; it can be administered pre-operatively, during the operation, post-operatively, or a combination of the above. Some national guidelines recommend a single prophylactic dose prior to start of surgery except under exceptional circumstances, such as prolonged surgery (more than four hours), major blood loss (in excess of 1.5 litres in adults or 25 ml/kg in children), or in specific types of surgery such as arthroplasty (SIGN 2008).

How the intervention might work

The mechanisms of action vary for different types of antibiotic, but in general terms they either destroy MRSA or prevent cell division (which prevents reproduction and hence multiplication in numbers of MRSA bacteria). The intervention might decrease the complications caused by MRSA infection by preventing MRSA infection. As described above, complications related to MRSA infection include mortality and serious adverse events such as bacteraemia, and chest infection; such outcomes are important in terms of clinical decision making (Fraser 2010; Reddy 2007; Sanjay 2010). Incidence of MRSA infection is considered to be an intermediate outcome that might influence mortality, and has serious adverse events.

Why it is important to do this review

One systematic review found that glycopeptide antibiotics (such as vancomycin and teicoplanin) did not influence post-operative surgical site infection rates compared with non-glycopeptide antibiotics (Chambers 2010). This is just of one of the comparisons included in this review. There has been no systematic review comparing other antibiotics. Loeb et al compared the use of different anti-microbial treatments in people colonised with MRSA either nasally or at extra-nasal sites and concluded that there was insufficient evidence to support use of topical or systemic (whole body) antimicrobial therapy for eradicating nasal or extra-nasal MRSA (Loeb 2003).There has been no systematic review or Cochrane review comparing the various antibiotics, other than the comparison between vancomycin and teicoplanin by Chambers 2010, in the prevention of MRSA infection and subsequent complications in people undergoing surgery. Such a review will be a useful guide for microbiologists, surgeons, and policy makers.

Objectives

To compare the benefits and harms of all methods of antibiotic prophylaxis in the prevention of MRSA infection and related post-operative complications in people undergoing surgery.

Methods

Criteria for considering studies for this review

Types of studies

We included randomised controlled trials (RCTs), irrespective of blinding, language, publication status, date of publication, study setting, sample size, or whether the incidence of MRSA infection was the primary outcome of the trial. We planned to include cluster randomised clinical trials if the effect estimate was available after adjusting for clustering effect. No other study designs (i.e. quasi-randomised studies and non-randomised studies) were included.

Types of participants

People undergoing surgery, irrespective of age, type of surgery, whether surgery was elective or emergency, and whether MRSA colonisation was identified by routine screening (in general, we expected that if MRSA colonisation was identified, it would have been eradicated before surgery). We excluded studies recruiting people with established MRSA SSIs, as these are covered in another review (Gurusamy).

Types of interventions

  1. Comparison of antibiotic prophylaxis (irrespective of the antibiotic) compared with placebo (or no treatment).

  2. Comparison of different antibiotic prophylaxis (and regimens). This includes different doses, routes, or timings of administration.

We included studies evaluating a combination of antibiotics in terms of the combined regimen rather than as single antibiotics.

Types of outcome measures

Primary outcomes
  1. All-cause mortality at maximal follow-up.

  2. Other serious adverse events (defined as any event that is life-threatening; requires inpatient hospitalisation; results in a persistent or significant disability; or any important medical event that might have jeopardised the patient or requires intervention to prevent it (ICH-GCP 1996), e.g. rates of bacteraemia, and other MRSA complications) at maximal follow-up due to surgery or MRSA infection or due to the use of antibiotics. Mild adverse events are unlikely to determine the clinical management, if there were no significant differences in the primary outcomes or one of the secondary outcomes.

  3. Quality of life (at maximal follow-up).

Secondary outcomes
  1. Total length of hospital stay (at maximal follow-up due to surgery or MRSA infection).

  2. Use of health care resources (e.g. hospital visits at maximal follow-up due to surgery or MRSA infection).

  3. Rates of SSIs (due to all organisms) within 30 days of surgery.

  4. Rates of SSIs due to MRSA within 30 days of surgery.

  5. Rates of infections due to MRSA at any site within 30 days of surgery.

We included only trials in which MRSA SSIs were reported.

Search methods for identification of studies

Electronic searches

In March 2013 we searched the following electronic databases to identify reports of relevant RCTs: 

  • The Cochrane Wounds Group Specialised Register (searched 28 February 2013);

  • The Cochrane Central Register of Controlled Trials (The Cochrane Library CENTRAL) (2013, Issue 1);

  • Database of Abstracts of Reviews of Effects (DARE) ) (The Cochrane Library 2013, Issue 1);

  • NHS Economic Evaluation Database ) (The Cochrane Library 2013, Issue 1);

  • Health Technology Assessment (HTA) Database ) (The Cochrane Library 2013, Issue 1);

  • Ovid MEDLINE (1948 to February Week 3 2013);

  • Ovid MEDLINE (In-Process & Other Non-Indexed Citations, 28 February 2013);

  • Ovid EMBASE (1974 to 2013 Week 08);

  • EBSCO CINAHL (1982 to 21 February 2013).

We used the following search strategy in The Cochrane Central Register of Controlled Trials (CENTRAL):

#1 MeSH descriptor Methicillin Resistance explode all trees
#2 MeSH descriptor Staphylococcal Infections explode all trees
#3 MeSH descriptor Staphylococcus aureus explode all trees
#4 (#2 OR #3)
#5 (#1 AND #4)
#6 MeSH descriptor Methicillin-Resistant Staphylococcus aureus explode all trees
#7 (methicillin NEXT resistan*) or (meticillin NEXT resistan*) or MRSA
#8 (#5 OR #6 OR #7)
#9 MeSH descriptor Wound Infection explode all trees
#10 MeSH descriptor Sepsis explode all trees
#11 MeSH descriptor Soft Tissue Infections explode all trees
#12 MeSH descriptor Surgical Wound Dehiscence explode all trees
#13 surg* NEAR/5 infect*
#14 surg* NEAR/5 wound*
#15 surg* NEAR/5 site*
#16 surg* NEAR/5 incision*
#17 surg* NEAR/5 dehisc*
#18 wound* NEAR/5 dehisc*
#19 (deep NEXT infection*) or "deep sepsis" or (infected NEXT
collection*)
#20 (#9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR
#18 OR #19)
#21 MeSH descriptor Anti-Bacterial Agents explode all trees
#22 MeSH descriptor Cephalosporins explode all trees
#23 MeSH descriptor Tetracycline explode all trees
#24 MeSH descriptor Penicillins explode all trees
#25 antibiotic* or penicillin* or beta-lactam* or cephalosporin* or clindamycin or trimethoprim* or tetracycline* or doxycycline or minocycline or linezolid or vancomycin or daptomycin or telavancin or rifampicin or gentamycin or gentamicin or fluoroquinolone
#26 (#21 OR #22 OR #23 OR #24 OR #25)
#27 (#8 AND #20 AND #26)

The search strategies for Ovid MEDLINE, Ovid EMBASE and EBSCO CINAHL can be found in Appendix 1. We combined the Ovid MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE: sensitivity- and precision-maximising version (2008 revision) (Lefebvre 2011). We combined the EMBASE search with the Ovid EMBASE filter developed by the UK Cochrane Centre (Lefebvre 2011). We combined the CINAHL searches with the trial filters developed by the Scottish Intercollegiate Guidelines Network (SIGN 2013). We did not restrict studies with respect to language, date of publication or study setting.

We searched the metaRegister of Controlled Trials (mRCT) (http://www.controlled-trials.com/mrct/), which includes the ISRCTN Register and the NIH ClinicalTrials.gov Register among others. We also searched the World Health Organization's International Clinical Trials Registry Platform (ICTRP) (http://apps.who.int/trialsearch/). The ICRTP portal includes national trial registry databases from a number of countries (last searched 11 December 2012).

Searching other resources

We searched the references of the included trials to identify further relevant trials. We also contacted experts in MRSA infection to identify further trials.

Data collection and analysis

We performed the systematic review following the instructions in the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011a).

Selection of studies

Two review authors (KG and RK) identified the trials for inclusion independently. We have listed the excluded studies with reasons for their exclusion. Any differences were resolved through discussion.

Data extraction and management

Both review authors extracted the following data independently:

  1. Year and language of publication of trial report.

  2. Country.

  3. Year in which trial was conducted.

  4. Inclusion and exclusion criteria.

  5. Sample size.

  6. Type of surgery.

  7. Details of antibiotic treatment including dose, route, frequency, and duration.

  8. Outcomes (described above).

  9. Risk of bias (described below).

  10. Source of funding.

When multiple reports had been published for a trial, we obtained information from all the reports. We contacted the authors of the individual trials to seek out any unclear or missing information. If there was any doubt about whether the trials shared the same participants, completely or partially (by identifying common authors and centres), we contacted the study authors to clarify whether the trial report had been duplicated. We resolved any differences in opinion through discussion.

Assessment of risk of bias in included studies

We followed the instructions in the Cochrane Handbook for Systematic Reviews of Intervention to assess risk of bias (Higgins 2011b). According to empirical evidence (Kjaergard 2001; Moher 1998; Schulz 1995; Wood 2008), the risk of bias of the trials was assessed according to the following bias risk domains:

Sequence generation
  1. Low risk of bias (the methods used are either adequate (e.g. computer-generated random numbers, table of random numbers) or unlikely to introduce confounding).

  2. Uncertain risk of bias (there is insufficient information to assess whether the method used is likely to introduce confounding).

  3. High risk of bias (the method used (e.g. quasi-randomised studies) is improper and likely to introduce confounding). Such studies will be excluded.

Allocation concealment
  1. Low risk of bias (the method used (e.g. central allocation) is unlikely to induce bias on the final observed effect).

