This service improvement study emerged from the experiences of the tissue viability nurses (TVNs) working in the acute care environment, who are referred patients following their discharge from hospital, many of whom presented with a surgical site infection (SSI) and some of whom had inappropriate or delayed assessment or treatment. The study aimed to gain some local insight into current practices and to inform and improve clinical practice for patients who develop SSIs after discharge. Relatively little attention has been given to the amount of SSIs that present following discharge (Woelber, 2016).
Background
An SSI is a wound infection that can develop after a surgical procedure (National Institute for Health and Care Excellence (NICE), 2019). Between 2013 and 2018 the incidence of SSIs following surgery (including readmissions) ranged between 0.5% for knee replacements and 8.7% for large bowel surgery, with coronary artery bypass graft (CABG) having an incidence of 3.5% (Public Health England (PHE), 2018), caused by bacteria multiplying in the wound.
The severity of SSIs varies. The Centers for Disease Control and Prevention (CDC) (2019) has defined three surgical site infection groups: superficial, deep incisional and organ-space SSIs (see Table 1). Only infections that develop within 30 days of surgery are classed as SSIs; however, this is extended to up to 1 year if the surgery involved implants. Wound infections can develop after surgery, but many are considered to have developed at the time the patient is in surgery, although these individuals may not develop any local or systemic signs of infection until after they have left hospital (NICE, 2019).
| Superficial incisional SSI | Deep incisional SSI |
|---|---|
|
Must meet the following 2 criteria:
|
Must meet the following 3 criteria:
|
| The following are not considered superficial SSIs | Organ-space SSIs |
|
|
Must meet the following 2 criteria:
|
Superficial infections are often easily treated with a short course of antibiotics. In cardiac surgery, superficial sternal wound infections often cause the patient some pain and discomfort, but they are likely to resolve in a short period with appropriate management, such as topical wound dressings and antibiotics. However, deeper sternal wound infections can cause serious morbidity and mortality, as well as bring significant financial costs to the health service (Gould, 2012).
Badia et al (2017) highlighted the economic burden caused by SSIs. They can result in extended hospital stays for patients or readmission to hospital. They require additional treatments, all with associated costs to the NHS; in addition, there will be costs for the patient related to their quality of life, increase in pain and a negative psychological and emotional impact (Tanner et al, 2012). Gottrup et al (2005) referred to SSIs as a burden for patients in terms of morbidity and mortality, and for health services in terms of the financial costs incurred. NICE (2019) has stated, due to the advances made in surgery and anaesthesia, more patients who are at higher risk of developing SSIs are today being offered surgery. With a greater number of high-risk individuals undergoing procedures there is, arguably, a need for more monitoring of such patients postoperatively and post discharge.
SSI surveillance
The aim of SSI surveillance is to provide accurate SSI rates that can be used to optimise and continuously improve the quality of patient care (Harrington, 2014). It can enable proactive and early interventions, reducing what is often considered to be a preventable harm to patients.
Local trust surveillance process
At Liverpool Heart and Chest NHS Foundation Trust, the infection prevention nurses collect and monitor SSI data relating to inpatients and readmissions, and use microbiology (laboratory) results to prompt review of additional patient records. The TVNs contribute to the data when patients are under their care, including those who are seen only after hospital discharge, by adding their details to a shared database of patients who have developed an SSI.
Prevalence of post-discharge SSI
Petherick et al (2006) described a global trend regarding shorter inpatient stays for surgical patients and more daycase surgery, meaning that more SSIs are likely to present after the patient has been discharged. More recently, Woelber et al (2016) conducted a systematic review of SSIs that develop after discharge. The review included 55 articles from 15 countries and estimated that post-discharge SSIs account for up to 60% of all SSIs. It is argued by Wilson (2013a) that planned, prospective surveillance will identify additional SSIs that would not otherwise be captured, thereby providing a more accurate SSI rate; active methods of surveillance would also ensure systematic and consistent data collection. However, for this type of surveillance to be undertaken, organisations need to ensure that they allocate appropriate resources. Where it is an additional role for clinical staff with other responsibilities, for example, infection prevention nurses, passive surveillance (using retrospective data from medical records) is more likely to occur, resulting in incomplete and inaccurate SSI rates (Wilson, 2013a).
