Does Fine Needle Aspiration Microbiology Offer Any Benefit Over Wound Swab in Detecting the Causative Organisms in Surgical Site Infections?

Objective. The objective of this study is to determine the role of fine needle aspiration microbiology (FNAM) in detecting the causative organisms of postoperative surgical site infections (SSIs) in comparison with the standard technique of surface swabbing. Materials and Methods. In this study, 150 patients with SSIs following elective and emergency operations were included. In all patients, FNAM was performed along with conventional surface swabbing to identify the causative microorganism. Sensitivity of surface swab and FNAM was calculated as the number of samples collected from the diagnosed case of SSI. Results. A total of 115 positive cultures were obtained from the 150 patients with SSIs; surface swab was positive in 110 cases and FNAM was positive in 94 cases. The mean number of organisms isolated by surface swab, and FNAM was 0.95 and 0.8, respectively. The sensitivity of surface swab was 94.3% in elective cases and 96.25% in emergency cases. The sensitivity of FNAM was 82.8% in elective cases and 82.5% in emergency cases. The sensitivity and negative predictive value of FNAM and surface swab did not significantly differ in clean elective cases. Conclusion. The overall sensitivity of surface swab and FNAM was 95.65% and 81.7%, respectively. Comparing the antibiotic susceptibility pattern, no difference was observed when the same organism was isolated by both methods, indicating that FNAM does not offer benefit over the conventional wound surface swab in detecting microorganisms in SSI in both elective and emergency surgeries. In certain cases with unexplained wound infections, FNAM can be used as an investigation to identify specific pathogens not detected by conventional surface swab.

Infection in the postoperative wound is a common problem in surgical patients leading to significant morbidity and mortality in the form of delayed wound healing, prolonged hospital stays, and an additional 10% to 20% in hospital costs.¹ Early detection and appropriate treatment of surgical site infections (SSIs) can have significant benefits in terms of patient comfort and medical resources used.²

Studies have shown wound infection is determined by a number of microbial-host interactions, particularly the number of bacteria, the virulence and pathogenicity of the organisms, and the immune capacity of the host.^3,4 A number of wound sampling techniques are available to obtain microbiological data from infected wounds. The important types are wound tissue biopsy, surface swabbing, and fine needle aspiration microbiology (FNAM).

Wound tissue biopsy is considered the gold standard for studying the microbiological profile of raw areas and other wounds. The bacterial load and individual causative pathogen can be identified by this method. However, use is restricted to acute situations where surgical debridement is required. It is not used in other conditions due to the technical difficulty and potential patient trauma, especially in those with postoperative wound infections.⁵

Being noninvasive, surface swabbing is the most widely used wound sampling technique, with limited use in cases of cellulitis and early abscesses where the skin is usually intact and surface swabs do not access the infected locus.⁶ Yet, in cases of open wounds or ulcers, the surface cultures are usually polymicrobial with the presence of skin commensals, whose pathogenic roles are uncertain.

Fine needle aspiration microbiology is a relatively newer technique considered an alternative for deeper wound biopsy. The technique aims to acquire wound tissue fluid from deeper parts of the wound through an area of normal skin. It reflects the pathogenic bacteria present in the deeper tissue planes, bypassing the superficial skin flora. The role of FNAM in identifying the microbiological profile of cellulitis and soft tissue infection has been well established; however, its role in SSIs is not clear.^7-12

To the best of the authors’ knowledge, there is only 1 study in the literature assessing the role of FNAM in SSIs that analyzed a small number (20) of patients to establish its advantage over swab.¹³ Thus, the study presented herein was conducted to compare FNAM with the standard technique of wound surface swabbing for the detection of causative organisms of postoperative SSIs.

The study was conducted in the Department of Surgery at Jawaharlal Institute of Postgraduate Medical Education & Research from August 2012 to July 2014. This was a prospective analytical study comparing 2 techniques for detecting the causative organism in patients who developed an SSI. In all patients, FNAM was performed along with the standard surface swab technique.

The sample size was calculated using OpenEpi software (Atlanta, GA). Based on a study by Parikh et al,¹³ sensitivity (detecting the presence of causative microorganisms) was considered the primary variable, with a 0.9 standard deviation of this parameter; with power of the study as 90%, the sample size was calculated to be 150.

