Diagnosis of Wound Infections: Current Culturing Practices of U.S. Wound Care Professionals

Introduction In recent years, substantive basic science and clinical research have been conducted to evaluate the mechanisms of wound healing, the efficacy of various modalities for treatment of wounds, and the best methods for diagnosing wound infection. A great deal of this effort has been directed toward evaluating the most accurate and reproducible methods for diagnosing chronic wound infection. From the surface, this seems like a fairly simple clinical question. However, it quickly becomes apparent that this question is quite involved. For example, chronic wounds often harbor bacteria at levels many times that which constitute infection in an acute surgical wound (i.e., >= 105 microorganisms/gm of tissue); yet, many of these chronic wounds go on to closure despite levels of microorganisms as great as 108/gm of tissue.1–3 Chronic wounds are thought to be able to tolerate these high numbers of microorganisms because of the type and content of the microorganisms present in the wound bed. Chronic wounds often contain three or more microorganisms. These organisms include both gram-positive and -negative bacteria as well as both aerobic and anaerobic bacterial species. In fact, the cohabitation of these organisms and the resultant competition for nutrients and space is thought to decrease their virulence. As a result, higher numbers of microorganisms can be tolerated because of their decreased ability to harm tissue.1,4 Because of these intrinsic differences in the way acute and chronic wounds respond to varying levels of microorganism numbers,4 a greater emphasis is currently being placed on holistic assessment with clinical signs and symptoms playing key roles in diagnosis of chronic wound infection. The classical signs of infection (erythema, edema, heat, purulent exudate, and pain) have been modified to include serous exudate with concurrent inflammation, delayed healing, discoloration of granulation tissue, friable granulation tissue, pocketing at the base of the wound, foul odor, and wound breakdown as these signs appear to be more predictive of chronic wound infection. Increasing pain and wound breakdown have been shown to be particularly good predictors of infection in the chronic wound.5 However, there are several intrinsic limitations to diagnosing a wound infection and establishing a treatment paradigm via clinical signs and symptoms alone. Of particular concern is the constantly evolving number of microorganisms with antibiotic resistance. While the evaluation of clinical signs and symptoms may prove to be a very cost-effective and expedient method for diagnosing chronic wound infection, the use of this method alone does not inform the wound care clinician of the most appropriate chemotherapeutic approach to treatment. Use of clinical signs and symptoms alone leaves the provider to select a therapeutic agent based on little specific information about the particular pathogen(s). As a result, broad-spectrum chemotherapeutic agents may be initiated that only serve to facilitate the development of antibiotic resistance. The use of clinical signs and symptoms to predict the need for wound culturing has been suggested in the literature.5,6 However, this points to another controversy in the practice of diagnosing wound infection—which methods are best? The method used for collection of wound specimens can influence the data obtained from microbiological culturing. Currently, collection of a biopsy specimen is the gold standard for determining the presence and identity of microorganisms within the wound bed tissue. However, there are limitations as to which healthcare providers can collect biopsies, availability of laboratories offering microbiological culture testing on biopsies, the expenses involved with performance of these tests, and the potential for further tissue damage and delay of healing when biopsies are taken. Needle aspiration is generally considered the next best method for microbiological culturing of abscesses and closed wounds, followed by the more common swabbing techniques for open wounds. However, there has been no standardization as to the best swabbing technique to use nor the best wound preparation technique. Over the last 30 to 40 years, many studies have been conducted to assess the sensitivity, specificity, and accuracy of swabs, aspirates, and biopsy tissue specimens in diagnosing wound infections of various etiologies.4,7–15 The authors refer the readers to an excellent review of many of these studies written by Stotts.16 Compounding the controversies outlined above is the fact that there are a variety of professionals providing wound care, including physicians, nurses, physical therapists, occupational therapists, physician assistants, and nurse practitioners, and each may have his or her own theoretical framework for wound care. As a result, the definition of “best practices” of wound care continues to evolve. In order to fully come to a consensus of best practices for diagnosis of chronic wound infection, an investigation as to the current practice patterns utilized by practitioners has been undertaken. The primary purpose of this study was to survey members of the Association for the Advancement of Wound Care (AAWC), a multidisciplinary association of wound care professionals, in order to assess the current methods utilized for diagnosis of wound infection (e.g., clinical signs/symptoms, wound culturing, or a combination thereof). Additionally, the preferred methods for culturing wound infections along with common treatment approaches were also studied. Materials and Methods A six-page, 