Ethical considerations
Although the study involved no direct patient contact, it involved processing personal data from the electronic patient records (EPR). Data protection and confidentiality practices were maintained. Permission was gained from the organisation.
Methods
A quantitative approach was used. Retrospective data collection from the EPR of all patients who had been seen in the tissue viability service wound clinic after discharge between 1 April 2017 and 31 March 2018 was used to determine how many individuals had an SSI, as well as patient demographics and other relevant data, as detailed in the collection form.
To retrieve reliable data, Gregory and Radovinsky (2012) stressed the importance of using a reliable and tested data collection form, and the use of a code book to guide the data abstractors. For this study, a pilot test was completed on a sample of 5 patients, which identified improvements to the form, ensuring easier and more effective data collection. It established the content validity of the data collection tool, and improved the data being collected, the scales used and its overall format. It tested inter-rater reliability, to establish that all 3 data abstractors were retrieving data consistently. A code book (Box 1) was developed specifically for the study, which was also tested, and changes made during the pilot. The code book included each item on the data collection form, a definition of what each means in this study, and the best source of data within the patient record.
The data were entered onto Microsoft Excel and analysed using the SPSSv25 statistical software package.
Results and data analysis
The study population was drawn from 112 patients who had been reviewed by the tissue viability service in the wound clinic: 77 were new patients referred to the service after discharge and 32 whose care continued after discharge (3 patients were excluded because they did not have a surgical wound) (Figure 1). Of these, 59 patients developed SSIs (within 30 days of surgery). This is the group discussed further in this article.
Some patients' wounds had already presented with delayed healing, dehiscence and/or infection before discharge, and were already under the care of the TVNs (n=32/112; 28.6%), so their management continued in the wound clinic in the outpatient department after they were discharged. The remaining 77 were referred to the service after discharge as follows:
Most of the 112 patients (n=86; 76.8%) who were seen post discharge had undergone coronary artery bypass graft (CABG) surgery, CABG plus valve surgery, or valve surgery alone. A smaller number had wounds relating to thoracic surgery or pacemaker sites (n=5; 4.5%); and a further group had ‘other’ wounds related to rewiring, removal of sternal wires and surgical debridement (n=11; 9.8%).
The most frequently seen wounds were sternal (n=69/112; 62%) with the next most frequent being donor leg wounds (n=28/112; 25%); the latter is created when the long saphenous vein is harvested for use in CABG surgery. Other patients had thoracotomy wounds (n=2/112; 2%) and a further group had wounds relating to drain sites, groin wounds and other procedures (n=11/112; 10%). For 2 (1.8%) patients, the wound type was not recorded.
Of the 112 patients who were reviewed by the tissue viability service in clinic after discharge, 29 (25.9%) patients had no infection, and this information was not recorded for 5 patients. A total of 59 patients were considered eligible for inclusion in the study as having developed an SSI (ie within 30 days of the date of surgery). The TVNs determined that 31 patients had not developed an SSI, in 17 cases it was not possible to determine whether there was an infection or not, and for 5 patients this information was not recorded. The missing data may have been because the nurse had not asked the patient during the consultation, had not recorded it in the patient record, or the patient may have been unable to recall information about the treatment relating to that period.
Patient demographics
The demographic information for the 59 patients who were seen post discharge who developed an SSI was as follows:
Characteristics such as gender, age, high BMI and diabetes are reported as risk factors for SSIs (Korol, 2017). In the final cohort of 59, more than twice as many men as women who were seen in outpatients had developed an SSI (n=42; 71%, versus n=17; 29%). In Liverpool Heart and Chest NHS Foundation Trust, the gender balance for cardiac surgery is about 70% men and 30% women
The patients were resident across England and Wales and the Isle of Man. Their post-discharge infections were more likely to be managed or readmitted locally. Consequently, not all patients would have been known to the tissue viability service and therefore would not have been captured within this study or within SSI data.