All patients in the surgical wards with postoperative SSIs following both elective and emergency surgeries were included in the study. Patients who could not unundergo primary wound closure, who underwent delayed primary suturing, and who received antimicrobial therapy prior to hospital admission also were excluded.

Patients diagnosed to have a SSI in the surgical wards using the Centers for Disease Control and Prevention (CDC) definitions were assessed for eligibility criteria.¹⁴ The presence of a SSI was clinically established by the classical 5 signs and symptoms of infection: pain, purulent discharge, erythema, edema, and local heat. Patient demographic profile, including the preoperative diagnosis, operative procedure performed, and grade of SSI according to CDC criteria, was noted. On the first occurrence of SSI symptoms, the 2 techniques (ie, surface swab and FNAM) were used to collect specimen to assess the microbiological profile of postoperative infected wounds.

Surface swab
Surface swab was taken from infected wounds using sterile cotton swabs. The swab was placed over a wound and rotated to collect the specimen. The collected specimen on the swab was transported immediately to the microbiology laboratory in the same institute in sterile containers, where plating of the specimen was done on 5% sheep blood agar and MacConkey agar for aerobic culture. The organisms grown on the culture and their antibiotic sensitivity pattern were duly reported.

Fine needle aspiration microbiology
About 5 cm² of skin around the infected wound was disinfected using 5% chlorhexidine solution; the area was allowed to dry for 60 seconds. A 21-gauge hypodermic needle connected to a 10-cc syringe was used for aspiration. The needle was inserted through the intact skin 1 cm from the edge of the wound. It was advanced tangentially 2 cm into the skin to reach the base of the wound. Using the syringe, aspiration was performed with negative pressure maintained for at least 10 seconds and the needle moving back and forth at different angles. Aspiration was not the same in all samples. The needle was then withdrawn and the aspirate was transferred onto a standardized sterile container. The specimen was transported immediately to the microbiology laboratory for aerobic culture. Local anesthetics were not used in this technique as it has been reported that they can have antimicrobial effects.¹⁵

Outcome measures
The culture and sensitivity information reported by the microbiology laboratory was analyzed. Wound infections were considered to be positive for pathogenic organisms if at least 1 of the wound sampling techniques (FNAM or surface swab) grew positive cultures for pathogenic organisms. Isolation of pathogenic organisms in a culture was considered a positive culture, whereas absence of growth or the isolation of only skin commensals in culture was considered negative.

Sensitivity of surface swab and FNAM in both elective and emergency cases was calculated as the number of samples collected from diagnosed cases of SSI that grew pathogenic organisms in each technique. Specificity was not determined, as detection of the organism was the primary aim rather than exclusion of it, and the positivity in either of the tests was considered the gold standard.

The Institutional Review Board and the Institute Ethics Committee approved this study, and all provisions of the Declaration of Helsinki and Good Clinical Practice were followed. The nature, methodology, and risks involved in the study were explained to the patient, and written informed consent was obtained. All information collected was kept confidential, and patients were given full freedom to withdraw at any point during the study.

Statistical analysis
Statistical analysis was calculated by using SPSS 19.0 for Windows (IBM Corp, Armonk, NY). The following variables were analyzed to identify factors associated with postoperative wound infections: gender, disease, surgical procedure, type of surgery, postoperative day wound infection was noted, presentation of wound infection, and bacterial pathogen grown.

Chi-square test was used for comparison of Enterococcus faecalis positive and E faecalis negative cultures obtained from surface swab and FNAM in emergency cases; P < .05 was considered statistically significant.

A total of 150 patients with SSIs were analyzed over a 2-year period, including 98 (65.3%) SSIs following emergency surgeries and 52 (34.7%) following elective procedures. Abdominal surgeries were the most common operative procedure (73.3%) followed by abdominal wall hernias (14%) and breast surgeries (11.3%) (2 cases of SSIs in varicose veins postop were not included in this study). While comparing the presentation of SSI, 59 (39.3%) patients had serous discharge, 42 (28%) with purulent discharge, 48 (32%) had erythema with induration of the wound and 1 patient (0.7%) had an open wound. Wound infection in the form of serous discharge is the most common presentation in both elective (44.2%) and emergency (36.7%) surgeries followed by presentation as redness (40.3%) in elective cases and as purulent discharge (35.7%) in emergency cases (Figure 1). Mean days before postoperative wound infection and subsequent sampling was 5.15. 