34-item questionnaire titled “Wound Culture Survey” was developed by the authors based on their cumulative experience in wound care, wound culturing, and clinical microbiology. The survey asked AAWC members to respond to questions about the following: 1) types of wounds seen in clinical practice; 2) whether wounds are cultured, when during wound care they are cultured, primary purpose of culturing, and final decision maker as to if and when wounds are cultured; 3) types of wound specimens collected and average number per wound; 4) preparation method for culturing; 5) factors that prompt a culture; 6) specific swab culture and biopsy techniques used; 7) specific lab tests ordered on wound specimens; 8) most common organisms isolated from wounds and type of growth report received; 9) routine approaches used in wound diagnosis; 10) most common treatment approaches used on infected wounds; and 11) demographic and professional information, including state of residence, profession, certifications related to wound care, specialty training and continuing education in wound care, proportion of practice time spent in wound care, practice setting, and payment systems of patients. The survey was reviewed by an expert panel of eight practicing wound care specialists from the AAWC membership. This panel included three registered nurses who hold doctoral degrees, two physicians, one podiatrist, and two physical therapists. The expert panel was chosen to include at least one professional from all professions surveyed among the AAWC membership. Slight modifications were made in the wording of three questions based on the input of the expert panel. Prior to dissemination of the survey, the survey instrument and research procedures were reviewed and approved by the University Policy and Review Committee for Human Research at East Carolina University. Completion and return of the survey was voluntary and constituted informed consent. All data were maintained and treated in a confidential and anonymous manner. Mailing labels for the AAWC membership were obtained from the organization office in September, 2000, and the final survey with explanatory letter and stamped return envelope were distributed to the 810 members who resided in the U.S. A second mailing was forwarded to nonrespondents in January, 2001. Data from all completed surveys received by May 1, 2001, were analyzed by frequency and, as appropriate, descriptive statistics, Chi-square, and T-test using the SPSS-PC+ version 9.0 software. Respondents who left some questions blank but answered the majority of items were included in the database along with those who answered the survey in its entirety. Results The response rate was 43 percent with a total of 345 returned surveys. Respondents who returned their survey prior to the second mailing were termed early respondents (n = 239) and those who responded after the second mailing were termed late respondents (n = 106). Demographic and professional profile of respondents. States were grouped by region using the National Center for Health Statistics’ classification system.17 The largest number of respondents were from the South (38.3%), followed by the West (25.7%), and an equal number of respondents were from the Midwest and Northeast (18.0% each). Relative to the individual states, the ones with the greatest number of respondents were California, Florida, and Texas with 51, 35, and 31, respectively. The professional background of respondents by geographic region and total sample is displayed in Table 1. By current profession, the largest number of respondents were registered nurses, followed by physical therapists, and thirdly followed by physicians. Over half of the respondents held one or more national certification(s) related to wound care (i.e., CWOCN/CETN, CWS, or CNSWC), had completed a formal program of training in wound care (e.g., workshops, institutes, credit courses), and reported attending at least one continuing education workshop on wound care per year. The respondents had a considerable experience base in wound care (mean = 10.9 years) with a mean of 64.1 percent of their professional work time being spent in wound care. The professional profile was similar across the regions. Over half of the respondents reported working in more than one setting where they performed wound care. The most common work setting was acute care hospital, followed by outpatient wound care center where respondents worked either full or part time. Most physicians and podiatrists reported private practice as a work setting, though a few nurse practitioners also reported this setting. A complete listing of work settings in which wound care was provided is displayed in Table 2. When asked about the financial coverage of their patients, 92.8 percent reported providing wound care for Medicare patients, 81.7 percent Medicaid, 85.2 percent private health insurance, 64.9 percent for patients in a health maintenance organization (HMO) plan, 62.3 percent self-pay, and 2.3 percent for military personnel. Wound care and culturing practices. The wound care professionals were asked to rate how often they cared for specific types of wounds. Over half the respondents indicated they care for the following types of wounds either frequently or very frequently: pressure, arterial insufficiency, venous insufficiency, and surgical. Results for all types of wounds reported are displayed in Table 3. The vast majority of the respondents (306, 88.7%) reported that they do culture at least some wounds. Data about reasons for culturing, when wounds are cultured, and decision making as to whether to culture are displayed in Table 4. The respondents were also asked what