Surgical site infections: What do we know?

Abstract

Though inherently linked to the act of surgery, surgical site infection (SSI) is an adverse event that places a heavy burden on surgery. It is every surgeon's nightmare, particularly since no matter how hard we try and prevent this complication, we do not seem to be able to completely prevent SSIs. But are we really trying? Surgical site infection has long been the main cause of surgical failure, historically leading to situations where surgery was considered an intervention of last resort, and the procedure worse than the actual pathology. In the middle of the 19th century abdominal interventions in man were banned due to the unacceptable mortality associated with the high infection rate 1. Historically, infectious complications like erysipelas, tetanus, pyaemia, septicaemia and gangrene caused mean mortality rates of 60% subsequent to the most commonly performed surgical procedure of amputation 2. Suppuration was considered normal, even a necessary and unavoidable step in the wound healing process. It is only with the pioneering work of Semmelweis, who published his book on the aetiology and prophylaxis of puerperal fever in 1861, that the scientific and medical world understood that human lives were not a necessary toll to be paid through what is now recognised as nosocomial infections. The simultaneous discovery of the germ theory led to the establishment of methods to combat sepsis, leading to the antisepsis era mainly promoted by Lister and his use of carbolic acid bandages. Soon after, pushed by Pasteur's saying, Instead of fighting bacteria in wounds, would it not just be better not to introduce them, surgeons like Koch and Halsted developed the methods to avoid the development of sepsis: the founding principles of asepsis and aseptic technique. Nowadays, these discoveries are considered to have had the greatest impact ever on patient survival with regard to infectious diseases or probably any other medical advancement. Yet, at the time, their peers sneered at those that we now consider ‘godfathers of surgery’. Despite the dramatic drops in mortality, the establishment of clean aseptic surgery was delayed by ignorance and disbelief of evidence together with the belief that personal experience was more essential than aseptic technique. For instance the filthier the jacket the greater the prestige, as surgeons accumulated blood stains from previous operations as a sign of experience. Even today, with the accumulation, understanding and apparent acceptance of all this knowledge, surgical site infection remains the most common cause of infection in surgical patients 3, both human and veterinary. In equine surgery, depending on procedure and wound classification, surgical site infection rates of 0% to more than 50% have been reported 4. Hampered by the lack of appropriate SSI definitions, absence of proper prospective surveillance programmes and small sample size, these reports probably provide a false insight into the SSI problem in equine surgery. The European CDC report on human SSI shows that up to 73% of SSIs are only diagnosed after hospital discharge 5, and the situation is unlikely to be very different in equine surgery. Large numbers of SSIs, particularly the superficial or mild ones, are probably never reported, as they are treated by veterinarians outside equine hospitals and never brought to the attention of the primary surgeon. Further, in the sole report of prospective SSI surveillance in veterinary medicine, active follow up of SSI cases and scrutiny of medical records have shown that approximately 35% of SSIs do not get reported in the medical record system 6. The source of bacterial contamination in a surgical wound can be endogenous and exogenous. The endogenous sources of contamination come from the patient's commensal microbiota at the surgical site or distant body sites. The exogenous sources of contamination are those originating from the surgical team, the environment, and the materials and instruments used. Based on human data, it is said that approximately half of all the identified SSI risk factors are endogenous and although some parameters (e.g. age, systemic disease, history of prior surgery) are difficult to modify in the direct operative period, careful selection (e.g. avoid operating on a patient with remote infections) and preparation of the patient can help control endogenous sources of contamination. New avenues in SSI prevention mainly focus on these endogenous factors, with screening for resistant staphylococci carriage and decolonisation prior to surgery potentially also being useful in equine medicine. Further, endogenous measures not targeting the transmission or elimination of pathogens but rather enhancing the host's immunity, like control of intraoperative hypothermia, tissue glucose levels, tissue oxygenation and fluid management all seem promising. The other half of risk factors for surgical site infections, however, are exogenous and can be adapted easily without need for new research. Past experience has shown that simple and low cost interventions that target the exogenous factors have the best chance of having an impact on reduction of SSI. A set of recommendations based mostly on historical intellectual rationales and experts' opinions dating back to the era of the ‘godfathers of surgery’ form the basis of SSI prevention today. Some measures such as surgical hand preparation, appropriate antimicrobial prophylaxis, postponing elective surgeries in case of remote infection, active and post discharge surveillance programmes and surgical etiquette have now passed the level of high evidence based guidelines. Although their use is commonly accepted, other methods such as patient preparation techniques, surgical attire, wound dressing, choice of suture material are only supported by low evidence, mostly due to difficulties in performing high evidence trials in this area of research. Blinded, randomised surgical trials with or without gloves for example would be ethically unacceptable. Further, it is unlikely that individual methods have a dramatic impact on SSI, but