All 17 breast surgeries included in the study were elective modified radical mastectomies, of which 64.7% of cases had redness as SSI presentation. Mean days of postoperative wound infection and subsequent sampling was 5.94. No growth was observed in 10 patients in both surface swab and FNAM groups. Pseudomonas spp was the most common organism identified in 3 patients by both methods.

Out of 21 abdominal wall hernia surgeries, 8 were elective and 13 were emergency surgeries. The most common SSI presentation was in the form of serous discharge in 12 (57.1%) patients. Mean days of postoperative wound infection and subsequent sampling was 4.9 in hernia surgeries. No growth was observed in 10 of these patients in both surface swab and FNAM groups. The common organism identified was Escherichia coli in 5 patients by surface swab and 6 by FNAM.

Majority of the wound infections following emergency surgeries were noted, and samples were taken on postoperative days 5 (36%) and 6 (23.3%). Appendicitis (20%) and perforation peritonitis (18.7%) were common in patients with SSIs following emergency surgery.

A total of 115 positive cultures were obtained from the 150 patients with a clinical SSI. Surface swab was positive in 110/115 cases, and FNAM was positive in 94 cases. Of the 150 patients, 35 had negative cultures. The negative cultures included 29 patients without any growth in the culture and 6 patients with only skin commensals grown in the culture.

In the surface swab group, 116 out of 150 samples had grown microorganisms. Skin commensals only grew in 6 of these, resulting in 110 cultures positive for pathogenic bacteria. In the FNAM group, 94 positive cultures were obtained, and no culture had grown skin commensals alone. E coli, grown in 71/150 patients (47.3%), was the most common organism identified in both elective and emergency cases. E coli was cultured in 64 surface swab specimens and 56 FNAM specimens (eTable 1). 

The yield of surface swabs was 66% in elective cases and 78.6% in emergency surgeries with an overall yield of 73.3%, whereas for FNAM the yield was 55.8% and 67.3% in elective and emergency surgeries, respectively, with an overall yield of 62.7%. The mean number of organisms isolated by surface swab and FNAM in this study was 0.95 and 0.8, respectively. The sensitivity in detecting pathogenic organisms of surface swabs is 94.3% in elective cases and 96.25% in emergency cases (eTable 2) and of FNAM is 82.8% in elective cases and 82.5% in emergency cases (eTable 3). The overall sensitivity of surface swabbing is 95.65% and is 81.7% for FNAM (eTable 4).

Subset analysis of the clean elective surgeries (27 cases) showed the sensitivity (84.6%) and negative predictive value (NPV) (87.5%) of surface swabs did not differ significantly with the sensitivity (76.9%) and NPV (82.3%) of FNAM (eTable 4). The number of positive cases for individual bacteria was almost the same in the surface swab and FNAM groups in both elective and emergency surgeries, except for E faecalis, which was grown in 10 FNAM samples of emergency surgery cases compared with only 3 positive surface swab cultures (eTable 5). While comparing the antibiotic susceptibility pattern, no difference was observed when the same organism was isolated by both methods. Of the 12 cultures that were positive for E faecalis, all were sensitive to vancomycin, and organisms in 4 of these cultures were resistant to all the other drugs including high-level gentamicin. All methicillin-resistant Staphylococcus aureus (MRSA) and E faecalis isolated were sensitive to vancomycin. 

The wound sampling techniques available to obtain the microbiological profile of infected wounds are mainly wound tissue biopsy, surface swabbing, and FNAM. Wound tissue biopsy is considered the gold standard in obtaining the causative bacterial pathogens in raw areas; however; it is potentially traumatic to patients and is technically difficult in wound infections with an intact skin and hence only suitable in acute wounds where debridement is required.⁵ The most widely used method of wound sampling is surface swabbing, and it has the advantages of being convenient and noninvasive. But surface swabbing has been challenged on the basis that superficial microbiology does not reflect deeper tissue and that it does not identify pathogenic bacteria in deeper tissues.⁶

