Rationale and impact
These sections briefly explain why the committee made the recommendations and how they might affect practice. They link to details of the evidence and a full description of the committee's discussion.
Evidence was identified on the use of mupirocin alone and mupirocin in combination with a chlorhexidine body wash. Mupirocin alone was effective in reducing Staphylococcus aureus infections caught in hospital in people who were identified as carriers of S. aureus. However, mupirocin alone did not reduce surgical site infections across all people undergoing surgery.
The evidence also showed that people identified as carriers of S. aureus who used nasal mupirocin in combination with a chlorhexidine body wash before surgery had fewer surgical site infections caused by S. aureus (including deep infections, methicillin-sensitive infections and infections caught in hospital) than those who did not have the intervention. However, the evidence was very limited and only covered S. aureus carriers.
Economic studies favoured the use of mupirocin alone. However, the studies were not UK-based and could not be applied to NHS practice (for example, because of the high cost of treating surgical site infections in US studies). An economic model based on UK data demonstrated that, compared with no treatment, using mupirocin with a chlorhexidine body wash before all operations was an efficient use of resources in most specialist surgeries. However, there was less certainty of cost effectiveness for surgery with a low risk of surgical site infections caused by S. aureus.
Because of the limited evidence, the committee were unable to make strong recommendations on nasal decolonisation before surgery and agreed that it should not be offered to all people having surgery. The committee applied their clinical understanding and experience of current best practice, and recommended that nasal mupirocin with chlorhexidine body wash should be considered before procedures that have an increased risk of surgical site infection caused by S. aureus, for which there would be the most benefit.
The recommendation does not define the surgical procedures in which S. aureus is a likely cause of a surgical site infection. The committee agreed that although cardiac and orthopaedic surgery can be considered high risk, decisions should be made locally through discussions between surgical and infection control teams, and should also take into account patient risk factors, such as whether the person is an S. aureus carrier and the potential impact of infection on the person, including the cost of managing the infection. The recommendation does not give an optimal timing for nasal decolonisation because of a lack of evidence. But the committee were aware that mupirocin with chlorhexidine can be given from 2 days before surgery to 3 days after surgery.
The committee also took into consideration the potential side effects of mupirocin (such as a burning sensation and local reactions) and the cautions identified for the use of chlorhexidine solution in people with existing skin conditions and in preterm newborn babies. The committee noted that the Medicines and Healthcare products Regulatory Agency (MHRA) has published MHRA advice on the use of chlorhexidine for skin disinfection in premature babies.
There was also a lack of evidence on antimicrobial resistance associated with the use of mupirocin and chlorhexidine body wash. The committee agreed that it would be helpful to encourage service providers to maintain surveillance on antimicrobial resistance associated with the use of mupirocin. This would allow any increase in resistance to be captured.
The committee developed a recommendation for research on the effectiveness of nasal mupirocin with chlorhexidine body wash across all surgical procedures to help determine whether this should be extended to all people having surgery. Antimicrobial resistance associated with the use of chlorhexidine body wash was also identified by the committee as an important area for research, so the committee made a recommendation for research on nasal decolonisation and antimicrobial resistance.
There is considerable variability in practice. In some services decolonisation is always offered before certain types of surgery, for example, before orthopaedic surgery. In other services decolonisation is offered only to people who are identified as methicillin-resistant S. aureus (MRSA) or methicillin-sensitive S. aureus (MSSA) carriers.
The new recommendation reflects best practice and allows services the flexibility to consider decolonisation for people who are likely to benefit the most. The recommendation may reduce surgical site infections in people undergoing surgical procedures for which the consequences of an infection are severe, such as cardiac surgery. The evidence suggests that any additional costs incurred in providing decontamination are likely to be more than recouped by savings associated with a lower incidence of surgical site infections. However, the committee acknowledged that there may be training implications for those implementing the recommendation.
Maintenance of surveillance systems assessing antimicrobial resistance associated with the use of mupirocin will reinforce good practice.
Based on their knowledge and experience, the committee agreed that an antiseptic should be used for skin preparation before surgery. Overall, the evidence showed that chlorhexidine in alcohol was associated with the lowest incidence of surgical site infections, whereas aqueous povidone-iodine was associated with the highest incidence. An economic analysis also showed that chlorhexidine in alcohol is likely to be cost effective. Based on the evidence, the committee agreed that an alcohol-based solution of chlorhexidine should usually be the first choice when deciding which antiseptic preparation to use. However, the quality of the studies was not good enough for the committee to make a strong recommendation for the choice of antiseptic preparation.
The committee discussed that alcohol-based solutions should not be applied to mucous membranes because of the risk of burns. For surgical procedures next to mucous membranes, they agreed to recommend an aqueous solution of chlorhexidine as an option for skin preparation. Because of the limited evidence, the committee were unable to make a strong recommendation.
There was little evidence to support the use of povidone-iodine, but based on their clinical experience the committee agreed that it should be an option when chlorhexidine is contraindicated, for example, in people with hypersensitivity to chlorhexidine.
There was no evidence on the use of skin antiseptics in babies. However, the committee were aware of risks, such as burns, associated with their use in this population, and wished to highlight this. The committee noted that the MHRA has published MHRA advice on the use of chlorhexidine for skin disinfection in premature babies.