  2. Uncertain risk of bias (there is insufficient information to assess whether the method used is likely to induce bias on the estimate of effect).

  3. High risk of bias (the method used (e.g. open random allocation schedule) is likely to induce bias on the final observed effect).

Blinding of participants, personnel
  1. Low risk of bias (blinding was performed adequately, or the outcome measurement is not likely to be influenced by lack of blinding).

  2. Uncertain risk of bias (there is insufficient information to assess whether the type of blinding used is likely to induce bias on the estimate of effect).

  3. High risk of bias (no blinding or incomplete blinding, and the outcome or the outcome measurement is likely to be influenced by lack of blinding).

Blinding of outcome assessors
  1. Low risk of bias (blinding was performed adequately, or the outcome measurement is not likely to be influenced by lack of blinding).

  2. Uncertain risk of bias (there is insufficient information to assess whether the type of blinding used is likely to induce bias on the estimate of effect).

  3. High risk of bias (no blinding or incomplete blinding, and the outcome or the outcome measurement is likely to be influenced by lack of blinding).

Incomplete outcome data
  1. Low risk of bias (the underlying reasons for missing data are unlikely to make treatment effects depart from plausible values, or proper methods have been employed to handle missing data).

  2. Uncertain risk of bias (there is insufficient information to assess whether the missing data mechanism in combination with the method used to handle missing data is likely to induce bias on the estimate of effect).

  3. High risk of bias (the crude estimate of effects (e.g. complete case estimate) will clearly be biased due to the underlying reasons for the data being missing, and the methods used to handle missing data are unsatisfactory).

Selective outcome reporting
  1. Low risk of bias (the trial protocol is available and all of the trial's pre-specified outcomes that are of interest in the review have been reported or similarly, if the trial protocol is not available, all the primary outcomes in this review are reported).

  2. Uncertain risk of bias (there is insufficient information to assess whether the magnitude and direction of the observed effect is related to selective outcome reporting).

  3. High risk of bias (not all of the trial's pre-specified primary outcomes have been reported).

We considered trials that were classified as having a low risk of bias in all the above domains to be low bias-risk trials.

Measures of treatment effect

For dichotomous variables, we calculated the risk ratio (RR) with 95% confidence interval (CI). We used Peto's odds ratios (Peto OR) for outcomes with a proportion of less than 1% events. For continuous variables, we planned to calculate the mean difference (MD) with 95% CI for outcomes such as hospital stay and standardised mean difference (SMD) with 95% CI for quality of life (where different scales might be used). For time-to-event outcomes such as survival at maximal follow-up, we planned to calculate the hazard ratio (HR) with 95% CI.

Unit of analysis issues

The unit of analysis was individual people undergoing surgical procedures. We did anticipate the need for people to have a second operation during the same admission within the trials. However, we decided that if we found any person having a second operation, provided that s/he did not have MRSA infection before the second surgery, we would consider each surgery to be a separate unit of analysis.

Dealing with missing data

We performed an intention-to-treat analysis whenever possible (Newell 1992). We imputed missing data for binary outcomes using various scenarios such as best-best scenario, worst-worst scenario, best-worst scenario, and worst-best scenario (Gurusamy 2009). In the best-best scenario, the people with the missing outcomes were considered not to have developed a complication. In the worst-worst scenario, the people with the missing outcomes were considered to have developed a complication. In the best-worst scenario, the people with the missing outcomes were considered not to have developed a complication in the intervention group and to have developed a complication in the control group. In the worst-best scenario, the people with the missing outcomes were considered to have developed a complication in the intervention group and not to have developed a complication in the control group.

For continuous outcomes, we planned to use available-case analysis. We planned to impute the standard deviation from P values according to the instructions in the Cochrane Handbook for Systematic Reviews of Intervention (Higgins 2011c), and use the median for the meta-analysis when the mean was not available. If it was not possible to calculate the standard deviation from the P value or the CIs, we planned to impute the standard deviation as the highest standard deviation in the other trials included under that outcome, fully recognising that this form of imputation would decrease the weight of the study for calculation of MD and bias the effect estimate to no effect in case of SMD (Higgins 2011c).

For time-to-event outcomes, where HR and 95% CIs were not reported, we planned to obtain the logarithm of hazard ratios (ln(HR)) and the standard error (SE) of ln(HR) according to the methods described by Parmar 1998.

Assessment of heterogeneity

We explored heterogeneity by means of the Chi2 test with significance set at P value 0.10, and measured the degree of heterogeneity by means of the I2 test (Higgins 2002). We also used overlapping of CIs on the forest plot to determine heterogeneity.

Assessment of reporting biases

We planned to use visual asymmetry on a funnel plot to explore reporting bias in the presence of at least 10 trials (Egger 1997; Macaskill 2001). We planned to perform the linear regression approach described by Egger 1997 to determine the funnel plot asymmetry.

Data synthesis

For the comparison of antibiotic versus placebo (or no intervention), we planned to perform the meta-analysis only if there was sufficient clinical homogeneity in terms of people included in the trials. This was based on our clinical judgement. For the comparison of different antibiotics, we planned to perform the meta-analysis only if there was sufficient clinical homogeneity in terms of people included in the trials and in terms of antibiotics used (i.e. similar class of antibiotics). We performed the meta-analyses using the software package RevMan 5 (RevMan 2011), following the recommendations of The Cochrane Collaboration (Higgins 2011a). We used both a random-effects model (DerSimonian 1986), and a fixed-effect model (DeMets 1987), for the meta-analyses. In case of discrepancy between the two models identified from the pooled estimates and their CIs, we have reported both results; otherwise we have reported the results of the fixed-effect model. Calculations for dichotomous outcomes to produce risk ratio (RR) or Peto OR do not include trials in which no events occurred in either group in the meta-analysis, whereas risk difference (RD) calculations do. We planned to report the RD when the results using this association measure were different from RR or Peto OR. However, RR or Peto OR were the measures that we used to arrive at conclusions, since they perform better than RD when there are differences in the control event rate (proportion of people who develop the event in the control). We planned to use the generic inverse variance method to combine the HRs for time-to-event outcomes.

Subgroup analysis and investigation of heterogeneity

We planned to perform the following subgroup analyses.

  1. Different antibiotics (or class of antibiotics).

  2. Different doses and durations of antibiotics.

  3. Different types of surgery including different types of wounds (clean, clean-contaminated, contaminated, and dirty or infected) (the risk of SSI varies according to wound types) (Garner 1986).

  4. Different patient characteristics (presence of systemic illness such as diabetes or other immunocompromised individuals).

  5. People routinely screened and treated versus those who were not routinely screened.

We planned to use a P value of less than 0.05 for the Chi2 test to identify the differences between subgroups and to investigate whether heterogeneity in effect estimates were attributable to differences in the above characteristics.

Sensitivity analysis

We performed a sensitivity analysis by imputing data for binary outcomes using various scenarios such as best-best scenario, worst-worst scenario, best-worst scenario, and worst-best scenario (Gurusamy 2009). We also planned to perform a sensitivity analysis by testing the effect of removing trials at unclear or high risk of bias and excluding the trials in which the mean and the standard deviation were imputed.

Presentation of results

We have presented the main results of the review in 'Summary of findings' tables, which provide key information concerning the quality of evidence, the magnitude of effect of the interventions examined, and the sum of available data on the main outcomes, as recommended by the Cochrane Collaboration (Schunemann 2011a). We planned to include the following in the 'Summary of findings' tables:

  1. All-cause mortality.

  2. Serious adverse events.

  3. Quality of life.

  4. Total length of hospital stay.

  5. Use of healthcare resources.

  6. Rates of surgical site infections due to MRSA within 30 days of surgery.

  7. Rates of surgical site infections due to all organisms within 30 days of surgery.

  8. Rates of overall MRSA infections.

We have included 'Summary of findings' tables for the above outcomes when they were available from the trials. Each 'Summary of findings' table includes an overall grading of the evidence related to each of the main outcomes, using the GRADE approach (Schunemann 2011b).

Results

Description of studies

Results of the search

We identified a total of 489 unique references through electronic searches. We excluded 454 clearly irrelevant references through reading titles and abstracts. We retrieved 35 references in full for further assessment. We did not identify any additional references to trials by scanning reference lists of included trials. We did not identify any new trials through other searches. We excluded 22 references because of the reasons mentioned in the Characteristics of excluded studies. Thirteen references to 12 trials met the inclusion criteria and were included in this review (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Ishibashi 2009; Ishida 2001; Kaiser 1987; Morimoto 2002; Saadeddin 2005; Saveli 2011; Shime 2007; Stone 2010; Vuorisalo 1998). The reference flow is shown in Figure 1.

Figure 1.

Study flow diagram.

Included studies

A total of 5284 people were randomised to different antibiotic regimens in the 12 trials (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Ishibashi 2009; Ishida 2001; Kaiser 1987; Morimoto 2002; Saadeddin 2005; Saveli 2011; Shime 2007; Stone 2010; Vuorisalo 1998). A total of 580 people were excluded from the analysis in eight trials (Hashizume 2004; Ishida 2001; Kaiser 1987; Morimoto 2002; Saadeddin 2005; Shime 2007; Stone 2010; Vuorisalo 1998). The proportion of drop-outs in these trials ranged from 2% to 33%. There were no post-randomisation drop-outs in one trial (Ishibashi 2009). Information on post-randomisation drop-outs was not available in three trials (Carsenti-Etesse 1999; Hashizume 2004; Saveli 2011). A total of 4704 people were included under the various comparisons. The average age of participants ranged between 54 years and 71 years in the five trials that reported this information (Ishibashi 2009; Ishida 2001; Morimoto 2002; Saadeddin 2005; Stone 2010). Four trials reported adult age as an inclusion criteria (Carsenti-Etesse 1999; Goldstein 2009; Saadeddin 2005; Stone 2010). One trial stated children as an inclusion criterion (Shime 2007). From the age range of the included participants, it was possible to identify that three trials included adults only (Ishibashi 2009; Ishida 2001; Morimoto 2002). The information on the age of those included was not reported in the remaining trials (Hashizume 2004; Kaiser 1987; Saveli 2011; Vuorisalo 1998). The proportion of females ranged between 20.7% and 100% in the six trials that reported this information (Ishibashi 2009; Ishida 2001; Morimoto 2002; Saadeddin 2005; Stone 2010; Vuorisalo 1998).