Referral and management
Table 2 shows the route of referral for the 59 patients who were recorded as having developed an SSI: 20 were already known to the tissue viability service and their care continued after discharge and 39 patients were referred to the service after discharge. Of these 39 patients, 15 (25%) had referred themselves; 7 (12%) were referred from follow-up clinic, attending a planned appointment and presenting with wound problems; 12 (20%) were referred by a GP or medical team (14% and 7% respectively). Had these 39 patients not been referred to the TVNs, they may not have been captured in the Trust's SSI data.
| Route of referral | Number of patients (%) |
|---|---|
| Not recorded | 2 (3) |
| At planned follow-up appointment | 7 (12) |
| Continued care by tissue viability nurse | 20 (34) |
| District nurse | 3 (5) |
| GP | 4 (7) |
| Medical team | 8 (14) |
| Patient | 15 (25) |
| Total | 59 (100) |
Table 3 shows that more than half the patients (n=20; 51%) were under the care of community nurses. It is important that community staff know who to refer back to if wounds fail to progress. The tissue viability service contact details are included in the discharge information provided to patients, but community staff may not be aware of these. Most patients (19/38; 50%) (1 patient was excluded due to insufficient information) continued to have ongoing wound problems 4 weeks after discharge before being referred by a health professional or self-referring for review by the tissue viability service. A recommendation has been made to develop supportive working relationships with community nurses, provide education where needed and improve the quality of patient information provided.
| Number of patients | Frequency | Percent |
|---|---|---|
| Not recorded | 9 | 23 |
| No | 10 | 26 |
| Yes | 20 | 51 |
| Total | 39 | 100 |
In cardiac surgery, 60% of patients develop SSIs within 30 days of surgery. However, the average time for a superficial infection to present is 10 days after surgery and 14 days for a deep or organ-space infection (Rochon et al, 2016). Patients are often in hospital for shorter periods and experience earlier discharges, before such wound infections have had time to develop, and will therefore present when the patient is back in the community (Melling et al, 2005). In this study, 26/39 (67%) of the patients who presented with SSIs after discharge had had an inpatient stay of less than 8 days (Figure 3).
Figure 4 shows the type of interventions and treatments that patients in this review (n=38) received for their wounds and wound infections. These ranged from additional antibiotics and topical wound dressings to wound reopenings, debridement, readmission, intravenous antibiotics and further surgery. Negative pressure wound therapy is often used in the management of surgical wounds that have not followed the normal wound healing process, with or without the presence of infection.
Discussion and recommendations
This study has provided local insight into activity in a wound clinic, with a specific focus on patients who developed SSI after discharge. Some patients did not present with signs of wound infection until after they had left hospital. This supports an argument for the Trust to consider participating in some form of post-discharge surveillance. Many of the patients also had an inpatient stay of less than 8 days, which increases the risk of a wound infection presenting after discharge.
Despite the fact that many patients had community nurse involvement, it was more than 4 weeks before the patient or nurse contacted the hospital for advice. Early diagnosis of an SSI is crucial in ensuring that the patient receives the most effective management (Keast and Swanson, 2014). The tissue viability service would suggest that it needs to receive referrals earlier in the patient's management. However, the information currently provided to patients following surgery does not include a referral pathway with recommended timeframes.
The study demonstrated that a variety of treatments are required as a result of a patient developing an SSI, highlighting the frequent use of additional interventions to promote wound healing, as part of treatment of an SSI. This not only increases costs, but also has an impact on the patient's quality of life.
These findings have been used to make recommendations to improve the quality of care for patients who develop an SSI after hospital discharge.
Recommendations
Review of the SSI surveillance process in the Trust
The Trust should consider a review of the current surveillance process, and consider options for future surveillance, including participating in post-discharge surveillance and different methods that could be used, such a such as patient questionnaire or follow-up telephone call. It has been argued that SSIs will be found if they are actively looked for (Downie, 2010), suggesting that passive surveillance misses some. Since up to 70% of SSIs can be identified after discharge there appears to be a strong argument for the Trust to consider participating in post-discharge surveillance.
