Surgical site infections (SSIs) account for 16% of nosocomial infections in humans and although there is currently no official surveillance system available in the veterinary industry it is thought that SSIs account for 0.8–18.1% of surgical complications in small animal species (Nelson, 2011). SSIs have many disadvantages such as client dissatisfaction, increased hospitalisation costs, revision surgeries and even patient morbidity all of which increase workload for clinical staff.
SSIs are infections that occur in a wound created by an invasive surgical procedure and can be classed as ‘superficial incisional’, ‘deep incisional’ (Figure 1) and ‘organ/space’ (Mangram et al, 1999). The first line of defense by a surgical wound against both endogenous and exogenous pathogens is oxidative killing of such pathogens by neutrophils, which as the name of the process suggests, requires a good oxygen supply (Paulikas, 2008). When a patient becomes hypothermic they experience peripheral vasoconstriction, the implication of this is that the blood flow is compromised thus decreasing the rate of oxygen transport to a surgical wound. It is within this blood flow that neutrophils and other immune substances are normally delivered to help with wound healing (Kirkbride and Buggy, 2003). Perioperative hypothermia is also known to decrease numbers of leukocytes and cytokines (Forbes et al, 2009). Leukocytes are white blood cells and the body's defence system against infection, and cytokines aim to decrease inflammation and promote wound healing (Dinarello, 2000). Kurz et al (1996) also found that there were more collagen deposits near the wounds in patients who remained normothermic throughout their surgery. Collagen deposits rely on a well oxygenated environment and provide wounds with much of their strength (O'Dwyer, 2007).
SSI risk factors such as duration of surgery, number of people in the operating theatre, contamination of surgical site and age have been identified in both human and animal studies (Eugster et al, 2004; Seamon et al, 2012). Perioperative hypothermia has also been identified as a risk factor, mainly in human literature (Kurz et al, 1996) and is defined by Forbes et al (2009) as a temperature below 36.0°C during the perioperative period.
How patients lose heat in the perioperative phase
Healthy, conscious animals respond to heat loss via temperature sensitive cells in the hypothalamus which initiate a response such as shivering and vasoconstriction (Murison, 2001). In the perioperative phase when patients are sedated or under general anaesthesia (GA) there are a number of ways in which patients can become hypothermic such as cold theatres, evaporation of moisture from body cavities, respiratory heat losses and surgical preparation of patients such as clipping and cleaning the surgical area. GA also depresses many of the body's mechanisms including the central nervous system which in turn affects the sensitivity of the hypothalamus to changes in temperature (Murison, 2001).
Initial heat loss in the perioperative phase occurs due to redistribution of heat from the patient's core to the periphery and this can result in a drop in temperature of 1–1.5°C which normally occurs shortly after induction of GA, often due to vasodilatory anaesthetic drugs (Armstrong et al, 2005).
Heat is lost from the body in four ways and radiation and convection account for up to 85% of heat loss (Paulikas, 2008). Radiation is the transfer of heat energy from hot to cold (Campbell, 2008), for example the loss of heat from an animal when anaesthetised to a cool operating theatre. Ideal theatre temperature as recommended by the National Institute for Health and Care Excellence (NICE) (2008) is 22°C and so this transfer of heat is always inevitable because of the temperature of the patient entering theatre. Convection is the loss of heat as air or water molecules move across the skin (Paulikas, 2008). This can be avoided by using passive means of patient warming for example covering the animal with a blanket. Implicated in this is the fact that many operating theatres do not allow the use of blankets because of the hair and dander which they might be carrying on them. Conduction and evaporation account for the other 15% of heat loss. Conduction is a method of heat loss from the body to a cold surface it may be in contact with such as a cold operating table and evaporation is the dissipation of heat from moisture in the body (Paulikas, 2008), for example as an animal breathes they lose heat in the moisture which is expired or when viscera is exposed, for example in laparotomies.
There are a number of ways in which heat loss can be reduced in the perioperative phase such as through the use of heat mats, bubble wrap, warm intravenous fluids and forced air warming (FAW) blankets, and it is imperative that these methods are adopted, not only to reduce the risk of hypothermia but also for patient comfort. Registered veterinary nurses (RVNs) have a duty of care towards patient welfare (Royal College of Veterinary Surgeons, 2012) and as advocates for their patients the RVN should always be striving for gold standards of care and so should play a large role in maintaining normothermia throughout the perioperative phase.
