Andrew Kingsley, RGN, CNS.
Infection Control and Tissue Viability Nurse.Wounds break the skin, which is the main non-specific defence mechanism to prevent invasion from micro-organisms and the potential for subsequent sepsis. While most wounds heal uneventfully, some become infected, causing pain, loss of function and delayed healing.
Wounds break the skin, which is the main non-specific defence mechanism to prevent invasion from micro-organisms and the potential for subsequent sepsis. While most wounds heal uneventfully, some become infected, causing pain, loss of function and delayed healing.
In this instance microbiological evaluation is warranted to guide and monitor intervention. Laboratory methodology was developed in the 19th century, but no universal sampling and processing protocol has been devised to date (Cooper and Lawrence, 1996).
Wound sampling is aimed at quantitative or qualitative analysis by examining wound tissue or wound fluid (Bowler et al, 2001) using mainly tissue, aspiration or swab culture (Stotts, 2000), with the latter predominant in clinical practice (Taddonio et al, 1990).
Most wound sampling in UK clinical practice is undertaken by the surface swab. There are three time sections involved in the microbiological process, all of which can influence the final report and contribute to the validity and reliability of the sampling method. These are:
- Sampling procedure
- Laboratory processing.
This paper discusses the sampling procedure of wound swabs. An audit was conducted to reflect the current knowledge of wound swab sampling among registered nurses in the community and hospitals of North Devon, and its findings are used to argue in favour of training and guidelines to ensure a higher priority for uniformity in sampling. Finally, we suggest protocols to facilitate this process.
In clinical practice wounds can be sampled for a variety of reasons (Box 1). In each of the five circumstances listed, subtle variation exists in the objective of sampling. In infection - the most common circumstance for taking a swab specimen - we consider that the objective is to identify the causative organism invading the tissues.
This enables its antibiotic sensitivities to be determined to check appropriateness of empirically initiated therapy. In healing delay, it is about characterising the whole polymicrobial population to help determine its influence on that indolence, and to guide antibiotic therapy if topical broad-spectrum antiseptics have failed to return the wound to normal healing progression.
Predictive information might guide a surgeon to use a delayed primary closure or to administer antibiotics as prophylaxis. The clinical objectives for sampling can be qualitative or quantitative, or both.
The limitations of swab sampling
Box 2 describes the limitations of swab sampling. These suggest that practitioners should not interpret a result of 'no growth' or 'no significant growth' as absence of infection, but rather a false negative, in a wound that has been swabbed because it displays the clinical signs of infection or critical colonisation.
These signs are outlined in Box 3.
Currently there is no consensus regarding the exact indications for swab sampling a wound, the technique of swabbing or preparation of the surfaces before sampling.
These are things that would enable a clearer description of the burden of wound infection in the UK to emerge (Lawrence and Ameen, 1998; Witherow, 2001). In the absence of a consensus method and formal teaching, specimen-taking consistency among nurses is likely to be poor (Donovan, 1998; Bowler et al, 2001). Despite this, swabbing remains the most common method of sampling wounds (Taddonio et al, 1990).
The tissue viability nurses at North Devon District Hospital became aware during their routine practice that patients admitted with wounds had them routinely swabbed without any sound rationale for this. This has time and cost implications for the laboratory and may encourage the inappropriate use of antibiotics.
A report of a similar swabbing practice questionnaire has recently been published (Bamberg and Sullivan, 2002). It recorded that clinical characteristics were the most common method for diagnosing infection, and 42% of those questioned indicated they routinely took swabs and biopsies.
The number of local registered nurses who completed and returned the North Devon questionnaire was 124. We distributed 190 forms, making the return rate 65%. However, of the 124, six returns did not specify which sub-group of nurses they were from, and were excluded from analysis.
The audit results
The overall findings demonstrated a wide variation in practice across all nurses and nursing sub-groups who answered the questionnaire (Figure 1). These findings show a need for practice to be standardised, indicating a role for the wound-swabbing protocols proposed later in this paper.
When to swab
Some nurses indicated that they would take a swab when there were signs of infection. Others agreed with this statement and also gave some examples of these signs, such as increased size, increased pain or inflammation. Respondents from all groups cited similar reasons for swabbing a wound.
A high proportion of acute nurses said they would take a swab at the request of a doctor. Not one community nurse gave this reason, indicating the potential autonomy of decision-making in this area of nursing practice.
The nurses were asked if they would choose to clean a wound before swabbing. The answer was fairly consistently 'no' from all groups, with 82% of acute nurses saying 'no', and 63% of community and 61% of practice nurses agreeing with that.
The nurses were asked: 'Is there a particular technique you would use when swabbing a wound?' Again, there was huge variation between and among the groups. Techniques cited included:
- Swab the whole wound
- Sweep from one side to other
- Swab from centre to outside
- Circular movement of swab to cover wound
- Swab infected area
- Start swabbing from the less infected end
- Swab the deepest part of the wound
- Swab the exudate
- Moisten (swab or wound not specified) with normal saline if dry
- Moisten with water if wound dry
- Moisten swab if dry (agent unspecified)
- Moisten with gel from swab (transport tube) if dry.