Four trials involved elective colorectal surgeries (Goldstein 2009; Hashizume 2004; Ishibashi 2009; Ishida 2001), one trial percutaneous endoscopic gastrostomy (Saadeddin 2005), one trial surgery for breast cancer (Morimoto 2002), two trials cardiac surgery (Shime 2007; Vuorisalo 1998), one trial surgery requiring sternotomy (Kaiser 1987), and two trials involved open fracture fixation (Carsenti-Etesse 1999; Saveli 2011). In one trial of cardiac surgery, all the participants were either positive for nasal carriage of MRSA or had previous MRSA infections (Shime 2007). However, it is not clear whether MRSA carrier status was treated prior to surgery. One trial clearly stated that the baseline MRSA screening status was not available for participants (Saadeddin 2005). One trial stated that routine screening was used, but, considering that the participants had open fractures, it is highly unlikely that MRSA was eradicated prior to surgery (Saveli 2011). Information about whether people were routinely screened and treated prior to operation was not reported in the remaining trials (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Ishibashi 2009; Ishida 2001; Kaiser 1987; Morimoto 2002; Stone 2010; Vuorisalo 1998). From the inclusion criteria in the trials, it appears that all the trials included at least some people without systemic illness such as diabetes.

Nine trials directly compared two prophylactic antibiotic regimens (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Ishibashi 2009; Ishida 2001; Morimoto 2002; Saveli 2011; Shime 2007; Vuorisalo 1998). Of the remaining three trials, one compared three prophylactic antibiotic regimens (Stone 2010), one compared four prophylactic antibiotic regimens (Kaiser 1987), and one compared a prophylactic antibiotic regimen with placebo (Saadeddin 2005). A total of 17 comparisons were possible from the data in the 12 included trials. The individual characteristics in trials including inclusion and exclusion criteria, post-randomisation drop-outs (and reasons for post-randomisation drop-outs), demographic characteristics, antibiotic regimens used, number of people included for analysis for the different antibiotic regimens and the outcomes reported are summarised in Characteristics of included studies.

Excluded studies

A total of 22 studies were excluded for the reasons stated in Characteristics of excluded studies. Three reports were not RCTs (Al-Mukhtar 2009; Diehr 2007; Kato 2006); five studies were quasi-randomised studies (Finkelstein 2002; Kanellakopoulou 2009; Salminen 1999; Tacconelli 2008; Tyllianakis 2010); 12 studies did not report the methicillin resistance status of the cultured organisms (Bluhm 1984; Dhadwal 2007; Frimodt-Moller 1982; Keighley 1979; Maki 1992; Marroni 1999; Mindermann 1993; Palmer 1995; Ramos 2008; Saginur 2000; Weaver 1986; Wilson 1988); and two studies did not report the methicillin resistance status for all staphylococcal infections (i.e. the methicillin resistance was stated for staphylococcal infections at some sites, but not at others) (Cann 1988; Poon 1998).

Risk of bias in included studies

All the trials were judged to be at high risk of bias as shown in Figure 2 and Figure 3.

Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies

Figure 3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study

Allocation

Four trials reported acceptable methods of random sequence generation and allocation concealment (Ishibashi 2009; Ishida 2001; Saadeddin 2005; Stone 2010). These four trials were also free from selection bias. One other trial reported an acceptable method of random sequence generation, but did not report on allocation concealment (Vuorisalo 1998). The remaining seven trials did not report on random sequence generation or allocation concealment (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Kaiser 1987; Morimoto 2002; Saveli 2011; Shime 2007). Eight trials were at unclear risk of selection bias (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Kaiser 1987; Morimoto 2002; Saveli 2011; Shime 2007; Vuorisalo 1998).

Blinding

Adequate blinding of participants, personnel and outcome assessors was reported in one trial (Saadeddin 2005); this trial was free from performance bias and detection bias (Saadeddin 2005). Two trials reported adequate blinding of outcome assessors (Ishibashi 2009; Ishida 2001), but had inadequate blinding of participants or personnel; these trials were free from detection bias but not performance bias. In one trial, blinding of participants, personnel or outcome assessors was not performed (Kaiser 1987); this trial was of high risk of performance bias and detection bias. Information on blinding was not available in the remaining eight trials (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Morimoto 2002; Saveli 2011; Shime 2007; Stone 2010; Vuorisalo 1998); these trials were at unclear risk of performance bias and detection bias.

Incomplete outcome data

One trial had no post-randomisation drop-outs and was free from attrition bias (Ishibashi 2009). In one trial, the imputation of missing outcome data under different scenarios did not influence the results and so the impact of post-randomisation drop-outs in this trial was minimal as shown in the section 'Sensitivity analysis' (Shime 2007). This trial was also considered to be low risk of attrition bias. There was unclear or high risk of attrition bias in the remaining trials (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Ishida 2001; Kaiser 1987; Morimoto 2002; Saadeddin 2005; Saveli 2011; Stone 2010; Vuorisalo 1998).

Selective reporting

Important outcomes such as mortality, antibiotic-related serious adverse events, overall SSIs and MRSA related SSIs were reported in three trials (Ishibashi 2009; Ishida 2001; Saadeddin 2005). These three trials were considered to be at low risk of reporting bias, while the remaining trials were considered to be at high risk of reporting bias.

Other potential sources of bias

There were no other potential sources of bias in the remaining trials.

Effects of interventions

See: Summary of findings for the main comparison Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: mortality; Summary of findings 2 Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: surgical site infection; Summary of findings 3 Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: MRSA surgical site infection; Summary of findings 4 Antibiotic prophylaxis for the prevention of methicillin-resistant Staphylococcus aureus (MRSA) infections and related complications in surgical patients: overall MRSA infections

There was not sufficient clinical homogeneity amongst the included studies to perform a meta-analysis. The effects of interventions are summarised in Summary of findings for the main comparison; Summary of findings 2; Summary of findings 3; and Summary of findings 4.

All-cause mortality at maximal follow-up

Four trials reported mortality (Ishibashi 2009; Ishida 2001; Saadeddin 2005; Vuorisalo 1998), with participants followed up until: discharge (Ishida 2001); seven days after the procedure (Saadeddin 2005); one month after the procedure (Ishibashi 2009); and one month after discharge (Vuorisalo 1998). One trial compared a prophylactic antibiotic regimen versus placebo for control (Saadeddin 2005), while the remaining trials compared different antibiotic regimens (Ishibashi 2009; Ishida 2001; Vuorisalo 1998). The overall mortality in these four trials was 14/1401 (1.0%) (Ishibashi 2009; Ishida 2001; Saadeddin 2005; Vuorisalo 1998). There were no significant differences in mortality between the intervention and control groups in any of the four comparisons (Analysis 1.1). Using Peto OR as the effect measure (as the proportion of people was fractionally less than 1.0%) did not alter the results.

Other serious adverse events

None of the trials reported on overall serious adverse events. Four trials reported serious adverse events related to antibiotics (Ishibashi 2009; Ishida 2001; Saadeddin 2005; Saveli 2011). One of these trials compared an antibiotic regimen with placebo as control (Saadeddin 2005), while the remaining trials compared different antibiotic regimens directly (Ishibashi 2009; Ishida 2001; Saveli 2011). There were no antibiotic-related serious adverse events in any of the 561 people randomised to the seven different antibiotic regimens in these four trials (Ishibashi 2009; Ishida 2001; Saadeddin 2005; Saveli 2011).

Quality of life

None of the trials reported quality of life.

Total length of hospital stay

None of the trials reported the total length of hospital stay.

Use of healthcare resources

None of the trials reported the use of healthcare resources.

Surgical site infections due to all organisms

Eleven trials reported SSIs in the different groups (Carsenti-Etesse 1999; Hashizume 2004; Ishibashi 2009; Ishida 2001; Kaiser 1987; Morimoto 2002; Saadeddin 2005; Saveli 2011; Shime 2007; Stone 2010; Vuorisalo 1998). One trial compared an antibiotic regimen with placebo as control (Saadeddin 2005), while the remaining trials directly compared different antibiotic regimens. Overall, 221/4032 (5.5%) people developed SSIs due to all organisms. There were no significant differences in the proportion of people who developed SSIs between any of the 15 comparisons involving direct comparison of one antibiotic regimen with another (Analysis 1.2). In the comparison involving antibiotic regimen with placebo, the proportion of people who developed SSIs was statistically significantly lower in people who received the antibiotic prophylaxis with co-amoxiclav or cefotaxime compared with placebo (RR 0.26; 95% CI 0.11 to 0.65; Analysis 1.2) (Saadeddin 2005).

MRSA surgical site infections

All 12 included trials reported SSIs due to MRSA in the different groups. One trial compared an antibiotic regimen with placebo as control (Saadeddin 2005), while the remaining trials compared different antibiotic regimens. Overall, 46/4704 (1.0%) people developed SSIs due to MRSA. There were no significant differences in the proportion of people who developed SSIs due to MRSA between any of the 16 comparisons involving comparison of one antibiotic regimen with another (Analysis 1.3). In the comparison involving prophylactic antibiotic regimen with placebo, the proportion of people who developed SSIs due to MRSA was statistically significantly lower in people who received the antibiotic prophylaxis with co-amoxiclav or cefotaxime compared with placebo (RR 0.05; 95% CI 0.00 to 0.83). Using Peto OR as the effect measure (as the proportion of people was fractionally less than 1.0%) did not alter the results.