Fine needle aspiration microbiology obtains tissue fluid from deeper parts of a wound through an area of normal skin. Hence, it is expected to obtain pathogenic bacteria from deeper tissues by bypassing superficial surface colonizers and is therefore considered equivalent to tissue biopsy without being significantly traumatic.¹² The role of FNAM in postoperative wounds and its sensitivity and predictive value in SSIs has not been clarified. Parikh et al¹³ assessed the role of FNAM in detecting the causative organism in the postoperative wound infections of 20 patients who underwent clean surgeries and reported a higher yield of FNAM for positive culture. Evidence to substantiate the role of FNAM in SSIs are sparse in the literature. The present study aimed at comparing FNAM to the commonly used surface swab technique for detecting causative organisms in postoperative wound infections. Postoperative wound infections were noted as per the criteria of CDC definition for SSIs.¹⁴ All the cases included in this study are superficial SSIs.

Both surface swabbing and FNAM were performed on each patient one after the other when wound infection was noted. Surface swabbing was done by gently rolling a cotton swab over the infected wound and FNAM by using a 10-cc syringe connected to a 21-gauge hypodermic needle by the technique described by Parikh et al.¹³ Local anesthetics were not used in this technique for pain control, as local anesthetics were shown to have antimicrobial property and could lead to false negative culture reports.¹⁵ Lee et al¹² followed the same technique for aspiration biopsy using a 22-gauge needle.

In the present study, E coli was grown either alone or with other organisms in 71 (47.3%) of the 150 patients. E coli was the most common organism identified by both surface swab and FNAM in patients who underwent abdominal surgeries and abdominal wall hernia surgeries (Figure 2). The common organism isolated among those who underwent modified radical mastectomy was Pseudomonas spp. In other studies, the most common organism isolated was S aureus in both surface swab and FNAM groups.^11,12,16,17 In the study by Brook and Frazier¹⁸ regarding SSIs following spine surgeries, E coli was the most common organism recovered from the wound site. 

The high incidence of wound infections due to E coli herein can be attributed to the fact that the majority of the SSIs followed abdominal surgeries. The source of wound infection following clean surgeries is likely due to contamination from exogenous environment or the patient’s skin flora, and these organisms are expected to be recovered from the wound surface including S aureus. In contaminated and dirty surgeries in emergency cases, the source is most likely the endogenous microflora of the intestine or the organ resected, more commonly E coli.¹⁹ High wound infection by E coli could also be due to the fact that most of these patients were given cephalosporins, which are more active against S aureus when compared with E coli.²⁰

In the present study, E faecalis was detected in 12 of 150 patients (8%). This is comparable with the results obtained in other studies; in the study by Giacometti et al,²¹ 5.6% of wound infections were caused by E faecalis, whereas it caused 11.3% of SSIs following liver transplantation in the study done by Asensio et al.²² Fine needle aspiration microbiology detected E faecalis in 10 cases of SSIs post emergency surgeries, which was detected in only 3 of the surface swab samples. Wound surface swab is less effective in detecting the presence of E faecalis in infected wounds following emergency surgeries when compared with FNAM.

The investigators herein found that surface swab was positive in 110 cases, providing a yield of 73.3%, whereas in FNAM the yield was only 62.7% with 94 positive cases. Parikh et al¹³ reported that surface swab and FNAM were positive in 30% and 80%, respectively, of SSIs. The cases included in the present study comprised both elective and emergency surgeries with wound class ranging from clean surgical wound to dirty surgical wound. All SSIs included in the Parikh et al study¹³ were following clean surgeries. Herein, the subset of SSIs that occurred in patients with clean surgical wounds showed positive FNAM in 37.04% and surface swab positivity in 40%, indicating there is no beneficial role of FNAM in SSIs following clean surgeries.

The mean number of organisms isolated by surface swab and FNAM was 0.95 and 0.8, respectively. The mean number of organisms isolated by Huovinen et al¹⁶ for surface swab and FNAM was 2.9 and 1.3, respectively. Lee et al¹² (2.4 for swab vs. 1.4 for FNAM) and Kessler et al¹⁷ (2.04 vs. 1.09) also isolated a similar number of organisms to Huovinen et al.¹⁶ These studies were performed on infected ulcers and raw areas, hence the expected higher number of organisms in the surface swab group.