Completing the documentation
Details from the audit returns of the clinical information that nurses would record also varied. Clinical signs were recorded by nurses on 25% of returned questionnaires, indicating that specimen forms in practice would lack the relevant details for the microbiologists to interpret the results. The following terms suggested use of the clinical signs and symptoms of infection to indicate the necessity of swabbing:
- Infection risk
- Change in wound
- Wound appearance
- Clinical signs of infection
If taken together, clinical signs were recorded by nurses on only 25% of swab specimen forms, indicating a significant lack of relevant detail for the microbiologists to interpret the results.
Other specimen form details recorded in varying combinations were:
- Wound site
- Wound type
- Duration of wound
- Patient history
- Current medication/antibiotics
- Recent antibiotics
- Type of swab
- Reason for swab
- Date of swab
- Time of swab
- Previous swabs
- Investigations required.
Early resolution for infected wounds relies on early recognition and prompt initiation of antimicrobial strategies. Early resolution signifies the best possible outcome that can be achieved for the patient in terms of the least pain and wound deterioration resulting from the infection episode.
Antimicrobial strategies - either systemic antibiotics and/or topical antiseptics - need to be initiated on clinical signs and symptoms and not left until the receipt of positive specimen results, as this will cause unnecessary exacerbation of infection.
The use of topical and systemic antimicrobials remains the subject of debate - problems and benefits exist for both groups. Concerns for topical antiseptics have revolved around cytotoxicity to fibroblasts, keratinocytes and leucocytes (Gulliver, 1999) and the shutting down of small capillaries (Brennan and Leaper, 1985). Much of this concern stems from laboratory study rather than clinical observation, but there is evidence for benefit, and slow-release formulations are likely to be safer (White et al, 2001). Though systemic antibiotics are the choice for obvious spreading clinical infection from the wound (White et al, 2001), their potential limitations are in wounds with poor blood supply such as pressure and leg ulcers, as therapeutic doses may not be reached or increased time will be taken for them to be achieved in the wound bed (Robson, 1997; Miller, 2001).
Of those answering the question, 54% of acute nurses initiated or had prescribed antimicrobials only after a positive microbiological result. This was true for 44% of community hospital nurses, 32% of community nurses and 35% of practice nurses. Only 18% of acute nurses started antimicrobials before receiving swab results. Nurses therefore have an easy opportunity to improve the care delivered to patients with infected wounds if suitable education/guidance is provided.
Suggested protocol 1
Suggested protocol 1 (Table 1) is aimed at identifying the infecting organism only. This, by definition, requires an infected wound, that is one which has the clinical signs and symptoms of infection evident.
This method is derived from Levine et al (1976), Cuzzell (1993) and Stotts (1996) and samples are taken from a 1cm2 area of clean tissue.
Difficulties can arise when there are no clean (slough and necrosis-free) areas in the wound base from which to sample, necessitating removal of surface tissue/debris by sharp debridement. This may be beyond the patient's wishes and the practitioner's skill, or outside local sharp debridement protocol (Fairbairn et al, 2002).
If an exposed area exists it may not be anywhere near the 'leading edge', which is the area on the wound bed adjacent to either the cellulitic flare on the surrounding skin or the wound extension in deep or tracking wounds which sometimes deteriorate without cellulitis in the skin.
The small sampling area may not collect the organism/s actually doing the tissue invasion. But if it does, then it will not provide information on the collective species and quantities that may have created the conditions that allowed the infecting organism to penetrate the surrounding tissue. Such knowledge may be valuable in considering how to manage the wound from the topical approach to prevent recurrence of infection.
For the practical reasons outlined this sampling method - suggested protocol 1 (see Table 1) - may only be useful for research rather than routine clinical practice.
Suggested protocol 2
Suggested protocol 2 (Table 2) is aimed at identifying the collection of species in the wound. As it is a non-specific methodology it is likely to be less successful at capturing anaerobic organisms which require more specialised techniques. From the clinical standpoint this method requires the medical microbiologist to interpret the result in the light of the findings and the clinical details provided on the specimen form.
This method may also only capture certain free living microbes, representing only a proportion of the species in the wound. This is because if biofilms occur in open wounds then swabbing is unlikely to capture species from these protected environments. This method may also be prone to interference by topical agents and dressing materials that may be bacterio-static or bactericidal if they contaminate the swab during use. This problem is most likely to be linked to topical antiseptics that maintain a slow release or residual activity, but might be linked to other products, for example hydrogels, which contain antimicrobial preservatives. However, if these agents are in use, pre-cleaning with tap water or sterile normal saline should reduce this effect.
To all intents and purposes this suggested protocol two is likely to be the simplest method on which to standardise clinical specimen-taking practice in wound healing by secondary intention (that is, whether the wound is clinically infected or critically colonised). Despite its limitations, it provides useful information on which to base decisions on antimicrobial action, as wounds treated with systemic antibiotics consistent with swab results most commonly get better. Current knowledge of wound microbiology, though undoubtedly incomplete, is still relevant.