MRSA infections at any site

Two trials reported overall MRSA infections in the different groups (Ishibashi 2009; Ishida 2001); both trials compared different antibiotic regimens. Overall, 19/478 (4.5%) people developed MRSA infections including SSI, chest infection and bacteraemia. There were no significant differences in the proportion of people who developed MRSA infections at any location in these two comparisons (Analysis 1.4).

Statistical variations

The issue of fixed-effect versus random-effects meta-analysis did not arise because only one trial was included for each comparison. As meta-analysis was not performed, we did not calculate the risk difference in order to assess the impact of the trials with zero events on the meta-analysis results.

Subgroup analysis

We did not perform any subgroup analysis, as only one trial was included for each outcome.

Sensitivity analysis

One trial had no post-randomisation drop-out (Ishibashi 2009). In three trials, it was not possible to determine whether there were any post-randomisation drop-outs (Carsenti-Etesse 1999; Hashizume 2004; Saveli 2011). In four trials, although there were post-randomisation drop-outs, the groups to which the drop-outs belonged were not reported (Goldstein 2009; Kaiser 1987; Morimoto 2002; Saadeddin 2005). Sensitivity analysis using imputation of missing outcome data under different scenarios was either not relevant, or could not be performed in these eight trials. Sensitivity analysis using imputation of missing outcome data under different scenarios was performed for six comparisons reported in four trials (Ishida 2001; Shime 2007; Stone 2010; Vuorisalo 1998). In one trial that reported SSI and MRSA SSI (Shime 2007), there was no change in the results when results were imputed under different scenarios (Analysis 2.4; Analysis 2.10). In another trial that reported mortality, SSI and MRSA SSI (Vuorisalo 1998), all the outcomes were sensitive to imputation under different scenarios (Analysis 2.2; Analysis 2.8; Analysis 2.14). In the remaining two trials (Ishida 2001; Stone 2010), some of the outcomes were sensitive to imputation under different scenarios (Analysis 2.3; Analysis 2.5; Analysis 2.7), while others were not sensitive (Analysis 2.1; Analysis 2.6; Analysis 2.9; Analysis 2.11; Analysis 2.12; Analysis 2.13; Analysis 2.15).

Reporting bias

Since only one trial was included for each comparison, we did not explore reporting bias using a funnel plot.

Discussion

Summary of main results

In this review, we compared the efficacy of different antibiotic prophylaxis regimens including no antibiotic prophylaxis in preventing MRSA and related postoperative complications in people undergoing surgery. A total of 16 comparisons of different antibiotic regimens were performed in 11 trials (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Ishibashi 2009; Ishida 2001; Kaiser 1987; Morimoto 2002; Saveli 2011; Shime 2007; Stone 2010; Vuorisalo 1998). Antibiotic prophylaxis was compared with no antibiotic prophylaxis in one trial (Saadeddin 2005). A total of 4704 people undergoing various types of surgery were included under the different comparisons in this review. Except for one trial where all the participants were either positive for nasal carriage of MRSA or had previous MRSA infections (no information provided on whether MRSA carriers were treated prior to surgery) (Shime 2007) , it does not appear that MRSA was eradicated prior to surgery; nor does it appear that there was high prevalence of MRSA carrier status in the people undergoing surgery. The overall short-term mortality after surgery in four trials that reported mortality was 1.0% (14/1401) (Ishibashi 2009; Ishida 2001; Saadeddin 2005; Vuorisalo 1998). Overall, 221/4032 (5.5%) people developed SSIs due to all organisms (Carsenti-Etesse 1999; Hashizume 2004; Ishibashi 2009; Ishida 2001; Kaiser 1987; Morimoto 2002; Saadeddin 2005; Saveli 2011; Shime 2007; Stone 2010; Vuorisalo 1998), and 46/4704 (1.0%) people developed SSIs due to MRSA (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Ishibashi 2009; Ishida 2001; Kaiser 1987; Morimoto 2002; Saadeddin 2005; Saveli 2011; Shime 2007; Stone 2010; Vuorisalo 1998).

One trial including 99 participants compared prophylactic co-amoxiclav with placebo in people undergoing percutaneous endoscopic gastrostomy (Saadeddin 2005). In this trial, the proportion of people who developed SSIs due to all organisms and those developed SSIs due to MRSA was lower in the antibiotic prophylaxis group compared with placebo (Saadeddin 2005). Although co-amoxiclav has significant in-vitro activity against some strains of MRSA (Alou 2004; Cantoni 1989; Prieto 1998), there is no evidence for significant clinical activity of co-amoxiclav and it is not one of the antibiotics recommended for MRSA infections (Liu 2011). The probable way that co-amoxiclav decreased MRSA infections was by decreasing the overall infection level, thereby preventing the wounds from becoming infected secondarily.

There were no significant differences in short-term mortality, SSIs, MRSA SSIs or overall MRSA infections in any of these comparisons. These comparisons included additional antibiotics or increased duration of antibiotics. There is currently no evidence that any of the important clinical outcomes are affected by the addition of antibiotics or increased duration of antibiotics. Although, there were no serious adverse events in any of the 561 people randomised to the seven different antibiotic regimens in four trials (Ishibashi 2009; Ishida 2001; Saadeddin 2005; Saveli 2011) (this information was not reported in the remaining eight trials (Carsenti-Etesse 1999; Goldstein 2009; Hashizume 2004; Kaiser 1987; Morimoto 2002; Shime 2007; Stone 2010; Vuorisalo 1998)), there are various concerns about increasing the duration or the number of antibiotics in the antibiotic combination. The obvious concern is the cost of the antibiotics. This is important in state-funded, insurance-funded, and private-funded healthcare systems, as unnecessary use of antibiotics increases the cost of procedures (besides the cost of the antibiotic per se, one should also consider the costs for administering the antibiotics). At the individual level, antibiotic-related adverse events may delay recovery after surgery. While adverse events such as nausea or other gastrointestinal upsets may not be serious adverse events, and are unlikely to be major determinants of the decision about the antibiotic regimens, these mild adverse events may prevent the participant from returning to normal activities, and hence may delay recovery. Although most adverse events related to antibiotics can be managed easily, some adverse events such as severe anaphylactic shock or antibiotic-related infections (by decreasing the gut flora - for example antibiotic-induced Clostridium difficile infection can cause pseudomembranous colitis) can be life-threatening (Schroeder 2005). At the community level, inappropriate and unnecessary use of antibiotics can lead to the development of drug resistance (Sieradzki 1999; Chua 2008), which could be problematic for future populations.

Overall completeness and applicability of evidence

A range of surgeries ranging from moderate to major surgeries were included in this review. So, this review is applicable to most types of surgery. This review is applicable only to people who do not have existing MRSA infections. The people included in this review did not appear to be particularly immunocompromised, so the results of this review are not applicable for immunocompromised people.

Quality of the evidence

The overall quality of the evidence is either low, or very low, as shown in Summary of findings for the main comparison; Summary of findings 2; Summary of findings 3; and Summary of findings 4. The risk of bias was high in all the trials. By using and reporting an appropriate method of randomisation, selection bias could be minimised. In most instances where different antibiotic regimens are compared, it is possible to blind the participants, personnel and outcome assessors by the use of an appropriate placebo. This minimises the performance and detection bias. By performing an intention-to-treat analysis, it is possible to decrease attrition bias. By reporting the important clinical outcomes that are likely to be measured routinely, such as mortality, morbidity, antibiotic-related adverse events, and length of hospital stay, selective outcome reporting could be minimised. Measuring the quality of life and health resource utilisation measures such as hospital or community nurse visits, wound dressings, etc would allow an health economic analysis to determine the cost-effectiveness of using different antibiotics. Apart from the risk of bias, the other quality issue was that the effect measures were imprecise, i.e. the confidence intervals were wide and one could not rule out a clinically significant benefit or harm related to different antibiotic regimens. This issue can be addressed by appropriate sample size calculations prior to conducting a trial. In spite of all the shortcomings, this is the best available evidence on this topic.

Potential biases in the review process

We have performed a thorough search of the literature without any restrictions regarding language or date of publication. In spite of this, we were not be able to identify any trials that were conducted, but not reported, in the pre-mandatory trial registration era. We included only trials in which MRSA infection was mentioned. It is highly likely that several other trials on antibiotic regimens have been conducted without assessing or reporting the methicillin resistance status of the organisms. The nature of this topic means that one has to be pragmatic and accept that such trials are unlikely to be identified electronically, and that contacting the authors of every single trial that has assessed prophylactic antibiotics to obtain information regarding whether they measured the MRSA status of the organisms, and to obtain information in a such way that it could be used for this systematic review would be resource intensive. The reliability of information obtained in this manner can also be a major concern.

Agreements and disagreements with other studies or reviews

There have been no previous systematic reviews on this topic.

Authors' conclusions

Implications for practice

Prophylaxis with co-amoxiclav decreases the proportion of people developing MRSA infections compared with placebo in people without malignant disease undergoing percutaneous endoscopic gastrostomy insertion, although this may be due to decreasing overall infection thereby preventing wounds from becoming secondarily infected with MRSA. There is currently no other evidence to suggest that using a combination of multiple prophylactic antibiotics or administering prophylactic antibiotics for an increased duration is of benefit in people undergoing surgery. However, the confidence intervals were wide.