percentage of wounds they treat without ever culturing with a mean of 69.7 percent reported. When asked what percentage of wounds they culture that they feel are infected based on clinical appearance and symptoms, a mean of 59.9 percent was reported. The respondents reported that of the wounds they do culture, a mean of 78.9 percent have a positive lab culture report. When asked what patient data is used in making a wound diagnosis, 98.3 percent indicated clinical characteristics (e.g., odor, color, visual appearance of the wound), 87.5 percent also listed patient-reported symptoms, and 7.8 percent listed Doppler results; only 70.1 percent also factored in available laboratory results of wound cultures into their diagnostic processes. The respondents provided detailed descriptions as to the techniques used for culturing and lab tests ordered on wound cultures (Tables 5 and 6). Swab is the more frequent type of specimen collected for a wound culture as compared to biopsy. For those who collected both swabs and biopsies (n = 128), wound factors that were reported by at least 50 percent of these respondents as always or almost always used in determining whether they would do a swab, a biopsy, or both, included number of previous failed treatments, length of time wounds have been treated, and wound drainage. Wound depth and odor were listed as factors used sometimes by at least 50 percent of the 128. Though multiple lab tests were routinely ordered on both swab and biopsy specimens, biopsies overall appear to have more microbiological tests ordered. For positive wound culture reports on both swab and biopsy specimens, over 50 percent of respondents indicated they routinely received semi-quantitative estimates for the isolated pathogen(s) (e.g., 1+ to 4+ or few, moderate, many) as opposed to either quantitative (i.e., number of colony forming units) or no estimate of growth, which were provided less frequently. The three most common organisms reported on positive wound culture reports were also indicated by 263 respondents (Table 7). As may be predicted, Pseudomonas species and Staphylococcus aureus (S. aureus) are the most common pathogens reported for positive wound cultures. Respondents were asked to indicate the most frequently used treatment approaches they utilized for infected wounds (Table 8). Systemic antimicrobials, debridement, and topical antimicrobials were clearly the most commonly used treatments for wounds with positive culture reports. Slightly over half of the respondents (51.2%) indicated that their treatment approach would be different if methicillin-resistant Staphylococcus aureus (MRSA) was reported on a wound culture. When asked to indicate how the treatment would change, the most common answers were more stringent contact isolation, more strict adherence to sterile precautions, more patient and caregiver education as to sterile precautions, more education of patient and his or her family on how to prevent cross-contamination, more aggressive treatment (i.e., stronger and longer), use of intravenous medications and topical applications as opposed to only oral medications, more frequent and aggressive debridement, use of antibiotic ointment, more frequent dressing changes, nonocclusive dressings, Dakin’s solution, and consultation with an infectious disease physician. Differences by demographic and professional characteristics. Results were also examined for statistical significance by comparing the responses by professional designation, by whether they held a national certification in wound care (yes or no), by whether they had received wound care training in a formalized program (yes or no), by whether they worked in an outpatient center devoted exclusively to wound care (yes or no), and by geographic region. For professional designation, several comparison analyses were run as follows: physicians and podiatrists versus registered nurses and nurse practitioners; physicians and podiatrists versus physical therapists; registered nurses versus physical therapists; and individual professional categories. Independent sample T-tests were performed on all numeric data variables, and Chi-square tests were performed on all nominal data variables for these group comparisons. Only statistically significant results at p < .05 are reported for these comparisons in Tables 9 through 13. Those with a national certification and/or formalized specialty training in wound care reported spending a significantly higher mean percentage of their work time in wound care as compared to those without these characteristics (Table 9). There were several significant differences between respondents who worked either partially or solely in an outpatient center devoted to wound care as compared to those who did not work in such a setting (Table 10), including greater utilization of culture results in wound diagnosis, culturing wounds more frequently, and allowing non-physician professionals greater decision-making authority in wound culturing. When comparing respondents by their current profession, several significant differences were found as displayed in Tables 11 and 12. Physicians and podiatrists were less likely to treat a wound without ever culturing, more likely to culture wounds that they assessed as infected based on clinical symptoms, and less likely to never culture a wound, as compared to respondents from all other professions. These results may be related to the fact that 71.6 percent of respondents indicated that the physician or podiatrist is the professional who makes the final decision as to whether a wound will be cultured as opposed to a consistent difference in wound diagnosis and management approaches between