In the present study, the sensitivity of surface swabbing in detecting pathogenic organisms is 94.3% in elective cases and 96.25% in emergency cases, whereas the sensitivity of FNAM was 82.8% in elective cases and 82.5% in emergency cases. Chakraborti et al²³ found the sensitivity of superficial swabs was 49% in lower extremity wounds. The overall sensitivity of surface swabs in the study herein was 95.65% with a NPV of 87.5% and for FNAM the sensitivity was 81.7% with 62.5% NPV.  The NPV was calculated with respect to the gold standard confirmatory test, not the clinical diagnosis of SSI. The sensitivity of surface swabbing was more than FNAM in detecting SSI microorganisms in both elective and emergency surgeries. The result indicates that FNAM does not provide any benefit over surface swabbing in terms of the number of positive cultures, sensitivity, and NPV.  This is contradictory to the results found by Parikh et al,¹³ that found FNAM to be more sensitive than surface swabbing; their study included only 20 patients and all were clean surgeries.

Specificity was not determined as detection of the organism was the primary aim rather than exclusion, and the positivity in either test was used as the gold standard. The choice to explore only sensitivity was also driven by the evidence-based recommendation to culture wounds only in the presence of signs of infection. A wound surface swab or FNAM is not performed in patients without SSI to prove absence of organisms.^24,25

Majority of SSIs occurring on clean surgical wounds are caused by skin surface microflora and clinically present as only inflamed surgical wounds without apparent discharge. Surface swabbing in such cases will be difficult and may provide polymicrobial culture that poses the risk of selecting multiple antibiotics, which then increases the risk of antibiotic resistance development. Fine needle aspiration microbiology could be advantageous in the clinical situation of infection without discharge or an open wound, and it may provide monomicrobial culture, hence selecting the appropriate antibiotic can be unchallenging. Considering these facts, it is apparent that the results shown by Parikh et al¹³ cannot be extrapolated for all SSIs where FNAM may not have advantage over the surface swab technique.

In the present study, the investigators found no difference in antibiotic susceptibility when the same organism was isolated by both methods. Resistance to ampicillin and high-level gentamicin was noted in 33.3% of cultured E faecalis strains. In the study by Johnston et al,²⁶ none of the isolated Enterococcus strains were resistant to tetracycline, gentamicin, or vancomycin. The isolation of 33.3% of cultures resistant to high-level gentamicin was similar to that seen in studies by Mathur et al²⁷ and Vergis et al.²⁸ In the study by Mathur et al,²⁷ 26% of E faecalis cultured was resistant to high-level aminoglycosides; in a multicenter study by Vergis et al,²⁸ 37% of isolated organisms were resistant. The only available alternative treatment is vancomycin, and there is rampant use of vancomycin in hospitals when Enterococcus and MRSA are cultured from samples. However, this poses an important risk factor for the development of vancomycin-resistant enterococci in the community.²⁷

This study has its own limitations. The technique of FNAM and surface swabbing were not compared with the gold standard technique of wound biopsy; rather, positivity in either of the tests was taken as the gold standard. Hence, specificity and positive predictive values were not used as a measure to compare the 2 diagnostic modalities.

The overall sensitivity of surface swabbing was 95.65% with 87.5% NPV; for FNAM, sensitivity was 81.7% with a NPV of 62.5%.  This shows surface swabbing is a more sensitive method than FNAM for detecting microorganisms in postoperative SSIs. The sensitivity of these methods did not differ significantly between elective and emergency cases. The overall yield of the surface swab technique was 73.3%, whereas the yield of FNAM was only 62.7%, indicating that FNAM does not offer a significant advantage over the conventional wound surface swab technique for identifying and isolating the causative microorganism in SSIs. The recovery of E faecalis was better with FNAM (10.2%) than with surface swab (3.1%) in wound infections following emergency surgeries (P = .0445).

Therefore, in certain cases with an unexplained wound infection, FNAM can be used as an investigation to identify specific pathogens not detected by conventional surface swab. Further studies analyzing the cost of performing FNAM in comparison to the costs related to the management of complications arising from wound infections may help in deciding the conscientious technique.

Affiliation: Jawaharlal Institute of Postgraduate Medical Education & Research (JIPMER), Pondicherry, Puducherry, India

Correspondence:
Sreenath GS, MS, FMAS
JIPMER,
Gorimedu,
Pondicherry, Puducherry 605006
India
dr.sreenathgs@gmail.com

Disclosure: The authors disclose no financial or other conflicts of interest.