Implications for research

Well designed randomised controlled trials are necessary on this topic. Such studies should use patient-oriented outcomes such as mortality, serious adverse events, quality of life, total length of hospital stay (due to surgery or MRSA infection), use of health care resources (e.g. hospital visits due to MRSA infection), and MRSA infections at sites other than the surgical site.

Acknowledgements

The authors would like to acknowledge the contribution of the peer referees, Wounds Group Editors (Julie Bruce) and referees (Caroline Main; Richard Kirubakaran; Jo Sutton) and copy editors Elizabeth Royle and Heather Maxwell.

Data and analyses

Download statistical data

Comparison 1. Comparison of different regimens of prophylactic antibiotics
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Mortality4 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
1.1 One dose of cefotiam or cefmetazol versus four doses of cefotiam or cefmetazol1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.2 Kanamycin, erythromycin and cefotiam versus cefotiam1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.3 Co-amoxiclav or cefotaxime versus placebo1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.4 Vancomycin versus cefuroxime1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2 Overall surgical site infection11 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
2.1 Cefamendole versus cefamendole and gentamycin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.2 Cefazolin and gentamycin versus cefamendole and gentamycin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.3 Cefazolin versus cefamendole1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.4 Cefazolin versus cefazolin and gentamycin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.5 Co-amoxiclav or cefotaxime versus placebo1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.6 Daptomycin and cefazolin versus cefazolin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.7 Kanamycin, erythromycin and cefotiam versus cefotiam1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.8 Levofloxacin versus ofloxacin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.9 One day of piperacillin, cefazolin, cefmetazole, or cefotiam versus three days of piperacillin, cefazolin, cefmetazole, or cefotiam1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.10 One dose of cefotiam or cefmetazol versus four doses of cefotiam or cefmetazol1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.11 Pefloxacin versus cefazolin and oxacillin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.12 Vancomycin and cefazolin versus cefazolin (open fractures)1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.13 Vancomycin and cefazolin versus cefazolin (vascular surgery)1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.14 Vancomycin and cefazolin versus daptomycin and cefazolin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.15 Vancomycin versus teicoplanin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.16 Vancomycin versus cefuroxime1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3 MRSA surgical site infection12 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
3.1 Cefamendole versus cefamendole and gentamycin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.2 Cefazolin and gentamycin versus cefamendole and gentamycin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.3 Cefazolin versus cefamendole1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.4 Cefazolin versus cefazolin and gentamycin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.5 Co-amoxiclav or cefotaxime versus placebo1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.6 Daptomycin and cefazolin versus cefazolin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.7 Ertapenem versus cefotetan1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.8 Kanamycin, erythromycin and cefotiam versus cefotiam1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.9 Levofloxacin versus ofloxacin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.10 One day of piperacillin, cefazolin, cefmetazole, or cefotiam versus three days of piperacillin, cefazolin, cefmetazole, or cefotiam1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.11 One dose of cefotiam or cefmetazol versus four doses of cefotiam or cefmetazol1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.12 Pefloxacin versus cefazolin and oxacillin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.13 Vancomycin and cefazolin versus cefazolin (open fractures)1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.14 Vancomycin and cefazolin versus cefazolin (vascular surgery)1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.15 Vancomycin and cefazolin versus daptomycin and cefazolin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.16 Vancomycin versus teicoplanin1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.17 Vancomycin versus cefuroxime1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4 Overall MRSA infections2 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
4.1 One dose of cefotiam or cefmetazol versus four doses of cefotiam or cefmetazol1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4.2 Kanamycin, erythromycin and cefotiam versus cefotiam1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 1.1.

Comparison 1 Comparison of different regimens of prophylactic antibiotics, Outcome 1 Mortality.

Analysis 1.2.

Comparison 1 Comparison of different regimens of prophylactic antibiotics, Outcome 2 Overall surgical site infection.

Analysis 1.3.

Comparison 1 Comparison of different regimens of prophylactic antibiotics, Outcome 3 MRSA surgical site infection.

Analysis 1.4.

Comparison 1 Comparison of different regimens of prophylactic antibiotics, Outcome 4 Overall MRSA infections.

Comparison 2. Sensitivity analysis
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Mortality (kanamycin, erythromycin and cefotiam versus cefotiam)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
1.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2 Mortality (vancomycin versus cefuroxime)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
2.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
2.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3 Surgical site infection (kanamycin, erythromycin and cefotiam versus cefotiam)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
3.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
3.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4 Surgical site infection (vancomycin versus teicoplanin)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
4.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
4.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
5 Surgical site infection (vancomycin and cefazolin versus daptomycin and cefazolin)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
5.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
5.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
5.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
5.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
6 Surgical site infection (vancomycin and cefazolin versus cefazolin)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
6.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
6.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
6.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
6.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
7 Surgical site infection (daptomycin and cefazolin versus cefazolin)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
7.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
7.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
7.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
7.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
8 Surgical site infection (vancomycin versus cefuroxime)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
8.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
8.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
8.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
8.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
9 MRSA surgical site infection (kanamycin, erythromycin and cefotiam versus cefotiam)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
9.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
9.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
9.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
9.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
10 MRSA surgical site infection (vancomycin versus teicoplanin)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
10.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
10.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
10.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
10.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
11 MRSA surgical site infection (vancomycin and cefazolin versus daptomycin and cefazolin)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
11.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
11.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
11.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
11.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
12 MRSA surgical site infection (vancomycin and cefazolin versus cefazolin)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
12.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
12.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
12.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
12.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
13 MRSA surgical site infection (daptomycin and cefazolin versus cefazolin)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
13.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
13.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
13.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
13.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
14 MRSA surgical site infection (vancomycin versus cefuroxime)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
14.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
14.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
14.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
14.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
15 Overall MRSA infections (kanamycin, erythromycin and cefotiam versus cefotiam)1 Risk Ratio (M-H, Fixed, 95% CI)Totals not selected
15.1 Best-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
15.2 Worst-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
15.3 Best-worst scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
15.4 Worst-best scenario1 Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
Analysis 2.1.

Comparison 2 Sensitivity analysis, Outcome 1 Mortality (kanamycin, erythromycin and cefotiam versus cefotiam).

Analysis 2.2.

Comparison 2 Sensitivity analysis, Outcome 2 Mortality (vancomycin versus cefuroxime).

Analysis 2.3.

Comparison 2 Sensitivity analysis, Outcome 3 Surgical site infection (kanamycin, erythromycin and cefotiam versus cefotiam).

Analysis 2.4.

Comparison 2 Sensitivity analysis, Outcome 4 Surgical site infection (vancomycin versus teicoplanin).

Analysis 2.5.

Comparison 2 Sensitivity analysis, Outcome 5 Surgical site infection (vancomycin and cefazolin versus daptomycin and cefazolin).

Analysis 2.6.

Comparison 2 Sensitivity analysis, Outcome 6 Surgical site infection (vancomycin and cefazolin versus cefazolin).

Analysis 2.7.

Comparison 2 Sensitivity analysis, Outcome 7 Surgical site infection (daptomycin and cefazolin versus cefazolin).

Analysis 2.8.

Comparison 2 Sensitivity analysis, Outcome 8 Surgical site infection (vancomycin versus cefuroxime).

Analysis 2.9.

Comparison 2 Sensitivity analysis, Outcome 9 MRSA surgical site infection (kanamycin, erythromycin and cefotiam versus cefotiam).

Analysis 2.10.

Comparison 2 Sensitivity analysis, Outcome 10 MRSA surgical site infection (vancomycin versus teicoplanin).

Analysis 2.11.

Comparison 2 Sensitivity analysis, Outcome 11 MRSA surgical site infection (vancomycin and cefazolin versus daptomycin and cefazolin).

Analysis 2.12.

Comparison 2 Sensitivity analysis, Outcome 12 MRSA surgical site infection (vancomycin and cefazolin versus cefazolin).

Analysis 2.13.

Comparison 2 Sensitivity analysis, Outcome 13 MRSA surgical site infection (daptomycin and cefazolin versus cefazolin).

Analysis 2.14.

Comparison 2 Sensitivity analysis, Outcome 14 MRSA surgical site infection (vancomycin versus cefuroxime).

Analysis 2.15.

Comparison 2 Sensitivity analysis, Outcome 15 Overall MRSA infections (kanamycin, erythromycin and cefotiam versus cefotiam).

Appendices

Appendix 1. Search strategies for Ovid MEDLINE, Ovid EMBASE and EBSCO CINAHL

Ovid MEDLINE

1 Methicillin Resistance/ (8026)
2 exp Staphylococcal Infections/ (23428)
3 exp Staphylococcus aureus/ (30286)
4 2 or 3 (39399)
5 1 and 4 (7604)
6 exp Methicillin-Resistant Staphylococcus aureus/ (5190)
7 (methicillin-resistan* or meticillin-resistan* or MRSA).tw. (15592)
8 or/5-7 (17196)
9 exp Wound Infection/ (14971)
10 exp Sepsis/ (51757)
11 exp Soft Tissue Infections/ (1929)
12 exp Surgical Wound Dehiscence/ (2990)
13 (surg* adj5 infect*).tw. (11123)
14 (surg* adj5 wound*).tw. (5735)
15 (surg* adj5 site*).tw. (8208)
16 (surg* adj5 incision*).tw. (4303)
17 (surg* adj5 dehisc*).tw. (395)
18 (wound* adj5 dehisc*).tw. (1856)
19 (deep infection* or deep sepsis or infected collection).tw. (1695)
20 or/9-19 (91225)
21 exp Anti-Bacterial Agents/ (215537)
22 exp Penicillins/ (16714)
23 exp Cephalosporins/ (12581)
24 exp Tetracycline/ (4066)
25 (antibiotic* or penicillin* or beta-lactam* or cephalosporin* or clindamycin or trimethoprim* or tetracycline* or doxycycline or minocycline or linezolid or vancomycin or daptomycin or telavancin or rifampicin or gentamycin or gentamicin or fluoroquinolone).tw. (156631)
26 or/21-25 (283068)
27 8 and 20 and 26 (1883)
28 randomized controlled trial.pt. (243797)
29 controlled clinical trial.pt. (39777)
30 randomized.ab. (198478)
31 placebo.ab. (92392)
32 clinical trials as topic.sh. (80116)
33 randomly.ab. (136484)
34 trial.ti. (73737)
35 or/28-34 (550390)
36 (animals not (humans and animals)).sh. (1629396)
37 35 not 36 (500869)
38 27 and 37 (117)