physicians and nurses. Some differences in culturing technique were also found. Physicians and podiatrists were less likely to collect only swab specimens or to routinely collect swab specimens by a 10-point diagonal method than physician extenders, nurses, and physical therapists. Physicians and podiatrists, though, were more likely than the other professional groups to routinely collect biopsy specimens by scalpel. If MRSA was isolated from a wound culture, the physicians, podiatrists, and physician extenders reported being more likely to treat the wound differently. Relative to regional differences in wound culturing (Table 13), respondents in the Northeast treated abscess wounds less often than in the other three geographic regions. Respondents in the South and West routinely ordered fungal cultures on biopsy specimens from wounds, presumably due to the warmer climate in the South and West. To determine if there were any differences in response (i.e., bias) by the early respondents versus the late respondents, an independent sample T-test was conducted on the seven numeric data variables and no statistically significant differences were found at p < .05. Chi-square analyses were performed on 10 nominal data variables comparing early and late respondents, and no statistically significant associations were found at p < 0.05. Discussion The findings in this study indicate that wound care clinicians are relying heavily on clinical characteristics for the diagnosis of wound infection. They used these findings 98 percent of the time followed by use of patient-reported symptoms (88%) and wound culturing (70%). This finding reflects current thinking in wound management education concerning chronic wound infection and suggests that advanced wound care professionals are aware of recent research findings and evolving recommendations for wound management. An increased awareness of current research and practice recommendations is not unexpected as most of the respondents have a national certification in wound management (65%), have attended a formal wound care training program (75%), have attended at least one continuing education workshop in the past year (97%), and are members of AAWC (100%). The utility of clinical signs in the diagnosis of chronic wound infection (most of the wound types seen by those surveyed) is supported by the surveyed wound care clinician’s experience that 79 percent of the wounds with positive clinical signs for infection also have positive cultures. This finding supports the practice of using physical signs as a quick screen to determine which wounds should be cultured. However, as many as 21 percent of wound patients with potential wound infections may go undiagnosed if clinical signs and symptoms alone are utilized in diagnosis. For example, immunocompromised patients may not exhibit typical signs and symptoms because their immune system may not respond normally. On the other hand, there are situations in which a patient may exhibit positive signs and symptoms of an infection but may not be truly infected. These include patients with unrelieved pressure, allergies to dressing components, or those who have chronic inflammation.18 Additionally, the finding that a majority of the clinicians surveyed relied heavily on clinical signs for wound infection diagnosis yet prescribed systemic antibiotics for putative infections is alarming. Current evidence from the microbiological literature and preferred antibiotic prescribing practice advocated by the Centers for Disease Control and Prevention warns against the indiscriminate use of antibiotics, especially those with a broad spectrum of activity because of the increased potential to generate antibiotic resistance.19–21 These findings point to the need to follow up clinical signs and symptoms assessment with culturing if antibiotics are to be prescribed. This approach may be less critical if the newer broad-spectrum antiseptics (time-released silver and iodine preparations), debridement, or physical agents are to be prescribed as these agents are not typically associated with the development of pathogen resistance. Interestingly, the data indicate that the practice of wound culturing is varied and lacks a consistent rationale for use. Many respondents reported they culture varyingly before or after treatment depending on the wound situation. Additionally, 20 percent of the surveyed wound practitioners who culture at least some wounds do so only after a treatment regimen has failed. Substantial delays in specimen collection after antibiotic therapy has been initiated can result in incorrect culture results. Antibiotic-treated pathogens may either fail to grow in culture or may exhibit atypical cultural characteristics, making them difficult to identify. Of the wound care professionals surveyed, physicians appeared to be the group with the highest likelihood of culturing wounds. Overall, they were about 23 percent less likely than nurses to prescribe a treatment for a putatively infected wound based on physical symptoms alone without a confirming culture. They more often ordered wound cultures prior to initiating treatment. While the reason for this difference is unknown, one might speculate that physicians may not be involved with day-to-day care of many individuals with wounds and may be consulted only on the most difficult cases. This is supported by survey results indicating that physicians had the lowest reported mean for percentage of work time spent performing direct wound care services (44%). Additionally, infection control protocols as well as physician education on appropriate antibiotic usage and potential for generation of antibiotic resistance may