In most situations where a swab is taken, the wound is clinically infected (Box 3) and the microbiological result enables systemic antibiotics, already initiated empirically and to local formulary instructions, to be adjusted, depending on organism and sensitivity findings. By the time the result is received it is likely to be clear whether the antibiotics chosen at the outset are correct in type and dose because of the clinical response seen.
On the few occasions when a critically colonised wound (Box 3), commonly seen as either an indolent 'black-yellow' or a 'yellow' wound (Gray et al, 2002), is unresponsive to appropriate wound care, such as debridement and use of relevant topical antiseptic formulations, systemic antibiotics may be considered for initiation.
Suggested protocol 2 (Table 2) uses the method proposed by Cooper and Lawrence (1996), with some further rationale on the issue of pre-cleaning wounds before swabbing.
The audit results and previous discussion highlight a variety of issues that make the swab an unreliable and potentially invalid tool, depending on user knowledge and technique. However, it is simple, convenient and non-invasive, making it likely to have a high acceptability score with patients. It is also inexpensive and can be universally processed by receiving laboratories, unlike some other methods, so is likely to remain the tool of routine use in clinical practice, despite its imprecision (Cooper and Lawrence, 1996; Stotts, 1996; Bowler et al, 2001; Witherow, 2001).
Currently there is no consensus regarding the practical use of the swab or preparation of the surface before sampling, matters that would enable a clearer description of the burden of wound infection in the UK to emerge (Lawrence and Ameen, 1989; Witherow, 2001). This appears to be reflected in the lack of consistent clinical practice expressed through the audit. Education is clearly needed to improve swabbing practice, but more importantly, to achieve improved recognition and treatment for wound infection.
The protocols suggested above aim to address this problem. A set of standard protocols for sampling wounds and the subsequent storage, transportation and processing of swabs would enable easier comparison of study data and may facilitate greater understanding of the relationships between culture results and clinical presentations in due course.
- Do you have any comments to give to the authors about the protocols suggested? If so, please email us at email@example.com
- The authors would like to acknowledge the assistance of Charlotte Edmondson from North Devon Clinical Audit Department with the questionnaire design, production and analysis.
Bamberg, R., Sullivan, K. (2002)Diagnosis of wound infections: current culturing practices of US wound care professionals. Wounds 14: 9.
Bowler, P., Duerden, B., Armstrong, D. (2001)Wound microbiology and associated approaches to wound management. Clinical Microbiology Reviews 14: 2, 244-269.
Brennan, S., Leaper, D. (1985)The effect of antiseptics on the healing wound: a study using the rabbit ear chamber. British Journal of Surgery 72: 780-782.
Cooper, R. (2002)Wound microbiology: past, present and future. British Journal of Nursing 7: 12, Silver Supplement part 3 S4-S6.
Cooper, R., Lawrence, C. (1996)The isolation and identification of bacteria from wounds. Journal of Wound Care 5: 7, 335-340.
Cutting, K., Harding, K. (1994)Criteria for identifying wound infection. Journal of Wound Care 3: 4, 198-201.
Cuzzell, J. (1993)The right way to culture a wound. American Journal of Nursing May: 48-50.
Donovan, S. (1998)Wound infection and wound swabbing. Professional Nurse 13: 11, 757-759.
Fairbairn, K., Grier, J., Hunter, C., Preece, J. (2002)A sharp debridement procedure devised by specialist nurses. Journal of Wound Care 11: 10, 371-375.
Gray, D., White, R., Cooper, P. (2002)The wound healing continuum. British Journal of Community Nursing 7: 12, Silver Supplement part 3 S15-S19.
Gulliver, G. (1999)Arguments over iodine. Nursing Times 95: 27, 68-70.
Kingsley, A. (2002)Wound infection: colonisation vs. infection. Professional Nurse 18: 1, Explaining wounds card series 6.
Lawrence, C., Ameen, H. (1998)Swabs and other sampling techniques. Journal of Wound Care 7: 5, 232-233.
Levine, N., Lindberg, R., Mason, A., Pruitt, B. (1976)The quantitative swab culture and smear: a quick simple method for determining the number of viable aerobic bacteria on open wounds. Journal of Trauma 16: 2, 89-94.
Miller, M. (2001)Wound infection unravelled. Journal of Community Nursing 15: 3, 31-36.
Robson, M. (1997)Wound infection: a failure of wound healing caused by the imbalance of bacteria. Surgical Clinics of North America 77: 3, 637-650.
Stotts, N. (1996)Clinical controversy: swabbing a wound. Advances in Wound Care 9: 5, 18.
Stotts, N. (2000)Wound infection: diagnosis and management. In: Bryant, R. (ed.). Acute and Chronic Wounds: Nursing management (2nd edn). St Louis, Mo: Mosby.
Taddonio, T., Thomson, P., Smith, D., Pradad, J. (1990)A survey of wound monitoring and topical antimicrobial therapy practices in the treatment of burn injury. Journal of Burn Care Rehabilitation 11: 423-427.
White, R., Cooper, R., Kingsley, A. (2001)Wound colonization and infection. British Journal of Nursing 10: 9, 563-578.
Witherow, A. (2001)Detecting surgical wound infections. Professional Nurse 16: 10, 1413-1416.