Ovid EMBASE

1 exp penicillin resistance/ (6486)
2 exp Staphylococcus infection/ (20926)
3 exp Staphylococcus aureus/ (71187)
4 2 or 3 (79541)
5 1 and 4 (3433)
6 exp methicillin resistant Staphylococcus aureus/ (24269)
7 (methicillin-resistan* or meticillin-resistan* or MRSA).tw. (23553)
8 or/5-7 (31101)
9 exp wound infection/ (19415)
10 exp sepsis/ (108581)
11 exp soft tissue infection/ (5213)
12 exp wound dehiscence/ (6928)
13 (surg* adj5 infect*).tw. (16679)
14 (surg* adj5 wound*).tw. (8175)
15 (surg* adj5 site*).tw. (12421)
16 (surg* adj5 incision*).tw. (6567)
17 (surg* adj5 dehisc*).tw. (551)
18 (wound* adj5 dehisc*).tw. (2660)
19 (deep infection* or deep sepsis or infected collection).tw. (2255)
20 or/9-19 (168503)
21 exp antibiotic agent/ (544500)
22 exp penicillin derivative/ (117164)
23 exp cephalosporin derivative/ (92762)
24 exp tetracycline derivative/ (65922)
25 (antibiotic* or penicillin* or beta-lactam* or cephalosporin* or clindamycin or trimethoprim* or tetracycline* or doxycycline or minocycline or linezolid or vancomycin or daptomycin or telavancin or rifampicin or gentamycin or gentamicin or fluoroquinolone).tw. (232206)
26 or/21-25 (615812)
27 8 and 20 and 26 (5216)
28 Randomized controlled trials/ (26835)
29 Single-Blind Method/ (15611)
30 Double-Blind Method/ (86092)
31 Crossover Procedure/ (31945)
32 (random$ or factorial$ or crossover$ or cross over$ or cross-over$ or placebo$ or assign$ or allocat$ or volunteer$).ti,ab. (944695)
33 (doubl$ adj blind$).ti,ab. (90521)
34 (singl$ adj blind$).ti,ab. (9694)
35 or/28-34 (979655)
36 animal/ (725595)
37 human/ (8657232)
38 36 not 37 (485192)
39 35 not 38 (947254)
40 27 and 39 (354)

EBSCO CINAHL

S28 S8 and S20 and S27
S27 S21 or S22 or S23 or S24 or S25 or S26
S26 AB (antibiotic* or penicillin* or beta-lactam* or cephalosporin* or clindamycin or trimethoprim* or tetracycline* or doxycycline or minocycline or linezolid or vancomycin or daptomycin or telavancin or rifampicin or gentamycin or gentamicin or fluoroquinolone)
S25 TI (antibiotic* or penicillin* or beta-lactam* or cephalosporin* or clindamycin or trimethoprim* or tetracycline* or doxycycline or minocycline or linezolid or vancomycin or daptomycin or telavancin or rifampicin or gentamycin or gentamicin or fluoroquinolone)
S24 (MH "Tetracyclines+")
S23 (MH "Cephalosporins+")
S22 (MH "Penicillins+")
S21 (MH "Antibiotics+")
S20 S9 or S10 or S11 or S12 or S13 or S14 or S15 or S16 or S17 or S18 or S19
S19 TI deep infection* or deep sepsis or infected collection or AB deep infection* or deep sepsis or infected collection
S18 TI wound* N5 dehisc* or AB wound* N5 dehisc*
S17 TI surg* N5 dehisc* or AB surg* N5 dehisc*
S16 TI surg* N5 incision* or AB surg* N5 incision*
S15 TI surg* N5 site* or AB surg* N5 site*
S14 TI surg* N5 wound* or AB surg* N5 wound*
S13 TI surg* N5 infection* or AB surg* N5 infection*
S12 (MH "Surgical Wound Dehiscence")
S11 (MH "Soft Tissue Infections")
S10 (MH "Sepsis+")
S9 (MH "Wound Infection+")
S8 S5 or S6 or S7
S7 TI ( methicillin-resistan* or meticillin-resistan* or MRSA ) OR AB ( methicillin-resistan* or meticillin-resistan* or MRSA )
S6 (MH "Methicillin-Resistant Staphylococcus Aureus")
S5 S1 and S4
S4 S2 or S3
S3 (MH "Staphylococcus Aureus+")
S2 (MH "Staphylococcal Infections+")
S1 (MH "Methicillin Resistance")

Contributions of authors

KS Gurusamy: wrote the review, assessed the trials for inclusion and extracted data on included trials.
R Koti: independently assessed the trials for inclusion and extracted data on included trials.
P Wilson and BR Davidson: commented critically on the review and approved the review.

Contributions of editorial base

Nicky Cullum: edited the protocol; advised on methodology, interpretation and review content. Approved the review for submission.
Susan O'Meara: Editor: advised on methodology, interpretation and content.
Sally Bell-Syer: co-ordinated the editorial process. Advised on methodology, interpretation and content. Edited and copy edited the review.
Ruth Foxlee: designed the search strategy and edited the Search methods section.

Declarations of interest

KS Gurusamy and BR Davidson are funded by a joint funding scheme between Department of Health and Wellcome Trust on a completely unrelated project.
R Koti: none known.
APR Wilson is a consultant microbiologist in the NHS advising on antibiotic use and advises some private hospitals on infection control. He is a member of a clinical trial drug safety monitoring board for a monoclonal antibody. He has been an expert witness in infection-related cases. He has a number of non-commercial grants for research in the area of transmission of infection. APR Wilson was part-funded by the UCLH/UCL Comprehensive Biomedical Centre with funding from the Department of Health's NIHR Biomedical Research Centres.

This project was funded by the National Institute for Health Research (NIHR).

Disclaimer

Department of Health disclaimer: The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS (National Health Service), or the Department of Health.

Sources of support

Internal sources

  • No sources of support supplied

External sources

  • National Institute for Health Research, UK.

    National Institute for Health Research, the health research wing of the UK Government Department of Health funds K Gurusamy to complete this review.

  • NIHR/Department of Health (England), (Cochrane Wounds Group), UK.

Differences between protocol and review

Peto's odds ratios were used for outcomes with a proportion of less than 1% events.
MRSA overall infections was included as a secondary outcome, since the antibiotics act on MRSA at any site rather than on the surgical wounds alone.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Carsenti-Etesse 1999

MethodsRCT
ParticipantsCountry: France
Number randomised: 616
Post-randomisation drop-outs: not stated
Revised sample size: 616
Average age: not stated
Male:female ratio: not stated
Inclusion criteria:
1. adults over 15 years of age with an open extra-articular fracture of the tibia requiring reduction fixation
2. wound closure that could be performed directly or with a muscle or skin plasty (grades 1 and 2)
3. delay between fracture and surgery < 12 hours
InterventionsParticipants randomly assigned to 2 groups:
Group 1: antibiotic 1 (n = 316), a single 800 mg dose of pefloxacin iv at anaesthetic induction
Group 2: antibiotic 2 (n = 300), 2-day course of cefazolin iv (4 x 1 g doses/day; total dose over 2 days = 8 g) from anaesthetic induction, followed by a 3-day course of oral oxacillin (3 x 1 g doses/day)
OutcomesSSI and MRSA SSI
Notes

We attempted to contact the authors in September 2012

Source of funding: not reported

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskComment: this information was not available
Allocation concealment (selection bias)Unclear riskComment: this information was not available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskQuote: "In a randomised double-blind trial…"
Comment: further details of how the blinding was performed were not mentioned
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskQuote: "In a randomised double-blind trial…"
Comment: further details of how the blinding was performed were not mentioned
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskComment: this information was not available
Selective reporting (reporting bias)High riskComment: some important outcomes, that are generally assessed, were not reported

Goldstein 2009

MethodsRCT
ParticipantsCountry: USA
Number randomised: 1002
Post-randomisation drop-outs: 330 (32.9%)
Revised sample size: 672
Average age: not stated
Male:female ratio: not stated
Inclusion criteria: people at least 18 years of age having elective colorectal surgery
InterventionsParticipants randomly assigned to 2 groups
Group 1: antibiotic 1 (n = 338), a single dose of ertapenem (1 g) over 30 min within 60 min prior to the initial incision
Group 2: antibiotic 2 (n = 334), a single dose of cefotetan (2 g) over 30 min within 60 min prior to the initial incision
OutcomesMRSA SSI
Notes

We attempted to contact the authors in September 2012
Reasons for post-randomisation drop-outs: not stated

Source of funding: not reported, however, some of the authors worked or belonged to the advisory board of a pharmaceutical company

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskComment: this information was not available
Allocation concealment (selection bias)Unclear riskComment: this information was not available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskComment: this information was not available
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskComment: this information was not available
Incomplete outcome data (attrition bias)
All outcomes
High riskComment: there were post-randomisation drop-outs
Selective reporting (reporting bias)High riskComment: some important outcomes, that are generally assessed, were not reported