that it is rather the combination of all that makes the difference. The subject of veterinary SSI prevention has been recently reviewed 7. Hand hygiene is considered the pillar of prevention of infectious diseases, and although strict randomised trials of surgeries with and without previous hand preparation are lacking, presurgical hand asepsis is the most effective preventive measure for surgical site infection 8. Historical evidence, such as the introduction of hand washing between the morgue and delivery ward by Semmelweis and the consequent dramatic drop in puerperal fever infections, is undeniable proof of its effectiveness. Since then, the evidence has grown and the literature on the basic science of hand hygiene has developed at a much faster pace than any literature available in the equine scientific field. Unfortunately, the basic knowledge about correct presurgical hand asepsis in veterinary surgeons is low (D. Verwilghen, unpublished data), even amongst the specialist community (ECVS and ACVS Diplomates), in which only 44% adhere to current state-of-the-art hand asepsis protocols based on hydro-alcoholic solutions 9. Antimicrobial prophylaxis (AMP), performed in the absence of established infection, is indicated in surgical procedures with the wound classification ‘clean-contaminated’ or ‘contaminated’ because of the increased risk of surgical site infection. The use of AMP in ‘clean’ wounds is controversial in both human and veterinary medicine. It is believed not to be necessary in clean human surgeries where the reported incidence of infection is lower than 5% 10 unless the operation lasts for more than 3 h, or the ASA score of the patient is 3 or more 11. Clean surgeries are considered those performed on an elective basis, that are closed by primary intention, non-traumatic of origin, uninfected, with no break in aseptic technique and where the respiratory, alimentary and genito-urinary tracts are not entered 4. A recent paper evaluating post operative septic arthritis following elective equine arthroscopies without AMP reported similar infections rates to those reported in populations operated with AMP 12. These data support a previous study in which post arthroscopy septic complications were not associated with administration or lack of antibiotics 13, suggesting that an approach similar to human recommendations can be used in equine medicine. Southwood 14, Ahern and Richardson 4 and Hollis and Wilkins 15 provide extensive reviews on the proper guidelines for use of antibiotics within equine surgical prophylaxis. While in the past, ‘no development of infection’ was the ultimate indicator of success, today we should consider ‘no development of infection and minimal risk of antimicrobial resistance emergence and antimicrobial adverse effects’ to be the ultimate desired goal. Consequently, in addition to the above guidelines, the principle of ‘less is more’, should be considered. While the extra benefit of AMP in reducing surgical site infection cannot be disputed 16, antimicrobials should not be used as the key component of the infection control plan. The risks of using antibiotics should be weighed against their benefits and not every patient undergoing surgery should receive AMP. Certainly, AMP should not be used to compensate for poor surgical technique or facilities 16 and adherence to Halsted principles of aseptic and atraumatic surgery should remain the foundation for prevention of SSI. Some argue that infection risks must be inherently higher in horses, and so the need for antimicrobials greater, because of their environment and recovery in a stall rather than a ‘clean’ hospital bed. Yet, the relative frequency of hospital-associated infections in humans and horses (an admittedly tenuous comparison) does not clearly indicate that hospital beds are a safer environment than barns and stalls. Furthermore, uncontrolled use of AMP eventually will lead to a more hostile environment with a higher infection pressure of multiresistant bacteria, and therefore a higher risk of SSI 17. Additionally, antimicrobial-induced colitis has been reported in 6.3% of horses receiving AMP with arthroscopic interventions 18. Compared with a SSI rate of 0.5% reported for arthroscopy, this is a high toll to pay. Operating room etiquette, surgical experience and technique: a set of words rarely heard in veterinary surgery. Yet the sound judgement and proper technique of the surgeon, and of the surgical and anaesthetic team, are an important factor in the prevention of surgical complications, including surgical site infection 19. Proper surgical training is not a day one competence and surgeons' experience and level of education have been shown to be related to the development of SSI. Closure of equine coeliotomy wounds by first or second year residents is a significant risk factor for SSI development 20 and survival after colic surgery is linked with board-certification status of the surgeon 21. Longer wound exposure times and longer tissue manipulation provide more opportunities for pathogens to seed into the wound and cause surgical site infection and at the same time allow for greater chance of wound desiccation or other forms of damage. This explains why surgery and anaesthesia time are often identified as risk factors in SSI studies. Whilst surgeon experience can be one factor in extending operation time, time spent in surgery will be dictated by the entire team. Surgical planning, availability of appropriate instrumentations, delivery of requested disposals in an efficient way, coordination with diagnostic imaging and myriad other factors can reduce surgical time and should be actively tackled in equine surgical settings. Although probably a surrogate outcome to other factors, operating room etiquette has been related to the development of surgical site infection. Noise level in the surgical theatre, for instance, has been significantly correlated with higher SSI rates 22. Talking about nonsurgery related topics is associated with higher sound levels and was shown to reduce concentration and increase surgical errors. A most interesting study on the subject