influence this practice of utilizing culture results to guide therapy. Controversy certainly remains as to the best culturing method for diagnosing wound infection. While the biopsy remains the gold standard for diagnosing wound infection, the results of this survey indicate that among well-educated and experienced wound care clinicians, the swab culture is the method most utilized. Approximately 54 percent of the respondents reported utilizing the swab only, followed by 42 percent that used both swab and biopsy methods for specimen collection. The biopsy method was used exclusively by less than five percent of those surveyed. These findings are consistent with the overall composition of the respondents, as over 63 percent of the individuals that responded to this survey were registered nurses or physical therapists, and these professionals do not typically perform biopsies. Concerning the method of taking a swab culture, it was interesting to note that 82 percent followed current practice standards (Agency for Healthcare Research and Quality) and cleansed the wound with saline prior to taking the sample. However, 10 percent of the respondents reported sampling from an unclean wound, which is a known confounder in overestimating microbial numbers. Also in relation to methodology, the 10-point diagonal was the most common method used for swab culturing (57% reported use of this method) followed by the one-point–rotation method reported by 38 percent of respondents. This survey was directed at advanced wound care practitioners in order to determine whether there was consistency in the U.S. as to the approach and methods that are being used in diagnosing chronic wound infection. Therefore, these results may not be reflective of wound practices in settings without advanced wound care professionals, such as in a family practice setting where a physician and nurse may treat wounds sporadically. It would be of interest to determine the practices utilized in these healthcare settings as well as those used by novice or non-certified healthcare professionals in wound care. The survey results did not reveal an absolute consensus in the approaches and methods used by the advanced wound care practitioners in diagnosing wounds through culturing. Though the results may preliminarily indicate that physicians and podiatrists are utilizing cultures in wound care more frequently than nurses who appear to rely more heavily on clinical symptoms, the results are too limited to say that there is a clear difference evolving in the approaches used by these professional groups or that one approach is better from a clinical efficacy view. The survey was not designed to answer such questions, but simply to gather information on what is being done in wound care practice by the experts. Research involving more healthcare settings and a wider spectrum of professionals, expert and novice, who treat wounds, will need to be conducted to address the issues of best practices. Conclusions The results of this nation-wide survey indicate that culturing of wounds may be a source of clinical information that is under utilized in healthcare. The majority of wounds (70%) are reported to be treated without ever culturing, which in itself, compromises the utility of results from post-treatment cultures. Only 12 percent of the respondents reported they routinely culture wounds before treatment regimens are begun and only 20 percent even culture when a treatment regimen has failed. Of the wounds that the respondents assess as being infected based on clinical signs and symptoms, only 60 percent of these are reported to be cultured. The respondents estimated that 79 percent of the ones they do culture are found to be infected based on a positive laboratory report, thus reinforcing the need for a higher frequency level of culturing wounds as balanced with cost-containment issues. The organisms most often isolated from infected wounds were reported by the respondents as Pseudomonas, Staphylococcus, and Streptococcus species as well as MRSA. Over half of the respondents indicated their treatment approach would be different if MRSA was isolated from a wound, again reinforcing the need for a higher level of routine culturing of wounds, as 31.6 percent of respondents listed MRSA as one of the three most common organisms isolated from infected wounds. The antibiotic sensitivity results from positive culture reports can be clinically useful information for the clinician, which is unavailable if a wound culture is not performed. Though the fact that these results are self reported and not validated by medical chart review constitutes a study limitation, this research does represent the collective reporting of a group of highly experienced and credentialed wound care professionals. The findings of this study indicate a need to educate professionals of the merits of using a combination approach of screening for wound infection by assessing for clinical signs and symptoms and then following with a culture for use in making a definitive diagnosis of wound infection and for establishing a treatment plan. In support of the need for such education for particularly physicians and nurses is a recent study that found physical therapists in the West reported that 66 percent of wound evaluations were done by nonphysical therapists, particularly by nurses or physicians.22 As respondents who worked full-time or part-time in outpatient centers devoted to wound care reported higher rates of wound culturing, increased development of such centers may aid in improving the amount of wound care guided by microbiological laboratory results in conjunction with patient symptoms and wound appearance.