Hashizume 2004

MethodsRCT
ParticipantsCountry: Japan
Number randomised: 521
Post-randomisation drop-outs: not stated
Revised sample size: 521
Average age: not stated
Male:female ratio: not stated
Inclusion criteria: people undergoing elective colorectal surgery
InterventionsParticipants randomly assigned to 2 groups:
Group 1: antibiotic 1 (n = 262), piperacillin, cefazolin, cefmetazole, or cefotiam for 1 day
Group 2: antibiotic 2 (n = 259), piperacillin, cefazolin, cefmetazole, or cefotiam for 3 days
OutcomesSSI and MRSA SSI.
Notes

We attempted to contact the authors in September 2012

Source of funding: not reported

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskComment: this information was not available
Allocation concealment (selection bias)Unclear riskComment: this information was not available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskComment: this information was not available
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskComment: this information was not available
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskComment: this information was not available
Selective reporting (reporting bias)High riskComment: some important outcomes, that are generally assessed, were not reported

Ishibashi 2009

MethodsRCT
Participants

Country: Japan
Number randomised: 275
Post-randomisation drop-outs: 0 (0%)
Revised sample size: 275
Average age: 68 years
Male:female ratio: 161:114 (41.5% female)

Inclusion criteria: elective surgery for colon cancer
Exclusion criteria: people with ileus or other conditions, who could not undergo colon preparation

InterventionsParticipants randomly assigned to 2 groups
Group 1: antibiotic 1 (n = 136), cefotiam or cefmetazol for 1 post-operative dose 1 h after surgery
Group 2: antibiotic 2 (n = 139), cefotiam or cefmetazol for 4 post operative doses over 2 days
OutcomesMortality, antibiotic-related serious adverse events, SSI, MRSA SSI, and overall MRSA infections
Notes

We contacted the authors in September 2012, and they provided prompt replies to our questions

Source of funding: no external funding (authors' replies)

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskQuote: "We prepared the same number of A and B in opaque and sealed envelopes, for example, total 100 opaque and sealed envelopes include 50 A and 50 B, and shuffled. Then the surgeon draws the envelope in turn when informed consent is obtained from patients on admission" (author replies)".
Allocation concealment (selection bias)Low riskQuote: "We prepared the same number of A and B in opaque and sealed envelopes, for example, total 100 opaque and sealed envelopes include 50 A and 50 B, and shuffled. Then the surgeon draws the envelope in turn when informed consent is obtained from patients on admission" (author replies)
Blinding of participants and personnel (performance bias)
All outcomes
High riskComment: the authors replied that the patients and healthcare providers were not blinded
Blinding of outcome assessment (detection bias)
All outcomes
Low riskComment: the authors replied that the outcome assessors were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskComment: there were no post-randomisation drop-outs
Selective reporting (reporting bias)Low riskComment: all important outcomes were reported

Ishida 2001

MethodsRCT
Participants

Country: Japan
Number randomised: 146
Post-randomisation drop-outs: 3 (2.1%)
Revised sample size: 143
Average age: 63 years
Male:female ratio: 89:54 (37.8% female)
Inclusion criteria: people undergoing surgery for colorectal diseases
Exclusion criteria:
1. not feasible to perform full mechanical bowel preparation

2. taken any antibiotics within 14 days before surgery

InterventionsParticipants randomly assigned to 2 groups
Group 1: antibiotic 1 (n = 72), kanamycin 2 g/day and erythromycin 1.6 g/day were given perorally, in 4 doses for 2 days before surgery and 6 doses of cefotiam (1 gm) within 3 days of surgery
Group 2: antibiotic 2 (n = 71), no pre-operative intervention and 6 doses of cefotiam (1 gm) within 3 days of surgery
All participants received mechanical bowel preparation
OutcomesMortality, antibiotic-related serious adverse events, SSI, MRSA SSI, and overall MRSA infections
Notes

We contacted the authors in September 2012, and they replied promptly with answers to our questions
Reasons for post-randomisation drop-outs: protocol violation

Source of funding: no external funding (authors' replies)

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskQuote: "We prepared the same number of A and B in opaque and sealed envelopes, for example, total 100 opaque and sealed envelopes include 50 A and 50 B, and shuffled. Then the surgeon draws the envelope in turn when informed consent is obtained from patients on admission" (authors' replies)
Allocation concealment (selection bias)Low riskQuote: "We prepared the same number of A and B in opaque and sealed envelopes, for example, total 100 opaque and sealed envelopes include 50 A and 50 B, and shuffled. Then the surgeon draws the envelope in turn when informed consent is obtained from patients on admission" (authors' replies)
Blinding of participants and personnel (performance bias)
All outcomes
High riskComment: the authors replied that the patients and healthcare providers were not blinded
Blinding of outcome assessment (detection bias)
All outcomes
Low riskComment: the authors replied that the outcome assessors were blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskComment: there were post-randomisation drop-outs. Imputation of missing outcome data under different scenarios showed change in conclusions
Selective reporting (reporting bias)Low riskComment: all important outcomes were reported

Kaiser 1987

MethodsRCT
ParticipantsCountry: USA
Number randomised: 1057
Post-randomisation drop-outs: 27 (2.6%)
Revised sample size: 1030
Average age: not stated
Male:female ratio: not stated
Inclusion criteria: people having median sternotomy
Interventions

Participants randomly assigned to 4 groups
Group 1: antibiotic 1 (n = 255), cefazolin, 2 g iv at induction of anaesthesia,1 g every 4 h during operation, and 1 g every 6 h after operation for 72 h

Group 2: antibiotic 2 (n = 253), as in group 1 plus gentamicin 1.5 mg/kg given iv at induction of anaesthesia
Group 3: antibiotic 3 (n = 259), cefamandole, 2 g iv at induction of anaesthesia, 1 g every 2 h during operation, and 1 g every 4 h after operation for 72 h
Group 4: antibiotic 4 (n = 263), as in group 3 plus gentamicin 1.5 mg/kg given iv at induction of anaesthesia

OutcomesSSI and MRSA SSI
Notes

We attempted to contact the authors in September 2012
Reasons for post-randomisation drop-outs: operations cancelled (11); failure to adhere to the antibiotic regimen (9); did not undergo median sternotomy (inadvertent entry into study (7)

Source of funding: not reported

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskComment: this information was not available
Allocation concealment (selection bias)Unclear riskComment: this information was not available
Blinding of participants and personnel (performance bias)
All outcomes
High riskQuote: "Placebo doses were not included"
Blinding of outcome assessment (detection bias)
All outcomes
High riskQuote: "Placebo doses were not included"
Incomplete outcome data (attrition bias)
All outcomes
High riskComment: there were post-randomisation drop-outs
Selective reporting (reporting bias)High riskComment: some important outcomes, that are generally assessed, were not reported

Morimoto 2002

MethodsRCT
ParticipantsCountry: Japan
Number randomised: 199
Post-randomisation drop-outs: 18 (9%)
Revised sample size: 181
Average age: 54 years
Male:female ratio: 181 women (100% female)
Inclusion criteria: women undergoing treatment for breast cancer
InterventionsParticipants randomly assigned to 2 groups
Group 1: antibiotic 1 (n = 99), levofloxacin 200 mg daily from day after operation for 5 days
Group 2: antibiotic 2 (n = 82), ofloxacin 200 mg daily from day after operation for 5 days
OutcomesSSI and MRSA SSI
Notes

We attempted to contact the authors in September 2012
Reasons for post-randomisation drop-outs: tumours proved to be benign

Source of funding: not reported

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskComment: this information was not available
Allocation concealment (selection bias)Unclear riskComment: this information was not available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskComment: this information was not available
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskComment: this information was not available
Incomplete outcome data (attrition bias)
All outcomes
High riskComment: there were post-randomisation drop-outs
Selective reporting (reporting bias)High riskComment: some important outcomes, that are generally assessed, were not reported

Saadeddin 2005

MethodsRCT
ParticipantsCountry: UK
Number randomised: 110
Post-randomisation drop-outs: 11 (10%)
Revised sample size: 99
Average age: 71 years
Male:female ratio: 53:46 (46.5% female)
Inclusion criteria: people over 16 years of age accepted for PEG insertion without malignant disease
Exclusion criteria: people who had received antibiotics within the 48 h preceding the PEG insertion
InterventionsParticipants randomly assigned to 2 groups
Group 1: antibiotic 1 (n = 51), a single dose of 2.2 g co-amoxiclav, made up of 1 g amoxicillin and 1.2 g co-amoxiclav (or 2 g cefotaxime if participant was penicillin-allergic)
Group 2: antibiotic 2 (n = 48), placebo injection
OutcomesMortality, antibiotic-related serious adverse events, SSI, and MRSA SSI
Notes

We contacted the authors in September 2012. Authors provided replies regarding serious adverse events related to antibiotic treatment
Reasons for post-randomisation drop-outs: failure to complete PEG insertion after randomisation (7 cases), protocol violation (3 cases) or PEG removal by patient (1 case)

Source of funding: no external funding (authors' replies)

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskQuote: "Randomization was done in advance using a random number generator"
Allocation concealment (selection bias)Low riskQuote: "study assignment cards with medication packs were kept in pharmacy"
Blinding of participants and personnel (performance bias)
All outcomes
Low riskQuote: "The endoscopy nurse prepared the medication and the syringe out of sight of the study investigator, and covered the syringe with an opaque sleeve so that both the study investigator and the patient were ‘blinded’"
Blinding of outcome assessment (detection bias)
All outcomes
Low riskQuote: "The endoscopy nurse prepared the medication and the syringe out of sight of the study investigator, and covered the syringe with an opaque sleeve so that both the study investigator and the patient were ‘blinded’"
Incomplete outcome data (attrition bias)
All outcomes
High riskComment: there were post-randomisation drop-outs
Selective reporting (reporting bias)Low riskComment: all important outcomes were reported