Develop supportive working relationships with community nurses and provision of education
Community nurses are involved in managing several types of wound and many factors can affect wound healing. These issues can make the management of different wounds in the community difficult and challenging.
Welsh (2018) stated that nurses may have insufficient wound-care knowledge, and follow practices that are sometimes ritualistic and out of date. Appropriate education for community nurses could contribute to improved knowledge and skills in surgical wound management and enable them to recognise early signs of infection, leading them to seek earlier involvement with the tissue viability service.
It is recommended that the tissue viability service forges closer working relationships with community staff, beginning with the local community Trust and delivers specific training in managing patients with wounds following cardiac surgery.
Review of patient information provided on discharge
All patients should be given consistent and understandable information about their wound when they are discharged from hospital. This should include how to care for their wound, recognising signs of an SSI and whom to contact if they have any concerns. Current wound-care patient information is included as part of a cardiac surgery discharge booklet. An additional leaflet is available, but it is not known whether all patients are provided with this; practices seem to vary from ward to ward. Sanger et al (2014) identified gaps in post-discharge care after surgery, which could impact negatively on clinical outcomes and the patient experience and their quality of life.
It is recommended that the Trust review information provided on discharge to support patients and community staff in recognising potential signs of wound infection and to enable easier access to the tissue viability service.
Consider implementing a quality improvement strategy such as ‘photo at discharge’
Andersson et al (2010) argued that a number of strategies can be used to support early diagnosis and treatment of SSIs. These can help improve the patient experience, by preventing or reducing the physical, social, psychological and financial effects that can occur as a result of an SSI.
One strategy implemented at Harefield Hospital is ‘photo at discharge’ (PaD). Rochon et al (2016) argued that the severity and duration of infection will be reduced if patients seek earlier advice and therefore receive earlier treatment. Providing a photograph of the patient's wound and some additional wound-care information at discharge gives the patient and any future healthcare provider, such as a community nurse or GP, a baseline to aid comparison. The information highlights key signs that could suggest there is a wound infection and provides advice on whom to contact. Rochon et al (2016) reported a significant reduction in readmissions for SSIs and associated costs.
Limitations
Although this study has identified some clinically interesting findings, it cannot contribute to meaningful statistical evidence relating to the significance of post-discharge SSIs at the hospital where the study was undertaken. It has demonstrated that a number of patients are presenting with wound infection after discharge, but what is unknown is what the actual numbers are or what proportion of SSIs they are likely to represent.
Conclusion
The aim of this study was to provide local insight into activity in TVN-led clinic in outpatients with the aim of improving the service that patients receive if they develop an SSI or wound infection. Retrospective data collection and analysis highlighted that patients who develop an SSI after discharge often ended up having a short inpatient stay. In addition, many patients and community nurses delayed seeking advice from a specialist TVN.
This study supports other research in demonstrating that many SSIs occur after hospital discharge. They can result in patients requiring additional treatment, including antibiotics, readmission to hospital and even additional surgery: this has consequences both for the patient, in terms of the impact on quality of life and experience, and to the NHS in terms of the financial costs. Downie (2010) questioned whether trusts can with confidence know what their actual SSI rate is because many patients will present after discharge from hospital and may therefore not be captured in reporting data (Tanner and Khan, 2008). Without proactive and accurate collection of SSI data, the true extent of the problem cannot be known, which could be limiting how resources are allocated to preventing SSIs in hospitals. Without complete surveillance data on infection rates (patient outcomes), poorer practice may not be identified and opportunities for making improvements in practice may be missed (Wilson, 2013b).
Several recommendations have been made. The Trust should consider the value of participating in some type of post-discharge surveillance. This could enable identification of patients experiencing wound-healing problems and instigate prompt assessment by the TVN and/or surgical team, supporting patients in accessing expert advice. Similarly, improving the quality of information provided to patients and community nurses, and developing closer working relationships with community trusts, could lead to prompt review, and to patients receiving earlier and appropriate treatment, which would potentially reduce the severity of infections and the number of readmissions.
Further research would need to assess the proportion of patients who develop an SSI after discharge. This would require prospective post-discharge surveillance applied in a continuous and consistent way.