reminds us that our daily work behaviour involves variable issues that impact outcomes, demonstrating that lapses in discipline by the surgical team were an independent risk factor for the development of SSIs 23. Increased movement in the theatre, exchange of surgical team members, noise and the presence of visitors in the operating room all independently contributed to increased infection rates. From the above it is clear that proper prevention measures are not an individual action, but involve adequate preparation of the surgical facility and the environment, surgical site, surgical and anaesthesia team, and surgical equipment. Basic rules are straightforward and simple to implement, but are unfortunately often not followed. Every member of a healthcare setting, including the surgeon, assistants, cleaning staff and the management team carry responsibility in the achievement of aseptic procedures and the corresponding success of surgeries. Adhering to all these practices builds the basis of what is known as ‘surgical and operating room team conscience’. Money, money, money…: Biosecurity measures cost money, but so do surgical site infections. Well documented through national health programmes, the cost to a hospital for SSI patients is calculated at twice the cost of a non-SSI patient 24. In horses SSI cost has not yet been formally established, but the cost of infection in tibial plateau levelling osteotomy (TPLO) procedures in dogs leads to a mean cost of SSI treatment of at least 50% of initial surgery cost 25. This does not take into account the negative reputational issues of SSI and the fact that ‘real billing’ for time and materials used is often not accounted for in in-house complications, putting an underestimated economical burden on the hospital. Even in the early days of SSI prevention, the cost issue was debated 2 and hampered the introduction of many of the infection control discoveries. Nowadays, practice owners and managers are often reluctant to change and introduce measures they consider extra cost (single-use gowns and drapes, better gloves …) ending in shortsighted micro-management. However, figures can talk. If a complication occurs with a prevalence of 10% and the associated treatment cost is €1000, a completely effective intervention against the complication would be justifiable economically at €100 per patient. Likewise if the intervention can halve the infection rate a cost up to €50 would be economically sound. Research into equine surgical site infections is mostly limited to colic surgeries where, amongst other factors, surgical time, use of staples, belly bands and suture pattern are reported as risk or protective factors 4. Confusingly, often factors that are protective in one population seem to be a risk factor in another, probably due to unstandardised approaches, lack of specific identical guidelines on surgical preparation methods, and variable definitions of SSI. This results in studies which are difficult to compare. Most interestingly, however, is the extremely wide variation in reported infection range (0.7–40.4% 26, 27) for colic surgery, suggesting that simple measures can reduce high SSI in certain populations. After all SSI is the most preventable of all the hospital-acquired infections 28. Yet, despite the available literature and the many published guidelines, protocol compliance rates remain highly unsatisfactory 28. Eventually, many identified risk factors can be attributed to our own behaviour and the prime and most threatening factor for development of SSI is probably ourselves. In a survey performed amongst human surgeons, 63% did not comply with the current recommended guidelines on preoperative bathing, hair removal, antimicrobial prophylaxis and intraoperative skin preparation 29. Furthermore, an observational study in companion animal clinics revealed inconsistent and poor compliance to well established surgical preparation practices 30. Recommended times for antiseptic soap cleansing of patient surgical site and surgeon hand preparation are at least 2 min; however, this was observed to be as short as 10 and 7 s, respectively, and nonsterile contact with the previously aseptically prepared surgical site occurred in at least 36% of cases30. So as long as compliance with basic protocols is not met, introducing more extensive measures will not have any significant impact on SSI reduction 23. Confirming that surgeons' behaviour in the operating theatre does not necessarily correlate with their scientific knowledge 31, 66% of respondents in a survey amongst ACVS and ECVS surgical specialists did not follow recommendations and their own stated beliefs that hydro-alcoholic solutions are superior for hand asepsis 9. With the discovery of antimicrobials, many thought that infectious diseases would become a thing of the past. However, with the effectiveness of the antimicrobial arsenal waning, infection prevention has to come to the forefront again. No pathogen has yet developed any resistance to aseptic technique, thus avoiding the introduction of pathogens instead of fighting them is more pertinent than ever in our journey to eradicate surgical site infections. The challenge is to do the right thing over and over again. As surgeons we embrace Halsted's ‘Cut Well, Sew Well and Do Well’ principles, but also recognise that success and failure of a surgical intervention does not lie solely in the expertise of surgeon's fingers, but depends on the surgical team. Hospital members must consider themselves as part of this team. Cynicism as a personal defence against engaging with the challenges that infection control brings to daily practice has to be avoided. The so-called ‘Semmelweis effect’, a metaphor for the reflex-like tendency to reject new evidence because it contradicts established norms and shakes the secure feeling of habits, is a real killer. It is clear that optimal SSI prevention will only be achieved by a multidisciplinary and multimodal approach in which the most important factor is the consciences of every individual involved. We are the biggest risk factor for the development of surgical site infections.