Saveli 2011

MethodsRCT
ParticipantsCountry: USA
Number randomised: 92
Post-randomisation drop-outs: not stated
Revised sample size: 92
Average age: not stated
Male:female ratio: not stated
Inclusion criteria: people with open fractures
InterventionsParticipants randomly assigned to 2 groups
Group 1: antibiotic 1 (n = 46), vancomycin (dose and frequency not stated) and cefazolin (dose and frequency not stated) from the time of presentation at ER to 24 h after surgery
Group 2: antibiotic 2 (n = 46), cefazolin (dose and frequency not stated) from the time of presentation at ER to 24 h after surgery
Both groups received cefazolin antibiotic prophylaxis
OutcomesAntibiotic-related serious adverse events, SSI, and MRSA SSI
Notes

We contacted the authors in September 2012; they replied, but no additional information was obtained

Source of funding: not reported

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskComment: this information was not available
Allocation concealment (selection bias)Unclear riskComment: this information was not available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskComment: this information was not available
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskComment: this information was not available
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskComment: this information was not available
Selective reporting (reporting bias)High riskComment: some important outcomes, that are generally assessed, were not reported

Shime 2007

MethodsRCT
ParticipantsCountry: Japan
Number randomised: 24
Post-randomisation drop-outs: 2 (8.3%)
Revised sample size: 22
Average age: not stated
Male:female ratio: not stated
Inclusion criteria:
1. children under 3 years old admitted for cardiac surgery to the PICU
2. carriage of MRSA in the nasal cavity, preoperatively confirmed by sampling the anterior nare, or a history of MRSA infection
InterventionsParticipants randomly assigned to 2 groups
Group 1: antibiotic 1 (n = 11), vancomycin 15 mg/kg, first dose after the induction of general anaesthesia, within the hour preceding the first surgical incision, and a second, identical dose, added to the priming solution of cardio-pulmonary bypass, based on an approximately doubled plasma volume at the onset of extracorporeal circulation. Postoperatively, the same dose was given upon admission to the PICU, and repeated every 8 h for 48 h
Group 2: antibiotic 2 (n = 11), teicoplanin 8 mg/kg, first dose after the induction of general anaesthesia, within the hour preceding the first surgical incision, and a second, identical dose, added to the priming solution of cardio-pulmonary bypass, based on an approximately doubled plasma volume at the onset of extracorporeal circulation. Postoperatively, the same dose was given upon admission to the PICU, and repeated every 24 h for 48 h
An intranasal ointment of mupirocin was applied three times a day, for 3 days before the operation
OutcomesSSI and MRSA SSI
Notes

We attempted to contact the authors in September 2012
Reasons for post-randomisation drop-outs: death 1 (vancomycin group); postoperative application of extracorporeal cardiopulmonary support (vancomycin group)

Source of funding: not reported

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskComment: this information was not available
Allocation concealment (selection bias)Unclear riskComment: this information was not available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskComment: this information was not available
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskComment: this information was not available
Incomplete outcome data (attrition bias)
All outcomes
Low riskComment: there were post-randomisation drop-outs, however, imputation of missing outcome data under different scenarios did not change the conclusions
Selective reporting (reporting bias)High riskComment: some important outcomes, that are generally assessed, were not reported

Stone 2010

MethodsRCT
ParticipantsCountry: USA
Number randomised: 181
Post-randomisation drop-outs: 12 (6.6%)
Revised sample size: 169
Average age: 65 years
Male:female ratio: 115:54 (32% female)
Inclusion criteria: people 18 years of age and older having vascular surgery and or vascular access (vascular fistula or vascular graft) procedure
Exclusion criteria:
1. allergy to daptomycin or vancomycin
2. chronic wounds
3. prior colonization with MRSA
4. a continuous inpatient stay of > 72 h prior to the surgical procedure
5. an active infection requiring antibiotics preoperatively
6. history of MRSA colonisation or infection
7. HIV
8. admission for > 3 months in an acute care centre or long-term care centre
9. dialysis
InterventionsParticipants randomly assigned to 3 groups.
Group 1: antibiotic 1 (n = 56), vancomycin (0 to 60 minutes prior to incision), weight-based dose with 1 g iv for people < 80 kg, 1.5 g iv for 80-199 kg, and 2 g iv for participants ≥ 200 kg. Participants given another dose in the OR if still there 6 h post-incision and had a GFR of > 60 (n = 56)
Group 2: antibiotic 2 (n = 51), 6 mg/kg iv daptomycin  (30 minutes prior to incision)
Group 3: no additional antibiotic (n = 62)
Further details: all participants received cefazolin (0 to 60 minutes prior to incision), weight-based dose with 1 g iv every 8 h for 24 h if weight < 80 kg, 2 g iv every 8 h for 24 h if weight 80-199 kg, and 3 g iv every 8 h if weight ≥ 200 kg. Participants given another dose in the OR if still there 3 h post-incision and had a GFR of > 60  
OutcomesSSI and MRSA SSI
Notes

We contacted the authors in September 2012: they replied, but no additional information was obtained
Reasons for post-randomisation drop-outs: active infection, dialysis, allergy to penicillin, no surgery

Source of funding: not reported, however, Dr Patrick A Stone, the primary author of this paper, is on the Speakers' Bureau for Cubist Pharmaceuticals

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskQuote: "Randomization was performed by staff from the Center for Health Services and Outcomes Research (CHSOR) using the statistical program SAS"
Allocation concealment (selection bias)Low riskQuote: "Randomization was performed by staff from the Center for Health Services and Outcomes Research (CHSOR) using the statistical program SAS"
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskComment: this information was not available
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskComment: this information was not available
Incomplete outcome data (attrition bias)
All outcomes
High riskComment: there were post-randomisation drop-outs. Imputation of missing outcome data under different scenarios showed change in conclusions
Selective reporting (reporting bias)High riskComment: Some important outcomes, that are generally assessed, were not reported

Vuorisalo 1998

  1. a

    Abbreviations
    < = less than
    > = greater/more than
    ≥ = greater/more than or equal to
    CABG = coronary artery bypass graft
    ER = emergency room
    g = gram
    GFR = glomerular filtration rate
    h = hour
    iv = intravenous
    MRSA = methicillin-resistant Staphylococcus aureus
    n = number of participants
    OR = operating room
    PEG = percutaneous endoscopic gastrostomy
    PICU = paediatric intensive care unit
    RCT = randomised controlled trial
    SSI = surgical site infection

MethodsRCT
ParticipantsCountry: Finland
Number randomised: 1061
Post-randomisation drop-outs: 177 (16.7%)
Revised sample size: 884
Average age: not stated
Male:female ratio: 701:183 (20.7% female)
Inclusion criteria: participants scheduled for CABG without valvular surgery
Exclusion criteria:
1. people with a known allergy to cephalosporins or vancomycin
2. presence of an active infection
3. received prescribed antibiotics within the previous 2 weeks
4. people having a repeat bypass operation
Interventions

Participants randomly assigned to 2 groups
Group 1: antibiotic 1 (n = 440), vancomycin 1 g slow iv infusion over 120 minutes beginning at the induction of anaesthesia plus 1 g after 12 h

Group 2: antibiotic 2 (n = 444), cefuroxime 1.5 g iv over 15 minutes at the induction of anaesthesia plus 2 additional doses of 0.75 g at 8-h intervals

OutcomesMortality, SSI, and MRSA SSI
Notes

We attempted to contact the authors in September 2012
Reasons for post-randomisation drop-outs: participant records missing (3); deviation from antibiotic regimens (78); active infection/antibiotic prescribed within previous 2 weeks (54); procedure other than CABG during the same operation (5); repeat bypass (12)

Source of funding: Satakunta Fund of the Finnish Cultural Foundation

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskQuote: "The patients were assigned by reference to a random number table"
Allocation concealment (selection bias)Unclear riskComment: this information was not available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskComment: this information was not available
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskComment: this information was not available
Incomplete outcome data (attrition bias)
All outcomes
High riskComment: there were post-randomisation drop-outs. Imputation of missing outcome data under different scenarios showed change in conclusions
Selective reporting (reporting bias)High riskComment: some important outcomes, that are generally assessed, were not reported

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    Abbreviation

    RCT = randomised controlled trial

Al-Mukhtar 2009Not an RCT
Bluhm 1984No information on methicillin resistance status of the cultured organisms
Cann 1988Methicillin resistance status of all staphylococcal infections was not reported
Dhadwal 2007No information on methicillin resistance status of the cultured organisms
Diehr 2007Not an RCT
Finkelstein 2002Quasi-randomised study (allocation by hospital number)
Frimodt-Moller 1982No information on methicillin resistance status of the cultured organisms
Kanellakopoulou 2009Quasi-randomised study (treatment chosen by the ward to which the patient was admitted)
Kato 2006Not an RCT
Keighley 1979No information on methicillin resistance status of the cultured organisms
Maki 1992No information on methicillin resistance status of the cultured organisms
Marroni 1999No information on methicillin resistance status of the cultured organisms
Mindermann 1993No information on methicillin resistance status of the cultured organisms
Palmer 1995No information on methicillin resistance status of the cultured organisms
Poon 1998Methicillin resistance status of all staphylococcal infections was not reported
Ramos 2008No information on methicillin resistance status of the cultured organisms
Saginur 2000No information on methicillin resistance status of the cultured organisms
Salminen 1999Quasi-randomised study (allocation by birth year)
Tacconelli 2008Quasi-randomised study (allocation according to the first letter of the patient's name)
Tyllianakis 2010Quasi-randomised study (allocation according to the room number to which the patient was admitted)
Weaver 1986No information on methicillin resistance status of the cultured organisms
Wilson 1988No information on methicillin resistance status of the cultured organisms

Ancillary