Within human medicine, NICE (2008) have published guidelines on perioperative hypothermia which recommend that measures should be undertaken to maintain patient temperature above 36°C in order to avoid complications associated with peri-operative hypothermia. The average temperature of a human is 37°C so with that in mind it could be said that perioperative veterinary patients should be maintained even higher than this.
Incidence of SSIs in the literature
A literature review of the incidences of SSIs reveals there is a degree of subjectivity in what constitutes an SSI, which is open to interpretation. A reliable and reputable study should therefore adopt the use of a wound scoring system such as the ASEPSIS wound scoring system. ASEPSIS stands for Additional treatment, presence of Serous discharge, Erythema, Purulent exudate, Separation of the deep tissue, Isolation of bacteria, and duration of inpatient Stay (Wilson et al, 1998).
There are many other factors which can contribute to SSIs so studies that look solely into perioperative hypothermia should have some sort of inclusion or exclusion criteria in the sampling process so that participants with existing risk factors were excluded. By not doing this there is the potential that the sample will not result in an accurate assessment of the topic being studied (Li Biondo Wood and Haber, 2006). For example, if animals who had endocrine diseases were included in the sample then the risk of SSI would potentially be greater as it is known that animals with endocrine disease are at a higher risk of SSI (Eugster et al, 2004).
A double blind study by Kurz et al (1996) investigated whether perioperative normothermia reduced the incidence of surgical wound infection in human patients. Their sample size of 200 was split into 104 and 96 in the normothermia and hypothermia groups respectively and they found that 19% of patients in the hypothermia group developed an SSI compared with 6% in the normothermia group. Within this study, the physicians were double blinded and so unaware of the group allocations. The wounds were evaluated using a wound scoring system (ASEPSIS) and when a wound was suspected of having an SSI this was confirmed by a positive culture. The two groups were similar with regards to variables such as smoking status, pre operative blood test results as well as intraoperative variables such as drugs used and fluid therapy. Rather worryingly this study concluded that core intraoperative temperatures two degrees below normal tripled the incidence of wound infection and prolonged hospitalisation by up to 20%. These patients were undergoing colorectal surgeries (clean contaminated) and so may have been at a higher risk of an SSI compared with those patients undergoing clean surgeries. The results of this study show how important maintaining normothermia is in high risk patients. Similarly, Seamon et al (2012) found in their study into the effect of perioperative hypothermia on SSIs following trauma laparotomies that 35°C was the temperature below which there was a profound impact on the rate of SSIs. Their study was retrospective and so did not have the strict controls which the Kurz et al (1996) study had, although they did have exclusion criteria.
There are few veterinary studies into the effects of perioperative hypothermia on SSI rates. Some papers are available which look into risk factors, of which perioperative hypothermia is one (Beal et al, 2000), but none look solely into perioperative hypothermia. It is well known in veterinary surgery that the longer the procedure takes the higher the risk of SSI (Eugster et al, 2004), although whether this could be due to perioperative hypothermia is yet to be established. Beal et al (2000) carried out a study into the effect of perioperative hypothermia and duration of anaesthesia on wound infection rate in clean wounds and found that perioperative hypothermia (mean lowest temperatures recorded were 36.3°c) was not a significant risk factor for SSIs. This study was based on clean wounds rather than the previous two human studies which focused on clean contaminated and contaminated surgeries. These studies show that the type of surgery and the risk factor for SSI should be taken into account by the RVN when considering patient comfort and normothermia controls. Beal et al's (2000) study was also retrospective and so lacked the strict controls which would be present in a more prospective study. There was no control group as all the patients were treated to maintain normothermia and once the animal had been discharged owner interpretation of the surgical wound was relied on, which is largely subjective.
Conclusion
SSIs are becoming highlighted more in day to day practice and it is important the RVNs have an understanding of the risk factors and their implications.
One veterinary study (Beal et al, 2000) conflicts with the studies for human counterparts by finding that perioperative hypothermia was not a significant risk factor for SSIs. However, aspects of the studies are very different (clean wounds vs clean-contaminated wounds) and it would be interesting to have some more veterinary studies looking into a potential link between perioperative hypothermia and SSIs in some of the other classifications of surgical wounds.
This part of the literature review can conclude that the veterinary profession lags behind human medicine in its investigations of SSIs, however, if the results of human studies can be extrapolated to veterinary patients then this evidence base can be used to improve standards of care because it can be seen from human studies, such as the one by Kurz et al (1996), that there is a link between hypothermia and SSIs in clean contaminated surgeries. More studies in the veterinary field would establish whether similar results can be yielded.