The committee also discussed that some operative procedures may require diathermy. This means that precautions must be taken when using alcohol-based antiseptic solutions because they are flammable and can result in burns. Along with using evaporation to dry antiseptic skin preparations and avoiding pooling, the committee also agreed that soaked materials, drapes or gowns should be removed before diathermy, excessive quantities of alcohol antiseptics should not be used and no excess product should be present before applying an occlusive dressing.
The committee agreed that further research is needed to establish the effectiveness of different concentrations of chlorhexidine in reducing the risk of surgical site infections. Therefore, the committee made a recommendation for research on antiseptic skin preparation to examine this further.
Antiseptic skin preparation before skin incision is standard practice although the type of antiseptic used varies depending on the type of surgery.
The recommendations follow current trends in practice and should reduce variation.
Limited evidence was identified on the intraoperative use of topical wound antiseptics before wound closure. Although this evidence suggested that topical povidone-iodine was effective in reducing surgical site infections, the studies were dated. This evidence also suggested that topical antiseptics, such as iodine in alcohol solution, are not effective in reducing surgical site infections.
The evidence on topical antibiotics before wound closure was varied, but also included several older studies. Some studies showed that antibiotics, such as ampicillin powder and cephaloridine, reduced the number of surgical site infections. However, the evidence for other antibiotics, such as vancomycin, which is widely used worldwide and commonly used in cardiac, orthopaedic and spine surgery, suggested no reduction in surgical site infections.
The committee agreed that the evidence was not current or clear enough to make a recommendation on the use of topical antiseptics and antibiotics before wound closure. The committee also took into account concerns about antimicrobial resistance and the potential for multidrug resistance, and agreed that without new conclusive evidence, use of intraoperative topical antibiotic and antiseptics should be stopped. They agreed that this is an important area for further research and recommended that they should be considered only in the context of further research to help limit unnecessary use and determine their clinical effectiveness. They also developed a recommendation for research on applying antiseptics and antibiotics before wound closure to determine the clinical and cost effectiveness of applying antiseptics and antibiotics before wound closure in reducing surgical site infection rates.
There was some economic evidence that antibiotic-loaded bone cement was cost effective when compared with plain cement. However, the committee were not confident that the evidence was applicable to current NHS practice. In addition, the clinical evidence suggested that antibiotic-loaded bone cement did not reduce the number of surgical site infections. The committee agreed that the evidence was too limited to make a recommendation for this intervention.
Evidence was also identified on the use of gentamicin implants before skin closure during different surgical procedures. In particular, the evidence suggested that gentamicin-collagen implants reduced the incidence of surgical site infections in people at 1 month and 2 months after cardiac surgery. Although the evidence was limited, cardiac surgery is associated with a high risk of surgical site infection, which is expensive to manage. Therefore, the committee agreed that gentamicin-collagen implants should be an option to reduce the risk of infection.
In practice, the use of topical antiseptics and antibiotics before wound closure varies. Limiting their use to clinical trials is likely to reduce their misuse in practice and encourage research in this area.
Although gentamicin-collagen implants are used in cardiac surgery, not all services currently use them. The new recommendation may help to reduce variation and standardise practice. Any additional costs are likely to be balanced by savings from a reduction in the number of surgical site infections.
Overall, the evidence suggested that staples increase the incidence of wound dehiscence when compared with sutures for wound closure across different types of surgery. However, when the studies were analysed according to the type of surgery, many of the studies showing this difference were found to be on wound closure after caesarean birth. The committee agreed that there was not enough evidence to recommend sutures over staples in all surgery, and decided to focus the recommendation on caesarean birth. The committee agreed that this was important in improving recovery for women having caesarean births, and that it should be reflected in the recommendations. However, the committee noted that the evidence did not capture all populations, for example obese women. Therefore, the recommendation was made to consider sutures rather than staples.
The committee discussed the evidence for antimicrobial triclosan-coated sutures and agreed that the evidence overall favoured triclosan-coated sutures over standard sutures for reducing surgical site infection. However, they noted that the studies covered many different types of surgery and were of variable quality, meaning that it was difficult to be confident of the benefit. Further analysis by the type of surgery showed a clear benefit of using triclosan-coated sutures only in paediatric surgery. The committee therefore agreed that they should be considered as an option for wound closure in all types of surgery, and that their use in paediatric surgery should be emphasised in particular. The committee also developed a recommendation for research on closure methods to better clarify which patients should have triclosan-coated sutures and which surgical layers they should be used for.
The recommendations are unlikely to have a major effect on current practice. Current practice in wound closure varies, so the new recommendations may help to reduce variation and standardise practice.
Using sutures rather than staples for wound closure in caesarean birth may lead to a reduction in the number of women experiencing wound dehiscence following surgery, which may reduce the costs of treatment. However, the committee acknowledged that there may be training implications for those implementing the recommendation.
Use of antimicrobial triclosan-coated sutures may increase, which may have cost implications because they are more expensive than standard sutures. However, it is likely that the increased cost will be outweighed by savings from a reduction in the number of surgical site infections, which are costly to treat.